Central Bank Policy: Theory and Practice 9781789737523, 9781789737516, 9781789737530

Table of contents :
Contents
List of Figures
List of Tables
About the Authors
Preface
Part I: General Review
Chapter 1: Introduction
1.1. The Central Bank and Economy
1.2. Central Bank, Academic Thinking, and Political Economy
1.3. Objectives of the Book
1.4. Systematics of the Writing
1.5. Utility of the Book
Chapter 2: Central Bank Evolution and Reform
2.1. Introduction
2.2. The Evolving Role of the Central Bank
2.2.1. Gold Standard Era and Real Bills Doctrine: 1840–1930
2.2.2. Central Banks under Government Control: 1940–1970
2.2.3. Central Bank Independence and Inflation Targeting: 1980–2007
2.2.4. Central Banks and the Global Financial Crisis of 2008/2009
2.3. Central Bank Policy Reform
2.3.1. Monetary Policy, Price Stability, and Exchange Rates
2.3.2. Exchange Rate Stability and Foreign Capital Flows
2.3.3. Monetary and Financial System Stability
2.3.4. Payment Systemic Stability
2.4. Central Bank Institutional Reform
2.4.1. Credibility, Independence, and Accountability
2.4.2. Transparency and Communication
2.4.3. Institutional Arrangements and Coordination
2.5. Concluding Remarks
Part II: Monetary Policy and Economy
Chapter 3: The Role of Money and Monetary Policy in the Economy
3.1. Introduction
3.2. Theoretical Review
3.2.1. The Role of Money in the General Equilibrium Analysis
3.2.2. Relationship between Money−Output and Imperfect Information
3.2.3. The Money−Output Correlation against Imperfect Market Competition and Price Rigidity
3.2.4. The Role of Money in the Monetarist Analysis Framework: “Money Matters” and the Keynesian-Neoclassical Synthesis
3.3. Empirical Evidence and Related Issues
3.3.1. Monetary Policy Modeling and Variable Selection Issues
3.3.2. Consensus Regarding the Role of Money and Monetary Policy
3.3.3. Empirical Evidence in Indonesia
3.4. Market Imperfections and the New Paradigm of Monetary Economics: Credit Matters
3.4.1. Credit Availability and Bank Behavior: Credit Rationing Equilibrium
3.4.2. A New Paradigm in Monetary Economics Theory
3.5. Concluding Remarks
Chapter 4: Exchange Rates and the Economy
4.1. Introduction
4.2. Exchange Rate Determination Theory
4.2.1. Mundell–Fleming Model: Policy Trilemma
4.2.2. Exchange Rate Model with Sticky Prices: Dornbusch’s Overshooting Model
4.2.3. Portfolio Balance Model and Central Bank Monetary Operations
4.2.4. Market Microstructure Models: Order Flows and Heterogeneous Information
4.3. Empirical Findings for Exchange Rates and the Economy
4.3.1. Exchange Rates and Trade
4.3.2. Exchange Rate and Inflation
4.3.3. Empirical Findings in Indonesia
4.4. Exchange Rate System and Policy
4.4.1. Choice of Exchange Rate System: Theoretical Review
4.4.2. Exchange Rate Systems in Practice: Bipolarization and Fear of Floating
4.4.3. Exchange Rate Crises: Causes and Impacts
4.4.4. Exchange Rate Stabilization Policy: Foreign Exchange Intervention
4.4.5. Exchange Rate System and Policy in Indonesia
4.5. Concluding Remarks
Chapter 5: Monetary Policy Transmission Mechanism
5.1. Introduction
5.2. Monetary Policy and the Transmission Mechanisms
5.2.1. MPTM Map
5.2.2. The Importance of Monetary Policy Transmission
5.2.3. Policy Rate and Monetary Operations
5.2.3.1. Interest Rate Policy. The interest rate policy adopted by the central bank will influence short-term interest rates on the interbank money market as well as market expectations of macroeconomic projections and the future direction of the central
5.2.3.2. Monetary Operations. The central bank performs monetary operations to influence liquidity conditions and, therefore, interest rates on the interbank money market in line with the policy rate set. To that end, the central bank projects bank liquid
5.2.3.2. Foreign Exchange Intervention. Exchange rate stability is very important for the economy, especially in EME and developing countries, therefore many central banks, including BI, pay a lot of attention to MPTM through the exchange rate channel. In
5.3. Monetary Transmission Channels: The Money View
5.3.1. Interest Rate Channel
5.3.2. Asset Price Channel
5.3.3. Exchange Rate Channel
5.3.4. Expectations Channel
5.4. Monetary Transmission Channels: The Credit View
5.4.1. Bank Lending and Bank Capital Channels
5.4.1.1. Bank Lending Channel. The first model of the bank lending channel was developed by Stiglitz and Weiss (1981) based on the assumption that the borrower has access to private information about the feasibility of his/her businesses. Despite producin
5.4.1.2. Bank Capital Channel. Bank capital also affects credit supply. Van den Heuvel (2002) studied the impact of the CAR and ability of banks to issue shares. Bank capital could be influenced by the ability to generate profit, the market price assessme
5.4.2. Balance Sheet Channel
5.4.2.1. External Finance Premium. The ideas of Bernanke, Gertler, and Gilchrist (1999) as well as Clastorm and Fuerst (2000) represent two salient references in the external finance premium model. For each business transaction, there is always the possib
5.4.2.2. Collateral Constraints. Financial accelerators can be analyzed using the collateral for loan transactions between lender and borrower. The seminal views of Kiyotaki and Moore (1997) represent an important reference in the analysis based on diffic
5.4.3. Risk-taking Channel
5.4.3.1. Risk-taking Dynamics. Financial theory in terms of investment teaches return optimization in line with the levels of risk of each respective alternative investment. This is achieved through risk diversification by forming an optimal investment po
5.4.3.2. The Role of Liquidity. Financial amplification through the risk-taking channel is also inextricably linked to liquidity, as an important element of financial transactions. There are two types of liquidity, namely funding (cash) liquidity and mark
5.5. Monetary Policy Transmission in Various Countries
5.5.1. Monetary Policy Transmission in Advanced Countries
5.5.1.1. Monetary Policy Transmission in the EU. Unification of the euro currency had an immediate impact on monetary policy transmission, with the loss of the exchange rate channel between countries in the EU. The economic and monetary union led to a uni
5.5.1.2. Monetary Policy Transmission in the United States. Bernanke and Blinder (1992) were among the first to apply a VAR model to map monetary policy transmission channels in the United States, revealing three important findings. First, the federal fun
5.5.2. Monetary Policy Transmission in EMEs
5.5.2.1. Interest Rate Channel, Asset Price Channel, Exchange Rate Channel, and Expectations Channel. A study by Moreno (2008) showed that, in many EME, the effect of the policy rate on deposit and lending rates is generally stronger and more persistent t
5.5.2.2. Credit and Balance Sheet Channels. Dominant banking sectors in many EME mean that the credit channel plays an important role in terms of influencing investment. The degree of influence varies, however, namely that it is stronger in Latin America
5.5.2.3. Monetary Policy Transmission in the Post-GFC Era. The GFC precipitated an important MPTM evolution in EMEs due to three salient factors, namely the relative role of banks on the bond market, money market globalization, and low long-term interest
5.5.3. Monetary Policy Transmission in Indonesia
5.5.3.1. Interest Rate, Exchange Rate, Asset Price, and Expectations Channels. A study by Kusmiarso et al. (2002) documented the growing role of the interest rate transmission channel in Indonesia. Specifically, using a money market structural model, the
5.5.3.2. Bank Lending and Balance Sheet Channels. Previous research provided complete empirical evidence concerning the bank lending channel in Indonesia (Agung, Morena, Pramono, and Prastowo, 2002a, 2002b). Three methods were used in the study. First, a
5.5.3.3. Transmission of Risk-taking Behavior and the Central Bank Policy Mix. Research performed by Satria and Juhro (2011) was the first empirical study concerning the impact of risk-taking behavior on lending by banks in Indonesia. The econometric mode
5.6. Concluding Remarks
Part III: Monetary Policy Framework
Chapter 6: Monetary Policy Strategic Framework
6.1. Introduction
6.2. Conceptual Dimension and Theoretical Models
6.2.1. Final Policy Target: Output Versus Prices
6.2.1.1. Targeting Long-term Policy Goals. Theoretically, the benefits of the respective policy strategies could be observed from the perspective of implementing monetary policy in the near term and long term. In this case, the characteristics of the econ
6.2.1.2. Targeting Short-term Policy Goals. In the near term, the monetary policy target selected, be it price stability or output stability, will be affected by shocks in the economy, irrespective of demand-side or supply-side sources. A simple analysis
6.2.2. Monetary Policy Regime
6.2.2.1. Exchange Rate Targeting. There are three alternative approaches to exchange rate targeting as a monetary policy strategy. First, by pegging the value of the domestic currency to certain commodities recognized internationally, such as gold (known
6.2.2.2. Monetary Targeting. In many countries, exchange rate targeting is not the primary choice of monetary policy strategy because of the requirements mentioned above, including monetary policy restrictions or flexibility to meet the domestic economic
6.2.2.3. Inflation Targeting. As the relationship between monetary aggregates and the final target of monetary policy faded, many countries began to adopt inflation targeting, which is implemented by publishing the medium-term inflation target and committ
6.2.2.4. No Explicit Anchor Targeting. Seeking to achieve the desired economic performance, monetary policy strategy can be implemented without an explicit anchor but still paying attention and providing commitment to achieve the final target of monetary
6.2.2.5. Nominal Income Targeting. Fundamentally, targeting nominal (output) income is based on the view that monetary policy can only affect nominal income growth rather than its determinants in the form of inflation and real output growth. In addition,
6.3. Empirical Studies and Related Issues
6.3.1. Short-term Output-prices Trade-off and the Phillips Curve Phenomenon
6.3.1.1. Presence of the Phillips Curve. The first issue relating to the emergence of empirical facts is stagflation, namely low economic growth coupled with high inflation, as experienced by industrialized nations in the 1970s, who opposed the trade-off
6.3.1.2. Linearity in the Phillips Curve. The second issue is based on the findings of Laxton, Meredith, and Rose (1995) concerning the asymmetric effect of economic activity on inflation in seven major OECD countries. The study provided an important conc
6.3.1.3. Formation of Expectations in the Phillips Curve. During development in the 1990s, studies of the Phillips curve tended to focus more on the effect of forming rational expectations on the trade-off between inflation and output per the Neo-Keynesia
6.3.2. Output-prices Trade-off in the Economic Cycle
6.3.3. Sacrifice Ratio: Cost of Controlling Inflation
6.4. Implementation in Several Jurisdictions
6.4.1. Final Target of Monetary Policy
6.4.1.1. A Shift in Priorities from the 1970s to 1980s. Economic dynamics around the world in the 1970s were characterized by the desire to stimulate economic growth and create job opportunities, consistent with the goal of economic development at that ti
6.4.1.2. From the 1990s Until the Present Day. Since the 1990s to the present day, nearly all central banks around the world have adopted price stability as the final target of monetary policy. In fact, most central banks have formally applied a monetary
6.4.2. Monetary Policy Regime
6.4.2.1. Exchange Rate Targeting. The exchange rate targeting regime was introduced by several countries as part of the efforts to control inflation during the 1980s. With the inherent advantages of the regime, exchange rate targeting successfully lowered
6.4.2.2. Monetary Targeting. Two countries that have seriously and successfully implemented monetary targeting are Germany and Switzerland. The success of such a policy on controlling inflation in those two countries is the reason why monetary targeting c
6.4.2.3. Inflation Targeting. As the relationship between monetary aggregates and the final policy target faded, inflation targeting emerged as a new monetary policy framework that appealed to countries previously plagued by inflation control problems in
6.4.2.4. No Explicit Anchor Targeting (Implicit Targeting). Fundamentally, several countries have practiced this strategy, evidenced by impressive macroeconomic performance (including low and stable inflation) without resorting to a nominal anchor, such a
6.5. The Monetary Policy Regime in Indonesia (1968–1998)
6.5.1. Economic Stabilization and Rehabilitation (1968–1972)
6.5.2. Era of Oil-based Economic Growth (1973–1982)
6.5.3. Period of Economic Deregulation, Debureaucratization, and Liberalization (1983–1997)
6.6. Concluding Remarks
Chapter 7: Monetary Policy Operational Framework
7.1. Introduction
7.2. Conceptual Dimension and Theoretical Models
7.2.1. Policy Instruments: Money Supply Versus Interest Rate
7.2.1.1. Standard Poole Model with Demand-side Shocks. Analyzing the ­conditions where each respective instrument is appropriate, Poole applied a simple Keynesian approach for goods market (investment savings (IS)) and money market (liquidity preference m
7.2.1.2. Poole Model Variations with Supply-side Shocks and the Role of ­Market Expectations. Fundamentally, the standard Poole model only calculated demand-side shocks (from the IS–LM equation). Poole model variations, however, also take into considerati
7.2.2. Policy Response: Rules Versus Discretion
7.2.2.1. Feedback Rule and Optimal Monetary Policy. Through development to the present day and consistent with money demand function instability (money multiplier), the use of constant rules, standard or soft, has started to fade, replaced by feedback rul
7.2.2.2. Simple Feedback Rule. One monetary rule, in the form of a simple feedback rule, is the monetary growth rule or McCallum rule. Where monetary growth reflects the policy response to the average change in base velocity, as a correction to the ongoin
7.2.3. Several Aspects of Optimal Monetary Policy Formulation
7.3 Application in Several Countries
7.3.1. Operational Targets and Monetary Policy Instruments
7.3.1.1. Operational Target: Monetary Aggregates Versus Interest Rates. In practice, the central bank may choose alternative operational targets, including short-term interest rates, monetary aggregates, and exchange rates. Departing from trends in advanc
7.3.1.2. Monetary Control Instruments. In terms of the monetary policy instruments used, in practice, there are two types of instruments employed by central banks around the world, namely direct and indirect market-based instruments. The choice of policy
7.3.2. Monetary Policy Response: Policy Rules
7.3.2.1. Implicit Interest Rate Policy Rule in Several Countries. A study by Clarida et al. (2000) provided a complete analysis framework for monetary policy implementation in the United States during the period from 1990 to 1996. The backward-looking Ta
7.3.2.2. Role of Monetary Aggregates When Applying an Implicit Rule. ­Congruent with the shift in the preferred operational target from monetary aggregates (money base) to short-term interest rates, particularly in advanced countries, use of the interest
7.4. Monetary Operations in Indonesia
7.4.1. Quantity-based Approach
7.4.2. The Shift from the Quantity-based Approach to the Price-based Approach
7.5. Concluding Remarks
Chapter 8: Inflation Targeting Framework: Concept and Implementation at Central Banks
8.1. Introduction
8.2. Conceptual Dimension and Theoretical Model
8.2.1.1. Rationale. Fundamentally, ITF is a framework where monetary policy can be directed toward achieving a future inflation target that is published transparently as a tangible form of central bank commitment and accountability. This definition contai
8.2.1.2. Characteristics. The rationale detailed above underlies the ITF-based monetary policy framework. In this regard, ITF is a monetary policy framework with a number of salient characteristics, namely an official statement from the central bank that
8.2.1.3. Advantages and Disadvantages. The growing number of countries applying ITF demonstrates that the framework offers several advantages over monetary policy regimes using money supply, the exchange rate or other anchors (Bernanke et al., 1999; Leide
8.2.2. Theoretical Model
8.2.2.1. Baseline ITF Macroeconomic Model. Consensus from both sides, New Keynesian and New Neoclassical Synthesis, has produced a viewpoint concerning the substance of ITF theory, namely that “inflation dynamics depend on expected (future) inflation and
8.2.2.2. ITF in the Inflation Forecast Targeting Format. According to Svensson (1997), in line with the forward-looking policy perspective, the substance of ITF could connote inflation forecast targeting. This has implications on the specification of the
8.3. Institutional and Operational Framework
8.3.1. Institutional Framework
8.3.1.1. Central Banking Laws. ITF implementation requires regulations in central bank laws stipulating price stability as the target of monetary policy and providing full authority to the central bank to achieve that target (instrument independence). In
8.3.1.2. Formulation of the Inflation Target. Formulation of the inflation target also varies among ITF countries in terms of the price index used, size of the target, target horizon, and publication method. Inflation target formulation primarily depends
8.3.1.3. Accountability and Transparency. All central banks applying ITF operate transparently because public accountability is an integral element of successful monetary policy. The importance of the accountability mechanism is clear because of the lag b
8.3.2. Operational Framework
8.3.2.1. Inflation Projections. In ITF, inflation projections play a key role in terms of monetary policy formulation, primarily considering the lag between monetary policy and its impact on inflation. Therefore, the ability to predict inflation will dete
8.3.2.2. Policy Transmission. As elucidated in the previous chapter, deep understanding of the monetary policy transmission mechanism is an integral element of implementing monetary policy. Such understanding becomes even more important in ITF because it
8.3.2.3. Policy Implementation. The operational target of monetary policy normally used by ITF central banks are short-term interest rates, usually overnight to 1 month. In addition to closely reflecting the actions of the central bank, short-term interes
8.4. Inflation Targeting Regimes
8.4.1. Regime and Rationale
8.4.2. Clarity of Commitment to the Inflation Target
8.4.3. Inflation Targeting Regime Credibility
8.4.4. Inflation Targeting Regime Classification
8.4.5. The Importance of Regime Classification
8.5. Inflation Targeting and Economic Performance
8.5.1.1. Inflation Targeting Successes. Mishkin and Schmidt-Hebbel (2001) are perhaps the best reference to draw several success stories during the first decade of ITF implementation in the 1990s from various previous empirical studies. Nevertheless, it i
8.5.1.2. Alternative Opinions. The conclusions made by Mishkin and Schmidt-Hebbel (2001) generally reflect the views of economists as well as central bankers applying ITF. The ITF success stories underlie the economists’ volition to propose ITF as the ide
8.5.2. Inflation Targeting after the GFC of 2008/2009
8.6. Concluding Remarks
Chapter 9: Inflation Targeting Framework: Implementation in Indonesia
9.1. Introduction
9.2. Institutional Framework
9.2.1. BI Act
9.2.2. Inflation Target Formulation
9.2.3. Accountability and Transparency
9.3. Operational Framework
9.3.1. Inflation Projection Techniques
9.3.2. Monetary Policy Transmission
9.3.3. Monetary Operations
9.4. Evolution Thus Far
9.4.1. Issues during the ITF Transition Period
9.4.2. An Evaluation of ITF Implementation since July 1, 2005
9.5. Strengthening the ITF Implementation Strategy after the Global Financial Crisis
9.6. Concluding Remarks
Part IV: Institutional Aspect of Central Bank Policy
Chapter 10: Monetary Policy Credibility and Time Consistency
10.1. Introduction
10.2. Conceptual Dimension of Policy Credibility
10.2.1 Causes of the Credibility Problem
10.2.2. Effort to Improve Credibility
10.3. Theoretical Model Concerning the Policy Credibility Problem: Time Inconsistency
10.3.1. Kydland and Prescott’s Model
10.3.2. Barro-Gordon Model
10.3.3. Credibility Problem Analysis with Supply Shocks
10.4. Empirical Studies: Measures of Credibility and Time Inconsistency
10.4.1. Parameterization Approach: Credibility Index
10.4.2. Qualitative Approach (Narrative)
10.4.3. Quantitative Modeling Approach
10.5. Assessing Monetary Policy Credibility in Indonesia
10.5.1. Policy Credibility in Indonesia during ITF Implementation
10.5.2. Empirical Studies of Time Inconsistency in Indonesia
10.6. Concluding Remarks
Chapter 11: Central Bank Independence and Accountability
11.1. Introduction
11.2. Conceptual and Empirical Dimensions of Independence and Accountability
11.2.1.1. Goal Independence. Fundamentally, the goal of monetary policy can be defined in broad terms, not only as the choice between price stability and output, but also the target horizon, the concrete indicators used, the value of the target and the es
11.2.1.2. Instrument Independence. Instrument independence implies that the central bank sets its own operational targets without government intervention. According to Bofinger (2001), independence has three important control elements as follows: (1) cont
11.2.1.3. Personal Independence. Despite enjoying goal independence and/or instrument independence, a central bank may struggle to achieve personal independence because the government could exert informal pressures on monetary policy implementation. For e
11.2.2. Central Bank Accountability
11.2.3. Correlation between Independence and Accountability
11.3. Theoretical Models Concerning Independence and Accountability
11.3.1. Delegation of Authority Theory: Conservative Agent
11.3.2. Central Bank Accountability Theory: Democratic Accountability
11.4. Empirical Studies and Related Issues
11.4.1. Central Bank Independence and Economic Performance
11.4.2. Related Issues
11.4.2.1. The Critical Issue of Central Bank Independence: Two Fallacies. There are at least two conclusions fundamental to the spectrum of studies on central bank independence, including the works of Rogoff (1985), Persson and Tabellini (1993), Alesina a
11.4.2.2. Ex Post and Ex Ante Accountability: The Difference between Accountability and Transparency. From the perspective of policy implementation timing, central bank accountability can fundamentally be debated in two forms, namely ex post and ex ante a
11.4.3. Policy Independence and Credibility during a Crisis Period
11.5. Bank Indonesia’s Independence and Accountability
11.5.1. Bank Indonesia’s Independence
11.5.2. Bank Indonesia’s Accountability
11.6. Concluding Remarks
Chapter 12: Policy Transparency and Communication Strategy
12.1. Introduction
12.2. Conceptual Dimension of Policy Transparency and Communication Strategies
12.2.1. Several Viewpoints Concerning Transparency
12.2.1.1. Scope of Transparency: Full Versus Limited. From an economic perspective, the main argument supporting the need for full transparency is that monetary policy would be most effective if correctly anticipated by the markets. That argument is base
12.2.2. Transparency, Monetary Policy Regime, and Democratic Accountability
12.2.2.1. Transparency and Democratic Accountability. The trend of democratization that has affected many countries, particularly in the past decade, has increased calls for transparent public policies, including monetary policy. The central bank is a pub
12.2.3. Monetary Policy Communication Strategy
12.2.3.1. Scope of Communication. The principle underlying monetary policy transparency is that the information communicated helps the public to understand, and anticipate, the central bank’s decisions as a logical conclusion to the policy sequence orient
12.2.3.2. Communications Methods. Sundarajan et al. (2003) proposed four basic dimensions that determine the credibility of policy transparency, namely: (1) the means or media used; (2) timeliness; (3) periodicity; and (4) the quality and scope of the inf
12.2.3.3. Communication Targets. To whom monetary policy transparency and communication is targeted is a reflection of the democratic accountability principles described previously. Blinder et al. (2003) suggested four main targets of central bank communi
12.3. Theoretical Models on Policy Transparency: Conservative Central Bank and Imperfect Transparency
12.3.1. Political Transparency: Public Uncertainty concerning the Preferred Weight of Output Stabilization (Parameter b)
12.3.2. Economic Transparency: Public Uncertainty to Changes in Economic Factors (value of parameter k)
12.3.3. Implications of Imperfect Transparency
12.4. Practices and Empirical Studies of Transparency and Communication Strategies in Different Countries
12.4.1. Assessing Transparency by the IMF
12.4.2. Practices at Several Prominent Central Banks
12.4.3. Monetary Policy Transparency and Economic Performance
12.4.3.1. Methodology. In their study, Chortareas et al. (2002) focused on the publication of economic forecasts by the central bank and its impact on inflation and output volatility. The hypothesis constructed from the nascent theories developing in the
12.4.4. Transparency and the Communication Strategy in a Crisis Period
12.5. Transparency and the Communication Strategy in Indonesia
12.5.2. Bank Indonesia’s Communication Strategy
12.6. Concluding Remarks
Chapter 13: Monetary Policy and Foreign Capital Flows
13.1. Introduction
13.2. Theoretical Dimension and Foreign Capital Flow Developments
13.2.1. Neoclassical Theory and PI Theory
13.2.2. Determinants of Foreign Capital Flows: Push and Pull Factors
13.3. Foreign Capital Flows and Economic Performance
13.3.1. Empirical Phenomenon: “Lucas Paradox”
13.3.2. Important Takeaways from the Lucas Paradox
13.3.2.1. Economic Fundamentals. The production and institutional structures in a particular country could create disparity between investment returns and the measures of marginal productivity of capital and labor. Several factors play an important role i
13.3.2.2. Domestic Financial Sector Deepening. The domestic financial sector plays a vital role in mobilizing savings and stimulating foreign capital flow intermediation efficiently for productive financing of economic growth. One neoclassical assumption
13.3.2.3. Imperfect International Finance. Imperfect international finance creates deviation between the actual and desired capital returns, thus engendering distortions between foreign capital flows, investment, and economic growth. Obstfeld and Rogoff (
13.4. Foreign Capital Flows and Monetary Stability
13.4.1. Empirical Studies on the Effect of Capital Flow Volatility
13.4.2. Monetary Policy Trilemma in an Open Economy
13.4.3. Foreign Capital Flows and Monetary Policy Dynamics in Indonesia
13.5. Foreign Capital Flow Management: Theories and Practices
13.5.1. Principles, Targets, and Instruments
13.5.1.1. Formulation and Implementation Principles. In general, the following conditions serve as a reference for when foreign CFM is applicable (IMF, 2012). First, foreign CFM is required when the space to make further macroeconomic policy adjustments b
13.5.1.2. Targets: Macroeconomic Stability and/or Financial System Stability. Foreign CFM requires a clear target, whether it be macroeconomic stability, financial system stability or both. The answer will depend upon the types and volatility of the capi
13.5.1.3. Instrument Selection: Prudential Regulations or Capital Controls. Which instruments are available to mitigate the risks associated with foreign capital flows? In general, the policy options depend on the conditions specific to the country invol
13.5.2. Foreign CFM Practices
13.5.2.1. Foreign CFM Effectiveness: Brazil and Colombia. Brazil and Colombia apply foreign CFM to mitigate the undesired impact on competitiveness and monetary policy autonomy (Ostry et al., 2011). In 2007, Colombia imposed unremunerated reserve requirem
13.5.2.2. Macroprudential Regulations on Capital Flows: South Korea. In South Korea, exporters – particularly shipbuilding companies with scheduled dollar receipts – were required to hedge against future export proceeds. Such exporters wished to sell thei
13.5.2.3. Prudential Regulations on Foreign Loans: Croatia. Foreign banks have dominated the Croatian banking system since the year 2000 and have played an important role in terms of foreign capital flows to the country. A sure of capital flows prompted i
13.6. Concluding Remarks
Chapter 14: Macroprudential Policy and Financial System Stability
14.1. Introduction
14.2. Conceptual Dimension of FSS
14.2.1. The GFC of 2008 / 2009 and FSS
14.2.2. Central Bank’s Role in FSS
14.2.3. Central Bank Macroprudential Policy
14.3. Theoretical Model and Empirical Evidence of Financial Procyclicality
14.3.1. Financial Accelerator Theory
14.3.2. Empirical Evidence for Financial Procyclicality
14.3.2.1. Asset Price Bubbles. Hunter, Kaufman, and Pomerleano (2003) and Evanoff, Kaufman, and Malliaris (2012) conducted literature reviews covering various aspects of asset price bubbles and the implications on monetary policy, regulation, and internat
14.3.2.2. Credit Booms and Busts. The importance of credit as an indicator to understand the financial cycle and its impact on a crisis was also emphasized by Schularick and Taylor (2012). Using data covering 140 years from 14 advanced countries for the p
14.3.2.3. Procyclical Capital Flows. Numerous studies have demonstrated procyclical capital flows. Kaminsky, Reinhart, and Vegh (2004), for instance, studied capital flow procyclicality and the interaction with fiscal and monetary policy. Using data from
14.4. Theoretical Model and Empirical Evidence for Interconnectedness and Financial Networks
14.4.1. Theory of Interconnectedness and Financial Networks
14.4.2. Empirical Evidence for Financial Interconnectedness
14.5. Macroprudential Policy Theory and Practices
14.5.1. Principles, Targets, and Instruments
14.5.2. Application in Various Jurisdictions
14.6. Concluding Remarks
Chapter 15: Central Bank Policy Mix
15.1. Introduction
15.2. Conceptual Dimension of the Central Bank Policy Mix
15.2.1. Integration of the Price Stability and Financial System Stability Targets
15.2.2. Basic Concepts of the Central Bank Policy Mix
15.2.3. Central Bank Policy Mix Transmission Mechanism
15.3. Structural Macroeconomic Modeling: Integrated Inflation Targeting
15.3.1. From Flexible to Integrated Inflation Targeting
15.3.2. Model Structure and Analysis Framework
15.3.3. Central Bank Policy Mix Formulation
15.4. DSGE Modeling with Macrofinancial Linkages
15.4.1. Macrofinancial Modeling Approaches in DGSE
15.4.2. Monetary Policy and Macroprudential Policy in DSGE Modeling
15.4.3. Central Bank Policy Mix Formulation
15.5. Bank Indonesia Policy Mix
15.5.1. Policy Mix Targets and Instruments
15.5.2. Interest Rate and Exchange Rate Policies
15.5.3. Foreign Capital Flow Management
15.5.4. Macroprudential Policy
15.5.5. Institutional Arrangements
15.6. Concluding Remarks
Bibliography
Index

Citation preview

CENTRAL BANK POLICY

Praise for Central Bank Policy: Theory and Practice The book rigorously discusses the theories, empirical studies and practices of international financial and monetary policy in emerging market economies, including Indonesia. Furthermore, the book is an invaluable policymaking input for the central bank and government, teaching material for lecturers and students as well as an important reference for scientific development in academia. Prof. Boediono, Vice President of the Republic of Indonesia (2009–2014); Governor of Bank Indonesia (2008–2009) The world has become increasingly challenging for central banks in both developed and emerging economies. While much has been written on central banks policies in the developed world, less known are the remarkable successes achieved by central banks in the emerging economies. This publication is a valuable source for policy makers in central banking. It not only offers knowledge on the theoretical foundations and institutional arrangements but also on the practical aspects of the policy tools that are at the disposal of central banks, particularly in the emerging world. Dr Zeti Akhtar Aziz, the 7th Governor (2000–2016) of Bank Negara Malaysia, Malaysia’s central bank This book is a major contribution to the theory and practice of central banking in emerging market economies. Drawing on the accumulated wisdom of many years of academic study and high-level policy experience, the authors provide an encyclopedic yet highly accessible survey and analysis hat bridges theory and practice. No stone is left unturned in this comprehensive study, drawing as it does on economic history, the international monetary system, globalization, and the political economy of macroeconomic policy making. The volume will be invaluable for a wide audience, including advanced undergraduate and graduate students, academic researchers, policy makers, financial market analysts, and anybody with an interest in contemporary macroeconomic challenges and issues. A highly recommended publication. Prof. Hal Hill, H.W. Arndt Professor Emeritus of Southeast Asian Economies, Australian National University When a pure academic writes a book, it lacks practical knowledge and connections; when a pure policy maker writes a book, it lacks academic rigor; but, when an academic who is also an experienced policy maker writes a book, the resultant is a balanced book. It is this balance that the literature lacks and for this reason alone this book, by two experienced academics who have decades of central banking policy making experience, will remain unique. This book is an amalgam of theoretical percepts, empirical case studies and practical policy debates. While existing reference material would either be too empirical or too theoretical and almost always short of practical policy discussions, this book is far from it. A gap in the history of central banking policies and practices is now immaculately covered. It makes understanding central banking policies and practices easy. The book is both inspirational and thought-provoking. The value and impact of this book will be long lasting. Prof. Paresh Kumar Narayan, Alfred Deakin Professor, Deakin University

CENTRAL BANK POLICY: THEORY AND PRACTICE

BY

PERRY WARJIYO AND SOLIKIN M. JUHRO

United Kingdom – North America – Japan – India – Malaysia – China

Emerald Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2019 Copyright © 2019 Emerald Publishing Limited Reprints and permissions service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. Any opinions expressed in the chapters are those of the authors. Whilst Emerald makes every effort to ensure the quality and accuracy of its content, Emerald makes no representation implied or otherwise, as to the chapters’ suitability and application and disclaims any warranties, express or implied, to their use. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-78973-752-3 (Print) ISBN: 978-1-78973-751-6 (Online) ISBN: 978-1-78973-753-0 (Epub)

Contents

List of Figures

ix

List of Tables

xi

About the Authors

xiii

Preface

xv

Part I  General Review Chapter 1 Introduction 1.1. The Central Bank and Economy 1.2. Central Bank, Academic Thinking, and Political Economy 1.3. Objectives of the Book 1.4. Systematics of the Writing 1.5. Utility of the Book Chapter 2 Central Bank Evolution and Reform 2.1. Introduction 2.2. The Evolving Role of the Central Bank 2.3. Central Bank Policy Reform 2.4. Central Bank Institutional Reform 2.5. Concluding Remarks

3 4 6 9 10 16 19 19 20 31 39 44

Part II  Monetary Policy and Economy Chapter 3 The Role of Money and Monetary Policy in the Economy 3.1. Introduction 3.2. Theoretical Review 3.3. Empirical Evidence and Related Issues 3.4. Market Imperfections and the New Paradigm of Monetary Economics: Credit Matters 3.5. Concluding Remarks

49 49 50 63 69 78

vi   Contents

Chapter 4 Exchange Rates and the Economy

81

4.1. Introduction 4.2. Exchange Rate Determination Theory 4.3. Empirical Findings for Exchange Rates and the Economy 4.4. Exchange Rate System and Policy 4.5. Concluding Remarks

98 101 112

Chapter 5 Monetary Policy Transmission Mechanism

115

5.1. Introduction 5.2. Monetary Policy and the Transmission Mechanisms 5.3. Monetary Transmission Channels: The Money View 5.4. Monetary Transmission Channels: The Credit View 5.5. Monetary Policy Transmission in Various Countries 5.6. Concluding Remarks

81 82

115 116 124 129 140 157

Part III  Monetary Policy Framework Chapter 6 Monetary Policy Strategic Framework 6.1. Introduction 6.2. Conceptual Dimension and Theoretical Models 6.3. Empirical Studies and Related Issues 6.4. Implementation in Several Jurisdictions 6.5. The Monetary Policy Regime in Indonesia (1968–1998) 6.6. Concluding Remarks Chapter 7 Monetary Policy Operational Framework 7.1. Introduction 7.2. Conceptual Dimension and Theoretical Models 7.3. Application in Several Countries 7.4. Monetary Operations in Indonesia 7.5. Concluding Remarks Chapter 8 Inflation Targeting Framework: Concept and Implementation at Central Banks 8.1. Introduction 8.2. Conceptual Dimension and Theoretical Model 8.3. Institutional and Operational Framework 8.4. Inflation Targeting Regimes 8.5. Inflation Targeting and Economic Performance 8.6. Concluding Remarks

161 161 162 170 176 186 190 193 193 194 205 216 220 223 223 224 234 247 252 259

Contents    vii

Chapter 9 Inflation Targeting Framework: Implementation in Indonesia 9.1. Introduction 9.2. Institutional Framework 9.3. Operational Framework 9.4. Evolution Thus Far 9.5. Strengthening the ITF Implementation Strategy after the Global Financial Crisis 9.6. Concluding Remarks Appendix 9.1: Amendments to the Bank Indonesia Act and their Implications Appendix 9.2: Bank Indonesia Macroeconomic Models

261 261 263 266 271 277 281 283 287

Part IV  Institutional Aspect of Central Bank Policy Chapter 10 Monetary Policy Credibility and Time Consistency

291

10.1. Introduction 10.2. Conceptual Dimension of Policy Credibility 10.3. Theoretical Model Concerning the Policy Credibility Problem: Time Inconsistency 10.4. Empirical Studies: Measures of Credibility and Time Inconsistency 10.5. Assessing Monetary Policy Credibility in Indonesia 10.6. Concluding Remarks

291 293

Chapter 11 Central Bank Independence and Accountability

319

11.1. Introduction 11.2. Conceptual and Empirical Dimensions of Independence and Accountability 11.3. Theoretical Models Concerning Independence and Accountability 11.4. Empirical Studies and Related Issues 11.5. Bank Indonesia’s Independence and Accountability 11.6. Concluding Remarks

319

Chapter 12 Policy Transparency and Communication Strategy

349

12.1. Introduction 12.2. Conceptual Dimension of Policy Transparency and Communication Strategies 12.3. Theoretical Models on Policy Transparency: Conservative Central Bank and Imperfect Transparency

349

298 306 311 318

320 327 337 344 346

350 365

viii   Contents

12.4. Practices and Empirical Studies of Transparency and Communication Strategies in Different Countries 12.5. Transparency and the Communication Strategy in Indonesia 12.6. Concluding Remarks

370 380 384

Part V  New Paradigm of Central Bank Policy Chapter 13 Monetary Policy and Foreign Capital Flows 13.1. Introduction 13.2. Theoretical Dimension and Foreign Capital Flow Developments 13.3. Foreign Capital Flows and Economic Performance 13.4. Foreign Capital Flows and Monetary Stability 13.5. Foreign Capital Flow Management: Theories and Practices 13.6. Concluding Remarks Chapter 14 Macroprudential Policy and Financial System Stability 14.1. Introduction 14.2. Conceptual Dimension of FSS 14.3. Theoretical Model and Empirical Evidence of Financial Procyclicality 14.4. Theoretical Model and Empirical Evidence for Interconnectedness and Financial Networks 14.5. Macroprudential Policy Theory and Practices 14.6. Concluding Remarks Chapter 15 Central Bank Policy Mix 15.1. Introduction 15.2. Conceptual Dimension of the Central Bank Policy Mix 15.3. Structural Macroeconomic Modeling: Integrated Inflation Targeting 15.4. DSGE Modeling with Macrofinancial Linkages 15.5. Bank Indonesia Policy Mix 15.6. Concluding Remarks

387 387 388 395 400 409 417 423 423 424 435 444 452 459 461 461 463 473 486 502 514

Bibliography

517

Index

553

List of Figures

Figure 3.1: Bank-optimal Rate. 73 Figure 3.2: Market Equilibrium Loanable Funds. 73 Figure 3.3: Equilibrium Credit Rationing. 74 Figure 4.1: Exchange Rate Determination Theory. 83 Figure 4.2: Policy Analysis in the Mundell–Fleming Model. 86 Figure 4.3: Policy Analysis According to Dornbusch’s Overshooting Model. 90 Figure 4.4: Analysis of Monetary Operations in the Portfolio Balance Model. 92 Figure 5.1: Monetary Policy and the Transmission Mechanism 118 Figure 5.2: Business Decisions without Collateral Constraints. 136 Figure 5.3: Business Decisions with Collateral Constraints. 136 Figure 6.1: Demand-side Shocks: Targeting Prices and Nominal Income. 165 Figure 6.2: Supply-side Shocks: Targeting Prices and Nominal Income. 166 Figure 6.3: Output Loss Linked to Disinflationary Policy. 176 Figure 10.1: Indifference Curve of the Time Inconsistency Problem.300 Figure 10.2: Policy Strategy and the Social Cost. 303 Figure 10.3: Core and Headline Inflation Trends. 312 Figure 11.1: Output Stabilization Preference Behavior. 331 Figure 11.2: Monetary Policy Behavior (c > 0). 335 Figure 11.3: Monetary Policy Behavior after Lowering c.336 Figure 11.4: Monetary Policy Behavior after Lowering h.337 Figure 11.5: Ex Post Accountability and Central Bank Reaction. 341 Figure 12.1: Measuring the Transparency Index using the Guttman Scale. 376 Figure 13.1: Aggregate GDP Per Capita of Capital Exporters (Weighted Current Account). 396 Figure 13.2: Foreign Capital Flows to EMEs and Developing Countries (Annual Average in Billions of USD). 397 Figure 13.3: Monetary Policy Trilemma. 402

x    List of Figures Figure 14.1: Linkages between Macro–Micro Policy and FSS. Figure 14.2: Financial Cycle and Economic Cycle Procyclicality. Figure 14.3: Incomplete Networking. Figure 14.4: Disconnected Network. Figure 14.5: Complete Network. Figure 14.6: Interconnectedness through Three Banks. Figure 14.7: Money Center Network. Figure 14.8: UK Banking Network. Figure 14.9: Global Banking Network. Figure 15.1: Integration of the Monetary Policy and Macroprudential Policy Framework. Figure 15.2: Macroprudential and Monetary Policy: Maintaining FSS in the Cyclical Dimension. Figure 15.3: Monetary and Macroprudential Policy Mix Cycle Maintains Macroeconomic and Financial System Stability in a Cyclical Dimension. Figure 15.4: Coordination and Losses after a Demand Shock. Figure 15.5: Coordination and Losses after a Supply Shock.

428 436 445 445 446 448 449 450 450 468 471 482 486 487

List of Tables

Table 6.1: The Central Bank’s Legal Monetary Policy Framework. 180 Table 6.2: Average Economic Growth and Inflation in Several Countries from 1991 to 2014 (%). 181 Table 7.1: Monetary Policy Operational Targets (2015). 206 Table 7.2: Monetary Policy Instruments in Several Countries (Until 2004). 210 Table 7.3: US Monetary Policy Reaction Function. 213 Table 8.1: Characteristics of Inflation Targeting. 227 Table 8.2: Provisions in Central Bank Laws. 235 Table 8.3: Price Indexes Used. 237 Table 8.4: Inflation Target: Point or Range. 238 Table 8.5: Inflation Target Horizon. 239 Table 8.6: Announcing the Inflation Target. 240 Table 8.7: Inflation Targeting Regime Classification. 251 Table 9.1: Actual and Target Inflation. 276 Table 11.1: Central Bank Personal Independence. 324 Table 11.2: Central Bank Accountability. 326 Table 11.3: Central Bank Independence and Accountability Index. 327 Table 12.1: Monetary and Financial Policy Transparency Index by Country. 373 Table 12.2: Economic Forecast Publication Profile of Central Banks.375 Table 12.3: Transparency and Inflation: Correlation Analysis. 377 Table 12.4: Transparency and Output Volatility: Correlation Analysis.377 Table 13.1: Global Foreign Capital Flows (% of GDP). 391 Table 13.2: Determinants of Foreign Capital Flows: Push Versus Pull Factors. 394 Table 13.3: Monetary Policy Trilemma Index in Indonesia. 407 Table 13.4: A Decomposition of the Dynamic Correlation between Exchange Rates Versus Foreign Capital Inflows and Interest Rates. 407 Table 14.1: Examples of Macroprudential Policy Instruments. 432 Table 14.2: Interaction between the Business Cycle, Risk-taking Behavior and the Financial Cycle. 438

xii    List of Tables Table 14.3: Macroprudential Instruments Applied in Asian Countries.454 Table 14.4: Macroprudential Instruments: Types and Risk Dimensions.458 Table 15.1: Effect of Monetary Policy of FSS. 472 Table 15.2: Integration of Price Stability and FSS: Three Modeling Approaches. 475 Table 15.3: Central Bank Framework: Integrated Inflation Targeting.478 Table 15.4: Three Approaches to Model Financial Frictions in DSGE. 489 Table 15.5: Four Cases of Price Stability and Financial System Stability. 506

About the Authors

Dr. Perry Warijiyo is currently the Governor of Bank Indonesia. Before serving as Governor, he was the Deputy Governor of Bank Indonesia (2013–2018). Previously, he also served as the Assistant Governor for monetary, macroprudential, and international policies. Dr Perry also held the prestigious position of Executive Director of the International Monetary Fund (IMF), representing 13-member countries in the SouthEast Asia Voting Group (2007–2009). He has a long, impressive, and outstanding career serving Bank Indonesia since 1984. His contributions have been primarily in economic and monetary policy research, international issues, organizational transformation and monetary policy strategies, education and research on central banks, and management of foreign reserves and external debts. Significantly, also, he has been the Editor-in-Chief of Bank Indonesia’s flagship journal, the Bulletin of Monetary Economics and Banking. He is also an active Postgraduate Lecturer at the University of Indonesia and at several top ranked universities in the country. In addition, he has authored and published a number of books, journals, and papers on economy, monetary, and international issues. He has offered numerous high-level keynote speeches that have influenced public policy. He received his Bachelor’s degree from Gajah Mada University, Indonesia, and Master’s and Ph.D. degrees in International Monetary Economics and Finance from Iowa State University, USA.

xiv    About the Authors Dr. Solikin M. Juhro is an Executive Director and the Head of Bank Indonesia Institute, Bank Indonesia. In his distinguished career at Bank Indonesia spanning more than 20 years, he has been extensively involved with monetary economic policy research and analysis, has spoken on academic and central banking issues at various national and international fora, and is an instrumental Editor of the Bank Indonesia flagship journal, the Bulletin of Monetary Economics and Banking. His current research focuses on macroeconomic transformation, central bank policy, and frontier issues in economic development, some of which have been published international outlets. Dr Juhro is currently a member of the SEACEN Advisory Group for Macroeconomic and Monetary Policy Management, representing Bank Indonesia. He was active in the Pacific Economic Outlook Structural Specialist projects in Osaka, Japan, from 2005 to 2015. He is also an active Postgraduate Lecturer at the University of Indonesia. He completed his Bachelor’s degree in Economics at the Airlangga University, a Master’s degree in Applied Economics at the University of Michigan, Ann Arbor, USA, a master’s degree in Economics at the University of Maryland, College Park, USA, and a doctoral degree in Economics (with cum laude) at the University of Indonesia, Indonesia.

Preface

Alhamdulillah, with all praise to Allah, the most glorified and exalted, we would like to present this book entitled Central Bank Policy: Theory and Practice to the readers, drawing fully on our extensive capabilities, knowledge, and experience to write a book that, to our understanding, does not exist elsewhere, at least in Indonesia. This book also represents an important contribution to central bank and government policymaking as well as teaching materials for lecturers and students and a salient reference for scholarly development in Indonesia and internationally. The idea for the book emerged from the experiences and concerns of the authors while teaching postgraduate international monetary and financial economics at the University of Indonesia. The teaching materials presented consist of a combination of theoretical studies, empirical reviews, and policy practices at central banks, which garnered an enthusiastic response from the students. Not only could the students explore and deepen the latest theories being developed in academia and empirically model the most interesting research topics, the inclusion of policy practices at central banks also embedded the students into the real world. Consequently, the courses were enlivened by interesting questions and discussions that enriched and enhanced the quality of the teaching. Unfortunately, we were unable to find a single reference book complete with a combination of theoretical studies, empirical reviews, and policy practices at central banks, thereby motivating us to write this book. This was, admittedly, no mean feat due to the rapid development of policy theories and practices at central banks as well as the fact that most of the materiel had not yet appeared in print as a reference for this book. Fortunately, we could draw on our vast experience and direct involvement with policymaking at Bank Indonesia (BI) as well as our active roles as resource persons or discussants at various international fora. Additionally, one of the author’s two-year tenure as Executive Director at the International Monetary Fund from 2007 to 2009 coincided with the global financial crisis, which further served to enrich the understanding and knowledge poured into this book, including the BI policy mix initiated in 2010 as a new policy paradigm at the central bank. This book is firmly believed to be a first in terms of comprehensively discussing the latest central bank policy theories and practices. The inclusion of the latest material, coupled with clear and logical systematics of the writing, ensures this book’s position as an unequivocal reference for policymakers at the central bank, practitioners, and academia. For academia, the book represents an important and leading reference for lecturers and students alike in terms of monetary and

xvi   Preface financial economics at the intermediate and advanced levels of master’s and doctoral programs as well as the latter stages of bachelor’s programs. The panoply of theoretical and empirical references contained in this book will facilitate future research among students and researchers. Similar benefits will also be available to economists and those in the business community. Although the discussions are predominantly on a philosophical conceptual level, accompanied by in-depth theory, the rationale and writing of this book is presented in a manner that is readable and easily digestible. To remain relevant and contemporary, the materials contained in this book will periodically be updated as per the latest developments and requirements in terms of new and innovative policy practices at the central bank. To that end, constructive criticisms and feedback from the readers are warmly welcomed in order to hone the contents of this book. In closing, the authors would like to acknowledge the important contributions, direct and indirect, of various parties to produce this book. On top of the gratitude felt for the publication of this book, the authors would also like to thank the contributors who could not be named individually, particularly colleagues at the BI Institute for their help in finalizing the book. Hopefully, this publication will expand scientific understanding and knowledge. Perry Warjiyo, S.E., M.Sc., Ph.D. and Dr Solikin M. Juhro, S.E., M.A.E., M.A. Jakarta, October 2018

Part I

General Review

This page intentionally left blank

Chapter 1

Introduction Any news concerning central bank policy decisions always garners public attention. Statements from the governor of the central bank in the United States, the Chair of the Federal Reserve System (the Fed), or indeed other members of the Federal Open Market Committee (FOMC), about the Federal Funds Rate (FFR), for instance, are always eagerly awaited by markets around the world.1 In fact, indications of the FFR direction, which are typically linked to statements from the Chair of the Fed or other FOMC members about US inflation and the economy, usually become a source of market speculation. Such conditions have become unequivocal since the Fed announced in May 2013 its plan to normalize the ultra-loose monetary policy stance adopted after the global financial crisis (GFC) of 2008/2009. This stance was to support the US recovery, which subsequently became known as the Fed’s Taper Tantrum. Actual decisions, or even just indications of the policy rate direction, directly influence money market interest rates, dollar exchange rates on the foreign exchange market, and stock prices on Wall Street. Such developments lead to reactions of interest rate, exchange rate, and stock prices in a number of countries, including Indonesia. Similarly, in Indonesia, statements released by Bank Indonesia (BI) are constantly in the news across various mass media outlets. The decisions of the BI Board of Governors regarding the BI Rate or even just indications about the possible future direction of the policy rate are increasingly becoming a reference for the markets and banks in terms of the financial transactions undertaken.2 In practice, non-resident investors, specifically those that invest in financial

1

See, for example, the “Transcript of Chair Yellen’s Press Conference,” December 16, 2015, Board of Governors of the Federal Reserve System (2015), http://www.federalreserve.gov. At the press conference, Chair Janet Yellen relayed the FOMC decision to hike the FFR by 25 bps, from 0.25% to 0.50%. This follows seven years of a near 0% rate in order to support the US recovery in the wake of the worst GFC in the United States since the Great Depression in the 1930s. 2 Pursuant to the BI announcement, dated April 15, 2016, effective from August 19, 2016 the BI Rate, which is equivalent to the 12-month money market rate, will be replaced by the BI 7-day (Reverse) Repo Rate, representing the seven-day money market rate. Central Bank Policy: Theory and Practice, 3–17 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191003

4    Central Bank Policy instruments such as government bonds (SUN), stocks, BI Certificates (SBI), and corporate bonds in Indonesia, base their investment decisions on statements regarding the BI Rate. In addition to policy rate decisions, the BI policy in relation to macroprudential regulation and supervision, including down payments on automotive loans as well as the maximum loan-to-value (LTV) ratio on property and housing loans, also attracts public attention. So too does payment system policy, such as money supply, clearing, card-based payment instruments, and electronic money. Decisions relating to the policy rate, macroprudential policy and the payment system are based on rigorous assessments of global and domestic macroeconomic and financial system developments and projections that are regularly delivered by BI.3

1.1. The Central Bank and Economy The brief overview in the preceding section demonstrates the importance of the central bank’s role in the economy. Since inception, the central bank has been mandated to maintain domestic currency stability (inflation and exchange rates) as well as financial system stability. Maintaining currency and financial system stability is the primary contribution of the central bank toward supporting sustainable economic growth. Therefore, the functions of monetary and payment system policy, as well as the regulation and supervision of financial institutions, fall under the auspices of the central bank. Initially, such functions were comparatively simple but have become increasingly complex over time in line with global and domestic economic and financial development. This can be gauged from the changing role of the monetary sector, from the commercial paper predominance (rediscounted by the Bank of England in the seventeenth century) to monetary operations targeting interest rates. The function of the payment system, on the other hand, has also progressed from merely printing and circulating banknotes and coins to the regulation and supervision of payment instruments, mechanisms, and infrastructure. Likewise, in terms 3

See, for instance, the BI Press Release, dated December 17, 2015, that announced the BI Rate would be held at 7.50% in anticipation of pressures from global financial markets, including the impact of the FFR hike. This resulted despite stable macroeconomic conditions at home. Indonesia, for instance, recorded low inflation and a sustainable current account deficit, which provided adequate space to ease the monetary policy stance. The BI Rate was subsequently lowered at the BI Board of Governors’ Meeting in January, February, and March 2016 by a total of 75 bps to 6.75%. At the BI Board of Governors’ Meeting in June 2016, in addition to reducing the BI Rate by 25 bps to 6.50%, BI also introduced macroprudential policy easing by raising the LTV ratio to boost lending and economic growth. In addition to news in the mass media, the decisions of the BI Board of Governors are published as press releases and posted to the official website of BI at http://www.bi.go.id. The BI also regularly publishes a quarterly Monetary Policy Report and monthly Monetary Policy Review that contains assessments of economic, financial, and monetary developments and projections that underlie BI Rate policymaking.

Introduction    5 of financial system stability, the function of the central bank has progressed from been a lender of last resort (LOLR) to macroprudential regulation and supervision. The monetary policy instituted by the central bank to influence money supply and interest rates is one determinant of macroeconomic stability; in particular, inflation and exchange rate stability. In the financial sector, the interest rate policy and exchange rate stabilization policy of the central bank directly influence bank funds and credit, as well as stock and bond prices on the capital market. Subsequently, through the influence of such monetary and financial developments on consumption, investment, exports, and imports, monetary policy also influences inflation, economic growth and, therefore, the creation of employment opportunities in addition to the balance of payments. In other words, the monetary policy stance adopted by the central bank has a major influence over public prosperity and welfare. Similarly, microprudential and macroprudential regulation and supervision are imperative in terms of maintaining financial system stability. Microprudential regulation entails microregulation and supervision of financial institutions and focuses on the soundness and performance of each individual financial institution. Meanwhile, macroprudential regulation involves macroregulation and supervision of financial institutions and focuses on systemic risk in order to achieve financial system stability. In many countries, macroprudential and microprudential regulation and supervision fall under the auspices of the central bank. In contrast, several countries, including Indonesia, Australia, and South Korea, transferred the microprudential function to a financial services authority, while the central bank maintained control over macroprudential aspects. In its implementation, central banks direct macroprudential policy toward dampening accelerators in the financial cycle, while preventing and detecting a build-up of systemic risk that leads to financial system instability. Regulations concerning the LTV for housing loans as well as down-payments on automotive loans in Indonesia are concrete examples of macroprudential policy to combat excessive credit growth in both sectors that could disrupt financial system stability. In terms of the payment system, currency is printed and circulated in line with the economic requirement and, therefore, supports monetary policy to achieve price stability and macroprudential policy to maintain financial system stability. In addition, payment system policy includes the reliable, efficient, and secure transfer of funds, clearing and financial transaction settlement, retail and wholesale, in the economy. Various payment instruments have also been developed in line with the advancement of financial product innovation and information technology development, such as automated teller machines (ATM), debit and credit cards, mobile and internet banking as well as electronic money. Even today, in the era of technological progress, the development of financial technology, or FinTech, has fundamentally changed the business models of various financial services offered by financial institutions to the public. Therefore, payment system development, monetary and financial system stability will affect each other, with all three considered crucial for economic advancement.

6    Central Bank Policy Due to the significant influence policy has on the economy and public welfare, it is understandable that the public, business, and financial communities, as well as government and parliament take an avid interest in the central bank. That interest extends beyond policy aspects to the institutional arrangements of the central bank in line with the ongoing change in the global economic and political landscape, with more and more countries applying a market-based economy and democratically elected governments. The institutional arrangements of the central bank are reinforced through the principles of good governance by strengthening the legal framework and in terms of policy implementation. This can be seen by the modernization of prevailing laws that provide a clear mandate for the central bank as well as independence in the execution of its duties. Furthermore, greater accountability and transparency from the central bank is now in increasing demand in terms of policy implementation. The various institutional arrangements of the central bank have been an ongoing concern for the past two decades and became increasingly important in the wake of the GFC. These changes reflect growing public awareness in various countries of the need to strengthen the role and standing of the central bank in order to support achievement of the economic policy targets.

1.2. Central Bank, Academic Thinking, and Political Economy From a central bank standpoint, the challenges faced in terms of carrying out the mandate are onerous. In fact, the challenges have become even more complex since the onset of the GFC. In the implementation of monetary policy, for instance, maintaining low and stable inflation has become increasingly important to support economic growth and ameliorate public welfare. The volatility of nonresident capital flows and exchange rates in emerging market economy (EME), including Indonesia, have also increased since the GFC with monetary policy divergence stemming from ultra-loose monetary policy in advanced countries and ubiquitous uncertainty blighting global financial markets. The GFC also showed that price and exchange rate stability alone are insufficient to maintain financial system and macroeconomic stability to support sustainable economic growth. Increasingly rapid and complex development in terms of product innovation and financial operations on the one hand has facilitated economic financing but, on the other hand, has also amplified the risk of financial system instability and affected the monetary policy transmission mechanism in the economy. Likewise, payment instrument development has necessitated increased regulation and supervision by the central bank in order to maintain a reliable, efficient, and secure payment system. Institutionally, the strengthening of consistency, independence, and accountability as well as transparency and communication have become progressively more important, not only in terms of governance but also to support policy effectiveness and credibility and, therefore, the reputation of the central bank. Throughout their evolution, economic conditions, political environment, and academic thinking have influenced how a central bank implements its mandate.

Introduction    7 At its inception in the seventeenth century, for instance, the role of the central bank in the economy emphasized creating and circulating currency, purchasing government debt, and functioning as LoLR for financial system stability. Application of the gold standard strengthened central bank credibility in terms of achieving price, exchange rate, and financial system stability because the central bank was charged with maintaining currency convertibility in line with the gold reserves held. Nonetheless, government control over the central bank to finance the post-World War recovery triggered soaring inflation, leading to an economic crisis. Furthermore, central bank credibility was also lost. Developments over the two decades prior to the GFC saw central banks focus shifted on price stability. This was in response to soaring inflation, while providing the central bank independence from government control. On the other hand, however, the focus on price stability was also based on the growing acceptance of New Neoclassical and New Keynesian synthesis, namely that monetary policy only influences inflation in the long term, despite a short-term trade-off between inflation and economic growth in line with the findings of the Phillips Curve. The economy is assumed to always be in equilibrium and the primary causes of imbalances are price and wage rigidities. Similarly, there is no friction in the financial system, therefore currency and credit are perfect substitutes that are affected by interest rates. Furthermore, non-resident capital flows freely and, therefore, a fully flexible exchange rate system is the optimal choice. Consequently, central bank policy merely needs to focus on stipulating the short-term interest rate in order to achieve price stability, while economic and financial equilibrium will generate economic growth and maintain macroeconomic and financial system stability. Central bank policy governance is strengthened by independence and is in line with rational expectations theory, policy consistency with the rules and the importance of transparency to form and anchor expectations. Such theories and ideas compelled central banks in many countries to target price stability, thereby adopting the Inflation Targeting Framework (ITF). The GFC, however, turned central bank practices and theories on their head; not because ITF policy had failed but, in contrast, because ITF had successfully lowered inflation in many countries, coupled with low interest rates that had stimulated rapid economic growth. The problem was that long-term stability and economic boom had led to excessive credit growth, asset price bubbles (stocks and housing) as well as high leverage. Financial accelerators caused the financial cycle to amplify the economic cycle. Furthermore, economic stability led to financial system instability that ultimately culminated in the worst crisis (namely, the GFC) since the Great Depression of the 1930s. Evidently, price stability alone is insufficient to ensure macroeconomic stability if financial system stability is not also maintained; “there is no macrostability without financial stability.” In reality, financial friction is inevitable due to asymmetric information, financial product innovation, price setting, and valuation as well as risk-taking behavior, hence the financial system is constantly in a state of flux that produces financial accelerators and turns fragility to systemic risk. Consequently, the GFC taught an invaluable lesson that the central bank should return to its original mandate, namely to achieve and maintain the stability of

8    Central Bank Policy domestic currency values (inflation and exchange rate), while supporting financial system stability. Central bank credibility and the framework that had been established through ITF became the foundation to achieve that mandate. Nevertheless, that was still not enough. Macrofinancial linkages demanded a macroprudential policy response from the central bank to mitigate procyclicality between the financial sector and economic activities that trigger economic and financial crises, such as the GFC. The policies of foreign exchange market intervention and nonresident capital flow management were also required to stabilize the exchange rate, while remaining in a flexible regime. Such developments encouraged many central banks to apply macroprudential policy and manage foreign capital flows in order to strengthen the effectiveness of monetary policy. In brief, the monetary and macroprudential policy mix became the new central banking paradigm in various countries after the GFC, including BI in 2010. This innovation in central bank policy practices also demands the development of underlying theories and empirical studies. The previous description illustrates that the theory underlying central bank policymaking has developed with rapidity. Likewise, the practice of policy formulation and implementation at various central banks has experienced a paradigm shift that requires the development of underlying theory. In other words, the advancements have had an advantageous reciprocal influence between the development of financial and monetary economic theory in the academic world and the various schools of thought that underlie practical central bank policymaking. Academia, in the pursuit of clarifying or offering solutions to prevent problems, has contributed greatly to conceptual thinking and theories concerning various phenomena and economic behaviors. The theoretical constructs have become advanced and deep on a philosophical conceptual level and using quantitative methods along with empirical studies, although in this case relying on certain assumptions that simplify complex economic behaviors. On the other hand, the central bank has also made various breakthroughs in terms of conceptual ideas and innovative policymaking as a solution to complex financial–economic problems. Oftentimes, policy innovations are also facilitated by close interaction between the central banking community at various global and regional forums and meetings. Theoretical development in academia is clearly a solid foundation for central bank policymaking. Nonetheless, the complexity of existing problems often necessitates an innovative policy response from the central bank, which also encourages the academic world to conduct theoretical studies and seek underlying empirical evidence. Rapid development in terms of policy concepts and institutional arrangements at the central bank, from the perspectives of both academia and the innovative policy response of the central bank, demands stringent reviews, studies, and documentation. The various concepts underlying the policy response of the central bank are a vital reference for the central bank and other policymakers to draw lessons on the general and best practices available to prevent the problems faced as well as for academia to seek clarification and develop the theories further. On the other hand, the relentless development of theories and ideas in academia are a helpful reference for the central bank in terms of policymaking in response to emerging problems.

Introduction    9 For academia, the synthesis (or indeed contradictions) that may appear between central bank policy theories and practices represent an interesting study that must be taught to students in order to enhance academic knowledge or, indeed, for further development. This is important to bridge the gap between theoretical ideas in the academic world and the practical policymaking process at the central bank. Moreover, academia, especially in Indonesia, has limited access to advanced books that discuss central banking policy theories and practices as a reference for postgraduate students, teachers, and researchers. In general, existing programs and subjects tend to focus on standard theories, which lack comprehensive and informed discussions on central bank policy practices and the underlying theories. Similarly, reviews of institutional aspects are also absent from academia.

1.3. Objectives of the Book This book aims to analyze various policy theories and practices adopted by central banks as well as the institutional arrangements underlying the principles of good governance in policymaking. In other words, this book explores the synthesis between theories, practices, and institutional arrangements that underlie central bank policy. The discussion focuses on philosophical conceptual theories and the policies themselves, supported by relevant quantitative analyses as required along with various issues and the latest problems as they emerge. This book aims to bridge central bank policy theories and practices at the intermediate and advanced levels, which, according to the authors, is currently lacking and, therefore, will provide an invaluable contribution to the world of knowledge. The goal is to produce a reference book on central bank policy theories and practices in order to advance the understanding and insight of policymakers and practitioners, as a study resource for researchers and academic teaching materials for teachers and students. To that end, the material discussed and contained in this book represents a combination of three perspectives. Firstly, a philosophical review and quantitative analysis of the theories underlying central bank policymaking. Secondly, the policy practices employed by central banks in various jurisdictions, including Indonesia. Thirdly, a study on the influence and implications of central bank policies and regimes to economic, monetary, and financial development in the countries concerned. The various empirical studies presented substantively explain and enrich the three dynamic perspectives. Admittedly, it is not a simple task to author and present such a reference book that bridges policy theories and practices at the central bank. Several challenges forced the authors to review and consider the most appropriate material that represents the overarching purpose. Onerous challenges were encountered not only due to the rapidity of theoretical and practical development, which have already been discussed in various other books, journals and papers, but also because of the panoply of central bank policy practices that do not yet appear in print as a reference for the writing of this book. In this case, the authors benefitted from extensive experience and direct involvement in the policymaking process at BI and various international forums, as well as academic activities linked to various

10    Central Bank Policy postgraduate programs at several universities in Indonesia. The authors seek to review various theories and constructs under development that are considered relevant and most clearly color the policy practices at the central bank, without rehashing existing discussions in more detail that can be found in the existing literature on monetary economic theory as taught in academia. During the writing process, one of the largest challenges was sorting the relevant theories for policymaking as practiced. More specifically, versatility when exploring the plethora of best policy practices employed by various central banks, while simultaneously raising the variety of debates and critiques under development, was imperative. This challenge stemmed not only from the comparatively new policy practices, for example, the Flexible Inflation Targeting (FIT) Framework, macroprudential policy or the institutional aspects of diverse central banks but also the review of empirical evidence concerning how policy and institutional arrangements contribute propitiously to central bank and economic performance.

1.4. Systematics of the Writing The systematics of this book are based on the relevance of policy theories and practices at central banks that the authors consider important for policymakers, practitioners, researchers, and as academic teaching materials. In addition to the experience gained from teaching postgraduate programs, the authors’ involvement in the policymaking process at BI, coupled with the rare opportunities to attend seminars, conferences, and a broad range of meetings with other central bankers in numerous jurisdictions are used as a key reference in the writing of this book. Through a process of contemplation, the authors ultimately decided to divide the book into five inseparable parts as the systematics of the writing. Part I explores the evolution, reforms, and latest issues concerning central bank policy, while Parts II–V discuss in depth the core of the material about central bank policy theories and practices. Those four parts include a theoretical review of the influence of central bank policy on the economy in Part II, central bank policymaking theories and practices in Part III, as well as the institutional theories and practices that underlie the principles of good governance in terms of central bank policy in Part IV. Part V of the book presents the latest perspectives and new paradigm for central banks in the wake of the GFC, namely the central bank policy mix consisting of monetary policy (including foreign capital flow management), macroprudential policy, and payment system policy, to achieve the dual mandate of price (and exchange rate) stability as well as financial system stability. Referring to the evolution and reformation of central banks since their inception in the seventeenth century up until the aftermath of the GFC, a number of important lessons can be given. Firstly, the central bank plays a significant role in supporting sustainable economic growth through price (and exchange rate) stability, while creating and maintaining financial system stability. This dual mandate has existed since the formation of central banks and was tangibly reinforced by the GFC. Secondly, in pursuance of the dual mandate, a central bank requires

Introduction    11 jurisdiction over monetary policy (including foreign capital flow management), the payment system as well as macroprudential regulation and supervision. Experience in the ITF era has shown that a focus on monetary policy by central banks contributed to financial instability that culminated in the GFC. Thirdly, central banks must be given autonomy to implement their authority in pursuit of the goals or mandate. Experience from when the government exercised control over the central bank also showed the adverse impact on macroeconomic stability and ultimately resulted in economic crisis. Fourthly, institutional arrangements need to be strengthened to ensure that central bank independence is accompanied by public accountability and transparency. Part II of this book, consisting of Chapters 3–5, reviews those theories considered relevant and lay the groundwork for the central bank policy practices. Chapter 3 discusses monetary economic theory in a closed economy, stressing the role of money and credit in the economy. Concerning the role of money, the authors put forward a debate on the neutrality versus non-neutrality of money from the perspective of Classical Theory and Keynesian Theory, the models that were subsequently developed as well as the empirical evidence for the influence of money in the economy. The theoretical discussions lead to a synthesis that money supply exerts more influence on prices in the long term, while acknowledging the trade-off between inflation and growth in the near term. This chapter closes with the new paradigm of monetary economic theory developed by J. E. Stiglitz, which emphasizes monetary analysis of credit quantity, not money! Therefore, monetary policy instituted through money supply or interest rates must be strengthened by banking regulations that directly affect credit, particularly through macroprudential policy. A monetary and macroprudential policy mix is required to support price and financial system stability. Chapter 4 takes the reader into monetary theory space for an open economy by exploring several theories that tackle the role of the exchange rate in the economy. The discussion focuses on studying models that determine exchange rates, which are discussed ad nauseam in the literature, from the Mundel–Flemming model, to the flexible price monetary model, Dornbusch’s overshooting model and the money substitution model. Various models and empirical evidence concerning the influence of the exchange rate on the economy, especially in terms of predicting exchange rate behavior and the effect of exchange rates on international trade, economic growth and inflation, are also explored. Specifically, a discussion is presented on the monetary policy trilemma in an open economy, namely between monetary policy independence for domestic purposes (price stability and economic growth), foreign capital flow mobility and exchange rate flexibility. The chapter closes by identifying some implications concerning the influence of exchange rates on monetary policymaking by the central bank, namely the requirement for an interest rate policy, foreign exchange market intervention, and capital flow management to overcome the aforementioned policy trilemma. Chapter 5 discusses the monetary policy transmission mechanism, in other words how monetary policy instituted by the central bank affects the financial system and economy. This represents a critical aspect underlying monetary

12    Central Bank Policy policymaking at the central bank in order to achieve the desired target but which, according to the authors, receives insufficient theoretical and empirical attention. The discussion also explores the two schools of thought regarding how monetary policy is transmitted to the economy, namely the money view by which policy is transmitted through the interest rate channel, exchange rate channel, asset prices, and expectations, as well as the credit view by which policy is transmitted through the bank credit and capital channel, the balance sheet and collateral constraints channel as well as the risk-taking behavior channel. In addition to theoretical discussions, various empirical studies are also explored concerning the workings of the monetary transmission mechanism in several countries, including advanced countries, EMEs, and Indonesia itself. It will be argued that the effect of monetary policy on price stability and economic growth is indirect and experiences a lag. In general, it will be argued that the channels of the money view continue to function as the principal transmission mechanism. Nonetheless, the GFC demonstrated the growing importance of the credit view channels due to financial frictions that stem from asymmetric information, financial product innovation and risk-taking behavior. Therefore, central bank monetary policy based on interest and exchange rates must be backed up by policy to manage credit through macroprudential policy. Part III of the book, containing Chapters 6–9, details the central bank policy framework applied during the two decades prior to the GFC. Like the explanation presented in Chapter 2, during this period, central banks tended to focus more on monetary policy to achieve price stability in support of economic growth. In many ways, the policy framework was a success and remains the backbone of central bank policy in various countries to this day. That framework underlies the theoretical and empirical discussions contained in Part II. Chapters 6 and 7 focus on the strategic and operational frameworks of monetary policy. Thereafter, Chapters 8 and 9 discuss the monetary policy framework with price stability as the overarching goal, which is practiced widely around the world, including in Indonesia, under the popular ITF. The discussion on the strategic monetary policy framework in Chapter 6 focuses on the strategic issues regularly debated to this day, namely the “duality” of accomplishing the ultimate monetary policy target: output stability (economic growth) or price stability (inflation)? Opening with the debate between economic growth and price stability as the final target of monetary policy, several policy regimes practiced by central banks as well as on a theoretical level are explored, such as exchange rate targeting, monetary targeting, inflation targeting, and nominal gross domestic product (GDP) targeting. Empirical evidence showing what such regimes are based on and the implications to economic performance in various countries is presented considering the historical perspective and evolution of policy orientation over time. In addition, several issues relating the Phillips Curve are also interpreted from the perspective of short- and long-term linkages between inflation and economic growth as well as the socio-economic costs (sacrifice ratio) that arise in the price and output stabilization process. Chapter 7 elaborates the monetary policy operational framework in terms of two other “duality,” namely: (1) Which are the preferable operational targets or

Introduction    13 policy instruments used to achieve the ultimate target of existing policy? Which are the quantity-based instruments or price-based instruments? (2) Which is the preferable policy response should be formulated; a rule-based response or discretion? The discussion commences with the selection of these monetary instruments to address various economic shocks; specifically, a distinction is made between demand shocks or supply shocks. With the operational target selected, the followup issue addressed is the optimal policy response to achieve the final target set. This chapter also presents a hypothesis for the existence of the state-contingent rule, which can accommodate rules and discretion, as an optimal policymaking response, particularly in the case of EMEs. As the most popular monetary policy framework adopted by central banks, the ITF discussion delivered in Chapters 8 and 9 focuses on its application at the central bank, including a case study of Indonesia. Institutional and operational regulations as well as their impact on central bank and economic performance are also explored rationally. Several ITF regimes practiced at central banks are also discussed and compared with other monetary policy regimes. The discussion in this chapter also addresses a number of misconceptions that persist at the central bank itself or in academia, such as how economic growth, exchange rates, and other aspects are accounted for under ITF. Part V of the chapter presents an assessment of ITF application in several developing countries and provides justification for the adoption of a FIT framework as the ideal format for a small open economy. Part IV of the book, incorporating Chapters 10–12, focuses on the salient aspects of implementing good governance in terms of the institutional arrangements and central bank policy. In addition to the monetary policy practices of central banks explored in Part III, the institutional aspects discussed in Part IV remain elusive and rarely taught academically. This is not only because such aspects include contemporary issues of central bank policy but also because it is important to discuss such issues in the context of the need to strengthen public institutions in line with the spread of democracy as well as policy transparency and accountability. Chapters 10 and 11 explore the institutional aspects salient to monetary policy, namely building policy credibility and striking the optimal balance between independence and accountability. Based on time inconsistency theory, in accordance with the models developed by Kydland–Presscott and Barro–Gordon, the discussion in Chapter 10 stresses the need for the central bank to follow a certain rule when formulating policy, for example, the Taylor rule, rather than discretion. The discussion continues with a number of approaches found in empirical studies to measure credibility as well as the contribution to central bank monetary policy performance with respect to supporting economic performance, including an assessment of monetary policy credibility in Indonesia. Chapter 11, meanwhile, presents the delegation of authority theory using the Rogoff model that emphasizes the importance of delegating authority by parliament to a conservative central bank that is inflation averse in order to stipulate an achievable inflation target. The theory is based on central bank independence since the 1980s in terms of policymaking to achieve the mandate, namely price stability. Central

14    Central Bank Policy bank independence also requires strong accountability to ensure credibility is achieved. Chapter 12 focuses on the growing importance of transparency and communication as an integral part of institutional strengthening and good governance at the central bank. This is not only increasingly demanded by the public as democracy flourishes but also because transparency and communication have become crucial instruments that support central bank policy credibility and effectiveness. This chapter shows that the phenomenon of increased policy transparency will persist moving forward and become a main menu item for central banks in the implementation of monetary policy. Therefore, transparency must be evaluated and strengthened coupled with a clear monetary policy communication strategy referring to international best practices, the monetary policy regime adopted and financial market development in the respective jurisdiction. Part V of the book, namely Chapters 13–15, presents a new paradigm that has evolved as central bank policy theories and practices since the GFC. Two important observations have emerged since the GFC. Firstly, the importance of the central bank returning to its original mandate at inception, namely price (and exchange rate) stability, while creating and maintaining financial system. Secondly, the need for monetary policy through interest rates to be reinforced by exchange rate stabilization policy and foreign capital flow management in an optimal policy mix along with payment system policy and macroprudential policy. The GFC demonstrated the need for the central bank to prevent crises through macroprudential regulation and supervision in order to maintain financial system stability from the perspective of macrofinancial linkages and systemic risk. In addition, the GFC also heightened the volatility of foreign capital flows from global financial markets to EMEs, thus necessitating more than just interest rate policy, namely exchange rate stabilization policy and foreign capital flow management to address the ever-present central bank policy trilemma. An optimal policy mix between monetary policy (including foreign capital flow management), payment system policy and macroprudential policy represents an innovative new central banking paradigm to achieve the dual mandate of price stability and financial system stability. Chapter 13 discusses central bank policy complexity and the response to heightened exchange rate and foreign capital flow volatility observed since the GFC, referring to the discussion presented in Chapter 4. The discussion begins by analyzing the push and pull factors of foreign capital flows to a country and the implications of volatility on monetary and financial system stability. The discussion then moves on to focus on the policy trilemma of achieving price stability, exchange rate stability, and free flows of foreign capital. The overarching message from the discussion in this chapter is that the policy response of the central bank, which is generally based on interest rates, must be reinforced by foreign exchange market intervention and foreign capital flow management in order to achieve price and exchange rate stability. The two-target approach (price and exchange rate stability), using two instruments (interest rates and foreign exchange market intervention), should strengthen central bank policy effectiveness and credibility.

Introduction    15 Chapter 14 discusses the role of the central bank in terms of financial system stability through macroprudential policy. The discussion reiterates the importance of financial system stability as a costly lesson of the GFC, and how the central bank supports financial system stability using macroprudential policy. Theoretical and empirical studies underlying two important areas of macroprudential policy are explored. The first relates to the phenomenon of financial accelerators as a characteristic the financial sector, which tends toward procyclicality, as well as excessive credit growth and asset bubbles. The second relates to interconnectedness and networking in the financial sector, such as that found on the interbank market and in the payment system, which could exacerbate and transform contagion risk into systemic risk that endangers financial system stability. The discussion also extends to the implementation of macroprudential policy frameworks in various countries, including advanced countries, EMEs and Indonesia itself, in terms of the principles, targets, and instruments employed. The chapter closes with a discussion of the importance of coordination between macroprudential policy and microprudential policy as well as macroeconomic policy (monetary and fiscal) in strengthening financial system stability. The new central bank policy paradigm is discussed in depth in Chapter 15. The discussion covers the conceptual dimensions of the central bank policy mix, including the integration of price stability and financial system stability, the mix of policy instruments used, as well as the transmission mechanism. Thereafter, the discussion focuses on two modeling approaches for the central bank policy mix. The first is the macroeconomic structural model known as IIF, which is an offshoot of FIT, covering the structure and analysis framework, as well as the central bank policy mix formulation model. The second is the Dynamic Stochastic General Equilibrium (DGSE) model, which expands the macroeconomic analysis with macrofinancial linkages based on financial frictions to integrate monetary and macroprudential policy when formulating the central bank policy mix. The chapter goes on to explore the policy mix instituted by BI as a new paradigm initiated in 2010, encompassing the targets and instruments, formulation of the interest rate, exchange rate, and macroprudential policy along with foreign capital flow management as well as strengthening institutional arrangements and coordination with the Government and other relevant authorities. As mentioned previously, the series of discussions in the chapters of this book are fundamentally inseparable. In that context, after the explanation of central bank evolution and reform since inception in the seventeenth century until after the GFC in Chapter 2, the authors deepen the theories on the role of money and credit in the economy, foreign capital flow mobility, and exchange rates, as well as the monetary transmission mechanism in Chapters 3–5. The policy framework based on ITF discussed in Chapters 6–9 becomes the subsequent foundation of good governance in the implementation of monetary policy and the implementation of the central bank policy mix. Likewise, institutional reform through independence on the one hand and strengthening accountability, transparency, and communication on the other, which is discussed in Chapters 10–12, becomes a fundamental step in terms of enhancing policy credibility and central bank reputation. To close, the modern central bank policy mix applied in the wake of

16    Central Bank Policy the GFC is discussed in Chapters 13–15 theoretically and as practiced at various central banks. It is important to note that the policy mix represents a new central banking paradigm since the GFC, and BI should be recognized as a pioneer in terms of its implementation since 2010.

1.5. Utility of the Book This book is considered a first in terms of comprehensively discussing the latest theories and practices of central bank policy. With the scope of material presented and systematics of writing employed, this book is believed to be a practical reference tool for policymakers at the central bank, practitioners, and academia. For the leaders, researchers, and staff at BI, as well as the financial-banking and business communities in general, the discussions contained in this book, especially in Parts III and IV, are fundamental to deepen the various concepts of central bank policy and theory. Similar benefits will also be enjoyed by economists and entrepreneurs in other sectors. Although the discussions are predominantly on a philosophical and conceptual level, accompanied by in-depth theory, the rationale, and writing of this book is presented in a manner that is readable and easily digestible. For academia, this book will be particularly useful as a primary and essential reference for university students of monetary and financial economics at the intermediate and advanced levels of Master’s degrees and Doctoral degrees. Meanwhile, for university students at the senior level of Bachelor’s degrees, this book will also be very useful as a reference on economic modeling and empirical evidence regarding the central banking issues discussed. Furthermore, this book will also be suitable as a foundation for the development of a postgraduate central banking curriculum. Besides, the book will also be particularly relevant for teachers and students to deepen central bank policy and theories in the central banking curriculum or, indeed, for other applicable subjects, such as the monetary sector and financial institutions. As stressed previously, this book will furnish teachers and students alike with a comprehensive and deep understanding of relevant economic theories, policy practices, and institutional aspects supported at the central bank. As a primary reference for financial and monetary economics students as well as postgraduate central banking programs, this book was designed to be taught in one full semester. The authors’ experience of teaching such subjects at postgraduate programs underlies that belief. Moreover, the chapters of the book are also presented in an order such that it aids the learning process for students. Nonetheless, each respective teacher maintains the scope to explore certain aspects in more or less depth as preferred. Students, including those taking monetary and financial economics or similar subjects at university, as well as those exploring central bank policy to write a thesis or even to deepen insight into economic policy, will have no problem understanding the materials and concepts put forward in this book.

Introduction    17 Over time, the materials presented in this book will be adjusted according to the prevailing requirements and developments in terms of policy practices at central banks as well as input from the readers in line with the requirements brewing among the public and in academia. To that end, inputs from readers are warmly welcomed for further refinement and advancement of this book. In closing, the authors sincerely expect this book to not only meet the needs of the readers in terms of a deeper understanding of the various salient aspects linked to central bank policy and theories but also to expand our scientific understanding and knowledge. Hopefully!

This page intentionally left blank

Chapter 2

Central Bank Evolution and Reform 2.1. Introduction The GFC reconfirmed the vital economic role played by the central bank. Each time a crisis event transpires, the central bank is the most active organization in the stabilization efforts to resolve the crisis and recover the economy. During the GFC, for example, the central banks of the United States and Europe injected liquidity and aggressively lowered interest rates to prevent further spillover and contagion of the systemic impacts to the global economy. Despite a sluggish and multispeed recovery, the US economy is now gaining momentum, with signs of recovery also emerging in Europe. The GFC also raised awareness that the central bank cannot merely apply monetary policy to achieve price stability and support sustainable economic growth, but also the importance of monetary policy for financial system stability. The central bank function as lender of last resort (LOLR) again saved the financial system from heightened systemic risk. The overarching goal of the central bank is to support sustainable economic growth through price (exchange rate) stability and financial system stability has been a solid early warning indicator for the past three centuries. Since its inception, for instance, the Bank of England has not only provided government debt and discounted loans on commercial papers to commercial banks, but also strived to prevent panic in the financial system as LOLR. Nonetheless, economic conditions, academic thinking, and political pressures have swung the pendulum of central bank focus over time. After World War II, most central banks fell under the auspices of government control to stimulate economic growth. Nevertheless, during the two decades that preceded the GFC, the policy pendulum of the central bank had swung toward emphasizing price stability through short-term interest rate instruments, namely application of the Inflation Targeting Framework (ITF). Central bank success in terms of catalyzing comparatively robust economic growth with low inflation and interest rates for two decades under ITF concealed the gradual build-up of risk and financial system fragilities that ultimately precipitated the GFC. Consequently, in the wake of the GFC, central banks were forced back to their original mandates, namely price (or exchange rate) stability and financial system stability. As the core purpose of the central bank, price stability can support stronger economic growth with maintained macroeconomic stability. Furthermore, Central Bank Policy: Theory and Practice, 19–45 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191004

20    Central Bank Policy numerous central banks opined that a focus on inflation control through monetary policy instruments such as ITF were no longer appropriate. In addition, the central bank must seek to maintain financial system stability in conjunction with the government and other relevant authorities. Crisis prevention is better than crisis resolution, of which the impact is more severe, costs more expensive, and management more complex. Ultimately, in the event of a crisis, the role of the central bank to stabilize and recover the economy is also vital. The central bank can actively seek financial stability through macroprudential regulation and supervision, while instituting monetary policy to lean against risk accumulation, and through payment system regulation and supervision. Moreover, the central bank also needs to tighten coordination with the government in the context of macroeconomic policy and structural reforms to support economic growth and macroeconomic stability. Similarly, central bank coordination with the government and relevant authorities, through the supervision of financial services and provision of deposit insurance, is also required to maintain financial system stability. This chapter discusses the evolving role, policy reform, and institutional aspects of the central bank. Following this introduction, the discussion is divided into three main sections. First, central bank evolution since central banking was established until the present day is presented as a series of important episodes in terms of price stability and financial system stability, namely the gold standard and era of the real bills doctrine, central banking under government control, the era of central bank independence and inflation targeting and, finally, central banks in the post-crisis landscape after 2008/2009. Second, central bank policy reform is discussed from the perspective of several central issues, namely: monetary policy, price and exchange rate stability, foreign capital management and exchange rate stability, monetary stability and financial system stability, as well as payment system stability. Third, the various central bank institutional reforms, which have gained impetus over time, especially since the 1980s, are discussed, covering: credibility, independence and accountability; transparency and communication; as well as institutional arrangements and coordination with other authorities. Before closing the chapter, several strategic issues on central bank theory and policy that are currently debated and practiced at central banks in different jurisdictions around the world are presented in the final section as the foundation for further discussion in the subsequent chapters of this book.

2.2. The Evolving Role of the Central Bank Central banking can be traced back to the seventeenth century, with the establishment of the Swedish Riksbank in 1668 as a private joint stock bank that functioned to provide loans to the government and as a clearing institution for financial transactions (Bordo, 2007). The Bank of England was subsequently established in 1694 to provide government debt and rediscounted loans through banks on commercial papers issued for trade, as well as to provide short-term liquidity facilities to banks, which later became known as LOLR (Goodhart, 2010). During the ensuing period in Europe, a number of central banks were established to finance ongoing conflicts and war. The Banque de France, for example, was established

Central Bank Evolution and Reform    21 by Napoleon in 1800 to stabilize the local currency in the wake of hyperinflation due to excessive printing of currency during the French Revolution. Another central banking model is to print and circulate money, known as a deposit bank, per the Wisselbank Amsterdam at the beginning of the 1800s (Ugolini, 2011). The Federal Reserve System (the Fed) in the United States is a central banking system that was replicated in several countries at the turn of the twentieth century (Watchel, 2010), primarily to consolidate various payment instruments (currency) into a national legal tender and to realize financial stability. The Fed was created in 1913 to replace two preceding central banks, namely the Bank of the United States (1791–1811 and 1816–1836). The Fed was mandated with providing uniform and elastic currency to accommodate seasonal, cyclical, and secular economic shifts, while also playing the role of LOLR to ensure financial system stability (Bordo, 2007). Of note, at their inception, central banks in developing countries were established to circulate currency by consolidating the existing payment instruments or issuing new currency as the legal tender, or as a prerequisite to creating a newly independent country. Similar conditions were found in Indonesia with the establishment of BI from the nationalization of De Javasche Bank, which assumed the function of a circulating bank, succeeding Unit IV of Bank Negara Indonesia (BNI) in July 1953. The discussion, thus far, has shown the evolution of central banks from circulating banks to the functions more commonly known and expected today. In general, central banks have been given a monopoly to print and circulate currency as the legal tender of a country. With the opening of commercial bank accounts, however, the central bank began to function as the banks’ bank, facilitating all transactions in the banking system, including clearing and settlement through commercial bank accounts held at the central bank. In addition to the payment system function, a central bank also undertakes monetary operations, like the commercial papers rediscounted by the Bank of England, which is now a function of monetary policy. Furthermore, as the banks’ bank, the central bank also functions as LOLR for banks facing liquidity problems, as an integral part of financial system stability. At their establishment, however, central banks were also responsible for government fiscal financing, either to finance war or purchase government debt, a function that is now restricted or even forbidden due to the adverse impact on inflation and economic stability. In general, the central bank is the authority responsible for policies that affect money supply and credit in a country. Although formulated differently in central bank regulations per jurisdiction and discussed in academia, the following four aspects and goals are regularly found in the history of central banking (Blinder, 2010; Bordo, 2007; Goodhart, 2010). First and foremost is achieving price (or exchange rate) stability in line with the monetary policy adopted, such as the gold standard, fixed exchange rate, or inflation targeting regime. Second is regulating and supervising the payment system, including the printing and distribution of currency, regulating clearing and interbank settlement, or regulating payment instruments. Third is maintaining financial system stability. In the early days of central banks, this function was linked to LOLR, but nowadays also covers macroprudential regulation and supervision. In contrast, the regulation and supervision

22    Central Bank Policy of individual banks (microprudential) can be performed by the central bank or transferred to a financial services authority. Fourth is stabilizing the real economy, which typically implies job creation and robust economic growth. Initially, this function was linked to the buying of government debt or even financing the fiscal deficit, but then evolved into macroeconomic policy coordination and structural reforms with the government. Where the pendulum swings between the main goals of the central bank evolve over time, influenced by economic conditions, political pressures, and academic thinking. Wherever the pendulum swings, however, stability will always be associated with the central bank, be it price (or exchange rate) stability, payment system stability, financial system stability, or real economic stability. In this context, Goodhart (2010) dived the focus and goals of central banking, since its inception, into three episodes: (1) the gold standard era from the 1840s to 1914, when price stability and financial stability were relatively well maintained; (2) the decades of central bank control under the auspices of the government from the 1940s to the end of the 1970s, with a focus on economic growth but prevalent economic instability; and (3) the period of central bank independence, with a focus on price stability (inflation targeting) and market mechanisms from the 1980s to 2007, when risks gradually accumulated in the financial system. The years in-between the three episodes were marred by economic distress or financial crises, including World Wars I and II from the 1920s to 1930s, the oil and government debt crises in the 1970s and the GFC. During normal times, “stability” reflected a focus on central bank policy to ensure monetary and financial system stability and, therefore, support sustainable economic growth. Meanwhile, during financial and economic crisis episodes, the term “stability” forced the central bank’s role in terms of stabilization, restructuring, and economic recovery.

2.2.1. Gold Standard Era and Real Bills Doctrine: 1840–1930 The salient developments of central bank evolution during the gold standard were maintained price and exchange rate stability, coupled with financial system stability. At their inception, central banks in various countries, including the Fed, were established to manage the gold standard, which was the prevailing international monetary system at that time. Congruent with the gold standard that remained in place until 1914, the value of a currency was defined in terms of gold and weighted according to the amount of gold reserves held. Central banks maintained large gold reserves to ensure their respective currency could be converted into gold. When gold reserves decreased due to a balance of payments (BOP) deficit or deteriorating domestic economic conditions, however, central banks were forced to raise the discount rate on their loans to commercial banks. Higher domestic interest rates attracted foreign investors and, therefore, brought in more gold to bolster central bank reserve assets. During the gold standard era, central banks complied with the rules to maintain convertibility of their respective currency to gold over all other considerations. Consequently, gold convertibility anchored economic stability and central bank credibility. This meant the amount of currency that could be created by the central

Central Bank Evolution and Reform    23 bank was restricted by the value of gold reserve assets held and growth of money supply and, therefore, price stability was maintained. Furthermore, inflation was pegged to gold, the long-term price of which was set by market mechanisms, so price expectations were also linked to the gold price. In other words, under the gold standard, central banks were avowedly committed to price stability. The gold standard also supported financial system stability, in addition to the central bank’s role as LOLR. Such conditions were supported by consistent limitations on total money supply in order to manage convertibility of the local currency into gold. Monetary policy with the gold standard also linked fund mobilization and lending by commercial banks to public income and real economic conditions. Nonetheless, monetary policy was constrained by the stringent rules of the gold standard; therefore central banks committed instruments to stimulate economic growth. In the event of an external shock, for instance an export decline that decreased (increased) the BOP surplus (deficit), the central bank would reduce money supply to maintain currency convertibility to the gold reserves. A tighter monetary policy stance would, therefore, undermine bank credit and real economic activities, including economic growth and the labor force. Similarly, a productivity shock that reduced aggregate supply at home would require tighter monetary policy to reduce aggregate demand and avoid exacerbating the BOP, which would disrupt convertibility to the gold standard. Consequently, under the gold standard, central banks focused solely on price stability and financial system stability, without considering the underlying impacts on overall economic stability and growth. A hot issue that is often debated is how to reconcile the gold standard while maintaining financial system stability, particularly under distress or panic conditions. As mentioned previously, monetary policy under the gold standard was achieved by regulating money supply to maintain currency convertibility to gold reserves. Operationally, however, this was achieved through the real bills doctrine, where the central bank would only purchase commercial paper from real businesses engaged in real productive activities (the real economy). Consequently, money supply was directly linked to real output/trade, therefore, monetary policy could create price stability. Similarly, exchange rate stability was also maintained because of the central bank’s monetary policy response to maintain currency convertibility in line with the gold reserves held. In fact, because money supply was linked directly to credit volume and real output/trade, the financial and business cycles moved in tandem without excessive credit expansion or asset price bubbles, thus supporting financial system stability. The central bank’s role in financial stability was also achieved through the function as LOLR. Per Bagehot’s Rule, the central bank provided short-term liquidity facilities to solvent but illiquid banks against readily convertible securities at higher rates than market interest rates. Problems surfaced, however, with the issuance of speculative securities or finance bills to purchase assets, particularly on the capital market. Consequently, the central bank’s monetary operations generally avoided purchasing finance bills to prevent a build-up of asset price bubbles that eventually burst, leaving an economic bust in their wake, including high inflationary pressures and financial system instability (Goodhart, 2010). Such conditions occurred in the United States during the 1920s and 1930s, when the Fed, in accordance with the real bills

24    Central Bank Policy doctrine, refused to buy finance bills, which were proliferating in line with the capital market boom, and prohibited banks from lending to purchase stocks. In other words, the Fed tightened monetary policy in 1928 to offset the Wall Street boom, which subsequently triggered a recession and the stock market crash in 1929 (Bordo, 2007). Another issue is the central bank buying government debt securities to finance war or stimulate economic growth and create job opportunities. One reason why the Fed objected to expansive monetary operations through the purchase of government debt during the Great Depression of the 1930s was due to inconsistency with the real bills doctrine (Meltzer, 2003).1 Although the gold standard was abandoned in 1914 and the real bills doctrine in the 1930s, several invaluable lessons also apply to central banks today. In general, both regimes produced a framework that integrates central banking theory and practices between monetary policy (price stability) and financial system stability (Goodhart, 2010). A significant aspect is the importance of central bank policy credibility, supported by the gold standard and real bills doctrine, which led to price stability and financial system stability. In the present-day context, such credibility is achievable through the ITF framework and Taylor rule to determine the appropriate interest rate policy response to achieve the inflation target. With a specific real interest rate based on the Taylor rule, the aggregate amount of real currency circulating in the economy is in line with real output, although the distribution of term deposits and bank loans is not always aligned to the real bills doctrine. In terms of financial system stability, per the real bills doctrine, while the discounted rate and bank credit are always pegged to real output/trade, or real bills, and not to capital market speculation or government debt, or finance bills, the goals of price stability and financial system stability can both be achieved simultaneously. This cannot be guaranteed in terms of monetary policy with ITF because fundamentally, setting the policy rate does not question how and where term deposits and bank credit are intermediated. Nevertheless, the real bills doctrine and inflation targeting face the same problem, namely financial product innovation outside the banking system that leads to asset price bubbles and, therefore, financial system instability risk. In this case, production innovation occurs outside the control of money supply and credit per the real bills doctrine, while according to inflation targeting, low short-term interest rates are often the drivers of product innovation. Consequently, the central bank requires other instruments to maintain financial stability, perhaps by leaning against the wind and/or macroprudential policy regulation and supervision.

2.2.2. Central Banks under Government Control: 1940–1970 During the two World Wars, the gold standard was overburdened with government demand for financing, with central banks returning to their inaugural function of

1

It is important to note that the real bills doctrine is extremely different to thinking today, which advocates central bank monetary operations by prioritizing the purchase of government debt securities.

Central Bank Evolution and Reform    25 buying government debt. Consequently, a combination of hyperinflation, recession, and exchange rate instability threatened the international monetary system. Efforts by several countries to save the gold standard failed. For countries lacking gold, such as the United Kingdom, a return to the gold standard would mean a tighter monetary stance to overcome hyperinflation, thus exacerbating already high unemployment. Through observations of labor market rigidity, Keynes recommended utilizing monetary policy to stabilize output. Oppositely, consistent with the real bills doctrine, the United States was forced to sterilize gold inflows by tightening monetary policy due to excessive growth of money supply, credit, and stock prices. The central banks’ failed commitment to the gold standard as the Great Depression descended in 1929 lead to a deep economic depression in the United States and Europe, coupled with the default and closure of thousands of US banks. Consequently, as is often the case when a crisis strikes, central banks were blamed. In many countries, including the United States and United Kingdom, the monetary policy function of the central bank was transferred to the government’s control and the financial system was subjected to strict regulation (Goodhart, 2010; Group of Thirty, 2015). Under the auspices of the government, monetary policy was used as an instrument to finance the fiscal sector and stimulate economic recovery after World War II. Consequently, inflation began to soar again despite a stronger economic recovery. Meanwhile, to avoid excessive growth of money supply, the government resorted to financial repression through interest rate policy and bank credit rationing. At the same time, deposit insurance and financial safety nets were created to reduce the number of defaulting banks and incidences of financial instability from the 1940s to 1980s. Of note, domestic financial repression marked the beginning of international banking operations, including the rapid proliferation of Eurodollar securities issued in London as well as bonds issued in Latin America. The monetary function was returned to the central banks in the 1950s, with a focus on price stability, which had been a salient issue in the United States and Europe around that time. The move was also consistent with the Bretton Woods system, agreed in 1944, to apply an adjustable fixed exchange rate system. The United States, as holder of the largest gold reserves, was the main proponent of the Bretton Woods system and the system, therefore, became synonymous with the US dollar exchange rate standard. Nonetheless, driven by the Phillips curve, which demonstrated that unemployment can be reduced only by small increases in inflation, the United States became inconsistent in its commitment to inflation, applying loose monetary policy to overcome unemployment. Other countries, therefore, such as Germany, which consistently maintained inflation by targeting money supply, began to protest US actions by threatening to exchange the dollar for gold. Consequently, the Bretton Woods system was formally abandoned in 1971, when President Nixon closed the gold window. The collapse of the Bretton Woods system was accompanied by heightened inflationary pressures and depression in several countries at the beginning of 1974. Freed from their international commitments, central banks focused monetary policy on reducing the spike in unemployment attributable to the oil price crisis in 1973. Nevertheless, inflation expectations accelerated rapidly, hence compelling

26    Central Bank Policy the central banks to apply monetary targeting, usurping unemployment. In reality, the strategy to target money supply was unable to control inflation due to, among others, unstable demand in line with financial sector development and foreign capital flows. In 1979, when the second oil price shock occurred, most central banks began to focus on inflation control rather than just leaning against the wind to anticipate bumps in unemployment. Another important aspect of the second oil price shock was the expansion of banking activities through international financial transactions to recycle oil proceeds from net producers to net importer EME. Several EMEs, particularly in Latin America during the 1970s, issued government bonds and/or took out large loans from United States and European banks, which subsequently triggered another crisis when interest rates increased. In advanced countries, the issue of financial stability was also raised among central banks during the middle of the 1970s due to several defaulting banks operating internationally, including the Franklin National Bank in the United States and Bank Herstatt in Germany in 1974.2

2.2.3. Central Bank Independence and Inflation Targeting: 1980–2007 High inflation compelled many central banks to focus more on price stability. Initially, monetary policy was directed toward controlling inflation through monetary targeting. Nevertheless, controlled inflation was unable to stimulate growth and, therefore, central banks in advanced and EME sought to bring inflation down to low levels. A new monetary policy framework, known as the ITF, which was introduced by New Zealand in 1988, became increasingly prominent. Numerous central banks around the world applied various iterations of ITF, from rigid to flexible, and with different weights for achieving the inflation target with a trade-off for economic growth (output gap). New Zealand was one example where a rigid ITF was applied, while the US Federal Reserve, European Central Bank (ECB), and Bank of Japan (BOJ) favored a more flexible approach that also considered economic growth and employment on top of the inflation target. Several characteristics of ITF supported central bank monetary policy commitment and credibility in terms of price stability (Bernanke, Gertler, & Gilchrist, 1999; King, 2004; Svenson, 1999). ITF shared a number of common aspects with central bank policy during the gold standard and real bills doctrine. First, commitment to achieve a published inflation target. Success in terms of

2

The need for central banks to maintain financial system stability in addition to price stability subsequently nurtured international cooperation through establishment of the Bank for International Settlements (BIS), which seeks to strengthen capital and bank supervision standards through the Basel Committee on Banking Supervision (BCBS) and the payment system through the Committee on Payments and Settlement Systems (CPSS). The BIS committees also reinforced two other multinational institutions, namely the International Monetary Fund (IMF), which promotes international monetary cooperation, and the World Bank, which advocates cooperation and financial development.

Central Bank Evolution and Reform    27 achieving the inflation target would anchor future inflation expectations. Per the gold standard, commitment to the inflation target was achieved through currency convertibility to the gold reserves held by the central bank. Second, monetary policy consistency to achieve the inflation target, by setting interest rates (for instance using the Taylor rule) to achieve the inflation target two years ahead. Under the gold standard, consistency was supported through monetary targeting in line with the reserve assets held to ensure exchange rate convertibility. Third, central bank independence in terms of monetary policy to achieve the inflation target. This is analogous with the central bank’s mandate to maintain currency convertibility under the gold standard. Fourth, central bank transparency and communication concerning inflation, macroeconomic projections, and even monetary policy direction. This aspect was strengthened under ITF in order to achieve strong central bank credibility, similar to that during the gold standard era. ITF, as a monetary policy framework, was considered a success when applied in most advanced countries and many EME (Berg, 2013; Taylor, 2014). Inflation and volatility decreased sharply, while interest rates also fell. Moreover, after the deep recessions felt in the 1970s, downturns in the 1980s and thereafter became less frequent and shallower (including in the United States). In fact, economic growth continued to accelerate, driving the rapid expansion of trade and investment between countries. The subsequent economic boom, coupled with low inflation, robust economic growth, lower unemployment and low interest rates, endured for the next two decades during an era that became known as the Great Moderation. Nevertheless, financial instability simultaneously returned. High credit growth, often linked to rapid property and housing development, was considered as the root cause. Robust economic growth, together with low inflation and interest rates prompted a domestic and global economic boom. Widespread financial deregulation in the 1980s further accelerated finance in the boom economy. In the United States, for example, removal of restrictions on deposit rates at financial institutions (savings and loans) in the 1980s accelerated short-term funding for long-term housing loans. Consequently, when policy rates were raised in the 1990s, savings and loans institutions were sent into bankruptcy. Furthermore, a US stock price shock in 1987 quickly spread to other countries. A housing credit and banking crisis also befell Japan at the beginning of the 1990s. In addition, several banks extending international loans, primarily to Latin America, experienced default in the 1980s, which again threatened financial instability. The subsequent boom in Asia, known as the East Asian Miracle, attracted an influx of foreign capital flows, including private external debt, which precipitated the Asian Financial Crisis in 1997/1998. Credit growth in the US private sector, however, particularly the housing sector, continued to outpace bank deposit funding for more than two decades (Schularick & Taylor, 2009). Financial innovation to meet the credit demand boom changed the banks’ business model as follows: (1) selling government debt assets (treasury bills and bonds); (2) applying “originate to distribute” strategy with credit securitization, particularly mortgage facilities, and selling the products to nonbank financial institutions and the capital market; and (3) relying on wholesale funding by issuing bonds and/or credit securitization (Goodhart, 2010). This change fostered greater risk-taking behavior and financial system fragilities.

28    Central Bank Policy Furthermore, during the economic boom and period of asset price bubbles, banks increased leverage through excessive credit growth based on rising asset prices. Although banks in the United States and Europe were subject to the capital requirements of Basel II, leverage was still increased by expanding asset portfolios (AAA) with mortgage-backed securities (MBS). Financial procyclicality increased due to risk-weighted assets and mark-to-market accounting. During the boom period, risk-weighted assets tended to be too low, while mark-to-market accounting expanded the capital base. Financial innovation, coupled with high leverage and risk-taking behavior during the two decades of economic boom ultimately triggered the GFC, the most severe crisis since the Great Depression, from September 2008 to March 2009, the fallout of which was felt worldwide.

2.2.4. Central Banks and the Global Financial Crisis of 2008/2009 The GFC of 2008/2009 forced central banks to stabilize the financial system again and rescue the economy. This time, however, the necessary central banking actions extended beyond LOLR. Central banks in advanced countries were also forced into post-crisis financial stabilization measures by expanding liquidity facilities to nonbank financial institutions and the financial markets, while also instituting accommodative monetary policy to save the economy. The US Fed, for example, issued four new financial stabilization instruments, namely the Term Auction Facility (TAF) to provide short-term liquidity using (1) securities issued by banks and nonbank financial institutions as collateral between December 2007 and April 2009; (2) loans to facilitate the purchase of Bear Sterns by JP Morgan; and (3) support for American International Group (AIG) and jurisdiction to repay interest on the statutory reserve requirement (Goodfriend, 2010). Consequently, the balance of Federal Reserve assets skyrocketed, from USD 0.9 trillion in July 2007 to USD 2.8 trillion in July 2011. The Fed subsequently adopted quantitative easing (QE) to recover the economy from crisis, prompting Fed assets to climb further to USD 4.2 trillion in April 2014 (Mehrling, 2015). The exorbitant cost of crisis recovery, combined with the far-reaching fallout, reaffirmed that the central bank mandate of monetary policy alone was insufficient, financial system stability also had to be maintained. In fact, the mandate to maintain financial system stability was present since the inception of central banks in the seventeenth century, namely through the central bank acting as the banks’ bank under normal conditions and functioning as LOLR under crisis conditions. In the present context, the severe impact and high cost of the GFC in 2008/2009 raised awareness that financial system stability is desirable more to prevent a crisis rather than overcome a crisis. The financial stability function of a central bank must be combined with monetary policy that has evolved to require sound credibility, similar to conditions under the gold standard and ITF eras. Nonetheless, central bank evolution in terms of monetary policy and financial system stability is heavily influenced by prevailing economic conditions, political pressure, and nascent theories. Oftentimes, the important role of central banks is forgotten or even ignored during economic boom periods, but when a crisis strikes, economic stabilization and recovery become crucial. That attitude toward central

Central Bank Evolution and Reform    29 banks needs to stop. Many important lessons have been learned from the evolution of central banking since its inception until the present day, while the GFC in 2008/2009 greatly reinforced the various efforts required moving forward as follows. First, price (or exchange rate) stability must remain the core mandate of the central bank. From past experience, the government’s exploitation of the central bank printing money to finance war, buy debt, or for post-war economic recovery left an indelible mark on the community through high inflation, low growth and economic uncertainty. On the other hand, central bank commitment and credibility to maintain currency convertibility during the gold standard era, and to control money supply in accordance with real output during the real bills doctrine, ensured price stability, business assurance, and economic growth. Likewise, central bank credibility through ITF in the decades since the 1980s helped to lower inflation and interest rates, while fostering robust economic growth during the era of Great Moderation. The need to anchor inflation in the future is also supported by a synthesis of Keynesian and Neoclassic theories, namely that money in the near term only affects prices but, in the long term, there is a trade-off between inflation and economic growth. The GFC of 2008/2009 clearly showed that low inflation due to monetary policy credibility could not guarantee macroeconomic and financial system stability. Prevailing thinking prior to the GFC, namely that if a central bank could stabilize inflation in the near term (e.g., up to two years, such as ITF) the economy would return itself to a steady state in the long term, proved to be erroneous (Borio, 2011). That perspective formed the basis of macroeconomic models at the time, stating the price rigidity was the only cause of economic deviation from its steady state (Walsh, 2010; Woodford, 2003). Conversely, the GFC also showed that a period of robust economic growth, coupled with low inflation and interest rates, due to successful monetary policy, actually creates financial system instability. Financial innovation and risk-taking behavior led to high leverage along with excessive credit growth and asset price bubbles, which ultimately resulted in the severe GFC of 2008/2009. Second, central banks must reinforce their mandates in terms of monetary policy, financial system stability, and the payment system, similar to when central banks were first established. As highlighted previously, the acute fallout and high cost of the GFC provided valuable insight that crisis prevention is better than crisis resolution. Experience of the GFC in 2008/2009 reversed the previously held view that the monetary policy function should be separate from financial system stability (Borio, 2010). Prevailing thought before the GFC stated that: (1) monetary policy should focus on price stability; (2) microprudential regulation and supervision, with a focus on individual financial institutions can and is sufficient to maintain overall financial system stability; (3) financial institution regulation and supervision should be separate from the central bank to avoid conflicts of interest; and (4) the central bank’s role as LOLR is adequate to ensure financial stability. The GFC decisively dispelled those viewpoints. Financial instability actually originated from financial system and macroeconomic linkages through intermediation. The crisis transpired because the financial cycle was bombarded by excessive credit growth, debt accumulation or high leverage, and uncontrolled

30    Central Bank Policy asset price bubbles. Therefore, the role of the central bank to maintain financial system stability and prevent crises must be strengthened through macroprudential regulation and supervision that focuses on a macrofinancial perspective and systemic risk (Borio, 2010). Moreover, the GFC of 2008/2009 also clarified that crisis resolution through low interest rates and large-scale monetary injections is not always effective and requires a protracted recovery period. Central banks in advanced countries must make every effort to expand their balance sheets in order to ensure healthy longterm interest rates and financial conditions when instituting ultra-loose monetary policy (Borio & Disyatat, 2010). The experience of the GFC reversed previously held beliefs, including ITF, that short-term interest rates would be sufficient to transmit the effect of monetary policy to the economy. Monetary policy, therefore, no longer focuses merely on controlling short-term interest rates, while simultaneously anchoring future expectations, but on setting the full interest rate structure for all subsequent tenors (Svensson, 2003). Prevailing view before the GFC even denied the possibility that central banks, such as the Fed, ECB, and BOJ, would lower nominal interest rates to 0% (Ahearne, Gagnon, Haltmaier, & Kamin, 2002; Orphanides & Wieland, 1998). The GFC also showed that financial frictions occur because of financial innovation, asymmetric information, valuation methods and price setting, and even risk-taking behavior, which prompted a fundamental change when determining interest rates and the monetary policy transmission mechanisms to the economy. Third, the GFC of 2008/2009 showed how rapidly spillover can propagate from central bank policy in one country to another and the large impacts thereof. Liquidity injections by central banks in advanced countries triggered massive foreign capital flows and high exchange rate appreciation in EME. Nonetheless, the normalization process of monetary policy in the United States since May 2013 and monetary policy divergence from the ECB and BOJ, has triggered widespread uncertainty in the global financial system, affecting foreign capital flow volatility and exchange rates in EME. The GFC demonstrated that monetary policy is becoming increasingly globalized. Despite not undermining its ability to control conditions on domestic financial markets and the national economy, the Fed now pays more attention to the global aspects underlying monetary policymaking (Bernanke, 2007). Financial interconnectedness between countries in the monetary and international financial system has also tended to amplify the accumulation of uncertainty (Borio, 2014), while flows of foreign capital led to excessive credit growth that subsequently triggered financial instability (Borio, McCauley, & McGuire, 2011). In fact, as external debt spiraled, along with USD denominated bonds issued by EME, including Asia, the impact of global spillovers from the Fed’s policy to economic and financial conditions in EMEs accelerated and increased in magnitude (McCauley, McGuire, & Sushko, 2015). The GFC experience has reversed the former doctrine of “keep-your-housein-order,” namely that if each central bank acted in accordance with their own respective economy, then global economic and monetary conditions would also be sound. This kind of doctrine was analogous with the view that if regulation and supervision could guarantee the soundness of individual financial institutions,

Central Bank Evolution and Reform    31 then financial system stability would also be maintained. This doctrine advocated that each central bank should focus on price stability in their respective domestic economy, through ITF application for instance, and follow a free-floating exchange rate regime (Rose, 2007). In many ways, such a viewpoint assumed that the international financial and monetary system worked efficiently and each financial asset was a perfect substitute, thus the law of one price and uncovered interest parity applied. In reality, however, the GFC demonstrated that such a viewpoint is not always correct and, therefore, many central banks, particularly in EME, must engage in exchange rate stabilization and foreign capital flow management to protect the domestic economy from the impact of global spillovers. Fourth, the evolving role of central banks, from their inception until the GFC in 2008/2009, has shown that central banks must be free from political intervention if they are to achieve price stability and financial system stability. Past experience from when the central bank was under government control, whether to finance war or purchase government debt, always resulted in high inflation and subsequent recession. The fiscal domination of central bank policy must be avoided. Nevertheless, central bank independence is never absolute. The central bank must remain avowedly committed to its mandate in order to remain credible, including in terms of developing a sound policy framework, strong leadership, and solid capabilities. The central bank must act to strengthen accountability and transparency through periodic reports, testimony in front of parliament, as well as communicating policy to the public and stakeholders. Moreover, independence does not absolve the central bank from building policy coordination with the Government and other relevant authorities. In contrast, central bank independence should be founded on macroeconomic and financial system policy interdependence as well as economic structural reforms to achieve robust economic growth and create broad job opportunities, supported by maintained macroeconomic and financial system stability.

2.3. Central Bank Policy Reform Coordination between monetary, macroprudential, and payment system policy with fiscal policy and government structural reforms was desperately required to accelerate the economic recovery in the wake of the GFC in 2008/2009. Hitherto, the onus for the GFC resolution measures and subsequent economic recovery have been placed on the central banks with monetary policy and financial stability; hence, El-Erian (2016) emphasized the difficult task of the central bank as “the only game in town.” Nonetheless, although the central banks worked diligently and successfully saved the financial system from total failure and economic collapse and from a protracted depression state, such as the Great Depression of the 1930s, the economies of advanced countries and the global economy have grown sluggishly against a backdrop of ubiquitous uncertainty blighting global financial markets. The role of the central bank must be strengthened further in order to stimulate robust economic growth and ensure lasting financial stability. That is still not enough, however. Comprehensive policy concerning structural reforms to more seriously nurture inclusive economic growth is required, along

32    Central Bank Policy with fiscal policy to stimulate demand, overcoming the debt overhang, as well as regional and international institutional arrangements (El-Erian, 2016). Several schools of thought have been proposed to strengthen modern central bank policy after the GFC of 2008/2009 (Bank for International Settlements [BIS], 2011; Borio, 2011; Committee on International Economic Policy and Reform, 2011; Georgsson, Vredin, & Sommar, 2015; Group of Thirty, 2015; Orphanides, 2011). Strengthening central bank policy encompasses several aspects to support sustainable economic growth through monetary and macroprudential policy, foreign capital flow management and payment system policy. Central bank evolution has provided a number of valuable lessons, while experience from before and after the GFC of 2008/2009 represents an important foundation to rebuild modern central banks.

2.3.1. Monetary Policy, Price Stability, and Exchange Rates Central bank evolution has shown that policy credibility is vital to create and maintain stability in the economy. During the gold standard era, commitment to currency convertibility in line with the gold reserves held combined three aspects of stability achieved by the central bank, namely; (1) exchange rate stability in line with the currency’s convertibility, (2) financial system stability because total money supply (and credit) was in line with real demand and the reserve assets, and (3) price stability due to currency convertibility and controlled money supply. During the ITF era, price stability was achieved through central bank commitment to the inflation target, with consistent short-term interest rates. Notwithstanding, exchange rate stability and growth of money supply (and credit) depended heavily on market mechanisms and behavior in the financial system. Under both regimes, economic growth is considered the second goal after price stability. The GFC of 2008/2009 exposed weakness in the rigid application of ITF and, therefore, many countries opted for a flexible ITF. Two issues emerged in its implementation, however: (1) the policy response to heightened exchange rate volatility, and (2) the accumulation of asset price bubbles, such as a stock price boom or property price boom. Regarding the exchange rate, central banks face a policy trilemma concerning interest rate policy efficacy in terms of achieving price stability with exchange rate volatility and foreign capital flows. In that context, several studies have shown how a central bank can overcome the policy trilemma. Calvo and Reinhart (2002), for example, pointed out the phenomenon of “fear of floating” in the case of exchange rate policy in various countries as more embraced an intermediate system rather than the hard pegged or purely floating systems. Meanwhile, empirical studies show that EMEs applying ITF employ a strategy where the central bank responds well to inflation and real exchange rate when setting the monetary policy rate (Aizenman, Hutchison, & Noy, 2011; Mohanty & Klau, 2004). The importance of exchange rate stabilization compelled several economists, including Ostry, Ghosh, and Chamon (2012a, 2012b), to recommend a dual-target regime, namely exchange rate stabilization to support attainment of the inflation target, in order to bolster central bank monetary policy credibility. The prerequisite,

Central Bank Evolution and Reform    33 while predicated on ITF, is that the exchange rate target is used to mitigate the undesirable impacts of capital flow shocks on inflation expectations, through the direct influence of exchange rate pass-through or indirectly through domestic demand. With the effect of exchange rate fluctuations included in the assessment and projections of the central bank, exchange rate targeting is directly more effective than an interest rate response to the impact of exchange rates on inflation. Ostry et al. (2012a, 2012b) recommended that central banks in EMEs apply the dual-target regime. Furthermore, central banks in EME also have a secondary instrument at their disposal in addition to the interest rate, namely foreign exchange intervention. The latest study by Blanchard, Adler, and Filho (2015) investigated the effectiveness of foreign exchange intervention as an instrument to mitigate exchange rate pressures originating from foreign capital flows. In other words, monetary policy with dual targets, namely the inflation target and exchange rate stabilization, using two instruments, namely interest rates and foreign exchange intervention, would increase, not decrease, central bank credibility. The latest empirical evidence after the GFC in 2008/2009 demonstrated the importance of central bank policy in response to asset price bubbles, particularly property prices. Jorda, Schularick, and Taylor (2014), for example, provided important evidence on the correlation between monetary conditions, house prices, and credit growth from historical data stretching back 140 years from 14 advanced countries. The discussion focused on how monetary policy could respond to the accumulation of risk, reflected in procyclical housing bubbles and credit booms. This issue also touches upon another debate, namely lean versus clean: whether it would be better for the central bank to lean against the wind to prevent the bubbles from bursting, or better to wait until the bubble has burst and then clean up the mess through aggressively loose monetary policy. The “clean” view was embraced by the Federal Reserve under the leadership of Alan Greenspan, otherwise known as the Jackson Hole consensus. A number of considerations underlie that view: an investment boom driven by higher productivity, bubbles due to lower risk premiums and irrational exuberance, as well as interest rate hikes that might be ineffective against the bubbles but could actually cause the bubble to burst and spillover into the economy (Greenspan, 2002). Nevertheless, the GFC of 2008/2009 dispelled that view after clearly showing potential risk from excessive credit and leverage, driven by the bubbles. Consequently, a leaning monetary policy was considered more appropriate to avoid bubbles, rather than cleaning up the fallout of a crisis. Experience in Australia, for example, showed that leaning monetary policy can be successfully implemented. In response to excessive credit growth in the housing sector in 2002 and 2003, the Reserve Bank of Australia gradually raised interest rates despite the inflation outlook remaining under control within the target (Bloxham, Kent, & Robson, 2011). Consensus now strongly supports central banks seriously considering aspects of financial stability and leaning against the wind despite financial stability not legally being part of the central bank’s mandate. Bernanke (2009), for instance, stated that the Fed played an influential role in overcoming the crisis in the United States and, therefore, its institutional ability to support financial stability and drive economic recovery with low inflation must be maintained.

34    Central Bank Policy 2.3.2. Exchange Rate Stability and Foreign Capital Flows As mentioned previously, not only did exchange rate volatility increase after the GFC, but also the volatility of foreign capital flows, which undermined the effectiveness of central bank policy to accomplish the domestic goals in line with the policy trilemma of an open economy. Exchange rate stabilization through foreign exchange intervention constitutes one option available to the central bank. Nonetheless, the magnitude of foreign capital flow volatility undermined the efficacy of the central bank’s interest rate policy and foreign exchange intervention, which also incurred considerable expense. In fact, exchange rate and foreign capital flow volatility can trigger financial imbalances and thus prompt macroeconomic and financial system instability (Borio & Disyatat, 2011). Consequently, the regulation of foreign capital flows represents another policy option available to the modern central bank (Eichengreen et al., 2011). Several follow-up studies show the importance of foreign capital flow management to help overcome the policy trilemma faced. Klein and Shambaugh (2013) investigated how far foreign capital flow management policy, through temporary and directed capital controls, or limited exchange rate flexibility, could realize full autonomy in terms of central bank policy. The findings showed that foreign capital flow management under a flexible exchange rate regime could improve monetary policy autonomy. Nevertheless, capital flow management under a fixed exchange rate regime would be less advantageous to monetary policy autonomy compared to full capital mobility. As has been applied worldwide, extensive and long-term capital controls would break the links between domestic and international interest rates under a fixed exchange rate system. Furthermore, inflation is generally more manageable under a fixed exchange rate regime and, therefore, the gains in terms of monetary policy autonomy are more limited. Rey (2013) also extolled that importance of foreign capital flow management to overcome the monetary policy trilemma based on the increased comovement of capital flows, asset prices, and economic growth between countries in the global financial cycle. Consequently, Rey (2013) urged central banks to change the trilemma into a dilemma, namely strengthening monetary policy autonomy and exchange rate stabilization through direct foreign capital flow management or with macroprudential policy. The study showed that the salient determinants of the global financial cycle are monetary policy in advanced countries, which influences capital flow leverage, and credit growth in the international financial system, the impact of which is transmitted to numerous countries, especially EMEs, thus interest rate policy and exchange rate flexibility are unable to protect an economy from the global financial cycle. Obstfeld (2015) went further, stating that the impact of financial globalization has not only created a monetary policy trilemma but also and financial trilemma. The study evaluated the ability of monetary policy in EMEs to mitigate the impact of global monetary and financial fluctuations. Congruent with the monetary policy trilemma, the study confirmed that EMEs applying a flexible exchange rate regime are better disposed toward monetary policy autonomy. Nevertheless, changes to the exchange rate regime are unable to protect the domestic economy from global monetary and financial shocks. The most important channels of

Central Bank Evolution and Reform    35 international monetary and financial policy transmission are long-term interest rates and risk premiums, while financial globalization has been shown to not only drive capital flows but also distribute risk across jurisdictions. Despite mitigating global spillovers through exchange rates, monetary policy cannot alleviate the impact of global capital flows and financial risks in the domestic financial system. Therefore, the monetary policy trilemma expanded to the financial trilemma, namely policy autonomy with the overarching influence of global financial integration while maintaining domestic financial system stability.

2.3.3. Monetary and Financial System Stability The GFC of 2008/2009 also raised awareness concerning the need to restore the mandate of financial stability to the central bank, on top of price stability that has been the goal since the beginning. The existing definitions of financial system stability are not always exactly the same between policymakers and academia, but fundamentally refer to conditions where the financial system functions soundly in an economy and demonstrates resilience to the various shocks that could emerge (Allen & Wood, 2006; Mishkin, 1999). Mandating the central bank with financial stability would also correct the policy reformulation and institutional reforms of the previous crisis. After the Asian Financial Crisis in 1997/1998, for example, various authorities focused on the most appropriate approach for that time. Central banks focused on price (and exchange rate) stability through monetary policy. Furthermore, the regulation and supervision of financial institutions was transferred away from the central bank to newly formed financial services authorities, such as in South Korea and Indonesia. The GFC in 2008/2009, however, showed just how important it is to look at the financial system from a macrofinancial perspective, in which systemic risk must be mitigated through macroprudential policy. The role of central banks in terms of financial stability is achieved directly through macroprudential policy and as LOLR, and indirectly through monetary and payment system policy (Haldane, 2004; Padoa-Scioppa, 2003). The central bank’s mandate to support financial stability is primarily accomplished through the macroprudential regulation and supervision of financial institutions from a macrofinancial perspective and a focus on systemic risk. Macrofinancial linkages between financial institutions and the economy often exacerbate financial procyclicality through a faster acceleration during periods of economic expansion and a deeper recession during an economic downturn. Procyclicality, if not managed accordingly, will accelerate the boom-bust periods of the financial cycle, which could result in a crisis. Therefore, macroprudential policy is directed toward managing financial procyclicality, with a strong focus on credit booms and asset price bubbles, to avoid macrofinancial imbalances that often lead to financial crises. In addition, macroprudential policy is also directed toward managing interconnectedness and financial networks, particularly through the interbank money market and financial infrastructure, to avoid systemic risk. Default at a financial institution can trigger a crisis, along with macrofinancial procyclicality as well as domestic or international economic shocks, while a systemic crisis to the entire financial system can occur rapidly due to interconnectedness and financial networks.

36    Central Bank Policy In addition to macroprudential policy, central banks also support financial stability through monetary and payment system policies as well as foreign capital flow management. Monetary policy supports financial stability by maintaining macroeconomic stability, and the primary target is to achieve price (and exchange rate) stability. This includes regulation and supervision of the money markets and foreign exchange market as an integral part of monetary policy, which also supports financial stability in terms of mitigating systemic risk from interconnectedness and financial networks. The same considerations also apply to the payment system, for which the primary goal is to provide a secure, reliable, and efficient payment system as a fundamental part of monetary policy effectiveness. Meanwhile, managing foreign capital flows protects financial stability from external spillovers, on top of the primary mandate to support exchange rate stability and avoid sudden-stop and BOP crises. The central bank is the most appropriate institution to conduct macroprudential regulation and supervision. The central bank has the capacity to perform macroeconomic and macrofinancial surveillance as well as the instruments to implement macroprudential policy (Kawai & Morgan, 2012). Furthermore, a BIS (2011) studied of 13 advanced and emerging market countries concluded that central banks must have direct involvement in financial system stability policymaking and implementation if the policy is to be effective based on three salient explanations. First, the monetary policy framework and task implementation focuses the central bank on macroeconomic analysis and projections, while complementing understanding of the financial markets, institutions and infrastructure, all matters of great import in terms of macroprudential policy implementation. Second, financial system instability can be caused by and have a significant impact on economic performance, with fundamental consequences on economic activities, price, and exchange rate stability, as well as monetary policy transmission effectiveness. Central banks are already familiar with evaluating financial system and economic interconnectedness, or the macrofinancial aspects, as the focus of macroprudential policy. Third, the central bank is a source of liquidity in the financial system and economy through monetary policy and the LOLR function, and the availability of liquidity is crucial when maintaining financial stability.

2.3.4. Payment Systemic Stability Printing and circulating currency have been the overarching purposes of the central bank since the beginning (Roberts & Velde, 2014) and, therefore, have become the basis of central bank duties in terms of monetary policy, the payment system, and financial system stability (Blinder, 2010; Goodhart, 2010). In the payment system, the central bank does not merely print currency as legal tender but also holds the accounts of the banking sector as final settlement assets for all their financial transactions, often referred to as currency outside bank or central bank money. Meanwhile, commercial banks create demand deposits by opening savings deposits for the public, otherwise known as inside money or commercial bank money. Both types of money have developed simultaneously in modern economic transactions, where public transactions are facilitated by commercial bank

Central Bank Evolution and Reform    37 money and then interbank transactions are facilitated by central bank money (BIS, 2003). Confidence in commercial bank money lies in the respective commercial bank’s ability to liquidate term deposits and other obligations to the public as required from the liquid assets held, or through financial transactions with other commercial banks and/or through their account held at the central bank. On the other hand, confidence in central bank money lies in the central bank’s ability to provide liquidity to commercial banks as required and maintain a stable value of the money created, namely price and exchange rate stability. The payment system encompasses all instruments and conventions to transfer funds from one bank to another (Borio, Russo, & van de Bergh, 1992). Therefore, the payment system consists of three main elements or core processes: (1) authorization to make payments from a customer to a bank; (2) exchange payment instructions between banks, known as clearing; and (3) transaction settlement through the respective accounts of each bank at the central bank (Sheppard, 1996). Large value transactions are typically settled through the real-time gross settlement system (RTGS), while small-value transactions are settled through the national clearing system. It is important to note that the payment system also includes payment mechanisms for government bond transactions (e.g., through the BI – Scripless Securities Settlement System, BI-S4) as well as stock and bond transactions on the capital market. Considering the importance of the payment system to accelerate the volume and efficiency of various transactions in the economy, central banks around the world have adopted best practices in the form of core principles when operating and regulating the payment system (BIS, 2001). The function of a central bank in the payment system correlates closely with its duties in monetary policy and financial stability (ECB, 2007). In general, monetary stability supports the continuity of various transactions and economic growth and, therefore, underpins financial stability and payment system reliability. Furthermore, a reliable payment system is a prerequisite to monetary policy implementation due to its support toward monetary operations and money market security and efficiency. Through monetary policy implementation, central banks set their policy rates and undertake monetary operations to influence market liquidity in line with the policy rates set. Effective monetary operations require: (1) an active interbank money market that transmits lending and borrowing between banks suffering from inadequate liquidity and banks with excess liquidity, which is also an indicator of monetary conditions, reflecting general economic activity; and (2) the central bank to predict the determinants of market liquidity, specifically the circulation of currency between the government and public, such as tax revenues and budget spending (Sheppard, 1996). A secure, reliable, and efficient payment system supports effective monetary operations and transmission to the economy in general (Gaspar & Russo, 2006). The payment system is also inextricably linked to the financial stability function of the central bank. The financial system is said to be stable if it is resilient to shocks and distress, thus preventing disruptions to money supply and credit, the intermediation function to allocate savings deposits to various investment activities, as well as the payment process in the economy. In this case, market infrastructure, including a sound, efficient, and resilient payment system, is a prerequisite

38    Central Bank Policy for the markets and financial institutions to function (BIS, 2000). Conversely, financial stability is crucial to the function of the payment system and, ultimately, the success of monetary policy, both of which represent the fundamental prerequisites for sustainable economic growth. Systemic disruptions can appear in the payment system due to, for instance, problems at a participating bank or because of technical difficulties with the system operator. Such disruptions can influence the position of liquidity at other banks, demand for currency, and ability to settle various financial instruments (Bogov, 2011). Eventually, such conditions spill over to interest rates, which subsequently impact trade activities and price setting mechanisms on the markets. Moreover, contagion risk could materialize due to the mutual transactions between banks, which hold the same positions as one another in the payment system. The inability (actual or perceived) of a bank to settle its obligations under financial market distress can undermine confidence and trigger systemic risk on the default of financial institutions. This is where the central bank plays a vital role in financial stability through monetary operations in the form of intervention when the markets fail to function as market maker of first resort, or by providing liquidity facilities to solvent banks as LOLR in accordance with Bagehot’s Rule (ECB, 2007). Over the past few decades, the payment system has developed rapidly in line with the evolution of the modern financial system (Fung, 2012). Concerning the retail payment system, payment innovation has progressed quickly, providing public access to various new payment instruments, such as the proliferation and widespread use of automated teller machines (ATM), debit and credit cards, mobile and internet banking, as well as electronic money. Central banks need to maintain the reliability and efficiency of the retail payment system, as a crucial element of a functioning economy, in addition to the influence of payment innovation on future public demand for currency. Meanwhile, concerning the large value payment system, interbank transaction settlement system security and efficiency are key to national financial system stability and resilience to shocks, such as the GFC. The large value payment systems around the world continue to function despite onerous liquidity issues in the financial system, even when faced with bankruptcies. Such conditions prove that market infrastructure resilience, including the payment system, is imperative for financial system stability and efficiency. The GFC of 2008/2009 also precipitated a new phenomenon that prompted a fundamental change in the current and future financial system, namely financial technology, or FinTech (Horn, Geerts, & Coolen, 2014; PwC, 2026; Walport, 2015; World Economic Forum, 2015). Against a backdrop of more stringent capital and prudential regulations for financial institutions, small-medium FinTech companies have sprung up that use information technology and telecommunications in very innovative ways to offer direct financial services to their customers. The business segments include: (1) using technology to directly present financial services to consumers, thereby competing and even replacing traditional financial services, such as a competitive peer-to-peer borrowing platform with banks when lending to micro, small, and medium enterprises (MSME); and (2) using

Central Bank Evolution and Reform    39 new technology to strengthen or optimize the outreach of banking markets, hedge funds, and other financial service businesses. Financial technology (Fintech) companies enable the digitization of financial products and services, thereby extending the outreach to customers, providing alternative solutions to improve the processes and supporting administration efficiency, including new technology to visualize and analyze the data. On one hand, the rapid emergence of FinTech has extended outreach, while increasing the volume and efficiency of financial services and the payment system, particularly the retail payment system. On the other hand, however, financial innovation outside of formal financial institutions through the rapid emergence of FinTech could create financial system fragmentation and financial system stability risks that demand vigilance and prudential regulation by the central bank. Payment system reliability and security are crucial for financial system stability, efficiency, and effectiveness. In this case, payment system failure could trigger potential financial systemic risk, particularly if the large value payment system is affected (Comotto, 2011). Payment default of an asset or transaction would: (1) cause the payment system participant to incur loss on the default payment (credit risk); and (2) exacerbate liquidity conditions at each respective payment system participant (liquidity risk). Problems at one payment system participant could inadvertently spill over to affect the customers, thereby creating a loss of confidence and squeezing market liquidity (market liquidity risk), which would propagate systemically through contagion to the rest of the financial system and even the economy. Payment system failure could originate at participating financial institutions because of inadequate liquidity or insolvency, or because payment default is caused by the issuing financial institution or due to a loss of confidence. Therefore, to maintain payment system reliability and security, central bank regulation and supervision aim to: (1) prevent a build-up of systemic risk and maintain financial stability; (2) enhance system efficiency and payment instruments; (3) safeguard security and public confidence in their payment assets; and (4) protect monetary policy transmission channels (Kokkola, 2010). In pursuance of its target, a central bank’s general functions are from various aspects of the payment system, as operator, initiator, and supervisory authority.

2.4. Central Bank Institutional Reform Central bank credibility is vital in pursuit of the goals, namely price (and exchange rate) stability, financial system stability and payment system stability. Under the gold standard, credibility was the central bank’s obligation through commitment to maintain currency convertibility in line with the gold reserves held. After a brief disappearance under government control, central bank credibility was restored under ITF through institutional reform that granted central bank independence, while strengthening policy accountability and transparency. Academic research also supported the institutional reforms, while actual practices at various central banks showed the institutional reforms had enhanced central bank credibility, particularly in terms of maintaining price stability.

40    Central Bank Policy 2.4.1. Credibility, Independence, and Accountability As discussed in the previous section, central bank policy practices and implementation during the ITF era were characterized by two types of institutional reform. First, providing greater independence or jurisdiction to the central bank. Second, refocusing on a single mandate, namely price stability. Both steps were taken in response to low central bank credibility during the era under government control. After that time, not only were central banks redirected toward stimulating economic growth but were also given autonomy over policymaking and implementation. Theoretically, the problem of central bank credibility correlates with monetary policy time inconsistency in terms of achieving price stability (Kydland & Prescott, 1977). Monetary policy inconsistency can occur because central banks assign larger weights to output (job opportunities) and/or formulate policy responses that are inconsistent with the inflation target (Barro & Gordon, 1983). For example, central banks initially directed monetary policy toward attaining a given inflation target. Nonetheless, political pressure to reduce interest rates and/or finance government spending through the creation of money led to inconsistent monetary policy and caused a higher inflation bias. In that context, monetary policy inconsistency is analogous to an inflation tax because of the drive toward stronger economic growth. From the central bank’s perspective, policy inconsistency can be interpreted as seigniorage, thus triggering inflation. A few measures are available to overcome policy inconsistency, including: (1) delegating authority or granting independence to the central bank; and (2)  introducing policy rules (Calvo, 1978; Kydland & Prescott, 1977; Rogoff, 1985). Fundamentally, independence is delegating authority from the public through the government to the policy target of the central bank. For the central bank, however, independence is required to limit political intervention in the legal mandate. In developing countries, referring to experience in many countries as well as comparatively underdeveloped political conditions, central bank independence is required to sustainably ensure the goal of economic development remains in line with the gravitas of stability. In an ITF context, for instance, setting the inflation target to be achieved by the central bank represents the delegation of public authority pursuant to prevailing laws. The inflation target may be set by the government itself or by the government based on the central bank’s recommendations. At several central banks, including New Zealand and the United Kingdom, the delegation of authority for the inflation target is contained within a remit to the central bank. Independence must be accompanied by central bank accountability for the mandate given, for example, attainment of the inflation target under ITF. Central bank credibility is only achievable if the inflation target is attained. In this case, a consistent monetary policy response toward achievement of the inflation target is vital for credibility. Consistency is often measured in terms of certain policy rules, such as the Taylor rule, when setting the monetary policy rate. How much stronger central bank interest policy must be from the policy rules depends on the credibility. Weaker credibility, for example, will require a stronger monetary policy response. In other words, while the central bank builds credibility in terms of attaining the inflation target, the policy response must be tighter than stipulated

Central Bank Evolution and Reform    41 by the corresponding policy rule and, therefore, the impact on economic growth tends to be more pronounced. The next level of central bank policy credibility is reputation (Barro & Gordon, 1983). Reputation is earned if the central bank can consistently build enough policy credibility with the public over a prolonged period. The central bank is said to have garnered a sound reputation if the public believes that the central bank has a proven track record in its avowed commitment to attaining the inflation target through an appropriate policy response. Consequently, public inflation expectations are anchored to the inflation target and, therefore, inflation bias due to policy inconsistency is avoided.

2.4.2. Transparency and Communication In addition to central bank credibility and policy consistency, reputation is also required to strengthen transparency and communication with the public. Therefore, extending monetary policy transparency at the central bank is an inevitability. This phenomenon is congruent with the trend of greater central bank independence against the backdrop of a democratic government, application of ITF and the importance of anchoring expectations on the financial markets. In this case, several interesting observations can be found. Firstly, efforts have been poured into formulating a set of best practices to be used as a reference for central banks to strengthen monetary policy transparency (Blinder, 2001; IMF, 1999). The IMF, for example, assesses four critical aspects of policy transparency, namely: (1) clarity of roles, responsibilities and objectives of the monetary policy authority; (2) open process for formulating and reporting monetary policy decisions; (3) public availability of information on monetary policy; and (4) accountability and assurances of integrity from the monetary authority. Secondly, a number of theories and analyses have been developed to provide formulae on transparency and its correlation with monetary policy, basic policy considerations and its impact on the economy. Fundamentally, the theories deal with asymmetric information in monetary policy, primarily in the context of the focus and clarity of the central bank’s preferences in terms of the objectives and relevance of monetary policy as well as the influence of market expectations on the monetary policy response and transmission, including the debate between rules versus discretion for monetary management. Thirdly, although lacking in number, several studies have provided empirical evidence to show the positive contribution of increased central bank transparency on economic performance. Specifically, increased transparency can help to control inflation, reflected by the negative correlation between transparency and inflation. On the other hand, increased transparency feeds through lower output volatility, or at least does not exacerbate output volatility. Greater transparency was demanded in the wake of the GFC in 2008/2009 and became a leading agenda item for central banks to implement monetary policy. An important development of central bank communication after the GFC was the emergence of forward guidance, defined as providing information and future monetary decisions for the market (Bihari, 2015). This practice was introduced by the US Federal Reserve at the announcement of the Federal Open Market Committee (FOMC) on December 16, 2008 as follows:

42    Central Bank Policy The Federal Reserve will employ all available tools to promote the resumption of sustainable economic growth and to preserve price stability. In particular, the Committee anticipates that weak economic conditions are likely to warrant exceptionally low levels of the federal funds rate for some time. The focus of the Committee’s policy going forward will be to support the functioning of financial markets and stimulate the economy through open market operations and other measures that sustain the size of the Federal Reserve’s balance sheet at a high level. After the Fed, other central banks, including the BOJ, Bank of England and ECB, followed suit and announced forward guidance as a central bank practice. Csortos, Lehmann, and Szalai (2014) analyzed the practice of forward guidance during a crisis and the experiences of the Fed and ECB. The study found that forward guidance is either forecast based or commitment based. In terms of the former, the central bank delivers forecast information based on currently available information and the most likely monetary policy response. Meanwhile, commitment-based forward guidance aims to influence through greater understanding of common temporary strategies, for example, the markets should be tolerant of higher inflation due to a low policy rate. In that context, the Fed and ECB apply commitment-based guidance, while the forward guidance of the BOJ and Bank of England tends to be more state-dependent. Filardo and Hofmann (2014) reviewed the effectiveness of the forward guidance applied by the Fed, ECB, BOJ, and Bank of England. The study measured the impact of the policy rate on financial markets and economy from three aspects: (1) in terms of forward guidance implying a looser future monetary policy stance compared to market expectations, thus influencing short and longterm interest rates; (2) in terms of the central bank providing greater clarity on the future direction of the policy rate, thus the volatility of market expectations concerning future interest rates would be lower and could even lower the risk premium; and (3) in terms of forward guidance mentioning a specific indicator, which would leave the markets more sensitive to the data released for that indicator. Furthermore, although the effectiveness of forward guidance varies over time and between central banks, the study stressed that central bank credibility was the most important factor for the success of monetary policy instruments. To ensure effectiveness, the central bank must clearly communicate forward guidance as a trusted form of credible central bank commitment and interpreted by the markets as intended by the central bank. Moving forward, transparency and communication will become increasingly important for the central bank and also academia. Evaluations and improvements must be undertaken by the central bank in terms of the monetary policy communication strategy and transparency, referring to international best practices, the monetary policy regime applied and current development of the financial markets. The issue of monetary policy transparency is also an interesting object for further study in academia. Moreover, a number of improvements are still required in terms of scientific and theoretical reviews linked to modeling and

Central Bank Evolution and Reform    43 analyzing the influence on economic performance, inflation and the central bank. Empirical studies are still required to analyze the impact of transparency on economic performance and financial market development.

2.4.3. Institutional Arrangements and Coordination Central bank credibility requires strong institutional arrangements at the central bank and through coordination with the government and other relevant authorities. At the central bank, the policy framework must be strengthened in terms of formulating an appropriate policy response to attain the inflation target and financial stability based on macroeconomic projections and macrofinancial linkages, particularly between the banking and external sectors. The integrated analysis and analytical framework underlies formulation of the monetary, payment system and macroprudential policy mix. Research must be strengthened in terms of the effectiveness of the policy framework as well as the accuracy of macroeconomic and macrofinancial projections. Similarly, research and analysis are required to broaden understanding on macrofinancial linkages, procyclicality, and systemic risk. The data and statistics must also be strengthened, including the development of inflation, macroeconomic, payment system stability and financial system stability indicators, and even development of an early detection system, as well as cross-sector balance sheet statistics at the national and provincial levels. The decision-making process must also be strengthened. A difference of opinion has emerged at various central banks, however, about whether it is more desirable to maintain or integrate the different committees for monetary policy and financial stability policy. Kohn (2015), for example, recommends maintaining the different committees, considering the disparate targets and focus, instruments, and accountability. This practice emerged at the Bank of England after the GFC in the form of establishing three committees outside of the monetary policy committee linked to financial stability, namely the Prudential Regulation Authority (PRA), created by the Bank of England for microprudential policy, the Financial Conduct Authority (FCA) for the financial markets, and the Financial Policy Committee for macroprudential policy. At BI, policy formulation and decisionmaking are conducted at the Board of the Governors meeting, concerning monetary, macroprudential and payment system policy. There are also three committees, namely monetary policy, financial stability, and the payment system, headed by the respective deputy government to oversee policymaking. A joint committee was also formed to strengthen the current policy mix, as well as the desired policy mix, through integrated analysis and projections from the three policy aspects. This differs considerably from the monetary policy framework based solely on ITF. Central bank independence is never absolute. Independence is based on policy autonomy pursuant to prevailing laws. Therefore, the effectiveness of central bank policy depends on interaction and coordination with other policies, be it fiscal policy and the real sector by the government, microprudential regulation and supervision by a separate authority or conditions in the business community and financial sector. In the case of Indonesia, BI coordinates closely with the Government and other relevant authorities. Fiscal and monetary policy coordination is necessary

44    Central Bank Policy between BI and the Government when formulating the national budget and various other macroeconomic management aspects, nationally and locally. Although BI has been granted independence under law, the policy mix implemented represents an integral part of the national policy mix, including macroeconomic policy, financial system stability, and structural reforms (Warjiyo, 2013). In terms of financial system stability, pursuant to Law No. 9 of 2016 on Prevention and Resolution of Financial System Crises, national coordination is achieved through the Financial System Stability Committee (KSSK), headed by the Minister of Finance, with the Governor of BI, Chairman of the Financial Services Authority (OJK), and Chairman of the Deposit Insurance Corporation (LPS) as members.

2.5. Concluding Remarks This chapter has discussed central bank evolution and reform since inception in the seventeenth century until after the GFC in 2008/2009 and several important lessons were discovered. First, the central bank plays a significant role in supporting sustainable economic growth by creating and maintaining price (and exchange rate) stability, while preserving financial system stability. This dual mandate has existed since the beginning of central banking but was reemphasized in the wake of the GFC. Second, to achieve the dual mandate, the central bank must be given jurisdiction for monetary policy (including management of foreign capital flows), the payment system as well as macroprudential regulation and supervision. Experience from the ITF era, however, revealed that a central bank focused solely on monetary policy allows financial instability to accumulate, thereby precipitating the GFC in 2008/2009. Third, the central bank needs autonomy to achieve its mandate. Experience from when central banks were under government control demonstrated the adverse effect on macroeconomic stability that led to economic crises. Fourth, institutional strengthening of the central bank is required to ensure that independence is accompanied by stronger public accountability and transparency. How can a central bank build credibility and reputation when pursuing its responsibilities to achieve the dual mandate? Herein lies the importance of theoretical research and empirical studies on the roles of currency, credit, and foreign capital flows in the economy. Through a synthesis of Neoclassical and Keynesian theories, money supply influences inflation in the long term, although there is a short-term trade-off between inflation and economy growth. An efficient financial system is always in equilibrium; thus, currency and credit are perfect substitutes. Likewise, foreign capital flows move freely between countries and, therefore, fully flexible exchange rates are required. Consequently, central bank policy is represented by price stability through setting short-term interest rates similar to monetary policy based on ITF. Nevertheless, the GFC of 2008/2009 reversed the prevailing thinking and assumptions. The financial system is not always in a state of equilibrium, financial frictions stem from information asymmetry, financial product innovation, evaluation methods, and risk management, as well as risk-taking behavior. Furthermore, the financial cycle tends to amplify the economic cycle by accelerating credit growth, asset price bubbles, and leverage, which heighten the risk of a crisis

Central Bank Evolution and Reform    45 occurring. Foreign capital flow mobility is also increasing, thus further amplifying the financial cycle and exchange rate volatility beyond fundamental values. Under such conditions, price stability alone is insufficient to maintain macroeconomic and financial system stability and, therefore, is unable to support sustainable economic growth. In addition, microprudential regulation and supervision to ensure the soundness of individual financial institutions would also not be able to guarantee financial system stability due to the factors that determine macrofinancial linkages in the amplification of the financial cycle. Modern central bank policy since the GFC in 2008/2009 to achieve price (and exchange rate) stability as well as financial system stability is characterized by a mix of monetary policy (including foreign capital flow management), payment system policy, and macroprudential policy. In that context, theoretical understanding of the function of currency and credit in the economy, foreign capital flow mobility, and the exchange rate, as well as monetary policy transmission mechanisms, as discussed in Chapters 3–5, is crucial. The ITF-based policy framework discussed in Chapters 6–9 represents the foundation of the central bank’s policy mix. Similarly, institutional reform through independence on one hand and stronger accountability, transparency and communication on the other is discussed in Chapters 10–12, which is constantly required to strengthen central bank credibility and reputation. The modern central bank policy mix framework since the GFC of 2008/2009 is discussed in full through Chapters 13–15 in terms of the theories and practices adopted at several central banks. It is important to note that policy mix theories and practices represent a new central banking paradigm that emerged after the GFC, and BI has pioneered such practices, especially since 2010.

This page intentionally left blank

Part II

Monetary Policy and Economy

This page intentionally left blank

Chapter 3

The Role of Money and Monetary Policy in the Economy 3.1. Introduction Precisely how monetary policy plays a role in the economy has always been the subject of public debate. Some support a monetary policy focus on price stability. Indonesia’s bitter experience during the 1960s showed that using monetary policy to finance lighthouse projects created hyperinflation, which culminated in an economic crisis and the failure of the economic development program during the Sukarno era. Meanwhile, another section of the public believes that monetary policy should play a role in stimulating output growth. Some hold this view due to political interests, while many others consider that demand for monetary stimuli during an economic downswing or even recession should support the fiscal stimuli of the Government. On a practical level at central banks, the core of the debate has become the essence of how monetary policy should offset the tradeoff between price stability, in the form of low and stable inflation, and fostering economic growth. The issue concerning the role of money and monetary policy in the economy has been the focus of studies for as long as monetary theory has been developed. The most fundamental debate concerns the neutrality of money in the economy, meaning that money only affects prices but is neutral in terms of real output.1 The crux of the debate begins with the proposition of Keynes (1936) in The General Theory, which aimed to attack classical thinking, namely Say’s law − “supply 1

Another proposal is the Classical Dichotomy, which separates the treatment of real economic theory on relative prices from monetary theory on prices. Therefore, in the analysis of factors influencing prices, the role of theory on relative prices can be debated (Johnson, 1967). Starting with the classical argument of economic flows, Lange (1942), that Say’s law − “supply creates its own demand” − logically breaks all the arguments in monetary theory because of the combination with Walras’ law − “the total supply of goods and money to the market must be equal to the total demand of goods and money” − which implies that excess demand of money is zero and realized in the absolute price (intermediate). Central Bank Policy: Theory and Practice, 49–79 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191006

50    Central Bank Policy creates its own demand.” Departing from the classical view of economic equilibrium (full employment) and, therefore, money only affects prices and is neutral in its effect on real output, the proposition of Keynes is based on the assumption that nominal wages are rigid, so if there is a change in the stock of money, real wages would also change and influence real output. In other words, the proposition stresses that monetary economics will reach equilibrium through several possibilities, where full employment is just one extreme condition. Such a concrete proposition has become the overarching focus of subsequent debates between Keynesian and Classical economists, and later with Monetarists, Neoclassical, and Neo-Keynesian economists. Consensus from the protracted debate between the Classical and Keynesian views concluded a synthesis that money only affects prices and is neutral in its effect on real output in the long run but can influence real output in the near term. Nonetheless, one core aspect − whether money is important or not − has not yet been satisfactorily addressed. In the evolution of monetary theory to the present day, opposing economists have developed approaches (economic models) assuming certain theories, which ultimately conclude the neutrality of money in the economy, or vice versa. Metzler (1951) as well as Gurley and Shaw (1960), for instance, showed that money affects real interest rates and output through government debt and open market operations by the central bank. Departing from the Classical and Keynesian economic views, Stiglitz & Greenwald (2003) developed a new paradigm of monetary theory based on the role of the stock of credit, not money, in an imperfect financial market due to asymmetric information between the banks and borrowers. According to such conditions, bank lending, in terms of quantity or interest rates, affects monetary aggregates and the real economy. The five sections of this chapter discuss the role of monetary policy in a closed economy. Following the introduction, the differences between theoretical views concerning the neutrality of money are reviewed in terms of general equilibrium, imperfect information and imperfect markets as well as the view of monetarists. The subsequent section presents the empirical evidence concerning the role of money and monetary policy in the economy, including the results of empirical studies in Indonesia. In the penultimate section, as a new theoretical paradigm, the views of Joseph E. Stiglitz are presented, wrapped up with some concluding remarks.

3.2. Theoretical Review Two issues relating to the structure of Classical theory, which treats relative prices as a factor affected by real demand and real supply, and the price of money as a factor dependent upon the stock of money, were the primary focus of the work of Don Patinkin in the 1950s (Johnson, 1967). In this case, the neutrality of money relates to conditions where an increase in the stock of money would raise all prices by the same magnitude, thus the relative price does not depend on the stock of money. If the effect of money is neutral, an increase in the stock of money would edge up prices without changing the relative price and interest rate and,

The Role of Money and Monetary Policy in the Economy    51 therefore, output. In the terminology of Pigou, money is merely a veil of real activity underlying the workings of a system. Patinkin’s argument on the neutrality of money, formulated as a short-term macroeconomic system with a structure à la Keynesian but based on the assumption of classical behavior, produced a classical paradigm that changes in the stock of money would not change real equilibrium in the economic system, as reflected by changes in the relative price and interest rate. The firm conclusion linked to the argument built on the assumptions required to realize money neutrality, namely wage and price flexibility, inelastic expectations, no money illusion, no distribution effect, homogenous securities, and no government debt or open market operations. The assumptions became a target of criticism from Metzler (1951) as well as Gurley and Shaw (1960), who described conditions where the effect of money was not neutral. Through government debt or open market operations by the central bank, Metzler argued that the relationship between worth and savings used in the Pigou effect analysis by Keynes had theoretical implications where changes in the stock of money could affect the interest rate and, therefore, output growth. Assuming that the real value of government bonds is fixed and the interest on government debt is paid as public income, Metzler showed that rising prices due to monetary expansion through open market operations caused the public stock of real assets to decrease and the propensity to save to increase, thereby lowering the equilibrium interest rate. A lower interest rate due to the distribution effect, assuming that private sector performance is affected by government real debt performance, would stimulate output growth. The seminal contribution of Gurley and Shaw (1960) to the debate on the neutrality of money was not only based on the assumptions of rigidity, the distribution effect, the money illusion, and expectations, but also based on analyses of structural and financial sector intermediation linkages with economic growth and monetary policy, as well as the differing liquidity characteristics of various assets. They differentiated two important concepts of money, namely inside money created by the monetary authority and banking industry in relation to private sector lending as well as outside money, in the form of currency outside of the private sector and not related to loans/claims, for instance banknotes and coins. Money supply that consists of inside and outside money implies that changes in the stock of money not only raise or lower general prices but also the relative price and, therefore, output. This is because if the public holds both types of money, then a change in the stock of one type without a corresponding change in the other would trigger a change in the assets and liabilities ratio, which would, in turn, change real variables in the economic system. In its evolution, the debate on money neutrality has splintered into two concepts, namely neutrality and superneutrality. In terms of general equilibrium analysis, the neutrality of money occurs if (Handa, 2000): (1) all prices rise proportionally; (2) the real value of the endowment does not change; (3) interest rates are paid on all components of money; and (4) no further price changes are anticipated. In this case, money is considered neutral if a once-for-all change in money supply does not affect the values of real economic variables, including

52    Central Bank Policy output. Meanwhile, superneutrality occurs if a continuous change in money supply (a change in money supply growth) has no real effect. In terms of general equilibrium analysis, the superneutrality of money, in addition to the influence of the first three factors of neutrality, is also affected by errors in the formation of inflation expectations. In this regard, the theoretical arguments generally agree that the neutrality of money may not occur, at least in the near term. In this case, money supply is non-neutral if economic imbalances are present. In fact, if there is an increase in the once-for-all of money supply, the non-neutrality of money on real variables can occur during an adjustment phase from one balance to another in the economic system. In this case, an increase in total money supply would not necessarily precipitate a proportional increase in the absolute prices of goods and nominal wages, thus the real price of goods and real endowment value would change and trigger a real change in the economy. In other words, increases in total money supply and inflation have a real effect, at least in the near term, when rigidity and imbalances occur in the economy. In economic theory, a debate emerged on whether the imbalances are merely temporary, followed by quick return to equilibrium, such that the consequences can be ignored. In general, Neoclassical and Modern Classical economists assume that economic dynamics quickly return to balance. The analysis has only focused on equilibrium conditions and, therefore, money is neutral on real economic variables. In contrast, Keynesian and Neo-Keynesian economists generally believe that protracted economic imbalances can persist, therefore money could affect real variables in the economy.

3.2.1. The Role of Money in the General Equilibrium Analysis Money plays a role in the economy in line with its primary functions, namely as a medium of exchange, as a store of value, and as a unit of account. How money affects real decisions and dynamic balance in an economy remains a topic of theoretical and empirical studies. In this respect, a general approach was developed to explain how money affects real decisions and dynamic balance in an economy by linking the presence of money based on certain assumptions in a general equilibrium analysis of an economic system. One important implication when evaluating different approaches to monetary economic theory is to observe how certain assumptions underlie the economic models developed, or on ways to introduce the role of money into an economic model. General equilibrium analysis in a closed economy developed by classical economists is based on equilibrium analysis in the real sector and on the money market. The formula is recursive, namely that balance in the real sector will affect money market equilibrium but not vice versa. Analysis of the real sector consists of two core equations, namely aggregate supply that reflects real output on the production side as a function of capital and human capital, and aggregate demand that explains real output on the expenditure side as a function of consumption, investment and government spending. Balance in the real sector will

The Role of Money and Monetary Policy in the Economy    53 determine real output, real wages, labor absorption, capital formation, consumption, and real investment. Meanwhile, analysis of the money market consists of balancing the supply of money by the central bank and demand as a function of real interest rates and real output. With the recursive formula, equilibrium analysis of the two economic sectors has shown that real output will determine the real interest rate. In other words, total money supply will only affect prices and remain neutral in terms of real output and other real economic variables. According to Keynesian economists, however, general equilibrium theory assumes nominal wage rigidity in the economic system. The assumption was born out of observations during economic recessions after the industrial revolution in Europe at the beginning of the nineteenth century, which demonstrated that labor contract negotiations did not always produce the same level of real wages per the view of classical economists. Therefore, with real wages affected by prices, the equation structure in the economic system becomes simultaneous. Equilibrium analysis in the real sector and money market will simultaneously determine not only the total money supply but also prices and, thereafter, labor absorption, real output, real interest rate, capital formation, as well as real consumption and investment. In other words, nominal wage rigidity implies that the effect of money is not neutral in the economy, meaning that total money supply not only affects prices, but also real output and other real economic variables. General equilibrium analysis concerning the neutrality of money in an economy will depend on the assumptions built into the economic system, including: wage and price flexibility, elasticity of expectations, the money illusion, distribution effect, homogenous securities as well as government debt or open market operations. Of the various approaches available, two are the focus of attention in this section, namely the overlapping generations (OLG) model based on the distribution effect of money between residents, and the money-in-the-utility (MIU) function that can demonstrate whether the money illusion is present in terms of public proclivity to hold currency in the analysis of money neutrality in an economy. Meanwhile, money neutrality analysis based on the effect of information availability on expectations as well as market imperfections and price rigidity will be discussed in the subsequent sections. (a) OLG Model: “Non-superneutrality of Money” A general equilibrium analysis approach that shows the non-neutrality of money based on the distribution effect in an economy is the OLG model. Built based on the consumption-loan model of Samuelson (1958), the OLG model structure appears simple: namely that time is discrete, individuals live in two periods, Nt individuals are born in period t, and the population grows at a rate, n, which can t be normalized to Nt = (1 + n) . Individuals born in period t are labeled young, and in period t + 1 they become old. Each individual is equipped with one unit (consumable) in their youth but not in old age. Individuals have access to storage technology, thus each unit stored in period t will produce 1+ r units in period t + 1. If productive units are stored, then   r > 0 , but if the stored unit is broken, then r = −1. The utility function of the individual born in period t is u (c1t , c2t +1 ).

54    Central Bank Policy For instance, if, during period zero, the government gives money totaling H to the old individual and the money can be exchanged for goods in period t with a price Pt, then the maximization problem for the individual born in period t, t ≥ 0 is Max u (c1t , c2t +1 )(1)



Subject to: Pt (1 − c1t) = Mdt and Pt+1c2t+1 = Mdt, where Mdt is individual demand for money. Considering the lack of natural uncertainty in the model, the condition of perfect foresight is assumed, in cases where the expected and actual prices in period t+1 are the same. The first-order condition for maximization is  P  Mtd = L  t (2)  Pt +1  Pt



which is a function of demand for money and simultaneously a function of savings. The rate at which money is acquired is therefore, Pt/Pt+1, or equivalent to 1 + gt, where gt is the rate of deflation. Per general equilibrium, Walras’ law, which equates demand of the young individual based on the function of demand for money (2) with the inelastic supply of the old individual based on the total money saved, H, is as follows: t

(1− n) Mtd = H (3)



Using (3) and (2) in the periods t and t+1 produces

−1

(1+ gt ) (1+ n) =

L (1+ gt )

L (1+ gt +1 )

(4)

In a steady state, g is constant so, according to Equation (4), g = n. Deflation must be the same as the rate of population growth, implying that prices experience a change at a rate where the supply of money expands at the same pace as demand for money, namely equivalent to population growth. The previous analysis requires constant nominal money stock growth. If nominal money stock grows at a rate, μ, then inflation will be μ − n in a steady state. The influence of money growth on the allocation of resources depends on how the additional money is calculated in the economy. In the case where the additional money is transferred in a lump sum to the old individual (consumables), the maximization problem must be modified in the constraint function of Equation (1) as follows:

Pt (1− c1t ) = Mtd

and

Pt +1c2 t +1 = ∆Mt + Mtd (5)

Therefore, considering that ΔMt is not affected by individual behavior, the solution for the first-order condition is

The Role of Money and Monetary Policy in the Economy    55

 P ∆M  Mtd t = L  t , (6)  Pt +1 Pt +1  Pt

Demand for money is a function of the rate at which money can be accumulated and transferred in the second period, which is exogenous. Meanwhile, the nominal stock of money grows, as an aggregate, at a rate, μ, thus:

Ht +1 = Ht + ∆Ht = (1 + µ) Ht(7)

Additional new money will be distributed equally to the old individual, ΔHt = NtΔMt, therefore, equilibrium on the money market and goods market in period t requires that:

Nt Mt = Ht(8)

In a steady state, the real cash balance per capita is constant, thus Ht/PtNt is also constant, implying that deflation must occur at a rate, g, where 1 + g = (1 + n)(1 + μ), where small values of n and μ are equivalent to n − μ. Considering the impact on inflation and, therefore, the rate at which money can be accumulated, money growth also influences the allocation of resources in the economy. Therefore, money is not “super-neutral” (non-superneutrality of money). Differing from the concept of money neutrality, which emphasizes the neutral impact of money on changes in the levels of real variables, the superneutral concept stresses the influence in the form of changes in growth. In other words, through the distribution effect of money between residents (in this case, between the younger and older generations), the OLG model shows that money supply growth will influence changes in prices (inflation), real output, and other real economic variables. (b) MIU Function Model: Superneutrality of Money General equilibrium analysis, by including the public preference to hold currency in the MIU model, demonstrates the neutral effect of money in the economy. The role of money was first introduced to the model based on the optimal behavior of economic agents by Sidrauski (1967), namely by assuming the direct presence of money in the utility function of each respective economic agent. This is based on the idea that introducing the role of money in the form of a demand function for money in the economy is inadequate, thus requiring that economic agents be shown to prefer holding currency. By including the preference to hold currency as a viable commodity in the utility function of economic agents (consumers), the MIU model can prove whether money influences the decisions of producers when determining the use of production factors (capital) and real output. This approach represents an advancement on the Neoclassical model (the Ramsey non-monetary model) and is recognized in the MIU analysis framework.

56    Central Bank Policy In the model of Sidrauski, each respective individual lives indefinitely, while the population, N, grows at a rate, n. The maximization problem is therefore expressed as follows:

Max Vs = ∫

∞ s

u( ct , mt ) exp[−θ(t − s )]dt ,

uc , um > 0, ucc , umm > 0(9)

where c and m are consumption (C) and real cash balance per capita. An individual can hold wealth in the form of cash (M) or capital (K). The budget constraint function is as follows:

C+

dK dM / dt + = wN + rK + χ (10) dt P

where P is the price, w and r are the level of wages and real interest rate, respectively. X is the transfer from the government. With total household net worth defined as A = K + M/P and transformation of the constraint function in the form of per capita (divided by N), the following equation is produced:

da = [( r − n )a + w + x ] − [ c + ( π − r ) m ](11) dt

where a is total net worth per capita and π is inflation. The equation shows that the magnitude of change in net worth per capita is the difference between income and consumption. Meanwhile, the non-Ponzi-game conditions are:

  t lim at exp−  ∫ ( rv − n ) dv = 0  (12) t →∞  0 

The design problem above requires an optimal control solution, where λt, as a co-state variable, is linked to constraint function above. The Hamiltonian function can be expressed as follows:

H = {u( c, m ) + λ[( r − n )a + w + x − c − ( π + r ) m ]}exp(−θt ).(13) The first-order condition are: uc ( c, m ) = λ um ( c, m ) = λ( π + r )



dλ (14) − θλ = −( r − n )λ dt lim atλt exp(−θt ) = 0

t →∞

The two first-order conditions imply that the marginal rate of substitution between consumption and real cash balance is the same as the nominal interest rate.

The Role of Money and Monetary Policy in the Economy    57 Steady-state analysis involves additional assumptions, per the decentralized Ramsey model, concerning conditions where corporate technology is a constant return to scale, market production factors are competitive, and there is the same lump-sum transfer with seigniorage from issuing currency, namely:2 r = f '( k )

w = f ( k ) − kf '( k ) x=

(15)

dM / dt  dM  M  =    = µm  M  PN  PN

where μ is the rate of money growth. Per steady state conditions, da/dt = dm/dt = dλ/dt = 0, where dm/dt = 0, implying that π = μ − n. Steady state conditions for capital stock, consumption, and real cash balance are as follows: f '( k*) = θ + n

c* = f ( k*) − nk *

(16)

um ( c*, m*) = ( θ + µ )uc ( c*, m*) The salient implication of such conditions is that money growth does not affect the real interest rate or wages as well as capital growth and real consumption, thus, per the steady state, money is superneutral (superneutrality of money). This differs from the common perception that by including the preference of economic agents to hold money, money will influence real output and other real economic variables. The previous analysis showed that money supply growth only affects inflation, in addition to the marginal utility of holding money and the real value of government transfers.

3.2.2. Relationship between Money−Output and Imperfect Information To explain the empirical factors of money neutrality to real output, Neoclassical economists have changed strategy through a different modeling approach and by expanding the assumptions to include imperfect information. This was first achieved by Lucas in 1973. In this case, assuming rational economic behavior in terms of optimizing decisionmaking, or rational expectations, Lucas showed how an equilibrium model using a decentralized market and imperfect information could explain the impact of nominal shocks on output. In the Lucas Island model, the economy consists of separate competitive markets spread over various islands. The shocks are distinguished into two categories, namely aggregate shocks that affect nominal money and idiosyncratic shocks that 2

Fundamentally, understanding seigniorage refers to state receipts, including government seigniorage and central bank seigniorage, which stem from issuing currency.

58    Central Bank Policy have a specific impact at the sectoral level. When determining the decisions of market participants, namely suppliers (producers-workers), in response to special shocks, considering the lack of sufficient information to differentiate, the respective responses of each market player will be different. Therefore the impact of the shocks, including nominal shocks, will influence output. For instance, the supply function for each respective market i = 1, …, n is as follows:

yti = b( pti − E [ pt | I ti ])(17)

where yit and pit are, respectively, the (log) output and output price on market i in period t. I it is the information set available on market i during period t, which contains complete information on economic conditions, including specific prices, P it. The supply function implies that an increase of supply would depend on the perception of the supplier in terms of the relative prices of the goods produced to general prices, Pt, which is known indirectly through an assessment based on information regarding distribution. Based on the preliminary information that the distribution of pt is normal with a prior mean and variance, observations will be made based on the following relationship:

pti ( z i ) = pt + zti (18)

where zi are special shocks, or relative demand-side shocks, with a total value for all markets of zero. The characteristic of relative shocks, zi, is white noise, distributed normally with variance σ2z. Through Bayes rule, the supplier forms a posterior mean from the general price as follows:

θ=

σ 2p σ 2p + σz2



In this case, the posterior mean represents the weighted average of the prior mean and special price on market i, with the weight influenced by the relative variance of the preliminary price distribution and z. If a price change is signaled due to relative demand-side shocks, limited information will cause supplier errors in their perception of which shocks require a response. If σ2z is small, however, the increase in Pit would be perceived more as a signal of higher general prices, rather than relative prices, pti − E [ pt | I ti ], which would cause a small, or no, change in the output produced. From the three equations, therefore, the following can be expressed: where β =

yti = b(1− θ )( pti − E [ pt | I t ]) = β ( pti − E [ pt | I t ])(19) bσz2 , and aggregating among markets, we have σ 2p + σz2

The Role of Money and Monetary Policy in the Economy    59

yt = β ( pt − E [ pt | I t ]) (20)

Equation (20) is known as the Lucas (aggregate) supply function, as explained by the Phillips curve under balanced conditions, namely that output is a positive function of unexpected increases (shocks) in general prices. The aggregate demand is specified simply based on the Classical quantity theory of money, assuming a constant income velocity, equal to zero, as follows (as a logarithm):

mt = pt + yt (21)

The aggregate supply and demand functions produce conditions where E [ pt | I t ]) = E [ mt | t ], and can substitute the function of output and prices on total money supply (actual and anticipated), namely:



 β   ( m − E [ mt | I t ]) yt =  1+ β  t  1   ( β E [ mt | I t ] − mt ). pt =  1 + β 

(22)

From the equation above, it can be concluded that output only responds to unanticipated money changes or money disturbances. The analysis explains the empirical fact of a correlation between money and real output as a short-term phenomenon because of imperfect information among economic agents due to unanticipated changes in money supply. Therefore, Neoclassical economics recommends formulating monetary policy based on rules rather than discretion, thus clarifying and providing assurance when anchoring economic expectations. According to this perspective, money neutrality requires monetary policy to focus on price stability, while real output must be driven by productivity gains and technological advancements. Through its evolution, Neoclassical analysis in the 1980s, which referred to equilibrium, was directed toward developing the real business cycle theory, where long-term economic fluctuations are attributable more to real shocks, such as government spending (Barro, 1986) as well as technology and productivity (Prescott, 1986), rather than nominal shocks, such as money and the erroneous perceptions of economic players.

3.2.3. The Money−Output Correlation against Imperfect Market Competition and Price Rigidity Rebutting the Neoclassical analysis of money neutrality under conditions of imperfect information, Neo-Keynesian analysis points to economic models that evidence a correlation between money and real output based on rational expectations, if nominal (and real) price and wage rigidity exists. This model supports the Keynesian viewpoint that monetary policy, through aggregate demand, affects output, and the long-term neutrality of money but not the short-run non-neutrality of money. Fischer (1977) and Taylor (1980) provided important contributions

60    Central Bank Policy by developing a wage-price mechanism that assumes nominal rigidity and rational expectations and has implications on the role of money and monetary policy in the economy. Fischer (1977) introduced his theoretical model, which can be expressed simply as follows:

y = ( m − p) + u 



y = −(w − p)(23)



w = E ( p | −1)

The first equation shows aggregate demand, with a unit elasticity of money held by the public to output, y, while u is the non-policy demand disturbance. The second equation shows aggregate supply, resolved from the maximization problem of corporate profit, which is assumed to have a unit elasticity to wages, w. The third equation determines wages, showing that nominal wages are set at the beginning of the period based on the information available (including the previous values of m and u) to achieve real wages and the job availability constant expected. Assuming rational expectations, the reduced form solution can be expressed as follows:

y = (1/ 2) ( m − E ( m | −1)) + (u − E (u | −1)) (24)

An important implication is that demand and money shocks can only affect output if the shocks are not anticipated. On one hand, the reduction equation is consistent with the Lucas model, but the channel used is the same as Keynes, namely that prices are flexible but nominal wages do not change, thus demand shocks raise prices, lower real wages and boost output. Considering that wages are only adjusted in one period, in the model, policymakers do not have any additional information than those setting wages and can, therefore, react flexibly, thus minimizing the impact of policy to stabilize output. Fischer’s model was subsequently modified to include staggered wage setting over two periods, which produced the same general conclusions, namely that output performance depended on unanticipated demand and money shocks. Taylor (1979, 1980) developed a staggered wage-setting model over two periods, very similar to Fischer’s model but with an important difference, namely that not only could wages be set earlier but the same nominal value could be set for two periods. The simple version of the model can be written as follows:

y = ( m − p) p = 1/ 2 (w − w (−1))(25)

The Role of Money and Monetary Policy in the Economy    61

w = (1/ 2)  p + E ( p (+1) | −1) + (1/ 2) a  E ( y | −1) + E ( y (+1) | −1) 

Differing from Fischer’s model, Taylor’s aggregate demand equation disregards other shocks except nominal money shocks. The second equation is a price-setting equation, representing the weighted average wage for two periods. In this case, each month, half of the labor force selects a wage for the two periods, namely the current period and subsequent period. Therefore, during a given period, half of the labor force will be paid a total of, w, while the other half will be paid w(−1). The final equation shows that if the labor force sets nominal wages for the two periods, it will depend on current prices and output as well as expected prices and output in the upcoming period. In this regard, money developments are only assumed to be known after wages have been set and, thus, cannot be used as an information set to determine wages. The equation to illustrate wage behavior is as follows:

w = (1/ 2) w (−1) + E (w (+1) | −1) + a  E ( y | −1) + E ( y (+1) | −1)(26)

In this case, in addition to workers taking into account relative wages, with w(−1) and E(w(+1)|−1), the wages are paid to half of the labor force in the current period and subsequent period. If money performance follows a stochastic process in the form of a random walk, the solution in the assumption of rational expectations implies the unanticipated dynamic influence of changes in money as follows:

w = kw (−1) + (1− k ) m (−1) p = 1/ 2 (w − w (−1))(27) y = ( m − p)

where k = (1 − a1/2)/(1 + a1/2) < 1, and k is a measure of real money persistence, while, a, reflects the effect of labor market conditions on wages. A smaller value of a will produce a value of k closer to 1, which indicates a longer persistence of real money influence. Considering that wages are determined prior to knowledge of money developments, such developments will not affect wages and prices in the current period and, therefore, will influence real output. Subsequently, in line with the corrections to wages and prices that occur, real output will gradually return to equilibrium following an exponential decline.

3.2.4. The Role of Money in the Monetarist Analysis Framework: “Money Matters” and the Keynesian-Neoclassical Synthesis The “money matters” proposition, namely that money influences the economy, was explicitly proposed by Milton Friedman and other Monetarists. Friedman

62    Central Bank Policy (1958) argued that changes in money supply influence nominal spending and output (not real output) in the long term. This was subsequently proved using nearly 100 years of US monetary history in Friedman and Schwartz (1963). The empirical evidence also demonstrated a lag concerning the influence of money supply on nominal output, implying the non-neutrality of money in terms of short-term real output in an economy adjusting to a new equilibrium. In addition, Friedman developed a doctrine based on empirical and theoretical studies that the demand for money function, including the relationship with income (income velocity), is stable; thus, strengthening the case for non-neutrality of money in terms of real output in the long term. That doctrine was widely accepted during the early 1960s and contributed to the emergence of the Keynesian-Neoclassical synthesis through investment savings-liquidity money (IS-LM) model framework development to determine aggregate demand. The synthesis emphasizes the need for monetary policy to focus on price stability in the long term, despite the need to also consider the effect on real output in the near term. Another contribution of Friedman, in terms of explaining the linkages between money supply and real economic variables, was through development of the natural rate hypothesis when analyzing the Phillips curve. As elaborated previously, the presence of the Phillips curve under conditions of economic equilibrium showed that output represents a positive function of unanticipated price shocks. In this case, Friedman argued that the natural rate of employment, or output under full-employment conditions, is independent of the anticipated or desired inflation rate, thus output and unemployment fluctuations indeed relate to actual inflation deviation from inflation expectations. This is known as Friedman’s expectations-augmented Phillips curve. Friedman’s proposition concerning the real effect of money was developed more specifically by monetarists through the St. Louis model as an empirical procedure to assess the linkages between national income and total money supply. The model was formulated as a reduced form of the short-term macromodel equation as follows:

Yt = α0 + ∑ i ai Mt−1 + ∑ j b j Gt− j + ∑ s cs Zt−s + µt (28)

where Y is nominal national income; M is the vector of the nominal value of monetary aggregates in the current period and previous period; G is the vector of the fiscal variable in the current and previous periods; Z is the vector of other explanatory variables; and μ is the disturbance term. An assessment conducted by Anderson and Jordan (1968) showed that monetary aggregates have a strong, positive, and rapid effect on nominal national income, which is more significant than the impact of fiscal policy. In general, these findings are in line with Friedman but the lagged effect of monetary policy was much shorter compared to the findings of Friedman and Schwartz (1963). The lag implies adequate space to utilize monetary policy for the stabilization of short-term real output, as proffered by the Keynesian school. The absence of lag would imply that total money supply affects nominal output immediately and,

The Role of Money and Monetary Policy in the Economy    63 therefore, money only affects rising prices and not real output, as found by the Classical school of thought. Consensus between the two economic schools of thought was achieved through the Keynesian-Neoclassical synthesis, which showed that monetary policy can influence real output in the near term but is neutral concerning the impact on the economy in the long term. The Keynesian-Neoclassical synthesis was the subject of a study conducted by Gali (1992), seeking to empirically validate the IS−LMPhillips curve model. To that end, the fundamental structure of the economy was represented through three equilibrium conditions, namely the goods market (IS), money market (LM), and price adjustment process to longterm equilibrium (Phillips curve): y = α + µS − σ (i − E ∆p+1 ) + µIS

m − p = ϕ y − λi + µMD ∆m = µMS

(29)

∆p = ∆p−1 + β ( y − µS ) where (y, i, p, and m) represent, respectively, real output, nominal interest rate, prices, and total money supply, while (μ…) indicate four different types of shock in the economy, namely: aggregate supply, aggregate demand, money demand, and money supply. An evaluation of that equation system for the US data from 1955 to 1987 identified the dynamic response to four economic shocks, which were generally consistent with the qualitative predictions of the IS−LM−Phillips curve analysis framework in line with the Keynesian-Neoclassical synthesis. An aggregate demand shock (at least in the short-term) affects output and other real economic variables due to the lagged adjustment of nominal variables. Monetary shocks (including money demand and money supply shocks) are transmitted to the real sector through changes in the real interest rate. Meanwhile, real output and prices move in the same direction in response to an aggregate demand shock, but in opposing directions in response to an aggregate supply shock.

3.3. Empirical Evidence and Related Issues 3.3.1. Monetary Policy Modeling and Variable Selection Issues How money plays a role in the economy is, ultimately, an empirical question. In this case, based on the empirical literature, consensus is generally reached among economists that money neutrality only occurs in the long term, while in the near term, monetary disturbances can have a significant impact on real variables, such as output. Furthermore, several approaches have been developed to assess the effect of money and monetary policy on real economic activities. The methodology used has evolved over time because of advancements in time-series econometric techniques as well as changes in the theoretical models used. The two main objectives of the testing are to prove whether changes in monetary policy affect real economic performance and to seek empirical justification and consistency

64    Central Bank Policy for the various theories that have been developed. The seminal studies in monetary economics include Friedman and Schwartz (1963), Sims (1972, 1980, 1992), Eichenbaum (1992) as well as Leeper, Sims, and Zha (1996). The study conducted by Friedman and Schwartz (1963), based on the US macroeconomic data covering a period of around 100 years, was based entirely on the timing pattern of the money supply growth and economic growth cycles, which supports the causal relationship between money supply and output fluctuations. From the findings, the data fully supported the conclusion that total money supply growth is a positive reference (with a long lead) of the economic cycle. In this case, faster growth of money supply would be accompanied by a corresponding increase in output above the normal trend and, vice versa, slower money supply growth would trigger a decline in output. Furthermore, variations in money supply would be accompanied by variations in real economic activity, with a long and diverse lead. Criticisms of that conclusion quickly emerged relating to the view that the evidence, based on precise movements and simple correlations, could not show the actual causality of the influence of money, considering a comparison of reference points of the economic cycle tended to ignore the information contained in time-series behavior of the variables of money supply, output and interest rates, thus making it difficult to assess the actual effect of monetary policy on output. Tobin (1970) opined that the correlation between money and output, fundamentally, is the result of reverse causality, namely that output influences money supply. Meanwhile, King and Plosser (1984) stressed the presence of inside money, a component of money supply that represents banking sector liabilities as the main factor explaining the correlation between money and output. Furthermore, it was concluded that the correlation between money supply (M1 and M2) and output emerges as a result of an endogenous response from the banking sector to economic shocks that do not correspond to the application of monetary policy. Efforts by Friedman and other monetarists seeking empirical evidence concerning the links between money and economic activity were not only based on developing graphical analyses and simple correlations. Friedman and Meiselman (1963) developed an econometric modeling approach, which was later developed and became known as the St. Louis model, to analyze the effect of monetary policy or fiscal policy on nominal income. Although more significant and stable empirical evidence was found concerning the effect of monetary policy, rather than fiscal policy, on output, the specification of the approach was criticized, particularly due to endogenous money, which could impact assessment consistency when merely using a simple regression technique. The issue of endogenous money was overcome by an important contribution from Sims (1972), namely by developing the Granger Causality approach for money and nominal income. Sims’ early study concluded that past money behavior (M1 and narrow money) helped explain future nominal income. In a subsequent study, Sims (1980), using the Vector Autoregression (VAR) approach, where the production index was used as a measure of real output and the nominal interest rate as another variable that impacts the economic system, showed that the role of money in terms of explaining variations in output declined

The Role of Money and Monetary Policy in the Economy    65 significantly. This implies that, in the near term, monetary policy could affect real output, but the effect fades in the longer term. Nevertheless, Sims conceded that the approach was very sensitive to which variable specifications should be considered. Relating to the trend of time-series variables, Eichenbaum and Singleton (1986) showed that using the first difference produced a conclusion regarding the effect of money, which was less significant than the specification of level. Departing from that study, Stock and Watson (1989) demonstrated that by systematically specifying the trend, the effect of money could explain output in upcoming periods. Applying the VAR approach to seeking empirical evidence concerning the linkages between money and output has received more attention since a largescale VAR modeling system was development, not just bivariate and multivariate systems. Sims (1992), by calculating several variables in the system, namely production, the consumer price index (CPI), short-term interest rates as a measure of policy impact, money supply, exchange rates, and the commodity price index, concluded that the real output response to the interest rate was in line with the advanced country studies, namely that monetary policy affects output, with a hump-shaped distribution. The conclusion implies that in the near term, monetary policy can influence real economic activity and, in the long term, the effect tends to fade. Another focus of the VAR approach (four variables) was developed by Eichenbaum (1992), namely by evaluating the effect of monetary policy using several alternative measures of monetary policy, such as money supply or policy rate shocks, as well as by analyzing how those alternatives produced puzzling results. If M1 shocks are used as a measure of monetary policy, then a positive M1 shock would be accompanied by an increase in the policy rate and lower output. This output puzzle challenges the standard model and shows that expansive monetary policy would push up output and that money demand is inversely proportional to the interest rate. In contrast, if interest rate shocks are used as a measure of monetary policy, then a positive policy rate shock (contractionary shock) would be accompanied by an output contraction and rising prices. The output puzzle would not occur in this case, rather the opposite would appear, known as the prize puzzle, where contractionary monetary policy would edge up prices. In general, current consensus lies in the fact that the variables used in the VAR modeling system do not represent the availability of information required by the monetary authority to set the policy rate. For instance, based on the prediction of heightened inflationary pressures in upcoming periods, the monetary authority would opt to raise the policy rate. If the policy response is inappropriate or poorly timed, then the higher policy rate would be accompanied by higher prices.

3.3.2. Consensus Regarding the Role of Money and Monetary Policy The general conclusion of the various studies previously conducted is that the role of monetary policy shocks in terms of explaining output follows a hump-shaped distribution. An important question, therefore, is how far the role of monetary

66    Central Bank Policy policy shocks can explain fluctuations in the economic cycle. Using monthly US economic data from the beginning of the 1960s to the beginning of 1996, Leeper, Sims, and Zha (1996) concluded that the role of monetary policy shocks was comparatively unimportant. Nonetheless, that conclusion demands further scrutiny considering that during the observation period, the Federal Reserve applied a different set of monetary policy operating procedures. This was evident in subsequent VAR models for two different sub-samples (1965.01−1979.09 and 1983.01−1994.12), which produced opposing conclusions concerning the role of monetary policy shocks to explain output fluctuations (Walsh, 2001). In this case, the role of monetary policy shocks was less important during the period when the Federal Reserve applied money supply as the operational target (1965.01−1979.09), but the reverse was true when the interest rate was used as the operational target (1983.01−1994.12), namely that monetary policy significantly influenced output fluctuations. Congruent with the criticisms of VAR modeling, studies into the role of monetary policy shocks were also conducted using structural models, large and small. One pioneering approach, using the large-scale econometric model developed by the Federal Reserve, applied a structure that allowed for simulations based on various assumptions concerning the formation of expectations (Brayton & Tinsley, 1996). Based on that approach, it was concluded that monetary policy using the interest rate had a significant impact on output and the influence followed a hump-shaped distribution like that of the simple VAR approach. Nevertheless, differing from the VAR model, which achieved the maximum influence after around two years, the large-scale econometric model achieved maximum influence in a much shorter period, namely around one year. Ultimately, although consensus was reached among economists concerning the role of money and monetary policy shocks in an economy, at least in the near term, no consensus has been reached regarding the transmission mechanism and role of a systematic monetary policy response. Several empirical questions arise, including whether money supply or the interest rate would more relevantly represent monetary policy? How long is the lag? And, how does the monetary policy transmission mechanism affect inflation, output and other real economic variables? This is primarily because of differing views concerning two aspects, namely what is the actual structure? And, to what extent is there interconnectedness between the variables in a policy regime? Operationally, the issues remain difficult to overcome through existing modeling approaches, namely VAR or structural econometric modeling.

3.3.3. Empirical Evidence in Indonesia Several studies exploring the influence of monetary policy on economic activity in Indonesia have been conducted concerning the issue of monetary policy transmission mechanisms (Muelgeni, 2004; Warjiyo & Agung, 2002). Those issues will be explored in more depth in Chapter 5. Using a different framework, Solikin (2005) studied the influence of monetary policy variables on the final policy target, namely output, inflation and job opportunities. The study developed a

The Role of Money and Monetary Policy in the Economy    67 structural cointegration vector autoregression (SVAR) approach. Inter-variable linkages and the magnitude of the monetary policy lag were observed using the impulse response function by identifying generalized functional innovation linkages. Nine variables were shown to influence the economic system, namely the policy rate (BI Certificates − SBI) and narrow money, as policy variables, output (GDP), inflation (CPI), job opportunities, exchange rate, London interbank offered rate (LIBOR), and Jakarta stock exchange (JSX). Using quarterly data, with an observation period from 1980.01 to 2003.04, the study concluded at least four empirical findings as follows: First, the interest rate has an adverse impact on economic growth, is neutral in terms of employment opportunities, and has a positive impact on inflation. The positive impact on inflation is linked to inflation expectation formation (supply side) and the possible accumulation of narrow money due to principal and interest payments on BI Certificates (SBI) sold in the previous period (demand side). The impact on inflation is considered more direct. There are also other channels of influence. On the demand side, this can be indicated through the impact of a higher interest rate on M0-real money balance (negative: 3−6 quarters), and the impact of M0 on aggregate demand and inflation (positive: 3−6 quarters). Therefore, an interest rate hike would anchor inflation expectations on the supply side and subdue inflation on the demand side. Second, the expansion of narrow money would positively impact output (only in the near term, not in the long term), and always impact positively on prices and employment opportunities. The indirect channels of influence could be explained through the impact of an increase to M0 on the interest rate (which is positive in the short term but negative thereafter), and subsequently, the impact of the interest rate on output (positive in the short term) and inflation (positive). Third, the influence of both policy variables on the ultimate policy target tend to be similar, which is due to the ongoing role of both policy variables in terms of operational monetary control during the observation period, where the impact tended toward complementary. During the post-crisis period, the two indicators were mutually responsive. In this case, from the causality test of the impact of the policy variables on the final target, namely output, a policy response in the form of hiking the interest rate during conditions of excess liquidity would be comparatively strong handed in comparison to raising monetary aggregates during an excessive monetary contraction by increasing the interest rate. In contrast, the opposite is true when using prices as the ultimate target of monetary policy. Fourth, by using the policy variables, SBI interest rates and the money base, complementarily, the lag in the impact of monetary policy on output and prices was identified as between one year, when the first extreme (peak) is reached, and two years, when the influence experiences a more persistent trend. Meanwhile, the impact on job availability gradually fades during the periods after the third year.3

3

The impact of monetary policy on job availability is difficult to identify due to, among others, the characteristics of the data, which contain conceptual and technical weaknesses.

68    Central Bank Policy Therefore, the average lag of monetary policy was identified at around 1.5 years, which is consistent with previous observations. One of the problems encountered, however, was that the influence of monetary policy on real economic activity and inflation, was transmitted through the various channels. Therefore, more rigorous and complete understanding of the monetary policy transmission mechanism in Indonesia is required to enrich the substance of the analysis. In relation to the results, it is important to understand the possible emergence of economic phenomena that deviate from theoretical predictions, known as the economic puzzle. In general, several of the fundamental empirical findings of this study corroborated the inter-variable linkages consistent with theoretical predictions. In this respect, the specification of the modeling system successfully negated the liquidity puzzle. Using monetary aggregates (M0) to gauge the general monetary policy stance could also negate the price puzzle and exchange rate puzzle. Of the various economic puzzles observed, only two were interesting enough as puzzles, namely when using the SBI rate to assess the monetary policy stance. The first concerns the impact of SBI rate innovation on higher inflation. Second is the impact of SBI rate innovation on exchange rate depreciation. Fundamentally, an alternative explanation of the two phenomena is interconnectedness between the variables, namely the interest rate, inflation, and exchange rate. This is possible because the exchange rate is also influenced by interest rate and inflation expectations at home and internationally. On the assumption that the influence of international interest rates and inflation is neutral (ceteris paribus), a domestic interest rate hike would (directly) raise inflation expectations and, therefore, exacerbate domestic inflation and exchange rate depreciation. That argument was supported by the findings of a previous study (Solikin & Sugema, 2004b), where a change in the interest rate was an important variable in determining inflation expectations in the manufacturing sector, primarily due to the component of output price setting on the supply side, including the cost of interest.4 Meanwhile, on the demand side, trends are inextricably linked to the use of SBI as the main instrument of monetary control through open market operations. On the other hand, however, is BI’s commitment to pay interest on SBI sold in previous periods. Therefore, the impact of monetary policy shocks through hikes to SBI rates during the initial period would undermine the nominal growth 4

According to Sims (1992), a corresponding increase in inflation and the interest rate (prize puzzle) could be explained by conditions where the monetary authority has sufficient information on possible future inflationary pressures. Therefore, seeking to dampen the inflationary pressures, contractionary monetary policy would be required through a higher interest rate. Under such conditions, prices would continue to rise after the monetary contraction, albeit not as significantly as if contractionary monetary policy had not been implemented. By arguing for such anticipatory measures, several research papers recommend using forward-looking variables, for instance international commodity prices, to overcome the prize puzzle.  In this research, considering that the monetary policy framework in Indonesia is primarily backward looking, the argument does not support the empirical findings.

The Role of Money and Monetary Policy in the Economy    69 of the monetary base because SBI sales would absorb money supply and, subsequently, lower inflation. Thereafter, assuming BI does not tighten monetary policy, the monetary base (and money supply) would accelerate as principal and interest payments are made on the BI Certificates (SBI). In addition to the resultant inflationary pressures, the consistent increase in the monetary base (and money supply) would also precipitate rupiah exchange rate depreciation.5 That argument is also supported by the findings of a previous study (Mochtar, 2003), which demonstrated an increase in SBI demand as the SBI rate begins to rise.6

3.4. Market Imperfections and the New Paradigm of Monetary Economics: Credit Matters The theory of monetary economics presented in the previous sections explained the role of monetary policy in an economy based on the transaction demand for money. On one hand, the Classical view regarding the quantity theory of money (MV = PT) and the Neoclassical view of the real business cycle à la Lucas (1972) or Kydland and Prescott (1977), where prices and wages are assumed to be fully flexible, concluded that money does not have a real impact. Furthermore, by also assuming rational expectations and full employment, money supply expansion would only be interpreted or anticipated by economic agents as an increase of inflation in upcoming periods; and, consequently, prices at that time would increase immediately, thus real income would not change. On the other hand, the Keynesian view, which assumes market failure and, hence, inflexible prices and wages (rigidity), suggests that money does have a real impact. With the presence of rigidity, money supply expansion would make economic agents feel an increase in real wages because prices would not yet have changed, thus driving up real economic activity as well. Although the two schools of monetary economic theory can generally explain the effect of money on the economy, the issue of whether money matters has not been definitively addressed. Specifically, monetary economic theory based on the bank intermediation function has not been developed in response to how money created by the central bank can subsequently be circulated in the financial sector and finance various economic activities. Joseph E. Stiglitz proposed a new paradigm of monetary theory based on the demand and supply of credit (Stiglitz & Greenwald, 2003). In that context, banks have access to asymmetric information concerning borrower conditions, thus encountering default risks when disbursing loans. Therefore, understanding the behavior and ability of banks to manage risk is crucial when faced with imperfect information in terms of lending. Considering

5

Theoretically, the phenomenon of rising inflation due to tight monetary policy in the previous period is known as the “unpleasant monetarist arithmetic” (Ljungvist & Sargent, 2000; Sargent & Wallace, 1981). 6 Mochtar (2003) showed that using government securities (T-Bills) as an instrument of open market operations had a different impact on inflation than raising the interest rate.

70    Central Bank Policy that credit performance affects monetary aggregates as well as output and other real variables, the banks’ behavior will therefore determine institutional economics. This represents a departure from traditional monetary theory that considers the performance of banking institutions as a given or exogenous. Several reasons have been proposed on why focusing on banking institutions is important, the most fundamental of which is the role of banks in terms of the intermediation function, namely by mobilizing deposits (demand deposits, savings deposits, and term deposits) from the public and subsequently disbursing the funds as loans to finance economic activity. In this case, monetary policy would have a direct effect on bank deposits and total money supply (M1 and M2) in the economy. Nevertheless, monetary policy affects not only credit performance but also bank lending behavior in the face of asymmetric information and credit risk. In other words, changes in monetary policy do not always feed through to proportional changes between bank deposit and loans and, therefore, the loanto-deposit ratio (LDR) is not always 100%. Credit performance is affected more directly, compared to the influence of money supply on output and other real economic variables, therefore, monetary economics should be based on the banks’ lending behavior. Rapid growth of the financial sector has also had a profound effect on monetary policy. Development of the money market account (MMA), for instance, through the securitization of long-term government debt securities in the United States will continue to replace term deposits at banks, which means the effectiveness of monetary policy through the quantity of money (M0, M1, and M2) will fade and, therefore, monetary policy should be based on its effect through the interest rate. Meanwhile, globalization and financial sector integration will bring parity between domestic and international interest rates, thus leaving standard monetary policy theory impotent in the case of a small open economy. Nevertheless, this does not mean that efforts to stimulate the economy through monetary policy cannot be applied, due to many factors other than interest rates that affect lending, for example, through the reserve requirement, LDR, prudential principles for lending and by allocating subsidies to local banks. Such thinking demonstrates the importance of monetary policy based on analysis of the structure of the banking sector in terms of implementing the intermediation function in a country. Does money matter? Stiglitz believes that money (monetary institutions and policy) matters; the only problem is that the influence of money is s­ometime too small to be detected.7 Furthermore, the definition of money is becoming

7

Following Keynes’ three motives of demand for money, most money is not under the control of the central bank. The very definition of money itself has become increasingly blurred with the advent of financial engineering. Differences between the checking account (typically low interest) and the T-Bill account (market rates) have also become unclear. In reality, the use of money is not linked to the income generating motive but to transactions on financial markets (decoupling). Nonetheless, it is not money that is used for many financial transactions but credit. Under such

The Role of Money and Monetary Policy in the Economy    71 increasingly blurred and much falls outside the control of the central bank due to financial transaction and product innovation, including, more recently, cardbased payment instruments and electronic money. With the emergence of such issues, monetary policy is still expected to influence the real sector due to: (1) market imperfections, for instance rigidity of nominal prices and wages; and (2) the redistributive effect of monetary policy through the financial sector. Therefore, the focus of alternative monetary theories should be directed toward credit availability, namely in terms of the supply and demand of credit, which is linked directly to imperfect financial markets and takes into consideration the distributive effect of credit on real sector activity. Therefore, per Stiglitz, the prevailing trend is credit matters rather than money matters.

3.4.1. Credit Availability and Bank Behavior: Credit Rationing Equilibrium Credit availability in an economy is closely related to bank behavior. Fundamentally, banks are risk averse because banks are limited in terms of diversifying and distributing the risks. This is linked to the very essence of the function as a trusted intermediation institution, which, with relatively little capital, can mobilize the funds of the public and disburse them in the form of loans and other economic financing. The comparatively small amount of capital leads to solvency risk at the bank, which underlies the numerous regulations issued and enforced by the relevant authority to ensure the soundness of individual banks (microprudential policy). The banks also face liquidity risk because the depositors can withdraw their funds at any time, but the funds themselves have already been disbursed as loans. Lending also contains risks (non-performing loans − NPL) because the borrowers may be unable to repay the loan due to economic or other reasons. With the risk-averse behavior of banks, any changes that occur in the economy will affect the availability of bank credit and economic activity in general. Similarly, the monetary policy response to influence the availability of credit (loanable funds) is also affected. During a recession, however, banks tend to apply stricter lending standards, which exacerbates economic conditions. Furthermore, monetary policy effectiveness in terms of influencing bank lending also fades. In this case, risk-averse behavior in the banking industry and imperfect information when lending can create a phenomenon known as credit rationing, as the price mechanism fails to bring about equilibrium in the market. Therefore, credit market equilibrium is reached in terms of total credit, where demand outstrips the supply of loanable funds and, subsequently, general economic equilibrium is

c­ onditions velocity is no longer constant or predictable as assumed. Meanwhile, even if the Neoclassical view always stresses that macroeconomic analysis must be based on microeconomics, the assumptions are still ad hoc, for example, money could suddenly appear in the utility or production function or be perceived as money donated to overcome the cash-in-advance (CIA) constraint.

72    Central Bank Policy reached at a level of real output that is below full employment and oversupply of labor or unemployment (Stiglitz & Weiss, 1981).8 Credit market equilibrium with credit rationing can be illustrated as follows. When lending, banks will consider the interest rates and risks. Despite monitoring and selecting prospective borrowers, the banks still have incomplete information concerning the borrowers’ actual conditions due to asymmetric information. Therefore, the banks tend to include credit risk (NPL) in their decision to offer how much credit and at which rate to the borrower. On the other hand, however, the interest rate stipulated by the bank affects the repayment capacity of the borrower. With full knowledge of the business conditions rather than the information known to the bank, the borrower will consider a lower level of credit risk (or no credit risk) when applying for a bank loan. With the influence of credit risk when setting the interest rate, conditions where demand for credit is the same as supply may not be reached. Generally, credit rationing occurs, where credit market equilibrium is achieved in terms of interest rates and total credit, with demand outstripping the supply of loanable funds. In addition to credit rationing, another lending phenomenon is known as adverse selection bias, which appears as a consequence of different borrowers having different repayment capacities. Considering the expected return of the bank depends on the repayment capacity of the borrower, the bank will strive to identify borrowers with the highest repayment capacity through various screening techniques. Nevertheless, considering how difficult it is to identify such borrowers, the banks screen potential borrowers referring to the interest rate the borrower could afford to pay (on average), the borrower who is willing to pay a higher interest rate may actually contain more risk. Adverse selection bias in terms of lending can manifest considering that the ability of the borrower to pay a high interest rate is possible due to the low probability of the borrower repaying the loan. Therefore, higher interest rates lead to a higher average degree of debtor risk, which erodes the profit obtained by the bank. Consequently, increases in the expected return are slower than increases in lending rates and, under certain conditions, the expected return could decline, as illustrated in Fig. 3.1. In this case, the interest rate at which the expected return is maximized is known as the bank-optimal rate, where there is clearly excessive demand for loanable funds. According to loanable fund theory, the interest rate is determined by the demand for and supply of loanable funds at a point where the demand for loanable funds curve intersects the supply curve of loanable funds (Fig. 3.2). During

8

Traditional monetary theory states that monetary policy is an effective means by which to control economic activity. In this case, the government or central bank can control the money supply, which affects interest rates and, thus, aggregate demand. Consequently, the economy is also believed to tend towards equilibrium. Per the new monetary economics paradigm, however, the self-adjustment mechanism is not fully functional, thus making economic equilibrium difficult to achieve or prompting a protracted recession.

The Role of Money and Monetary Policy in the Economy    73

Fig. 3.1:  Bank-optimal Rate.

Fig. 3.2:  Market Equilibrium Loanable Funds. a recession, however, the demand curve (influenced by the demand for investment goods) shifts leftwards. Meanwhile, the supply curve (influenced by savings deposits) also moves to the left because savings will also decline in line with decrease of income. During a recession, the supply decline is more significant than the drop in demand (implying excessive demand). Under such conditions, the interest rate would not be determined by the intersect between the supply and demand curves because the borrower will face a higher real interest rate, which exacerbates matters. After understanding the behavior of the equilibrium (optimal) interest rate and equilibrium in the loanable funds market, the next stage is to understand how determining credit market equilibrium is linked to credit rationing, as presented in Fig. 3.3. Considering that demand for funds depends on the interest rate set by the bank, r, while the supply of funds is influenced by the average expected return, P, using the conventional supply and demand diagram is not possible. Quadrants I and IV represent interest rate equilibrium as elaborated previously, while Quadrant III shows a positive correlation between the supply of loanable funds and the expected return. If a bank is free to compete for depositors, then P is the interest rate received by the depositors. From Fig. 3.2, credit rationing equilibrium occurs when demand for loanable funds exceeds supply. When the bank sets the interest rate, the income per unit

74    Central Bank Policy

Fig. 3.3:  Equilibrium Credit Rationing. of currency of the funds borrowed will decline. In this case, the amount of excess demand for funds is estimated as Z. It is important to stress that there is a rate at which the demand for loanable funds is the same as the supply of loanable funds, namely rm, but the rate is not the equilibrium interest rate. Therefore, the bank could increase its profits by setting rˆ *, not rm. Through comparative statics, excess demand for loanable funds will decrease in line with an increase in the supply of loanable funds. Nonetheless, the interest rate determined by the bank will not change as credit rationing is maintained. Ultimately, congruent with the increase in the supply of funds, Z can decline and reach zero, prompting lower market rates.

3.4.2. A New Paradigm in Monetary Economics Theory Stiglitz’s criticism of traditional monetary theory was not due to a perceived lack of persuasiveness of the theoretical foundation but more because of the erroneous direction of the approach to analyzing monetary policy; with a focus on the incorrect policy variable, for instance money supply or the interest rate, which fundamentally have an unstable relationship with output and other real variables, particularly during a crisis.9 Meanwhile, an analysis focused more on credit, 9

Per traditional monetary economics theory, the emphasis of the role of money, as reflected by the presence of the LM curve (the transactions motive for money demand), received criticism due to the unstable nature of the relationship between money and output.

The Role of Money and Monetary Policy in the Economy    75 rather than money, and bank behavior, better represents how monetary policy influences the economy. Stiglitz’s proposition ushered in a new theory paradigm and monetary policy based on credit and bank behavior. The focus was not merely limited to how credit and the interest rate affect output and other real economic variables but also on how bank behavior transmits the effects of monetary policy in the economy. Furthermore, the analysis is based on the factual phenomenon that financial markets are always in an imperfect state (rather than general equilibrium) due to various factors, including asymmetric information, moral hazard, default risk, the influence of net worth, and so on. Stiglitz and Greenwald (2003) provided a detailed elaboration of their new monetary economics paradigm. First, that the quantity of credit is clearer and directly affects (depending on the quantity if money) output and other real variables. The quantity of credit disbursed by the banking industry is clearly linked to the need for financing for production (capital and workforce) and spending (consumption, investment, exports, and imports) that creates output. On the other hand, the correlation between the quantity of money and output is becoming weaker and more difficult to predict. In addition to currency, the definition of money supply continues to develop in line with rapid financial innovation, therefore including products that are only traded on the financial markets and not just linked to transactions and economic financing. Fluctuations in the quantity of money are operational target always consistent with the quantity of credit. In banking terminology, such conditions are reflected in a LDR that is not always one or fixed over time. Second, the interest rate is not a viable instrument or indicator to represent monetary policy. The spread between the funding rate (demand deposits, savings deposits, and term deposits) and the lending rate is not always stable because of several factors, including operating costs, profit margin, and the risk premium. Moreover, the interest rate only represents one factor in the decision of the bank to extend a loan. In general, credit is not allocated through an auction or market mechanisms but through a selection process by the bank using an assessment of net worth based on the relevant risk factors. In this case, the fundamental characteristic of the credit market is imperfect competition because the bank may not have complete and accurate information on the potential risks of the prospective borrower. As a result of asymmetric information and internal conditions, credit rationing has become a phenomenon in the banking industry, thus interest rates and the quantity of credit are determined during excess demand on the credit market. Such conditions represent a complete departure from traditional monetary theory, which typically assumes general equilibrium, where the interest rate will offset the demand and supply of credit. Third, bank behavior and internal conditions also affect credit performance and, therefore, impact the economy. Internal conditions, including capital, liquidity, profitability, and NPLs, strongly influence bank lending ability. Similarly, when anticipating the economic outlook, the banking sector tends toward procyclical behavior, implying more lending during an economic upswing (boom) and extreme prudence during an economic downswing (recession). The banks’

76    Central Bank Policy funding preference and strategy (retail a or wholesale) as well as credit strategy (retail or corporate) will also have an influence. Therefore, the interest rate and quantity of bank funds (demand deposits, savings deposits, and term deposits), which generally represent the interest rate and money supply in traditional monetary economics theory, cannot always explain credit developments and the influence on the economy. Understanding bank behavior and internal conditions are crucial in terms of economic and monetary policy theory. Fourth, the availability of complete and accurate information on the general conditions of the economy, business dynamics, and internal conditions of the customer is also vital for the banking sector. Despite screening and monitoring prospective and existing borrowers, the banks will never be able to fully detect and anticipate potential business failure or the risk of loan default. The agency problem between bank and borrower could surface due to asymmetric information and other factors. The borrower’s business could fail due to an inappropriate business strategy, moral hazard or general economic conditions. Consequently, the influence of information availability on bank lending behavior is an important element of the new paradigm of economic and monetary policy theory. The new paradigm of monetary economics has several fundamental implications for central bank monetary policy. First, the monetary policy stance (tight/ loose) can be measured more precisely based on credit availability, rather than total money supply or real interest rates. To that end, a methodology needs to be developed to measure the optimal level of credit availability in the economy, which should be linked to inflation, output, and other macroeconomic variables. The range and threshold would help determine how consistent the credit gap is with the goal of price stability as well as the trade-off with economic growth. The credit gap can be used as an important indicator of monetary policy in conjunction with the output gap (the gap between actual and potential output) and the interest rate gap (the real interest rate compared to the natural rate of interest) when formulating monetary policy. Second, the interest rate can still be used as a monetary policy instrument by strengthening its effectiveness. The central bank implements its monetary policy stance by influencing interest rates (and liquidity) on the money market, which is subsequently transmitted to term deposit rates, lending rates and interest rates on other financial assets, thus affecting aggregate demand, output, and inflation. Nevertheless, the effectiveness of monetary policy transmission through the interest rate channel depends on banking and financial sector efficiency. Operating costs, profit margin, and the risk premium affect the magnitude of spread between lending and funding rates (demand deposits, savings deposits, and demand deposits) as well as money market rates. Therefore, market deepening efforts to boost efficiency are required to narrow the interest rate spread, including regulations concerning transparency and market conduct for banks and other financial institutions when setting interest rates. Third, bank regulatory policy plays a salient role in terms of strengthening the use of traditional monetary instruments to influence lending. In addition to the interest rate and reserve requirement as monetary instruments, bank regulations on capital, liquidity, profitability, and loan loss provisions have a significant

The Role of Money and Monetary Policy in the Economy    77 impact on credit. Furthermore, development of the credit information system is also important to reduce the problem of asymmetric information faced by banks when disbursing loans. It is important to stress that bank regulatory policy, in terms of monetary policy, must be based on bank behavior theory. The application of an inappropriate approach will be inefficient (pareto) and also counterproductive, which could actually exacerbate the risks faced by the banking industry. If banking supervision and monetary policy are conducted separately by two institutions, sound coordination is required between the two. Furthermore, bank regulatory policy and monetary policy are best applied not only paying due consideration to aggregate demand but also to aggregate supply. Fourth, there is a greater need for policymakers to understand how monetary policy influences bank behavior. Banks tend to be procyclical, meaning they offer more credit during an economic upswing and are more prudent lenders during a downswing. Consequently, during a period of economic moderation, to prevent exacerbating conditions further, the monetary authority must respond with extra prudence by tightening the monetary policy stance to avoid the probability of bank default. Relaxing banking regulations, if possible, may be considered under such circumstances. Conversely, to control aggressive bank lending during an economic boom, the central bank may consider a combination of raising the interest rate and tightening bank regulatory policy. Under both sets of conditions, the central bank must meticulously monitor bank conditions and behavior. Fifth, it is becoming increasingly important for the central bank to use micro, disaggregated, and granular analyses. Credit is heterogeneous, where excess funds at one bank cannot substitute shortfalls at another. The market and industry microstructure can cause frictions and disruptions to the circulation of funds and credit between banks and economic sectors. Therefore, policy that refers to aggregates tends to be misleading. The sectoral impact of monetary policy is not proportional, therefore, an overreliance on monetary policy may be distortive and, consequently, undermine effectiveness. Micro, disaggregated, and granular analyses are required to determine how far the traditional monetary instruments, such as the interest rate, need to be complemented by bank regulatory policy to reduce the distortions that emerge. Finally, central bank or government intervention is possible if distortions cannot be corrected using traditional monetary instruments or bank regulatory policy. In general, loans are not disbursed through auctions and market allocation is not pareto-efficient, therefore intervention could raise prosperity to the pareto-optimal level. Under such conditions, the government could create a credit institution or scheme to complete the missing parts of the system, particularly in terms of the allocative and redistributive impact of credit on public welfare, for instance through interest rate subsidies or providing credit programs for micro, small, and medium enterprises (MSME). During crisis conditions, particularly during periods of financial system restructuring, the government could introduce measures to maintain the intermediation function by providing a blanket guarantee scheme to prevent bank runs and tax incentives for credit restructuring, or by showing willingness to purchase high-quality financial assets that the market cannot absorb.

78    Central Bank Policy

3.5. Concluding Remarks In this chapter, monetary policy and theory have been discussed in the context of a closed economy. From the theoretical studies and empirical evidence, the neutrality of money in an economy fundamentally depends on the assumptions used. The lengthy debate between Classical and Keynesian economists has produced a synthesis that money is neutral in the long run but not in the near term. Therefore, in the long term, total (growth of) money supply only influences prices (inflation) and not real output (economic growth), but affects both variables in the short term. Consequently, monetary policy should be directed toward achieving price stability, in other words low and stable inflation, while considering the impact on economic growth, particularly in the near term. To what extent there is a trade-off between inflation and economic growth in the near term depends on the assumptions underlying monetary economics in a country’s economy. In an economy with perfect markets (high economic flexibility) and rational economic agents (with complete information), per the assumptions of classical theory, the short-term trade-off between inflation and economic growth tends to be small. Nonetheless, those assumptions are not always present in an economy, as observed by Keynesian economists. Price and wage rigidity, incomplete information and irrational economic behavior are all present, which means a market is not always in equilibrium and, therefore, has implications on quantity and prices. Total money supply also has distributive and allocative effects through the impact of real interest rates on output, among others, due to holdings of government bonds and central bank monetary operations. The various factors explain why monetary policy not only affects prices but also output and other real economic variables. The discussion on theory and empirical findings also showed that the real interest rate is a more appropriate monetary policy instrument than money supply in terms of its effect on the economy. The relationship between money supply and output is not stable and difficult to predict due to the innovation of products similar to money that may only be traded in the financial sector and do not relate directly to economic transactions. In fact, reverse causality has been found, meaning that economic growth has more of an impact on demand for money supply. Consequently, central bank monetary operations should be directed more toward influencing interest rates on the money market, which would be expected to affect interest rates in the financial sector and, thus, aggregate demand, inflation and economic growth. Therefore, how effectively the policy rate is transmitted to various interest rates in the financial sector and economic variables becomes significant. In this case, the assumptions debated between Classical and Keynesian economists, including the flexibility of economic sectors, the completeness of information, and the rationality of economic agents, also influence the effectiveness of the interest rate in the economy. Nobel prize winner, Joseph Stiglitz, proposed a new paradigm of monetary economics theory, believing that the availability of bank credit has a direct effect on various economic activities. Monetary policy using credit availability, concerning quantity and the interest rate, is more effective in terms of influencing inflation

The Role of Money and Monetary Policy in the Economy    79 and economic growth than monetary policy using money supply. Furthermore, imperfect markets and asymmetric information in the credit market cause the phenomenon of credit rationing, where credit quantity and interest rates are determined under conditions of excess demand. Therefore, credit availability is not only determined by the interest rate, but also by internal bank conditions and the information available concerning general economic conditions and the conditions of the customers. Consequently, the effectiveness of monetary policy transmitted through the interest rate channel should be strengthened through bank regulatory policy, the provision of general economic information and a debtor information system, as well as greater understanding of business microconditions and bank behavior. Understanding the various aspects regarding the role of money and monetary policy presented in this chapter is crucial for the discussions in subsequent chapters. The mechanisms and effectiveness of monetary policy transmission, in terms of influencing the economy through the money supply channel, interest rate channel and bank credit channel will be discussed in Chapter 5. The strategic and operational frameworks of monetary policy as applied by the central bank will then be explored in Chapters 6 and 7. Those two chapters will lay the foundation for the discussions in Chapters 8 and 9 on the ITF commonly applied by central banks, including BI. Bank regulatory policy to influence credit performance as a macroprudential policy measure has become more widespread since the GFC in 2008, which will be discussed in Chapters 15, along with the linkages to monetary policy in a central bank policy mix.

This page intentionally left blank

Chapter 4

Exchange Rates and the Economy 4.1. Introduction Monetary policy and theory in an open economy are dimensionally disparate from monetary policy and theory in a closed economy on two levels. First, the flow of foreign exchange from international trade and investment activities manifests on the balance of payments (BOP). Foreign exchange also affects total money supply in a country and is transmitted to monetary and real sector variables. Second, exchange rates also have an economic impact due to domestic and external factors. The exchange rate also influences the relative prices of goods and financial assets, thus affecting the economic decisions of economic players. Consequently, flows of foreign exchange and the exchange rate demand constant vigilance and represent important considerations underlying monetary policy formulation by the central bank. Since the GFC of 2008/2009, external influences on domestic economic developments have increased and become harder to predict. Through the trade channel, global spillovers have occurred because the sluggish global economic recovery has undermined export performance and, therefore, the magnitude of foreign exchange flows from international trade. Meanwhile, through the financial channel, flows of foreign exchange from international trade have become more fluctuative, especially during the period of US monetary expansion from 2008 to 2013, which subsequently subsided but became less certain after the Federal Reserve announced its plan to normalize monetary policy in May 2013. Increasing exchange rate volatility and foreign exchange flows since the global crisis have become a complex challenge for central banks when formulating appropriate policy to mitigate the impact of global spillovers on domestic monetary and financial system stability. The above description demonstrates the close interconnectedness between the exchange rate, foreign capital mobility and the central bank’s monetary policy response in an open economy. In the literature for monetary economics and international finance, that interconnectedness has become known as the impossible trinity. More specifically, the theory states that exchange rate stability, foreign capital mobility and monetary policy autonomy to achieve domestic price stability are three goals that are impossible to attain simultaneously in an

Central Bank Policy: Theory and Practice, 81–113 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191007

82    Central Bank Policy open economy.1 Only two of the three goals are attainable at any one time. If an authority chooses to pursue exchange rate stability and foreign capital mobility, for instance, monetary policy autonomy to achieve domestic price stability would be difficult to attain because the significant influence of foreign capital flows on money supply and, subsequently, domestic prices. Meanwhile, if foreign capital mobility and monetary policy autonomy are desired to achieve domestic price stability, exchange rates would tend to fluctuate in line with the magnitude of foreign capital flows in order to maintain control over domestic money supply without affecting price stability. The issue of the policy trilemma has dominated central bank discussions on monetary policy and theory in an open economy. This chapter will present in-depth discussions on exchange rate policy and theoretical aspects, while the discussion on foreign capital flows in terms of the policy trilemma in an open economy will be discussed in Chapter 14. The discussion in this chapter is arranged into five sections. After the introduction, exchange rate theory will be discussed in the second section based on an analytical framework in relation to exchange rates, foreign capital flow mobility, and the monetary policy response in an open economy. Various relevant theories, including the Mundell–Fleming model, Dornbusch’s overshooting model, and exchange rate models with imperfect information and an imperfect foreign exchange market microstructure, shall be discussed. Then, the empirical findings of the impact of the exchange rate on inflation, growth, and other economic variables in various countries and Indonesia shall be presented in the third section . The fourth section will discuss the exchange rate system and exchange rate policy practiced in several countries, highlighting the debate on bipolarization of the exchange rate system, namely to apply a pegged or floating system. The causes and fallout of exchange rate crises, as well as the appropriate policy response to stabilize the exchange rate will also be discussed in the conclusion.

4.2. Exchange Rate Determination Theory Numerous theories have been developed in the literature concerning international economics and finance to explain the exchange rate and relationships between variables in an open economy. Initially, exchange rate determination theory tended to emphasize international trade on the BOP. Nevertheless, as global financial markets became more developed and financial liberalization was pursued in various countries, exchange rate determination theory tended to evolve with a focus on capital

1

In this case, monetary policy autonomy is the central bank’s ability to formulate monetary policy to achieve price stability, particularly in the face of external shocks such as the impact of economic developments, inflation, and external interest rates. The term should not be confused with the institutional independence of the central bank, namely to use monetary policy instruments (instrument independence) and remain free from intervention by a third party (institutional independence), as discussed in Chapter 12.

Exchange Rates and the Economy    83 transactions on the BOP. Modern approaches view the exchange rate as an asset on the international financial market, which emphasize monetary aspects, currency substitution, and cross-border investment portfolio balance. Fig. 4.1 clearly illustrates the evolution of exchange rate determination theory. The analyses contained in various theories are not only limited to exchange rate determination theory but also the impact of the exchange rate on various economic variables and the implications on the macroeconomic policy pursued. Although not all theories successfully explain the complex exchange rate puzzles and inter-variable relationships in an open economy, the various theories described below provide a framework to analyze the linkages between three important aspects of international finance and economics, namely exchange rate stability, cross-border capital mobility, and the macroeconomic policy response, particularly monetary policy in an open economy. Exchange rate determination theory was initially developed through purchasing power parity (PPP), which is also recognized as exchange rate inflation theory. Fundamentally, PPP theory states the law of one price for various products traded internationally.2 This theory is based on the concept of flows in terms of

Fig. 4.1:  Exchange Rate Determination Theory Source: Gartner (1993, p. 28), modified.

2

The law of one price states that if calculated in the same currency, freely traded commodities have the same price assuming perfect market mechanisms (namely small transaction costs, tax free, homogenous products, and no uncertainty). If the prices of homogenous products are different from one another, arbitrage will occur to take advantage of differing prices for the same asset until the prices are the same.

84    Central Bank Policy exchange rate determination from international trade activities, where the flow of demand for foreign exchange to pay for imports is the same as the flow of supply of foreign exchange proceeds from exports.3 If there are no barriers to international trade or arbitrage taking advantage of foreign exchange trade, the prices of goods in each country will be the same when prevailing exchange rates are taken into consideration in absolute terms, namely P = P* + S, where S is the nominal exchange rate, while P and P* are the domestic and international prices of goods. In relative terms, however, if the methods and basket of goods used to calculate inflation are not too dissimilar, then according to PPP, changes in exchange rates will mirror the differences in inflation between countries, namely ΔS = π − π*, where π and π* are domestic and international rates of inflation, as put forward by Cassel (1918).4 As trade has expanded to many countries, PPP theory has become the foundation for measuring weighted exchange rates in real terms, otherwise known as the real effective exchange rate (REER), namely REER = Σ ωjπ*j /π, where ω is the respective value of trade as a share of each trading partner. In general, REER is expressed as an index with a base year that reflects normal conditions of the country in question. The REER index can be used as a measure of currency misalignment from fundamental conditions. Furthermore, the REER index is also used to measure real exchange rate competitiveness, thus facilitating analysis of the impact on exports and imports. Departing from PPP, which is based on the trade balance, interest rate parity (IRP) theory emphasizes the concept of capital flows between countries in the capital account. With the presence of the international asset market, IRP conditions are congruent with the law of one price of cross-border interest rates. Such conditions apply, assuming that the foreign exchange market is efficient in terms of transforming information into exchange rate movements and does not face significant transaction barriers to perfect market competition. Perfect capital mobility is also present with assets traded on relatively homogenous financial markets with perfect market substitutability.5 Under such conditions, the returns on investment or interest rates in different countries will be the same after exchange rates are taken into consideration. IRP theory can be expressed in the form of covered interest rate parity (CIRP) or uncovered interest rate parity (UIRP). Per UIRP, the return on investment or domestic interest rates will be the same as international interest rates after market

3

PPP is considered a flow concept because it is based on the flow of goods and services in the trade balance to determine exchange rates. 4 If speculative behavior is taken into consideration, Roll (1979) suggested an alternative PPP equation in the form of expectations, namely in absolute terms: E(P) = E(P*) + E(e), or in relative terms, where π  −  π* = E(et+1  −  et), where E(.) is the ­expectations operator. 5 Examples of capital market barriers include restrictions on cross-border capital ­mobility (such as foreign exchange controls), transactions costs, tax as well as d ­ isparate risks when investing in securities.

Exchange Rates and the Economy    85 expectations of interest rate fluctuations are taken into consideration, namely r = r* + [E(S) − S]/S, where {r,r*} = domestic and international interest rates and {E(S), S} = expected and actual spot exchange rate. Meanwhile, CIRP indicates IRP for cross-border investment hedged against currency risk using forward transactions, thus r = r* + (F − S)/S, where F = forward exchange rate. IRP may occur because arbitrage on international financial markets will prevent abnormal profits from investment transactions or borrowing on the domestic asset market compared to the international market. Therefore, interest rate disparity between countries will be lost due to capital mobility and arbitrage.

4.2.1. Mundell–Fleming Model: Policy Trilemma Departing from the PPP and IRP theories that use partial modeling, various exchange rate determination theories were subsequently developed in the 1970s from the perspective of an open macroeconomic system. According to the IS-LMPhilips Curve, the relationships between domestic and international economic variables can be analyzed using the BOP, with the exchange rate represented as an asset price on the international financial market, through a monetary, money substitution, or portfolio balance approach (PBA). The monetary model, for instance, emphasizes the use of money only in the domestic economy, whereas the money substitution model assumes the internationalization of money and in the portfolio balance model there are is money along with other financial assets. In this case, there are differences in dealing with price behavior and capital mobility between countries. The neoclassical model of Mundell (1963) and Fleming (1962) was the first monetary model to determine exchange rates assuming fixed prices (thus possible to ignore) due to underemployment. In an open economy, this model assumes that the transportations costs between countries are very low, thus satisfying PPP conditions, while the risk premium and UIRP are effective and cross-border capital mobility is perfect. Therefore, the Mundell–Fleming model applies the IS-LM approach for a closed economy and includes the relationship of an open economy with PPP and UIRP theories. The model’s structure is based on analyzing the balance between three markets, namely the goods market, money market and foreign exchange market. An IS curve indicates balance on the goods market as follows:

Y = C + I + G + NX (1)

where Y = gross domestic product (GDP), C = consumption that depends on the interest rate (r) GDP, I = investment that is also influenced by the interest rate (r) and GDP, G = government expenditure, NX = net exports that are influenced by international GDP (Y*) and the real exchange rate Q = E[P/S·P*] with S = nominal exchange rate and (P·P*) = domestic and international prices. Meanwhile, the LM curve indicates balance on the money market as follows:

M d / P = L (Y , r ) and M d = M S = M (2)

86    Central Bank Policy where Md = demand for money, Ms = supply of money, M = total money supply, and r = nominal interest rate. On the other hand, the foreign exchange (FE) curve indicates balance on the foreign exchange market, which illustrates balance on the BOP as follows: (a) Balance of payments6: BOP = CA + KA(3) (b) Balance of trade: CA = NX = PX – SP * Z (4) (c) Capital account: KA = K (r − r * − E ( ∆ S ))(5) where r* = international interest rate, and E[ΔS] = expected nominal exchange rate depreciation. Analysis per the Mundell–Fleming model focuses on fundamental factors of exchange rate determination, and the extent of monetary and fiscal policy effectiveness in terms of output stabilization (not price stabilization), which was the overarching concern of stabilization policy in the 1970s. The exchange rate and foreign exchange systems will influence the analysis results of the model. In this regard, the exchange rate is determined by various factors that influence balance on the domestic goods market and, pursuant to PPP theory, exchange rate adjustments will occur through changes to imports and exports in the trade balance. In more detail, Fig. 4.2 presents the Mundell–Fleming model for two specific cases. The first case looks at the influence of expansive monetary policy, assuming a flexible exchange rate regime and perfect capital

Fig. 4.2:  Policy Analysis in the Mundell–Fleming Model. Source: Gartner (1993, p. 10 and 14). 6

In practice, the balance of payments (BOP) consists of: (1) balance of trade (CA), (2) capital account (CP), and (3) changes in reserve assets (OR), thus BOP = CA + CP + OR. OR is omitted from this model.

Exchange Rates and the Economy    87 mobility and the second case refers to the influence of expansive fiscal policy, assuming a fixed exchange rate regime and system of foreign exchange control (imperfect capital mobility). The IS curve has a negative slope, contrasting the positive slopes of the LM and FE curves. The FE slope depends on crossborder capital mobility. In the first case, with perfect capital mobility and assumed exchange rate flexibility, the FE curve is horizontal. In this example, expansive monetary policy will effectively drive output. An increase in money supply would shift the LM curve to the right, thereby lowering domestic interest rates and, with foreign capital outflows, trigger exchange rate depreciation. The trade balance would increase, with rising exports and declining imports due to exchange rate depreciation, thus output would also increase. The IS curve would move to the right, hence the domestic and international interest rates would return to parity (UIRP assumption). Meanwhile, fiscal stimuli would not influence output. Increased government spending would move the IS curve to the right and edge up domestic interest rates, thus prompting exchange rate appreciation due to foreign capital inflows. The trade balance would deteriorate and return the IS curve to its original position without affecting the level of output. In the second case, however, under a system of perfect capital control, the FE curve is vertical at a level of output that offsets the BOP. Under such ­conditions, expansive fiscal policy would effectively raise output.7 Increased government expenditure would push the IS curve to the right, causing higher interest rates and appreciatory pressures on the exchange rates. To maintain the desired exchange rate, the central bank would purchase foreign exchange using domestic currency, thus moving the LM curve to the right to achieve a new balance. In contrast, monetary policy would not effectively drive output under a fixed exchange rate regime. In the event of depreciatory pressures, for instance if money supply outstripped the requirement, the central bank would have to intervene by selling enough foreign exchange for the exchange rate to return to the predetermined level. Using the above analysis framework, another interesting feature of the Mundell–Fleming model is the presence of the policy trilemma in an open economy. Differing from the Tinbergen rule for a closed economy, namely that there should be at least the same number of instruments as there are targets, in an open economy, focusing one policy on one target must also take into consideration the exchange rate regime and foreign exchange system applied. Monetary policy would only be effective in terms of influencing output if the country in question adheres to a flexible exchange rate regime and perfect capital mobility. Oppositely, fiscal policy as an output stabilization policy instrument would be very effective under a fixed exchange rate regime and perfect capital control.

7

Fisher (1994) stated that under a fixed exchange rate regime, only fiscal policy is effective, thus limiting central bank independence in terms of implementing monetary policy to achieve the domestic economic goals, such as inflation and economic growth.

88    Central Bank Policy 4.2.2. Exchange Rate Model with Sticky Prices: Dornbusch’s Overshooting Model Increasing exchange rate volatility as globalization accelerated along with international financial markets at the beginning of the 1980s attracted the attention of economists and policymakers alike. Consequently, the analysis of monetary models shifted from output stabilization policy to the phenomenon of exchange rate volatility and its implications on domestic price stability. The overshooting model proposed by Dornbusch (1976) explained that short-term exchange rate volatility was caused more by price rigidity (sticky prices) on the domestic goods market, while the exchange rate quickly reacts to economic shocks on the foreign exchange market. In the long term, however, prevailing monetary opinion is that PPP theory is valid and that inflation represents a monetary phenomenon. Dornbusch’s overshooting model differs from the Mundell–Fleming model by assuming sticky prices that cause a trade-off between inflation and economic growth in the near term since the inception of the Philips curve. In this regard, price corrections to non-traded goods occur with a lag to a new equilibrium level in response to a shock in the economy, while the prices of trade goods rise proportionally with changes to money supply. Therefore, prices increase less than money supply, causing currency supply to outstrip demand. Under such conditions, economic shocks would trigger exchange rate overshooting in the near term, implying greater exchange rate volatility than would fundamentally be expected, before returning to a new balance in the long term. A simple version of Dornbusch’s overshooting model can be found in Gartner (1993) in the log-linear equation, which mathematically illustrates balance on the goods market, money market, and foreign exchange market. Balance on the goods market is indicated by the Philips curve and aggregate demand curve as follows: Philips curve: π = β ( y d − y )(6) Aggregate demand: y d = δ ( s − p ) + σ y + g(7) Meanwhile, balance on the money market is indicated by demand for currency and equilibrium as follows: Currency demand: m d = p + ϕ y + λr (8) Money market equilibrium: m d = m s = m(9) On the other hand, equilibrium on the foreign exchange market is indicated through uncovered interest parity (UIP) with the formation of exchange rate expectations as follows: Foreign exchange market: r = r * + E [ ∆ s ](10) Adaptive expectations formation: E [ ∆ s ] = θ( sˆ − s )(11)

Exchange Rates and the Economy    89 where π = inflation, ρ = price, yd = aggregate demand, y = output (exogenous), s = short-term exchange rate, s = long-term exchange rate, Δs = exchange rate depreciation, r, r* = domestic and international interest rates, md = demand for currency, ms = supply of currency, g = government spending, and E[·] = operator expectations. The exchange rate overshooting model can be analyzed by expressing the model in two equations, namely an IS equation as goods market equilibrium and an MS equation that represents money market and foreign exchange market equilibrium. Therefore, because yd = y and π = o, the IS curve is as follows:

p = s + g / δ − (1− σ ) y / δ (12)

with a positive 45° slope, which demonstrates price and exchange rate homogeneity as contained in PPP theory. Meanwhile, the MS curve is as follows:

p = m − ϕ y + λr * + λθ( sˆ − s )(13)

with a negative 45° slope that is consistent with the quantity theory of money (QTM). Simply analyzing the influence of monetary policy in this model is as follows: In the long term, s = s and p = p′, therefore price equilibrium is:

p′ = m − ϕ y + λr * (14) with a (nominal) exchange rate equilibrium at:



sˆ = p′ − g / δ + (1− σ ) y / δ = m + λr * − g / δ + (1− σ − ϕδ ) y / δ (15) and (real) exchange rate equilibrium level of:



( sˆ − p′ ) = −g / δ + (1− σ ) y / δ(16)

Therefore, the long-term influence of monetary policy is as predicted by PPP theory and the QTM, namely that an increase of money supply only impacts (homogenously) the exchange rate (∂ sˆ / ∂ m = 1) and price level (∂p′/∂m = 1), while the real exchange rate is not affected (∂ sˆ − p′) / ∂ m = 0). In the near term, however, the behavior of the (nominal) exchange rate is shown by the following equation:

s = sˆ + ( m − p − ϕ y ) / λθ + r * / θ (17)

Therefore, an increase of money supply would cause the exchange rate to overshoot in the short term, namely excessive fluctuations beyond the long-term balance: [∂s/∂m]ST = ∂ŝ/∂m + 1/λθ = (1 + 1/λθ) > 1 = [∂ŝ/∂m]LT. The magnitude of

90    Central Bank Policy

Fig. 4.3:  Policy Analysis According to Dornbusch’s Overshooting Model. short-term exchange rate misalignment is more attributable to sticky prices in terms of adjusting to economic shocks as follows:

s − sˆ = [( m − ϕ y − λr * ) − p ] / λθ = [ p′ − p ] / λθ(18)

In other words, the preliminary results of Dornbusch’s overshooting model confirmed the impact of price rigidity on heightened short-term exchange rate volatility before returning to a new equilibrium in the long term. These movements are presented graphically in Fig. 4.3 as follows: The monetary policy trilemma in an open economy can also be analyzed using Dornbusch’s overshooting model. When exchange rate stability close to the longterm equilibrium is desired, monetary policy will have less of an influence on exchange rates if cross-border capital flows are restricted though capital controls. In this case, the goal of cross-border capital mobility, such as through the application of perfect capital mobility, will not be achieved and the impact on prices will also be more pronounced (lack of monetary policy autonomy). On the other hand, when price stability close to the long-term equilibrium is desired, monetary policy autonomy to minimize price fluctuations could be achieved if capital mobility and a flexible exchange rate regime are applied. In such as case, exchange rate fluctuations would be of greater magnitude to absorb (shock absorber) the impact of cross-border capital mobility on domestic money supply.

4.2.3. Portfolio Balance Model and Central Bank Monetary Operations Kouri (1976) was the first to expand the PBA in terms of exchange rate determination and BOP analysis. The model analyzed the exchange rate based on public

Exchange Rates and the Economy    91 behavior in terms of financial asset portfolio diversification to the domestic currency (M), domestic bonds (B), and foreign bonds (F) taking into consideration the returns (price and interest rate) of each respective asset type. Trade balance conditions will influence the impact of the exchange rate on other economic variables. Therefore, the influence of monetary policy will be determined by how the central bank conducts monetary operations, whether through the purchase of domestic bonds or intervention on the foreign exchange market. In the near term, the total of domestic bonds is assumed to be fixed and held by the public and central bank, thus B = BP + BA. Similarly, foreign bonds are also held by the public and central bank, hence F = FP + FA, but the total will fluctuate in line with the current account surplus or deficit as follows:

CA = ∂ F / ∂t = NX (S / P,Y ) + r * ( FP + FA ) (19)

Primary funds represent the total of domestic and foreign bonds held by the central bank, thus M = BA + SFA. Therefore, economic worth can be expressed as follows:

W = M + BP + SFP = B + SF (20)

In the model, balance on the money market, which is usually expressed in the following form: M  =  m(r, E(ΔS), Y, W), can be expressed in the relationship between the exchange rate and interest rate with the ME curve as follows:

S ME = m( r, M ,Y ,W )(21)

with a positive slope: ∂S / ∂ r = −( ∂ m / ∂ r ) / ( ∂ m / ∂W ) FP > 0. Implying that an increase in the interest rate on the money market would reduce demand for currency and trigger exchange rate depreciation. Balance on the domestic bond market can be expressed in the following form: BP = b( r, E ( ∆ S ),Y ,W ) , which can also be expressed in the relationship between the exchange rate and interest rate with the BE curve as follows:

S BE = b( r, BP ,Y ,W )(22)

with a negative slope: ∂S / ∂ r = −( ∂b / ∂ r ) / ( ∂b / ∂W ) FP < 0. Implying that an increase in the interest rate on the domestic bond market would reduce central bank demand for domestic bonds and trigger exchange rate appreciation. Similarly, balance on the foreign bond market can be expressed in the following form: SFP = f ( r, E ( ∆ S ),Y ,W ), which can also be expressed in the relationship between the exchange rate and interest rate with the FE curve as follows:

S FE = f ( r, FP ,Y ,W )(23)

92    Central Bank Policy with a negative slope: ∂S / ∂ r = (∂ f / ∂ r ) / (1−∂ f / ∂W ) FP < 0. Implying that an increase in the interest rate on the foreign bond market would reduce demand for foreign bonds and trigger exchange rate appreciation. Short-term analysis of monetary policy’s impact is determined by the monetary operations implemented by the central bank. Three cases serve as examples. First, open market operations (OMO), namely when the central bank increases money supply through the purchase of domestic bonds held by the public, thus: ΔM  =  −ΔBP  =  ΔBA. Second, non-sterilized foreign exchange intervention, namely when the central bank increases money supply through the purchase of foreign bonds, thus: ΔM = −SΔFP = SΔFA. Third, sterilized foreign exchange intervention, namely when a central bank purchases foreign bonds held by the public, while simultaneously selling domestic bonds, thus the amount of money supply remains unchanged and: −SΔFP = ΔBA and ΔM = 0. Fig. 4.4 presents an impact analysis of the three types of monetary operations available to the central bank in terms of the exchange rate and interest rate. In the first case, OMO would increase money supply, thus moving the ME curve to the left, while the purchase of domestic bonds would move the BE curve downwards, which would lower interest rates and cause exchange rate depreciation. The FE curve would not move because OMO merely create an exchange between currency and domestic bonds. In the second case, central bank monetary operations through non-sterilized foreign exchange intervention would create excess currency supply, thus moving the ME curve to the left, while the purchase of foreign bonds would move the FE curve to the right, which would lower interest rates and lead to exchange rate depreciation. Meanwhile, sterilized foreign exchange intervention is a combination of OMO and non-sterilized foreign exchange intervention, which would not affect money supply and, therefore, the ME curve would not move. The purchase of foreign bonds would move the FE curve to the right, while the sale of domestic bonds would also move the BE curve to the right. Consequently, interest rates would increase and the exchange rate would depreciate. Exchange rate depreciation

Fig. 4.4:  Analysis of Monetary Operations in the Portfolio Balance Model.

Exchange Rates and the Economy    93 would occur because excessive demand for foreign bonds would depreciate the exchange rate back to equilibrium. The above analysis of monetary operations could be considered when formulating monetary policy to achieve the short-term economic targets required. For example, if monetary policy is required to overcome a trade deficit and stimulate economic growth, monetary operations through OMO and non-sterilized foreign exchange intervention would be the first choice because both trigger exchange rate depreciation and lower interest rates, while domestic inflation remains under control. Meanwhile, monetary operations through sterilized foreign exchange intervention would be the optimal choice to improve the trade balance and control inflation due to the resultant exchange rate depreciation and higher interest rates, while also considering the impact on growth.

4.2.4. Market Microstructure Models: Order Flows and Heterogeneous Information The models presented above explain exchange rate theory and the impact on the economy from the perspective of macroeconomic fundamentals. The question, therefore, is how far the theoretical models can explain exchange rate fluctuations on the market? A review of the literature concluded that the predictive power of the models is no better than the random walk estimation of exchange rate fluctuations on the market (Meese & Rogoff, 1983). In other words, the theoretical models can better explain exchange rate fluctuations in the long term but not in the short term. There is some confusion when explaining short-term exchange rate fluctuations on the market, a phenomenon known as the exchange rate puzzle. The exchange rate theories developed thereafter were based more on a market perspective, known as market microstructure analysis, namely to study how trade and foreign exchange market transactions determine exchange rate fluctuations on the market in the short term (O’Hara, 1995). The foreign exchange market microstructure consists of two segments.8 The first segment is between the customers and dealers, where transactions are quote driven, while the second segment is between dealers, where the bid-ask mechanism is used to achieve transaction agreement. In general, the dealer is from the treasury division of a commercial bank and conducts foreign exchange transactions to serve the customer directly or as required by the bank itself. Meanwhile, the customers can be divided into two groups, namely financial customers in the form of a financial manager, another bank, the central bank or multilateral financial institutions, as well as corporate customers that buy or sell foreign exchange in line with their economic activities, such as exports-imports, external debt, or dividend proceeds. The transactions are based on an exchange rate quotation by a certain dealer, through bidask spreads, where the consumer or another dealer/broker can place an order of a specific amount. Order flows and transactions on the foreign exchange market

8

Refer to Osler (2008), for instance, for further explanation about foreign exchange market microstructure.

94    Central Bank Policy can occur at any time and continue indefinitely, thus influencing exchange rate fluctuations on the market. The magnitude of impact of market microstructure on exchange rate fluctuations is in line with the results of direct surveys of foreign exchange market players (Cheung & Chinn, 2001; Taylor & Allen, 1992). The surveys showed that market players hold three beliefs that are completely different from the assumptions used in the standard exchange rate models. First, dealers believe that the exchange rate moves in response to trading flows (orders and transactions) on the market. This implies that order and transaction flows developing on the market will influence the behavior of dealers and brokers when setting their daily trade strategy and exchange rates. Such conditions are vastly different from the standard model that focuses on FX holdings, not trading flows, and assumes that PPP and UIRP effectively explain exchange rate fluctuations on the market. Second, the surveys also showed that dealers consider private information an important aspect of the foreign exchange market. For example, Cheung and Chinn (2001) reported that dealers viewed large banks as having excess information because of the number of customers and extent of the networks in place. Furthermore, many big banks conduct macroeconomic and financial market research and analysis when setting their portfolio strategy and determining foreign exchange market transactions. Such conditions are very far removed from the assumptions of the standard model that all information is in the public domain and, therefore, accessible to all market players. Third, market players consider that trading flows are a medium to transmit the impact of private information to exchange rate fluctuations. This implies that without trading flows, private information would not influence exchange rates on the market. With transactions available at any time, each order is based on the latest private information and then determines the exchange rate transacted. Thus, order process, private information and the exchange rate transacted move quickly and continuously. Such conditions are very different from the standard market efficiency model, which assumes that news causes immediate exchange rate fluctuations without considering whether there are transactions on the foreign exchange market that could influence exchange rates. The influence of order flows on exchange rate fluctuations has been the object of several theoretical and empirical studies (Bionne & Rime, 2005; Evans & Lyons, 2005; Lyons, 1995; Payne, 2003). Based on UIRP theory, the model is as follows:

∆ St +1 = βi ( rt − rt* ) + βzOt (24)

where ΔS  =  exchange rate fluctuations (appreciation or depreciation), also known as the exchange rate return, (r − r*) = disparity between domestic and foreign exchange rates as a fundamental indicator, and O = order flows indicator as a measure of the number of purchases minus the number of sales of foreign exchange on the market. In this context, the time interval (t, t  +  1) does not denote calendar days but transactions on the market.

Exchange Rates and the Economy    95 The previous equation shows the extent of imbalances on the foreign exchange market per the order flows indicator, which can explain whether excessive exchange rate fluctuations on the market are attributable to fundamental factors, as measured by differences in the interest rate indicator. A study by Evans and Lyons (2002) found that the equation could explain around 70% of exchange rate fluctuations on the market. Furthermore, the empirical results demonstrated that an aggressive increase in purchases between dealers of US dollar (USD) 1 billion would cause 0.5% appreciation in the USD against the Deutsche eMark (DEM). Similarly, Froot and Ramadorai (2002) confirmed those results, adding that future short-term interest rate fluctuations also influence order flows and exchange rate fluctuations. The promising empirical findings for order flows prompted further development of several exchange rate models based on market microstructure (Bacchetta & van Wincoop, 2003; Breedon & Vitale, 2004; Evans & Lyons, 2005; Froot & Ramadorai, 2005). In addition to order flows, several models were developed with the inclusion of behavioral factors and information heterogeneity regarding the market players. Such models differed greatly from the previous exchange rate theories, which assumed that all market players have access to identical information, understand the models well and utilize all information in the decision-making process. In reality, although foreign exchange market players constantly monitor the news concerning economic fundamentals, their understanding or perception is heterogeneous. Different behaviors and reactions to information also have an influence, for instance carry traders generally base their decisions on differences between interest rates and currency risks. Similarly, despite involvement in market transactions, market players may not have access to complete information regarding order flows on the market. Such factors precipitated model development to include interconnectedness between indicators of economic fundamentals, the magnitude of order flows as well as the behavior of market players and information heterogeneity to explain short-term exchange rate fluctuations on the market. Evans and Lyons (2005), for example, began their analysis by referring to the determination of exchange rate fundamentals in the monetary model for two countries as follows:

∞

st = (1− λ )∑ λ j Et ft + j (25) j =0

where ft = ( mt − mt* ) − ϕ ( yt − yt* ) + qt − ( rt − rt* + αρt ) , namely the exchange rate is determined by disparity in terms of money supply, economic growth and interest rates, while paying due consideration to the risk premium. The equation can be used to predict the influence of fundamental factors on exchange rate fluctuations on the market as follows:

∆ st +1 =

1− λ ( st − Et ft ) + εt +1 λ

96    Central Bank Policy where

∞

εt +1 = (1− λ )∑ λ j ( Et +1 − Et ) ft + j +1(26) j =0

In this case, if st = Etft, then exchange rate fluctuations will follow the random walk. Alternatively, if fundamental changes follow AR(1):

∆ ft = φ∆ ft−1 + ut (27)

then exchange rate deviation from the fundamental value will also follow AR(1). Differing from the macromonetary model above, the market microstructure model determines market player expectations using fundamental information available at the beginning of the transaction period:

∞

st = (1− λ )∑ λ j Etm ft + j (28) j =0

therefore, exchange rate fluctuations on the market can be represented by the following equation:

∆ st +1 =

1− λ ( st − Etm ft ) + εtm+1 λ

where

∞

εtm+1 = (1− λ )∑ λ j ( Etm+1 − Etm ) ft + j +1(29) j =0

In this case, exchange rate fluctuations will be influenced by two factors: (1) public information that creates exchange rate deviation from the fundamental value at the beginning of the transaction; and (2) private information, namely the perceptions of each respective market player concerning future information that will influence the exchange rate. The empirical findings relating to the influence of order flows on exchange rate fluctuations, as found previously, show that market players form private information from transaction flows occurring on the market, in addition to the public information available from the authorities, data, analysis, and market news. This can happen due to two factors. First, market players base their transaction behavior on the fundamental information available and, therefore, will be contained in the order flows on the market. Furthermore, there is a lag between the time when information becomes available from order flows and the time when market players process and understand that information. This is because market players only know some of the transactions when transacting during the first session, and only fully understand all order flows in the next transaction session.

Exchange Rates and the Economy    97 Ordering to such conditions, changes in fundamental factors that influence exchange rate fluctuations on the market not only limit existing public information but also the perceptions of market players in terms of new public information and the information gleaned from order flows. ∆ ft = φ∆ ft−1 + ut + δvt (30) Ot = γOt−1 + vt



Assuming news is public information, μt, that is known at the time of transaction and order flows information, vt, is only understood by market players with a time lag of one period, then Etm ft−1 = ft−1 and Etm ft = (1 + φ ) ft−1 − φ ft−2 + ut thus ( ft − Etm ft ) = δvt . Therefore, the exchange rate equation becomes as follows:

st = Etm ft +

λφ 1 λφ δ Etm ∆ ft = ft − ft−1 − vt(31) 1− λφ 1− λφ 1− λφ 1− λφ

Such that spot exchange rate innovation, εtm+1 ≡ ( st +1 − Etm st +1 ), is: 1 λφ δ vt +1 ( ft +1 − Etm ft +1 ) − ( ft − Etm ft ) − 1− λφ 1− λφ 1− λφ (32) 1 [1+ φ(1− λ )]δ = ut +1 + vt 1− λφ 1− λφ

εtm+1 =

Substituting the order flows equation produces the following equation:

∆ st +1 =

1− λ 1 [1+ φ(1− λ )]δ ( st − Etm ft ) + ut + (Ot − γOt−1 )(33) λ 1− λφ 1− λφ

This equation explains that there are three factors influencing exchange rate fluctuations on the market, namely: (a) the perception of market players to exchange rate misalignment from its fundamental value; (b) news appears from fundamental information; and (c) information is derived from order flows. In their empirical study, Evans and Lyons (2005) used the following two equations:

∆ st +1 = α0 + α1OtA + et +1 ∆ st +1 = α0 + ∑ j +1 α j OSj ,t + et +1 6

(34)

where the first equation was estimated for order flows as an aggregate of all customer segments, while the second equation was estimated for each respective customer segment. In general, the empirical results showed that the market microstructure model has a higher prediction power than the macromodel or random walk. Moreover, exchange rate fluctuations on the market are not only influenced by public information relating to economic fundamentals but also by the private information of market players concerning current news on economic fundamentals and from studying order flows on the market.

98    Central Bank Policy

4.3. Empirical Findings for Exchange Rates and the Economy The various theoretical reviews presented in previous sections provide a rationale to analyze the exchange rate and its impact on the economy. How far can these theories explain empirical economic conditions in various countries? Specifically, can the exchange rate improve the trade balance and, therefore, stimulate economic growth? Also, to what extent does the exchange rate influence inflation? This kind of empirical review represents an important input for the central bank to formulate monetary policy to achieve price stability and also stimulate economic growth.

4.3.1. Exchange Rates and Trade Theoretically, the Marshall–Lerner condition states that exchange rate depreciation will only cause a balance of trade improvement if export elasticity to the exchange rate is greater than import elasticity to the exchange rate. In general, such conditions are found in countries where the composition of manufactured commodities in exports is larger than in imports. Dynamic analysis of the elasticity concept in the Marshall–Lerner condition reveals a J-curve phenomenon concerning the exchange rate’s influence on the trade balance. In the near term, exchange rate depreciation would undermine trade balance performance due to the negligible decline of imports to meet domestic demand. As time goes by, however, commodities to substitute the imports would emerge in the country and, therefore, exchange rate depreciation would reduce imports. The trade balance would improve as exports increased and imports declined in response to the exchange rate depreciation. Numerous empirical studies have corroborated the positive impact of exchange rate depreciation on the trade balance. From a panel data study of around 100 countries for a period of 10 years (2000–2009), for instance, Nicita (2013) showed that exchange rate misalignment from the fundamental value (measured using the REER) triggers a 1% increase in world trade. Auboin and Rut (2011) found that the influence of the exchange rate on the trade balance was more pronounced in highly competitive countries, along with countries that are part of global/regional supply chains or countries experiencing domestic price rigidity, including those using foreign currencies in the production process. Meanwhile, the empirical evidence remains inconclusive about whether exchange rate volatility would adversely affect the trade balance (Clark, Laxton, & Rose, 2004). This might be due to weak predictive power of the theoretical analysis because, in general, exchange rate volatility is linked to economic shocks, such as technology, preferences, and exchange rate systems. Empirically, the ambiguous influence of exchange rate volatility can be associated with hedging and other risk mitigation practices undertaken in the corporate sector, the large amount of raw materials imports as well as the proliferation of affiliated corporations operating internationally. Concerning developing countries, Rodrik (2008) showed that undervaluation of the nominal exchange rate (or a high real exchange rate) could stimulate

Exchange Rates and the Economy    99 economic growth. From a study of seven countries in the period of 1950–2004, the empirical results showed that a 20% undervaluation of the real exchange rate could accelerate economic growth by around 0.4%. Exchange rate undervaluation was shown to be transmitted primarily through the tradable sector, especially the manufacturing industry, which is generally competitive and profitable, thereby avoiding the market failures and institutional default that can plague developing countries. How can real exchange rate undervaluation be achieved through various policies, such as fiscal policy (large structural surplus), income policy (redistributing income to large savers through the compression of real wages), saving policy (mandatory saving schemes and pension reforms), capital account management (tax foreign capital inflows, liberalize capital outflows), or exchange rate intervention (with adequate reserve assets)? Experience in East Asia showed that countries targeting real exchange rates to enhance competitiveness successfully increased exports, which must be supported by deregulation and other structural reforms to boost economic competitiveness. The positive influence of real exchange rates on economic growth was also empirically proven in ASEAN5 countries (Abudalu, Ahmed, Almasaied, & Elgazoli, 2014). Using quarterly data for the period of 1996–2006, the study showed the respective coefficients of real exchange rate elasticity to GDP growth in Thailand (0.11), Indonesia (0.14), Malaysia (0.16), the Philippines (0.49), and Singapore (0.88). Meanwhile, Onafowora (2003) empirically demonstrated the Marshall–Lerner condition with a J-curve in the near term regarding trade between ASEAN5 with Japan and the United States for the period from 1980 to 2001. For Indonesia, for example, exchange rate depreciation would reduce the real trade balance for a duration of around three quarters and then subsequently increase with an elasticity of 0.35 for Japan and 0.24 for the United States. Similarly, in Thailand’s case, the real trade balance would reduce for around four quarters before increasing with an elasticity of 1.08 for Japan and 1.65 for the United States. Nonetheless, Hadad and Pancaro (2010) advised against enhancing competitiveness through real exchange rates in the long term because such policy would raise inflation and undermine structural reforms to boost real sector competitiveness, which would be more important to foster long-term growth.

4.3.2. Exchange Rate and Inflation The influence of the exchange rate on inflation (exchange rate pass-through – ERPT) is an important aspect of monetary policy formulation. Thus far, observations have shown that developing countries experience a greater decline of ERPT than advanced countries. Frankel, Parsley, and Wei (2005) conducted an empirical study concerning this phenomenon and its determinants using a sample of 76 countries in the period of 1990–2001. Price indicators were differentiated between port prices, retail prices, and CPI. The results showed that countries with a history of high inflation experienced an ERPT reduction from around 0.79 during the period of 1990–1996 to just 0.20 in 1996–2001. In contrast, countries with historically low inflation recorded a decline from 0.50 to 0.12 over the same period. For all price indicators, ERPT was around 0.63 for port prices, 0.41 for

100    Central Bank Policy import prices and 0.22 for CPI. The proliferation of post-globalization international trade was cited as the main cause of lower transportation costs between countries. Meanwhile, exchange rate volatility was shown to have an asymmetrical impact on ERPT, meaning that depreciation exceeding 25% could raise ERPT but exchange rate appreciation may not necessarily reduce ERPT. The empirical study also stressed the role of monetary policy in terms of lowering long-term inflation as an important aspect of reducing ERPT. How monetary policy framework performance in terms of price stabilization (Inflation Targeting Framework – ITF) reduced ERPT was the focus of an empirical study by Edwards (2006), who revealed several interesting findings. First, countries applying ITF experienced a decline in the exchange rate influence on inflation, most significantly in Brazil, decreasing from 0.719 to 0.056, followed by Chile, Israel, and México. Meanwhile, the decline in South Korea was not significant due to the relatively low level of ERPT (0.020). Second, ITF application did not exacerbate exchange rate volatility (nominal or real). Nonetheless, the three countries that applied a floating exchange rate regime, namely Brazil, Chile, and Israel, experienced an uptick in exchange rate volatility. Third, ITF countries, especially those facing comparatively high inflation, included nominal exchange rate factors into their monetary policy. Estimating monetary policy rate determination using the Taylor rule revealed the significance of the exchange rate variables (depreciation and volatility), with coefficients varying from 0.79 for México and 0 for Chile. Regarding ASEAN5 countries, Corinthas (2007) conducted an empirical ERPT study bilaterally between two countries, linked to the possibility of currency unification using data for the period from 1968 to 2001. The results showed that ERPT to inflation was significant and relatively large (0.10) for Indonesia– Philippines, but not significant for Thailand–Singapore or significant but close to zero for Thailand–Malaysia and Malaysia–Singapore. In terms of ERPT to import prices, the study found significant ERPT for Indonesia–Malaysia and Indonesia–Singapore. The study concluded that currency unification was only feasible between Malaysia and Singapore. For Indonesia, currency unification was shown to benefit monetary policy effectiveness in terms of lowering inflation to the regional level.

4.3.3. Empirical Findings in Indonesia The influence of the exchange rate on the economy has also been the object of empirical studies in Indonesia. Winarno (2014), for instance, studied the Marshall–Lerner condition and J-curve effect in Indonesia using monthly data from 2008 to 2012. The results proved that the Marshall–Lerner condition was present in Indonesia, namely that each 1% depreciation of the rupiah exchange rate would raise the current account, total trade balance and non-oil and gas trade balance by 0.51%, 0.72%, and 0.77%, respectively, in the long term. Furthermore, the study also proved the J-curve phenomenon in Indonesia, namely that the exchange rate influence would reduce the current account, total trade balance, and non-oil and gas trade balance for around 3–4 months before increasing to

Exchange Rates and the Economy    101 its long-term trajectory. The J-curve effect was also demonstrated by Adiningsih, Siregar, and Hasanah (2013), using quarterly data from 1996 to 2011, for bilateral trade between Indonesia–China and Indonesia–Japan but not for Indonesia–United States. In another study, Affandi and Mochtar (2015) showed that the effect on the current account since 2000 has been more due to temporary real exchange rate shocks rather than permanent. The ERPT analysis has also been conducted by BI and through other empirical studies. Kuncoro (2015), using monthly data from 2003 to 2013, found that ERPT has tailed off since ITF application in Indonesia. The study reported that ERPT to CPI and import prices for each 10% of exchange rate depreciation was 6% and 3% prior to ITF implementation, dropping thereafter to around 3% and 1.5%. That level of ERPT exceeded those proposed in the empirical studies conducted by BI, namely around 0.7%–0.3% for each 10% of exchange rate depreciation. The empirical results point to the need to rupiah exchange rate stabilization policy as an integral part of monetary policy to achieve the inflation target set.

4.4. Exchange Rate System and Policy The discussion on exchange rate determination in the previous section highlighted an interesting problem for further review that the relationship between macroeconomic variables and monetary policy effectiveness in an open economy will depend on the exchange rate regime adopted. For example, under a flexible exchange rate system, Mundell–Fleming model analysis showed that monetary policy could very effectively influence output. Meanwhile, Dornbusch’s overshooting model showed that the impact of monetary expansion on exchange rate depreciation is larger in the near term before returning to equilibrium in the long term. Similarly, in the context of the monetary policy trilemma, exchange rate stability may be achieved under a fixed exchange rate system with capital controls, while price stability is more achievable under a flexible exchange rate regime with capital mobility.

4.4.1. Choice of Exchange Rate System: Theoretical Review Observations of the international monetary system after the collapse of Bretton Woods in 1942 showed that exchange rate systems vary from one country to another, from a fixed (pegged) exchange rate system at one end of the spectrum to a free-floating exchange rate regime on the other. As stated by Frankel (1999), “No single currency regime is right for all countries or at all times.” There are numerous considerations when determining the exchange rate system in accordance with prevailing economic and financial conditions in the country (Later, 1999). The factors include: (1) the relative size of the economy compared to other countries; (2) trade integration between countries or groups of countries; (3) flexibility of the economic structure; (4) production factor mobility, including labor and capital; (5) capacity to absorb or confront internal and external economic shocks; (6) economic homogeneity with other countries

102    Central Bank Policy or groups of countries; (7) availability of external sources of financing to overcome BOP shortfalls; (8) availability of monetary policy that collectively supports exchange rate system stability; and (9) exchange rate credibility to anchor or standardize expectations. In general, it can be said that for countries with a comparatively small economy and high factors listed from points 2 to 9, a pegged exchange rate regime should be considered. Such a system is even more important in countries with historically high inflation and low central bank credibility in terms of maintaining price stability. The application of a fixed exchange rate system under such economic conditions would help the country to lower inflation toward the country or group of countries used as the standard for exchange rate determination. Nonetheless, economic developments in the affected country would also mirror the other countries, with output fluctuations and higher unemployment, while monetary policy autonomy to achieve the domestic goals would be forfeited. Theoretically, the selection of the exchange rate system is generally based on three salient criteria (Ghosh et al., 2001), namely (1) insulation; (2) economic integration; and (3) monetary policy credibility. The first approach emphasizes the ability of the exchange rate system to insulate the domestic economy from external shocks. In this regard, Freidman (1953) suggested that a flexible exchange rate regime is generally better because it supports the role of automatic adjustments from the nominal exchange rate to imbalances in the BOP. Similarly, the exchange rate system can also be selected based on the sources of economic shocks, in which case a floating system is typically better to protect domestic output if shocks tend to originate from the real economy, while a pegged exchange rate regime would be better if the nominal side is more dominant. The second approach emphasizes exchange rate system selection based on the level of economic integration with another country or group of countries. In this context, optimum currency area theory (McKinnon, 1963; Mundell, 1961) states that economic integration between two countries facing a similar and closely correlated output shock could opt for an exchange rate system that applies a common peg. Under such conditions, the function of the nominal exchange rate as an adjustment instrument to shocks will be lost due to economic integration and application of the common peg. The adjustment function would have to be replaced by flexible prices and wages, production factor mobility and the availability of a fiscal transfer system contained in the economic integration. If the requirements are met, application of a common peg with a common currency would be optimal. In contrast, a flexible exchange rate regime would better suit countries with disparate economic characteristics. The third approach stresses monetary policy credibility to achieve the domestic economic goals. Analysis is focused on how far the applied exchange rate regime can nurture central bank credibility in terms of controlling inflation. This theoretical approach refers to the monetary policy credibility and optimal inflation target

Exchange Rates and the Economy    103 theory developed by Barro and Gordon (1983) for an open economy.9 Analysis emphasizes the ability of a fixed exchange rate regime as a pre-commitment to low inflation, thus limiting the propensity of the central bank to create surprise inflation to stimulate output. In this context, a fixed exchange rate system would be preferable if central bank credibility is low so that the costs incurred by surprise inflation would be higher than the cost of losing the exchange rate as an adjustment instrument in the economy in the form of more prominent output fluctuations and higher unemployment. Conversely, a flexible exchange rate system should be applied if central bank credibility is high in terms of controlling inflation, hence the exchange rate can function as an adjustment instrument to shocks in the economy. In practice, exchange rate system classification is a country is also a problem in and of itself. One approach, the de jure approach, classifies the exchange rate system based on the official statement of the central bank submitted annually to the IMF. Nevertheless, the official statement of the central bank cannot always be used as a basis to classify the exchange rate system in accordance with theoretical analysis of exchange rate fluctuations. The second approach, therefore, known as the de facto approach, classifies the exchange rate regime based on observations of nominal exchange rate behavior and several monetary indicators that are in line with the theoretical concept. One such approach is exchange rate system classification based on the magnitude of nominal exchange rate volatility, fluctuations in reserve assets, nominal exchange rates, and other monetary indicators. Another technique, known as natural classification, is based on survey results, the official statement of the central bank, interviews, and so on that are subsequently compiled into an exchange rate system index (Rogoff et al., 2003).

4.4.2. Exchange Rate Systems in Practice: Bipolarization and Fear of Floating Applying different approaches has implications on the results of assessments into the classification of exchange rate systems adopted around the world. Using a de jure approach, Ghosh et al. (2003), for instance, evaluated exchange rate system evolution in 165 IMF member countries for the period of 1970–1999. The exchange rate systems were classified into three main groups, namely pegged, intermediate, and floating, with varying policy practices contained within. The 9

In brief, the model of Barro and Gordon (1983) characterized an economy where nominal wages are determined prior to the inflation target and the central bank seeks to create surprise inflation to stimulate output. The model shows that such central bank behavior would not be optimal to achieve the inflation target and would therefore undermine central bank credibility and output. The optimal solution for the inflation target is achieved if the central bank remains committed to maintaining low inflation so that wage determination can be trusted in the decision-making process. Central bank credibility can therefore be achieved by: (1) setting an optimal inflation target, namely one that minimizes real shocks; (2) central bank commitment to the inflation target; and (3) achievement of the inflation target set.

104    Central Bank Policy first group consisted of the currency board system, or a hard peg to a certain currency or basket of currencies. The second group included floating systems with intervention pursuant to certain rules or masked (intermediate). The third group included free floating systems with little or no intervention. A number of salient observations can be delivered. First, fewer countries are applying pegged systems, dropping from 84.8% in 1970–1979 to just 46.6% in 1990–1999, while countries favoring more flexible exchange rate systems are increasing, with intermediate systems accounting for 26.4% and floating systems for 27.0% in 1990–1999. Second, among the pegged systems, there has been a shift away from a peg to a single currency toward a hard peg to a basket of currencies. Similarly, flexible exchange rate systems have trended toward intermediate systems with masked intervention or floating systems with little or no intervention. Differing from Ghosh et al. (2003), Rogoff et al. (2003) used de facto classifications to evaluate exchange rate systems in various countries from 1973 to 1999. The classification was differentiated into six groups, namely hard peg, other peg, limited flexibility, managed floating, free floating, and free-falling. The results revealed significant differences between the official stance and the actual system applied, with the results for only around half of the countries the same using the two approaches. The differences became even more stark when classifying the floating system, with only 20% of the sample countries actually applying a free-floating system, while 60% adopted hard-peg to managed floating systems, and the other 20% were in a state of free fall.10 On the other hand, all de jure hard peg exchange rate systems were indeed hard peg systems and around 60% of the countries applied limited flexibility and managed floating systems as per the official stance. The study also analyzed exchange rate system bipolarization, which is often found in the literature. The study showed that since the middle of the 1990s, numerous EMEs have been expected to shift to either end of the spectrum, either hard peg or floating. The growth of hard peg and free-floating exchange rate systems in the study supported the prevailing view of exchange rate system bipolarization. Nonetheless, the trend has not been accompanied by a reduction in the number of intermediate systems, a phenomenon known as “hollowing of the middle.” Meanwhile, several countries opted for free-floating systems (South Korea, Thailand, Indonesia, and South Africa) or pegged systems (Argentina and Malaysia), while many others preferred intermediate systems (Brazil, Peru, Poland, Russia, and Venezuela). In addition, the transition from a de facto pegged system in the 1990s tended to favor a soft peg (China, Egypt, and Jordan) rather than a hard peg. The de jure and de facto classifications raised an interesting question about whether countries that officially embraced a floating exchange rate system were actually applying a floating system. To that end, Calvo and Reinhart (2002) analyzed the behavior of exchange rates, reserve assets, monetary policy, interest rates, and goods prices between 39 countries using monthly data from January

10

The free-falling exchange rate regime is used to denote periods when an exchange rate experiences large-scale depreciation due to unfavorable macroeconomic dynamics, similar to what occurs prior to a crisis episode.

Exchange Rates and the Economy    105 1970 to November 1999, which illustrated around 154 different exchange rate systems. The results implied that countries officially adopting a floating system actually had a fear of floating. Relative to the countries more committed to floating systems – such as the United States, Australia, and Japan – exchange rate volatility observed in other countries was very low. In fact, nominal and real economic shocks in such countries were larger than those found in the United States and Japan. In other words, low exchange rate volatility was more attributable to the stabilization policies instituted by the relevant authorities. The analysis also looked at the magnitude of reserve assets, which should be smaller under a flexible exchange rate regime. Greater interest rate variability (real and nominal) showed that the authorities were not entirely reliant on foreign exchange market intervention to stabilize exchange rates, a problem also caused by a lack of central bank policy credibility. Monetary policy was also more variable, such as interest rate variability, indicating procyclical monetary policy to stabilize exchange rate fluctuations. The empirical evidence also showed that prices (in the domestic currency) fluctuated more, reflecting that authorities only allowed limited exchange rate adjustments to absorb real sector shocks. In fact, many countries showed no strong correlation between commodity prices and the exchange rate, congruous with the view that the exchange rate does not permit adjustments in response to terms of trade (ToT) shocks. Another factor that drove monetary authorities to stabilize exchange rates was the fact that comparatively large exchange rate depreciation could be linked to recessions in EME due to the rapid increase in exposure to external obligations accompanied by the depreciation. The adjustment process in the current account to exchange rate fluctuations was also shown to be more significant in EME as a result of the relatively large import content in their export industry. Access to the international credit market is also affected by exchange rate volatility. Similarly, exchange rate stabilization is also driven by the desire for monetary policy to control inflation. Another question raised in the classification study was whether economic performance in one country was affected by the choice of exchange rate regime adopted. This question was addressed by Rogoff et al. (2003) by investigating several aspects, especially inflation, economic growth, and developmental level (developing, emerging, or advanced) of the country involved. The study concluded that exchange rate flexibility would be more beneficial to countries with greater access to international capital markets and could also help build healthy financial systems. Furthermore, the economic improvements would be larger in countries that applied sound and consistent macroeconomic policy.

4.4.3. Exchange Rate Crises: Causes and Impacts As more countries have adopted more flexible exchange rate regimes, exchange rate volatility has increased along with vulnerability to exchange rate crises. Consequently, crises in several EME have attracted the attention of economists offering a number of theories regarding the causes and impacts of such crises. The first-generation crisis models (Flood & Garber, 1980; Krugman, 1979)

106    Central Bank Policy explained exchange rate crises from the perspective of differences between government behavior, assuming inconsistent policy, and private investor behavior, with access to adequate information and acting rationally. For example, under a fixed exchange rate regime (and assuming money supply is controlled), excessive fiscal expansion would deplete reserve assets and encourage investors to launch a speculative attack and, as foreign exchange reserves became more depleted, could trigger a crisis. On the other hand, under a flexible exchange rate system (and assuming no change in reserve assets), excessive fiscal expansion would increase money supply and lead to uncontrolled inflation. Such conditions, if allowed to persist, would exacerbate exchange rate depreciation and, as investor confidence declined, could trigger a crisis. Although simple, the model reminds us that rational investor behavior and confidence in economic fundamentals can change into a speculative attack on the exchange rate due to imprudent policy instituted by the government. The second-generation models tried to explain crises in terms of changes in market perception to government policy, which subsequently becomes a self-fulfilling crisis. The conditions that cause self-fulfilling crises depend on the specific conditions found in a particular country. Referring to the attack on the exchange rate mechanism (ERM) at its inception when currencies in Europe were unified, Flood and Garber (1984) as well as Obstfeld (1986), explained that crises can occur if the markets perceive (perhaps erroneously) the government to persevere with expansive monetary policy after a speculative attack on a fixed exchange rate system. Similarly, Calvo and Mendoza (1995) showed that the 1994 crisis in México was consistent with the idea that issuing short-term securities and anticipating a government bail-out of a weak banking system would be vulnerable to changes of market perception. Similar dynamics also explained the 1997 Asian crisis, which was triggered by market panic stoked by high levels of short-term private external debt, with the risk of a currency mismatch and maturity mismatch likely, thus undermining confidence in the government maintaining exchange rate stability. Both crisis models above emphasized government policy credibility that was not aligned with market perception. Nonetheless, several crisis models that have been developed since have tended to highlight private sector behavior on the international financial markets. The first approach maintained the rational behavior of investors, but stressed the reality that capital markets in EME cannot offer comprehensive financial products and that investors do not have access to the same information. Limited financial products on the market cause suboptimal investment portfolios due to inefficient price setting, inadequate liquidity, and a lack of diversity. Limited information on financial products, economic dynamics, and government policy can cause exposure to international investor perception. Such product and information distortion could spark a crisis if unexpected changes in conditions occurred and/or if investor confidence in the government’s ability to stabilize the financial markets was lost. The second approach emphasizes differences in the behavior of various investor types and attempts to interpret the data to illustrate such behavioral changes. This analysis is often linked to the types of capital flows in the BOP, namely direct capital investment, bank loans, and portfolio investment, which is then broken

Exchange Rates and the Economy    107 down as long or short term to observe the differences in investor behavior. The goal is to analyze how different types of investors react to changes in financial product yields and risks as well as domestic and international economic developments, and then sorts the investors into groups, including real money and noise traders. In other words, this approach focuses on the reactions of noise traders to securities that oftentimes spurs high volatility in short-term capital flows and tends to be speculative. Such investors and private capital flows more commonly spark off a crisis than other types of investors. Crisis analyses according to the models above are important in terms of understanding not only the contributing factors but also the policy alternatives required. In that context, it is important to note that previous crises in EME, particularly in terms of BOP crises and banking crises, had a deleterious effect on economic performance, while incurring huge costs for the government and public. The cost of restructuring the banking sector can exceed 20% of GDP and the loss of output after a crisis can reach around 14%. Deeper analysis of the contributing factors and crisis fallout such as this is crucial to formulate a policy response that can prevent and/or minimize the impact of similar crises. Referring to the crisis models above, in addition to acknowledging the importance of strengthening economic fundamentals and the financial system, the various policies required for crisis prevention include the significance of government policy consistency, information transparency to avoid self-fulfilling investor expectations, domestic financial market deepening or development, the importance of understanding the difference between short-term noise traders and long-term real money investors, external debt controls, and foreign exchange flow management to control foreign capital flow volatility. Furthermore, the development of systems and mechanisms to prevent and mitigate crises that can detect existing fragilities in the economy early as well as the availability of a financial system safety net are crucial in order to remain prepared and able to institute the policy measures required.

4.4.4. Exchange Rate Stabilization Policy: Foreign Exchange Intervention The discussions in previous sections showed that exchange rate fluctuations in the near term tend to create misalignment from the fundamental value, accompanied by higher volatility. Consequently, the authorities in many EME have adopted exchange rate stabilization measures for numerous reasons, a phenomenon known as fear of floating in countries that apply a floating exchange rate regime. In other words, foreign exchange market intervention is not something that can be eliminated from emerging economies. Specifically, foreign exchange market intervention can be used, as part of monetary policy, to stabilize market expectations, alleviate market shocks, and control undesired exchange rate fluctuations from short-term economic shocks. Intervention is also available in line with policy to unwind macroeconomic imbalances, such as the current account deficit. To that end, intervention can support efforts to maintain macroeconomic stability by minimizing excessive exchange rate fluctuations, when there is credible commitment to the macroadjustments required.

108    Central Bank Policy Theoretically, the impact of foreign exchange market intervention on exchange rate stabilization can be explained using the portfolio balance model or the impact of policy signals on the market (Sarno & Taylor, 2001). In accordance with the portfolio balance model, foreign exchange intervention will influence the investor composition between domestic and foreign financial asset portfolios. Therefore, the exchange rate will be affected because changes in the portfolio will influence the yield and value of financial asset portfolios denominated in the home currency held by foreign investors. Foreign exchange intervention can also signal the market regarding the central bank’s stance to the exchange rate fundamental value and the policy direction pursued. Such signals, coupled with order flows per the market microstructure above, will also influence exchange rate fluctuations on the market in the near term. Chutasripanich and Yetman (2015) studied foreign exchange intervention strategies and their effectiveness. In general, intervention can correct exchange rate misalignment from the fundamental value or help to lean against the wind. Intervention effectiveness was evaluated based on five criteria: exchange rate stabilization, fewer current account imbalances, reducing speculation, minimizing reserve assets, and curtailing intervention costs. This study showed that intervention that can reduce exchange rate volatility can also lower the risk of speculation. Nonetheless, uncertainty concerning the fundamental value will undermine the effectiveness of intervention. Leaning against the wind that is not based on an assessment of exchange rate fundamentals could lower exchange rate volatility but would also exacerbate exchange rate misalignment. Furthermore, the cost of foreign exchange intervention will be higher is exchange rate fluctuations are caused by inconsistent interest rate policy. Regarding which approach is most effective, whether transparent or not, will depend on the desired objectives. If exchange rate stability is the goal, leaning against the wind would be optimal. If the aim is to reduce misalignment, however, the most effective approach would be transparent targeted policy. Nonetheless, both approaches would reduce speculation. Furthermore, more aggressive foreign exchange intervention by the central bank would erode the reserve assets and, therefore, reduce the effectiveness of the intervention itself. Mohanty and Berger (2015) conducted a survey of foreign exchange intervention practices by central banks in Asia and the Pacific. Although the goals of monetary and financial system stability remained the primary consideration, a change has occurred since the GFC, namely to mitigate the economic risks. Most intervention aims to prevent currency speculation, while alleviating the impact on inflation and capital flow volatility. The survey also showed that the target of intervention is to curb exchange rate volatility rather than achieve a specific target. The basic strategy of intervention has not changed much, namely: monitoring information on the position of international investors, focusing on the most liquid market segments, and the proclivity for masked intervention to maximize the effect. Most central banks believe that intervention is effective in terms of achieving the desired exchange rate goals, although differences emerge concerning the magnitude and duration. In terms of which channel is effective, most central banks believe that intervention works best through the signaling channel. The success of intervention is also supported by macroprudential policy and capital controls.

Exchange Rates and the Economy    109 Does intervention, in practice, support exchange rate stabilization in the wake of the GFC? Blanchard, Adler, and Filho (2015) conducted an empirical study using a sample of 35 EME for the period from 1990 to 2013. The results showed that, consistent with the predictions of the portfolio balance channel, intervention in the form of buying foreign exchange created less exchange rate appreciation during influxes of foreign capital flows after the GFC. Furthermore, the disparity between interveners and floaters was found to be large, namely approaching 1% of quarterly GDP (0.25% of annual GDP) would result in 1.5% lower exchange rate appreciation. The positive outcome would persist for three to four quarters before fading. Similarly, gross capital inflows reacted the same or more notably to intervention by interveners rather than floaters, showing the effect of exchange rate appreciation on lower capital inflows. No evidence was presented, however, for differences in interest rate behavior between interveners and floaters, indicating that central banks did not use the interest rate to mitigate capital flows. Fratzscher, Gloede, Menkhoff, Sarno, and Stöhr (2015) presented further empirical evidence using daily data for intervention conducted in 33 countries from 1995 to 2011. Effectiveness was measured based on two motives for the intervention: to achieve the desired exchange rate direction (moving) or just to reduce exchange rate volatility (smoothing). The assessment distinguished between flexible and narrow band exchange rate system. The results showed that, for flexible exchange rate systems, the success rate was around 60% for moving and 80% for smoothing. Nonetheless, the success rate of moving could be improved to 80% if the intervention was large. Meanwhile, for narrow band exchange rate systems, the success rate was higher at 84% for moving. The effectiveness of intervention could be improved for both groups, however, if accompanied by oral intervention, especially during periods of excessive exchange rate pressures. In general, the empirical evidence showed that central banks in various countries have successfully improved the effectiveness of their foreign exchange intervention to influence exchange rate fluctuations on the market. Nevertheless, it is important to note that intervention is not itself an instrument because the effectiveness depends on the consistency of the exchange rate desired through general macroeconomic policy (Canales-Klijenko, 2003). The policy trilemma mentioned above provides a valuable lesson about how perfect capital mobility prevents exchange rate policy and monetary policy from being instituted independently. Intervention will be ineffective if exchange rate fluctuations are due to persistent macroimbalances. Large-scale and long-term intervention is clearly unsustainable and such conditions require a change of exchange rate policy or other macroeconomic policies to improve the external balance. Similarly, a financial sector crisis, particularly a banking sector crisis, or capital flight would necessitate adjustments to several policy aspects, including the exchange rate, banking, monetary, and even fiscal policy to recover the external balance.

4.4.5. Exchange Rate System and Policy in Indonesia As mentioned previously, the most appropriate exchange rate system is not only determined by the economic and financial dynamics specific to a particular

110    Central Bank Policy country, but also the implications on monetary policy and foreign exchange flow management (foreign capital flows). Consistent with the monetary policy trilemma concept of an open economy, if a country applies a fixed exchange rate system, foreign capital inflows/outflows will occur that influence monetary policy effectiveness, while the reverse is true for a floating exchange rate system. If capital controls are applied, however, capital mobility to and from the country will be less and, thus, support more effective monetary policy, and vice versa for a free-floating exchange rate system. The monetary policy trilemma also implies that monetary policy effectiveness depends on the exchange rate and foreign exchange systems in place. Exchange rate policy in Indonesia, in addition to supporting sustainable economic development, also supports monetary policy effectiveness. In the history of the Indonesian economy, a fixed exchange rate system, managed exchange rate system and floating exchange rate system have all been used. The fixed exchange rate system was applied from 1973 to March 1983.11 A tightly managed exchange rate system was applied from March 1983 to September 1986. During that period, the Government introduced rupiah devaluation policy as follows: (1) devaluation in November 1978 from Rp425 per USD to Rp625 per USD; (2) devaluation in March 1983 from Rp625 per USD to Rp825 per USD; and (3) devaluation in September 1986 from Rp1,134 per USD to Rp1,644 per USD. Thereafter, a more flexible managed floating exchange rate system was implemented in Indonesia from September 1986 to January 1994 and with intervention bands from January 1994 to August 1997. During this period, the following exchange rate policies were introduced in the country: (1) BI released the exchange rates each day at midday. (2) The intervention bands were expanded eight times, namely from Rp6 (0.25%) to Rp10 (0.5%) in September 1992, to Rp20 (1%) in January 1994, to Rp30 (1.5%) in September 1994, to Rp44 (2%) in May 1995, to Rp66 (3%) in December 1995, to Rp118 (5%) in June 1996, to Rp192 (8%) in September 1996 and to Rp304 (12%) in July 1997. (3) BI intervened on the foreign exchange market to ensure the rupiah exchange rate remained within the intervention bands set, through the purchase of USD when the exchange rate approached the floor of the intervention band and through the sale of USD when the exchange rate approached the ceiling of the intervention band.

11

Prior to 1973, Indonesia applied a multiple exchange rate system through the Proof of Export System from 1957. The associated regulations contained trade and foreign exchange flow restrictions. Then, in 1967, the Export Bonus System was introduced that aimed to stimulate exports by permitting the public to freely transfer or trade export proceeds on the market at constantly changing prices (floating exchange rate system).

Exchange Rates and the Economy    111 A floating exchange rate system was introduced in Indonesia on August 14, 1997, which remains in place to the present day. The new exchange rate system was introduced in response to government policy to confront speculative attacks in the foreign exchange market from July to August 1997. Demand for foreign exchange was high to meet international obligations, due to large private sector external debt in Indonesia, and for speculative attacks from certain domestic and foreign parties seeking to exploit exchange rate shocks for personal gain. Consequently, BI could no longer keep pace with such high demand for foreign exchange, after eroding the reserve assets held, to sustain the managed exchange rate system. If the managed exchange rate system was maintained, the already lower position of reserve assets stoked concerns that foreign exchange reserves could dry up. Several neighboring countries, including South Korea and Thailand, also introduced a floating exchange rate system. By applying a floating exchange rate system, the rupiah began to move in line with market supply and demand and was relatively more stable (less sharp fluctuations) in accordance with economic fundamentals. Nonetheless, the exchange rate was often influenced by volatility non-economic domestic factors, particularly during the early part of the recovery after the devastating Asian financial crisis of 1997/1998. Attempting to stabilize the rupiah, BI intervened at specific moments on the foreign exchange market, namely when faced with excessive currency shocks and/or when exchange rate fluctuations were expected to pass through to inflation. It was not the intention of BI, however, to use foreign exchange market intervention to achieve a specific exchange rate target on the market. The exchange rate regime in Indonesia is regulated by Act No. 24 of 1999 concerning foreign exchange flows and the exchange rate system. Pursuant to that law, the exchange rate system in Indonesia is determined based on the recommendations put forward by BI, considering that any change to the regime would entail considerable ramifications, not only in terms of monetary policy and the financial sector, but also real economic activities, such as investment and international trade. Consequently, any change to the exchange rate system must be based on established ways of thinking and backed up by robust research that pays due consideration to the economic, political, and social aspects. In this context, BI is required to submit its recommendations concerning any planned changes to the existing exchange rate regime, due to its experience and knowledge in this area as well as the influence on monetary, banking, and payment system policy. BI’s implementation of exchange rate policy is not only stipulated in Act No. 24 of 1999, but also in more detail through the BI Act No. 23 of 1999, which has been amended several times, most recently by Act No. 6 of 2009. According to Article 7 of the BI Act, the mandate of overarching purpose of BI is unequivocally specified, namely to maintain rupiah (price) stability, which also contains provisions for exchange rate policy. Therefore, pursuant to Article 12, BI is required to implement exchange rate policy in line with the exchange rate system adopted as follows:

112    Central Bank Policy (a) under a fixed exchange rate system through devaluation or revaluation of foreign currencies; (b) under a floating exchange rate system through market intervention; and (c) under a managed floating exchange rate system through daily exchange rate determination and adjusting the intervention bands. When maintaining rupiah stability, Bank Indonesia sets exchange rate policy based on the floating exchange rate system by implementing various strategies to manage the exchange rate, including, but not limited to, OMO. The elucidation of Article 10, paragraph (1), letter b, point 1 of the BI Act also states that OMO include foreign exchange market intervention by Bank Indonesia to stabilize the rupiah.

4.5. Concluding Remarks This chapter has discussed, in depth, monetary policy and theory in an open economy, particularly through the exchange rate and its impact on the economy. First, the theoretical and empirical studies demonstrate that exchange rate fluctuations are influenced by fundamental factors, such as money supply, the interest rate, BOP and economic growth, as well as foreign exchange market microstructure dynamics, specifically order flows and information heterogeneity. Although long-term exchange rate dynamics can be explained by fundamental factors, short-term developments are extremely volatile and difficult to predict. Therefore, the exchange rate not only influences inflation and growth, but the volatility can trigger monetary and financial system instability. Second, the exchange rate correlates positively with the trade balance and, therefore, economic growth. Theoretically, this occurs if the Marshall–Lerner condition is met, such as if export elasticity to the real exchange rate exceeds the corresponding import elasticity. Empirically, however, that positive exchange rate effect can be observed to increase in countries where the export structure is dominated by competitive and hi-tech commodities, as part of the international trade chain, and the domestic industry structure can reduce dependence on imports. The use of exchange rates to boost competitiveness and, thus, improve the trade balance and stimulate growth is often a polemic in international economic politics, which has manifested in the ongoing debate on currency wars. Nonetheless, solid competitiveness to support trade and growth is required, through improvements to the investment climate, infrastructure and structural reforms in the real sector. Third, exchange rate dynamics directly and indirectly impact price stability and, therefore, domestic economic growth. The empirical findings in many countries have evidenced a dwindling exchange rate effect on inflation as a positive consequence of post-globalization international trade as well as monetary policy credibility. A more significant decline has been found in countries that apply an ITF but the central bank must still pay due consideration to the influence of the exchange rate on inflation, especially the asymmetric nature of the impact, meaning that the influence is stronger during periods of high depreciation.

Exchange Rates and the Economy    113 Consequently, the central bank must consider the impact of the exchange rate when formulating monetary policy to achieve low inflation, while also considering economic growth. Fourth, exchange rate volatility in many countries increased in the wake of the GFC, with capital flow volatility the inauspicious outcome of expansive monetary policies to stimulate economic recovery in advanced countries. In fact, more volatile exchange rates and capital flows have complicated policy response formulation in EME to maintain domestic monetary and financial system stability. The policy of foreign exchange market intervention to stabilize exchange rates has been widely adopted in EME. Although the empirical data points to a relatively high level of effectiveness, exchange rate volatility is still a risk that demands the attention of central banks. Intervention also becomes more effective if supported by policy to manage capital flows and macroprudential policy to maintain financial system stability. Understanding the role of the exchange rate and monetary policy, as explained in this chapter, is crucial for the discussions in subsequent chapters. Monetary policy transmission effectiveness and the mechanisms to influence the economy through the exchange rate channel, money supply, the interest rate channel and bank credit channel are discussed in Chapter 5. Thereafter, the strategic and operational framework of monetary policy set by central banks is discussed in Chapters 6 and 7. The two chapters underlie the discussion in Chapters 8 and 9 on the ITF, as applied by many central banks, Indonesia included. Finally, foreign intervention supported by capital flow controls to address the policy trilemma in an open economy, as well as the linkages with monetary policy in the central bank policy mix, is discussed in Chapters 14 and 15.

This page intentionally left blank

Chapter 5

Monetary Policy Transmission Mechanism 5.1. Introduction The Monetary Policy Transmission Mechanism (MPTM) is always an important and interesting topic in terms of the theory of monetary economics and its practice at central banks. Like a roadmap, MPTM illustrates the process of how the monetary policy implemented by the central bank influences various economic and financial activities to achieve the desired target, namely price stability and economic growth (Taylor, 1995; Warjiyo, 2004).1 For the central bank, understanding MPTM is crucial when determining the monetary policy stance, the instruments as well as the timing and magnitude of the measured and apposite response. For the financial community, corporate sector, and public, understanding MPTM is required to anticipate and consider the effect of monetary policy on their economic decisions. For academia, MPTM issue poses a challenge to the underlying theoretical thinking and empirical studies. MPTM represents a relatively complex process involving interactions between the central bank, financial sector and economic players as well as the policies instituted by the government and other relevant domestic and international authorities. During the two decades leading up to the GFC, for instance, monetary policy credibility in various jurisdictions contributed to lower inflation and stimulated economic growth (Berg et al., 2013; Taylor, 2014). The era of Great Moderation in the US, for example, was characterized by low inflation and robust economic growth for nearly two decades. Nevertheless, the US economic boom also facilitated product innovation in the financial sector, which became increasingly complex and disparate from the traditional intermediation function through deposits and credit, with a proliferation of derivatives that contained risks and for 1

A detailed explanation of the monetary policy transmission mechanism and the empirical findings in Indonesia is presented in Perry Warjiyo, The Monetary Policy Transmission Mechanism in Indonesia, Central Banking Series No. 11, Centre for Education and Central Banking Studies, May 2004. Those unfamiliar with the topic are recommended to read this book before proceeding. Central Bank Policy: Theory and Practice, 115–158 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191008

116    Central Bank Policy which the transmission mechanisms were difficult to comprehend. Global financial integration has also accelerated and, hence, tightened the linkages in terms of financial transmission and international capital flows. The ongoing developments demand greater attention, particularly in understanding MPTM, to ensure greater monetary policy effectiveness. This chapter explores MPTM, focusing on the periods before and after the GFC. In many ways, this chapter updates the description of MPTM in Indonesia presented by Warjiyo (2004) and its linkages with monetary policy (Warjiyo & Solikin, 2003) for the periods before and after the Asian Financial Crisis in 1997/98. The discussion is divided into six main sections. After the introduction, Section 5.2 presents the MPTM map as well as its relevance to monetary policy, interest rate policy, and monetary operations as the starting point of MPTM. Section 5.3 discusses the transmission channels based on the money view, while the credit view is examined in Section 5.4, including the influence of imperfections or financial frictions with the emergence of the risk-taking channel. Section 5.5 outlines the empirical studies of MPTM in several countries, including advanced countries, developing countries, and Indonesia, with an emphasis on the various theoretical factors discussed in the previous two sections, together with developing the economic and financial dynamics. Section 5.6 concludes the chapter with some important lessons from MPTM theory and empirical studies, while the implications for monetary policy at the central bank are presented in the subsequent chapters.

5.2. Monetary Policy and the Transmission Mechanisms MPTM mapping strives to provide theoretical and empirical answers to two salient questions asked by Bernanke and Blinder (1992), namely: (1) how can monetary policy influence the real economy, in addition to the impact on prices; and (2) through which mechanisms is monetary policy transmitted to the economy? Both questions represent an important issue in the monetary policymaking at the central bank as well as in the theoretical discussions on monetary economics by economists.

5.2.1. MPTM Map Initially, MPTM theory referred to the role of money in the economy, as explained by the quantity theory of money (Fisher, 1911). After further development, consistent with the advancement of non-bank financial sectors combined with increasing global financial integration, six monetary policy transmission channels emerged in the literature (Cecchetti, 1995; De Bondt, 2000; Kakes, 2000; Mishkin, 1996). The seven monetary policy transmission channels are the direct monetary channel, interest rate channel, exchange rate channel, asset price channel, credit channel, balance sheet channel, and expectations channel. Furthermore, observations of imperfections and financial frictions during the decade before and after the GFC compelled several economists (Allen & Carletti, 2008; Borio & Zhu, 2008) to view the risk-taking channel as another monetary policy transmission channel.

Monetary Policy Transmission Mechanism     117 The workings of MPTM begin with monetary policymaking at the central bank through interest rate policy and other monetary instruments, such as monetary operations, foreign exchange intervention, reserve requirement, and other instruments (Fig. 5.1). Subsequently, the mechanism is influenced by activities in the financial sector and real economy through the various monetary policy transmission channels, namely the interest rate channel, exchange rate channel, asset price channel, credit channel, balance sheet channel, and expectations channel. Monetary policy is transmitted through two stages of interaction in the economy, namely: (1) interaction between the central bank and the banks as well as other financial institutions in terms of various financial sector transactions; and (2) interaction between the banks and other financial institutions with economic players in the real sector through financial intermediation concerning various domestic and international economic activities. The uppermost panel in Fig. 5.1 presents the various MPTM stages from the central bank, to the financial sector, real economic activities, and the ultimate target. In the financial sector, monetary policy affects interest rates, exchange rates, bond yields, and stock prices, as well as the volume of private funds deposited in the banking industry, bank lending and placements in bonds, stocks, and other securities. Meanwhile, monetary policy influences the real economy through aggregate demand, including domestic demand (consumption and investment) as well as external demand (exports and imports). Furthermore, monetary policy also affects aggregate supply through the cost of production capital as well as the decisions on wages and the duration of work contracts. Ultimately, the magnitude of the gap between aggregate demand and aggregate supply determines the pace of economic growth and rate of inflation as the final targets of monetary policy. Additionally, monetary policy also influences inflation directly through the effect of the exchange rate on the prices of imported goods and services and indirectly through the current account as well as the capital and financial account in the balance of payments (BOP). The central panel of Fig. 5.1 illustrates the interactions between various financial and economic variables in the financial sector and real economy. In reality, MPTM is a complex process and, therefore, is often known as the “black box” by monetary economists (Bernanke & Gertler, 1995). MPTM is predominantly influenced by three factors, namely: (1) behavioral changes in the central bank, government, and banking industry or among economic agents in terms of their economic and financial activities; (2) the lag from when the monetary policy is implemented by the central bank until the time when the effect manifests in terms of economic growth and the final policy target is achieved; and (3) a change occurs in the monetary policy transmission channels due to a change in (1) above in accordance with economic and financial developments in the country involved. The lower panel in Fig. 5.1 shows the determinants of MPTM beyond the control of the central bank, including changes in the risk premium, bank capital, global economy, fiscal policy, and commodity prices.

118    Central Bank Policy

Fig. 5.1:  Monetary Policy and the Transmission Mechanism. 5.2.2. The Importance of Monetary Policy Transmission Behavioral changes in the central bank, government and banking industry or among economic agents will clearly affect the interactions of various economic and financial activities and, therefore, prompt a change in MPTM. Furthermore, due to changes in behavior and expectations, MPTM is replete with uncertainty and is relatively difficult to predict. If the central bank lowers the interest rate, for instance, bank behavior and balance sheet dynamics will conspire to lower the interest rates and adjust the volume of deposits and loans. Such interest rate policy would also influence financial market, exchange rate, and stock price expectations. Additionally, household behavior and balance sheet dynamics also influence how much interest rate policy affects deposits and demand for new bank loans. Similarly, a behavioral change in the banking industry, concerning financial operations and product innovation, such as a reluctance to lend or a proliferation of derivatives in foreign exchange transactions, would also determine how the monetary policy of the central bank influences various economic activities. In general, the effect of MPTM on economic growth and inflation has a relatively long and varied lag (Friedman & Schwartz, 1963). The impact of monetary policy on economic growth and inflation typically takes six to eight quarters. During the preliminary stage in the financial sector, the impact of lowering policy rate on lending and deposit rates usually takes around three to six months because the banking industry must wait for existing term deposits and loan agreements with customers to mature. On the other hand, foreign exchange market and stock market respond more quickly through exchange rate depreciation and elevated stock prices, resulting in changes in the composition of the investment portfolio. Meanwhile, during the second stage, from the financial sector to the

Monetary Policy Transmission Mechanism     119 real economy, household, and corporate demand for new bank loans takes time to manifest. The corporate sector requires assurance that the business outlook is promising enough to pay loan installments to the banking industry. The complexity of MPTM also stems from changes in the role and processes of monetary policy transmission channels in the economy. For economies which are still dominated by the banking industry, MPTM through the interest rate, money supply, and credit channels will play an important role. As financial markets become more developed, however, the role of the interest rate channel and asset prices, such as stocks and bond yields, become more important. Likewise, open economies with perfect capital mobility and a floating exchange rate regime facilitate a faster exchange rate channel than countries applying foreign exchange controls (imperfect capital mobility) and a fixed exchange rate regime. In an open economy, MPTM is also influenced by economic and financial developments in other countries, among others, through changes in the exchange rate, export and import volume, as well as the magnitude of capital inflows and outflows. The speed and scale of MPTM also differ during periods of economic boom and bust (asymmetric). Interest rates rise faster than they fall. Similarly, credit is more expansive (contractive) during an economic upswing (downswing), a phenomenon known as the financial accelerator. MPTM complexity demands analysis and empirical studies to map the various existing transmission channels, including the money supply, credit, interest rate, exchange rate, asset price, or expectations channels, in line with economic and financial development. In terms of monetary policy and theory, MPTM studies generally review two salient aspects. First, to investigate the most dominant transmission channel in the economy to from the underlying base for monetary policy-making strategy. If the money supply channel is dominant, for example, a monetary policy strategy targeting money supply is still relevant. On the other hand, if the interest rate channel or exchange rate channel were dominant, the central bank would have to consider targeting the interest rate or exchange rate as the monetary policy strategy. In general, the exchange rate channel is important, specifically in the case of EME. Second, to investigate the magnitude and duration of the lag of each respective transmission channel, from when the monetary policy decision is taken by the central bank, the effect through each respective transmission channel, to inflation and economic growth. Such analysis is important to determine the strongest economic and financial variables as leading indicators for inflation targeting and future economic growth, as well as to determine the operational target of monetary policy. Consequently, if money supply, credit, interest rate, and exchange rate have strong influences over future economic dynamics and inflation, those variables should be used as leading indicators. Of those leading variables, the most appropriate one is selected as the operational target, for instance the monetary base or interest rates. Furthermore, analysis of the lag profile of each transmission channel is also required to formulate a forward-looking monetary policy strategy. By understanding the lag of the MPTM, the central bank could better formulate forward-looking and pre-emptive monetary policy in the current period to steer future inflation and economic growth in line with the desired targets.

120    Central Bank Policy Comprehensive and in-depth understanding of the MPTM is vital in terms of reviewing the monetary policy-making practices of the central bank as well as to understand and develop monetary economics theory in academia. For the central bank, MPTM mapping facilitates monetary policy formulation in terms of setting the operational target (policy rate and monetary operations), the intermediate target (interest rate, money supply, credit, exchange rate, asset prices, and expectations) and the effectiveness of achieving the final target (inflation and economic growth). For academia, MPTM is an interesting topic for empirical and theoretical study, linked to the debate between monetary economists in line with the monetarists’ viewpoint, credit, and even imperfections and financial frictions.

5.2.3. Policy Rate and Monetary Operations The processes of MPTM begin with the monetary policy decision taken by the central bank. When formulating monetary policy, the central bank generally sets the interest rate to influence macroeconomic projections in line with the desired target, namely price stability and economic growth. Interest rate policy is implemented through monetary operations by the central bank to influence liquidity and, therefore, interest rates in the money market. In addition, the central bank also conducts monetary operations in the foreign exchange market when intervention is required to stabilize exchange rate fluctuations. Other monetary instruments include the deposit facility (DF), lending facility (LF), or discount window, reserve requirement and lender of last resort function using high-quality and liquid securities as collateral. 5.2.3.1. Interest Rate Policy.  The interest rate policy adopted by the central bank will influence short-term interest rates on the interbank money market as well as market expectations of macroeconomic projections and the future direction of the central bank’s interest rate policy. Such developments will subsequently affect bank deposit and lending rates, asset prices on financial markets, including stock prices and bond yields, the exchange rate and long-term interest rates. Using a simple mathematical approach, interaction between the policy rate and bank interest rates can be expressed as follows: (1) transmission from the central bank interest rate (rCB) to the interbank rate (rP) is expressed as the following function: rP = f(rCB, other factors2); (2) transmission from the interbank rate (rP) to the deposit rate (rD) is expressed as the following function: rD = g (rP, other factors); and (3) transmission from the deposit rate (rD) to the lending rate (rK) is expressed as the following function: rK = h (rD, other factors). Confirming previous findings, the transmission process typically contains a lag, primarily due to the maturity structure as well as the loan and deposit agreements between the banks and their customers. Consequently, the reduced form

2

Other factors generally relate to internal conditions in the banking sector.

Monetary Policy Transmission Mechanism     121 equation to investigate the transmission of the policy rate (rCB) to lending rates can be expressed as follows3:

rk = α + ∑ β j rCB ,t− j + ∑ γ j Z j −t + εt(1)

where Z = internal bank conditions, such as deposit composition, liquidity (loanto-deposit ratio – LDR), capital (capital adequacy ratio – CAR), non-performing loans (NPL), and so on, and εt = random shocks in the econometric equation. 5.2.3.2. Monetary Operations.  The central bank performs monetary operations to influence liquidity conditions and, therefore, interest rates on the interbank money market in line with the policy rate set. To that end, the central bank projects bank liquidity conditions (bank reserves) and base money on a weekly basis as a guideline for monetary operations. Liquidity on the interbank money market is monitored, while base money projections are made using a central bank balance sheet equation as follows:

∆B = ∆NFA + ∆NCG + ∆NCP −∆NOI ± ∆OPT (2)

where ΔB = change in base money, ΔNFA = change in net foreign assets, ΔNCG = net claims to government, ΔNCP = net claims to private sector,4 and ΔNOI = net other items. In Equation (2), the magnitude of weekly monetary operations (ΔOPT) represents the difference between the expected position of base money and the policy rate set. When projecting base money, the central bank must consider whether the total money supply in the economy is adequate. In general, the procedure is as follows. First, total money supply (M1 and M2) is projected annually in line with the requirement of the real economy based on the inflation target set, economic growth predictions, and the estimated income velocity in the economy. This can be achieved simply by forecasting the demand for money by the public, for instance, using the following function:

M / P = h( y, r, ∆s )(3)

where y = economic growth, r = interest rate and Δs = change in the exchange rate. Second, using the projection of money supply and estimation of the money multiplier, the position, and growth of base money can be predicted on an annual, quarterly, monthly, and weekly basis using the following equation: 3

m = M / B (4)

The time lag is a characteristic that must be monitored to understand monetary policy transmission. This is applicable to various equations mentioned in this chapter. To simplify understanding, the lag is not always mentioned in the equations. 4 If applicable.

122    Central Bank Policy The weekly projection of base money is subsequently used as a guideline for monetary operations at the central bank to control liquidity and, therefore, interbank rates in line with the policy rate determined by Equation (2). It is important to note that differing from monetary policy using the interest rate as the operational target, as outlined above, monetary policy targeting money supply can also be achieved through monetary operations with Equation (2) based on the projections of base money Equation (4) and money supply Equation (3). In practice, MPTM through the money channel such as this is generally applied by central banks in low-income countries with an underdeveloped financial sector. In this case, base money is controlled through Equation (2) in line with projection Equation (4), which is subsequently transmitted to total money supply (M1 and M2) congruent with public demand Equation (3) and, ultimately, money supply influences inflation and economic growth. The effectiveness of monetary policy in targeting the monetary base depends on several factors. First, there is a stable and predictable correlation between total money supply and inflation and economic growth, reflected in the predictability of income velocity or money demand function stability. Second, there is a causal relationship from total money supply to economic growth and inflation but not vice versa. And third, there is a stable and predictable relationship between base money and total money supply, reflected in the predictability of the money multiplier. As the financial sector develops, however, not only do income velocity and the money multiplier become less stable and more difficult to predict, but an inverse causal relationship often emerges, namely from economic growth and inflation to total money supply. Such conditions impede the control of money supply as the basis of implementing monetary policy based on the money channel in many countries, including EME and advanced countries.5 In addition to the implications for monetary policy, the MPTM through the money channel is also one reason for restrictions on central bank lending to the government. As found in Equation (2), central bank lending to the government to finance the fiscal deficit would increase total base money and liquidity in the economy. On top of influencing interest rate policy effectiveness, this would also increase total money supply in the economy beyond demand, leading to uncontrolled inflation.6 5.2.3.2. Foreign Exchange Intervention.  Exchange rate stability is very important for the economy, especially in EME and developing countries, therefore many central banks, including BI, pay a lot of attention to MPTM through the exchange rate channel. In this case, the exchange rate affects international trade and investment, such as exports, imports, foreign capital investment, portfolio

5

This is evidenced by the growing number of central banks that have switched from targeting money supply to interest rate targeting as the operational target of monetary policy. 6 If the central bank reabsorbed its loans to the government, interbank rates would rise and, therefore, exacerbate investment, and economic growth. In this case, investment distortions would appear from the private sector to the government, which is often known as the crowding out effect in the literature.

Monetary Policy Transmission Mechanism     123 investment, and external debt. Furthermore, the exchange rate and non-resident capital flows influence inflation and economic growth. The exchange rate also affects financial (balance sheet) conditions in the banking industry, household sector, and corporate sector, namely through foreign currency assets and liabilities. Moreover, wide currency fluctuations create economic uncertainty and could even threaten political stability. With greater economic openness, accompanied by a floating exchange rate system and perfect capital mobility, comes a larger influence of exchange rates and non-resident capital flows. In addition to interest rate policy, the central bank also implements monetary policy through foreign exchange market intervention to stabilize exchange rates. Central bank intervention has a direct impact on the supply of foreign exchange and, therefore, exchange rate developments on the market. In this regard, there are several motives and targets for foreign exchange intervention, namely to control exchange rate volatility, prevent misalignment from the currency’s fundamental value and to anchor future exchange rate expectations (Moreno, 2005; Neely, 2001). An empirical study concerning the targets of foreign exchange intervention could be conducted using the following equation (Adler & Tovar, 2014; Neely, 2005):

INTt = α0 + α1∆st−1 + α2 ( st−1 − st*−1 ) + α3 VOLt−1 + α4 INTt−1 + εt(5)

where INT = foreign exchange intervention, Δs = changes to the exchange rate (appreciation or depreciation), (s − s*) = exchange rate misalignment from the fundamental value, and VOL = exchange rate volatility. A change in the monetary policy rate would affect the interest rate differential and then the exchange rate due to the size of foreign capital flows beyond supply and demand in the foreign exchange market. As explored in Chapter 4, exchange rate transmission is shown through international interest rate parity, namely: (1) based on uncovered interest parity (UIP), r − r* = ( s e − s ) / s, or (2) based on covered interest parity (CIP), r − r* = ( F − s ) / s = ( s e − s ) / s + ρ , where {r, r*, s, F, ρ} = {domestic interest rate, international interest rate, spot exchange rate, forward exchange rate, and risk premium}. Several studies have also been conducted to measure the effectiveness of interest rate policy and foreign exchange market intervention in terms of exchange rate stability (Galati & Disyatat, 2005; Gerl, 2006; Miyajima & Montoro, 2013; Sarno & Taylor, 2001). For example, the following model can be estimated:

∆st = β0 + β1 ( rt − rt* ) + β2 ρt + β3 INTt + εt(6)

where Δs = a change in the exchange rate, (r − r*) = interest rate differential, p = risk premium, and INT = foreign exchange market intervention.7 In this case, β3 measures the effectiveness of foreign exchange market intervention in terms of exchange rate stability. 7

Empirically, intervention and changes in the exchange rate are mutually reinforcing and, therefore, Equations (5) and (6) contain heteroscedasticity. In general, GARCH is the estimation technique for the instrumental variables used in empirical studies.

124    Central Bank Policy

5.3. Monetary Transmission Channels: The Money View In monetary economics theory, a debate rages about how the MPTM affects the financial sector and real economy. Per the former theory, known as the money view (Cecchetti, 1995) or neoclassical channels (Boivin, Kiley, & Mishkin, 2010), monetary policy through short-term interest rates and/or base money control influences consumption and investment through long-term interest rates, asset prices, and the exchange rate. A financial system is assumed to be efficient and, therefore, there is no need to discuss the banks’ role in the MPTM. Per the latter theory, however, which is often referred to as the credit view or the non-neoclassical channels, financial system imperfections affect the MPTM in the economy through the credit channel and balance sheet channel. Furthermore, the implications for monetary policy are also different. The assumption of financial market efficiency per the money view implies that monetary policy does not have a distributive effect on the various segments of the real economy along with price setting. In contrast, financial frictions per the credit view cause the impact of monetary policy to vary by segment in the financial sector and real economy, depending on bank behavior and internal conditions, the lending relationship between lender and borrower, as well as risk-taking behavior among economic players. The workings of the MPTM per the money view on investment, consumption and international trade were developed based on the theories that emerged in the middle of the twentieth century, such as the neoclassical investment theory (Jorgenson, 19963; Tobin, 1969), the lifecycle and permanent income models (Ando & Modigliani, 1963; Brumberg & Modigliani, 1954; Friedman, 1957), as well as the international IS/LM analysis models (Fleming, 1962; Mundell, 1963). The MPTM works through the interest rate channel, asset price channel, and exchange rate channel. In this case, monetary policy affects investment via the interest rate channel in terms of the cost of capital and the asset price channel in terms of the value of the firms, which is also known as “Tobin’s q channel.” The influence on consumption occurs through the interest rate channel due to the intertemporal substitution effect and through the asset price channel because of the wealth effect. Meanwhile, the exchange rate channel facilitates direct and indirect exchange rate pass-through to inflation and the international trade channel through changes in the relative prices of exports and imports.

5.3.1. Interest Rate Channel The most commonly cited MPTM channel in macroeconomic modeling involves the influence of the interest rate on corporate and household propensity to invest, including investment in fixed assets, such as land and buildings, as well as durable assets, such as motor vehicles, machinery, and equipment. Therefore, macroeconomic models take into account the influence of the interest rate on private consumption. Transmission through the cost of capital on investment. The standard neoclassical investment model shows that the user cost of capital is a key determinant that

Monetary Policy Transmission Mechanism     125 affects demand for investment. The user cost of capital, μc, is usually expressed by the following equation:

µc = pc [i − ( E ( πc ) − δ )]

with

E ( πc ) = ∑ i = 0 βt Et ( Πt +i )(7) α

where pc = relative price of new capital asset (compared to the replacement), i = nominal interest rate, E(πc) = expected capital asset price hike, and δ = rate of depreciation. The equation shows that the external finance premium represents the interest rate cost to obtain new capital minus the increase in the value of the capital asset after depreciation. In terms of corporate investment, the expected capital asset price hike may represent the current market value of the asset or profit rate (II) or current and future cash flow. In this case, firms or households will only take on new investments using bank loans if the user cost of capital is negative. Consequently, expected capital asset price hike, after depreciation, would exceed the interest payments of the loan received. Therefore, the investment demand function can be expressed as follows:

I d = g ( E (Y ), µc , Z )(8)

where I = investment, E(Y) = economic growth outlook, μc = user cost of capital in Equation (7), and Z = vector of other variables that affect investment, such as business risk, ease of licensing, level of competition, and so on. Demand for investment would increase if the cost of capital was low or decreasing and the business outlook was promising. Several factors demand attention when analyzing the effect of monetary policy on investment through the interest rate channel. First, the long-term interest rate has a greater effect on investment, considering the decision to invest is long term and, therefore, firms will consider interest rate fluctuations and expected increases in asset prices over the same horizon as the investment. Second, monetary policy is determined by and works through short-term interest rates, therefore, MPTM effectiveness through the interest rate channel is also affected by the term structure of interest rates, which is linked to short- and long-term interest rates. When the policy rate is reduced, for instance, the long-term interest rates also decrease due to price rigidity in the near term. Nevertheless, this only occurs if the financial markets can form the term structure of interest rates and if the central bank’s monetary policy is credible and, hence, able to anchor the expectations of economic players in terms of their business decisions. Transmission through the substitution effect on consumption. Interest rates affect consumption through the intertemporal substitution effect. In other words, interest rates influence household preferences or decisions about whether it would be better to increase consumption in the current period or postpone the increase in consumption until a future period. This can be expressed using the following equation:

C = c (Y d , rD , rK )(9)

126    Central Bank Policy where C = consumption, Yd = disposable income, rD = term deposit interest rate, and rk = consumer loan lending rate. If the policy rate was lowered, for instance, lending rates would also be reduced, which would therefore drive household consumption. This would depend on the respective preferences and financial conditions of each household, namely as savers or borrowers. In Equation (9), in addition to the impact on lending rates, the central bank’s interest rate policy would also affect deposit rates and, therefore, interest income for the savers. The income effect and substitution effect determine the influence of the interest rate channel on consumption.

5.3.2. Asset Price Channel Monetary policy, implemented through interest rates, monetary operations, and foreign exchange intervention or other instruments, also affects other asset prices, including financial asset prices, such as bond yields and stock prices, as well as physical asset prices, particularly property prices and gold. This transmission occurs because funds in the investment portfolio are not only in the form of term deposits held at a bank and other investment instruments on the rupiah money market and foreign exchange market, but also as bonds, stocks and physical assets. Therefore, changes in the interest rate and exchange rate as well as the magnitude of investment on the rupiah money market and foreign exchange market would affect the volume and prices of bonds, stocks and physical assets. For financial assets, the basic theory applied is investment portfolio formation by the investor, à la Markowitz model, as an alternative investment in central bank instruments with a risk-free interest rate of rCB, bank term deposits, bonds and stocks as follows:

MaxU (C , A) with constraint :

∑ pC i

i

= W0 + ∑ ri / σi Ai (10)

where {U, C, A, W, p, r, σ} = {utility, consumption, assets, wealth, consumer goods prices, asset yield, asset risk}. From the investment portfolio formation, the demand solution for each respective financial asset in equilibrium produces a yield relationship for each respective financial asset after calculating the risk factor as follows:

rB / σB = f ( rCB , rD / σD , dS / σS )(11a)



dS / σS = g ( rCB , rD / σD , rB / σB ) (11b)

where rD and σD = term deposit rate and risk, rB and σB = bond yield and risk, dS and σS = stock yield and risk. Equation (11) shows that the central bank interest rate affects the yields of bonds and stocks. Transmission through Tobin’s q on investment. The investment decisions of firms and households can be analyzed using Tobin’s q ratio. For business investment, Tobin’s q is defined as the ratio of the market value of a company’s assets

Monetary Policy Transmission Mechanism     127 divided by the replacement cost of the company’s assets. A high Tobin-q value, for instance during an economic upswing, when the company’s market value is high and/or the policy rate is lowered, thereby reducing the replacement cost, implies that corporate demand for investment would increase. Companies would, therefore, expand bank borrowing or issue stocks or bonds to finance the investment. A similar analysis could be applied to household investments, for instance property, because of rising asset prices as lending rates fall. Tobin-q analysis, which emphasizes transmission through the asset price channel, may be linked to the user cost of capital approach, which emphasizes transmission through the interest rate channel as explained previously. Transmission through the wealth effect on investment. Rising asset prices could also drive consumption due to an increase in net worth. The lifecycle hypothesis of saving and consumption shows that current spending on consumption is determined by the resources at hand to the consumer over his/her lifetime, encompassing not only future income but also net worth, such as property, stocks, and other assets. A reduction to the policy rate would stoke demand for assets, such as property and stocks, thereby edging up asset prices. Similarly, a reduction in the interest rate means a lower discount rate, hence elevating the current value of income flows and returns on investments in stocks, property and other assets received at a future date. An increase in net worth would drive household consumption and aggregate demand. Therefore, the wealth effect through asset prices represents an important MPTM channel, especially in countries where household investment in financial assets, such as stocks and bonds, or property is high.

5.3.3. Exchange Rate Channel According to the money view, monetary policy transmission through the exchange rate channel affects the real economy through the export and import component of aggregate demand as well as the imported prices of goods on inflation. Exchange rate transmission on international trade. A reduction to the policy rate, consistent with the interest rate parity theory mentioned previously, precipitates exchange rate depreciation because the return on domestic assets declines compared to the return on assets available offshore. Exchange rate depreciation would stimulate exports and, simultaneously, reduce imports. Nonetheless, the impact of net exports is not always desirable, depending on export elasticity to exchange rates relative to the import elasticity, otherwise known as the BalassaSamuelson effect:

xt = α + βτt + γ ( yt − yt* ) + εt

with

τt = ∆pt − ( ∆st + ∆pt* )(12)

where {x, τ, y, y*, s, p, p*) = {net exports, terms of trade (ToT), domestic economic growth, global economic growth, exchange rates, domestic prices, international prices}. The exchange rate channel plays an important role in small open economies, with a more favorable impact on economic growth if the export structure is dominated by manufacturing commodities rather than primary goods.

128    Central Bank Policy Exchange rate transmission on inflation. In addition to indirect pass-through via net exports and growth, as mentioned above, there is also direct pass-through to inflation. Direct pass-through occurs because exchange rate developments affect corporate price setting and the public’s inflation expectations, particularly for imported goods as finished goods, raw materials, and capital goods. This can be explained using purchasing power parity (PPP), namely sP*/P = 1, which can be estimated using the following equation:

∆pt = α∆St + β∆pt* + εt (13)

where Δpt = domestic inflation, Δst = exchange rate depreciation, ∆pt* = global inflation, εt = random shock, with α = β = 1 for PPP. In general, the effect of the exchange rate on inflation is lower in countries with a flexible exchange rate system and sufficient monetary policy credibility to anchor inflation expectations.

5.3.4. Expectations Channel As uncertainty continues to accumulate in the economy and financial sector, the expectations channel is becoming increasingly important in terms of monetary policy transmission to the real sector. Economic players, such as banks and companies, will base their business decisions on the economic and financial outlook. They form certain expectations from the performance of various economic and financial indicators. How economic players form their expectations is influenced by their anticipation of central bank and government policies, as well as global economic developments. Consequently, monetary policy credibility to anchor inflation expectations plays an important role in terms of the MPTM. Inflation expectations are often used by economic players as an indicator of the central bank’s interest rate policy. The problem is that inflation expectations are unobservable. Therefore, one estimation method applies Fisher theory, namely8:

rte+ j = ite+ j + πte+ j 14a If there is a forward interest rate on the market for tenor j, then:



πte+ j = rt +f j − ite+ j 14b

where {re, rf, ie, πe} = {nominal interest rate expectations, nominal forward interest rate, real interest rate expectations, and inflation expectations}. Mishkin (1990) used this equation to measure future inflation expectations as follows: 8

Another method would be to use the inflation expectations of the business community and consumers contained in various surveys, such as the Business Survey and Consumer Expectations Survey published by BI.

Monetary Policy Transmission Mechanism     129

[ πt + j − πt ] = αt + j + βt + j [ rt + j − rt ] + εt (15)

where βt+j†0 shows the inflation expectations information contained in the interest rate for period j. Inflation expectations affect various real sector activities. The effect on aggregate demand occurs due to the impact on real interest rates, which determine the magnitude of demand for consumption and investment. Meanwhile, inflation expectations affect aggregate supply through corporate product price setting. In addition, the effect of inflation expectations on aggregate supply and demand will also determine the rate of inflation and real output in an economy. Greater monetary policy credibility will reduce the deviation in inflation expectations from the inflation target set by the central bank. Consequently, the distortions that emerge in terms of real output and monetary policy effectiveness to achieve the inflation target become smaller.

5.4. Monetary Transmission Channels: The Credit View As explained previously, the money view theory of MPTM assumes an efficient financial system. The interest rate structure on financial markets transmits the short-term monetary policy rate to longer term interest rates, which subsequently influences relative prices, the components of aggregate demand (investment, consumption, and international trade) and inflation. The financial system also supplies financial products, thus making it possible for companies and households to build an optimal investment portfolio and/or access financing sources other than bank loans, such as issuances of stocks and bonds. Similarly, the foreign exchange market also works efficiently. Exchange rate fluctuations, therefore, are not misaligned from the currency’s fundamental value, including when faced with volatile non-resident capital flows. The various aspects of financial system efficiency influence the effectiveness of monetary policy transmission through the money view channels. Departing from the money view, the credit view of MPTM theory is based on imperfect financial markets due to asymmetric information and moral hazard in various financial transactions (Bean, Larsen, & Nikolov, 2002; Cecchetti, 1998). The effect of financial frictions on the MPTM was observed by Bernanke and Gertler (1995), who found three anomalies in the aggregate demand response to changes in the interest rate as follows: (1) composition, namely that changes in the short-term interest rate affected durable goods, such as houses, which should be more responsive to long-term interest rates; (2) propagation, namely that the real economy continues to respond after the short-term interest rate has been raised and even when the interest rate has been lowered; and (3) amplification, namely that the interest rate triggers a faster change in output even though the investment expenditure of an individual firm does not significantly respond to the cost of capital. In the context of the exchange rate channel, international interest rate parity is not really explained by the economic cycle (Meese & Rogoff, 1983). The effect of the exchange rate on inflation is also somewhat subdued, with a portion,

130    Central Bank Policy therefore, spilling over to undermine corporate profit margins. In addition, there were indications of an asymmetric impact of monetary policy on real output and inflation between tightening and easing, which cannot be explained by the interest rate or exchange rate channels. Such observations compelled the credit view of MPTM theory to propose three alternative transmission channels, namely the bank lending channel and bank capital channel, balance sheet channel and risk-taking channel. The bank lending and bank capital channels explain how the MPTM works through bank lending behavior in the face of asymmetric information concerning borrower conditions and bank modeling requirements. The balance sheet channel focuses on MPTM analysis in terms of the loan relationship between lender and borrower, which is influenced by the external finance premium and collateral constraints due to moral hazard. Meanwhile, the risk-taking channel analyzes the effect of monetary policy on risk management and financial product innovation in the financial system.

5.4.1. Bank Lending and Bank Capital Channels The bank lending and bank capital channels play an important banking role in the intermediation function in an economy to overcome the problems associated with asymmetric information concerning the actual feasibility of debtor projects. To that end, the banks are selective regarding applications for new loans (demand for credit) and then monitor the borrower to avoid loan delinquency and default. The ability to select borrowers and the amount of capital required to mitigate credit risk will affect the MPTM through the supply of credit by banks. 5.4.1.1. Bank Lending Channel.  The first model of the bank lending channel was developed by Stiglitz and Weiss (1981) based on the assumption that the borrower has access to private information about the feasibility of his/her businesses. Despite producing the same expected return, the borrower’s business project contains different probabilities of success. With limited liability, pursuant to the loan agreement from the bank, the borrower could choose to ignore loan delinquency if the business fails. Therefore, if the central bank tightened monetary policy by applying a higher policy rate, only the debtors with a low chance of success (high risk) would opt to borrow from a bank, leading to a low rate of return on the loan. Consequently, as explained in Chapter 3, the credit market is always in a state of imbalance and the banks are inclined to ration credit. The phenomenon of credit rationing affects the transmission effectiveness of interest rate policy through the bank lending channel on investment and consumption in the real economy. Several studies propose solutions concerning the need for a selection instrument to reveal the actual business feasibility of a prospective borrower. In the present context, the selection instrument may be in the form of a credit information system that contains the business conditions of borrowers and their credit history. This view was proposed by Rothschild and Stiglitz (1976), who found that such a selection instrument would overcome credit rationing. Meanwhile,

Monetary Policy Transmission Mechanism     131 Bester (1985) proposed the need for credit guarantees as a selection instrument, thereby only safe borrowers would apply for new loans rather than risky borrowers. Nevertheless, as sophisticated as the credit selection process may be, or even with credit guarantees in place, credit risk will always emerge after the loan has been disbursed. Business failure is always possible due to a plethora of reasons, including management negligence, moral hazard, or general economic conditions. Banks continue to face the problem of adverse selection when extending loans. In a bank’s credit portfolio, there are always borrowers whose businesses succeed and those who contain the risk of default. Several later studies linked credit supply to the bank funding structure. Bernanke and Blinder (1988), for instance, showed that the funding structure of retail banks would influence bank lending. Such conditions are typically found in countries dominated by the banking sector and with an underdeveloped capital market. The banks will rely on retail funding, while the corporate sector also depends on bank loans. The business scale is usually small and medium enterprises, which do not meet the requirements to issue stocks or bonds on the capital market. Therefore, the banks could face funding shortfalls if the central bank tightened its monetary policy stance because retail funding is not normally sensitive to the interest rate, while a dip in income could reduce bank funding. Consequently, the banking industry would ration more credit, with small and medium borrowers most affected.9 Bernanke and Blinder (1988) demonstrated that monetary policy will tend to amplify the real economy under such a bank credit and funding structure, implying that credit will be too tight when the economy experiences a downswing and too loose during an economic upswing. 5.4.1.2. Bank Capital Channel.  Bank capital also affects credit supply. Van den Heuvel (2002) studied the impact of the CAR and ability of banks to issue shares. Bank capital could be influenced by the ability to generate profit, the market price assessment of net assets on the balance sheet and the mandatory reserves required to mitigate credit risk and default of the securities held. The ability of a bank to issue shares on the capital market also affects bank capital. More bank capital implies greater lending capacity. Capital constraints in terms of credit supply become more binding if the level of capital approaches or falls below the capital requirements set by the central bank or relevant supervisory authority. Such factors in the bank lending and capital channels influence the supply behavior of banks. In general, the bank supply function is determined not only by the lending rate (rk) and economic outlook (y), but also by credit risk (ρk) liquidity (LDR) and capital (CAR) at each respective bank as follows:

9

K i = f ( y, rki , ρ ki ,LDR,CAR)

with

rki = rd + c i (16)

Under such conditions, the LDR is not always constant and, therefore, monetary policy would have an asymmetric impact on bank lending if the policy stance is changed. This is very different to the money supply channel that assumes no funding constraints, thus changes in lending will always mirror changes in money supply.

132    Central Bank Policy In the credit supply function (16), banks may set different lending rates for different borrowers depending on the cost of monitoring (ci) and credit risk (ρki), despite the same funding rate (rd). Such conditions show that monetary policy has a distributive effect on loan allocation by borrower group. Similarly, the internal bank conditions, such as LDR and CAR, also affect the credit supply function and, therefore, monetary policy transmission at each respective bank. Furthermore, the workings of the MPTM through the bank lending and capital channels are more complex and tend to be asymmetrical. During a period of monetary policy easing, lower interest rates may drive credit supply in several ways. First, a lower policy rate would increase the net interest margin (NIM) and increase profits, thereby improving the balance sheet over time. Second, monetary policy easing also pushes up asset prices and directly boosts capital. Third, credit risk would also dissipate due to sounder business conditions among the borrowers during an economic upswing and, therefore, reduce the need for reserves and increased bank capital. In general, through the bank capital channel, monetary policy easing could increase bank capital and lending beyond that possible by lowering the interest rate and, therefore, drive aggregate demand, especially by facilitating bank-dependent borrowers to increase spending on investment and consumption. Nonetheless, an economic downswing, or financial crisis, would exacerbate conditions. A decline in asset value, increase of NPL and lower profits would undermine bank capital, thus compelling the bank to reduce lending. Issuing stocks to maintain bank capital would not be a feasible option because assets prices would track a downward trend and there would be fewer investors to purchase the stocks as the economy moderates. Less credit availability would exacerbate economic conditions and also, therefore, the business conditions and balance sheets of the debtors, with a higher intensity than the effect of the interest rate, which would subsequently compound bank capital due to escalating credit risk and lower profits. Such financial sector deleveraging would undermine economic conditions further, particularly due to a decrease of corporate spending that depends on bank loans. According to such conditions, the central bank’s interest rate policy would also become less effective.

5.4.2. Balance Sheet Channel Differing from the bank lending and capital channels, which stress adverse selection and credit rationing in the face of (ex ante) asymmetric information, the bank balance sheet channel emphasizes the issue of moral hazard because of (ex post) asymmetric information in terms of lending transactions between lender and borrower. The borrower, after receiving the loan from the lender, may be unable to fulfill their obligations because of moral hazard. Due to the costly state of verification of the borrower’s business feasibility and/or limited liabilities in terms of enforcing the loan agreement, this channel proposes two solutions to overcome moral hazard. First, is the external finance premium and the second is to demand additional collateral (collateral constraints) in the loan agreement.

Monetary Policy Transmission Mechanism     133 5.4.2.1. External Finance Premium. The ideas of Bernanke, Gertler, and Gilchrist (1999) as well as Clastorm and Fuerst (2000) represent two salient references in the external finance premium model. For each business transaction, there is always the possibility of default, where it would be more efficient for the borrower not to honor the loan agreement. Nevertheless, there is also moral hazard because of the limited liability contained in the loan agreement. Due to the costly state of verification of a project, the borrower must pay a cost to ensure actual business feasibly. Therefore, the lender imposes an external finance premium on the borrower to compensate the cost of verification. The external finance premium creates imperfections, or financial frictions, in the lending transaction between the lender and borrower. For safe businesses, with a feasible project, the external finance premium implies it is better to use internal funding rather than external finance. A highly profitable project would encourage the business to increase investment, thus driving aggregate demand and real output. Meanwhile, for higher-risk businesses and those facing balance sheet constraints to internal finance, investment funding in the form of lending is an alternative option that incurs the cost of the external finance premium. In other words, the external finance premium is a veil that creates financial frictions between capital finance and investment. The issue is that the external finance premium lowers the profitability of the borrower and, therefore, increases the risk of default. Furthermore, Bernanke, Gertler, and Gilchrist (1999) showed that the eternal finance premium also triggers the financial accelerator, which amplifies procyclicality between the financial cycle and economic cycle. Such analysis is significantly different from the real business cycle because of the productivity effect without including financial aspects (Kydland & Prescott, 1982; Long & Plosser, 1983; Lucas, 1980). The financial accelerator phenomenon fundamentally changes MPTM analysis in the financial system and real economy. In the case of monetary policy easing, for instance, a reduction to the interest rate would spur increased investment at feasible and profitable businesses, thus strengthening the economic cycle. Increased corporate investment by less feasible or unfeasible businesses, however, would exacerbate the risk of default due to the external finance premium or business failure. Consequently, not only does the financial accelerator affect the financial and economic cycles, the risk that an economic upswing could reverse into a recession, or boom become bust, also increases, which could potentially spur a crisis. The reverse is true concerning the effect of tighter monetary policy on investment and real output. A hike to the interest rate would undermine cash flow and, therefore, the profitability of the borrower’s business. In addition, a higher interest rate would also prompt the lender to raise the external finance premium imposed on the borrower, which would further compound the risk of business default and loan delinquency. A similar analysis was also applied to lending transactions in the form of loans between firms and the banks. Amalgamating MPTM analysis through the corporate balance sheet channel with the bank lending and bank capital channels would better explain the financial accelerator that occurs due to the interest rate policy of the central bank (Bernanke & Gertler, 1995). Analysis

134    Central Bank Policy of how the financial accelerator exacerbates the impact of monetary policy on the financial and economic cycles is an integral part of understanding the boom-bust cycle, which often culminates in a crisis, including the GFC. 5.4.2.2. Collateral Constraints.  Financial accelerators can be analyzed using the collateral for loan transactions between lender and borrower. The seminal views of Kiyotaki and Moore (1997) represent an important reference in the analysis based on difficulties enforcing the loan agreement due to limited liability, amounting to the nominal value of the loan. Therefore, to prevent moral hazard by the borrower, the lender typically requests collateral in the form of physical assets in addition to collateral to ensure business feasibility. The size of the loan offered by the lender is, therefore, based on an assessment of assets to be used as collateral by the borrower and will hence fluctuate in line with the market value of the collateral. Subsequent MPTM analysis, using collateral constraints, was similar to the financial accelerator analysis conducted by Bernanke, Gertler, and Gilchrist (1999) mentioned above. A reduction to the central bank’s policy rate would improve economic performance and raise asset prices. A higher market value of the assets used as collateral would raise the credit ceiling from the borrower for the lender to offset the cost of investment. The financial accelerator would then occur between the higher asset prices, size of loan, demand for investment, economic growth, and so on. The reverse is true if the monetary policy rate is raised. Less economic activity would lower the market price of assets as well as the value of collateral and investment, and further undermine economic performance. Consequently, the financial and economic cycles could change from boom to bust and, therefore, exacerbate the risk of a crisis occurring. A simple analysis of the financial accelerator is presented in the following section, including the external finance premium and collateral constraints. Supposing a business requires an input, x, for the upcoming production cycle with a production function, y1 = z1 F ( x1 ) , where y = output and z = productivity. To pay for the production input, the business could use internal finance from the sales of production in the previous period, p0y0, or borrowed capital with instalments (1+r)L. Therefore, the financing constraints to procure the production input can be expressed as follows:

x1 ≤ [ p0 y0 − (1+ r0 )L 0 ] + L1(17)

To receive the loan, the business must provide additional collateral, for instance in the form of buildings (W), with a market value, q1 and α as the (inverse) of the loan-to-value (LTV) ratio. Therefore, the loan received by the business, with the collateral constraints can be written as follows:

L1 ≤

q1W (18) α(1+ r )

Monetary Policy Transmission Mechanism     135 Under such conditions, the optimal level of credit for the business is selected to maximize profit to ensure production can continue even if the internal capital is insufficient as follows:

Max ( L1 ) .Π = p1 z1 F ( x1 ) − (1− r1 )L1(19)

with constraints (17) and (18). If the market value of the collateral is adequately large that the business no longer faces any collateral constraints, profit maximization is, therefore, as follows:

Max ( L1 ) .Π = p1 z1 F ( p0 y0 − (1 + r0 )L0 + L1 ) − (1− r1 )L1(20)

Therefore, the first-order condition of optimization requires the same marginal product value of the input as the cost of borrowed capital as follows:

p1 z1 Fx ( x*) = (1+ r1 )(21) with the optimal loan size as follows:



L1 = x1* − [ p0 y0 − (1+ r0 )L0 ](22)

Therefore, if the value of collateral is adequate, the business will be able to overcome a lack of funding from the bank loan if production declines. In other words, a temporary shock, for instance from the effect of monetary policy, would not disrupt production, merely erode some of the income between the business and bank. Conditions would be different if the market value of the collateral was insufficient for the business to secure a loan of the desired amount, thus the business would face collateral constraints per Equation (18). In this case, the scale of the production inputs purchased by the business would be determined by the maximum loan feasible, namely:

x1 = x1 = [ p0 y0 − (1+ r0 )L0 ] +

q1W < x1* (23) α(1+ r1 )

In this regard, production would be suboptimal due to less corporate investment. In other words, monetary policy would lower the value of collateral, leaving the business unable to secure an adequate loan, which would not only affect the size of the loan but also production and real output. Therefore, the impact of monetary policy would exceed the effect of the interest rate. Both credit conditions and the impact on real output are illustrated in Fig. 5.2 if the collateral constraints are not binding and in Fig. 5.3 if they are. The previous description shows that collateral is required to protect the lender if the borrower defaults on their loan. Through collateral, the banks can offer a

136    Central Bank Policy

Fig. 5.2:  Business Decisions without Collateral Constraints.

Fig. 5.3:  Business Decisions with Collateral Constraints. lending rate approaching the risk-free rate. This does not mean, however, that the lending rate will be the same as the risk-free rate. While credit rationing is present due to asymmetric information, the marginal value of each additional loan is generally larger than the market value, namely the lending rate. This is congruent with the external finance premium explained by Bernanke and Gertler (1999). In Fig. 5.2, if the collateral constraints increase, conditions are determined by the business that can only procure x1 < x1* input units. In this case, there is an external finance premium that must be borne by the borrower, which exceeds the cost of using internal funds (or equivalent to the net return on production) equal to r1 < r1. The external finance premium will become smaller as corporate balance sheet conditions improve, measured by net capital and liquidity. Collateral constraints and the external finance premium create financial accelerators in the economy. During an economic recession and financial crisis, for

Monetary Policy Transmission Mechanism     137 instance, corporate balance sheets tend to decline, while the banks tighten lending standards and raise the external finance premium because of several factors, including less production and lower prices and, thus, less internal business capital, a lower market value of property and other financial assets and, therefore, the value of collateral also declines and the banks offer fewer loans (LTV ratio is elevated) due to the higher-risk business outlook, while the costs of monitoring and resolving bad loans increase. In contrast, during an economic upswing, the uptick of credit expansion tends to outstrip demand due to more profitable businesses, higher market value of collateral and more loans offered by the banking industry due to the promising business outlook and less stringent lending standards. MPTM analysis through the balance sheet and collateral constraints channels can also be applied to borrowing transactions in the household sector. For example, rising house or property prices would raise the value of collateral, thus facilitating larger loans and more lenient lending standards. In other words, higher property asset prices could lower the external finance premium and unwind the collateral constraints to the amount of credit available to households. Financial accelerators occur in the financial and economic cycles, namely the chain of rising property prices, credit, demand for investment and property development, and increasing economic activity, which subsequently edge up property prices and so on. A housing sector boom, accompanied by a low monetary policy rate, was prevalent during the era of Great Moderation in the United States, which subsequently culminated in the worst crisis since the Great Depression in the 1930s.

5.4.3. Risk-taking Channel The periods before and after the GFC showed that the risk-taking channel in the financial system contributes to fragilities, contagion and asset price bubbles. Risk amplification by the financial system could occur due to several reasons, including financial product innovation, the capital and accounting valuation methods, ease of funding, risk-tolerance as well as global financial integration. Financial shocks leading up to and after the GFC proved that the financial system is not necessarily a shock absorber but can also perform as a shock amplifier (Allen & Carleti, 2008). From many perspectives, risk amplification in the financial system is inextricably linked to monetary policy. Monetary stability and readily available liquidity during an economic boom (upswing) period facilitates greater financial product innovation, while exacerbating risk-taking behavior and financial fragilities. In contrast, during an economic downswing, financial amplifiers due to risk-taking behavior compound a financial crisis and the contagion, thereby complicating the appropriate monetary policy response required. Specifically, the risk-taking channel is defined as the impact of changes in policy rates on either risk perceptions or risk-tolerance and, hence, on the degree of risk in the portfolios, on the pricing of assets and on the price and non-price terms of the extension of funding.

138    Central Bank Policy 5.4.3.1. Risk-taking Dynamics. Financial theory in terms of investment teaches return optimization in line with the levels of risk of each respective alternative investment. This is achieved through risk diversification by forming an optimal investment portfolio, congruent with the return and risk profiles. Financial product innovation was created to be return enhancing in terms of investment portfolio optimization, such as through housing loan securitization with mortgage-backed securities and asset-backed securities. Financial derivatives, including forward, swap, option and credit default swaps (CDS), were also created to mitigate and hedge the risks. Various measurement techniques and techniques to set prices per the risks were also developed, such as value at risk (VaR) and the probability of default (PoD). Similarly, risk-management techniques were also strengthened, which led to the innovation of financial products as if they were risk free. From the investors’ perspective, financial product innovation and forming the investment portfolio could boost yields and enhance risk-management, which would subsequently drive further innovation of financial derivatives that become more developed and complex. In addition, financial product innovation has also changed business models in the financial system, including the banking business model, by facilitating the proliferation of credit securitization strategies and wholesale funding, thereby facilitating the further creation of liquidity and lending, which has become known as the originate to distribute strategy in the United States and United Kingdom. Increasingly complex financial product innovation, coupled with a lack of investor understanding, flourished during the decades of Great Moderation prior to the GFC in the United States, supported by low inflation and monetary policy rates. Omitted from the previous analysis is the fact that financial product innovation and monetary stability have increasingly encouraged risk-taking behavior in the financial system in the pursuit of higher yields during an economic boom. In fact, such observations were also made by Minsky (1982), namely that the capitalist economic system, based on the trade of capital as investment funding in the economy, tends to create financial instability, with uncontrolled debt accumulation during a boom period. Expectations of further profit gains and increasing net worth compared to the burden of loan installments during an economic boom causes more investors to speculate and become involved in Ponzi schemes rather than hedge. This is also in line with the observations of three MPTM analysis anomalies in the economy by Bernanke and Gertler (1995), which were explained through the bank lending and capital channels as well as the balance sheet channel in two subsections above. How does the risk-taking channel strengthen financial accelerators and the amplification of financial and economic cycle dynamics? Borio and Zhu (2008) alluded to three ways in which the risk-taking channel affects the MPTM in the financial system and the economy. First, the effect of the interest rate in terms of valuations, revenues, and cash flow from the investments. The mechanisms of the risk-taking channel are nearly the same as the financial accelerator through the balance sheet channel explained previously, and the two channels also tend to be mutually reinforcing. A reduction to the interest rate, for instance, would

Monetary Policy Transmission Mechanism     139 edge up the valuation of assets prices and collateral, cash flow and profits, thus strengthening the financial accelerator in bank lending and other financial transactions. Similarly, the perception of bad loans would also decline during an economic upswing, hence lowering the external finance premium and easing tolerance of lending standards. Second, the link between the interest rate and target yield is generally used as a benchmark to evaluate the performance of financial investments (Rajan, 2005). A higher yield compared to the benchmark implies a larger bonus for the investment manager. The nominal yield target does not often change due to the nature of the contracts with investment managers, namely in pension funds and insurance. Consequently, a reduction to the policy rate would create a large disparity with the target yield and encourage investors to seek alternative investments with a higher yield (search for yield). Investor perception, which tends to overestimate asset prices above the fundamental value during an economic boom, also coined as irrational exuberance by Alan Greenspan in 1996, is yet another reason for financial accelerators. Third, the effect of monetary policy communication and transparency by the central bank. As an integral part of the monetary policy framework, many central banks are becoming increasingly transparent in their communications on future inflation and macroeconomic projections as well as the interest rate policy and underlying policy considerations. The direction of monetary policy is also communicated, including several key indicators underlying future interest rate policy-making decisions, often known as forward guidance, or statements from central bank officials. Monetary policy transparency can anchor the expectations of economic players and lower the risk premium on the financial markets. Economic stability and a clear monetary policy direction strengthen business assurance and, therefore, nurtures investment activity and various other economic activities, especially by prudent and productive economic players. 5.4.3.2. The Role of Liquidity.  Financial amplification through the risk-taking channel is also inextricably linked to liquidity, as an important element of financial transactions. There are two types of liquidity, namely funding (cash) liquidity and market liquidity. Funding liquidity represents the investors’ ability to obtain cash for the assets held, through selling or use, to receive external finance, as loan collateral. Meanwhile, market liquidity indicates conditions where the volume and price setting on the market work efficiently, thereby each investor can transact at market prices. From those two perspectives, liquidity conditions reflect an important dimension of the effect of financial conditions on the real economy. Easy access to funding liquidity and market liquidity removes the constraints to investment and other real economic activities. Such analysis is similar to collateral constraints from the perspective of Kiyotaki and Moore (2001) above. Liquidity and risk-taking behavior are closely interconnected and mutually reinforcing. When the central bank eases monetary policy, for example, the lower interest rate also loosens funding liquidity and market liquidity conditions. Simultaneously, the lower interest rate also improves the risk perception and eases risk-tolerance, thus encouraging economic players to engage in higher-risk

140    Central Bank Policy investments. The interaction between risk-taking behavior and liquidity during an economic upswing can create financial amplifiers. The opposite is true when the economy is moderating or in recession, when risk perception and risk-tolerance tighten. Lower asset prices trigger sales in order to realize a profit or, at least, prevent the risk of further losses. Consequently, more investors are looking to sell assets rather than buy, which further tighten funding liquidity and market liquidity. Asset prices subsequently tumble, thus encouraging more investors to sell. Fire sales begin to appear as the majority of investors prefer to hold cash rather than assets, a phenomenon known as cash is king. The declines are amplified, which creates financial system instability and could spur a crisis. A clear example of this is the case of bad housing loans that led to the subprime mortgage debacle in the United States and culminated in the GFC. The explanation above demonstrates the importance of risk-taking behavior in the financial system in MPTM analysis of the economy. The presence of and interaction with other MPTM channels creates financial accelerators and amplifiers, driven by the increasing rapidity and complexity of financial product innovation, which is not easily understood by risk-taking behavior and the implications on the financial system and real economy. The risk-taking channel strengthens the bank lending and bank capital channels as well as the balance sheet channel in terms of financial accelerators and amplifiers. Similarly, interaction between the risk-taking channel and funding liquidity and market liquidity could create a liquidity multiplier. The boom-bust financial and economic cycles resulting from interactions between the various MPTM channels often cause fragilities and crisis risks in the financial system and economy. Theoretical and empirical understanding of this is crucial in terms of monetary policy analysis and formulation at the central bank.

5.5. Monetary Policy Transmission in Various Countries Numerous MPTM studies have been performed by central banks and academia in a range of advanced countries and EME. This corroborates the importance of understanding MPTM as well as how the transmission channels interact and are affected by the specific economic and financial conditions of a country. In terms of monetary policy formulation at the central bank, empirical MPTM studies are vital to deepen our understanding of how the monetary policy response, including the interest rate policy, exchange rate policy, or other instruments, is transmitted to the financial system and its influence on the final targets of inflation and economic growth. For academia, many empirical MPTM studies have been conducted to find empirical evidence for a particular theory, or even propose a new theoretical idea from the empirical evidence observed in terms of financial system and economic behaviors. The approach used in empirical studies depends on the objectives, methodology and availability of data. In terms of the objectives, empirical MPTM studies are typically conducted in the form of mapping and investigating the relative effectiveness of each respective MPTM channel, with a focus on MPTM channels and specific cases, such as the workings of the bank lending and capital channels

Monetary Policy Transmission Mechanism     141 or the corporate balance sheet channel, or a comparison between countries, for example, monetary policy transmission in EME. In general, the estimation method applies a vector autoregressive (VAR) model to map the transmission channels or single equation model using panel microdata for specific channel case studies (Amato & Gerlach, 2001). The latest empirical MPTM studies have also applied dynamic stochastic general equilibrium (DSGE) models to investigate monetary policy transmission on financial accelerators in the financial and economic cycles (Boivin et al., 2010). Data availability is also very important, not only time series data but also disaggregated data at the micro level for banks, corporations, and households. The simultaneous effect between monetary policy and financial and real economic variables demands attention in the estimation model selected. In general, several empirical MPTM studies have produced interesting findings. First, the traditional transmission channels, namely the interest rate, asset price and exchange rate channels, are still the core monetary policy transmission channels. The interest rate and asset price channels play a greater role in advanced countries, while monetary policy in EME is generally transmitted through the interest rate and exchange rate channels. Second, in a more developed financial system, monetary policy transmission through the bank lending and capital channels, corporate and household balance sheet channels, as well as the risk-taking channel tend to amplify the effect of monetary policy on the financial and economic cycles. Furthermore, the effect tends to be asymmetric, with a stronger effect felt when the financial system is under pressure. Third, global financial integration facilitates the stronger transmission of monetary policy from advanced countries to EME through global spillovers from the interest rate, exchange rate and asset prices. Liquidity conditions and risk-taking behavior on global financial markets also amplify the effect of monetary policy spillovers.

5.5.1. Monetary Policy Transmission in Advanced Countries In advanced countries, most empirical studies have been conducted for the European Union (EU), with a focus on changes in monetary policy transmission related to the monetary union launched in 1999, increasing financial innovation and the impact of the GFC (Angeloni, Anil, Benoit, & Daniele, 2002; ECB, 2010). Meanwhile, studies of financial transmission in the United States have primarily focused on the economic boom and low interest rates during the 20 years of Great Moderation and the impact thereafter of the GFC. Various studies have tried to capture and reveal financial system behavior in terms of amplifying and accelerating the financial and economic cycles, which often culminates in a crisis. 5.5.1.1. Monetary Policy Transmission in the EU.  Unification of the euro currency had an immediate impact on monetary policy transmission, with the loss of the exchange rate channel between countries in the EU. The economic and monetary union led to a uniform exchange rate channel between EU member states (Boivin, Giannoni, & Mojon, 2008). Furthermore, eliminating currency risk in the euro area reduced the cost of trade transactions and increased capital market integration. Nevertheless, although the benefits of the economic and monetary

142    Central Bank Policy union depended on country-specific conditions, several studies demonstrated a 5–10% increase in trade after unification of the euro currency (Baldwin, DiNino, Fontagné, De Santis, & Taglioni, 2008). Bank holdings and cross-border transactions also increased significantly after currency unification (Kalemli-Ozcan, Papaioannou, & Peydró, 2009). The centralization of monetary policy at the European Central Bank (ECB) brought stronger credibility to the new regime and a clear mandate on price stability. The immediate positive effects included controlled inflation and anchored expectations to the targets set. This was evidenced by long-term bond yields and surveys concerning inflation expectations (Beechey, Johannsen, & Levin, 2007; Ehrmann, Fratzscher, Gurkaynak, & Swanson, 2007). In addition, evidence was put forward that the Phillips curve was becoming flatter, indicating a weaker correlation between inflation and the output gap (Calza, 2008). This was also linked, however, to more credible monetary policy under the ECB in terms of anchoring expectations. The effects on wages and prices were also more controlled. In general, central bank credibility benefits monetary policy transmission in terms of maintained price stability, which drives real economic activities (Erceg & Levin, 2003; Paries & Moyen, 2009). Several empirical studies have produced differing results concerning the effect of monetary unification under the ECB on output and inflation. Meanwhile, the VAR model developed by Gerke, Weber, and Worms (2009) did not find significant differences between the periods before and after currency unification. Cecioni and Neri (2010), using a DSGE model, showed that monetary policy was more effective in terms of economic stabilization by reducing nominal rigidity and anchoring inflation. Angeloni, Kashyap, Mojon, and Terlizzese (2003) as well as Angeloni and Ehrmann (2003) showed that monetary policy transmission through the interest rate channel was still relevant, with a significant impact on output after around one year. Nonetheless, monetary policy transmission through the financial system began to play a more important role through the bank lending channel and innovation on financial markets. The importance of financial intermediation in terms of monetary policy transmission was also stressed by Beck, Colciago, and Pfajfar (2014). Monetary policy transmission in the EU has also been influenced by financial innovation since currency unification. Asset securitization increased, including bank loans, and banks could easily obtain non-term deposit funds by issuing bonds. The ease of funding drove robust credit expansion in the EU, which was also facilitated by working around the capital constraints through derivative instruments such as CDS to contain the interest rate and credit risks. Consequently, financial innovation tended to strengthen non-loan financing in the economy, while bank lending was influenced more by the bank balance sheet channel rather than the interest rate channel (Altunbas, Gambacorta, & Marquéz-Ibañez, 2009; Loutskina & Straham, 2006). This phenomenon was also bolstered by monetary policy credibility, with low inflation and interest rates, coupled with increasing financial integration in the euro area. On the other hand, however, greater dependence on wholesale funding from the capital market with financial innovation rather than traditional retail funding

Monetary Policy Transmission Mechanism     143 also exposed the banks to financial market dynamics. This occurred, specifically, when the markets were experiencing financial distress. Bank difficulties securing wholesale funding exacerbated liquidity condition, and then lending, and even spurred a crisis in the financial system (Hempell & Sorensen, 2009). Conditions were then compounded by capital constraints, which could no longer be overcome under financial distress because of difficulties issuing bonds and/or hedging against the risks. Several studies have shown banks with limited capital undertaking less lending when the economy moderates (Altunbas, de Bondt, & MarquézIbañez, 2004; Maddaloni & Peydro, 2010). Financial innovation also expanded the role of the risk-taking channel in the banking sector in response to monetary policy, but the empirical evidence for this transmission channel is difficult to prove in the EU and the United States (Altunbas, Gambacorta, & Marquéz-Ibañez, 2010). As mentioned previously, low shortterm interest rates would encourage risk-taking behavior in the banking industry in terms of lending quantity (volume and value of approved loan applications) as well as prices (low lending rates). Jimenez, Ongena, Peydro, and Saurina (2012), for example, using credit registry data from Spain showed that banks with lower capital and liquidity ratios engaged in riskier lending during normal periods and were, therefore, the hardest hit during a crisis episode. The risk-taking channel has been proven to be stronger during normal periods because of low short-term interest rates and the proliferation of financial innovation (Maddaloni & Peydro, 2010). The rapid increase of asset securitization and ease at which risk was transferred through derivative instruments during the decade before the GFC, exacerbated risk-taking behavior in the banking sector with looser credit standards and oversight. 5.5.1.2. Monetary Policy Transmission in the United States. Bernanke and Blinder (1992) were among the first to apply a VAR model to map monetary policy transmission channels in the United States, revealing three important findings. First, the federal funds rate (FFR) is a solid measure of the effect of monetary policy. Second, the effect of the FFR on real economic variables (inflation and GDP) is more dominant than money supply, T-Bill rates, and bond yields. Third, monetary policy affects bank asset composition and, thus, lending to the real economy. Follow-up empirical studies showed that the effectiveness of interest rate transmission fades with financial innovation and monetary policy credibility to form expectations in the real economy (Kuttner & Mosser, 2002). The bank lending and balance sheet channels become more important, however, in terms of monetary policy transmission (Boivin et al., 2010; Endut, Morley, & Tien, 2013). Boivin et al. (2010) also stressed that the core channels of monetary policy transmission through short-term interest rates to asset prices and the exchange rate remain an important element of structural macroeconomic models through to the latest DSGE models. The impact of the interest rate on investment is generally modeled through the user cost of capital in the macroeconomic structural model or through Tobin’s q in a DSGE model. Nonetheless, the empirical findings indicate that investment elasticity to the user cost of capital is relatively small, at around minus 2–10%, thus several economists (such as Bernanke & Gertler, 1995) questioned the importance of this transmission channel. The effect of monetary

144    Central Bank Policy policy on consumption in a macroeconomic model is generally modeled through the wealth effect with a comparatively small elasticity at around 3–4% for property and stocks. Meanwhile, Case and Shiller (2003) showed that rising asset prices, such as property, had a significant impact on investment and consumption. The importance of the housing market, in terms of transmitting monetary policy to the real sector, including during the US housing crisis in 2007/08, was also shown by Mishkin (2007). The latest empirical studies have tended to focus on the role of the banking industry in terms of the MPTM by separating the influence of supply and demand for loans. On the supply side, the bank funding structure responds differently to monetary policy. For example, using US bank balance sheet microdata, Kashyap and Stein (2000) showed that the credit portfolios of small and illiquid banks were most affected by changes in monetary policy. Similarly, banks with a small ratio of capital to assets and banks that have not yet gone public were also more affected by monetary policy (Kishan & Opiele, 2000). On the supply side, several studies have shown diverse impacts of changes in monetary policy on the corporate sector. Gertler and Gilchrist (1994) found that investment at small corporations was more susceptible to monetary policy due to greater dependence on bank loans. Using US corporate microdata for the period from 2003 to 2008, Ippolito, Ozdagliy, and Perez (2015) showed monetary policy transmission to corporate balance sheets through bank lending using floating interest rates. Companies more dependent on bank loans and not engaged in hedging activity, and were the most sensitive to the response of stock prices, selling, liquidity, inventory, and investment to changes in monetary policy. Meanwhile, Beck et al. (2012) demonstrated that monetary policy also influences aggregate demand through household balance sheets, especially those reliant on bank loans. These studies serve to confirm the findings of Ciccarelli, Maddaloni, and Peydro (2010, 2013) for the case of the EU, who found a stronger influence in the balance sheet channel during normal periods, while the bank lending channel was more influential during periods of financial system distress. Like conditions in the EU, financial innovation also plays an important role in the MPTM through the risk-taking channel in the United States (Smets, 2013). Angeloni et al. (2011) showed that expansive monetary policy increases leverage in the banking sector and, therefore, credit exposure. Securitization tends to increase incentives for risk-taking at the banks, facilitated by low interest rates and the proliferation of derivative instruments. Norden, Buston, and Wagner (2012) showed that banks transfer risk with derivative instruments to borrower institutions. Meanwhile, Buch, Eickmeier, and Prieto (2013), using data from the Fed’s credit survey for the period from 1997 to 2008 showed that small banks increase credit exposure after monetary policy easing, while big banks extend riskier new loans without adjusting the composition of their credit portfolios. These empirical findings are consistent with the study by Dell’Ariccia, Laeven, and Suarez (2016), who also showed that a lower policy rate increased risk-taking behavior in terms of lending. Nevertheless, differing from the findings of Jimenez, Ongena, Peydro, and Saurina (2012) for the case of Spain, this study showed that

Monetary Policy Transmission Mechanism     145 securitization and derivatives allowed banks with a large capital base to transfer the risk to other financial players, while banks with a small capital base tended to face credit exposure. Several studies also investigated the effectiveness of monetary policy during a crisis period. Theoretically, monetary policy can be effective during a financial crisis if able to overcome financial market failures, for instance, by easing the credit constraints or restoring confidence. If not, deleveraging and uncertainty would undermine monetary policy effectiveness in terms of driving the economic recovery. Using data from the 1980s, Jannsen, Potjagailo, and Wolters (2014) proved that expansive monetary policy could drive output during a recession due to a financial crisis. In contrast, the effect of monetary policy on output fades during the crisis recovery period. Dalhaus (2014) also showed that monetary policy is more effective and persistent during a period of financial distress in terms of the macroeconomic impact on output, consumption, and investment compared to normal periods. The impact is transmitted through the bank lending channel, as confirmed by other studies (Gambacorta & Marques-Ibanez, 2011).

5.5.2. Monetary Policy Transmission in EMEs Studies have also been conducted concerning the MPTM in EME. Mohanty and Turner (2008), for example, reviewed MPTM in EME based on previous research by Kamin et al. (1998), considering several fundamental changes that had occurred, namely greater monetary policy independence and credibility with a strong emphasis on inflation control, more developed financial markets and an economic structure that had changed toward freer international trade and investment. Thereafter, Mohanty and Rishah (2016) analyzed the impact of the GFC on MPTM in EME. 5.5.2.1. Interest Rate Channel, Asset Price Channel, Exchange Rate Channel, and Expectations Channel.  A study by Moreno (2008) showed that, in many EME, the effect of the policy rate on deposit and lending rates is generally stronger and more persistent than that of bond yields. The effect of the monetary policy rate on long-term bond yields tends to be temporary. In EMEs with strong demand for external financing, such as in Latin America, monetary policy could influence the inflation risk premium on debt denominated in the domestic currency and even the country’s risk premium. The asset price channel is also becoming more significant in the Asian region, particularly in terms of property prices. In several countries, including China, Hong Kong and Singapore, property prices have generally mirrored bank lending since 2005. In fact, in Hong Kong, the interest rate channel through property prices has a stronger impact on inflation than household consumption and net worth. In Southeast Asia, asset price sensitivity to the interest rate has also changed. In South Korea, house prices have become more sensitive to changes in monetary policy and bank lending since the Asian crisis. In Singapore, the effect of the interest rate on the property market cycle plays a significant role in terms of the consumption cycle, while stock prices do not. Oppositely, in Thailand, the interest rate has more of an impact on stock prices than property prices.

146    Central Bank Policy The influence of the interest rate on the exchange rate has also evolved. A low and stable risk premium due to sound macroeconomic dynamics would elicit a more predictable exchange rate response to the monetary policy rate. In general, an unexpected 25 bps hike to the policy rate would trigger immediate exchange rate appreciation of around 0.5–1.0%. Similarly, the impact of the exchange rate on inflation has also been observed to fade in many EME (Mohanty & Klau, 2008), which is linked to more credible monetary policy in terms of anchoring inflation expectations. Nevertheless, more open trade has amplified the effect of the ToT on the real exchange rate in the medium-long term due to a change in the composition of trade from primary goods to manufacturing goods. Consequently, the central bank is required to monitor structural changes such as these when considering the real exchange rate underlying monetary policy. The expectations channel has also become more effective. In many EME, including the Czech Republic, México, and South Africa, private Southeast inflation expectations tend to hover around the inflation target of the central bank. The financial markets also react more strongly to policy rate announcements, as is the case in Indonesia, Thailand, and the Philippines. In fact, in Thailand, the housing market and bond market respond faster to changes in monetary policy proceeded by changes in housing loan rates than by changes in the central bank’s policy rate. With financial markets that are becoming better equipped to overcome changes in monetary policy and with more anchored inflation expectations, the need to adjust the policy rate has dwindled. In other words, with stronger monetary policy credibility, the financial markets often do the central bank’s work. The effect of monetary policy on output in many EME is relatively small, at 1–8%, and temporary, around one to two years (Mohanty & Turner, 2008). This provides compelling evidence for the neutrality of money in the long term. In contrast, the effect of monetary policy on inflation has increased but to a different extent in each respective country. In Indonesia, México, and South Africa, the inflation response to monetary policy has increased from around 2–5% in the period from 1998 to 2003 to around 10–30% from 2004 to 2008. On the other hand, the corresponding response in India, South Korea, Thailand, Chile, the Czech Republic, and Poland has remained relatively unchanged at around 2–3%. The empirical findings also show lower variability in terms of inflation and output, which is consistent with monetary policy credibility to control inflation and, therefore, reduce the variability in real output. More anchored expectations also play an important role. 5.5.2.2. Credit and Balance Sheet Channels. Dominant banking sectors in many EME mean that the credit channel plays an important role in terms of influencing investment. The degree of influence varies, however, namely that it is stronger in Latin America and Eastern Europe than in Asia, but the experience of each individual country also varies. In Asia, more developed financial markets, as alternative investment financing, do not always weaken the bank lending channel. In the Philippines, for instance, although the commercial paper market has flourished recently and many banks have committed a credit ceiling to corporate customers, the effect of monetary policy through the bank lending channel

Monetary Policy Transmission Mechanism     147 remains important due to the size of NPL in the banking system. In Thailand, post-Asian crisis financial diversification has dampened the response of inflation and output to bank lending. This is also the case in Singapore, where small and medium enterprises are increasingly issuing stocks and bonds for financing. Meanwhile, in China and India, the bank lending channel remains dominant. In China, monetary policy is directed toward controlling the quantity of money supply and there remain policy controls on credit, therefore, monetary policy influences demand primarily through changes in credit supply. In India, studies have shown that small banks tend to reduce credit supply more than big banks during a period of monetary policy tightening. The household balance sheet channel is now playing a more important role in line with the proliferation of consumer loans for the nascent middle class in EME. The surge in leverage has exacerbated the balance sheet constraints in terms of household consumption decisionmaking and, therefore, the temporal substitution effect of monetary policy has gained impetus. Furthermore, changes on the household balance sheet could amplify the wealth effect of monetary policy, mainly through home-ownership. The wealth effect is reinforced when using the property as loan collateral. Another implication regarding the influence of the balance sheet is linked to the effect of cash flow from monetary policy on consumption and investment through the nominal interest rate, scale of financial assets and liabilities or household loan agreements. In general, household loans are short term and, therefore, more sensitive to the monetary policy rate, moreover when the loan applies a variable interest rate. In addition, the interest rate on housing loans in many countries is generally tied to the base lending rate or monetary policy rate. With the dominance of bank loans in corporate financing, the corporate balance sheet channel also plays a salient role. Furthermore, corporate leverage is more significant in EME than in advanced countries. In Asia, for instance, leverage averaged around 38% during the period from 1993 to 2003, exceeding the 24% reported in the United States, Europe, and Japan. Higher leverage in Asia is due to the comparatively low market value of corporations, which compels the corporate sector to finance investment using debt rather than issuing shares. Nonetheless, in some EMEs, including Indonesia, a large portion of firms rely on internal financing, which implies low leverage. The implications of corporate financial conditions on MPTMs are difficult to gauge. Low levels of debt ease corporate cash flow constraints and, therefore, undermine the investment response to monetary policy. A sound corporate balance sheet, however, would also undermine the role of financial accelerators in the economy. On one hand, financial market development with alternative sources of financing to bank loans tends to weaken the balance sheet channel in the transmission of monetary policy. The propagation of derivative instruments has facilitated hedging of interest rate and currency risks in the corporate sector, thus unwinding balance sheet fragilities to changes in monetary policy. Financial market liberalization also opens corporate access, especially large corporations, to international sources of financing, including debt as well as issuances of stocks and bonds. On the other hand, a better capital market function could reinforce the

148    Central Bank Policy influence of the monetary policy rate on the prices of various financial assets and, therefore, strengthen the direct cost impact of changes in monetary policy on investment. Similarly, a more liquid market and active trading of securities strengthens the impact of changes in the interest rate on the market valuation of corporate balance sheets, while investment spending could become more responsive to the monetary policy rate. Changes in expectations of the monetary policy rate are also now playing a more important role in terms of corporate financial conditions than previously. As mentioned above, greater banking system soundness and efficiency in many EMEs have influenced MPTMs. On one hand, such conditions have strengthened the direct influence of monetary policy. In Malaysia, for instance, the longterm elasticity of the money market rate to the lending rate has continued to rise, increasing from 0.3 in 1998 to 0.5 in 2005, in line with increasing competition and banking system efficiency (Ooi, 2008). On the other hand, less bank balance sheet vulnerability has also reduced non-price distortions to credit supply and, therefore, undermined the bank lending channel in the MPTM of many countries. A solid capital base and greater bank access to alternative sources of funding through certificates of deposit and long-term bonds, such as the case in Chile, have had a similar effect in terms of unwinding the funding constraints in the banking sector, particularly during a period of tighter monetary policy. A change in bank balance sheet conditions could intensify market risk and the effect of monetary policy on risk-taking behavior and market exposure. In several EME, the share of variable interest rate loans exceeds the composition of term deposits. The impact of the monetary policy rate is stronger, therefore interest rate exposure on the bank balance sheet increases if the average funding rate is not adequately adjusted to the change in the monetary policy rate. In many ways, such market risk can be mitigated through an appropriate hedging strategy. In South Africa, for instance, banks tend to transfer the market risk to their customers by offering variable rate term deposits. In addition to market risk, the banking sector also faces maturity risk because the average maturity of term deposits is shorter than the maturity of loans, such is the case in Latin America and Indonesia, where the average maturity if term deposits is less than six months, Another important source of exposure from changes in the monetary policy rate could appear on the banks’ investment portfolio. During the decade prior to the global crisis, commercial bank investments in government and central bank securities increased in terms of share to total assets. In a number of countries, that trend was linked to central bank intervention in order to prevent exchange rate appreciation by selling government securities or central bank issued securities to absorb the excess liquidity. An increase of government and central bank securities in the banks’ asset portfolios would increase exposure to market price valuations on the balance sheet, thus exacerbating the financial accelerators. Banks could aggressively expand lending during a period of monetary policy easing because of elevated asset prices and the profits from trading bonds. In contrast, during a period of tighter monetary policy, the banks tend to ration credit in order to compensate for the decline in the prices of their financial assets. Another implication is the impact on the central bank in terms of setting the monetary policy rate

Monetary Policy Transmission Mechanism     149 because of paying due consideration to the impact on overall financial system stability. That observation was more visible during the GFC, when global uncertainty and spillovers besieged the financial systems of many EMEs. 5.5.2.3. Monetary Policy Transmission in the Post-GFC Era.  The GFC precipitated an important MPTM evolution in EMEs due to three salient factors, namely the relative role of banks on the bond market, money market globalization, and low long-term interest rates (Mohanty & Rishab, 2016). From 2004 to 2013, the ratio of total economic financing in the form of domestic and offshore bank loans and bond issuances to GDP in EMEs increased rapidly and then accelerated again in the wake of the global crisis due to low interest rates and abundant excess liquidity on global financial markets originating from the expansive monetary policies adopted in the United States and Europe. Growth was even more rapid in EME with perfect capital mobility and/or stable exchange rates compared to imperfect capital mobility and/or a flexible exchange rate regime. Hong Kong is a good example, although China has also experienced similar dynamics due to the attractiveness of its economy. Similarly, South Korea, Malaysia, and Singapore all noted an uptick in total financing after the GFC. Although bank credit has remained dominant, its share of total economic financing has decline, especially in Asia. China, for instance, experienced a 21% drop in the share of credit from 2004 to 2013, a trend that has been mirrored throughout Asia. How do global interest rates and excess liquidity affect the interest rate, exchange rate and bank lending transmission channels in EME? Several studies have shown that low long-term interest rates on global financial markets are affecting long-term interest rates in EME rather than short-term interest rates (King & Low, 2014). Furthermore, Miyajima, Mohanty, and Chan (2015) found that the response of the benchmark 10-year bond yield in 11 EME to a 100 bps hike in the 10-year US T-Bill yield was to increase from just 31–53 bps after the GFC. During periods of global financial market turbulence, however, such as the Fed’s Taper Tantrum in May and June 2013, the effect increased to more than 100 bps. Using more recent data, Sobrun and Turner (2015a) reported similar findings, namely that the weak correlation between bond yields in EME and United States yields from 2000 to 2004 strengthened and became significant after 2005. In another study, Kharroubi and Zampolli (2015) demonstrated that the long-term bond yield correlation was around 60 bps, while the response of the short-term interest rate was just 20 bps. Sobrun and Turner (2015b) also reported similar findings, noting that, like Miyajima, Mohanty, and Yetman (2014), the effect of the term premium on bond yields was larger than the effect of future expectations of short-term interest rates. How do long-term global interest rates influence bank lending rates in EME? Differing from the impact on bond yields, bank lending rates are not particularly responsive to long-term interest rates. The average response of interest rates on household loans and corporate loans was just 16 and 34 bps, respectively, due to the funding structure of the banking industry, which is typically short term, with a median of just four months in the 11 EME studied at the end of 2013 (Ehlers & Villar, 2015). In addition, the share of variable interest rate term deposits was shown to be relatively high, at around 50–70%, implying that term deposit

150    Central Bank Policy rates are effectively indexed to the policy rate of the central bank. In terms of credit, variable interest rate housing loans account for around 70−99% of total housing loans in Asia, contrasting conditions in Latin America, where fixed-rate loans account for nearly 100% in Brazil and around 70–96% in Argentina, Chile, Colombia, and México. Changes have also occurred in the effect of global spillovers on the exchange rate channel. Prior to the global crisis, monetary policy independence and credibility, as the ITF gained popularity, reduced the impact of the exchange rate on inflation. Previous crisis experience also helped to control currency risk and maturity risk in the banking and corporate sectors in terms of their external debt. Nevertheless, the global crisis reaffirmed the need for central banks in EME to stabilize exchange rates. In addition to the effect of global spillovers, increasing exchange rate volatility was also linked to the significant spike in external debt at non-financial institutions in EME (Chui, Kuruc, & Turner, 2016). Exchange rate fluctuations had a notable impact on economic and financial conditions in many Latin American countries, including Brazil, as well as in several Asian countries, including India and Indonesia, and in Eastern European countries, such as Turkey and Hungary, but with a smaller intensity. Another factor affecting exchange rate volatility was commodity prices, particularly in net exporters (BIS, 2014). In addition, the correlation between exchange rates in EMEs and global risk perception indicators, such as the volatility index (VIX) market volatility index, increased significantly after the GFC (Rajan, 2014). In fact, Miranda-Agrippino and Rey (2013) as well as Rey (2013) found that changes in the perception of global investors had created global convergence in terms of exchange rate and asset price fluctuations. One effect of exchange rate fluctuations, while external debt is accumulating due to excess global liquidity, is that the credit cycle in EMEs tends to mirror the dollar exchange rate. Bruno and Shin (2014) called this phenomenon “the risktaking channel of the exchange rate” to illustrate the important effect of global financial markets on the transmission of external shocks through the distribution of liquidity on the US money market to the banking systems in EME. USD depreciation improved the balance sheets of banks indebted in USD, therefore, the credit risk faced effectively eased, which prompted further credit growth. In contrast, a period of strong USD appreciation was accompanied by a contraction of the bank balance sheet and widespread USD shortages. Evidence also pointed to the effect of the exchange rate on funding and financing conditions, although without financial imbalances. The mechanism could operate through interactions between the foreign exchange market and bond market, possible caused by a change in the portfolio strategies of speculative investors and professional asset managers (Feroli, Kashyap, Schoenholtz, & Shin, 2014). Turner (2012) showed different returns on yields in EME between those hedged and those not. The lack of UIP implied that non-resident investors affect the risk premium and cause large fluctuations in domestic monetary conditions. When the exchange rate appreciates, investors engage in carry trade on the bond market for profit taking due to future exchange rate appreciation, which reduces the risk premium and bond yields in EMEs to very low levels. In contrast, under conditions of market distress, exchange rate depreciation and uncertainty would increase, hence

Monetary Policy Transmission Mechanism     151 non-resident investors would withdraw, pushing up yields and tightening financing conditions further. The bank lending channel is also influenced by global spillovers. One effect is through banking globalization, which has eroded the link between monetary policy and credit (Cettorelli & Goldberg, 2012). The ability of global banks to efficiently lend across borders through their networks has undermined the impact of the domestic policy rate on credit, implying that credit conditions in EME are more vulnerable, especially to operational target changes in US monetary policy. Cettorelli and Goldberg (2012) reported that a 100 bps hike in the FFR would reduce commercial bank lending in the United States to EME by around 2.2%. In addition, a larger USD debt burden in EMEs has created a tighter correlation between domestic credit conditions and the availability of dollar liquidity. USD scarcity after the Lehman Brothers collapsed in 2008 spurred intense deleveraging pressures on EME (McGuire & von Peter, 2009). Consequently, as corroborated by Borio et al. (2011), the sharp spike in USD obligations in EME caused the EME credit cycle to synchronize more with the global credit cycle. During an economic boom, global credit tends to increase more rapidly than credit overall, with banks seeking large-scale funding. In contrast, a higher US policy rate and lower USD debt burden undermined lending, which spread to EME. Increased risk-taking behavior made the bank lending channel more important to financial system stability in many EME. Unfortunately, differing from studies in advanced countries, research into the risk-taking channel in EMEs remains limited due to a lack of individual borrower and lender data. Nonetheless, studies using aggregate data have provided some preliminary indications. Using a crossborder panel model, for example, Kohlscheen and Rungcharoenkitkul (2015) found that external factors, such as the USD exchange rate and US capital market volatility (VIX Index), were significant drivers of credit growth in EME after the GFC compared to pre-crisis conditions. Consistent with the risk-taking and exchange rate channels, this study showed that 10% exchange rate appreciation in EME against the USD would boost credit growth in EMEs by 85 bps in the near term and by 135 bps in the long term.

5.5.3. Monetary Policy Transmission in Indonesia The first general study on MPTMs conducted by BI was in 2001 and documented by Warjiyo and Agung (2002). The study focused on empirically mapping the interest rate channel, bank lending channel, corporate balance sheet channel, exchange rate channel, asset price channel, and expectations channels. The study covered the period from 1990 to 2001 and was divided into the periods before and after the Asian Financial Crisis of 1997/98. The research methodology used a VAR model, structural equations and surveys. In general, the results provided invaluable empirical findings on MPTM in Indonesia and showed how the mechanism evolved due to the Asian Financial Crisis of 1997/98. Several empirical studies were subsequently conducted by BI and academia regarding specific transmission channels, such as the interest rate, exchange rate, lending, and balance sheet

152    Central Bank Policy channels. In fact, several studies also investigated monetary policy transmission through the risk-taking channel and the transmission of BI’s policy mix. 5.5.3.1. Interest Rate, Exchange Rate, Asset Price, and Expectations Channels.  A study by Kusmiarso et al. (2002) documented the growing role of the interest rate transmission channel in Indonesia. Specifically, using a money market structural model, the study showed that the interest rate of BI Certificates (SBI) and liquidity conditions in the banking sector influence interbank rates (PUAB), term deposit rates, and lending rates. Furthermore, changes to the interest rates would also affect investment and consumption through the cost of capital as well as the substitution effect and income. Investment growth was shown to be very responsive to lending rates, while consumption growth was responsive to term deposit rates. In another study, using data from January 1988 to June 1997 and January 1999 to December 2010, Tai, Sek, and Har (2012) demonstrated an asymmetric response, namely that the term deposit rate was more responsive than the lending rate to BI’s policy rate after the Asian Financial Crisis in 1997/98. A study by Natsir (2008) also concluded that transmission through the interest rate channel was effective in terms of achieving the final target of monetary policy. Using a VAR model and data for the period 1990:2 to 2007:1, Natsir showed that the SBI rate, through the interbank rate, could explain around 63.1% of the variation in inflation with a lag of 10 quarters. Concerning the exchange rate channel, Siswanto, Kurniati, Gunawan, and Binhadi (2002) showed that, with a more flexible exchange rate system after the Asian Crisis, exchange rate transmission increased. The direct impact of the exchange rate on inflation (through changes in the import prices of goods) was stronger and instant from the first month compared to the indirect impact (through aggregate demand) that had a lag of two months. This empirical finding was linked to the magnitude of the risk premium due to high country risk and suboptimal market mechanisms due to problems in the banking industry at that time. The findings were also confirmed by surveys of banks, businesses and households, which showed that non-economic factors, limited supply compared to demand for foreign exchange and global exchange rate developments were the three main drivers of rupiah exchange rate fluctuations. Most banks regarded foreign exchange market intervention as BI’s most powerful monetary policy instrument in terms of influencing the exchange rate. Only a few banks cited the SBI rate as having an influence over the exchange rate. With monetary policy credibility at BI since implementation of ITF in 2005, coupled with increased macroeconomic stability, subsequent studies have shown that the effect of the exchange rate on inflation is fading. Kuncoro (2015), for example, using monthly data from 2003 to 2013 proved that with ITF implementation, the effect of 10% rupiah depreciation on the inflation of import prices and producer prices declined respectively from 6% and 3% to 3% and 1.5%. That estimate is slightly higher than BI’s own estimates of the effect on inflation, namely around 0.7–1.2% for each 10% of rupiah depreciation. The diminishing effect of the exchange rate on inflation after ITF implementation was also evidenced by Siregar and Goo (2009).

Monetary Policy Transmission Mechanism     153 A study on the asset price channel was conducted by Idris, Yanuarti, Iskandar, and Darsono (2001), which focused on share prices due to the lack of data on property and land prices. In general, the study concluded that stock prices were comparatively weak in terms of transmitting monetary policy to the real sector. Although monetary policy could influence share prices and the size of the financial asset portfolio, the pass-through effect to inflation was weak. In other words, changes in share prices have not displayed evidence of the wealth effect in the economy. This may be due to the relatively small share of stocks (only around 5%) in the investment portfolio compared to alternative sources of funding, especially deposits held at a bank as well as property and land assets. Meanwhile, Yuniarsa and Chang (2015) used monthly data from 2000 to 2013 to show that the correlation between the interest rate, exchange rate volatility, and stock prices had increased since the GFC. Wuryandani, Ikram, and Handayani (2001) focused their study on the inflation expectations channel using inflation expectations data from the BI Business Survey. The study concluded that inflation expectations are affected more by exchange rate fluctuations, inflation in the previous period (inertia) and the interest rate.10 The empirical evidence was further substantiated by a survey of banks, businesses, and households. Inflation in the previous period was again shown to influence the formation of inflation expectations, evidencing the importance of enhancing monetary policy effectiveness in terms of controlling inflation. On the other hand, the effect of the exchange rate on the formation of inflation expectations tends to be asymmetric. In the corporate sector, there is downward price rigidity when firms set prices, implying that businesses are generally reluctant to lower prices when the rupiah experiences appreciation. In contrast, exchange rate depreciation beyond a given threshold would force companies to raise prices. Nevertheless, a subsequent empirical study on the inflation expectations channel by Tjahjono, Harmanta, and Nugroho (2012) produced some interesting findings after implementation of the ITF in Indonesia. First, the expectations indicator contained in the Consensus Forecast (CF) was superior to other measures of inflation expectations, including the survey results. In addition to the availability of month data, the CF indicator provided inflation expectations for the upcoming one to four quarters. Second, the study did not reveal heterogeneous inflation expectations between economic players and professional projections for the upcoming one month and, therefore, the inflation expectations could be considered homogeneous. Third, causality analysis showed that consumer inflation expectations were influenced by retailers’ inflation expectations, while professional inflation expectations were more affected by wholesalers’ inflation expectations.

10

Testing was also conducted using other indicators to measure inflation expectations, for example, the Consumer Expectations Survey performed by BI as well as Fisher Theory. The results confirmed that the inflation expectations indicator contained in the Business Survey behaved consistently.

154    Central Bank Policy Concerning the transmission of the GFC, Raz, Indra, Artikasih, & Citra (2012) performed an empirical study comparing the impact on economic growth of the Asian Financial Crisis of 1997/98 and the GFC. In general, the study showed that East Asian countries were more resilient to the GFC than the Asian financial crisis of 1997/98, primarily due to stronger economic fundamentals, a more prudent banking sector and more credible monetary policy. Banking sector reforms instituted in the wake of the Asian financial crisis of 1997/98 have strengthened bank capital and oversight by the regulator. Corporate transparency has also played an important role in terms of strengthening economic fundamentals, while larger reserve assets have contained global crisis spillovers. Meanwhile, Nurhidayah, Kaluge, and Pratomo (2014) compared the roles of the BI Rate and exchange rate on inflation and economic growth in Indonesia prior to and after the GFC. Using monthly data from 2005:10 to 2012:12 and a Vector Error Correction Model (VECM), the study showed that the contribution of the BI rate, measured using variance decomposition, had declined from 22.3% to 13.6% before and after the global crisis. Conversely, the contribution of the exchange rate to inflation had increased from 0.5% to 1.95% after the GFC. 5.5.3.2. Bank Lending and Balance Sheet Channels. Previous research provided complete empirical evidence concerning the bank lending channel in Indonesia (Agung, Morena, Pramono, and Prastowo, 2002a, 2002b). Three methods were used in the study. First, a VAR model was used to analyze the impact of changes in the monetary policy rate on bank behavior concerning term deposits, loans and securities. Second, the demand and supply of loans was analyzed to determine whether credit market imbalances were more attributable to credit supply as indicated by the bank lending channel. Third, the panel data of individual banks was investigated to discover whether monetary policy had differing impacts on the different bank characteristics, reviewed from the perspective of capital strength and asset size. The evidence from aggregate data analysis confirmed that monetary policy influences bank lending with a lag because banks tend to avoid reducing the volume of term deposits rather than selling securities to meet their credit obligations. Further empirical evidence showed that after monetary policy was adjusted, flight to quality was observed from national banks to foreign banks and joint venture banks. The phenomenon explains why loans from foreign and joint venture banks are not sensitive to changes in monetary policy compared to national private banks. Furthermore, disaggregated bank lending data to the individual bank and corporate level showed that corporate loans were relatively less sensitive to changes in monetary policy, while individual loans experienced a sharp drop off. Such developments can be explained by the phenomenon of flight to quality in terms of borrower selection by the banks if tighter monetary policy is perused, namely that only viable firms would be eligible to receive a loan, while less feasible businesses and individuals would face credit rationing. In terms of capital-based bank groups, the study showed that the impact of monetary policy was more pronounced for banks with a relatively small capital base. Furthermore, time series data and panel data analysis in the study demonstrated a more asymmetric impact of monetary policy during a recession

Monetary Policy Transmission Mechanism     155 compared to boom conditions. Less effective monetary policy on bank lending during the period prior to the Asian Financial Crisis of 1997/98 was caused by the banks’ ability to access offshore funds. After the Asian Financial Crisis of 1997/98, however, characterized by less bank capital and heightened credit risk, a higher policy rate due to a tighter monetary policy stance, increased the PoD and, therefore, the banks were more reluctant to extend loans. Meanwhile, an empirical study on the corporate balance sheet channel focused on two salient questions. First, which balance sheet conditions play an important role in influencing corporate investment decisions? Second, how does monetary policy affect corporate balance sheets and their investment decisions? In the case of Indonesia, the previous study showed that the importance of balance sheet dynamics, especially. cash flow and leverage, in terms of influencing corporate investment decisions, while the impact was more pronounced for small business rather than large corporations (Agung et al., 2002a, 2002b). Investment sensitivity to corporate balance sheet dynamics actually increased during periods of contractionary monetary policy compared to expansionary monetary policy. The empirical evidence pointed to the importance of the corporate balance sheet channel in the transmission of monetary policy in Indonesia. Several other studies have also investigated the bank lending and balance sheet channels. Deriantino (2013), for example, researched the impact of bank competition on monetary policy transmission effectiveness using panel data for 95 banks from 2001 to 2012. The empirical results showed that more competitive banks tended to be less responsive in terms of adjusting their lending rates after a change in the monetary policy rate. Although the elasticity of term deposit rates and lending rates to the BI rate were 0.52% and 0.30%, respectively, the level of bank competition did not significantly change monetary policy transmission. Similar conditions were found concerning bank capital. Meanwhile, Yarasevika, Tongato, and Muthia (2015) showed that the bank lending channel was influenced more by past economic growth and credit behavior rather than lending rates and reserve requirements. In a study using monthly microdata for foreign banks from 2002 to 2007, Fazzaalloh and Sasongko (2014) showed that foreign banks with comparatively smaller total assets or capital reacted more strongly in response to changes in monetary policy when lending compared to the big banks. Concerning the transmission of the Asian Financial Crisis of 1997/98 to credit, Silalahi, Wibowo, and Nurliana (2012) studied the impact of foreign bank loans in Indonesia by combining macro push and pull factors of foreign capital flows. In general, the study showed that the global crisis had an impact on foreign bank lending in Indonesia, and that the impact was more significant on foreign bank branches compared to joint venture banks. In addition, foreign bank loans were also influenced by push and pull factors of foreign capital flows, such as economic growth, risk factors as well as domestic and global liquidity dynamics. From a micro perspective, foreign bank loans were also affected by balance sheet conditions and the banks’ asset portfolios. 5.5.3.3. Transmission of Risk-taking Behavior and the Central Bank Policy Mix.  Research performed by Satria and Juhro (2011) was the first empirical

156    Central Bank Policy study concerning the impact of risk-taking behavior on lending by banks in Indonesia. The econometric model used the following error correction model (ECM) specification: ∆Kti = α( Kti−1 − β1GDPt−1 − β2 ) + γ1∆GDPt−1 + γ2 ∆Kti−1 + γ3 At + γ 4 DDt + γ5STANCEt + γ6 (A t ×STANCEt ) + γ7 (DDt ×STANCEt ) + εt

(24)

where Kt = real credit disbursed by the banking sector with the market equilibrium interest rate (investment loans, working capital loans, and consumer loans), GDPt = Real GDP, At = bank portfolio risk perception index, DDt = distance to default, and STANCEt = monetary policy stance (tight or loose). The bank risk perception index is a measure of the risks contained in the asset portfolio as follows: A=



E ( rp ) − rf y ×σ 2p

(25)

where (E(rp) – rf) = the difference between the yield of the portfolio at risk and the yield of the risk-free portfolio, y* = bank total assets at risk (excluding SBI and SUN), and σp2 = asset yield variance. Meanwhile, the risk of bank default was measured using the distance to default of the assets’ market values compared to the minimum capital requirements as proposed by Chan-Lau and Sy (2007) as follows:

DDt =

ln(VA t / X t ) + ( µ − σA2 )T (26) σA T

where VAt = market value of at-risk assets, Xt = total liabilities, λ = (1 − CAR), where CAR is the minimum capital requirements pursuant to banking regulations, μ = average asset value growth, σA = standard deviation of asset value, and T = maturity date of corporate loans. The study produced several interesting findings. The risk-taking channel plays a significant role in terms of monetary policy transmission through the bank lending channel in Indonesia, where credit and risk-taking behavior tend to be procyclical, while monetary policy is countercyclical. Furthermore, the risk perception index of the bank portfolio and risk of default tend to weaken the effectiveness of the monetary policy stance. If a loose monetary policy stance is adopted, the banks are more prudent when lending and, oppositely, the banks are more inclined to take risks when a tighter monetary policy stance is pursued. These findings contradict Taylor (2009) for advanced countries, where bank risktaking behavior tends to exacerbate lending procyclicality in response to monetary policy. One explanation for the case in Indonesia is the banks perhaps detect signals that the economy would be compounded by monetary policy easing, an explanation that requires further empirical evidence. Another interesting empirical study, by Wimanda, Maryaningsih, Nurliana, and Satyanugroho (2014), evaluated transmission of the BI policy mix. As discussed in Chapters 13–15, after the GFC, central banks, including BI,

Monetary Policy Transmission Mechanism     157 implemented a mix of monetary policy foreign capital flow management and macroprudential policy. The study reviewed how the BI policy mix is transmitted to the economy, particularly the final targets of price inflation) stability and financial system stability. In this research, monetary policy and macroprudential policy were analyzed. Empirically, monetary policy was shown to influence inflation and financial system stability with respective lags of 18 and 10 months. All monetary policy transmission channels identified with the interest rate channel were observed to be more dominant in terms of the impact on inflation, while the asset price channel was more dominant in terms of financial system stability. In general, macroprudential policy had a moderate and temporary impact on the intermediate targets but was not shown to effectively influence the final targets of inflation and financial system stability. Monetary policy was also shown to be transmitted through the corporate balance sheet channel, becoming more sensitive after the application of tighter monetary policy, which evidenced greater utilization of internal funds for investment in line with less bank loan availability. Such developments were primarily found at small businesses that typically face financial constraints.

5.6. Concluding Remarks The theoretical and empirical studies elaborated in this chapter revealed several determinants of MPTM dynamics and evolution from one period to another. The first factor is central bank monetary policy credibility and consistency. Experience since the 1990s has shown that monetary policy at many central banks has become more credible, especially since implementation of price stability as the final target through the ITF. Monetary policy was shown to lower inflation in many advanced countries and EMEs and support robust economic growth (Beck et al., 2015). Monetary policy consistency strengthens the effectiveness of the interest rate channel, which is subsequently transmitted to the other channels. In fact, monetary policy credibility, supported by central bank transparency and communication, could successfully anchor inflation expectations and, thus, reduce the interest rate policy response required. The second factor is the condition of the financial system in terms of banking and financial product innovation. The supply of bank loans is not only determined by interest rates and the economic outlook but also by liquidity conditions, the funding structure and strength of the capital base. Therefore, the central bank’s interest rate policy will have different impacts on the economy, in line with the internal conditions of banks and the corporate sector, and an asymmetric impact in terms of looser or tighter monetary policy. Meanwhile, financial product innovation, such as credit securitization and derivative instruments, may unwind internal bank constraints to the supply of credit, especially during an economic upswing. Nevertheless, the simultaneous accumulation of risk in the financial system would also accelerate and could culminate in crisis if the economy moderated. In brief, MPTM analysis should focus on understanding the dynamics of the financial and economic cycles that tend to amplify and accelerate between economic boom and bust.

158    Central Bank Policy The third factor is increasing global financial integration. Foreign capital flows volatility and risk-taking behavior in the economy, and global finance impacts significantly on monetary policy transmission effectiveness in EME with an open economic system and perfect capital mobility, including Indonesia. How spillovers from the United States crisis led to the massive and protracted GFC is the latest tangible example. The impacts of global spillovers are transmitted through the trade channel and financial channel. In terms of the trade channel, economic moderation and lower world trade volume (WTV) undermine external sector performance in the form of declining exports and current account performance, thus exacerbating pressures on macroeconomic stability and economic growth in EMEs. In terms of the financial channel, global spillovers increase exchange rate and foreign capital flow volatility, which complicate monetary policy formulation in EME. Both conduits of global spillovers influence monetary policy transmission effectiveness through nearly all channels, not only the money view channels, such as the interest rate, exchange rate, asset price, and expectations channels, but also the credit view channels, namely the bank lending and balance sheet channels. In fact, global spillovers also affect monetary policy transmission through the risk-taking channel. Those three factors have several salient implications in terms of monetary policy formulation. Monetary policy consistency and credibility are crucial to maintain macroeconomic stability and support economic growth, which not only requires a solid monetary policy framework, as explained in Chapters 6–9 on ITF-based monetary policy, but also institutions that emphasize independence, consistency, and transparency, as described later in Chapters 10–12. Furthermore, the effectiveness of monetary policy formulation and implementation requires comprehensive understanding of the financial system, including the implications of financial frictions, the proliferation of financial product innovation, and spread of risk-taking behavior. The stronger impact of increasing global financial integration on domestic economic and financial dynamics must be well understood and anticipated when formulating monetary policy. Such conditions cannot be overcome merely by strengthening economic fundamentals. The central bank’s policy mix response, in the form of interest rate policy, exchange rate policy, foreign capital flow management, and macroprudential policy, is required to achieve price stability and support maintained financial system stability, as explained in Chapters 13–15.

Part III

Monetary Policy Framework

This page intentionally left blank

Chapter 6

Monetary Policy Strategic Framework 6.1. Introduction Fundamentally, monetary policy encompasses the regulation of money and credit (in terms of volume and/or interest rates) in the economy to achieve the final target of price (and exchange rate) stability and sustainable economic growth. Through implementation, as elaborated in Chapter 3, monetary policy in the long term has a stronger effect on price stability, contrasting the short-term trade-off between inflation and economic growth. In an open economy, as explained in Chapter 4, the ability of monetary policy to achieve the target of domestic price stability and sustainable economic growth shall be affected by exchange rate stability and foreign capital flow mobility. How the monetary policy process influences various economic and financial variables prior to achieving the economic targets can be illustrated as transmission mechanisms, described in Chapter 5. How does the central bank develop a framework to formulate and implement monetary policy in line with the strategic and operational targets? Theoretically, from a monetary policy perspective, this question is fundamentally linked to three aspects, often cited as dualities, namely (1) What is the best policy target: output stability (economic growth) or prices (inflation)? (2) What operational target or policy instruments should be applied to achieve the final target: quantity-based instruments or price-based instruments? (3) What is the best policy response: rules versus discretion? This chapter reviews various issues relating to the first duality, while the second and third dualities are presented in subsequent chapters of the book. In other words, the discussion on the monetary policy strategy framework is inextricably linked to the presence of a final target and the relationship with the policy regime applied. This chapter will review the strategies in five sections. After this introduction, the discussion focuses on conceptual dimensions and theoretical models related to the final policy target and monetary policy regime using a nominal anchor. Several issues concerning the Phillips Curve phenomenon are also elaborated in terms of long-term linkages and the socio-economic costs that arise. In addition, the application of monetary policy strategy frameworks in various countries is also presented from a historical perspective along with policy evolution over time. The concluding remarks at the end of the chapter describe the monetary policy regime trends over the past few years.

Central Bank Policy: Theory and Practice, 161–191 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191010

162    Central Bank Policy

6.2. Conceptual Dimension and Theoretical Models 6.2.1. Final Policy Target: Output Versus Prices The selection of a policy strategy that prioritizes output (economic growth) or price (inflation) stability is a salient central bank issue when analyzing monetary policy that continues to this day.1 The history of monetary policy economics shows that many central banks around the world prioritized output stabilization, especially during the 1970s and 1980s. Beforehand, price stability was the prevailing priority during the 1950s and 1960s under the auspices of the Bretton Woods Agreement. The debate over the past decade, however, has centered on the differences in views between central bankers and academics. In general, central bankers favor price (inflation) stability as the main objective, while academics lean toward inflation and real output stability. In fact, since the beginning of the 1990s, nearly all central banks have adopted price stability as the final target of monetary policy (Bofinger, 2001). The priority between price stability and economic growth as the final targets of monetary policy, similar to general macroeconomic policy, is to improve public welfare (prosperity). Therefore, to evaluate whether a monetary policy is optimal, the main reference point is achieving the desired level of prosperity. A traditional approach to this issue is through the social welfare function. Per the ideal version, the social welfare function is very difficult to evaluate because it depends on the public’s welfare preferences, which are very broad, now and in the future, as well as the alternative policies that are required to achieve maximum welfare. Over time, the approaches used have focused on the simpler objective function, in the form of a macrowelfare function, which links directly to several macroeconomic variables, such as output, employment, and prices/inflation. A macroeconomic welfare function, which plays an important role in monetary economics as a quadratic loss function that must be minimized by the central bank, can be formulated simply based on the trade-off between price-output stabilization as follows:

L=

1 ωp ( πt − π*)2 + ωy ( y − y*)2 (1) 2

The argument is, therefore, inflation deviation (π) from its target (M*) and output deviation (y) from its potential (y*). Fundamentally, determining the quadratic function implies that positive or negative deviation from the target is 1

The implementation of monetary policy strategy, as an integral part of macroeconomic policy, can be achieved using multiple targets, namely to achieve the final targets of sustainable economic growth, full employment, price stability, and BOP equilibrium. In practice, however, the central bank tends to focus more on efforts to achieve dual targets, namely economic growth (output stability) and price stability. This is achieved with firm confidence in the close linkages between the achievement of economic growth and more employment, as well as between price stability and exchange rate appreciation and the pursuit of external balance.

Monetary Policy Strategic Framework     163 equally as harmful to the public. In this case, higher deviation receives a larger weight than lower deviation. Meanwhile, ωp and ωy, respectively, reflect the preferences or weight of achieving the objective based on the final target sought by policymakers, namely prices or output. In addition to that loss function, there is another variation of the loss function, which can be expressed as follows:

1 2 L = [ ( πt + ∆yt ) − ( π + ∆y ) *] (2) 2

The argument of that function is nominal output growth deviation (the sum of inflation and real output growth) from the nominal output growth target. Conceptually, the first loss function was primarily used in monetary policy analysis where the strategy framework/regime was based on inflation targeting, while the second loss function was used based on nominal income targeting. 6.2.1.1. Targeting Long-term Policy Goals.  Theoretically, the benefits of the respective policy strategies could be observed from the perspective of implementing monetary policy in the near term and long term. In this case, the characteristics of the economy in the near term are not assumed to be influenced by demand-side or supply-side shocks, which is consistent with general prevailing theory, namely the long-term neutrality of money. From the literature, Walsh (2001) proposed some interesting findings on the relationships between money supply, inflation, and output, namely that in the long term, the linkages between money supply growth and inflation are perfect, while the linkages between money growth or inflation and real output growth fade toward zero. The findings demonstrate consensus that, in the long term, monetary policy only affects inflation and not real output. Meanwhile, consensus in the literature regarding the short-term influence of money shows that a monetary policy shock would prompt a hump-shaped distribution in terms of real economic activities, implying that, in the near term, monetary policy affects real economic activity but the influence fades over time. Therefore, the choice of low inflation (price stability) as the single target of monetary policy is fundamentally a feasible policy goal in the long term as well. The neutrality of money hypothesis is, therefore, less applicable if output growth or nominal income are defined as the long-term targets of monetary policy, considering nominal income is affected by real factors, hence monetary policy cannot be applied to overcome the problem of dwindling growth in the long term. In addition, it would not make sense to compensate the decline in economic growth by increasing inflation in the long run. Consequently, when determining the long-term monetary policy target, a strategy to achieve price stability would be considered more appropriate than a strategy targeting output/nominal income stability. 6.2.1.2. Targeting Short-term Policy Goals.  In the near term, the monetary policy target selected, be it price stability or output stability, will be affected by shocks in the economy, irrespective of demand-side or supply-side sources. A simple analysis of the benefits of each respective policy strategy can be performed

164    Central Bank Policy using aggregate demand (AD)–aggregate supply (AS) equilibrium analysis, such as that conducted by Bradley and Jansen (1989). For simplicity, the policy target is defined in level form, thus the respective targets (achievement of policy targets) can be expressed as P = P* (prices) and PY = PY* (nominal income). Targeting a certain (constant) level of nominal income, for instance PY = K , thereK fore P = . Y Supposing point a represents equilibrium with price stability (P0) and fullemployment output (Y t) with a zero output gap. In the event of a negative demandside shock (Fig. 6.1), which reduces investment activity, the investment saving (IS) curve would move to the left, from IS0 to IS1, implying a shift in the AD curve to the left, from AD 0 to AD 1 (the central bank adopts a wait-and-see attitude at the beginning of the period). Such developments produce a new equilibrium at point b and, consistent with the decrease in prices from P0 to P1, total money supply increases, reflected by a shift in the LM curve from LM0 to LM1. In the case of a strategy framework targeting nominal income, the central bank must undertake monetary expansion, which moves the LM curve to the right, to drive AD to its previous level, point a, which is the intersection between the AS curve and constant nominal income curve. That process ultimately returns to equilibrium under conditions of full employment output (Y t) and prices P0. In the case of a strategy framework targeting prices, the monetary policy response will be the same as the previous policy response, namely monetary expansion to raise AD and prices to the original equilibrium. Therefore, the conflict between achieving price stability and stimulating output growth and employment can be avoided. Meanwhile, the consequences and policy response to supply-side shocks (Fig. 6.2) would be different. In this case, a positive shock on the supply side, which would move the AS curve to AS1, on one hand would raise prices from P0 to P1 but on the other hand would also lower output from Yf to Y1. That trade-off forces the central bank to strive toward achieving one of the targets, implying a decline in the other target. For example, expansive monetary policy (M = Mf) could strive to return output growth to its original equilibrium, Yf , at the expense of driving up prices in Pf. If lower prices are sought, however, contractive monetary policy would be required thereafter. Under such conditions, applying a monetary policy framework targeting prices would undermine the desired result after driving further output declines. Meanwhile, a policy framework targeting nominal income would facilitate a compromise, or flexibility, in terms of achieving the dual targets, namely by striking an optimal balance between higher prices and lower output. In this case, nominal income may be targeted through monetary policy that is not too tight (M = Mb), thus the rise in prices would not be too steep (P = Pb), while a drastic decline in output would also be prevented (Y = Yb). The prominent policy implication of the analysis above is that the final target of monetary policy, whether it be price stability or output growth, could be achieved by setting policy target priorities for different horizons. In this regard, price stability is more appropriate as the policy target in the long term, while nominal

Monetary Policy Strategic Framework     165

Fig. 6.1:  Demand-side Shocks: Targeting Prices and Nominal Income.

income stability could be applied as the near-term policy target. Although, theoretically, the dual targets are a distinct possibility, this is extremely hard to apply in practice, considering the difficulties encountered when identifying the cut-off point between the short and long term as well as the policy response to the prevailing shocks. If nominal income is targeted (in the near term) the central bank will confront difficulties when gauging the appropriate monetary policy response to extremely short-term shocks, particularly on the supply side. In addition, targeting nominal income would also erode central bank policy effectiveness in terms of anchoring inflation expectations. In general, therefore, central banks have applied a pragmatic solution in the form of achieving the long-term policy target of price

166    Central Bank Policy

Fig. 6.2:  Supply-side Shocks: Targeting Prices and Nominal Income.

stability, defined flexibly considering the possibility of shocks in the near term. Ignoring that problem for a moment, however, there are still advantages and disadvantages to each strategic policy target.

6.2.2. Monetary Policy Regime In principle, there are several monetary policy implementation strategies available to the central bank to achieve the final policy target. The fundamental difference between the strategies lies in the different indicators used as nominal anchors or intermediate targets to achieve the policy goal.2 In the literature, strategy selection depends on the corresponding monetary policy regime applied.

2

In the literature of traditional monetary economics, the monetary policy framework is generally translated through the relationships between instruments, operational targets, intermediate targets, and final targets. Over time, congruent with greater use of the interest rate as the operational target, with the direct influence of the final policy target, the presence of intermediate targets has become less relevant. Therefore, the focus has centered more on nominal anchors.

Monetary Policy Strategic Framework     167 According to Bordo and Shwartz (1999), the monetary policy regime is a set of monetary institutions and regulations, accompanied by public expectations concerning the policymakers’ measures on one hand and the policymakers’ expectations regarding the public’s reaction to the policy measures taken on the other. Per Mishkin (1999), however, the main characteristic of all monetary regimes is the use of nominal anchors in various forms. In addition, the success of different monetary policy regimes depends on steering the policy-making measures toward a certain condition in such a way that long-term price stability is the target that must be achieved. That definition has nearly the same understanding as the strategy or standard targeting rule. Utilizing the targeting strategy is believed to have a number of benefits in the form of commitment to limit discretion in policy decisions and, thus, inflation bias.3 Another benefit is to protect economic uncertainty from the political cycle. Therefore, the sound characteristics of targeting strategy will consider several aspects, including: (1) optimization, namely the best compromise between output and price stability; (2) stability, namely unforeseen contingency proof; and (3) simplicity, namely that the targeting strategy is best achieved transparently and accountably. Several targeting strategies are practiced at central banks, including: (1) exchange rate targeting, (2) monetary targeting, (3) inflation targeting, and (4) no explicit anchor targeting. In addition to those four targeting strategies, there is also a hypothetical targeting strategy that has not yet been put into practice, namely: (5) nominal income targeting.4 The following sections present a brief overview of each monetary policy strategy. 6.2.2.1. Exchange Rate Targeting.  There are three alternative approaches to exchange rate targeting as a monetary policy strategy. First, by pegging the value of the domestic currency to certain commodities recognized internationally, such as gold (known as the gold standard). Second, by pegging the value of the domestic currency to the hard currencies of larger countries with low inflation (known as the currency board system). Third, by pegging the value of the domestic currency to the currencies of specific other countries, where the domestic currency is allowed to fluctuate in line with the differences in inflation between the two countries (crawling peg). In Asia, the second alternative is applied by Hong Kong and Brunei, while Singapore applies the third alternative.

3

Representing a situation when, because of growth-oriented economic policy, real output does not increase, while inflationary pressures accumulate. In general, inflation bias is linked to time inconsistency in terms of policy-maker behavior. This issue is discussed further in Chapter 10 of this book. 4 For a more in-depth description of empirical observations concerning monetary policy implementation strategy in several jurisdictions, refer to Mishkin (1999).

168    Central Bank Policy Whichever alternative is selected, exchange rate targeting credibility depends on monetary authority discipline in terms of maintaining domestic money supply and credit to remain consistent with optimization of the exchange rate formation process. Similarly, fiscal policy discipline is also required to manage the budget deficit in line with macroeconomic stability. Flexibility in the real economy, in terms of production and labor, is also required to overcome the external shocks faced. Consequently, this regime is typically embraced by small and very open economies in terms of trade and services, such as Singapore and Hong Kong. Exchange rate targeting also has the following key advantages. First, exchange rate targeting may dampen inflation originating from changes in import prices of goods. Second, exchange rate targeting may anchor public inflation expectations. Third, exchange rate targeting is a rules-based monetary policy strategy that ensures monetary policy implementation discipline. Fourth, exchange rate targeting is sufficiently uncomplicated and unambiguous, and thus easy to understand by the public. In addition to the advantages, exchange rate targeting also contains some inherent weaknesses. First, when exchange rate targeting is pursued in a very open economy with foreign capital mobility, monetary policy cannot be implemented independently.5 Second, exchange rate targeting may cause each structural shock in certain jurisdictions to be transmitted or impact domestic economic stability directly. Third, exchange rate targeting is vulnerable to speculative attacks on domestic currency holdings. Fourth, in the event of a balance sheet mismatch, and if there is a need to devaluate with a high position of external debt, this strategy will trigger fragilities and a possible financial crisis, which could subsequently squeeze fund disbursements in the financial system. 6.2.2.2. Monetary Targeting. In many countries, exchange rate targeting is not the primary choice of monetary policy strategy because of the requirements mentioned above, including monetary policy restrictions or flexibility to meet the domestic economic goals. Consequently, a number of jurisdictions favor monetary targeting, namely by stipulating total money supply growth as an intermediate target, for instance, narrow money (M1), broad money (M2) as well as credit. The main advantage of monetary targeting over exchange rate targeting is the possibility of independent monetary policy, thus allowing the central bank to focus on the targets set, such as low inflation and sustainable economic growth. In addition, as with exchange rate targeting, monetary targeting is completely transparent and allows the public to instantly know the monetary policy stance adopted by the central bank. Such signals are expected to anchor public inflation expectations to future inflation and, hence, alleviate inflationary pressures. Nonetheless, this strategy has an inherent weakness in the form of dependence on a stable relationship between monetary aggregates and the final policy target (price and output), in addition to exchange rate stability, which may not be

5

Departing from the concept of institutional independence, independence in this context relates to the extent of monetary freedom or neutrality in terms of shocks originating from the external sector.

Monetary Policy Strategic Framework     169 maintained. Increasingly developed financial instruments and greater integration between the domestic and global economies has disrupted the stability of the relationship between monetary aggregates and the final policy target, as reflected by income velocity instability. Furthermore, if standard monetary targeting is applied without a feedback mechanism or economic response, the strategy would be unable to achieve output stability despite achieving price stability. Such conditions are the main reasons central banks cite for not applying the standard approach (with a trade-off between credibility and flexibility) or even for abandoning the strategy all together. 6.2.2.3. Inflation Targeting.  As the relationship between monetary aggregates and the final target of monetary policy faded, many countries began to adopt inflation targeting, which is implemented by publishing the mediumterm inflation target and committing to achieve price stability as the long-term goal of monetary policy. By targeting inflation as the nominal anchor, the central bank is more credible and focused on achieving price stability as the final target.6 Although inflation is targeted, this strategy does not neglect the attainment of other monetary policy goals, such as output and employment. In this case, the central bank always considers output stability and employment (at a certain level) in the near term. In addition, to minimize any output decline, the central bank periodically hones the medium-term inflation target toward a lower long-term inflation target. Nonetheless, if there is a conflict between inflation and output, inflation is the priority. The main advantage of this strategy, in addition to a focus on domestic interests (namely inflation), is policy transparency that reduces uncertainty as well as accountability to prevent inflationary pressures. In contrast, a salient weakness of the strategy is that inflation becomes a policy target rather than an intermediate target, which is difficult to control. Furthermore, there is a lag in the monetary policy transmission mechanism, consequently inflation targeting does not provide an immediate signal concerning the direction of monetary policy. A more in-depth description of this framework is presented specifically in Chapters 8 and 9 of this book. 6.2.2.4. No Explicit Anchor Targeting. Seeking to achieve the desired economic performance, monetary policy strategy can be implemented without an explicit anchor but still paying attention and providing commitment to achieve the final target of monetary policy, including low and stable inflation. This regime provides flexibility to the central bank when determining which target to prioritize in a given period, whether it be inflation, economic growth or creating job opportunities, depending on the prevailing economic dynamics and challenges. Additionally, this strategy is forward-looking, implying that monetary policy is applied by prudently monitoring the signals of indicators for the final policy target, accompanied by anticipatory measures.

6

A more in-depth description of the ITF is presented in Chapters 8 and 9 of this book.

170    Central Bank Policy Although successfully applied in several jurisdictions, this strategy is considered less transparent, hence the public tend to make assumptions regarding the purpose and goal of policy issued by the central bank, which could trigger market uncertainty concerning the output and price outlook. The ambiguity of the strategy could also undermine central bank accountability in the eyes of the public and parliament due to the lack of success criteria, which are typically announced beforehand. 6.2.2.5. Nominal Income Targeting.  Fundamentally, targeting nominal (output) income is based on the view that monetary policy can only affect nominal income growth rather than its determinants in the form of inflation and real output growth. In addition, a lack of comprehensive understanding regarding the factors influencing those two determinants implies that the central bank should be oriented toward achieving a certain level of nominal income growth, not controlling inflation and real output separately. According to McCallum (1988), although this targeting strategy has never been put into practice, counterfactual experiments have shown that this strategy might be superior to the others. Several empirical studies (Hall & Mankiw, 1993; McCallum, 1988) have shown that if this strategy was applied in the past, inflation and output would have hypothetically performed more smoothly than the historical trends. From a theoretical point of view, nominal income targeting is an alternative strategy garnering widespread attention.

6.3. Empirical Studies and Related Issues The priority between inflation and output as the final target, as well as the most appropriate monetary policy regime for a particular country will depend on the specific economic dynamics prevailing at a given time. In this regard, numerous studies have been performed to investigate the relationship between inflation and economic activity in the short and long term. This is where the debate emerges, not only because inflation does not always impact long-term output adversely as theory would predict, but also because the short-term relationship between the two is not always consistent over time in one country or between jurisdictions at the same time. In addition to the structure and flexibility of the real economy, the sources and magnitude of shocks in the economy, demand-side or supply-side, also have an effect. Furthermore, the trade-off between inflation and output is also influenced by the policies instituted by the government and monetary authority. Consequently, when formulating the monetary policy strategy framework, the achievement of price stability and output growth are typically sought in tandem, paying due consideration to the degree of flexibility between both in the near term, while prioritizing price stability in the long term. According to Fischer (1996), several empirical facts point to a trend of giving the central bank responsibility for controlling inflation. Nevertheless, when implementing that mandate, the central bank also considers economic growth because monetary policy can indeed influence output, at least in the near term. Furthermore, the understanding of economic players is based more on monetary policy signals and their effect

Monetary Policy Strategic Framework     171 on economic growth. In this regard, it is important to understand that monetary policy support for economic growth will be best if the central bank consistently maintains low inflation and supports financial system stability.

6.3.1. Short-term Output-prices Trade-off and the Phillips Curve Phenomenon The empirical study of Phillips (1958) concerning the trade-off between output and price stability represents seminal findings that remain pertinent to the present day, namely the Phillips Curve. The study demonstrated the trade-off between inflation and output, which manifests as deviation from the long-term trend, although this trade-off is comparatively weak in the near term. As explained in Chapter 3, the empirical evidence for the Phillips curve subsequently became the foundation of the synthesis between Classical economics, which opines that monetary policy only influences inflation in the long term, and Keynesian economics, which believes monetary policy can effectively drive economic growth. In several ways, this study underlies monetary policy strategy to actively stabilize output, a regime implemented in several countries during the 1970s and beginning of the 1980s. Nevertheless, in the 1980s, the pendulum swung toward monetary policy with more of a focus on price stability. The description above shows how important empirical studies of the tradeoff between inflation and output are in terms of underlying the monetary policy strategy framework. This is primarily linked to three salient issues underlying the existing research, namely: (1) Whether the relationship or trade-off can be proven empirically and how the relationship or trade-off occurs in the short and long term? (2) Whether the relationship or trade-off in linear (symmetrical) or nonlinear (asymmetric)? (3) What are the implications of the relationship or trade-off in terms of implementing economic policy? 6.3.1.1. Presence of the Phillips Curve.  The first issue relating to the emergence of empirical facts is stagflation, namely low economic growth coupled with high inflation, as experienced by industrialized nations in the 1970s, who opposed the trade-off shown in the Phillips curve. Stagflation simultaneously demonstrates that, in the long term, the trade-off in the Phillips curve is not always supported by consistent empirical evidence and that high inflation is not necessarily a constraint to robust growth. In fact, research in the past decade (Barro, 1995; Fischer, 1993; Ghosh & Phillips, 1998) concluded a significant inverse correlation or trade-off between output growth and inflation, implying that prevailing economic conditions and the policies implemented will also influence the relationship between inflation and output. In the context of multivariate relationships, by including demand-side variables, such as money supply and government spending, as well as supply-side variables, such as productivity, the fundamental relationship between output growth and inflation is extremely complex. This is possibly because of the variable interactions and mutual interconnectedness between the determinants of inflation and determinants of output growth. Therefore, it can be concluded that even if the close relationship between inflation and output growth is present, the

172    Central Bank Policy correlation is more complex than a linear relationship with a constant parameter (monotonic). In other words, although the Phillips curve could be used as a reference when prioritizing inflation and economic growth as the final targets of monetary policy, the trade-off between the two and the strategy to attain the target is not easily explained theoretically. 6.3.1.2. Linearity in the Phillips Curve.  The second issue is based on the findings of Laxton, Meredith, and Rose (1995) concerning the asymmetric effect of economic activity on inflation in seven major OECD countries. The study provided an important conclusion, namely that there is no linear or constant tradeoff between economic activity and inflation in the long term. Specifically, strong evidence for the non-linear relationship between inflation and output in the Phillips curve is reflected in the stronger impact of excess demand shocks in terms of driving inflation rather than excess supply shocks in terms of dampening inflation. Several empirical studies have also provided evidence for the non-linearity of the Phillips curve, including the research of Clark, Laxton, and Rose (1996), Debelle and Laxton (1997), and Fisher, Mahadeva, and Whitley (1997). A number of policy implications may be drawn from the empirical studies mentioned above. First, policymakers may align the achievement of output and price stability in the long term more optimally by separating the effect of demand-side shocks and supply-side shocks through supply-management policies, such that achieving robust output growth in the long term would not exacerbate inflationary pressures that could culminate in economic losses. Second, the non-linear relationship between economic activity and inflation has an important implication for demand-management policies, namely that in response to demand-side shocks, and to maintain control over inflation, average real economic growth must be maintained below its potential. Consequently, if real output exceeds its potential rate, the central bank must accelerate its monetary policy response to control the early symptoms of inflation, namely a shock in the form of an increase in AD. 6.3.1.3. Formation of Expectations in the Phillips Curve.  During development in the 1990s, studies of the Phillips curve tended to focus more on the effect of forming rational expectations on the trade-off between inflation and output per the Neo-Keynesian theoretical framework. Departing from the traditional version of the Phillips curve, the new iteration predicted inflation to have a positive correlation with the output gap when forming forward-looking inflation expectations. This proposition was based on analyzing the heterogeneity of nominal price formation between producers (staggered price setting) inspired by the study conducted by Taylor (1980). The general specification was developed by Calvo (1983), who showed the Phillips curve with forward-looking expectation formation, where the current rate of inflation is linked to future inflation expectations and the marginal cost or output gap. Therefore, inflation in the previous period is not relevant when determining inflation in the current period, implying that while the central bank remains committed to stabilizing the output gap, in other words maintaining economic growth near potential output, the economy could mitigate inflation itself at a low and stable rate. Expressed in a different way, there is no longer a stable trade-off between prices and the output gap.

Monetary Policy Strategic Framework     173 The problem is that not all economic players act rationally and always form future expectations. The formulation of expectations in the Phillips curve, be they forward-looking or backward-looking, is the subject of another debate. As stressed by Fuhrer and Moore (1995), the dynamic behavior of inflation and output in the forward-looking specification appears awkward if linked to the empirical evidence. That specification implies that inflation anticipates the output gap but the empirical evidence points to the opposite, namely that the output gap tends to drive inflation.7 In this regard, the empirical evidence appears consistent with the traditional version of the Phillips curve. Meanwhile, the latest studies from Sbordone (1999), Gali and Gertler (1999), and Gali, Gertler, and LópezSalido (2000) provided significant evidence of forward-looking natural inflation behavior as well as linkages between inflation and the real marginal cost or output gap. Regarding both phenomena, Gali and Gertler (1999) and Gali et al. (2000) proposed a hybrid Phillips curve model, namely one that also considered certain fractions of backward-looking expectations in addition to forward-looking expectations in the relationship between inflation and real output. In Indonesia’s case, evaluating and testing the empirical model produced several significant findings that could be concluded as the Phillips curve phenomenon, where the presence and behavior evolved over time in line with changes in the structure of economic fundamentals, particularly due to the economic crisis in 1997 (Solikin, 2004). Specifically, the formation of expectations and linearity of the Phillips curve experienced a significant change between the pre- and post-crisis periods. Pre-crisis, the characteristics of the Phillips curve showed that: (1) the effect of output gap pressures on inflation was moderate (0.2–0.3); (2) forward and backward-looking expectations were relatively well balanced; and (3) the relationship tended to be linear. In contrast, the post-crisis characteristics of the Phillips curve showed that: (1) the effect of output gap pressures on inflation tended to increase (0.4–0.6); (2) forward-looking expectations became dominant; and (3) the relationship tended to be non-linear (convex). The empirical findings showed how crisis conditions have exacerbated the trade-off between inflation and economic growth, which must be considered in the monetary policy strategy framework.

6.3.2. Output-prices Trade-off in the Economic Cycle Studies into co-movement between output and prices have also become an area of focus when analyzing the economic (business) cycle. One objective of such research is to seek evidence concerning the types of structural shocks that influence comovement. Previous observations have shown that price developments are procyclical in terms of output (Lucas, 1977; Mankiw, 1989), consistent with analysis into the dominance of demand-side shocks in the economic cycle. This is also congruent with the standard Phillips curve findings that provide a rational basis for positive comovement between real output and inflation. On the other hand,

7

At least if using detrended log GDP as a proxy of the output gap.

174    Central Bank Policy however, Cooley and Ohanian (1991) suggested that, fundamentally, the trade-off between inflation and output depends on the relative effect between two shocks, namely demand shocks and supply shocks. In this context, demand shocks cause output and prices to move in the same direction, while supply shocks cause output and prices to move in opposing directions. That price developments explain economic cycle procyclicality was also refuted by Kydland and Prescott (1990) because the correlation between output and prices after World War II was generally inverse, or positive but weak. Several follow-up studies showed that the trade-off between output and prices could be explained in the standard model assuming sticky prices and only demand shocks (Ball & Mankiw, 1994; Chadha & Prasad, 1993; Judd & Trehan, 1995). Judd and Trehan (1995) showed that the correlation coefficient between inflation and output depended not only on the magnitude of the relative effect between demand and supply shocks but also the dynamic response of each respective variable to the shocks. In addition, Rotemberg (1996) found that the inverse correlation between projected output and prices was consistent with the simple sticky price model using only demand shocks. Nevertheless, the trade-off between inflation and real output is difficult to analyze based on a model assuming flexible prices, considering that that would require different money supply behavior in response to real shocks, forecastable or not. The intuition of the description above is as follows. Under conditions where the immediate response of prices is slow (sticky prices), nominal (and positive) demand shocks would edge up output in the near term but the trend would not be maintained in the long run. Real output would, therefore, fall back to the steady state value after spiking in the near term because expansive monetary policy would spur demand-side shocks. On the other hand, the response of higher prices to such demand shocks would not appear in the near term after expansive monetary policy because of sticky prices. Nonetheless, the price increase would only become permanent in the long term with output surpassing its long-term trend. Such post-shock economic cycle dynamics explain the trade-off between output and prices in the long run. Motivated by previous findings, Den Hann (2000) analyzed the comovement of output and prices in post-WWII United States using a different strategy. He argued that the main cause of uncertainty concerning the studies on comovement stemmed from the use of just one correlation coefficient. In his study, Den Hann applied a vector autoregression (VAR) approach to analyze the correlation function of forecast errors over different horizons. At least two important findings were concluded from the study. First, using the statistical data proposed, Den Hann found that comovement between output and prices were positive in the near term and negative in the long run. Second, Den Hann showed that the results could be explained using a model where demand shocks were dominant in the near term and supply shocks were dominant in the long term. Therefore, this study provided empirical evidence for the relative effect of demand and supply shocks on the correlation between real output and inflation in the economic cycle in the theoretical models developed previously by Chadha and Prasad (1993) and Rotemberg (1996).

Monetary Policy Strategic Framework     175 6.3.3. Sacrifice Ratio: Cost of Controlling Inflation In general, the main objective of monetary policy is to achieve low and stable inflation, which is expected to contribute to sustainable economic growth in the long term. A related practical question, therefore, is why the central bank does not reduce inflation drastically? The answer is because disinflationary policy may prompt a more significant decline in real output and, therefore, a large sacrifice ratio for the sustainability of the economy in general. For example, disinflationary policy by significantly slowing the growth of money supply would drive up interest rates and trigger a recession. The central bank dilemma of controlling inflation at the cost of lower output is a controversial topic in economics. The traditional view states that lower inflation would cause a recession, as mentioned above. Opposing that viewpoint, however, Neo-Classical economists believe that the cost is negligible or not even present. According to Sargent (1983), the inflation cost could be negated if anticipated. For example, the central bank could announce its intention to reduce the supply of money supply to subdue inflationary pressures. If the economic players could anticipate that information, the response would be a decline in the rate of price change in the same period, hence real money supply would not change and the recession could be avoided. Even if the decline in money supply growth was not anticipated, if businesses could adjust prices quickly, the effect on output would be comparatively small and quick to dissipate. Therefore, it can be concluded that the effect of disinflationary policy depends on the fundamental assumptions used, namely: (1) whether the change in monetary policy could be anticipated, meaning whether monetary policy could be announced clearly and whether that announcement is trusted by the markets; and (2) whether the adjustment process to shocks could be implemented quickly enough. Using different assumptions and approaches, several empirical studies have been conducted to observe the inflation cost of disinflationary policy. Fundamentally, the cost of disinflationary policy can be measured several ways. Using a simple concept, namely the sacrifice ratio, Okun (1978) was the first to measure the sacrifice cost, in the form of output loss, of disinflationary policy based on an assessment of the Phillips curve. Using a back-of-the-envelope approach based on estimates of the linear Phillips curve parameters, Okun (1978) concluded that the sacrifice ratio is 10, implying a 1% decline in inflation would prompt a corresponding 10% decline in economic growth. This concept is illustrated in Fig. 6.3. Meanwhile, Gordon and King (1982) applied a more complex approach in the modeling structure for inflation, output and the linear Phillips curve, concluding that the sacrifice ratio was much lower than that proposed by Okun at just 3. On the other hand, Ball (1994) criticized potential problems in the assessment based on the linear Phillips curve and estimated the sacrifice ratio using an atheoretical approach. By measuring directly the output loss in periods when inflation decelerated, Ball concluded that the sacrifice ratio was 2.3. Relevant studies have also tackled the non-linear Phillips curve, including Clark et al. (1995), Laxton et al. (1995), and Filardo (1998). The empirical

176    Central Bank Policy

Fig. 6.3:  Output Loss Linked to Disinflationary Policy.

facts concerning the non-linear correlation in the Phillips curve created additional complexity when estimating the sacrifice ratio. Christiano, Eichenbaum, and Evans (1944) estimated the sacrifice ratio using a VAR modeling system approach, where the ratio was defined as the ratio of output response relative to the inflation response in the event of policy rate shocks. This involved complex estimations due to dependence on a change in police regime, whether contractive (output below the trend) or expansive (output above the trend). Using structural equations that took into consideration the piecewise non-linearity of the Phillips curve, Filardo showed that the sacrifice ratio during an economic downswing was 5.0, compared to 2.1 when the economy was overheating. Nevertheless, Filardo’s figures differed from the sacrifice ratio estimated using a linear Phillips curve equation, namely 5.7.

6.4. Implementation in Several Jurisdictions The evolution of the central bank’s role and reforms over time was elaborated in Chapter 2, showing how monetary policy is influenced by economic dynamics, the political configuration, and academic thinking. Therefore, the monetary policy

Monetary Policy Strategic Framework     177 strategy framework applied by central banks has also experienced evolution over the years. The pendulum of the final policy target swung from economic growth to inflation, while the monetary policy regime has also evolved from exchange rate targeting to inflation targeting.

6.4.1. Final Target of Monetary Policy How is monetary policy practiced around the world to ameliorate public welfare? Is the focus on price stability or also directly linked to stimulating economic growth? The following section describes how political preferences also influence the selection of the final target of monetary policy. 6.4.1.1. A Shift in Priorities from the 1970s to 1980s. Economic dynamics around the world in the 1970s were characterized by the desire to stimulate economic growth and create job opportunities, consistent with the goal of economic development at that time, namely post-WWII recovery in advanced countries and post-independence economic advancement in developing countries. In general, the central bank was still under government control as a member of the cabinet and/or the formation of a monetary board. Consequently, although the legal and institutional frameworks (laws) of the central bank pointed to several monetary policy final targets, many central banks around the world in the 1970s prioritized stimulating economic growth and creating job opportunities rather than controlling inflation and the balance of payments (BOP). In other words, monetary policy was viewed as an instrument of output stabilization policy at that time. This is congruent with Keynes’ view developed based on observations of the UK economy after the Great Depression in the 1930s. Output stabilization policy and monetary policy were supported by the correlation between output and inflation per the Phillips curve, which meant the central banks had to determine the desired trade-off in terms of stimulating economic growth on rising inflation that was politically tolerable. Thereafter, in the 1980s, a fundamental change was witnessed as the pendulum of the central bank’s final policy target swung again, this time from economic growth to price stability. Using monetary policy as a pro-growth instrument in the 1970s proved to be a failure because the economy did not grow but inflation soared, leading to stagflation in industrialized nations in 1973–1974, which was followed by soaring inflation in 1979–1980 as the global oil price skyrocketed. Consequently, central banks once again prioritized controlling inflation. Despite remaining under government control, central banks began to consolidate their control over money supply, including for government fiscal financing, which had previously prioritized stimulating economic growth. Reprioritizing the final target of monetary policy back toward explicitly controlling inflation began in the UK. By applying a medium-term policy strategy, monetary policy at the beginning of the 1980s sought to control inflation and support sustainable economic growth and job creation. In that regard, the implementation of extremely tight monetary policy in the 1980s fed through to pound sterling appreciation and triggered an economic recession. The tight monetary policy stance did, however, strengthen monetary policy credibility in terms of

178    Central Bank Policy easing wage-push inflationary pressures on the labor market that were previously uncontrolled, while the impact on the financial sector was subdued by government policy. The recession experienced at the beginning of the 1980s ultimately created solid anti-inflation policy credibility in the UK. In the United States, drastic measures to tighten monetary policy in terms of controlling inflation were generally avoided throughout the 1970s because of the huge sacrifice cost involved in the form of declining output.8 Consequently, inflation was unacceptably high throughout the 1970s but such conditions were acceptable socio-politically. Radical measures were only implemented in 1979, when the Federal Reserve decided to end accommodative monetary policy. That move triggered a deep recession, with unemployment soaring to around 11% at the end of 1982. Seeking to overcome the issue, the US Administration instituted a number of policies, including expansive fiscal policy and a loose monetary policy stance. Supported by greater wage flexibility, unemployment was eventually reduced gradually. In general, controlled inflation and lower unemployment in the US became a less prominent issue concerning the shift in the final target priority of monetary policy than in the UK and other European countries. In Germany, unemployment was high in the 1980s, increasing two-fold on the level recorded in the 1970s. Despite such conditions, the central bank was not inclined to implement policy to boost demand, although inflation remains very low, approaching zero. Such conditions demonstrated good monetary policy discipline in terms of controlling inflation in Germany. Meanwhile in France, although a platform to explicitly reduce unemployment was introduced at the beginning of the 1980s, high unemployment remained, exceeding 6.6% in 1980. Once France decided to join the European Monetary System (EMS) in 1982, the country also committed to controlling inflation to reduce inflation disparity with other EMS members. In fact, stimulating economic growth and creating job opportunities were temporarily set aside in terms of the priority final target of monetary policy. Compared to other advanced countries, in the middle of the 1970s, Japan successfully maintained a relatively low level of unemployment, while inflation simultaneously experienced a sharp decline. At the end of the 1970s and throughout the 1980s, marginal increases in the level of unemployment were observed, albeit moderate, to a level of around 3%. For developing countries, however, the issue of shifting the final target of monetary policy from economic growth or job creation to controlling inflation was not a core issue and therefore did not receive much attention during 1970s–1980s. In general, the policy orientation of developing countries focused on creating a climate conducive to production and employment. Even if controlling inflation was considered, this was achieved jointly in the context of sustainable economic growth and job creation. In this regard, the strategy to determine the final target

8

From the study by Arthur Okun (1978), the cost of disinflationary policy (sacrifice ratio) was estimated at 10:1, implying a 10% decline in real output would be required to lower inflation by 1%. A full description of the sacrifice ratio is presented in this chapter.

Monetary Policy Strategic Framework     179 of monetary policy was conducted without explicitly prioritizing a certain target. Such developments were reflected in the relatively high pace of economic growth recorded in developing countries, including Indonesia, Thailand, and Malaysia at the end of the 1980s through to the beginning of the 1990s. Meanwhile, although not achieving a particularly low rate, (moderate) single-digit inflation was achieved and maintained. 6.4.1.2. From the 1990s Until the Present Day.  Since the 1990s to the present day, nearly all central banks around the world have adopted price stability as the final target of monetary policy. In fact, most central banks have formally applied a monetary policy framework with controlling inflation as the final target. This phenomenon began with the proliferation of the Inflation Targeting Framework (ITF) during the 1990s at many central banks in advanced countries and EME. Fundamentally, ITF emphasizes the desired inflation target, monetary policy consistency to achieve said target as well as stronger central bank independence and transparency to the public. ITF is considered viable to overcome the high inflation experienced in many countries in the near term, while simultaneously providing a solid foundation to achieve sustainable economic growth. A more thorough description of ITF and how it is practiced at various central banks is presented in Chapters 8 and 9. In practice, a number of central banks focus on inflation, while other central banks select inflation and exchange rate stability as the dual final targets of monetary policy. In many jurisdictions, including Indonesia, there has not only been a shift in priority but also the target focus from multiple objectives, namely economic growth, inflation control, and job creation, to a single objective, namely rupiah currency stability (inflation and the exchange rate).9 In many countries, however, the generic target of price stability could be defined more concretely, including the use of relevant price indexes, such as the Consumer Price Index (CPI). For example, the Bank of Canada targets 1–3%, the Bank of England targets 2.5 ± 1% and the Reserve Bank of Australia targets 2–3%. In general, the inflation target is set by the government or through coordination between the government and central bank. The formal final targets of monetary policy in various jurisdictions are presented in Table 6.1. It is important to note several exceptions when setting price stability as one of the final targets of monetary policy. One is the case of Germany, where despite not using inflation control as the final target of monetary policy, the Deutsche Bundesbank implicitly targets inflation in the 1.5–2% corridor. Similarly, although many perceive the US Federal Reserve to focus on maintaining low inflation, price stability is not more dominant. In this case, the Federal Reserve Act legally

9

As contained in Act No. 13 of 1968 concerning the Central Bank, the duty of Bank Indonesia is to assist the government regulate, create, and maintain rupiah stability, while also stimulating production and development as well as creating job opportunities to ameliorate public welfare. Meanwhile, based on the Bank Indonesia Act (No. 23) of 1999, the objective of Bank Indonesia is to create and maintain the value of the rupiah.

180    Central Bank Policy Table 6.1:  The Central Bank’s Legal Monetary Policy Framework. Final Target of Monetary Policy

Country

Domestic price stability as the final target

Columbia, Czech Republic, Hungary, Iceland, South Korea, México, New Zealand, Norway, Poland, Sweden, Thailand, UK

Currency stability as the final target

Finland,a Indonesia, Norway

Currency stability and other targets

Brazil, Australia, Canada, Chile, Israel, South Africa

Source: Carare, Shaechter, Stone, and Zelmer (2002). a Also maintains the stability and security of the monetary system and currency in circulation.

stipulates that the goal of monetary policy is to “achieve long-term economic growth to potentially drive production, job opportunities, price stability and moderate long-term interest rates.” The question, therefore, is how to observe the linkages between setting the final target and monetary policy performance? A simple explanation concerning those linkages based on indicator data for the final monetary policy target (among others), economic growth, and price stability, is presented as follows along with indicator performance over the past 25 years. From the Table 6.2, at least two salient empirical facts emerge as follows: (1) First, there is a common trend, namely the phenomenon of declining inflation, both in advanced and developing countries. In the case of Brazil, however, despite experiencing increasing average inflation for more than the past decade in the 5–8% range, this represents a remarkable improvement on conditions at the beginning of the 1990s, when hyperinflation topped 100%. In Indonesia, average inflation remains quite high but has decreased significantly from the position in the 1990s until early 2000s. Sticky inflation in Indonesia stems from potentially intense inflationary pressures on administered prices (AP) in line with rising fuel prices over the past few years. (2) Second, opposing accelerating and decelerating economic growth trends were observed in advanced and developing countries. In the US, UK, Germany, and México, for example, average economic growth tended to accelerate during the second half of the 1990s, while Japan and Brazil experienced relatively stable average growth throughout the 1990s, contrasting conditions in Thailand and Indonesia, two EME in Southeast Asia that were hit by a severe economic crisis in 1997/98. During the latter half of the 1990s, the economies of both countries, which had previously posted solid average growth (around 8%), suddenly plummeted to just 1%. Meanwhile, during the five years after the GFC, flagging economic growth was observed in advanced countries, especially the United States and United Kingdom.

1.83

2.62

2001–2007 2.76

2010–2014 1.94

2.06

2.45

4.10

2.47

yg

1.70

2.74

2.54

2.79

π

Source: CEIC data. yg: annual economic growth; π: annual inflation.

2.71

1996–2000 3.14

π

3.02

yg

1991–1995 1.68

Period

United Kingdom United States

2.06

1.43

1.69

1.34

yg

1.47

1.79

1.42

3.40

π

Germany

0.31

1.01

π

1.47

0.66

1.41 –0.17

1.41

1.52

yg

Japan

3.32

3.52

2.24

3.10

yg

π

yg

π

Mexico

6.11

7.67

5.93 3.36

2.37

3.97

4.35

8.35 16.66

> 100 2.57 19.55

Brazil

Developed Countries

Table 6.2:  Average Economic Growth and Inflation in Several Countries from 1991 to 2014 (%).

3.85

5.38

0.65

7.84

yg

2.50

2.79

3.76

4.87

π

Thailand

8.75

π

5.80

5.07

6.17

9.08

1.02 20.85

7.89

yg

Indonesia

Monetary Policy Strategic Framework     181

182    Central Bank Policy Those two salient facts, namely the downward inflation trend and multispeed economic growth, fundamentally support the previous statement concerning a gradual evolution in the final target of monetary policy, namely toward controlling inflation. Nevertheless, although congruent with the advantages of applying the ITF, it cannot be concluded that the framework is an unmitigated (absolute) success in terms of overcoming the problem of high inflation in many countries, considering that many countries, including the US, Germany and Japan, successfully achieved low inflation without explicitly applying the ITF. In fact, three of the countries did not even mention that domestic price stability was a final monetary policy target in their legal framework.

6.4.2. Monetary Policy Regime Of the various monetary policy regimes practiced by central banks, exchange rate targeting, monetary targeting, and inflation targeting are the most common. In their evolution, economic, and political dynamics along with academic thinking have shaped the regime change from exchange rate targeting to money supply and inflation targeting. 6.4.2.1. Exchange Rate Targeting. The exchange rate targeting regime was introduced by several countries as part of the efforts to control inflation during the 1980s. With the inherent advantages of the regime, exchange rate targeting successfully lowered inflation in many industrialized countries at that time. For example, the UK and France successfully applied exchange rate targeting to suppress inflation by pegging their currencies to the German mark. In 1987, when France first pegged its currency to the German mark, inflation in the country was 3%, which was 2% above the rate recorded in Germany. In 1992, however, inflation in France had decelerated to 2%, a level not only considered consistent with stability, but also below inflation in Germany. Inflation in France and Germany finally converged in 1996 at slightly below 2%. A similar trend occurred in the UK, where inflation fell from 10% in 1990 to 3% in 1992 after the pound sterling was pegged to the German mark. The successful implementation of exchange rate targeting to lower inflation was not merely limited to advanced countries but also transpired in several EME. Argentina is an important case study, where the currency board arrangement was initiated in 1991. Accordingly, the central bank applied a fixed peg between the USD and (new) peso at a level of 1:1.10 At the beginning of implementation in Argentina, the currency board was considered very successful. Inflation fell

10

The currency board was a form of fixed exchange rate regime, with certain criteria. The currency board demanded greater and clearer commitment to the exchange rate target compared to the standard fixed exchange rate regime because the monetary authority was mandated to exchange domestic currency for foreign currency at a certain rate as required by the public. To build credibility, the currency board maintained more than 100% reserve assets as support for the domestic currency, while permitting the monetary authority to deviate from discretion in absolute terms.

Monetary Policy Strategic Framework     183 dramatically from more than 100% in 1989 and 1990 to below 5% at the end of 1994, while economic growth accelerated with rapidity, averaging 8% from 1991 to 1994. In practice, however, exchange rate targeting has many inherent weaknesses. During German reunification in 1990, for instance, the attention of policymakers tended to focus on the intense inflationary pressures stemming from expansive fiscal policy to rebuild the former East Germany. Such developments edged up shortterm and long-term interest rates in 1991. Subsequently, the shocks that occurred in the German economy, as an anchor country in the exchange rate mechanism (ERM), were transmitted to other countries with a currency pegged to the German mark because interest rates increased in line with interest rates in Germany. Consequently, such conditions culminated in significant economic moderation and increased unemployment in many countries, especially France and UK. The main problem that emerged in relation to reunification and the exchange rate crisis was speculative attacks on the money market. This is well-founded considering how committed ERM member countries were to maintain the value of their currencies in line with the German mark, while, on the other hand, spiraling unemployment became more questionable. Consequently, a growing trend emerged in other countries, including France, Spain, Sweden, Italy, and UK to depreciate their currencies against the German mark. Therefore, large-scale speculative attacks commenced in September 1992. In reality, only France remained committed to maintain its peg against the German mark. In general, other countries were less inclined to maintain the peg because of the adverse impacts on the real economy, which ultimately led to currency depreciation. In opposition to France, the UK elected to leave the ERM in 1992, replacing the regime with inflation targeting, a framework that subsequently benefitted the implementing economies through increased economic growth, lower unemployment and low inflation. Meanwhile, the implementation of exchange rate targeting, with a certain degree of flexibility, in EME culminated in spectacular economic crises, such as the 1982 crisis in Chile, 1994 in México, and 1997 in Southeast Asia. The implementation of exchange rate targeting in EME, generally in the form of a fixed or managed exchange rate regime, precipitated a trend of spiraling foreign debt in the corporate and financial sectors. The long-term exchange rate stability associated with this regime was perceived as a guarantee against currency risk by the authorities on their foreign loans. The dominant composition of foreign debt was a common feature of the institutional structure in Latin America and Southeast Asia prior to the crises. Economic crises are triggered by structural problems in the economy that spur pressures on the domestic currency and, thus, drive strong depreciatory pressures or even devaluation.11 Domestic currency depreciation against foreign currencies

11

In several cases, the scale of the economic crisis is exacerbated by a political crisis that prevents the monetary authority or government from taking the most optimal and timely policy measures as required, which occurred in México and Indonesia.

184    Central Bank Policy also compounded the burden of the foreign debt held by the corporate and financial sector. Meanwhile, considering most assets are generally in denominated in the domestic currency, the value of corporate assets failed to accrue. Such conditions undermined the balance sheets and net worth of domestic corporations and financial institutions. In the following series of events, aggravated by a loss of public confidence in economic activity, investment, and economic activities nosedived, as occurred in Chile, México, and several Southeast Asian countries, including Indonesia. 6.4.2.2. Monetary Targeting.  Two countries that have seriously and successfully implemented monetary targeting are Germany and Switzerland. The success of such a policy on controlling inflation in those two countries is the reason why monetary targeting continues to receive support. This represents a grand departure from the experiences of other countries that have been less successful with monetary targeting, including the US, UK, and Canada, because serious efforts were lacking in its implementation and the relationship between monetary aggregates and the final policy target was unstable. The key to successful monetary targeting implementation in Germany and Switzerland was simplicity, avoiding overly technical implementation, and relying on a monetary policy communication strategy that focused on the long-term considerations and efforts to control inflation. Therefore, the considerations underlying monetary targeting variations became a public exercise, paying due attention to the following aspects: (1) setting the inflation target as a range or corridor based on a quantity framework; (2) providing regime flexibility, in practice, by considering attaining other targets, such as output and exchange rates; and (3) providing avowed commitment by the monetary authority as well as a clear public communication strategy. The success of the regime to control inflation in Germany convinced several other countries to follow suit, providing one indication why monetary targeting was more popular than exchange rate targeting in a number of countries. One important success that should be noted was the success of Germany in controlling inflation because of post-reunification economic expansion at the beginning of the 1990s, where, in 1995, the inflation rate had finally dropped below the normative 2% rate of the Bundesbank.12 The implementation of monetary targeting in Germany also received serious criticism in relation to the asymmetric Bundesbank reaction when failing to hit the target set (Clarida & Gertler, 1997). In this respect, when overshooting the monetary target, the Bundesbank instituted policy to raise interest rates but when undershooting, efforts to lower interest rates were not taken. Consequently, the policy direction of the Bundesbank in the middle of the 1990s, when inflation was below the normative target of 2%, appears overly tight, which not only exacerbated German unemployment but also in other EMS members, such as France.

12

With the characteristics of the monetary policy framework adopted and the successful achievement of relatively low and stable inflation, there was a presumption that the Bundesbank was fundamentally an inflation targeter.

Monetary Policy Strategic Framework     185 Compared to implementation in Germany, monetary targeting in Switzerland was relatively problematic. With a small open economy (SOE), the institutional structure of the Swiss financial system experienced a significant change. In 1977– 1978, the Swiss franc experienced hyper-appreciation of 40%, which was intolerable to the Swiss National Bank (SNB). Under such conditions, Switzerland temporarily abandoned monetary targeting (1978–1979) in favor of exchange rate targeting. In 1980, however, monetary targeting was formally reintroduced, with a change in the target indicator used from M1 to broad money. In 1989–1992, monetary targeting in Switzerland encountered more problems, with average inflation accelerating beyond 5%. SNB failure to maintain low inflation was due to currency shocks and a shift in the demand for money function as a consequence of monetary expansion to depreciate the Swiss franc and policy to stimulate financial system development. Consequently, the SNB found it difficult to predict and achieve the inflation target set previously, hence inflation continued to soar beyond an acceptable rate. The problems encountered when implementing monetary targeting led to ambitious monetary policy easing in Switzerland at the beginning of the 1990s, thus rendering the advantages of the regime, as an instrument to signal the direction of monetary policy, impotent. Seeking to overcome that constraint, the SNB adopted a more flexible framework, by announcing a medium-term (3–5 years) inflation target as well as setting the basis for the expansion path at the end of 1992. Thereafter at the end of 1994, the SNB reiterated its mediumterm broad money expansion path for 1995–1999. With a more flexible framework, monetary policy in Switzerland successfully brought inflation back under control at around 1%. Two invaluable lessons can be taken from the German and Swiss experiences. First, monetary targeting can be used to control inflation in the long term, despite the possibility of target inaccuracies. Therefore, rigid rules are considered unnecessary to achieve the inflation target. Second, the main reason why monetary targeting was a success in Germany and Switzerland, despite the inaccuracies, was because the final monetary policy target was announced clearly, with the Bundesbank and SNB fully committed to a public communication strategy, which improved central bank transparency and accountability of public perception. 6.4.2.3. Inflation Targeting.  As the relationship between monetary aggregates and the final policy target faded, inflation targeting emerged as a new monetary policy framework that appealed to countries previously plagued by inflation control problems in the 1990s. New Zealand was the first country to formally adopt the inflation targeting regime in 1990. Over time, New Zealand proved how successful inflation targeting could be at driving the real economy, while also controlling inflation, which piqued central bank confidence in the new framework. After New Zealand, the following countries adopted inflation targeting: Canada in 1991, UK in 1992, Sweden in 1993, Finland in 1993, Australia in 1994, and Spain in 1994. As a monetary policy framework that has received broad attention from academia and central bankers, the practice of inflation targeting is discussed in more depth in Chapters 8 and 9 of this book.

186    Central Bank Policy 6.4.2.4. No Explicit Anchor Targeting (Implicit Targeting). Fundamentally, several countries have practiced this strategy, evidenced by impressive macroeconomic performance (including low and stable inflation) without resorting to a nominal anchor, such as exchange rate targeting, monetary targeting or inflation targeting. The real success of this just-do-it strategy is apparent in the US experience, namely when the Fed successfully controlled US inflation from double digits in 1980 to around 3% at the end of 1991. Since then, US inflation has been maintained at around or below 3%. Strategy implementation was characterized by successful pre-emptive measures, particularly from February 1994 until the beginning of 1995, when the Federal Reserve raised the federal funds rate (FFR) incrementally from 3% to 6% despite stable inflation at the time. Economic expansion followed shortly thereafter, which brought unemployment down to below 5%, an achievement not attained since the 1960s. Nevertheless, despite large economic expansion inflation dropped to below 2%, while robust economic growth in the 1990s endured. The success of this strategy in the United States was not without incident. Many times, the policy measures were not communicated to the public, therefore the markets regularly second-guessed the policies taken by the Fed, which triggered financial market shocks and uncertainty concerning the economic outlook. Likewise, The Fed’s accountability in front of Congress and the public was also called into question, considering the lack of clear criteria to evaluate policy performance, which created time inconsistency problems. One potential problem with the strategy was that success depends entirely on the preferences, capabilities and individuals at the top of the central banking organization. Regarding the US experience, Alan Greenspan, as Governor of the Federal Reserve, was considered a strong individual on the back of successful policy to control US inflation, thus garnering public confidence in the Fed. Nonetheless, there are no guarantees that subsequent Governors will be as strong as Greenspan, while the situation faced has perhaps become more complex. In the face of political dynamics, US experience has also shown that central bank leaders play a key role in the success of monetary policy. Experience from the 1970s showed that strong political pressures can spur inflationary policies from the Fed and, using no explicit anchor targeting, the potential problems could return. Political pressures were also observed in 1997, when, after reducing the FFR several times in the previous period, the Fed hiked the interest rate target by 25 bps. Despite the relatively small bump in the interest rate, with robust economic performance at that time, the Fed faced a storm of criticism from Congress and the public in general.

6.5. The Monetary Policy Regime in Indonesia (1968–1998) Three decades prior to enactment of the BI Act (No. 23) of 1999, which provided the foundation for the ITF, Indonesia experienced monetary policy regime changes from one period to the next. Starting with the period of economic stabilization and rehabilitation in the 1960s and the era of oil-based economic growth in the 1970s, as well as the period of economic deregulation, debureaucratization,

Monetary Policy Strategic Framework     187 and liberalization in the 1980s to the period before the Asian Financial Crisis of 1997/98. This section describes the monetary policy regimes implemented in Indonesia during those three broad periods, while the ITF is explored specifically in Chapters 8 and 9.

6.5.1. Economic Stabilization and Rehabilitation (1968–1972) Experience during the period after national independence until the middle of the 1960s provided invaluable lessons on the importance of prudential principles when implementing macroeconomic policy. First, fiscal policy must control the budget deficit at normal levels. To that end, budget spending must be selected stringently and prioritized to spending that stimulates real sector activity, thereby avoiding excessive and non-strategic expenditure. Second, monetary policy may not be used to finance the budget deficit on the fiscal side. Monetary policy must remain focused on controlling inflation and, therefore, printing money to offset the government deficit would threaten price stability in particular and monetary stability in general. Third, fiscal policy and monetary policy must be well coordinated, while maintaining the independence of both institutions to ensure policy synergy in terms of maintaining economic stability for development sustainability. In the subsequent period, namely from the end of the 1960s, economic and financial development were maintained. Initially, government policy prioritized economic stability, which came under threat in the middle of the 1960s. Budget spending was selected judiciously, the government’s budget deficit was under control and financing was sourced from soft foreign loans, which avoided economic instability, especially in terms of controlling inflation. In the monetary sector, no more currency was printed to finance the government’s budget deficit and total money supply was brought under control. Such endeavors were accompanied by the provision of much needed goods and services. Through fiscal and monetary discipline, economic stability was quickly restored, with inflation falling rapidly to single digits, which also restored confidence in the economic recovery. After successfully restoring economic stability, the Government subsequently commenced long-, medium-, and short-term national development planning, which gradually brought structure to the national economy and accelerated growth. The structure of the economy, particularly the monetary and banking sectors, was solidified by the Central Bank Act (No. 13) of 1968. Pursuant to the Central Bank Act, BI was tasked with assisting the Government in two areas as follows: (1) regulate, create, and maintain rupiah currency stability; and (2) promote development, while creating job opportunities to ameliorate public welfare. Monetary policy was formulated by the Monetary Board and BI undertook its monetary policy duties in line with the decisions of the Monetary Board. It is important to note, on one hand, institutional regulation such as this was considered positive because of monetary policy integration and coordination with fiscal policy and other macroeconomic policies. On the other hand, however, such a regulatory framework blurred task focus, discipline and accountability at both institutions when formulating and implementing policy. There was also a lack of checks and balances between monetary policy, fiscal policy, and other

188    Central Bank Policy macroeconomic policies. Furthermore, such institutional regulation still facilitated the exploitation of monetary policy to finance the fiscal side, thereby undermining prudential principles and macroeconomic policy discipline.

6.5.2. Era of Oil-based Economic Growth (1973–1982) National economic growth subsequently accelerated on the back of increasing oil proceeds at the beginning of the 1970s. The discovery of oil fields in Indonesia had advantages and disadvantages. On one hand, oil proceeds provided a windfall in terms of state receipts and could be used to finance routine and development spending on the fiscal side. With an active role and dominated by the government, fiscal policy could stimulate real economic activity. On the other hand, however, increasing oil foreign exchange revenues and government spending also expanded total money supply on the fiscal side, which demanded monetary expansion absorption on the fiscal side to avoid excess liquidity in the economy that could push up inflation. Against that backdrop, the Government began to implement selective credit policy on the monetary side in 1974. The goal was to control total money supply and, thus mitigate rising inflation, primarily by regulating the magnitude of credit expansion permitted by the banking industry, often known as the net asset expansion ceiling. Each year, therefore, BI prepared a national credit expansion plan, paying due consideration to total money supply in line with predicted inflation and output growth. The banks were subsequently required to submit their credit plans to BI for the central bank to set the credit ceiling for the upcoming year for each respective bank. Ultimately, the ceiling for each individual bank was used as the basis for credit disbursements/liquidity provided by BI in line with the predetermined sector/program. Banking regulations also restricted deposit rates and lending rates. At the monetary and banking level, credit and interest rate limits became known as financial repression, which was also practiced in various other countries. Although banking sector conditions were less than favorable due to scarce funding sources after private savings evaporated and restrictions were placed on lending and interest rates during the period of financial repression, investment activity persisted, particularly government investment. Consequently, to provide more wiggle room for banks to utilize funds, especially in terms of disbursing loans to the private sector, BI halved the reserve requirement in 1978 from 30% to 15%.

6.5.3. Period of Economic Deregulation, Debureaucratization, and Liberalization (1983–1997) At the beginning of the 1980s, the oil price slumped on the international market due to the onset of a global recession, which squeezed state revenues to offset the State Budget. Government dominance to support increasing economic activity was no longer tenable and, consequently, the sustainability of national development was threatened. Therefore, the Government instituted a series of

Monetary Policy Strategic Framework     189 policy reforms in the economy to overcome the crisis threat due to tumbling oil prices. The goal was to grow, nurture, and expand the role of the private sector in every aspect of economic life to replace the role of the Government in terms of maintaining national development. As a result, at the beginning of the 1980s, the Government introduced deregulation, debureaucratization, and liberalization policy to various economic sectors, including the banking and financial sectors, trade, investment, and so on. On June 1, 1983, the Government issued bank deregulation policy, signaling an era of liberalization in the banking sector specifically and financial sector in general. Policy focus was to liberalize interest rate setting by the banks as well as limit the net asset ceiling to just the credit disbursement program. This policy prompted rapid banking and financial sector development in Indonesia, by mobilizing private funds not only in the form of demand deposits, saving deposits and term deposits, but also in the form of loans and other types of financing offered by the banking sector to the business community. Likewise, the financial sector experienced rapid development in terms of financial transaction volume as well as the variety of financial products available (stocks, bonds, securities, and derivative products) to trade. Consequently, more funds circulated in the financial sector, which affected the relationship between money, inflation, and output compared to previous periods. Prevailing economic conditions, particularly the financial sector, had fundamental implications on the implementation of monetary policy by BI. Monetary policy, which had been implemented directly using selective credit policy began to shift to more indirect and market-oriented channels, including open market operations to control liquidity in the economy. Monetary controls tended to follow monetary targeting by setting total money supply (M1 and M2) as the intermediate target and narrow money (M0) as the operational target. Meanwhile, money market operations were conducted through auctions of BI Certificates (SBI), which were first issued in 1984 as the main instrument of monetary policy. Rupiah money market intervention helped to control liquidity by providing short-term lending facilities, from overnight to seven days. Money market operations were directed toward achieving the operational target (M0), which itself was geared toward keeping money supply (M1 and M2) under control in line with current projections. Thereafter, driving domestic economic activity in the face of global competition, the Government issued a policy package on October 27, 1988, containing regulations to refine financial, monetary and banking policies.13 Deregulation focused on the ease of opening a bank office or branch as well as expanding bank operations, including foreign exchange activities. Therefore, deregulation a­ ccelerated

13

Unhealthy economic development occurred in 1987, when the public speculated on foreign currencies after expecting the Government to initiate rupiah devaluation. To overcome that problem, the Government introduced policy measures known as “Gebrakan Sumarlin,” which reduced bank liquidity dramatically and dampened speculative activities.

190    Central Bank Policy fund mobilization, lending, product development, and various other banking and financial sector business activities. In addition, a more conducive climate of competition was created in the banking sector through less stringent licensing to establish a new bank or joint venture bank. Concerning the efforts to enhance monetary control effectiveness, the measures introduced included reducing the reserve requirement from 15% to just 2%. In other words, loose deregulation policy precipitated a fundamental change in terms of rapid banking and financial sector development as well as monetary policy implementation in Indonesia. Congruent with rapid banking and financial sector development in Indonesia, new problems emerged in terms of implementing monetary policy, particularly linked to efforts to control total money supply (M1 and M2). Bank operations and products, to mobilize funds and finance the business community, were not only limited to demand deposits, savings deposits, term deposits, and loans, but also diversified into various forms of money market instrument, such as negotiable certificates of deposit (NCD), commercial papers, promissory notes, automated teller machines (ATMs), and so on. On the other hand, capital market development began to accelerate with rapidity in the form of transaction volume and the types of securities traded. Consequently, decoupling between the financial sector and real sector occurred, creating a more tenuous relationship between money supply and inflation and real output, especially in the near term. In addition, as a consequence of financial sector liberalization, the inflow of funds to Indonesia, particularly in the form of private foreign loans, surged rapidly. The influx of funds exploited the current economic boom at that time and was supported by the nascent wave of globalization in the financial sector, trade, and investment. On one hand, the deluge of foreign capital flows closed the saving-investment gap, which drove economic growth and national development. On the other hand, however, the influx of non-resident capital spurred a number of problems too. Most foreign capital was in the form of short-term private foreign loans that were not hedged against currency risk and were primarily utilized to fund long-term private projects that did not produce foreign exchange. On the monetary side, the magnitude and mobility of foreign funds also complicated monetary policy implementation by BI. Striving to prevent the adverse impact of money supply expansion originating from foreign capital flows on higher inflation and rupiah stability, BI absorbed the excess liquidity in the economy, which edged up domestic interest rates. Higher domestic interest rates, however, attracted more foreign capital, primarily in the form of short-term securities. Consequently, total private foreign loans spiraled in various forms and with different maturities. Such conditions were exacerbated further by the implementation of private projects financed by the foreign loans in compliance with good corporate governance, which has been the main cause of crisis since 1997.

6.6. Concluding Remarks In practice, selection of the monetary policy regime by a central bank is based on the perceived benefits of that regime in terms of achieving the monetary policy targets, adjusted to local economic conditions. Nevertheless, there are two other

Monetary Policy Strategic Framework     191 aspects that underlie monetary regime selection, namely policy commitment and prudence (discretion). In this regard, the success of monetary policy regime implementation, in terms of achieving price stability and robust growth in the long term, fundamentally depends on striking the optimal balance between policy commitment and prudence (Stone & Bhundia, 2004). According to a previous viewpoint, Mishkin (1999) classified several monetary policy regimes, as practiced, based on the selection of various nominal anchors, namely exchange rate targeting, monetary targeting, inflating targeting, and no explicit anchor targeting. Of course, the diverse preferences of the various policymakers in different jurisdictions provided a hotbed for other policy regime designs, with different combinations of commitment and prudence. Stone and Bhundia (2004) proposed a new classification of monetary policy regimes, defined as the choice and clarity of the nominal anchor. Departing from the classification proposed by Mishkin, the emphasis of the new classification system, according to Stone and Bhundia, was not only from the aspect of choice, but also clarity of the nominal anchor used. In this case, the different taxonomy of regimes that could be classified based on a combination of commitment and prudence included monetary non-autonomy, exchange rate peg, full-fledged inflation targeting, implicit price stability anchor, inflation targeting lite, weak anchor, and money anchor. Monetary non-autonomy is a regime where a country does not issue currency independently but is tied to certain regulations, for instance in the form of a currency board arrangement. The exchange rate peg uses a nominal anchor in the form of the domestic exchange rate to other currencies, usually trading partner currencies with low inflation. This regime is the focus of exchange rate targeting classification according to Mishkin. Meanwhile, the two iterations of inflation targeting are full-fledged, where there is clear and formal commitment to the inflation target, as well as inflation targeting lite, where the commitment is not fully toward achievement of the inflation target but also other variables, such as the exchange rate and monetary targets. Fundamentally, the implicit price stability anchor is an eclectic regime with the same characteristics as the implicit nominal price anchor per Mishkin’s classification. Likewise, the money anchor regime is the same as monetary targeting, where monetary aggregates are used as the nominal anchor. Finally, the weak anchor regime is aimed at countries without a nominal anchor in practice, where inflation is typically north of 40% year on year. The gradation of regime classification implies that the empirical analysis performed by Stone and Bhundia stressed a trade-off between commitment and prudence, induced by a shift or refinement in the policy regime, thereby creating space for the emergence of regimes with a new classification. Currently, there is a tendency among countries using the implicit price stability anchor to increase their commitment to achieving the inflation target, thus giving the impression of full-fledged inflation targeting. On the other hand, however, as credibility is more important when applying a full-fledged inflation targeting regime, there is a trend toward softening commitment to the inflation target, thereby increasing the element of prudence in terms of seeking output growth and financial system stability. Therefore, what occurs is a regime that stands somewhere in between implicit price stability and full-fledged inflation targeting.

This page intentionally left blank

Chapter 7

Monetary Policy Operational Framework 7.1. Introduction In line with the nascent issue of “duality” in terms of achieving the final goal of monetary policy, namely output (economic growth) stability or price (inflation) stability, as explored in Chapter 6, the longer discussion on how monetary policy should be implemented also occurs in theory and practice. When controlling price or output stability, the central bank has two options. It can apply money supply (monetary base) or the interest rate as the operational target. In relation, a critical review by Poole (1970) found that during certain periods, the monetary authority was unaware of the types of shocks or level of output in the economy. Therefore, during such periods, the monetary authority would have the option to target money supply or the interest rate, while accommodating demand for currency. Furthermore, in terms of the monetary policy operational framework, the monetary authority would also have to consider other equally important factors, including the monetary policy response itself. To that end, intensive discussions since the middle of the twentieth century have highlighted the importance of applying rules or discretion, representing opposite ends of the same spectrum, as a reference for the central bank to determine an optimal monetary policy response. Naturally, the selection of policy instruments and the most appropriate monetary policy response to prevailing economic dynamics is not a simple task. It is considering that in the monetary policy implementation the monetary authority is not only faced with a range of contextual alternatives, but also factual. In practice, therefore, it is very possible that a central bank may be less assertive concerning the foundation of the policy that must be instituted. In order to explore that issue further, this chapter is organized into six parts. After the introduction, this capture presents the conceptual dimensions and theoretical models concerning the existence and role of operational targets, instruments, and the policy response in the monetary sector. The section will also cover aspects relating to the formulation of an optimal monetary policy framework. The following section will highlight monetary management as practiced in several countries as well as in Indonesia. This chapter will end by presenting hypotheses on state-contingent rules to formulate an optimal monetary policy response, especially for developing economies.

Central Bank Policy: Theory and Practice, 193–221 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191011

194    Central Bank Policy

7.2. Conceptual Dimension and Theoretical Models 7.2.1. Policy Instruments: Money Supply Versus Interest Rate In general, from the various literature, it can be concluded that the selection of optimal monetary policy instruments is determined by, among others, the origin of the shock and how the monetary policy transmission mechanism functions in an economy. An intuitive elaboration of the issue was presented by Poole (1970). According to Poole, when selecting the optimal policy instrument, the monetary authority does not necessarily know the confronted shocks faced or the current level of output, and is thus faced with the decision to target money supply or the interest rate, while also accommodating demand for currency. The standard Poole model fundamentally only accommodates demand-side shocks in the form of monetary shocks and spending shocks. Over time, however, several variations of the Poole model have been developed to include supply-side shocks and the rational expectations theory (Beare, 1978; Walsh, 2001) as follows: 7.2.1.1. Standard Poole Model with Demand-side Shocks. Analyzing the ­conditions where each respective instrument is appropriate, Poole applied a simple Keynesian approach for goods market (investment savings (IS)) and money market (liquidity preference money supply (LM)) equilibrium as follows:

Y = Yr r + u (1) M = LyY + Lr r + v

The two equations represent the investment savings - liquidity money (IS-LM) equation with respective stochastic residuals of u (spending disturbance) and v (monetary disturbance). Meanwhile, the expected loss function of the monetary authority is targeting output stability is formulated simply as follows1:

E ( L ) = E [(Y −Y *)2 ](2)

In this case, further analyses were conducted to compare the expected loss function of monetary authorities when applying monetary policy through money supply or the interest rate, such that the level of output or income would arrive at the expected level. In the deterministic model, where σu2 = σv2 = 0 , both choices are equivalent but different in the stochastic model. If the interest rate is determined in r* and Y*  = Yrr* thus Y = Y* + u, then the expected loss function can be expressed as follows:

1

E ( L )r = r* = E (Y −Y *)2  = E ( u )2  = σu2

(3)

During initial development of the Poole model, the objective function was formulated referring only to the final target, namely output stabilisation, because that was the most popular issue at the time.

Monetary Policy Operational Framework     195 If the monetary authority targets money supply, the solution to the IS–LM equation is as follows: M − LyY − v r= Lr (4) L u −Yr v Y =Y * + r Lr +Yr Ly and the expected loss function is: E ( L ) m= m* = E [(Y −Y *)2 ]

2  2 2 2 2  Lr u −Yr v   Lr σu − 2ρ LrYr σu σv +Yr σv (5) = E    = ( Lr +Yr Ly )2  Lr +Yr Ly    

where ρ is the correlation coefficient between u and v. Therefore, the strategy employed will be the policy that incurs the lowest level of loss, namely by comparing Equations (3) and (5) as follows:

E ( L ) m= m* L2r − 2ρ LrYr σv / σu +Yr2 σv2 / σu2 (6) = ( Lr +Yr Ly )2 E ( L )r=r*

If the ratio is less than 1, the policy selected is M = M*. In contrast, if the ratio is more than 1, the policy selected is r = r*. If σv / σu = Ly and ρ = −1, then the numerator and denumerator values will be the same or have a ratio equivalent to 1. Empirical evidence is available by referring to the behavior of demand for currency as well as demand for goods and services. If the demand function for currency is relatively stable compared to the demand function for goods and services, then σv (monetary shocks) will be smaller than σu (spending shocks), thus policy strategy using total money supply is superior to interest rate instruments to achieve the final target of monetary policy, and vice versa. 7.2.1.2. Poole Model Variations with Supply-side Shocks and the Role of ­Market Expectations.  Fundamentally, the standard Poole model only calculated demand-side shocks (from the IS–LM equation). Poole model variations, however, also take into consideration supply-side shocks, while alternative strategies price targeting can be formulated through the demand–supply model as follows:

Y = −αr + u M − P = βY − λr + v (7) Y = γP + w

where the final equation is the aggregate demand equation. The stochastic residuals, u and v, are respective demand-side shocks, while w represents shocks on the supply side. When calculating the influence of supply-side shocks, w than u and v, are assumed to be zero, while the expected loss function of the monetary authority when setting the price stabilization strategy is as follows:

E ( L ) = E [( P − P*)2 ](8)

196    Central Bank Policy Therefore, if the monetary policy strategy is to achieve output stability, then through the stages of finding an equilibrium solution as previously done, we get:

E ( L )r = r* = E [(Y −Y *)2 ] (9) = E [( u )2 ] = σu2 E ( L ) m= m* = E [(Y −Y *)2 ]



  αw  = E   α + γ ( αβ + λ )  

2

 (10) α2 2 =  [ α + γ ( αβ + λ )]2 σw 

Meanwhile, if the monetary policy strategy is to achieve price stability, the following equations are used:

E ( L )r = r* = E [( P − P*)2 ]

(11) = E [(−w / γ )2 ] = σw2 / γ 2

E ( L ) m= m* = E [( P − P*)2 ]

 ( αβ + λ )w 2  (12) ( αβ + λ )2   =  = E  σw2   2  α + γ ( αβ + λ )  + + α γ αβ λ [ ( )]  

The optimal monetary policy strategy is selected the policy which resolution the smallest loss, namely by comparing Equations (9) and (10) if the strategy is output stability, and by comparing Equations (11) and (12) if the strategy is price stability. Therefore, the following can be concluded: (i) If output stability is the target, targeting the interest rate is superior to targeting the money supply, because:

α2 σw2 > 0 (13) [ α + γ ( αβ + λ )]2

(ii) If price stability is the target, targeting the money supply is superior to targeting the interest rate, because:

E ( L ) m= m* [ γ ( αβ + λ )]2 = < 1(14) E ( L )r= r* [ α + γ ( αβ + λ )]2

The model was developed further by taking into account rational expectations. Blanchard and Fischer (1989) applied nearly an identical model framework as that used by Sargent and Wallace (1975), a supply–demand model as follows: yt = −b[it − ( E [ pt +1 | t ] − pt )] + zt

mt − pt = yt − ait − vt yt = β ( pt − E [ pt | t −1] + ut )

(15)

Monetary Policy Operational Framework     197 where the last equation is the aggregate supply equation. It was assumed that the three shocks are white noise, and total money supply approaches zero, thus E [ pt | t −1] and E [ pt +1 | t ] are zero. If the final target of monetary policy is output stability, the solution for each respective instrument, money supply and the interest rate, is as follows:

yt ( m |m= 0 ) =

{βbvt + b(1+ a )β ut + a β zt } (16) [ a β + bβ + b + ba ]

yt (i |i = 0 ) =

{βbut + β zt } (17) (β + b)

The two equations generate three solutions.



First, it is conclusive with the Poole analysis, namely that if the policy aims to dampen output fluctuations, and if the primary source of uncertainty is currency demand shocks (v), then an interest rate instrument would be superior because when applying such a strategy, output is not influenced by the shocks. Second, the output response to demand-side shocks of goods and services (z) is larger when applying an interest rate strategy than the same response using a money supply strategy. Third, the output response to supply-side shocks is larger in the interest rate strategy if b > 1 because such conditions guarantee a flatter aggregate demand curve than aggregate supply.

7.2.2. Policy Response: Rules Versus Discretion The policy instrument, money supply or the interest rate, to achieve the final target, either output stability or price stability, can be selected in different ways relating to the selection of optimal policy measures in response to economic fluctuations. The conceptual background also implies that the policy variables of macroeconomic stabilization policy can be influenced based on rules that consider permanent feedback in the relationships among the economic variables. That is an alternative to active strategy, or discretion, which is based more on certain considerations and assessments as well as fine-tuning by the policymakers. An alternative way of understanding discretion is as an anti-rules strategy. Fundamentally, analysis of framing the response is the manifestation of the long-standing debate known as “rules versus discretion.” Interpretation and the extreme differences of “rules versus discretion,” the terminology and approach of which were first developed by Kydland and Prescott (1977) and echoed by Barro and Gordon (1983), have also fundamentally experienced a shift in line with

198    Central Bank Policy the ongoing debate over the past few decades.2 Analytically, Barro and Gordon (1983) found that a rule-based monetary policy response (or selecting the policy instruments based on rules) is more appropriate in response to current dynamics, which have already been taken into account in the previous formulation of policy instruments. In contrast, a discretion-based monetary policy response is based on evaluations over time of ongoing dynamics, while disregarding past developments and policies. Meanwhile, Taylor (1993), explained that, departing from discretion-based policy, rule-based policy is symmetrical, meaning that it is based on methodologies and planning, not casual and random measures.3 The debate on “rules versus discretion” also takes into consideration the formulation of more rigid rules, where the instrument variables are determined using a constant value, not feedback. This implies the presence of non-active policy, namely policy that is not adjusted in line with the short-term stabilization needs. Per the Keynesian framework, monetary policy is best implemented actively, and is directed by a formal feedback rule or discretion as well as an informal assessment by the policymakers while considering the economic stabilization goals. In contrast, Friedman and other monetarists referred more to the k-percent rule, where money supply is increased based on a constant percentage over a certain period. The standard version of the rule is based on the Quantity Theory of Money (QTM), which implies that the central bank must maintain money supply growth at a minimum of k%, even in the near term; while the soft version allows deviation over a longer period. 2

The debate on “rules versus discretion” now refers to a new argument that highlights the time-inconsistency problem when framing policy strategy. The time-inconsistency problem refers to the differences in (optimal) policy measures announced by the central bank to the public – if the central bank’s credibility is sound – with the upcoming policy measures to be implemented by the central bank after the public takes a decision based on their expectations (Kydland & Prescott, 1977). For example, the central bank announces its intention to achieve a given inflation target, and the public executes contracts or work agreements based on that announcement. Under such conditions, the central bank has an incentive not to fulfil its promise by seeking to achieve stronger output growth, which would exacerbate inflationary pressures. Ultimately, however, the public would find out what happened and would raise their inflation expectations, which would constrain real output. If this chain of events was to reoccur, inflationary bias would emerge, a situation when real output does not increase but inflationary pressures heighten. The issue of monetary policy credibility and time inconsistency is elaborated in Chapter 10. 3 Although both responses can be developed based on the optimisation method, optimisation for rule-based policy and discretion-based policy are completely different. Woodford (1999) stressed that, departing from discretion-based policy that uses evaluations over time, the characteristics of rule-based policy are grounded in rules with a timeless perspective, namely that the rule is not influenced by current macroeconomic developments. If new information is obtained, however, which causes the central bank to revise its rules, the adjustment can be made without creating a bias in the achievement of the policy target (McCallum, 2000).

Monetary Policy Operational Framework     199 The preference for a constant rule is based on the consideration that economic conditions, fundamentally, are unstable, thus making forecasting to frame stabilization policy extremely difficult. Consequently, it is possible that active stabilization policy could destabilize the economy due to imprecise policy timing and uncertainty. Assuming rational expectations, however, uncertainty leads to policy effects that were not anticipated by the real sector. Conversely, a constant rule can steer public perceptions and, thus, the direction of monetary policy, thereby reducing output fluctuations. 7.2.2.1. Feedback Rule and Optimal Monetary Policy. Through development to the present day and consistent with money demand function instability (money multiplier), the use of constant rules, standard or soft, has started to fade, replaced by feedback rules. Furthermore, due to market imperfections and uncertainty, feedback rules when framing the monetary policy response are believed to achieve the policy goal more optimally. An important initial contribution was put forward by Theil (1961), based on the analysis conducted by Tinbergen (1952). The simple version is as follows.4 Assuming that the relationship between the final policy goal (policy variable) and policy instrument (instrument variable) can be expressed in a stochastic equation, namely: Y = α + βX + u



= Y p + u (18)

where Y and Yp are, respectively, the actual and target values of the final policy target variable (GNP for instance). X is the policy instrument variable, which is under the full control of the monetary authority. α and β are parameters assumed to be accurately known, while μ is the stochastic variable with an average value of zero and variance of σμ2. It is assumed that the loss function of the monetary authority can be expressed as follows: L = ϕ1 (Y −Y *)2 + ϕ2 ( X − X *)2(19)



where φ1 and φ2 are the fixed weights, and Y* and X* are the respective desired levels of the final target and policy instruments. Considering that μ is a stochastic variable, then Y is also a stochastic variable, thus the loss function is expected as follows: E ( L ) = E [ϕ1 (Y p + u −Y *)2 + ϕ2 ( X − X *)2 ]



(20) = ϕ1σu2 + E [ϕ1 ( α + β X −Y *)2 + ϕ2 ( X − X *)2 ]

Optimal policy is determined by minimizing the loss function, with the following first-order condition: X=

4

ϕ1β ϕ2 (Y * −α ) + X *(21) ϕ1β 2 + ϕ2 ϕ1β 2 + ϕ2

For clarification, refer to Beare (1978).

200    Central Bank Policy Fundamentally, the equation is an optimal linear decision rule, showing that the optimal value of X is a linear function of the desired level of the final target (Y*) and policy instruments (X*). If no costs are incurred using policy instruments, φ1 = 1 and φ2 = 0, then the optimal conditions are as follows:

X=

(Y * −α ) (22) β

From the two equations above, the inclusion of uncertainty through the stochastic variable μ, will influence the desired level of the loss function. On the other hand, however, such conditions do not affect the rule to determine the optimal policy taken, thus the monetary authority implements a strategy of certainty equivalence. Under extreme circumstances, the feedback rule can be determined using target-based rules, for instance the inflation target and nominal-income target, or through instrument-based rules such as the money growth rule and interest rate rule. Therefore, variations when stipulating an optimal monetary policy rule are fundamentally based on linkages between instruments and the monetary policy target, for instance in the form of correlation between the structural equations derived from the simple monetary policy transmission model, combined with the quadratic objective function that analyses the costs-benefits of achieving the policy target. To fulfill the criteria as an ideal rule, the correlation must be simple, thus the solution of the corresponding optimization problem can easily be understood and resolved. 7.2.2.2. Simple Feedback Rule.  One monetary rule, in the form of a simple feedback rule, is the monetary growth rule or McCallum rule. Where monetary growth reflects the policy response to the average change in base velocity, as a correction to the ongoing changes due to amended regulations and technological advances, as well as the cyclical development of nominal output (McCallum, 1988). The McCallum rule can be expressed as follows:

∆bt = ∆x * −∆vta + λ (∆x * −∆xt−1 )(23)

where bt is the total monetary base, xt and x*t are the nominal output and desired level respectively, and Δx* is the constant parameter that reflects the average nominal output growth per period in a year. Δvat is the average growth of base velocity, where vat can be expressed as (1/k)Σi=1,.., k (xt-I – bt-i). Meanwhile, λ is the positive parameter that reflects the stimulative effect of cyclical nominal output on base money growth. Although the formula is very intuitive, due to the instability problem of linkages between the monetary aggregates and the final policy target (output), the McCallum rule is less desirable. As a replacement, however, another monetary policy rule has gained attention recently, namely the interest rate rule or Taylor

Monetary Policy Operational Framework     201 rule, where interest rate developments reflect the response of output and inflation (Taylor, 1993). The original formula of the Taylor rule is as follows:

it = i * +∆pta + α ( yt − y *) + β (∆pta − π *)(24)

where it is the short-term interest rate as a monetary policy instrument, yt is real output, Δpat or πt is average inflation, i* is the long-term interest rate equilibrium, Y* is the potential output, and π* is the inflation target. Meanwhile, α and β, respectively, are the positive parameters that reflect the impact of real output and inflation on the interest rate. According to Ball (1997) and Svensson (1997a), the original version of the Taylor rule is not an ordinary equation but fundamentally reflects feedback that can be resolved through optimization by the central bank, calculating the Phillips curve (backward-looking) on the supply side and IS curve on the demand side. Meanwhile, the general objective function of monetary policy is to minimize the loss function, namely the weighted expectation value of output deviation from its potential and inflation from its target. Nonetheless, the Taylor rule equation is not, fundamentally, forward looking. The interest rate response to output deviation from its potential (output gap) and inflation deviation from its target, are contemporaneous. The response to contemporaneous shocks aims to stabilize current output and inflation, which is not relevant considering the time lag of monetary policy (Svensson, 1997a). Meanwhile, in the context of an open economy, the Taylor rule equation can also be modified with other variables, particularly changes in the exchange rate due to the influence of certain exogenous variables (Ball, 1997). The design of monetary policy rules varies greatly in line with the theoretical background it is based on. Clarida, Gali, and Gertler (2000), for instance, included interest rate smoothing and exogenous shocks in the system. Meanwhile, in their study of the inflation targeting framework (ITF) for Canada’s economy, Macklem-Srour (2000) used a forward-looking orientation for an explicit horizon of six to seven quarters, referring to the monetary policy lag and efforts to mitigate excessive output shocks. In general, the above rules have failed to include possible non-linearity or asymmetry in the Philips curve. Schaling (1999) developed the inflation-forecast-based rule proposed by Svensson (1997a), by calculating a convex Philips curve. Using the optimal control technique, the derived asymmetric monetary policy rules are expected to correct interest rate determination in the inflation-forecast-based rule that has a tendency to underestimate.

7.2.3. Several Aspects of Optimal Monetary Policy Formulation In terms of utilizing monetary policy instruments, money supply and the interest rate, in general, are not targeted rigidly. Similarly, the policy response, using rules or discretion, is more flexible in its implementation. Irrespective of whether there are inconsistencies between the theoretical conclusions and empirical practices, it is generally understood that such inconsistencies are due to two factors,

202    Central Bank Policy namely inconsistent theoretical assumptions and the unique fundamentals of an economy that underlie the behavior and patterns of linkages between the economic variables. In relation to the use of rules, the benefits or advantages of applying a pure (unmodified) policy rule can be viewed from at least three aspects, namely transparency, consistency, and predictability as follows:



First, the application of a rule provides a mechanism to analyze how committed the monetary authority is to that rule because, technically, the rule can be verified by economic players. Second, the application of a rule guarantees policy consistency, thus, if the monetary authority is credible, the application of a mechanistic rule will reduce uncertainty surrounding the measures available to the monetary authority. Third, affects predictability. Although comparatively simple, the application of a rule when analyzing monetary modeling, in general, generates sound recommendations for the monetary authority.

On the other hand, the disadvantages of applying a pure policy rule are as follows:





First, the results generated may not be optimal because the design of a rule only responds to a small part of the information on economic dynamics. Such conditions are reflected by the Taylor rule, which, in general, explains no more than two-thirds of empirical interest rate developments (Judd & Rudebusch, 1998). Second, there are many situations of deviation from the rule that necessitate judgment. In fact, Taylor (1993) himself conceded discretionary departures from the rule. Consequently, a rule is considered incomplete because it does not accommodate the misaligned rule. Third, although empirical studies support the need to apply a policy rule, for instance the Taylor rule, no benchmarks are available to test the success of the rule considering that no central banks are fully committed to the rule (Svensson, 2002).

Consensus was reached after a long debate between economists in relation to the selection of policy response, rule versus discretion, stating that the central bank cannot fully implement monetary policy based on discretion. Conversely, rules are a prerequisite for the implementation of sound monetary policy, thus the application of monetary policy without using a certain rule could possibly lead to the opposite. As alluded to previously, when using rules, economists now traditionally focus their observations on two types of rule, namely money growth rules (McCallum, 1988) and interest rate rules (Taylor, 1993). The best type of rule is a question that remains unanswered. Nonetheless, it has generally been agreed that rule-based policy can be applied with a certain level of discretion. The opposite is also true,

Monetary Policy Operational Framework     203 although ideal conditions have been met, discretion is still recommended, while paying due consideration to the rules. In the design of policy rules, namely growth rules and interest rate rules, two distinct ways of thinking are prevalent. First, policy rule design fundamentally reflects linkages between the final policy goal (output and prices) and the operational target or policy instrument (base money and short-term interest rates). In general, if output growth is selected as the final policy target according to GDP nominal targeting, then the monetary growth rule will be the first choice. Conversely, if price stability is selected as the final policy target per the ITF, then the interest rate rule will be the first choice. Nevertheless, McCallum and Nelson (1999) stressed that the role of the final target indicator in the designs of both rules are interchangeable, thus allowing the possibility, for instance, of designing a rule with the monetary base as the operational target/instrument, where its performance is a response to output and inflation. Similarly, it is also possible to set the interest rate rule based on feedback from output developments. Second, the analysis of monetary policy rules does not have to reflect optimal central bank behavior. If the objective of the analysis is to seek an optimal policy, then the design of the policy rule should be produced from optimization measures that refer to the function of the central bank’s goal, which is based on public utility or behavior; where, in practice, no central bank explicitly states the objective function. Therefore, not all analyses recommended for use must assume optimal measures from the central bank. In this case, it is reasonable to expect positive analysis, examining the impact of hypothetical rules that reflect various alternative approaches, is more beneficial (McCallum, 2001). Irrespective of the diverse views described above, understanding the relationships between output and price stability play an extremely important role in the analysis of monetary policy, at least from the perspective of two concepts, namely to explain the influencing factors (determinants), while simultaneously forecasting inflation, as well as to formulate an optimal monetary policy framework with the single target of price stability. Technically, the essence of the first concept relates directly to elements of uncertainty and forming market expectations to influence the dynamic behavior of inflation. Therefore, careful observation of both elements through an appropriate econometric approach is crucial. Meanwhile, the essence of the second context relates more to which monetary policy is required and how to implement it. The policy framework selection strategy is based on various considerations or choices, for instance the choice between credibility and flexibility as well as between forward-looking or backward-looking, and the need to consider a cost-benefit analysis of disinflationary policy by the central bank, especially in relation to the ever-present problem of output variability when pursuing price stability. Therefore, the monetary policy framework selection strategy should fundamentally be directed toward an optimal monetary policy framework, namely policy that is based directly on the assumption of optimal behavior among the economic agents. To conclude that a policy is optimal, an evaluation is typically required of changes in the welfare of each individual economic agent because of applying the alternative policy, which can be derived through a simple modeling system or

204    Central Bank Policy general equilibrium analysis. Theoretically, several studies, including Theil (1961), Ball (1997), Svensson (1997a), Rudebusch and Svensson (1998) as well as Blake, Weale, and Young (1998), showed that optimal policy is generally formulated by considering (at least) the basic elements, namely a cost-benefit analysis of applying the policy, for instance output or price stabilization, based on optimization measures in the decision-making process of the central bank. In this regard, the analysis could be described with reference to the loss function that must be minimized by the central bank, with the determination of policy instruments as the subject and several constraint functions relating to intervariable structural linkages in the economic system. Based on such conditions, optimal monetary policy can be mathematically resolved as an optimal reaction function by the central bank, otherwise known as the optimal decision rule or optimal policy rule, which stipulates the presence of policy instruments as a function of all the information that is available. Independent of the basic elements above, to achieve the best policy application framework and to facilitate monitoring, policymakers specifically need to consider two important aspects: First, forward-looking policy while paying due consideration to the time lag of monetary policy (Svensson, 1997a). In this respect, Svensson submitted a proposal on implementing monetary policy based on inflation forecasts rather than actual values. The advantage of this theoretical approach, in addition to producing a response equivalent to optimal conditions, is to increase monetary policy credibility by focusing on inflation expectations. Second, informational requirements in policy formulation while paying due consideration to the short-term behavioral adjustment mechanism (Blake, 2000). In this regard, central banks were urged to adopt more generic policy rules, considering the factors used to make the forecasts. With the possibility of using a more detailed model specification, per Blake, model representation with an error correction mechanism could produce the best policy control framework. In the study, Blake et al. (1998) showed that a simple rule, for instance the Taylor rule, is inferior to other, more optimal and more complex rules, estimated paying due consideration to deviation from the simple rule in the event of an economic shock. The fundamental issue is how to implement an explicit monetary policy target in the utilization of instruments and choice of policy response. This is linked to the empirical condition that the actual economic behavior observed may appear consistent with the behavior of policymakers, based on optimizing the simple or relatively more complex policy response. With the conditions outlined above, studies into how monetary policy should be applied optimally in a realistic economic modeling framework, congruent with the economic characteristics or fluctuations, are a strategic choice. From the perspective of monetary policy directed toward inflation targeting, where the principles of policy transparency and credibility are imperative, the selection of rule-based policy is considered more relevant than discretion-based policy (Hubbard, 2002; McCallum, 2000). To that end, an alternative approach to study the application of optimal monetary policy is to look at alternative ways to determine the optimal monetary policy rule, namely by considering the aforementioned elements and forward-looking policy procedures.

Monetary Policy Operational Framework     205

7.3 Application in Several Countries 7.3.1. Operational Targets and Monetary Policy Instruments In terms of monetary policy implementation, the central bank must choose an operational target along with monetary policy instruments to achieve the policy target. In line with the common operational framework, the policy instruments must be under the control of the central bank, while the transmission of changes in policy instruments to changes in the operational target, and the attainment of the final policy target must be managed well. In this context, the choice of the operational target and policy instruments depends on the financial market structure in each respective country. 7.3.1.1. Operational Target: Monetary Aggregates Versus Interest Rates. In practice, the central bank may choose alternative operational targets, including short-term interest rates, monetary aggregates, and exchange rates. Departing from trends in advanced countries (Australia, Canada, UK, and Japan), where short-term interest rates are the preferred operational target, but where monetary aggregates were previously used, developing countries, at least until the early 2000s (including México, India, Thailand, and Indonesia) continued to apply monetary aggregates. It is important to note that in line with commitment to apply Inflation Targeting, South Korea and Thailand also take short-term interest rates into consideration, along with bank reserves, as the operational target. In Indonesia, during the transition to inflation targeting through to July 2005, the operational target was still monetary aggregates, while short-term interest rates were used as a supporting operational target. Exchange rates are generally only used as a supporting operational target. Nonetheless, only countries with very open economies, such as Singapore and Hong Kong, still apply exchange rates as the operational target. Monetary policy, as applied in several developing countries using bank reserves or broader indicators, such as the monetary base, is starting to garner attention. This possibly reflects the strong perception that bank reserves or base money can be used to predict broader monetary aggregate trends, such as M1, M2, and credit. Another interpretation is possibly that signals from price indicators are not as sufficiently strong as in advanced countries, where the financial systems are more established and stable. Particularly in countries that have experienced an economic crisis, including México, Thailand, and Indonesia, the drastic switch to interest rates when vulnerable to inflationary pressures would trigger shocks or distortions in terms of monetary policy implementation. Congruent with the domestic structural economic changes and significant financial sector deregulation/liberalization policy, many central banks have finally realized that the monetary base is quite fluctuative and does not always accurately mirror M1 or M2 developments or economic dynamics in general. In addition, financial sector deregulation/liberalization tends to expand the role of the interest rate in the monetary policy transmission mechanism. Therefore, nearly all central banks ultimately selected short-term interest rates as the operational target, or at least as a supporting operational target, when applying monetary policy day-today. Among the few countries using monetary aggregates, interest rate fluctuations are still carefully taken into account to monitor daily liquidity.

206    Central Bank Policy Experience in México is the exception. In general, interest rates do not play a role in the monetary policy framework of México, even during the currency crisis in 1994/95. Strengthening the value of the peso, the Bank of México targeted the settlement balances of commercial banks at the central bank (positive or negative). Therefore, interest rates and exchange rates were left to fluctuate in line with market forces. Experience of this operational framework through to the end of 1998 showed that changes in the cumulative reserves target can drive faster and more significant price adjustments on the money and foreign exchange markets. Nonetheless, such conditions were disrupted when the financial sector in México faced disruptions at the end of 1998. The use of operational targets and supporting operational targets in several countries is presented in Table 7.1. In line with the growing trend of using short-term interest rates as the operational target, the question of which interest rate is best to use has emerged. In general, advanced countries and several developing countries apply the overnight (O/N) interbank rate as their first choice because the O/N rate represents the shortest term or maturity in the term structure of interest rates, which can be controlled relatively easily by the central bank. In addition, the O/N interbank money market is very liquid, with a dominant transaction volume in the economy. Theoretically, in its role as supplier or through its ability to influence demand for bank reserves through various policy instruments, the central bank can influence the O/N rate relatively accurately as well. According to the expectations hypothesis Table 7.1:  Monetary Policy Operational Targets (2015). Main Target

Supporting Targets

Advanced Countries United Kingdom Overnight interest rate

–

United States

Overnight federal funds rate

–

Germany

Monetary Base

Money market rate

Japan

Overnight interbank rate

–

Australia

Overnight interbank rate (cash rate)–

Canada

Overnight interbank rate

–

Emerging Market Economies México

Overnight interbank rate

–

Brazil

Overnight repo rate

–

India

Overnight repo rate

Repo/reverse repo

South Korea

7-day repo rate

Exchange rate

Thailand

1-day repo rate

–

Overnight interbank rate

–

a

Indonesia

Source: Several IMF and BIS publications. a Interest rate used as the operational target since July 2005.

Monetary Policy Operational Framework     207 of the term structure of interest rates, changes in the O/N rate are transmitted to longer tenors, including deposit and lending rates. Ultimately, policy transmission through the interest rate channel is expected to effectively influence real sector activity. Many central banks, however, tend not to focus exclusively on the O/N rate because it might not play a dominant role in the monetary policy transmission mechanism based on the structure and characteristics of the respective financial system. In this case, a central bank exercising control over the O/N rate may not produce the expected results for all maturities and, similarly, in terms of economic performance. Consequently, longer tenors are considered more relevant.5 The Bank of England in the 1990s, for instance, utilized short-term interest rates with a maturity of two weeks because the base rates of commercial banks correlated closely with relatively longer money market interest rate maturities. Nevertheless, the Bank of England currently uses the O/N rate. Considering the emergence of the FFR as the money market benchmark, it has naturally become the focus of the Federal Reserve when implementing monetary policy in the United States. Similar practices are also applied in several emerging market economies (EME), reflected in the tactical selection of each respective central bank, where the O/N interbank rate is typically favored. Another interesting phenomenon is the proclivity of all central banks that apply full-fledged inflation targeting, with the exception of México, to use shortterm interest rates as the operational target, with maturities ranging from O/N to 3 months. This trend implies that such central banks prefer to accommodate fluctuations in bank reserves or settlement balances. In this regard, a number of central banks adjust the short-term interest rate by considering the response to deviation in terms of inflation or inflation expectations to the target and output gap.6 In practice, however, the central bank complements that approach with judgment based on all information that supports the inflation forecasts. Meanwhile, central banks in EME rely more on qualitative information because model development in emerging markets tends to contain a comparatively larger amount of uncertainty than models developed in industrialized nations. 7.3.1.2. Monetary Control Instruments. In terms of the monetary policy instruments used, in practice, there are two types of instruments employed by central banks around the world, namely direct and indirect market-based instruments. The choice of policy instrument by each central bank depends on

5

Conceptually, using interest rates with longer maturities (than overnight) has its own disadvantages. First, it is possible that the central bank wishes to dominate the supply side and demand has dwindled, therefore control over bank reserves would also decline significantly. Second, because the central bank uses information on interest rates to gauge market expectations, the central bank pays less attention to the actual formation of interest rates. Third, a concentration on longer maturities tends to exacerbate overnight interest rate shocks. Refer to Van ’t Dack (1999). 6 Interest rate determination based on the response is fundamentally similar to the Taylor rule.

208    Central Bank Policy respective financial system development. At the end of the 1970s, central banks in advanced/industrialized countries began to move away from direct instruments for monetary control in favor of indirect instruments. This trend was followed in subsequent years by developing countries and transition economies7 due, primarily, to increasingly open global economic dynamics, which drove financial market development and subsequently made direct instruments ineffectual because of inefficiencies and the trend of disintermediation. Naturally, the transition from direct to indirect policy instruments required indepth preparations. The IMF, in 1995, published the results of a study regarding the experiences of 19 developing and transition countries that had changed from using direct to indirect instruments. The transition period was assumed to begin when central bank or government securities were sold (auctioned) for the first time and ended when interest rate policy controls were terminated and the credit program accounted for less than 25% of total credit in the economy. The study concluded that the average transition period was 3.7 years from direct to indirect policy instruments, substantiating the fact that a protracted period of time is required to refine the expertise and institutional arrangements to use indirect instruments, as well as the behavioral change required in terms of interest rate determination and competition on financial markets. In addition, other supporting policies are required, including payment system development, enhanced bank supervision quality, overcoming the barriers to competition in the banking sector, and technical development of monetary analysis. In practice, the common direct instruments include credit ceilings, controlling interest rates, direct credit, and statutory liquidity ratios. On the other hand, the most common indirect policy instruments include primary reserves, reserve requirements, discount facility and open market operations (OMO). Besides, rediscount facility, credit auctions, and foreign exchange operations (swaps) are also commonly used.8 Nonetheless, the main instruments used as the backbone of monetary policy implementation are somewhat different. Most central banks apply OMO as the primary policy instrument, while some jurisdictions have used other instruments, such as the discount facility applied in Brazil. The widespread use of OMO by most central banks in advanced as well as developing countries is understandable because OMO are very flexible and the involvement of banks and brokers (second parties) is not binding. Furthermore, central banks can control the frequency of OMO and the number of auctions desired, which is very beneficial in terms of stabilizing the monetary base or short-term interest rates. In addition, of no less importance is that OMO do not place a tax burden on the banks. The main instrument of OMO are government debt obligations, such as T-Bills, complemented with central bank securities. The

7

For more information, refer to “The Adoption of Indirect Instruments of Monetary Policy,” IMF Occasional Paper No. 126, 1995. 8 A detailed review of monetary policy instruments is available in “Monetary Control Instruments,” by Ascarya, Centre for Central Banking Studies and Education, Bank Indonesia, 2002.

Monetary Policy Operational Framework     209 use of government securities will considerably alleviate the cost of OMO borne by the central bank. Table 7.2 presents the general monetary policy instruments used in various countries. From the table, developing countries tend to follow advanced countries in terms of selecting monetary policy instruments consistent with the respective level of financial market development. Nevertheless, developing countries generally also use other supporting instruments in relation to the transition from direct to indirect instruments per the level of development in the respective jurisdiction. For example, India used a Switching Facility instrument until 1992. Through the switching facility, the Reserve Bank of India (RBI) purchased notes at a low coupon rate and sold/replaced them with notes at a higher coupon rate in order to improve the portfolios of banks and other financial institutions. In addition, more specific policy instruments are still applied to attain certain targets. Thailand, for instance, uses the credit ceiling to allocate financing to priority economic sectors. Concerning the application of reserve requirements, however, developing countries tend to use lower ratios than advanced countries. Meanwhile, since the beginning of the 1990s, advanced and developing countries have been inclined to lower their respective reserve requirement. For example, the reserve requirements applied in the United States, Germany, South Korea, and India have decreased significantly over the past two decades, while primary reserves in the UK are approaching zero. Therefore, primary reserves are not actually part of monetary policy implementation, merely a form of tax revenue from the banking sector. In fact, several countries, including New Zealand, Canada, and Sweden, have abolished primary reserves as the effectiveness of the instrument faded, in terms of monetary policy, in line with financial sector innovation. Experience has shown that the position of bank reserves determined by the central bank tends to decrease toward zero when the central bank applies short-term interest rates as the operational target of monetary policy. Conceptually, lower reserve requirements have two implications that must be addressed by the central bank to apply effective monetary policy. First, in an environment of non-binding reserve requirements, there are fundamental linkages between the structure of the payment system and the implementation procedures of monetary policy. In this regard, if strong demand for central bank balances is driven to meet payment needs rather than the reserve requirement, a change in the payment system structure becomes important in the design of monetary policy implementation. Second, a system in which the reserve requirement is low or even zero will exacerbate short-term interest rate shocks because of the relative difficulty in predicting demand for reserves or the lower demand sensitivity of reserves to interest rates.9 From the experience of New Zealand, Canada, and Sweden, it can be concluded that there are tight linkages between changes in payment system structure and monetary policy operating procedures. Nonetheless, there is also a weak correlation between reserve requirements and the degree of short-term interest rate volatility. In this case, interest rate volatility occurs when

9

For more comprehensive analysis, refer to Sellon and Weiner (1997).

1%

0.75%

Not applicable

Not applicable

Germany

Japan

Australia

Canada

Short-term, as required

Overnight

Per bank, with discretion

3 Months, mediumterm, long-term

Short-term, as required

3% and 10%

United States

Discount Facility LOLR

Reserve Requirement

United Kingdom Not Applicable

Advanced Countries

Country

Open Market Operations

Government securities, private securities

Government securities, private securities

Government securities, bonds, CDs

Government securities

Government securities, bonds, notes

Government securities, private securities

Table 7.2:  Monetary Policy Instruments in Several Countries (Until 2004).

LVTS

Foreign Exchange Swaps

As required

Credit extension facilities

Not applicable

Not applicable

Other Instruments

210    Perry Warijiyo and Solikin M. Juhro

10.5%

0–7%

6%

6.5% SR: 1–3%

4%

México

South Korea

Thailand

Indonesia

India

Short term, 4–8 weeks

Emergency, short term 0) was apparent in the pre-Volker era. Meanwhile, after 1982, the influence of the output gap had a negative sign and was not significant. The high ρ value shows that interest rate smoothing plays an important role in the decision-making process of the Federal Reserve. This strategy became even more important in the subsequent periods, except from 1979 to 1982, when the Fed operationally practiced monetary targeting (monetary base), which ultimately spurred larger interest rate shocks. Furthermore, observations failed to pick up clear indications that the output gap played an important role in terms of monetary policy implementation by the Federal Reserve during the period after 1979. It is interesting to note that large deviation was observed in the interest rate compared to the normative channel based on the Taylor rule, particularly during several episodes prior to 1987 and in 1992, when John B. Taylor revised his rule. Despite monetary targeting, operationally, monetary policy in Germany was not directed by a simple rule linked to the monetary base as the operational target, for instance the McCallum rule. Nonetheless, certain trends were believed to steer decisionmaking at the Bundesbank. In that respect, as occurred in other industrialized nations, the Taylor rule was considered to play an important role in the operational framework of monetary policy in Germany, where the short-term interest rate was used as a supporting target. Analysis conducted by the Bundesbank (Monthly Report, 51(4), 1999) also showed that during the period after 1987, interest rates in Germany were more aligned with the normative channel determined by the Taylor rule, assuming an average real interest rate of 3.4%. Meanwhile, using the original Taylor rule, Bofinger (2001) showed that during the 1970s and first half of the 1980s, interest rate behavior, as determined by the Bundesbank, was relatively aligned with that determined by the Taylor rule. In terms of the UK economy, as experienced by several other central banks as well, the decision-making process at the Bank of England was also directed by an iteration of the Taylor rule. McCallum (2000) analyzed the behavior of UK interest rates based on the Taylor rule, using an average interest rate value, r*, of 2.25% and inflation target, π*, of 2%. The analysis showed that monetary policy in the 1970s was considered too loose, contrasting monetary policy from 1983 to 1987, which was deemed excessively tight. Furthermore, during the period after 1987, with 1994 as the only exception, the monetary policy response instituted by the Bank of England was considered relatively optimal. Bofinger (2001) also showed that short-term interest rate trends in the UK after the application of inflation targeting could be replicated using the original version of the Taylor rule. The Taylor rule also explains the salient periods of monetary policy application in Japan (Bofinger, 2001). After the collapse of the Bretton Woods system (1973–1978), nominal interest rates were too low and, therefore, real interest rates became negative. Consequently, inflation soared, primarily during the years after the first oil price shock (1973–1975). During the 1980s, however, monetary policy

Monetary Policy Operational Framework     215 was comparatively restrictive, where the nominal interest rate was higher than the normative interest rate per the Taylor rule. From 1985 to 1995, the Taylor rule could also explain the interest rate policy of the Bank of Japan (BOJ). Since 1995, Japan has applied a zero interest-rate policy (ZIRP), close to the normative interest rate per the Taylor rule. This demonstrates the BOJ’s efforts to provide strong monetary stimuli in an economy experiencing deflation. In that context, relatively small deviation of the actual interest rate to the normative interest rate per the Taylor rule from 1989 to 1995 also showed that persistent deflation in Japan during the 1990s was not caused by erroneous monetary policymaking. 7.3.2.2. Role of Monetary Aggregates When Applying an Implicit Rule.  ­ Congruent with the shift in the preferred operational target from monetary aggregates (money base) to short-term interest rates, particularly in advanced countries, use of the interest rate-based Taylor rule has outstripped use of the McCallum rule based on the money base. Nonetheless, this does not imply that the McCallum rule has no place in monetary policy application. To test that role, in a counterfactual study, McCallum (2001) compared the historical performance of the Taylor rule and McCallum rule in three advanced countries, namely the United States, United Kingdom, and Japan. In the US case, both rules indicated excessively tight monetary policy during the 1970s, with the McCallum rule providing stronger indications than the Taylor rule. During the 1980s and 1990s, disparities emerged between the two rules, while during the period after 1987, both rules indicated undesirable monetary policy. In the UK, both rules showed excessive inflationary pressures during the 1970s but differences appeared in the 1980s, when the McCallum rule indicated excessively loose monetary policy during the middle and toward the end of the decade, while the Taylor rule did not. In Japan, the focus of attention concentrated on the period since the 1990s. In this case, both rules indicated overly tight monetary policy in 1998, but the McCallum rule indicated excessively tight monetary policy from 1990 to 1998, while the Taylor rule did not. Operationally, the role of monetary aggregates can also be seen in the context of applying monetary policy in situations that require the central bank to support financial system stability. Interestingly, McCallum (1994) found that central banks possibly apply a macro-oriented monetary policy rule together with the role as lender of last resort (LOLR) in the financial system. In this respect, McCallum proposed a procedure using both interest-based rules and monetary based rules simultaneously in the case of the United States. Technically, the system consisted of: (1) a policy rule for the FFR designed to achieve the monetary base target; and (2) a policy rule that highlights the response of monetary growth to changes in the FFR. From the observations, McCallum concluded that the procedures explored could accommodate the macroeconomic goals together with the central bank acting as LOLR during the periods from 1974 to 1979 (September) and 1988 to 1991.

216    Central Bank Policy

7.4. Monetary Operations in Indonesia 7.4.1. Quantity-based Approach Monetary policy implementation prior to the crisis referred to Act No. 13 of 1968 concerning the Central Bank, which stipulated the final targets (plural) of monetary policy at BI. In addition to low and stable inflation, BI was also mandated to achieve robust economic growth, low unemployment and a zero balance of payments (BOP). Operationally, however, BI relies heavily on base money (M0) as the operational target, with the nominal exchange rate target used as the policy anchor. In this case, the exchange rate is tightly controlled within a narrow band and depreciated at a relatively constant rate (managed exchange rate system). According to that monetary policy strategy framework, a controlled monetary base is expected to translate into controlled money supply (M1 and M2) and, therefore, economic activities (economic growth and prices). The quantity-based approach may be adopted with some considerations. First, when economic and financial conditions are undergoing structural change, as in the case of Indonesia, BI needs an indicator that can be controlled, namely the money base. Second, base money developments also influence money supply, output and inflation. Operationally, the process of monetary control begins with monetary programming. Fundamentally, monetary programming is policy planning to control total money supply in order to achieve the final target of monetary policy. Monetary programming includes determining the operational target of monetary policy.11 Thereafter, BI determines the measures required and the instruments to be used to influence the operational target. The monetary instruments available to influence the operational target are OMO, the discount facility, reserve requirements and appeals. OMO are conducted through auctions of securities, which aim to increase or decrease liquidity on the money market. Meanwhile, the discount facility is a credit facility offered to financial institutions with the discount determined by BI. The reserve requirement is the minimum amount of liquid assets that must be maintained by the banks held at BI. Finally, appeals are made by BI so that all banks are on the same page and can follow the monetary policy measures expected by BI.12

11

When preparing monetary programming, the operational target is determined using the following assumptions: (a)  Policy and developments in other sectors (fiscal, trade, investment, etc.) will progress as expected. (b)  A stable relationship between money supply (as the intermediate target) and the real economy (as the final target). Such conditions require stability in terms of income velocity and demand for money. (c)  A stable relationship between the monetary base (as the operational target) and money supply (as the intermediate target). Such conditions require money multiplier stability. 12 For an in-depth explanation, refer to the Central Banking Book Series, No. 3, ­Monetary Control instruments, by Ascarya, PPSK-Bank Indonesia, 2002.

Monetary Policy Operational Framework     217 It is understood that the relationship between monetary policy aggregates, output, and inflation in the near term may not have been stable during past episodes of financial liberalization and the recent crisis periods.13 Congruently, a new way of thinking emerged concerning an alternative approach considered more appropriate to the monetary policy system in Indonesia, namely the price-based approach.14 This approach emphasizes the role of prices in terms of monetary aggregates, namely the interest rate as a salient variable in transmitting the impact of monetary policy to economic activity.15 A new way of thinking emerged with the development of the financial sector. With the emergence of the financial sector came features such as advanced financial product innovation securitization, and the loss of linkages between the monetary sector and real sector, which had adverse implications for monetary policy implementation. Several supporting research papers, focused on Indonesia, showed that rapid change in the financial sector of Indonesia caused a twofold increase in money creation by the financial system beyond that created by the central bank. Such developments created an unstable money multiplier.16 Furthermore, instability not only affected the money multiplier but also income velocity and the demand for money function.17

7.4.2. The Shift from the Quantity-based Approach to the Price-based Approach Understanding that the ultimate targets are diverse, focused monetary policy is difficult to implement. Moreover, operationally, monetary policy that relies on the monetary base also has several inherent problems. Although the quantitybased approach is considered effective over a longer time horizon, the approach has begun to face stiffer challenges, especially since the beginning of the 1990s. Extremely rapid money market development due to deregulation and greater economic integration internationally has destabilized the relationship between monetary aggregates and output and inflation. Consequently, quantity-based monetary policy has become less effective. Addressing that challenge, BI subsequently adopted a more pragmatic (eclectic) policy framework. Without abandoning the quantity-based approach, more attention was put on interest rate developments, while the extent of BI intervention bands on the foreign exchange

13

Such phenomena have long been found in advanced countries and are the primary reason the quantity-based approach was abandoned in the monetary policy framework. 14 This idea was delivered by Boediono at the seminar on monetary policy mechanisms in Indonesia, held in May 1998. Refer to the Monetary and Banking Economic Bulletin (1998). 15 The rationale to use the interest rate as the operational target emerged prior to the crisis in 1997. 16 Sarwono (1996). 17 Solikin (1998).

218    Central Bank Policy market under a managed exchange rate regime was broadened to reduce the cost of monetary policy. Ultimately, in August 1997, BI abandoned the intervention bands and inevitably applied a floating/flexible exchange rate system. In 1998, however, BI, which had lost control of exchange rates when forced to disburse large-scale liquidity facilities to the banking industry to unwind the Asian financial crisis of 1997/98, began to apply a tight monetary policy stance. The monetary policy stance was tightened to prevent an escalation of inflationary pressures, by attempting to stop all forms of monetary expansion to prevent excess liquidity in the economy. In this case, monetary base targeting was reinforced through a series of policy innovations and improvements. A stiff penalty system for negative bank balances at BI was also introduced to dampen bank demand for the liquidity facilities provided by BI. Limits were placed on deposit rates and interbank lending guaranteed by the Government, which eased the risk of liquidity flows from Bank Indonesia. Meanwhile, the auction system for Bank Indonesia Certificates (SBI) through OMO was changed from targeting the interest rate to targeting the quantity. Thereafter, the “rupiah intervention” instrument was created (subsequently named Bank Indonesia Facilities – FASBI) as an additional contractionary and expansionary instrument to alleviate interest rate shocks on the money market. Under the floating exchange rate system, rupiah exchange rate fluctuations on the market became particularly vulnerable to market sentiment. To reduce excessive currency shocks, BI conducted intervention policy, at that time, by focusing on the lower bound of Net International Reserves (NIR). The actual NIR value is announced publicly on a weekly basis and has never dropped below the lower bound. In terms of the relationship with the management of monetary aggregates, foreign exchange sterilization has a contractionary impact on the money base and, therefore, can be considered an instrument to bolster OMO. Crisis experience provided several invaluable lessons for BI. In terms of monetary management, after losing control of the exchange rate in August 1997, monetary policy lost its exchange rate anchor during the crisis. Consequently, BI applied monetary base targeting as the concurrent operational target and nominal anchor of monetary policy. The effectiveness of the monetary base as the sole indicator and, therefore, anchor of monetary policy in Indonesia was also called into question. As was alluded to previously, the quantity-based approach was used as the monetary policy framework during the crisis period because of monetary leakage that had to be absorbed due to excess liquidity in the banking industry as a result of the liquidity facilities provided through BI’s function as LOLR, not because of the more fundamental consideration of the stable relationship between inflation and monetary aggregates. In fact, several research papers have debunked the assumptions underlying monetary base targeting, for instance that the money multiplier and velocity of money are inadequately stable in the case of Indonesia. The monetary policy framework using this program was subsequently formalized as part of the IMF program, where achievement of the monetary base target was considered a viable performance criteria indicator. Monetary base targeting could not guarantee sound attainment of the inflation target however. The low performance of monetary base targeting was primarily

Monetary Policy Operational Framework     219 due to difficulties predicting public behavior in terms of holding currency. Postcrisis public demand for currency experienced a structural change that was difficult to explain from the motive perspectives of transactions or for a rainy day. Control over the monetary base was also considered more difficult when the bank intermediation function was not normal. Under such conditions, raising the SBI interest rate monetary instrument (policy rate) to absorb currency back into the banking system was oftentimes constrained by the limited deposit rate response, which meant that coercing currency back into the banking system required a hike to the policy rate of higher magnitude that would normally be expected. This reality caused BI to regularly face a monetary policy dilemma. On one hand, to reduce demand for base money, BI had to adopt a tight policy stance, which raised interest rates. On the other hand, however, the ineffective transmission of monetary policy implied that the tight monetary policy stance would require a much larger hike to the interest rate for a positive impact on the economic recovery. When banking and corporate sector conditions were comparatively weak, however, the adverse impact of higher interest rates on the economy was much more pronounced than under normal conditions (Warjiyo & Agung, 2002). The dilemma made it difficult for BI to fulfill its monetary policy commitments to achieve the monetary base target set. Difficulties controlling the monetary base were also found when the money base fell well short of its target, as transpired in 2002. Efforts to grow the monetary base were ineffective when the banking industry was still weak and real sector risks were considered high. While the bank intermediation function was not considered normal, additional economic liquidity through the banking sector would flow back to the central bank. Therefore, the monetary base is said to be more demand determined than supply determined. The problems that emerged in relation to monetary base targeting as the operational target of monetary policy corroborated the implications presented in preliminary studies, which recommended a price-based approach to monetary policy, using the interest rate as the operational target (Solikin, 2005; Warjiyo & Agung, 2002). This is also supported by the empirical fact that, in the case of an emerging economy, the characteristics of base money are more endogenous and, therefore, the indicator is less flexible as the operational target. In contrast, however, considering that the monetary base and interest rate mutually respond to one another, simultaneously targeting the money base and interest rate, as has occurred recently, tends to trigger problems relating to policy transparency and consistency. Fundamentally, the implications stress the importance of BI focusing the orientation of the operational target on just one indicator, namely the interest rate. Congruent with ITF application since July 2005, BI officially applies the interest rate as the operational target of monetary policy in Indonesia. Consequently, two questions emerge, namely which interest rate indicator should be used as the operational target? And which interest rate is representative of the policy rate and can therefore reflect the monetary policy stance? Studies specifically addressing the choice of an operational target for monetary policy under an inflation targeting regime found that the O/N interbank rate was a feasible operational target of monetary policy based on controlling the interest

220    Central Bank Policy rate under an inflation targeting regime. In addition to the findings of previous research that concluded several advantages of using the O/N interbank rate per international best practices, such as measurability, ability to affect the ultimate target and easily understandable by the public, the study also found strong indications that BI could control the O/N interbank rate (controllability) through the signaling and liquidity adjustment mechanisms. Other important empirical evidence includes low O/N interbank rate volatility during the observation period, despite not targeting the interest rate in the operational framework at that time (BI, 2004b). In terms of selecting a representative interest rate for the policy rate, there are several schools of thought, specifically concerning the criteria and essence of the policy rate itself. There are at least three salient criteria for interest rate indicators that could be used as the policy rate, namely that the interest rate indicator selected must be controllable by BI, can influence the cost of funds in the banking industry, and can effectively be used as a reference rate by economic players in their economic decision-making process. Those criteria are congruous with the essence of the policy rate, namely that the policy rate is part of the (transparent) communication strategy, is a variable that reflects the monetary policy direction (as the simplest form of central bank communication), can be steered to influence the (pricing) expectations behavior of market players, indicates the level of short-term interest rates desired by the central bank, reflects the central bank’s response to various macroeconomic indicators, and is not the actual interest rate on the money market. Based on those considerations, BI chose the BI rate as the policy rate. In terms of OMO, the BI rate was used, until the middle of 2008, as the reference rate for auctions of 1-month SBI. Since 2008, however, the BI rate has been used as a reference rate for the O/N interbank rate.

7.5. Concluding Remarks Academically, analysis of and support for rules-based policy strategy using the Taylor rule (Taylor, 1993), at least in the context of advanced countries, has increasingly reverberated considering the main problems associated with fine tuning a discretion-based policy strategy. The issues include difficulties formulating stabilization policy in terms of policy timing and elements of uncertainty in terms of the policy influence, as well as the tendency for time-inconsistency and inflationary bias problems to emerge. Meanwhile, in the case of emerging economies, Taylor (2000) stressed that rules are also very useful but require modification to the unique characteristics of the economy involved. In fact, Taylor (1996) also emphasized the flexibility by referring to conditions where central banks have abandoned rules, in addition to the possibility of applying discretion combined with rules. For the Indonesian economy, there remains broad research scope concerning the possible application of monetary policy strategy using rules, or even

Monetary Policy Operational Framework     221 state-contingent policy rules as mentioned by Keynes three decades ago.18 What is meant by state-contingent rules? King (1997) explained that The choice of monetary strategy is often described as a choice between rules and discretion … The optimal strategy is a statecontingent rule, which allows flexibility in the response of policy to shocks while retaining a credible commitment to price stability. Meanwhile, Woodford (2003) stressed that “The fully optimal policy commitment would lead not only to the optimal long-run average values of target variables, but also to the optimal responses to disturbances.” Therefore, statecontingent rules are considered an optimal policy strategy that allows the central bank to respond to shocks without neglecting the target of price stability. The choice of policy strategy will depend on the characteristics of the respective economy and the monetary policy response. For example, Indonesia’s economy is unique and has experienced economic gyrations over the past three decades.19 Paying due consideration to large macroeconomic fluctuations as a consequence of structural changes in Indonesia, particularly during the past decade, along with the reorientation of monetary policy strategy with the final target of price stability, monetary policy strategy formulation using state-contingent rules, which can accommodate rules and discretion, is incredibly relevant. Therefore, one salient issue for further study is how, technically and operationally, to formulate the state-contingent rules. In his pioneering frontier study on formulating state-contingent rules for the Indonesian economy, Solikin (2004b) showed that policy response formulation based on state-contingent rules successfully represented an optimal design of policy rule for the Indonesian economy. The policy rule design accommodated the long and short-term policy responses with an error correction mechanism, as well as the presence of specific shocks that surfaced due to sudden structural changes and dynamic changes because of elements of uncertainty that affected monetary policy response volatility.

18

Boughton. Several studies on the (possible) use of rules (Taylor rule) in Indonesia have been conducted, including Darsono et al. (2001) and Hutabarat (2002) of the Directorate of Economic Research and Monetary Policy – BI. the studies focused on the relevance and use of rules with the possible implementation of the ITF. 19

This page intentionally left blank

Chapter 8

Inflation Targeting Framework: Concept and Implementation at Central Banks 8.1. Introduction Since the beginning of the 1980s, a growing number of central banks in various jurisdictions have applied a monetary policy regime or framework based on inflation targeting. In advanced countries, the Inflation Targeting Framework (ITF) has been implemented, among others, in New Zealand (1990), Canada (1991), Israel (1991), United Kingdom (1992), Sweden (1993), Finland (1993), Australia (1993), and Spain (1994). On the other hand, in emerging market economies (EMEs), ITF has been applied in several Latin American countries, including Chile and Brazil, in Eastern European countries, such as Poland and the Czech Republic, as well as in a number of Asian countries, like South Korea, Thailand, the Philippines, and Indonesia. Furthermore, ITF has also become a hot topic in numerous discussions and research papers authored by economists and as educational material to be explored at various universities. In the context of Indonesia, ITF was implicitly introduced in the year 2000 upon enactment of the BI Act (No. 23) of 1999 on May 17, 1999. Pursuant to that law, BI was mandated with achieving rupiah currency stability, meaning price (inflation) stability, and rupiah exchange rate stability. Seeking to achieve that goal, monetary policy is consistently directed toward attaining the published inflation target. In addition, Bank Indonesia also independently formulates and implements monetary policy as well as the other duties stipulated in the Bank Indonesia Act. Notwithstanding, BI cannot extend loans to the Government. Those provisions represent the salient characteristics and form the legal basis of ITF implementation by BI in Indonesia. Indirectly, the legislation contained within the BI Act mandates that monetary policy be based on the framework known as ITF in theory and as practiced by other central banks. Nevertheless, congruent with the protracted economic recovery process after the Asian Financial Crisis of 1997/1998 that persisted into the 2000s, ITF was implemented gradually through a transition period and officially applied in full on July 1, 2005. This chapter will discuss the ITF-based monetary policy framework along with its application at various central banks, while implementation in

Central Bank Policy: Theory and Practice, 223–260 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191012

224    Central Bank Policy Indonesia will be explored in the subsequent chapter. The discussion is divided into seven sections. After this introduction, the second section will present a detailed overview of the ITF, including the rationale and characteristics as well as advantages and disadvantages. The third section will discuss the institutional arrangements, focusing on provisions in the central bank statutes, inflation target formulation as well as accountability and transparency. The fourth section will describe the operational arrangements of ITF, specifically inflation projection techniques, monetary policy transmission mechanisms and monetary operations. The fifth section will discuss the most common ITF regimes applied around the world, namely full-fledged inflation targeting (FFIT), eclectic inflation targeting (EIT), and inflation targeting lite (ITL). The sixth section will explore the contribution ITF implementation has had on economic performance in various countries, particularly inflation, macroeconomic performance, and in terms of the central banks’ proclivities to monetary policy. To conclude, the seventh section will present several important takeaways from ITF application.

8.2. Conceptual Dimension and Theoretical Model 8.2.1. Rationale, Characteristics, Advantages, and Disadvantages 8.2.1.1. Rationale.  Fundamentally, ITF is a framework where monetary policy can be directed toward achieving a future inflation target that is published transparently as a tangible form of central bank commitment and accountability. This definition contains several implications. First, there must be an official statement, supported by central banking laws, that the final goal mandated by the public in terms of monetary policy is to achieve and maintain price stability/low inflation. Second, a medium–long term inflation target must be determined and announced as an anchor of monetary policy. Third, there must be commitment and discipline from the central bank to steer monetary policy toward the inflation target set, while enjoying full independence to formulate and implement monetary policy, without interference from the government or a third party. And fourth, there must be monetary policy transparency to the public and accountability mechanisms that hold the central bank accountable for its monetary policy to the public in pursuance of the price stability mandate given. In theory and in practice at various central banks, there is different rationale underlying ITF implementation. First, a growing body of empirical evidence has shown that monetary policy only influences prices in the long term, although monetary policy can affect economic growth (real output) in the near term. Second, there is an increasing requirement to base monetary policy formulation and implementation on a specific framework to enhance central bank discipline, consistency, and credibility. Third, transparency and accountability of public policies, including monetary policy, have become increasingly important as the embodiment of good governance and democratic accountability in the government’s institutional reforms, including the central bank. The following section will explain the rationale behind ITF implementation.

ITF Concept and Implementation at Central Banks     225 In Chapter 3, the effect of money on the economy was discussed, specifically the impact on prices and real output. Various empirical evidences have shown that money only influences prices in the long term, thus proving the neutrality of money on real output congruent with the quantity theory of money. Meanwhile, in the near term, money can influence real output to differing degrees depending on the various definitions of money supply, the analysis period, and cross-border analysis. The studies have provided empirical evidence that in the medium–long term, monetary policy must be directed toward price stability. Nonetheless, there is a trade-off in the near term between inflation and real output when price stability is set as the final target of monetary policy. Over the subsequent few chapters, the need for consistent and transparent monetary policy implementation to enhance effectiveness and credibility in terms of influencing inflation and real output shall be discussed. The time-inconsistency theory proposed by Kydland and Prescott (1977) showed that monetary policy directed toward a medium–long term inflation target is more optimal than policy oriented toward a short-term inflation target because a short-term policy target would trigger time inconsistency and not take into consideration the policy implications on long-term public welfare. Meanwhile, Barro-Gordon (1983) proposed several requirements for monetary policy credibility as follows: (a) the central bank must publish its inflation target and the monetary policy direction taken to achieve said target; (b) the public must have confidence that the central bank will consistently orient monetary policy toward achieving the inflation target; and (c) the central bank can successfully achieve the inflation target set. Monetary policy transparency could be contained in rational expectation hypothesis theory, indicating that only anticipated monetary policy has an influence on real output,1 considering that under conditions of increasing uncertainty, economic players will tend to act rationally and take into account the direction of monetary policy implemented by the central bank when taking business decisions. Therefore, policy transparency will facilitate decisionmaking among economic players and avoid an unnecessary build-up of uncertainty that could arise if the direction and target of monetary policy are unknown. In other words, transparency can enhance monetary policy effectiveness in terms of influencing economic and financial activities to achieve the inflation target set. 1

This theory is discussed in more detail in Chapter 12 along with other theories underlying central bank monetary policy transparency. Fundamentally, the theories are linked to the problem of asymmetric information in monetary policy, primarily in the context of the focus and clarity of the central bank’s propensity towards the goal of monetary policy, the relevance and magnitude of market expectations in response to and in terms of transmitting monetary policy as well as the debate between rules versus discretion in monetary management. The various theories have been motivated by the phenomenon of increasing monetary policy transparency at various central banks in line with the proliferation of central bank independence under democratic government rule, the application of a monetary policy framework based on inflation targeting, as well as the importance of managing expectations in increasingly advanced and complex financial markets.

226    Central Bank Policy Meanwhile, the importance of central bank independence and monetary policy accountability is based on agency theory in various public policies as a form of good governance and democratic accountability in the government system. In this respect, the public, as the principal, gives a mandate to the central bank, as the agent, to achieve and maintain price stability as the goal of monetary policy. On the other hand, to implement its public mandate, the central bank must be given full independence to formulate and execute its monetary policy. Per agent theory, the principal–agent relationship must also be accompanied by accountability mechanisms for the independence given to the agent, namely the central bank, in monetary policy to implement the mandate given by the agent, namely the public, toward the final goal of price stability. Furthermore, because public policy is involved, the principal–agent relationship in terms of mandating the goal, monetary policy independence as well as the accountability mechanisms are stipulated in central banking laws formulated and enacted through parliament.2 8.2.1.2. Characteristics.  The rationale detailed above underlies the ITF-based monetary policy framework. In this regard, ITF is a monetary policy framework with a number of salient characteristics, namely an official statement from the central bank that the final goal of monetary policy is to achieve and maintain low inflation in the medium–long term, as well as to publically announce the inflation target. The announcement implies that the central bank is avowedly committed and assures the public that every monetary policy issued shall consistently strive to achieve the inflation target, while ensuring that the central bank is accountable for its monetary policy if the inflation target is not achieved. In other words, ITF application satisfies the principles of monetary policy oriented toward a medium– long term inflation target, with central bank commitment to focus its monetary policy on achieving the inflation target as well as giving independence together with increased transparency and accountability in the monetary policy implementation process by the central bank. In more depth, the characteristics of inflation targeting per Bernanke, Laubach, Mishkin, Posen (1999) are presented in Table 8.1. Table 8.1 presents the general characteristics associated with ITF. First, there is an official statement that price stability is the final goal of monetary policy and the inflation target is announced to the public. Regarding who sets the inflation target, whether it be the government or central bank itself, remains the subject of debate in theory and in practice. In addition, the inflation target should be determined in the medium term. Second, the central bank is given independence to formulate and implement monetary policy (instrument independence) to achieve the inflation target set. In this case, the monetary policy response will be based on inflation and output projections and could apply rules in its application. Third, there is monetary policy transparency with intensive public communication, not only concerning the inflation target but also explaining how the central bank

2

A more detailed description of the theories and practices associated with independence and accountability at several central banks and at BI will be discussed in Chapter 11 of this book.

ITF Concept and Implementation at Central Banks     227 Table 8.1:  Characteristics of Inflation Targeting. Criteria

Bernanke et al. (1999)

Svensson (2000)

King (2004)

1

Price stability as the final goal of monetary policy

Yes

Yes

Yes

2

Inflation target announcements

Yes

Yes

Yes

3

Medium-term inflation target

Unclear

Yes

Yes

4

Intensive communication with the public

Yes

Yes

Yes

5

Use of a specific monetary policy rule

Unclear

Inflation projection targeting

Inflation targeting + supply-side response

6

Published inflation and output projections

Not required

Yes

Unclear

7

Government-set target (goal dependent)

Yes

Yes

Not required

8

Instrument independent

Yes

Yes

Yes

Sources: Bofinger (2001) and King (2004).

conducts monetary policy along with performance in terms of achieving previous inflation targets. Mishkin (2004) suggested five core elements that reflect ITF application as follows: (1) the public announcement of medium-term numerical targets for inflation; (2) an institutional commitment to price stability as the primary goal of monetary policy, to which other goals are subordinated; (3) an information inclusive strategy in which many variables, and not just monetary aggregates or the exchange rate, are used for deciding the setting of policy instruments; (4) increased transparency of the monetary policy strategy through communication with the public and the markets about the plans, objectives and decisions of the monetary authorities; and (5) increased accountability of the central bank for attaining its inflation objectives. Those five aspects explain the key elements of ITF, which cover more than just announcing the future inflation target. It is important to understand, however, especially for EMEs, most of which routinely report the inflation target or other economic projections as part of the government’s economic plan for the upcoming year, that such monetary policy cannot be categorized as ITF because it does not meet the four other requirements. As emphasized by Bernanke et al. (1999), ITF is not a hard and fast rule but more of a framework for the central bank to formulate and implement its monetary policy. Furthermore, ITF application does not necessarily imply that

228    Central Bank Policy monetary policy only considers inflation, per the “inflation nutters” moniker suggested by King (1996).3 As explained in Chapter 7, the term “rule” in the monetary policy implementation strategy is linked to the theoretical thinking behind applying one operational target based on one rule or instrument. For example, the money rule à la Milton Friedman established that growth of the monetary base at a certain rate is based on the predicted requirement for money supply in the economy as the operational target of monetary policy. In addition, the interest rate rule à la Taylor linked a specific short-term interest rate as the operational target based on future projections of inflation and real output. As a monetary policy framework, ITF represents a comprehensive regulatory framework to formulate and implement central bank monetary policy concerning institutional arrangements, monetary policy implementation, and organizational implications for the central bank (Schaechter, Stone, & Zelmer, 2000). As an institutional framework, ITF encompasses central banking legislation that stipulates price stability as the main target of monetary policy, central bank independence as well as restrictions on the central bank financing the Government’s budget deficit. In addition, determining the inflation target as well as the accountability and transparency mechanisms are also regulated. As an operational framework, ITF covers monetary policy formulation and implementation strategies to achieve the inflation target set, including regulations on modeling inflation projections, monetary policy transmission mechanisms and the monetary operations available. In principle, based on an assessment of inflation projections (as well as other economic and financial variables) and the expected transmission mechanisms, monetary policy is formulated in such a way to guarantee that future economic and financial fluctuations remain within a certain corridor to achieve price stability. In its implementation, monetary operations are oriented toward achieving the given operational target, usually shortterm interest rates, based on a specific rule, such as the Taylor rule. Nevertheless, the use of a rule is only one consideration underlying monetary policy formulation by the central bank in addition to various other economic and financial indicators. In other words, rules are only used as a guideline for consideration when determining the operational target of monetary policy, otherwise known as rule-constrained discretion in ITF implementation. ITF also covers the organizational/institutional arrangements at a central bank required to support effective and successful implementation. This includes regulations concerning the monetary policy formulation process, which can guarantee that the policies to be instituted have been through a rigorous analysis process considering all the relevant data and information. In addition, committees are required with specific decision-making regulations to enhance the quality of

3

In theory and in practice, when the inflation target is set, its impact on economic growth is always considered. When setting the inflation target in the context of the trade-off between inflation and output per the Phillips curve, refer to Chapter 7. The empirical evidence described in the sixth section of this chapter also shows how real output is considered when applying ITF in various countries.

ITF Concept and Implementation at Central Banks     229 monetary policy. Various adjustments to the internal organization of the central bank are also required, including data collection, inflation and economic projection modeling techniques as well as policy analysis, in addition to enhancing the competence of human resources. 8.2.1.3. Advantages and Disadvantages. The growing number of countries applying ITF demonstrates that the framework offers several advantages over monetary policy regimes using money supply, the exchange rate or other anchors (Bernanke et al., 1999; Leiderman & Svensson, 1995; Svensson, 1997) as follows: ⦁⦁ First, ITF helps divert public attention away from short-term interventionist

policies, where monetary policy plays a smaller role, to the achievement of low and stable inflation in the medium–long term to support macroeconomic stability and economic growth. ⦁⦁ Second, ITF fundamentally increases monetary policy and fiscal policy accountability and discipline. ⦁⦁ Third, established ITF implementation provides space for the central bank to respond to short-term developments with a smaller risk to inflation creditability in the long term. ⦁⦁ Fourth, ITF helps to motivate institutional reforms at the central bank, while also encouraging more general structural reforms, especially in the context of lowering inflation (disinflation). Of course, ITF implementation also has several potential weaknesses, including instrument instability if the framework is applied too strictly, the need for a relatively strong fiscal position and the possibility of higher short-term exchange rate volatility, especially in a small open economy (Schaechter et al., 2000).

8.2.2. Theoretical Model As described in previous sections, there are at least three salient aspects in the implementation of monetary policy. First, expectations play a fundamental role in the workings of monetary policy. This is evidenced when determining the policy rate, for example. Expectations concerning how monetary policy will manifest in upcoming periods as well as the current interest rate are given the same degree of importance by policymakers. Second, there is understanding that economic dynamics will constantly evolve. Therefore, there are no hard and fast rules to support optimal monetary policy strategy formulation. Third, the complexity of optimal economic agent behavior allows the central bank to consider the possibility that the public applies simple and practical rules. Congruent with that perspective, macroeconomic theory, particularly monetary theory, has developed with rapidity over the past three decades. Evidence that theoretical advancements have had a significant effect on monetary policy practices is relatively simple to find with a foothold in historical evolution and tied to the long-term future perspective. The growing trend of central bank independence, ITF implementation and the use of policy rules to guide monetary policy implementation represent tangible examples of the impact macroeconomic

230    Central Bank Policy theory has had on monetary policy practices. Although the substance or rationale behind ITF, especially in relation to building monetary policy credibility, can be explained by theories existing prior to the emergence of ITF, for instance, the theory of credibility and time consistency by Kydland and Prescott (1977), forming credibility and reputation by Barro-Gordon (1983), credibility and conservative/independent central banks by Rogoff (1985), the general view states that the growing number of central banks applying ITF is part of the “practices ahead of theory” phenomenon (King, 2005). The practice ahead of theory phenomenon is based on guiding the achievement of today’s monetary policy goal on efforts to maintain price stability, a rationale that did not develop during the three decades prior to ITF proliferation. From the end of WWII until the mid-to-late 1970s, most economists, academics and policymakers held the same view, that monetary policy was less related to inflation and an ineffective instrument of monetary management. The underlying theoretical basis of that opinion is not particularly clear, but empirical experience has shown that price instability causes fluctuations in real output and expensive job opportunities. Meanwhile, monetary policy oriented toward price stability has been shown to be a key to success for aggregate demand management. Congruent with the empirical facts, theoretical support for monetary policy practices through evolution has been a significant acknowledgment during the past two decades, reflected by the accelerating development of theoretical views based on ITF implementation in various countries. ITF, which was initially applied as an “art of central banking” with a grounding in strengthening the institutional framework of monetary policy to support policy credibility and manage expectations (Bernanke et al., 1999), gradually began to be considered through the foundational development of macroeconomic theory. In this case, there are two main schools of thought providing a foundation for ITF macroeconomic theory, namely New Keynesian (Clarida, Gali, & Gertler, 1999; Rottemberg & Woodford, 1997; Svensson, 1997) and New Neoclassical Synthesis (Goodfriend & King, 1997). New Keynesian represents supply-side equilibrium, via the theory of inflation expectations in the New Keynesian Phillips Curve (NKPC) and demand side (IS Curve), as well as policy rule behavior. Meanwhile, New Neoclassical Synthesis offers a formal opinion on the achievement of price stability based on achieving prosperity. 8.2.2.1. Baseline ITF Macroeconomic Model.  Consensus from both sides, New Keynesian and New Neoclassical Synthesis, has produced a viewpoint concerning the substance of ITF theory, namely that “inflation dynamics depend on expected (future) inflation and output gap” (Clarida et al., 1999) and “the central bank should seek to keep output at its potential by targeting the minimum mark-up consistent with near-zero core inflation” (Goodfriend & King, 1997). In general, such conditions are reflected in a system of three fundamental equations (baseline model) considered as a form of modern macroeconomic thinking, namely (Walsh, 2003):



Supply equation (NKPC) πt = Et πt +1 + αxt + εt (1)

ITF Concept and Implementation at Central Banks     231

Demand curve (IS) xt = Et xt +1 − δ (it − Et πt +1 ) + ut(2)



Interest-based policy rule it = φπ πt + φx xt + φi it−1 + υt (3)

By emphasizing the optimal policy rule design, an additional two equations were created in the system as follows:



∞

t Central bank loss function Lt = Et ∑ δ τ ( πt + τ − π )2 + λxt2+ τ  (4) τ =t



λ α

Targeting rule πt + xt = 0(5)

where π is inflation, x is the output gap, π is the inflation target, and I is the nominal interest rate. Several fundamental points emerged when developing the equation system above as follows: ⦁⦁ First, the modeling system above is based on a microfoundation that funda-

mentally and explicitly can be derived from the optimal behavior of economic agents (representative agents), namely the central bank as the policymaker, producers as prices setters, and the public or households as the end users of goods and services (consumers). In this case, the optimal policy taken aims to maximize public welfare. ⦁⦁ Second, the equations in the system are still in basic form (generic), where NKPC and the IS Curve are fully forward looking. In the empirical literature, however, further development facilitated a hybrid NKPC with a mixture of forward- and backward-looking expectations. Meanwhile, the policy rule can be stated explicitly in the form of the Taylor rule. ⦁⦁ Third, NKPC and IS Curve behavior are fundamentally already in equilibrium, thus interaction between the two is theoretically in equilibrium. Nonetheless, a system with forward-looking expectations triggers multiple equilibria as a form of behavior or expectations, which directly or indirectly become self-fulfilling expectations identified in the rational expectations model. Such conditions create indeterminacy. From a general perspective, the problem of indeterminacy can be overcome by proposing an active policy rule design, namely with a reaction parameter of the policy rate to changes in inflation of more than 1. Therefore, the presence of a policy rule that emerges as the third equation is the optimal policy rule design, which is typically oriented toward efforts to achieve system stability in line with prevailing policy commitment. ⦁⦁ Fourth, the central bank’s preference, be it inflation or output stability, will influence the impact of the monetary policy taken. Theoretically, strict ITF application, where λ = 0, will provide a larger short-term impact on inflation, output and interest rates compared to flexible ITF, where λ ≠ 0.

232    Central Bank Policy 8.2.2.2. ITF in the Inflation Forecast Targeting Format.  According to Svensson (1997), in line with the forward-looking policy perspective, the substance of ITF could connote inflation forecast targeting. This has implications on the specification of the inflation target and how monetary policy must operate to achieve that target. The view interpreted per optimal policy rule can be presented with the following basic considerations: (1) Bearing in mind the time lag inherent with monetary policy, the target of monetary policy must be based on (expected) inflation forecasts, not current or past inflation. (2) Inflation targeting implies inflation forecast targeting, where the weight allocated to output stabilization determines the speed at which the inflation forecast adjusts to the inflation target. With those basic considerations, adapting the optimal policy rule based on modifications to the ITF equation system as the baseline model can be expressed as follows: Economic system:

Et [ πt +1 ] = πt + α yt + εt +1(6)



Et [ yt +1 ] = β1 yt − β2 (it − πt ) + ηt +1(7) Loss function (a normalized natural rate of output at zero): L ( πt , yt ) =



1 ( πt − π * )2 + λ yt2   2

Intertemporal loss function: ∞

Et ∑ δ τ t L ( πτ , yτ ) 



τ =t

With the given lag of monetary policy (1 year for output and 2 years for inflation), then: Et [ πt +2 ] = a1πt + a2 yt − a3it + ( εt +1 + εt +2 + α1ηt +1 )(8)

where a1 = 1+ α1β2 a2 = α1 (1+ β1 ) a3 = α1β2 and

∂ πt +2 = −a3 ∂it

ITF Concept and Implementation at Central Banks     233 Case 1: There is only one policy target (inflation target), λ = 0 Minit Et δ 2 L (πt +2 )

FOC:

 ∂ Et δ 2 L ( πt +2 ) ∂π  = Et δ 2 ( πt +2 − π * ) t +2  (9)  ∂it ∂it  





= −δ 2 a3 ( Et [ πt +2 ] − π * ) = 0



⇒ Et [ πt +2 ] = π *



⇒ π * = a1πt + a2 yt − a3it The optimal policy rule can be expressed as follows: it = πt + b1 ( πt − π * ) + b2 yt (10)

where

1 α1β2 1+ β1 b2 = β2 b1 =

Case 2: Considering other policy targets (output stability), λ ≠ 0

Minit Et δ 2 L ( πt +2 ) + δ L ( yt +1 )

FOC: ⇒ Et [ πt +2 ] − π * = −

λ Et [ yt +1 ](11) δα1 k

2   λ(1− δ )  4λ  1  λ(1− δ ) 1 +  where k = 1− +  + 2   2  a1  δα12 δα12    1 From (6), Et [ yt +1 ] = ( Et [ πt +2 ] − Et [ πt +1 ]) is obtained. α1

The FOC can be written as follows:

Et [ πt +2 ] = c π * + (1− c ) Et [ πt +1 ](11’)

234    Central Bank Policy where c=

δα12 k λ + δα12 k



From (7): ⇒ Et [ πt +2 ] = πt + α1 (1+ β1 ) yt − α1 (it − πt ) (12) The optimal policy rule is derived from (11) and (12) as follows:





⇒ it = πt +

δα1 k β Et [ πt +2 ] − π * ) + 1 yt (13) ( β2λ β2 ∧

∧

= πt + b1 ( πt − π * ) + b 2 yt

where c α1β2 ∧ c + β1 b2 = β2 ∧

b1 =

After solving the optimal policy rule above, fundamentally the Taylor rule could be modified into a forward-looking, inflation-forecast-based rule. A forward-looking policy orientation that considers the monetary policy lag supports the relevance of implementing monetary policy based on the inflation forecast, not actual inflation. The theoretical advantage of this approach, in addition to producing a response equivalent to optimal conditions, is to enhance monetary policy credibility by focusing on inflation expectations.

8.3. Institutional and Operational Framework An institutional framework that supports ITF application encompasses provisions in central banking laws, formulation of the inflation target as well as accountability and transparency mechanisms. As mentioned previously, ITF has been implemented in the following advanced countries: New Zealand, Canada, Israel, United Kingdom, Sweden, Finland, Australia, Spain, and South Africa. On the other hand, ITF has been adopted by EMEs in Latin America, such as Chile and Brazil, in Eastern Europe, including Poland and the Czech Republic, as well as Asian countries, like South Korea, Thailand, the Philippines, and Indonesia. In this case, there are several iterations of ITF practiced in full at different central banks around the world. In general, EMEs tend to apply comparatively more institutional regulations than advanced countries.

ITF Concept and Implementation at Central Banks     235 8.3.1. Institutional Framework 8.3.1.1. Central Banking Laws.  ITF implementation requires regulations in central bank laws stipulating price stability as the target of monetary policy and providing full authority to the central bank to achieve that target (instrument independence). In this case, some variation arises concerning the clarity of the monetary policy target and the weight given to the attainment of price stability in the regulations of various central banks (Table 8.2). In many advanced countries that have applied ITF, including New Zealand, United Kingdom, Spain, Sweden, and Finland, price (currency) stability as the target of monetary policy is stipulated in the statutes of the central bank law. In other advanced countries, however, including Australia and Canada, although the central bank’s mandate covers broader economic goals, in practice, price stability has become the main monetary policy target of central banks. In EMEs, the goal of monetary policy is legislated in central banking laws, which facilitate the internal and external dimensions of currency stability. Therefore, laws typically stipulate currency stability as the goal of monetary policy, encompassing price (inflation) stability and exchange rate stability, which reflects the role of a managed or floating exchange rate system in several EMEs. In practice, however, the focus of the monetary policy target has shifted away from domestic price stability in line with the proliferation of jurisdictions adopting a flexible exchange rate system. Table 8.2:  Provisions in Central Bank Laws. Legal Provisions

Country

Monetary policy goal • Currency stability as the primary goal • Price stability as the primary goal • Currency stability and other goals

• Finland, Czech Republic, and Indonesia • New Zealand, Poland, Spain, Sweden, and United Kingdom • Australia, Canada, Israel, Brazil, Chile, and South Africa

Instrument independence

• New Zealand, Australia, Canada, and United Kingdom, Poland, Spain, Sweden, and Israel, Finland, Czech Republic, Brazil, Chile, South Africa, and Indonesia

Financing the fiscal deficit • Restricted • Proscribed

Source: Schaechter et al. (2000, p. 7).

• Canada, Czech Republic, and South Africa • Brazil, Chile, Finland, Israel, Poland, Spain, Sweden, and Indonesia

236    Central Bank Policy Instrument independence in monetary policy is generally safeguarded in central banking laws. All EMEs stipulated instrument independence in prevailing laws prior to ITF implementation. In advanced countries, the United Kingdom and Sweden amended their central banking laws to accommodate instrument independence. In Australia, Canada, and New Zealand, however, although prevailing central banking laws permit a parliamentary amendment to the monetary policy decisions of the central bank, none of the respective governments have taken such action in over 30 years (Schaechter et al. 2000). Central banking laws in EMEs also explicitly regulate restrictions on the central bank financing the government’s budget deficit. This includes the purchase of government debt securities by the central bank on the primary market due to previous experience of hyperinflation stemming from the overprinting of money to finance the deficit. Government debt is relatively large in EMEs compared to their advanced counterparts, which demonstrates lower inflation volatility to monetary expansion to offset the deficit (Kumhoff, 2000). Nevertheless, it is important to note that restrictions on financing the government’s budget deficit do not extend to buying government debt securities on the secondary market for monetary operations. In many countries, ITF implementation has been reinforced by amendments to central banking laws. When to amend and the magnitude of the change to central banking laws vary from country to country, reflecting a trade-off between the efforts required to garner public support for ITF adoption compared to the increase of credibility brought about by the new legislation (Walsh, 1995). In advanced countries, including New Zealand, Spain, and Sweden, legal amendments were implemented in conjunction with ITF implementation. Conversely, in United Kingdom, Canada, and Israel, public support for a change in legislation was obtained after ITF credibility was proven. On the other hand, EMEs have been more inclined to amend prevailing central banking regulations prior to ITF application. In other words, the legal amendments were considered a way to enhance monetary policy credibility in terms of stabilizing inflation, which, in general, has historically been comparatively higher in EMEs than advanced countries. Of course, the contribution of central banking laws to improving monetary policy credibility must be taken in perspective because monetary policy credibility needs to be built through the ability of the central bank to achieve and maintain low and stable inflation (Blinder, 1998). 8.3.1.2. Formulation of the Inflation Target.  Formulation of the inflation target also varies among ITF countries in terms of the price index used, size of the target, target horizon, and publication method. Inflation target formulation primarily depends on several factors specific to the country involved, including the level of central bank credibility, economic vulnerability to shocks as well as data availability. In several cases, a number of countries have adjusted how the inflation target is formulated consistent with changes in those specific factors. What is most important, however, is that when formulating and specifying the inflation target, the central bank also considers the relationship and impact on various other macroeconomic variables, especially economic growth. Selection of the price index. Fundamentally, the choice of which price index to use as the inflation target is affected by the influence monetary policy has on

ITF Concept and Implementation at Central Banks     237 controlling prices (controllability) and the level of public trust in the price index selected (credibility). In this case, the CPI, also known as headline inflation, is a sound measure of inflation that provides a better level of credibility than the other price indexes.4 CPI is the earliest price indicator available periodically, the focus of expectations, easy to understand by the public and not thought to be subject to manipulation by the central bank. On the other hand, however, CPI often contains components that are beyond the control of the central bank (administered prices (APs), terms of trade (ToT), and indirect taxes), which could, therefore, undermine the influence of monetary policy. Consequently, the use of CPI could spur excessively tight monetary policy if the price components that are beyond the control of the central bank are comparatively significant. Another available price index is core inflation, which fundamentally shows the components of CPI that are affected by monetary policy. Various methods have been developed to measure core inflation (Roger, 1998). The exclusion method omits the price components beyond the control of the central bank from the CPI. The trimmed method statistically excludes price components within certain percentiles of the CPI distribution. Although this price index is more indicative of inflation that can be controlled through monetary policy and is, therefore, the responsibility of the central bank, this method is hampered because it is difficult to understand by the public and the measure could reduce the credibility of the inflation measure used. In general, EMEs applying ITF typically use the CPI to formulate the inflation target because it is easy to understand by the public and supports credibility (Table 8.3). Nonetheless, core inflation is also generally measured and monitored, not only because it is a measure that underlies monetary policymaking but also because core inflation indicates future CPI trends. Furthermore, core inflation could fundamentally be used in terms of central bank accountability by showing which components of inflation are influenced by monetary policy and which are Table 8.3:  Price Indexes Used. Price Index

Country

Consumer price index (CPI)

• Australia, New Zealand, Brazil, Chile, Israel, Poland, Spain, Sweden, and Indonesia

Core inflation

• Canada, United Kingdom, Finland, Czech Republic, and South Africa

Source: Schaechter et al. (2000, p. 10).

4

An alternative to CPI is the GDP deflator, which is a measure of the level of prices of all new, domestically produced, final goods and services in an economy. A weakness of the GDP deflator in terms of ITF is the availability of monthly data, which could impair its transparency and credibility.

238    Central Bank Policy outside the control of the central bank. In this case, to enhance credibility, core inflation should be measured by a third party, independent of the central bank, usually an institution dealing with economic statistics, based on a method jointly agreed and announced to the public. Subsequently, CPI could be replaced with core inflation as the inflation target after core inflation and the application of ITF have been widely accepted by the public, as has occurred in Australia (1998) and New Zealand (1999). Size of the inflation target. The inflation target can be set at a certain point or within a specific range. Using an explicit tolerance band provides greater flexibility to the central bank in response to shocks in the economy and allows discretion concerning the other policy targets in addition to inflation. Furthermore, the breadth of the inflation target corridor indicates how tolerant the central bank is to inflation fluctuations around the midpoint. On the other hand, however, a target range undermines central bank accountability and can complicate the formation of public inflation expectations. A broader range also reduces the focus on the formation of public inflation expectations and the central bank’s commitment to its monetary policy. Several countries prefer an inflation target anchored to a specific point, such as in United Kingdom, Spain and Finland or to a point with a stipulated margin, such as in Brazil and Sweden, because both cases are easier to understand by the public (Table 8.4). Conversely, a number of other countries prefer an inflation target specified as a certain range. In this regard, the breadth of the range will depend on the frequency and magnitude of the shocks in the economy, in addition to central bank credibility. Many central banks set the inflation target range at 2% or less, excluding Brazil and South Africa, where a larger range is preferred to cover price changes that are beyond the control of the central bank. Target horizon. The target horizon indicates the period during which the central bank is committed to achieve the inflation target set. In this context, as inflation represents the long-term target, the target horizon illustrates the lag of monetary policy transmission in terms of influencing inflation as well as the desire to avoid excessive changes in the use of monetary instruments and the impact on output. In Australia, New Zealand, United Kingdom, Sweden, and Chile, where the inflation target is set at a level approaching long-term inflation, the target horizon is set at more than one year (Table 8.5). A longer target horizon allows the central Table 8.4:  Inflation Target: Point or Range. Point or Range

Country

Target point

• United Kingdom, Spain, and Finland

Target point and margin

• Brazil, Sweden, and Indonesia

Target range of 2% or less

• Australia, Canada, Chile, Czech Republic, Israel, and Poland

Target range of more than 2%

• New Zealand and South Africa

Source: Modified from Schaechter et al. (2000, p. 11).

ITF Concept and Implementation at Central Banks     239 Table 8.5:  Inflation Target Horizon. Target Horizon

Country

Annual target

• Israel, Poland, and South Africa

Annual target announced several years in advance

• Brazil, Czech Republic, Spain, and Indonesia

Medium–long target horizon or not announced

• Australia, Canada, Chile, Finland, New Zealand, Sweden, and United Kingdom

Source: Modified from Schaechter et al. (2000, p. 8).

bank to respond to economic shocks and catalyze relatively more stable economic growth, in addition to anchoring public inflation expectations. On the other hand, a shorter target horizon could trigger instability, primarily if shorter than the lagged effect of monetary policy on inflation. In countries where inflation is above its long-term trend, a shorter target horizon could accelerate disinflation. Monetary policy can respond to shocks by adjusting the inflation target rather than through a change in policy stance. Many EMEs pursue a disinflation path in conjunction with a long-term inflation target. Some countries announce the disinflation path years in advance, such as in Czech Republic and Indonesia, while others announce the path annually, as in Poland. In general, setting an annual inflation target makes it easier to assess the success of efforts to reduce inflation, especially if disinflation is achieved more quickly than expected. Furthermore, an annual target allows monetary policy to absorb short-term inflationary pressures and output fluctuations, in addition to preventing an overly ambitious inflation target that would be impossible for the central bank to attain. Periodic monitoring of the inflation target is required, usually through the publication of an inflation report or monetary policy report, monthly or quarterly for instance. Inside, the reports describe annual inflation developments (year on year) compared to the inflation target, the determinants of inflation and current monetary policy. Furthermore, the reports contain future inflation projections (for instance with an inflation forecast fan chart) and the macroeconomic outlook in order to explain the future monetary policy direction and targets. In addition to monetary policy transparency in the eyes of the public, periodic monitoring, and the publication of an inflation report or monetary policy report will simplify the central bank’s accountability mechanism in terms of achieving the inflation target set. Announcing the inflation target. The central bank may announce the inflation target itself, or the government, or the government in conjunction with the central bank, depending on the mechanism used to determine the inflation target (Table 8.6). Announcing the inflation target by the government or together with the central bank has several advantages, including enhancing the credibility of the inflation target because the government also displays its commitment to ITF implementation by the central bank as well as fiscal policy implementation to

240    Central Bank Policy Table 8.6:  Announcing the Inflation Target. Institution

Country

Government

• Brazil, Israel, and United Kingdom

Central Bank

• Chile, Finland, Poland, Spain, Sweden, and Indonesia

Government and Central Bank

• Australia, Canada, Czech Republic, New Zealand, and South Africa

Source: Modified from Schaechter et al. (2000, p. 8).

support attainment of the inflation target. Theoretically, Debelle and Fischer (1994) suggested that the government must instruct the central bank of the target because the government represents the interests of the people, while the central bank, depending on the level of independence enjoyed, could set an inflation target considered too strict by the government. On the other hand, the government could prioritize short-term interests in terms of stimulating economic growth, thus setting an inflation target well above its long-term trend. Both of these issues can be overcome if the inflation target is set by the government after consultation with the central bank and announced jointly to the public. The inflation target is typically announced by the central bank alone in countries where the law explicitly mandates price stability as the final goal of monetary policy and provides jurisdiction to the central bank to set the inflation target, such as in Poland, Spain, and Sweden. Meanwhile, the central banks in Chile and Finland, where the goal of monetary policy is stated more implicitly, namely exchange rate stability, announce their inflation targets to show the ability of the central bank to interpret the legal mandate as domestic price stability. By announcing the inflation target, the central bank affirms its accountability to the public and is expected to enhance monetary policy credibility, considering that announcing the inflation target represents a symbol of delegating authority to set the goal of monetary policy from the government to the central bank. Escape clauses. Formulating the inflation target may also encompass certain requirements that clearly state the level of tolerance to inflation fluctuations (escape clauses). Inflation target formulation must also consider the trade-off between flexibility and credibility. For example, Canada, Czech Republic, New Zealand, and South Africa stipulate escape clauses in ITF to increase flexibility. In this context, the escape clauses state under which conditions the inflation target may not be achieved, while also requiring the central bank to indicate the horizon needed to achieve the inflation target again or communicate the new direction of the inflation target. One weakness of the escape clause is that it could be interpreted as a policy based merely on judgment and unable to anticipate all potential shocks. Therefore, the use of an escape clause in an ambiguous situation could undermine ITF accountability and credibility. Bernanke et al. (1999) showed that, thus far, escape clauses have only been triggered by New Zealand in 1990–1991 and 1993–1994, when the prices of oil and wood were omitted from the inflation basket.

ITF Concept and Implementation at Central Banks     241 8.3.1.3. Accountability and Transparency.  All central banks applying ITF operate transparently because public accountability is an integral element of successful monetary policy. The importance of the accountability mechanism is clear because of the lag between monetary policy and inflation, which complicates the monitoring of policy commitment in the near term. This differs from a regime targeting money supply or the exchange rate, where attainment of the inflation target can be monitored in the short term. In addition, accountability may also help safeguard monetary policy from outside political pressures. To improve accountability and transparency, central banks applying ITF periodically clarify their stance on achieving the inflation target and existing monetary policy through regular publications, monthly or quarterly, of the inflation report or monetary policy statement, submitted to the government, parliament, and public at large. The reports detail the latest economic and financial developments and their impact on inflation compared to the inflation target, clarify the rationale underlying monetary policy and explain the recent developments as well as discuss inflation projections for the upcoming 12–24 months. Such measures are also accompanied by an elucidation from the central bank to parliament concerning various important aspects as required. In several cases, including Czech Republic and Sweden, the central bank publishes the minutes of monetary policy discussions. Transparency is also enhanced through the immediate communication of the monetary policy measures decided by the central bank. In general, the announcement is made through a press release, stating the monetary policy decisions made and a summary of the underlying factors. High-ranking officials from the central bank also give presentations on the latest economic and financial developments, attainment of the inflation target as well as the monetary policy stance adopted to various parties, including the mass media. Such measures are required to improve broad and comprehensive understanding of monetary policy objectives and central bank performance. In Brazil, officials from the central bank give lectures and presentations to the private sector and media regarding the goal and implementation of monetary policy, while in Canada, in addition to media briefings and presentations through various forums, meetings are also held with various private sector groups to discuss the latest economic and financial dynamics (Schaechter et al., 2000). Various other forms of communication are also available to improve accountability. Central banks in advanced countries, and in Brazil, Chile, Israel, and Poland, also publish their research to enhance understanding of the various economic and financial dynamics underlying inflation projections. In addition to posting online, central banks also sponsor seminars and workshops where their researchers can discuss compelling issues with external experts. Hitherto, only central banks from advanced countries have published detailed models used to generate inflation projections. Meanwhile, not many central banks in EMEs have published their models because they are still under development (Schaechter et al., 2000).

242    Central Bank Policy 8.3.2. Operational Framework ITF implementation requires the central bank to hone its monetary policy strategy and operational framework. In general, monetary policy at ITF central banks is implemented as follows (Schaechter et al., 2000): ⦁⦁ Regular inflation projections are prepared and updated based on the latest eco-

nomic data, market sentiment indicators, modeling results, and judgment.

⦁⦁ The direction of monetary policy, in the form of an operational target required

to minimize deviations between inflation projections and the inflation target paying due regards to the output gap and other factors, is determined based on modeling and judgment. ⦁⦁ Other factors, like international developments and political events, which could influence the timing and magnitude of changes in the operational target, must also be taken into consideration. ⦁⦁ Changes in policy direction are announced and the operational target is adjusted according to the desired level.

Fundamentally, ITF-based monetary policy requires three elements, namely a model or method to project future inflation, a review of the policy transmission mechanism, and effective monetary control when setting the operational target and monetary instruments to be used. In general, central banks in advanced countries and EMEs implement a similar monetary policy strategy. In all cases, the central bank must rely on judgment when formulating monetary policy. Countries that have already developed statistical models can base their monetary policymaking decisions on such models. In other countries, however, particularly EMEs, many factors must be considered because of underdeveloped statistical models as well as limited data availability and ongoing structural changes in the economy. Furthermore, emerging economies also face far more economic and financial shocks, in addition to underdeveloped financial markets. 8.3.2.1. Inflation Projections.  In ITF, inflation projections play a key role in terms of monetary policy formulation, primarily considering the lag between monetary policy and its impact on inflation. Therefore, the ability to predict inflation will determine how well monetary policy can steer the economic and financial outlook in line with the inflation target set. In many cases, inflation projections can be viewed as an intermediate target to achieve the inflation target as understood in the monetary policy strategy of monetary economics (Svensson, 1997). In practice, the ability to project inflation is based on a combination of three factors: namely inflation indicators, economic modeling, and qualitative judgment (Schaechter et al., 2000). In this case, central banks applying ITF typically assess the accuracy of projections from economic modeling as well as the various inflation and economic indicators selected. Subsequently, the central bank makes a judgment based on the qualitative information obtained through contact with various economic sectors to decide the most appropriate monetary policy direction.

ITF Concept and Implementation at Central Banks     243 Inflation indicators. Monetary inflation indicators, encompassing an observable data and information set that can signal the direction of future changes in inflation, are used to formulate monetary policy (Schaechter et al., 2000).5 Numerous inflation indicators are available in countries that apply ITF. Fundamentally, inflation indicators cover aggregate demand and supply variables, measures of interest rates and exchange rates, inflation, price measures, and expectations. Many EMEs tend to focus on indicators of aggregate demand and supply, considering the large economic shocks encountered and the lack of alternative indicators to reflect market conditions. Furthermore, new inflation indicators need to be created and old ones improved. Many central banks in EMEs work in conjunction with existing statistical institutions to measure core inflation, which can signal inflation trends. Central banks also identify or create new inflation indicators based on market conditions. Surveys of inflation expectations conducted by third parties or the central bank itself are also often utilized as inflation indicators. In this case, such surveys are designed to reveal inflation expectations at the corporate and consumer levels (New Zealand). Central banks also summarize the results of existing surveys to indicate inflation expectations (United Kingdom). Private sector qualitative surveys are also used in Canada and New Zealand. Furthermore, the central bank could also consider informal information gleaned from discussions or meetings with various private parties. Government bond yields are also an indicator of inflation. Shifts in spread between various bond yields have also been used in Australia, Canada, Israel, New Zealand, and United Kingdom to assess the impact of monetary policy and various other factors on inflation expectations and the inflation premium contained in the bond yield spread. Inflation expectations can also be measured using the spread between nominal and real forward interest rates. Nevertheless, this kind of data must be used with caution considering that spread between bond yields or forward interest rates contain the markets’ perception of default risk. Economic models. Structural economic modeling is also often used by central banks to calculate inflation projections. In general, such models cover the aggregate demand curve for open economies, international asset market equilibrium and a monetary policy reaction function (Bogdanski, Springer, Goldfjan, & Tombini, 2000). These elements have been applied in various empirical research on central bank behavior, monetary policy rules, and the sacrifice ratio (Taylor, 1999). In addition, the equations are oftentimes incorporated in structural macroeconomic models at central banks, including the Quarterly Projection Model at the Bank of Canada and the FRB model at the US Federal Reserve Board.

5

Per the monetary policy framework, there are fundamentally two types of monetary policy indicators (Bofinger, 2001; Handa, 2000). The first are economic and financial indicators that show the direction of future inflation fluctuations, also known as leading indicators. The second are the economic and financial indicators selected as policy indicators. In terms of ITF, inflation indicators are considered leading indicators.

244    Central Bank Policy The use of large structural macroeconomic models is common in advanced countries. Such models have helped guide monetary policy formulation for a long time and maintain a relatively stable relationship. Canada and New Zealand use dynamic models to produce future inflation and economic projections (Drew & Hunt, 1998). The models are calibrated to generate simulations for theoretical assessment as the models are based on historical data, which is less useful for policy simulations. The development of large-scale models requires a protracted amount of time, while experience of ITF implementation remains comparatively brief. Many advanced countries also utilize small structural models to analyze specific sectors or key variables (Drew & Hunt, 1998). Meanwhile, many EMEs use judgment combined with economic models. Relatively brief experience of ITF implementation impairs the comparatively stable relationship between economic variables when projecting inflation, on top of the structural changes that often occur in the economy (Leiderman & Baror, 2000). Nevertheless, an increasing number of central banks in EMEs are benefitting from using such models to formulate monetary policy and have, therefore, allocated more resources to the development of economic models. Brazil, Czech Republic, and Israel have developed models with three or four key equations that are often used in advanced countries (Bodganski et al., 2000; Clinton, 2000). In addition to producing inflation projections, economic models can facilitate monetary policy in many significant ways (Issard & Laxton, 2000). Small-scale models can help the central bank rationalize policy transmission through the various channels considered dominant. More importantly, the models can help as frameworks to discuss policy and even facilitate the presentation of inflation projections to increase transparency. Fan charts, which show the distribution of inflation projections covering various scenarios, are the latest upgrade in terms of representing inflation projections based on economic models (Allen, 1999). Time series models are thought to place fewer restrictions on economic structure but make short-term projections, as well as cross-checking the inflation projections produced by large-scale models. Univariate models for an aggregate inflation index, the components of inflation or other key indicators are comparatively easier to build and facilitate more cross-checking with the direction of inflation. In New Zealand, the results of such models are used to generate projections for the last two quarters after the historical data have ended. Multiple time series models, such as vector autoregressive (VAR) models, also provide simultaneous interactions between key variables together with fewer structural restrictions. Such models are used by the central banks in Brazil, Chile, and Israel. 8.3.2.2. Policy Transmission. As elucidated in the previous chapter, deep understanding of the monetary policy transmission mechanism is an integral element of implementing monetary policy. Such understanding becomes even more important in ITF because it will determine the effectiveness of monetary policy in terms of steering the future direction of inflation in line with the target set. In other words, considering that the impact of monetary policy on the economy and inflation occurs after a lag, inflation targeting necessitates pre-emptive monetary policy.

ITF Concept and Implementation at Central Banks     245 Issues arise during implementation. Monetary policy has an imperfect influence on inflation. Not only because sources of inflation originate from non-monetary factors, such as APs as well as disruptions to the supply and distribution of goods and services, but also, in the words of Milton Friedman, monetary policy works with a relatively long and varied lag. As elaborated in Chapter 5, the channels through which monetary instruments influence inflation include money supply, interest rates, exchange rates, and expectations. Central bank understanding of the monetary policy transmission mechanisms is a prerequisite to formulate monetary policy. Studies on monetary policy transmission mechanisms are also required to comprehend the relative effect of each channel in an economy. Such studies can be used to identify and create important indicators for monetary policy formulation because they show the magnitude and lag of monetary policy’s impact on inflation and real output through each respective channel. Considering that the use of intermediate targets (other than inflation projections), as proffered by Svensson (1997), is not in keeping with the ITF, most central banks underscore several economic and financial indicators as the basis of monetary policy formulation. Such indicators cover a variety of economic and financial variables that contain a wealth of information on the future direction of inflation.6 Although not stressing which indicators should be used, many central banks depend on various information sources when setting monetary policy.7 In an open economy, the transmission channels between monetary policy actions and inflation work directly and indirectly (Ball, 1999; Svensson, 1998). In the near term, the direct impact of monetary policy occurs through changes in the exchange rate, where a hike to the interest rate directly reduces inflationary pressures through a stronger exchange rate, which feeds through to import prices. Over the longer term, however, a hike to the interest rate indirectly lowers inflation, first by lowering inflation expectations and, therefore, corporate investment, and secondly by reducing consumption and, thus, prompting exchange rate appreciation that diverts spending from traded to non-traded goods. Of course, a central bank cannot rely solely on the exchange rate to transmit monetary policy because economic players will gauge the extent to which monetary policy influences inflation through the interest rate or exchange rate. Besides, monetary policy can also affect inflation indirectly through changes in wealth and aggregate demand. Furthermore, in the case of underdeveloped and unestablished financial markets, monetary policy transmission channels can be compromised by corporate balance sheets and bank lending behavior. In the Czech Republic, for example, monetary policy transmission was impaired by the inelasticity of bank lending policy to the interest rate due to high non-performing

6

Examples include the monetary conditions index developed by the central banks of New Zealand and Canada, and the broad money (M2) indicator used by the central bank of Spain. 7 In other words, the most common monetary regime practiced in ITF countries is constrained discretion.

246    Central Bank Policy loans (NPL) and lending to unproductive businesses. In Poland, the credit channel was undermined by underdeveloped financial markets and a deluge foreign funds. The monetary policy transmission channels in EMEs, where inflation is typically high, are characterized by sticky prices (price rigidity) and the rapid impact of the exchange rate on inflation. Brazil, Chile, and Israel, which had a long history of high inflation, downward price inertia and an almost immediate exchange rate impact on inflation prior to ITF, have become more credible (Bogdanski et al., 2000). Exchange rate pass-through to inflation has also faded in advanced countries, to around six to eight quarters (Menon, 1995). In New Zealand, the short-term direct exchange rate channel was observed to decline with the advent of hedging and the longer duration impact of monetary policy on inflation. In the case of Brazil, the indirect channels of monetary policy through demand have a lag of two to three quarters on inflation. 8.3.2.3. Policy Implementation.  The operational target of monetary policy normally used by ITF central banks are short-term interest rates, usually overnight to 1 month. In addition to closely reflecting the actions of the central bank, short-term interest rates are used because they can be transmitted to the entire interest rate spectrum and, thereafter, to the economy and inflation. Furthermore, the growing popularity of interest rates as the operational target is a corollary of the general central banking trend to move away from targeting money supply in monetary policy that might not be consistent with ITF (Borio, 1997). Several central banks in countries with an open economy utilize a combination of interest rates and exchange rates, contained in an index known as the monetary condition index (MCI), as the operational target or to signal monetary policy to the markets. MCI is used because of the complexity involved with differentiating monetary policy transmission through the interest rate channel and exchange rate channel in terms of the influence on inflation in a small open economy. In practice, however, the use of MCI also entails complications due to market expectations that all exchange rate fluctuations, even daily changes, will elicit a response from the central bank through changes in interest rates to maintain MCI and the fact that exchange rate fluctuations can impact MCI differently (Freedman 1994; Hunt, 1999). Canada first used such an index at the end of the 1980s but subsequently reduced its role as a signaling tool in 1998. New Zealand used to announce its desired MCI direction and range but shifted to an interest rate operational target in the middle of 1998 to avoid short-term interest rate fluctuations. In the implementation of monetary policy, all ITF central banks use instruments based on market mechanisms to maintain the interest rate operational target at the desired level. Open market operations, outright and repo, are the main instruments used by central banks to maintain market liquidity consistent with the published interest rate target. OMO involve buying and selling government securities or securities issued by the central bank itself. Meanwhile, direct monetary instruments, such as the reserve requirement or lending requirements, are rarely used because they are considered less effective in terms of overcoming inflationary pressures or as a signal of monetary policy.

ITF Concept and Implementation at Central Banks     247 The interest rate was set as the operational target in reaction to deviations of inflation projections from the target and the output gap. In this case, the interest rate response in ITF is more forward looking compared to other monetary policy frameworks. In general, interest rates are set based on a mechanism often referred to in monetary economics as the Taylor rule. In practice, however, central banks bolster this approach with judgment based on various information that support future inflation projections. Judgment is more commonly used in EMEs because, among others, economic projection models remain under development or contain unstable inter-parameter relationships, while the corresponding economies are often blighted by relatively large shocks and data availability is limited.

8.4. Inflation Targeting Regimes Currently, around 40 countries, including advanced countries and EMEs, have applied ITF as their central bank’s monetary policy framework.8 In general, the 40 countries apply a more flexible exchange rate system and explicitly acknowledge their commitment to controlling inflation in their monetary policy. Furthermore, the countries avoid implementing a fixed exchange rate system in order to limit economic vulnerability to speculative exchange rate attacks, while also increasing their monetary policy independence from external influences. Simultaneously, monetary policy based on targeting money supply is becoming less effective because of unstable demand for money, among others.

8.4.1. Regime and Rationale In practice, there are variations in the extent to which ITF is applied from country to country depending upon the specific conditions of the country involved. In this case, Carare and Stone (2003) classified three ITF regimes, namely FFIT, EIT and ITL. The classification was based on the clarity and credibility of the central bank’s commitment to the inflation target. Clarity was reflected in the announcement of the inflation target to the public and with institutional arrangements that support accountability to achieve the target. On the other hand, credibility was measured using the achievement of actual inflation and the domestic government’s debt rating. FFIT is the most prominent ITF regime. FFIT countries have a medium-high level of credibility, clear commitment to the inflation target and institutional commitment in the form of a transparent monetary policy framework to support central bank accountability to attainment of the inflation target. For such countries,

8

In the selection of the 42 countries, Carare and Stone (2003) used the following assumptions: (1) that a country applying a flexible exchange rate regime is committed in some way to an inflation target; and (2) a total of 89 small countries with GDP below USD 4 billion and GDP per capita below USD 720 were omitted from the 185 IMF member states.

248    Central Bank Policy applying FFIT empowers the central bank to maintain monetary policy consistency in terms of controlling inflation at a certain pace of economic growth. An EIT regime also enjoys a relatively high level of monetary policy credibility to maintain low and stable inflation without the need to fully emphasize transparency and accountability through a specified inflation target. Low and stable inflation, coupled with financial system stability, allows the central bank to pursue output stabilization and price stabilization. Countries adopting ITL announce the inflation target but due to low credibility are unable to maintain inflation at the announced target. Relatively low credibility reveals vulnerability to economic shocks and financial instability as well as a weak institutional framework. ITL can be viewed as a transitional regime, where the authorities are implementing the structural reforms required to credibly apply a single inflation target. ITF regimes can be viewed as a combination of policy objectives to maximize social welfare, each based respectively on the level of credibility enjoyed by the country involved. The final goal of monetary policy is to maximize social welfare through robust and stable economic growth in the long run. Monetary policy can support long-term economic growth through a combination of single-digit inflation, financial stability, and output stability. The combination of those three policy objectives to maximize social welfare in the formulation of monetary policy depends on the level of credibility garnered in a country. Therefore, the three ITF regimes mentioned above can be explained by the weight given to the three policy objectives and clarity of commitment to the inflation target. FFIT has a medium-high level of credibility and solid financial stability but cannot maintain low and stable inflation without commitment to the inflation target. The clarity and transparency of such commitment incurs a cost in the form of less flexibility to stabilize output. EIT countries have already maintained low inflation and a stable financial system and, therefore, can credibly achieve price stability and enjoy flexibility to stabilize output. The dual objectives mean that EIT countries cannot operate as transparently as FFIT countries. On the other hand, ITL countries do not have enough credibility to maintain low inflation and are vulnerable to financial and cyclical shocks. Consequently, EIT countries are unable to steer their monetary policy toward achieving an inflation target, but still require the flexibility to confront shocks.

8.4.2. Clarity of Commitment to the Inflation Target The clarity of commitment to the inflation target demonstrates a fundamental difference between ITF regimes and other monetary regimes based on exchange rates or monetary aggregates. In terms of exchange rate targeting, the central bank can easily be held accountable if monetary policy fails to stabilize or achieve a certain exchange rate because exchange rates are easy to monitor in real time. Similarly, in a regime targeting money supply, it is easy to measure whether the target has been achieved because money supply statistics are available with a lag of only one or two months. On the other hand, however, achievement of the inflation target per ITF can only be observed after a lengthy and variable lag

ITF Concept and Implementation at Central Banks     249 because of the time required from when the policy action is taken until its effect on inflation can be observed. Therefore, inflation target clarity manifests in stages determined by application of the ITF monetary policy framework. Inflation target clarity contains two elements. The first is clarity and public communication from the central bank concerning the inflation target to be achieved. In general, central banks that apply ITF announce their inflation target. The difference is in the clarity of the level or range of the inflation target, the time horizon, the type of price index used and the monetary policy that will be and has been pursued to achieve the inflation target. Ex-ante and ex-post monetary policy clarity, communicated effectively and in detail, will facilitate central bank accountability in terms of achieving the target. The second is transparency concerning the institutional arrangements required so that the public and markets can monitor monetary policy and hold the central bank accountable for the target. Institutional transparency is reflected by the preferred communications media used by the central bank, including the publications, frequency, and depth of the analysis contained in the inflation report or similar, the announcement of changes to the direction and target of monetary policy contained in a press release or other form, an assessment of inflation target attainment and the monetary policy implemented, whether the inflation projection model has been published, as well as the various other forms of communication media and publications used.9

8.4.3. Inflation Targeting Regime Credibility Despite clear commitment to inflation, a country is not always considered a serious proponent of ITF if the inflation target set is comparatively high and negative market sentiment exists toward the central bank. In this case, the credibility of commitment to achieving the inflation target is the main differentiator of ITF. Carare and Stone (2003) used two indicators to measure the credibility of inflation targeting regimes, the first being inflation performance using average inflation over several years or a comparison between inflation realization and the target or the level of public inflation expectations. Low and stable inflation supports robust and stable economic growth in the long run (Sarel, 1996) and, therefore, monetary policy that supports long-term economic growth is deemed more credible. Consequently, countries that maintain a relatively low rate of inflation are considered to have more credible monetary policy. The second indicator used was the rating of government securities denominated in the domestic currency, for instance the rating affirmed by Standard and Poor’s or another international rating agency. This indicator negated the weaknesses inherent to the inflation achievement indicator, for which the data were relatively short term in the few countries new to ITF application, because the rating directly captures market perception regarding confidence in long-term financial

9

The issues of central bank transparency and communication strategy will be discussed in Chapter 12.

250    Central Bank Policy stability, which is ultimately the responsibility of the central bank. Furthermore, this indicator also illustrates the strength of the government’s fiscal position, which also determines the credibility of commitment to the inflation target. In countries suffering from a comparatively large fiscal deficit, offset by government issuances of debt securities, any fiscal considerations will influence the central bank’s monetary policy and, therefore, undermine the achievement of no fiscal dominance as a prerequisite of ITF implementation.

8.4.4. Inflation Targeting Regime Classification Based on the analysis of commitment clarity and credibility to inflation targeting, using inflation data from 1999 to 2002 and government debt ratings from 2001, Carare and Stone (2003) classified 43 countries into three inflation targeting regimes (Table 8.7). The first regime consists of 18 countries with explicit commitment to an inflation target and applying a transparent monetary policy framework to ensure central bank accountability to achievement of said target. These countries were classified as FFIT because they are fully committed to achieving the inflation target set. It is interesting to note that the 18 FFIT countries include small and medium sized advanced countries as well as medium and large EMEs. The second group consists of five central banks with sound credibility, which allows for more implicit commitment to the inflation target. These countries have been very successful in maintaining low inflation but implement different monetary policy frameworks, particularly in terms of price stability and monetary operations. They appear to enjoy more flexibility to orient monetary policy toward goals other than price stability, such as economic growth or exchange rate stability and are, therefore, known as EIT. All members of this group are advanced countries. Countries in the third group announce an inflation target, but due to low credibility cannot offer full commitment to the target. They are also heterogeneous in terms of formulating the goals or monetary operations. This regime is known as ITL because the practicing countries cannot explicitly give credible commitment to the inflation target. A total of 19 countries were classified per this regime, all of which are EMEs.

8.4.5. The Importance of Regime Classification In addition to explaining the practical variations of ITF implementation at different central banks, inflation targeting regime classification also reinforces the analysis of countries moving from one regime to another. From discussions on the criteria of each respective regime, the ideal conditions to be considered include which requirements are necessary for an ITL country to enhance commitment credibility to the inflation target, thus raising its status to FFIT. In this case, a comparative study of FFIT and ITL EMEs would provide a clearer picture. In terms of the macroeconomic conditions, Carare and Stone (2003) showed that the level of economic and financial development in FFIT EMEs (EMFFIT)

ITF Concept and Implementation at Central Banks     251 Table 8.7:  Inflation Targeting Regime Classification. FFIT

EIT

ITL

Advanced countries

Advanced countries

EMEs

1. Australia

1. US

1. Peru

2. Canada

2. ECB

2. Venezuela

3. New Zealand

3. Japan

3. Uruguay

4. United Kingdom

4. Switzerland

4. Guatemala

5. Iceland

5. Singapore

5. Jamaica

6. Norway

6. Dominican Republic

7. Sweden

7. Honduras 8. Russia

EMEs

9. Croatia

8. Brazil

10. Slovenia

9. Chile

11. Slovakia

10. Colombia

12. Albania

11. México

13. Kazakhstan

12. Czech Republic

14. Rumania

13. Hungary

15. Philippines

14. Poland

16. Indonesia

15. Israel

17. Sri Lanka

16. South Africa

18. Mauritius

17. South Korea

19. Algeria

18. Thailand Source: Excerpted from Carare and Stone (2003).

was generally higher and more advanced than in ITL EMEs (EMITL). First, the average GDP in EMFFIT was six times that of EMITL and GDP per capita was three times that of EMITL. Second, monetization and financial markets are more advanced in EMFFIT compared to EMITL. The ratio of money supply to GDP was also around 40% larger in EMFFIT. Capital market capitalization was also much higher and real interest rates lower in EMFFIT. More advanced financial sector conditions explain why EMFFIT can more effectively implement monetary policy through indirect and interest rate instruments as the operational target. Third, EMFFIT countries maintain a stronger fiscal position with clear restrictions on central bank financing of government operations. Although the average fiscal surplus/deficit is not particularly disparate between the two regimes, EMITL suffer from a larger position of government debt, which is important as a reflection of the no-fiscal-dominance prerequisite of ITF application.

252    Central Bank Policy A comparison can also be made of the operational targets and monetary policy instruments used by the central bank. In general, despite using the interest rate as the operational target and open market operations as the instrument, the clarity of the operational framework is different between EMFFIT and EMITL regimes. In EMFFIT countries, the elements of the monetary policy operation framework are explicit, namely short-term interest rates as the operational target and open market operations as the instrument. Conversely, EMITL countries generally apply a mix of operational targets and monetary instruments. In addition to short-term interest rates, other operational targets are also used, such as the exchange rate and the monetary base. Some EMITL countries also still use direct instruments, including the reserve requirement and lending requirements. On top of comparatively underdeveloped financial sectors, the use of several operational targets and monetary instruments is possibly linked to the prevailing dual goals of monetary policy. The analysis above shows which preparations are required by EMITL countries to upgrade their status to the EMFFIT regime. Specifically, EMITL countries must introduce structural and policy changes prior to FFIT implementation. Carare and Stone (2003) suggested a number of interesting observations. First, EMEs generally apply FFIT during periods of disinflation and to build credibility. Second, the ratio of government debt to GDP is typically tracking a downward trend thanks to sounder debt management and disinflation. And third, the ratio of money supply to GDP is increasing, as reflected by the growing role of the financial sector. These observations support the preconditions proposed by Bernanke et al. (1999) and other economists, who showed that prior to FFIT implementation, EMEs strived to lower inflation, consolidate government fiscal conditions, and deepen the financial sector. In other words, such conditions illustrate that EMEs need to try harder to build credibility in terms of ITF implementation.

8.5. Inflation Targeting and Economic Performance 8.5.1. Inflation Targeting Successes and Alternative Opinions With conceptual understanding, institutional and operational frameworks as well as an inflation targeting regime as described in previous sections, the question arises about whether ITF implementation, as a monetary policy framework, contributes positively and significantly to economic performance and the central bank. Specifically, does ITF implementation generate lower inflation and inflation expectations? If yes, how much disinflation must be achieved with lower real output and volatility, particularly economic growth and unemployment, to interest rates in the country involved? Does ITF implementation trigger changes in the way the central bank implements its monetary policy? Is economic performance better in countries applying ITF than in other countries? Several studies have researched ITF application in different countries and its impact on the central bank and economic performance. Some of the studies analyzed the basic features of ITF, including Haldane and Simon (1995), Masson,

ITF Concept and Implementation at Central Banks     253 Savastano, and Sharma (1997), Agenor (2000), Blejer et al. (2000), Fry, Julius, Mahadeva, Roger, and Sterne et al. (2000), Schaechter et al. (2000), Carare, Shaechter, Stone, and Zelmer (2002), Sterne (2002), Amato and Gerlach (2002), and Mishkin (2004). Other cross-border studies focused on a comparison of macroeconomic performance and inflation stabilization using a sample of ITF countries, including Schmidt-Hebbel and Werner (2002) for Brazil, Chile, and México, as well as Kahn and Parrish (1998), Mishkin and Posen (1997), Kuttner and Posen (1999), and Siklos (1999) for industrialized nations. Other studies have compared macroeconomic performance in ITF countries with countries practicing other monetary policy regimes, including studies by Bernanke et al. (1999), Cecchetti and Ehrmann (2000), Corbo, Landerretche, and Schmidt-Hebbel (2000), Mishkin and Schmidt-Hebbel (2001), Landerretche, Corbo, and Schmidt-Hebbel (2001), Neumann and Von-Hagen (2002) as well as Ball and Sheridan (2003). This section will review a number of seminal empirical studies, particularly those containing cross-border comparisons. The discussion will emphasize two main points to provide a more objective view concerning the pros and cons of ITF implementation. First, opinions that sum up success stories of ITF implementation as summarized from various empirical studies of ITF countries, such as those suggested by Mishkin and Schmidt-Hebbel (2001). Second, opinions that criticize the methodologies and empirical findings of previous studies, such as the latest study by Ball and Sheridan (2003). Reviews of other relatively recent studies are available in Launderretche et al. (2001) as well as Neumann and VonHagen (2002). 8.5.1.1. Inflation Targeting Successes. Mishkin and Schmidt-Hebbel (2001) are perhaps the best reference to draw several success stories during the first decade of ITF implementation in the 1990s from various previous empirical studies. Nevertheless, it is important to note that because of differences in the methodologies used, countries sampled, and periods analyzed, the conclusions made by the two economists tended to emphasize a number of salient and relevant empirical findings that provide valuable lessons. In other words, the review was not intended as an incontrovertible conclusion or a substitute for existing empirical studies or those still required. In general, Mishkin and Schmidt-Hebbel (2001) concluded that ITF implementation was proven as a very successful new monetary policy framework according to the experiences of the countries involved and a viable alternative to the monetary regimes applied in advanced countries during the 1990s. In summary, the two economists put forward eight success stories as follows: (1) ITF has helped implementing countries to lower inflation, albeit not below the levels attained in non-ITF advanced countries. The empirical evidence shows that ITF countries lowered long-term inflation beyond what would have been possible without ITF implementation. Nevertheless, the introduction of ITF did not create inflation lower than that achieved in advanced countries applying other monetary policy regimes, as evidenced by Bernanke et al. (1999) and Mishkin (1999). The empirical evidence shows that ITF implementation can be interpreted as an investment in strengthening monetary policy

254    Central Bank Policy to initially produce significant disinflation, equivalent to the performance of other advanced countries that have successfully controlled inflation. (2) ITF has been proven to overcome economic shocks. The 1990s, excluding the period when a devastating financial crisis befell EMEs from 1997 to 1999, was a golden age for the global economy, with relatively low inflation and the US economy driving expansion. Such conditions galvanized the opinions of several economists that ITF implementation was untested in terms of overcoming economic shocks that undermined the achievement of low and stable inflation in the affected countries. Mishkin and Schmidt-Hebbel (2001) were of another opinion, however. Many small open economies applying ITF faced significant shocks in the wake of the Asian Financial Crisis of 1997/1998 compared to advanced countries that were largely unaffected. A combination of financial and ToT shocks experienced by Australia, Chile, Israel, and New Zealand, in addition to other ITF countries, led to significant exchange rate devaluation and served as the prefect proving ground for ITF implementation. Those countries successfully overcame the relatively small external pass-through shocks from exchange rates to inflation in their respective jurisdictions. Furthermore, the oil price shocks of 1999–2000 were another rigorous test for ITF implementation, especially for large importers of oil, such as Brazil, Czech Republic, Poland, and others. A significant spike in imported inflation, through rising energy prices and exchange rate devaluation, should have threatened the inflation targets set. Nonetheless, evidence from ITF countries showed that the impact of the oil price shocks on core inflation was comparatively small, while the impact on higher headline inflation was small and temporary.10 (3) ITF has helped reduce the sacrifice ratio and output volatility in practicing countries to levels approaching the performance of non-ITF advanced countries. The cost of disinflation can be measured using the sacrifice ratio, namely the percentage loss of output per the percentage decline of inflation. Technically, the ratio is calculated using the deviation of potential output from the actuals during a given period divided by CPI disinflation during the same period. In other words, the sacrifice ratio is a measure of the costs associated with slowing down economic output to bring inflation toward the target set. Bernanke et al. (1999) found that ITF implementation did not lower the cost of disinflation per the sacrifice ratio (and Philipps curve) in ITF countries, which was not observed to change dramatically after ITF implementation nor was it different to the levels found in non-ITF advanced countries. Notwithstanding, based on the latest evidence from a larger sample of ITF countries, Corbo et al. (2000) concluded that the sacrifice ratio experienced a decline, especially in EMEs after ITF implementation. They also found that output volatility decreased in EMEs and advanced countries after

10

Similar findings were also put forward by Neumann and von Hagen (2003).

ITF Concept and Implementation at Central Banks     255 ITF implementation to a level on par with (and sometimes lower than) that observed in non-ITF advanced countries. (4) ITF has helped anchor and influence inflation expectations and overcome inflation shocks. According to Almeida and Goodhart (1998) as well as Bernanke et al. (1999), ITF application does not immediately lower inflation expectations, only gradually. Corbo et al. (2000) reported that erroneous inflation projections based on VAR models in ITF countries were reduced consistently by ITF implementation to the low levels found in non-ITF advanced countries. They also found that inflation persistence had declined significantly among ITF countries during the 1990s, showing that by setting and announcing an inflation target, ITF had successfully anchored inflation expectations compared to formulation based on previous inflation. (5) ITF-based monetary policy is adequately flexible in its symmetrical response to inflation shocks and in terms of accommodating temporary inflation shocks, which do not affect attainment of the medium-term target. ITF countries are not considered “inflation nutters” per King (1996), because they generally react symmetrically to positive and negative shocks in the economy, strive to gradually lower inflation and react to temporary output shocks as well. The evidence discovered by Cecchetti and Ehrmann (2000) showed that output deviation has a positive weight in all objective functions of a central bank applying ITF. (6) Monetary policy has a clearer focus on inflation and can be more tightly applied through ITF implementation. The mandate of the central bank to maintain price stability was reinforced by ITF implementation (Bernanke et al., 1999). Cecchetti and Ehrmann (2000) showed that central bank aversion to inflation shocks (relative to output shocks) was strengthened by ITF implementation, a conclusion that was corroborated by Corbo et al. (2000). (7) Central bank independence could be strengthened by ITF implementation. Experience from several countries in the 1990s showed that greater independence was afforded to the central bank in support of ITF implementation. In several countries, ITF was introduced after formal independence and instruments were given to the central bank, such as in the cases of New Zealand and Chile. In other countries, including United Kingdom, instrument independence was given after ITF implementation. That positive correlation was further confirmed by the empirical findings concerning formal independence and instruments, but not for target independence. (8) Communication, transparency, and accountability are strengthened through ITF implementation. In general, ITF implementation is accompanied (or sometimes preceded) by improvements underlying central bank communications with the public and markets to enhance monetary policy transparency. Most central banks applying ITF have begun publishing inflation reports, monetary policy statements, minutes of board of governors’ meetings, central bank models, and inflation projections. Such underlying communication efforts employed by central banks appear to be strengthened by ITF application compared to other monetary policy regimes, primarily because of the crucial role played by policy credibility and the management of inflation expectations to achieve the inflation target per ITF (Bernanke et al., 1999).

256    Central Bank Policy 8.5.1.2. Alternative Opinions.  The conclusions made by Mishkin and SchmidtHebbel (2001) generally reflect the views of economists as well as central bankers applying ITF. The ITF success stories underlie the economists’ volition to propose ITF as the ideal framework for a new monetary policy regime. Proponents of this monetary policy framework refer to the various advantages associated with ITF implementation, such as those described above. With such a range of advantages, ITF appears to be a suitable new monetary policy paradigm for implementation in other countries, including the US and ECB. In their latest study, however, Ball and Sheridan (2003) revealed some empirical findings that could change or, at least, soften this view. In general, the two economists concluded that, if the focus is merely on ITF countries, macroeconomic performance has indeed improved with the implementation of the ITF. Nevertheless, macroeconomic performance in non-ITF countries also improved during the 1990s. Therefore, the question arises: were the macroeconomic gains reported in ITF countries due to ITF implementation or other more general reasons that were also found in other countries. Seeking to investigate this issue more rigorously, Ball and Sheridan (2003) researched the impact of ITF implementation on macroeconomic performance in 20 OECD member countries, seven of which applied ITF prior to 1999 (New Zealand, Australia, Canada, United Kingdom, Finland, Sweden, and Spain) and 13 others that did not. The study analyzed whether ITF affected several dimensions of macroeconomic performance, such as inflation, output growth, and interest rates, which differed between ITF and non-ITF countries. Departing from previous studies based on VAR models or structural macroeconomic models, Ball and Sheridan (2003) based their methodology on the “differences in differences” approach. To that end, the observation sample was divided into several subsamples of before and after ITF implementation. By separating the subsamples, the impact of ITF on inflation, output growth, and interest rates could be analyzed, while also controlling the specific influence of countries, observation periods, and ITF implementation itself. The results of this study not only confirmed the previous empirical findings but also provided some different conclusions. In general, the salient empirical results concerning the impact of ITF implementation on inflation, economic growth, and interest rates can be summarized as follows. First, ITF’s impact on inflation. Similar to the findings of several other studies, ITF implementation can lower the rate and variability of inflation to levels comparable in non-ITF advanced countries. Likewise, ITF implementation can reduce inflation persistence toward the target set rather than previous inflation (inertia). Nonetheless, when compared to non-ITF advanced countries, Ball and Sheridan (2003) found no irrefutable evidence that ITF implementation provided additional advantages. In terms of disinflation, for instance, if the influence of initial inflation conditions is omitted, ITF does not provide any different advantages than in the other countries. Furthermore, ITF implementation was also not proven to decrease inflation variability any more than in non-ITF countries. Similarly, the decline of inflation persistence found in ITF countries was also observed in non-ITF countries.

ITF Concept and Implementation at Central Banks     257 Second, ITF’s impact on output growth. Ball and Sheridan (2003) found that ITF implementation was not unequivocally shown to stimulate real economic growth, as effused by ITF supporters (e.g., Mishkin, 1999). Average economic growth was indeed observed to accelerate after ITF implementation, rather than decline as in the case of non-ITF countries. Nevertheless, the difference between the two groups was not significant. Similarly, Ball and Sheridan (2003) found no empirical evidence that the implementation of flexible ITF stabilized inflation and output, or that output variability increased if ITF was applied too strictly. During the period when ITF was first implemented, output variability was observed to increase in ITF countries. Nonetheless, output stabilized after ITF implementation and was not significantly different to output variability reported in non-ITF countries. Third, ITF’s impact on interest rates. ITF implementation was said to reduce inflation permanently toward the target and, therefore, lower long-term interest rates. Ball and Sheridan (2003), however, found no evidence for this. Although long-term real interest rates decreased in ITF countries, the improvements were not particularly different from those enjoyed in non-ITF countries. Similarly, ITF implementation has not been shown to significantly change central bank behavior in terms of applying interest rate monetary instruments. Despite higher shortterm interest rate volatility in ITF countries, which subsequently declined, similar observations were also found in non-ITF countries. With their empirical findings, Ball and Sheridan (2003) concluded that ITFbased monetary policy did not significantly improve economic performance in the sample countries. Both economists offered several reasons why ITF-based monetary policy did not significantly improve economic performance. One possibility was that ITF and non-ITF countries implemented similar interest rate policies, previously identified by several studies as using a rule, such as the Taylor rule. If this is the case, ITF implementation, fundamentally, does not change the behavior of monetary policy instruments and, therefore, it is not surprising that economic performance would also not experience a significant change. These findings, however, are not supported by the institutional arrangements or formal ITF regulations, such as inflation target announcements, inflation reports, and central bank independence, which have also been implemented to some extent in non-ITF advanced countries.

8.5.2. Inflation Targeting after the GFC of 2008/2009 During the past decade, especially after the GFC, the global economy has confronted various difficult challenges. Furthermore, the global economy has not yet fully recovered from the crisis, therefore, the challenges to ITF implementation have also became more complex, particularly in terms of confidence in credible ITF implementation on one hand and determining the priority between price stability and output growth to maintain economic recovery momentum on the other. Another pertinent issue concerns how far existing ITF implementation has overcome the structural and specific problems of the respective economies.

258    Central Bank Policy Fundamentally, the GFC also provided valuable lessons for monetary policy. First, in an open economy, monetary policy faces increasingly complex challenges, thus the instruments used must be diverse and part of an instrument mix. A policy mix of monetary policy, capital flow, and exchange rate management as well as macroprudential policy allows the central bank to overcome the various dilemmas faced.11 In terms of monetary policy, monetary policy complexity through the interest rate can be overcome by applying monetary policy quantitatively. When confronting capital flows, for instance, the exchange rate should be managed, albeit flexibly, to avoid misalignment from the currency’s fundamental value because such conditions could threaten macroeconomic stability. Simultaneously, efforts to accumulate reserve assets are also imperative as part of the strategy “to prepare the umbrella before it rains” considering short-term capital flows are particularly vulnerable to the risk of reversal. Regarding capital flow policy, by maintaining a floating exchange rate regime, various macroprudential policy options are available to overcome excessive capital flow cycles, particularly short-term and volatile. In addition, macroprudential policy that aims to maintain financial system stability could mitigate bubble risks in the economy. Second, the global crisis reinforced belief that price stability should remain the final goal of central bank monetary policy. Nevertheless, the global crisis also taught us that low inflation alone is not enough to achieve macroeconomic stability. Several crises that have occurred in recent decades have shown that macroeconomic stability primarily originates from the financial system sector. Financial markets are always inherently characterized by imperfections, which create excessive macroeconomic fluctuations. Therefore, the key to managing macroeconomic stability is not only success in terms of controlling goods (inflation) and external (balance of payments) imbalances, but also imbalances in the financial sector, such as excessive new loan growth, asset price bubbles, and procyclical risk-taking behavior in the financial sector that is vulnerable to changes in perception. Third, the need to clearly position the role of the exchange rate in the ITF. Per the ITF monetary policy framework, a floating exchange rate system represents the optimal choice for an economy. This policy direction is required because the exchange rate system will act as an economic shock absorber. Nonetheless, exchange rates in an environment of globally integrated financial markets have been shown to move independently. In this case, the most dominant exchange rate dynamics will influence changes in investor risk perception on global financial markets, which would intensify short-term foreign capital flows more than fundamental factors. Learning from experience, in general, all parties have agreed that the main goal of monetary policy should remain focused on achieving price stability or low inflation. The issue is that when faced with a complex challenge, the standard ITF-based monetary policy framework is ineffective. For example, according to standard ITF, the interest rate is the only monetary policy instrument that

11

A more detailed description concerning the central bank policy mix will be presented in the closing remarks of Chapter 15.

ITF Concept and Implementation at Central Banks     259 influences aggregate demand and the output gap, while anchoring inflation expectations to the target. Notwithstanding, in an open economy, hiking the interest rate is often ineffective due to the imminent deluge of foreign capital inflows that flood the economy with liquidity. Without sterilization, the additional liquidity would intensify inflationary pressures and create asset price bubbles that undermine financial system stability. Therefore, new instruments are required to avoid that dilemma faced by central banks. In that context, the central bank can effectively maintain macroeconomic stability if able to control financial sector behavior. Consequently, ITF needs to be enriched to capture financial sector dynamics by stressing the importance of risk management in the macroeconomic and macroprudential policy frameworks as well as paying due consideration to the tail risks that would be detrimental to the economy. Fundamentally, ITF can still be relied upon as a monetary policy strategy. Referring to a comprehensive evaluation of nine basic principles of monetary policy, including ITF, which were considered a general consensus prior to the crisis, Mishkin (2011) concluded that “none of the lessons from the financial crisis in any way undermines the nine basic principles of science of monetary policy.” The nine principles are as follows: (1) inflation is always and everywhere a monetary phenomenon; (2) price stability has important benefits; (3) there is no long-run trade-off between unemployment and inflation; (4) expectations play a crucial role in the determination of inflation and in the transmission of monetary policy to the macroeconomy; (5) real interest rates need to rise with higher inflation, that is, the Taylor principle; (6) monetary policy is subject to the time-inconsistency problem; (7) central bank independence helps improve the efficiency of monetary policy; (8) commitment to a strong nominal anchor is central to producing good monetary policy outcomes; and (9) financial frictions play and important role in business cycles. Referring to those nine principles, Mishkin (2011) also indirectly stressed that, theoretically, ITF, which is oriented toward achieving low inflation with greater transparency, is still relevant to the monetary policy goal of achieving price stability.

8.6. Concluding Remarks ITF proliferation throughout numerous central banks has been strengthened by its status as a new monetary policy framework that could viably be implemented in other countries, especially those confronting inflation problems to the detriment of economic advancement and the amelioration of public welfare. The strength of this new framework is not merely reflected in its goal-focused approach, coupled with avowed central bank commitment and monetary policy discipline, but also in terms of the institutional arrangements that reflect the

260    Central Bank Policy modern practices of an independent, accountable, and transparent central bank. The institutional arrangements are more in line with the spread of democracy around the world. Various empirical studies have documented the benefits of ITF implementation to improve economic performance. ITF can help lower inflation to the levels achieved in advanced countries, reduce inflation persistence, anchor inflation expectations to the target as well as lower output volatility and the sacrifice ratio, while remaining flexible enough to overcome short-run economic shocks and improve the discipline, commitment, and credibility of the monetary policy instituted by the central bank. Moreover, ITF implementation can strengthen monetary policy independence, transparency, and accountability. Several empirical studies have also provided evidence that ITF implementation is not always successful in terms of improving economic performance. Nevertheless, such studies tend to stress that the various success stories enjoyed by ITF countries were also enjoyed by non-ITF countries. In their conclusion, Ball and Sheridan (2003) also found empirical evidence that could not be interpreted as a rejection of ITF. First, ITF can provide benefits politically rather than the economically. As suggested by Bernanke et al. (1999), ITF implementation creates more open and transparent monetary policy formulation, thereby supporting a central bank’s role that is more consistent with the principles of democracy. Second, ITF implementation may improve economic performance in the future. Despite the number of economic shocks experienced in the past, many central banks still chose to apply an untested ITF to overcome the underlying shocks. In the future, policymakers may face supply-side shocks similar to what took place in the 1970s or strong political pressures for inflationary policies. At that time, we shall observe how well the central banks applying ITF are able to face the challenges compared to more effective central banks and those adopting a just-do-it policy approach. Fundamentally, although ITF can still be relied upon as a sound monetary policy strategy, the specific issues that arose after the GFC made central bankers realize the need to strengthen the monetary policy framework through refinements to ITF strategy. To that end, assessments of ITF implementation in various developing countries have provided sufficient justification for the need to apply flexible ITF as an ideal format for a small open economy.12

12

The term “flexible ITF” was first popularized by Svensson (1999) by juxtaposing strict ITF versus flexible ITF. Per strict ITF, or the so-called “inflation nutters,” the central bank only focuses on the inflation gap (deviation between inflation and its target) (King, 1997). In contrast, per flexible ITF, the central bank pays due consideration to the inflation gap, as well as the output gap and/or interest rate smoothing. Nevertheless, no central banks currently apply strict ITF (refer to Svensson, 2010; Walsh, 2008).

Chapter 9

Inflation Targeting Framework: Implementation in Indonesia 9.1. Introduction Enactment of the new Bank Indonesia (BI) Act, namely Act No. 23 of 1999, which was subsequently amended by Act No. 3 of 2004 and Act No. 6 of 2009, prompted a fundamental change underlying monetary policy formulation and implementation in Indonesia.1 The new legislation contained a strong legal foundation for modern practices at an independent central bank, primarily concerning provisions on the goal of BI, monetary policy independence as well as public accountability and transparency consistent with the implementation of good governance. Act No. 23 of 1999 stipulated a clear mandate to achieve and maintain rupiah stability as the overarching goal of BI. Accordingly, BI’s duties focused on three aspects, namely formulating and instituting monetary policy, regulating and supervising the banking industry, and regulating and operating the payment system.2 The statutes also provided a strong legal basis for central bank independence, including goal independence, instrument independence, and institutional independence. Implicitly, the regulations also mandated the implementation of monetary policy based on what was known in central banking theory and practice as the Inflation Targeting Framework (ITF) and, therefore, BI initiated various

1

Two amendments to the Bank Indonesia Act have been implemented, in 2004 and 2009. The first amendment, contained in Act No. 3 of 2004, strengthened several institutional aspects and had significant implications on the implementation of monetary policy, as described in Appendix 9.1. 2 With the enactment of Act No. 21 of 2011 concerning the Financial Services Authority (OJK), the bank regulation and oversight function was transferred from Bank Indonesia to OJK. Bank Indonesia, as the macroprudential authority, in conjunction with other authorities (OJK, Ministry of Finance and Deposit Insurance Corporation (LPS)), plays its role in terms of maintaining financial system stability. A detailed description of the role central bank macroprudential policy plays in terms of maintaining financial system stability will be presented in Chapter 14. Central Bank Policy: Theory and Practice, 261–288 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191013

262    Central Bank Policy implementation measures in Indonesia. As part of Act No. 23 of 1999 implementation, after the year 2000, BI set and announced its desired inflation target to be achieved through monetary policy. Furthermore, various endeavors were implemented to prepare the preconditions required for full ITF application in the future. Meanwhile, monetary policy per the operational target of base money under the auspices of the IMF was continued, but research was conducted to develop a new monetary policy framework consistent with ITF. Regular Board of Governors (RDG) meetings were held, as the highest decision-making forum at BI, to formulate monetary policy. For accountability and transparency, BI submitted periodic reports to the People’s Representative Council and communicated routinely with the public regarding monetary policy implementation by BI. The path to full ITF implementation in Indonesia was fraught with difficulties. Experience since 2000 has shown that inflation is constantly influenced by factors beyond the scope of monetary policy, while the monetary policy transmission mechanism to inflation and the economy has been impaired by several constraints. Some of the constraints were linked to prevailing economic and financial sector dynamics in Indonesia at that time, which were undergoing structural changes in the wake of the Asian Financial Crisis of 1997/1998 that blurred the monetary policy transmission process, making it difficult to predict. Other constraints stemmed from the monetary policy framework still under development consistent with ITF. Furthermore, support from the stakeholders, including the People’s Representative Council, government and public, remained weak for BI in terms of building credibility for ITF implementation. The various economic and political dynamics prevalent at that time help to explain why ITF implementation in Indonesia was replete with difficulties. After adequate preparations had been completed, BI officially introduced (full) ITF on July 1, 2005, using the interest rate as the operational target. Reorientation of the monetary policy target per ITF was also due in part to macroeconomic recovery policy to evade the devastating economic crisis. Nonetheless, this was not easily achieved considering the recent catastrophic economic crisis, the recovery from which was more expensive and prolonged than in other similarly affected countries in the region. Gradually, ITF implementation was accepted by the stakeholders after BI successfully built up adequate policy credibility in terms of controlling inflation in Indonesia. During the past decade or so, when the economic situation had not fully recovered from the problems that emerged due to the global financial crisis (GFC), the challenges to implementing a new monetary policy paradigm based on the ITF have become more onerous and complex. This was particularly true concerning confidence in applying a credible ITF on one hand, and prioritizing price stability or output growth to maintain economic recovery momentum, which was blighted by ubiquitous uncertainty, on the other. Such developments compelled the central bank to strengthen its monetary policy framework by honing the future ITF implementation strategy.

ITF Implementation in Indonesia    263 This chapter will review ITF implementation in Indonesia since 2000.3 The discussion will focus on the institutional arrangements and operational frameworks found in the literature (Schaechter, Stone, & Zelmer, 2000).4 In terms of the institutional arrangements, the salient aspects of the new BI Act will be reviewed, along with inflation target formulation, accountability, and transparency. In terms of the operational framework, inflation projection techniques, monetary policy transmission, and monetary operations will be explored. The subsequent section will evaluate monetary policy performance in terms of controlling inflation and the various constraints faced, particularly the salient issues that emerged during the transition phase and during the early stages of ITF implementation. Before the concluding remarks, the substance and rationale behind strengthening the ITF implementation strategy in Indonesia after the GFC will be discussed.

9.2. Institutional Framework As elaborated in the previous chapter, ITF-based monetary policy requires underlying institutional arrangements, including provisions in the central bank act, inflation target formulation as well as accountability and transparency. In the case of Indonesia, institutional arrangements are contained in the new BI Act. Furthermore, BI has also implemented incremental measures since 2000.

9.2.1. BI Act The BI Act of 1999 provides a solid legal foundation concerning central bank goal clarity, monetary policy independence and restrictions on central bank financing of the government’s fiscal deficit. The overarching purpose of BI is to achieve and maintain rupiah stability, encompassing price (inflation) stability and rupiah exchange rate stability. This represents a grand departure from the previous BI Act, which mandated that BI was tasked not only with maintaining rupiah stability but also with stimulating production activity and expanding employment opportunities. With the floating exchange rate system embraced by Indonesia, this goal theoretically meant that price stability was the overriding objective, while exchange

3

To review ITF implementation in Indonesia during the transition phase, refer to Warjiyo (2002), Alamsyah et al. (2000), and Sitorus et al. (2000). Moreover, to review ITF performance after the official introduction in July 2005, refer to Juhro et al. (2009) and Juhro (2015). 4 A review of the institutional implications of ITF implementation will be presented in Part IV of this book, which explores the various institutional aspects of monetary policy in detail, including credibility and time consistency, independence, and accountability, as well as policy transparency and communication strategies.

264    Central Bank Policy rate stability was an intermediate target or indicator in monetary policy formulation.5 Such conditions were strengthened by laws that directed monetary policy toward achieving the inflation target. In this regard, the inflation target, pursuant to Act No. 23 of 1999, set by BI (goal independence), was amended by Act No. 3 of 2004, according to which the inflation target was set by the government after consultation with BI (goal dependence). To achieve the inflation target, BI was given independence to formulate and implement monetary policy. In other words, the prerequisite of instrument independence demanded by ITF implementation was safeguarded by law. The monetary instruments include open market operations, the discount rate, reserve requirement and bank lending regulations. BI can also intervene on the FX market to stabilize the rupiah exchange rate as part of open market operations as well as provide short-term funding facilities as lender of last resort. In addition, considering the rapid development of sharia banking in Indonesia, monetary policy implementation is also in compliance with sharia principles. The new BI Act also clearly stipulates that it is illegal for the central bank to finance the government deficit. Pursuant to Act No. 23 of 1999, BI was prohibited from extending loans to the government. The regulation, however, permits the purchase of government debt securities on the secondary market for open market operations. Thereafter, in accordance with the amendment, Act No. 3 of 2004, the purchase of government debt securities on the primary market was permitted to bolster stock for monetary instruments but limited to instruments of up to one year. The purchase of government securities on the primary market by BI to provide a financial safety net and, therefore, maintain financial system stability is only possible upon approval from the People’s Representative. The regulations provide adequate restrictions to ensure no fiscal dominance, as required by sound monetary policy principles, especially ITF.

9.2.2. Inflation Target Formulation Act No. 23 of 1999, as amended by Act No. 3 of 2004 and Act No. 6 of 2009, provides discretion to the Government and BI in terms of formulating the inflation target as the final goal of monetary policy, specifically in terms of the price indexes used, size of the target and horizon. Prevailing laws merely state that the inflation target must be determined: (1) based on a calendar year; (2) with consideration of price developments directly influenced by monetary policy; and (3) with consideration of the macroeconomic developments and outlook. Meanwhile, the announcement of the inflation target was changed from BI itself in pursuance of Act No. 23 of 1999 to the Government after consultation with BI in accordance with Act No. 3 of 2004. 5

Nevertheless, a lack of understanding concerning this issue often places pressure on monetary policy implementation to achieve exchange rate stability as the final target. This was linked to rupiah exchange rate performance, which depreciated and fluctuated over time. With the relatively stable and appreciating rupiah of the past few years, however, such pressures have begun to ease.

ITF Implementation in Indonesia    265 As a country transitioning to full ITF implementation, inflation target formulation in terms of monetary policy in Indonesia has experienced a number of changes since the year 2000. At first, the inflation target was set by BI with a 2% range for CPI beyond the influence of government policy in terms of prices and incomes. For example, the inflation target was set at 3%–5% in 2000 and 4%–6% in 2001. In addition, BI also announced the projected impact of government policy in terms of prices and incomes on inflation, namely at 2% in 2000 and 2%–2.5% in 2001. Therefore, by adding the two, BI’s projection of headline inflation was 5%–7% in 2000 and 6%–8.5% in 2001. Since 2002, however, the inflation target was set for all CPI components to facilitate greater public understanding. Furthermore, BI also began to announce a medium-term target, namely 6%–7% for achievement in 2006. The inflation target, in the form of a value and horizon, was considered optimal, where the target could be oriented toward a low rate, the attainment of which would not impair the economic recovery process in Indonesia.6 In that context, taking into consideration the relatively high inflation in Indonesia in 2001, the central bank strived to meet the inflation target through disinflation by setting a lower target each year toward the medium-term target. Therefore, the indicative inflation target was set at 9% for 2002 and 2003, 8% for 2004, 7% for 2005, and 6% for 2006. In addition, bearing in mind the high level of uncertainty in terms of achieving the inflation target and the effectiveness of monetary policy to control inflation in the near term, the annual inflation target was accompanied by a reliable level of deviation, set at ±1%. Currently, the government’s inflation target,7 determined after coordination with BI, is set at 4% ± 1% for 2016 and 2017, and 3.5% ± 1% for 2018. Using CPI as the inflation target ensures that core inflation can play more of a role as an indicator underlying monetary policy formulation, considering core inflation is a more accurate measure of the inflationary pressures that can be influenced by monetary policy. Therefore, BI has refined its methods to calculate core inflation using various techniques, including exclusion as a complement to core inflation with the trimmed method statistical technique. In this case, core inflation is calculated by omitting administered prices (AP) and volatile supply shocks. Using this technique, CPI inflation projections can be made by combining core inflation projections, which require a monetary policy response, with the projections of AP and volatile supply shocks. In addition to facilitating public transparency by explaining the various sources of inflationary pressures, this approach also helps to avoid an excessive monetary policy response, particularly in terms of AP and volatile supply shocks.

9.2.3. Accountability and Transparency As explained in Chapter 8, accountability and transparency are integral elements of ITF implementation. One advantage of this monetary policy framework is 6

Determining the BI Inflation Target, BI Annual Report, 2001. Minister of Finance Regulation (PMK) No. 93. PMK.011/2014 concerning the ­Inflation Target. 7

266    Central Bank Policy central bank commitment to use all instruments at its disposal to achieve the inflation target. That is why the central bank must enjoy instrument independence and be held accountable to the inflation target. Moreover, the central bank must remain transparent and routinely communicate with the public to convey the monetary policy measures instituted to achieve the inflation target. By building credibility in terms of achieving the inflation target, the central bank can fundamentally anchor public inflation expectations consistent with the inflation target and, therefore, ultimately achieve the inflation target. To that end, BI has implemented several measures to enhance monetary policy accountability and transparency. At the beginning of each year, after the discussions and decisions of the Board of Governors meeting, BI publicly explains its appraisal of monetary policy performance in the previous year and announces the inflation target and direction of monetary policy for the upcoming year through press conferences, supplements published in the mass media, as well as through discussions with academics, bankers and the business community. Furthermore, a formal report on the assessment and direction of monetary policy is also submitted to the People’s Representative Council as a performance appraisal of BI and the Board of Governors, as well as to the Government as a valuable form of information. BI also publishes an annual report that contains an evaluation of the economy and the monetary policy direction pursued. Each quarter, BI also communicates its assessment of inflation and economic performance, along with the monetary policy direction for the upcoming period, through press releases, supplements published in the mass media and discussions with the market if required. Pursuant to prevailing laws, BI also submits a report to the People’s Representative Council containing an evaluation of economic and inflation performance, monetary policy, macroprudential policy, and payment system policy. The report is also submitted to the government as a form of information and published as the Monetary Policy Statement to increase transparency. In terms of a monthly review, the decisions of the Board of Governors meeting are also communicated to the public through the Monetary Policy Review. In addition to publishing and submitting periodic reports to the People’s Representative Council, BI also constantly seeks to enhance monetary policy transparency through seminars, workshops, and other communication media for academics, the mass media and public to explain the latest economic developments and monetary policy instituted by BI. In addition, BI also collaborates with numerous universities to include the various policies taken by BI, including monetary, macroprudential, and payment system policies, along with the relevant theories and rationale into the curriculum to enrich student understanding of applying the theories learned in practice at the central bank.

9.3. Operational Framework As explained in Chapter 8, ITF-based monetary policy credibility is determined by the success of the central bank in terms of attaining the inflation target and public trust in the central bank’s commitment to its monetary policy. This clearly requires the support of a solid monetary policy operational framework. First is

ITF Implementation in Indonesia    267 the ability to project future inflation, thus current monetary policy can anchor inflation projections consistent with the inflation target. This is determined by the availability of various reliable techniques and models to produce inflation and macroeconomic projections. Second is a deep understanding of the monetary policy transmission mechanisms, which is important for the central bank to determine the transmission channel, the strength, and duration for monetary policy actions to be transmitted and influence the future direction of the economy and inflation within the target corridor. Finally, the central bank must strengthen monetary operations using the various instruments available to signal the markets, thus ensuring the operational target of monetary policy is consistent with the inflation target.

9.3.1. Inflation Projection Techniques BI utilizes survey outcomes as well as formal economic models to project inflation and other monetary economic variables. In this context, due to the limited amount of surveys conducted by other institutions, BI performs various surveys to detect and capture public perceptions and expectations, including the banking industry, corporate sector, and households, regarding the direction of inflation and other key indicators. BI’s surveys include the Business Survey (SKDU), Consumer Expectation Survey (SK), Retail Sales Survey, Property Survey, and Banking Survey. Inflation expectations are primarily captured by the Business Survey and Consumer Expectation Survey, which reflect public expectations at the corporate and consumer levels. Research has also been conducted to develop formal economic models to project future inflation and other economic variables. Accordingly, the small-scale structural macroeconomic (SSM) model, medium-scale structural macroeconomic model (Short-Term Forecast Model for the Indonesian Economy [SOFIE]), and large-scale structural macroeconomic model (Macroeconomic Model of Bank Indonesia [MODBI]) have been developed. In addition, to strengthen the projection system and simultaneously analyze macroconsistent policy, BI has also adopted the Forecasting and Policy Analysis System (FPAS) developed by the IMF. In this regard, BI has also developed a dynamic “semi” general equilibrium macroeconomic model to simulate policy (ARIMBI) as well as specific ad hoc models to project certain macroeconomic variables, including inflation, exchange rates, and money supply.8 The various models are mutually complementary and adequate to deliver consistent future economic and inflation projections in the near and medium-long term. The projections generated by the various models

8

For a brief explanation of each macroeconomic model used by BI, refer to the appendix. In practice, the various models are used together to cross-check the consistency of other projections. Furthermore, the results are benchmarked against the projections of other institutions. Besides, discussions are also held with various relevant parties, including government institutions as well as economists from other institutions, to garner greater understanding of the latest real economic developments.

268    Central Bank Policy can be used to determine the inflation target as well as evaluate and project future inflation and economic dynamics, internally at BI to assist in formulating monetary policy and as inputs for the government when preparing the state budget assumptions and medium-term macroeconomic scenarios in terms of the Annual Development Plan (Repeta) and National Development Program (Propennas). Several reviews have also been conducted to prepare key indicators as required for monetary policymaking, including Leading Indicators for Inflation (LII) and Leading Economic Indicators (LEI). LII consist of real money supply, real credit, interest rates, stock prices and other variables, which provide critical information on the future direction of inflation to confirm the projections of the formal models. Meanwhile, the information contained in the LIE indicates the future direction of economic activity, particularly on the demand side. Considering that demand-side pressures will affect inflation, LIE also function as inputs to project inflation.

9.3.2. Monetary Policy Transmission Monetary policy implementation in Indonesia during the period before the Asian Financial Crisis of 1997/1998 and during the IMF’s program was fundamentally based on the money channel. Therefore, BI would annually project total money supply required for the economy in pursuance of the inflation target and assumptions of economic growth, interest rates, and the exchange rate in the demand for money function. Based on the projected total money supply as the intermediate target, BI would subsequently set the annual, weekly, quarterly, and monthly monetary base targets paying due consideration to seasonal public demand for money trends. The monetary base target was then used as the operational target of monetary operations at BI.9 As mentioned in Chapter 5, the monetary channel as the basis underlying monetary policymaking has been deemed less effective. Therefore, BI has performed various studies to review and comprehend the monetary policy transmission mechanisms in Indonesia. The latest comprehensive study conducted by BI showed various interesting changes, especially in the wake of the Asian Financial Crisis of 1997/1998. First, the credit channel was shown to significantly influence monetary policy transmission because of the bank disintermediation problem. Bank lending to the real sector is primarily affected by internal bank conditions on the supply side as well as corporate cash flow and leverage on the demand side. Differing from the money channel, imbalances on the credit market have impacted monetary policy transmission from the banking industry to the real sector. The phenomenon of less effective monetary policy transmission through the credit channel has been reinforced by empirical evidence on transmission through

9

Under the IMF Program, the quarterly and annual monetary base targets were used as performance criteria for monetary policy implementation in Indonesia. According to such a monetary policy framework, Indonesia could be said to be applying monetary targeting.

ITF Implementation in Indonesia    269 the interest rate channel. On one hand, changes in lending rates have had a negative and significant impact on investment (due to the cost of capital) and consumption (due to the substitution and income effects). On the other hand, however, the transmission of short-term interest rates, which can be influenced by monetary instruments, to term deposit rates and lending rates contains a lag and is asymmetric. The empirical evidence for the credit channel and interest rate channel demonstrates the importance of monetary policy, accompanied by efforts to overcome bank disintermediation and, therefore, enhance transmission effectiveness in terms of the economy and inflation. It is also important to consider the empirical evidence for the exchange rate channel when formulating monetary policy. On one hand, exchange rate developments have a strong influence on inflation due to the direct impact of changing import prices of goods and services (direct pass-through effects) and the indirect impact through the output gap (indirect pass-through effects). On the other hand, however, the ability of monetary instruments to affect exchange rates is constrained by the impact of the risk premium on exchange rate developments. The empirical evidence demonstrates the importance of monetary policy to stabilize exchange rates through FX intervention, not to achieve a given exchange rate target but to minimize the impact on inflation. The inflation expectations channel is also important to consider. The empirical evidence has shown that inflation expectations are influenced strongly by inflation realization in previous periods (inertia) and exchange rates, in addition to the inflation target. This not only corroborates the importance of exchange rate stabilization measures in the exchange rate channel, but also to enhance monetary policy credibility. Such efforts are required not only to garner commitment to achieving the inflation target, but also to improve the explanations given to the public concerning the monetary policy taken by BI to restore public confidence in the central bank’s commitment to the inflation target. Increasingly dynamic economic developments ultimately manifest in the domestic economy as more complex challenges that could potentially disrupt the monetary policy transmission mechanisms, particularly in terms of the asymmetric impact of monetary policy. Observations have shown that the asymmetric influence of monetary policy is linked to procyclical financial sector behavior as well as risk perception in terms of affecting the transmission mechanism.10 Under normal conditions, when the economic cycle is expansionary, monetary policy sensitivity to aggregate macroeconomic variables is consistent with the general concept. A reduction to the policy rate would be followed by lower interest rates and faster credit growth, which is in line with the hypothesis of the risk-taking channel, where financial institutions tend to lower their risk perception while simultaneously increasing tolerance to lending risk standards.

10

Financial sector procyclicality is financial system behavior that drives stronger economic growth during an economic upswing and exacerbates the downturn during a contractionary phase. With procyclical behavior, the financial system exacerbates macroeconomic instability by creating output fluctuations.

270    Central Bank Policy Nevertheless, conditions changed during the GFC, when the sensitivity of the (lower) lending rate response to BI Rate (reductions) faded. Similar conditions occurred in the response of (increasing) bank lending to (lower) lending rates. Observations also showed that financial institutions became less sensitive to the transmission of lower interest rates. The banking industry tended to delay reductions to interest rates and ration credit despite significant easing of the policy rate (Juhro et al., 2009).

9.3.3. Monetary Operations Monetary operations, as applied by BI from when Indonesia was still bound by the IMF Program until 2003, used base money as the operational target. Fundamentally, this strategy was initiated by setting the base money target at the beginning of the year in accordance with the projections of total money supply pursuant to demand for liquidity in the economy, which was calculated by projecting the demand for money function in line with the inflation target and other macroeconomic assumptions, particularly economic growth, interest rates, and exchange rates. The annual base money target was translated into quarterly, monthly, and weekly base money targets per the seasonal public demand for money. The weekly base money target was subsequently applied as the operational target of monetary operations at BI, primarily through open market operations, including weekly SBI auctions and other BI facilities to absorb market liquidity. Supporting exchange rate stabilization efforts, BI also conducted FX market intervention as necessary. As mentioned previously, the monetary policy operational framework using base money as the operational target was eventually considered ineffective due primarily to the fading correlation between money supply and real sector activities and inflation, as well as the increasing complexity of controlling base money due to unstable and unpredictable demand for money. In July 2005, BI began to apply interest rates in its monetary operations. In addition to the dwindling effectiveness of base money in terms of monetary management to achieve the inflation target, the interest rate was required as the operational target of monetary policy in line with BI’s efforts to officially implement an ITF-based monetary policy framework in full. Various efforts have been undertaken to ensure more effective implementation of the interest rate as the operational target of monetary policy. To control inflation with the interest rate as the operational target, there must be a stable and predictable relationship between the current interest rate and future direction of inflation. In general, the central bank sets the policy rate to anchor the performance of the shortest tenor interest rate available on the market, and then depends on the financial sector to transmit the rate to the whole spectrum of interest rates on the market and, subsequently, to various real economic activities and, ultimately, to inflation. The operational target, therefore, must be set on the shortest tenor of the entire interest rate spectrum available. In the case of Indonesia, the overnight interbank rate or 1-month SBI rate was used. During preliminary ITF implementation in 2005, using the BI Rate as the policy rate,

ITF Implementation in Indonesia    271 BI continued to apply the 1-month SBI rate as the operational target. Since the middle of 2008, however, BI has used the overnight (O/N) interbank rate as the operational target of monetary policy. There are at least two aspects for BI to consider when ensuring the interest rate is feasible as the operational target or monetary policy. The first is linked to the importance of forming the term structure of interest rates on the markets and the second is connected to efforts to overcome bank disintermediation. Regarding the term structure of interest rates, the secondary market for government bonds has begun to develop with rapidity after several government and BI initiatives to deepen the domestic financial markets. This market for government debt securities importantly underlies formation of the yield curve as a signal of BI’s monetary policy. In addition, the availability of such securities also supports BI’s ongoing efforts to diversify from SBI to government debt securities as monetary instruments. Another critical BI effort is the use of models or other analytical tools that can simulate the direction of changes in the interest rate in response to deviations in future inflation projections from the target. Intensive studies have been performed in relation to this, primarily to formulate a forward-looking monetary policy rule. Further refinements of this model have been made. Relevant studies have also been conducted to develop a policy instrument mix and market signaling because policy practices that depended on SBI for open market operations were considered too expensive in terms of the monetary control costs incurred by BI. In addition to optimizing the use of government debt securities (SUN) as monetary instruments (supported by the nascent repo market), the utilization of other policy instruments to manage liquidity and FX flows was also a priority. Furthermore, the central bank is also required to build clearer and more regular communication channels regarding the direction of monetary policy as a form of market signaling, which is important not only to improve transparency but also to build BI monetary policy credibility.

9.4. Evolution Thus Far 9.4.1. Issues during the ITF Transition Period As mentioned toward the beginning of this chapter, although Act No. 23 of 1999 provided a solid legal foundation for BI to introduce ITF, economic and financial conditions in Indonesia at that time, which had experienced seismic structural changes after the Asian Financial Crisis of 1997/1998, complicated full ITF implementation. Sources of inflationary pressures not only originated from the monetary sector but also from adjustments to AP as well as disruptions to the supply and distribution of goods and services in the economy, which are categorized as volatile foods (VF). The monetary policy transmission mechanism has also met numerous constraints, thereby undermining monetary policy effectiveness in terms of its influence on inflation and the real sector. The problem of bank disintermediation amidst conditions of growing excess liquidity has skewed the effect of changes in interest rate monetary instruments through the interest rate

272    Central Bank Policy and credit channels. Such conditions have been complicated further by a vulnerable exchange rate that occasionally threatens an adverse impact on inflation. Several constraints and requirements must be overcome prior to full ITF implementation. Several opinions have been put forward concerning this topic, including Warjiyo (2002), Alamsyah et al. (2001), Sitorus (2000), Boediono (2000), and Felman (2000). In other words, a number of teething problems must be considered and overcome during the transition period for the successful implementation of full-fledged ITF in Indonesia. First, the ongoing political transition process in Indonesia has undermined the effectiveness of various public policies, including monetary policy. Therefore, credible ITF implementation can only be achieved if conducted in conjunction with improvements in public perception of the government’s general policy climate. Specifically, support from the government, People’s Representative Council and the public to commitment for controlling inflation as the goal announced per the monetary policy of the central bank is also a key factor. The amendment to the BI Act, which stipulates that the inflation target is now determined by the Government after consultation with BI, is expected to enhance public selection, commitment, and coordination to controlling inflation by BI. Second, there remain concerns over government financial sustainability, which affects monetary policy formulation at BI. As a consequence of the crisis, the Indonesian Government faced a deepening debt crisis, with domestic debt from issuances of bank recapitalization bonds amounting to around half of GDP. Around two-thirds of the recapitalization bonds were issued with variable interest rates adjusted to the three-month SBI rate. Consequently, if inflationary pressures accumulated, thus necessitating a hike to the SBI rate in order to achieve the inflation target, it would have a direct impact on the government servicing its domestic debt in the state budget. Although such conditions could happen in other countries, this issue cannot be neglected by BI when formulating monetary policy. Third, fragilities remain in the Indonesian banking system, which is still in a state of recovery. Although credit growth has accelerated, the bank intermediation function remains a pertinent issue. The credit crunch is a proven phenomenon in Indonesia and the banking industry’s interest rate response to the SBI rate is slow with a protracted lag. Considering the dominance of the banking system in Indonesia’s financial system, monetary policy effectiveness has been undermined by constraints in the transmission mechanism through the banking sector, which has edged up the cost of monetary controls, meaning that a larger SBI interest rate response is required to prompt any movement in bank interest rates. Furthermore, this issue has been compounded by corporate sector restructuring and, therefore, bank liquidity tends only to circulate in the financial sector. Such conditions not only create decoupling between the financial and real sectors, but also expose the financial sector and complicate monetary policy implementation. Fourth, housekeeping was still required by BI to hone its monetary policy framework, particularly from an operational perspective, to support ITF implementation. Details of this were explored in previous sections. The salient points, however, include the ability to accurately project inflation and other macroeconomic variables, understanding of the monetary policy transmission mechanism,

ITF Implementation in Indonesia    273 and determining the operational target required to achieve the inflation target. Various economic models are still under development, while economic and financial structural changes are also underway, which has complicated monetary policy implementation. And finally, Indonesia’s inclusion in the IMF Program also restricted space to implement the ITF. As mentioned previously, the monetary policy framework put in place by the IMF was based on the money channel, using base money as the operational target. Although using base money as the operational target was tried based on demand for money in the economy in line with the inflation target and other macroeconomic projections, the behavior of base money itself has become more difficult to control because of changes in public demand for money and seasonal factors, such as school holidays, Eid-ul-Fitr, Christmas, and New Year. Furthermore, it has become more complicated to control and link the base money target with the monetary policy framework to achieve the inflation target. At the BOG meeting on March 4, 2004, the BI Board of Governors finally decided to gradually implement a monetary policy framework based on ITF in full with the interest rate as the operational target. This was a positive decision, not only to remove any existing doubts concerning BI’s stance, but also to enhance monetary policy commitment and credibility moving forward. In addition, the amendment to the BI Act through Act No. 3 of 2004, which precipitated a number of opportunities and challenges, as well as the improving economic and financial conditions, were expected to support the full implementation of ITF. There are several important observations that support successful ITF implementation as follows. First, pursuant to the amendment to the BI Act, the inflation target is now set by the Government after consultation with BI. On one hand, the amendment garnered government and public ownership and support toward the inflation target and provided a clear mandate for BI to direct monetary policy toward the inflation target. Such conditions are reinforced if BI and the Government can demonstrate tight coordination in terms of setting and announcing the inflation target, for instance through a Memorandum of Understanding (MoU) or monetary and fiscal policy implementation. On the other hand, however, the amendment also stoked demand for greater accountability and transparency at BI in terms of achieving the inflation target and implementing monetary policy. This is a logical consequence of a clear relationship between the principal (Government) and agent (BI) in terms of providing a mandate and independence to achieve the inflation target as the final goal of monetary policy in pursuance of the amendment to the BI Act. Second, despite evidencing sound developments and relatively well maintained macroeconomic stability, the economic and financial structural changes are still far from complete. Consequently, funds tend to circulate in the financial sector, with little trickling down to the real sector due to decoupling, thus leaving monetary stability vulnerable, which requires monitoring and maintaining. It is also important for BI to observe various microaspects of the financial and real sectors to explore inflation processes and monetary policy transmission mechanisms. This is especially important in countries that have experienced a crisis, such as Indonesia, where various microproblems have emerged. For example, a study concerning

274    Central Bank Policy the price setting mechanisms in several major sectors is required or the behavior of important CPI components. Of no less importance would be a study into bank investment portfolio behavior or rupiah and FX money market conditions to explore interest rate and exchange rate behavior. In general, a variety of microside studies on economic and financial conditions would enrich and complement other studies on formal macroeconomic modeling approaches and, therefore, help monetary policy formulation and implementation effectiveness at BI. Third, monetary policy must be strengthened to support the success of applying the interest rate as the operational target. This is not merely limited to the ability to formulate a rule mechanism to determine the interest rate target, but also how the instruments and monetary operations could support attainment of the interest rate operational target. This issue demands more attention in terms of allowing BI to purchase tradeable government securities (SBN) on the primary market to bolster its stock of monetary instruments and to provide the Government with emergency financing facilities. Despite the existing MoU between BI and the Government concerning this matter, which would be strengthened through additional legislation, the provision of emergency financing facilities would prompt monetary policy expansion and undermine the prerequisite of no fiscal dominance to support successful ITF implementation. Last but not least is increased monetary policy transparency at BI through a more directed and continuous communication strategy. This encompasses communications with the academic community, markets, and public regarding the inflation targeting monetary policy framework to ensure complete understanding as the basis for forming inflation expectations or taking business decisions. A sound communication strategy would ultimately simplify and enhance the effectiveness of monetary policy implementation at BI.

9.4.2. An Evaluation of ITF Implementation since July 1, 2005 During the decade since ITF implementation on July 1, 2005, the Indonesian economy has faced various issues of no less import, linked primarily to the perfect storm of financial and economic crises that reappeared to blight the global and domestic economies just after Indonesia entered its fifth year of ITF implementation. With the current economic situation, which is yet to fully recover from the problems triggered by the GFC, the institutional challenges to implementation of an ITF-based monetary policy paradigm are becoming more complex, especially in terms of confidence in a credible ITF on one hand and determining the priority between price stability and output growth to maintain recovery momentum in an economy overshadowed by uncertainty on the other. Consequently, the reviews of ITF application in Indonesia for nearly the first five years of implementation have already recorded several successes, including more disciplined monetary policy management, accompanied by enhanced quality in line with best practices, theoretical rationale and empirical conditions in Indonesia (Juhro et al., 2009). In this case, compared to conditions prior to ITF implementation, the successes include setting and announcing the inflation target, the institutional and operational frameworks, policy coordination as well

ITF Implementation in Indonesia    275 as policy analysis and research quality. In general, the positive assessment can be linked to the various aspects of business as usual. Specifically, however, ITF implementation has recorded a number of successes in the form of substantive improvements to economic fundamentals, which differentiates the benefits of ITF from other policy frameworks as follows: (1) maturation of institutional existence; (2) policy signal clarity; and (3) increased policy credibility. Regarding the maturation of institutional existence, ITF application has successfully improved BI institutionally in terms of implementing symmetrical and structured monetary policy based on the principles of good governance when formulating public policy. Such developments are evidenced by transparent and independent policy-making processes and procedures, as well as public accountability to the inflation target. Consequently, BI has changed from an inward-oriented organization to more of an outward-oriented organization. In terms of policy signal clarity, through a gradual learning process, supported by more intensive public communication, ITF has successfully strengthened monetary policy transmission through expectations. The public now understands the processes underlying monetary policy-making decisions and can better capture monetary policy signals, which strengthens and accelerates monetary policy transmission. This represents a clear departure from conditions prior to ITF implementation, when BI’s policy signals, using base money, were not accurately detected by the markets, which, under certain circumstances, would alter or even exacerbate inflation expectations. Congruent with the two important successes mentioned above, monetary policy credibility can slowly but surely be improved. Several indicators support that conclusion. First, observations using surveys and empirical testing showed a previous or ongoing shift in the formation of public inflation expectations from backward looking to forward looking, which has had a positive impact on reducing the degree of inflation persistence. Second, consistent with increasing monetary policy credibility, the regular announcement of BI’s policy stance through the policy rate (BI Rate) became a crucial reference economic indicator for the markets and wider business community. In addition, regardless of the different perceptions regarding forecasting inflation and other macroindicators by BI and the Government, the public generally understood why BI and the Government used different calculation methods. Nevertheless, monetary policy credibility requires constant improvement because attainment of the inflation target, as the overriding objective of the central bank, is not as easy to achieve as is generally thought. Several supply-side structural shocks have occurred in recent years, which pushed inflation beyond the target corridor in 2005, 2008, 2010, 2013, and 2014. Inflation in 2005, 2008, 2013, and 2014 soared beyond the upper bound of the target in line with the government’s energy reform policy to hike fuel prices.11 Inflation accelerated in 2010 11

Fuel prices were hiked twice in 2005, the first time by an average of 30% in March and then again by an average of 96% in October. Fuel prices were hiked again in May 2008 by around 33% and then again by 48% in July 2013. In 2014, the government raised fuel prices by 34% in November.

276    Central Bank Policy slightly beyond the target corridor after international commodity prices skyrocketed and inclement weather affected agricultural produce (Table 9.1). By observing potential inflationary pressures in future periods, if no structural or integrated policy measures were taken, inflation would be difficult to steer within the long-term target of 3% in the immediate future, as occurred in advanced countries as well as neighboring countries in the ASEAN region. Consequently, many parties welcomed the government’s energy reforms, primarily by reducing/removing fuel subsidies. Such policy consistency eased inflationary pressures in future periods in terms of AP because a change (increase) in the global oil price would no longer be such a massive strain on the government’s budget,

Table 9.1:  Actual and Target Inflation. Year

Inflation Target

2005

6% ± 1%

17.1

9.7

9.17

2006

8% ± 1%

6.6

6.03

11.83

2007

6% ± 1%

6.6

6.29

8.56

2008

5% ± 1%

11.06

8.29

8.67

2009

4.5% ± 1%

2.72

4.09

6.5

2010

4.5% ± 1%

6.39

4.29

6.5

2011

4.5% ± 1%

3.8

4.1

6.0

2012

4.5% ± 1%

4.3

4.4

5.8

2013

4.5% ± 1%

8.4

5.0

7.5

Fuel price hike, inclement weather and rupiah depreciation

2014

4.5% ± 1%

8.4

7.5

Fuel price hike, inclement weather

2015

4% ± 1%

3.4

7.5

Source: Bank Indonesia.

Actual Inflation

Core Inflation

SBI Rate (BI Rate)

Main Determinants Global shocks, fuel price hikes in March and October

Fuel price hike (May) Rising international commodity prices, inclement weather

ITF Implementation in Indonesia    277 thereby obviating further energy/oil subsidy cuts, which would subsequently drive up domestic fuel prices. By naturally absorbing global oil price fluctuations through price adjustments to fuel consumed directly at home, a smoother inflation trend was expected, while minimizing any potential drastic inflation spikes. In the medium-long term, if the funds saved by reducing subsidies could effectively be allocated to infrastructure development and to support energy security, the impact on inflation would be propitious. Difficulties achieving the inflation target also stemmed from structural issues on the supply side, which caused persistent and fluctuating inflation in Indonesia. Persistently high and fluctuating inflation was due to structural rigidity and the expanding role played by international commodity prices (during the economic boom) in the inflation structure of Indonesia. A body of research exists to corroborate such findings.12 Relatively high inflation persistence in Indonesia was inextricably linked to structural problems, particularly on the supply side. Domestically, the issues were primarily related to microstructural constraints, such as inefficiency, infrastructure limitations, and the low-skilled labor market, which undermined supply-side responsiveness to policy stimuli on the supply side, the imperfect market structure and inefficient trade channel. Externally, however, international commodity price fluctuations influenced domestic inflationary pressures. On top of the structural issues plaguing the supply side, difficulty achieving the inflation target also stemmed from the complexity of problems in the monetary sector faced by BI. As experienced during the past five years, BI has intervened on the FX market to overcome the deluge of capital inflows and avoid excessive appreciatory pressures on rupiah exchange rates, which boosted liquidity on domestic money markets. Such conditions have precipitated excess liquidity that must be reabsorbed by BI to curb future inflationary pressures. Of course, BI’s efforts to maintain macroeconomic stability were not without consequence, raising the cost of monetary operations and, thus, affecting the financial performance of BI.

9.5. Strengthening the ITF Implementation Strategy after the Global Financial Crisis In addition to the success stories outlined above, several fundamental refinements have been made to the ITF implementation strategy, including a behavioral change in the financial sector on one hand and calls for ITF to support the domestic economic recovery process after the GFC on the other. The issues culminated in an increasingly complex monetary policy management dimension

12

Several research papers conducted by BI concluded a high degree of core inflation persistence. Alamsyah (2008) confirmed the findings for high inflation persistence. The degree of headline inflation persistence and a disaggregation based on groups of goods and services was recorded at 0.8–0.9, despite tracking a downward trend between the periods before and after the crisis.

278    Central Bank Policy in Indonesia, thereby requiring BI to more flexibly respond to the uncertainties appearing in the economy, beyond conventional wisdom. Referring to the previous assessments, it can be concluded that ITF is still fundamentally reliable as the de facto monetary policy strategy in Indonesia. Notwithstanding, due to the issues explained above, BI must reinforce its monetary policy framework through further refinements to the ITF implementation strategy. To that end, an assessment of ITF implementation in Indonesia has also provided sufficient justification for the application of Flexible ITF as the ideal format for the Indonesian economy (Juhro et al., 2009).13 That rationale implies three fundamental consequences in terms of ITF design and the significance of flexibility, particularly in relation to the monetary policy preferences and response as follows: (1) In the near term and with maintained macroeconomic stability, there can be a potential shift in the focus of monetary policy toward greater emphasis on economic growth. This fundamentally confirms the idea that achieving price stability as the overriding goal does not imply that the central bank can neglect efforts to maintain economic growth momentum. In contrast, in the near term, the preference of monetary policy can be oriented toward specific considerations that maintain the economic recovery process, while striving toward price stability in the long term as a prerequisite for sustainable economic growth. (2) Congruent with some of the lessons garnered from the GFC, a more flexible monetary policy response is required when considering the strategic role and dynamics of the financial sector. To that end, an optimal and effective monetary and macroprudential policy instrument mix is required. Despite the need for greater consideration of the strategic role and dynamics of the financial sector, the strategic monetary policy framework in Indonesia maintains price stability as the main factor underlying the monetary policy response. (3) In terms of policy implementation, flexibility also contains institutional aspects considering that the root of the problems faced by the central bank is not always within the scope of monetary policy influence. Therefore, strong policy coordination between BI and the Government is imperative. It should be reiterated that, in the context of ITF implementation, BI already implicitly applies Flexible ITF, meaning that in the near term, BI strives to stabilize inflation and the real economy together. ITF flexibility is reflected in the

13

The term “Flexible ITF” was first popularized by Svensson (1999) by contrasting strict ITF versus flexible ITF. According to strict ITF implementation, otherwise known as inflation nutters, the central bank only focuses on the inflation gap (deviation between inflation and the target) (King, 1997). Conversely, per flexible ITF, the central bank not only focuses on the inflation gap but also the output gap and/or interest rate smoothing. Nevertheless, strict ITF is not currently applied in any country (refer toSvensson, 2010; Walsh, 2008).

ITF Implementation in Indonesia    279 orientation to achieve the overriding goal, namely inflation, but while also paying due consideration to the dynamic performance of other macroeconomic variables in the short term, such as economic growth, exchange rates, and financial sector developments. To that end, five important factors are always considered when determining the policy rate, namely: (1) the two-year inflation forecast and consistency with the inflation target; (2) the two-year economic growth forecast; (3) the exchange rate forecast and affecting factors (including capital flows), as well the impact on the inflation and economic growth forecasts; (4) bank credit and interest rates; and (5) asset valuation in the financial sector. The first two factors maintain BI Rate consistency with attainment of the inflation target, while considering the inflation-growth trade-off to maximize public welfare. The final three factors balance monetary stability and financial system stability, while also assessing the monetary policy transmission mechanism. Similarly, from a policy perspective in terms of managing internal and external balance in line with economic dynamics after the GFC, BI has, over the past five years, also implemented a policy mix, encompassing interest rate policy, exchange rate policy and macroprudential policy.14 For example, the orientation of the policy mix in 2001 was as follows. First, the BI Rate was required to signal BI’s avowed commitment to achieve the future inflation target. Second, to help ease inflationary pressures, BI embraced more appreciatory rupiah management, approaching the degree of appreciation found in neighboring countries, in line with the currency’s fundamental path. Appreciatory rupiah management was aligned with an optimal accumulation of reserve assets to bolster self-assurance against the risk of a capital reversal. Third, to control the dynamic flow of foreign capital and simultaneously mitigate capital reversal risk, BI applied macroprudential policy to manage foreign capital flows, including a foreign currency reserve requirement along with a minimum holding period for BI Certificates (SBI) and mandatory hedging. Fourth, considering the excess liquidity in the banking system, BI prepared options to manage the excess liquidity, including a rupiah reserve requirement. Fifth, BI strengthened policy coordination with the Government to control inflation by providing recommendations to stabilize VF and regarding the impact of AP on inflation, while also optimizing the role of the National Inflation Task Force (TPI) and Regional Inflation Task Forces (TPID). Based on the lengthy experiences and lessons of implementing a monetary policy strategy while exposed to a financial crisis, the in-depth literature and research on ITF per the characteristics of the Indonesian economy, as well as an assessment of ITF best practices at different central banks, the monetary policy strategy was strengthened in accordance with the domestic economy according to the format of Flexible ITF. In this context, the scope of flexibility not only refers to the policy preferences toward price and output stability, as contained in the literature, but also other strategic aspects of monetary policy. Therefore, based on the elements inherent to ITF, the fundamental components of flexible ITF were formulated per the following rationale (Juhro, 2015; Warjiyo, 2013c ):

14

A discussion on the central bank policy mix is presented specifically in Chapter 15.

280    Central Bank Policy (1) Inflation targeting as the strategy underlying monetary policy: The main substance of inflation targeting is to control inflation per the target as the overriding objective of monetary policy. The trade-off between economic growth, exchange rate stability, and financial system stability is taken into consideration when formulating monetary policy but, if a conflict emerges, the inflation target is prioritized. In addition, the institutional elements of ITF must also be strengthened, including monetary policy independence, accountability, and transparency. (2) Monetary and macroprudential policy integration to strengthen policy transmission and support macroeconomic stability: The linkages between the monetary stability and financial system stability frameworks can be strengthened through monetary policy integration with macroprudential policy. The global crisis provided an invaluable lesson concerning the financial accelerator in terms of monetary policy. Financial system stability determines the effectiveness of monetary policy transmission. Likewise, the monetary policy response influences financial system stability. Therefore, a policy instrument mix is crucial. In that regard, the policy rate response, as the main policy stance, must be supported by macroprudential policy to manage foreign capital flows and domestic liquidity. (3) The role of policy to manage exchange rates and foreign capital flows in support of macroeconomic stability: Strengthening exchange rate policy to achieve price stability. In this case, exchange rate stability is managed to remain consistent with attainment of the inflation target and macroeconomic stability. An optimal solution to the impossible trinity is sought by observing the linkages between exchange rate stabilization policy and foreign capital flow management, as well as the implications on macroeconomic stability. (4) Strengthening policy coordination between BI and the Government to control inflation as well as maintain monetary and financial system stability: Strengthening the coordination framework is imperative considering that, in addition to the demand side, sources of inflationary pressure also originate from the supply side and strategic commodities. Furthermore, more limited economic capacity and constraints to the implementation of infrastructure programs necessitate an integrated policy strategy between the policy authorities. Besides, from a broader perspective, policy coordination can be applied to manage foreign capital/portfolio flows because such flows are vulnerable to short-term shocks and issues that could trigger a reversal. (5) Strengthening the policy communication strategy as part of the policy instruments: Communicating monetary policy is no longer merely a way to increase transparency and accountability but is becoming an important monetary policy instrument. The communication policy is designed to anchor public and market expectations, reduce uncertainty, decipher the noise, and increase predictability, thereby alleviating financial market volatility, while simultaneously providing greater public understanding of the goals of monetary policy, the monetary policy framework and operational framework as well as monetary policy transmission.

ITF Implementation in Indonesia    281 Referring to the five elements detailed above, in the achievement of the overriding objective, ITF and Flexible ITF are substantively the same, namely by targeting inflation. A new dimension that has emerged since the GFC is the central bank’s role in maintaining financial system stability, which requires an integrated mandate to also achieve price stability. Furthermore, there is a need to strengthen the ITF implementation strategy itself; in this case in terms of embodying the element of flexibility, namely by providing adequate space to integrate monetary and financial system stability through a mix of monetary policy, macroprudential policy, exchange rate policy, and foreign capital flow management; as well as institutional reforms to optimize the role of policy coordination and communication.

9.6. Concluding Remarks Over the past 15 years, with the economic situation yet to fully recover from the issues that emerged due to the GFC, the monetary policy challenges in terms of the ITF strategy in Indonesia have been onerous. Various challenges emerged as problems rooted in the dynamics of the strategic environment globally and nationally. Externally, the problems stemmed from the GFC along with shortterm capital flow mobility, which had a strong impact on exchange rate developments. In addition, changing behavior in the financial system, which was linked to the growing role of risk perception and procyclicality, also exacerbated the complexity of monetary control and complicated decisionmaking. At home, the issues related to structural rigidity on the supply side, which fundamentally disrupted the policy transmission mechanism and placed additional pressures on monetary stability. Against such a dynamic backdrop, it was no longer a simple matter for BI to steer inflation toward the target and, therefore, build monetary policy credibility. Such conditions were congruous with the ongoing academic debate and empirical studies that extolled the feasibility of ITF, especially in developing countries, which were typically undergoing structural change and in the early throes of transition or institutional development. Approaching its 25th anniversary, the legitimacy and compatibility of ITF has also been an arduous challenge since the GFC, which slashed global economic growth until the present day. No one could have predicted the rapidity and depth at which the financial crisis struck. In the quarterly and even monthly data, global growth projections by international organizations (IMF and World Bank) were continuously revised downwards beyond existing predictions. For example, in the middle of 2008, the IMF still projected global economic growth at around 3.9% but ubiquitous uncertainty and risk compelled the international institution to revise down its projection again and again. Currently, global economic growth is around 3% and expected to track a decelerating trend, primarily due to the deep contractions experienced in advanced countries. Weaker economic growth was coupled with an unforeseen decline in potential output, especially when compared to conditions before 2007, as characterized by robust economic growth along with relatively stable inflation.

282    Central Bank Policy In general, from the important lessons learned from those salient experiences, all parties concurred that the overriding goal of monetary policy must remain focused on achieving price stability or low inflation. Nevertheless, the substance of stability could be interpreted into a broader rationale, namely by considering financial system stability indicators, demonstrating that (flexible) ITF implementation was necessary but not sufficient. To that end, the successes of ITF implementation must be supported by a macroprudential policy regulatory framework. This would simultaneously overcome one of the prerequisites to successful ITF implementation, namely a sound and efficient financial system. The fundamental substance must receive attention, namely that when formulating a post-crisis monetary policy strategy, the central bank must strengthen financial system stability to ensure a stable economy and financial system in terms of the macroeconomy and financial sector. The shift or emphasis in the mandate of the central bank to maintain financial system stability created the need for a new paradigm and a central bank policy mix.15 Therefore, the format of BI’s mandate to concurrently maintain monetary and financial system stability needs to be formulated appropriately, especially in terms of formulating an optimal monetary and macroprudential policy mix. Chapter 15 discusses the new central bank policy mix paradigm and its implementation in Indonesia since 2010.

15

Departing from the existing format and governance of monetary policy that was widely understood in terms of ITF implementation, the new format began development and implementation at central banks in the wake of the GFC in 2008/2009.

ITF Implementation in Indonesia    283

Appendix 9.1: Amendments to the Bank Indonesia Act and their Implications Act No. 3 of 2004, as an amendment to the BI Act (No. 23) of 1999, contained several fundamental changes to the regulatory, institutional, and policy frameworks of BI. The amendment stipulated a total of four major changes in terms of monetary policy implementation, namely determining the inflation target, the purchase of government debt securities (SUN) on the primary market for monetary operations, the purchase of SUN on the primary market to provide government funds for emergency financing, as well as accountability and transparency.

Setting the Inflation Target by the Government A major change was implemented with the introduction of Act No. 3 of 2004, namely that the inflation target formally set by BI (goal independence) was now determined by the government after consultation with BI (goal dependence). On one hand, the loss of independence to set the inflation target reduced BI’s discretion in terms of determining the inflation target and formulating monetary policy. Per Act No. 23 of 1999, BI had full discretion to set the short- and long-term inflation targets, the type of inflation (headline or core) and the form of the target (point or range). The inflation target could also be adjusted to the ability of monetary policy to achieve the target. After an amendment to the BI Act, in the form of Act No. 3 of 2004, BI no longer enjoyed such discretion. On the other hand, however, Act No. 3 of 2004 also contained a number of advantages. Several arguments were proposed in the monetary economics literature that emphasized the benefits of the government setting the inflation target. The main argument was supported by the institutional context of the economic authority under a democratic system, as well as the pragmatic consideration that an inflation target set by the government would provide additional credibility for the central bank’s monetary policy. Mishkin (2000), for instance, recommended that the inflation target should be determined by an elected government because it allowed the public some oversight of government actions and, therefore, ensured accountability.16 In the context of democratic accountability, it is reasonable for the government to set the inflation target to guide the national economy because they are democratically charged with a public mandate. Debelle and Fischer (1994) further revealed some other advantages if the government were to set the inflation target, with or without consulting the central bank. In general, the government’s inflation target would be considered to 16

The fifth guiding principle, that the public must be able to exercise control over government actions and that policymakers must be accountable, so basic to democracy, strongly suggests that the goals of monetary policy should be set by the elected government. In other words, the central bank should not be goal dependent. The corollary of this view is that the institutional commitment to price stability should come from the government in the form of an explicit, legislated mandate for the central bank to pursue price stability as its overriding, long-run goal.

284    Central Bank Policy represent the preferences of the public, while the inflation target set by the central bank would be considered overly tight.17 Furthermore, an inflation target set by the government would also bring another important “advantage” to the central bank, namely greater alignment of fiscal policy with monetary policy. If the government sets the inflation target as a mandate that must be achieved by the government, it would be harder to pressure the central bank into instituting expansive short-run monetary policy because it would be inconsistent with the achievement of price stability. Besides, “a government commitment to price stability also is a commitment to making monetary policy dominant over fiscal policy, ensuring a better alignment of fiscal policy with monetary policy.” What must be avoided if the inflation target is set by the government is the possibility of a bias toward populist policy. In other words, a loose inflation target would be set to accommodate short-term economic growth, thus making it difficult to achieve low and stable inflation in the long run. In the case of developing countries, such concerns could be stoked further if government expertise – or experience at least – in terms of determining the inflation target is considered weak. If the government casually sets the inflation target for the monetary authority to strive toward, the economy could encounter unnecessary fluctuations that are counterproductive in the long term. Moreover, if the government applied a short-term timeframe when setting the inflation target, the change in policy stance required to achieve the inflation target would have to be of greater magnitude, which would amplify the economic gyrations further. With the short-run viewpoint of the government, an inflation target set by the government could exceed the optimal long-run target socially. To reduce this difference, the inflation target should be set in conjunction or after consultation with the central bank as the most ideal situation. Of the 19 countries implementing ITF, 10 central banks set the inflation target in conjunction or after consultation with the government (Mishkin & Schmidt-Hebbel, 2001). In the context of Indonesia, what became a challenge with the amendment, Act No. 3 of 2004, was how to align the views of BI and the Government regarding the importance of achieving low inflation in the medium-long term. The inflation rate is a crucial element of raising national economic competitiveness. What has the government achieved so far? The government successfully reduced the fiscal deficit to close to zero in 2006 and has maintained a surplus ever since, which is an important factor in the efforts to lower inflation. A long-term inflation target also helps the formation of inflation. In this regard, BI must be able to furnish the government with a comprehensive picture to announce a long-term inflation target, which has never been done previously. Aligning the viewpoints of BI and the Government in terms of controlling inflation is a strength of the ITF. Through ITF, the government will have ownership in the program to control inflation. Congruently, government commitment to achieve the inflation target is also expected to increase. Considering the causes 17

The democratic legitimacy of the government to set the inflation target would be strengthened if the leadership of the central bank was selected by parliament and not through other political processes.

ITF Implementation in Indonesia    285 of inflation beyond the control of the BI can be resolved through government policy, achieving the inflation target will be enhanced.

SUN Purchases on the Primary Market for Open Market Operations Monetary policy effectiveness is determined by the monetary instruments available to the central bank. The more instruments available, the easy it is for the central bank to control monetary policy in order to achieve price stability. In that context, prior to the amendment of Act No. 23 of 1999, BI was only permitted to purchase securities on the secondary market for monetary operations. After the amendment, however, pursuant to Article 55 (4), BI may purchase short-term securities (up to one year) on the primary market for monetary operations. In addition, the amendment also paved the way for BI to purchase government debt securities (SUN) on the secondary market as an additional alternative monetary instrument in the future. There are a number of advantages to BI purchasing SUN on the secondary market. First, purchasing SUN will boost its stock as a monetary instrument, as a replacement for Bank Indonesia Certificates (SBI). For use as a monetary instrument, BI must be able to trade SUN on the secondary market. Second, purchasing SUN will mitigate the risk of losses incurred by the central bank because the cost is passed on to the government through the state budget (currently, the SUN coupon is larger than the discounted SBI). Third, SUN purchases will expedite domestic secondary market development and, hopefully, create a benchmark yield for the financial markets. Nonetheless, Indonesia still faces limitations considering SUN supply depends on the government’s commitment and financial posture.

SUN Purchases for Emergency Financing Facilities Efforts to improve financial system stability also play an important role in terms of supporting monetary stability. Fundamentally, monetary policy is transmitted through the banking and financial sectors, therefore, financial system stability will determine the effectiveness of monetary policy implementation by the central bank. Consequently, if a bank should encounter financial difficulties, which is expected to have a systemic impact and could therefore trigger a crisis and threaten the financial system, pursuant to Act No. 3 of 2004, Article 11 (4), BI may provide an Emergency Funding Facility (FPD), funded by the state budget, something that was not regulated previously. Therefore, the government shall issue new SUN as required for emergency financing in accordance with the financial safety net to ensure the integrity of the overall financial system.18 To that end,

18

The provision of emergency financing facilities differs from the short-term funding facility offered by Bank Indonesia as lender of last resort pursuant to Article 11 (1) of Act No. 3 of 2004. The emergency financing facility is provided to banks experiencing difficulties that threaten the integrity of the overall financial system, while the shortterm funding facility is provided to banks to overcome short-term liquidity shortfalls that would not affect financial system stability in general.

286    Central Bank Policy BI is permitted to purchase SUN on the primary market, the proceeds of which can be used by the government to help banks experiencing financial difficulties. An important aspect that demands attention is how SUN purchases on the primary market by BI could potentially create fiscal dominance, which would complicate policy management and undermine achievement of the inflation target. Furthermore, the time lag between BI mobilizing the funds and the Government issuing SUN also demands attention. BI disbursing the funds to the bank before receiving the SUN from the Government19 would raise the cost of open market operations. Therefore, a faster approval process by the People’s Representative Council would be better for BI and for monetary control.

Increasing Transparency and Accountability In terms of monetary policy management, emphasizing the importance of transparency and accountability has gained traction. Prior to the amendment, at the beginning of each fiscal year BI would be required to disclose information to the public through the mass media containing: (1) an evaluation of monetary policy implementation in the previous year and (2) the monetary policy plan and targets for the upcoming year, paying due consideration to the inflation target as well as economic and financial developments. After the amendment, however, pursuant to Act No. 3 of 2004, Article 58, BI now reports its responsibilities in terms of task implementation to the People’s Representative Council, which previously embodied the public in a broad scope. The report submitted to the People’s Representative Council (and also the Government) at the beginning of each year contains: (1) task implementation and authority in the previous year and (2) the policy plan, targets and measures for the upcoming year, paying due consideration to the inflation trend as well as economic and financial developments. Furthermore, BI also submits a report on task implementation and authority to the People’s Representative Council and Government on a quarterly basis. Both reports (annual and quarterly) are used for the annual performance appraisals of the Board of Governors and BI.20 The report on task implementation and the policy plan, setting the targets and implementation measures is not only delivered to the People’s Representative Council and Government, but also to the public (including the markets). This is important for the public to know the direction of policies to be taken as well as to discipline BI in terms of achieving the inflation target. Consequently, BI credibility is expected to improve. 19

SUN issuances require time because People’s Representative Council approval must first be sought. 20 The demand for greater accountability increased with the formation of the Supervisory Board, which has access to Bank Indonesia’s budget, therefore the Board may also wish to question policy aspects. The duties of the Supervisory Board include: (1) reviewing the annual financial statements of Bank Indonesia; (2) reviewing the operational and investment budgets of Bank Indonesia; and (3) reviewing the decisionmaking procedures for operational activity that falls beyond the remit of monetary policy and asset management at Bank Indonesia.

ITF Implementation in Indonesia    287

Appendix 9.2: Bank Indonesia Macroeconomic Models In-depth understanding of the overall macroeconomy is crucial when formulating monetary policy. The ability of macroeconomic modeling will determine the accuracy of macroeconomic projections, including inflation, economic growth and credit growth, as well as the monetary policy response required in line with the inflation target. In this regard, BI has developed structural macroeconomic models to analyze and project future economic and inflation trends. One of the large-scale structural macroeconomic models developed by BI is known as MODBI, which currently consists of five blocks/sectors (aggregate demand and supply (GDP), price, monetary, fiscal and external) and 87 equations, comprised of 57 behavioral equations and 30 identity equations. MODBI, which was first developed at the beginning of the 1990s, is often used to analyze different macroeconomic variables and their projections in the medium-long run (five-year horizon). The passing of Act No. 23 of 1999, which laid the groundwork for ITF implementation in Indonesia, created greater demand to develop larger macroeconomic models, primarily to support the implementation of ITF-based monetary policy. Larger macroeconomic models were needed for inflation and economic analysis and projections as the materials underlying monetary policymaking at the Board of Governors meeting (RDG). Consequently, BI began to develop and strengthen several macroeconomic models for short-term analysis and projections, the first of which was the SSM.21 As an inflation projection model, the basic model reflects an ITF-based monetary policy framework. The model consists of seven equations, namely the short-term Phillips curve, aggregate demand, output gap, demand for money function, traded inflation, exchange rate, and Taylor rule to determine the operational target.22 For quarterly macroeconomic analysis and projections, BI also developed a medium-scale macroeconomic structural model known as SOFIE. The SOFIE model consists of six blocks/sectors (demand/GDP, price, monetary, fiscal, banking, and external), with 279 equations comprised of 170 behavioral equations and 79 identity equations. The basic construction of the model also reflects the theoretical and practical implementation of ITF. The model also encompasses the short-term Phillips curve, aggregate demand, aggregate supply, the monetary sector, interest rate, and exchange rate behavior as well as a type of Taylor rule. The SOFIE model is more disaggregated than the SSM model because the goal is to illustrate the behavior of several macroeconomic variables consistent with the inflation projections, as well as use an error-correction technique as the econometric approach. The model is used to help researchers analyze and project 21

The preliminary version of this model is presented in Appendix 9.2, The Inflation Forecasting Model, of the paper by Alamsyah et al. (2001). 22 For exchange rate analysis and projections, a simple model was developed known as the Behavioral Model of Effective Exchange Rate (BEER). The model consists of several equations to illustrate exchange rate behavior as influenced by the terms of trade and risk premium (measured using the spread between Government Bond Yields issued on the New York market with US Treasury bills).

288    Central Bank Policy macroeconomic variables in the short term (two years) on a quarterly basis, which serves as materials for the Board of Governors meetings. The model is currently under further development to detect clearer behavioral trends concerning monetary policy transmission and determining the interest rate operational target in line with the Taylor rule. In 2010, BI strengthened SOFIE through integration with the SSM model. For simulations, a dynamic stochastic macroeconomic model was developed known as Dynamic General Equilibrium Model of Bank Indonesia (GEMBI). In a departure from the structural models, GEMBI uses a modern approach based on dynamic stochastic macroeconomic theory. By using certain stylized facts, the model can be calibrated to produce long-term projections of macroeconomic variables, including economic growth, inflation, exchange rates, and so on. The goal is more simulation-based to illustrate the medium-long term behavior of various macroeconomic variables in response to a number of alternative monetary policy scenarios. Nevertheless, using the GEMBI model on too large of a scale to accommodate the economic system in its entirety would ultimately encounter numerous constraints/challenges, particularly in terms of recalibrating the behavioral parameters (parameterization) and the linkages between certain variables without a firm theoretical or empirical footing. The GEMBI model was eventually simplified and is currently no longer in use. During the early days of ITF implementation, the four macroeconomic models mentioned above were mutually complementary in terms of analyzing and projecting inflation along with other macroeconomic variables. The SOFIE and MODBI models produced quarterly and annual projections, which were subsequently used as the initial values in the GEMBI model to simulate the dynamic long-term behavior of key macroeconomic variables in the medium-long term according to several alternative monetary policy scenarios. Meanwhile, the SSM model was more commonly used to analyze the behavior of inflation specifically. To simultaneously strengthen the projection and policy analysis system consistent with post-global crisis conditions, BI adopted the FPAS approach recommended by the IMF in 2010. Consequently, BI developed a dynamic “semi” general equilibrium macroeconomic model for policy simulations, namely the ARIMBI model (Aggregate Rational Inflation – Targeting Model for Bank Indonesia). The ARIMBI model was developed based on New Keynesian theory and consists of six blocks/sectors, namely aggregate demand, aggregate supply, price, monetary, financial, and external. Over the past three years, congruent with the increasingly complex policy challenges, the ARIMBI model has been strengthened as the main modeling system in the FPAS, using other existing models, such as SOFIE and various ad hoc models to expand the scope of policy analysis through the integration of external or balance of payments (BOP) variables (including the current account as well as capital and financial account), the financial sector (including credit and default risk) and the endogenous behavior of risk variables. Consequently, it is now possible to comprehensively simulate BI’s policy mix using monetary instruments (interest rate) and macroprudential instruments (RR and LTV).

Part IV

Institutional Aspect of Central Bank Policy

This page intentionally left blank

Chapter 10

Monetary Policy Credibility and Time Consistency 10.1. Introduction During the 1970s and 1980s, the general theme of monetary policy discussions was oriented more toward the selection of appropriate strategic and operational frameworks, as explained in Chapters 6 and 7, while the institutional aspects of policy implementation did not receive much attention. Only in the 1990s did intuitional aspects become the subject of intensive discussions, which was partially precipitated by the emergence of the long debate concerning the formation of the European Central Bank (ECB) at the beginning of the decade. One area that received attention was how to design the right legislative framework for the existence of a central bank so that all parties mandated with implementing monetary policy could act to achieve the policy objectives. In this case, the policy goal was, typically, to achieve price stability (low inflation) in the medium–long term, while paying due consideration to supply and demand-side shocks or output performance in the near term.1 An important lesson could be extracted from the empirical experience of policy implementation by nearly all central banks around the world, namely that inflation bias was primarily attributable to the political perspective, which sought to achieve monumental changes in a short period of time. Such conditions necessitated a legislative framework that provided central bankers independence from political interference, coupled with relatively long tenure.2 In this case, longer 1

The increasing shift in policy preference towards achieving price stability was based on the belief that price stability (low inflation) was a prerequisite of sustainable economic growth and, therefore, public welfare. 2 The issue of central bank independence has been around since the inception of central banks. David Ricardo voiced his support for an autonomous central bank proscribed from financing the government’s budget deficit in 1824 (Fraser, 1994).   In the dictionary, independence is defined as free from outside influences, instructions or control. When applied to central bank independence, Meyer (2000) interpreted independence as free from government influence, instruction and control, both Central Bank Policy: Theory and Practice, 291–318 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191015

292    Central Bank Policy tenures were believed to protect the central banker from political interference. Nevertheless, the status of central bank independence, in many countries, raised the question of how the institution could effectively be monitored by the public and representative members of parliament. Monitoring was required because of the risk of incompetent policymaking at the central bank, or the possibility of introducing policy that conflicted with the mandate of the central bank. Therefore, mandating central bank independence had the consequence of necessitating accountability.3 A study focusing on both institutional aspects would be pertinent considering the crucial role both aspects play in terms of improving central bank policy credibility in the eyes of the public. Policy credibility can be defined as central bank commitment to comply with the rules or policy goals articulated in a transparent manner. Simply put, therefore, policy credibility is built through consistency between what is said and what is done. In essence, for a central bank mandated with controlling inflation, credibility is a function of the difference between actual inflation and the central bank’s inflation target. Why is credibility important? Blinder (2000) stated that with credibility a central bank could reassure the public that, through the policy measures taken, the policy goals will be achieved. Therefore, higher credibility at a central bank is generally believed to enhance the implementation effectiveness of monetary policy itself because if a central bank is credible and the public have rational expectations, then delivering commitment or a statement on the planned policy to control inflation will elicit a positive public response, hence the achievement of price stability could be strived toward without sacrificing economic growth. According to Perrier and Amano (2000), a credible monetary authority could form more accurate judgments on economic capacity in terms of producing goods and services, thereby creating job opportunities without stoking concerns regarding the emergence of inflationary pressures. This is an important benefit of policy credibility, particularly when there is uncertainty concerning the output gap. Although the important role of credibility in terms of monetary policy implementation has long been recognized, modern studies only began to appear at the end of the 1970s through the seminal works of Kydland and Prescott (1977) as well as Calvo (1978). Both studies showed that the rational expectations hypothesis has fundamental implications underlying macroeconomic policy credibility in general and monetary policy credibility specifically. One implication is that

executive and legislative. Fraser (1994), on the other hand, defined central bank independence as freedom for the central bank to implement monetary policy without political considerations. 3 According to the dictionary, accountability is defined as the obligation to justify actions or decisions, being responsible.There are two opposing opinions concerning the relationship between independence and accountability. Fraser (1994) opined that accountability could help preserve central bank independence. In contrast, Meyer (2000) suggested a trade-off between independence and accountability, implying that something which increased independence would reduce accountability and vice versa.

Monetary Policy Credibility and Time Consistency     293 monetary policy inconsistency would lead to inflation exceeding its target, or an inflation bias, because the central bank is seeking stronger economic growth. Therefore, if the central bank is not pre-committed to achieving the inflation target, then the subsequent trends that transpire will be inconsistent with monetary policy implementation and a lack of credibility will be conveyed through the policy statement of the central bank. In this chapter, aspects of central bank credibility will be discussed along with several measures that could be taken to improve credibility. The following two sections will describe several conceptual dimensions and present a survey of the literature, as well as the theoretical models linked to policy credibility, specifically in terms of monetary policy inconsistency. The fourth part will review the empirical observations and several important related issues, such as measures of the credibility index and time inconsistency. An empirical assessment is presented in the fifth section regarding monetary policy credibility in Indonesia, followed by the concluding remarks. It is also important to note that improving central bank policy credibility is fundamentally linked to the development of institutional aspects, such as policy independence, accountability and transparency, as discussed in Chapters 11 and 12.

10.2. Conceptual Dimension of Policy Credibility On a pragmatic level, the issue of credibility can be observed from two perspectives, namely policy credibility and target credibility. Policy credibility relates to how consistent public expectations are of policy implementation by the central bank compared to the policy plan previously announced by the central bank. This depends on whether policy implementation by the central bank deviates from the previous statement due to outside interference or economic changes. Meanwhile, target credibility relates to how consistent public expectations are of achieving the future policy target compared to the target announced previously by the central bank. This depends on the level of policy credibility and whether the economic structure, including the public response, facilitates achievement of the previously announced target. Therefore, policy and target credibility are fundamentally determined by at least three aspects: (1) the public response reflected in the formation of expectations; (2) central bank capacity to maintain and commit to the policy strategy over time; and (3) the structure and development of the economy that facilitates sound policy transmission mechanisms. For example, if the public feel the central bank is unable to implement policy or could implement inconsistent policy, the policy will probably fail. The more often the central bank fails in its policy implementation, the lower its credibility will be in the eyes of the public. Consequently, it will become increasingly difficult for the central bank to implement policy, which is often referred to as the “credibility problem.”

10.2.1 Causes of the Credibility Problem Analytically and practically, there are several causes of the credibility problem on the policymaking side as follows:

294    Central Bank Policy (1) Time inconsistency Time inconsistency occurs when the policy authority faces a dilemma in determining the policy preferences in relation to the time dimension. For instance, the policy authority pledges to implement policy to subdue inflationary pressures in a given period. After the economic agents respond to the decision by setting nominal wages based on the expectation of low inflation, however, the policy authority could be tempted to tolerate higher inflation in the subsequent period in order to increase output and allow the economy to absorb more labor. This could occur if the policymakers are confronted with the dilemma of controlling inflation or lowering unemployment. Implementing such a policy would quickly erode the credibility of the policy authority because when setting the same policy in the subsequent period the economic agents will be less inclined to trust the authority and will anticipate conditions based on previous experience by setting higher nominal wages. (2) Internal inconsistency Internal inconsistency occurs if the policy authority faces a policy dilemma related to sectoral interests. In general, the issues relate to a lack of policy coordination, including policies in a certain policy authority’s domain as well as polices between the policy authorities, which leads to contradictory policies and, therefore, the credibility problem. An example would be deflationary policy coupled with an unrestricted increase in the budget deficit, which could occur if policymakers instituted tight money policy together with an excessive budget deficit policy. Another example is the inconsistent implementation of an economic reform program or a consistent program but with asymmetrical implementation. (3) Asymmetric information The third source of the credibility problem is asymmetric information or incomplete information concerning the policy authority. Economic agents may not be aware that the policy authority is seriously pursuing a deflationary policy, thus when the policy is announced, the economic agents who are not convinced of the policy authority’s commitment require time to learn and convince themselves that the policy authority will really implement the desired policy. Such conditions typically occur in developing countries, where the policy authority tends to be inconsistent and implement inconsistent policies that confuse the economic agents. Consequently, the economic agents are not sure about the goal of the policy. Furthermore, as the economic agents do not have access to symmetric information, they are also unable to monitor what the policy authority is actually trying to achieve. Therefore, it becomes difficult to build a solid reputation that the policy authority is seriously pursuing a particular economic policy. (4) Stochastic economic shocks Economic shocks that are difficult to predict (stochastic) could cause a policy program to fail in the achievement of its target despite sound design and consistent implementation because such shocks could create off-target public expectations concerning an economic indicator set by the policy authority. In the context of economic stabilization policy, the various shocks that appear

Monetary Policy Credibility and Time Consistency     295 could lead to an inaccurate economic stabilization program. If the economic agents subsequently assume an inaccurate economic stabilization program, their expectations of the program will be low and the program could reasonably run into problems or even fail. (5) Political uncertainty The presumption among economic agents that the policymakers are unable to implement an economic policy package due to a lack of domestic political support could also exacerbate the credibility problem, which could occur if the government is a coalition of various political parties or if the government’s legitimacy is questioned by the public. Consequently, even if the economic stabilization policy package issued by the government is sound, the lack of political support, coupled with vested policy-making interests, could trigger the credibility problem. In practice, several of the factors mentioned above combine to influence the problem of policy-making credibility. Consequently, it is nearly impossible to measure the magnitude of the credibility problem caused by each respective aspect, especially in terms of certain economic packages or programs. In general, credible programs are assumed to succeed, while programs lacking credibility tend to fail. Nevertheless, the sources of the failure are difficult to separate individually.

10.2.2. Effort to Improve Credibility Improving credibility requires serious efforts by the policy authority. The various determinants of the credibility problem listed above demand vigilance in terms of time and sector inconsistency, information availability, anticipation of the elemental ideas, as well as process assurance. Several efforts can be applied to improve policy credibility. (1) Using a rule Using a specific policy rule is thought to prevent the policy authority from “deceiving” the markets. The use of a rule is expected to reduce uncertainty concerning future policies and, therefore, stabilize expectations and macroeconomic variables. In practice, an example of a rule is through commitment to a fixed exchange rate. By implementing a fixed exchange rate system, the central bank is expected to alleviate inflationary pressures on the domestic currency due to external pressures and simultaneously ease inflation bias because the public understands the consequences of applying a rule to monetary policy. Therefore, if the public and policymakers are jointly committed to a specific rule, then its use will help overcome the credibility problem. Conversely, if commitment to a rule is not fulfilled, the relevance of the implemented policy would be undermined or even become counterproductive. Operationally, a rule could be applied using an operational target or policy response that adheres to rules for certain variables, for example based on monetary aggregates (Friedman rule and McCallum rule) or based on the interest rate (Taylor rule), as discussed in Chapter 7.

296    Central Bank Policy (2) Building reputation The academic community has explored various solutions to the time inconsistency problem. One solution was proposed by Barro and Gordon (1983), namely by building reputation. They explained that the time inconsistency problem emerges because the public knows that the central bank has an incentive to accommodate measures to achieve low inflation (to anchor low public inflation expectations) and subsequently tends to shock the public with high inflation to stimulate economic growth or reduce unemployment. Nonetheless, if the central bank seriously and constantly strives to build its reputation (which typically requires a protracted period), by meeting its commitments, the public will gradually begin to trust the actions taken by the central bank. Consequently, if the central bank announces its commitment to maintain low inflation, the public will trust the move and keep their inflation expectations low. Oppositely, by setting high inflation, the central bank will damage its reputation and find it more difficult to catalyze economic growth in future periods. It is important to note that when building reputation, the policy authority must signal its seriousness to the economic agents in terms of economic policy implementation. In many cases, this seriousness is evidenced by bold policy measures as shock therapy with significant implications on the workings of the economy. In addition, an appropriate policy sequence is required for a policy to be sustainable. For example, a microeconomic adjustment and institutional change are required before macroeconomic policy is implemented. If the policy sequence is appropriate, however, the economic agents will see that any misalignment would be detrimental to the policy authority itself and, therefore, they trust that the policy authority will not adjust the economic policy sequence. (3) Delegation of authority Concerning efforts to improve policy credibility through independence, one practical solution was also proposed by Rogoff (1985) through the delegation of authority. Based on the principal-agent rule, Rogoff stressed that the reigns of central bank leadership should be delegated to inflation-averse or inflation-conservative individuals rather than other types of economist. By delegating authority, a mechanism to guarantee monetary policy implementation consistent with the hopes of the public is expected.4 For example, if the central bank is given a high level of independence, the central bank is also expected to protect monetary policymaking from political processes. To that end, Alesina and Gatti (1995) argued that central bank independence could reduce dynamic influence and political uncertainty on output.

4

After the seminal theoretical contribution of a “conservative central bank” (Rogoff, 1985), the literature on independence was enriched by other studies, including “insulation from political uncertainty” (Alesina & Gatti, 1995) as well as “optimal contract” (Walsh, 1995). Meanwhile, a prominent theoretical argument regarding accountability is contained in “democratic accountability” (Eijffinger, Hoeberichts, & Schaling, 2000).

Monetary Policy Credibility and Time Consistency     297 In addition, the central bank could join a monetary union and apply a fixed exchange rate system, thereby submitting to the monetary policymaking of the monetary union. By capitulating to the exchange rate policy and monetary policy of a large and credible central bank, the credibility of the policy authority itself could also be improved. This is an effective method for relatively small countries or those suffering from comparatively low policy credibility. (4) Contract determination Walsh (1995) opined that if inflation bias is present, namely that if a central bank is tempted to precipitate economic change that triggers inflationary pressures, this may be due to a lack of incentives. Therefore, an optimal contract must be designed for the central bank to prevent, or at least reduce, potential inflation bias. In this context, the government could offer a contract, for instance through performance-based salaries for central bankers, in accordance with achieving the inflation target or another economic variable at a certain level, for example, money supply growth. In this case, central bankers would be rewarded if inflation is successfully controlled but also penalized in failure or a lack of accountability. Such an optimal incentive structure in terms of controlling inflation is congruent with the current trend of setting an inflation target as the overriding goal of monetary policy. (5) Political support for the rule of law The benefits of an economic policy must ultimately be redistributed to the wider community. If the policy authority does not enjoy the support of society at large, the economic policies taken will not run smoothly. In general, policy authority credibility depends upon support from the political party in power, legitimate in the eyes of the public. Such political support is typically obtained when a new government administration is elected. The public tends to accept the policies of the new administration even though such policies could spur economic pressures in the near term. Nevertheless, support can quickly turn to opposition in the event of an extended recession, which would force the policy authority to adjust the contents of the economic policy package. Meanwhile, laws applicable to policymakers can also be viewed as an appropriate measure to improve policy credibility. This may be in the form of state laws contained in the constitution. Furthermore, unwritten rules can also apply, such as social norms or social contracts that have a strong legal foundation. In this regard, it is believed that standard behaviors or ethics when serving the public may be unwritten but, if acceptable to the public, could have a strong effect. Based on that rationale, the importance of commitment through a rule-based policy response, reputation building or the delegation of authority is the primary aspect underlying policy credibility oversight. This is also based on the recommendations of several research papers that the central bank should adopt a monetary policy framework which prioritizes good governance, for instance inflation targeting. Through inflation targeting, as discussed in Chapters 8 and 9, a policy

298    Central Bank Policy framework can be built that orients central bank discipline and commitment to achieve the inflation target in the long run, while simultaneously improving monetary policy credibility in the eyes of the public.

10.3. Theoretical Model Concerning the Policy Credibility Problem: Time Inconsistency The model of Kydland and Prescott (1977) and further development by Barro and Gordon (1983) represent two seminal references in terms of policy credibility theory. Furthermore, both models underlie the theories of independence and accountability regarding policy transparency to be discussed in Chapters 11 and 12.

10.3.1. Kydland and Prescott’s Model Evidence for the problem of time inconsistency was first proposed by Kydland and Prescott (1977). In their seminal work, Kydland and Prescott analyzed the time inconsistency problem to show economic policy formulation that generally relies on a rule is better than relying on discretion. Kydland and Prescott used a dynamic game theory approach, à la Stackelberg, between the policymaker (leader) and markets (followers). They showed that although policymakers can proclaim an optimal policy and the markets may trust the claim, there is the possibility that in subsequent periods the policy implemented is no longer optimal if there is a change from the original plan. A change in policy previously considered optimal over time is known as time inconsistency and, therefore, a policy measure relying on discretion is considered suboptimal or failing to maximize public prosperity. According to the approach, the two periods to determine the policy and social objective function of the policy are assumed as follows:

S = S (x1 , x2 , π1 , π2 )(1)

where, S, x, and π represent the policymakers’ preferences, economic agents’ decision variable and policymakers’ (instrument) decision variable. If, in the economic decision-making process, the economic agent is assumed to consider the policies formulated by the authority, then:

x1 = x1 (π1 , π2 )

and      x2 = x2 (x1 , π1 , π2 )(2)

Seeking an optimal intertemporal solution, the decision of both policy instrument variables is taken in period 1. Therefore, by substituting the values of x1 and x2, the objective function of the policymaker (1) can be expressed as follows:

S (.) = S (x1 (π1 , π2 ), x2 (x1 (π1 , π2 ), π1 , π2 ), π1 , π2 )(3)

Monetary Policy Credibility and Time Consistency     299 Therefore, the first-order conditions for optimal decisionmaking based on a rule, namely π1* and π2* are as follows: ∂S (⋅) / ∂ π2 = [ ∂S / ∂ x1 ⋅ ∂ x1 / ∂ π2 + ∂S / ∂ x2 ⋅ ∂ x2 / ∂ x1 ⋅∂ x1 / ∂ π2 ]



+[ ∂S / ∂ x2 . ∂ x2 / ∂ π2 + ∂S / ∂ π2 ] = 0 Thus : ∂S / ∂ x2 ×∂ x2 / ∂ π2 + ∂S / ∂ π2  +∂ x1 / ∂ π2 [ ∂S / ∂ x1 + ∂S / ∂ x2 ×∂ x2 / ∂ x1 ] = 0

(4)

Meanwhile, for discretion-based policy, the policymaker in period 1 determines π1* and π2* in the same way as above but in period 2, π1* and π2* refer to the value already realized. Consequently, in the second period, the policymaker will take another decision based on the following conditions: S = S (x1 , x2 , π1 , π2 )(5)

with

x2 = x2 (x1 , π1 , π2 );  x1 = x1 , dan           π1 = π1*(6) Thus, the first-order condition for period 2 with discretion-based policy π2** is:



∂S / ∂ x2 . ∂ x2 / ∂ π2 + ∂S / ∂ π2 = 0 (7)

From the two solutions above, the first-order condition with discretion will only be the same as the first-order condition for the rule if ∂x1/∂π2 [∂S/∂x1 + ∂S/∂x2 × ∂x2/∂x1] = 0. In reality, such conditions are difficult to find and therefore, the solution for a rule (π1*, π2*) will be different to the solution for discretion (π1*, π2**). As the solution for a rule (π1*, π2*) maximizes the intertemporal utility, the result is optimal. Meanwhile, discretion-based policy (π1*, π2**) will not be optimal because discretion-based policy in period 2 would not consider the impact of the decision taken in period 1 on (x1, π1). Kydland and Prescott provided a graphical illustration concerning the application of the monetary policy time inconsistency problem in terms of a dilemma, linked to the trade-off between inflation and unemployment. In this case, the relationship between inflation and unemployment is assumed to follow the Expected Augmented Phillips Curve (EAPC) as follows:

ut = u* – α (πt – π e )(8)

where u* is the natural rate of unemployment. The exogenous parameter α is positive (>0), while the inflation expectations for period t are formed in period t − 1. As inflation exceeds the expectations, real wages decline, thereby stimulating

300    Central Bank Policy

Fig. 10.1:  Indifference Curve of the Time Inconsistency Problem. Source: Kydland and Prescott (1977). demand for labor (reducing unemployment). Meanwhile, the objective of the policymaker is to maximize the objective function: S (ut , πt )(9)



where S reached maximum at the point where μt < μ* and πt = 0. Specifically, the formal form of S is represented by the quadratic function5: S = ( ut − kut* )2 + γπt2 (10)



Parameter y represents the weight of the inflation preference (relative to unemployment) chosen by the policymaker. Meanwhile, k < 1 reflects distortion that compels the government to target unemployment below equilibrium. The indifference curve representing S is presented in Fig. 10.1. Assuming no uncertainty, πe = πt and, therefore, ut = u*. If, in period t − 1, the policymaker (government) is committed to controlling inflation, remembering ut = u*, the optimal rate of inflation is πt = 0. Therefore, the optimal ex ante choice is (u*, 0), namely at point O. Nevertheless, the commitment is time inconsistent because in the subsequent period (ex post), when the economic agents’ expectations have already been formed (in line with inflation set by the government in period t), the government 5

The S curve is concave with a negative partial derivative to both arguments, ut and πt.

Monetary Policy Credibility and Time Consistency     301 is tempted to push up inflation to reduce unemployment further. Ultimately, this will edge up inflation and create inflation bias, namely at point C. It is said to be a bias because with the rational expectations of economic agents, the main preference of the policymaker to reduce unemployment and also the commitment to control inflation, only higher inflation would be achieved, while unemployment would not be reduced.

10.3.2. Barro-Gordon Model The ideas of Kydland and Prescott were more clearly interpreted by Barro and Gordon (1983) in a model which showed that through commitment (rule), the central bank could achieve a desired level of inflation consistent with public expectations, while incurring minimal losses. With discretion, however, public expectations constantly change in line with the inconsistency of the central bank’s measures, thus creating inflation bias. Barro and Gordon (1983) analyzed the time inconsistency problem in monetary policy through game theory, à la Nash Equilibrium, between the central bank and public or private sector in the economy with the following conditions: (1) The central bank is assumed capable of managing economic activity and orienting monetary policy toward achieving social prosperity/public welfare. (2) The public only acts in accordance with trusted inflation expectations. (3) Time inconsistency can occur because on one side the public must form their inflation expectations at the beginning of the period and maintain them until the end of the period, while, on the other hand, the central bank has full discretion to set strategy over time. Under such conditions, the inflation target set at the beginning of the period may not be optimal by the end of the period and will create gains/incur costs for the central bank and public. The model was formulated as follows. The central bank minimizes the social loss function (social costs):

L = b(U −U * )2 + π 2 (11)

with parameter b > 0 and initial inflation target of zero (π* = 0). The relationship between unemployment (U) and inflation follows the EAPC as follows:

U = U n – a (π – π e ) with a > 0 and U * = k U n

where 0 < k < 1(12)

The central bank can control inflation through its monetary policy, thus inflation is in line with money supply growth, namely π = m. In other words, there are

302    Central Bank Policy no problems in terms of monetary policy transmission. With that assumption, the central bank’s problem is to minimize the social costs as follows: Z = {b[(1− k )U n – a ( π − π e )]2 + π 2 }(13)



From the first-order condition, the social costs would be minimized if inflation is reduced optimally as follows: ∂ Z / ∂ π = −2ab[(1− k )U n − a (π – π e )] + 2π = 0



π ** = ab(1− k )U n / (1 + a 2 b) + a 2 b π e / (1 + a 2 b) = Φ ( π e )

(14)

The solution indicates time inconsistency, namely that the initial inflation target of zero (π* = 0) is no longer optimal by the end of the period. The presence of EAPC causes the optimal rate of inflation to be influenced by inflation expectations. In general, inflation expectations are positive (πe > 0), therefore optimal inflation is also positive (π** > 0). In fact, if inflation expectations are zero (πe = 0), optimal inflation would still be positive (π** > 0) because of the social cost factor (b) and unemployment phenomenon (k) as well as inflation deviation from the target (a), which is positive. Why? This is because public inflation expectations influence actual inflation, which can be explained by the sources, namely: (a) Surprise inflation. Inflation will still occur even if the central bank sets the initial inflation target at zero, π* = 0, and public expectations are also zero, πe = 0, as long as the central bank can ignore inflation deviation from the target because of real sector dynamics. Under such conditions, actual inflation is as follows:

π s = ab(1− k )U n / (1+ a 2 b)(15) Thus, the social cost of inflation is as follows:

LS = {b[(1− k )U n – a ( ab(1− k )U n ) / (1 + a 2 b)  2 +  ab(1− k )U n / (1 + a 2 b)]2

LS = b[(1− k )U n ]2 / (1 + a 2 b)

(16)

(b) Inflation bias. Although the public form rational expectations (congruent with the rational expectations hypothesis), thereby knowing the central bank loss function and Phillip’s curve behavior (game theory à la Stackelberg), the initial inflation target, π* = 0 would not be credible. Under such conditions, inflation expectations would be the same as the central bank’s target, π* = πe. Therefore, actual inflation would become:

π REH = ab(1− k )U n(17)

Monetary Policy Credibility and Time Consistency     303 And the social cost:

LREH = b[(1− k )U n ]2 + ab(1− k )U n  2 = b (1+ a 2 b)  (1− k )U n ]2 (18)  

(c) Inflation rule. Barro and Gordon suggested that with rational expectations and the behavior/game above, the social cost could be reduced by giving the central bank a rule that inflation = 0 and money supply growth = μ. Under such conditions:

π RULE = 0 and LRULE = b[(1− k )U n ]2 (19)

The analysis above indicates that the smallest social cost occurs when there is surprise inflation and largest when there is inflation bias, while the inflation rule is somewhere between the two or:

LS < LRULE < LREH(20)

Therefore, in the context of game theory, the Barro-Gordon model tends to produce the prisoner’s dilemma, meaning that the optimal strategy for both agents (central bank and public) would be detrimental to both agents. The discussion above shows that while there is deviation between the inflation target and actual inflation, optimal inflation will not be achieved, which incurs a social cost. Consequently, the central bank is required to orient policy to achieve the target and the public form inflation expectations in line with the target set by the central bank, thus π* = πe = π**. This is where the issue of monetary policy credibility becomes important. To explain the benefits of monetary policy credibility, the Barro-Gordon model can be reanalyzed as follows: (a) If monetary policy is not credible, the central bank is forced to assume public inflation expectations as an (exogenous) explanatory variable. Therefore, as in the previous analysis, the social cost will be minimized if:

Fig. 10.2:  Policy Strategy and the Social Cost. Source: Bofinger (2000).

304    Central Bank Policy

π ** = ab(1− k )U n / (1+ a 2 b) + a 2 b π e / (1+ a 2 b) = Φ (π e )(21)

In this case, the central bank would expect optimal inflation policy to be influenced by inflation expectations and the level of unemployment, hence the inflation target would change over time. The level of unemployment at optimal inflation is as follows: U ** = U n – a[{ab(1− k )U n / (1+ a 2 b) + a 2 b π e / (1+ a 2 b)}– π e ]

= [1 – a 2 b(1− k ) / (1+ a 2 b) U n – a  a 2 b / (1+ a 2b)−1]π e So U ** = (1+ a 2 bk ) / (1+ a 2 b)U n +  a / (1+ a 2 b) π e    

(22)

*****Therefore, the actual level of unemployment depends on the target (value) of desired unemployment, k, and inflation expectations. The natural rate of unemployment under economic equilibrium will be achieved if the target k = 1 and inflation expectations πe = 0. (b) On the other hand, if monetary policy is credible, thus π = π* = πe, then the social cost is:

L = {b[(1− k )U n ]2 + π 2 }(23) From the first-order condition:



∂ L / ∂ π : 2π = 0 (24)

Therefore, if monetary policy is credible, optimal inflation is π** = 0. Such conditions also guarantee achievement of the natural rate of unemployment.

10.3.3. Credibility Problem Analysis with Supply Shocks Sources of the policy credibility problem are very diverse and complex. From a theoretical perspective, that rationale highlights the role of changes in the preferences of the policymakers to stimulate output or reduce unemployment (parameters b and k) as well as inflation deviation due to the need to stimulate output or reduce unemployment (parameter a). What follows is an analysis of how the credibility problem could be exacerbated by shocks in the economy, for instance supply-side shocks. This is important to monitor, primarily by considering the economic characteristics of developing countries, where supply shocks are comparatively dominant in terms of influencing monetary policy effectiveness. Therefore, the Barro-Gordon model was developed further, namely by including supply shocks. This implies tighter requirements for the central bank when formulating monetary policy to garner credibility. To analyze monetary policy

Monetary Policy Credibility and Time Consistency     305 credibility under conditions of supply shocks, the Phillip’s curve is assumed to behave as follows:

U = U n – a (π – π e ) + ε(25)

where ε = supply shock. Therefore, the problem of minimizing the social cost by the central bank becomes:



L = [ b(U −U * )2 + π 2 ] = {b[ (U n – a (π – π e ) + ε) − kU n ]2 + π 2 } (26) L = {b[(1− k )U n – a (π – π e ) + ε )]2 + π 2 }

Consequently, three scenarios can be analyzed as follows: According to the first scenario, the central bank cannot guarantee that actual inflation will be zero but can be committed to maintaining average inflation at zero (πR = 0). The public trusts the central bank and forms expectations based on the average inflation, thus πe = πR = 0. Therefore, the social cost becomes:

L = [ b((1− k )U n – a π + ε )2 + π 2 ] (27) From the first-order condition:



∂ L / ∂ π =− 2ab((1− k )U n – a π + ε ) + 2π = 0 (28) π = ab((1− k )U n + ε ) / (1+ a 2 b)



And average inflation is :   E (π) = ab(1− k )U n / (1+ a 2 b )(29)

According to this scenario, therefore, to ensure average inflation is equal to zero (E(π) = 0), the central bank would need to not only overcome the supply shocks but also orient monetary policy to achieve full employment output, or k = 1. Monetary policy credibility would produce an optimal rate of inflation π** = 0. Therefore, although credibility could be achieved at any level of inflation, optimal inflation policy (inflation = zero) would create full credibility. Under such conditions, the natural rate of unemployment could also be achieved. According to the second scenario, the central bank guarantees that actual (not average) inflation would be equal to zero and the public trusts the central bank, thus πe = π =0. Therefore, the social cost becomes:

L = [ b((1− k )U n + ε )2 + π 2 ](30) And the inflation solution is π = 0

306    Central Bank Policy If the central bank could achieve actual inflation of 0, the policy would be considered credible although difficult to implement because the contained supply shocks are difficult to predict. Per the third scenario, the central bank treats public inflation expectations (πe) as exogenous and the public act rationally. The social cost therefore becomes:

L = [ b((1− k )U n – a ( π − π e ) + ε )2 + π 2 ](31)

From the first-order condition:

∂ L / ∂ π = −2ab((1− k )U n – a ( π − π e ) + ε ) + 2π = 0 (31)

Actual inflation becomes:

π = ab((1− k )U n + a π e + ε ) / (1+ a 2 b)(32)

And average inflation:

E (π) = ab((1− k )U n + a π e ) / (1+ a 2 b)(33)

According to this scenario, it is more difficult for the central bank to achieve monetary policy credibility because the central bank must meet several requirements: (1) achieve the inflation target and earn public trust, thus the inflation expectations are equal to zero (πe = 0); (2) overcome supply shocks; and (3) orient monetary policy toward achieving full employment output, or k = 1. From the three scenarios, the second was considered the best. Therefore, monetary policy would gain credibility if the inflation target was achieved along with public trust in the formation of inflation expectations. Nevertheless, the impact on output and unemployment would also have to be analyzed.

10.4. Empirical Studies: Measures of Credibility and Time Inconsistency As mentioned previously, numerous studies have been conducted regarding the credibility problem and time inconsistency but most are hypothetical, namely with a fundamental understanding that the time inconsistency problem would create inflation bias and then reduce the effectiveness of monetary policy. Very few empirical studies deal with measuring the extent of the costs incurred due to time inconsistency. In general, the empirical studies deal with the influence of institutional aspects, for instance independence and transparency, in terms of the impact on monetary policy effectiveness. Several relevant empirical studies dealing

Monetary Policy Credibility and Time Consistency     307 with measures of credibility and time inconsistency, using various approaches, are detailed as follows.

10.4.1. Parameterization Approach: Credibility Index Cobham, Papadopoulos, and Zis (2001) analyzed the credibility of monetary policy implementation in the UK for the period from 1994 to 1997 using an ex ante and ex post credibility index. The ex ante credibility index was compiled based on the frequency of agreement/disagreement between the Minister of Finance as the decisionmaker and the Bank of England (BoE) as the provider of recommendations concerning changes to the interest rate. Meanwhile, the ex post credibility index was calculated based on the yield differential between the UK and Germany (or the US). Arguing that political intervention to manipulate central bank preferences would reduce monetary policy credibility, they hypothesized that there is a correlation between both aspects. For example, a lack of agreement from the Minister of Finance regarding the recommendations put forward by the BoE (published with a certain lag) could be interpreted by the financial markets as a difference in preferences between the government and central bank (mandated with maintaining price stability). This could be further interpreted as evidence that the political interests of government to realize a separate goal would undermine monetary policy credibility, thereby raising the long-term interest rate. Using the long-term interest rate relates to the understanding that the long-term interest rate contains a risk premium, which reflects the possibility that the timing and magnitude of interest rate change is influenced by political considerations. In that context, long-term interest rate data is taken from the yield data of 10-year government debt securities. Meanwhile, the classification of agreement/disagreement used to compile the ex ante credibility index refers to the five categories obtained from the minutes of the meeting, namely a firm vote to raise the interest rate, propensity to raise the interest rate, impartiality to raise/lower the interest rate, propensity to lower the interest rate, and a firm vote to lower the interest rate. Statistically, the index with an interval of [0.1] has a number of characteristics. First, this index is sufficiently persistent, where the value is influenced by the previous value (smoothing parameter of 0.75) and adjusted incrementally to the latest information regarding new consensus. Second, the index behaves asymptotically toward the extremes (between 0 and 1), implying that for the next agreement/disagreement to occur, the progressive increase/decrease in the index will become smaller. The next step is to assess the equations of long-term correlation of the two credibility indexes. The assessment evidenced a negative correlation between the two variables, namely that monetary policy credibility in the UK is symmetrically linked to the agreement/disagreement between the Minister of Finance and BoE. Those findings imply that the costs incurred due to political intervention on monetary policy implementation are significant. It was concluded from the simulations that if the Minister of Finance agrees with the recommendations put forward by the BoE, the average long-term interest rate will be 1.5% lower than

308    Central Bank Policy the average actual rate. In other words, if the long-term interest rate in Germany was 5.87% (April 1997), without political intervention, the long-term interest rate in UK would be 6.11%, not the 7.61% recorded. Interesting results are found relating to an alternative assessment if different long-term interest rates in the United Kingdom and United States are used. The assessment found that there was not a symmetrical correlation between different long-term interest rates in the United Kingdom and United States with the credibility index, which reflects differences in terms of monetary policy implementation or other specific factors that influence yields in the United States and Germany. In this case, they found that United States yields were more fluctuative than yields in Germany. Ultimately, the differences between the United States and Germany were easier to understand in a more strategic context, where the United States monetary policy regime was considered to lack a clear nominal anchor, more commonly referred to as a “just do it” regime.

10.4.2. Qualitative Approach (Narrative) Researching the relevance of time inconsistency theory in terms of setting monetary policy in the United States, Chappel and McGregor (2004) applied a narrative approach, namely by using historical evidence in the form of recorded conversations at the Federal Open Market Committee (FOMC) contained in the Memorandum of Discussion to evaluate the verbal arguments used by policymakers to support their desired policy direction. Empirically, the analysis focused on the period from 1970 to 1978 (when the Fed was under the leadership of Arthur Burns), a period when strong inflationary pressures blighted the United States. The main advantage of the narrative approach is its original ability and fairness to explore the rationality and motives of the policymakers through their arguments delivered to support a specific policy direction. Nevertheless, the approach does not allow the testing of a hypothesis against other alternative hypotheses, while the arguments observed were chosen beforehand in the context of the analysis. The analysis based on the Memorandums of Discussion and various other transcripts obtained from official documentation (Ford Library) explored the journeys of the policies containing time inconsistency during the 1970s, which could be grouped into four aspects as follows: (1) public and political preferences concerning economic performance; (2) the short and long-term trade-offs of the Phillips curve; (3) the policymaking characteristics of the FOMC over time; and (4) the role of expectations under equilibrium conditions. Broadly speaking, based on those four aspects, the narrative analysis produced the following conclusions: (1) In general, the FOMC under Arthur Burns very rarely achieved politically acceptable levels of output and unemployment. Technically, in the standard Barro-Gordon model, inflation performed beyond the desired level (b > 0), while political pressures forced the Fed to constantly strive toward a level of unemployment below the natural rate (k < 1).

Monetary Policy Credibility and Time Consistency     309 (2) The FOMC adhered to the Phillips curve behavior assumed in the time inconsistency theory. In the short term, the sacrifice ratio was relatively high but in the long term, the inflation expectations formed contributed to high inflation and unemployment. In addition, from the FOMC’s perspective, the rising cost of controlling inflation, triggered by expectations, was considered exogenous at each meeting. Consequently, although inflation was already high, the rational measures taken by the policymakers tended to perpetuate high inflation. (3) During the 1970s, FOMC members faced problems in terms of determining the policy focus for the subsequent period as well as considering the long-term consequences of that policy. One member complained about the briefness of the time allocated to formulate and deliver alternative policies to the FOMC, while also believing a longer period was required to evaluate the future inflationary impact of the policies proposed. Furthermore, the FOMC desired contractionary monetary policy with the consequence of a possible recession. Nevertheless, pressures in the form of public and political opinion blocked that decision. Such conditions demonstrated how the FOMC was forced into policies with short-term goals, despite knowing there would be long-term consequences. (4) Although FOMC members tended to regard public expectations as inertia, they also felt that the financial markets often reacted excessively and irrationally in response to changes in policy. Under such circumstances, the FOMC tended to be too prudent in response to market behaviors when interpreting a policy and tried to avoid excessive and irrational market reactions. Consequently, the FOMC realized that the benefits of the monetary stimuli issued would be more effective if applied without a corresponding increase in public expectations. From the narrative, the problem of time inconsistency was ultimately concluded to play an important role in terms of perpetuating inflation persistence in the United States during the 1970s. In this case, the Fed admitted that the Phillips curve trade-off and political pressures had made it more difficult for the Fed to introduce disinflationary policies in each period. On the other hand, however, such conditions also exacerbated the already excessive expansionary monetary policy due to an extremely short policy planning horizon at each FOMC meeting.

10.4.3. Quantitative Modeling Approach Analyzing the phenomenon of time inconsistency in the United States, Surico (2003) assessed the magnitude of inflation bias in monetary policy implementation for the period from 1960 to 2002 using a hypothetical model, representing the reduced form of the policy response (policy rule). Surico developed a strategy to identify the relevant asymmetric parameters in the objective function of the central bank, where policymakers were more concerned about output contraction rather than expansion, and then interpreted that into a measure of

310    Central Bank Policy time inconsistency. In addition, by comparing solutions with commitment and solutions using discretion, average inflation could be decomposed in the form of a target and bias. This approach provided an important new contribution in the literature about optimal monetary policy because it could measure the average inflation bias that appeared from the asymmetrical preferences of the central bank on output stabilization. The central bank loss function in exponential linear form is specified as follows:

Lt =

1

2

(πt –

π * ) + λ[( exp( γ yt ) – γ yt – 1) / γ 2 ](34) 2

Meanwhile, the feedback rule of the empirical model, solved by optimizing the central bank objective function is as follows:

πt = π * + α yt + β yt 2 + vt(35) That equation has a linear parameter, α = −λθ and β − λθ λ/2 With the error terms defined as follows:



vt = – {α( yt – Et−1 yt ) + β [ yt 2 – Et−1 ( yt 2 )] + et }

where y is the output gap, λ is the relative weight of the central bank preference to output stability, θ reflects the effect of inflation deviation (on expectations) in the output formation of the Phillips curve, and γ refers to the characteristics of the asymmetric central bank preferences. Meanwhile, rational expectations are formed in period t − 1. The complexity of the parameters in the equation represents the preferences of the policymaker and explains the structure of the economy. By solving the reduced form, the asymmetric preferences of the central bank can be identified further, γ = 2β /α and average inflation bias, βσy2. If the costs that are incurred are larger during output contraction than output expansion, the model predicts γ < 0, α < 0 (because γ and θ are positive) and β > 0. By including the behavior in the Phillips curve, the model implies that the central bank has an incentive for surprise inflation to negate the possibility of declining economic activity. Therefore, the asymmetric preference for output stability would precipitate prudence in terms of expansion. This in congruent with the model behavior that predicted a positive correlation between average inflation and variations in the output gap (σy2). Observations were also performed covering two different regimes, namely before and after the leadership of Paul Volker (1979 breakpoint) as well as during the period of leadership under Alan Greenspan (after 1987). Using the Generalized Method of Movement (GMM), the results concluded there were significant policy (regime) changes during 40 years of monetary policy implementation in the United States, as measured by changes in the monetary policy preferences of the Federal Reserve. Specifically, the results showed that in addition to a reduction in average inflation target from 3.42% to 1.96%, inflation bias, which was estimated at around 1.01% prior to 1979, was minimal (approaching zero) in the

Monetary Policy Credibility and Time Consistency     311 final two decades of the observation period. The rationality of the results can be linked to the magnitude and asymmetry of the monetary policy preference toward output stabilization found in the period of Paul Volker’s leadership and before, but not in the period thereafter. Despite other affecting factors, including better policy formulation and the minimal impact of supply shocks, the analysis successfully provided a new notion in the empirical evidence and quantitative measures of US policy behavior.

10.5. Assessing Monetary Policy Credibility in Indonesia A change to the mandate, policies and institutional arrangements of Bank Indonesia after the passing of Act No. 23 of 1999 eventually oriented monetary policy in Indonesia toward implementation of the inflation targeting framework (ITF). Bank Indonesia reforms consolidated the institutional dimension (governance) in terms of central bank practices and simultaneously represented an integral part of macroeconomic recovery policy to escape the clutches of crisis.

10.5.1. Policy Credibility in Indonesia during ITF Implementation Per the ITF regime, forward-looking monetary policy to future inflation implies that future inflation must be controlled in line with the existing target. To support policy formulation, the central bank must be able to capture or take into account expectations as well as accurate and reliable information concerning future inflation projections to set the desired monetary policy response, be it neutral, tighter or looser. This surely requires high policy credibility at the central bank to provide avowed commitment to achieving low and stable inflation. If the central bank is credible, the inflation expectations of economic agents could be oriented toward the inflation target set by the central bank. As explained previously, theoretically, if monetary policy is fully credible, inflation expectations and actual inflation will converge on the inflation target.6 Observations of the Indonesian case have shown that during the past decade, inflation has tracked a downward trend toward the era of single-digit inflation. In this case, congruent with increasing inflation control policy credibility by Bank Indonesia working in conjunction with the Government during the ITF era since 2005, core inflation has declined significantly from an average of 8.1% before 6

In practical terms, however, policy credibility can be observed in terms of target achievement and policy implementation. Concerning policy target achievement, credibility can be assessed based on whether the central bank can attain the inflation target. Meanwhile, regarding the policy implementation process, policy credibility can be evaluated based on the policy response, whether implemented consistently or not. Theoretically, both aspects are closely related, implying that a well-planned policy implementation process will support achievement of the desired policy target. In practice, however, the central bank may implement sound and consistent policy but, due to influences beyond the control of the central bank, the inflation target cannot be achieved optimally.

312    Central Bank Policy

Fig. 10.3:  Core and Headline Inflation Trends. ITF implementation to an average of 5.2% after ITF implementation. Likewise, headline CPI inflation has also experienced significant disinflation from an average of 8.7% to 6.2%. It is important to note that during a few episodes, headline inflation exceeded the target set by the Government, after consultation with Bank Indonesia, due to extreme inflationary pressures stemming from administered prices (AP) and volatile foods (VF). Moving forward, however, the government’s energy reforms have bolstered commitment to controlling inflation on the supply side, therefore, inflationary pressures from AP and VF are predicted to remain minimal. Such developments are also in line with the results of empirical studies showing that prior to ITF implementation, monetary policy credibility was relatively low in Indonesia and this was one of the determinants of high inflation persistence (Harmanta, 2009). Nonetheless, congruent with ITF implementation, monetary policy credibility improved, reflected by a Kalman gain7 from around 0.2 in the period before ITF implementation (July 2005) to approximately 0.4 after ITF implementation. It could be argued that if monetary policy credibility was still low, similar to conditions before ITF implementation, inflation would be influenced by the backward-looking behavior of economic agents, thus creating higher inflation persistence and delaying convergence with the central bank’s inflation target. This is because monetary policy that is not yet credible prolongs the learning process of economic agents in the context of the inflation target. In contrast, when monetary policy credibility improves, such as after ITF implementation, inflation 7

Technically, the Kalman gain captures the forecasts of economic agents in terms of projecting the inflation target of the monetary authority in the Taylor rule equation through a derivation of the Kalman filter. The Kalman gain reflects the learning process of economic agents towards the inflation target of the monetary authority, which can be used as a measure of the degree of monetary policy credibility, where a faster learning process implies a greater degree of monetary policy credibility. The value of the Kalman gain is between 0 and 1, where a value closer to 1 indicates a greater degree of monetary policy credibility.

Monetary Policy Credibility and Time Consistency     313 is then influenced more by forward-looking expectations, which lowers inflation persistence and anchors inflation expectations to the monetary authority’s target that is also forward looking.8 The dimensions of monetary policy credibility can be observed from the implementation of inflation (disinflation) control strategy. Theoretically, ITF aims to achieve low and stable inflation.9 Empirically, based on the experience of several ITF countries, there are a number of strategies that could be employed to achieve low and stable inflation. In general, when ITF was first implemented, advanced countries had typically already achieved low (single-digit) inflation, thereby facilitating an aggressive long-term disinflation strategy to around 2.0%–3.0%. In contrast, developing countries were typically experiencing double-digit inflation when ITF was first implemented, thus a more gradual disinflation strategy was implemented with a short- and medium-term target and the pace of disinflation was set at around 0.8% per annum prior to convergence with the long-term inflation target of 2.0%–3.0% (Roger & Stone, 2005).10 Observations of the Indonesian case have shown that the disinflation process pursued by Bank Indonesia during the ITF era, including the transition phase and after official ITF implementation, was applied gradually. The rate of disinflation stood at around −0.5% per annum, slightly lower than the average disinflation rate recorded in developing countries at −0.7% per annum (Roger & Stone, 2005). This is in line with monetary policy in Indonesia, which is not fully credible, thus the learning process of economic agents in terms of reducing the inflation target to low and stable inflation was slightly more protracted. In addition, the 8

This is congruent with numerous empirical observations showing that the formation of natural inflation expectations oriented towards the inflation target as the main monetary policy target in the ITF era would facilitate the learning process of economic agents compared to conditions before ITF application, when multiple targets were possible. That process would create more forward-looking inflation expectations among economic agents after ITF implementation, which is in line with the findings of Juhro (2004), Yunuarti (2007), and Alamsyah (2008). Furthermore, that research is also consistent with Siregar and Goo (2008), who showed that inflation inertia in Indonesia declined, for tradeable and non-tradeable goods, after ITF implementation. 9 Experience from several ITF countries showed that the inflation target has a number of traits: (1) sufficiently low; (2) small fluctuations; (3) challenging for the central bank to achieve; and (4) can be achieved with minimal output loss. 10 Several studies have shown that the disinflation strategy is explicitly subject to central bank monetary policy credibility. Advanced countries, where monetary policy is deemed more credible, tend to apply a more aggressive disinflation strategy because monetary policy can anchor the inflation expectations of economic agents and actual inflation with greater rapidity towards the inflation target, thus the interest rate response will be optimal and avoid any large output loss. Developing countries, however, where monetary policy is considered less credible, tend to apply a more gradual disinflation strategy because monetary policy is unable to steer inflation expectations and actual inflation as rapidly towards the inflation target, hence the interest rate response is not optimal and could trigger a large output loss.

314    Central Bank Policy gradual disinflation process implemented in Indonesia was also due to previous experience of double-digit inflation as well as the moderate degree of inflation persistence, while not fully forward looking or anchored to the inflation target. Gradualism in terms of the disinflation process in Indonesia was fundamental because of imperfect credibility, therefore overly aggressive monetary policy oriented toward disinflation within a brief period of time would have led to an unacceptably high sacrifice ratio. In the context of a small open economy, the impact of rapid disinflation against a backdrop of imperfect credibility by hiking the interest rate would spur excessive appreciatory pressures on the exchange rate, thus exacerbating the trade-off between inflation and output.11 With the gradual disinflation strategy adopted in Indonesia, achieving the long-term target of low and stable inflation was expected to require a prolonged period. Nevertheless, this was also affected by the process to build policy credibility itself and the dynamic public learning process. In relation, the empirical study conducted by Cukierman (2006) showed that a characteristic of inflation stabilization from double digits toward the medium- and long-term inflation target was gradual disinflation. Experience in several countries, including Chile, has also shown that disinflation toward low and stable inflation under the ITF requires a prolonged period, namely around 36 quarters, or 9 years. In fact, in addition to Indonesia, there remain several ITF developing countries that are yet to complete the disinflation process, including Colombia – since 1999, Rumania – since 2005, and Turkey – since 2006.

10.5.2. Empirical Studies of Time Inconsistency in Indonesia Harmanta (2009) reported that high inflation persistence in Indonesia is caused by imperfect monetary policy credibility. Furthermore, research by Revenna (2005) also substantiated the previous claims based on a survey of 82 countries that placed monetary policy in Indonesia in the “low credibility” category because the inflation target had not been achieved. The issue of monetary policy credibility in Indonesia relates closely to the time inconsistency problem, (Goeltom, 2005). Time-inconsistent policy has the potential to undermine monetary policy credibility, thus the economic agents will form higher inflation expectations than the target announced by the authorities. If this is the case, does the credibility problem in Indonesia indicate time inconsistency in terms of Bank Indonesia’s monetary policy? Goeltom (2005) stated that from 1990 to 2003, monetary policy in Indonesia was subject to the problem of time inconsistency, as reflected in the

11

Due to imperfect credibility, a gradual interest rate decline when shocks occur in terms of lowering the inflation target and actual inflation, as reflected in a tight bias monetary policy stance, could lead to a larger output loss. Tight bias monetary policy would influence the consumption habits of economic agents and, hence, weaken aggregate demand, which would feed through to the production side and output would therefore not increase. Such conditions could create a larger output loss, as reflected in the sacrifice ratio.

Monetary Policy Credibility and Time Consistency     315 suboptimal monetary policy, which manifested in an overly loose policy stance during an economic upswing and excessively tight monetary policy when the economy flagged. Studies on the institutional arrangements of the central bank are a nascent topic today and, therefore, not enough research has been done, especially for the Indonesian case. Studies on time-inconsistent monetary policy in Indonesia remain scarce. One study, conducted by Budiyanti (2009) revealed the problem of time-inconsistent monetary policy in Indonesia before and after the Asian Financial Crisis of 1997. Budiyanti used the Barro-Gordon standard linear quadratic model to explain the problem of time-inconsistent monetary policy in Indonesia. Fundamentally, the model assumed a symmetrical central bank preference to the output gap, implying that the central bank was indifferent to a positive or negative output gap. Nevertheless, the assumption of symmetrical preferences received criticism from academics and monetary practitioners on a theoretical and practical level (Blinder, 1998; McCallum, 1997). Empirically too, a large body of research doubted the assumptions of the standard linear quadratic model to explain the time inconsistency problem (Cukierman, 2000; Surico, 2003). Cukierman, for instance, questioned whether central banks really had the same preference toward negative and positive output deviation from its potential? Observations showed that the central bank and the public tend to prefer a negative output gap during an economic downturn rather than a positive output gap during an economic upswing. Therefore, such asymmetric preferences have become an important issue and interesting study topic, particularly in terms of time-inconsistent monetary policy by the central bank. The study by Rahmahdian and Warjiyo (2013) was the first to measure asymmetric preferences in terms of time-inconsistent monetary policy in Indonesia. They found at least three strategic issues in their research. First, was there an asymmetric preference parameter to the magnitude of the output gap, thus indicating time-inconsistent monetary policy in Indonesia before and after Bank Indonesia’s independence? If yes, was there a difference in the magnitude of the asymmetric preference, thus indicating different degrees of time inconsistency before and after Bank Indonesia’s independence? Second, what were the implications of time-inconsistent monetary policy or policy with different degrees of time inconsistency before and after Bank Indonesia’s independence in terms of influencing the output gap on setting the inflation target in Indonesia? Third, how did inflation perform in Indonesia before and after Bank Indonesia’s independence in the context of time-(in)consistent monetary policy or with differing degrees of time inconsistency? The analysis period was separated into two subsamples, namely the period prior to Bank Indonesia’s independence (1990:1 to 1999:4) and the period after independence (2000:1 to 2009:4). This was congruent with a fundamental change in the institutional arrangements and application of monetary policy in Indonesia in 2000, marked by a change in Bank Indonesia’s status from dependent to independent pursuant to Act No. 23 of 1999, which explicitly stipulated that monetary policy would only focus on the overriding goal of price stability.

316    Central Bank Policy The sample was also separated in reference to Rogoff (1985), who stated that central bank independence could overcome the time inconsistency problem. Likewise, the degree of time inconsistency was also suspected to differ in the periods before and after Bank Indonesia’s independence. The research by Rahmahdian and Warjiyo (2013) revealed several salient conclusions. First, the problem of time inconsistency was detected in Bank Indonesia’s monetary policy prior to independence, with an asymmetric preference to the output gap, contrasting the time-consistent monetary policy instituted after Bank Indonesia’s independence, with a symmetrical preference to the output gap. Second, time-inconsistent monetary policy prior to independence triggered intense output gap pressures on inflation, while time-consistent monetary policy after independence eliminated the influence of the output gap on inflation. Third, the implementation of consistent ITF after independence has successfully realized lower inflation, although the disinflation process has been prolonged and not quite as expected. The empirical findings of an asymmetric preference toward the output gap and its impact on time-inconsistent monetary policy before independence was reputedly due to government control over Bank Indonesia at that time. Pursuant to Act No. 13 of 1968, Bank Indonesia was mandated with multiple objectives, namely: first, to control, create and maintain rupiah stability and second, to stimulate production and development as well as expand job opportunities to ameliorate the public standard of living. That double mandate led to an asymmetric preference toward the output gap because, in addition to maintaining price stability, Bank Indonesia was also an agent of development, obligated to create job opportunities. These findings corroborated the previous research conducted by Budiyanti (2009), who found that during the period before the crisis (1990–1997), Bank Indonesia faced the time inconsistency problem in the short and long term. Furthermore, the research was also consistent with Goeltom (2005), who stated that Bank Indonesia’s monetary policy from 1990 to 2003 still faced the problem of time inconsistency, which manifested in too tight and sometimes too loose monetary policy. Regarding the subsample after Bank Indonesia’s independence, the time inconsistency problem was no longer detected. Bank Indonesia’s preference to the output gap had become symmetrical, positive or negative, thus reflecting consistent monetary policy and commitment to achieve low inflation by reducing the element of discretion in response to current economic dynamics (boom or recession). These findings were ostensibly due to the single mandate of rupiah stability and independent status of Bank Indonesia in accordance with Act No. 23 of 1999. Furthermore, monetary policy consistency, coupled with a symmetrical preference toward the output gap, was also observed to improve ITF performance. More consistent monetary policy led to more credible monetary policy in the eyes of economic agents. The empirical evidence for monetary policy consistency after independence was also accompanied by increased credibility at Bank Indonesia. Harmanta (2009) reported that monetary policy credibility increased at Bank Indonesia after ITF implementation despite imperfect credibility, which is understandable considering full ITF implementation is relatively new.

Monetary Policy Credibility and Time Consistency     317 The aforementioned three conclusions have salient implications for monetary policy in Indonesia. First, Bank Indonesia is required to constantly improve monetary policy consistency. The effectiveness of achieving the final target of monetary policy with the single mandate of price stability depends on how committed Bank Indonesia is to avoiding the inconsistency problem in the near term in order to realize low and stable inflation, for instance by stimulating stronger economic growth. It would be better to implement more consistent monetary policy based on clear rules, while reducing the element of discretion. Monetary policy consistent with ITF has been proven to lower inflation despite the current delays. Therefore, to restore public confidence in the reputation of Bank Indonesia, monetary policy consistency to achieve stability must constantly be improved. The second implication is the need to increase coordination. Considering that not all components of inflation are influenced by monetary policy, in particular AP and VF, close coordination between monetary policy and government policy, including the central government and regional administrations, is required. Existing coordination between the Government and Bank Indonesia, especially through the National Inflation Task Force (TPI) and Regional Inflation Task Forces (TPID), must be improved to achieve price stability as a prerequisite for ameliorating public welfare. In addition to minimizing the inflationary pressures originating from rising AP and VF, policy coordination is also critical to strengthen synergy in terms of general economic management. The third implication is the need to increase communication. Anchoring inflation expectations is an integral element of the ITF-based monetary policy framework, considering how far inflation expectations determine actual inflation. Public inflation expectations are still sometimes backward looking; therefore, more transparent monetary policy is required to reduce the asymmetric information between Bank Indonesia and the economic agents. Bank Indonesia must increase monetary policy communication in order to orient public inflation expectations toward the future (forward looking) pursuant to the requirements of ITF. The main objective of strengthening the communication strategy is to help gradually lower and anchor public inflation expectations to the inflation target. In the Indonesian case, congruent with high inflation persistence and uncertainty, the effectiveness of achieving the final target of monetary policy with the overriding goal of price stability will depend on Bank Indonesia’s commitment and credibility in terms of achieving low and stable inflation within a given horizon. In addition, monetary policy is better when instituted consistently and by reducing the element of discretion. This is linked to the large “uncertainty cost” of monetary policy implementation. Therefore, the plan to implement ITF-based monetary policy should be seen as a means to improve central bank credibility in order to overcome persistent inflationary bias and reduce the cost of controlling inflation. Last but not least, in terms of ITF implementation, the institutional framework must be strengthened, primarily through policy coordination with the government, while aligning the spirit of independence supported by sound accountability. In addition, increased policy transparency requires the support of an effective communication strategy for all societal strata and relevant stakeholders. In this

318    Central Bank Policy case, the moral of the story is that macroeconomic stability should not be taken for granted but should be seen as the fruit of prudent and measured economic policy measures. An in-depth description of other institutional aspects, relating to independence, accountability and transparency as well as the monetary policy communication strategy is presented in Chapters 11 and 12.

10.6. Concluding Remarks This chapter has discussed the salient institutional aspects of the central bank, namely policy credibility and consistency. A credible central bank can form public expectations that the policies instituted will always be oriented toward achieving the final target and that the final target can be achieved. If credibility has taken a prolong period to attain and, therefore, the central bank’s reputation has already been built, monetary policy will be more effective, whether the response is through interest rates or controlling money supply. In fact, a reputable central bank and thus, strong public expectations, would be consistent with monetary policy and achievement of the final target, which could reduce and even replace the interest rate and monetary policy response. The capacity of the central bank to build policy credibility will depend on efforts to overcome several issues, namely time-inconsistent or sector-inconsistent policy as well as asymmetric information among economic agents, economic shocks, and even political uncertainty. Therefore, to build credibility, the central bank is recommended to base the policy response on a specific rule rather than discretion, build reputation through consistent policies and strengthen political and public support. This is no mean feat because the internal institutional arrangements of the central bank itself must be strengthened, including professionalism, governance, and strong leadership to ensure avowed commitment to the mandate in accordance with prevailing regulations. Furthermore, building public perception through transparency and a clear communication strategy is also crucial.

Chapter 11

Central Bank Independence and Accountability 11.1. Introduction Chapter 10 showed that the introduction of credibility through commitment and policy consistency will have a positive economic effect by steering the formation of public expectations. Based on public expectations, the central bank will be better equipped to boost policy effectiveness in terms of achieving price stability. Furthermore, it has been suggested that one effort to overcome the problem of policy credibility is by providing a clear mandate to conservative central bankers accompanied by full authority (independence) of central banking. In general, providing a clear mandate and independence from the principal (public) to the agent (central bank) tends to focus on the achievement of low and stable inflation. In this case, with a mandate clearly stipulated in law, the central bank is expected to enjoy independent monetary policy implementation to achieve the mandated goal. It is important to note, however, that in line with the delegation of authority, a mechanism is also ultimately required to guarantee monetary policy implementation is consistent with public expectations, which is achieved in the form of policy accountability and transparency by the central bank. Therefore, improving policy credibility at the central bank fundamentally relates to institutional aspects, such as policy independence and accountability as well as transparency and communication. This chapter discusses aspects of central bank independence and accountability in six sections. After this introduction, the second and third sections outline the conceptual dimensions and present a literature review as well as the theoretical models linked to the need for policy independence and accountability. The fourth section will highlight several critical issues relating to the institutional aspects, including the influence of independence on economic performance and the challenges of policy independence during crisis conditions. The fifth section provides a snapshot of policy independence and accountability in Indonesia, followed by the concluding remarks, which confirm the importance of policy credibility and growing need to interpret the important role of central bank independence when confronting the complex challenges faced by the central bank.

Central Bank Policy: Theory and Practice, 319–347 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191016

320    Central Bank Policy

11.2. Conceptual and Empirical Dimensions of Independence and Accountability 11.2.1. Central Bank Independence In general, central bank independence implies that the central bank has the freedom to determine and implement monetary policy in pursuance of prevailing regulations without political considerations or interference from a third party. In fact, the issue of independence is as old as central banking itself. David Ricardo (1871) advocated central bank independence and suggested that the central bank should not finance the government’s budget deficit. Considering that central bank independence is an integral part of achieving the overriding purpose of the central bank, central bank independence is often linked to institutional development and performance. Furthermore, the independence of a central bank is often tied to its ability to achieve the final target, low inflation for example. Nevertheless, the influence of independence of central bank, good and bad, remains a hot topic of debate. In practice, central bank independence is categorized differently by various experts. Fraser (1994), Debelle and Fischer (1995), and Meyer (2000) divided central bank independence into goal independence and instrument independence. Goal independence means the central bank itself sets the goals to be achieved, while instrument independence implies that the central bank has sufficient jurisdiction to implement its duties in order to achieve the goals set. On the other hand, Grilli, Masciandaro, and Tabellini (1991) and Elgie (1995) divided central bank independence into political independence and economic independence. Political independence implies that the central bank has the freedom to set its goals or policy decisions free from government interference, while economic independence means that the central bank is free to use all policy instruments at its disposal, without government restrictions, to achieve the goals.1 Baka (1994) and Mboweni (2000) specifically added another aspect of independence, namely financial independence, implying that the central bank has access to adequate financial resources and exerts full control over its budget. In general, financial independence is not always considered when assessing the variables of independence despite several studies evidencing a positive relationship between financial independence and central bank performance and policy credibility. Other categories have been developed, which generally contain nearly the same aspects, together with the advantages and disadvantages of each. Bofinger (2001) classified three categories as a synthesis of the above, namely: (1) goal independence, requiring that the government has no direct influence when setting the goals of monetary policy; (2) instrument independence, requiring that 1

Political independence also extends to nomination, recruitment, dismissal, qualifications, tenure, reappointment, and other officers than the governor, deputy governor, and members of the BOGs as well as the decision-making process, which encompasses decisionmaking, directives from the government, veto from government representatives, BOG remuneration, and central bank capital ownership.

Central Bank Independence and Accountability     321 the central bank can autonomously set the operational targets of monetary policy; and (3) personal independence, requiring that the decision-making body of the central bank exists without formal direction or informal pressure from the government.2 11.2.1.1. Goal Independence.  Fundamentally, the goal of monetary policy can be defined in broad terms, not only as the choice between price stability and output, but also the target horizon, the concrete indicators used, the value of the target and the escape clauses.3 Therefore, goal dependence is reflected by different conditions. In this case, a central bank could enjoy independence over the full scope of monetary policy goals, while another central bank may enjoy independence over several aspects only. The central bank of the United States, the Federal Reserve (the Fed), owns the highest degree of goal independence among central banks around the world. The mandate of the Fed is formulated in broad terms, therefore the work activities relate to economic growth and creating job opportunities. By mandating the achievement of several monetary policy goals for the Federal Reserve System and Federal Open Market Committee (FOMC) through prevailing laws, without allocating a specific weight for each policy goal, the Fed owns full independence to determine its preferences, for instance, whether to seek price stability or create more job opportunities. In contrast, the European Central Bank (ECB) is considered to have a more limited degree of goal independence, which is evidenced by the emphasis on price stability as the monetary policy goal mandated by law. Nevertheless, considering the formulation of goal achievement does not require a horizon, the ECB is free to choose whether to achieve the target in the near term or over a longer horizon. In terms of political aspects, prevailing regulations explicitly proscribe political interference from governments, parliaments, or the EU Commission in ECB task implementation. Such goal independent regulations are also found in other countries, including Japan and Sweden. Central banks with low goal independence include the UK, Canada, and New Zealand. In those countries, the overriding goal of monetary policy, as mandated by prevailing laws, is the achievement of price stability but the discrete value of the target is in the hands of the respective government. In New Zealand, price stability is defined in the Policy Target Agreements (PTA) negotiated periodically between the Minister of Finance and Central Bank Governor. In the UK, the law stipulates that the Minister of Finance is responsible for defining and publishing what is meant by price stability. In Canada, however, the central bank and federal government jointly announce the inflation-control target. Going beyond the degree of independence experienced by the central banks listed above, the literature also supports central banks that are not goal

2

Quantitatively identifying the degree of independence, in the form of an independence index, has been achieved in several studies, including Cukierman (1992) and Bofinger (2001). 3 A clause in a contract that specifies the conditions under which one party can be freed from an obligation.

322    Central Bank Policy independent (Bernanke et al., 1999; Fischer, 1995) because the final policy goal must fundamentally reflect the will of the people and, thus, should be set by officials elected representatively. Notwithstanding, Lohmann (1992) stressed that placing trust in elected officials could also prompt the risk of inflation bias because the government, or parliament, has a short-term orientation, hence its influence on monetary policy implementation could trigger shocks that exacerbate inflationary pressures. In addition, short-term supply-side shocks also encourage the government, or parliament, to tolerate deviation between target achievement and the medium-term target set. Consequently, there is a trade-off between monetary policy flexibility in response to supply-side shocks (with full commitment to achieving price stability) and the degree of independence (politically). Experience from the central bank of Germany, the Bundesbank, which is known as a particularly conservative central bank, showed that the risk of inflation bias was considered more relevant than the risk of an independent central bank that is unwilling to accommodate shortterm supply-side shocks. The Bundesbank, for example, prefers to respond to supply-side shocks more flexibly. In fact, the Bundesbank was open to an implicit inflation target, set at 2%, for a number of periods. 11.2.1.2. Instrument Independence.  Instrument independence implies that the central bank sets its own operational targets without government intervention. According to Bofinger (2001), independence has three important control elements as follows: (1) controlling short-term interest rates as the main operational target; (2) controlling the exchange rate as a supplementary operational target in the context of an open economy; and (3) credit restrictions from the central bank to the government, which could undermine control of the monetary base and, similarly, short-term interest rates. It is important to note that the first element assumes the propensity of the central bank is generally toward using short-term interest rates, rather than the monetary base (base money), as the main operational target. In practice, monetary policy implementation by the central bank can be directed autonomously to control interest rates. Nonetheless, as in the case of goal independence in several countries, such as the UK, Canada, and New Zealand, the government can intervene in the decisions taken by the central bank. With respect to controlling exchange rates, there are currently no central banks in the world with unrestricted authority to regulate exchange rates. The ECB owns broad jurisdiction, where prevailing laws have mandated authority to formally regulate the exchange rate system for members of the European Council (consisting of the Ministers of Finance of the member nations of the European Union), after consultation with the European Parliament. A more limited mandate would be given if a formal exchange rate system was not strived for, namely if the European Council could formulate general exchange rate policy orientation after receiving ECB recommendations. Differing from conditions at the ECB, other central banks generally receive limited exchange rate policy authority, which is typically mandated to the government. Concerning lending restrictions to the government, central bank laws generally regulate specific clauses. Despite the possibility of disbursing loans to the

Central Bank Independence and Accountability     323 government, depending on the economic conditions and political dynamics in each respective country, most central banks are restricted in terms of lending to the government because the impact could lead to accelerating inflation. In general, however, the central bank is guaranteed to achieve the monetary policy goals if politically independent and given the authority to decide when and how much to lend to the public sector. 11.2.1.3. Personal Independence.  Despite enjoying goal independence and/or instrument independence, a central bank may struggle to achieve personal independence because the government could exert informal pressures on monetary policy implementation. For example, the governor of the central bank could be removed at any time without a firm reason or the top of the central bank’s leadership structure may only enjoy the support of one or two individuals, thus the government could appoint an individual with the same common interests as the government. Therefore, observing the degree of personal independence at a central bank involves two main elements, namely the number of members serving on the highest decision-making forum (Board of Governors (BOG) for instance) and length of tenure of each respective member. More BOG members, or longer tenures, imply a greater degree of independence from government intervention. The following table presents the conditions of personal independence at several central banks based on the two elements mentioned. In general, compared to other central banks, the US Federal Reserve, and ECB benefit from the greatest degree of independence. In contrast, the Reserve Bank of New Zealand (RBNZ) has the lowest degree of independence. Empirically, there are several different methods to measure the level of independence at a central bank in comparison to other central banks and over time. Experts have identified and developed independence indicators used to subsequently measure the degree of independence at a central bank. For example, Parkin (1978, 1987) developed an approach to differentiate the various types of central bank independence. Burdekin, Clas, and Thomas (1992) as well as Masciandaro and Spinelli (1994) developed an approach to rank independence. Meanwhile, Cukierman, Webb, and Neyapti (1992) developed a precise measure of central bank independence, using 16 indicators grouped into eight indicator sets, which were weighted between 0.20 for the four indicator sets concerning the government and 0.05 for the indicator sets containing one indicator regarding loans disbursed by the central bank. The method to measure the degree of personal independence employed by Robert Elgie (1995), who refined the method of Cukierman et al., shall be explored in more depth. Elgie used 29 indicators of political independence, compared to just five used by Cukierman et al. Furthermore, Elgie also used seven indicators for the governor, compared to just four per Cukierman et al. Finally, Elgie used five indicators for the internal decision-making process, eight indicators for the deputy governor and nine for the BOGs. Concerning economic independence, Elgie only used seven indicators, while Cukierman et al. used 11. Eight of the 11 were linked to central bank restrictions on lending to the government, however, which Elgie combined into just one indicator. In general, Elgie developed a group of economic independence indicators

324    Central Bank Policy Table 11.1:  Central Bank Personal Independence. Central Bank

Highest Decision-making Forum

Europe

Governing Council

Total Members

Tenure

6 Executive Board Min 5 years: Central Members, 11 Central Bank Governors from Bank Governors from member states member states Min 8 years: Executive Board Members United States Federal 7 BOG members, 5 7 years: BOG Open Market Presidents of Federal members Committee Reserve Banks United Monetary Policy 1 Governor, 2 Deputy 5 years: Governor and Kingdom Committee Governors, 6 special Deputy Governors members 3 years: Directors Canada Executive 1 Governor, 1 Deputy 5 years: Governor and Committee Governor, 5 Directors Deputy Governor (Governing 3 years: Directors Council) Japan Policy Board 1 Governor, 2 Deputy 5 years Governors, 6 special members New Zealand Governor Governor 5 years Source: Bofinger (2001).

that provided a more comprehensive picture. Moreover, the refined methodology consists of far more comprehensive political and economic independence indicators.

11.2.2. Central Bank Accountability Congruent with the degree of independence enjoyed, in terms of increasing credibility, a central bank must be held publicly accountable and clarify monetary policy implementation. In general, accountability can be observed from two aspects. First, from the single goal that makes the central bank more trustworthy because it clarifies the measures and accountability. Second, the appointment process and short tenure of the central bank governor as well as possible reappointment could improve accountability because it would allow for performance appraisals. In practice, central bank accountability could also be observed from the definition of accountability per de Haan, Amtenbrink, and Eijffinger (1999). The following table presents a comparison of central bank accountability in

Central Bank Independence and Accountability     325 several countries based on at least 13 aspects linked to three characteristics of accountability, namely the final policy goal, ultimate accountability, and political transparency. From Table 11.2, we can see that, if each aspect with a “Yes” condition is given a weight of 1 (one), it could be concluded that the Bank of England has the highest degree of accountability. Legally, the ECB is seen to have the lowest degree of accountability. It is important to note that, in practice, for several aspects (3, 4, and 11), the ECB has applied principles far beyond those mandated by central banking laws. Consequently, in practice, the degree of ECB accountability is higher than just based on prevailing laws. Per Issing (1999), however, such conditions do not make the ECB “among the most transparent and accountable central banks in the world.”

11.2.3. Correlation between Independence and Accountability Nowadays, people believe that central bank independence could encourage price stability. Therefore, achieving central bank independence is not the final goal but a means to achieve the policy objectives stipulated in law. Providing a mandate that requires accountability implies that the central bank must apply sound accountability principles to the policy-making process, meaning the central bank must inform and justify several major aspects, such as the progress of achieving the policy goals, the monetary policy strategy and current policy measures. How far, therefore, can an independent central bank meet those requirements? Answering that question, de Haan, Amtenbrink, and Eijffinger (1999) conducted an empirical study linking central bank independence and accountability in advanced countries. An accountability index was created by scoring each of the 13 aspects covering the three characteristics of accountability, namely the final goal, ultimate accountability, and transparency. Meanwhile, an accountability index was also developed based on the assessments performed by Eijffinger and Schaling (1993) as well as Eijffinger and Van Kuelen (1995), per three aspects, namely the weight of holding the ultimate accountability for monetary policy (central bank/government/central bank and government), the presence of government representatives on the BOGs (yes/no), as well as the fraction of board members appointed by the government (more/less than half). The total independence and accountability index is presented in Table 11.3. Based on the data, the linkages were assessed using a regression approach between the independence index and each respective accountability characteristic index and total accountability index. The assessment produced several observations as follows: (1) There is a negative correlation between independence and two characteristics of accountability, namely ultimate accountability and transparency. (2) There is a positive but insignificant correlation between independence and ultimate accountability. (3) In general, there is a negative correlation between independence and accountability, albeit not as strong as the respective correlation in point (1).

Source: de Haan et al. (1999).

  (1)  Stipulated in the central banking act  (2) Goal prioritization   (3)  Clear goal definition   (4)  Goal quantification (per prevailing laws) (B)  Ultimate Accountability   (5)  Monitored by parliament   (6)  Government right to provide instructions   (7)  Procedures to implement a policy rejection mechanism   (8)  Possibility to request government instruction   (9)  Possibility of amending laws by parliament majority (10) Using past performance to consider removal of central bank governor from office (C)  Political Transparency (11) Required to publish a monetary policy report (inflation report) (12)  Publishes minutes of the Board meeting (13) Required to provide public clarification if the target is achieved

(A)  Final Policy Goal

Accountability Aspects

Table 11.2:  Central Bank Accountability.

Yes – – – Yes – Yes – Yes –

Yes Yes Yes

Yes – – – – –

– (Yes) – Yes

United States

Yes Yes – (Yes) – (Yes)

Europe

Yes Yes

Yes

Yes Yes Yes – Yes –

Yes Yes Yes Yes

United Kingdom

– Yes

Yes

Yes Yes Yes – Yes –

Yes – – –

Canada

– Yes

–

Yes Yes Yes – Yes –

Yes – – –

Japan

326    Central Bank Policy

Central Bank Independence and Accountability     327 Table 11.3:  Central Bank Independence and Accountability Index. Country

Independence

Accountability Final Goal

Australia Belgium Canada Denmark France Germany Italy Japan Netherlands New Zealand Spain Sweden Switzerland UK US ECB

1 3 1 4 4 5 2 3 4 3 3 2 5 3 3 5

1 0 1 1 3 2 0 1 1 4 3 0 1 4 1 3

Ultimate Transparency Accountability 5 3 4 2 2 1 3 4 4 4 2 3 1 4 2 0

1 0 2 1 1 0 1 1 0 2 2 2 0 3 3 1

Source: de Haan et al. (1999).

In general, the results showed an inverse correlation between central bank independence and accountability. Nevertheless, the results do not imply that it would be impossible to simultaneously achieve independence and accountability. The general belief at central banks that independence and accountability can be strived for together is also very strong, in line with the prevailing opinion that accountability, in terms of democratic accountability, should not be seen as a mechanism that limits independence at a central bank; but should be seen as a mechanism to reinforce the legitimacy of central bank independence. In other words, even if it is difficult to consider as a requirement, accountability should be considered complementary to independence.

11.3. Theoretical Models Concerning Independence and Accountability Several studies have explored the theoretical evidence and conducted empirical testing regarding the importance of institutional development in terms of central bank policy implementation. Nonetheless, in contrast to the earlier and already established studies that focused on aspects of independence, studies on aspects of

328    Central Bank Policy accountability remain at a relatively embryonic stage.4 The following section presents the evolution of theory concerning independence, or delegation of authority, and central bank accountability.

11.3.1. Delegation of Authority Theory: Conservative Agent In one of his seminal pieces, Kenneth Rogoff (1985) stated that an independent and inflation-averse central bank would tend to lower average inflation but increase output variability. In other words, a conservative central bank could reduce inflation bias caused by the time inconsistency problem but also play a more limited role in terms of output stability. This was explained through a model popularized by Barro and Gordon (1983), which can generally be developed as follows. Output performance (y) is formulated based on the supply curve specified à la Lucas:

yt = a( πt − πte ) + εt (1)

where πt is inflation, πte is inflation expectations, and εt is independent and identically distributed (iid) shocks with an average of 0 and variance σε2. a reflects the influence of fallacies on output. Expectations are formed before the presence of shocks and before policymaker determine πt. The policymakers’ objective function is:

Lt = πt2 + b( yt − y*)2(2)

where, b > 0 and the natural level of output, y* > 0. Substituting (1) for (2) produces:

Lt = πt2 + b( a ( πt − πte ) + εt − y*)2 (2’) The first-order condition minimizes the loss function as follows:



4

 aby *   a 2 b  e  ab   π − πt =  +  ε (3) 1+ a 2 b  1 + a 2 b  t 1+ a 2 b  t

After the seminal theoretical contribution of the “conservative central bank” (Rogoff, 1985), the literature on central bank independence was enriched by other studies, including “insulation from political uncertainty” (Alesina & Gatti, 1995) and “optimal contract” (Walsh, 1995). Meanwhile, a prominent theoretical argument on accountability is contained in “democratic accountability” (Eijffinger, Hoeberichts, & Schaling, 1998).

Central Bank Independence and Accountability     329 Assuming πte = Et−1 ( πt ), produces:

Et−1 ( πt ) = aby* > 0(4)



 ab  πt = aby * −  ε (5) 1+ a 2 b  t



 1  yt =   ε (6) 1+ a 2 b  t

Policy rule (5) contains the component known as inflation bias aby* and component of stabilization (ab/(1+a2b))ε. From equations (4)–(6), the following unconditional moments are obtained:

E ( πt ) = aby* ⇒

∂ E ( πt ) > 0(7) ∂b



E ( yt ) = 0 (8)



 ab  ∂ var( πt ) > 0(9) var( πt ) =   ⇒ 1 + a 2 b  ∂b



 1  ∂ var( yt ) var( yt ) =  < 0 (10)  ⇒ 1+ a 2 b  ∂b

2

2

From the results above, it could be concluded that monetary policy based on discretion would cause high inflation without stimulating output beyond its natural level, also known as stagflation. In addition, there is a stabilization trade-off, where a smaller weight of output stabilization (b) would lead to lower and more stable inflation, but accompanied by output volatility. In this case, inflation bias could be overcome through future (ex ante) commitment from policymaker to abide by an optimal policy rule, depending on the realization of ε. Realistically, such commitment is not a choice available to policymaker due to difficulties identifying the characteristics of the shocks, thus efforts to eliminate market distortions and frictions could incur a large cost. Therefore, to approach optimal social results, as if there were commitment from the policymaker, the practical solution is the delegation of authority. In this case, Rogoff suggested that public prosperity could be improved if the policymaker delegate the (ex ante) choice of monetary policy to a previously selected independent agent. Independent, in this case, means that the agent cannot be removed ex post, when determining the policy choice. The policymaker could, however, select an agent with different preferences (public preferences), reflected in parameter b. System behavior is each respective period is as follows: First, the policymaker will select an agent from the population with parameter bt. then, after the expectations have been formed and the shocks have occurred, the agent determines policy.

330    Central Bank Policy Meanwhile, the agent’s preferences are formulated as follows: Lt = πt2 + bˆi ( yt − y*)2(11)



If a certain agent is selected with parameter bj, then inflation and output shall form as follows:

 ˆ  ˆ * − ab  ε (12) πˆ t = aby t  2 1+ a bˆ 



 1   ε (13) yˆt =  1+ a 2 bˆ  t

In each subsequent period, the selected agent may be different but because the behavior in each period is identical, the optimal choice of the policymaker (and the public, if aligned) is as follows:

Min E ( L ( b, bˆ )) = E ( πˆ t2 + b( yˆt − y*)2 )(14)



2 2   abˆ    1     ˆ          ε − y * (15) = E aby * −  ε  + b     1 + a 2 bˆ  t 1 + a 2 bˆ  t   



2  abˆ   1  2 2 2  ˆ     = ( aby*) +  σε + by *2 (16)  σ + b  1 + a 2 bˆ  ε 1 + a 2 bˆ 

bˆ

2

Equation (15) shows that when selected, the agent will follow rule (5) with preference parameter b. Considering the policymaker acts earlier, before πte and εt have been realized, the policymaker sets expectations based on the shocks that have occurred. The first-order condition of the solution to the minimization of the loss function problem is as follows:

 y * ∂L = 0 ⇒  2  (1 + a 2 bˆ )3 bˆ = b − bˆ(17)  σε  ∂bˆ

Such conditions show the presence of parameter b in the search for solution f(b) = b − b, which can be represented graphically as shown in Fig. 11.1 (Walsh, 2001): The solution is therefore:

0 < bˆ < b (18)

Central Bank Independence and Accountability     331

Fig. 11.1:  Output Stabilization Preference Behavior.

From the solution, we can conclude that the policymaker could improve welfare (and therefore public welfare if aligned) by delegating monetary policy to a more inflation-averse or conservative agent than the policymaker itself. By analogizing that the policymaker is represented by the government or parliament, and the agent is represented by the decisionmaker and accountability of the central bank, the fundamental implications of the delegation of authority theory are as follows. In comparison to countries where the central bank is not independent, independent central banks will experience: (1) lower and more stable average inflation; and (2) higher output volatility as a consequence of efforts to control inflation, but average output will not change.

11.3.2. Central Bank Accountability Theory: Democratic Accountability In reality, the delegation of authority, in terms of monetary control, by the government to the central bank is not without risk. Therefore, there should be a mechanism to guarantee that monetary policy implementation by the central bank is aligned with the desires/hopes of the public. In other words, the central bank must demonstrate sound accountability to the institutions democratically elected by the public. How to understand the meaning of accountability in a democratic society? In a study on the issue of accountability, de Haan et al. (1999) differentiated three main characteristics in the meaning of central bank accountability, namely: (1) Who takes the decision on the final goal of monetary policy? (2) Who is ultimately accountable for monetary policy implementation? (3) How politically

332    Central Bank Policy transparent is the monetary policy?5 Eijffinger et al. (1998) developed a theoretical model concerning central bank accountability, focusing their analysis on two characteristics of accountability, namely political transparency and ultimate accountability.6 Meanwhile, the issue of determining the final goal is considered in relation to the goal independence of the central bank. The theoretical model of accountability was developed based on the studies by Lohmann (1992) as well as Nolan and Schaling (1996). The model scenarios are as follows: (1) The government has delegated monetary policy implementation to a conservative central bank. Nevertheless, the government and public are not exactly sure the central bank’s relative preferences to price stabilization and output. (2) Subsequently, after the central bank has proposed its inflation preferences, the government could decide to review or reject the proposal, thus incurring a certain cost and, therefore, the government could set the inflation target. In this case, the central bank is considered partially independent. The cost of rejecting the proposal relates to the question of who is most accountable for monetary policy implementation. If the government is unwilling to bear the cost, the central bank is considered the most accountable for monetary policy implementation. In contrast, if the cost is considered small, the government is held accountable for monetary policy implementation. In concrete terms, the theoretical model developed can be expressed in the following way, with output determined by the supply function à la Lucas and simplified as follows:

5

y = π − π e + ε, dengan

ε ~ N (0, σε2 ) (19)

Geraats (2000) differentiated five aspects of transparency that support accountability, namely political, procedural, policy, operational, and economic transparency. Political transparency encompasses formal goals, quantitative targets, and clarification of institutional structure. Furthermore, political transparency is the most important aspect because it provides the criteria and identifies who is accountable. Economic, procedural, and policy transparency are required to understand the background/motivation behind the policies instituted, while operational transparency is required to comprehend the constraints to successful policy achievement. 6 Accountability and transparency are inextricably linked. The central bank must be held accountable for task implementation to its stakeholders. A more transparent institution shall also increase central bank accountability and, ultimately, boost central bank performance through greater market efficiency and increased decision-making clarity (Poole, 1999, 2003). Meanwhile, better performance would improve accountability at the institution/central bank involved. Nevertheless, transparency is a necessary condition for accountability but not a sufficient condition because accountability is also determined in terms of monetary policy implementation.

Central Bank Independence and Accountability     333 The objective function of the government is:

1 1 LG = π 2 + ( y − y*)2 + δc (20) 2 2

where y* > 0 is the government’s target output. If the government rejects the inflation target proposed by the central bank, the government will bear the cost, c, determined by the prevailing political institutions. δ is a dummy variable with a value of 1 if the government rejects the central bank’s proposal and 0 if the government accepts the central bank’s proposal. Meanwhile, the central bank has random (stochastic) preferences, where the degree of conservatism is calculated in the contract (quadratic function) with parameter f. In addition, the central bank has private information concerning the realization of shocks in its preferences, with a uniform distribution. Therefore, the loss function of the central bank is formulated as follows:

LCB =

1− x 2 1 f π + ( y − y*)2 + π 2 , (21) 2 2 2

with x ∼ U[−h, h] and h < f. Without delegating monetary policy, the government sets (with discretion) an inflation target that minimizes the loss function, namely: πG =



y * +π e − ε (22) 2

If monetary policy has been delegated to the central bank, the inflation target is set based on minimizing the loss function of the central bank, namely: πCB =



y * +π e − ε (23) 2−x + f

Considering that the central bank is more conservative, or inflation-averse, than the government (f − x > 0), the policy preferences of the central bank would feed through to lower inflation bias than from the government but, on the other hand, would also be less active in response to supply-side shocks. If the government decided to reject the inflation target proposed by the central bank (23), however, the government would then use its own inflation target (22). Therefore, the loss borne by the government would be (δ = 1):

1 LG ( πG ) = ( y * +π e − ε )2 + c (24) 4

Meanwhile, if the government decides to accept the central bank’s proposed inflation target, the loss function would therefore be (δ = 0):

LG ( πCB ) =

1 + ( x − 1− f )2 ( y * +π e − ε )2 (25) 2( x − 2 − f )2

334    Central Bank Policy The two choices show whether the government rejects or accepts the central bank’s proposal, namely if: LG ( πG ) < LG ( πCB )(26)



If the government decides the cost of rejecting the central bank’s proposal outweighs the benefits of the inflation target set by the government, the central bank will be independent. By substituting equations (24)–(26), we can conclude that the central bank will be independent if: c>



( x − f )2 ( y * +π e − ε )2 (27) 4( x − 2 − f )2

Therefore, depending on shock realization, the central bank would face two conditions upon acting, namely independence ((x, e) ∈ I) or accommodative ((x, e) ∈ A). If the central bank is accommodative, the inflation target would represent the weighted average of inflation based on the government’s preference and inflation based on the central bank’s preference, as follows:

πACC = ϕπG + (1− ϕ )πCB =

2 − ϕ( x − f ) ( y * +π e − ε )2 (28) 2(2 − x + f )

Accordingly, substituting equation (28) for (20) and using equation (19) shows that the loss incurred by the government is as follows:

L ( πACC ) =

1+ ( x −1− f )2 − ϕ ( x − f )2 (1− 21 ϕ ) ( y * +π e − ε )2 (29) 2(2 − x + f )2

The central bank will always take accommodative actions; hence the government is indifferent to rejecting the central bank’s proposal. Therefore, the central bank will have φ, thus LG ( πACC ) < LG ( πG ). By combining equations (20) and (29):

ϕ = 1−

2(2 − x + f ) c , if ( x, ε ) ∈ A , and φ = 0 if ( x, ε ) ∈ I (30) | y * +π e − ε || x − f |

By substituting equation (20) for (28), we obtain a rate of inflation if the central bank implements accommodative measures, namely:

πACC =

y * +π e − ε − c sgn( f − x )sgn( y * +π e − ε )(31) 2

Assuming that the central bank will always act conservatively irrespective of the shocks occurring (f > x), then equation (31) could be expressed as follows:

πACC =

y * +π e − ε − c sgn( y * +π e − ε )(32) 2

Central Bank Independence and Accountability     335

Fig. 11.2:  Monetary Policy Behavior (c > 0). Based on that interpretation, the actual behavior of monetary policy, with the rejection cost, can be explained by Fig. 11.2. We can explain that the area around y* + πe = ε is the area where the central bank is independent in terms of setting the inflation target per its own preferences. Outside of that area is where the central bank has to accommodate an inflation target based on the preferences of the government. In this area, the government sees that the sacrifice ratio of stabilization measures due to supply-side shocks is large, therefore an inflation target set by the central bank would be considered insufficient. Parallel to the government’s reaction function are two reaction functions that occur due to the cost of rejecting the central bank’s inflation proposal. The intersection between that reaction function and the reaction function of the central bank determines the area of independence. Based on such behavior, we can observe the implications of one of the characteristics of accountability, namely responsibility, on economic developments, particularly the impact on inflation and shock stabilization.7 In this case, one way to improve central bank accountability is to delegate (technically, a shift in direction) the burden of responsibility for monetary policy from the central bank to the government. In the theoretical model, this can be achieved by reducing the rejection cost (c). Fig. 11.3 shows that the distance between the government’s reaction function and two parallel lines is narrowing and, consequently, the area of central bank independence is also narrower. We can see that a lower rejection cost: (1) 7

A more in-depth discussion on the aspects of transparency is presented in the following chapter on Policy Transparency and Communication Strategy.

336    Central Bank Policy

Fig. 11.3:  Monetary Policy Behavior after Lowering c. undermines central bank independence; (2) raises inflation expectations; and (3) increases stabilization to shocks. Therefore, increasing accountability by delegating the burden of responsibility of monetary policy from the central bank to the government is especially applicable to countries suffering from a relatively serious flexibility problem (high σε2) compared to a credibility problem (high y*). Meanwhile, another aspect of accountability, namely transparency, can be explained by its correlation with the element of uncertainty in the behavior (preferences) of the central bank. In this case, increasing the degree of transparency at a central bank relates closely to reducing the degree of uncertainty in the central bank’s preferences (lowering the value of h in equation (21)). Lowering the degree of uncertainty would make the central bank more conservative, as reflected in the reaction function of the central bank, which becomes flatter, therefore, central bank transparency would also lower inflation expectations. Another consequence of lowering the degree of uncertainty (lowering the value of h) would be to reduce the accommodative nature of the central bank to efforts to stabilize shocks (Fig. 11.4).8 Therefore, increasing accountability by delegating the burden of responsibility for monetary policy from the central bank to the government is especially applicable to countries suffering from a relatively serious credibility problem (high y*) compared to a flexibility problem (high σE2). It is important to note that although the area of independence narrows, increasing accountability through greater transparency does not effectively undermine central bank independence because greater transparency ensures the macroeconomic benefits obtained by the central bank are aligned with the central bank’s preferences.

8

Mathematically, the reduction can be seen in Eijffinger and Hoeberichts (2000).

Central Bank Independence and Accountability     337

Fig. 11.4:  Monetary Policy Behavior after Lowering h.

11.4. Empirical Studies and Related Issues 11.4.1. Central Bank Independence and Economic Performance There are numerous studies on central bank independence that support or reject the benefits of central bank independence based on different arguments. Those who support the benefits of independence are based on the evidence that: (1) Greater central bank independence leads to lower and more stable inflation, which can drive long-term economic growth. (2) Greater central bank independence leads to a lower long-term budget deficit due to a separation of authority between the printing and spending of money. Those who reject the benefits of independence are based on the evidence that: (1) Although there is a correlation between independence and low inflation, it does not mean that through greater independence, lower inflation will be achieved. (2) Monetary policy is a part of overall economic policy, therefore, there is no reason to separate fiscal policy, monetary policy, trade policy, and other policies. (3) If the central bank officers are not appointed democratically, then the decisions taken on interest rates, exchange rates, inflation and other monetary issues are not representing the interests of the public. Consequently, independence could create an abuse of authority.

338    Central Bank Policy Regardless of the differing opinions, the reality is that more and more central banks in advanced countries and developing countries are becoming independent. Several studies have been conducted to observe the empirical linkages between independence and economic performance achieved by the central bank, including Grilli et al. (1991), Cukierman (1992), Alesina and Summers (1993), and Bofinger (2001). Grilli et al. (1991) concluded that political and economic independence have an inverse correlation to inflation. Meanwhile, several other forms of independence correlate but to a varying degree, implying that by giving the central bank independence, lower inflation could be achieved. Grilli et al. also concluded that political and economic independence do not have a significant correlation with real output growth. In general, therefore, central bank independence could result in stronger economic performance, in the form of lower inflation, without sacrificing economic growth. While that conclusion has been corroborated by a number of other studies, the research conducted by Alesina and Summers (1993) produced a rather different conclusion. Alesina and Summers concluded that there was a significant inverse correlation between central bank independence and average inflation in western industrialized nations from 1955 to 1988. Nevertheless, no symmetrical correlation was found between central bank independence and average economic growth. It is important to note that the study used relatively old data (excluding the 1990s). Bofinger (2001) performed a similar study using an independence index based on three elements, namely goal, instrument, and personal independence.9 That study produced a conflicting conclusion to the previous research, namely that central bank independence did not correlate with the achievement of low inflation in advanced countries during the period from 1996 to 2000. The surprising results were linked to the comparatively brief observation period, thus demanding some caution. First, central banking laws that emphasis aspects of independence were first promulgated in the middle of the 1990s, therefore, observations and comparisons between central banks over a longer horizon are not possible. Second, the period from 1996 to 2000 was a time when central bank policy around the world was oriented toward achieving price stability and, thus, is often considered the base period to start comparisons of central bank performance. The conceptual problem of conflicting conclusions in the existing empirical research on the linkages between independence and inflation fundamentally lies in the condition that central bank independence is ideally required in situations where the government strives to prioritize inflationary policy in order to reduce unemployment. Therefore, a politically independent central bank will only create inflation bias if:

9

It is important to note that the independence index can be estimated based on several criteria (elements of independence) by calculating the weight of each respective criterion. Subjectively, therefore, using an independence index can be different between different studies.

Central Bank Independence and Accountability     339 (1) The economy is facing a serious unemployment problem. (2) The inflation problem is not taken seriously enough. (3) There is a short-term political agenda to maintain power in a subsequent term. In other words, countries that maintain a low level of unemployment, prefer price stability, and enjoy political stability should have a negligible inflation bias, even with a central bank that is not independent. Such conditions were personified by the economic cases of Japan and France in the 1980s and 1990s, when policy was oriented toward price stability, while neither central bank was independent until 1993 (Banque de France) and 1998 (Bank of Japan). With the current policy preference oriented toward price stability, the implication is that central bank independence has become less critical. Nevertheless, the empirical evidence showed a potential inflation spike during periods of low inflation, for instance during the 1950s and 1960s. Therefore, mandating central bank independence is considered a strategic preventative measure for sound monetary policy implementation.

11.4.2. Related Issues 11.4.2.1. The Critical Issue of Central Bank Independence: Two Fallacies.  There are at least two conclusions fundamental to the spectrum of studies on central bank independence, including the works of Rogoff (1985), Persson and Tabellini (1993), Alesina and Gatti (1995), and Walsh (1995). First, the lack of (pre) commitment in policy implementation by an independent central bank would not avoid discretionary policy that leads to inflation bias. Second, a contract submitted by the government to the central bank could overcome the time inconsistency problem, which is the root of inflation bias. The conclusions above received criticism from McCallum (1995), who showed two fallacies in the interpretations, the first being inconsistencies between the analysis and reality. In this case, the range of studies on central bank independence tended to underestimate the sound monetary policy achievements of an independent central bank. In addition, there were general errors when interpreting the beneficial effects of the contracts made by the central bank. Several studies have concluded, however, that when determining the optimal policy response, the central bank is faced with two alternative choices in the form of a solution based on discretion and a solution based on a rule.10 According to McCallum, there is no justification for the opinion which assumes that, with commitment, an independent central bank will tend to act based on discretion rather

10

Compared to setting inflation on a period-by-period basis or based on discretion, the central bank bases inflation on a pre-existing policy rule when the inflation expectations are formed.

340    Central Bank Policy than a policy rule.11 In practice, there are no meaningful impediments to a central bank acting based on a policy rule, which facilitates better results. Furthermore, even without pre-commitment as the basis of assured future measures, this does not mean that the policy measures instituted by the central bank would not be feasible for implementation. Therefore, McCallum continued to recommend the adoption of a policy rule by central banks. It is important for central banks to avoid inflation bias by believing that the continuous exploitation of expectations formed in the current period (but reflecting the response to central bank measures taken in the previous period) is not beneficial, and that the goal of the central bank will be better achieved if the central bank ignores the expectations, which are temporary anyway.12 In other words, if the central bank applies a rule, it is important to believe, after behaving differently in the past, that expectations are never immediately aligned with actual conditions. A lag is required to align the expectations to the rule applied. Therefore, an independent central bank must implement the rule consistently, considering that after several periods, expectations will align with actual conditions pursuant to the assumption of rational expectations. Operationally, emphasizing such central bank measures must refer to the long-term characteristics of each policy procedure. Similarly, studies on the contracts between the government and central bank have concluded that if the government provides the contract (incentives) to the central bank, where remuneration depends upon attaining the inflation target, optimal central bank performance may be achieved because it is achieving the solution based on a rule. Nevertheless, McCallum deemed this to be the second fallacy of studies on independence considering that several aspects relating to the study outcomes fundamentally fail to ignore the motivation for inconsistency, only linking that problem to the contemporaneous presence of the government and central bank. Specifically, there is an alternative case, where the government could stipulate in the contract that the government will reduce the central bank’s budget if inflation becomes too high, for example. Under such conditions, as it is relatively difficult for the central bank to avoid inflationary measures, the government also has an incentive to avoid this alternative. In other words, if the lack of precommitment is considered a serious problem, this should apply to an institution that encompasses the contemporaneous presence of the government and central bank. Furthermore, the problem faced will fundamentally not be overcome (analytically) by merely recommending a specification for the central bank’s objective function to the constitutional phase of the political process. The problem requires

11

The literature (Alesina & Tabellini, 1987; Chari et al., 1989; Debelle & Fischer, 1995) states that “there is no pre-commitment technology available to the unconstrained central bank.” This is due to the view that central bank behavior is in line with discretion. 12 Expectations relate to output formation behavior (supply function), which mirrors the expectational Phillips Curve.

Central Bank Independence and Accountability     341 enforcement, thus the constitution must be effective for the government and central bank. 11.4.2.2. Ex Post and Ex Ante Accountability: The Difference between Accountability and Transparency.  From the perspective of policy implementation timing, central bank accountability can fundamentally be debated in two forms, namely ex post and ex ante accountability (Bofinger, 2001). Ex post accountability implies that the central bank must provide ex post justification if the policy target is not achieved. Meanwhile, ex ante accountability implies that the central bank must justify deviations in the forecast from the target to be achieved. To that end, the central bank must be forward looking. In terms of ex post accountability, a scenario of successful monetary policy implementation in response to demand and supply-side shocks is illustrated in Fig. 11.5. We can now see that (ex post) accountability implies that the central bank must justify why it has been unable to avoid the adverse aspects, namely: (1) positive/negative demand-side shocks relating to over/undershooting of the inflation target; and (2) positive/negative supply-side shocks while achieving price stability but relating to a positive/negative output gap. Regarding supply-side shocks, in terms of over/undershooting the inflation target and a negative (positive) output gap, there is an assumption that the central

Fig. 11.5:  Ex Post Accountability and Central Bank Reaction. Source: Bofinger (2001).

342    Central Bank Policy bank reacts inappropriately but the assessment depends on the size of the output gap and inflation deviation from its target. It is important to add that of the reactions available to the central bank, the “Yes” reaction displayed in the box represents a monetary policy reaction considered successful. A similar analysis was performed for ex ante accountability but the indicators used in this case were the inflation forecast and output gap. Notwithstanding, as the indicator represents a forecasted value, the analysis is more complicated. Moreover, considering the information contained in the forecast may also be provided by parties external to the central bank, if the public considers the central bank lacking objectivity when announcing the forecast, the public may instead trust the forecasts provided by external parties. Such public distrust in the central bank’s ability to provide accurate forecasts directly relates to another institutional aspect of the central bank, namely transparency. Is the view that accountability may be ventured through transparency valid? Castellani (2002) showed that, fundamentally, there is a difference between accountability and transparency. Despite evidence that transparency plays a role in the accountability problem, the traditional “accountability through transparency” approach is considered incomplete. Several approaches, which focus on transparency as a precondition for accountable monetary policy implementation, are generally based on formal accountability concepts. The fact of the matter is that such approaches tend to fail in terms of clearly differentiating accountability and transparency. In his study, Castellani clarified the difference, namely that accountability relates more to policy implementation (deeds), while transparency relates more to the communication strategy adopted by the central bank (words). In other words, accountability can be represented as ex post political intervention relating to government observations of monetary policy implementation. Meanwhile, transparency can be described as the ex ante decisions of the central bank in relation to the communication strategy. This is congruent with the view of Duisenberg (1999), namely that the publication of forecasts does not contribute to increase central bank accountability. In this case, central bank performance in terms of controlling inflation in the medium term is the main factor considered by the public when assessing the success, and therefore accountability, of the central bank.

11.4.3. Policy Independence and Credibility during a Crisis Period The independence and credibility enjoyed by a central bank EBB and flow in line with the dips and gyrations of economic and non-economic factors. ITF implementation at the beginning of the 1990s was considered an underlying factor of increasing central bank independence and policy credibility in numerous countries. Meanwhile, the GFC forced many central banks to intervene as lender of last resort and institute policy to maintain financial stability. That broader mandate demanded close policy coordination between the central bank and other policy authorities, especially the fiscal authority. In addition to the benefits of harmonious policy coordination, several observers have opined that the situation precipitated a policymaking compromise, which influenced the independence and

Central Bank Independence and Accountability     343 policy credibility of the central bank and could also spur a more serious issue if the coordination hampered mandate implementation by the central bank to credibly achieve low inflation. Would intense coordination or cooperation between the monetary and fiscal policy authorities trigger a compromise and, therefore, undermine central bank independence and policy credibility? Under normal conditions, there is a clear separation between the domains of central bank monetary policy and government fiscal policy. Nevertheless, the opposite can occur under crisis conditions. According to Blinder (2012), this can occur because of various changes under a crisis situation. First is the time horizon. Under crisis conditions, when the central bank is faced with rapidly changing financial sector dynamics, the time horizon shrinks from the medium-long term to align with the shorter time horizon faced by the government. Consequently, such conditions have a propitious impact as cooperation becomes more natural. Second relates to a shift in the objectives of the central bank, namely that it is no longer (just) combatting inflation because that could be counterproductive if the economy continues to slump; but to be jointly responsible for maintaining financial system stability. A combination of urgency and the different interests between the central bank and government requires coordination even though the central bank and government have strong differences in terms of process implementation, capabilities, legal aspects, and institutional arrangements. With the comparative advantages enjoyed by each respective institution, policy coordination between the two could better overcome the economic problems. Third, policy coordination is critical to restore order to financial markets blighted by panic, not only operationally but also politically and psychologically. Although the central bank has the means and capability to implement policy, government support is required to ensure political legitimacy. Harmonious policy coordination between the central bank and government, and possibly other relevant policy authorities, would not only pacify market psychology but also the legislature. Fourth, policy coordination is critical when deciding which troubled financial institutions to resolve; namely which financial institutions should be allowed to survive through bail-out funds and which should be left to close. In this case, political legitimacy in the joint decision-making process is vital. If the problem is politicized, there is a distinct possibility that the central bank or government would have to be available to take the unenviable consequences. Therefore, the different comparative advantages of the central bank and government must be exploited optimally. Finally, per the worst-case scenario, under serious crisis conditions that threaten the solvency of the central bank’s balance sheet, with potentially massive financial losses expected, the government is expected to stand firm in its backing of the central bank. We can, therefore, conclude that, under crisis conditions, central bank policy independence could be interpreted narrowly as impossible and undesirable. Consequently, independence must encompass the broader interests through policy coordination. Nonetheless, after the crisis period has been overcome, the central

344    Central Bank Policy bank must prudently extricate itself from the policy compromise with the government and gradually rebuild its independence, the pace of which shall depend on the prevailing circumstances. If financial system stability has been substantially restored, the central bank must reorient monetary policy from financial system stability toward policy independence and credibility. Therefore, similar to policy independence, policy credibility must also encompass the broader interests. In this case, anti-inflation policy credibility is (only) one aspect of credibility. Meanwhile, commitment to undertake whatever is necessary to overcome a financial crisis (as advocated by Ben Bernanke, 2008), and then meeting that commitment, is also important. Consequently, a central bank equipped to handle a financial crisis must understand that success, although not explicitly inflation control, is achieved in terms of building future policy credibility. Considering that policy credibility is built through consistency between what is said and what is done, financial crisis resolution demands not only smart policy actions but also a high level of transparency and broad communication policy because the public, markets, and politicians have the right to a coherent explanation of the unconventional or untested policy measures adopted.

11.5. Bank Indonesia’s Independence and Accountability 11.5.1. Bank Indonesia’s Independence The concept of central bank independence has been widely discussed since the establishment of Bank Indonesia as the central bank of the Republic of Indonesia in the 1950s. Mr Sjafruddin Prawiranegara, the first Governor of Bank Indonesia, hinted at potential disruptions to independence through formation of a monetary board. He stated: it is precisely because of the nature of the work of circulation banks that their leadership cannot be influenced by the whims of political interests, therefore the government should not be given absolute power over the circulation bank. The danger is that the circulation bank might be used for the interests of political parties, which have power over the state… With the introduction of the Bank Indonesia Act (No. 23) of 1999, which has been amended by Act No. 3 of 2004, Bank Indonesia is the central bank of the Republic of Indonesia, which has the status of an independent state institution. Bank Indonesia independence is based on the three aspects of central bank independence put forward by Bofinger (2001) as follows: (1) Goal Independence Functionally, Bank Indonesia has the right to decide all aspects of monetary policy and price stability in accordance with prevailing laws. Nonetheless, Bank Indonesia is not goal independent because Bank Indonesia itself

Central Bank Independence and Accountability     345 does not set the targets to be achieved, the goals are clearly stipulated in law, namely to achieve and maintain rupiah stability. In addition, it is the government that sets the inflation target, after consultation with Bank Indonesia.13 The lack of goal independence at Bank Indonesia is consistent with the prevailing trend of central bank independence around the world, although the degree of independence is different. (2) Instrument Independence Bank Indonesia has adequate jurisdiction/authority to achieve the operational target per prevailing laws. To that end, Bank Indonesia has control over the instruments used to influence inflation, while paying due consideration to the impacts on the economy in general, as stipulated in prevailing laws. Furthermore, Bank Indonesia is also proscribed from disbursing loans to the government. (3) Personal Independence Pursuant to prevailing laws, third parties are not allowed to interfere with task implementation at Bank Indonesia (BOG), and Bank Indonesia (BOG) must reject or disregard all forms of third-party interference. In addition, the Bank Indonesia BOG cannot be dismissed, except in cases of resignation, prolonged absence, or unlawful acts. The BOG (and/or Bank Indonesia Officers) is also immune to prosecution due to the decisions taken or policies implemented in good faith consistent with their duties and authority. The tenure of the members of the BOG is five years, with reappointment possible. The BOG is proposed and appointed by the President after approval from the People’s Representative Council.

11.5.2. Bank Indonesia’s Accountability The Bank Indonesia Act (No. 23) of 1999, amended by Act No. 3 of 2004, also requires accountability in terms of task implementation, jurisdiction, and budget. The stipulation for Bank Indonesia accountability intends to ensure all stakeholders can oversee all policy measures implemented by Bank Indonesia. In addition, accountability can also be used by the People’s Representative Council to appraise Bank Indonesia performance. The principles of accountability are stipulated in accordance with prevailing laws in terms of Bank Indonesia task implementation and jurisdiction through direct public discourse in the mass media at the beginning of each year concerning an assessment of monetary policy implementation in the previous year as well as the planned monetary policy direction and monetary policy targets for the upcoming year. That information is submitted in writing to the President and People’s Representative Council. Furthermore, Bank Indonesia is also required to report to the People’s Representative Council on a quarterly basis or as required in line with the oversight function of the People’s Representative Council.

13

Prior to the amendment, the Bank Indonesia Act (No. 23) of 1999 stipulated that Bank Indonesia would set the inflation target. With the amendment in 2004, however, the government became responsible for setting the inflation target after coordination with Bank Indonesia.

346    Central Bank Policy In terms of the budget, before the fiscal year begins, Bank Indonesia is required to submit its annual budget to the People’s Representative Council and government, accompanied by an evaluation of budget implementation in the current year. The operating budget of Bank Indonesia requires People’s Representative Council approval through a simulated policy budget submitted to the People’s Representative Council as information. Moreover, the Annual Financial Statements of Bank Indonesia is also submitted to the Audit Board of the Republic of Indonesia for auditing and the resultant opinion is published through the mass media. Bank Indonesia is also required to prepare a weekly balance sheet to be published in the State Gazette of the Republic of Indonesia. Accountability and transparency are closely related. The central bank must be held accountable to its stakeholders for its task implementation. A more transparent institution will increase central bank accountability and, ultimately, improve central bank performance (Poole, 2001). Therefore, to increase accountability and transparency, Bank Indonesia routinely publishes various reports and publications, including the Weekly Report, Monthly Economic, and Financial Statistics of Indonesia, Monthly Monetary Policy Review, Quarterly Monetary and Economic Developments, Quarterly Monetary Policy Report, and the Annual Financial Statements.

11.6. Concluding Remarks In this chapter, we have discussed two salient institutional aspects of the central bank, namely independence and accountability, as integral parts of strengthening monetary policy credibility to achieve price stability. Chapter 10 explained that time-consistent and rule-based policy responses are crucial to build central bank policy credibility. Central bank independence, as described in this chapter, represents a form of learning the authority to achieve the price structure mandate of a competent central bank that can implement its duties and guarantee the partnership between policy and setting the rules. Meanwhile, accountability is the logical consequence of independence in pursuance of the mandate. In other words, policy consistency, including rule-based policy, independence, and accountability are all vital to build central bank policy credibility. Relating to the principles of independence, providing a mandate of independence to the central bank, accompanied by accountability, is required ensure the effectiveness of policies to achieve the final target of professional integrity. Referring to the experiences of various countries and the underdeveloped political conditions in developing countries (typically applying a multi-party political system), central bank independence is necessary to subdue the interests of the elite that are counterproductive to the attainment of sustainable economic growth. Although the legal facts show that central bank independence has been set proportionally in prevailing laws, implementation must be observed in the context of integrated policy coordination. This implies that the central bank’s policy cannot run by itself. On one hand, central bank policy is determined as the manifestation of autonomous professionalism. On the other hand, however, the institutional and political constellation, which is oriented toward improving public

Central Bank Independence and Accountability     347 welfare, requires that the central bank not be sterile to the political process altogether. To that end, central bank policy implementation must be strengthened by harmonious policy coordination between the central bank and government or other relevant stakeholders. The critical importance of policy coordination becomes even more apparent during crisis conditions. Numerous aspects change and problems emerge during a crisis, therefore central bank policy independence could be interpreted narrowly as impossible and undesirable. Consequently, independence must encompass the broader interests through policy coordination between the central bank and government to jointly overcome the crisis. Nonetheless, after the crisis period has been overcome, the central bank must prudently extricate itself from the policy compromise with the government and gradually rebuild its independence.

This page intentionally left blank

Chapter 12

Policy Transparency and Communication Strategy 12.1. Introduction Transparency and communication are integral parts of strengthening monetary policy credibility and central bank policy in general. How the statements of the chairman of the US Federal Reserve are always anticipated and influence developments on the US and global financial markets and economies is a solid example. Over the past decade, central banks around the world have highlighted transparency by improving their communication strategies and providing more information to the public. Senior central bank officers routinely clarify the monetary policies taken to the public. Furthermore, most central banks publish the inflation and economic forecasts underlying their monetary policy orientation. In fact, many central banks clarify how the monetary policy framework is applied, including publishing the projection models and results of other research. Notwithstanding, there remains an ongoing debate in theory and in practice over how transparent a central bank should be. Theoretically, economists tend to suggest that there is never too much information given to the public (Blinder, Goodhart, Hildebrand, Lipton, & Wyplosz, 2001). Since the emergence of classical economics, imperfect information has long been viewed as a source of market imbalances that create distortion in the allocation of economic resources, which translates into suboptimal public welfare.1 Likewise, in the context of the debate between classical and Keynesian economics with rational expectations theory, although surprise or unanticipated monetary policies influence the real economy, the central bank’s desire to manipulate expectations will not be optimal for the general economy (Barro, 1977; Cukierman & Meltzer, 1986; Lucas, 1972; Sargent, 1973; Sargent & Wallace, 1975). In a broader context, demand for monetary policy transparency as a public policy is the embodiment of democratic accountability (Blinder et al., 2001). In the words of Milton Friedman, as cited 1

Refer, for example, to the market for lemons theory developed by Akerlof (1963), which analyzes market imperfections due to asymmetric information. Central Bank Policy: Theory and Practice, 349–384 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191017

350    Central Bank Policy by Faust and Svensson (2000), for instance, the reluctance of the central bank to communicate monetary policy is “avoiding accountability on the one hand and achieving public prestige on the other.” The policymakers, including the central bank, view information differently. In the past and at some central banks even today, secrecy remains a key concern, with the central bank restricting external and internal communications. In general, the rationale and framework underlying monetary policy is the preserve of the most senior leadership of the central bank. Such information is viewed as a recipe or instrument private to the central bank because it determines monetary policy effectiveness in terms of influencing the economy. Concerning the theory of asymmetric information, this view can be explained by the superiority of the information held by the central bank as the most important element of monetary policy effectiveness. Similarly, information confidentiality also supports monetary policy effectiveness because only surprise/unanticipated monetary policy has an effect on the economy. Institutionally, many central banks in the past were not independent, thus demand for transparency as a key to fulfilling monetary policy accountability was also not forthcoming. Nevertheless, it seems that that era has changed. Convergence between central bank policy transparency theory and practice, particularly monetary policy, has already begun, especially as more and more central banks expand transparency and their monetary policy communication strategy. This chapter specifically focuses on that problem over six sections. After the introduction, the second section puts forward several concepts and empirical dimensions underlying the importance of transparency and its implications on policy effectiveness and credibility. The third section explores the theoretical models developed that underlie policy transparency at numerous central banks. Thereafter, the practices and empirical studies on policy transparency and communication strategies in several countries will be presented. Then, before the concluding remarks, Bank Indonesia’s efforts to increase policy transparency and improve the communication strategy are presented.

12.2. Conceptual Dimension of Policy Transparency and Communication Strategies There are at least three factors that can explain the phenomenon of increasing monetary policy transparency and improving communication strategies in many countries (Amato, Morris, & Shin, 2003). First, greater central bank independence demands increased accountability. In general, increasing accountability translates into more central bank communication concerning the direction of monetary policy operations and the confidence therein. Only through increasing transparency can a fair appraisal of central bank performance be conducted. Therefore, regulating the relationship between an independent, transparent and accountable central bank and the government and public is becoming more important. Establishment of the European Central Bank (ECB), which avoided parliament and government interference, is a good example.

Policy Transparency and Communication Strategy    351 Second, more and more central banks in advanced countries and emerging market economies (EMEs) are implementing the Inflation Targeting Framework (ITF) as their monetary policy framework. As discussed in Chapter 8, although the regimes applied by central banks vary, all central banks that formally implement ITF stress the importance of communicating monetary policy to the public. Through regular public communication, the monetary policies instituted by the central bank to achieve the inflation target will be better understood by the public. This is important not only to increase central bank credibility and accountability but also to anchor public inflation expectations to the desired inflation target. Third, financial markets are becoming more developed in many countries. Transactional behavior and price formation in financial markets are predisposed to market expectations concerning the economic outlook, including inflation and the direction of the central bank’s policy rate. The discussion on monetary policy transmission mechanisms presented in Chapter 5 evidenced the importance of the expectations transmission channel. Furthermore, monetary policy influences transactions and financial asset prices through other transmission channels, such as the interest rate, money supply, credit, exchange rate, and asset price channels. Managing and manipulating expectations in financial markets, therefore, are important parts of monetary policy, which will clearly not be achieved without transparency and a sound communication strategy by the central bank. As a contemporary issue of monetary economics, various viewpoints have been developed on what, why, the extent and how monetary policy transparency and communication is undertaken by a central bank. Nevertheless, the differing views fail to reach the same conclusion. Consequently, it is worth reviewing the different opinions contained in the literature before discussing transparency practices at central banks and their contribution to economic performance in more depth.

12.2.1. Several Viewpoints Concerning Transparency In general, Poole (2003) defined monetary policy transparency as “accurately conveying accurate information, including all the information market participants need to form opinions on monetary policy that are as complete as possible.” In the context of the Code of Good Practices in Monetary and Financial Policies developed by the IMF, Sundarajan, Das, and Yossifov (2003) provided more concrete understanding that monetary and financial transparencies refer to an environment in which the objectives of the policy; its legal, institutional and economic framework; policy decisions and their rationale; data and information related to monetary and financial policies; and the accountability of the policy making body are provided to the public in an understandable, accessible and timely way. This is consistent with the views of Geraats (2001), who placed transparency in the phases of providing information on monetary policy to the public.

352    Central Bank Policy Geraats (2001) separated transparency into five aspects, namely: (1) disclosure of the final policy goal, such as the explicit inflation target (political transparency); (2) disclosure of the economic data, models, and forecasts delivered by the central bank (economic transparency); (3) information relating to the monetary policy strategy and the associated internal considerations, for example, by publishing the minutes and voting records (procedural transparency); (4) communication of the policymaking process, for example, a change in the policy rate and statements regarding the future policy direction (policy transparency); and (5) disclosure of policy implementation, market interventions, and control errors (operational transparency or commonly referred to as market transparency). Differentiating transparency facilitates a review of how the views have developed in the literature in order to understand the motives and effectiveness of transparency itself. Political transparency has been analyzed by Faust and Svensson (2001) as well as Nolan and Schaling (1996). The analyses showed that clarification regarding the central bank’s preferences and desired goals of monetary policy had the propitious impact of reducing inflation bias as well as fluctuations in inflation and job opportunities. This is because such assurance informs the public of the central bank’s preferences in terms of the ultimate goal of monetary policy, in this case inflation, and, therefore, reduces the disparity between the rate of inflation desired by the central bank and the inflation expectations of the public. Meanwhile, procedural and policy transparency, specifically in terms of publishing the minutes, voting records and policy directives, has been discussed by Goodfriend (1986), Buiter (1999), and Issing (1999). Formally, the publication of individual voting records was analyzed by Gersbach and Hahn (2000). Regarding policy transparency, in particular clarifying or publishing the direction and decisions of monetary policy, most economists view the positive contribution on expectation formation by economic agents and the effectiveness of monetary policy itself. Nonetheless, concerning procedural transparency, specifically the publication of minutes and voting records, there remains a difference of opinion between economists and central bankers. Hitherto, only a few central banks have published the minutes of their meetings and voting records of the monetary policymakers. Several models of operational transparency have also been developed. In their seminal paper, Cukierman and Meltzer (1986) proposed operational ambiguity in terms of monetary policy to improve controlling errors when striving to achieve the goals set. Meanwhile, Faust and Svensson (2000, 2001) provided a theoretical foundation for the differences between the inability to control monetary policy and operational transparency, while also showing that transparency could ameliorate public welfare. In a similar model to Faust and Svensson (2001), using New Keynesian economics rather than the standard Phillips curve, Jenson (2002) found that increasing transparency could undermine public welfare and, therefore, advocated confidentiality in terms of monetary policy. Economic transparency has also been discussed in relation to several models based on central bank publications of economic and inflation forecasts as well as assessments of public expectations and the formal models used.

Policy Transparency and Communication Strategy    353 Gersbach (1998), Cukierman (1999), and Geraats (2001) argued that publishing the central bank’s projections eliminates asymmetric information concerning shocks in the economy. Notwithstanding, differences in the assumptions regarding monetary policy transmission channels and information on the central bank’s preferences have produced disparate conclusions, where Gersbach (1998) and Cukierman (1999) showed how transparency undermined public welfare but Geraats (2001) showed how transparency improved public welfare. On the other hand, Tarkka and Mayes (1999) argued that not only do the public not know the central bank’s preferences but also do not have access to the central bank’s assessment of public inflation expectations. Therefore, by publishing its forecasts, the central bank could eliminate such public uncertainty and, hence, improve monetary policy effectiveness moving forward. 12.2.1.1. Scope of Transparency: Full Versus Limited. From an economic perspective, the main argument supporting the need for full transparency is that monetary policy would be most effective if correctly anticipated by the markets. That argument is based on the three characteristics of effective monetary policy signaling (Blinder et al., 2001). First, monetary policy generally works through variables that are moved by market expectations. Consistent with Blinder (1998), short-term interest rates are the most controllable variable by the central bank in many countries and also have a broad effect on the economy. Meanwhile, longterm interest rates, asset prices and exchange rates, despite also having a strong and wide-ranging impact on the economy, are difficult to control by the central bank. The relationships between various financial variables and short-term interest rates involve the expectations formed in the markets. Less volatility in the expectations leads to a more stable and easier-to-predict relationship between monetary policy and its impact on the economy. To some extent, market expectations are influenced by the future direction of central bank actions, therefore, monetary policy would be more effective if easily and accurately anticipated by the markets. Second, the transmission channel from the interest rate to the real economy and inflation involves price and wage formation at the producer and consumer levels. Part of the influence of monetary policy lies in its impact on economic dynamics, through the traditional backward-looking Phillips curve, as explained by Gordon (1998), and partly through expectations in the forward-looking Phillips curve channel, as found by Clarida, Gali, and Gertler (1998). The first channel typically operates more slowly and disruptively, while the second channel is faster yet more difficult to predict. Consequently, transparency would improve credibility and credibility would strengthen the second channel and, therefore, improve the effectiveness, predictability, and speed of monetary policy transmission. In fact, price and wage formation would adjust quickly if what the central bank is doing and trying to achieve are known. Third, transparency also reduces the costs incurred by changing policy direction. If economic dynamics change or differ from what was anticipated, the direction of monetary policy may need to be adjusted. In this regard, central bank concerns that a policy change would confuse the public and undermine credibility are understandable. On the other hand, however, the original policy

354    Central Bank Policy direction could delay the policy adjustment and lead to suboptimal policy. Therefore, it is fundamental for the public to fully comprehend the policy regime adopted so that a policy change would be easier to accept and could reduce the costs incurred. Based on those three arguments, Blinder et al. (2001) concluded that the basis for monetary policy transparency is incredibly robust. Fundamentally, the rationale lies in the phenomenon of asymmetric information, namely that the central bank has greater knowledge about itself, its instruments and desires compared to the public. Such conditions create a misunderstanding that undermines monetary policy effectiveness as described above. Although the markets will constantly strive to understand central bank behavior, if the central bank fails to clarify its knowledge of the economic situation and monetary policy direction pursued, misunderstandings will continue to occur. It would be better for the central bank to transparently select the regime and procedures of monetary policy implementation. In other words, this view recommends, as a general rule, that all aspects must be communicated publically, except in special cases where confidentiality is necessary. Considering the clear and fundamental need for broad transparency, why did central banks limit their communication with the public for so long? Why did central banks in the past keep the considerations underlying monetary policy a secret or prefer creative ambiguity? A number of considerations can explain such developments. First, only unanticipated monetary policy is effective in terms of influencing the economy. The seminal argument for limited transparency was proposed by Cukierman and Meltzer (1986) by assuming that, in an economy with rational players, the central bank’s actions would have already been considered when forming the prices of goods, wages, and interest rates in the markets. Under such conditions, monetary policy that has been anticipated would not effectively influence the economy and lead to the short-run neutrality of money. Consequently, to successfully steer economic developments in the desired direction, the central bank must limit transparency in order to act as leader in this game, implying that the central bank must formulate monetary policy that is full of surprises. According to Blinder et al. (2001), this argument, although common in economic theory, lacks the support of convincing empirical evidence. Furthermore, there is also a lack of cogent reports that central banks try to create surprises when implementing monetary policy. Second, limited transparency improves monetary policy’s ability to influence economic growth and job opportunities in the near term. This argument is based on the view of inflationary bias proposed by Kydland and Prescott (1977), namely that if the central bank wishes to stimulate economic growth, or reduce unemployment, inflation surprises must be created from time to time. This ability would be lost, however, if monetary policy was implemented transparently. Of course, the need for transparency would fade if the central bank already enjoyed a high level of credibility. With a high level of central bank credibility, the public would continue to trust the actions of the central bank and, therefore, the central bank would have more space to create inflation surprises and influence output.

Policy Transparency and Communication Strategy    355 According to Blinder et al. (2001), there is no firm evidence concerning inflationary bias from central banks in advanced countries. Despite several cases of inflationary policies implemented by central banks to stimulate short-term output, such moves typically stem from past political pressure and usually fail to hit the target. With the growing trend of central bank independence around the world, such arguments lack a solid foundation. Third, however broad the respective central bank’s scope of transparency is, the markets and public may still not be able to accurately capture the corresponding monetary policy signals and, therefore, asymmetric information would still persist. In fact, for the skeptics and speculative investors, such information from the central bank could be manipulated to create inaccurate and exaggerated expectations in pursuit of profit. Financial market volatility could increase, coupled with panic and herding behavior, ultimately holding the central bank hostage to market sentiment. According to Blinder et al. (2001), that argument is baseless, showing that the typical sources of market confusion are insufficient and inaccurate information form the central bank. Fundamentally, disclosure would eliminate misunderstanding, the cost of which, for monetary policy effectiveness, is prohibitive. A better solution would be to provide clear and complete information concerning the future direction of monetary policy, such as the direction of the interest rate, so that the markets, especially the bond market, can formulate accurate expectations of the interest rate term structure in the markets. In the words of Tarkka and Mayes (1999) and Winkler (2000), full public transparency regarding the central bank’s preferences would precipitate convergence between the heterogeneous views found in the financial markets and accurate expectations.

12.2.2. Transparency, Monetary Policy Regime, and Democratic Accountability The various opinions among economists described above demonstrate the different considerations underlying the rationale of monetary policy transparency. Likewise, in practice at many central banks, the basic considerations and communication strategies differ. On one extreme, communication could merely be limited to specific aspects of monetary policy. On the other extreme, however, communication could be oriented toward achieving the maximum degree of transparency concerning various aspects of monetary policy. In this respect, it would be interesting to discuss the linkages between transparency and the monetary policy regime applied by the central bank, as described by Blinder et al. (2001). To that end, a central bank could be considered transparent if it provided adequate information to the public at all times to understand the monetary policy regime, thereby facilitating cross-checks and providing a clear assessment of whether the actions of the central bank were consistent with the regime applied. Theoretically and as practiced by central banks, the monetary policy regime could vary from rules to discretion-based. With such understanding, communication seeks to provide various detailed information on the procedures, data, policy formulation process, and so on, as applied by the central bank.

356    Central Bank Policy The currency board regime with a simple automated rule occupies one extreme of policy transparency. Under such a regime, monetary policy is based on an exchange rate peg with a simple rule for the central bank to provide the exact same amount of money supply as the change in reserve assets required to support the fixed exchange rate. As the regime converts to an exchange rate peg, simple transparency in the form of an exchange rate rule is maintained but the central bank enjoys additional monetary policy flexibility due to the possibility of instituting devaluation policy or revaluating the exchange rate, while being exempt from providing the same amount of money supply as the change in reserve assets. A crawling peg or managed floating regime affords greater flexibility because the exchange rate is allowed to fluctuate within a given range, in addition to the advantages offered by the exchange rate peg mentioned previously. Therefore, monetary policy transparency depends on the width of the exchange rate range (upper and lower bounds). Meanwhile, clarification regarding the discretionbased foreign exchange intervention conducted by the central bank is not usually published. In the context of monetary policy, flexibility is required to increase central bank autonomy in terms of orienting monetary policy toward the domestic economic goals, particularly inflation, away from external influences. With a more flexible exchange rate system, the monetary policy regime is more often oriented toward the domestic economic goals. As explored in Chapter 8, an inflation targeting regime, in this case the ITF, entails transparency at least in terms of the ultimate goal and instruments of monetary policy. The final goal of monetary policy is price stability, while also determining and announcing the medium–long term inflation target. Meanwhile, monetary instrument transparency is reflected in the interest rate direction and operational target in the current period to achieve the future inflation target. In practice, flexibility per this regime generally lies in the need to publish forward-looking economic and inflation forecasts, the models used as well as other types of operational transparency. A monetary policy regime based more on discretion than a rule involves more complex transparency. This is usually applied by central banks with multiple mandates, not only to achieve price (inflation) stability but also to stimulate economic growth and create job opportunities for instance. Furthermore, the complexity of the transparency not only lies in the lack of clarity concerning the final target, whether it be inflation or economic growth or something else, but also in the linkages between the strategy and operational target set by the central bank. In general, central bank communication emphasizes qualitative clarification regarding the future direction of the economy and inflation based on the current operational target of monetary policy. The Federal Reserve, for instance, generally announces its decisions and the desired Federal Funds Rate (FFR) based on qualitative assessments of the economic and inflation outlooks, without providing a detailed explanation concerning the formal framework used to formulate monetary policy. 12.2.2.1. Transparency and Democratic Accountability.  The trend of democratization that has affected many countries, particularly in the past decade, has

Policy Transparency and Communication Strategy    357 increased calls for transparent public policies, including monetary policy. The central bank is a public institution and, therefore, must be held accountable for all actions and policies implemented. The principles of accountability, therefore, are the basis of the requirement for information on the monetary policy implemented by the central bank to be published, unless there is a strong reason not to. That exemption could be justified if the information submitted periodically by the banking industry and financial institutions to the central bank, along with the variety of other sensitive information, could undermine the integrity of financial market operations. Beyond that, other information is required by the public as the manifestation of democratic accountability (Blinder et al., 2001). The application of democratic accountability rests on the central role of the communication strategy to increase central bank accountability. Whichever legal mechanisms must be followed to fulfill monetary policy accountability, the central bank will never be able to soundly implement its duties without the backing of the stakeholders. Under certain circumstances, the central bank must implement unpopular policies to achieve the goals, such as price stability. For example, the monetary tightening required to control inflationary pressures would raise the interest rate, which could have an adverse impact on short-term economic growth. Pursuing such a policy would fail to garner public support if trust in the central bank, including the monetary policy direction, targets and commitments, has not been built. In other words, monetary policy effectiveness can only be improved if the motives and procedures of the central bank are clearly understood by the public. The application of democratic accountability principles emphasizes the importance of a central bank communicating constantly and transparently, during the good times and bad, with the relevant stakeholders to clarify the monetary policy stance. In practice, there are variations in the application of democratic accountability principles in terms of central bank monetary policy transparency from one country to the next, depending on the democratic progress achieved and prevailing social norms. In several countries, as discussed in Chapter 11, the process of democratization has been accompanied by broader central bank independence. In terms of the regulatory, intuitional and monetary policy frameworks, greater independence has led to a more obligations in terms of transparency and accountability. This is congruous with the proliferation of the ITF in many countries. In this case, clearer transparency guidelines regarding the price stability goal of monetary policy, announcing the inflation target, clarifying the monetary policy framework, publishing economic and inflation forecasts as well as announcing the direction and operational target of monetary policy are integral parts of ITF implementation.2 In several other countries, particularly where the central bank does not enjoy independence in practice or independence is not contained in prevailing laws, the application of democratic accountability in terms of monetary policy transparency is not clearly observable. Under such conditions, the central bank remains

2

Refer to the discussion on ITF implementation at various central banks in Chapter 8.

358    Central Bank Policy an extension of the government and, therefore, monetary policy is formulated and applied as a part of the government’s economic policy. Consequently, clarity concerning the goals, targets, and monetary policy framework is not as forthcoming as from a central bank that enjoys a high degree of independence. Although some of these central banks announce their monetary operations, the linkages to the ultimate target of monetary policy, desired and prioritized, remain opaque.

12.2.3. Monetary Policy Communication Strategy The discussion above demonstrates the strength of the argument for monetary policy transparency by the central bank, although full transparency is not possible to achieve and, therefore, market misunderstanding could possibly persist. The question is therefore, what scope of information should be communicated, by which methods and to whom? 12.2.3.1. Scope of Communication.  The principle underlying monetary policy transparency is that the information communicated helps the public to understand, and anticipate, the central bank’s decisions as a logical conclusion to the policy sequence oriented toward achieving the final target. Per that rationale, Blinder et al. (2001) substantively opined that the scope of information communicated to the public in accordance with the corresponding priority involves the following aspects: (1) goals; (2) methods; and (3) decision-making; (1) Goals: The central bank must clarify what it hopes to achieve from its monetary policy, including the final target (such as price stability) or the short-term goals (namely the operational target, such as the short-term interest rates, and intermediate targets, such as money supply, medium-term interest rates, and the exchange rate). The public understands that determining the targets represents a political decision and often entails a dilemma. The public wants to see rational targets, particularly in support for public prosperity, and whether the central bank is consistent and serious in terms of achieving the goals when implementing monetary policy. The most fundamental aspect to be communicated is the long-term goal because this will have the biggest impact on public welfare. Therefore, the scope of information will depend on the monetary policy regime adopted. This is comparatively clear and simple for countries applying a pegged exchange rate system because the exchange rate, point or range, is the goal of monetary policy. Similarly, in ITF countries, price stability is usually stipulated in prevailing laws and the inflation target, whether it be set by the government or the central bank itself, is also announced publically. Conditions become less clear for central banks with a more ambiguous goal, for instance “achieving price stability, while stimulating sound national economic development,” such as the Bank of Japan (BOJ), or when there is a dual mandate for monetary policy, for example, “achieving price stability and creating broad job opportunities,” such as the US Federal Reserve.

Policy Transparency and Communication Strategy    359 The short-term goals must also be communicated. Monetary policy affects inflation and economic growth after a lag, therefore information on the short-term goals should encompass the operational and intermediate targets to be achieved. Transparency, in terms of the operational target, for example, the short-term interest rate target, patently needs to be communicated because the information contained within could be utilized by the public as a measure of the monetary policy actions taken by the central bank to achieve the long-term target. The case becomes less clear regarding the intermediate targets, not only because the degree of uncertainty is linked to the operational and final targets, but also because of the ever-present dilemma. Previously, we discussed the argument for transparency concerning the intermediate target, primarily to form market expectations and improve monetary policy efficiency. Nevertheless, such transparency, especially if the dilemma between controlling inflation and stimulating job opportunities is present, could risk public opposition and open the door to political intervention from the government or parliament. (2) Method: A difference of opinion also prevails concerning the extent to which a central bank needs to be transparent about the methods used to determine monetary policy, particularly in terms of the economic forecasts and models used. In general, the opinion to limit transparency is based on the argument that the economic models and projections use specific assumptions, contain a degree of uncertainty and could have a potential impact on central bank credibility if proven erroneous. Blinder et al. (2001) recommended the opposite by advocating transparency, particularly in terms of the key features and basic assumptions used in the economic forecasts and models, without necessarily detailing the minutia. In addition to forming expectations on the financial markets and enhancing public understanding of the central bank’s monetary policy, the public would also gradually understand that the projections still contain the risk of inaccuracies. On the other hand, the central bank is generally transparent in terms of the practical methods employed, namely the monetary operations applied to achieve the operational target. Each central bank that announces its operational target decisions, for instance the level or direction of changes in shortterm interest rates, also directly announces its monetary operations, oriented to achieve the operational target through open market operations, discount facilities and other facilities. Exemptions are typically put in place regarding foreign exchange intervention by the central bank, which is usually undertaken based on certain considerations by the central bank. The importance of such procedural transparency is based on the need to avoid and minimize shocks in the markets, even within a brief period. (3) Decisionmaking: Transparency concerning the operational target decisions made by the central bank is always monitored publically, not only in terms of forming

360    Central Bank Policy expectations but also to check the consistency of monetary policy. In this case, central banks in several countries usually announce changes in the operational target, for instance decisions relating to short-term interest rates, immediately after the decision has been taken. The announcement is made through a press release, sometimes with an accompanying explanation and sometimes without. Such practices are becoming more common at central banks, a positive development in terms of monetary policy transparency. Nevertheless, there remains a difference of opinion regarding the need for the central bank to announce its preliminary plans about future monetary policy. The traditional viewpoint argues that forward-looking information is still considered taboo because of concerns that such information could be misinterpreted by the markets, thus triggering excessive expectations regarding the future orientation of monetary policy. On the other hand, to increase transparency and improve the formation of expectations, a more moderate view tends to support central bank measures to clarify the direction of monetary policy with the markets, politicians, and public. In this case, transparency involves clarifying the future direction of the economy and inflation, the concerns that need to be addressed, as well as the direction and considerations underlying monetary policymaking. Several central banks have already begun to announce the forward direction of their monetary policy. The most advanced is the Reserve Bank of New Zealand, which has published its monetary policy contingency plan for the upcoming three years since 1997. The contingency plan emphasizes that only monetary policy decided in the current period will be implemented, while the future policy direction will depend on the unfolding dynamics. Likewise, the US Federal Reserve began clarifying its monetary policy direction in May 1999. Although qualitatively in line with confidence concerning the future direction of the economy and inflation, the move is often referred to as a preventative action. On the other hand, the Bank of England and ECB are more vague when indicating the forward direction of monetary policy. Meanwhile, in the context of financial and monetary policy transparency, Sundarajan et al. (2003) proposed three fundamental elements of sound policy transparency, namely: (1) clearly and consistently stated policy goals, with periodic clarification regarding performance and the underlying considerations; (2) a solid foundation, legally, institutionally and economically, for the policies instituted; and (2) data and information availability to create a sound view of the monetary and financial policies implemented as well as the implications on the financial choices of individuals and companies. The three basic elements outlined above underlie the “Code of Good Practices on Transparency in Monetary and Financial Policies,” which was developed by the IMF in 1999 and is now adhered to by many member states (IMF, 2003).

Policy Transparency and Communication Strategy    361 In more detail, the scope of monetary policy transparency contained in the document is as follows3: ⦁⦁ Clarification regarding the role, jurisdiction, and goals of the monetary policy

authority, interpreting and explaining the final goal and institutional framework of monetary policy, as well as the institutional relationship between the monetary policy authority and other financial authorities. ⦁⦁ Disclosure of the monetary policy formulation and reporting processes, clarifying the framework, instruments, and targets set to achieve the final target of monetary policy; changes in the monetary policy instruments; as well as the progress in terms of achieving the final target of monetary policy and the outlook. ⦁⦁ Public availability of monetary policy information, explaining the presentation and publication of central bank data in accordance with the prevailing standards; balance sheet reporting and annual financial statements, as well as the salient economic, monetary, banking and financial sector developments. ⦁⦁ Accountability and integrity of the monetary authority, covering specific regulations concerning the accountability mechanism, such as periodic monetary authority clarification before parliament or the public regarding monetary policy implementation; availability of audited financial statements; internal procedures per good governance; as well as behavioral guidelines (code of ethics) and legal protection for monetary authority employees. 12.2.3.2. Communications Methods.  Sundarajan et al. (2003) proposed four basic dimensions that determine the credibility of policy transparency, namely: (1) the means or media used; (2) timeliness; (3) periodicity; and (4) the quality and scope of the information submitted. On top of those four dimensions, Ehrmann and Fratzscher (2005) added another: timing. The five dimensions of transparency can be explained as follows: ⦁⦁ There are various means and media used in terms of monetary policy transpar-

ency, such as: (a) clarification through the publication of official documents; (b) clarification through the mass media or before parliament; (c) direct clarification with the public; and (d) other forms of clarification. The various means and media can be used simultaneously per the desires of the monetary authority to effectively expand its transparency. In many ways, transparency can be expanded through public discourse to seed holistic understanding of the monetary policy instituted by the central bank. ⦁⦁ Timeliness in terms of submitting information is crucial to effectively create transparency. That information must be useful and relevant in terms of monetary policy implementation by the central bank. The information should be published immediately after a decision has been taken or when the data become 3

For a detailed explanation on financial policy transparency, refer to IMF (2003).

362    Central Bank Policy available. Delaying the publication of such information not only makes it outdated but also contains the risk of a leak and could trigger spurious rumors about the actual monetary policy decisions and data. ⦁⦁ Periodicity in terms of submitting information is also vital to maintain the credibility of monetary policy transparency. The publication schedule for monetary policy decisions, once set, should be maintained and respected. For example, if the central bank has already resolved to announce the results of the decision on the operational target of monetary policy, be it short-term interest rates or money supply, on a monthly basis, then the announcements must be delivered consistently on a monthly basis. This is important to formulate market expectations and expectations among economic agents. Delaying the announcement must be avoided, especially if the news is bad, because this would not only challenge the spirit and meaning of transparency but could also undermine monetary policy credibility. ⦁⦁ The scope and quality of the information provided is also critical to support monetary policy transparency. The scope of information will be determined by a central bank assessment of the types of monetary policy information to be published. The emphasis of transparency must be placed on the materiality and relevance of the information presented. The goal of transparency will not be attained if contradictory information is published or if the language used is too technical or ambiguous. ⦁⦁ The timing of publishing information is also considered important to monetary policy effectiveness. In this respect, it is important to understand whether the publication of information under certain circumstances is required, when it is required and the intensity required, considering that the policies communicated could be understood differently by financial market players before and after a policy is announced to the public. 12.2.3.3. Communication Targets. To whom monetary policy transparency and communication is targeted is a reflection of the democratic accountability principles described previously. Blinder et al. (2003) suggested four main targets of central bank communication, namely: (i) the mass media and public; (ii) the government and parliament; (iii) the financial markets; and (iv) central bank observers. The scope of information and the communication means would, therefore, depend on the four targets. (i) Mass media and public: In general, the public does not know in detail, nor can the public anticipate, the monetary policy of the central bank. Nevertheless, the public understands the importance of price stability and will complain if inflation increases. Furthermore, the public also monitors interest rate and exchange rate developments, especially if large fluctuations occur. Ultimately, the public sees the central bank as the authority with control over economic growth and job availability. Such societal conditions do not exclude the public from being the target of central bank communications. In fact, public sentiment has become the focus

Policy Transparency and Communication Strategy    363 of central bank communication programs in many countries, using the various information channels available. A range of brochures are provided to the public, whether published, explained directly or through the mass media and official website of the central bank. Furthermore, central bank officers routinely appear at press conferences, interviews and briefings to relay the performance and background behind the monetary policies implemented. In general, the public gets its information from the non-specialized mass media, including the television, radio, and newspapers. The specific challenge facing central banks lies in the communication strategy through the specialized mass media, providing financial analyses, including the central bank’s monetary policy. In this case, the specialized mass media wishes to provide more detailed commentary and analysis concerning the pros and cons of monetary policy. Nonetheless, the content of the analysis, especially on the television, is sometimes inaccurate and leads to misunderstanding. That is why it is important for the central bank to furnish reporters with the background and a greater understanding through periodic workshops, briefings and other means. Equally as important is to provide observers of central bank policy a clear perspective and analysis to help mould public opinion. (ii) Government and parliament: In terms of democratic accountability, the government and parliament represent the public but do not always have the same agenda as the monetary policy of the central bank. On one hand, the government has a keen interest in monetary policy because of its effect on several important economic indicators, such as inflation, interest rates, exchange rates, and economic growth. Nevertheless, greater government attention on populist policies sometimes places the government’s focus on the near term, particularly in terms of achieving robust economic growth and creating broad job opportunities despite reducing the weight and importance of controlling inflation and the implications on the central bank’s policy rate. Under such conditions, conflicting views between the government and central bank could emerge when determining the inflation target and its implications on interest rates and economic growth. Furthermore, the proliferation of greater central bank independence has occasionally raised questions regarding the most appropriate relationship format between the government and central bank. Herein lies the importance of a close relationship and communication strategy between the central bank and government. However central bank independence is implemented, monetary policy effectiveness will be influenced by synergy with fiscal policy because both are integral elements of macroeconomic policy. Consequently, a communication strategy that strengthens monetary-fiscal policy coordination is fundamental, among others, through periodic meetings between the Minister of Finance and Governor of Bank Indonesia as well as their subordinates at both institutions to discuss macroeconomic projections as state budget assumptions as well as the various aspects of monetary, fiscal, and other macroeconomic policies based on the duties and authority of each respective institution.

364    Central Bank Policy Meanwhile, parliament generally represents the public in terms of monitoring and appraising central bank performance, including monetary policy implementation. To that end, the central bank is typically required to submit periodic reports to parliament. In many cases, a senior central bank officer will routinely appear before parliament and follow the dialog covering various policy aspects in the interests of the public. The quality of the dialog will depend on the support staff in parliament as working partners of the central bank. The dialog between the central bank and parliament is heavily reported in the mass media and, therefore, complete, unambiguous, and accurate clarification would support the credibility of monetary policy transparency in the eyes of parliament and the public. (iii) Financial markets: The financial markets always maintain a keen interest in monetary policy because of its strong influence on market performance. In terms of monetary policy, the financial markets are an important channel where the monetary decisions of the central bank are transmitted to the economy and, eventually, to the achievement of the final monetary policy target. Nonetheless, this channel is heavily influenced by expectations, therefore the central bank’s ability to reassure the financial markets is critical. That is why the central bank always allocates a large portion of its communications to the financial markets. Blinder et al. (2001) noted several aspects that demand attention when the central bank communicates with the financial markets. First, financial system stability (FSS) fundamentally underlies the central bank’s attention because of its role in terms of financial intermediation for the real sector as well as monetary policy transmission. It is important to remember that a focus on FSS cannot shackle the central bank when formulating its monetary policy. In many cases, however, particularly when there is a tangible threat to FSS, it is best for the central bank to clarify the situation to the public.Second, it is important to realize what may be considered sound for the financial markets may not always be best for the public. Fundamentally, the financial markets like volatility, for example, interest rate or exchange rate volatility, as a source of profit, while in the interest of the public, financial stability is preferred. Moreover, the financial markets operate on a shorter horizon, from second to second even, while the public cares about the effect of monetary policy on inflation and economic growth in subsequent periods, in addition to interest rate and exchange rate developments. In this respect, monetary policy transparency can bridge the sometimes disparate interests of the financial markets and public. Besides, greater transparency feeds through to less volatility on or created by the financial markets from the formation of expectations that are misaligned with the monetary policy direction. In addition, transparency also allows observers to assess the performance and direction of monetary policy as a correction if erroneous expectations are formed on the financial markets.

Policy Transparency and Communication Strategy    365 (iv) Central bank observers: Central bank observers encompass the financial markets, media and academia. They have deep expertise and understanding regarding the technical aspects, which allows them to interpret the central bank’s monetary policy. Therefore, such observers play an important role in terms of clarifying the actions of the central bank and disseminating the various information communicated by the central bank to the public. In fact, some of the observers will have the ability to forecast the direction of monetary policy, thus playing an important role in terms of transmitting monetary policy to the financial markets or forming public opinion in general. Such abilities, given adequate information through an appropriate communication strategy, will play a strategic role in determining the effectiveness of the central bank’s monetary policy. For the central bank, such observers could be considered a friend or a foe. The key lies in how far the central bank can communicate and persuade the observers regarding the rationale and orientation of monetary policy implementation. A high level of transparency would reduce the risk of misinterpretation by a section of the observers, thus supporting the clarification and dissemination of information by the central bank. Nonetheless, providing too much information, especially confidential information, is not always beneficial because the risk of misusing that information to form public opinion could paint the central bank into a corner. Therefore, the most appropriate strategy would be to provide as deep of an explanation as possible concerning the aspects of monetary policy that should be in the public domain and select more limited information regarding the more confidential aspects.

12.3. Theoretical Models on Policy Transparency: Conservative Central Bank and Imperfect Transparency This section shows the possible effects of imperfect transparency on economic dynamics. Conceptually, the problem of imperfect transparency could occur when the central bank has certain information on the characteristics of a shock and the policy response to influence the economy is not known by the public (Cukierman, 1992, 2000); or when the central bank ambiguously states its purpose (Cukierman & Meltzer, 1986); or when the public faces uncertainty regarding the central bank’s preferences (Nolan & Schaling, 1996). To accommodate such possibilities, when developing their theoretical model, Demertzis and Hughes Hallett (2003) observed imperfect transparency from two aspects, namely: (1) political imperfect transparency concerning the policy goals and priorities; and (2) economic imperfect transparency concerning the information relating to policy-making, shocks, target values, and so on. For both respective conditions, imperfect transparency triggers disruptions that impair the correct formation of public expectations, thus influencing more general economic activity.

366    Central Bank Policy Fundamentally, the theoretical model of transparency developed was adopted from the basic model developed by Rogoff to explain the issue of a conservative central bank (Rogoff, 1985), as explored in the previous chapter. In this case, the central bank’s loss function and supply constraint function, à la Lucas, were formulated as follows:

L=

1  2 E π + b( y − k )2  (1) 2  y = π − π e + ε (2)



where y and π, respectively, are output and inflation expressed in terms of deviation from their steady state value. Meanwhile, b > 0 is the output stabilization weight. The solution for the policy variables of the model are as follows: π = bk −



b ε(3a) 1+ b

π e = bk (3b)



y=



b e(4) 1+ b

Assuming that there is only one player in the system (one policy authority, namely the monetary authority) and no uncertainty in transmitting the effect of monetary policy. In this regard, k is defined as economic factors that affect the achievement of suboptimal unemployment (socially), for instance tax and insurance. And k > 0 reflects the central bank’s desire to correct the distortion.

12.3.1. Political Transparency: Public Uncertainty concerning the Preferred Weight of Output Stabilization (Parameter b) As explained previously, the central bank’s loss function is formulated as follows:

L=

1  2 E a π + b( y − k )2  (5) 2 

Departing from the Rogoff restriction, where k > 0, Demertzis and Hallett also suggested that perfect (full) political transparency is not only conditioned by policy preference weight in the loss function of the central bank (a and b) explicitly, but also the relative weight of both parameters (a/b) as well as the Marginal Rate of Substitution, including (aΔπ/bΔy). The issue of transparency appears when public perception of the central bank’s preferences, for example, the weight of output stabilization (β), differs from the weight set by the central bank. Consequently, there is a false perception (η), considering β = b + η, where E(η) = 0 and V(η) = ση2. This shows that in general, the (average) public perception is correct but not in specific cases. The consequence of uncertainty regarding the value of b is uncertainty in terms of the relative role

Policy Transparency and Communication Strategy    367 of controlling inflation by the central bank in the eyes of the public. By assuming that a + b = 1, then α = a − η. Therefore, the relative weight of the central bank’s preferences is formulated as follows: β b+η = (6) α a−η β b β b where the possibility of full transparency requires E   = or = + ξ , where α a α a ( a + b )η ξ= , and a( a − η )



 ( a + b )η  ( a + b )η (−a )( a + b )ση2 a 2 ( a + b )η 2 = E (ξ ) = E  − + 3 ση (7)  a( a − η )  a( a − η ) a 2 ( a − η )2 a ( a − η )3  

We can see that the assumption, E(η) = 0, is insufficient for the realization of full transparency. In this case, full transparency requires E(η) = 0 or V(η) = ση2 = 0. To simplify the analysis without changing the qualitative conclusion, we then assume a = 1. Consequently, the problem of transparency relates more to uncertainty in the actual value of parameter b. With the perceived value of β, the public expects the central bank to set: π = βk −



β ε (8a) 1+ β

π e = β k(8b)

based on optimization:

Min L = π

1  2 E π + b( π 2 − β k + ε − k )2  (9) 2 

produces the solution of inflation in equilibrium, namely

π* =

b(1 + β ) k b − ε(10) 1+ b 1+ b

The solution is in line with rational expectations, considering that E(π*) = βk = E(πe). Meanwhile, assuming E(η, e) = 0, inflation variance is formulated as follows: 2



 b  2 2 2 V ( π * ) = E [ π * − E ( π * )]2 =   ( k ση + σε ) 1+ b 

(11)

Consequently, inflation variance is a function of imperfect transparency based on public perception (ση2). If b and k are not equal to 0, with a higher value of ση2 (transparency failure), inflation variance would also increase, considering: 2



∂V ( π * )  b  2 =   k > 0 (12) 1 + b  ∂ση2

368    Central Bank Policy Meanwhile, average inflation behaves independently to ση2, hence: ∂ E( π * ) = 0 (13) ∂ση2



by searching for further solutions, the output equilibrium solution is: y* =



ε − µk (14a) 1+ b E ( y* ) = 0 (14b)

and Meanwhile,



V ( y* ) =

σε2 + k 2 ση2 (1 + b )2

(15)

The results imply that the average level of output equilibrium is unaffected by changes in the degree of transparency, while output variance would reduce in line with increasing transparency, considering:

k2 ∂V ( y* ) = > 0(16) 2 (1+ b )2 ∂ ση

Another important implication from the solutions above is the relative effect of imperfect transparency on inflation and output stability, by comparing Equations (12) and (16). We can conclude that if the central bank behaves conservatively, namely is b < a = 1, then:

∂V ( y* ) ∂V ( π * ) (17) > ∂ση2 ∂ση2

Therefore, imperfect political transparency will have a more adverse impact on output stability compared to inflation variability.

12.3.2. Economic Transparency: Public Uncertainty to Changes in Economic Factors (value of parameter k) Like political transparency, in the context of economic transparency, the public have a different perception of the target set by the central bank, k. The public perceive the target as c = k + v, where E(v) = 0 and V(v) = σv2. In this case, full transparency requires E(v) = 0 or V(v) = σv2 = 0. The public expects the central bank to specify:

π = βc −

β ε(18a) 1+ β

π e = βc (18b)

Policy Transparency and Communication Strategy    369 Therefore, the optimization solution of the central bank’s loss function is as follows: Min L =



π

1  2 E π + b( π 2 − βc + ε − k )2  (19) 2 

Where the solution of inflation in equilibrium is as follows: π* =



b( bc + k ) b − ε (20) 1+ b 1+ b

Considering that E(c) = k, then the solution is also in line with rational expectations theory because E(π*) = bk = E(πe). Meanwhile, inflation variance is as follows: 2

 b  2 2 2 V ( π * ) = E [ π * − E ( π * )]2 =   ( b σv + σε )(21) 1+ b 



Consequently, inflation variance is directly proportional to variance v and, therefore, inversely proportional to transparency. 2

∂V ( π * )  b  2 =   b > 0 (22) 1 + b  ∂σv2



Then, output equilibrium is as follows: y* =



e − bv (23) 1+ b

With E(y*) = 0, we can conclude that average output is independent of σv2. Meanwhile, output variance is: V ( y* ) =



σε2 + b 2 σv2 (24) (1+ b )2

Therefore:

b2 ∂V ( y* ) = > 0(25) 2 (1 + b )2 ∂σv

The solutions above imply that economic transparency is the same as political transparency, namely that if the central bank acts conservatively, if b < a = 1, then:

∂V ( y* ) ∂V ( π * ) (26) > ∂σv2 ∂σv2

Therefore, imperfect economic transparency would have a more adverse impact on output stability rather than inflation variability.

370    Central Bank Policy 12.3.3. Implications of Imperfect Transparency From the theoretical model developed above, we can draw a number of conclusions or implications from the presence of imperfect transparency as follows. First, imperfect transparency, political and economic, does not lead to biases in terms of inflation or output. Therefore, public inflation and output expectations will align with actual conditions. Second, in the short run, the public believes that the central bank will behave more conservatively than the actual conditions (β < b and/or c < k), with the temporary possibility of experiencing lower inflation and higher output than the respective trends and vice versa for a more liberal central bank. Third, imperfect transparency, politically and economically, influences inflation and output. In this case, for a conservative central bank, less transparency would exacerbate inflation and output variability; with a more pronounced impact on output variability and vice versa for a more liberal central bank. Fourth, considering average inflation and output are not affected, and the ratio of output variance to inflation variance is the same for both types of transparency (political and economic), hypothetically there is no difference in the results when testing transparency models. Different results may be produced by the data/ information used in the testing. Fifth, the results of empirical testing in several OECD member countries found evidence supporting the conclusion that average inflation and output were not influenced by the degree of transparency. Nonetheless, differing degrees of transparency could explain approximately 50% of inflation variability, while the linkages between transparency and output variability were not as clearly defined.

12.4. Practices and Empirical Studies of Transparency and Communication Strategies in Different Countries As mentioned in Chapter 2, central banks have increased monetary policy transparency over the past decade. In addition to the three salient factors described previously, namely greater central bank independence, the proliferation of ITF implementation and financial market advancement, such developments have also been advocated by international institutions through various endeavors, including promoting monetary and financial policy transparency since 1999.

12.4.1. Assessing Transparency by the IMF Driven by the importance of policy transparency to minimize the risk of a crisis, such as those that have struck Asia and Latin America, the IMF initiated and developed transparency guidelines for monetary and financial policies, commencing in 1999, as the “Code of Good Practices on Transparency in Monetary and Financial Policies (MFP),” to which many members now adhere (IMF, 2003). Participation, in terms of implementing the transparency guidelines, is voluntary and began with 23 member countries in the year 2000, subsequently increasing to 57 of the 184 IMF members. Four important aspects of policy transparency

Policy Transparency and Communication Strategy    371 were evaluated, namely: (1) clarity regarding the role, jurisdiction, and purpose of the monetary policy authority; (2) disclosure concerning the monetary policy formulation and reporting process; (3) public availability of monetary policy information; and (4) monetary authority accountability and integrity. The assessment revealed that, in 2000, monetary policy was most transparent in terms of clarifying the role, jurisdiction and purpose of the monetary policy authority. In nearly all participating countries, the final goal of monetary policy, the duties of the central bank and authority to utilize monetary policy instruments were stipulated and regulated in prevailing laws. Central bank accountability and integrity, however, showed a number of key weaknesses. Based on that assessment, the IMF recommended several improvements to the general scope and forms of transparency (for instance, the quality and timeliness of information, publication frequency, etc.) and identified the weaknesses in terms of guaranteeing central bank accountability and integrity. Subsequently, the assessment conducted in 2003 showed that despite remaining high, monetary policy transparency had declined on conditions in 2000. Congruent with the IMF (2003), this could be explained by several factors. First, the assessment in 2000 included several countries as pilot projects employing self-assessment techniques that were typically less stringent than the direct evaluation performed by the IMF itself. Second, a number of low-income countries were included in the 2003 assessment, where monetary policy is often less transparent. Third, IMF experience in terms of assessing aspects of transparency and the countries involved also improved. Consequently, the quality of the assessment in 2003 was better than in 2000. The assessment in 2003 showed that public availability of information on monetary policy was the most transparent and exceeded the achievements of 2000. Such developments reflected the growing number of communication channels exploited by central banks to communicate with the public, not merely relying on periodic publications but also uploading data and information to websites to expand public access to information. Nevertheless, the assessment in 2003 also revealed a number of weaknesses, particularly in terms of transparent practices at central banks. Similar to the previous assessment, central bank practices and integrity remained the weakest link of monetary policy transparency. A cross-border study on monetary policy transparency was conducted by Sundarajan et al. (2003) based on the results of a survey disseminated to 160 central banks of the 178 IMF members. The participation rate of the survey was comparatively high, with 75% of the surveyed central banks completing the questionnaire, representing a broad range of economic and political development. Transparency was assessed based on various elements of the four transparency aspects contained in the MFP Transparency Code. Specifically, the study proposed an index based on the average implementation of four monetary policy transparency dimensions, namely the means, timeliness, periodicity as well as form and content of the information submitted.4

4

For a more detailed description of the method used to calculate the index, refer to Appendices I and II in Sundarajan et al. (2003).

372    Central Bank Policy The degree of transparency per the country classifications based on economic development and political conditions is presented in Table 12.1. The table shows that monetary policy is more transparent in higher-income countries, including OECD members and advanced countries. Transparency concerning the role, jurisdiction, and purpose of the monetary policy was adequate, however, in most countries. Most countries displayed weaknesses in terms of monetary policy authority accountability and integrity. Conspicuous differences were found, however, in terms of disclosing the policy formulation process and public availability of information on monetary policy, which are typically lower in low-income and developing countries.

12.4.2. Practices at Several Prominent Central Banks Blinder et al. (2001) evaluated monetary policy transparency practices at several central banks located in advanced countries, specifically the Federal Reserve, ECB, BOJ, Bank of England, Bank of Canada (BoC), Reserve Bank of New Zealand (RBNZ), and Swedish Riksbank. Aspects of transparency were evaluated, including economic forecasts and the direction of monetary policy, information disclosure concerning monetary policy decisionmaking, as well as communication with and accountability to parliament. All central banks in advanced countries publish periodic information regarding monetary policy and economic developments, encompassing economic and monetary policy reports as well as macroeconomic projections. In general, such information is reported on a quarterly and monthly basis or as required, while the macroeconomic forecasts are published in brief. Only the RBNZ and Swedish Riksbank publish detailed macroeconomic projections. Variations were found between the seven central banks, however, in terms of publishing information on monetary policy decisionmaking. In general, the central banks in question announced their interest rate decisions immediately after the decision had been made, accompanied by an explanation of the considerations underlying any changes in the decisions taken. There were differences, however, in the publication of records from the monetary policy meetings held at the central banks, including the minutes and verbatim records. The Fed, BoJ, BoE, Swedish Riksbank, and RBNZ published their minutes, while the other central banks did not. On the other hand, prevailing laws in advanced countries required the respective central bank to submit a monetary policy report to parliament, albeit at different frequencies (Table 12.4). Likewise, the central banks were required to provide clarification to parliament periodically, through a, generally open, plenary hearing or commission, by the Governor of the Central Bank or other senior official. In general, however, parliament or the commission were not authorized to appraise central bank performance or dismiss the Governor and/or members of the Board of Governors for the decisions taken.

74.6 78.0 73.9 92.1

• Low-middle income

• Upper-middle income

• High income, non-OECD

• High income OECD

72.6 81.3 77.6

• Developing

• Transition

Total

Source: Sundarajan et al. (2003).

90.4

• Advanced

Economic/political characteristics

72.0

• Low income

Stage of economic development

83.3

85.8

80.4

90.2

90.3

86.9

82.1

81.0

80.9

65.3

78.5

55.6

85.0

88.6

57.1

63.9

59.9

58.6

68.6

66.7

64.7

82.5

84.6

61.9

72.4

66.4

60.2

59.8

60.2

57.5

66.1

66.7

58.2

56.2

57.8

59.6

Implementing All Clarity Concerning the Monetary Policy Public Central Bank Elements of the Role, Jurisdiction, and Formulation, and Availability of Accountability MFP Transparency Goal of the Monetary Reporting Process Information on and Integrity Code Policy Authority Disclosure Monetary Policy

Table 12.1:  Monetary and Financial Policy Transparency Index by Country.

Policy Transparency and Communication Strategy    373

374    Central Bank Policy 12.4.3. Monetary Policy Transparency and Economic Performance As a contemporary issue, most of the literature available concerning monetary policy transparency focuses on the theoretical analysis, as reviewed in Section 12.2. Not many studies explore the impact of monetary policy transparency on economic performance. The few studies available focus on the financial markets using time series data. Nonetheless, Chortareas, Stasavage, and Sterne (2002) conducted an empirical study concerning the impact of monetary policy transparency on inflation and output using data from 87 countries. Their methodology and results are described as follows. 12.4.3.1. Methodology.  In their study, Chortareas et al. (2002) focused on the publication of economic forecasts by the central bank and its impact on inflation and output volatility. The hypothesis constructed from the nascent theories developing in the literature revealed two important aspects. First, that publishing central bank forecasts improved inflation performance, considering that a greater degree of transparency increased the sensitivity of the central bank’s reputation to the actions taken, which therefore reduced the incentives for the central bank to implement inflationary policy. Second, that publishing central bank forecasts reduced output variability, considering that with a greater degree of transparency, the central bank underscored the achievement of low inflation and, therefore, limited its ability to stabilize the economy, for instance, using surprise policies to overcome demand-side shocks in the economy. The study utilized the results of a survey conducted by Fry, Julius, Mahadeva, Roger, and Stone (2000) of central banks and monetary authorities from 94 countries. The survey questioned various aspects of transparency, such as the publication and clarification of macroeconomic forecasts, explaining the monetary policy formulation process, analysis quantity, research activity and the speeches of central bank officers. The survey also assessed various aspects of accountability, independence, and target setting, which contribute to monetary policy framework clarity and transparency. Chortareas et al. (2002) emphasized the publication and clarification of macroeconomic projections by the central bank as the aspect most closely related to various other studies on monetary policy transparency. Most of the central banks surveyed (79% of 94 central banks) published some form of forecast or forward analysis, although the form and scope varied. Most clarified the direction of the economy, such as economic growth and its components, inflation, interest rates, exchange rates, and monetary aggregates. Some of the central banks published money supply targets as part of their forward analysis as intermediate targets rather than operational targets. Some also provided a more detailed explanation of the risks contained in the forecasts and the inaccuracies found in previous projections. The profiles of the surveyed central banks, reviewed in terms of the types of forecasts published and quality of the forward analyses, are presented in Table 12.2. To measure the degree of central bank transparency in terms of publishing economic forecasts, Chortareas et al. (2002) applied a Guttman scale. Fundamentally, the index applied a dichotomous format to four questions concerning

Policy Transparency and Communication Strategy    375 Table 12.2:  Economic Forecast Publication Profile of Central Banks. Questions Form of publication of forecasts

Forward-looking analysis in standard bulletins and reports

Discussion of past forecast errors

Categories of All Industrial Transition Developing Answers Countries Words and numbers

35

16

 5

14

Either words or numbers

25

 8

 6

11

Unspecified

13

 0

 4

 9

None

21

 4

 7

10

More than annually

39

18

 7

14

At least annually

24

 4

 4

16

Unspecified

10

 2

 4

 4

Otherwise

21

 4

 7

10

Yes

21

 8

 3

10

Sometimes

 9

 7

 2

 0

No

64

13

17

34

 9

 7

 2

 0

Either words or numbers

23

 9

 4

10

None

62

12

16

34

Risks to forecast Words and published numbers

Source: Chortareas et al. (2002, p. 103).

forecast publication transparency. Therefore, the index would produce a higher value for a central bank that practices all four aspects of transparency mentioned in the survey. The index of central bank transparency in terms of publishing economic forecasts based on the Guttman scale is presented in Fig. 12.1. Of the 82 countries assessed, 25 produced a Guttman scale of 0, while eight had a scale of 1, 24 had a scale of 2, six had a scale of 3 and 19 had a scale of 4. The effect of transparency on inflation and output variability was analyzed in terms of the correlation to each transparency question. To control the effect of other variables, transparency was also analyzed using a regression model that included the influence of GDP, degree of economic openness, exchange rate system, announcement of the inflation and money supply targets, and degree of political instability. In addition, to analyze the effect of economic, political and legal conditions, the impact of transparency was measured using indicators of democracy, GDP per capita, central bank independence, and central bank analysis quality.

376    Central Bank Policy

Fig. 12.1:  Measuring the Transparency Index Using the Guttman Scale. Source: Chortareas et al. (2002, p. 105). (1) Transparency and inflation: Correlation between the Guttman scale for the four transparency questions and the publication of economic forecasts by the central bank is presented in Table 12.3. From the table, we can see that inflation correlates inversely and significantly with all transparency indexes. This is empirical evidence that a greater degree of monetary policy transparency could control inflation. The strongest correlation was recorded for transparency in terms of publishing forecasts and clarifying the risks contained within. Regression analysis further confirmed the empirical evidence that transparency could improve inflation performance.5 A stronger impact was recoded for countries that implemented a flexible exchange rate system. In such countries, where inflation began in double digits at around 12% per annum, the central bank’s decision to publish routine inflation forecasts (a bump in the index from 0 to 2) triggered disinflation of around 1.8%–7% per annum. Conversely, in countries that implemented a fixed exchange rate system, the effect of transparency on disinflation was minimal, with inflation moderating slightly from 12% to 11.8% per annum. (2) Transparency and output volatility: The correlation between transparency and output volatility was also analyzed using the standard deviation of GDP, as presented in Table 12.4. From the table, we can see that the correlation between transparency and output volatility is generally negative, especially in terms of quarterly output volatility, but with a comparatively low level of significance. Although this analysis fails to provide strong evidence that transparency could reduce output volatility, there is also no evidence that greater monetary policy transparency would exacerbate output volatility. The regression analysis of the transparency index against output volatility, including the variable, terms of trade, also failed to provide conclusive evidence that transparency correlated with output volatility. 5

Refer to Chortareas et al. (2002, p. 107). In addition to the transparency index, the variables log GDP per capita, openness, political instability, dummy exchange rate peg (model 1) were included in the regression model, along with pegxtransparency (model 2) as well as inflation targetxtransparency and money targetxtransparency (model 3).

Policy Transparency and Communication Strategy    377 Table 12.3:  Transparency and Inflation: Correlation Analysis. Log Inflation Guttman scale of transparency

−0.37 (p 0) are useful to bridge the gap between savings deposits and investment (S − Y < 0) to stimulate economic growth. Meanwhile, in advanced countries that typically enjoy a current account surplus (CA > 0), foreign capital outflows to capitalize on excessive savings deposits (S − I >) to invest in another country would also stimulate economic growth. Based on the neoclassical theory, where countries face a constant return to scale, homogenous capital and labor factors, coupled with truly efficient and open financial markets, capital flows from rich countries to poor based on the law of diminishing returns so that the marginal product of capital is higher in poorer countries. For instance, the production function follows Cobb-Douglas: y = Ak β , where y is income per worker and k is capital per worker, then the marginal product of capital is: r = Aβ k β−1 or equivalent to r = β A1/ β k ( β −1)/ β if expressed as production unit per worker. Therefore, the influx of foreign capital from advanced countries, such as the United States, to EMEs, such as Indonesia, shows that the return on capital investment is higher in Indonesia than in the US, which is mutually beneficial to both economies. Several assumptions underlie neoclassical theory to better understand the behavior, benefits and risks of foreign capital flows. First, foreign capital flows are required for investment in the real sector, supported by high productivity, thus catalyzing national economic growth. This implies that foreign capital flows play a complementary (rather than substitutionary) role in terms of domestic savings for economic financing. It is also possible that a country can still capitalize on foreign capital inflows, while simultaneously investing a portion of savings offshore as long as productivity is increased, thus driving real economic growth. In short, the propitious impact of foreign capital flows on investment and economic growth will be enjoyed if supported by productivity, a favorable investment climate and other factors of competition in the receiving country. Second, the financial sector in the receiving country has developed efficiency to transmit the foreign capital flows into real sector economic financing. In the case of FDI, the investment climate must draw foreign capital flows and channel them

390    Central Bank Policy to productive sectors in the economy. In terms of PI and external debt, domestic financial institutions and financial markets are sound conduits in the form of credit instruments, bonds or stocks for economic financing. Regarding government bonds, financing the fiscal deficit would not crowd-out the financial markets or investments in the real sector. More efficient transmission leads to greater benefits from foreign capital flows, implying that most of the foreign capital can be channeled to the real sector and only a little, as excess liquidity, will circulate and be traded in the financial sector. In brief, financial market deepening in terms of the instruments, setting interest rates, liquidity, and infrastructure are crucial to channel foreign capital flows to productive economic financing. Third, the foreign exchange system implemented in the receiving country could facilitate foreign capital inflows (perfect capital mobility), especially useful for productive economic financing with low volatility. In other words, current account liberalization (CAL) must be carefully designed in terms of the order, scope, and timing to achieve both goals. This is not a simple matter because the preliminary conditions in countries are very diverse. Countries still applying foreign exchange controls enjoy more freedom to design the order, scope, and timing of CAL policy. This is especially true of low-income developing countries. Nevertheless, many EMEs have already implemented a freer foreign exchange system and, therefore, are no longer in a position to optimally design CAL policy. For such countries, the best option would be CFM to optimize the benefits for economic financing and mitigate the adverse effects of the volatility that may appear. It is more appropriate to apply neoclassical theory to the analysis of foreign capital flows that successfully boost productivity and are generally longer term, such as FDI and long-term external debt. Meanwhile, short-term foreign capital flows are typically determined by the return and risks of a particular investment. Such analyses can be applied to flows of PI on equity, securities, bonds and other financial securities. Theoretically, foreign PI is based on the International Capital Asset Pricing Model (IAPM), which optimizes the portfolio of domestic and global financial assets in line with the return profile and level of risk: E ( RP ) = RF + βW RW + ∑ i =1 δi Si  k



(4)

where E(RP) = expected return of the investor’s portfolio, RF = return on riskfree assets (such as the interest rate on US T-Bills), Rw = global benchmark PI return and Si = magnitude of change in the exchange rate. Forming a global portfolio is based on optimization per Markowitz theory that produces an Efficient Frontier as follows: RW = ∑ i =1 ωi Ri k

therefore :

σW2 = ωi2 σ 2j + 2∑ i =1 k

∑

l j =1

ωi ω j σi σ j σij 

(5)

where ωi = share of asset i in the global portfolio, Ri = return on financial assets in country i, and σ2 = risk associated with financial asset i. From the theory, we can see that international PI could prompt greater diversification beyond the financial assets of the home country. Therefore, to optimize

Monetary Policy and Foreign Capital Flows    391 the portfolio return, investors will also include financial assets from EMEs, which offer a return and contain risks that do not correlate with the financial assets in the investor’s home country (advanced country). Due to the cross-border nature of the portfolio, investors will pay due regard to exchange rate fluctuations, in addition to the returns on the financial assets. How far the investor considers the return and the exchange rate would depend on the motive of the investor. Longterm investors (real money) will prioritize the yield and exchange rate stability, while short-term investors (carry trade) will maximize profit from exchange rate fluctuations. Furthermore, due to asymmetric information and other factors, there is a tendency for investors to perpetuate a home bias rather than a foreign bias.

13.2.2. Determinants of Foreign Capital Flows: Push and Pull Factors From the two theories mentioned previously, international financial integration, or financial globalization, can be measured by the magnitude of cross-border capital flows. By this measure, the scale of global foreign capital flows has continued to increase since the end of World War II, accelerating in the 1990s and then really taking off in the final decade prior to the GFC. Measured as a percentage of GDP, foreign capital flows have increased from 5.7% in 1980–1989 to 6.2% in 1990–1999 before spiking at 13.3% in 2000–2007 and then falling dramatically back to 6.2% in 2008–2012 (Table 13.1). In terms of composition, PI and bank funds have exploded, while FDI has accelerated more gradually. From Table 13.1, flows of PI and bank funds have increased from 2.3% and 1.9% of GDP, respectively, in 1990–1999 to 4.2% and 5.0% in 2000–2007, before dropping back to 1.4% and 0.4% of GDP in 2008–2012. Meanwhile, after accelerating robustly in the previous two periods, FDI flows have remained relatively stable at 2.9% of GDP in 2000–2007 and in 2008–2012. Congruent with the global economic developments, foreign capital flows to EMEs have also posted significant gains since 2004. After plummeting in 2008 due to the GFC, foreign capital inflows subsequently skyrocketed until 2013 before fading due to the Fed’s Taper Tantrum. In fact, net foreign capital flows to EMEs halved from USD 1,074 billion in 2014 to just USD 548 billion in 2015 (Institute Table 13.1:  Global Foreign Capital Flows (% of GDP). Annual Average 1980–1989

1990–1999

2000–2007

2008–2012

FDI

1.0

1.5

2.9

2.9

Portfolio investment

1.2

2.3

4.2

1.4

Other investment

2.7

1.9

5.0

0.4

Reserve assets

0.8

0.5

1.2

1.5

Total

5.7

6.2

13.3

6.2

Source: James, McLoughlin, and Rankin (2014).

392    Central Bank Policy of International Finance, 2015). As a percentage of GDP, foreign capital inflows accounted for just 2%, down from the peak of 8% in 2007. Sharp declines were recorded in terms of PI and OI, particularly the flow of bank funds, while FDI has remained comparatively stable. Furthermore, foreign capital outflows from EMEs have also increased because of global financial market turbulence caused by US monetary policy uncertainty. Such conditions exacerbated pressures on reserve assets, exchange rates, and asset prices in EMEs. Consequently, in 2015, foreign capital flows to EMEs were negative for the first time since 1988, recording a net outflow totaling USD 540 billion. The developments described above raise an important question: which factors influence foreign capital flows to EMEs? The literature review revealed a number of global push factors and domestic pull factors (Koepke, 2015). The push factors include global economic growth, international interest rates, and the risk premium in global financial markets, while the pull factors include domestic economic growth, investment returns (such as interest rates, bond coupons, and stock dividends), exchange rates, and domestic risk indicators. In general, the relationship between foreign capital flows and the push and pull factors can be expressed as follows;

Ft = α + β X t + δYt + φ Zt 

(6)

where Ft = foreign capital flows (FDI, PI, and OI), Xt = push factor vector, Yt = pull factor vector, and Zt = other factors, such as the foreign exchange system, the depth of the financial markets and institutional conditions (legal, political, and so on). Numerous empirical studies have been performed to show which factors were more dominant in terms of influencing foreign capital flows to EMEs, with a different focus and findings. After foreign capital flows to EMEs accelerated in the wake of the crisis in México in the 1980s, for instance, Ghosh and Ostry (1993) proved that fundamental factors were more dominant in relation to the inflation stabilization program, privatization and capital market liberalization undertaken in EMEs. Meanwhile, Calvo, Leiderman, and Reinhart (1992) focused on the cyclical decline in the United States as the most dominant factor. Nevertheless, the continuation of foreign capital inflows to EMEs during the US economic recovery in the 1990s evidenced structural factors as the most dominant. The World Bank (1997) referred to this structural change in terms of the proliferation of institutional investors in advanced countries, financial market deregulation in EMEs and the impact of information technology. Thereafter, in the 2000s, Baek (2006) showed that the interest rate and risk premium were the most dominant push factors in terms of explaining PI trends. Meanwhile, De Vita and Kyaw (2008) demonstrated that pull factors, particularly productivity and competitiveness, were the dominant factors of FDI inflows to EMEs, especially Emerging Asia. Several empirical studies have also explored the relative effect of push and pull factors for each type of foreign capital flow to EMEs. Of the three types of foreign capital flow, FDI was least affected by global economic fluctuations. Of the pull factors, domestic economic growth was the biggest draw for FDI. Meanwhile, asset yield indicators on financial markets (such as interest rates and dividends) were not an important factor because FDI is typically long term. In fact, aspects

Monetary Policy and Foreign Capital Flows    393 of governance, including the investment climate, legal assurance, and political stability were significant FDI pull factors. Likewise, exchange rate stability was another important pull factor. Concerning the push factors, several empirical studies have shown that global economic growth was not always a significant factor. The positive impact of world economic growth was shown to be significant for vertical FDI, namely investment as part of the global or regional production chains, but not significant for horizontal FDI, namely to satiate the domestic markets. In terms of international interest rates, most empirical studies found an insignificant effect. Meanwhile, the impact of global risk indicators on FDI inflows was inconclusive, with some studies showing a positive effect of the Volatility Index (VIX) and others evidencing the adverse impact of country risk indicators.1 In terms of PI to EMEs, the push factors were more dominant. All studies showed that PI to stocks and bonds in EMEs was significantly, adversely, and immediately affected by changes in the perception of global risk, as measured by the VIX or CDS spread.2 Similarly, international interest rates (usually based on US interest rates) had an adverse and more dominant impact on bonds than stocks. In relation to the effect of economic growth in advanced countries, not many studies have demonstrated a significant direct impact, except improving the probability of PI inflows to EMEs (Forbes & Warnock, 2012). Of the pull factors, nearly all studies have shown domestic economic growth performance as an important pull factor, but the elasticity declines as the data frequency increases. Likewise, yield indictors were a critical pull factor, especially stock and bond yields. The monetary policy rate has not always been shown as an important factor because the effect has already been detected through bond yields and stock prices, except when the monetary authority implements a change of policy stance (Ahmed & Zlate, 2014), while real exchange rate volatility was shown to have a negative impact (Baek, 2006; World Bank, 1997). Indicators of external vulnerability also had an inverse impact, reflected by the external debt to GDP ratio (World Bank, 1997) and sovereign credit rating (Kim & Wu, 2008). Foreign capital flows through the banking system (banking flows) were shown to respond negatively to the perception of global risks, along with interest rates in advanced countries, as the push factors. The impact was subsequently strengthened, however, by the risk-taking channel, where risk perception could exacerbate the impact of interest rates during periods of high global financial market volatility. Meanwhile, the influence of economic growth in advanced countries on banking flows proved inconclusive. Of the pull factors, strong evidence was provided for the effect of

1

VIX is an abbreviation for the Chicago Board of Exchange (CBOE) Volatility Index as a measure of volatility implied by Standard & Poor (S&P) index options. Colloquially referred to as the “fear index,” VIX measures market expectations of stock market volatility for the upcoming 30 days. 2 CDS or credit default swap is a swap agreement where the CDS seller will compensate the buyer (in general, a creditor or bondholder) in the event of default by the borrower or bond issuer. Non-resident investors that invest in the bonds of EME can purchase CDS to hedge against default risk.

394    Central Bank Policy domestic economic growth, yields and risk indicators on banking flows to EMEs. The positive impact of yields was shown through various indicators, such as stock yields, exchange rate appreciation and, specifically, banking sector stock performance. Risk indicators, such as a decline in the external debt to GDP ratio or upgraded sovereign rating were also shown to significantly increase banking flows to EMEs. In summary, the effect of push and pull factors on foreign capital flows to EMEs is presented in Table 13.2. In addition to the studies mentioned above, a number of empirical studies have also explored the effect of international investors’ behavior on foreign capital flows to EMEs. Calvo and Reinhart (1996), for instance, showed that contagion could occur due to declines in domestic vulnerability indicators combined with the flight-tosafety phenomenon on global financial markets. The effect of contagion originates from changes in the behavior of investment managers due to changes in investment perception by institutional investors. Another study examined the impact of changes in investor behavior in terms of adjusting financial asset portfolio allocation due to yield indicator performance on domestic financial markets, otherwise known as rebalancing versus return chasing (Curcuru, Thomas, Warnock, & Wongswan, 2011). Meanwhile, the effect of asymmetric information and transaction costs were not detected in the analysis of push and pull factors (Portes & Rey, 2005). The empirical studies have provided important inputs in terms of understanding the determinants of various types of foreign capital flows to EMEs. In general, FDI is more influenced by domestic economic fundamentals, while global factors have a greater impact on foreign capital flows in the form of PI and banking flows. Nevertheless, the relative effect of push and pull factors can change due to short-term developments in the global economy. Fratzscher (2012), for instance, showed that push factors were more dominant in terms of PI during the GFC, while PI was pushed more by macroeconomic fundamentals, institutional Table 13.2:  Determinants of Foreign Capital Flows: Push Versus Pull Factors. Portfolio Equity

Portfolio Debt

Banking Flows

FDI

Negative strong

Negative strong

Negative strong

Inconclusive

Interest rates in advanced countries

Negative strong

Negative strong

Negative

Inconclusive

GDP in advanced countries

Positive

Positive

PULL Domestic GDP

Positive

Positive

Positive strong

Positive strong

Domestic yields

Positive

Positive

Positive strong

Inconclusive

Sovereign credit rating

Negative

Negative

Negative strong

Negative

PUSH Global risk perception

Inconclusive Inconclusive

Monetary Policy and Foreign Capital Flows    395 factors and EME policy in the period following the GFC. Such conditions confirm the importance of maintaining economic fundamentals as well as instituting an appropriate and prudent macroeconomic policy response to maximize the benefits and minimize the risks of foreign capital flows.

13.3. Foreign Capital Flows and Economic Performance There is a general belief, as explained by neoclassical theory, that foreign capital will flow from rich countries to poor and stimulate economic growth in both. Nonetheless, the empirical evidence does not always concur. In fact, during the GFC, foreign capital flows, which had previously flooded EMEs, reversed back to advanced countries. Despite recording gains since the GFC, foreign capital flows to EMEs and developing countries experienced a net outflow in 2015. There was also disparity among the EMEs and developing countries, namely most foreign capital flowed to Emerging Asia (predominantly China) and only a small portion went to Africa, which contradicts the neoclassical prediction that foreign capital flows would flow to countries with additional productivity from higher levels of capital per worker, in this case Africa, followed by Asia and then advanced countries.

13.3.1. Empirical Phenomenon: “Lucas Paradox” Critical observations of cross-border foreign capital flows that contradict the neoclassical theory were raised by Lucas (1990), which subsequently became known as the Lucas Paradox. Namely, that globalization and cross-border financial international do not always benefit economic performance in developing countries and EMEs or the global economy. This issue has been more enthusiastically debated internationally since the GFC in relation to the global economic rebalancing agenda between advanced and developing countries. Lucas (1990) showed that average income (relative to the US) in net exporters of capital has declined well below that of net capital importers. This implies that capital indeed flows from poor countries to rich, not “downwards” but “upwards.” The combination of a US current account deficit and current account surplus in China is an important cause of why capital flowed upwards. Nonetheless, many advanced countries run a deficit and, similarly, many developing countries maintain a current account surplus. This is not a new phenomenon because similar trends were observed in the 1980s. As explained by neoclassical theory, capital should flow to developing countries and EMEs that achieve robust growth and, therefore, are attractive draws for investment. Prasad, Rajan, and Subramanian (2006) analyzed foreign capital flows and economic growth trends in 59 developing countries for the period from 1970 to 2004.3 During this period, many advanced countries were experiencing a

3

Growth trends were grouped from low to high and the 59 developing countries were divided into three categories with a balanced population, with China and India the only exceptions. After including the current account deficits, foreign capital flows of USD were calculated in real terms using CPI.

396    Central Bank Policy

Fig. 13.1:  Aggregate GDP per Capita of Capital Exporters (Weighted Current Account). Source: Prasad et al. (2006). current account deficit, while many developing countries and EMEs were enjoying a current account surplus. As illustrated by Fig. 13.1, advanced countries (deficit countries) actually posted increases in income per capita, while the developing countries and EMEs (surplus countries) suffered declining income per capita. This implies, contradicting neoclassical predictions, that developing countries and EMEs, which were more dependent on foreign capital flows, did not achieve stronger economic growth in the long term. Furthermore, among the developing countries and EMEs, as illustrated in Fig. 13.2 during the period from 1970 to 2004 net foreign capital flows to the group of countries with the highest economic growth was actually lower than to the low- and middle-income growth countries. The high-growth group of EMEs, averaging of 4.1% (yoy), received foreign capital flows totaling around USD400 million, while the medium- and low-growth countries, averaging 0.3% (yoy) and 1.6% (yoy), respectively, received around USD 900 billion and USD 600 billion in foreign capital flows. For a shorter observation period, the results were even more stark, with China and India as well as high and medium-growth EMEs all exporting capital to advanced countries, while the low-growth EMEs received significant foreign capital flows. Therefore, the flow of foreign capital contradicted the growth trend predicted by the neoclassical theory.

13.3.2. Important Takeaways from the Lucas Paradox The Lucas paradox might be less controversial if analyzed in more depth. There were various issues discovered in developing countries and EMEs that meant the risk-adjusted return was not always the same as the measures of marginal productivity of capital and labor. In general, there are three explanations for the Lucas paradox. First, fundamental factors linked to different production structures

Monetary Policy and Foreign Capital Flows    397

Fig. 13.2:  Foreign Capital Flows to EMEs and Developing Countries (Annual Average in Billions of USD). in the respective economies of each country. Second, domestic financial market deepening in terms of mobilizing savings and financial intermediation for productive financing in-country. Third, imperfect global financial markets, including asymmetric information and home-bias investors as well as disparate sovereign ratings and credit default risks. 13.3.2.1. Economic Fundamentals.  The production and institutional structures in a particular country could create disparity between investment returns and the measures of marginal productivity of capital and labor. Several factors play an important role in the production structure, including technology, infrastructure, and other production factors, as well as government policy, such as tariffs, taxes, FX controls, and other impediments to investment, which affect total productivity but are not reflected in the measures of capital and labor productivity. Similarly, the institutional structure, including legal assurance and political stability, also influences the returns on investment from foreign capital flows. Other than capital and labor, another important production factor is human capital as a factor of education, technology as well as research and development. Lucas (1990) included human capital in the production function, expressed as: y = Ak β h γ , where y is income per effective worker and h is human capital per worker. Basically, human capital was interpreted as a labor productivity multiplier at all skill levels, such as its function as a multiplier of total factor productivity. By including human capital, Lucas (1990) estimated the ratio of investment returns in India and the US. The results showed that the ratio of returns, which had previously been 58 times higher, was nearly the same if human capital was taken into consideration. Despite acknowledging that measuring human capital and the linkages with productivity was not easy, the analysis showed that one cause of the paradox of foreign capital flows and economic growth was a production factor that omitted land, human capital, and so on.

398    Central Bank Policy The institutional structure of each respective country also affected the desired returns (ex ante) and the actual returns (ex post). Although a number of technologies are available to various countries, implementation constraints, or the inefficient utilization of such technology have a significant impact of the rate of return of capital investment. Under such circumstances, similar yields of capital productivity per worker, measured using Ait f 1 ( kit ) = r = Ajt f 1 ( k jt ), do not mean that the capital will flow from rich countries to poor. One important contributing factor of total factor productivity in a country is institutional quality. The study by Alfaro, Kalemli-Ozcan, and Volosovych (2011) provided empirical evidence that institutional quality was an important factor to explain the Lucas paradox, after omitting other factors such as education, income per capita and constraints to capital flows. Likewise, Ju and Wei (2006) demonstrated the importance of financial system quality and the protection of copyright laws in terms of explaining the Lucas paradox. Such rationale and empirical evidence proves that policy to strengthen political stability, improve bureaucratic quality, ensure copyright protection, eradicate corruption, and strengthen law enforcement are critical factors to increase investment and productivity from foreign capital flows in terms of stimulating economic growth. 13.3.2.2. Domestic Financial Sector Deepening.  The domestic financial sector plays a vital role in mobilizing savings and stimulating foreign capital flow intermediation efficiently for productive financing of economic growth. One neoclassical assumption is the close relationship between savings and investment, in terms of the total and productivity, to stimulate production gains and economic growth. The function of foreign capital flows should be complementary (not a substitution) to the savings and investment gap. If the domestic financial sector works efficiently, public funds will be mobilized optimally with an interest rate (after calculating the normal intermediation cost) that is relatively the same as the return on investment. Such conditions draw foreign capital flows to the country involved. This rationale underlies the theoretical and empirical study performed by Gourinchas and Jeanne (2013), who explained the Lucas paradox due to a misallocation in the relationship between savings, investment and economic growth. Two causes could occur, namely an investment wedge or a saving wedge. The investment wedge encompasses the various distortions that cause non-resident investors to receive a lower return than the return on investment in the country involved due to the cost of bureaucracy, corruption, and barriers to the flow of foreign exchange. Meanwhile, the saving wedge encompasses various factors that distort the public’s propensity to save in order to maximize their current and future levels of consumption, such as banking interest rate regulations and limited financial products for investment. The theoretical model developed was subsequently calibrated using data from 68 developing countries and EMEs for the period from 1980 to 2000 to analyze the relationship between net foreign capital flows, productivity, the investment wedge, and the saving wedge. The study produced a number of interesting findings. First, consistent with the Lucas paradox, countries with a net inflow of foreign capital tended to have lower

Monetary Policy and Foreign Capital Flows    399 total factor productivity.4 Second, countries with a high investment wedge tended to enjoy a productivity catch-up to the low average level of productivity of all 68 countries. In other words, the investment wedge tended to exacerbate the misallocation of foreign capital flows in terms of productivity catch-up, as demonstrated by the Lucas paradox. Third, countries with a high level of productivity catch-up were those that could mobilize domestic savings, while those countries that could not mobilize savings would remain left behind in terms of productivity. Therefore, the ability to optimally and efficiently mobilize domestic savings was shown to be a leading determinant of cross-border foreign capital flow misallocation and the ability of developing countries to boost investment in order to improve productivity and stimulate economic growth. The study emphasized the importance of overcoming the various distortions in the relationship between savings and investment, especially in the domestic financial sector. A shallow financial sector could limit gains in terms of domestic savings and investment to stimulate economic growth. Therefore, the public’s propensity to save for future consumption would also be restricted. Similarly, inefficient financial intermediation would make investment less responsive to productivity and economic growth. Misallocation could occur if the banks guaranteed a company with low productivity. A shallow financial system could encourage the public to save excessively for precautionary motives rather than for productive investment purposes. 13.3.2.3. Imperfect International Finance. Imperfect international finance creates deviation between the actual and desired capital returns, thus engendering distortions between foreign capital flows, investment, and economic growth. Obstfeld and Rogoff (2010), for instance, showed that the Lucas paradox occurred because of the relationship between international trade and finance. Consequently, Ait f 1 ( kit )(1− τit ) = r = Ajt f 1 ( k jt )(1− τit ), where t reflects tax, foreign exchange controls, tariffs, transportation costs, and other constraints. Likewise, Razin, Rubenstein, and Sadka (2003) explained the large start-up cost as a cause of distortions in terms of foreign capital flows. That cost means investment in small firms becomes unfeasible despite achieving marginal productivity that could generate an attractive return for foreign capital flows. Asymmetric information could also impede foreign capital flows despite high returns on investment, therefore investors preferred investing in their home country and in assets or countries that offer sufficient investment alternatives, a phenomenon known as home bias. Ahearn, Griever, and Warnock (2004) conducted a study on investor behavior in the United States in terms of offshore equity investments and proved that countries in which the corporate sector would fail to meet US regulations tended to receive a smaller portfolio allocation (underweight). Concerning FDI), a study by Levis, Muradoglo, and Vasileva (2007) showed that

4

South Korea, for example, which achieved average TFP growth of 4.1% per annum and average rate of investment of 34% per annum, was not a recipient of large foreign capital flows, while Madagascar, with average TFP growth of 1.5% and average investment of less than 3% received foreign capital flows to the tune of around 7% of GDP.

400    Central Bank Policy corporate investors favored neighboring countries (home bias) or countries in the same economic union or with the same legal system. In other words, investors are more confident and upbeat about investing in countries that they already know despite the chance of higher returns in other countries. Default risk is another factor that distorts foreign capital flows from investment congruent with the level of productivity. Reinhart and Rogoff (2004) explained default risk as the main reason why not much foreign capital will flow from advanced countries to poor countries. In effect, countries that experience default, generally the poorer countries, will find it difficult to borrow offshore. Foreign capital flows in the form of FDI and PI would also be low due to the negative reputation of the country after defaulting on a loan. This evidence is robust in terms of explaining the Lucas paradox. In other words, the problem is not so much the flow of foreign capital from rich countries to poor, but too much foreign capital flowing to countries with high default risk.

13.4. Foreign Capital Flows and Monetary Stability Capital flow volatility, especially PI and short-term external debt, may induce risks to monetary, financial system, and macroeconomic stability in general. The risks are transmitted through various channels. Foreign capital flows directly influence supply and demand on the foreign exchange market and, therefore, capital flow volatility would directly affect exchange rate volatility. Capital flows as PI also influence the development of domestic financial asset prices, including stock prices and bond yields. Similarly, the impact of liquidity from capital flow volatility would exacerbate liquidity risk and undermine banking industry capacity to extend credit for the economy. Furthermore, capital flow volatility and the risk-taking behavior of international investors could intensify systemic risk in the domestic financial system.

13.4.1. Empirical Studies on the Effect of Capital Flow Volatility Various empirical studies have shown the adverse effect of capital flow volatility on monetary, financial system and macroeconomic stability.5 Fuertes, Phylaktis, and Yan (2014), for instance, showed that capital flow volatility to EMEs was primarily attributable to flows of PI and short-term foreign loans, otherwise known as hot money. The study utilized monthly PI flow data (stocks and bonds) from the US to 18 EMEs (nine in Asia and nine in Latin America) for the period from 1988:1 to 2012:12, as well as quarterly data on banking flows for the same period. A Kalman filter was also used to separate temporary and short-term capital flows from permanent and long-term capital flows.

5

Box 13.1 also contains several empirical studies that confirmed an increase in capital flow volatility due to global spillovers in the wake of the global financial crisis in 2008/2009, including the adverse impact of monetary policy uncertainty from the US Federal Reserve.

Monetary Policy and Foreign Capital Flows    401 The empirical results of the study showed that flows of bank loans and PI from the US to EMEs in the 1990s were more permanent and long term, later usurped by temporary and short-term flows. The empirical findings were made even more robust with the inclusion of various push and pull factors, combined with other temporary and permanent decomposition techniques. The increasing domination of hot money contributed to the Asian Financial Crisis of 1997/1998 and served as the main conduit of global spillovers from the US subprime mortgage debacle that subsequently unfolded into the GFC, of which the adverse effects quickly spread to EMEs. Consequently, various macroeconomic and macroprudential policies were required, along with foreign CFM, to mitigate the harmful impact of short-term PI and bank loans on macroeconomic and financial system stability. In a different study, Caporalea, Ali, Spagnolo, and Spagnolo (2015) provided empirical findings for the effect of equity portfolio flows (stocks and bonds) on exchange rate volatility in EMEs. The study was based on bilateral monthly data between the US and EMEs as well as developing countries in Asia (India, Indonesia, South Korea, Pakistan, Hong Kong, Thailand, the Philippines and Taiwan) for the period from 1993:01 to 2012:11. The results showed that such portfolio flows exacerbated exchange rate volatility. Moreover, using a Markov-switching model, the study also showed that a net stock (bond) portfolio flow precipitated high (low) exchange rate volatility. Specifically, bond portfolio flows raised the probability of low volatility in Pakistan, Thailand, and the Philippines, while in Indonesia the level of volatility remained high. Meanwhile, stock portfolio flows were shown to increase the probability of high volatility in India, Indonesia, South Korea, Hong Kong, and Taiwan. The empirical evidence provided by that study reinforced the return-chasing hypothesis for international investors and, consequently, foreign CFM of PI has become an effective instrument to stabilize the exchange rate. Ananchotikul and Zhang (2014) also conducted a study regarding the effect of portfolio flows and global risks on asset prices in 17 EMEs using weekly portfolio flow data from the Emerging Portfolio Fund Research (EPFR) for the period from the beginning of 2013 until February 2014. Global risk perception was measured using the VIX, while asset prices included stock yields, bond yields and exchange rates. Three salient findings emerged from the study. First, PI flows had a significant impact on asset prices in the domestic financial markets, which increased during the GFC. The impact increased five- to tenfold during the period of global distress due to the simultaneous effect of a foreign capital reversal, tight market liquidity and herding behavior among investors. Second, the empirical results also demonstrated that global risk perception had a significant effect on all three components of asset prices in EMEs, with the magnitude depending on the specific characteristics of each country. The impact of VIX on stock price volatility was shown to increase as financial openness in the country increased, measured using total international financial obligations to GDP. As a country opened itself up to international capital flows, the impact of volatility from global risk spillovers to the domestic stock market also increased. Nevertheless, a similar pattern was not found for bond investments. The sensitivity of bond yield volatility to VIX was shown to correlate more closely with domestic economic fundamentals in terms of macroeconomic stability, particularly inflation and the current account, with a stronger influence as the duration of bond tenors increased.

402    Central Bank Policy In terms of exchange rate volatility, the impact of spillovers from perceived global risks depended on the exchange rate system applied, with the most significant impact affecting flexible exchange rate systems and a relatively more moderate effect in managed exchange rate systems. The empirical evidence showed the importance of exchange rate stabilization policy to mitigate to adverse impacts of global risk perception on exchange rate volatility. Intervention policy on the foreign exchange market could mitigate the impact of such global spillovers on exchange rate volatility, but must be supported by stabilization policy on the bond and stock markets. More importantly, solid macroeconomic fundamentals play a significant role in terms of strengthening resilience to international shocks. Specifically, maintaining low inflation and preventing a large current account deficit would significantly reduce domestic financial asset price sensitivity to global spillovers.

13.4.2. Monetary Policy Trilemma in an Open Economy The various studies mentioned above showed that foreign capital flow volatility created additional complexity for the central bank, known as the monetary policy trilemma or impossible trinity. Chapter 4 explained a theoretical review of the impossible trinity, using the Mundell-Flemming model, Donbusch’s overshooting model and the portfolio balance model. Basically, with perfect capital mobility, the exchange rate system will determine the ability of the central bank’s monetary policy to achieve the domestic goals. The central bank must consider its preferences in terms of three policy goals that are not aligned, namely monetary autonomy for the domestic goals of price stability and economic growth, exchange rate stability or foreign capital flow mobility (Fig. 13.3). The simple view of the impossible trinity shows that a country embracing a free foreign exchange system but with a fixed exchange rate regime will lose its monetary policy autonomy to achieve the domestic economic goals. A more moderate view considers the trilemma as a policy trade-off and, therefore, the central bank must optimize monetary policy to control inflation with policy to stabilize the exchange rate and apply foreign CFM. Obstfeld, Shambaugh, and Taylor (2005) studied the monetary policy trilemma in 103 countries covering a period of 130 years. Monetary policy autonomy was measured by comparing domestic interest rates with interest rates in other countries. Using a de facto and de jure approach as explained in Chapter 4, the exchange

Fig. 13.3:  Monetary Policy Trilemma.

Monetary Policy and Foreign Capital Flows    403 rate system was categorized as a pegged system or non-pegged system, while the foreign exchange system was divided into capital mobility and capital control. Furthermore, the study was divided into three periods, namely the Gold Standard (1870–1914), Bretton-Woods (1959–1970), and post-Bretton-Woods (1973–2000). Based on the international theory of uncovered interest rate parity (UIP), the policy trilemma was proven in the study using the following equation:

∆Rit = α + β∆Rit* + εit (7)

where R = domestic interest rate and R* = comparable interest rate. Peer countries using a pegged system included: UK for the Gold Standard period, US for the Bretton-Woods period, and various countries in the post-Bretton-Woods era. In general, the study confirmed that the policy trilemma was indeed valid as a monetary policy framework guideline in an open economy. First, the pegged exchange rate system was shown to create a closer link between the domestic interest rate and interest rate in a peer country rather than a non-pegged exchange rate regime. Interest rates in a pegged country tended to react more and have a stronger long-term correlation to the interest rate in the peer country. Second, by easing controls on foreign capital flows, pegged countries experienced a decline of monetary policy autonomy. Simultaneously, non-pegged countries enjoyed a greater degree of autonomy, particularly in the post-Bretton-Woods period, even under conditions of perfect capital mobility. Third, pegged countries, in reality, did not always actually peg their exchange rate due to devaluation or maintaining a range. On the other hand, non-pegged countries were also guilty of not always maintaining a floating exchange rate and often followed the peer country when setting the domestic interest rate. Several other studies have shown how the central bank overcomes the policy trilemma. Calvo and Reinhart (2002), for instance, revealed a phenomenon they called a fear of floating in the exchange rate policy pursued by various countries. As explained in Chapter 4, classifying the exchange rate systems adopted in various countries showed that more and more countries are embracing an intermediate system rather than a hard peg or purely floating. The fear of floating is a way to overcome the policy trilemma by stabilizing the exchange rate. In fact, in countries that implemented a flexible exchange rate system, the central bank also intervened on the foreign exchange market to stabilize the exchange rate. Several empirical studies have explored the extent to which monetary policy considered the exchange rate when determining the policy rate under the Inflation Targeting Framework (ITF). The methodology included the real exchange rate when determining the policy rate á la Taylor rule (Batini, Harrison, & Millard, 2003; Taylor, 2001) with the following model specification:

ii = it* + ρit−1 + (1− ρ )[λ1 ( πt − πt* ) + λ2 ( yt − yt* ) + λ3 ( qt − λ4 qt−1 )](8)

where qt = real exchange rate deviation in period t from the equilibrium. The specification enabled the researchers to examine the effect of volatility on determining the policy rate: (1) if λ4 = 0, the central bank reacted symmetrically to deviations in the real exchange rate from equilibrium; (2) if λ4 = 1, the central bank reacted

404    Central Bank Policy to fluctuations in the real exchange rate consistently by limiting exchange rate volatility; and (3) if 0 < λ4 < 1, the central bank gave a certain weight to rapid changes in the exchange rate and limited exchange rate misalignment. Aizenman, Hutchison, and Noy (2011), for instance, studied a modified Taylor rule in the determination of the monetary policy rate using panel data for 16 EMEs during the period from 1989:Q1 to 2006:Q4. The results showed that ITF EMEs applied a strategy where the central bank would respond to inflation and the real exchange rate when setting the policy rate. A study by Mohanty and Klau (2004) confirmed similar results using a modified Taylor rule for 13 EMEs, where the central bank reacted to actual inflation, the output gap and exchange rate fluctuations. In general, the results proved that central banks reacted to real exchange rate volatility. All countries, with Chile the only exception, were shown to lean against the wind by raising the interest rate in the event of exchange rate depreciation. Furthermore, the was a high degree of persistence in the monetary policy response to exchange rates in South Korea, India, México, Peru, Thailand, and South Africa. The monetary policy response to the exchange rate was, in fact, larger than that to inflation and the output gap, further proof of the fear of floating. Foreign exchange market intervention is a tool generally employed by the central bank for stabilization purposes to ensure exchange rate flexibility remains in line with economic fundamentals and supports the achievement of price stability. Theoretically, the effectiveness of foreign exchange intervention can be analyzed using the portfolio balance model, as discussed in Chapter 4. Most economists agree that sterilized intervention is effective in terms of influencing the exchange rate but the effectiveness of non-sterilized intervention is considered inconclusive from a theoretical and empirical standpoint. In this case, the effectiveness of sterilized intervention on the exchange rate works through the signal channel because intervention is seen by the markets as a change in the intention of the central bank’s monetary policy as well as the portfolio balance channel because intervention affects the asset composition of investors. The latest study by Blanchard, Adler, and Filho (2015) explored the effectiveness of foreign exchange intervention, particularly sterilized foreign exchange intervention, as an instrument to mitigate exchange rate pressures originating from foreign capital flows. Consistent with the portfolio balance channel, the empirical findings of the study showed that relatively large intervention created comparatively lower exchange rate appreciation in response to foreign capital inflows. The magnitude of effect proved that foreign exchange intervention was a valid instrument of macroeconomic management. The importance of exchange rate stabilization persuaded a number of economists, including Ostry, Ghosh, and Chamon (2012a), to opine that under certain conditions, dual targeting – namely exchange rate stabilization to support achievement of the inflation target – could strengthen monetary policy credibility at the central bank. The conditions are, based on ITF, that exchange rate targeting is used to mitigate the undesired impacts of capital flow shocks on inflation expectations, either directly through exchange rate pass-through or indirectly through domestic demand. With the effect of exchange rate fluctuations included in the inflation assessment and projections of the central bank, direct exchange rate targeting was more effective than an interest rate response to the impact of

Monetary Policy and Foreign Capital Flows    405 the exchange rate on inflation. Ostry, Ghosh, and Chamon (2012b) recommended that central banks in EMEs pursued dual targets. Moreover, central banks in EMEs have access to a second instrument, namely foreign exchange intervention. In other words, dual target monetary policy, targeting inflation, and exchange rate stabilization, using two instruments, namely interest rates and foreign exchange intervention, would increase, not decrease, central bank credibility. Several subsequent studies demonstrated the importance of foreign CFM to circumvent the policy trilemma. Klein and Shambaugh (2013) examined how far foreign CFM policy, such as directed and temporary capital controls or limited exchange rate flexibility, could realize full monetary policy autonomy. The results showed that foreign CFM under a flexible exchange rate regime could increase monetary policy autonomy. Nevertheless, foreign CFM under a fixed exchange rate system tended not to increase monetary policy autonomy compared to perfect capital mobility. As is widely practiced, extensive and long-term capital controls tend to break the relationship between domestic and international interest rates under a fixed exchange rate system. In addition, inflation is typically more controllable under a fixed exchange rate system and, therefore, the increase to monetary policy autonomy is not significant. Rey (2013) also extolled the importance of foreign CFM to overcome the monetary policy trilemma. This was based on the increasing comovement of capital flows, asset prices and economic growth across borders in the global financial cycle. Consequently, the central bank was recommended to convert the policy trilemma into a dilemma, namely to strengthen monetary policy autonomy and exchange rate stabilization through direct capital controls or macroprudential policy. The study revealed the dominant determinants of the global financial cycle, namely monetary policy in advanced countries, which influenced capital flow leverage, as well as credit growth in the international financial system. The impact was transmitted to monetary conditions in many countries, especially EMEs. In fact, interest rate policy and exchange rate flexibility were unable to protect a country’s economy from the global financial cycle. Obstfeld (2015) went further, stating that the impact of financial globalization has not only created a monetary policy trilemma but also a financial trilemma. The study evaluated the ability of monetary policy in EMEs to dampen the effect of global financial and monetary fluctuations. Consistent with the monetary policy trilemma, the study confirmed that EMEs with a flexible exchange rate were in a better position in terms of monetary policy autonomy. Nonetheless, changes in the exchange rate itself were unable to protect the domestic economy from global financial and monetary shocks. The important channels of international monetary and financial transmission operate through long-term interest rates and risk premiums, while financial globalization was shown not only to spur capital flows but also propagate the risks across jurisdictions. Although it would be possible to mitigate global spillovers through exchange rates, monetary policy cannot mitigate the exacerbating impact of capital flows and global financial risks on risks in the domestic financial system. The monetary policy trilemma has expanded into a financial trilemma, namely policy autonomy with global financial integration now a strong influence on maintaining domestic financial system stability.

406    Central Bank Policy 13.4.3. Foreign Capital Flows and Monetary Policy Dynamics in Indonesia In the case of Indonesia, as a small open economy, the deluge of foreign capital flows6 heightened the complexity of monetary control challenges. First, exchange rate fluctuations became more responsive to foreign capital flows rather than current account developments, which intensified domestic financial market volatility and, in turn, strengthened the transmission of further shocks (shock amplifier). Second, persistent foreign capital inflows reduced the potential effectiveness of monetary controls, considering the measures taken by Bank Indonesia to control liquidity in the economy were ultimately offset by the significant surge of foreign capital inflows. The two salient challenges outlined above had fundamental implications for ITF-based monetary policy implementation in the context of a small open economy, like Indonesia, considering, on one hand, the assumed role of the exchange rate as a shock absorber was reversed, while, on the other hand, there was a tendency to orient monetary policy, directly and indirectly, toward managing exchange rate fluctuations within a certain range, in line with economic fundamentals. With the influx of foreign capital, a policy orientation toward managing external balances could be counterproductive to policy to manage the internal balances. Therefore, it is important to monitor the degree of monetary policy autonomy enjoyed by Bank Indonesia in terms of managing rupiah stability against external influences. As is widely accepted, in line with changes in the global environment over the past few years due to behavioral changes in the financial system or the impact of financial crises in the last two decades, international monetary policy management in numerous EMEs has not been fully anchored to the impossible trinity, with the rationale starting to shift toward the need for a possible trinity.7 In Indonesia’s case, the policy trilemma index estimated by Aizenman, Chinn, and Ito (2008) showed that during the past two decades, against a backdrop of increasing financial market integration (foreign capital flow dynamics), although depreciating on the period prior to the Asian Financial Crisis of 1997/1998 and monetary policy autonomy, the rupiah exchange rate tended to be more stable (Juhro, 2015). Such developments revealed misalignment between the theoretical perspective and empirical facts in Indonesia (Table 13.3). Observations of foreign capital flow behavior (Net Foreign Assets – NFA), exchange rates and the interest rate differential also revealed the start of a shift in the monetary policy paradigm of Indonesia, moving away from the impossible 6

Over the past few years, particularly after the GFC, the economy of Indonesia, as with other EME, faced the problem of excessive foreign capital flows. The period of widespread quantitative easing policy to exit from the crisis in various advanced countries, including the US, EU, and Japan, triggered an influx of foreign capital to Indonesia’s economy. In contrast, however, subsequent indications of monetary policy normalization reduced foreign capital inflows and heightened the risk of a sudden capital reversal. 7 The possible trinity implies that sub-optimal monetary policy (not independent) could be formulated by restricting overly flexible exchange rate fluctuations, coupled with measures to restrict the mobility of foreign capital flows.

Monetary Policy and Foreign Capital Flows    407 Table 13.3:  Monetary Policy Trilemma Index in Indonesia. Policy Trilemma Index

1997–2000 2001–2005 2006–2008 2009–2013 Global Asian Crisis ITF Transition Pre-crisis ITF Financial Crisis

Exchange rate stability

0.11

0.27

0.25

0.28

Monetary policy autonomy

0.45

0.30

0.50

0.57

Financial market integration

0.74

0.69

0.69

0.71

Sources: Juhro (2015, spring); Juhro & Goeltom (2015, February)

Table 13.4:  A Decomposition of the Dynamic Correlation between Exchange Rates Versus Foreign Capital Inflows and Interest Rates. Decomposition of the Correlation with Exchange Rates

1997–2000 Asian Crisis

2001–2005 ITF Transition

2006–2008 2009–2013 Pre-crisis Global Financial ITF Crisis

Foreign capital inflows (NFA)

0.86

0.74

0.56

0.53

Interest differential

0.14

0.26

0.44

0.47

Source: Juhro & Goeltom (2015, February)

trinity. Rolling standard deviation data was used to show the trend of comovement between the three variables, which have evolved over time since the introduction of a floating exchange rate system in August 1997. This implies a decline in the degree of monetary policy autonomy in Indonesia due, among others, to monetary policy orientation, which not only strived to control inflation per se, but also to manage foreign capital flows and maintain rupiah exchange rate stability through active intervention. In other words, exchange rate dynamics were not completely removed or influenced by market forces, but strongly influenced by domestic monetary policy. Fundamentally, this phenomenon was also reported in numerous EMEs. Quantitatively, a decomposition of comovement between the dynamics of exchange rate fluctuations and foreign capital flows experienced a significant decline from 86% to 54%, while comovement between the dynamics of exchange rate fluctuations and interest rate differential experienced a significant increase from 14% to 46% (Table 13.4). The decline in the degree of monetary policy autonomy also represents another portrait of this unique phenomenon, where a surge of foreign capital flows could undermine the effectiveness of monetary controls. This could occur when the central bank intervenes with foreign capital inflows and subsequently increases liquidity in the domestic economy, which must be reabsorbed (sterilized) in order to mitigate the inflationary pressures from total money supply growth. Nonetheless, measures to control liquidity will cause the interest rate to rise

408    Central Bank Policy (and interest rate differential), thus drawing more foreign capital inflows to offset the decline of total domestic liquidity. For Indonesia’s case, Juhro (2010) examined the magnitude of offset and sterilization for the period from 2000 to 2010 through the reaction function of the central bank’s monetary policy, referring to Kouri-Porter (1974) and Cumby and Obstfeld (1982). An offset coefficient of 0.7 showed that the high degree of economic openness in Indonesia allowed external factors to subvert monetary policy implementation by Bank Indonesia because around 70% of the liquidity absorbed by Bank Indonesia (monetary controls) was offset by more foreign capital inflows. On the other hand, however, a comparatively low sterilization coefficient of 0.5 implied that around 50% of rupiah liquidity expansion, stemming from foreign capital flow intervention policy, was reabsorbed (neutralized) by Bank Indonesia. The relatively high offset coefficient compared to the sterilization coefficient was symptomatic of a low degree of monetary policy autonomy in Indonesia in terms of controlling foreign capital flow dynamics. There are at least three fundamental policy implications. The first relates to how to formulate an optimal possible trinity against a backdrop of ubiquitous uncertainty blighting the international environment. To that end, calculations on the use of monetary policy instruments, exchange rate developments and reserve asset management cannot merely be based on macroeconomic considerations but also integrated with the microstructure of the financial markets. The second relates to efforts to formulate an optimal strategy of monetary control instruments within the ITF despite the relatively limited role played by the interest rate. With the deluge of foreign capital inflows and excess liquidity on the financial markets, non-interest rate instruments (such as reserve requirements and other macroprudential instruments) may be used as the main or supporting instruments to control liquidity in the economy. The third relates to the need for coordinated liquidity control measures on the domestic financial markets by managing potential foreign capital inflows. Coordination is critical, not only in terms of efforts to unwind external and internal imbalances but also to manage the impact of monetary policy optimally, without exacerbating or negating conditions. The need for Bank Indonesia’s monetary policy response to symmetrically react to deviation in the real exchange rate from the equilibrium value was supported by the data. Juhro and Mochtar (2009) performed a simple analysis of Bank Indonesia’s monetary policy response for the period from 2000 to 2009, with and without including real exchange rate dynamics in the Taylor rule. The specification of the simple standard Taylor rule was expressed as follows:

it = δ + ρit−1 + (1− ρ )  γ π πt + γ y yt  (9)

where δ = (1− ρ )α Then, by including the ad hoc role of the exchange rate in the Taylor rule, the hypothetical reaction function of the enhanced Taylor rule, with the real exchange rate, was formulated as follows:

Monetary Policy and Foreign Capital Flows    409

k=K   it = δ + ρit−1 + (1− ρ )  γ π πt + γ y yt + ∑ γek et−k (10)   k=0

where et is the expression of exchange rate deviation from the equilibrium value and represented by the real exchange rate. If the monetary policy regime adopted is pure ITF, hypothetically γe 0 = ... = γeK = 0. The adjusted rule represents an interesting format because it takes into consideration the flexibility of the exchange rate’s role (enhanced rule). Despite some uncertainty concerning the application of that format, which is considered inconsistent with the basic rationale of ITF, empirical observations of economies with large exchange rate pass-through as well as relatively high and unstable inflation demonstrated the feasibility of the enhanced rule (Edwards, 2006). Congruently, Bask (2006) concluded that, technically, for small open economies, the addition of an exchange rate variable in the design of the Taylor rule could facilitate system stability, as long as the data used in the rule was contemporaneous. In the case of lagged data, such an addition would be unnecessary. Solving the two aforementioned equations produced three salient conclusions. First, despite different coefficients of determination (R-squared), namely 90% for the simple rule and 98% for the enhanced rule, it was generally concluded that the behavior of both rules was adequately explained by the data. The independent variables were in line with the theoretical predictions and statistically significant. All parameters in the two equations, namely inflation deviation, output gap, (effective) real exchange rate, and interest rate lag, were statistically significant at 5%. Second, inflation deviation was two to three times higher than output growth, therefore, ITF implementation in Indonesia was consistent with the fundamental rule of ITF, that controlling inflation was the overarching priority of monetary policy implementation. Furthermore, the large autoregressive interest rate parameter (0.7–0.8) indicated a conservative monetary policy response through interest rate smoothing (gradualism). Third, the short-term impact of real exchange rates was estimated from the design of the enhanced rule at 0.3 for the case of Indonesia, compared to zero in Chile but 0.8 in México. In general, the results provided sufficient justification to include the behavior of exchange rates in the implementation of monetary policy based on the ITF, which was supported by the data. In fact, during certain periods, the policy response in the design of the enhanced rule allowed a more optimal countercyclical policy response compared to the simple rule.

13.5. Foreign Capital Flow Management: Theories and Practices A theoretical and empirical review of the various countries mentioned in previous sections showed that foreign capital flows, in addition to stimulating economic growth, could also trigger macroeconomic and financial system stability risks. Consequently, a mix of macroeconomic policies (fiscal and monetary), financial system stability and structural reforms are critical at the national level. The combination of the policy mix will depend on local conditions in the country involved. Furthermore, structural reform policy is crucial to amplify the benefits

410    Central Bank Policy of foreign capital flows in terms of productivity and economic growth. Macroeconomic policy must also be adjusted, especially during periods of pressure on macroeconomic stability, such as high inflation or a wide current account deficit. Consequently, prudent and consistent monetary and fiscal policies play an important role as the first line of defense against foreign capital flows. Thereafter, adequate FX reserves, coupled with bilateral and multilateral swap arrangements, also play an important role as the second line of defense. In addition, strengthening and deepening the financial sector, as well as improving institutional capacity have been shown to improve a country’s ability to face foreign capital flows.

13.5.1. Principles, Targets, and Instruments In the context of the central bank, a monetary policy mix, exchange rate stabilization, and foreign CFM are required to achieve an optimal balance in terms of addressing the policy trilemma described above. Experience since the global financial crisis of 2008/2009 has shown that interest rates alone, as the main policy instrument, cannot respond to the volatility and magnitude of foreign capital flows to EMEs, even when complemented with foreign exchange intervention as a secondary instrument. In the face of an influx of capital flows, for instance, foreign exchange intervention would also increase the accumulation of reserve assets. Despite reinforcing the external resilience of EMEs, the cost of accumulating FX reserves is large because the central bank is, therefore, required to sterilize the resultant liquidity. Under such conditions, foreign CFM must be considered by the central bank. Foreign CFM is also required in the event of large foreign capital outflows because the interest rate response is not always effective, while foreign exchange intervention could also restrict the scope of the reserve assets owned by the central bank. 13.5.1.1. Formulation and Implementation Principles.  In general, the following conditions serve as a reference for when foreign CFM is applicable (IMF, 2012). First, foreign CFM is required when the space to make further macroeconomic policy adjustments becomes more limited, for instance after tightening monetary or fiscal policy, but foreign capital inflows remain beyond control. Such conditions are typically found during an episode of economic overheating or when signs appear of asset price bubbles, accumulating external debt, overvalued exchange rates as well as excessive and expensive reserve assets. Second, foreign CFM is required if there is a lag in the macroeconomic policy adjustment process or in terms of the economic impact. For example, a change in the budget as part of fiscal policy generally requires time for parliamentary approval. Monetary policy effectiveness could also be delayed by the transmission mechanism in terms of influencing inflation expectations and domestic demand. Third, foreign CFM is also required if the surge in foreign capital inflows amplifies the risk of financial system instability. Macroprudential policy would be more appropriate for overcoming systemic risk that does not originate from foreign capital flows. Nevertheless, an influx of foreign capital could also prompt excessive lending due to expansive liquidity conditions as well as increasing risktaking behavior. Under such conditions, restricting foreign capital flows through

Monetary Policy and Foreign Capital Flows    411 foreign CFM supports macroeconomic, monetary, and macroprudential policy effectiveness in terms of maintaining macroeconomic stability, monetary stability, and financial system stability. Foreign CFM implementation must take into consideration aspects of effectiveness and efficiency (IMF, 2013a). Likewise, the design and implementation of foreign CFM must be transparent, targeted, temporary and, wherever possible, must not discriminate between residents and non-residents. ⦁ Transparent and targeted: clear communication on the targets and instruments

of foreign CFM is critical to avoid market distortions and misinformed public expectations. Foreign CFM, targeting certain aspects that pose the most risks, such as PI flows as well as short-term external debt and speculation, is considered more effective. Striking a balance between the scope, effectiveness and sideeffects of foreign CFM requires an assessment specific to the country involved. Targeted foreign CFM, therefore, is easier to monitor and avoids triggering the undesired side-effects. ⦁ Temporary: As applied, foreign CFM could be tightened if foreign capital inflows increase and loosened as the flows ebb or even reverse. Foreign CFM can be maintained as required to support financial system stability, but not the BOP. ⦁ Non-discriminatory: In general, foreign CFM that does not discriminate between residents and non-residents is preferred. Nevertheless, if such efforts undermine the effectiveness of foreign CFM, instruments that discriminate against non-residents (otherwise known as capital controls) may be applied. The preference for non-discriminatory foreign CFM is based on the fair and impartial treatment expected of all IMF members. 13.5.1.2. Targets: Macroeconomic Stability and/or Financial System Stability.  Foreign CFM requires a clear target, whether it be macroeconomic stability, financial system stability or both. The answer will depend upon the types and volatility of the capital flows, combined with the economic fundamentals and financial sector dynamics in the respective country. As mentioned previously, volatile PI flows impact exchange rate, stock price, and bond yield volatility. If a country is not suffering from inflation and current account issues, sound monetary policy may be implemented through the interest rate or foreign exchange market intervention to mitigate asset price volatility. In this case, foreign CFM to achieve macroeconomic stability would support monetary policy, especially if inflation is relatively low and the position of FX reserves is inadequate to stabilize the exchange rates. Nonetheless, if inflation and the current account deficit are high, foreign CFM is required to complement the macroeconomic policies required, through monetary policy, fiscal policy or policies in the real sector. Under other circumstances, however, capital flow volatility may not only threaten macroeconomic stability but also spur pressures on financial system stability. Excessive asset price volatility would exacerbate market risk. Similarly, pressures on financial system stability may also appear from expansive liquidity and bank lending due to excessive foreign capital inflows. Furthermore, external debt in the banking industry would not only impact macroeconomic stability

412    Central Bank Policy but also financial system stability. The extent of the impact would depend upon financial sector resilience as well as the effectiveness of the macroprudential and microprudential policies instituted. In this regard, foreign CFM to control capital flows that targeted macroeconomic stability would support macroprudential policy effectiveness in terms of maintaining financial system stability. Occasionally, there is the perception that foreign CFM encompasses macroprudential policy, even though they both have different targets. Foreign CFM aims to restrict capital flows, while macroprudential policy aims to mitigate systemic risk and maintain financial system stability. Besides, foreign CFM is oriented toward controlling the total and composition of foreign capital flows, while macroprudential policy aims to mitigate any accumulation of systemic risk, irrespective of whether the risks originated externally or domestically. For example, taxes imposed on certain types of foreign capital flows constitute foreign CFM and only indirectly affect financial system stability. On the other hand, the additional capital surcharge and countercyclical buffer imposed on systemically important financial institutions are macroprudential policies and only indirectly affect capital flows. Nevertheless, there are several situations where foreign CFM and macroprudential policy become mutually complementary (IMF, 2015a). If the capital flows trigger systemic risk in the financial sector, both foreign CFM and macroprudential policy may be used. For example, capital flows to the banking sector in the form of external debt or PI could prompt a credit boom and domestic asset price bubbles. Regulations to limit external debt in the banking industry, through reserve requirements, debt-to-capital ratio, net open position (NOP) or even applying different risk weights to capital requirements, could restrict capital flows and dampen exchange rate volatility, which would simultaneously control excess liquidity in the banking industry as well as excess credit growth and help prevent domestic asset price bubbles from forming. Under such conditions, the regulations instituted to control capital flows and mitigate systemic risk in the financial sector may be considered as foreign CFM and macroprudential policy. Another important consideration is how and when to unwind or exit from foreign CFM and macroprudential policy. When the capital flows have ebbed and the associated volatility has subsided, foreign CFM could become an undesirable economic expense or even become ineffective and must, therefore, be removed. Nonetheless, certain macroprudential policies should be maintained to mitigate systemic risk in the financial sector that may persist after the influx of capital flows and period of high volatility. It is important to formulate macroprudential policy that targets financial system stability and not to control capital flows, a topic that will be discussed in Chapter 14. Additionally, in the case of persistent or permanent foreign capital flows, further reform policies would be required to deepen the financial sector, boost investment, and productivity in the real sector and strengthen the institutional aspects. 13.5.1.3. Instrument Selection: Prudential Regulations or Capital Controls.  Which instruments are available to mitigate the risks associated with foreign capital flows? In general, the policy options depend on the conditions specific to the country involved. Several instruments are available to control capital flows in the

Monetary Policy and Foreign Capital Flows    413 form of administrative instruments, including capital controls and prudential regulations. Administrative instruments directly target specific types of capital flows, while prudential regulations aim to improve the ability of the financial or corporate sector to mitigate the associated risks. The administrative instruments of foreign CFM are instituted through certain policies targeting the capital transactions of residents or non-residents in the form of specific taxes imposed on the capital flows of non-residents, additional reserve requirements without interest remuneration, special licensing policies, and even restrictions or outright bans. Such regulations could be applied to all capital flows, or specifically target flows based on type or period (debt, equity, FDI; short term vs long term). Furthermore, the regulations could target the economy as a whole or specific sectors (normally the financial sector) or specific industries (e.g., strategic industries). In general, the terminology of foreign CFM implies nondiscriminatory instruments, meaning the requirements should apply to residents and non-residents alike. Nevertheless, if there are differences between residents and non-residents, the term capital controls is used. For example, a higher reserve requirement applied to non-residents than residents would be a form of capital control but differentiating the reserve requirement based on currency, applicable to residents and non-residents alike, would constitute foreign CFM. In general, prudential foreign CFM regulations target foreign exchange transactions based on currency and not the residency of the parties engaged in transactions. The regulations are imposed on domestic financial institutions, specifically the banking industry. The instruments include restrictions on the NOP based on capital or restrictions on banking sector investment in foreign currency assets. The banks may also be limited in terms of disbursing foreign currency loans, particularly to customers that do not hedge, or face different reserve requirements on obligations in the domestic currency and a foreign currency. In addition to financial institutions, prudential foreign CFM regulations may also be applied to the corporate sector, primarily to mitigate the risks associated with external debt. The corporate sector may have to meet hedging regulations to mitigate currency risk and liquidity regulations to ensure repayment capacity as well as a maximum debt-to-equity ratio or minimum credit rating to mitigate corporate default risk. The various foreign CFM instruments are typically applied to control foreign capital inflows. There are, however, a few regulations that can be used to mitigate the adverse impact of foreign capital outflows on macroeconomic stability (IMF, 2015b). For example, taxes or reserve requirements may be loosened or removed altogether to overcome the risk of a capital reversal, depending on the intensity of the problems faced. Several other foreign CFM instruments, particularly the prudential regulations, such as the NOP and restrictions on foreign loans, may be maintained to strengthen risk management for capital flow volatility. In addition, other foreign CFM instruments are also available to mitigate foreign capital outflows, including: (1) restrictions on resident investments in offshore financial instruments (Iceland); (2) a minimum holding period for non-residents to sell their investments (Chile, 1990s); a minimum holding period before transferring the proceeds of an investment sale (Ukraine); a tax on the transfer of the proceeds of an investment sale (Malaysia); (3) restrictions on converting and

414    Central Bank Policy transferring assets in the domestic currency (Iceland) as well as restrictions on withdrawing term deposits (Argentina, Greece). Restricting non-resident access to the local currency would also make speculation more difficult. Similar to the case of foreign capital inflows, the effectiveness of foreign CFM instruments to control foreign capital outflows would be improved by integrating the measures into a comprehensive policy package, supported institutionally and by a healthy legal system. In other words, foreign capital outflow management is not a replacement for sound macroeconomic policy, particularly if the foreign capital outflows are due to domestic economic fragilities. Specifically, foreign capital outflow management would be more effective if formulated appropriately and implemented consistently. To be effective, comprehensive foreign CFM is necessary, tailored to the specific problems faced.

13.5.2. Foreign CFM Practices The previous section dealt with various foreign CFM instruments together with the macroprudential policies available to the central bank in order to mitigate the adverse impacts of capital flows on monetary and financial system stability. How could the various instruments be applied to EMEs and how effective would they be? The answer depends on the specific conditions of the country involved, particularly the complexity of the problems faced and the ability of the relevant authority to respond. In general, there are a number of key takeaways. First, foreign CFM instruments are typically not applied across the board but are targeted and adjusted over time (either loosened or tightened) depending on the issues faced. Second, instrument choice depends on the institutional arrangements and jurisdiction of the authorities in each respective country. In general, the instruments under the preserve of the central bank are simpler and faster to implement or adjust as required. Third, the effectiveness of foreign CFM instruments is influenced by the ability of the relevant authorities to detect regulatory loopholes exploited by the banking sector and market players early and, therefore, the authorities must be ready to refine the regulations as necessary. Fourth, foreign CFM instruments and macroprudential policy are complementary and, therefore, oftentimes implemented in tandem to enhance the effectiveness of controlling capital flows. And, most importantly, prudent macroprudential policy, on the monetary and fiscal sides, along with structural reforms to increase economic and financial sector capacity is the most optimal response in the face of foreign capital flows. 13.5.2.1. Foreign CFM Effectiveness: Brazil and Colombia.  Brazil and Colombia apply foreign CFM to mitigate the undesired impact on competitiveness and monetary policy autonomy (Ostry et al., 2011). In 2007, Colombia imposed unremunerated reserve requirements (URR) of 40% at six months on external borrowing as well as portfolio inflows, stocks and bonds. The URR was increased to 50% in May 2008 and the scope expanded to inflows of FDI to remain in the country for a minimum of two years in order to overcome strong exchange rate appreciation. Brazil, in March 2008, imposed a 1.5% tax on portfolio inflows to stocks and

Monetary Policy and Foreign Capital Flows    415 bonds, which was subsequently removed due the onset of the global financial crisis in 2008. In October 2009, however, the tax was reintroduced at 2% on portfolio inflows to stocks and bonds but the rate was lowered to 1.5% if the non-resident investors converted American Deposit Receipts (ADR) into domestic stocks. Although the instruments used were different, there were a number of similarities between foreign CFM implementation in Brazil and Colombia. Despite initially only targeting foreign loans using URR in Colombia, both countries subsequently expanded the scope to include portfolio flows to stocks and bonds in order to prevent non-resident investors from entering the stock market and then converting to bonds. From the beginning, Colombia exempted stock issuances through ADR, while Brazil initially imposed the same regulation but then changed to a tax on ADR issuances to close the regulatory loophole by converting ADR to local stocks. Onshore and offshore derivatives markets began to develop in both countries to circumvent the regulations. For example, because the URR and tax were imposed on a mandatory margin on a derivative transaction, foreign investors tended to seek profit by selling dollar forwards and speculating on exchange rate fluctuations (carry trade). Likewise, non-resident investors could cooperate with the domestic banks through branch offices operating internationally because such transactions were not subject to the URR or tax requirements. Foreign investor loans to international branches could subsequently be exchanged into the domestic currency and then used to repay the loan to the foreign investor. Colombia closed this loophole by restricting the banks’ position on derivative instruments, a good example of prudential regulations to strengthen the URR instrument. Meanwhile, the tax instrument used in Brazil was unable to close the loophole through such a derivative instrument. In other words, foreign CFM effectiveness was also influenced by the ability of the authorities to detect and close loopholes exploited by the markets through other foreign CFM instruments or with prudential regulations. 13.5.2.2. Macroprudential Regulations on Capital Flows: South Korea. In South Korea, exporters – particularly shipbuilding companies with scheduled dollar receipts – were required to hedge against future export proceeds. Such exporters wished to sell their dollars through forward contracts to receive Korean won to pay for their operating activities. When strong won appreciation was expected, however, exporters tended to over-hedge in order to profit from a level of appreciation that exceeded the interest rate differential. Domestic banks (typically foreign bank branches) as the counterparty, purchased dollar forwards and sold Korean won forwards, which were paid for by loans from banks abroad, including subsidiary banks, exchanged into Korean won and invested in domestic assets to fulfill the forward to the exporter. In this case, domestic banks were fully protected from foreign currency risk and sought profit from carry trade on exchange rate fluctuations. To mitigate the rapid increase of short-term foreign loans and its impact on currency risk and financial system stability, in June 2010, the Bank of Korea promulgated a comprehensive package of prudential regulations to prevent over-hedging and sudden stop risk from the short-term loans. The package strengthened regulations issued in November 2009, requiring banks

416    Central Bank Policy to increase their share of long-term foreign loans from 80% to 90% and then to 100%. In addition, the 2010 package also contained restrictions on the banks’ position of foreign exchange derivatives, tight restrictions on provisions for foreign currency loans and a more stringent liquidity ratio imposed on domestic banks. The various regulations outlined above successfully restricted foreign capital flows in the form of short-term foreign loans and bank exposure to foreign exchange. The regulations were imposed on domestic banks; therefore, the exporters were still able to meet their hedging requirements through offshore banks. To that end, the offshore banks would generally cover their short position on Korean won from non-deliverable forwards (NDF) through investments in government bond portfolios. In other words, although the regulations successfully limited foreign loans in the domestic banking sector, foreign capital inflows to South Korea were maintained in other forms, including government bonds. 13.5.2.3. Prudential Regulations on Foreign Loans: Croatia. Foreign banks have dominated the Croatian banking system since the year 2000 and have played an important role in terms of foreign capital flows to the country. A sure of capital flows prompted issues with macroeconomic and financial system stability, particularly the large share of household consumer loans. The problems became more complex because the central bank of Croatia (Croatian National Bank – CNB), as a member of the Eurosystem, had limited access to monetary instruments, while fiscal policy was constrained by structural budget problems. Consequently, the CNB had to rely on prudential regulations imposed on the financial sector, including foreign CFM instruments and capital controls. In 2003, the CNB imposed limits on credit growth disbursed by the banking industry and required a minimum level of retained earnings to be maintained if the limits were exceeded. Nonetheless, the banks were able to bypass the regulations by selling a portion of their credit portfolio to an affiliated leasing company, thereby transferring the risk to the balance sheet of the parent bank abroad. In July 2004, the CNB changed the regulations into capital controls on domestic banks by introducing URR in the form of a marginal reserve requirement (MRR) to be deposited at the CNB if the position of foreign currency liabilities exceeded the upper bound as of June 2004. In the subsequent period, the CNB tightened the URR regulation and expanded its scope to close the loophole. In addition, the capital controls were also tightened. Consequently, the banks were required to meet a minimum Capital Adequacy Ratio (CAR) of 12% if credit growth was maintained below 12%, with CAR progressively increasing if credit growth accelerated. The experiences of Croatia show how difficult it can be to control foreign capital flows if the central bank has limited space to maneuver as part of a monetary union and fiscal policy is also constrained. This implies that prudent macroeconomic policy, as the first line of defense, would be unavailable. Consequently, capital flows expose the domestic economic and financial cycles to the global financial cycle. In fact, economic advancement in Croatia created complex problems as foreign capital flows increased. As a member of the Eurosystem, Croatia was also not in a position to use exchange rate policy to control the foreign capital flows. Prudential regulations and capital controls successfully lowered credit growth in the banking industry and temporarily eased the deluge of capital flows.

Monetary Policy and Foreign Capital Flows    417 Nevertheless, credit growth remained high because domestic loans were replaced by direct foreign loans, thus changing the structure of foreign capital flows that went beyond the reach of prudential regulations and capital controls.

13.6. Concluding Remarks This chapter has discussed various aspects relating to the surge and dynamism of foreign capital flows in line with increasing global economic and financial integration. The benefits and risks of capital flows depended of the conditions specific to each respective country. Macroeconomic resilience, particularly low inflation, a sound current account, controlled fiscal deficit, and healthy financial sector, especially the banking sector, are critical. An appropriate macroeconomic policy response, from the monetary or fiscal side, is necessary to ensure that domestic economic fundamentals are maintained. Equally important is structural reform policy, encompassing the investment climate, increasing productivity, and competitiveness, including technology and human capital, as well as institutional strengthening of the government sector and corporate sector. Such measures would redirect the foreign capital flows to long-term and productive investments, hence supporting economic growth. Financial market deepening is also important to mobilize domestic savings deposits and stimulate more efficient intermediation of foreign capital flows to productive economic financing. From the central bank’s perspective, a mix of interest rate policy, exchange rate stabilization, and foreign CFM would produce the most optimal results in terms of monetary and financial system stability. Interest rate policy must still be oriented toward achieving price stability by paying due consideration to general economic and financial developments. Furthermore, exchange rate flexibility could help as a domestic economic shock absorber to external shocks. Notwithstanding, if exchange rate fluctuations and volatility deviate from their fundamentals, foreign exchange intervention would be necessary to achieve price stability and, therefore, dampen the impact on economic growth. Foreign CFM and macroprudential policy, individually or combined, could be implemented to mitigate the impact of capital flows on monetary and financial system stability. This chapter has also demonstrated the close relationship between monetary stability and financial system stability. Global financial integration has tightened the linkages between the domestic economic and financial cycles with the global economic and financial cycles. Foreign capital flows not only affect exchange rates and other financial asset prices, but also impact liquidity conditions and credit growth in the banking sector. Furthermore, global risks also influence capital flow volatility as well as the behavior of the banking industry and domestic financial markets. Monetary stability not only correlates closely with financial system stability, but both are heavily influenced by global financial and economic developments. Consequently, monetary and macroprudential policy integration has become more important in terms of strengthening monetary and financial system resilience in the face of greater global financial integration. Chapters 14 and 15 will provide a detailed and comprehensive discussion on financial system stability, the macroprudential policy response required as well as integration with monetary policy in a central bank policy mix.

418    Central Bank Policy

Box 13.1:  The Impact of Global Spillovers from US Monetary Policy The global financial crisis of 2008/2009 precipitated extraordinary foreign capital flows to emerging market economies in terms of their scale and volatility. Several studies have examined the impact of global spillovers from the monetary policy of the US Federal Reserve. Two periods, in particular, have received special attention, namely the period of unconventional monetary policy (UMP) after the global financial crisis to stimulate the US recovery, and the period of monetary policy normalisation from May 2013 until the Fed’s proposal to hike the Federal Funds Rate (FFR). In general, the impact of global spillovers from the Fed’s UMP on foreign capital flows, asset prices and economic activity in emerging market economies was transmitted through various channels (Lavigne, Sarker, & Vasishtha, 2014). First, the portfolio balance channel, namely the purchase of long-term assets, such as government bonds and mortgage-backed securities (MBS) by the Fed, which reduced supply and the risk premium of such assets on the market. Second, the signaling channel, namely Fed commitment to loosen monetary policy after the global financial crisis, which lowered long-term bond yields. A combination of those two factors pushed investors to adjust their portfolio composition through the purchase of assets in emerging market economies with a higher risk-adjusted return, otherwise known as rebalancing versus return chasing. Third, the exchange rate channel, namely expansionary monetary policy by the Fed, which led to USD depreciation. In addition to the deluge of foreign capital flows to emerging market economies, such developments decreased US imports and EME exports to the United States through the trade channel. Several studies have examined the impact of the Fed’s expansionary monetary policy on foreign capital flows to emerging market economies. Ahmed and Zlate (2014), for instance, studied the determinants of foreign capital flows to EMEs by including the effect of the Fed’s UMP, using net portfolio investment flow data to 12 emerging market economies from Asia (India, Indonesia, South Korea, Malaysia, the Philippines, Taiwan, and Thailand) and Latin America (Argentina, Brazil, Chile, Colombia, and México). The sample period, from 2002:Q1 to 2013:Q2, covered the Fed’s UMP through to the Taper Tantrum in the middle of 2013. The specification of the empirical model using panel data is as follows: ( F / Y )i ,t = β0 + β1 ( gi ,t − gtAE ) + β2 ( Ri ,t − RtUS ) + β3 LSAPt + β4 RA t + εt (11) where F/Y = net portfolio investment to GDP ratio, ( gi ,t − gtAE ) = economic growth differential with advanced countries, ( Ri ,t − RtUS ) = interest rate differential with US, LSAPt = the Fed’s UMP through large-scale asset purchases (LSAP), and RAt = global risk perception. The research produced several interesting findings. First, as predicted, the Fed’s expansionary

Monetary Policy and Foreign Capital Flows    419

monetary policy through LSAP increased portfolio investment to emerging market economies. Second, the economic growth differential and interest rate differential as well as global risk perception had a significant impact of foreign capital flows to EMEs. Third, there was a significant change in the dynamics of portfolio investment before and after the global financial crisis of 2008/2009, particularly in terms of becoming more sensitive to the interest rate differential. Fourth, the capital controls introduced by a number of emerging market economies since the global financial crisis reduced portfolio investment flows to EMEs. Anaya, Hachula, and Offermanns (2015) examined how global spillovers from the Fed’s UMP affected macroeconomic conditions and the policy response in EMEs. A Global VAR (GVAR) empirical model was used, thus facilitating interactions between push factors from the US, including the Fed’s UMP, and pull factors in emerging market economies. Such interaction would be indicative of a contagion effect from the global spillovers. Monthly data was used from 2008 to 2014 for 20 EMEs: Argentina, Brazil, Chile, Colombia, Hungary, India, Indonesia, Israel, Korea, Mexico, Malaysia, Peru, the Philippines, Poland, Russia, Singapore, South Africa, Thailand, and Turkey. The empirical findings confirmed that the Fed’s UMP significantly increased portfolio flows to emerging market economies. The surge of portfolio flows elevated real economic growth, stock returns, and exchange rate appreciation, while lowering lending rates in emerging market economies. Furthermore, EMEs reacted by easing monetary policy in response to US expansionary monetary policy in order to address the monetary policy trilemma. Meanwhile, several studies have examined the impact of the Fed’s plan to normalize monetary policy since May 2013 on foreign capital flows to emerging market economies. Gauvin, McLoughlin, and Reinhardt (2014), for instance, investigated the extent to which macroeconomic policy uncertainty in advanced countries triggered global spillovers to EMEs through portfolio flows to stocks and bonds. Monthly portfolio data was used covering 20 EMEs for the period from 2004 to 2011. The model was expressed as follows:

Fi ,t = ∑ αn Fi ,t−n + β0 + β1∆PUt + β2 X i ,t + δi + εi ,t (12)

where Fi,t = portfolio flows to bonds and stocks in EMEs, ΔPU = policy uncertainty vector in the US and EU, Xi,t = vector of the control variable, and δi = country-specific fixed effects. Policy uncertainty was measured using the difference between economists’ predictions and a Big-Data search of newspaper articles. The control variable Xi,t, covered the push factors, including risk perception (VIX), stock returns, global liquidity (M2 growth in the US, EU, and Japan), interest rates and oil prices, as well as the pull factors, including CDS spreads, stock returns, and domestic interest rates.

420    Central Bank Policy

The empirical findings showed that the effect of policy uncertainty on portfolio flows to EMEs was very different depending on the origin, namely from the US or EU. Policy uncertainty in the US reduced portfolio flows to stocks and bonds in EMEs. Conversely, policy uncertainty in the EU reduced portfolio flows to bonds but increased portfolio flows to stocks. The level and direction of the global spillovers was not linear, namely that the impact would increase during periods of high global risk. Regarding portfolio flows to stocks, the level of risk in each respective country was also significant, namely that EMEs with a low level of risk would experience an increase of such flows despite high global risks. Koepke (2014) looked at the effect on portfolio investment flows to EMEs of market expectations concerning FFR hikes by emphasising the element of surprise in US monetary policy. The empirical model used in this study included two additional dummy variables, namely D1 = 1 for months when the FFR was expected to rise and D2 = 1 for months when the FFR was expected to fall. Market expectations were measured using FFR futures contracts. Ft = α0 + α1 Ft−1 + β1D1MPt + β2 D2 MPt + ∑ λi PULLi ,t + ∑ δi PUSHi ,t + εt (13)

The results confirmed that changes in market expectations for US monetary policy were a significant determinant of portfolio flows to emerging market economies, particularly bonds. Furthermore, the effect was asymmetric, with expectations of monetary policy tightening having a larger adverse impact on portfolio flows to bonds (and relatively lower on stocks) compared to the positive impact of expectations of monetary policy easing. In other words, the emphasis on the surprise view in this study showed that changes in market expectations had a greater impact on foreign portfolio flows to EMEs than the actual impact of the Federal Reserve hiking the Federal Funds Rate (FFR). Rai and Suchanek (2014) examined the impact of four Federal Reserve tapering announcements in 2013 on foreign portfolio flows and the financial markets in 19 emerging market economies. Using an event-study approach, the empirical model was estimated as follows:

Rit = β1Zit + β2 Zit Sit + β3Sit + β4 X t + εit (14)

where R = vector of exchange rate fluctuations, stock prices, bond yields, and net portfolio flows to EMEs; Z = vector of the dummy variable for each announcement date (May 22, 2013, June 19, 2013, September 18, 2013, and December 18, 2013); S = vector of specific EME variables, such as GDP growth, inflation and the current account balance; and X = vector of global variables, namely global stock and bond yields. The empirical results showed that the Fed’s announcements in May and June 2013 had the adverse effect of precipitating foreign capital outflows, exchange rate

Monetary Policy and Foreign Capital Flows    421

depreciation, higher bond yields, and lower stock prices in numerous emerging market economies. In contrast, the very same indicators reversed after the Fed announced its plans to delay tapering in September 2013. Reaction to the tapering announcement in December 2013 was more muted, however, because the markets had already anticipated the move and the FOMC’s forward guidance had improved. In addition, more severe impacts were observed in more vulnerable EMEs, namely those with lower economic growth, a large current account deficit, high external debt and a low position of reserve assets. The impact manifested in larger capital outflows (capital flight), exchange rate depreciation and elevated bond yields. This research confirmed that solid economic fundamentals and an appropriate macroeconomic policy response could help to mitigate the global spillovers stemming from the Fed’s plan to normalize its monetary policy stance.

This page intentionally left blank

Chapter 14

Macroprudential Policy and Financial System Stability 14.1. Introduction In addition to the influence it exerted on foreign capital flow volatility as elucidated in Chapter 13, the global financial crisis (GFC) of 2008/2009 also raised awareness concerning the importance of maintaining financial system stability (FSS). A crisis can occur at anytime, anywhere, in advanced and developing countries. Differing from previous crisis episodes, however, the GFC that originated in the US stemmed from an accumulation of leverage in the housing sector, facilitated by securitized bank loans traded extensively in the financial sector. Subprime mortgage securities, coupled with structured products as well as excessive risk-taking behavior, created a build-up of leverage and mortgage delinquencies soared, bringing down the entire financial system. The leveraging process was also facilitated by financial institutions in Europe that exploited low interest rates stateside. Consequently, the crisis that began in the US quickly spread into a crisis in Europe and the GFC of 2008/2009. The adverse effect of the GFC was extraordinary, with an economic and financial depression felt around the world in advanced and developing countries, which took a protracted period to resolve. The crisis exposed the harmful side of the capitalist economy, which had spread worldwide through financial liberalization and globalization. Maintaining FSS incorporates wide-ranging dimensions in the national interest and, therefore, cannot be the preserve of just one policy or institution. The sound of financial institutions is indeed important, both banks and nonbanks alike. The Lehman Brothers’ crisis in the US showed that a crisis can occur due to fragilities at nonbank financial institutions. Nonetheless, sound financial institutions are not enough. A financial crisis can occur due to four salient factors: (1) financial sector fragilities; (2) imprudent macroeconomic policy; (3) deteriorating governance in the corporate sector and government; and (4) volatile capital flows (Eichengreen, 2004). Excessively expansive fiscal and monetary policy overheats the economy, prompts inflation to soar, and leads to runaway fiscal and current account deficits, which ultimately trigger a crisis. Similarly, foreign capital

Central Bank Policy: Theory and Practice, 423–459 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191020

424    Central Bank Policy flow volatility creates pressures on the balance of payments (BOP) and exchange rates, thus exacerbating the crisis. FSS must be able to anticipate systemic risk because of financial system and macroeconomic (macrofinancial) linkages, while facilitating prevention measures to avoid a crisis. Therefore, the early detection of FSS fragilities to the aforementioned factors is critical. Close policy coordination between the government, central bank, supervisory authority and deposit insurance corporation is, therefore, vital. This chapter discusses various aspects of macroprudential policy in terms of supporting FSS. Following the introduction, the systematic discussion is divided into four sections. First, an explanation of FSS policy and the concept, including the central bank’s role in supporting FSS with an emphasis in macrofinancial linkages. Second, a discussion on the theory and empirical evidence for financial procyclicality, which is considered an important trigger of financial crises, emphasizing the threat from asset price bubbles and excessive credit growth (credit boom). Third, an explanation of the theory and empirical evidence for interconnectedness and financial networks that oftentimes accelerate crisis propagation, emphasizing the interbank money market and systemic banking crises. Fourth, a summary of the macroprudential policy theory and practices implemented at various central banks, including the targets, approaches, and instruments used as well as the importance of coordination with the microprudential supervisory authority. The discussions in this chapter and in Chapter 13 form an integral part of the central bank’s policy mix as discussed in Chapter 15.

14.2. Conceptual Dimension of FSS The GFC of 2008/2009 again exposed the fragilities of the capitalist economy based on the trade of currency and capital through the financial system in order to finance investment activities. We witnessed the crisis in Latin America during the 1980s caused by excessive fiscal expansion and an accumulation of government debt. We felt the Asian Financial Crisis of 1997/1998, driven by an economic boom in the wake of financial liberalization during the 1980s, which fed through to excessive credit growth, property bubbles and a build-up of private external debt. And we were impacted by the GFC of 2008/2009 that originated in the US due to massive default on mortgage-backed securities (MBS) that imploded the financial system. The GFC of 2008/2009 and the previous crisis episodes have had severe consequences on the financial system, economy, fiscal burden and public prosperity. That the roots of crises stem from an excessive build-up of debt, public and private, has been proven throughout the history of human civilization (Kindleberger, 1978; Reinhart & Rogoff, 2009). Likewise, that procyclical property bubbles and credit booms precede and trigger crises in various countries is also not a new idea (Claessens & Kose, 2013). And that crises are becoming more frequent and multidimensional, comprising of currency crises, debt crises and financial system crises, has also been proven in various jurisdictions (Bordo, Eichengreen, Klingebiel, & Peria, 2001). The problem, therefore, is why we have not learned the various causes of crisis in order to anticipate or prevent a crisis? And why have

Macroprudential Policy and Financial System Stability    425 we not focused on efforts to maintain FSS, protecting against macroeconomic instability, financial sector fragilities, weak corporate and government governance as well as risks from capital flow volatility in order to prevent the reoccurrence of crisis episodes?

14.2.1. The GFC of 2008 / 2009 and FSS What is FSS and what is the central bank’s role in terms of supporting FSS? Although academic references to financial instability are available from the thinking and ideas of Minsky (1982), the GFC of 2008/2009 brought such thinking to the fore, becoming the focus of serious attention by policymakers around the world. According to Minsky (1982), financial instability is a consequence of the capitalist economy where rising asset prices and the build-up of debt are potentially unchecked. During an expansionary economic period, supported by price (inflation) stability and low interest rates, such as during the two decades of Great Moderation in the US prior to the outbreak of the GFC in 2008/2009, the returns on investment and rising asset prices (financial and physical) outstripped lending rates or the cost of capital. The misplaced perception that returns would remain high during the economic boom period led to a proliferation of trade in debt and capital to finance the investment. The accumulation of debt, from bank loans and foreign capital, became increasingly uncontrollable. Speculative investors and Ponzi schemes began to dominate and outnumber prudent investors engaged in hedging activities. Ultimately, speculation and the Ponzi schemes incurred huge losses, triggered a credit crunch and prompted financial instability. Stability is destabilizing due to the boom-bust financial cycle, which eventually pulls the economy into a crisis. Definitions of FSS are not always identical between policymakers and academics. Fundamentally, however, FSS refers to conditions where the financial system functions soundly in an economy and is resilient to the various shocks that could emerge (Allen & Wood, 2006; Mishkin, 1999b ). Several academic definitions contrast FSS with conditions that could trigger a crisis. Borio and Drehmann (2009), for example, departed from the description of financial instability when defining FSS as follows: financial instability is a set of conditions that is sufficient to result in the emergence of financial distress/crises in response to normal-sized shocks. These shocks could originate either in the real economy or the financial system itself. Financial stability is then defined as the converse of financial instability. Meanwhile, Schinasi (2004) defined FSS in broader terms: a financial system is in a range of stability whenever it is capable of facilitating (rather than impeding) the performance of an economy, and of dissipating financial imbalances that arise endogenously or as a result of significant and unanticipated events.

426    Central Bank Policy Central banks apply more practical definitions. For example, the European Central Bank (ECB) defines financial stability as a condition in which the financial system – intermediaries, markets and market infrastructures – can withstand shocks without major disruptions in financial intermediation and in the general supply of financial services. In the same vein, Act No 9 of 2016 concerning Financial System Crisis Prevention and Resolution (UU-PPKSK), defined “financial stability as a condition in which the financial system functions effectively and efficiently and is resilient to domestic and global shocks,” with reference to the financial system as a system consisting of financial services institutions, financial markets and financial infrastructures, including the payment system, which interact to facilitate the accumulation of funds from the public and allocation to support national economic activities. From the aforementioned definitions, at least five important aspects must be stressed in terms of maintaining FSS and preventing crises as follows. First, the soundness of individual financial institutions is necessary but not sufficient. FSS is linked to how the financial system functions to resist and survive shocks in the economy. The key, therefore, lies in the macrofinancial linkages of the financial system with the economic activities rather than the soundness of individual financial institutions. Second, history has shown four types of macrofinancial linkages that often occasion a crisis as follows: (financial and property) asset bubbles; credit booms; excessive leverage; and sudden-stop capital reversals (Claessens & Kose, 2013; Reinhart & Rogoff, 2009). Instability that emerges from those four macrofinancial linkages often leads to financial procyclicality, which accelerates economic growth during a boom period and, conversely, undermines economic growth and even exacerbates the economic cycle during periods of recession. The phenomenon of procyclicality, if left unchecked, will amplify the boom-bust cycle that can ultimately result in a crisis. Furthermore, financial procyclicality often foregoes financial crisis episodes (Claessens, Kose, & Terrones, 2011; Jorda, Schularick, & Taylor, 2011) and, therefore, must be anticipated and prevented early. We witnessed this phenomenon during the crises in Latin America, Asia, US, and Europe as well as the GFC of 2008/2009. Third, policy to dampen shocks in the domestic economy and the ability to overcome external shocks are pivotal to support FSS. When domestic demand outstrips economic capacity, economic overheating is oftentimes the result, marked by high inflation and/or a current account deficit and/or a fiscal deficit. Such macroeconomic imbalances can either trigger or accompany macrofinancial imbalances as discussed previously. Furthermore, the risk of default, in terms of FSS, can originate from the spillover of global and/or regional economic shocks. The trigger may be in the form of a sudden crisis or change to policy in another

Macroprudential Policy and Financial System Stability    427 jurisdiction. Such an external shock could precipitate a sudden-stop capital reversal, thus prompting distress and a currency crisis, payment crisis, external debt crisis as well as financial and economic crises. Consequently, maintaining FSS requires macroeconomic policy (fiscal and monetary), structural reform in the real sector and foreign capital flow management. Fourth, while a crisis can be triggered by default at one financial institution, the bursting of a bubble due to procyclical macrofinancial imbalances, or international or domestic economic shocks, the contagion process leading to a systemic crisis event throughout the entire financial system can progress with rapidity due to interconnectedness and financial networks in the markets and infrastructures, including the payment system (Allen, Babus, & Carletti, 2010; Acemoglu, Ozdaglar, & Tahbaz-Salehi, 2015; De Bandt & Hartmann, 2000). A currency crisis, for example, could be triggered by a foreign capital reversal or global or regional spillovers that propagate through contagion due to interconnectedness and collapse of the foreign exchange market (Calvo & Reinhart, 2000). In this case, investors are looking to buy but the supply of foreign exchange has evaporated, except from the central bank. Likewise, the default of an individual bank could prompt bank runs and spread to the banking system due to interconnectedness and insufficient liquidity on the interbank money market (Freixas, Parigi, & Rochet, 2000; Morris & Shin, 2004). Globalization has also exacerbated contagion worldwide due to interconnectedness and networks between international financial institutions, markets and infrastructures. Fifth, a crisis typically peaks when contagion through interconnectedness and financial networks are accompanied by herding behavior and information contagion (Acharya & Yorulmazer, 2003; Bikhchandani & Sharma, 2001). Panic on the markets and among depositors, or even financial institutions, due to herding behavior and information contagion would further compound a financial crisis, resulting in fire sales, inadequate liquidity, and financial market collapse. The US subprime mortgage debacle demonstrated how contagion can propagate rapidly through fire sales on financial markets and tight credit in the banking system (Diamond & Rajan, 2010). Herding behavior and information contagion quickly turned the subprime mortgage fiasco into a full-blown and complex financial crisis, not only in the US but also in Europe and around the world. The widespread and devastating impact of the crisis reinforced the need for joint measures and accountability. No one institution can, by itself, be accountable for the entire mandate of maintaining FSS. A mechanism is therefore required, or institutional policy coordination, to maintain FSS nationally, which has become common practice in many jurisdictions. The supervisory authority for banks and financial services is responsible for overseeing the soundness of individual financial institutions in order to protect the depositors, investors and consumers through microprudential regulation and supervision. The central bank, therefore, is responsible for macroprudential regulation and supervision to mitigate the risk of macrofinancial imbalances and systemic risk in the financial system, as well as lender of last resort (LOLR). A deposit insurance corporation also plays an important role in terms of mitigating information contagion from bank runs to protect depositors, as well as active involvement in the early resolution actions to

428    Central Bank Policy

Fig. 14.1:  Linkages between Macro–Micro Policy and FSS.

resolve a problem bank. Furthermore, the government, through the Ministry of Finance, needs to lead national FSS coordination efforts to prevent the onset of an economic crisis and the exorbitant fiscal and public costs incurred. Fig. 14.1 illustrates how the financial system looks from four dimensions and the required policy responses, namely: microfinancial sector, macronational, macrointernational and macrofinancial linkages. The first three dimensions were already the subject of serious attention prior to the GFC in 2008/2009, while the fourth dimension has emerged in the wake of the catastrophic GFC. Before the GFC of 2008/2009, as presented in the first three parts of the schematic, attention in terms of economic management and crisis prevention were primarily focused on: (1) microprudential regulation and supervision to maintain the soundness of individual financial institutions; (2) macroeconomic (fiscal and monetary) policy for macroeconomic stability (inflation, exchange rates, fiscal deficit and current account deficit); and (3) foreign capital flow management to ensure resilience against contagion and international shocks. The new dimension, which has developed since the GFC of 2008/2009, as presented in the lower part of the schematic, is the need to include the financial system in the function of the economy, which could trigger macrofinancial instability. This must be overcome through macroprudential policy, namely the regulation and supervision of financial institutions from a macrofinancial and systemic risk perspective. The linkages between the various policies are required to support sustainable economic growth, while maintaining macroeconomic and FSS.

Macroprudential Policy and Financial System Stability    429 A coordinated crisis prevention and management protocol between institutions is, therefore, crucial to maintain FSS. Placing FSS in the context of national macro–micro policy interconnectedness was seen as a correction to the reformulation of policy and institutional reforms of previous crises. After the Asian financial crisis of 1997/1998, for instance, the prevailing focus of many authorities was considered appropriate for that time. The Ministry of Finance was responsible for fiscal policy, while the central bank focused on its mandate of price (and exchange rate) stability through monetary policy. The regulation and supervision function of financial institutions has now been transferred away from the central bank, however, to a newly formed authority, such as what happened in South Korea and Indonesia. In fact, the independence of each respective authority was strengthened to ensure more effective policies at both. On one hand, this approach was motivated by the democratic movement that has increasingly become accepted as the global model. On the other hand, the change was driven by the academic community recommending the need to focus on one instrument for one goal, congruent with the Tinbergen rule. Nevertheless, the GFC exposed this approach to be inappropriate. The rule failed to acknowledge that various polices are mutually interconnected and, therefore, require a coordination mechanism to achieve robust economic growth and maintain stability. Furthermore, the GFC showed how important it is to consider the financial system from a macrofinancial perspective and that systemic risk must be mitigated through macroprudential policy. The central bank is the most suitable institution to issue macroprudential policy, on top of monetary and payment system policy as well as LOLR.

14.2.2. Central Bank’s Role in FSS The previous discussion showed that the central bank requires a mandate to maintain FSS, in addition to the mandate of price and exchange rate stability (Haldane, 2004; Padoa-Schioppa, 2003). The mandate of price (and exchange rate) stability through monetary policy (including foreign capital flow management) is already well known and accepted. The central bank’s policy framework is already well established, supported by a credible Inflation Targeting Framework (ITF) as discussed in Chapters 8 and 9. That policy framework has already lowered long-term inflation as well as supported robust economic growth and lower interest rates in many jurisdictions (Berg, Hallsten, Queijo, HeideKen, & Söderström, 2013). The salient features of the policy framework that supports monetary policy credibility include: a clear inflation target to be achieved, solid analytical models and macroeconomic projections, interest rate policy consistency to attain the inflation target, a formal and regular policy formulation process, as well as published inflation projections and other communications to anchor inflation expectations. The central bank also fundamentally applies the same policy framework to foreign capital flow management, as referenced in Chapter 13. Such a policy framework, applying macroeconomic analysis and projections, can easily be expanded to include macrofinancial linkages, particularly through the financial system, foreign capital flows and macroprudential policy by the central bank.

430    Central Bank Policy The central bank’s mandate to support FSS is primarily achieved through macroprudential regulation and supervision of financial institutions from a macrofinancial perspective and a focus on systemic risk. As mentioned previously, macrofinancial linkages between financial institutions and the economy often create financial procyclicality, thus accelerating economic growth during an expansionary period but also exacerbating the downturn during a recession. Procyclicality, if not well managed, accelerates the boom-bust of the financial cycle, which could ultimately trigger a crisis. Macroprudential policy is, therefore, directed toward managing financial procyclicality, especially credit booms and asset bubbles, to avoid macrofinancial imbalances that often lead to financial crises. In addition, macroprudential policy is also directed toward managing interconnectedness and financial networks, in particular through the interbank money market and financial infrastructures, to mitigate systemic risk. While default at one financial institution could trigger a crisis, along with macrofinancial procyclicality, or international or domestic economic shocks, a systemic crisis event throughout the financial system could propagate with rapidity due to interconnectedness and financial networks. In addition to macroprudential policy, the central bank also supports FSS through monetary and payment system policy as well as foreign capital flow management. Monetary policy supports FSS through avoidance of economic shocks as part of the third aspect of FSS mentioned previously, in addition to the overarching goal of price (and exchange rate) stability. This includes the regulation and supervision of money markets and the foreign exchange market as an integral part of monetary policy, which also supports FSS by mitigating systemic risk that could surface due to interconnectedness and financial networks as the fourth aspect of FSS. Similar considerations underlie central bank authority in terms of the payment system, in addition to the core purpose of maintaining a secure, efficient, and reliable payment system, which is inextricably linked to effective monetary policy. Meanwhile, the authority of the central bank to manage foreign capital flows supports FSS by avoiding economic shocks per the third aspect, on top of the main target to support exchange rate stability, while preventing sudden-stop and BOP crises. The central bank is the most appropriate institution to implement macroprudential regulation and supervision. The central bank has the capacity for macroeconomic and macrofinancial surveillance, coupled with the required macroprudential policy instruments (Kawai & Morgan, 2012). Moreover, a Bank for International Settlement (BIS) (2011) study of 13 advanced and emerging market countries concluded that central banks must have direct involvement in FSS policy-making and implementation if the policy is to be effective based on three explanations. First, the monetary policy framework and task implementation focuses the central bank on macroeconomic analysis and projections, while complementing understanding of the financial markets, institutions, and infrastructure, all matters of great import in terms of macroprudential policy implementation. Second, financial system instability can be caused by and have a significant impact on economic performance, with fundamental consequences on

Macroprudential Policy and Financial System Stability    431 economic activities, price, and exchange rate stability, as well as monetary policy transmission effectiveness. Central banks are already familiar with evaluating financial system and economic interconnectedness, the macrofinancial aspects as the focus of macroprudential policy. Third, the central bank is a source of liquidity in the financial system and economy through monetary policy and the LOLR function, and the availability of liquidity is crucial when maintaining financial stability.

14.2.3. Central Bank Macroprudential Policy Central bank urgency to strengthen the FSS framework requires solid financial infrastructures, including an adequate inspection and supervision function to support domestic financial market integration into an increasingly complex financial system. In this context, Borio (2003) stressed the need to strengthen the regulatory framework or macroprudential policy to limit risks if pressures or distress emerge on the financial markets for a protracted period, which could precipitate a real output slump in the economy. Conceptually, macroprudential policy contains prudential regulatory instruments that aim to maintain FSS as a whole, not the soundness of individual financial institutions. Analogously, microprudential policy consists of prudential regulatory instruments that aim to maintain the soundness of individual financial institutions. Macroprudential policy seeks to develop, oversee and deliver appropriate policy responses to the financial system as a whole. It aims to enhance financial system resilience and dampen systemic risks that spread through the financial system. (The G-30) Therefore, macroprudential policy is used to prevent a boom-bust cycle in the supply of credit and liquidity that could trigger economic instability. Through its role of maintaining a stable supply of financial intermediation, macroprudential policy also supports the goals of monetary policy in terms of maintaining price stability and output. There are two salient dimensions of macroprudential policy. First is the timeseries dimension, namely that macroprudential policy is directed toward dampening the risk of excessive procyclicality in the financial system. To that end, macroprudential policy must be designed in such a way as to avoid or mitigate procyclicality. The principal is how to encourage financial institutions to prepare an adequate buffer during an expansionary economic phase, when imbalances tend to appear, and then how to utilize that buffer when the economic slumps. The second is the cross-section dimension that shifts the focus from prudential regulation of individual financial institutions to the regulation of the financial system as a whole. History has shown that most financial crises around the world occur not because of problems at individual banks that subsequently propagate to the financial system. In contrast, large crises occur because of simultaneous

432    Central Bank Policy exposure to macrofinancial imbalances by most of the financial system. Therefore, a more holistic view of the financial system and linkages to the macroeconomy from various perspectives is desperately required. In line with changes to the landscape of the financial sector, especially since the GFC of 2008/2009, many central banks have begun to apply broader macroprudential policy instruments. In this regard, several instruments previously thought of as monetary instruments (such as reserve requirements) have been used to prevent systemic risk and maintain FSS in the economic cycle, particularly for targeted economic sectors. The instruments are not focused, however, on efforts to mitigate risk at individual banks. Therefore, the policy instruments can be categorized from a broader macroprudential perspective. Examples of common macroprudential policy instruments are presented in Table 14.1.1 In addition to the policy goal (target) set and policy instruments available to achieve that target, another important aspect to be formulated in the macroprudential policy framework is the policy response. When formulating the macroprudential policy response, rule versus discretion is a prominent consideration. Like monetary policy, there is always a trade-off between using rules or discretion. Rules provide assurance to the markets and credibility to the central bank. Nonetheless, rules are too rigid, thus removing the flexibility to respond to structural changes or uncertainty that often befall financial markets. Considering the advantages and disadvantages or rules and discretion, decision-making models that combine the two are often preferred.2 Table 14.1:  Examples of Macroprudential Policy Instruments. Problem Leverage (potential procyclicality)

Instrument Adjust risk weighting in capital requirements Apply capital ratio to risk-weighted assets

Credit (debtor linkages Apply countercyclical provisioning (on certain loans) and characteristics, Restrict the loan-to-value ratio for certain sectors pressure on macrostability (with potential bubbles) – interconnectedness) Limit lending to certain sectors (for instance property, credit cards) Adjust the reserve requirement, either across the board or targeted Liquidity (potential risk on certain aspects)

Apply a buffer to reduce dependence on risky sources of funds Apply the loan-to-deposit ratio

Sources: Borio and Shim (2007), Hannoun (2010), and G-30 (2010). 1

A more detailed discussion on the various types and utilization of macroprudential policy instruments in various jurisdictions is presented in Section 14.5 of this chapter. 2 More detailed discussion on this issue is presented in Box 14.1.

Macroprudential Policy and Financial System Stability    433

Box 14.1:  Macroprudential Policy Response: Rule Versus Discretion When seeking the most optimal macroprudential policy response, it is imperative to have comprehensive understanding of business cycles and financial cycles. The main problem faced by central banks in the financial sector is identifying when the economic cycle enters an expansionary or contractionary phase and, thus, the policy response required. Departing from the existing monetary policy framework that has a firm foundation in academic thinking, theory and established practices, the macroprudential policy framework remains in the early stages of maturation, especially since the global financial crisis of 2008/2009. Consequently, prevailing thinking on designing the appropriate macroprudential policy response is an extremely interesting topic to discuss. In terms of policymaking, this helps to organize the various perspectives on certain macroprudential policies, while increasing policy transparency and reinforcing policy communication. Referring to existing concepts, the macroprudential policy response can be categorized into two polar opposites: discretion and rule. As explained in Chapter 7, understanding discretion refers to a policy response based on certain judgments in accordance with current (past) developments, while rule refers to a policy response based on existing rules that have previously been implemented. In a simple case, namely the application of countercyclical macroprudential policy (for instance a countercyclical capital buffer), the ultimate goal of policy is to prevent excessive credit imbalances or to smooth the financial cycles. In this case, the macroprudential policy response can be applied through a rule or based on discretion. Based on discretion, the central bank may require the banks to build capital buffers above the normal rate during an economic boom period. Based on pre-determined variables, the central bank will announce that the economic cycle has entered a boom period and decide if the boom has triggered excessive and dangerous risktaking behavior, thus requiring the banks to provision additional capital as a buffer should the economy moderate in future periods. Conversely, when the economy contracts, based on central bank discretion, the banking industry shall be permitted to release or relax its capital buffers. On one hand, countercyclical macroprudential policy eases procyclicality risk but, on the other hand, could increase the risk of default at individual banks. In fact, capital buffers are considered more appropriate for macroprudential purposes, namely by targeting risks and the wellbeing of the overall financial system, which is independent of the solvency situation of individual banks as the preserve of microprudential policy. In a broader context, discretion also suffers weaknesses in the form of adverse effects from the warnings and recommendations put forward by the central bank,

434    Central Bank Policy

which could trigger self-fulfilling prophesies and, ultimately, significant political pressures on the central bank. On the other hand, a rule-based response seeks to establish macroprudential policy per automatic mechanisms and triggers, thus avoiding time inconsistencies, even more so when the policy authority is forced to identify (subjectively) an optimal balance between the micro and macroobjectives in an integrated manner. Unfortunately, rules can be overly complex and difficult to apply, particularly for new policies and those not considered international best practices. As a pragmatic measure, the use of simple (feedback) rules has been explored as stylized designs à la Taylor rule in monetary policy. Table 14.4.1:  Monetary and Macroprudential Policy Rules. Risks Inflation

Rule Design Monetary Policy Rule: Policy Rate = f(inflation gap; output gap)

Cyclical risks Balance sheet composition

Macroprudential Policy Rule: Instruments representing the structure of the balance sheet = f(target cyclical riska) For instance: countercyclical capital buffer = f(credit to GDP ratio)

Terms and conditions of financial transactions

Instruments representing the terms and conditions = f(target cyclical riskb) For instance: loan to income (LTI) ratio = f(property or house prices)

Cross-sectional Macroprudential Policy Rule: risks Instruments representing market structure = (target Market structure cross-sectional riskc) Source: Adapted from Fic, Tatiana (2012). a

Instruments representing the structure of the balance sheet: countercyclical capital buffer, sectoral capital requirements, leverage ratio, liquidity buffer and so on. Targets: credit to GDP ratio, credit growth, credit gap, and so on. b

Instruments representing the terms and conditions: LTV, LTI. Targets: property or house price growth, house price deviation from equilibrium price, and so on. c

Instruments representing market structure: wholesale funding taxation. Targets: composite index of interconnectedness, information from surveys, and so on.

Referring to the various rule designs, future macroprudential research and policymaking will certainly need to be directed towards setting targets and instruments in line with the specific risk (instability) category, namely balance sheets, terms and conditions of financial conditions, and

Macroprudential Policy and Financial System Stability    435

structural risks. Several macroprudential policy instruments introduced by the Bank of England, such as the countercyclical capital buffer (CCyB), sectoral capital requirements, and leverage ratios, fundamentally only overcome (directly) the risks associated with balance sheet composition. If a quantitative policy target has been set, however, empirical verification is still required to show which instruments are the most effective in meeting the policy target. Further identification is required of the macroprudential policy transmission channels, which is not a simple task. With the distinct advantages and disadvantages of each macroprudential policy response based on rule or discretion, prevailing thought suggests a combination of the two, as is the case with monetary policy. The Bank of England (2009), for instance, proposes a constrained discretion macroprudential regime, namely to apply a response that is largely discretional to allow policymakers to adapt as they learn, but remains systematic, transparent and accountable, in line with the pre-determined policy goals and constraints. Such a regime requires the support of clear policymaking targets and frameworks as well as measures of accountability that are acceptable to the stakeholders. A combined response could also place more weight on using rules rather than discretion. Such a policy response would use a simple rule as the main reference, as is the case with the countercyclical capital buffer. For example, several key financial sector variables, which might fall under the auspices of different financial sector policy authorities, must be announced clearly and transparently (ex ante). The relevant policy authorities have the option to deviate from the rule, allowing the banks to adjust (raise/lower) the buffer in line with their information pertaining to economic conditions. This, however, must be the exception, rarely used and clearly explained to the markets. Therefore, the general policy response will adhere to the rule in daily practice, while discretion shall be available when absolutely necessary (Quagliariello & Libertucci, 2010).

14.3. Theoretical Model and Empirical Evidence of Financial Procyclicality Per the Financial Stability Board (FSB, 2009), the term “procyclicality” relates to: the dynamic interactions (positive feedback mechanisms) between the financial and the real sectors of the economy. These mutually reinforcing interactions tend to amplify business cycle fluctuations and cause or exacerbate financial instability. Therefore, procyclicality is a phenomenon where the financial cycle accelerates the economic cycle, as illustrated in Fig. 14.2, which shows a procyclical financial

436    Central Bank Policy

Fig. 14.2:  Financial Cycle and Economic Cycle Procyclicality. cycle accelerating the booms and busts of the economic cycle. During a period of economic expansion (boom), the financial cycle tends to outpace the economic cycle. Bank credit expands rapidly, while financial and property asset price bubbles appear, excessive debt accumulates and an influx of foreign capital floods domestic financial markets, all of which exceed what is actually required per economic fundamentals and, therefore, risks build and fragilities appear in the financial system. Oppositely, during a period of economic contraction (bust), the financial cycle exacerbates the downswing in the economic cycle. Consequently, a domestic or international shock, against a backdrop of financial system fragilities due to a build-up of risk, often leads to financial and economic crises. Procyclicality is a consequence of dynamic interactions between the financial and the real sectors of the economy and macrofinancial linkages, caused by factors in the microfinancial sector and macroeconomy. Concerning the microfinancial sector, several factors lead to procyclical lending and other capital financing in the economy. First, asymmetric information between the lender and borrower causes credit rationing or the external finance premium. To overcome information asymmetry, the lender requests collateral on the loan or investment capital extended to the borrower, in the form of a business project or fixed asset, such as land of property. As the value of business projects and fixed assets fluctuate with the economic cycle, the loan or investment capital will also fluctuate. During an expansionary phase of the economic cycle, business prospects improve and corporate wealth increases, thus reducing credit rationing and lowering the risk premium. As investment also increases and accelerates economic growth, the acceleration becomes more pronounced due to (collateral) asset price bubbles during the boom period. In contrast, during an economic downswing, credit rationing, and the risk premium increase. Investment declines, along with economic growth, thus exacerbating the downturn. In the literature, this phenomenon is known as the financial accelerator. Second, regulation and accounting methods can also compound procyclicality. In the banking industry, for instance, capital requirements are based on risk

Macroprudential Policy and Financial System Stability    437 contained on the bank balance sheet. Difficulties when evaluating the fair value of assets tend to underestimate risk during an expansionary phase and overestimate risk as the economy declines. Consequently, capital requirements based on risk assessments can aggravate procyclicality if the under/overestimates of risk are not adjusted. Fair value accounting can also create problems if the market value of the asset deviates far from the fundamental value. Such conditions can occur if asset price bubbles appear during an economic upswing. Price corrections and weak markets, as the economy moderates, heighten systemic risk that are often late to overcome. Third, changing perceptions among economic players can also exacerbate asset price volatility in the financial and the economic cycles. During an economic boom, market confidence is high and asset prices are expected to keep rising. This occurs in the property sector and for financial assets, including stocks and bonds. As the markets increasingly tend to believe favorable economic news, interest grows in purchasing assets, the prices of which increase, leading to herding behavior. Consequently, asset price bubbles appear and continue to grow as economic optimism increases. Then, as the economy begins to moderate, a reversal of perception encourages investors to release assets in order to prevent further price declines. Once again, herding behavior, this time to release assets, exacerbates the downward price trend. Furthermore, investors may begin to panic, thus prompting fire sales to save themselves. Consequently, large price corrections occur and trigger systemic risk. In terms of the macroeconomy, there are several determinants of the intensity of procyclicality in the financial system. In general, fiscal expansion accelerates during an economic upswing due to larger tax revenues, easier access to government debt and other factors. The impact of monetary policy on financial system procyclicality occurs through changes to interest rates, exchange rates, credit or risk-taking behavior in the financial sector. Stringent and lenient regulation and supervision by the authorities also affects financial sector behavior and activities in response to corporate and household demand for financing, as well as other financial services. Such conditions are typically linked to financial sector liberalization in a given jurisdiction. Likewise, foreign capital flows in various forms, including external debt, are normally relatively large during an economic upswing but vulnerable to a sudden reversal when the economy moderates. Macroeconomic developments, due to the various aforementioned factors, interact dynamically with the microfinancial sector factors and mutually reinforce financial system procyclicality in the economy. Therefore, procyclicality is not merely the result of dynamic interactions between the business cycle and financial cycle, but also stems from the risktaking cycle, namely behavior marked by excessive optimism during an economic upswing and excessive pessimism during a downswing (Nijathaworn, 2009). The business cycle is said to expand when the economy enters a growth phase and contract when the economy moderates. The financial cycle is characterized by more expansive banking behavior in the form of increased leverage in line with the expansionary phase of the business cycle. Conversely, the banks become more conservative through deleveraging as the business cycle

438    Central Bank Policy contracts. This interaction between the business cycle and financial cycle is determined by the behavior of economic agents to risk, which is also influenced by economic expectations, risk perception, regulations, and incentives. These changes to risk explain why investor behavior can change so drastically from extreme optimism when risk is low to extreme pessimism in order to avoid risk. Such changes to risk-taking behavior are based on sudden changes in financial system and economic activities. Interaction between the three cycles, moving in the same direction and mutually reinforcing, create procyclicality in the financial sector (Table 14.2). The interactions are typically illustrated in the context of the boom-bust cycle. Initially, when the economy expands, characterized by macroeconomic stability and stronger economic growth, investor confidence boosts optimism in the economy, which encourages risk-taking behavior and triggers demand for credit and higher asset prices. During a period of optimism, risk in the financial sector dissipates, interest rate spread narrows and loan loss provisions are reduced because banks are inclined to look at shorter horizons rather than the longer term. Higher asset prices raise the value of collateral, which boosts lending and, therefore, elevates market confidence and prompts risk-taking behavior, as reflected by increased leverage. As banks are more inclined to lend, the corporate sector undertakes more investment activities and households consume, which stimulates economic Table 14.2:  Interaction between the Business Cycle, Risk-taking Behavior, and the Financial Cycle.

Expansionary Phase

Business Cycle

Risk-taking Behavior

Financial Cycle

• Macroeconomic stability • Stronger economic growth

• Greater confidence and optimism • More risk-taking behavior • Stronger demand for credit

• Lower risk assessment, narrower interest rate spread • Rising asset prices raise value of collateral • Leverage increases • Influx of foreign capital • More lending

Contractionary • Heightened Phase macrovolatility • Less economic activity

Source: Modified from Nijathaworn (2010).

• Lower market • Bank deleveraging confidence •L  arger loan loss • Risk averse provisions • Less demand for • Broader interest rate credit spread • Less lending • Decline of foreign capital inflow

Macroprudential Policy and Financial System Stability    439 growth. In contrast, when confidence in the economy slumps, the behavior changes and investors become risk averse, thereby precipitating lower asset prices and the value of collateral declines. In response, the banks deleverage, shifting their portfolios away from risky loans to low-risk assets, such as Bank Indonesia Certificates (SBI) and government debt securities (SUN), in order to maintain adequate capital. The banks also reinforce loan loss provisions in anticipation of deteriorating credit quality. Such developments reduce lending and, in turn, exacerbate the economic downswing.

14.3.1. Financial Accelerator Theory The literature shows that imperfections and asymmetric information in the financial sector accelerate the business cycle and economic cycle, a phenomenon known as the financial accelerator. The mechanism operates through capital availability or changes to the risk premium. In the banking sector, for example, the acclaimed work of Stiglitz and Weiss (1983) showed that credit rationing is a normal phenomenon. Information asymmetry between banks and borrowers concerning actual business conditions leads to adverse selection. Only borrowers with a viable business outlook can obtain loans from a bank and, therefore, imbalances appear on the credit market with the credit volume and lending rates below demand for loans. Meanwhile, Williamson (1987) suggested that credit rationing occurs at different volumes and interest rates in accordance with an assessment of the borrower’s business prospects. On the other hand, there is moral hazard among the borrowers which causes the banks to demand additional collateral beyond project feasibility (Bester & Hellwig, 1987). Consequently, the phenomenon of credit rationing in the banking sector could be caused due to limited supply from the banks, per Stiglitz and Weiss, or in line with project feasibility per Williamson, or even because the additional collateral is in the form of land, buildings or other fixed assets. Whatever the reason, the financial accelerator affects bank loans because the project value and additional capital fluctuate with the economic cycle due to fiscal and monetary policy as well as other macrofactors. Greenwald and Stiglitz (1993, 2003) analyzed the dynamic interactions of credit rationing with the macroeconomy. Departing from the aforementioned factors of capital availability through credit rationing, Bernanke, Gertler, and Gilchrist (1999) analyzed the financial accelerator due to risk premium factors in terms of investment financing. Asymmetric information causes firms seeking external capital, such as bank loans, to incur the external risk premium. Therefore, the net worth of the project, based on the net-present value of profits obtained now and at a future date, is larger than the value of the loan received after calculating the interest. During an economic upswing, corporate net worth is generally higher due to higher financial expectations and asset market prices, which increase compared to the level at equilibrium. Such conditions create procyclicality. The procyclical net worth of borrowers, due to profit and asset price procyclicality, also leads to procyclical bank lending and lending rates.

440    Central Bank Policy The model developed by Kiyotaki and Moore (1997) explained the influence of fluctuating asset prices on project value and collateral in terms of financial system procyclicality. Lenders, including banks, can only enforce loan agreements through foreclosure on corporate net worth, including additional collateral in the form of land, buildings and other fixed assets. Therefore, the maximum loan value is determined by corporate net worth as collateral. Simultaneously, the value of collateral is influenced by total credit as an aggregate. In other words, there is a mutually reinforcing relationship between total credit, collateral value and economic growth. The financial accelerator and procyclicality occur. Fluctuations in the value of collateral affect credit volume, corporate performance and economic growth, asset prices, and so on, which exacerbates the impact of the financial cycle on the economy, thus triggering systemic risk. Several important conclusions can be drawn from the financial accelerator phenomenon that creates procyclicality in the financial sector and economy. Early detection is crucial to measure how much procyclicality is excessive, among others, by comparing differences between credit and financing growth in the financial sector and the level considered normal in line with fundamentals. Likewise, rising financial asset and property prices demand vigilance to detect the presence of bubbles. Macroprudential policy can combat excessive procyclicality by, for instance, using the loan-to-value ratio (LTV) or more stringent capital requirements. Microprudential regulation and supervision are also vital for risk management and sound credit procedures. Furthermore, the availability of transparent information on corporate conditions must be improved, among others through a credit information system. Such measures are required to dampen the impact of procyclicality on systemic risk.

14.3.2. Empirical Evidence for Financial Procyclicality Experience has shown that crises commonly occur with procyclicality in three key sectors: (1) financial asset prices, including property and securities; (2) bank credit and innovative structured products; and (3) foreign capital flows, particularly foreign debt and portfolio investment. The following studies provide empirical evidence for procyclicality in those three sectors. In essence, procyclical asset prices, lending and foreign capital flows cause an excessive accumulation of debt as the main root of financial crises (Reinhart & Rogoff, 2009). 14.3.2.1. Asset Price Bubbles.  Hunter, Kaufman, and Pomerleano (2003) and Evanoff, Kaufman, and Malliaris (2012) conducted literature reviews covering various aspects of asset price bubbles and the implications on monetary policy, regulation, and international policy. Congruent with economic theory, asset price bubbles were said to occur if: “the market price of an asset exceeds its price determined by fundamental factors by a significant amount for a prolonged period.” In general, market price (Pt) shows the net-present value of return (Rt) on assets received now and at a future date and can be expressed as follows:

Pt = ∑ i = 0 Et Rt +i + f ( x )(1) ∞

Macroprudential Policy and Financial System Stability    441 where P* = ΣEt Rt +i is the fundamental price of the asset and f(x) is market price deviation from the fundamental value. The return on the asset is in the form of capital gain and the proceeds from investment in the asset. If the investment is in a financial asset, such as stocks, the return originates from higher share prices and dividends. If the investment is in property, the return is the increase in property price and any rental fees collected. Asset price bubbles can be identified by studying market price deviation from the fundamental value, namely a function of f(x) above. On an efficient market, prices generally fluctuate around the fundamental value. The function f(x) is random with a normal distribution. Such conditions indicate an absence of asset price bubbles. Nonetheless, if market price deviation from the fundamental value increases, an asset price bubble is said to have occurred. In the economic literature, this phenomenon is known as growing bubbles. The spread of price deviation can occur at a constant speed or exponentially. History has shown that price bubbles are frequently observed prior to, and are therefore an indication of, a financial crisis. Studies by Claessens et al. (2011) and Drehmann et al. (2010) showed how property price bubbles are an accurate indication of an impending crisis after the financial cycle has peaked. Similarly, from a study of crises in five advanced countries and 15 emerging market economies (EME), Kaminsky and Reinhart (1999) found that a crisis is preceded by excessive credit growth, accompanied by around a 40% jump in stock prices each year above normal levels. Property and other asset prices also bubble prior to a crisis. At some point, however, the bubbles burst, hence share and property prices collapse. Such developments are typically followed by bad credit (loss loans), default, and tight liquidity on the interbank money market. Furthermore, systemic risk in the banking industry, at financial institutions and on the financial markets also intensifies. Crises triggered by asset price bubbles, specifically property price bubbles, and excessive credit growth, were also found in many countries, including Japan, Latin America, and Europe (Hunter et al., 2003). Such observations were clearly visible when the GFC of 2008/2009 began in the US, which was foreshadowed by a housing price bubble, accompanied by a price bubble of structured products at nonbank financial institutions, which subsequently triggered systemic risk and a financial crisis. Driven by strong demand, rising house prices were followed by a construction boom (Scherbina, 2013). Many new houses were built to supply the market at exorbitant prices, particularly in areas beset by rapidly rising prices. From 2004 to 2006, the Federal Reserve raised its policy rate 17 times from 1% to 5.25% to control inflation. Default on mortgage-backed loans became common. Then, when demand and prices began to decline, the inventory of unsold new homes prompted foreclosures by the banks, thus depressing prices further. What began as a housing crisis, credit crisis and structured products crisis, subsequently evolved into a financial crisis and economic crisis in the US (Brunnermeier, 2009). Experience from the Asian financial crisis in 1997/1998 also revealed property price bubbles and a credit boom prior to the crisis (Collyns & Senhadji, 2003). Property prices skyrocketed and the banking industry had the capacity to lend to the sector using rising property prices as collateral thanks to financial

442    Central Bank Policy liberalization in the 1980s. Nevertheless, this study showed that the impact of property price bubbles on the crisis event was also determined by the resilience of the banking industry in each respective jurisdiction. Weak banking sectors in Thailand, South Korea, and Indonesia were vulnerable to speculative attacks on the foreign exchange market that triggered a property crisis, external debt crisis, banking crisis, and economic crisis. Conversely, Hong Kong, Singapore, and Malaysia survived the speculative attacks due to stronger banking systems and sounder regulatory and supervisory systems. Consequently, when the property price bubble burst, those three countries avoided a deep economic crisis. Such experiences from the Asian Crisis demonstrated that sound banking regulation and supervision is a strong barrier to a crisis event when an asset price bubble bursts. 14.3.2.2. Credit Booms and Busts. The importance of credit as an indicator to understand the financial cycle and its impact on a crisis was also emphasized by Schularick and Taylor (2012). Using data covering 140 years from 14 advanced countries for the period of 1870–2008, this study showed that a shift in the role of money supply and credit has occurred since World War II. Financial liberalization has increased the importance of credit, while monetary policy using interest rates has proven incapable of controlling the credit cycle in the economy. During an expansionary economic period, credit growth was shown to boom, exceeding 7%, but then bust at −2% during an economic downswing in 14 advanced countries. Furthermore, the boom and bust periods of the credit cycle were shown to be inextricably linked to financial crisis episodes. The study recommended that central banks place more emphasis on assessing bank credit performance, in terms of pace of growth, application of lending standards, and risk accumulation. Consequently, an interest rate policy response would be insufficient, instead requiring control over bank credit, including macroprudential regulation and supervision. Mendoza and Terrones (2008) conducted a study of 27 credit booms in advanced countries and 22 EMEs for the period of 1960–2006 using microcorporate and macroeconomic data. The results showed that credit booms correlate closely with economic expansion, elevated property and stock prices, real exchange rate appreciation and a broader current account deficit, followed by the reverse conditions when the economy slumps. The same dynamics were observed in corporate indicators of leverage, worth, and dependence on external financing, as well as banking indicators of asset quality, profitability, and credit activity. Moreover, credit booms have tended to synchronize internationally, peaking in the debt crisis in the 1980s, the ERM crisis in 1992 as well as external crises in EMEs. In addition to the various consistencies, several differences in the characteristics of credit booms between advanced countries and EME were also revealed as follows: (1) macroeconomic and microcorporate indicator fluctuations tended to be more pronounced, more persistent, and asymmetric in EMEs, with a strong correlation to the non-tradeable sector; (2) not all credit booms culminate in crisis, but many crises in EMEs were preceded by credit booms; and (3) credit booms in EMEs were often preceded by a deluge of foreign capital inflow rather than domestic financial reform or increased productivity, while credit booms in

Macroprudential Policy and Financial System Stability    443 advanced countries were generally prefaced by a bump in productivity or financial reform. In the case of Asia, Craig, Davis, and Pascual (2006) conducted a study concerning the influence of financial system characteristics on accelerating procyclicality in five advanced countries (Australia, Hong Kong, Japan, New Zealand, and Singapore) and six EMEs (China, Indonesia, South Korea, Malaysia, the Philippines, and Thailand). The study evidenced strong but asymmetric procyclicality between property prices, credit growth, and GDP, with a stronger correlation during a cyclical downswing. Analysis also showed that asset prices and credit growth have deviated significantly from long-term trends, peaking during several episodes and, thus, triggering financial instability. Panel estimations using macro and individual data from nearly 300 banks found that the structural characteristics of financial systems in Asia were an important source of procyclicality, namely: excessive dependence upon credit guarantees, delayed recognition of loss loans (bad credit) and insufficient loan loss provisions, loose credit risk assessments that fed through to excessively small lending margins to cover the risk, lower quality credit assessments during a cyclical upswing, loan concentration at state-owned enterprises and credit groups, bank dependence upon foreign sources of funds, and financial liberalization. 14.3.2.3. Procyclical Capital Flows. Numerous studies have demonstrated procyclical capital flows. Kaminsky, Reinhart, and Vegh (2004), for instance, studied capital flow procyclicality and the interaction with fiscal and monetary policy. Using data from 104 countries for the period of 1960–2003, the study revealed several salient aspects of procyclical capital flows. First, capital flows were proven to be procyclical, namely that foreign loans would increase during an economic upswing and decrease during a downswing in most Organization for Economic Co-operation and Developmen (OECD) and developing countries. Second, fiscal policy is also procyclical, namely that government spending increases during an economic upturn and decreases during a downturn in the majority of developing countries. Third, in EMEs, monetary policy tends to be procyclical, such that interest rates are lowered as the economy improves and raised when the economy moderates. Fourth, in developing countries and, specifically, in EMEs, periods of foreign capital inflow correlate to expansive macroeconomic policy, while foreign capital outflow is associated with contractive macroeconomic policy. In other words, in EMEs and developing countries, when it rains it pours. Meanwhile, Calderon and Kubota (2012) found that private capital flows could spur a credit boom. Based on quarterly data concerning foreign capital inflows, credit to the private sector and other macrofinancial indicators for 71 countries for the period of 1975–2010, the study revealed several interesting findings: (1) a surge of private capital flows is a relevant predictor of a credit boom; (2) the probability of a credit boom is more likely if the foreign capital flows are driven by other private investments and, with a smaller coefficient, by portfolio investment; (3) a deluge of inflow is also a relevant predictor of a bad credit boom, namely one that culminates in a financial crisis; (4) a bad credit boom is more commonly linked to capital inflows as other investments, but

444    Central Bank Policy mitigation is possible through an increase of foreign direct investment (FDI); (5) the predictive ability of other investment flows is primarily determined by external debt; and (6) an accumulation of leverage and overvaluation of real exchange rates strengthen the predictive capacity of credit booms that culminate in a systemic crisis. Empirical evidence that foreign capital inflows prompt a credit boom and might culminate in a financial crisis was also put forward by Sa (2006) and Amri et al. (2013). Arena, Bouza, Dabla-Norris, Gerling, and Njie (2015) conducted a further study into foreign capital flows, credit booms, and macroeconomic dynamics in low and middle-income countries. Using data from 135 developing countries for the period of 1960–2011, the study provided several invaluable lessons. Foreign capital flows were shown to trigger a credit boom, with a similar frequency and level between the two groups of developing countries. Nevertheless, the impact on macroeconomic dynamics was more pronounced in middle-income countries, with larger fluctuations noted in terms of output, investment, private and public consumption, real exchange rates, money supply, and the current account balance. Furthermore, countries with a more rigid, or less flexible, exchange rate regime, experienced faster credit growth during a period of foreign capital inflow and, therefore, were more vulnerable to capital reversals. Experience has also shown that financial liberalization could prompt a credit boom if not accompanied by prudential banking regulations. The study reminded developing countries to carefully consider the implications of financial liberalization on the potential influx of foreign capital, which have been empirical shown to trigger a credit boom and create fragilities in terms of FSS.

14.4. Theoretical Model and Empirical Evidence for Interconnectedness and Financial Networks A crisis can originate from bank default that propagates into bank runs. Furthermore, a crisis can also stem from a property bubble bursting, credit crunch, default on external debt payments, or sudden capital reversal. Nonetheless, the rapidity and propagation of bank default or macrofinancial imbalances into a systemic crisis is determined by interconnectedness and financial networks in financial markets and infrastructures, in addition to the spread of information. Experience showed the speed at which the speculative attack on the foreign exchange market in Thailand spread to South Korea and Indonesia, which quickly grew into the Asian Financial Crisis of 1997/1998. The public witnessed first-hand how defaults on housing loans and mortgaged-backed securities in the US culminated in the bankruptcy of Lehman Brothers in September 2008, which quickly evolved into the financial crisis that befell the US, Europe and the rest of the world.

14.4.1. Theory of Interconnectedness and Financial Networks Financial interconnectedness is indicative of a network consisting of credit exposure, trade relationships, and other transactional linkages, as well as interdependence between financial players (Allen & Babus, 2007; DTCC, 2015). In general,

Macroprudential Policy and Financial System Stability    445 financial interconnectedness refers to the relationships between economic players created by through financial transactions and the supporting agreements. In the financial system, interconnectedness specifically occurs from linkages between: (1) financial institutions, banks and nonbanks; (2) providers of financial infrastructure services; and (3) supply and third-party linkages supporting the financial institutions. For example, banks that provide loans to and borrow from other banks become interconnected through credit exposure on the interbank money market. Similarly, contractual obligations between financial institutions, such as ownership, loans, derivatives and other contracts, also form mutual connections. Furthermore, when financial institutions invest in the same asset, they are interconnected through the same exposure to that asset. Interconnectedness is extremely important because it facilitates contagion, which can exacerbate financial system instability and trigger a systemic crisis (Allen & Gale, 2000). The impact of default at a financial institution, a bank for instance, which is highly interconnected, will quickly propagate to the entire financial system. The strength of the interconnectedness through the financial network between financial institutions will determine how far the default will spur systemic risk in the financial system as a whole. In that context, financial networks can be categorized as complete, incomplete, or disconnected networks, as presented in Figs. 14.3–14.5, respectively, as follows: In the case of a complete network, as shown in Fig. 14.3, all financial institutions are mutually interconnected to one another bilaterally or through another financial institution. Under such conditions, default at any of the financial

Fig. 14.3:  Incomplete Networking.

Fig. 14.4:  Disconnected Network.

446    Central Bank Policy

Fig. 14.5:  Complete Network. institutions would spread quickly to the others, thus creating systemic risk. Vigilance is also required in the case of incomplete networks, as presented in Fig. 14.4. To prevent a systemic impact from the default of one financial institution, for example, Financial Institution No. 3, prevention efforts must be directed toward Financial Institution No. 4 in order to prevent contagion at Nos. 1 and 2. Conditions are very different in the case of a disconnected network, as shown in Fig. 14.5. Default at a financial institution, for example, No. 3 could only propagate to No. 4 and, therefore, systemic risk is low. Interconnectedness through financial networks can occur directly and indirectly (DTCC, 2015). Direct financial interconnectedness occurs through transactions, obligations, contracts and agreements as well as other financial linkages as follows. First, interbank credit exposure occurs through lending and borrowing activities on the interbank money market; swap, forward and other financial derivative transactions on the foreign exchange market; as well as certificates of deposit or securities issued by one bank and held by another. Second, direct interconnectedness through financial market infrastructure linkages, specifically payments, clearing, and transaction settlement through the payment system. Third, through supply and other third-party linkages, providing financial services to several banks, for instance information technology. Fourth, dependence on other financial services between banks, for example, the derivatives market in the case of hedging, custodian banks, and guarantee companies. Meanwhile, indirect financial interconnectedness is possible through various channels, where problems at one financial institution can affect another despite no direct financial or transactional links between the two. First, exposure to securities or other financial assets jointly held by financial institutions, such as bonds, foreign exchange and other securities. Price fluctuations, including interest rates and exchange rate, as well as heightened risk on the securities, especially when the financial markets are distressed, create financial interconnectedness between the financial institutions. Second, mark-to-market losses that trigger fire sales on financial markets also create indirect financial interconnectedness. Fire sales occur when numerous financial assets are sold during a brief period due to capital shortfalls, sudden withdrawals (bank run) due to inadequate liquidity and so on. When a fire sale occurs, tumbling prices and tight liquidity persist. Third, margin calls and haircuts also create indirect financial

Macroprudential Policy and Financial System Stability    447 interconnectedness. Lower collateral prices erode liquidity at lending banks that must meet the margin calls. Fourth, interconnectedness through shadow banking activities. Although most counterparties are still banks, financial product innovation has expanded the scope of counterparties to include nonbanks, such as mutual funds, pension funds, insurance and hedge funds. Housing loan securitization, as occurred in the US, for instance, began in the banking sector but trade in those securities, as well as various structured products, quickly spread to the shadow banking system. Furthermore, such transactions spread to the international financial markets, particularly in Europe. The problem is that nonbank regulation and supervision is not as rigid as for the banking industry in any one jurisdiction or across borders. Consequently, default on a financial security could occur at a nonbank financial institution and then quickly propagate to the conventional banking system and to the entire financial system in the country involved or even to other jurisdictions, thereby triggering a chain reaction of systemic risk from financial interconnectedness. Fifth, financial interconnectedness through information spillovers. Difficulties or default at one bank or financial institution could be perceived by the markets as a negative signal on other banks or financial institutions. Such developments could spark a crisis of confidence on the markets and then trigger a contagious reaction by market players. Such information spillovers can occur quickly not only from one market player to another but also through the mass media, including the television, print media and online. Default at one bank could also trigger bank runs. For example, the default on a structured financial security from housing loans in the US triggered a contagious reaction on the financial markets that culminated in the failure of the entire financial system.

14.4.2. Empirical Evidence for Financial Interconnectedness Realizing the importance of preventing systemic risk in the financial system, several approaches and analytical empirical studies of interconnectedness and financial networks were developed by central banks and academia (ECB, 2010). Central banks, including the Bank of England, Deutsche Bundesbank, ECB and Reserve Bank of India (RBI), used financial network models to analyze interconnectedness on interbank money markets for systemic risk assessment in the banking system as well as the impact on the financial system as a whole. In particular, central banks used financial network models to: (1) understand and visualize the linkages and interactions of interbank lending and borrowing that are not clearly explained by the numbers; (2) assess the risk of credit exposure, liquidity, bank behavior and probability of distress that could emerge from transactions on the interbank money market; and (3) simulate potential systemic risk from default at one bank on the other banks and financial system as a whole, including probability, contagion, and losses incurred. Mapping interbank financial networks and understanding the topology are crucial to formulate regulations and design the restructuring required to mitigate systemic risk (Haldane, 2009). The data used for the analysis is relatively detailed

448    Central Bank Policy and varied, such as: (1) direct bilateral exposure between banks on the interbank money market; (2) various types of transaction on the financial markets, for instance unsecured and secured loans, foreign currency exposure, swaps, forwards, and securities; (3) counterparty data; and (4) data from the payment system. The analyses are conducted using the nominal value and frequency, namely daily, weekly, and monthly. From the data, the first step is typically to map the financial networks. Each bank is represented as a node on the network in the interbank market, the size of which corresponds to the share of transactions at the respective bank. A larger market share equates to a larger node. Meanwhile, the interbank transactional linkages are illustrated as a link, the boldness of which corresponds to the transaction value. The links represent the direction of the transaction. In this case, the direction of the network shows the interconnectedness between banks in terms of assets and liabilities and, therefore, measures total transaction exposure (gross, not netting). The network links are weighted from the total transaction value on the interbank market, or are not weighted. Mapping the financial network helps to analyze banks where potential systemic risk could appear. This is achieved through centrality analysis of the role of each respective bank in terms of transactions on the interbank money market. Fig. 14.6 shows that banks 1, 2, and 3 act as money centers on the interbank money market, while the other banks are interconnected through those three banks. The three banks exacerbate interconnectedness on the interbank money market. Consequently, banks 1, 2, and 3 could potentially trigger systemic risk in the event of default. For instance, if bank 1 experienced default, not only would banks 2 and 3 be at risk but also the other banks that transact with those three.

Fig. 14.6:  Interconnectedness through Three Banks.

Macroprudential Policy and Financial System Stability    449 To deepen our understanding of interbank system dynamics, financial network analysis also explains the role of the bank on the interbank money market using indicators of cohesion, centrality and distance (Allen & Babus, 2007). Such indicators help investigate the statistical characteristics of the interbank system structure. The centrality indicator is useful to understand the distribution of banks on the financial network and determine the strength, effect and control of the respective bank. In Fig. 14.6, bank N is the centrality on the interbank market. The centrality may be in the form of central lender, central borrower, central counterparty or can also imply a balance between lending and borrowing. The cohesion indicator expresses the key linkages on the interbank market in terms of connectivity. Meanwhile, the distance indicator shows the extent of the financial network and how various information propagates through the interbank market. Several models and empirical studies have revealed significant findings in the analysis of financial networks. In addition to the aforementioned study by Allen and Gale (2000), Sachs (2010) investigated the effect of credit exposure distribution between banks on the interbank money market. The study was motivated by observations of strong interconnectedness between large banks on a core network in the interbank money market with numerous small periphery banks that were connected to just one core but not the other banks. This kind of financial network became known as a money center. The study showed that FSS not only depends on the completeness and interconnectedness of the financial network, as revealed in the study by Allen and Gale (2000), but also on the distribution of credit exposure between banks in the interbank money market. Financial networks with a money center, as illustrated in Fig. 14.7, where the assets are concentrated at the core banks, tend to be more stable compared to a system where the banks are all of similar size in a random network.

Fig. 14.7:  Money Center Network.

450    Central Bank Policy

Fig. 14.8:  UK Banking Network.

Fig. 14.9:  Global Banking Network. Several central banks use financial network analysis to assess systemic risk in the financial system. The Bank of England, for example, evaluates the network of large interbank exposures to assess systemic risk in the banking system in the UK, as presented in Fig. 14.8 using Q2/2009 data. Large exposures are defined as a value of liabilities that exceeds 10% of the capital required (Tier 1 and Tier 2). In that network, the nodes illustrate each bank in the UK, the size of which indicates total exposure of one bank to the other banks and nonbank financial

Macroprudential Policy and Financial System Stability    451 institutions in the banking system network. The boldness of the links is weighted proportionally from the exposure of a lending bank. Minou and Reyes (2012) conducted an interesting study of the global banking network in 184 countries for the period of 1978–2010. In Fig. 14.8, the links show capital flows intermediated by banks between individual countries in 2007. The thin lines indicate a capital flow US$ 1–5 billion, the medium lines indicate a capital flow of US$ 5–10 billion and the bold lines indicate a capital flow exceeding US$ 10 billion. The diagram shows the interconnectedness of banking activity in 15 countries, which explains the effect of spillovers and systemic risk from default at a bank in one country on banks in other countries. Moreover, the intensity of the global banking network also tends to be procyclical, implying that expansion and contraction follow the global capital flow cycle. Interconnectedness in the global network tends to increase prior to debt and banking crises and then declines dramatically post crisis. Using financial network analysis, Haldan (2009) found several characteristics with important implications in the assessment of the impact on systemic risk. First, financial networks often appear robust yet fragile. To a certain degree, interconnectedness in a financial network acts as a shock absorber by diversifying and distributing the risks of various transactions among financial institutions. Nevertheless, in the event of default at a financial institution, financial networks also exacerbate conditions and accelerate propagation to the entire financial system. The US subprime mortgage debacle is a good example of this. The long-tailed distribution structure of the financial network was proposed as the cause. As presented in Fig. 14.5, default at a large bank, as a money center on the interbank money market, would trigger a systemic event throughout the banking system. In fact, default at a periphery bank could also have a systemic impact, known as the small-world phenomenon. This would occur if default at the periphery triggered problems at the respective systemic bank. Second, financial network fragilities accelerate under market distress due to hoarding liabilities and fire sales of financial assets. Similar to contagion, reaction can be to hide or flight from the spillovers. The initial reaction of the banks when market conditions are under stress is to withdraw from transactions that create liabilities, holding on to liquid assets. The second reaction, however, is flight by selling assets to avoid deeper losses. In the financial system, the reaction to build liquidity, combined with fire sales, triggers a chain reaction through the financial network, which exacerbates and accelerates the propagation of bad assets or default at one bank into systemic failure. Third, interconnectedness and transaction complexity in a financial network can create uncertainty and undermine price setting, risk assessment and counterparty transaction risk. In terms of credit default swaps (CDS), for instance, direct counterparty transaction risk assessment does not stop at the counterparty. A larger financial network implies that counterparty transactions could have many other counterparty transactions as well. Consequently, assessing the risk of default of counterparty transactions becomes more complex and less assured due to the potential effect on subsequent counterparties in the financial network chain.

452    Central Bank Policy Fourth, financial innovation also affects the complexity of the financial network chain. In the decade prior to the GFC of 2008/2009, for example, structured products exploded, thus breaking up the risk of financial assets and issuing new financial products, such as MBS and credit default orders (CDOs). Financial product innovation precipitated interconnectedness among increasingly complex claims contracts, thus blurring price setting and risk assessment. The problem is that innovative financial products are issued by an institution with a high rating, which are easy to sell due to guaranteed higher returns and risks that appear low. As more products are issued, however, risk assessment becomes less precise. Consequently, as a financial product turns bad, network complexity through financial product innovation exacerbates potential systemic risk. Fifth, the benefits of financial product innovation diversification and phased risk management are lost due to replication and imitation on the financial markets. In the financial sector, diversification through financial product innovation, such as the securitization of housing loans and CDS, is desirable to increase returns and as a part of risk management. As such products begin to provide higher returns, however, other financial institutions imitate, leading to a proliferation of trade on the financial markets. Commercial banks begin to act as investment banks, for example, through housing loan securitization, selling such instruments on the financial markets as part of a new originate and distribute business model. Investment banks create financial products through fund managers, targeting maximum returns with risks that appear to be mitigated through financial product innovation. These managed funds are subsequently sold to various investors, including mutual funds, pension funds and insurance. What remains unknown, however, is that the risk of default is increasing, interconnectedness through the financial network is more complex and systemic risk is building. The five characteristics are clearly apparent in the case of the massive US housing credit crisis along with the proliferation of structured products. In general, such developments created interconnectedness and complex financial networks with all five of the characteristics. During an economic upswing, a financial network flourishes due to the attractive returns offered, which are imitated and traded extensively on the financial markets, and appear robust yet are actually vulnerable to shocks. Therefore, a small shock of bad financial products triggers a chain reaction on the financial markets and subsequently feeds through to systemic failure due to complex interconnectedness in the financial network. Spreading, the markets hoard liquidity and initiate fire sales of their financial assets, thus exacerbating the decline and propagation into a large financial crisis. Such characteristics of interconnectedness and financial networks demand vigilance as an integral element of maintaining FSS.

14.5. Macroprudential Policy Theory and Practices As explained previously, the central bank is the most appropriate institution to conduct macroprudential (regulation and supervision) policy. The central bank’s mandate to maintain FSS is primarily achieved through such macroprudential policy. In addition, the central bank supports FSS through monetary policy,

Macroprudential Policy and Financial System Stability    453 payment system policy and foreign capital flow management, including jurisdiction to regulate and supervise the money markets and foreign exchange market as an integral part of monetary policy.

14.5.1. Principles, Targets, and Instruments As mentioned at the beginning of this chapter, macroprudential policy encompasses the macroregulation and supervision of financial services institutions, with a focus on systemic risk to foster FSS. The primary target is to mitigate the risks that emerge from procyclical macrofinancial linkages and the accumulation of systemic risk that appears from interconnectedness and the networks between and in financial institutions, markets and infrastructures, including the payment system. The primary target of macroprudential policy is to prevent the build-up of risk from boom-bust financial cycles because of factors in the financial system itself as well as interaction with the domestic and international economies (time series dimension). The second target is directed toward reinforcing financial system resilience and mitigating contagion risk from interconnectedness and networks in the financial system (cross-section dimension).3 The two main targets of macroprudential policy, from a macroperspective and focusing on systemic risk in terms of FSS, differ from the targets of microprudential regulation and supervision, which are to maintain the soundness of individual financial institutions, including banks and nonbanks. The sound of individual financial institutions is necessary but insufficient to maintain FSS. Default at an individual financial institution, if not considered systemic, must be viewed as an insolvency problem due to mismanagement, weak risk management or lack of competitiveness. The closure of an individual financial institution due to default can be seen as a natural process of banking system restructuring and, simultaneously, as a test of overall financial system resilience. It is different for systemically important banks (SIB). In accordance with the four criteria stipulated by BIS (2012), namely size, interconnectedness, complexity, and substitutability, the PPKSK Act of April 2016 defines systemic banks as banks that, because of the size of their assets, capital and liabilities; broad network or transaction complexity of banking services; as well as interconnectedness with other financial sectors, could precipitate the failure of some or all other banks or financial sectors, operationally and financially, in the event of distress or default. In that context, failure of a domestic systemically important bank (D-SIB) would likely trigger systemic risk and failure of the financial system as a whole. Per the 3

In addition to the two primary targets, another target of macroprudential policy, as conducted by the European Central Bank (ECB), is to develop prudential regulations that, from a general system perspective, can facilitate market incentives and disincentives (structural dimension).

454    Central Bank Policy Table 14.3:  Macroprudential Instruments Applied in Asian Countries. SIN HKG INA MAL KOR IND THA PHI PRC Total

%

Credit related

13

5

11

6

23

 6

15

 1

 9

 92  49.2

Liquidity related

 0

0

 7

3

 3

18

 1

10

31

 79  42.2

Capital related

 1

1

 1

0

 2

 4

 0

 6

 1

 16   8.6

Total

14

6

19

9

28

28

16

17

41

187

100

Source: Lee, Asuncion, and Kim (2015).

PPKSK act, systemic banks are required to: (a) meet specific capital and liquidity adequacy ratios; and (b) prepare an action plan that at least contains the obligations of the controlling shareholders and/or other parties to inject capital and convert certain types of debt into capital. Which macroprudential policy instruments can be used? Lim et al. (2011) proposed ten instruments, as listed in Table 14.3, to control procyclicality and systemic risk in relation to credit exposure, foreign exchange, liquidity and capital. In terms of credit procyclicality, the most common instruments are the LTV, debt-to-income ratio (DTI) and restrictions on credit growth to certain sectors. Concerning foreign exchange exposure, the instruments include the net open position (NOP), restrictions on foreign currency loans as well as mandatory hedging and regulating the maturity of external debt. Regarding liquidity instruments, the reserve requirement is generally used, which can be adjusted in accordance with prevailing liquidity conditions. Meanwhile, capital can be strengthened to mitigate procyclicality and systemic risk through a countercyclical capital buffer (CCyB), loan loss provisions in line with procyclical credit risk and profit-sharing regulations. Galati and Moesner (2014) categorized macroprudential policy instruments according to risk, namely the risks associated with leverage/credit boom/asset price bubbles, liquidity/market risks as well as interconnectedness/market structure risks in line with the risk dimension, namely dynamically over time (time series) or statically across sectors (cross section). In general, LTV policy and the CCyB are used to mitigate procyclicality in the time series dimension for the first type of risk. Time series instruments are also used to overcome liquidity/market risks, such as the loan-to-deposit ratio (LDR) in accordance with prevailing conditions as well as additional liquidity requirements for SIB, in addition to static instruments, for example, additional capital requirements for derivative debt and levies on non-core liabilities. In terms of interconnectedness and market structure risks, cross-sectional instruments are typically employed, including more stringent liquidity and capital requirements for SIB or additional deposit insurance premiums to mitigate systemic risk. Macroprudential instruments can be loosened and tightened as required in accordance with the level of risk anticipated from procyclicality and systemic

Macroprudential Policy and Financial System Stability    455 risk. Claessens, Gosh, and Mihet (2014), categorized macroprudential policy instruments that can be applied during periods of economic expansion and contraction or to mitigate contagion risk from interconnectedness and financial networks. The instruments can take the form of lending restrictions, limitations on instruments and activities or restrictions on financial sector balance sheets and buffer-based policies. For example, during a period of economic expansion, the LTV, RR and CCyB ratios could be raised but then subsequently relaxed during a period of economic contraction. Adjustments can also be made to regulations concerning provisions, margins and haircuts. Furthermore, contagion could be mitigated through more stringent regulations on banks or systemically important financial institutions. Several macroprudential and microprudential regulation and supervision instruments may, in fact, be the same because they are both based on assessments of liquidity, market and credit risks. Nonetheless, the targets and perspectives are different. Macroprudential instruments aim to maintain FSS from a macrofinancial (or procyclicality risk) perspective and focus on systemic risk. Conversely, microprudential instruments target the soundness of individual financial institutions with a focus on consumer protection, which include liquidity ratios, minimum capital requirements per the risk profile and prudent risk management. For the few instruments that have both a macroprudential and microprudential dimension, the following three regulatory aspects must be considered: individual soundness, systemic risk and procyclicality. Therefore, the instruments can be set incrementally in line with the three regulatory dimensions based on assessments of prevailing financial system and macroeconomic conditions. This kind of approach solves many of the issues associated with instruments that have macroprudential and microprudential targets. Capital regulations, as stipulated by the FSB in November 2015, require a minimum level of Total Loss Absorbency Capacity (TLAC) for Global Systemically Important Banks (G-SIBs), as an example of this kind of approach. The magnitude of TLAC ensures that G-SIBs maintain an adequate level of capital to absorb losses and recapitalization capacity in the event of business default or resolution, thus the critical bank functions are maintained without incurring losses to the state or threatening FSS. Consequently, TLAC is typically set at between 16% and 20% for capital requirements based on risk-weighted assets (RWA) and 6%–6.75% based on total bank exposure. The regulations are applicable to G-SIBs specified prior to 2015. First, Basel III stipulates a capital requirement of 8% in accordance with the risk profile, which must be met by all banks, systemic or otherwise. The 8% threshold aims purely to ensure the soundness of individual banks. Second, various capital buffers, under Basel III, are also required, including: the capital conservation buffer for systemic risk, CCyB to mitigate procyclicality, and a capital surcharge for G-SIBs. As a macroprudential instrument, the rate of the CCyB can be raised or lowered in accordance with assessments of procyclicality in the current period. Third, the additional capital required and subordinated debt instruments with a maturity of more than one year may be included in the calculation of TLAC. A structured regulatory approach based on individual

456    Central Bank Policy soundness targets, systemic risk and procyclicality, such as TLAC, can also be applied to other prudential banking instruments that have a micro and macroprudential perspective.

14.5.2. Application in Various Jurisdictions After the GFC of 2008/2009, central banks in various jurisdictions applied different macroprudential policies to mitigate macrofinancial risks (procyclicality) and systemic risk that could undermine FSS, in addition to monetary policy and foreign capital flow management to control inflation, stabilize exchange rates and maintain a healthy BOP. The same phenomenon occurred at central banks in Asia, which proved that the application of macroprudential instruments supports FSS using different types of instrument to mitigate macrofinancial and systemic risks in line with prevailing conditions in each country (Lee, Asuncion, & Kim, 2015). Table 14.3 presents the macroprudential instruments applied in 10 Asian countries to control credit, including LTV and DTI, which were most commonly applied in Indonesia, South Korea, Singapore and Thailand. Meanwhile, liquidity-related macroprudential instruments, such as RR and NOP, were applied in China, India and the Philippines. In contrast, capital-related macroprudential instruments, such as the CCyB, at that time, were not commonplace, except in India. Of the three instrument types, credit-related macroprudential instruments were the most popular among central banks in Asia at that time. Of the 10 countries, China applied the most macroprudential policy instruments from 2000 to 2013 to regulate balance sheet exposure at financial institutions, rein in the credit boom and escalating house prices, and simultaneously avoid systemic risk. From 2006 to 2008, the People’s Bank of China (PBoC) and China Banking Regulatory Commission simultaneously implemented macroprudential instruments to reinforce FSS. Macroprudential policy was tightened as many as 40 times to offset heightened financial system instability risk at that time, including 31 times using liquidity-related instruments, specifically the reserve requirement. The experience demonstrated that tightening credit-related macroprudential policy instruments successfully dampened the credit boom and rising house prices but failed to reduce debt accumulation (leverage). Meanwhile, liquidity-related macroprudential instruments initially succeeded in reducing leverage, but the impact on rising new house prices only become apparent several periods thereafter. South Korea systematically applied a number of macroprudential instruments prior to the GFC of 2008/2009 and even before the Asian financial crisis of 1997/1998, including several liquidity ratios to overcome fragilities in the banking system and in terms of foreign currency transactions. In November 2009, for example, the Bank of Korea promulgated regulations requiring the banks to increase their share of long-term foreign loans from 80% to 90% and then subsequently to 100%. In addition, the Bank of Korea also introduced restrictions on foreign currency derivatives at banks and reserves for foreign currency loans, as well as tighter liquidity ratios for domestic banks. Furthermore, the Bank of Korea tightened macroprudential policy in 2006 and 2007 to overcome

Macroprudential Policy and Financial System Stability    457 spiraling house prices, which were exacerbated by the US crisis in 2007–2008. In fact, from 2000 to 2013, the Bank of Korea issued macroprudential instruments approximately 28 times, most of which were credit related. The impact of tighter macroprudential policy quickly became apparent in terms of slowing credit expansion and leverage, but only subsequently did the impact on rising house prices become evident. In India, bank credit growth has become an important consideration underlying monetary and macroprudential policymaking at the RBI. Consequently, macroprudential policy has been instituted since 2004 through several instruments, including liquidity and capital-related instruments, to maintain FSS. The primary focus is concentrated on excessive credit growth in the housing sector. To that end, capital-related macroprudential policy, as applied by the RBI, focuses on the banking industry through application of the CCyB. In addition, tighter regulations on risk weights and asset reserves have also been enforced for housing loans. RBI also applied liquidity-related macroprudential instruments, primarily by raising the reserve requirement. Although credit growth persisted, the range of macroprudential policy instruments controlled excessive leverage and rising house prices and, therefore, helped to prevent an accumulation of systemic risk and asset price bubbles. The challenge was, however, to control financing to the housing sector from the shadow banking system, including from developers, as well as rising house prices (Table 14.4). In Singapore, although the Monetary Authority of Singapore (MAS) maintains stringent regulation and supervision over the financial system, financial system instability risk has increased since 2010 due to rising house prices, which have already exceeded the peak recorded in 2008. Higher property prices have stoked concerns of rising inflation expectations and systemic risk because most property is purchased using comparatively cheap bank loans. Housing loans have continued to accelerate and household debt has also escalated over the past three years to 76% of GDP at the end of 2012, with large exposure at domestic banks to the housing sector. In response, MAS quickly strengthened surveillance and assessments of FSS as well as the policies required, one of which was to tighten regulations in the property sector by raising the loan installment-to-income ratio (IIR), lowering the threshold of the LTV, introducing an additional stamp duty levied on property purchases and increasing the minimum cash downpayment for property purchases. The regulations specifically targeted the speculative segment of the property market. In general, tighter macroprudential policy successfully reduced house prices and housing loans and, therefore, become an integral element of maintaining FSS in Singapore. In addition to the four countries mentioned, numerous other Asian countries have also applied macroprudential policy, including Hong Kong, Malaysia, Philippines, Thailand and Indonesia. In general, macroprudential policy is used to mitigate systemic risk and, thus, maintain FSS during periods of excessive credit growth and rising property prices, in addition to other sectors, such as the automotive sector in Indonesia. Such problems were also linked to economic booms and/or the influx of foreign capital flows to the Asian region from 2010 to 2013 due to loose monetary policy in various jurisdictions after the GFC in 2008/2009,

458    Central Bank Policy Table 14.4:  Macroprudential Instruments: Types and Risk Dimensions. Risk Types

Risk Dimension: Time Series or Cross-sections

Leverage/credit/ asset price booms

Time Series: • Countercyclical capital buffers • Through-the-cycle valuation of margins or haircuts for collateral used in securitized funding markets (like repos) • Countercyclical change in risk weights for exposure to certain sectors • Time-varying LTV, Debt-To-Income (DTI) and LoanTo-Income (LTI) caps • Time-varying caps and limits on credit or credit growth • Dynamic provisioning • Rescaling risk-weights by incorporating recessionary conditions in the probability of default assumptions (PDs)

Liquidity/market risk

Time Series: • Time-varying systemic liquidity surcharges • Levy on non-core liabilities • Time-varying limits in currency mismatch or exposure (e.g., real estate) • Time-varying limits on loan-to-deposit ratio • Stressed VaR to build additional capital buffer against market risk during a boom Cross-sections: • Capital charge on derivative payables • Levy on non-core liabilities

Interconnectedness/ market structure/ financial infrastructrure

Cross-sections: • Higher capital charges for trades not cleared through CCPs • Systemic capital surcharges • Systemic liquidity surcharges • Powers to break up financial firms on systemic risk concerns • Deposit insurance risk premia sensitive to systemic risks • Restrictions on permissible activities (e.g., ban on proprietary trading for SIBs)

Source: Galati and Moessner (2014).

as explained in Chapter 13. Macroprudential instruments vary from one country to another, including credit, liquidity and capital-related instruments, depending upon the specific problems faced and prevailing financial system conditions. The most common instruments include the LTV and DTI ratios, reserve requirement and CCyB. The application of macroprudential policy has proven effective in terms of controlling excessive credit growth and rising property prices and, therefore, is an integral element of preventing build-ups of systemic risk, while also helping to maintain FSS.

Macroprudential Policy and Financial System Stability    459

14.6. Concluding Remarks This chapter has elucidated an invaluable lesson from the GFC of 2008/2009, namely the need to maintain FSS as well as macroeconomic stability in order to support robust and sustainable economic growth. The chapter focused on interconnectedness between the financial system and economy, or macrofinancial linkages, as reflected in financial procyclicality and systemic risk. The GFC of 2008/2009 and previous crisis episodes were all preceded and caused by an accumulation of systemic risk that surfaced due to excessive credit growth, asset price bubbles, particularly property prices, uncontrolled external debt, and sudden capital reversals. It is impossible, however, to prevent the various crisis triggers through microprudential regulation and supervision, macroeconomic policy (fiscal and monetary policy) and foreign capital flows management. Macroprudential policy encompasses the regulation and supervision of financial institutions from a macroperspective, namely macrofinancial linkages, with a focus on systemic risk to maintain FSS. There are two overarching targets: preventing financial procyclicality (time series dimension) and mitigating systemic risk (cross-section dimension) that appear from macrofinancial linkages. The instruments used to control procyclicality and mitigate systemic risk are creditrelated, foreign currency-related, liquidity-related, and capital-related. Macroprudential instruments can also be adjusted per the types of risk, namely leverage/ credit boom/asset price bubble risks; liquidity/markets risks; and interconnectedness/market structure risks in line with the risk dimension, either dynamically over time (time series) or statically across sectors (cross section). Furthermore, macroprudential policy can be loosened or tightened according to the level of risk anticipated from potential procyclicality and systemic risk. The central bank is the most appropriate institution to implement macroprudential policy. In general, macroprudential policy instruments have become commonplace at central banks in most jurisdictions, including Asia, in addition to monetary and exchange rate policy as well as foreign capital flow management. The GFC of 2008/2009 ultimately confirmed the importance of deepening and implementing policy to control macrofinancial linkages and to avoid a build-up of systemic risk that not only endangers FSS but also threatens macroeconomic stability and sustainable economic growth. The problems are becoming increasingly complex as foreign capital flows gain mobility through stronger global financial integration. The challenge is how central banks integrate monetary policy, macroprudential policy, and foreign capital flow management into an optimal policy mix to achieve price and exchange rate stability on one hand, and also support sustainable economic growth on the other. This issue will be explored and addressed in Chapter 15 concerning the central bank’s policy mix.

This page intentionally left blank

Chapter 15

Central Bank Policy Mix 15.1. Introduction The global financial crisis of 2008/09 had fundamental implications on the central bank’s mandate, policy and institutional arrangements. In terms of the mandate, central banks can no longer merely focus on price stability alone but also on maintaining financial system stability. To achieve that dual mandate, central banks institute monetary policy, foreign capital flow management, and macroprudential policy within an optimal policy mix, in addition to payment system policy. Institutionally, the central bank strengthens internal capacity through assessments, policy scenarios, and decision-making that underlie the policy mix. Furthermore, the central bank also coordinates with the government in terms of macroeconomic policy as well as with other relevant authorities to maintain financial system stability. During the two decades preceding the global financial crisis of 2008/09, central banks in various jurisdictions focused on price (inflation) stability to support economic growth. This was driven by academic consensus at the time regarding the trade-off between inflation and economic growth in the short term and monetary policy neutrality in the long term, per a synthesis of Keynesian and Neoclassical economics, as elaborated in Chapter 3. From a political economic perspective, that focus represented a reformulation of central banking policy and institutional reforms through crisis resolution in the wake of the Asian financial crisis of 1997/98. In fact, the bank supervision function was transferred away from the central bank in several countries, including South Korea and Indonesia. In practice, a monetary policy framework that targets price stability, otherwise known as the inflation targeting framework (ITF), became popular and was embraced in various advanced and emerging market economies (EMEs) alike, as elucidated in Chapters 8 and 9. The global financial crisis of 2008/09 fundamentally changed the perspective of the central bank’s mandate and policy practices but not because ITF-based monetary operations had failed. In contrast, ITF had successfully achieved low inflation, stimulated economic growth, and reduced interest rates in many jurisdictions. In fact, monetary stability had spurred economic booms in many countries, including the United States, during an era that became known as the

Central Bank Policy: Theory and Practice, 461–515 Copyright © 2019 by Emerald Publishing Limited All rights of reproduction in any form reserved doi:10.1108/978-1-78973-751-620191021

462    Central Bank Policy Great Moderation. The problem was that a focus on price stability alone, made central banks less responsive to crisis risks contained in macrofinancial linkages. On the other hand, microprudential regulation and supervision focuses on the soundness of individual financial institutions, especially the banks. The US crisis is a concrete example of where macrofinancial linkages during an economic boom in a superpower heightened systemic risk that culminated in a devastating financial crisis and impacted all corners of the globe. In addition, the global financial crisis of 2008/09 also had other implications for central bank policy, namely the influx of volatile foreign capital flows to EMEs as a consequence of monetary expansion in advanced countries to stimulate the economic recovery process. Such conditions further complicated the central banks’ policy response to achieve price (and exchange rate) stability and to maintain financial system stability. In brief, to support sustainable economic growth, the central bank cannot merely rely on monetary policy to achieve price (and exchange rate) stability alone. Indeed, the monetary policy strategy of the central bank, as described in Chapters 6–9, as well as institutional strengthening, as explained in Chapters 10–12, are no longer considered sufficient. Central banks are now required to support financial system stability through macroprudential regulation and supervision from a macrofinancial perspective and with a focus on systemic risk as elucidated in Chapter 14. Furthermore, the central bank is also required to apply foreign capital flow management in order to provide the maximum gains in terms of sustainable economic growth, without triggering the risk of macroeconomic and financial system instability due to a balance of payments crisis, external debt crisis, and/or sudden-stop capital reversal crisis, as explored in Chapter 13. The question, therefore, is how the central bank can integrate and combine those three policies, namely monetary policy, macroprudential policy, and foreign capital flow management, into an optimal policy mix? This chapter discusses the latest ideas and practices regarding the importance of the central bank’s policy mix to achieve price stability and support financial system stability. Various ideas that originated at international forums and the central banking community will be proposed, along with several theories currently being developed in the academic community. Following on from this introduction, the discussion is broken down into four main topics. The first part deals with the conception of the central bank policy mix in terms of integrating the dual targets of price stability and financial system stability, formulation of the monetary policy mix, macroprudential policy, and foreign capital flow management, as well as the transmission mechanisms to attain the dual targets. The second and third sections introduce and examine two common modeling approaches used to formulate the policy mix, namely the structural macroeconomic model and the Dynamic Stochastic General Equilibrium (DSGE) model. The fourth section explains how the policy mix at Bank Indonesia has been formulated and applied during three distinct episodes since 2010. In general, the policy mix applied by Bank Indonesia plays an important role in terms of strengthening national economic resilience to global economic and financial shocks and uncertainty since the global financial crisis of 2008/09.

Central Bank Policy Mix     463

15.2. Conceptual Dimension of the Central Bank Policy Mix Fundamentally, the central bank policy mix represents the optimal integration of monetary policy, macroprudential policy, and foreign capital flow management, as applied by the central bank to achieve price stability and maintain financial system stability. The terminology has a broader scope than the Flexible Inflation Targeting Framework (FITF) proposed in several studies (Agénor & da Silva, 2013; Svensson, 2009; Ito, 2010; Juhro, 2015; Warjiyo, 2013a ). First, and departing from FITF, the ultimate target is not only price stability but also maintaining financial system stability. Second, and this is similar to FITF, the instruments used are monetary policy, macroprudential policy, and foreign capital flow management in one optimal mix. And third, which is also like FITF, policy mix formulation requires an analysis framework and macroeconomic projections that take into consideration macrofinancial linkages. Formulation of the central bank policy mix also requires assessments of the more complex transmission mechanisms, not only of the three component policies on price stability but also on support for maintaining financial system stability. Three aspects that are the fundamental concepts of the central bank policy mix framework are discussed in the following subsections.

15.2.1. Integration of the Price Stability and Financial System Stability Targets Chapter 14 discussed the need for central banks to maintain financial system stability. Therefore, support has recently gained momentum for central banks to have a dual mandate, namely price stability and supporting financial system stability (Acharya, 2015; Bank for International Settlements (BIS), 2012; Kawai & Morgan, 2012; Sinclair, 2000). How can a central bank integrate both mandates? Several ideas have been developed in the context of the close correlation between monetary stability and financial system stability. Regarding the first mandate, namely to achieve price stability, consensus has been reached among EMEs that central banks should not only focus on inflation but also maintain exchange rate stability, as explained in Chapter 13. The nascent issue is, therefore, how a central bank can address other price dimensions, such as financial asset prices (stocks and bonds) as well as property prices. Concerning the second mandate, namely to support financial system stability, as explained in Chapter 14, central banks institute macroprudential policy through the regulation and supervision of financial institutions from a macrofinancial linkages perspective and with a focus on systemic risk, in addition to monetary policy (including foreign capital flow management) and payment system policy. The importance of price stability, as an integral part of macroeconomic stability and financial system stability to support sustainable economic growth, is inevitable. One dimension of price stability is goods and services, commonly referred to as inflation, which is measured using the price change above the base

464    Central Bank Policy Consumer Price Index (CPI).1 Another dimension of price stability is with respect to other currencies, commonly referred to as exchange rates. Unlike conditions in advanced countries, the role of exchange rate stability is crucial in EMEs due to a number of salient considerations. In general, foreign exchange markets in EMEs remain inefficient, thus often leading to volatile fluctuations because of domestic economic shocks and/or foreign capital flows. The exchange rate impacts inflation directly through import prices and indirectly through domestic demand. Exchange rate fluctuations also influence financial system stability through their impact on the balance sheets of financial institutions as well as stock prices and bond yields. In fact, exchange rate (currency) shocks often stoke concerns in the business community and among the public and, therefore, are a good barometer of political stability in many EMEs. The various considerations outlined ensure that exchange rate stability remains an important part of the price stability mandate, in addition to inflation, as is the case in Indonesia. This is achieved through monetary policy, using interest rate instruments, foreign exchange intervention, and foreign capital flow management, as discussed in Chapters 4 and 13. A new dimension has become the focus of post-global crisis discussions between central bankers and academia, namely how the central bank should deal with other asset prices, such as stock and bond prices as well as property prices? Several economists suggested the central bank should maintain focus on price stability in support of maintaining financial system stability (Bordo, 2007; Schwartz, 1995). Nevertheless, the global financial crisis of 2008/09 proved that price stability was insufficient and, therefore, the central bank should also support financial system stability (Borio & Lowe, 2002; Borio et al., 2003; Shirakawa, 2012). In fact, the global financial crisis demonstrated that too great of a central bank focus on short-term price stability to maintain economic stability actually created instability due to asset price bubbles, debt accumulation, excessive credit growth and other macrofinancial imbalances that culminated in a crisis. This is the essence of the financial instability hypothesis in a capitalist economy, as coined by Minsky (1982). Concerning the linkages between monetary policy and property prices, an interesting debate continues to rage regarding whether the success of the Federal Reserve in terms of achieving low and stable inflation and interest rates for two decades during the Great Moderation was a contributing factor to the financial crisis that befell the United States. Taylor (2010), for example, stated that the Fed’s policy during that time was excessively loose and the Federal Funds Rate (FFR) was too low for a prolonged period, which precipitated a housing sector boom and, therefore, macrofinancial imbalances that culminated in a crisis. Bernanke (2010), on the other hand, disagrees with that point of view. He argued that

1

CPI measures prices at the consumer level as a weighted average of each good or service. Inflation can also be measured at the producer level using the Producer Price Index (PPI). Another measure of inflation is the GDP deflator, which measures the level of prices of all components in an economy in the current year against a base year when calculating nominal and real GDP.

Central Bank Policy Mix     465 the main cause of excessive housing bubbles was the various types of cheap and attractive credit and mortgage-backed securities against a backdrop of weak lending and underwriting standards. Therefore, a more appropriate solution would be prudential regulation rather than monetary policy. Meanwhile, Filardo (2001) found that the monetary authority must respond to rising asset prices while it has access to trusted information on inflation and output, although it is difficult to differentiate between the cause of rising asset prices, namely due to fundamental factors or bubbles. The latest empirical evidence after the global financial crisis of 2008/09 has shown the importance of central bank policy in response to asset price bubbles, particularly property prices. Jorda, Schularick, and Taylor (2014), for instance, proved the importance of linkages between monetary conditions, house prices and credit growth from data stretching over 140 years from 14 advanced countries. Specifically, the empirical evidence showed that a 1% decline in the policy rate would raise house prices (in the ratio to income) and housing loans (in the ratio to GDP) by more than 4% and 3%, respectively, over a horizon of 4 years. History has shown that potential instability from loose monetary policy must be considered seriously in the narrow terms of price stability on CPI inflation compared to the adverse impact that may result on the economy. Meanwhile, Williams (2015) suggested that there is an expensive trade-off using monetary policy to mitigate house prices when there is a conflict between the targets of macroeconomic stability and financial system stability. Using the same data, a 1% increase in the monetary policy rate was shown to depress real house prices (after considering inflation) by more than 6% over 2 years, while real GDP per capita fell by nearly 2%. Such empirical evidence confirms that central banks must equally consider price stability and financial system stability when formulating monetary policy. The discussion then leans toward the issue of how monetary policy responds to macrofinancial imbalances and a build-up of systemic risk, which manifest in procyclical housing bubbles, credit booms, external debt, and volatile foreign capital flows. This issue is also linked to another debate, lean vs clean: namely whether it would be better for a central bank to lean against the wind in order to prevent bubbles from bursting or better to wait for the bubble to burst before cleaning up the mess through aggressively loose monetary policy. The “clean” view was embraced by the Federal Reserve under Alan Greenspan. Some of the basic considerations include that an investment boom is driven by increased productivity, bubbles are due to a lower risk premium and irrational exuberance, and raising interest rates might be ineffective in mitigating the bubbles and actually burst them, thus adversely impacting the economy (Greenspan, 2002). Nonetheless, the global financial crisis of 2008/09 dispelled that view. The Crisis clearly showed potential risk from excessive credit growth and leverage, driven by bubbles. Therefore, it is considered more appropriate for the central bank to lean against the wind to prevent bubbles rather than cleaning up the devastating crisis impact. Experience from the Australian case, for instance, showed that leaning monetary policy can be successful. In response to excessive credit growth in the housing sector during 2002 and 2003, the Reserve Bank of Australia gradually

466    Central Bank Policy raised its policy rate despite inflation remaining under control and on target (Bloxham, Kent, & Robson, 2011). The justification for a tighter monetary policy stance in terms of the ITF was that the Reserve Bank of Australia was increasingly concerned about excessive credit growth. Consensus now strongly supports leaning against the wind and for central banks to seriously consider aspects of financial system stability, even though financial system stability is not part of the mandate. Bernanke (2009), for instance, stated that the Federal Reserve played a significant role in terms of mitigating the crisis in the United States and, therefore, must maintain its institutional capacity to support financial system stability and foster economic recovery with low inflation. The second central bank mandate, namely to maintain financial system stability, also contains several applicable aspects and policy instruments. Chapter 14 has already explained that maintaining financial system stability is determined by several salient factors, namely the soundness of individual financial institutions, controlling procyclical lending and housing bubbles, domestic macroeconomic stability (inflation and fiscal deficit) and external stability (exchange rate, current account deficit, external debt, and risk of sudden capital reversal), interconnectedness and financial networks, as well as controlling the risk of information contagion. Those aspects, in general, form part of the FSS indicators, such as the Financial Soundness Indicator (FSI), as well as Early Warning Indicators (EWI).2 Furthermore, those aspects also show where a central bank supports financial system stability through policy coordination with the government and other relevant authorities. The central bank’s mandate to support financial system stability is primarily achieved through macroprudential policy, namely the regulation and supervision of financial institutions from a macrofinancial perspective with a focus on systemic risk. As discussed in Chapter 14, macroprudential policy is principally directed toward two main targets to maintain financial system stability. The first target is to prevent an accumulation of risk from a boom-bust financial cycle, 2

Several studies have developed measures for the key aspects of FSS. In general, however, simple FSS indicators, covering default risk and asset prices, are inadequate. Concerning the soundness of individual financial institutions, the IMF developed the FSI, which encompasses indicators of capital adequacy, liquidity, credit quality, income and profitability, as well as sensitivity to market risk (IMF, 2001). The IMF and FSB also collaborated to develop and analyze Early Warning Exercises (EWE), which are typically presented each year in April and October. This assessment covers: (1) sectoral and market fragilities in the external, fiscal and corporate sectors, asset prices, market valuation, and bubble detection, as well as risk-taking behavior on the financial markets; (2) modeling sovereign risks, especially systemic risks; and (3) assessing the systemic implications of spillover and contagion analysis, analysis of Large Complex Financial Institutions (LCFI) as well as global scenarios (IMF, 2010). Another ambitious IMF project is to develop a more comprehensive and integrative measure of systemic risk, known as “SysMo” for six key risks, namely financial institutions, asset prices, sovereign risks, the broader economy, cross-border linkages, and crisis risks (Blancher et al., 2013).

Central Bank Policy Mix     467 specifically credit booms and housing bubbles, either due to internal factors of the financial system itself or interaction between the domestic and international economies. The main macroprudential policy instruments available to control financial procyclicality include: the loan-to-value (LTV) ratio or debt-to-income ratio (DTI), countercyclical capital buffer (CCB), and loan-to-deposit ratio (LDR) as well as additional liquidity requirements for systemic banks. Meanwhile, the second target of macroprudential policy is to strengthen resilience and mitigate contagion risk from interconnectedness and financial networks in the financial system. The instruments available include mapping, regulation and interbank money market restructuring to control systemic risk from the perspective of interconnected exposure in terms of liquidity, credit, counterparty transactions, and the operational failure of financial infrastructure, including the payment system. In addition, more stringent liquidity and capital requirements can be applied to systemic banks or additional deposit insurance premiums to mitigate systemic risk. In addition to macroprudential policy, central banks also support financial system stability through monetary and payment system policy and foreign capital flow management. Monetary policy supports financial system stability by preventing economic shocks that could threaten financial system stability through controlled inflation, exchange rate stability, and a managed current account deficit. Monetary policy instituted through interest rate instruments and the statutory reserve requirement can be used to lean against risk accumulation due to financial procyclicality, particularly excessive credit growth, and housing bubbles. In addition, monetary policy also covers the regulation and supervision of money markets and the foreign exchange market, which also supports financial system stability from the perspective of mitigating systemic risk that can emerge from interconnectedness and financial networks, especially on the interbank money market and foreign exchange market. The same considerations are at play concerning central bank jurisdiction in the payment system, which supports financial system stability in terms of interconnectedness and networks in the case of uninterrupted transactions, instrument security, and payment system infrastructure resilience. Meanwhile, the central bank’s authority regarding foreign capital flow management also supports financial system stability by preventing economic shocks from currency instability risk, sudden-stop, and balance of payments crises as well as excessive external debt accumulation.

15.2.2. Basic Concepts of the Central Bank Policy Mix The previous section explained the close linkages between price stability and financial system stability as the dual mandate of the central bank after the global financial crisis of 2008/09, and how monetary policy, macroprudential policy, and foreign capital flow management are mutually complementary in pursuance of that dual mandate. The question, therefore, is how the central bank integrates those instruments into an optimal policy mix to achieve the dual mandate? As described in Chapters 6–9, the monetary policy framework established at various central banks has been proven to achieve price stability but with a trade-off in

468    Central Bank Policy terms of economic growth. The most common instruments are interest rates and liquidity on the interbank money market to influence domestic demand through the interest rate, exchange rate, money supply, credit, risk-taking behavior, and market expectations channels. That monetary policy framework is supported by an analysis system and macroeconomic projections, accompanied by simulations of a consistent policy response, scheduled policymaking process, as well as transparent communications, as explained in Chapters 10–12. Furthermore, as elucidated in Chapter 13, the impact of the global economy through exchange rates, foreign capital flows, and the balance of payments has also already been included in the macroeconomic projections. In fact, a policy mix response between interest rate instruments, intervention on the foreign exchange market and foreign capital flow management is practiced by various central banks in EMEs to overcome the monetary policy trilemma in an open economy. The established monetary policy framework could easily accommodate macroprudential policy to manage macrofinancial imbalances and systemic risk in order to support financial system stability as mentioned in Chapter 14. The existing framework must be expanded to cover macrofinancial linkages in the financial system. In addition to the existing analysis as well as macroeconomic and external projections, analyses, and projections of macrofinancial imbalances are required along with interconnectedness in the macroeconomic assessment. To that end, macrofinancial assessments should focus on four types of procyclicality and systemic risk that are often cited as the causes of a crisis, namely assets (financial and property), price bubbles, credit booms, external debt accumulation, and volatile foreign capital flows. Macroeconomic and macrofinancial analyses and projections should underlie formulation of the optimal policy mix to be instituted by the central bank. Fig. 15.1 illustrates how the macroprudential policy framework is integrated into the monetary policy framework already established at the central bank.

Fig. 15.1:  Integration of the Monetary Policy and Macroprudential Policy Framework.

Central Bank Policy Mix     469 The upper panel illustrates the commonly understood monetary policy framework to achieve price (and exchange rate) stability, including: inflation as the final target, macroeconomic analysis, and projections, assessment of monetary policy transmission and the monetary policy instruments available. The external dimension of the monetary policy framework not only covers assessments of exchange rates and BOP conditions in the macroeconomic analysis and projections, but also the impact of global spillovers and the foreign capital flow management required. Such a policy framework refers to FITF, which has been applied by numerous central banks, including Bank Indonesia. The integration of macroprudential policy into that policy framework is presented in the lower panel, covering: the complementary targets to support financial system stability, analysis, and projections of macrofinancial linkages from the financial system in the economy, assessments of procyclicality and systemic risk, as well as the macroprudential policy instruments available. In implementation, three salient aspects demand attention, namely: (1) how the dual targets of price stability and financial system stability can be integrated; (2) the mix of policy instruments to use; and (3) the effectiveness of the transmission mechanism. The description in the upper part explains how price stability and financial system stability can be integrated into the dual mandate of the central bank. In terms of the instruments, integration between monetary policy, macroprudential policy, and foreign capital flow management is required to ensure mutual complementariness when pursuing the dual mandate of the central bank. Accordingly, three fundamental policymaking frameworks underlie formulation of the central bank policy mix as follows. First, monetary policy is still directed toward achieving price stability, while also simultaneously paying more attention to (financial and property) asset prices in order to maintain financial system stability. Regarding the exchange rate, as elucidated in Chapter 13 and reiterated above, interest rate policymaking already takes into consideration the impact of the exchange rate on achieving the inflation target through an extension of the Taylor rule. In fact, many EMEs, including Indonesia, adhere to two instruments-two targets, namely the interest rates and foreign exchange market intervention to achieve the inflation target and exchange rate stability. Meanwhile, the interest rate policy response to other asset prices, particularly property prices, is achieved by leaning against asset price movements that are shown to deviate from fundamentals. The monetary policy response to asset prices, including exchange rates, stock prices, and bond yields, as well as property prices, is formulated to achieve price stability and support financial system stability. Second, macroprudential policy covers the regulation and supervision of financial services institutions from a macrofinancial perspective, with a focus on systemic risk to nurture financial system stability. As noted in Chapter 14 and reiterated above, macroprudential policy to control financial procyclicality, specifically credit booms and housing bubbles, is achieved by applying LTV, DTI, CCB as well as other instruments. Meanwhile, systemic risk is mitigated using macroprudential instruments in the form of interbank money market regulations and financial infrastructure, as well as enforcing liquidity and capital requirements on

470    Central Bank Policy financial institutions shown to have a high level of interconnectedness and intricate financial networks. The range of macroprudential policy instruments used to maintain financial system stability also reinforce the effectiveness of monetary policy transmission to achieve price stability. During an economic upswing, for example, interest rate policy effectiveness to control excessive credit growth could be strengthened through more stringent LTV or DTI enforcement for sectors less sensitive to interest rates, such as the property sector. Third, foreign capital flow management could further strengthen monetary policy effectiveness to achieve price stability, or macroprudential policy effectiveness to support financial system stability from global spillovers. As mentioned in Chapter 13, foreign capital flows exert influence on exchange rates, asset prices, liquidity, and risk-taking behavior at home. Similarly, external risks to domestic macroeconomic conditions can also originate from external debt, especially short-term external debt and that used to finance the domestic-oriented economy. The risks associated with foreign capital flow volatility and external debt, in general, heighten during periods of economic expansion and also go hand-inhand with financial procyclicality and systemic risk that emerge in the domestic financial system. Therefore, foreign capital flow management is crucial to support exchange rate stability and also represents an important element of BOP and suddenstop crisis prevention efforts, which can often culminate in a financial crisis. In addition to integration of the three policies mentioned, financial market deepening is another vital aspect of supporting the efficacy of the central bank’s policy mix. Mature and efficient financial markets enhance the effectiveness of monetary policy transmission through the interest rate channel, exchange rate channel, liquidity channel and bank credit channel, as well as in terms of risktaking behavior. Financial market deepening also supports intermediation of foreign capital flows for economic financing, as well as provides exchange rate flexibility and mitigates the risks to financial system stability that appear. Furthermore, financial product innovation and investment portfolio diversification through financial market deepening supports monetary and financial system stability. Nonetheless, more stringent prudential regulations and market conduct are required to accompany financial market development. As will be explained in subsequent sections, financial product innovation and portfolio diversification also increase interconnectedness and financial networks and, therefore, exacerbate systemic risk. Similar considerations are also at play concerning central bank jurisdiction over the payment system. Increased interconnectedness and instrument innovation in the payment system would enhance the reliability and efficiency of various financial and economic transactions, in addition to improving monetary policy effectiveness. Therefore, more stringent regulation and supervision of the payment system would be required to ensure such positive developments do not trigger systemic risk to financial system stability.

15.2.3. Central Bank Policy Mix Transmission Mechanism Effective transmission of the central bank policy mix is crucial to achieve price stability and maintain financial system stability. Chapter 5 explained how the

Central Bank Policy Mix     471 monetary policy transmission mechanism influences domestic demand and the attainment of price stability through the interest rate, exchange rate, asset price, liquidity, and money supply as well as credit channels along with risktaking behavior and market expectations. How does the central bank policy mix transmission mechanism support financial system stability? In this context, two policy dimensions are discernible in terms of sources of financial system instability, namely short-term cyclical factors and long-term structural factors. The former policy dimension can manifest as monetary policy, capital flow management and/or macroprudential policy, congruent with the current financial cycle, and can target specific sources of risk to financial system stability. Meanwhile, the latter policy dimension can be in the form of microprudential policy and/or macroprudential policy, directed toward strengthening financial system resilience to possible shocks through regulations pertaining to liquidity, credit risk exposure, market risk exposure, including currency risk, as well as capital requirements. Furthermore, stronger structural policy reduces the expense of applying cyclical policy. Policy to control cyclical risk from financial system stability is formulated in stages, as illustrated in Fig. 15.2 (IMF, 2015). The first phase determines whether FSS risk is excessive. Several financial variables are available, including rapidly rising asset prices during an economic boom. Other variables, such as credit growth and high leverage or corporate debt can also be regarded as potential sources of economic disruptions. In general, rapidly rising asset prices also drive excessive credit growth, thus precipitating asset price corrections and default risk. How far FSS risk is considered excessive will depend on the respective conditions in each jurisdiction, as well as the resilience of the corresponding financial system. Generally, such conditions occur when several of the financial variables simultaneously indicate excessive or rapid escalation. The second stage is to determine the appropriate policy response to mitigate excessive FSS risk. Typically, the first option focuses on macroprudential policy, in particular because such policy can be effectively directed toward certain sources of financial system fragilities (Blanchard, Dell’Ariccia, & Mauro, 2010;

Fig. 15.2:  Macroprudential and Monetary Policy: Maintaining FSS in the Cyclical Dimension. Source: Adapted from IMF (2015).

472    Central Bank Policy IMF, 2013b ). For example, FSS risk often surfaces due to rising house prices and excessive credit growth in the property sector, therefore, LTV and DTI can effectively target this sector specifically. Other policies can also be considered to reinforce macroprudential policy, such as tighter fiscal policy. Tighter monetary policy to achieve price stability would also help maintain financial system stability, for instance during an economic boom, when inflation generally accelerates and/or the current account deficit widens, which is often accompanied by high credit growth and external debt accumulation. When FSS risk builds up but inflation remains under control, monetary policy can be considered to lean against the wind in terms of maintaining financial system stability and macroeconomic stability if indications of economic overheating are present. Transmission between monetary policy and financial system stability is facilitated through two links, namely between the interest rate and financial variables, and between the financial variables and the probability of a shock to macroeconomic conditions (IMF, 2003). As presented in Table 15.1, a change to the monetary policy stance would influence financial system stability through the balance sheet channel due to tighter loan constraints and the probability of losses that could arise. For example, loose monetary policy removes guarantee constraints because asset prices increase and borrower credit worthiness improves, while the cost of external finance decreases and, therefore, credit conditions ease in general (Gertler & Gilchrist, 1994; Jiménez, Ongena, Peydró, & Saurina, 2009). Conversely, raising interest rates could threaten loan instalments and, therefore, exacerbate default risk and financial instability. A change of monetary policy stance could also affect risk-taking behavior at financial institutions (Jiménez, 2009; Merrouche & Nier, 2010). With asymmetric information, low interest rates provide an incentive for the banks to more aggressively extend loans by lowering lending standards. Furthermore, low interest rates could also shift risk to other economic players more willing to take risks in pursuit of higher gains (Landier, Sraer, & Thesmar, 2011). The adverse influence Table 15.1:  Effect of Monetary Policy of FSS. Predicted Impacts (Improves Stability) Sources of Financial Instability

Channel

Borrowing Constraint

Balance sheet (default) channel

Risky Behavior of Financial Institutions

Risk taking channel

Externalities through Aggregate Prices

Asset price channel

Risk shifting channel

Exchange rate channel

Source: Adapted from IMF (2003).

r

r

Central Bank Policy Mix     473 tends to be stronger if monetary policy is too accommodative during a boom period for a prolonged period, as more and more investors favor speculative and Ponzi actions rather than hedging activity, as foretold by Minsky (1982). On the other hand, aggressively easing monetary policy during a recession to stimulate the real economy and financial system tends to have a stronger impact due to the incentives thus available to financial institutions and economic players to take on more risk. Monetary policy can exacerbate asset price and exchange rate fluctuations and, therefore, the balance sheets and net worth of financial institutions and corporations. Low interest rates can raise asset prices as collateral, thus increasing leverage and asset price booms, which would compound the financial cycle (IMF, 2009). Oppositely, a tight monetary policy stance could constrain collateral due to a loss of corporate worth and/or net worth, which could prompt fire sales by the borrowers to meet repayments and by financial institutions to meet the requirement for liquidity. This phenomenon often occurs when economic conditions and the financial system are under distress. In addition, for EMEs and developing countries, higher interest rates can attract a deluge of foreign capital inflows and spur exchange rate appreciation. From there, the liquidity from non-resident capital tends to accelerate credit growth and foreign debt accumulation and, thus, heightens risk to financial system stability from currency exposure, maturity, and default (Hahm, Mishkin, Shin, & Shin, 2012; Merrouche & Nier, 2010). This phenomenon was found in eastern European countries prior to the crisis with European banks exploiting low rates in the United States. After the global financial crisis and before the US Taper Tantrum, the influx of foreign capital to EMEs such as Brazil, Turkey, South Africa, and Indonesia also drove credit growth, external debt accumulation, and rising house prices. Vigilance and an appropriate response are required in terms of monetary policy, macroprudential policy, and foreign capital flow management in order to pre-emptively analyze, project, and control the domestic and external economic risks as well as the risks contained in the financial system to achieve price stability and maintain financial system stability.

15.3. Structural Macroeconomic Modeling: Integrated Inflation Targeting Several studies have already developed models to support formulation of the central bank’s policy mix. The models include macrofinancial linkages and the influence of foreign capital flows in the macroeconomic analysis and projections. Several models were even developed prior to the global financial crisis under the auspices of the FITF, using a structural macroeconomic economic model approach based on the New-Keynesian Philips Curve (NKPC). In several EMEs, model development was initially motivated to accommodate the impact of foreign capital flows and exchange rates when formulating monetary policy to achieve price stability. As explained in Chapter 13, the model accommodated the external sector, including the influence of global risks, and was extended to determine the interest rate by including the real exchange rate in the Taylor rule.

474    Central Bank Policy After the global financial crisis of 2008/09, the financial sector was included in several models through measures of the credit gap and credit risk to support the integration of macroprudential policy in the formulation of monetary policy and foreign capital flow management in the FITF.

15.3.1. From Flexible to Integrated Inflation Targeting Flexible inflation targeting is already practiced by many central banks in EMEs to accommodate the effect of foreign capital flows and exchange rates in monetary policymaking to achieve price stability. The global crisis, however, forced central bankers to discuss how to best include financial system stability with the goal of price stability. Consequently, a consensus shift occurred away from Jackson Hole in 2009 to leaning monetary policy in order to support financial system stability and/or even to place the goal of financial system stability on equal standing to price stability, as presented in Table 15.2 (Smets, 2014). The first approach was based on a viewpoint that emerged in advanced countries prior to the crisis, as a modification of Jackson Hole Consensus. The approach asserts that monetary policy should remain focused on price stability, while FSS should be achieved using effective and credible macroprudential policy. FSS aspects are considered indirectly in terms of monetary policymaking if the assessment of macrofinancial linkages and systemic risk demonstrates an impact on price stability and real economic activities. This view is based on the conviction that the targets, instruments and transmission mechanisms between monetary policy and macroprudential policy can easily be separated. The considerations are that: interaction between the two policies is limited; monetary policy does not contribute significantly to macrofinancial imbalances that cause crises; and short-term interest rate policy is not an effective instrument to overcome those imbalances. Nonetheless, this approach recognizes the importance of exchanging information and assessments between monetary and macroprudential policymakers, irrespective of whether the two policies fall under different committees at the central bank or different authorities altogether (Adrian, Covitz, & Liang, 2013). The global financial crisis of 2008/09 changed the prevailing Jackson Hole Consensus view to the second or third approaches, especially in EMEs and several advanced countries. The central bank must take into consideration price stability and financial system stability when formulating monetary and macroprudential policy. In general, various studies have shown that: (1) monetary and macroprudential policy mutually reinforce one another when leaning against the financial cycle, driven by excessively optimistic expectations and control of risk-taking behavior; (2) there are potential coordination problems because of “push-me, pull-you” factors between the two policy instruments if separated; and (3) including macroprudential policy does not change but actually enriches the central bank’s monetary policy framework (Smets, 2014). In their study, Cecchetti and Kohler (2012) showed that interest rate policy and macroprudential policy (such as LTV or bank capital requirements) mutually interact and reinforce one another in terms of their influence on bank credit on the supply and demand

Granular and effective

Limited interaction and easily separable targets, instruments and so on

• Coordination? • Lender of last resort?

Svensson (2012, 2013); Collard et al. (2013)

Macroprudential Policy

Interaction

Issues

Model

Source: Adapted from Smets (2014).

• Framework largely unchanged; • Limited impact on credit and risk-taking behavior • Interest rate is not an instrument for macrofinancial imbalances

Monetary Policy

Modification of Jackson Hole Consensus

• Twin targets with the same considerations • Prevents macrofinancial imbalances during economic upswing and deteriorating balance sheet

Price Stability and FSS as Targets

Borio and White (2003); Cecchetti, Genberg, Lipsky, and Wadhwani (2000); Cecchetti, Genberg, and Wadhwani (2002)

• Coordination? • Encumber monetary policy?

Financial fragilities influence monetary policy transmission and price stability

Woodford (2011, 2012)

Problem of time inconsistency?

Price stability and FSS are inextricably linked

Does not fully control the financial Indistinguishable from monetary cycle; arbitration policy; coordinated

• FSS as the second target; policy horizon extended • Influences credit and risk-taking behavior • Gets in all of the cracks

Leaning against the Wind Approach

Table 15.2:  Integration of Price Stability and FSS: Three Modeling Approaches.

Central Bank Policy Mix     475

476    Central Bank Policy sides. Consequently, this study emphasizes coordination between the two policies to be more optimal and effective. Various studies have shown the need for central banks to lean against the wind in terms of asset prices in order to maintain financial system stability and achieve price stability (Borio & Lowe, 2002). Rajan (2005) found that price stability was insufficient to maintain financial system stability and recommended that central banks lean against financial imbalances through tighter monetary policy even when there is no risk to price stability. This view is consistent with the European Central Bank (ECB), as found by Trichet (2004) and Issing (2011). The financial cycle interacts closely with the economic cycle and, therefore, cannot be overcome using macroprudential policy alone. Monetary policy affects risk-taking behavior in the financial sector and, in contrast, financial sector fragilities influence the effectiveness of policy transmission and attainment of price stability. Therefore, financial system stability must be included as the second target of central bank policy. Woodford (2011, 2012) developed a model to analyze the implications of financial imbalances on monetary policy, revealing that an optimal response would take into consideration financial system stability, such as the marginal risk of financial crisis or credit gap, in addition to the inflation gap and output gap. This model recommends extending the policymaking horizon at the central bank to beyond 2 years in order to capture the financial imbalances. Fahr, Motto, Rostagno, Smets, and Tristani (2013) modeled macroeconomic simulations using financial frictions in the euro area and showed how monetary policy that leans against the credit gap could reduce the tradeoff between price stability and the output gap, thus improving macroeconomic performance in general. The third view recognizes that price and financial system stability are interconnected and difficult to separate. Fundamentally, monetary policy strives to stabilize the financial system and deepen the financial markets, hence strengthening the effectiveness of monetary policy transmission. An inefficient financial system clearly influences monetary policy and, therefore, coordination with macroprudential policy is a must. Brunnermeier and Sannikov (2012) developed a model to test the close correlation between monetary policy and financial system stability based on financial frictions in the monetary policy transmission mechanism. The model showed that financial frictions create close linkages between price stability and financial system stability. When an economy moderates or enters a recession, the optimal monetary policy response must identify and overcome balance sheet declines in the financial system that clog up the flow of funds to productive sectors in the economy. Less effective accommodative monetary policy in Europe and Japan, with interest rates approaching zero, is an example of such balance sheet issues. In contrast, during an economic boom, vigilance is crucial to identify macroeconomic and macrofinancial imbalances in order to institute monetary and macroprudential policy that prevents a crisis from unfolding. All three views clearly stress the importance of linkages between price stability and financial system stability but differ in terms of which monetary policy response would be best to reinforce macroprudential policy and vice versa. It all depends on conditions in each respective jurisdiction. First, if the interaction

Central Bank Policy Mix     477 between price stability and financial system stability increases, monetary and macroprudential policy coordination will be more effective under one roof, namely at a central bank with the dual mandate of price stability and financial system stability. Second, if the macroprudential policy instruments are not fully effective in controlling the financial cycle, monetary policy could reinforce such instruments in pursuance of the central bank mandate to support financial system stability. And third, if the expense of monetary policy to overcome a crisis becomes too large, there is a strong argument to give the central bank the dual mandate of price stability and financial system stability. In practice, the various monetary policy instruments could bolster financial system stability, for instance through the minimum reserve requirement and intervention on the foreign exchange market. Based on various studies and the previously mentioned ideas, Agénor and da Silva (2013) recommended that central banks in EMEs apply a new policy framework, known as the Integrated Inflation Targeting Framework (IITF). IITF is defined as a flexible IT regime in which the central bank’s mandate is explicitly extended to include a financial stability objective, the policy interest rate is set to respond directly to a (well-defined) measure of excessively rapid credit expansion and monetary and macroprudential policies are calibrated jointly to achieve macroeconomic (price) and financial stability. Table 15.3 presents a comparison between IITF and FITF before and after the global financial crisis of 2008/09. Calibration must be based on macroeconomic models that calculate macrofinancial linkages with the monetary and macroprudential policy response, which can strengthen the transmission mechanism to the financial system and economy. IITF application requires the central bank to understand the macroeconomic and macrofinancial interactions in the financial system and economy. Modeling is required, including analysis of the transmission mechanism. In general, commercial banks tend to dominate the financial systems in EMEs, therefore, analysis of credit and risk-taking behavior is also crucial, including the impact on the financial system and economy as well as the resilience to domestic shocks and global spillovers. The monetary and macroprudential policy response will depend on the macroeconomic and macrofinancial analyses as well as the transmission mechanisms. Departing from standard ITF and FITF, the policy response under the new framework is not merely determined by the inflation gap, output gap and exchange rate volatility, but also by the credit gap as a proxy of macrofinancial imbalances. Measurement of the credit gap is based on a comparison between actual credit growth and a reference of optimal credit growth to achieve sustainable economic growth. It is important for the central bank to analyze, predict and detect excessive credit expansion, or credit booms, as a key indicator of macrofinancial imbalances, which generally trigger financial system instability and can culminate in crisis. Furthermore, as discussed in Chapter 13, central banks in

FITF (Pre-crisis)

Free-floating exchange Floating exchange rate regime rate regime with FX intervention

Source: Adapted from Agénor and da Silva (2013).

Micro–Macroprudential Instruments (MiP and MaP)

External stability

Exchange rate (FX) volatility

IITF (Post-crisis)

Floating Exchange Rate regime with FX intervention and foreign capital flow management

Monetary and macroprudential instruments calibrated together (Mon+MaP) and Coordinated Micro–Macroprudential instruments (MiP+MaP)

Taylor-rule with I, O and Central bank rule with I, O, C, and FX factors FX factors

FITF (Post-crisis)

Microprudential instruments (MiP)

Inflation gap (I) Taylor-rule with I Output gap (O) and O

Macroeconomic Indicator

Financial stability Credit gap (C)

Price stability

Macroeconomic Target

Table 15.3:  Central Bank Framework: Integrated Inflation Targeting.

478    Central Bank Policy

Central Bank Policy Mix     479 EMEs must also consider exchange rate volatility through inclusion of the real exchange rate gap when determining the appropriate policy response. Central banks in EMEs must also take into consideration aspects of credibility and expectations that could be influenced through application of the new policy framework. Specifically, the considerations underlying the policy response formulation process must also be communicated clearly to the public, including stressing the importance of controlling excessive bank credit expansion to maintain financial system stability and sustainable economic growth, without neglecting price stability, which has remained an integral part of existing monetary policy frameworks. The issue of credibility is important because of possible inconsistencies between monetary policy and the dual mandate of price and financial system stability. Smets (2014), for instance, stated two challenges that must be confronted by the central bank. First, central bank policy cannot fully prevent a financial crisis, which could undermine the central bank’s reputation and, therefore, affect monetary policy independence and credibility. Second, when both objectives are merited with the same degree of importance, inconsistencies could appear where macroprudential policy does not lean adequately against credit booms and housing bubbles because of political pressures and the reliance on monetary policy to overcome some of the macrofinancial imbalances.

15.3.2. Model Structure and Analysis Framework Several studies have addressed how the central bank formulates its policy mix using structural macroeconomic modeling and expanding the Taylor-rule policy response. Accordingly, there are three dimensions that must be considered when modeling. First, the central bank policy response in terms of exchange rate stability to achieve price stability. This was discussed in Chapter 13, which recommended that the central bank utilize two instruments, namely the interest rate and foreign exchange market intervention, to achieve two objectives, namely the inflation target and exchange rate stability (Aizenman, Hutchison, & Noy, 2011; Blanchard, Adler, & Filho, 2015; Edwards, 2006; Mohanty & Klau, 2004; Ostry, Ghosh, & Chamon, 2012a). Second, the central bank policy response in terms of asset prices. To that end, a monetary policy response taken in accordance with the Taylor rule, extended to address asset price stability, such as stock prices and property prices, would increase macroeconomic stability (Cecchetti et al., 2000, 2002). Third, the central bank policy response in terms of financial system stability. In this regard, several studies recommend that the central bank should consider credit growth or financial system instability risk in the monetary policy response (Curdia & Woodford, 2009; Woodford, 2012). The central bank’s policy mix response has also been modeled according to a structural macroeconomic analysis framework, using the NKPC approach (Cecchetti et al., 2000; Woodford, 2012) evolved from the forward-looking structural model developed by Batini and Nelson (2000). This approach is also practiced at various central banks, including México (Penaloza, 2011) and Indonesia (Harmanta, Tarsidin, Bathaluddin, & Idham, 2012). The structure and analysis framework of the structural macroeconomic model are described as follows.

480    Central Bank Policy The IS equation was expanded to capture the impact of the exchange rate, asset prices, and credit gap as follows:

yt = β1 yt−1 + β2 yt +1 − β3 rt + β4 yt* + β5 zt + β6 crt + εty(1)

where y = output gap, r = real interest rate gap, y* = foreign demand gap, z = real exchange rate gap, and c = credit gap. Gap measurement is based on the potential level, long-term value or fundamental value of each respective variable.3 Meanwhile, the Philips-Curve equation is expanded to include the real exchange rate as follows:

πt = λ1πt−1 + (1− λ1 ) Et πt +1 + λ3 yt + λ4 zt + εtπ (2)

where π = inflation gap, measured from the target set, which is influenced by backward-looking and forward-looking expectations, output gap, and real exchange rate gap. In this context, the real exchange rate gap influences nominal exchange rate fluctuations as follows:

Et ∆St = σ∆St +1 + (1− σ ) zt (3)

where ΔS = change in nominal exchange rate. Macro-financial imbalances are modeled in the structural macroeconomic model framework through inclusion of the credit gap, in addition to the output gap and real exchange rate gap, which is also influenced by interest rate spread as well as two policy variables, namely LTV ratio and reserve requirement (RR) as follows:

crt = δ1crt−1 + (1− δ1 )( δ2 rt − δ3 spreadt + δ4 yt + δ5 LTVt − δ6 RRt ) + εtcr(4)

In this case, interest rate spread is influenced by loss loans or default risk in the following equation:

spreadt = ν1spreadt−1 + (1− ν1 )v2 deft + εtspread (5)

where default credit risk is determined by:

deft = θ1deft−1 + (1− θ1 )( θ2 spreadt + θ3 crt−1 − θ4 yt ) + εtdef (6)

Fundamentally, this equation shows that default credit risk will escalate as spread and the credit gap increase, and the output gap declines. 3

Output gap is measured according to potential output value, the real interest rate gap according to the long-term value or real rate of interest and the real exchange rate gap according to the fundamental value, and the credit gap is measured according to the optimal level of credit growth.

Central Bank Policy Mix     481 The central bank formulates its policy mix through monetary policy with interest rate instruments using the Taylor rule equation, expanded with the real exchange rate gap, as well as macroprudential policy instruments, including LTV and RR. The Taylor rule equation to determine the (nominal) monetary policy interest rate, expanded with the real exchange rate gap can be expressed as follows:

it = ρ1it−1 + (1− ρ1 )( ρ2 πt + ρ3 yt + ρ4 zt ) + εti (7)

While LTV and RR are set to control the credit gap as follows:

LTVt = µ1LTVt−1 + (1− µ1 )(−µ2 crt ) + εtLTV (8)



RRt = κ1RRt−1 + (1− κ1 )(κ2 crt ) + εtGWM (9)

It is important to note that macroprudential policy formulation per equations (15.8) and (15.9) must be supported by bank credit procyclicality analysis in order to determine the timing and magnitude required for pre-emptive and forwardlooking control of excessive credit growth.

15.3.3. Central Bank Policy Mix Formulation The aforementioned structural macroeconomic model can be used as a macroeconomic and macrofinancial projection and analysis framework as a solid basis to formulate the optimal mix of monetary policy, exchange rate stabilization and macroprudential policy by the central bank. In general, this approach is applied by central banks to bridge more complex DSGE modeling approaches. Several salient aspects must be stressed when formulating the policy mix as follows. First, to ensure a pre-emptive and forward-looking policy mix, the macroeconomic and macrofinancial projection horizon must be extended from the typical 2–3 years or more in order to: (1) detect indications of macrofinancial imbalances in the financial cycle that generally have a longer lag than the macroeconomic cycle; and (2) provide a sufficient lead time for interest rate and macroprudential policies to unwind the macroeconomic and macrofinancial imbalances. In general, the lag for inflation and economic growth to react to interest rate policy is usually six to eight quarters. Second, in-depth research and modeling are required to broaden understanding of macrofinancial linkages in order to strengthen policy mix formulation at the central bank based on the structural macroeconomic modeling framework mentioned previously. As elucidated in Chapter 14, financial procyclicality often precipitates macroeconomic and financial system instability risk that generally originates from: bank credit expansion, asset and housing bubbles, external debt accumulation, and sudden capital reversal. In the structural macroeconomic model above, the influence of macrofinancial linkages from bank credit expansion and foreign capital flows are shown by adding the variables: credit gap and

482    Central Bank Policy real exchange rate gap. Risks in the form of asset and housing bubbles as well as external debt accumulation can be detected through both variables. Nevertheless, a separate model is required to investigate the behavior of the financial cycle in terms of the four macrofinancial linkages mentioned in order to detect indications of risk in the form of macroeconomic, macroexternal and macrofinancial imbalances. This will facilitate the pre-emptive and forward-looking policy mix required. Moreover, research such as this is also crucial to determine the appropriate timing and magnitude required of the policy mix. Fig. 15.3 shows how the monetary and macroprudential policy cycle must be directed to control the financial cycle in line with the economic cycle, thus avoiding macroeconomic and macrofinancial imbalances. To that end, the policy mix must be tightened at the appropriate time, when there are indications and projections that the financial cycle could spur risks in the form of macroeconomic and macrofinancial imbalances during an economic upswing or boom period. The opposite is also true, namely that the policy mix must be loosened to avoid the adverse effects of financial procyclicality during an economic downswing or bust period, or when the bubbles burst. To that end, further research into the behavior of the financial and economic cycles, as described in Chapter 14, is critical. Such research, into the four aspects of financial procyclicality that oftentimes trigger macroeconomic and macrofinancial imbalances, should be conducted as an aggregate and by sector. Furthermore, research into corporate and household behavior, with a more micro approach, would also benefit the understanding of macroeconomic and macrofinancial imbalances. Third, consistency and coordination when formulating the central bank policy mix are crucial to ensure that monetary policy credibility, which has thus far remained strong, can be maintained (Angelini et al., 2011; Cecchetti & Kohler, 2012; Curdia & Woodford, 2010; Smets, 2014). Similar to the findings above, Smets (2014) emphasized the importance of monetary and macroprudential policy

Fig. 15.3:  Monetary and Macroprudential Policy Mix Cycle Maintains Macroeconomic and Financial System Stability in a Cyclical Dimension.

Central Bank Policy Mix     483 consistency, especially when divergence between the price stability and financial system stability targets arises, in order to maintain central bank credibility. In this regard, several models have been developed that include macrofinancial linkages in the formulation of the monetary and macroprudential policy mix. Angelini et al. (2011) modeled the FSS target in the form of a policy mix response to stabilize the credit-to-GDP ratio, like the credit gap in the structural macroeconomic model above. Curdia and Woodford (2010) favored credit spread when modeling the impact of financial frictions on public prosperity. Credit spread shows the financial frictions in bank intermediation due to the costs incurred by the banks when selecting credit demand and credit default because a portion of borrowers do not repay their loans. Meanwhile, Cecchetti and Kohler (2012) analyzed under which conditions monetary policy, through the interest rate, and macroprudential policy, through the CCB, would be mutually complementary or substitutes to achieve price stability and support maintained financial system stability. In general, their research demonstrated that tight coordination between monetary policy and macroprudential policy is crucial to optimally achieve both targets without reducing monetary policy credibility to attain price stability. In more detail, the analysis framework of Cecchetti and Kohler (2012) was based on a structural macroeconomic model with the inclusion of macroeconomic and macrofinancial linkages and, thus, could better capture the transmission mechanism of the monetary and macroprudential policy mix through the interest rate, exchange rate, asset price, bank credit, corporate balance sheet, and household worth channels. In this case, aggregate demand was expressed in the following equation:

y d = −α( ρ − π e ) − β (i − π e ) − δπ + η(10)

where yd = aggregate demand, i = short-term nominal interest rate, ρ = nominal lending rate, πe = inflation expectations, π = inflation, and η = random demand shocks. The short-term nominal interest rate (1) is determined by the central bank, while the lending rate (ρ) is determined by credit market equilibrium. For simplification purposes, the magnitude of credit is assumed to be determined by the amount of bank capital, which can be expressed as follows:

LS = −κ.k + τ .B (11)

where Ls = bank credit supply, k = capital requirements, and B = bank capital. Moreover, the real value of bank capital is assumed to rise as the real value of output increases as follows:

B = by(12) Meanwhile, credit demand depends on the real lending rate and real output as flows:



LD = −φ( ρ − π e ) + ω y (13)

484    Central Bank Policy On the other hand, aggregate supply shows output determined by the difference between inflation and inflation expectations pursuant to the Philips Curve as follows: y s = γ ( π − π e ) + ε(14)



where ys = aggregate output supply and e = random supply shock that does not correlate with the random demand shock in equation (15.10) above. Ultimately, the model shows a balance between the real sector and financial sector as follows:

yS = y d = y

and

LS = LD = L (15)

When analyzing the monetary and macroprudential policy mix, Cecchetti and Kohler (2012) also considered interest rate spread in addition to achieving the inflation target and considered economic growth in the objectives of the central bank as follows:

min.Wbauran = π 2 + λ. y 2 + ζ ( ρ − i )2 (16) (i ,k )

where (ρ − i) = spread between lending rate and the monetary policy rate, and ζ = FSS weight in the objective function of the central bank. In this regard, the central bank minimizes losses to public welfare by selecting an optimal policy mix between interest rate instruments on the monetary policy side and capital requirements on the macroprudential policy side. Through follow-up analysis, Cecchetti and Kohler (2012) compared the optimal solutions that had the least adverse impact on public welfare if monetary policy and macroprudential policy were independent, coordinated or partially coordinated. In general, the results showed that tight coordination between monetary and macroprudential policy provides the optimal solution. In other words, achievement of price stability and financial system stability requires close coordination between both policies to provide the optimal solution toward public betterment. In more detail, when no coordination exists between monetary and macroprudential policy, determining the policy rate to minimize the adverse impact on the public (15.16) can be expressed as follows:

 α( ω − bτ ) + φ(1− γδλ )  1 *  .ε + .η (17) iMPindependen = − 2   (1 ) βφ + γ λ β  

Meanwhile, the capital requirements determined in the optimal solution are as follows:

* kFSindependen =

( δε + γη )( ω − bτ ) (18) ( γ + δ )κ

In this regard, because monetary policy does not take into consideration the impact of changes to capital requirements, the impact on public welfare is more

Central Bank Policy Mix     485 significant. Increasing demand, for instance, tends to exaggerate the interest rate response because tightening the capital requirements by raising the CCB rate is not considered. The opposite is also true, a negative supply-side shock also tends to amplify the reduction to the policy rate because loosening the capital requirements is not considered. In fact, excessive loosening of both instruments would be a risk to maintained financial system stability. Similarly, the target of maintaining financial system stability would also be undermined if both policy instruments were independent or not coordinated. The analysis above demonstrates the importance of close coordination between monetary and macroprudential policy to improve public welfare. If both policies are coordinated, the optimal solution when determining the monetary policy rate can be expressed in the following equation:

  (1− γδλ ) 1 *  .ε + iMPkoordinasi .η (19) = −  ( α + β )(1+ γ 2λ )  (α + β )   While the optimal solution to determine the capital requirements is as follows:



 ( α + β )( ω − bτ ) + ϕ(1− γδλ )  ϕ *  .ε + kFSkoordinasi .η (20) = − 2   ( )(1 ) + α + β + γ λ κ α β )κ (  

The optimal solution precipitates the same social losses as the optimal monetary policy solution under the ITF as discussed in Chapters 6–9, namely:

 λε2  *  WBauran =  (1+ γ 2λ )   

* for i = iMPkoordinasi

* and k = kFSkoordinasi (21)

From the optimal solution, macroeconomic imbalances (inflation gap and output gap) and macrofinancial imbalances (interest rate spread) can be overcome through close coordination between determining the policy rate and determining the capital requirements. The interest rate will not be too tight (loose) if macroeconomic imbalances occur during an economic boom (bust) period. Likewise, the capital requirements will not be too tight (loose) if macrofinancial imbalances occur. In other words, coordination between the two policies would produce an optimal solution to achieve price stability and support financial system stability. A similar analysis can be conducted for partially coordinated monetary and macroprudential policy. In this regard, as with game theory, the monetary policy response can be determined after the macroprudential policy response to overcome macrofinancial imbalances. Oppositely, the macroprudential policy response can be determined after the monetary policy response to overcome macroeconomic imbalances. The two alternative policy responses do not produce the optimal solution, however, because interaction between the macroeconomic and macrofinancial imbalances is not overcome through an optimal policy mix. Visually, Figs. 15.4 and 15.5 present a comparison of the influence of monetary policy and macroprudential policy formulation, which is independent, coordinated, or partially coordinated, on public losses.

486    Central Bank Policy

Fig. 15.4:  Coordination and Losses after a Demand Shock. The two graphs show that the optimal solution of the monetary and macroprudential policy mix produce the lowest public loss compared to policies that are independent or partially coordinated. This is applicable to a demand-side shock, as shown in Fig. 15.4, as well as a supply-side shock as shown in Fig. 15.5. In fact, public losses from that policy mix are lower than if monetary policy is formulated independently. The two graphs also show that the solution from partially coordinated policies is lower than policies that are independent to a certain point but thereafter the impact tends to amplify. Such conditions are inextricably linked to the behavior of the financial cycle, which tends to exacerbate the economic cycle during an economic bust period, and vice versa.

15.4. DSGE Modeling with Macrofinancial Linkages The main criticism of the NKPC approach when modeling FITF is the assumed homogeneity of financial institutions and economic players, thereby leading to aggregate analyses. The global crisis demonstrated the importance of heterogeneity among financial institutions, especially banks, similar to liquidity, capital, and risk-taking behavior, as well as among economic players, similar to project feasibility, demand for external financing and risk-taking behavior. Heterogeneity

Central Bank Policy Mix     487

Fig. 15.5:  Coordination and Losses after a Supply Shock. among banks and economic players creates financial procyclicality and systemic risk when analyzing financial system stability. Similarly, a number of studies have developed macroeconomic models with macrofinancial and external sector linkages using the DSGE approach. Several central banks have also developed the DSGE approach to support formulation of the mix between monetary policy, macroprudential policy, and foreign capital flow management (Brázdik, Hlaváček, & Maršál, 2011; Roger & Vleck, 2012). In advanced countries, the DSGE approach is used for policy formulation at the ECB using the NAWM model (Christoffel, Coenen, & Warne, 2008) and CMR model (Christiano, Motto, & Rostagno, 2010); at the US Federal Reserve using the EDO model (Chung, Kiley, & Laforte, 2010); at the Reserve Bank of New Zealand using the KITT model (Lees, 2009); at the Bank of Canada using the ToTEM model (Murchison & Rennison, 2006); at the Bank of England using the BEQM model (Harrison et al., 2005); at the Bank of Japan (BOJ) using the JEM model (Fujiwara, Teranishi, & Hara, 2004); at the Sveriges Riksbank using the SIGMA model (Erceg, Guerrieri, & Gust, 2006) and RAMSES model (Adolfson, Laseen, Linde, & Villani, 2007); and at the Banca d’Italia (Angelini et al., 2011; Gerali, Neri, Sessa, & Signoretti, 2010). Meanwhile, in EMEs, DSGE development was found in Brazil (de Carvallo & de Castro, 2015; Ferreira &

488    Central Bank Policy Nakane, 2015); in India (Anand, Peiris, & Saxegaard, 2010; Banerjee & Basu; 2015); and Bank Indonesia (Harmanta, Purwanto, Rachmanto, & Oktiyanto, 2013).

15.4.1. Macrofinancial Modeling Approaches in DGSE There are three basic approaches commonly applied in studies that include financial frictions in DSGE models, namely financial accelerator, collateral constraints, and through explicit modeling of financial intermediation, as presented in Table 15.4 (Roger & Vleck, 2012). Two changes to New Keynesian economics are required to include financial frictions in DSGE models. The first is to change economic players from homogenous (representative) to heterogeneous. This provides a basis for the analysis of borrowing between economic players due to differences in terms of consumption, productivity, and financial constraints. Second is to introduce asymmetric information between economic players that creates financial frictions due to borrower selection and the costs incurred to assess business feasibility and loan monitoring, which manifests as the external financing premium. The financial accelerator approach departs from the assumption of asymmetric information regarding project feasibility, thus creating the external financing premium above the cost of internal capital incurred by a firm seeking a loan, congruent with the mindset of Bernanke, Gertler, and Gilchrist (1999). This is modeled using DSGE with heterogeneous economic players according to the value to net worth and, therefore, the external financing premium will correlate inversely with the net worth of the borrower and limit the size of the loan. During an economic upswing, a borrower has a higher net worth, thus improving business feasibility and reducing the cost of external funding. Conversely, during an economic downswing, the loss of net worth reduces business feasibility and raises the cost of borrowing. Therefore, the external financing premium causes countercyclical behavior, creates an amplification mechanism and accelerates the investment and real output response to the shocks that emerge. For example, higher output due to technological advancement is amplified by rising asset prices and corporate net worth. Thereafter, such improvements reduce the external financing premium and amplify the output and asset price gains. The collateral constraints approach includes financial frictions in the DSGE model through net worth, which becomes collateral when receiving a new loan per the model developed by Kiyotaki and Moore (1997). Departing from the financial accelerator approach, net worth of the borrower has a direct influence on the size of the loan received and an indirect impact through the external financing premium. This approach assumes that the legal enforcement of loan agreements in the event of default is somewhat weak, thus necessitating the banks to request physical collateral on the disbursed loans, including land, property, or capital. In the DSGE model, economic borrowers (impatient agents) have a lower discount factor than other economic agents, and prefer to borrow at current interest rates to finance consumption using existing assets as collateral. Congruent with the financial accelerator approach above, the investment and output amplification

Bank Intermediation

Amplification of asset price shocks. Moderate amplification of non-financial shocks. Strong amplification is very sensitive to model assumptions.

Borrower net worth fluctuations determine the cost of external financing.

Transmission Mechanism

Implications of Amplification of asset price financial frictions shocks. Moderate amplification of non-financial shocks. Strong amplification is very sensitive to model assumptions.

Source: Adapted from Roger and Vleck (2012).

Asset price fluctuations influence collateral value and, therefore, credit constraints.

External financing premium

Key Variables

Credit limits and spreads

Shock transmission is weak because of procyclical interest rate spreads, except if combined with rigid interest rates or another financial friction approach.

Interest rate spread determines size of loan.

Interest rate spreads

Advance production costs required or assume a fixed portion of the loan is used for investment.

Heterogeneous economic players: households as savers and corporations as borrowers.

Motivation to borrow and lend

Heterogeneous economic players in the form of different discount factors because of consumption preferences.

Cost of financial intermediation increases with volume of financial services because of loan selection and/or credit default.

No explicit financial intermediation. No explicit financial intermediation. Explicit financial intermediation through the banking sector.

Collateral Constraints

Sources and External financing premium sensitive Loans determined by collateral characteristics of to net worth of economic players due value of economic players due to financial frictions to ex post asymmetric information. weak loan agreements.

Financial Intermediation

Financial Accelerator

Table 15.4:  Three Approaches to Model Financial Frictions in DSGE.

Central Bank Policy Mix     489

490    Central Bank Policy mechanism of bank credit manifests through fluctuating asset prices as collateral. For example, lower asset prices would translate to a lower value of collateral and, therefore, the bank would reduce the credit ceiling offered to the borrower. Consequently, spending by the borrower would also subsequently decrease and this would exacerbate the decline of asset prices, investment, and output. Initially, DSGE modeling per the financial accelerator and collateral constraints approaches assumed that the borrower could access direct funding from the lender without an intermediary. Nonetheless, inclusion of the banking sector enriched the DSGE model analysis, despite the inherent complexities and difficulties. Furthermore, financial frictions could be enriched by impact analysis of the intermediation costs, balance sheet conditions, and risk-management behavior. Modeling typically differentiates wholesale banking from retail banking. In general, DSGE modeling focuses on: (1) endogenously determining the interest rate spread per the business cycle in line with the risk-taking behavior of economic players, be they bankers, corporations or households; (2) modeling bank capital channels in the financial and economic cycles, thus facilitating analysis of the transmission mechanism and impact of capital requirements on the banks’ balance sheets; and (3) modeling bank risk-management. Several academic and central bank studies combine two or more of the approaches depending on the objectives of the analysis and desired policymaking requirements. The KITT model at the Reserve Bank of New Zealand, for instance, applies the collateral constraints approach, where households borrow using the house as collateral and there are collateral constraints to using offshore loans (Lees, 2009). At the Bank of England, the DSGE model applies the financial frictions approach with an external financing premium that is enhanced by the banks incurring costs on the capital accumulated (Markovic, 2006). The Banca d’Italia applies the collateral constraints approach, with the cost of capital accumulation leading to interest rate spread and the banks also face capital requirements (Gerali et al., 2010). At the ECB, Angeloni and Faia (2009) modeled the risk of bank runs, the probability of which increased as bank capital decreased, while Christiano et al. (2010) used the financial accelerator approach, augmented by the banking sector, which provided liquidity services. Meanwhile, the Bank of Canada developed a multi-country DSGE model with international financial flows, where each country had a heterogeneous banking sector and credit faced capital requirements (Beaton, Lalonde, & Snudden, 2010).

15.4.2. Monetary Policy and Macroprudential Policy in DSGE Modeling Monetary and macroprudential policy analysis using DSGE modeling is not a simple process. A balance is required between the structural complexity of the DSGE models themselves and the importance of analyzing both policies. In this regard, modeling in macroprudential policy studies using DSGE models has been developed by Gerali et al. (2010). In that model, the economy was characterized by a number of enterprises, heterogeneous households, and monopolistic banks. Patient households (savers) placed their savings at the banks, while impatient

Central Bank Policy Mix     491 households (borrowers) and enterprises borrowed from the banks despite collateral constraints. Firms produced consumer goods and investment using labor from households and capital. Bank assets included corporate and household loans, while the liabilities were savings deposits and capital. Banks strived to maintain capital in pursuance of the specific capital requirements, vt, determined exogenously by the authorities, such as the Basel requirements. Box 15.1 explores the DSGE model developed by Gerali et al. (2010) in more detail.

Box 15.1:  DSGE Models with Financial Frictions DSGE model with financial frictions was developed by Gerali et al. (2010). They introduced two types of financial instrument in the banking system, namely savings (term deposits) and loans (credit). When seeking a loan from a bank, economic players face credit constraints in relation to the value of collateral held, which could be in the form of a house for individuals/ households or physical capital (net value of fixed assets) for entrepreneurs/ businesses. The heterogeneity of the economic players determines the sustainability of intermediation in the economy. The banking sector operates under monopolistic competition, meaning that the banking industry sets deposit and lending rates to maximise profits. In summary, this model is able to demonstrate that macro-financial linkages, together with the boom and bust economic and financial cycles, are exacerbated by collateral constraints when extending credit as well as differentiation in deposit and lending rates due to monopolistic competition in the banking industry. A. Households and Entrepreneurs/Businesses In the model setting of Gerali et al. (2010), there are two types of household, namely patient households as savers (P) and impatient households as borrowers (I), as well as one group of entrepreneurs as businesses (E). Patient households have a higher discount factor concerning their proclivity to consume in a future period (βP) than impatient households (βI), which is assumed to be the same as entrepreneurs (βE). Patient household, i, maximises the intertemporal utility expectation function as in Equation (15.1.1) depending on the current level of individual consumption, cp, and aggregate consumption in the previous period scaled by group-specific consumption habits, αP, household services, hp, and working hours, lp. Preferences are influenced by two shocks, consumption, εc, and household demand, εh. The real budget constraints in Equation (15.1.2) encompasses current consumption, accumulated housing services at real prices, qh, and the current position of term deposits, dt. Meanwhile, income is derived from salaries and wages, wPlP, interest receipts in the previous period, (1 − dt-1)dt-1/πt, and a lump-sum transfer in the form of

492    Central Bank Policy

dividend payments from a business or the bank (with the patient households as the owners) after deducting labour costs and insurance. ∞  l P (1+φ )  max E0 ∑ βPt (1− α P )εtc log( ctP,i − α P ctP−1 ) + εth log htP,i − t ,i 15.1.1 1 + φ  t=0 

s.t ctP,i + qth ∆htP,i + dt ,i = wtP ltP,i + (1+ rtd−1 )dt−1,i / πt + stP,i 15.1.2 Impatient household maximizes the intertemporal utility expectation function as in Equation (15.1.3) depending on consumption, cI, household services, hI, and working hours, 1I. The parameter, αI, measures the effect of habits specific to this group, while εc and εh are consumption and housing shocks. The real budget constraints in Eq. 15.1.4 consists of current consumption, accumulated housing services as well as interest and principal payments on loans in the previous period, bt–1, with lending rates for households, rbH, determined by the bank. In addition to salaries and lump-sum transfers, income is also derived from loan withdrawals, bI. ∞  l I (1+φ )  max E0 ∑ βIt (1− α I )εtc log( ctI,i − α I ctI−1 ) + εth log htI,i − t ,i 15.1.3 1 + φ  t=0 

I I I I I s.t ctI,i + qth ∆htI,i + (1 + rtbH −1 )bt −1,i / πt = wt lt ,i + bt ,i + st ,i

15.1.4

(1 + rtbH )btI,i ≤ mtI Et [ qth+1htI,i πt +1 ]15.1.5 Furthermore, impatient households face credit constraints as in Equation (15.1.5), which stated that the expected value of the property used as collateral in period t must be sufficient as a guarantee for the principal and interest payments, using mI as the LTV ratio for housing loans. From a macro perspective, the magnitude of the LTV ratio determines the value of the bank loan extended to a household based on the value of the house used as collateral. From a micro perspective (1 − mI) could be interpreted as the proportional cost of the guarantee held by the bank in the case of loan delinquency or default. The central bank is assumed to set the LTV ratio in order to influence the availability of credit in the economy and is, therefore, exogenous to the household. Entrepreneur maximizes the intertemporal utility expectation function as in Equation (15.1.6). The real budget constraint equates the total spending and total income, with consumption, cE, physical capital, kE, bank loans, bE, capacity utilisation, ut, labor input, lE, depreciation, δ, the price

Central Bank Policy Mix     493

of each physical capital unit measured from consumption, qk, real utility cost, ψ(μt)kt–1, and production function, yE. Moreover, similar to impatient households, the size of the loan a bank can guarantee to an entrepreneur is limited by the value of the collateral in the form of physical capital as in Equation (15.1.8), where mE = the LTV ratio for loans extended to the corporate sector. ∞

max E0 ∑ βEt log( ctE,i − α E ctE−1 )

15.1.6

t=0

E k E E s.t ctE,i + wtP ltE,i ,P + (1+ rtbE −1 )bt −1,i / πt + qt kt ,i + ψ( ut ,i ) kt −1,i



=

ytE,i xt

+ btE,i + qtk (1− δ ) ktE−1,i



15.1.7

(1+ rtbE )btE,i ≤ mtE Et [ qtk+1πt +1 (1− δ ) ktE,i ]15.1.8

Demand for deposits and loans were modeled in Gerali et al. (2010) using the Dixit–Stiglitz monopolistic competition framework to describe the market power of the banking industry. Hence, corporate and household deposit and credit contracts were differentiated slightly by each bank branch, j, with a constant elasticity of substitution (CES) of εd, εbH, and εbE, respectively, for term deposits, household loans and corporate loans. Therefore, not only could each bank offer different deposit and lending rates for each customer but also apply a narrower (wider) spread from the monetary policy rate during an economic upswing (downswing). The credit demand functions for impatient households and entrepreneurs in Eq. 15.1.9 as well as the deposit demand function in Eq. 15.1.10 illustrate that the size of the loan/deposit would depend on the overall volume and interest rates borne/ paid by bank, j, relative to the average in the economy. −etbH



 rtbH  ,j  b =  bH   rt  I t, j

−etbE

 rtbE  ,j  b =  bE   rt 

I t

E t, j

b

btE 15.1.9

−etd



dt , j

 rtd, j  =  d   rt 

dt15.1.10

B. Banking Industry Bank intermediation is modeled in Gerali et al. (2010) by using deposits as the sole financial asset for saving and credit as the only financial asset for borrowing, to reflect emerging market economies (EMEs) and developing

494    Central Bank Policy

countries which are still dominated by the banking industry. There are two important characteristics relating to financial frictions in the model that produce financial procyclicality: (1) the monopolistic competition; and (2) capital requirement constraint. Monopolistic competition allows the banks to differentiate deposit rates for each patient household and lending rates for each impatient household and entrepreneur, in addition to credit volume based on the value of collateral. In addition, industry dynamics also allow the banks to set deposit and lending rates in response to shocks and other cyclical factors in the economy. Consequently, this model facilitates impact analysis of financial frictions on bank behavior in terms of the interest rate channel and credit channel of the monetary policy transmission mechanism. The balance sheet constraint faced by the banks stated that each bank could fund its loan disbursements by mobilising funds and capital, which can only be accumulated from retained earnings. The capital requirements constraint stated that the minimum capital-to-asset ratio must equal the requirements set by the central bank. Hence, bank capital acts as a channel that influences reciprocity between the financial system and real economy. When macroeconomic conditions improve, bank profit and capital increase along with the banks’ ability to disburse loans. In contrast, during an economic recession, bank losses and decreasing capital tend to limit bank lending. The structure of the banking industry consists of two types of retail bank (loan and deposit) and one type of wholesale bank. Loan retail bank disburses loans to impatient households and businesses, while deposit retail bank mobilizes the funds deposited by patient households. Both types of retail bank operate under monopolistic competition and, thus, can differentiate deposit and lending rates with different spreads for each customer. Meanwhile, wholesales banks operate under perfect competition, maintaining the group’s capital position through financial intermediation and preventing violations of the capital requirements set by the central bank. Wholesale bank branch operates under perfect competition by collecting bank capital, Kb, and wholesale term deposit mobilisation, Dt, and disbursing wholesale credit, Bt, from and to retail bank branches. The objective is to choose the amount of loans and deposits to maximise the current value of all future cash flows with balance sheet constraints as in Equation (15.1.11), subject to the balance sheet constraint as in Equation (15.1.12). The bank faces an adjustment cost, AKb, if the capital-to-assets ratio, Kb/Bt, deviates from the capital requirements set by the central bank, vb. Bank capital is accumulated from retained earnings in each period as in Equation (15.1.13), where Πb = overall profit of the three bank branches, (1 − ωb) = the bank’s dividend policy and δb = funds used to manage bank capital. Dividend policy is assumed to be fixed and, therefore, not a choice for the bank. Wholesale lending and deposit rates, Rb and Rd, are determined exogenously.

Central Bank Policy Mix     495

(1 + Rtb )Bt − Bt +1 + Dt +1 − (1+ Rtd )Dt +    2 b Max E0 ∑ Λ P0,t     A K b b b Kb  t { Bt , Dt }  ∆Kt +1 −  15.1.11 − v  Kt t=0      2  Bt    ∞





s.t. Bt = Dt + Ktb15.1.12



Ktb = (1− δ b ) Ktb−1 + ω b Πtb−115.1.13

Solving the problem using the budget constraint and assuming banks have access to funds from the central bank’s lending facilities at the monetary policy interest rate so that the deposit rate will be the same as the interbank rate, Rd=rt, yields Equation (15.1.14) which links wholesale lending rates on the interbank money market to the monetary policy rate and capital-to-assets ratio. The equation determines bank loan availability, depending on capital conditions. Furthermore, rearranging the equation, the spread between the lending rate and monetary policy rate in Equation (15.1.15) correlates inversely with the capital-to-assets ratio, implying that if the bank’s capital is limited, the spread would be wider and, thus, the ability to lend would become more limited. 2

Kb  K b  R = rt − AKb  t − v b  t  15.1.14  B   B b t



t

t

2

Kb  K b  S ≡ R − rt = −AKb  t − v b  t  15.1.15  B   B W t



b t

t

t

Retail loan bank branch obtains a wholesale loan, Bt,j, at a wholesale lending rate, Rb, and subsequently differentiates the loan into credit for impatient households or businesses with different spreads. The objective is to maximize profit as in Eq. 15.1.16 subject to the demand for credit in Eq. 15.1.9, with Bt , j = bt , j = btI, j + btE, j . Retail loan bank incurs a cost when adjusting the lending rate per the parameters, AbH and AbE, which is proportional to the aggregate lending rate. ∞

I bE E b Max E0 ∑ Λ P0,t  rtbH , j bt , j + rt , j bt , j − Rt Bt , j bH bE

{ rt , j , rt , j }

t=0

−

AbH 2

2 2  15.1.16  bE  rtbH    , j −1 r bH b I − AbE  rt , j −1 r bE b E   r bH  t t   t t 2  rtbE  t−1, j   −1, j 

496    Central Bank Policy

For each respective lending rate offered to households and businesses (indexed using s = {I , E }), log-linearizing the first order condition produces Equation (15.1.17). The equation shows that the lending rate set by the retail bank incorporates the current value of the wholesale interest rate. Adjustments to changes in the wholesale interest rate are inversely proportional with the intensity of the adjustment cost (measured using Abs) and positively proportional with the level of competition in the bank credit sector (measured using the inverse of εbs). rˆt bs =

Abs βP Abs rˆt−bs1 + bs Et rˆt bs +1 + εbs −1+ (1+ βP ) Abs ε −1+ (1+ βP ) Abs εbs −1 Rˆtb εbs −1+ (1+ βP ) Abs

15.1.17

In the case of flexible interest rates, interest rate determination yields Equation (15.1.18), which basically shows that retail loan interest rates are set as a mark-up from the wholesale loan interest rate. Furthermore, analysing the spread between the lending rate and monetary policy rate as in Equation (15.1.19), The retail lending rate spread increases if the policy rate increases and is proportional to the wholesale spread, SW. Furthermore, a larger market share (namely a lower substitution elasticity, εbs) would determine a wider absolute spread, reflecting the bank’s ability to innovate in pursuit of interest rate margin.



 ebs  rtbs =  bs t  Rtb 15.1.18  et −1 Stbs ≡ rtbs − rt =

etbs 1 StW + bs rt 15.1.19 etbs −1 et −1

Retail deposit bank branch mobilises the deposits, dt,j, of patient households and subsequently deposits the funds at a wholesale bank branch at an interest rate, rt. The objective is to maximize profit as in Equation (15.1.20) subject to the demand for deposit in Equation (15.1.10), with Dt , j = dt , j. Retail deposit bank incurs a cost when adjusting the deposit rate per the parameter, Ad, and is proportional to the aggregate deposit rate.

 ∞ A  Max E0 ∑ Λ P0,t  rt Dt , j − rtd, j dt , j − d d {rt , j } 2  t=0 

2   rtd, j    rtd dt 15.1.20 1 −  r d    t−1, j  

Central Bank Policy Mix     497

Log-linearizing the first-order condition produces Equation 15.1.21. The equation shows that the deposit rate set by the retail bank incorporates the current value of the monetary policy rate. Adjustments to changes in the wholesale interest rate are inversely proportional with the intensity of the adjustment cost (measured using Ad) and positively proportional with the level of competition in terms of mobilising bank funds (measured using the inverse of εd). Ad βP Ad rˆt d = rˆt−d 1 + Et rˆt +d 1 + 1+ ε d + (1+ βP ) Ad 1 + ε d + (1 + βP ) Ad 1+ ε d rˆt 1 + ε d + (1 + βP ) Ad

15.1.21

In flexible interest rates, interest rate determination yields Equation (15.1.22), which basically shows that retail deposit interest rates are set as a mark-down from the monetary policy rate.  ed  rtd =  d t  rt 15.1.22  et −1



In general, the real profit of a bank is the total net revenue (intermediation margin minus the other costs) of the wholesale bank and retail bank branch. After omitting the inter-branch transactions, bank profitability is determined as in Equation (15.1.23), with AB the adjustment cost of changes in the deposit and lending rates. 2

 A Kb Π = rt b + rt b − rt dt − Kb  t − v b  Ktb − AtB 15.1.23  2  Bt b t

bH

I t

bE

E t

d

Model Estimation Estimation can be performed using a linearization model around the steady-state and proceed with Bayesian approach. Gerali et al. (2010) uses 12 observational variables: real consumption, real investment, real house prices, real deposits, real household and corporate credit, overnight rate, deposit rate, as well as the lending rates for households and businesses. For further clarification and estimation results, refer to Gerali et al. (2010).

When analyzing the interaction between monetary policy and macroprudential policy, Angelini et al. (2011) used two macroprudential instruments, namely the LTV ratio and the CCB, namely vt. In this context, banks accumulate capital naturally by acquiring retained earrings as follows:

K b,t = (1− δb ) K b,t−1 + Πb,t−1(22)

498    Central Bank Policy where Kb,t = bank capital, Π = retained earnings, and δ = capital depreciation. The model assumes that banks only accumulate capital from retained earnings. The lending rates offered to households (H) and enterprises (E) are determined by the first form equation of bank profit maximization as follows:  K  RtB ,i = µtB ,i Rt + φ  b ,t − νt  + AtB ,i i  =  H ,  E (23)   Lt   where μt = time-varying mark-up applied by banks due to their monopolistic nature, Rt = monetary policy rate, ϕ = the cost incurred by the bank if the ratio of capital to credit (L) is less than the capital requirement, vt, as determined by the central bank as the macroprudential authority, and At = cost of adjusting the bank interest rates due to rigidity. Equation (15.23) can be interpreted as a function of bank credit supply. If credit increases, the ratio of capital to assets decreases below vt, thus compelling the bank to raise lending rates, which, in turn, reduces demand for new loans. The time deposit rate, RD, as the interest rate for savers to choose between consumption and saving is determined as follows:



RtD = µtD Rt + AtB ,D(24)

where μt = time-varying mark-down below the monetary policy rate that can be determined by the banks due to their monopolistic nature, and At = cost of adjusting the term deposit rate. Equations (15.23) and (15.24) show that in this model, monetary, and macroprudential policy play independent roles. The monetary policy rate, Rt, has an immediate impact on lending and deposit rates, while the macroprudential instrument, vt, only affects lending rates. The independent formulation of both instruments creates increasingly significant misalignment between deposit rates and lending rates, which will ultimately have a larger impact on consumption, saving and investment. The second macroprudential instrument is the LTV ratio, which also has an impact on the economy. In this case, the optimal solution regarding the level of consumption/saving by saver households (s) and by borrower households (b) is determined using the Euler equation to maximize the utility function of each respective household type.

 1 (1+ R D )  1 s t  (25) = β E t  c s πt +1  cts t +1`



 1 (1+ R B ,H )  1 t  + (1+ RtB ,H )λt(26) = β b Et  b b  ct +1` πt +1  ct

where βi = is the discount factor with βs > βb, c = consumption, and μ = inflation. For savers (Equation 15.25), the optimal level of consumption depends on the current real deposit rate and future expectations. On the other hand, for

Central Bank Policy Mix     499 borrowers (Equation 15.26), the optimal level of consumption depends on the current real lending rate and future expectations, in addition to the Lagrange multiplier, λt, of the loan constraints as follows:

(1+ RtB ,H )btH ≤ m H Et [ qth+1ht πt +1 ](27)

where qt = house prices, ht = current stock of houses, bt = value of bank loan, and mt = LTV ratio for housing loans. Equations (15.25)–(15.27) show how macroprudential policy can influence the economy indirectly of monetary policy. The monetary policy rate, Rt, influences deposit and lending rates through Equations (15.23)–(15.24) and, thus, the choice to consume or invest by savers through Equation (15.25.) and the decision to consume or borrow by borrowers through Equation (15.26). By stipulating the LTV ratio, mt, macroprudential policy influences loan constraints, λt, and, hence, the level of consumption and borrowing by borrowers through Equation (15.26) but not the level of consumption and investment by savers. When collateral constraints tighten and λt increases, the ability of borrowers to cover household consumption and investment decreases and, therefore, the corresponding level of spending on both items also declines. It is important to note that the role of independent macroprudential policy is due to collateral and borrowing constraints. When such constraints are not present, monetary and macroprudential policies are mutually influential.

15.4.3. Central Bank Policy Mix Formulation Using the structural DSGE model above, Angelini et al. (2011) analyzed monetary and macroprudential policy formulation with the following considerations: (1) sources of economic shock, whether technological or demand-side; and (2) the formulation process of both policies, whether independent or coordinated into a mix. In harmony with the analysis of Cecchetti and Kohler (2012), using the structural DSGE model above, Anglenia et al. (2011) also showed that the coordinated formulation of a monetary and macroprudential policy mix provides the most optimal solution with a DSGE model. In this regard, the central bank formulates monetary policy by determining the policy rate according to the Taylor rule as follows:

Rt = (1− ρR )R * +(1− ρR )  ωπ ( πt − π*) + ω y ( yt − y*) + ρR Rt−1(28)

where ρR = monetary policy rate adjustment inertia, and ωi = weight of the central bank’s preferred inflation and economic growth targets. The monetary policy response is directed toward minimizing the loss function on public welfare as follows:

W MP = σπ2 + ky,MP σ y2 + kR σ∆2 R (29)

500    Central Bank Policy where σ2 = variance of inflation, output and changes to the monetary policy rate. A positive kR weight indicates a cost of adjusting the monetary policy rate, meaning that the central bank will avoid an excessive change to the policy stance in order to prevent an excessive impact on volatility. Meanwhile, macroprudential policy is formulated by setting the CCB rate or LTV ratio. Each instrument is analyzed for simplicity. In this regard, the capital requirements are determined as follows:

vt = (1− ρv )v * +(1− ρv )ωv X t + ρv vt−1(30)

where the parameter v* measures the capital requirement vt at a steady state. The level of v* can be interpreted as the capital requirements necessary to ensure the soundness of individual banks, with adjustments to the capital surcharge for systemically important banks (SIB), as discussed in Chapter 14. The CCB, vt, is adjusted with the dynamics of the macroeconomic variable Xt, which is measured by the level of output. Therefore, the CCB rate can be raised when output is projected to climb and, thus, macroprudential policy is directed toward controlling procyclical bank lending by requiring additional capital. The reverse is also true, namely that if a recession is expected, the CCB rate can be eased to stimulate bank lending and support stronger economic growth. With a similar mechanism, macroprudential policy using the LTV ratio can be determined as follows:

mt = (1− ρ m ) m * +(1− ρ m )ω m X t + ρ m mt−1(31)

where parameter m* measures the capital requirement mt at a steady state. Like the capital requirements, the level of m* can be interpreted as the LTV ratio required to ensure the soundness of individual banks, while the LTV ratio, mt, is adjusted over time with the dynamics of the macroeconomic variable Xt, which is measured with house prices or housing loans. Therefore, the LTV ratio can be raised (lowered) to control (stimulate) bank lending if rising (falling) house prices or housing loans are projected to accelerate during a period of economic overheating (recession). When formulating macroprudential policy, the central bank minimizes the loss function on public welfare as follows:

W FS = σl2/ y + ky, FS σ y2 + km σ∆2 m (32)

where σ2 = variance of the credit to output ratio, output and changes to macroprudential instruments. In the equation, macroprudential instruments are represented by the LTV ratio. The same analysis can be conducted for the CCB instrument. Equation (15.32) shows that the desired target of the central bank to maintain financial system stability is controlling bank credit expansion, indicated by variability in the credit to output ratio. A positive weight, km, indicates a cost

Central Bank Policy Mix     501 of adjusting the macroprudential instrument, meaning that the central bank will avoid an excessive change to the policy stance in order to prevent an excessive impact on volatility. Formulating a coordinated monetary and macroprudential policy mix is analyzed through joint calibration of both instruments to minimize the total loss function on public welfare as follows: W Mix = W MP +W FS = σπ2 + σl2/ y + ( ky ,MP + ky , FS )σ y2 + kR σ∆2 R + km σ∆2 m (33) Several observations can be found from the central bank’s objective function above. First, the target of the central bank policy mix is to achieve price stability (in this case controlling inflation within the target corridor) and financial system stability (in this case a stable credit to output ratio). Second, when formulating the policy mix, the central bank also nurtures sustainable economic growth toward the potential output. Third, the central bank must also consider the adjustment cost of changing policy stance to avoid excessive volatility. The optimal solution of the central bank policy mix produces a set of parameters (ωπ, ωy, ωm, ρR, ρm) to minimize the objective function (15.33) above. When the two policies are formulated independently, however, the monetary policy response is determined by minimizing the function (15.29), assuming the macroprudential policy instrument is congruent with Equations (15.30) and (15.31). And vice versa, the macroprudential policy response in Equation (15.32) is determined assuming monetary policy is consistent with Equation (15.28). There is no joint calibration between the two policy instruments. This lack of coordination can occur at a central bank if the committee to formulate monetary policy is separate from the committee to formulate macroprudential policy. Conditions will become even less coordinated, or more independent, if the responsibility for macroprudential policy is not under the auspices of the central bank. Independent policies cause economic players, households and firms, to be passive, meaning that their behavior and interactions are not considered when formulating policy and merely accept the policy response set by the central bank as a given. A comparison between the alternative policymaking approaches by Angelini et al. (2011) revealed several interesting findings. Under normal conditions, meaning that economic dynamics are influenced more by technology shocks and supply-side shocks, actively applying macroprudential policy instruments, such as LTV or the CCB, would not have a significant impact on macroeconomic stability. In fact, if both policies are formulated independently, the central bank could actually formulate conflicting policies and, therefore, trigger excessive interest rate and LTV/capital volatility without significantly strengthening macroeconomic stability (inflation and output) and financial system stability (credit to output variability). This kind of conflict demonstrates the mutual influence of both policies on interest rates and bank lending, but the impact on inflation, output, and credit to output can vary. The results are very different if the macroeconomic fluctuations are more attributable to financial or demand-side shocks. Under such conditions,

502    Central Bank Policy macroeconomic stability is strengthened significantly by macroprudential policy derived from monetary policy. Monetary and macroprudential policy coordination produces broader benefits in terms of macroeconomic stability (especially output) and financial system stability (credit to output variability). The LTV macroprudential instrument is significantly effective if the financial shocks occur in a specific sector, although this may be accompanied by slightly higher inflation and interest rate volatility because monetary policy supports macroprudential policy to stabilize credit and output. The benefits of stronger macroeconomic and financial system stability will not be realized if the two policies are formulated independently (not coordinated). The results explain why many countries apply macroprudential policy, particularly in the wake of the global financial crisis, namely due to economic shocks from the financial side (advanced countries) or demand side. It is important to note that macroprudential policy is not a substitute for monetary policy, nor can both policies be considered a panacea to overcome all problems. A mix of the two should be applied and formulated in a coordinated way in accordance with the sources of shocks in the economy, particularly financial shocks and shocks in specific sectors. Moreover, independent policy formulation under normal conditions may not safeguard macroeconomic or financial system stability, and could even exacerbate conditions. In reality, technology shocks and financial shocks can occur, which both influence the economy. During an economic upswing due to productivity factors, for instance, higher corporate net worth could lower the external financing premium or higher collateral asset prices could raise the bank credit ceiling. Consequently, technology shocks to increase output could be accompanied by financial shocks, thus exacerbating the amplification of the financial cycle by the economic cycle, as found by Bernanke et al. (1999), Kiyotaki and Moore (1997), and Minsky (1982). The analysis above shows that a monetary and macroprudential policy mix calibrated through close coordination could control the financial cycle to remain synchronized with the economic cycle and, thus, not spur macroeconomic and financial system instability risk. This is the essence and advantages of a central bank policy mix rather than traditional monetary policy.

15.5. Bank Indonesia Policy Mix In terms of monetary policy, Bank Indonesia applies the ITF introduced in the early 2000s, which has been officially implemented since the middle of 2005. The final goal of monetary policy is price stability (inflation). To that end, exchange rate stability is consistent with attainment of the inflation target and supports general macroeconomic stability, including management of the current account deficit and non-resident capital flows. The primary policy instrument is the interest rate, which is supported by exchange rate policy instruments, foreign capital flow management, statutory reserve requirements, and other prudential regulations. Although directed toward price stability, monetary policymaking pays due consideration to macroeconomic and financial system conditions in general, particularly economic growth, the balance of payments and monetary policy transmission effectiveness.

Central Bank Policy Mix     503 ITF-based monetary policy has facilitated a downward inflation trend within the 4±1% target corridor since 2015.The Government introduced energy reforms at the end of 2014 – the main driver of higher inflation – but inflation remained under control and continues to decline toward the medium-term target of 3±1%. Institutionally, ITF has reinforced monetary policy credibility at Bank Indonesia. Disciplined, regular and formal economic assessments underlying policy formulation at the monthly BI Board of Governors’ Meeting has placed Bank Indonesia at the forefront of macroeconomic conditions and the national economic outlook, including the necessary policy response. The regular economic evaluations are also very helpful to Bank Indonesia when coordinating with the government (central government and regional administrations) in terms of monetary and fiscal policy as well as structural reforms to underpin sustainable economic growth and macroeconomic stability. Furthermore, aggressive communication by Bank Indonesia has proven an effective instrument to anchor inflation expectations and relay the macroeconomic outlook. In general, the established monetary policy framework represents a solid foundation for Bank Indonesia to institute macroprudential policy and, thus, support financial system stability through a policy mix. Several domestic and external challenges remain, however, that must be addressed to ensure a successful policy mix. First, Indonesia’s economy is spread out across a vast archipelago and continues to rely on commodities, and hence, often faces inflation shocks and volatile food prices, along with balance of payments uncertainty. Mitigating the internal and external balance requires a monetary policy that is closely coordinated with the government’s fiscal policy and structural reforms. Second, the financial system remains dominated by the banking sector with an immature financial market and, therefore, the economic cycle is replete with financial cycle accelerators in the banking system. Mitigating procyclicality and systemic risk in the banking system through macroprudential (regulation and supervision) policy is critical not only to support financial system stability but also in terms of monetary policy effectiveness. And third, Indonesia’s economy is comparatively small with a very open capital account system, thus foreign capital flow management is a crucial element of ensuring that the benefits of non-resident capital reach the economy, while simultaneously avoiding economic instability. The three challenges must be considered in the Bank Indonesia policy mix, moreover because all three are interconnected and mutually reinforcing. During an economic upswing, for instance due to financial deregulation, a commodity price boom or propitious global economic conditions, expanding domestic demand accelerates credit growth, creates property bubbles, intensifies inflationary pressures, widens the current account deficit and increases government external debt. We have experienced boom and bust periods in the economic and financial cycles during mini crisis and large crisis episodes throughout Indonesia’s economic history. The economic bonanza after financial sector liberalization from 1983 to 1988 subsequently culminated in the Asian financial crisis of 1997/98, which raised the curtain on various systemic risks and macrofinancial imbalances that had accumulated in previous periods. Developments during the decade before and then after the global financial crisis of 2008/09 also revealed a

504    Central Bank Policy similar pattern, albeit at a comparatively lower level. The mini crisis of 2005, for instance, was distinguishable for rapid growth of domestic demand, bank credit and foreign capital inflows from the economic boom due to high international commodity prices. Similar conditions occurred again in 2010 to 2013 because global commodity prices stoked the current account deficit when they fell. Macroeconomic and macrofinancial imbalances such as these cannot be overcome though monetary policy, foreign capital flow management or even microprudential regulation and supervision. Thus, macroprudential policy is an additional instrument of critical importance, which must be integrated, along with monetary policy and foreign capital flow management, into the Bank Indonesia policy mix.

15.5.1. Policy Mix Targets and Instruments As explained in the previous section, two important aspects must be considered when including the FSS dimension and macroprudential policy into the central bank’s policy mix, namely: (1) the need to broaden price stability beyond inflation to also encompass (financial and property) asset prices; and (2) the necessary response to macrofinancial linkages in terms of financial system stability, specifically in relation to procyclicality and systemic risk accumulation in the financial system. Concerning the first aspect, in addition to CPI inflation, Bank Indonesia also focuses on assessments of exchange rates, bond yields and stock prices, as well as property prices. Bank Indonesia considers the exchange rate for macroeconomic analysis and projections as well as the required policy response. Consistent with ITF, the interest rate is the primary instrument used to achieve the inflation target. Nonetheless, Bank Indonesia is not averse to intervention on the foreign exchange market in the case of exchange rate volatility or misalignment from the currency’s fundamental value. Regarding other asset prices, Bank Indonesia uses separate models to enhance the macroeconomic and macrofinancial assessments and projections, as well as monetary and/or macroprudential policy instruments considered more appropriate to tackle the risks faced. Referring to the second aspect, to deepen understanding of macrofinancial linkages, Bank Indonesia has expanded its macroeconomic projection model to include the external sector, incorporating external default risk as a proxy of sudden-stop risk, as well as the banking sector, containing the credit gap as a measure of procyclicality (Harmanta et al., 2012, 2013). The model presents various policy scenarios using interest rates and reserve requirements on the monetary policy side and/or LTV on the macroprudential policy side. The model is forward-looking with a horizon of 2 years, therefore, the projections provide valuable input when determining the timing and magnitude of the appropriate policy response to lean against price stability risk, capital reversal risk as well as FSS procyclicality and systemic risk using monetary instruments, foreign capital flow management or macroprudential policy instruments or, indeed, a combination of the three. To hone the understanding of procyclical behaviors and other macrofinancial linkages, particularly during credit booms and property bubbles, models and studies have been developed to assess cyclical behaviors and systemic risk

Central Bank Policy Mix     505 accumulation, as an aggregate and by sector (Alamsyah, Adamanti, Yumanita, & Astuti, 2014; Harun, Taruna, Nattan, & Surjaningsih, 2014). From the perspective of financial system stability, several approaches and technical systemic risk assessments have been developed as recommended in the literature (Bisias, 2012) or as practiced by central banks (such as by the European Banking Authority (EBA), 2015). The focus is on systemic risk assessment (not individual soundness) of SIB, namely banks that, due to the size of assets, capital and liabilities; scope of network or transaction complexity of banking services; as well as interconnectedness with other financial sectors, could trigger default in some or all other banks or the financial sector as a whole, operationally or financially, if the bank in question were to experience disruptions or default. (PPKSK Act, 2016) Interconnectedness in the interbank money market and payment system is also evaluated. Credit risk is analyzed using several methods, including the “probability of default” and “transition matrix.” The banking system is also regularly stress tested to analyze resilience to macroeconomic shocks through an integrated and/ or balanced approach based on the results of risk surveys conducted. Corporate sector and household risks are also assessed in terms of financial performance and the impact on banking system performance. To assess macrofinancial linkages between the financial system, from a procyclicality and systemic risk standpoint, balance sheet statistics were developed that highlight (asset and liability) linkages between economic players and the financial system, public and private, nationally, and domestically. Based on a holistic assessment of the macroeconomic analysis and projections, macrofinancial linkages as well as time series and cross-section dimensions of systemic risk, Bank Indonesia’s policy mix is formulated using the following four instruments (Warjiyo, 2014a, 2015b). First, like ITF, interest rate policy is directed to ensure forward-looking inflation projections are in the 4±1% target corridor for 2016 and 2017. Second, exchange rate policy is directed toward maintaining exchange rate stability on the market in line with the currency’s fundamental value to remain consistent with achieving the inflation target and to guard against excessive volatility that could trigger financial system and macroeconomic instability. Third, foreign capital flow management is used to support exchange rate policy as well as macroeconomic and financial system stability, especially during an influx of foreign capital or a sudden capital reversal. Fourth, macroprudential policy is directed toward supporting financial system stability and the effectiveness of monetary policy transmission. Bank Indonesia also maintains close policy coordination with the central government and regional administrations in terms of macroeconomic management. A coordination mechanism also exists between Bank Indonesia, the Financial Services Authority (OJK) and Deposit Insurance Corporation (LPS) for financial system stability matters. Clear communication is crucial for the success of the policy mix.

506    Central Bank Policy How does the policy mix respond to the various risks associated with attaining the inflation target and supporting financial system stability in the future? This issue has been a hot topic of debate due to possible conflicts between monetary policy to achieve price stability and macroprudential policy to support financial system stability. The policy mix is also said to violate the Tinbergen rule, namely that there should be at least the same number of instruments as there are targets. Nevertheless, recent understanding is that in many cases the two policy instruments are mutually complementary to achieve the two targets (Yellen, 2014). Table 15.5 presents four cases of price stability risk and FSS risk that could emerge from the assessment of macroeconomic and macrofinancial projections over a 2-year horizon as well as the corresponding monetary and macroprudential policy mix stances available. In the first quadrant, where low price stability and FSS risks are projected, the monetary policy and macroprudential policy stance is neutral or loose. Conversely, in the fourth quadrant, where high price stability and FSS risks are projected, a tight monetary and macroprudential policy stance should have been adopted. Potential conflicts can occur in the second and third quadrants. In the second quadrant, where price stability risk is projected low but FSS risk is projected high, a tight macroprudential policy stance is clearly required. Under such conditions, monetary policy can support the effectiveness of macroprudential policy by leaning against the projected future FSS risk. This case is similar to conditions in the United States prior to the global crisis, which has been debated by Taylor (2010) and Bernanke (2010) as explained above. In the third quadrant, where price stability risk is projected high but FSS risk is projected low, a tight monetary policy stance is required. Under such conditions, macroprudential policy Table 15.5:  Four Cases of Price Stability and Financial System Stability. Projected Price Stability Risk Low Projected Financial HIGH System Stability (FSS) Risk

LOW

High

Quadrant II Quadrant IV • NEUTRAL/ • TIGHT monetary LEANING policy monetary policy • TIGHT • TIGHT macroprudential macroprudential policy policy Quadrant I Quadrant III • NEUTRAL/ • TIGHT monetary LOOSE monetary policy policy • NEUTRAL/ • NEUTRAL/ LEANING LOOSE macroprudential macroprudential policy policy

Central Bank Policy Mix     507 can support the effectiveness of monetary policy by leaning against the projected future price stability risk. How far and the macroprudential policy instruments applied depend on the factors causing price stability risk. A reasonable choice would be macroprudential policy that can strengthen monetary policy transmission to achieve price stability. For example, when price stability risk stems from strong domestic demand, driven by rapid bank credit growth in the property sector, adjusting the LTV ratio would be the most appropriate policy action. The factual issues that occur in the real world, however, are not always as simple as described in Table 15.5. Nonetheless, the four cases presented in the table can be referred to as guiding principles that help assess the existing problems and select a policy mix that addresses the potential conflicts that arise in the dual mandate of price stability and maintaining financial system stability. It is important to reiterate that the choice of monetary and macroprudential instruments depends on an assessment of the factors causing the projected high risks to price stability and financial system stability. This kind of approach is also used to address potential conflicts between price stability and autonomous monetary policy, exchange rate stability and free foreign capital flows, namely the policy trilemma that plagues an open economy and is regularly explained in international finance (Obstfeld, 2015). The following sections will explore how the policy mix of Bank Indonesia is applied to reinforce monetary stability and financial system stability during periods of economic turbulence and in terms of international finance since the global financial crisis of 2008/09.

15.5.2. Interest Rate and Exchange Rate Policies In pursuance of the monetary policy strategy based on the ITF, interest rate policymaking is directed toward ensuring inflation projections remain within the target corridor. The issue, therefore, is how to address exchange rate fluctuations on the market that can edge inflation projections outside of the target range. Such conditions have been faced by EMEs when confronting highly volatile foreign capital flows since the global financial crisis of 2008/09. Differing from advanced countries, exchange rate stability is crucial for EMEs due to various factors, including underdeveloped financial markets, the monumental impact on banking conditions and financial system stability, as well as rigidity in the economy. Under such conditions, targeting the exchange rate as an element of achieving the inflation target will strengthen monetary policy credibility based on ITF (Ostry et al., 2012b). Specifically, targeting the exchange rate can help mitigate the undesirable impact of foreign capital flows on achieving the inflation target, either directly through exchange rate pass-through or indirectly through domestic demand. Many EMEs already consider the exchange rate when formulating interest rate policy using the Taylor rule (Aizenman et al., 2011; Mohanty & Klau, 2004). Foreign exchange intervention represents another option. If foreign capital flows are creating a misalignment between the exchange rate and its fundamental value and inflation has moved outside of the target range, an interest rate response combined with foreign exchange intervention would be more effective and, thus, reinforce monetary policy credibility.

508    Central Bank Policy Bank Indonesia introduced such an approach in 2010, which has proven superior to the standard ITF application that is merely based on interest rate instrument. Three episodes since the global financial crisis of 2008/09 have backed up such claims, including the period from 2010 until the Fed’s Taper Tantrum in May 2013, the period after the Taper Tantrum until the middle of 2015 and then the period thereafter. During the first period, Indonesia enjoyed the benefits of global spillovers, in particular high commodity prices and an influx of foreign capital flows (Warjiyo, 2013b). Economic growth peaked at 6.5% in 2011 and remained relatively robust at 6.3% in 2012. Furthermore, inflation was successfully brought down to its historically lowest level of 3.8% in 2011, even dipping below the target corridor of 5±1%, but accelerating thereafter to 4.3% in 2012. Indonesia received a deluge of foreign capital flows during that period, driven by excess global liquidity seeking higher yields and a promising domestic economic outlook. The exchange rate appreciated due to the foreign capital flows, coupled with a current account surplus due to high commodity prices. The challenge was how to manage the foreign capital flows in order to prevent and mitigate systemic risk to financial system stability, similar to conditions when bank credit growth soared to 23% per annum from 2010 to 2012. Such conditions mirrored the second quadrant, where price stability risk was low but FSS risk was relatively high, as discussed above. Consistent with ITF, Bank Indonesia lowered its policy rate by 75bps from 6.5% in 2010 to 5.75% in 2012. Further reductions would have been inconsistent with ITF because inflation had already hit its lowest level in history, while being ineffectual in terms of overcoming the torrent of foreign capital inflows that were driven more by push factors than pull factors (Indawan, Fitriani, S., Permata, & Karlina, 2013). In addition, further reductions to the policy rate would have also been inconsistent with the goal of supporting financial system stability due to excessive credit growth. Therefore, Bank Indonesia intervened (purchased) on the foreign exchange market to stem the flow of foreign capital and strengthen the exchange rate. In addition, to more effectively sterilize the impact on liquidity expansion in the domestic banking industry, the statutory reserve requirement was raised from 5% to 8% in November 2011. The position of official reserve assets increased significantly from just USD 66.2 billion at the beginning of 2010 to peak at USD 112.8 billion in 2012, which provided a larger buffer to overcome the foreign capital outflows after the Fed’s Taper Tantrum in the middle of 2013. The situation reversed in the second period. Foreign capital flowed out of Indonesia with rapidity after the Federal Reserve announced plans to normalize monetary policy, which stoked concerns and rattled the markets from May to August 2013. The sudden and large capital reversal from government bonds and stocks triggered herding behavior that exacerbated the risks to monetary and financial system stability (Warjiyo, 2014b). The problems were compounded by a current account deficit that swelled to 4.4% of GDP as well as declining export performance as commodity prices tumbled, while imports remained high on the back of strong domestic demand. Inflation spiked to 8.4% in 2013, as the government raised fuel prices in July, and was subsequently recorded at 8.3% in 2014 as the government reformed its energy policy and scrapped fuel subsidies, while

Central Bank Policy Mix     509 limiting diesel subsidies in October. In terms of financial system stability, bank credit growth remained high at 21.4% in 2013. Such conditions reflect the fourth quadrant, where the risks to price stability and FSS are high, as described above. Bank Indonesia immediately responded to stabilize the situation by raising the policy rate and tightening macroprudential policy. Bank Indonesia was one of the first authorities to raise its policy rate after the Fed’s Taper Tantrum in May 2013. Bank Indonesia first raised its policy rate in June 2013, followed by further hikes in the following months, totaling 175 bps to 7.50%, over a 6-month period through to November 2013. The main goal of aggressively raising the policy rate was to pre-empt intense inflationary pressures when the government hiked fuel prices in July 2013. Furthermore, the move was part of coordinated macroeconomic and fiscal policy to reduce the excessive current account deficit. The timing of the increase was critical because of the need to simultaneously control the foreign capital reversal after the Fed’s Taper Tantrum. The aggressive interest rate response provided a clear signal to the markets, thereby reinforcing monetary policy credibility. Bank Indonesia also intervened (sold) on the foreign exchange market to overcome the capital reversal and stabilize the exchange rate after the Taper Tantrum. The exchange rate stabilized thereafter in September 2013. Consequently, the position of foreign exchange reserves decreased sharply to USD 92 billion in September 2013 before rebounding to USD 99 billion at the end of the year. Foreign exchange intervention was also accompanied by the purchase of government bonds on the secondary market, a tactic known as dual intervention (Warjiyo, 2013c). Fundamentally, the tactic makes sterilization more effective because the purchase of bonds on the secondary market prevents liquidity from evaporating as it is absorbed due to the intervention (selling) of foreign exchange. Dual intervention, as a tactic, also strengthens the effectiveness of foreign exchange intervention in terms of stabilizing the exchange rate. Through this tactic, Bank Indonesia provided a clear signal that it would not hesitate to supply foreign exchange and was prepared to buy the bonds sold by non-resident investors, thus preventing herding behavior and contagion from the capital reversal. Moreover, dual intervention to safeguard monetary stability is more consistent with the need to maintain financial system stability. By stabilizing the foreign exchange market and government bond market, dual intervention helped stabilize the financial markets in general. The aggressive and timely monetary policy response bore fruit and reinforced credibility. Market confidence quickly recovered and foreign capital inflows returned at the end of 2013 and throughout 2014. Furthermore, macroeconomic and financial system stability were maintained. In fact, inflation decelerated from 8.3% after the energy reforms in 2014 to 3.3% at the end of 2015, while the current account deficit reduced from 3.3% to 2.0% of GDP over the same period. Such conditions are like those found in the first quadrant of the table, where the risks to price stability and financial system stability are low. Nevertheless, economic growth slumped from 5.2% in 2014 to 4.8% in 2015 and credit growth was squeezed to around 10%. Therefore, with stability maintained and clearer communication from the Fed that the normalization process would be gradual, Bank

510    Central Bank Policy Indonesia eased its monetary policy stance to reinforce looser macroprudential policy taken the year earlier. Bank Indonesia reduced its policy rate three times in the first 3 months of 2016, totaling 75 bps, to 6.75%. The reserve requirement was also lowered by 50 bps in November 2015 and again by 100 bps to 6.5% in February 2016. Bank Indonesia was confident that easing monetary and macroprudential policy would strengthen the fiscal stimuli introduced to catalyze economic growth with inflation controlled in the target range for 2016 and 2017 at 4±1%. Through coordinated macroeconomic and FSS policies, combined with accelerating the government’s structural reforms, economic growth is expected to improve on that posted in 2015, and then continue to accelerate in 2016 and 2017.

15.5.3. Foreign Capital Flow Management Foreign capital flow management in Indonesia aims to complement, not substitute, sound macroeconomic policy. The best defense to reap the benefits and mitigate the adverse effects of foreign capital flows is to strengthen economic fundamentals, institute prudent FSS, and macroeconomic policies as well as accelerate structural reforms in the real sector. Congruently, foreign capital flow management in Indonesia is based on three guiding principles. First, the goal is to mitigate the adverse impact of foreign capital flows in terms of exchange rate, monetary, financial system and macroeconomic instability in general. Second, selective foreign capital flow management is applied to specific types of foreign capital, namely short-term and speculative, while long-term and productive foreign capital is promoted to finance the national economy. And third, foreign capital flow management instruments and regulations remain consistent with the open capital account regime. As far as possible, the regulations do not differentiate between residents and non-residents. Enforcement is temporary when required, namely by tightening regulations when foreign capital flows are too large and vice versa. The following section outlines several foreign capital flow management regulations that have been implemented. When foreign capital flows inundated the domestic financial system after the global financial crisis through to the Taper Tantrum, Bank Indonesia managed foreign capital flows in the form of a 6-month holding period for Bank Indonesia Certificates (SBI) and capped short-term offshore loans at 30% of capital. This meant that investors had to wait 6 months before releasing their SBI, thereby limiting short-term foreign capital flows and encouraging longer terms. Similarly, restrictions on short-term foreign loans prevented the banks from using short-term funds to extend long-term loans, while simultaneously mitigating currency risk. During the subsequent period, as foreign capital inflows subsided and outflows were even recorded in the wake of the Fed’s Taper Tantrum in 2013, foreign capital flow management was eased by relaxing the holding period to just 1 month for SBI, while a number of transactions were excluded from the restrictions on short-term foreign loans by banks. The measures helped to manage short-term and speculative foreign capital flows and, hence, were consistent with and supported the achievement of price (and exchange rate) stability as well as financial system stability.

Central Bank Policy Mix     511 Bank Indonesia also introduced regulations at the end of 2014 to strengthen risk management for non-bank corporate external debt. In Indonesia, government external debt has been restricted pursuant to prevailing regulations that limit the fiscal deficit to 3% of GDP. For the banking industry, in addition to the restrictions on short-term offshore loans mentioned above, each foreign loan withdrawn by a bank would have to be preapproved by Bank Indonesia in order to remain consistent with the goals of maintaining macroeconomic and financial system stability. Applicable to non-banks, the new regulations promulgated by Bank Indonesia at the end of 2014 strengthened risk management for external debt by requiring: (1) a minimum currency hedging ratio of 25% of net foreign currency liabilities maturing in the upcoming three and 6 months; (2) a minimum liquidity ratio of 50% (after calculating foreign currency assets in the hedging ratio) of net foreign currency liabilities maturing in the upcoming three and 6 months; and (3) a minimum credit rating of one level below investment grade. Monitoring of more than 2,000 firms that submitted financial statements to Bank Indonesia showed that 88% had met the requirements. In addition, the new regulations also encouraged corporations to hedge against currency risk, evidenced by the rapid increase of swap and forward transactions on the foreign exchange market.

15.5.4. Macroprudential Policy As described above, the application of macroprudential policy in Indonesia can and must strengthen the effectiveness of monetary policy to achieve price (and exchange rate) stability as well as financial system stability. Consequently, monetary policy transmission can be more effective, especially through the banking system, and financial system stability can also be better maintained. If monetary and macroprudential policies were not mutually reinforcing, the interest rate response on the monetary policy side to achieve price (and exchange rate) stability and support financial system stability would have to be notably larger. For example, to control excessive bank credit expansion, a much larger hike to interest rates would be required if not accompanied by tighter macroprudential policy. The impact would also not only trigger economic losses but also financial system losses. At Bank Indonesia, in addition to expanding macroeconomic analyses and projections by including macrofinancial linkages when formulating the policy mix, several approaches and models have also been developed to assess the optimal level of bank credit growth (Utari, Arimurti, & Kurniati, 2012). Models such as these are applied as an aggregate and by loan type (consumer loans, working capital loans, and investment loans) as well as by economic sector. Comparisons can be drawn between actual credit growth and the optimal level of credit growth, which reveal whether bank credit growth as an aggregate, by loan type and by economic sector, is excessive. Such assessments are critical to determine when and which macroprudential policy instruments should be applied to reinforce monetary policy and foreign capital flow management to achieve price (and exchange rate) stability as well as support financial system stability.

512    Central Bank Policy Bank Indonesia followed such an approach by enforcing an LTV ratio averaging 70% for the automotive and property sectors in June 2012 (Warjiyo, 2015a). As explained above, while inflation was kept under control, FSS risk began to build on strong credit growth and an influx of foreign capital flows. Raising the policy rate was not an option because the move would attract even more foreign capital and drive credit growth further. Therefore, to control foreign capital inflows, Bank Indonesia intervened (purchased foreign exchange), raised the statutory reserve requirement and enforced several foreign capital flow management regulations. Macroprudential regulations, namely the LTV ratio, strengthened the Bank Indonesia policy mix. Furthermore, reinforcing policy to maintain macroeconomic and financial system stability in the wake of the Fed’s Taper Tantrum, Bank Indonesia tightened the LTV ratio in August 2013 on purchases of a second and additional property, applicable to specific house and apartment types. This policy measure was part of monetary and fiscal policy tightening to control high inflation and the current account deficit. Tighter macroprudential policy enhanced the effectiveness of raising the policy rate, which obviated the need for further interest rate hikes. Furthermore, the policy move also paid due consideration to lending rate rigidity in response to changes in the monetary policy rate with a wide spread between deposit and lending rates. The tightening of macroprudential policy was also accompanied by tighter supervisory actions imposed on certain banks, primarily those with excessively high property credit growth. Thus far, experience has shown that macroprudential regulation and supervisory actions have strengthened the effectiveness of monetary policy transmission and supported financial system stability (Purnawan & Nasir, 2015; Wimanda, Permata, Bathaludin, & Wibowo, 2012; Wimanda, Maryaningsih, Nurliana, & Satyanugroho, 2014a ). In general, loans extended to the property sector declined from around 25% in 2012 to 15% in 2015 and to 11% more recently. Growth of housing loans also fell from around 45% in 2012 to 15% in 2015 and to 7% recently. Meanwhile, real estate loans began to decelerate after the LTV was tightened in August 2013, falling from 35% at that time to 15% in 2015 but then accelerating again to around 20% recently. The recent surge in real estate loans since 2015 has been attributed to increased capital spending as an integral part of the Government’s fiscal stimulus package, which also fed through to growth of construction loans at around 28%. It is important to note that the automotive and property sectors contain a large import component and, hence, controlling credit to both sectors helped narrow the current account deficit, which was a macroeconomic sticking point in 2013. During the third period, as part of the policy to stimulate economic growth, while maintaining macroeconomic stability and financial system stability, Bank Indonesia relaxed macroprudential regulations by raising the LTV ratio (or, reducing downpayments on loans) to an average of around 10% in June 2015. As described above, macroeconomic and macrofinancial projections at the time indicated that future inflation risk and FSS risk would be kept under control, similar to conditions in the first quadrant. Nonetheless, reductions to the monetary policy rate were constrained by ubiquitous uncertainty surrounding the proposed FFR hike through to December 2015. Therefore, Bank Indonesia began to ease

Central Bank Policy Mix     513 its policy mix by relaxing macroprudential policy instruments in June 2015 and subsequently lowering the reserve requirement in November 2015 before finally reducing the monetary policy rate at the beginning of 2016. Easing the policy mix by lowering the policy rate and reserve requirement on the monetary side and raising the LTV ratio (lowering loan downpayments) on the macroprudential side, coupled with fiscal stimuli and accelerated structural reforms by the government, catalyzed economic growth, while maintaining macroeconomic and financial system stability. In terms of macroprudential policy, Bank Indonesia began to apply the CCB at the end of 2015. Congruent with the ongoing spirit of loosening the policy mix, the CCB rate was set at 0% and is reviewed every 6 months. CCB application by Bank Indonesia, including international macroprudential standards, is critically important.

15.5.5. Institutional Arrangements Central bank policy mix effectiveness requires strong institutional arrangements at the central bank and in terms of coordination with the government and relevant authorities. To that end, Bank Indonesia has refined its macroeconomic projection and policy analysis framework by including macrofinancial linkages, specifically for the banking and external sectors. The integrated analysis and projection framework is used as the basis underlying policy mix formulation, including interest rate policy, exchange rate policy, the reserve requirement, and macroprudential policy (LTV). Research has also been conducted to investigate the behavior of foreign capital flows and financial procyclicality, particularly in terms of bank lending and the housing sector. More analysis and research is required, however, to broaden our understanding of macrofinancial linkages, procyclicality and systemic risk. The data and statistics have also been enhanced through inclusion of FSS indicators and an early warning system as well as cross-sector financial balance sheet statistics at the national and regional levels. The decision-making process has also been strengthened. In practice at several central banks, there are disparate views about whether to maintain separate monetary policy and FSS policy committees or integrate them. Kohn (2015), for instance, recommended maintaining separate committees due to the different targets and focus, instruments, and accountabilities. Such conditions are found at the Bank of England, where three committees relating to FSS have been formed, separate to the monetary policy committee, since the global financial crisis, namely the Prudential Regulation Authority (PRA) dealing with microprudential policies, the Financial Conduct Authority (FCA) for the financial markets, and the Financial Policy Committee (FPC) to formulate macroprudential policies. At Bank Indonesia, monetary, macroprudential, and payment system policy formulation and decision-making are the preserve of the monthly BI Board of Governors’ Meeting. The process does involve three committees, however, namely the monetary policy committee, the FSS committee and the payment system committee, led by the respective deputy governor responsible for policy formulation. A joint committee was also established to strengthen the policy mix by integrating

514    Central Bank Policy the projections and analyses of the three policy aspects along with formulating the policy mix required. Strengthening the projection models and policy mix formulation process has enriched our understanding of macroeconomic conditions as well as the macrofinancial linkages in the financial system, along with formulating a more optimal policy mix to achieve price stability and support financial system stability. This is very different to the standard monetary policy framework that is merely based on the ITF. Bank Indonesia also coordinates closely with the government and other relevant authorities. Monetary and fiscal policy coordination between Bank Indonesia and the Government is applied to state budget formulation as well as various other aspects of macroeconomic management, centrally and through regional administrations. Although Bank Indonesia was granted independence to execute its tasks and duties pursuant to prevailing regulations, the Bank Indonesia policy mix represents an integral part of the national policy mix, particularly in terms of macroeconomic policy, financial system stability and structural reforms (Warjiyo, 2013a). Concerning financial system stability, coordination is facilitated through the Financial Stability Policy Committee (KKSK), chaired by the Minister of Finance and with the Governor of Bank Indonesia, Chairman of the Financial Services Authority (OJK), and Chairman of the Deposit Insurance Corporation (LPS) as members. Furthermore, the recently enacted Financial System Crisis Prevention and Resolution Act (PPKSK Act) provides a legal foundation for FSS policy coordination, details the respective jurisdictions of each respective institution and contains procedures for the handling of systemic banks as well as the crisis prevention and resolution mechanism. Bilaterally, coordination is also implemented between the macroprudential regulation and supervision conducted by Bank Indonesia and the microprudential regulation and supervision by the Financial Services Authority (OJK).

15.6. Concluding Remarks This chapter has explained the concepts, theories, and implementation of the central bank’s policy mix in pursuance of the dual mandate of achieving price stability and maintaining financial system stability. The policy mix consists of four key elements, namely interest rate policy, exchange rate policy, foreign capital flow management, and macroprudential policy. The central bank’s dual mandate of achieving price stability and financial system stability is mutually complementary. Four cases were presented to demonstrate the consistency between the targets of price and financial system stability, followed by the appropriate policy mix response. The ITF-based monetary policy framework represents a solid basis to support the policy mix of the central bank. The key, however, is to expand and extend the macroeconomic projection and policy analysis framework under ITF to include the macrofinancial linkages found in the financial system, particularly through assessments of procyclicality and systemic risk, as well as a policy mix that is consistent with the specific challenges faced. The structural macroeconomic theoretical foundations and modeling approaches per IITF and DSGE, as the basis of central bank policy mix formulation, were also described.

Central Bank Policy Mix     515 Bank Indonesia’s experience of applying a policy mix since 2010 has evidenced its superiority over the standard ITF. Three episodes of macroeconomic and FSS issues in Indonesia have shown that application of the BI policy mix has supported economic resilience in Indonesia and successfully maintained macroeconomic and financial system stability. Underpinning the policy mix formulation process, Bank Indonesia has also expanded the existing macroeconomic projection models to include macrofinancial linkages, particularly in the banking system and external sector. Research and analysis has also been developed to broaden our understanding of macrofinancial linkages, specifically in terms of credit, property and procyclicality, external debt, foreign capital flow volatility as well as systemic risk assessment in the banking system, particularly in terms of interconnectedness and financial networks in the financial markets and payment system. The policymaking process was further reinforced by the establishment of a joint committee to oversee monetary, macroprudential, and payment system policy, which integrates all the three aspects and provides optimal policy mix response recommendations. Close coordination is also maintained with the Government and other relevant authorities. Nationally, to achieve robust economic growth, while preserving macroeconomic and financial system stability, a national economic mix is required, consisting of: macroeconomic policy, FSS policy, and structural reforms. Accelerating structural reform policy aims to boost economic growth from a national production capacity (potential output) perspective. Furthermore, macroeconomic policy coordination intends to manage the economic cycle and prevent internal imbalances, namely high inflation, and external imbalances, such as an excessive current account deficit. Meanwhile, FSS policy coordination aims to manage the financial cycle to dampen boom-bust cycles and prevent a build-up of systemic risk that could culminate in a financial crisis.

This page intentionally left blank

Bibliography AbuDalu, A., Ahmed, E. M., Almasaied, S. W., & Elgazoli, A. I. (2014). The real effective exchange rate impact on ASEAN-5 economic growth. International Journal of Economics & Management Sciences, 3(2). Acemoglu, D., Ozdaglar, A., & Tahbaz-Salehi, A. (2015). Systemic risk and stability in financial networks. American Economic Review, 105(2), 564–608. Acharya, V. (2015, April). Financial stability in the broader mandate for central banks: A political economy perspective. Hutchins Center Working Paper No. 11. Brookings, Washington, D.C. Acharya, V., & Yorulmazer, dan T. (2003). Information contagion and inter-bank correlation in a theory of systemic risk. CEPR Discussion Paper 3743. Adiningsih, H., Siregar, H., & Hasanah, H. (2013). Does the J-curve phenomenon exist in Indonesia’s bilateral trade balances with major trading countries? ASEAN Journal of Economics, Management and Accounting, 1(1), 13–22. Adler, G., & Tovar, C. E. (2014). Foreign exchange interventions and their impact on exchange rate levels. Monetaria, 1( January–June), 1–48. Adolfson, M., Laseen, S., Linde, J., & Villani, M.. (2007). RAMSES: A new general equilibrium model for monetary policy analysis. Economic Review, 2, 5–40. Adrian, T., Covitz, D., & Liang, N. (2013, February). Financial stability monitoring. Federal Reserve Bank of New York Staff Report No. 601. Affandi, Y., & Mochtar, F. (2013). Current account and real exchange rate dynamics in Indonesia. Procedia Economics and Finance, 5, 20–29. Agenor, P. R. (2000). Monetary policy under flexible exchange rates: An introduction to inflation targeting. Policy Research Working Paper No. 2511. The World Bank, Washington. Agénor, P. R., & da Silva, L. A. P. (2013). Inflation targeting and financial stability: A perspective from the developing world. Banco Central Do Brazil, Working Paper No. 324, September. Agung, J. (1998). Financial deregulation and bank lending channel of monetary policy in developing countries: The case of Indonesia. Asian Economic Journal, 12(3), 273–294. Agung, J. (2000). Financial constraints, firm’s investment and channel of monetary policy for Indonesia. Buletin Ekonomi Moneter dan Perbankan, 3(1), 146–178. Agung, J., Kusmiarso, B., Pramono, B., Hutapea, E. G., Prasmuko, A., & Prastowo, N. J. (2001). Credit crunch in Indonesia in the aftermath of the crisis. Jakarta: Direktorat Riset Ekonomi dan Kebijakan Moneter Bank Indonesia. Agung, J., Morena, R., Pramono, B., & N. J. Prastowo. (2002a). Bank lending channel of monetary transmission in Indonesia. In: P. Warjiyo & J. Agung (Eds.), Transmission mechanism of monetary policy in Indonesia. Jakarta: Bank Indonesia. Agung, J., Morena, R., Pramono, B., & Prastowo, N. J. (2002b). Monetary policy and firm investment: Evidence for balance sheet channel in Indonesia. In P. Warjiyo & J. Agung (Eds.), Transmission mechanism of monetary policy in Indonesia. Jakarta: Bank Indonesia. Ahearne, A. G., Gagnon, J., Haltmaier, J., & Kamin, S. (2002). Preventing deflation: Lessons from Japan’s experience in the 1990s. International Finance Discussion Papers, No 729, Board of Governors of the Federal Reserve System.

518   Bibliography Ahearne, A. G, Griever, W. L., & Warnock, F. E. (2004). Information costs and home bias: An analysis of US holdings of foreign equities. Journal of International Economics, 62, 313–336. Ahmed, S., & Zlate, A. (2014). Capital flows to emerging market economies: A brave new world? Journal of International Money and Finance, 48, 221–248. Aizenman, J., Chinn, M. D., & Ito, H. (2008). Assessing the emerging global financial architecture: Measuring the trilemma’s configurations overtime. NBER Working Paper No. 14533. Aizenman, J., Hutchison, M., & Noy, I. (2011). Inflation targeting and real exchange rates in emerging markets. World Development, 39(5), 712–724. Alamsyah, H. (2008). Persistensi inflasi dan implikasinya terhadap pilihan kebijakan moneter di Indonesia. Ph.D. Dissertation, University of Indonesia, Indonesia. Alamsyah, H., Adamanti, J., Yumanita, D., & Astuti, R. I. (2014, December). Financial cycles in Indonesia. Bank Indonesia Working Paper No. WP/8/2014. Alamsyah, H., Joseph, C., Agung, J., & Zulverdy, D. (2001). Towards implementation of inflation targeting in Indonesia. Bulletin of Indonesian Economic Studies, 37(3), 309–324. Alesina, A., & Gatti, R. (1995). Independent central banks: Low inflation at no cost? American Economic Review, 85. Alesina, A., & Summers, L. W. (1993). Central bank independence and macroeconomic performance: Some comparative evidence. Journal of Money, Credit, and Banking, 25. Alesina, A., & Tabellini, G. (1987). Rules and discretion with non-coordinated monetary and fiscal policies. Economic Inquiry, 25(4), 619–630. Alfaro, L., Kalemli-Ozcan, S., & Volosovych, V. (2011). Sovereigns, upstream capital flows, and global imbalances. NBER Working Paper No. w17396. Allen, W. A. (1999). Inflation targeting: The British experience. Bank of England handbooks in Central Banking Lecture Series No. 1. London: Bank of England. Allen, F., & Babus, A. (2007). Networks in finance. In: P. Kleindorfer & J. Wind (Eds.), Network-based strategis and competencies. Philadelphia, PA: Wharton School Publishing. Allen, F., Babus, A., & Carletti, E. (2010, July). Financial connections and systemic risk. NBER Working Paper No. 16177. Allen, F., & Carletti, E. (2008, July). Financial system: Shock absorber or amplifier? BIS Working Papers No. 257. Allen, F., & Gale, D. (2000). Financial contagion. Journal of Political Economy, 108(1), 1–33. Allen, W. A., & Wood, G. (2006). Defining and achieving financial stability. Journal of Financial Stability, 2(2), 52–72. Alexander, W. E., Balino, T. J. T., & Enoch, C. (1995). Adopting indirect instruments of monetary policy. IMF Occasional Paper No. 126. Washington, DC. Altunbas, Y., de Bondt, G., & Marquéz-Ibañez, D. (2004, May). Bank capital, bank lending and monetary policy in the Euro area. Kredit und Kapital. Altunbas, Y., Gambacorta, L., & Marquéz-Ibañez, D. (2010). Does monetary policy affect bank risk-taking? BIS Working Papers No. 298. Altunbas, Y., Gambacorta, L., & Marquéz-Ibañez, D. (2009). Securitization and the bank lending channel. European Economic Review, 53(8), 996–1009. Amato, J. D. (2005). Risk aversion and risk premia in the CDS market. BIS Quarterly Review, December, 55–68. Amato, J. D., & Gerlach, S. (2002, November). Inflation targeting in emerging market and transition economies: Lessons after a decade. Center of Economic Policy Research (CERP) Discussion Paper No. 3074. Amato, J. D., Morris, S., & Shin, H. S. (2003, January). Communications and monetary policy. BIS Working Paper No. 123 .

Bibliography    519 Almeida, A., & Goodhart, C. (1998). Does the adoption of inflation targets affect central bank behavior? Banca-Nazionale-del-Lavoro-Quarterly-Review, 51(204), 19–107. Ammer, J., & Freeman, R. T. (1995). Inflation targeting in the 1990s: The experiences of New Zealand, Canada, and the United Kingdom. Journal of Economics and Business, 47(2), 165–192. Amri, P. D., Chiu, E. M. P., Richey, G. M., Willet, T. D. (2013). Do financial crises discipline future credit growth?. Journal of Financial Economic Policy, 9(3), 284–301. Anachotikul, N., & Zhang, L. (2014). Portfolio flows, global risk aversion and asset prices in emerging markets. IMF working paper. Anand, R., Peiris, S., & Saxegaard, M. (2010, January). An estimated model with ­macrofinancial linkages for India. IMF Working Paper WP/10/21. Anaya, P., Hachula, M., & Offermanns, C. (2015, November). Spillovers of U.S. unconventional monetary policy to emerging markets: The role of capital flows. Discussion Paper 2015/35. Freie Universitat Berlin, Berlin. Andersen, L. C., & Jordan, J. L. (1968). Monetary and fiscal actions: A test of their relative importance in economic stabilization. FRB of St. Louis Review, 50, November 1968. Angelini, P., Neri, S., & Panetta, F. (2011, March). Monetary and macroprudential policies. Banca D’Italia Working Papers No. 801. Angeloni, I., Anil, K., Benoit, M., & Daniele, T. (2002). Monetary transmission in the Euro area: Where do we stand? ECB Working Paper No. 114. Angeloni, I., & Ehrmann, M. (2003, July). Monetary policy transmission in the Euro area: Any changes after EMU? ECB Working Paper No. 240. Angeloni, I., & Faia, E. (2009). A tale of two policies: Prudential regulation and monetary policy with fragile banks. Kiel Institute for the World Economy Working Papers 1569. Angeloni, I., Faia, E., & Duca, M. L. (2011, December). Monetary policy and risk taking. Bruegel Working Paper 2011-00. Angeloni, I., Faia, E., & Winkler, R. (2011). Exit strategies. Kiel Institute for the World Economy Working Papers 1676. Angeloni, I., Kashyap, A. K., Mojon, B., & Terlizzese, D. (2003, September). Monetary transmission in the Euro area: Does the interest rate channel explain all? NBER Working Paper 9984. Annanchotikul, N., & Chang, L. (2014, August). Portfolio flows, global risk aversion, and asset prices in emerging markets. IMF Working Paper WP/14/156 . Archer, D. (2005). Central bank communication and the publication of interest rate projections. Sveriges riskbank conference on inflation targeting, Stockholm, June. Arena, M., Bouza, S., Dabla-Norris, E., Gerling, K., & Njie, L. (2015, January). Credit booms and macroeconomic dynamics: Stylized facts and lessons for low-income countries. IMF Working Paper WP/15/11. Ariyoshi, A., Habermeier, K., Laurens, B., Otker-Robe, I., Canales-Kriljenko, J. I., & Kirilenko, A. (2000). Capital control: Country experiences with their use and liberalization. Occassional Paper No. 190. Washington, DC: International Monetary Fund. Auboin, M., & Rut, M. (2011, October). The relationship between exchange rates and international trade: A literature review. Staff Working Paper ERSD-2011-17. World Trade Organization. Baek, I. M. (2006). Portfolio investment flows to Asia and Latin America: Pull, push or market sentiment? Journal of Asian Economics, 17(2), 363–373. Baka, W. (1994). Please respect the National Bank. Central Banking, 5, 65–72. Batini, N., & Haldane, A. (1999). Monetary policy rules and inflation forecast. Bank of England Quarterly Bulletin, 39, 60–67. Batini, N., Harrison, R. H., & Millard, S. P. (2003). Monetary policy rules for an open economy. Journal of Economic Dynamics and Control, 27(11–12), 2059–2094.

520   Bibliography Batini, N., & Nelson, E. (2000). Optimal horizons for inflation targeting. Bank of England Working Paper. Baldwin, R., DiNino, V., Fontagné, L., De Santis, R. A., & Taglioni, D. (2008). Study on the impact of the euro on trade and foreign direct investment. Economic Papers 321. European Commission. Balino, T. J. T., & Zamalloa, L. M. (Eds.). (1997). Instruments of monetary management: Issues and country experiences. Washington, DC: IMF. Ball, L. (1994). What determines the sacrifice ratio? In N. G. Mankiw (Ed.), Monetary policy. National Bureau of economic research studies in business cycles (Vol. 29). Chicago, IL: The University of Chicago Press. Ball, L. (1997, March). Efficient rule for monetary policy. NBER Working Paper No. 5952. Ball, L. (1999). Policy rules for open economies. In J. Taylor (Ed.), Monetary policy rules. Chicago, IL: The University of Chicago Press. Ball, L., & Mankiw, N. G. (1994, March). Asymmetric price adjustment and economic fluctuations. The Economic Journal, 104. Ball, L., & Mankiw, N. G. (1995, February). Relative price changes as aggregate price shocks. Quarterly Journal of Economics, 110. Ball, L., & Sheridan, N. (2003, June). Does inflation targeting matter? IMF Working Paper No. WP/03/129. Banerjee, S., & Basu, P. (2015, May). A dynamic stochastic general equillibrium model for India. NCAER Working Paper No. WP 109 . Bank Indonesia (BI). (1999a). Pengukuran Target Inflasi dalam Rangka Melaksanakan Kebijakan Moneter secara forward looking. Divisi Sektor Riil. Jakarta: Direktorat Riset Ekonomi dan Kebijakan Moneter Bank Indonesia. Bank Indonesia (BI). (1999b). Mekanisme Pengendalian Moneter dengan Inflasi sebagai Sasaran Tunggal. Jakarta: Divisi Analisis dan Perencanaan Kebijakan, Direktorat Riset Ekonomi dan Kebijakan Moneter Bank Indonesia. Bank Indonesia (BI). (2000a). The general equilibrium model of Bank Indonesia (GEMBI). Jakarta: Divisi Studi Makroekonomi, Direktorat Riset Ekonomi dan Kebijakan Moneter Bank Indonesia. Bank Indonesia (BI). (2000b). Kerangka Kerja Operasional Kebijakan Moneter Berbasis Pengendalian Suku Bunga dalam Mencapai Sasaran Inflasi. Jakarta: Divisi Analisis Kebijakan Moneter, Direktorat Riset Ekonomi dan Kebijakan Moneter Bank Indonesia. Bank Indonesia (BI). (2001a). Short-term forecast model of Indonesian economy (SOFIE). Jakarta: Divisi Studi Makroekonomi, Direktorat Riset Ekonomi dan Kebijakan Moneter Bank Indonesia. Bank Indonesia (BI). (2001b). Laporan Tahunan 2000, 2001, 2002, 2003. Jakarta: Bank Indonesia. Bank Indonesia (BI). (2004a). Undang-undang Republik Indonesia Nomor 23 tahun 1999 tentang Bank Indonesia, Bank Indonesia, Jakarta, 1999, yang diamandemen dengan Undang-undang Republik Indonesia Nomor 3 Tahun 2004. Bank Indonesia (BI). (2004b). Blue print inflation targeting framework. Jakarta: Direktorat Riset Ekonomi dan Kebijakan Moneter Bank Indonesia. Bank for International Settlement (BIS). (2000). The contributions of payment systems to financial stability. Paper presented in Payment System Workshop, CEMLA, Mexico City, May. Bank for International Settlement (BIS). (2001, January). Core principles on systemically important payment system. Committee on Payment and Settlement System (CPSS). Bank for International Settlement (BIS). (2003, August). The role of central bank money in payment systems. Committee on Payment and Settlement Systems (CPSS). Bank for International Settlement (BIS). (2011a, May). Central bank governance and financial stability. Study Group Report by Stefan Ingves, Governor, Sveriges Riksbank.

Bibliography    521 Bank for International Settlement (BIS). (2011b, January). The future of central banking under post-crisis mandates. Ninthook BIS annual conference 24–25 June 2010. BIS Papers No. 55. Bank for International Settlements. (2012). A framework for dealing with domestic systemically important banks. Basel Committee on Banking Supervision. Bank for International Settlement (BIS). (2014). The transmission of unconventional monetary policy to the emerging markets. BIS Papers, No. 78. Bansal, R. (1997). An exploration of the forward premium puzzle in the currency market. The Review of Financial Studies, 10, 369–403. Barro, R. J. (1977). Unanticipated money growth and unemployment in the United States. American Economic Review, 67. Barro, R. J. (1986). Reputation in a model of monetary policy with incomplete information. Journal of Monetary Economics, 17(1). Barro, R. J. (1995, May). Inflation and economic growth. Bank of England Quarterly Bulletin. Barro, R. J., & Gordon, D. B. (1983). Rules, discretion, and reputation in a model of monetary policy. Journal of Monetary Economics, 17(1), 101–122. Bask, M., Liu, T., & Widerberg, A. (2007). The stability of electricity prices: Estimation and inference of the Lyapunov exponents. Physica A: Statistical Mechanics and its Applications. 376(c), 565–572. Bean, C., Larsen, J., & Nikolov, K. (2002, January). Financial frictions and the monetary transmission mechanism: Theory, evidence, and policy implications. European Central Bank Working Paper Series No. 113. Beare, J. B. (1978). Macroeconomics: Cycles, growth, and policy in a monetary economy. New York, NY: Macmillan Publishing Co. Inc.. Beaton, K., Lalonde, R., & Snudden, S. (2010). The propagation of U.S. shocks to Canada: Understanding the role of real-financial linkages. Bank of Canada Working Paper No. 2010–40. Beck, T., Buyukkarabacak, B., Rioja, F. K., & Valev, N. T. (2012). Who gets the credit? And does it matter? Household vs. firm lending across countries. Journal of Macroeconomics, 12(1), 1–46. Beck, T., Colciago, A., & Pfajfar, D. (2014, April). The role of financial intermediaries in monetary policy transmission. (DNB Working Paper No.420). Retrieved from https://www.dnb.nl/binaries/Working%20Paper%20420_tcm46-306757.pdf. Beechey, M. J., Johannsen, B. K., & Levin, A. (2007). Are long-run inflation expectations anchored more firmly in the Euro area than in the United States? CEPR Discussion Papers No. 6536. Berg, C., Hallsten, K., Queijo, V., HeideKen, V., & Soderstrom, U. (2013). Two decades of inflation targeting: Main lessons and remaining challenges. Sveriges Riksbank Economic Review, 3(Special Issue). Bernanke, B. S. (1992). The federal funds rate and the channels of monetary transmission. American Economic Review, 82(4), 901–921. Bernanke, B. S. (2007, March). Globalization and monetary policy. In Fourth economic summit, Stanford Institute for Economic Policy Research. Bernanke, B. S. (2009). The right reform for the Fed. Washington Post, 29 November. Bernanke, B. S. (2010). Monetary policy and the housing bubble. Paper presented at annual meeting of the American Economic Association, Atlanta, January. Bernanke, B. S., & Blinder, A. (1988). Credit, money, and aggregate demand. American Economic Review, 78, 435–439. Bernanke, B. S., & Gertler, M. (1995). Inside the black box: The credit channel of monetary transmission. Journal of Economic Perspective, 9(4), 27–48. Bernanke, B. S., & Gertler, M. (2000, February). Monetary policy and asset price volatility. NBER Working Paper No. 7559 .

522   Bibliography Bernanke, B. S., Gertler, M., & Gilchrist, S. (1999). Financial accelerator in quantitative business fluctuations framework. In J. B. Taylor & M. Woodford (Eds.), Handbook of macroeconomics (Vol. 1). Amsterdam: Elsevier. Bernanke, B. S., Laubach, T., Mishkin, F., & Posen, A. (1999). Inflation targeting: Lessons from the international experience. Princeton, NJ: Princeton University Press. Bester, H. (1985). Screening vs rationing in credit markets with imperfect information. American Economic Review, 75(4), 50–55. Bester, H., & Hellwig, M. (1987). Moral hazard and equilibrium credit rationing: An overview of the issues. In G. Bamberg & K. Spremann (Eds.), Agency theory, information and incentives. Berlin, Germany: Springer. Bihari, P. (2015). An Odysseian journey: Experience with forward guidance. Economic Review, 62(7–8), 749–766. Bikhchandani, S., & Sharma, S. (2001). Herd behaviour in financial market. IMF Staff Papers, 47(3). Bisias, D., Flood, M., Lo, A. W., & Valavanis, S. (2012). A survey of systemic risk analytics. Office of Financial Research Working Paper #0001. Blake, A. P. (2000). Optimality and Taylor rules. National Institute of Economic Review, 4, 80–91. Blake, A. P. Weale, M., & Young, G. (1998). Optimal monetary policy. National Institute Economic Review, 164(1), 100–109. Blanchard, O., Dell’Ariccia, G., & Mauro, P. (2010). Rethinking macroeconomic policy. IMF Staff Discussion Note 10/03. Blanchard, O. J., Adler, G., & Filho, I. C. (2015, November). Can foreign exchange intervention stem exchange rate pressures from global capital flow shocks? Peterson Institute of International Economics No. WP 15-18. Blanchard, O. J., & Fischer, S. (1989). Lectures on macroeconomics. Cambridge, MA: MIT Press. Blancher, N., Mitra, S., Morsy, H., Otani, A., Severo, T., & Valderrama, L. (2013, July). Systemic risk monitoring (SysMo) toolkit: A user guide. IMF Working Paper WP/13/168. Blejer, M. I., Ize, A., Leone, A. M., Werlang, S. (2000). Inflation Targeting in Practice. International Monetary Fund, Washington. Blejer, M. I., Ize, A., Leone, A. M., & Werlang, S. (Eds.). (2001). Inflation targeting in practice: Strategic and operational issues, and application to emerging economies. Washington, DC: IMF. Blejer, M. I., Leone, A. M., Rabanal, P., & Schwartz, G. (2002). Inflation targeting in the context of IMF-supported adjustment programs. IMF Staff Papers, 49(3), 313–338. Blinder, A. (2000). Central-bank credibility: Why do we care? How do we built it?. American Economic Review, 90(5), 1421–1431. Blinder, A. S. (1998). Central banking in theory and practices. Cambridge, MA: MIT Press. Blinder, A. S. (2010). How central should the central bank be? Journal of Economic Literature, 48(1), 123–133. Blinder, A. S. (2012, September). Central bank independence and credibility during and after a crisis. Princeton University Griswold Center for Economic Policy Studies Working Paper No. 229. Blinder, A. S., Goodhart, C., Hildebrand, P., Lipton, D., & Wyplosz, C. (2001). How do central banks talk? Switzerland: International Center for Monetary and Banking Studies. Blinder, A. S., & Krueger, A. B. (2004). What does the public know about economic policy? And how does it know it?. Brookings papers on economic activity, Economic studies program, the Brookings Institutions, 35(1), 327–397.

Bibliography    523 Bloxham, P., Kent, C., & Robson, M. (2011). Asset prices, credit growth, monetary and other policies: An Australian case study. NBER Working Paper 16485. Boediono. (1998). Merenungkan Kembali Mekanisme Transmisi Moneter di Indonesia. Buletin Ekonomi Moneter dan Perbankan, 1(1). Boediono. (2000). Comments on inflation targeting in Indonesia. In C. Joseph & A. Gunawan (Eds.), Monetary policy and inflation targeting in emerging economies. Jakarta: Bank Indonesia. Bofinger, P. (2000). A framework for stabilizing the euro/yen/dollar triplet. The North American Journal of Economics and Finance, 11(2), 137–151. Bofinger, P. (2001). Monetary policy: Goals, institutions, strategies, and instruments. Oxford: Oxford University Press. Bogdanski, J., Springer, P., Goldfjan, I., & Tombini, A. (2000). Inflation targeting in Brazil: Shocks, backward-looking prices, and IMF conditionality. In N. Loayza & R. Soto (Eds.), Inflation targeting: Design, performances, and challenges. Santiago: Central Bank of Chile. Bogov, D. (2011). Modern payment systems’ impact on monetary policy and financial stability. Welcome speech di 16th international conference of clearing institutions in Central and Eastern Europe, Skopje, October. Boivin, J., Giannoni, M. P., &. Mojon, B. (2008). How has the euro changed the monetary transmission? NBER Working Papers No. 14190. Boivin, J., Kiley, M. T., & Mishkin, F. S. (2010, April). How has the monetary transmission mechanism evolved over time? NBER Working Paper No. 15879. Bordo, M. D. (2007, December). A brief history of central banks. Cleveand, OH: Federal Reserve Bank of Cleveand, Economic Commentary. Bordo, M. D., Eichengreen, B., Klingebiel, D., & Peria, S. M. M. (2001). Is the crisis problem growing more severe? Economic Policy, 32, 51–82. Bordo, M. D., & Flandreau, M. (2003). Core, periphery, exchange rate regimes, and globalization. In M. D. Bordo, A. M. Taylor, & J. G. Williamson (Eds.), Globalization in historical perspective. Chicago, IL: University of Chicago Press. Bordo, M. D., & Schwartz, A. (1984). A retrospective on the cassical gold standard (pp. 1821–1931). Chicago, IL: University of Chicago Press. Bordo, M. D., & Wheelock, D. C. (2007). Stock market booms and monetary policy in the twentieth century. Federal Reserve Bank of St. Louis Review, 89(2), 91–122. Borio, C. (2003). Towards a macroprudential framework for financial supervision and regulation? BIS Working Papers No. 128. Borio, C., English. B., & Filardo, A. (2003). A tale of two perspectives: old or new challenges for monetary policy?, BIS Working Papers No. 127. Borio, C., McCauley, R. N., and McGuire, P. (2011). Global credit and domestic credit booms. BIS Quarterly Review, 43–57. Borio, C., & Shim, I. (2007). What can (macro-)prudential policy do to support monetary policy?. BIS Working Paper No. 242l. Borio, C. E. V. (1997). The implementation of monetary policy in industrial countries: A survey. BIS Economic Papers, 57, 411–425. Borio, C. E. V. (2010). Implementing a macroprudential framework: Blending boldness and realism. Keynote address for BIS-HKMA Research Conference about “Financial stability: Towards a Macroprudential Approach,” Honk Kong SAR, July. Borio, C. E. V. (2011a, September). Central banking post-crisis: What compass for uncharted waters? BIS Working Papers No. 353. Borio, C. E. V. (2011b, May). Global imbalances and the financial crisis: Link or no link? BIS Working Papers No. 346. Borio, C. E. V. (2014, September). The international monetary and financial system: Its Achilles heel and what to do about it, BIS Working Papers No. 456.

524   Bibliography Borio, C. E. V., & Disyatat, P. (2010, November). Unconventional monetary policies: An appraisal. BIS Working Papers No 292. Borio, C. E. V., & Drehmann, M. (2009, June). Towards an operational framework for ­financial stability: “Fuzzy” measurement and its consequences. BIS Working Papers No. 284. Borio, C. E. V., & Lowe, P. (2002). Asset prices, financial and monetary stability: Exploring the Nexus. BIS Working Paper No. 114. Borio, C. E. V, McCauley, R., & McGuire, P. (2011). Global credit and domestic credit booms. BIS Quarterly Review, September. Borio, C. E. V., Russo, D., & van de Bergh, P. (1992). Payment system arrangements and related policy issues: A cross-country comparison. SUERF Papers on Monetary Policy and Financial Systems 13, Société Universitaire Européene de Recherches Financiéres, Tilburg. Borio, C. E. V., & White, W. R. (2003, February). Whither monetary and financial stability? The implications of evolving policy regimes. BIS Working Paper No. 147. Borio, C. E. V., & Zhou, H. (2008, December). Capital regulation, risk taking and monetary policy: A missing link in the transmission mechanism. BIS Working Paper No. 268. Bradley, M. D., & Jansen, D. W. (1989). Understanding nominal GNP targeting. FRB of St. Louis Review, 71(6), 31–40. Brayton, F., & Tinsley, P. (1996, October). A guide to the FRB/US: A macromodel of the United States. Federal Reserve Board, Finance and Economics Discussion Series. Brázdik, F., Hlaváček, M., & Maršál, A. (2011). Survey of research on financial sector modeling within DSGE models: What central banks can learn from it. Czech National Bank (CNB) Research and Policy Notes 3/2011. Britton, A. (2001). Monetary regimes of the twentieth century. Cambridge: Cambridge University Press. Broaddus, J. A., Jr. (2002, July/August). Transparency in the practice of monetary policy. FRB of St Louis Review, 84(4), 161–166. Brunnermeier, M. K. (2009). Deciphering the liquidity and credit crunch 2007–2008. Journal of Economic Perspectives, 23(1), 77–100. Brunnermeier, M. K., & Sannikov, Y. (2012). A macroeconomic model with a financial sector. Princeton, NJ: Princeton University, Mimeo. Bruno, V., & Shin, H. S. (2014). Capital flows and the risk-taking channel of monetary policy. NBER Working Paper No. 18942. Buch, C. M., Eickmeier, S., & Prieto, E. (2013). In search for yield? Survey-based evidence on bank risk taking. Tübingen, Germany: University of Tubingen, Mimeo. Budiyanti, E. (2009). Time Inconsistency dalam Kebijakan Moneter Kasus Indonesia: Sebelum dan Sesudah Krisis. Tesis. Fakultas Ekonomi – Universitas Indonesia. Buiter, W. H. (1999). Alice in Euroland. Journal of Common Market Studies, 37(2). Calderon, C., & Kubota, M. (2012). Gross inflows gone wild: Gross capital inflows, credit booms and crises. London: World Bank Group, Economic Research. Calvo, G. A. (1978, November). On the time inconsistency of optimal policy in a monetary economy. Econometrica, 46(6), 1411–1428. Calvo, G. A. (1983). Staggered prices in a utility maximising framework. Journal of Monetary Economics, 12(3), 383–398. Calvo, G. A., Leiderman, L., & Reinhart, C. M. (1992). Capital inflows to Latin America: The 1970s and the 1990s. IMF Working Paper No. WP/92/85. Calvo, G. A., & Reinhart, C. M. (1996). Capital flows to Latin America: Is there evidence of contagion effects? World Bank Policy Research Working Paper No. 1619. Calvo, G. A., & Reinhart, C. M. (2000). When capital inflows come to a sudden stop: consequences and policy options. In P. B. Kenen & A. K. Swoboda (Eds.), Reforming the international monetary and financial system. Washington, DC: International Monetary Fund.

Bibliography    525 Calvo, G. A., & Reinhart, C. M. (2001). Fixing for your life. In S. M. Collins & D. Rodrik (Eds.), Brookings trade forum 2000. Washington, DC: Brookings Institution. Calvo, G. A., & Reinhart, C. M. (2002). Fear of floating. Quarterly Journal of Economics, 117(May), 379–408. Calza, A. (2008). Globalisation, domestic inflation and global output gaps. Working Paper Series No. 890, ECB. Canales-Kriljenko, J. I., Guimaraes, R., & Karacadag, C. (2003). Official intervention in the foreign exchange market: Element of best practice. Working Paper No. WP/03/152. Washington, DC: IMF. Caporalea, G. M., Ali, F. M., Spagnolo, F., & Spagnolo, N. (2015, November). International portfolio flows and exchange rate volatility for emerging markets. DIW Berlin Discussion Papers No. XXXX. Carare, A., Shaechter, A., Stone, M., & Zelmer, M. (2002). Establishing initial conditions in support of inflation targeting. IMF Working Paper No. 102. Carare, A., & Stone, M. R. (2003, January). Inflation targeting regimes. IMF Working Paper No. WP/03/9. Carlstorm, C. T., & Fuerst, T. S. (2000). Monetary shocks agency cost, and business cycles. Federal Reserve Bank of Cleveland Working Paper No. 0011. Case, K. E., & Shiller, R. J. (2003). Is there a bubble in the housing market? Brookings Papers on Economic Activity, 34(2), 299–342. Cassel, G. (1918). Abnormal deviations in international exchanges. Economic Journal, 28(112), 413–415. Castellani, F. (2002). A model of central bank’s accountability. HEI Working Paper No. 04/2002. Cecchetti, S. G. (1995, May/June). Distinguishing theories of the monetary transmission mechanism. Federal Reserve Bank of St. Louis Review. Cecchetti, S. G. (1998, June). Policy rules and targets: Framing the central banker’s problem. FRBNY Economic Policy Review, 4(2), 1–14. Cecchetti, S. G., & Ehrmann, M. (2000). Does inflation targeting increase output volatility? An international comparison of policymakers’ preferences and outcomes. Working Paper No. 69. Central Bank of Chile, Chile. Cecchetti, S. G., Genberg, H., Lipsky, J., & Wadhwani, S. (2000, May). Asset prices and central bank policy. Geneva Report on the World Economy No. 2. Cecchetti, S. G., Genberg, H., & Wadhwani, S. (2002, June). Asset prices in a flexible inflation targeting framework. NBER Working Paper 8970. Cecchetti, S. G., & Kohler, M. (2012, May). When capital adequacy and interest rate policy are substitutes (and when they are not). BIS Working Paper No. 379. Cecioni, M., & Neri, S. (2010). The monetary transmission mechanism in the Euro area: Has it changed and why? Paper presented at Seminar ECB, March. Cettorelli, N., & Goldberg, L. (2012). Banking globalization and monetary transmission. Journal of Finance, 67(5), 1811–1843. Chada, B., & Prasad, E. (1993). Interpreting the cyclical behavior of prices. IMF Staff Papers, 40, 266–298. Chan-Lau, J. A., & Amadou, N. R. (2007). Distance to default in banking: A bridge too far? Journal of Banking Regulation, 9(1), 14–24. Chang, R. (1998, 1st Quarter). Policy credibility and the design of central bank. FRB of Atlanta Economic Review, 83(1), 4–15. Chappel, H. W., & McGregor, R. R., Jr. (2004). Did time inconsistency contribute to the great inflation? Evidence from the FOMC transcripts. Economics and Politics. Chari, V. V., Kehoe, E. J., & Prescott, E. C. (1989). Time consistency and policy, In R. J. Barro, (Ed.), Modern Business Cycle Theory. Cambridge, MA: Harvard University Press.

526   Bibliography Czech National Bank. (2000). Inflation targeting in transition economies: The case of the Czech Republic. Prague: Czech National Bank. Chortareas, G., Stasavage, D., & Sterne, G. (2002, July/August). Does it pays to be transparent? International Evidence from central bank forecasts. FRB St Louis Review, 84(4), 99–118. Christiano, L. J., Eichenbaum, M. S., & Evans, C. L. (1994a, May). Identification and the effects of monetary policy shocks. Paper No. WP-94-7. FRB of Chicago, Chicago, IL. Christiano, L. J., Eichenbaum, M. S., & Evans, C. L. (1994b, April). The effects of monetary policy shocks: Some evidence from the flow of funds. NBER Working Paper No.4699. Christiano, L. J., Eichenbaum, M. S., & Evans, C. L. (1998, February). Monetary policy shocks: What have we learned and to what end? NBER Working Paper No. 6400. Christiano, L. J., Motto, R., & Rostagno, M. (2010). Financial factors in economic fluctuations. European Central Bank Working Paper No. 1192. Christoffel, K., Coenen, G., & Warne, A. (2008). The new area-wide model of the euro area: A micro-founded open-economy model for forecasting and policy analysis. European Central Bank Working Paper No. 944. Chui, M., Kuruc, E., & Turner, P. (2016). Emerging market companies: Currency mismatches, leverage and profitability. New York, NY: Mimeo. Chung, H., Kiley, M., & Laforte, J. P. (2010). Documentation of the estimated, dynamic, optimization-based (EDO) model of the U.S. economy: 2010 Version. Board of Governors of the Federal Reserve System (U.S.). Finance and Economics Discussion Series 2010–29. Chutasripanich, N., & Yetman, J. (2015, March). Foreign exchange intervention: Strategies and effectiveness. BIS Working Papers No. 499. Ciccarelli, M., Maddaloni, A., & Peydro, J. L. (2013). Heterogeneous transmission mechanism: Monetary policy and financial fragility in the Eurozone. Economic Policy, 28(75), 459–512. Claessens, S., Gosh, S. R., & Mihet, R. (2014, August). Macro-prudential policies to mitigate financial system vulnerabilities. IMF Working Paper No. WP/14/55. Claessens, S., & Kose, M. A. (2013). Financial crises: Explanations, types, and implications. IMF Working Paper No. WP/13/28. Claessens, S., Kose, M. A., & Terrones, M. E. (2011). How do business and financial cycles interact? IMF Working Paper No. WP/11/88. Clarida, R. H. (2001, June). The empirics of monetary policy rules in open economies. Seminar on Monetary Policy in Open Economics, IMF. Clarida, R. H., Gali, J., & Gertler, M. (1997). Monetary policy rules in practice: Some international evidence. Paper presented at the 20th international seminar on macroeconomics. Clarida, R. H., Gali, J., & Gertler, M. (1998). Monetary policy rules in practice: Some international evidence. European Economic Review, 42(6), 1033–1067. Clarida, R. H., Gali, J., & Gertler, M. (1999, May). The science of monetary policy: A new Keynesian perspective. NBER Working Paper No. W7147. Clarida, R. H., Gali, J., & Gertler, M. (2000, January). Monetary policy rules and macroeconomic stability: Evidence and some theory. The Quarterly Journal of Economics, 115(1), 147–180. Clarida, R. H., Gali, J., & Gertler, M. (2001, January). Optimal monetary policy in open versus closed economies: An integrated approach. Working Paper. Clark, P. B., Laxton, D., & Rose, D. (1995). Capacity constraints inflation and the transmission mechanis: Forward-looking versus myopic policy rules (International Monetary Fund Working Paper No. 95/75). Retrieved from http://www.imf.org/external/pubs/ cat/longres.aspx?sk=1271

Bibliography    527 Clark, P. B., Laxton, D., & Rose, D. (1996). Asymmetry in the US output–inflation nexus: Issues and evidence. IMF Staff Papers, 43(1), 216–251. Cobham, D., Papadopoulos, A., & Zis, G. (2001). The cost of political intervention in monetary policy. Discussion Paper Series. England: Department of Economics, University of St. Andrews. Collard, F., Dellas, H., Diba, B., & Loisel, O. (2013). Optimal monetary and prudential policies. Mimeo, University of Bern. Collyns, C., & Senhadji, A. (2003). Lending booms, real estate bubbles, and the Asian crisis. In W. C. Hunter, G. G. Kaufman, & M. Pomerleano (Eds.), Asset price bubbles: Implications for monetary, regulatory, and international policies. Cambridge, MA: MIT Press. Comotto, R. (2011, September). The interconnectivity of central and commercial bank money in the clearing and settlement of the European Repo market. European Repo Council. Corbo, V., Landerretche, O., & Schmidt-Hebbel, K. (2000). Does inflation targeting make a difference? Paper presented in the central bank of Chile conference on ten years of inflation targeting: Design, performance, challenges, Santiago, November. Corbo, V., & Schmidt-Hebbel, K. (2000). Inflation targeting in Latin America. Paper presented at the Latin American conference on financial and fiscal policies, Stanford University, November. Cooley, T. F., & Ohanian, L. E. (1991). The cyclical behavior of prices. Journal of Monetary Economics, 28, 25–60. Cortinhas, C. (2007). Exchange rate pass-through in ASEAN: Implications for the prospects of monetary integration in the region. The Singapore Economic Review, 54(4), 657–687. Craig, R. E., Davis, E. P., & Pascual, A. G. (2006). Sources of pro-cyclicality in East Asian financial system. In S. Gerlach & P. Gruenwald (Eds.), Procyclicality of financial systems in Asia. London: Palgrave MacMillian. Csortos, O., Lehmann, K., & Szalai, Z. (2014, July). Theoretical considerations and practical experiences of forward guidance. MNB Bulletin, 9(2), 45–55. Cukierman, A. (1992). Central bank strategy, credibility, and independece: Theory and evidence. Cambridge, MA: MIT Press. Cukierman, A. (1999). Accountability, credibility, transparency and stabilization policy in the Eurosystem. Tel-Aviv Israel: Tel-Aviv University, Mimeo. Cukierman, A. (2000). Establishing a reputation for dependability by means of inflation targets. Economics of Governance, 1(1). 53–76. Cukierman, A. (2003, July/August). Are contemporary central banks transparent about their economic models and objectives, and what different does it makes? FRB St Louis Review. Cukierman, A. (2006). Central bank independence and monetary policymaking institutions: Past, present, and future. Working Papers Central Bank of Chile 360. Central Bank of Chile, Chile. Cukierman, A., & Meltzer, A. H. (1986). A theory of ambiguity, credibility, and inflation under discretion and asymmetric information. Econometrica, 54(5), 1099–1128. Cukierman, A, Webb, S. B., & Neyapti, B. (1992). Measuring the independence of central banks and its effect on policy outcomes. The World Bank Economic Review, 6(3), 353–398. Cumby, R. E., & Obstfeld, M. (1981). Capital mobility and the scope for sterilization: Mexico in the 1970s. NBER Working Paper No. 770. Cumby, R. E., & Obstfeld, M. (1984). International interest rate and price linkages under flexible exchange rate: A review of recent evidence. In J. F. O. Bilson & R. C. Marton (Eds.), Exchange rates: Theory and practice. Chicago, IL: Chicago Press.

528   Bibliography Curcuru, S. E., Thomas, C. P., Warnock, F. E., & Wongswan, J. (2011). US international equity investment and past and prospective returns. American Economic Review, 101(7), 3440–3455. Curdia, V., & Woodford, M. (2009, March). Credit frictions and optimal monetary policy. BIS Working Papers No. 278. Curdia, V., Woodford, M. (2011). The central bank balance sheet as an instrument of monetary policy. Journal of Monetary Economics, 58(1), 54–79. Dahlhaus, T. (2014, June). Monetary policy transmission during financial crises: An empirical analysis. Bank of Canada Working Paper No. 2014-21 . De Bandt, O., & Hartmann, P. (2000, January). Systemic risk: A survey (pp. 249–298). ECB Working Paper No. 35, European Central Bank, Germany. De Bondt, G. J. (2000). Financial structure and monetary transmission in Europe. Cheltenham: Edward Elgar Publ. de Carvallo, & de Castro. (2015, November). Macroprudential and monetary policy interaction: A Brazilian perspective. Banco Central do Brazil Working Papers No. 405. De Vita, G., & Kyaw, K. S. (2008). Determinants of capital flows to developing countries: A structural VAR analysis. Journal of Economic Studies, 35(4), 304–322. Debelle, G. (1997). Inflation targeting in practice. Working Paper No. 97/35. International Monetary Fund, Washington, DC. Debelle, G. (2001). The case for inflation targeting in East Asian countries. In Future directions for monetary policies in East Asia (pp. 65–87). Sydney: Reserve Bank of Australia. Debelle, G., & Fischer, S. (1994). How independient should a central bank be? In Proceedings of a conference, North Falmouth, Boston, MA. Debelle, G., & Fischer, S. (1995). How independent should a central bank be? In J. Fuhrer (Ed.), Goals, guidelines and constraints facing monetary policymakers (pp. 195–221). Federal Reserve Bank of Boston: Boston, MA. Conference Series, 38. Debelle, G., & Laxton, D. (1997). Is the Phillips curve really a curve? Some evidence for Canada, the United Kingdom and the United States. IMF Staff Papers, 44, 249–282. Debelle, G., & Wilkinson, J. (2002, January). Inflation targeting and the inflation process: Some lessons from an open economy. Reserve Bank of Australia Research Discussion Paper No. 2002-01. Dell’Ariccia, G., Laeven, L., & Suarez, G. A. (2016, May). Bank leverage and monetary policy’s risk-taking channel: Evidence from the United States. European Central Bank Working Paper Series No. 1903. Demertzis, M., & Hallett, A. H. (2003, January). Central bank transparency in theory and practice. De Nederlansche Bank Working Paper. Deriantino, E. (2013, May). Banking competition and effectiveness of monetary policy transmission: A theoretical and empirical assessment on Indonesia case. International Journal of Economic Sciences, Prague: University of Economics. Diamond, D. W., & Rajan, R. G. (2010, March). Fear of fire sales and the credit freeze. BIS Working Papers No. 305. Domínguez, K. M. E., & Frankel, J. A. (1993). Does foreign-exchange intervention matter? The portfolio effect. American Economic Review, 83(5), 1356–1369. Dooley, M. (1996). A survey of literature on controls over international capital transactions. IMF Staff Paper, 43(4), 639–687. Dornbusch, R. (1976). Expectation and exchange rate dynamic. Journal of Political Economy, 84, 1161–1176. Dornbusch, R. (1986, September). Special exchange rates for capital account transactions. World Bank Economic Review, 1(1), 3–33. Dornbusch, R. (1988). Purchasing power parity. In New Palgrave: A dictionary of economics. New York, NY: Stockton Press. Dornbusch, R, Fischer, S., & Samuelson, P. (1977). Comparative advantage, trade and payments in Ricardian model with a continuum of goods. American Economic Review, 67(5), 823–839.

Bibliography    529 Downes, P., & Vaez-Zadeh, R. (1991). The evolving role of central banks. Washington, DC: International Monetary Fund. Drehman, M., Borio, C., Gambacorta, L., Jimenez, G., & Trucharte, C. (2010). Countercyclical capital buffers: exploring options. BIS Working Papers No. 317. Drew, A., & Hunt, B. (1998). The forecasting and policy system: Preparing economic projections. RBNZ Discussion Paper No. G98/7. DTCC. (2015, October). Understanding interconnectedness risks: To build a more resilient financial system. Duisenberg, W. (1999). Monetary policy in the Euro area. Paper presented at new year reception organized by Deutsche Borrse, January. Retrieved from http://www.ecb.int Edwards, S. (2006, April). The relationship between exchange rates and inflation targeting revisited. NBER Working Paper No. 12163. Edwards, S. (2007). The relationship between exchange rates and inflation targeting revisited. In F. S. Mishkin & K. Schmidt-Hebbel (Eds.), Monetary policy under inflation targeting (pp. 373–413). Santiago: Central Bank of Chile. Ehlers, T., & Villar, A. (2015). The role of banks. BIS Papers No. 83. Ehrmann, M., & Fratzscher, M. (2005). The timing of central bank communication. European Central Bank Working Paper Series No. 565. Ehrmann, M., Fratzscher, M., Gurkaynak, R. S., & Swanson, E. T. (2007). Convergence and anchoring of yield curves in the Euro area. ECB Working Paper Series No. 817. Eichenbaum, M. (1992, June). Comments: Interpreting the macroeconomic time series facts: The effects of montary policy. European Economic Review, 36(5), 1001–1011. Eichenbaum, M., & Evans, C. L. (1995). Some empirical evidence on the effects of shocks to monetary policy and exchange rates. Quarterly Journal of Economics, 110(1995), 975–1010. Eichenbaum, M., & Singleton, K. J. (1986). Do equilibrium real business cycle theories explain postwar U.S. business cycles? In Fischer (Ed.), NBER macroeconomics annual. Cambridge, MA: MIT Press. Eichengreen, B. (2004). The challenge of financial instability. Copenhagen Consensus. Cambridge: Cambridge University Press. Eichengreen, B., & Musa, M. (1998). Capital account liberalization: Theoretical and practical aspects. Occassional Paper No. 172. Washington, DC: IMF. Eichengreen, B., et al. (2011, September). Rethinking central banking. Washington, DC: Committee on International Economic Policy and Reform. Eijffinger, S., & Schalling, E. (1993): Central Bank independence in twelve industrial countries, Banca Nazionale del Lavoro Quarterly Review, No. 184, Rome, Banca Nazionale del Lavoro. Eijffinger, S. C. W. (1997, September). The new political economy of central banking. Paper presented at monetary theory as a basis for monetary policy, Italy. Eijffinger, S. C. W., & Hoeberichts, M. M. (2000). Central bank accountability and transparency: Theory and some evidence. Discussion Paper No. 6/00. Economic Research Centre of the Deutsche Bundesbank. Eijffinger, S. C. W., & Hoeberichts, M. M., & Schaling, E. (2000). A theory of central bank accountability. CEPR Discussion Paper No. 2354. Eijffinger, S. C. W., & van Keulen, M. (1995). Central bank independence in another eleven countries. Banca Nazionale del Lavaro Quarterly Review, 192, 39–83. El-Erian, M. A. (2016). The only game in town: Central banks, instability, and avoiding the next collapse. New York, NY: Penguin Random House. Elster, J. (Ed). (1986). Rational choice. Oxford: Blackwell. Elgie, R. (1995). Core executive-central bank relations: Central bank independence: What it is and how to compare it. Unpublished political studies association 1995 annual conference paper, Political Studies Association. Endut, M., Morley, J., & Tien, P. L. (2013, August). The changing transmission mechanism of U.S. monetary policy. New York, NY: Mimeo.

530   Bibliography Engle, C. (1999). On the foreign exchange risk premium in sticky-price general equilibrium models. NBER No. 7067. Retrieved from www.nber.org/papers/w7067 Erce, C. J., & Levin, A. T. (2003). Imperfect credibility and inflation persistence. Journal of Monetary Economics, 50(4), 915–944. Erceg, C., Guerrieri, L., & Gust, C. (2006). SIGMA: A new open economy model for policy analysis. International Journal of Central Banking, 2(1), 1–50. European Banking Authority (EBA). (2015). Risk assessment of the European banking system. Luxembourg: Publications Office of the European Union. European Central Bank (ECB). (2007, November). Payments and monetary and financial stability. ECB-Bank of England joint conference. European Central Bank (ECB). (2010a, May). Monetary policy transmission in the Euro area: A decade after introduction of the Euro. Monthly Bulletin, 85–98. European Central Bank (ECB). (2010b, January). Recent advances in modeling systemic risk using network analysis. Frankfurt: ECB. Evans, G. W., & Honkapohja, S. (2002, February). Monetary policy, expectations and commitment. ECB Working Paper No. 124. Evans, M., & Lyons, R. (1999). Order flow and exchange rate dynamics. NBER 7317. Retrieved from www.nber.org/papers/w7317 Evanoff, D., Kaufman, G., & Malliaris, A. G. (Eds.). (2012). New perspectives on asset price bubbles. New York, NY: Oxford University Press. Fahr, S., Motto, R., Rostagno, M., Smets, F., & Tristani, O. (2013). A monetary policy strategy in good and bad times: Lessons from the recent past. Economic Policy, 28(74), 245–288. Faust, J., & Svensson, L. (2000). The equilibrium degree of transparency and control in monetary policy. CEPR Discussion Paper No. 2195. Faust, J., & Svensson, L. (2001). Transparency and credibility: Monetary policy with unobservable goals. International Economic Review, 42(2), 369–397. Fazaalloh, A. M., & Sasongko. (2014). Whether the bank lending channel can work? Evidence from foreign banks in Indonesia. Review of Integrative Business Econonomic Research, 4(1), 440–451. Felman. (2000). Comments on inflation targeting in Indonesia. In C. Joseph & A. Gunawan (Eds.), Monetary policy and inflation targeting in emerging economies. Jakarta: Bank Indonesia. Feroli, M, Kashyap, A., Schoenholtz, K., & Shin, H. S. (2014, February 28). Market tantrums and monetary policy. Report for the 2014 US Monetary Policy Forum, New York City. Ferreira, L. N., & Nakane, M. I. (2015, November). Macroprudential policy in a DSGE model: Anchoring the countercyclical buffer. Banco Central do Brazil Working Papers No. 407. Filardo, A. J. (1998). New evidence on the output cost of fighting inflation. Federal Reserve Bank of Kansas City Economic Review, 83(3), 33–61. Filardo, A. J. (2001, July). Should monetary policy respond to asset price bubbles? Some experimental results. Federal Reserve of Kansas City No. RWP 01-04. Filardo, A. J., & Hofmann, B. (2014, March). Forward guidance at the zero lower bound. BIS Quarterly Review. Financial Stability Board (FSB). (2009). Report of the financial stability forum on addressing procyclicality in the financial system. Fischer, S. (1977). Long term contracts, rational expectations, and the optimal money supply rule. Journal of Political Economy, 85(1), 191–205. Fischer, S. (1985). Central bank independence revisited. The American Economic Review, 85(2), 201–206. Fischer, S. (1993). The role of macroeconomic factors in growth. Journal of Monetary Economics, 32(3), 485–512.

Bibliography    531 Fischer, S. (1995). Central bank independence revisited. American Economic Review, Papers and Proceedings, 85, 201–206. Fischer, S. (1996). Central banking: The challenges ahead: Maintaining price stability. Finance and Development. Fisher, P., Mahadeva, L., & Whitley, J. (1997). The output gap and inflation: Experience at the Bank of England. Paper in BIS Model Builders Meeting, Basel, January. Fleming, J. M. (1962). Domestic financial policies under fixed and under floating exchange rate. IMF Staff Papers, 9, 369–379. Flood, R. P., & Garber, P. M. (1984). Collapsing exchange rate regimes: Some linear examples. Journal of International Economics, 17(1–2), 1–13. Forbes, K. J., & Warnock, F. E. (2012). Capital flow waves: Surges, stops, flight, and retrenchment. Journal of International Economics, 88(2), 235–251. Frankel, J., Parsley, D., & Wei, S. J. (2005, March). Slow passthrough around the world: A new import for developing countries? NBER Working Paper No. 11199. Frankel, J. (1978). Purchasing power parity: Evidence from the 1920’s. Journal of International Economics, 8(2), 169–191. Frankel, J. (1999). No single currency regime is right for all countries or at all times. NBER No. 7338. Retrieved from http://www.nber.org/papers/w7338 Frankel, J., & Froot, K. A. (1989). Forward discount bias: Is it an exchange risk premium? The Quarterly Journal of Economics, 77(1), 133–153. Frankel, J., & Levich, R. (1975). Covered interest arbitrage: Unexploited profits? Journal of Political Economy, 83(2), 325–338. Fraser, B. W. (1994). Central bank independence: What does it means? Speech on 20th SEANZA Central Banking Course, Karachi, 23 November. Fratzscher, M. (2012). Capital flows, push versus pull factors and the global financial crisis. Journal of International Economics, 88(2), 341–356. Fratzscher, M., Gloede, O., Menkhoff, L., Sarno, L., & Stohr, T. (2015, November). When is foreign exchange intervention effective? Evidence from 33 countries. Discussion Paper No. 1518, DIW Berlin. Freedman, C. (1994). The use of indicators and of the monetary condition index in Canada. In T. J. T. Balino & C. Cottarelli (Eds.), Frameworks for monetary stability: Policy issues and country experience. Washington, DC: IMF. Freedman, C. (2002, July/August). The value of transparency in conducting monetary policy. FRB of St Louis Review. Freeman, R. T., & Willis, J. L. (1995). Targeting inflation in the 1990s: Recent challenges. International Finance Discussion Paper No. 1995-525. Board of Governors of the Federal Reserve System. Freixas, X., Parigi, B., & Rochet, J. C. (2000). Systemic risk, interbank relations and liquidity provision by the central bank. Journal of Money, Credit and Banking, 32(3), 611–638. Friedman, B. M. (1968). The role of monetary policy. American Economic Review, 58, 1–17. Friedman, B. M. (1969). The supply of money and changes in prices and output, 1958, republish on Milton Friedman. In The optimal quantity of money and other essays. Chicago, IL: Aldine Publishing. Friedman, B. M. (1975). Targets, instruments, and indicator of monetary policy. Journal of Moneteray Economics, 14, 443–473. Friedman, B. M., & Hann, F. K. (1990). Handbook of monetary economics (Vols. I and II). New York, NY: Elsevier Science B.V. Friedman, B. M., & Meiselman, D. (1963). The relative stability of monetary velocity and the investment multiplier in the United States, 1897–1958. In Stabilization policies. Englewood Cliffs, NJ: Prentice Hall. Friedman, B. M., & Schwartz, A. (1963). A monetary history of the United States, 1867–1960. Princeton, NJ: Princenton University Press.

532   Bibliography Fry, M. J. (1998). Assessing central bank independence in developing countries: Do actions speak louder than words? Oxford Economic Papers, 50(3), 512–529. Fry, M. J., Goodhart, C. A. E., & Almeida, A. (1996). Central banking in developing countries: Objectives, activities and independence. London: Routledge. Fry, M. J., Julius, D., Mahadeva, L., Roger, S., & Sterne, G. (2000). Key issues in the choice of monetary policy framework. In L. Mahadeva & G. Sterne (Eds.), Monetary frameworks in global context. London: Routledge. Fuertes, A. M., Phylaktis, K., & Yan, C. (2014). Hot money in bank credit flows to emerging markets during the banking globalization era. London: Cass Business School, City University, Mimeograph. Fuhrer, J., & Moore, G. (1995). Inflation persistence. Quarterly Journal of Economics, 110(1), 127–159. Fujiwara, I., Teranishi, Y., & Hara, N. (2004). The Japanese economic model: JEM. Econometric Society, 2004 Far Eastern Meetings 723. Fung, B. (2012). Conference summary: New developments in payments and settlement. Bank of Canada Review, Spring. Galati, G., & Disyatat, P. (2005, March). The effectiveness of foreign exchange market intervention in emerging market countries: Evidence from the Czech Koruna. BIS Working Papers No. 172. Galati, G., & Moessner, R. (2014, September). What do we know about the effects of macroprudential policy? De Nederlandsche Bank NV, Working Paper No. 440. Gali, J. (1992, May). How well the IS-LM model fit the post-war US data? Quarterly Journal of Economics, 107(2), 709–738. Gali, J., & Gertler, M. (1999). Inflation dynamics: A structural econometric analysis. Journal of Monetary Economics, 44(2), 195–222. Gali, J., Gertler, M., & López-Salido, J. D. (2000). European inflation dynamics. European Economic Review, 45(7), 1237–1270. Gambacorta, L., & Marques-Ibanez, D. (2011, May). The bank lending channel: Lessons from the crisis. BIS Working Papers No. 345. Garfinkel, M. R., & Oh, S. (1995). When and how much to talk: Credibility and flexibility in monetary policy with private information. Journal of Monetary Economics, 35(2), 341–357. Gartner, M. (1993). Macroeconomics under flexible exchange rates. Manchester: Harvester Wheatshealf Publication. Gaspar, V., & Russo, D. (2006, November). Evolving payment systems, central bank money and the conduct of monetary policy. Paper presented at 9th annual research conference De Nederlandsche Bank, Amsterdam. Gauvin, L., McLoughlin, C., & Reinhardt, D. (2014, September). Policy uncertainty spillovers to emerging markets: Evidence from capital flows. Bank of England. Working Paper No. 512. Genberg, H. (1978). Purchasing power parity under fixed and flexible exchange rate. Journal of International Economics, 8, 247–276. Georgsson, M., Vredin, A., & Sommar, P. A. (2015). The modern central bank’s mandate and the discussion following the financial crisis. Sveriges Riksbank Economic Review, 1. Geraats, P. M. (2001). Why adopt transparency? The Publications of Central Bank Forecasts. European Central Bank Working Paper No. 41. Geraats, P. M. (2002, March). Central bank transparency. Cambridge: University of Cambridge. Gerali, A., Neri, S., Sessa, L., & Signoretti, F. (2010). Credit and banking in a DSGE model of the Euro area. Working Papers No. 740. Bank of Italy, Economic Research Department. Gerke, R., Weber, A., & Worms, A. (2009). Has the monetary transmission process in the Euro area changed? Evidence based on VAR estimates. BIS Working Papers No. 276.

Bibliography    533 Gerl, A. (2006). Testing the effectiveness of the Czech National Bank’s foreign-exchange interventions. Czech Journal of Economics and Finance, 56, 9–10. Gersbach, H. (1998). On the negative social value of central banks’ transparency. Heidelberg, Germany: University of Heidelberg, Mimeo. Gersbach, H., & Hahn, V. (2000). Should individual voting records of central bankers be published? Heidelberg, Germany: University of Heidelberg, Mimeo. Gertler, M., & Gilchrist, S. (1994). Monetary policy, business cycles, and the behavior of small manufacturing firms. The Quarterly Journal of Economics, 109(2), 309–340. Ghosh, A., Gulde, A., & Wolf, H. C. (2003). Exchange rate regimes: Choices and consequences. Cambridge, MA: The MIT Press. Ghosh, A. R., & Ostry, J. D. (1993). Do capital flows reflect economic fundamentals in developing countries? IMF Working Paper WP/93/34. Ghosh, A. R., & Phillips, S. (1998, December). Warning: Inflation may be harmful to your growth. IMF Staff Papers, 45(4), 672–710. Goeltom, M. S. (2005). Perspectives of Time Consistency and Credibility in Monetary Policy: The Case of Indonesia. Paper for Bank Indonesia International Conference on Marrying Time Consistence in Monetary Policy with Financial Stability. Dalam Essay in Macroeconomic Policy: The Indonesian Experience. Gramedia Pustaka Utama. Jakarta. Gourinchas, P. O., & Jeanne, O. (2013). Capital flows to developing countries: The allocation puzzle. Review of Economic Studies, 80, 1484–1515. Goodfriend, M. (1986). Monetary mystique: Secrecy and central banking. Journal of Monetary Economics, 17(1), 63–92. Goodfriend, M. (2010, August). Central banking in the credit turmoil: An assessment of federal reserve practice. Carnegie Mellon University and National Bureau of Economic Research (NBER), Mimeo. Goodfriend, M., & King, R. G. (1997, January). The new neoclassical synthesis and the role of monetary policy. In B. S. Bernanke & J. Rotemberg (Eds.), NBER macroeconomics annual 1997 (Vol. 12). Cambridge, MA: MIT Press. Goodhart, C. A. E. (2010, November). The changing role of central banks. BIS Working Papers No. 326. Gordon, R. J. (1998). Foundations of the goldilocks economy: Supply shocks and the timevarying NAIRU. Brookings Papers on Economic Activity No. 2. Gordon, R. J., & King, S. R. (1982). The output cost of disinflation in traditional and vector autoregressive models. Brookings Papers on Economic Activity, 13, 205–244. Gray, S., Hoggarth, G., & Place, J. (2001). Introduction to monetary operation (revised 2nd ed.). Handbooks in Central Banking No. 10. England: CCBS Bank of England. Greenspan, A. (2002). Opening remarks in rethinking stabilization policy. Symposium sponsored by The Federal Reserve Bank of Kansas City. Greenwald, B. C., & Stiglitz, J. E. (2003). Towards a new paradigm in monetary economics. Cambridge: Cambridge University Press. Greenwald, B. C., & Stiglitz, J. E. (1993). Financial market imperfections and business cycles. Quarterly Journal of Economics, 108(1), 77–114. Grenville, S. (1997, July). The evolution of monetary policy: From money targets to inflation targets. Conference about monetary policy and inflation targeting, RBA . Grilli, V., Masciandaro, D., & Tabellini, G. (1991). Political and monetary institutions and public financial policies in the industrial countries. Economic Policy, 13, 341–392. Group of Thirty (G-30). (2010). Enhancing financial stability and resilience: Macroprudential policy, tools, and systems for the future. Washington D.C. Group of Thirty. (2015, October). Fundamentals of central banking: Lessons from the crisis. Washington, DC. Gurley, J. G., &. Shaw, E. S. (1960). Money in a theory of finance. Washington, DC.

534   Bibliography Haddad, M., & Pancaro, C. (2010, June). Can real exchange rate undervaluation boost exports and growth in developing countries? Yes, but not for long. Economic Premise, World Bank. Hahm, J.-H., Mishkin, F. S., Shin, H. S., & Shin, K. (2012). Macroprudential policies in open emerging economies. NBER Working Paper Series No. 17780. Haldane, A. G. (2004). Defining monetary and financial stability. Unpublished data. Bank of England, London. Haldane, A. G. (2009, April). Rethinking the financial network. Speech at the Financial Student Association, Amsterdam, Bank of England. Haldane, A. G., & Salmon, C. K. (1995). Three issues in inflation targets. In A. G. Haldane (Ed.), Targeting inflation (pp. 170–201). London: Bank of England. Hall, R. E., & Mankiw, N. G. (1993, August). Nominal income targeting. NBER Working Paper No. 4439. Handa, J. (2000). Monetary economics. London: Routledge. Haan, J. D., Amtenbrink, F., & Eijffinger, S. C. W. (1999). Accountability of central banks: Aspects and quantification. Banca Nazionale del Lavoro Quarterly Review, 52, 209, 169–193. Hann, W. J. (2000). The comovement between output and prices. Journal of Monetary Economics, 46, 3–30. Hannoun, H. (2010). Towards a global financial stability framework. Bank for International settlements. Presented at 45th SEACEN Governors’ Conference, Siem Reap Province, Cambodia, 26-27 February 2010. Hansen, P., & Hodrick, R. J. (1980). Forward exchange rates as optimal predictors of future spot rates: An econometric analysis. Journal of Political Economy, 88(5), 829–853. Harmanta. (2009). Kredibilitas Kebijakan Moneter dan Dampaknya Terhadap Persistensi Inflasi dan Strategi Disinflasi Di Indonesia: Dengan Model Dynamic Stochastic General Equilibrium (DSGE). Disertasi, Universitas Indonesia. Harmanta, T., Bathaluddin, M. B., & Idham. (2012). ARIMBI with macroprudential policy. Jakarta: Economic Research Group Bank Indonesia. Harmanta, H., Purwanto, N. M. A., & Oktiyanto, F. (2015). Internalisasi Sektor Perbankan dalam Model DSGE. Buletin Ekonomi Moneter dan Perbankan, 17(1), 23–59. Harmanta, H., Purwanto, N. M. A., Rachmanto, A., & Oktiyanto, F. (2013). Monetary and macroprudential policy mix under financial frictions mechanism with DSGE model. Jakarta: Bank Indonesia. Harrison, R., Nikolov, K., Quinn, M., Ramsay, G., A. Scott, & Thomas, R. (2005). The Bank of England quarterly model. England: Bank of England. Harun, C. A., Taruna, A. A., Nattan, R. R., & Surjaningsih, N. (2014, December). Financial cycle of Indonesia: Potential forward looking analysis. Bank Indonesia Working Paper No. WP/9/2014. Heijdra, B. J., & Van der Ploeg, F. (2002). The foundations of modern macroeconomics. Oxford: Oxford University Press. Hempell, H. S., & Sørensen, C. K. (2010, November). The impact of supply constraints on bank lending in the Euro area: Crisis-induced crunching? ECB Working Paper No. 1862. Heuvel, V. (2002). The bank capital channel of monetary policy. Philadelphia, PA: University of Pennsylvania, Mimeo. Ho, C., & McCauley, R. N. (2003). Living with flexible exchange rate: Issues and recent experience in inflation targeting emerging market economies. BIS Working Papers No. 130, BIS, Basel. Hong Kong Monetary Authority. (2008). The housing market channel of the monetary transmission mechanism in Hong Kong. BIS Papers No. 35. Bank for International Settlements. Horn, D., Geerts, P., & Coolen, J. (2014). Financial technology taking over. Compact, 2014, 4 Hubbard, R. G. (1994). Money, the financial system, and the economy (3rd ed.). Boston, MA: Addison-Wesley Publishing Co., Inc.

Bibliography    535 Hubbard, R. G. (2002). Money the financial system and the economy (4th ed.). Reading, MA: Addison Wesley Co. Hunt, B. (1999). Inter-forecast monetary policy implementation: Fixed-instrument vs. MCI-based strategies. RBNZ Discussion Paper No. G99/1. Hunter, W. C., Kaufman, G. G., & Pomerleano, M. (2003). Asset price bubbles: Implications for monetary, regulatory, and international policies. Cambridge, MA: MIT Press. Idris, R. Z., Yanuarti, T., Iskandar, C. L., & Darsono. (2001). Asset price channel of monetary policy transmission mechanism in Indonesia. In P. Warjiyo & J. Agung (Eds.), Transmission mechanism of monetary policy in Indonesia. Jakarta: Bank Indonesia. Iljas, A. (1999). Peranan Bank Indonesia dalam Pengendalian Inflasi. Paper presented at Seminar about the role of Bank Indonesia in inflation control, December 6. Indawan, F., Fitriani, S., Permata, M. I., & Karlina, I. (2013, January). Capital flows in Indonesia: The behavior, the role, and its optimality uses for the economy. Bank Indonesia: Bulletin of Monetary, Economics and Banking. International Monetary Fund (IMF). (1999). Code on good practices of transparency in monetary and financial policies. Washington, DC: IMF. International Monetary Fund (IMF). (2001, June). Financial soundness indicators: Policy paper. Monetary and Exchange Affairs Department and the Statistics Department. International Monetary Fund (IMF). (2003). Assessments of the IMF code on good practices of transparency in monetary and financial policies: Review of the experience. Washington, DC. Retrieved from http://www.imf.org/external/np/mae/mft/index.html International Monetary Fund (IMF). (2009). Balance of payments and international investment position manual (6th ed.) (BPM6). Washington, DC: IMF. International Monetary Fund (IMF). (2010, September). The IMF-FSB early warning exercise: Design and methodological toolkit. Washington, DC: IMF. International Monetary Fund (IMF). (2012, November). The liberalization and management of capital flows: An institutional view. Washington, DC: IMF. International Monetary Fund (IMF). (2013a, April). Guidance note for the liberalization and management of capital flows. Washington, DC: IMF. International Monetary Fund (IMF). (2013b, September). The interaction of monetary and macroprudential policies. IMF Policy Paper. Washington, DC: IMF. International Monetary Fund (IMF). (2015a). Measures which are both macroprudential and capital flow management measure: IMF approach. Washington, DC: IMF. International Monetary Fund (IMF). (2015b, December). Managing capital outflows: Further operational considerations. Washington, DC: IMF. International Monetary Fund (IMF). (2015c, September). Monetary policy and financial stability. IMF Policy Paper. Washington, DC: IMF. Institute of International Finance. (2015, October). Capital flows to emerging markets. Ippolito, F., Ozdagliy, A. K., & Perez, A. (2015, June). Transmission of monetary policy through bank lending: The floating rate channel. Boston, MA: Universitat Pompeu Fabra and Febderal Reserve Bank of Boston, Mimeo. Isard, P., & Laxton, D. (2000). Inflation-forecast targeting and the role of macroeconomic models, In Warren C. (Ed.), inflation targeting in transition economies: The case of the Czech Republic (pp. 105–137). Prague: Czech National Bank. Issing, O. (1999). The Eurosystem: Transparent and accountable. Journal of Common Market Studies, 37, 503–520. Issing, O. (2011). Lessons for monetary policy: What should the consensus be? IMF Working Paper No. 11/97. Ito, T. (2010, July). Monetary policy and financial stability: Is inflation targeting Passé? ADB Working Paper Series No. 206 . Ioannidou, V. P., Ongena, S., & Peydro, J. L. (2007, October). The impact of short-term interest rates on risk-taking: Hard evidence. Retrieved from https://voxeu.org/article/ impact-short-term-interest-rates-risk-taking-hard-evidence

536   Bibliography Jadresic, E. (1999, April). Inflation targeting and output stability. IMF Working Paper No. 61. James, E., McLoughlin, K., & Rankin, E. (2014, June). Cross-border capital flows since the global financial crisis. Reserve Bank of Australia Bulletin, 65–72. Jannsen, N., Potjagailo, G., & Wolters, M. H. (2014, September). Monetary policy during financial crises: Is the transmission mechanism impaired? New York, NY: University of Kiel and Kiel Institute for the World Economy, Mimeo. Jensen, H. (2002). Optimal degrees of transparency in monetary policy making. Scandinavian Economic Journal, 104(3), 399–422. Jiménez, G., Ongena, S., Peydró, J.-L., & Saurina, J. (2009). Hazardous times for monetary policy: What do twenty-three million bank loans say about the effects of monetary policy on credit risk-taking? Bank of Spain Working Papers 833. Jimenez, G., Ongena, S., Peydro, J., & Saurina, J. (2012). Credit supply versus demand: Bank and firm balance-sheet channels in good and crisis times. Discussion Paper No. 2012-005. Tilburg, The Netherlands: Tilburg University, Center for Economic Research. Johnson, H. G. (1967). Essays in monetary economics. London: George Allen & Unwin Ltd. Johnson, H. G. (1972). Macroeconomic and monetary theory: Lectures in economics 1. Chicago, IL: Aldine Publ. Co.. Johnston, R. B. (1998). Sequencing capital account liberalization and financial sector reform. Paper on Policy Analysis and Assesment No. PPAA/98/8. Washington, DC: IMF. Johnston, R. B., & Sundarayan, V. (1999). Sequencing financial sector reform: Country experience and issues. Washington, DC: IMF. Johnston, R. B., & Tamirisa, N. T. (1998, December). Why do countries use capital control? IMF Working Paper No. WP/98/181. Jorda, O., Schularick, M., & Taylor, A. M. (2011). Financial crises, credit booms, and external imbalances: 140 years of lessons. IMF Economic Review, 59(2), 340–378. Jorda, O., Schularick, M., & Taylor, A. M. (2014, December). Betting the house. NBER Working Paper 20771. Joseph, C., & Gunawan, A. (Eds.). (2000). Monetary policy and inflation targeting in emergin economies. Jakarta: Bank Indonesia. Ju, J., Wei, S. (2006). A Solution to Two Paradoxes of International Capital Flows (July 2006). IMF Working Paper No. 06/178, pp. 1-39, 2006. Available at SSRN: https:// ssrn.com/abstract=926233 Judd, J. P., Rudebusch, G. (1998). Taylor’s rule and the fed, 1970-1997. Economic Review, 3–16. Juhro, S. M. (2010, November). The vicious circle of rising capital inflows and effectiveness of monetary control in Indonesia. Indonesia: Catatan Riset, Departemen Kebijakan Ekonomi dan Moneter, Bank Indonesia. Juhro, S. M. (2015, Spring). The role of the central bank in promoting sustainable growth: Perspectives on the implementation of flexible ITF in Indonesia. Afro-Eurasian Studies, 4(1). Juhro, S. M., & Goeltom, M. S. (2015, February). Monetary policy regime in Indonesia: Towards a post-GFC framework. In A. Kohsaka (Ed.), Macro-financial linkages in in Pacific region. New York, NY: Routledge. Juhro, S. M., Harmanta, Herdiawan, J., Mochtar, F., & Asih, K. N. (2009). Review penerapan inflation targeting framework di Indonesia. Working paper. Bank Indonesia. Juhro, S. M., & Mochtar, F. (2009, January). Peran Nilai Tukar dalam Flexible ITF: Pemikiran Alternatif mengenai Policy Rules. Indonesia: Catatan Riset, Departemen Kebijakan Ekonomi dan Moneter, Bank Indonesia. Juhro, S. M., et al. (2009, December). Review Penerapan Inflation ITF di Indonesia. Indonesia: Departemen Kebijakan Ekonomi dan Moneter, Bank Indonesia.

Bibliography    537 Judd, J. P., & Trehan, B. (1995). The cyclical behavior of prices: Interpreting the evidence. Journal of Money, Credit, and Banking, 27, 789–797. Kahn, G. A., & Parrish, K. (1998). Conducting monetary policy with inflation targets. Federal Reserve Bank of Kansas City Economic Review, Third Quarter, 83, 5–32. Kakes, J. (2000). Monetary transmission in Europe: The role of financial markets and credit. Cheltenham: Edward Elgar. Kalemli-Ozcan, S., Papaioannou, E., & Peydró, J. L. (2009). What lies beneath the Euro’s effect on financial integration: Currency risk, legal harmonization, or trade? NBER Working Papers No. 15034. Kamin, S., Turner, P., & Van’T Dack, J., (1998). The transmission mechanism of monetary policy in emerging market countries: an overview. BIS Policy Papers, No. 3, 5−64. Retrieved from http://www.bis.org/publ/plcy03.pdf Kaminsky, G. L., & Reinhart, C. M. (1999). The twin crises: the causes of banking and balance of payments problems. American Economic Review, 89(3), 473–500. Kaminsky, G. L., Reinhart, C. M., & Végh, C. A. (2004). When it rains, it pours: Procyclical capital flows and macroeconomic policies. NBER macroecnomics annual 2004. Kashyap, A. K., & Stein, J. C. (2000). What do a million observations on banks say about the transmission of monetary policy? American Economic Review, 90(3), 407–428. Kawai, M., & Morgan, P. J. (2012). Central banking for financial stability in Asia. Asian Development Bank Institute (ADBI) Working Paper No. 377. Keefer, P., & David, S. (1998, August). When does delegation improve credibility? Central bank independence and the separation of powers. Working Paper No. WPS/98-18 . Keynes, J. M. (1923). A tract on monetary reform. London: Macmillan and St. Martin’s Press for the Royal Economics Society. Keynes, J. M. (1930). A treatis on money (Vol. 1). London: Macmillan. Keynes, J. M. (1936). The general theory of employment, interest and money. London: Macmillan. Kharroubi, E., & Zampolli, F. (2015, August). Interest rate pass-through and co-movement in the global financial system: What drives country heterogeneity? Paper presented at Bank Indonesia and BIS Research conference. Kim, C. (2014). Macroprudential policies in Korea – key measures and experiences. Financial Stability Review, Banque de France, 18, 121–130. Kim, S. J., & Wu, E. (2008). Sovereign credit ratings, capital flows and financial sector development in emerging markets. Emerging Markets Review, 9(1), 17–39. Kindleberger, C. P. (1978). Manias, panics, and crashes: A history of financial crises (pp. xii, 271). New York, NY: Basic Books. Kindleberger, C. P., & Aliber, R. Z. (2005). Manias, panics, and crashes: A history of financial crises. Edisi Kelima: John Wiley & Sons. King, M. (1996). How should central banks reduce inflation? Conceptual issues. Federal Reserve Bank of Kansas City Economic Review, 81, 26–52. King, M. (1997). Changes in the UK monetary policy: Rules and discretion in practice. Journal of Monetary Economics, 39(1), 81–97. King, M. (2004). Practical experience with inflation targeting. Paper presented at International Conference Czech National Bank, May. King, M. (2005). Monetary policy: Practice ahead of theory. Speech delivered at The Mais Lecture, Cass Business School, May, London. King, M., & Low, D. (2014). Measuring the ‘world’ real interest rate. NBER Working Papers No. 19887. King, R. G., & Plosser, C. (1984). Money, credit and prices in a real business cycle. American Economic Review, 74(3), 363–380. King, R. G., Plosser, C. I., Stock, J. H., & Watson, M. W. (1991). Stochastic trends and economic fluctuations. American Economic Review, 81, 819–840.

538   Bibliography King, R. G., & Watson, M. W. (1997). Testing long-run neutrality. Economic Quarterly FRB of Richmond, 83(3), 69–101. Kishan, R. P., & Opiela, T. P. (2000). Bank size, bank capital, and the bank lending channel. Journal of Money, Credit and Banking, 32(1), 121–141. Kiyotaki, N., & Moore, J. (1997). Credit cycles. Journal of Political Economy, 105(2), 211–248. Klaus, V. (2002, July/August). The value of transparency in monetary policy: The Czech experience. FRB of St Louis Review. Klein, M. W., & Shambaugh, J. C. (2013). Rounding the corners of the policy trilemma: Sources of monetary policy autonomy. NBER Working Paper No. 19461. Koepke, R. (2014, May). Fed policy expectations and portfolio flows to emerging markets. MPRA. Koepke, R. (2015, April). What drives capital flows to emerging market? A survey of the empirical literature. MPRA Paper No. 62770. Kohlscheen, E., & Rungcharoenkitkul, P. (2015). Changing financial intermediation: Implications for monetary policy transmission. BIS Papers, No. 83. Kohn, D. (2015, October). Implementing macroprudential and monetary policies: The case for two committees. Washington, DC: Brooking Institute. Kokkola, T. (Eds.). (2010). Payments system: Payments, securities, derivatives, and the role of Euro system. European Central Bank. Kouri, P. (1976). The exchange rate and the balance of payments in the short run and in the long run: A monetary approach. Scandinavian Journal of Economics, 78(2), 280–304. Kouri, P., & Porter, R. (1974). International capital flows and portfolio equilibrium. Journal of Political Economy, 82, 443–467. Krugman, P. R. (1979). A model of balance of payments crises. Journal of Money, Credit and Banking. Krugman, P. R. (1991). Target zones and exchange rate dynamics. Quarterly Journal of Economics, 106(August), 669–682. Kuncoro, H. (2015, August). Inflation targeting, exchange rate pass-through, and monetary policy rule in Indonesia. International Journal of Business, Economics and Law, 7(3). Kunter, K., & Janssen, N. (2002). Credibility of monetary regimes: Is inflation targeting different? Paper presented at CCBS Conference on Inflation and Monetary Regimes, January. Kumhof, M. (2000). A quantitative exploration of the role of short-term domestic debt in balance of payments crises. Journal of International Economics, 51(1), 195–215. Kusmiarso, B., Sukowati, E., Prasmuko, A., Pambudi, S., Angkoro D., & Hafid, I. S. (2002). Interest rate channel of monetary transmission in Indonesia. In P. Warjiyo & J. Agung (Eds.), Transmission mechanism of monetary policy in Indonesia. Jakarta: Bank Indonesia. Kuttner, K. N., & Mosser, P. C. (2002, May). The monetary transmission mechanism in the United States: Some answers and further questions. Federal Reseve Bank of New York Economic Policy Review. Kuttner, K. N., & Posen, A. S. (1999, October). Does talk matter after all? Inflation targeting and central bank behavior. Federal Reserve Bank of New York Staff Report, No. 88. Kydland, F. E., & Prescott, E. C. (1977). Rules rather than discretion: Time inconsistency of optimal plans. Journal of Political Economy, 85(1), 473–491. Kydland, F. E., & Prescott, E. C. (1982). Time to build and aggregate fluctuations. Econometrica, 50, 1345–1370. Kydland, F. E., & Prescott, E. C. (1990). Business cycle: Real facts and a monetary myth. FRB of Minneapolis Quarterly Review, 14(1), 3–18. Landerretche, O, Corbo, V., & Schmidt-Hebbel, K. (2001, September). Does inflation targeting make a difference? Bank of Chile Working Paper No. 106.

Bibliography    539 Landier, A., Sraer, D., & Thesmar, D. (2011). The risk-shifting hypothesis: Evidence from sub-prime originations. Presented at The 12th IMF Jacques Polak annual research conference. Lange, O. (1942). Say’s law: A restatement and criticm. In O. Lange, F. McIntyre, & T. O. Yntema (Eds.), Mathemetical economics and econometrics. Chicago, IL: University of Chicago Press. Laubach, T., & Posen, A. S. (1997, April). Some comparative evidence on the effectiveness of inflation targeting. Research Paper 9714. Federal Reserve Bank of New York, New York. Lavigne, R., Sarker, S., & Vasishtha, G. (2014). Spillover effects of quantitative easing on emerging-market. Bank of Canada Review, Autumn, 23–33. Laxton, D., Meredith, G., & Rose, D. (1995). Asymmetric effects of economic activity on inflation: Evidence and policy implications. IMF Staff Papers, 42(2), 344–374. Lee, M., Asuncion, R. C., Kim, J. (2015). Effectiveness of macroprudential policies in developing Asia: An empirical analysis. Emerging Markets Finance and Trade, 52(4), 923–937. Leeper, E. M., Sims, C. A., & Zha, T. (1996). What does monetary policy do? Brookings Papers on Economic Activity, 27(2), 1–78. Lees, K. (2009). Introducing KITT: The Reserve Bank of New Zealand new DSGE model for forecasting and policy design. Reserve Bank of New Zealand Bulletin, 72, 5–20. Leiderman, L., Bar-or, H. (2000). Monetary policy rules and transmission mechanisms under inflation targeting in Israel. Bank of Israel. Leiderman, L., & Svensson, L. E. O. (1995). Inflation targets. London: Centre for Economic Policy Research. Leone, A. M. (1993). Institutional and operational aspect of central bank losses. Papers on Policy Analysis and Assessments No. PPAA/93/14. IMF, Washington, DC. Levi, M. (1990). International finance (2nd ed.). New York, NY: Mac Graw-Hill Publishing Company. Levis, M., Muradoglo, G., & Vasileva, K. (2007). Near home bias in foreign direct investments. New York, NY: Mimeo. Levy-Yeyati, E., & Sturzenegger, F. (2002). Classifying exchange rate regimes: Deeds vs. words. Working Paper, Universidad Torcuato Di Tella. Retrieved from http://www. utdt.edu/∼fsturzen Lim, C., Columba, F., Costa, A., Kongsamut, P., Otani, A., Saiyid, M., …, Wu, X. (2011, October). Macroprudential policy: What instruments and how to use them? IMF Working Paper No. WP/11/238. Loayza, N., & Soto, R. (2001, November). Ten years of inflation targeting: Design, performance, challenges. Bank of Chile Working Paper No. 131. Lohman, S. (1992). Optimal commitment in monetary policy: Credibility versus flexibility. American Economic Review, 82(1), 273–286. Long, J., & Plosser, C. (1983). Real business cycles. Journal of Political Economy, 91, 39–69. Loutskina, E., & Strahan, P. E. (2006). Securitisation and the declining impact of bank finance on loan supply: Evidence from mortgage acceptance rates. NBER Working Papers No. 11983. Lucas, R. E., Jr. (1972, April). Expectations and the neutrality of money. Journal of Economic Theory, 4(2), 103–124. Lucas, R. E., Jr. (1973, June). Some international evidence on output-inflation tradeoffs. American Economic Review, 63(3), 326–334. Lucas, R. E., Jr. (1977). Understanding business cycles. In K. Brunner & A. H. Meltzer (Eds.), Stabilization of the domestic and international economy. Carnegie-Rochester Conference Series on Public Policy, Vol. 5. North-Holland. Lucas, R. E., Jr. (1980). Methods and problems in business cycle theory. Journal of Money, Credit, and Banking, 12, 696–715.

540   Bibliography Lucas, R. E., Jr. (1990). Why doesn’t capital flow from rich to poor countries? American Economic Review, 80(2), 92–96. Mankiw, N. G. (1989). Real business cycles: A new Keynesian perspective. Journal of Economic Perspectives, 3, 79–90. Macklem, T., & Srour, G. (2000, July). Monetary policy rules in an inflation targeting framework lessons from Canada. In BI-IMF conference on monetary policy and inflation targeting in emerging economies. Maddaloni, A., & Peydró, J. L. (2010, March). Bank lending standards and the origins and implications of the current banking crisis. ECB Research Bulletin, 9, 6–9. Maddaloni, A., & Peydró, J. L. (2011). Bank risk-taking, securitisation, supervision and low interest rates: Evidence from lending standards. Review of Financial Studies, 24, 2121–2165. Mahadeva, L., & Sterne, G. (2002). The role of short-run inflation targets and forecasts in disinflation. Bank of England Working Paper No. 167. Markovic, B. (2006). Bank capital channels in the monetary transmission mechanism. Bank of England Working Papers No. 313. Masciandro, D., & Spinelli, F. (1994). Central banks independence: Institutional determinants, rankings and central bankers’ views. Scottish Journal of Political Economy, 41(4), 434–443. Masson, P., Savastano, M., & Sharma, S. F. (1997, October). The scope for inflation targeting in developing countries. IMF Working Paper No. 97/130. Mboweni, T. T. (2000). Central bank independence. Speech delivered at The Reuters Forum Lecture, Johannesburg, 11 October. Retrieved from http://www.stlouisfed.org/news/ speeches/1999/11_04_99.html McCallum, B. T. (1988). Robustness properties of a rule for monetary policy. CarnegieRochester Conference Series on Public Policy, Autumn. McCallum, B. T. (1994, April). Monetary policy rules and financial stability. NBER Working Paper No. 4692. McCallum, B. T. (1995). Choice of target for monetary policy. Economic Affairs, 15(4). McCallum, B. T. (1997). Crucial issues concerning central bank independence. Journal of Monetary Economics, 39(1), 99–112. McCallum, B. T. (2000). Alternative monetary policy rules: A comparison with historical settings for The United States, The United Kingdom, and Japan. NBER Working Paper No. 7725. McCallum, B. T. (2000a, September). The present and future of monetary policy rules. NBER Working Paper No. 7916. McCallum, B. T. (2000b, June). Alternative monetary policy rules: A comparison with historical settings for the United States, The United Kingdom, and Japan. NBER Working Paper No. 7725. McCallum, B. T. (2001, April). Should monetary policy respond strongly to output gaps? NBER Working Paper No. 8226. McCallum, B. T., & Nelson, E. (1999). Performance of operational policy rules in an estimated semi-classical structural model. In J. B. Taylor (Ed.), Monetary policy rules. University of Chicago Press for NBER, Chicago, IL. McCauley, R., McGuire, P., & Sushko, V. (2015). Global dollar credit: Links to US monetary policy and leverage. Economic Policy, 30, 187–229. McCauley, R. N. (2001). Setting monetary policy in East Asia: Goals, developments and institutions. In Future directions for monetary policies in East Asia (pp. 7–55). Sydney: Reserve Bank of Australia. McGuire, P., & von Peter, G. (2009). The US dollar shortage in global banking and the international response. BIS Working Papers No. 291. Meese, R. A., & Rogoff, K. (1983). Empirical models of exchange rate: Do they fit out of sample. Journal of International Economics, 14(1–2), 3–24.

Bibliography    541 Mehrling, P. (2015). Why does central banking need to be re-imagined? BIS Papers No. 79. Meltzer, A. H. (2003). A history of the Federal Reserve (Vol. 1, pp. 1913–1951). Chicago, IL: University of Chicago Press. Mendoza, E. G., & Terrones, M. E. (2008, July). An anatomy of credit booms: Evidence from macro aggregates and micro data. International Finance Discussion Papers Number 936. Board of Governors of the Federal Reserve System . Mendoza, E. G., & Uribe, M. (1999). The business cycles of balance-of-payment crises: A revision of a mundellian framework. NBER 7045. Menon, J. (1995). Exchange rate pass-through. Journal of Economic Surveys, 9, 197–231. Merrouche, O., & Nier, E. (2010). What caused the global financial crisis? Evidence on the drivers of financial imbalances 1999–2007. IMF Working Paper No. 10, 265. Metzler, A. H. (1960, November). Merchantile credit, monetary policy and size of firms. Review Economic Statistics, 42(4), 429–437. Meyersson, P., & Karlberg, P. P. (2012). A journey in communication: The case of the sveriges riksbank. SNS Förlag. Meyer, L. H. (2000). The politics of monetary policy: Balancing independence and accountability. Speech delivered at The University of Wisconsin, LaCrosse, Wisconsin, 24 October. Retrieved from http://www.federalreserve.gov/boarddocs/ speeches/2000/20001024.htm Minoiu, C., & Reyes, J. A. (2011, April). A network analysis of global banking: 1978–2009. IMF Working Paper No. WP/11/74. Minsky, H. P. (1982). The financial instability hypothesis. Levy Economics Institute Working Paper No. 74. Miranda-Agrippino, S., & Rey, H. (2013). World asset markets and global liquidity. London: London Business School, Mimeo. Mishkin, F. S. (1996a). Understanding financial crises: A developing country perspective. NBER Working Paper No. 5600. Mishkin, F. S. (1996b, February). The channels of monetary transmission: Lessons for monetary policy. NBER Working Paper No. 5464. Mishkin, F. S. (1999a). International experiences with different monetary policy regimes. Journal of Monetary Economics, 43, 579–605. Mishkin, F. S. (1999b). Global financial stability: Framework, events, issues. Journal of Economic Perspectives, 13(Fall), 3–20. Mishkin, F. S. (2002, July/August). Commentary. FRB St Louis Review. Mishkin, F. S. (2004). Can inflation targeting work in emerging market countries? Presented at conference tributed to Guillermo Calvo, April 15 and 16, di International Monetary Fund, Washington, DC. Mishkin, F. S. (2007). Housing and the monetary transmission mechanism. Finance and Economics Discussion Series, 2007-40, Federal Reserve Board. Mishkin, F. S. (2011, February). Monetary policy strategy. NBER Working Paper No. 16755. Mishkin, F. S., & Posen, A. S. (1997, August). Inflation targeting: Lessons from four countries. Federal Reserve Bank of New York Economic Policy Review, 9–117. Mishkin, F. S., & Savastano, M. A. (2000, March). What should central banks do? NBER Working Paper. Mishkin, F. S., & Savastano, M. A. (2001, June). Monetary strategies for emerging markets countries: Case studies from Latin America. Paper prepared for the 7th Dubrovnik Economic Conference, Current issues in emerging market economies, Croatia. Mishkin, F. S., & Schmidt-Hebbel, K. (2001, July). One decade of inflation targeting in the world: What do we know and what do we need to know? Central Bank of Chile Working Paper No. 101. Miyajima, K., & Montoro, C. (2013, October). Impact of foreign exchange interventions on exchange rate expectations. BIS Paper No. 73.

542   Bibliography Miyajima, K., Mohanty, M., & Chan, T. (2015). Emerging market local currency bonds: Diversification and stability. Emerging Markets Review, 22, 126–139. Miyajima, K., Mohanty, M., & Yetman, J. (2014). Spillovers of US unconventional monetary policy to Asia: The role of long-term interest rates. BIS Working Papers No. 478. Mochtar, F. (2003, June). SBI, T-Bills dan Pengendalian Inflasi. Working Paper No. DKM . Mohan, R. (2005, October). Communications in central banks: A perspective. Reserve Bank of India Bulletin. Mohanty, M. S., & Berger, B. (2015). Central bank views on foreign exchange intervention. BIS Papers No. 73. Mohanty, M. S., & Klau, M. (2004). Monetary policy rules in emerging market economies: Issues and evidence. BIS Working Papers No. 149. Mohanty, M. S., & Rishabh, K. (2016, March). Financial intermediation and monetary policy transmission in EMEs: What has changed post-2008 crisis? BIS Working Papers No. 546 . Mohanty, M. S., & Turner, P. (2008, January). Monetary policy transmission in emerging market economies: What is new? BIS Papers No. 35. Monetary Authority of Singapore. (2001, February). Singapore’s exchange rate policy. Retrieved from http://www.mas.gov.sg Monetary Authority of Singapore. (2005, October). Assessing market response to MAS’ monetary policy statements. Macroeconomic Review. Moreno, R. (2005, May). Motives for intervention. BIS Papers No. 24. Morris, S., & Shin, H. S. (2004). Liquidity black holes. Review of Finance, 8(1), 1–18. Muchlinski, E. (2002). Against rigid rules: Keynes’s economic theory. Discussion Paper No. XXXX, Frein Universitat Berlin. Muelgini, Y. (2004). Pemetaan Mekanisme Transmisi Kebijakan Moneter di Indonesia. Disertasi Doktor, Tidak dipublikasikan, Fakultas Ekonomi Universitas Indonesia. Mundell, R. A. (1963). Capital mobility and stabilization policy under fixed and flexible exchange rate. Canadian Journal of Economics and Political Sciences, 29(4), 475–485. Murchison, S., & Rennison, A. (2006). ToTEM: The Bank of Canada’s new quarterly projection model. Bank of Canada Technical Reports No. 97. Mussa, M. (1976). The exchange rate, the balance of payments and monetary and fiscal policy under a regime of controlled floating. Scandinavian Journal Economics, 78(2), 229–248. Natsir, M. (2008). Peranan Jalur Suku Bunga Dalam Mekanisme Transmisi Kebijakan Moneter di Indonesia. New York, NY: Mimeo. Neely, C. J. (2001, May/June). The practice of central bank intervention: Looking under the hood. Federal Reserve Bank of St. Louis Review. Neely, C. J. (2005, November/December). An analysis of recent studies of the effect of foreign exchange intervention. Federal Reserve Bank of St. Louis Review. Neumann, M. J. M., & von Hagen, J. (2002, July/August). Does inflation targeting matter? FRB St Louis Review. Nicita, A. (2013). Exchange rates, international trade and policies. Policy Issues Series No. 56. UNCTAD, Geneva. Nijathaworn, B. (2009). Rethinking procyclicality – what is it now and what can be done?. Presented at the Bank for International Settlements/Financial Stability InstituteExecutives’ Meeting of East Asia-Pacific Central Banks (BIS/FSI-EMEAP) High Level Meeting on “Lessons Learned from the Financial Crisis – An International and Asian Perspective”, Tokyo, 30 November 2009. Nijathaworn, B. (2010). Macroprudential policies and capital flows - managing under the new globalization. Conference on macroprudential policies, Shanghai October 18, 2010. Nolan, C., & Schaling, E. (1996, October). Monetary policy uncertainty and central bank accountability. Bank of England Working Paper No. 54.

Bibliography    543 Norden, L., Buston, C. S., & Wagner, W. (2012). Financial innovation and bank behavior: Evidence from credit markets. Paper Presented at the sixth annual risk management conference, Singapore. Nurhidayah, L., Kaluge, D., & Pratomo, D. S. (2014). The role of BI rate and exchange rate affect inflation and economic growth in Indonesia (before and after the global financial crisis). Journal of Basic and Applied Scientific Research, 4(6), 249–260. Obstfeld, M. (1986, March). Rational and self-fulfiling balance of payments crises. American Economic Review, 76(1), 72–81. Obstfeld, M. (2015). Trilemmas and trade-offs: Living with financial globalisation. BIS Working Papers No. 480, January. Obstfeld, M., & Rogoff, K. (1995). The mirage of fixed exchange rates. Journal of Economic Perspectives, 9(Fall), 73–96. Obstfeld, M., & Rogoff, K. (2010). Global imbalances and the financial crises: Product of common causes. In Federal Reserve Bank of San Fransisco’s Asia Economic Policy conference on asia and the global financial crisis, San Fransisco. Obstfeld, M., Shambaughy, J. C., & Taylor, A. M. (2004). The trilemma in history: Tradeoff among exchange rates, monetary policies, and capital mobility. Paper No. C04-133. Center for International and Development Economics Research, University of California, Berkeley, CA. Obstfeld, M., Shambaugh, J. C., & Taylor, A. M. (2005). The trilemma in history: Tradeoffs among exchange rates, monetary policies and capital mobility. Review of Economics and Statistics, 87(3), 423–438. Onafowora, O. (2013). Exchange rate and trade balance in East Asia: Is there a J−curve? Economics Bulletin, 5(18), 1–13. Okun, A. M. (1962). Potential GNP: Its measurement and significance. American statistical association proceedings of the business and economic statistics section, pp. 98–104. Okun, A. M. (1978, May). Efficient disinflationary policies. American Economic Review, 68(2), 348–352. Orphanides, A. (2011, November). New paradigms in central banking? Central Bank of Cyprus Working Paper No. 2011-6. Orphanides, A., & Wieland, V. (1998, June). Price stability and monetary policy effectiveness when nominal interest rates are bounded at zero. Discussion Series 98-35. Board of Governors of the Federal Reserve System. Ostry, J. D., Ghosh, A. R., & Chamon, M. (2012a). Two targets, two instruments: Monetary and exchange rate policies in emerging market economies. IMF Staff Discussion Note No. SDN/12/01. Ostry, J. D., Ghosh, A. R., & Chamon, M. (2012b, May 27). On inflation targeting and forex intervention: Are two targets better than one?. Ostry, J. D., Ghosh, A. R., Habermeier, K., Laven, L., Chamon, M., Qureshi, M. S., & Kokenyne, A. (2011, April). Managing capital inflows: What tools to use? IMF Staff Discussion Note No. SDN/11/06. Padoa-Schioppa, T. (2003). Central banks and financial stability: Exploring the land in between. In V. Gaspar, P. Hartmann, & O. Sleijpen (Eds.), The transmission of the European financial system. Frankfurt: European Central Bank. Palley, T. I. (2009). Rethinking the economics of capital mobility and capital controls. Brazilian Journal of political Economy, 29(July–September), 15–34. Pandit, B. L., Mittal, A., Roy, M., & Ghosh, S. (2006, January). Transmission of monetary policy and the bank lending channel: Analysis and evidence for India. Development Research Group Study No. 25. Reserve Bank of India. Pariès, M. D., & Moyen, S. (2009). Monetary policy and inflationary shocks under imperfect credibility. Working Paper Series No. 1065. ECB.

544   Bibliography Park, S. J. (2006). Asset prices and monetary policy: Korean experience. Paper presented at The BIS Autumn Economists’ Meeting, October. Parkin, M. (1987). Domestic monetary institutions and deficits. In J. M. Buchanan, C. K. Rowley, & R. D. Tollison (Eds.), Deficits. New York, NY: Blackwell. Parrado, E. (2004, February). Inflation targeting and exchange rate rules in an open economy. IMF Working Paper No. WP/04/21. Penaloza, D. S. (2011, December). In the quest of macroprudential policy tools. Bank of Mexico. Pencegahan dan Penanganan Krisis Sistem Keuangan (PPKSK). (2016). Undang-Undang Republik Indonesia No. 9 Tahun 2016. Retrieved from https://jdih.kemenkeu.go.id/ fullText/2016/9TAHUN2016UU.pdf Perrier, P., & Amano, R. (2000). Credibility and monetary policy. Bank of Canada Review, 11–17. Persson, T., & Tabellini, G. (1993). Designing institutions for monetary stability. CarnegieRochester Conference Series on Public Policy, 39, 53–84. Phillips, A. W. (1958). The relation between unemployment and the rate of change of money wage rates in the United Kingdom, 1861–1957. Econometrica, 25, 283–299. Poole, W. (1970). Optimal choice of monetary policy instruments in a simple stochastic macro model. Quarterly Journal of Economics, 84(2), 197–216. Poole, W. (1999). Central bank transparency: Why and how. Speech delivered at University of Missouri, Columbia, 4 November. Retrieved from http://www.stlouisfed.org/ news/speeches/1999/11_04_99.html Poole, W. (2001). Central bank transparency: why and how?. Speech 51, Federal Reserve Bank of St. Louis. Poole, W. (2003, November/December). Fed transparency: How, not whether. FRB of St Louis Review. Portes, R., & Rey, H. (2005). The determinants of cross-border equity flows. Journal of International Economics, 65(2), 269–296. Posen, A. (2002, July/August). Commentary. FRB St Louis Review. Prasad, E., Rajan, R., & Subramanian, A. (2006, August). Foreign capital and economic growth. Washington, DC: Research Department IMF . Prescott, E. (1986). Theory ahead of bussiness cycle measurement. FRB of Minneapolis Staff Report, p. 102. Price Waterhouse and Cooper (PWC). (2016, March). Blurred lines: How FinTech is shaping financial services. Global FinTech Report. Purnawan, M. E., & Nasir, M. A. (2015). The role of macroprudential policy to manage exchange rate volatility, excess banking liquidity and credits. Bulletin of Monetary, Economics and Banking, 18(1), July. Rahmahdian, R., & Warjiyo, P. (2013). Measuring the time inconsistency of monetary policy in Indonesia. Bulletin of Monetary Economics and Banking, 15(4). Rai, V., & Suchanek, L. (2014, November). The effect of the Federal Reserve’s tapering announcements on emerging markets. Bank of Canada Working Paper No. 2014-50. Rajan, R. (2005). Has financial development made the world riskier? Paper presented at The Greenspan Era: Lessons For the Future, Federal Reserve Bank of Kansas City, Jackson Hole, Wyoming, 25–27 August. Rajan, R. (2014, April 10). Competitive monetary easing: Is it yesterday? Speech delivered at Brookings Institution, Washington DC. Raz, A. F., Indra, T. P. K., Artikasih, D. K., & Citra, S. (2012, October). Global financial crises and economic growth: Evidence from East Asian Economies. Bulletin of Monetary, Economics and Banking, 15(2), 1–20. Razin, A., Rubenstein, Y., & Sadka, E. (2003). Which countries export FDI, and how much? Cornell University.

Bibliography    545 Reinhart, C. M., & Rogoff, K. S. (2004). The modern history of exchange rate arrangements: A reinterpretation. Quarterly Journal of Economics, 119(February), 1–48. Reinhart, C. M., & Rogoff, K. S. (2009). This time is different: Eight centuries of financial folly. Princenton, NJ: Princenton University Press. Revenna F. (2005). Inflation Targeting with Limited Policy Credibility. University of California- Santa Cruz April. Rey, H. (2013). Dilemma not trilemma: The global financial cycle and monetary policy independence. Proceedings of the Federal Reserve Bank of Kansas City Economic Symposium at Jackson Hole, August. Ricardo, D. (1871). Plan for the Establishment of a National Bank (1824). Reprinted in J.R. McCulloch (ed.), The Works of David Ricardo - With a Notice of the Life and Writings of the Author. London: John Murray. Roberds, W., & Velde, F. R. (2014, August). Early public banks. Federal Reserve Bank of Atlanta Working Paper No. 2014-9. Robert, B., Withborg, C., & Willen, T. D. (1992). A monetary constitution case for an independent European central bank. The World Economy, 15(2). Rodrik, D. (2008). The real exchange rate and economic growth. Brookings Papers on Economic Activity, Fall. Reinhart, C. M., Rogoff, K. S. (2009a). The Aftermath of Financial Crises. American Economic Review, 99(2): 466–72. Roger, S. (1998). Core inflation: concepts, uses and measurement. Reserve Bank of New Zealand Discussion Paper No. G98/9. Roger, S., & Stone, M. (2005). On target? The international experience with achieving inflation targets. IMF Working Paper No. WP/05/163. Roger, S., & Vleck, J. (2012, January). Macofinancial modeling at central banks: Recent developments and future directions. IMF Working Paper No. WP/12/21. Rogoff, K. (1985). The optimal degree of commitment to an intermediate monetary target. The Quarterly Journal of Economics, 100, 1169–1189. Rogoff, K. S., Husain, A. M., Mody, A., Brooks, R., & Oomes, N. (2003). Evolution and performance of exchange rate regimes. IMF Working Paper No. WP/03/243. Roll, R. (1979). Violation of PPP and their implication for efficient commodity market. In M. Sarnat & G. Szego (Eds.), International finance and trade. Cambridge, MA: Ballinger. Romer, D. H. (1996). Advanced macroeconomics. New York, NY: McGraw-Hill. Romer, D. H., & Romer, C. D. (1996). Federal Reserve private information and the behavior of interest rates. NBER Working Paper No. 5692. Rose, A. K. (1996). Explaining exchange rate volatility: An empirical analysis of ‘the holy trinity’ of monetary independence, fixed exchange rates, and capital mobility. Journal of International Money and Finance, 15(6), 925–945. Rose, A. K. (2007). A stable international mmonetary system emerges: Inflation targeting is bretton woods, Reversed. Journal of International Money and Finance, 26(5), 663–681. Rotemberg, J. J. (1996). Prices, output, and hours: An empirical analysis based on a sticky price model. Journal of Monetary Economics, 37, 505–534. Rotemberg, J., & Woodford, M. (1997). An optimization-based econometric framework for the evaluation of monetary policy. NBER Macroeconomics Annual 1997, 12, 297–361. Rothschild, M., & Stiglitz, J. E. (1976). Equillibrium in competitive insurance markets: An essay on the economics of imperfect information. Quarterly Journal of Economics, 90(4), 630–649. Rudebusch, G. D., & Svensson, L. E. O. (1998, April). Policy rules for inflation targeting. NBER Working Paper No. 6512. Sa, S. (2006, December). Capital flows and credit booms in emerging market economies. Banque de France, Financial Stability Review, 9, 49–66.

546   Bibliography Sachs, A. (2010). Completeness, interconnectedness and distribution of interbank exposures: A parameterised analysis of the stability of financial networks. Frankfurt, Germany: Deutsche Bundesbank. Samuelson, P. A. (1959). An exact consumption-loan model of interest with or without the social contrivance of money. Journal of Political Economy, 66(6), 467–482. Sarel, M. (1996). Nonlinear effects of inflation on economic growth. International Monetary Funds, 43(1), 199–215. Sargent, T. J. (1973). Rational expectations, the real rate of interest, and the natural rate of unemployment. Brokings Papers on Economic Activity, 2, 429–472. Sargent, T. J. (1983). Stopping moderate inflations: The methods of Poincare and Thatcher. In R. Dornbusch & M. H. Simonsen (Eds.), Inflation, debt and indexation. Cambridge, MA: MIT Press. Sargent, T. J. (1987). Dynamic macroeconomic theory. Cambridge, MA: Harvard University Press. Sargent, T. J., & Wallace, N. (1975). Rational expectations, the optimal monetary instrument, and the optimal money supply rule. Journal of Political Economy, 83(2), 241–254. Sarno, L., & Taylor, M. P. (2001). Official intervention in the foreign exchange market: Is it effective and, if so, how does it work? Journal of Economic Literature, 39(September), 839–868. Sarwono, H. A. (1996). Seeking a new monetary management paradigm in Indonesia. Makalah SESPIBI Angkatan XXI. Sarwono, H. A., & Warjiyo, P. (1998). Mencari paradigma baru manajemen moneter dalam sistem nilai tukar fleksibel: Suatu pemikiran untuk penerapannya di Indonesia. Buletin Ekonomi Moneter dan Perbankan, 1(1), 5–23. Satria, D., & Juhro, S. M. (2011, January). Risk behavior in the transmission mechanism of monetary policy in Indonesia. Bulletin of Monetary, Economics and Banking, 13(3), 251–280. Sbordone, A. M. (1999). Prices and unit labour costs: A new test of price stickiness. Newark, NJ: Rutgers University, Mimeo. Schaechter, A., Stone, M. R., & Zelmer, M. (2000, December). Adopting inflation targeting: Practical issues for emerging market countries. Ocassional Paper No. 202. International Monetary Fund. Schaling, E. (1999). The non-linear Phillips curve and inflation forecast targeting. London: Bank of England. Scherbina, A. (2013, February). Asset price bubbles: A selective survey. IMF Working Paper No. 13/15. Schinasi, G. J. (2004, October). Defining financial stability. IMF Working Paper No. WP/04/187. Schmidt-Hebbel, K., & Tapia, M. (2002, June). Monetary policy implementation and results in twenty inflation-targeting countries. Central Bank of Chile Working Paper No. 166. Schmidt-Hebbel, K., & Werner, A. (2002). Inflation targeting in Brazil, Chile, and Mexico: Performance, credibility, and exchange rate. Central Bank of Chile Working Paper. Schularick, M., Taylor, A. M. (2012). Credit booms gone bust: Monetary policy, leverage cycles, and financial crises 1870-2008. American Economic Review, 102(2), 1029– 1061. Schwartz, A. (2002, November). Asset price inflation and monetary policy. NBER Working Paper No. 9321. Schwartz. A. J. (1995). Why financial stability depends on price stability. Economic Affairs, 15(4). Sellon, G. H., & Weiner, S. E. (1996). Monetary policy without reserve requirements: Analytical issues. Economic Review. Sellon, G. H., & Weiner, S. E. (1997). Monetary policy without reserve requirements: Case studies and options for the United States. Economic Review.

Bibliography    547 Sheppard, D. (1996). Payment systems. In Handbooks in central banking No. 8. Centre for Central Banking Studies, Bank of England. Shirakawa, M. (2012). Central banking: Before, during, and after the crisis. Remarks at A conference sponsored by the federal reserve board and the international journal of central banking. Sidrauski, M. (1967, May). Rational choice and patterns of growth in a monetary economy. American Economic Review, 57(2), 534–544. Siklos, P. L. (1999, March/April). Inflation-target design: Changing inflation performance and persistence in industrial countries. Federal Reserve Bank of St. Louis Review, 81(2), 46–58. Siklos, P. L., & Bohl, M. T. (2004). Do action speak louder than words?: Evaluating monetary policy at the Bundesbank. Research Paper. Silalahi, T., Wibowo, W. A., & Nurliana, L. (2012, October). Impact of global financial shock to international bank lending in Indonesia. Bulletin of Monetary, Economics and Banking. Sims, C. A. (1972, September). Money, income and causality. American Economic Review, 62(4), 540–552. Sims, C. A. (1980, May). Comparison of interwar and postwar bussiness cycles. American Economic Review, 70(2), 250–257. Sims, C. A. (1992, June). Interpreting the macroeconomic time series facts: The effects of montary policy. European Economic Review, 36(5), 975–1000. Sinclair, P. J. N. (2000, November). Central banks and financial stability. Quarterly Bulletin. Siregar, R. Y., & Goo, S. (2008). Inflation targeting policy: The experience of Indonesia and Thailand. Centre for Applied Macroeconomic Analysis, The Australian National University. Working Paper No. 23/2008. Retrieved from http://cama.anu.edu.au Siregar, R. Y., & Goo, S. (2009, September). Effectiveness and commitment to inflation targeting policy: Evidences from Indonesia and Thailand. Munich Personal RePEc Archive (MPRA). Siswanto, B., Kurniati, Y., Gunawan, & Binhadi, S. H. (2002). Exchange rate channel of monetary transmission in Indonesia. In P. Warjiyo & J. Agung (Eds.), Transmission mechanism of monetary policy in Indonesia. Jakarta: Bank Indonesia. Sitorus, T. (2000). Towards the implementation of inflation targeting in Indonesia: A review of operational issues. In C. Joseph & A. Gunawan (Eds.), Monetary policy and inflation targeting in emergin economies. Jakarta: Bank Indonesia. Smets, F. (2013). Financial stability and monetary policy: How closely interlinked? Sveriges Riksbank Economic Review, 3(Special Issue), 121–160. Smets, F. (2014). Financial stability and monetary policy: How closely interlinked? International Journal of Central Banking, 10(2), 263–300. Sobrun, J., & Turner, P. (2015a). Bond markets and monetary policy dilemmas for emerging market economies. BIS Working Papers No. 508. Sobrun, J., & Turner, P. (2015b). Low long-term rates as a global phenomenon. New York, NY: Mimeo. Solikin. (1998). The stability of income velocity, demand for money and money multiplier in Indonesia. Working Paper at the University of Michigan. Unpublished data. Solikin. (2004a, March). Kurva Phillips dan Perubahan Struktural di Indonesia: Keberadaan, Pembentukan Ekspektasi, dan Non-Linearitas. Buletin Ekonomi Moneter dan Perbankan (BEMP). Bank Indonesia. Solikin. (2004b, November). Respons Kebijakan Moneter yang Optimal di Indonesia: The State-Contingent Rule? Bank Indonesia Working Paper. The Center for Central Banking Studies (PPSK – BI). Solikin. (2005, June). Fluktuasi Makroekonomi dan Kebijakan Moneter yang (Sub)Optimal: Studi Kasus di Indonesia. Disertasi Doktor, Tidak dipublikasikan, Fakultas Ekonomi Universitas Indonesia.

548   Bibliography Solikin & Sugema, I. (2004, September). Rigiditas Harga-Upah dan Efektivitas Kebijakan Moneter di Indonesia. Buletin Ekonomi Moneter dan Perbankan (BEMP). Bank Indonesia. Sriphayak, A., & Vongsinsirikul, P. (2006). Asset prices and monetary policy transmission in Thailand. Paper Presented at The BIS Autumn Economists’ Meeting, October. Stella, P. (1997, July). Do central bank need capital? IMF Working Paper No. WP/97/83. Stella, P. (2002, August). Central bank financial strength, transperancy, and policy credibility. IMF Working Paper No. WP/02/137. Sterne, G. (2002). Inflation targeting in global context. In N. Loayza & R. Soto (Eds.), Inflation targeting: Design, performances, and challenges. Central Bank of Chile. Stiglitz, J. E. (1998). Central banking in a democratic society. De Economist, 146(2), 199– 226. Stiglitz, J. E. (1999). On liberty, the right to know, and public discourse: The role of transparency in public life. Oxford Amnesty Lecture. Retrieved from http://www.worldbank. org/html/extdr/extme/jssp012799.htm Stiglitz, J. E., & Greenwald, B. (2003). Toward a new paradigm in monetary economics. Cambridge, MA: Cambridge University Press. Stiglitz, J. E., & Weiss, A. (1981). Credit rationing in markets with imperfect information. American Economic Review, 71, 393–410. Stiglitz, J. E., & Weiss, A. (1983, June). Credit rationing in markets with imperfect information. American Economic Review, 71(3), 393–410. Stock, J. H., & Watson, M. W. (1989, January). Interpreting the evidence on money-income causality. Journal of Econometrics, 40(1), 161–181. Stone, M. R. (2003, January). Inflation targeting lite. IMF Working Paper No. WP/03/12. Stone, M. R., & Bhundia, A. J. (2004). A new taxonomy of monetary regimes. IMF Working Papers No. 04/191. Sturm, J.-E., & de Hann, J. (2001, June). Inflation in developing countries: Does central bank independence matter? CESifo Working Paper No.511. CESifo Institute for Economic Research. Munich, Germany. Sundarajan, V., Das, U. S., & Yossifov, P. (2003). Cross-country and cross-sector analysis of transparency of monetary and financial policies. IMF Working Paper No. WP/03/94. Surico, P. (2003). Measuring the time-inconsistency of US monetary policy. European Central Bank Working Paper No. 291. Svensson, L. E. O. (1997a). Inflation forecast targeting: Implementing and monitoring inflation targets. European Economic Review, 41, 1111–1146. Svensson, L. E. O. (1997b). Open economy inflation targeting. Paper available at SSRN: https://ssrn.com/abstract=30745 or http://dx.doi.org/10.2139/ssrn.30745. Svensson, L. E. O. (1997c). Optimal inflation targets, conservatives central banks and linear inflation contracts. American Economic Review, 87, 98–114. Svensson, L. E. O. (1999a). Price-level targeting versus inflation targeting: A free lunch. Journal of Money, Credit and Banking, 31, 277–295. Svensson, L. E. O. (1999b). Price-level targeting versus inflation targeting: A free lunch. Journal of Money, Credit and Banking, 31, 277–295. Svensson, L. E. O. (1998). Inflation targeting as a monetary policy rule. NBER Working Paper No. 6790. Svensson, L. E. O. (2002, May). Inflation targeting: Should it be modeled as an instrument rule or a targeting rule? NBER Working Paper No. 8925. Svensson, L. E. O. (2003). Escaping from a liquidity trap and deflation: The foolproof way and others. Journal of Economic Perspectives, 17(4), 145–166. Svensson, L. E. O. (2009, September). Flexible inflation targeting: Lessons from the financial crisis. Speech delivered at Workshop Towards a New Framework for Monetary Policy? Lessons from the crisis. Netherlands Bank, Amsterdam.

Bibliography    549 Svensson, L. E. O. (2010). Inflation targeting, Handbook of monetary economics. In, B. M. Friedman & M. Woodford (Eds.), Handbook of Monetary Economics, 1ed (Vol. 3, pp. 1237–1302). Netherlands: North Holland. Svensson, L. E. O. (2012). Central-banking challenges for the Riksbank: Monetary policy, financial-stability policy and asset management. CEPR Discussion Papers No. 8789, C. E. P. R. Discussion Papers. Svensson, L. E. O. (2013). Some lessons from six years of practical inflation targeting. CEPR Discussion Papers No. 9756, C.E.P.R. Discussion Papers.

Tai, P. N., Sek, S. K., & Har, W. M. (2012). Interest rate pass-through dan monetary transmission in Asia. International Journal of Economics and Finance, 4(2). Tarkka, J., & Mayes, D. (1999). The value of publishing official central bank forecasts. Bank of Finland Discussion Paper No. 22/99. Taylor, J. B. (1979). Staggered price setting in a macro model. American Economic Review, 69(2), 108–113. Taylor, J. B. (1980). Aggregate dynamics and staggered contracts. Journal of Political Economy, 88(1), 1–23. Taylor, J. B. (1993). Discretion versus policy rules in practice. Carnegie-Rochester Conferences Series on Public Policy, 39, 195–214. Taylor, J. B. (1995). The monetary transmission mechanism: An empirical framework. Journal of Economics Perspectives, 9(4), 11–26. Taylor, J. B. (1996). How should monetary respond to shocks while maintaining longrun price stability: Conceptual issues. Conference paper, Federal Reserve Bank of Kansas. Taylor, J. B. (1999). A historical analysis of monetary policy rules. In J. B. Taylor (Ed.), Monetary policy rules (pp. 319–341). Chicago, IL: The University of Chicago Press. Taylor, J. B. (2000a). Using monetary policy rules in emerging market economies. Conference paper, Banko de Mexico. Taylor, J. B. (2000b). Recent developments in the use of monetary policy rules. In C. Joseph & A. H. Gunawan (Eds.), Monetary policy and inflation targeting in emerging economies. Proceedings of a conference, July (pp. 207–219). Taylor, J. B. (2001). The role of the exchange rate in monetary-policy rules. American Economic Review, 91(2), 263–267. Taylor, J. B. (2009). The financial crisis and monetary response: An empirical analysis of what went wrong. NBER Working Paper Series No. 14631. Taylor, J. B. (2010, May/June). Getting back on track: Macroeconomic policy lessons from the financial crisis. Federal Reserve Bank of St. Louis Review. Taylor, J. B. (2014). Inflation targeting in emerging markets: The global experience. Keynote address at the South African Reserve Bank Conference Series. Theil, H. (1960). Best Linear Index Numbers of Prices and Quantities. Econometrica, 28(2), 464-480. doi:10.2307/1907734 Thornton, D. L. (2002). Monetary policy transparency: Transparent about what? FRB St Louis Working Paper No. 2002-028B. Tim Inflation Targeting. (1999). Survey hasil-hasil penelitian mengenai inflation targeting di Bank Indonesia. A survey on inflation targeting research in Bank Indonesia, Mimeo. Tjahjono, E. D., Harmanta, & Purwanto, N. M. A. (2012, April). Survey measures of inflation expectation. Bulletin of Monetary Economics and Banking. Tobin, J. (1970, May). Money and income: Post hoc ergo proctor hoc? Quarterly Journal of Economics, 84(2), 301–317. Trichet, J. C. (2004, October 5). Key issues for monetary policy: An ECB view. Keynote at the National Association of Business Economics, Philadelphia. Turner, P. (2012). Weathering financial crises: Domestic bond markets in EMEs. BIS Working Papers No. 63.

550   Bibliography Ugolini, S. (2011). What do we really know about the long-term evolution of central banking? Evidence from the past, insights for the present. Norges Bank Working Paper 2011-15. Utari, G. A. D., Arimurti, T., & Kurniati, I. N. (2012, October). Optimal credit growth and macroprudential policy in Indonesia. Bulletin of Monetary, Economics and Banking. Van ’t Dack, J. (1999). Implementating monetary policy in emerging market economies: An overview of issues. BIS Working Paper No. 5. Vayid, I. (2013). Central bank communications before, during and after the crisis: From openmarket operations to open-mouth policy. Working Paper/Document de travail, 201341, Bank of Canada. Villafuerte, J., & Yap, J. T. (2015, November). Managing capital flows in Asia: An overview of key issues. ADB Economics Working Paper Series No. 464. Vredin, A. (2015, July). Inflation targeting and financial stability: Providing policymakers with relevant information. BIS Working Papers No. 503 . Wachtel, P. (2010, November). Central banking for the 21st century: An American perspective. New York, NY: Stern School of Business New York University, Mimeo. Walport, M. (2015, March). FinTech futures: The UK as a world leader in financial technologies. Report oleh Government Chief Scientic Adviser. Walsh, C. E. (1995). Optimal contract for central bankers. American Economic Review, 85, 150–167. Walsh, C. E. (2001). The science (and art) of monetary policy. FRBSF Economic Letter. 13. Walsh, C. E. (2002, July/August). Commentary. FRB St Louis Review. Walsh, C. E. (2003). Implications of a changing economic structure for the strategy of monetary policy. Proceedings Economic Policy Symposium, Jackson Hole, Federal Reserve Bank of Kansas City. Walsh, C. E. (2008). Inflation targeting: What have we learned? Paper presented at International Experience with the Conduct of Monetary Policy under Inflation Targeting, Bank of Canada. Walsh, C. E. (2010). Monetary policy and theory (3rd ed.). Cambridge, MA: The MIT Press. Warjiyo, P. (2002, January). Towards inflation targeting: The case of Indonesia. In Inflation targeting: Theories, empirical models and implementation in Pacific basin countries. Seoul: Bank of Korea. Warjiyo, P. (2004, May). Mekanisme Transmisi Kebijakan Moneter di Indonesia. Buku Seri Kebanksentralan No. 11, Pusat Pendidikan dan Studi Kebanksentralan (PPSK), Bank Indonesia. Warjiyo, P. (2013a, May). Macroeconomic policy mix for promoting sustainable and inclusive growth. Keynote Speech delivered at ESCAP High Level Policy Dialogue and Eleventh Bank Indonesia Annual International Seminar, Yogyakarta. Warjiyo, P. (2013b, January). Indonesia’s monetary policy: Coping with volatile commodity prices and capital inflows. BIS Papers No. 70. Warjiyo, P. (2013c, October). Indonesia: Stabilizing the exchange rate along its fundamental. BIS Papers No. 73. Warjiyo, P. (2014a). US monetary policy normalization and EME policy mix: The Indonesian experience. Speech delivered at NBER 25th Annual East Asian Seminar on Economics, Tokyo, June. Warjiyo, P. (2014b, July). Flexible ITF in Indonesia: Framework, modeling, and policy making. International conference on economic modeling, Bali, Indonesia. Warjiyo, P. (2014c, August). The transmission mechanism and policy responses to global monetary developments: The Indonesian experience. BIS Papers No. 78 . Warjiyo, P. (2015a). Indonesia: Changing patterns of financial intermediation and their implications for central bank policy. BIS Papers No. 83. Warjiyo, P. (2015b, November). Indonesia: Global spillover and policy response. Paper presented at The Asia Economic Policy Conference (AEPC), Federal Reserve Bank of San Francisco (FRBSF).

Bibliography    551 Warjiyo, P., & Agung, J. (2002). (Eds.). Transmission mechanism of monetary policy in Indonesia. Jakarta: Bank Indonesia. Warjiyo, P., & Solikin. (2003). Kebijakan Moneter di Indonesia. Buku Seri Kebanksentralan No. 6, Pusat Pendidikan dan Studi Kebanksentralan (PPSK), Bank Indonesia. White, L. H. (1999). The theory of monetary institutions. London: Blackwell Publishers. Williams, J. C. (2015, August). Measuring monetary policy’s effect on house prices. Federal Reserve Bank of San Fransisco Economic Letters No. 2015-28. Williamson, S. D. (1987). Costly monitoring, Loan contracts, and equilibrium credit rationing. The Quarterly Journal of Economics, 102(1), 135–145. Wimanda, R. E., Maryaningsih, N., Nurliana, L., & Satyanugroho, R. (2014a, December). Evaluasi Transmisi Bauran Kebijakan Bank Indonesia. Bank Indonesia Working Paper No. WP/3/2014. Wimanda, R. E., Maryaningsih, N., Nurliana, L., & Satyanugroho, R. (2014b, December). Evaluating the transmission of policy mix in Indonesia. Bank Indonesia Working Paper No. WP/3/2014. Wimanda, R. E., Permata, M. I., Bathaludin, M. B., & Wibowo, W. A. (2012, December). Study on implementing macroprudential policy in Indonesia. Bank Indonesia Working Paper No. WP/11/2012. Winarno, T. (2014). The sources of Indonesia’s current account balance fluctuations: An empirical test of the vector error correction model (VECM). Journal of Emerging Economies and Islamic Research, 2(1). Winkler, B. (2000, August). Which kind of transparency: The need for clarity in monetary policy making. European Central Bank Working Paper No. 26 . Woodford, M. (1999, July). Optimal monetary policy Inertia. NBER Working Paper No. 7261. Woodford, M. (2003). Interest and prices: Foundations of a theory of monetary policy. Princenton, NJ: Princenton University Press. Woodford, M. (2011). Monetary policy and financial stability. Columbia University Working Paper presented at the 2011 NBER Summer Institute, Cambridge, MA. Woodford, M. (2012). Inflation targeting and financial stability. Sveriges Riksbank Economic Review, 2012(1), 7–32. World Bank. (1997). Private capital flows to developing countries: The road to financial integration. World Bank Policy Research Report. Oxford: Oxford University Press. World Economic Forum. (2015, June). The future of financial services: How disruptive innovations are reshaping the way financial services are structured, provisioned and consumed. An Industry Project of the Financial Services Community. Wuryandani, G., Ikram, A. M., & Handayani, T. (2001, July). Monetary policy transmission through inflation expectation channel. In P. Warjiyo & J. Agung (Eds.), Transmission mechanism of monetary policy in Indonesia. Jakarta: Bank Indonesia. Yanuarti, T. (2007). Persistensi Inflasi di Indonesia. Working Paper Bank Indonesia. Yarasevika, S., Tongato, A., & Muthia, A. C. (2015). Bank lending channel in Indonesia’s monetary policy transmission mechanism: A VECM approach. Proceedings of ISER 5th international conference, Singapore, September. Yellen, J. L. (2014, July). Monetary policy and financial stability. Remarks delivered at The 2014 Michel Camdessus Central Banking Lecture. Washington, DC: International Monetary Fund. Yetman, J. (2003, November). The credibility of monetary policy: Free lunch. ECB Working Paper No. 284. Yuniarsa, S. O., Chuan, J., & Chang, D. (2015). Exploring the relationships among interest rate, exchange rate, and stock market in Indonesia. Global Journal of Business and Social Science Review, 4(1), 440–447.

This page intentionally left blank

Index Note: Page numbers followed by “n” with numbers indicate footnotes. Academic thinking, 6–9 Accountability, 40–41, 241, 265–266, 286, 292n3, 335, 341–342, 361 Accountable central bank, 350 Ad hoc models, 288 role, 408 Adaptive expectations formation, 88 Administered prices (AP), 180, 237, 265, 312 Administrative instruments, 413 Adverse selection bias, 72 Aggregate demand (AD), 164, 287 curve, 88 shock, 63 Aggregate Rational Inflation– Targeting Model for Bank Indonesia model (ARIMBI model), 267, 288 Aggregate supply (AS), 146 Amalgamating MPTM analysis, 133–134 American Deposit Receipts (ADR), 415 Amplification, 129 Annual Development Plan, 268 Annual Financial Statements of Bank Indonesia, 346 Annual monetary base targets, 268n9 Argentina, monetary policy regime in, 182–183 Asian Financial Crisis (1997/1998), 27, 35, 106, 111, 116, 151–152, 155, 262, 268, 271, 315, 401, 424, 441–442, 503–504

Asset price channel, 126–127, 351 in EMEs, 145–146 in Indonesia, 152 Asset prices, 140, 480 booms, 473 bubbles, 424, 436, 440, 465 Asset(s) bank, 491 market prices, 439 portfolios, 28 securitization, 142–143 Authority theory, 328–331 Automated teller machines (ATM), 5, 38, 190 Back-of-the-envelope approach, 175 Backward-looking Taylor rule, 212 Bagehot’s Rule, 23 Balance of payments (BOP), 22, 81, 85, 86n6, 117, 177, 216, 288, 388, 424 Balance sheet conditions, 490 constraint, 494 Balance sheet channel, 132, 154–155 collateral constraints, 134–137 in EMEs, 146–149 external finance premium, 133–134 Balassa-Samuelson effect, 127 Bank for International Settlements (BIS), 26n2, 32, 36, 430–431, 463 Bank Indonesia (BI), 3–4, 15–16, 110–111, 345 communication strategy, 382–384 independence and accountability, 344–346 officers, 345

554   Index Bank Indonesia Act (BI Act), 111, 179n9, 186, 261–264, 266, 380, 382 amendments to, 283–286 aspects for, 271 Board of Governors, 273 macroeconomic models, 287–288 rate consistency, 279 surveys, 267 Bank Indonesia Certificates (SBI), 67, 69, 152, 189, 218, 285, 439, 510 Bank Indonesia Facilities (FASBI), 218 Bank Indonesia policy mix, 502 foreign capital flow management, 510–511 institutional arrangements, 513–514 interest rate and exchange rate policies, 507–510 macroprudential policy, 511–513 policy mix targets and instruments, 504–507 Bank lending, 50 behavior, 76, 130 channel, 130–131, 151, 154–155 Bank Negara Indonesia (BNI), 21 Bank of Canada, 490 Bank of England (BoE), 20–21, 207, 307, 447, 450 Bank of Japan (BOJ), 26, 215, 358, 487–488 Bank(ing) assets, 491 BANK-optimal rate, 73 behavior, 71–74 capital channel, 130–132 credit growth, 457 crisis, 442 deregulation policy, 189 industry, 493–497 interest rate regulations, 398 intermediation function, 219 regulatory policy, 76–77 role, 70

sector, 272 supply function, 131 survey, 267 Bankruptcy, 444 Barro-Gordon model, 301–304 standard linear quadratic model, 315 Basel Committee on Banking Supervision (BCBS), 26n2 Baseline ITF macroeconomic model, 230–231 Bayes rule, 58 Behavioral Model of Effective Exchange Rate (BEER), 287n22 Bipolarization, 103–105 BI–Scripless Securities Settlement System (BI-S4), 37 Board of Governors meeting (RDG), 287 Bonds, 463 Boom-bust financial cycle, 425 Brazil, Foreign CFM effectiveness in, 414–415 Bretton Woods system, 25–26 period (1959–1970), 403 Business cycle, 439 decisions without collateral constraints, 136 failure, 131 survey, 267 Capital, 486–487 account management, 99 capital-based bank, 154–155 controls, 403, 411, 412–414 mobility, 403 Capital adequacy ratio (CAR), 121, 132, 416 Capital flow management (CFM), 387 Capital flows, 84 macroprudential regulations on, 415–416 management, 34 volatility, 400–402, 413–414

Index    555 Capital market barriers, 84n5 development, 190 function, 147–148 Capital requirements, 436–437, 471, 484–485 constraint, 494 Cash-in-advance constraint (CIA constraint), 71n7 CCB (see Countercyclical capital buffer (CCyB)) Central bank, 10–11, 36, 216, 243, 266, 270–271, 357, 372, 387, 452 academic thinking, political economy and, 6–9 aspects and goals of, 21–22 assessment, 362 behavior, 354 credibility, 40–41, 103n9, 293 and economy, 4–6 evolution, 20 exchange rate stability and foreign capital flows, 34–39 GFC, 19–20, 28–31 gold standard era and real bills doctrine, 22–24 under government control, 24–26 inflation targeting, 26–28 institute monetary policy, 461 institutional arrangements and coordination, 43–44 institutional reform, 39–44 leaders, 186 legal monetary policy framework, 180 LOLR, 20–21 loss function, 231 macroprudential policy, 431–435 monetary and financial system stability, 35–36 monetary operations, 90–93 monetary policy, price stability, and exchange rates, 32–33 observers, 365 payment systemic stability, 36–39

policy, 3 policy reform, 31–39 role in FSS, 429–431 transparency and communication, 41–43 Central Bank Act, 187 Central bank independence and accountability, 26–28, 40–41, 226, 291n2, 292, 331–337 Bank Indonesia’s independence and accountability, 344–346 central bank accountability theory, 331–337 critical issue, 339–341 delegation of authority theory, 328–331 and economic performance, 337–339 empirical dimensions, 320–327 empirical studies, 337 ex post and ex ante accountability, 341–342 policy independence and credibility during crisis period, 342–344 theoretical models concerning, 327 Central bank policy mix, 467 Bank Indonesia policy mix, 502–514 conceptual dimension, 463 DSGE model, 462 DSGE modeling with macrofinancial linkages, 486–502 formulation, 481–486, 499–502 integration of monetary policy and macroprudential policy, 468 ITF, 461 price stability and financial system stability targets, 463–467 structural macroeconomic modeling, 473–486, 487 Taylor rule, 469 transmission mechanism, 470–473

556   Index Central banking laws, 235–236 in EMEs, 236 provisions in, 235 Chicago Board of Exchange (CBOE), 393n1 China Banking Regulatory Commission, 456 Classical economics, 349–350 Classical quantity theory of money, 59 Classical theory, 11, 50–51 Collateral constraints, 134–137, 488 Colombia, Foreign CFM effectiveness in, 414–415 Committee on Payments and Settlement Systems (CPSS), 26n2 Communication, 41–43, 358–361 effectiveness, 382 methods, 361–362 monetary policy, 351 policy, 280 targets, 362–365 Consensus Forecast (CF), 153 Conservative agent, 328–331 Conservative central bank, 296n4, 328n4, 365–370 Constant elasticity of substitution (CES), 493 Constant rule, 199 Consumer inflation expectations, 153 protection, 456 Consumer Expectation Survey, 267 Consumer price index (CPI), 65, 179, 237–238, 464 inflation projections, 265 Contagion process, 427 Core inflation, 237–238, 243, 254, 265, 276, 311–312 Corporate balance sheet channel, 147 investment, 125 technology, 57 transparency, 154 Cost of capital on investment, 124–125 of controlling inflation, 175–176

of crisis recovery, 28 of disinflationary policy, 178n8 Countercyclical capital buffer (CCyB), 435, 454, 467 Covered interest parity (CIP), 123 Covered interest rate parity (CIRP), 84–85 Credibility, 24, 39–41, 292, 319 during crisis period, 342–344 effort to improving, 295–298 index, 307–308 macroprudential policy, 474 measures, 306 parameterization approach, 307–308 problem, 293–295 problem analysis with supply shocks, 304–306 qualitative approach, 308–309 quantitative modeling approach, 309–311 Credit, 491 availability and bank behavior, 71–74 booms and busts, 442–443 cards, 38, 79, 113 ceilings, 208 crunch, 272 exchange rate, 351 gap, 480, 481–482 growth, 417, 511 guarantees, 131 information system development, 77 market, 75, 130 performance, 70 risk, 72, 132, 505 risk-taking behavior, 468 supply function, 132 Credit channel, 268–269, 272 bank, 470 in EMEs, 146–149 In Poland, 246 Credit default orders (CDOs), 452 Credit default swaps (CDS), 138, 393n2, 451 Credit rationing, 130, 136, 439 equilibrium, 71–74

Index    557 Credit view of monetary transmission (see also Money view of monetary transmission) aggregate demand response, 129–130 balance sheet channel, 132–137 bank capital channel, 130–132 bank lending channel, 130–131 risk-taking channel, 137–140 Credit-to-GDP ratio, 483 Crisis (see also Global financial crisis (GFC)) analyses, 107 Asian Financial Crisis (1997/1998), 27, 35, 106, 111, 116, 151–152, 155, 262, 268, 271, 315, 401, 424, 441–442, 503–504 banking, 442 economic, 183–184, 442 exchange rates, 105–107 external debt, 442, 462 financial, 22, 132, 344, 379, 423–424 México crisis (1994), 106 payments, 462 post-global, 464 prevention, 20 sudden-stop capital reversal, 462 US, 462 Croatia, prudential regulations on foreign loans in, 416–417 Croatian National Bank (CNB), 416 Cross-border capital mobility, 90 Cross-border investment, 388–389 Crowding out effect, 122n6 Currency, 5, 36 board, 182n10 demand, 88 fluctuations, 123 Current account liberalization (CAL), 390 Cyclical behaviors, 504–505 De facto approach, 103–104, 402–403 De facto monetary policy strategy, 278

De jure approach, 103–104, 402–403 Debt-to-capital ratio, 412 Debt-to-income ratio (DTI), 454, 467 Demand for credit, 72 curve, 231 for deposits and loans, 493 for foreign exchange, 111 for investment, 125 for money, 55, 70, 287 Demand-side shocks, 164–165 standard Poole model with, 194–195 Democratic accountability, 296n4, 331–337, 355–358 Deposit(s), 491 bank, 21 insurance corporation, 427–428 Deutsche Bundesbank, 179, 184n12, 184–185, 214, 322, 447 Deutsche eMark (DEM), 95 Direct credit, 208 instruments, 208 monetary instruments, 246 pass-through effects, 269 Disinflation process, 314 Domestic bond market, 91 Domestic currency, 167 depreciation, 183–184 stability, 4 values, 8 Domestic economy, 401, 407–408, 467 Domestic financial institutions, 413 markets, 401 repression, 25 sector deepening, 398–399 Domestic systemically important bank (D-SIB), 453 Domestic-oriented economy, 470 Dornbusch’s overshooting model, 82, 88–90, 101 Dual mandate, 10, 14, 358, 461, 463, 467, 469, 477, 479, 507 Dynamic analysis of elasticity, 98

558   Index Dynamic General Equilibrium Model of Bank Indonesia (GEMBI), 288 Dynamic Stochastic General Equilibrium model (DGSE model), 15, 141–143, 462 formulation, 499–502 with macrofinancial linkages, 486 macrofinancial modeling approaches, 488–490 monetary policy and macroprudential policy, 490–499 Eclectic inflation targeting (EIT), 224, 248, 250 Economic cycle, 436 output-prices trade-off in, 173–174 Economic factors, 342–343, 366 public uncertainty to changes in, 368–369 Economic performance, 337–339, 374–377, 395 empirical phenomenon, 395–396 Lucas Paradox, 396–400 Economic/economy, 57–58 Central bank and, 4–6 crisis, 183–184, 442 decision-making process, 298 deregulation, period of, 188–190 developments, 335 downswing, 132 exchange rates and, 81–112 forecasts, 349 fundamentals, 397–398 growth, 12, 162, 250, 356, 374, 387–388, 402, 461, 468 imperfect transparency, 365 independence, 320, 323–324 indicators, 243n5, 363 integration, 102 models, 243–244 puzzle, 68 recovery momentum, 262 resources, 349–350 shocks, 88, 294–295

stability, 7 stabilization and rehabilitation, 187–188 system, 67, 232 theory, 440 transparency, 352, 368–369 upswing (see Financial accelerator) Emergency Funding Facility (FPD), 285 SUN purchases for, 285–286 Emerging market economies (EMEs), 6, 26, 207, 244, 351, 387, 441, 461, 493–494 asset price channel, 145–146 credit and balance sheet channels, 146–149 exchange rate channel, 145–146 expectations channel, 145–146 interest rate channel, 145–146 monetary policy transmission in, 145 monetary policy transmission in post-GFC era, 149–151 Emerging Portfolio Fund Research (EPFR), 401 Empirical studies on effect of capital flow volatility, 400–402 of transparency and communication strategies in countries, 370–380 Equilibrium, 59 analysis, 53 credit rationing, 74 model, 57 Error correction model (ECM), 156 European Central Bank (ECB), 26, 142, 291, 321, 350, 426, 447, 453n3, 476 European Monetary System (EMS), 178 European Union (EU), 141 monetary policy transmission, 141–143 Eurosystem, 416

Index    559 Ex ante, 398, 435 accountability, 341–342 credibility index, 307 monetary policy clarity, 249 Ex post, 398 accountability, 341–342 credibility index, 307 monetary policy clarity, 249 Exchange rate channel, 127–128, 418 in EMEs, 145–146 in Indonesia, 152 Exchange rate determination theory (see also Monetary economic theory), 82 CIRP, 84–85 exchange rate model with sticky prices, 88–90 market microstructure models, 93–97 Mundell–Fleming model, 85–87 portfolio balance model and central bank monetary operations, 90–93 PPP, 83–84 Exchange rate mechanism (ERM), 106–107, 183 Exchange rate pass-through (ERPT), 99–101 Exchange rate(s), 32–33, 81, 287, 391, 429, 462, 464, 468, 480, 507, 511 appreciation, 394 choice of exchange rate system, 101–103 crises, 105–107 depreciation, 92–93, 421 determination theory, 82–97 developments, 269 empirical findings for, 98 empirical findings in Indonesia, 100–101 and inflation, 99–100 intervention, 99 models, 82, 88–90 monetary policy and theory, 81–82 overshooting model, 89

policy, 507–510 stability, 34–39, 122–123, 235, 250, 402 stabilization, 404–405 stabilization policy, 107–109 system and policy, 101, 109–112 systems, 103–105, 403 targeting, 167–168, 182–184 theory, 82, 93 and trade, 98–99 volatility, 364, 402 Expectations channel, 128–129 in EMEs, 145–146 in Indonesia, 152 market, 468 Expected Augmented Phillips Curve (EAPC), 399 Export Bonus System, 110n11 Fear index, 393n1 Fear of floating, 32, 103–105, 107 Federal Funds Rate (FFR), 3, 143, 356, 418, 420, 464–465 Federal Open Market Committee (FOMC), 3, 41, 308, 321 Federal Reserve Act, 179–180 Federal Reserve System (Fed), 21, 33 Financial account, 389 amplification, 139 crisis, 22, 132, 344, 379, 423–424 DSGE models with financial frictions, 491–497 globalization, 405 indicators, 243n5 innovation, 27–28, 143–144 instability, 425 interconnectedness, 444–445 liberalization, 82 policy transparency, 373 repression, 188 system, 124, 141, 447 system instability, 430–431 theory, 138 transparency, 351 variables, 353

560   Index Financial accelerator, 15, 119, 133–134, 137, 488 theory, 439–440 Financial asset(s), 127, 391, 437 portfolio diversification, 91 prices, 126, 440, 463 Financial Conduct Authority (FCA), 43 Financial market(s), 146, 364, 392–393 development, 147 liberalization, 147 volatility, 355 Financial network analysis, 449, 451 chain, 452 Financial Policy Committee (FPC), 513 Financial procyclicality, 426, 435 empirical evidence for, 440–444 financial accelerator theory, 439–440 financial cycle and economic cycle procyclicality, 436 interaction between business cycle, 438 Financial products, 189 innovation diversification, 452 Financial sector, 474 liberalization, 190 procyclicality, 269n10 Financial Services Authority (OJK), 44, 261n2, 505, 514 Financial Soundness Indicator (FSI), 465 Financial Stability Report (FSR), 378–379, 379n7 Financial System Crisis Prevention and Resolution Act (PPKSK Act), 426, 453, 505, 514 Financial system stability (FSS), 19–20, 24, 28–29, 35–36, 280, 364, 411–412, 423, 472, 476–477, 502 central bank’s role in, 429–431

conceptual dimension, 424 GFC of 2008/2009 and, 425–429 interconnectedness and financial networks, 444–452 macroprudential policy theory and practices, 452–458 effect of monetary policy, 472 risk, 471 targets, 463–467 theoretical model and empirical evidence of financial procyclicality, 435–444 Financial technology (FinTech), 5, 38–39 Fixed exchange rate regime, 87, 102 system, 102, 110, 405 Flexible Inflation Targeting Framework (FITF), 10, 278n13, 279, 463 Foreign bond market, 91 Foreign capital flows, 34–39, 387 and economic performance, 395–400 foreign CFM, 409–417 impact of global spillovers from US monetary policy, 418–421 management, 34, 461–463, 470, 487, 503, 510–511 and monetary policy dynamics in Indonesia, 406–409 and monetary stability, 400–409 neoclassical theory and PI theory, 388–391 push and pull factors, 391–395 theoretical dimension and, 388 volatility, 423 Foreign CFM, 409 practices, 414–417 principles, targets, and instruments, 410–414 Foreign exchange, 81 derivatives, 416 intervention, 107–109, 122–123, 507

Index    561 Foreign exchange market, 88, 118, 129, 427, 467 intervention, 404 microstructure, 93 GARCH, 123n7 “Gebrakan Sumarlin”, 189n13 General equilibrium analysis framework, money role in, 52 MIU function model, 55–57 OLG model, 53–55 real sector analysis, 52–53 Germany in controlling inflation, 184 Global economy, 387 developments, 391–392 Global financial crisis (GFC) (see also Inflation), 3, 6–7, 12, 14, 19–20, 32–33, 35, 38, 81, 137, 150, 154, 258, 262, 418, 423, 461, 465 of 2008/2009 and FSS, 425–429 central banks and, 28–31 inflation targeting after GFC of 2008/2009, 257–259 strengthening ITF implementation strategy after, 277–281 Global spillovers, 387 from US monetary policy, 418 Global Systemically Important Banks (G-SIBs), 455 Government, 295 bond yields, 243 debt securities, 271, 285, 439 financial sustainability, 272 fiscal stimulus package, 512 and parliament, 363–364 Great Depression (1930), 7, 24, 28, 31, 137 Great Moderation, 27, 462 Gross domestic product (GDP), 12, 85 deflator, 237n4 growth, 420–421 Hamiltonian function, 56 Headline inflation (see Consumer price index (CPI))

Hedge funds, 447 Herding behavior, 427 Heterogeneous economic players, 488 Heterogeneous information, 93–97 Household balance sheet channel, 147 and entrepreneurs/businesses, 491–493 Hybrid Phillips curve model, 173 Imperfect information, 57 money–output correlation against, 59–61 relationship between money– output and, 57–59 Imperfect international finance, 399–400 Imperfect transparency, 365–370 Implicit interest rate policy rule, 212–215 Implicit targeting, 186 Independence (see also Central bank independence and accountability), 40–41, 43–44 goal, 320–322, 344–345 index, 325 institutional, 168 instrument, 236, 320, 322–323, 345 performance, 338 personal, 323–324, 345 policy, 319, 342–344 political, 320 Indonesia assessing monetary policy credibility in, 311 asset price channel, 152 balance sheet channel, 154–155 Bank Indonesia’s communication strategy, 382–384 bank lending channel, 154–155 banking system, 272 economic stabilization and rehabilitation, 187–188 empirical studies of time inconsistency in Indonesia, 314–318

562   Index era of oil-based economic growth, 188 exchange rate channel, 152 exchange rate on economy in, 100–101 exchange rate system and policy in, 109–112 expectations channel, 152 foreign capital flows and monetary policy dynamics in, 406–409 inclusion in IMF Program, 273 interest rate channel, 152 ITF, 186–187, 262–263 monetary operations in, 216–220 monetary policy in, 66–69, 186 monetary policy transmission in, 151 period of economic deregulation, debureaucratization, and liberalization, 188–190 policy credibility in Indonesia during ITF implementation, 311–314 political transition process in, 272 transmission of risk-taking behavior and central bank policy mix, 155–157 transparency and communication strategy in, 380 transparency to strengthen monetary policy framework, 380–382 Inflation (see also Global financial crisis (GFC)), 162, 349, 376, 387–388, 420–421, 463–464 bias, 167n3, 302–303 controlling cost, 175–176 deviation, 409 exchange rate and, 99–100 exchange rate transmission, 128 expectations, 128–129, 153n10, 267 indicators, 243 inflation-averse central bank, 328 inflation-forecast-based rule, 201

nutters, 278n13 projections, 242–244, 267–268 rule, 303 stability, 236 variance, 367–368 Inflation targeting, 169, 185 clarity of commitment to, 248–249 classification, 250–251 credibility, 249–250 and economic performance, 252 after GFC of 2008/2009, 257–259 importance, 250–252 regimes, 247 regimes and rationale, 247–248 successes and alternative opinions, 252–257 Inflation Targeting Framework (ITF), 7, 19, 26–27, 33, 100, 179, 186–187, 201, 223, 261, 278n13, 311, 351, 403, 429, 461 actual and target inflation, 276 announcing, 239–240 formulation, 236–240, 264–265 by government, 283–285 implementation, 262–263, 274–277 in inflation forecast targeting format, 232–234 institutional framework, 235–241, 263–266 issues during transition period, 271–274 ITF-based monetary policy framework, 223–224, 226 macroeconomic theory, 230 operational framework, 242–247, 266–271 point or range, 238 policy credibility in Indonesia during ITF implementation, 311–314 rationale, characteristics, advantages, and disadvantages, 224–229 size, 238

Index    563 strengthening ITF implementation strategy after GFC, 277–281 theoretical model, 229–234 Institutional arrangements, 513–514 and coordination, 43–44 Instrument independence, 320, 322–323, 345 in monetary policy, 236 Interbank money market, 449 system dynamics, 449 Interconnectedness, 453 empirical evidence for financial, 447–452 and financial networks, 444 theory, 444–447 Interest rate, 67, 75–76, 194–196, 205–207, 247, 258–259, 351, 357, 364, 468, 502 lag, 409 policy, 120–121, 507–510 rules, 203 Interest rate channel, 124–126 in EMEs, 145–146 in Indonesia, 152 Interest rate parity theory (IRP theory), 84–85 International Capital Asset Pricing Model (IAPM), 390 International Monetary Fund (IMF), 26n2, 270, 388 assessing transparency by, 370–372 Investment capital (see Loan—capital) cost of capital on, 124–125 demand function, 125 wealth effect on, 127

Keynesian-neoclassical synthesis, 29, 61–63 Kydland and Prescott’s model, 298–301

J-curve effect, 100–101 “Just-do-it” strategy, 186, 308

Large-scale models, 244 structural macroeconomic model, 267 Leading Economic Indicators (LEI), 268 Lender of last resort (LOLR), 5, 7, 19–21, 28–29, 35, 38, 215, 427–428 Lending rate, 495 requirements, 246 Lending facility (LF), 120 Linearization model, 497 Liquidity, 139–140, 486–487 liquidity-related macroprudential instruments, 456 market, 39, 139 Liquidity preference money supply (LM), 194 Loan agreements, 488–489 capital, 436 demand and supply of, 154 loanable fund theory, 72–73 Loan-to-deposit ratio (LDR), 70, 121, 132, 454, 467 Loan-to-value ratio (LTV ratio), 4, 134, 137, 440, 467, 513 Log-linear equation, 88 first-order condition, 497 Loss function, 232 of monetary authority, 199 Lucas Island model, 57–58 Lucas Paradox, 395–400 Lucas supply function, 59

k-percent rule, 198 Kalman filter, 312n7, 400 Keynesian and neoclassical economics, 349–350, 461

Macroeconomic policies, 409–410 stability, 410–412, 463–465, 502 welfare function, 162

564   Index Macroeconomic Model of Bank Indonesia (MODBI), 267, 287 Macrofinancial linkage, 428 Macrofinancial modeling approaches in DSGE, 488–490 Macrointernational linkage, 428 Macronational linkage, 428 Macroprudential policy, 35–36, 258, 410–411, 424, 431, 461– 462, 466–467, 469–470, 487, 511–513 application in various jurisdictions, 456–458 central bank, 431–435 in DSGE modeling, 490–497 framework, 468 interconnectedness and financial networks, 444–452 loan constraints, 499 principles, targets, and instruments, 453–456 response, 433 theoretical model and empirical evidence of financial procyclicality, 435–444 theory and practices, 452–458 time deposit rate, 498 Macroprudential regulation, 5 on capital flows, 415–416 Market cycle, 145 equilibrium loanable funds, 73 imperfections, 69–77 liquidity, 39, 139 microstructure analysis (see Exchange rate theory) microstructure models, 93–97 signaling, 271 transparency (see Operational transparency) Market expectations channels, 468 Poole model variations with, 195–197 Markov-switching model, 401 Markowitz model/theory, 126, 390

Marshall–Lerner condition, 98–100 Mass media and public, 362–363 McCallum rule, 200–201, 212, 214 México crisis (1994), 106 Micro, small, and medium enterprises (MSME), 38, 77 Microfinancial sector, 428 Microprudential policy (see also Macroprudential policy), 377–378n6, 431 Microprudential regulation, 5 Microprudential supervisory authority, 424 Monetarist analysis framework, money role in, 61–63 Monetary aggregates, 205–207, 215 Monetary and Financial Policies (MFP), 370–371 Monetary Authority of Singapore (MAS), 457 Monetary condition index (MCI), 246 Monetary economic theory (see also Exchange rate determination theory), 11, 69, 115, 124 credit availability and bank behavior, 71–74 money, 70–71 new paradigm in, 74–77 Monetary policy, 5, 23, 26, 28–30, 32–33, 35, 49, 105, 117, 125, 126, 131, 161, 187, 189, 225, 274, 337, 377n6, 387, 462, 464–465, 469, 472, 487 accountability, 226 autonomy, 82n1 communication strategy, 358 communication targets, 362–365 communications methods, 361–362 consensus regarding role of, 65–66 consistency, 27 in DSGE modeling, 490–499 dynamics in Indonesia, 406–409 in economy, 49–50 effectiveness, 272, 285

Index    565 empirical evidence in Indonesia, 66–69 foreign capital flows and monetary stability, 400–409 framework, 468 impact of global spillovers from US monetary policy, 418–421 inconsistency, 40 instruments, 205–212 market imperfections and new paradigm of monetary economics, 69–77 monetary policy modeling and variable selection issues, 63–65 normalization process of, 30 optimization, 387 rate, 393, 495 regime, 355–358 scope of communication, 358–361 theoretical review, 50–63 and theory, 81–82, 119 transmission, 268–270 transmission channels, 353 transparency, 42–43, 139, 225, 349–350–351, 373–377 Monetary policy credibility, 102–103, 225, 275, 429, 482–483 assessment in Indonesia, 311–318 conceptual dimensions of policy credibility, 293–298 empirical studies, 306–311 time inconsistency, 298–306 Monetary policy operational framework, 12–13, 193 application in several countries, 205–215 conceptual dimensions and theoretical models, 194–204 monetary operations in Indonesia, 216–220 selection strategy, 203 Monetary policy response implicit interest rate policy rule in several countries, 212–215 monetary aggregates, 215

Monetary policy strategy framework, 161 from 1990s to present day, 179–182 empirical studies and related issues, 170 exchange rate targeting, 167–168, 182–184 final policy target, 162, 177 formulation, 170–171 implementation in several jurisdictions, 176 inflation targeting, 169, 185 monetary policy regime in Indonesia, 186–190 monetary targeting, 168–169, 184–185 no explicit anchor targeting, 169–170, 186 nominal income targeting, 170 output-prices trade-off in economic cycle, 173–174 Phillips curve phenomenon, 171–173 quadratic function, 162–163 regime, 166–167, 182 sacrifice ratio, 175–176 shift in priorities, 177–179 short-term output-prices trade-off, 171–173 targeting long-term policy goals, 163 targeting short-term policy goals, 163–166 Monetary policy transmission mechanism (MPTM), 11–12, 115–116, 118–120, 271 in advanced countries, 141 in countries, 140 in EMEs, 145–151 empirical MPTM studies, 140–141 in EU, 141–143 in Indonesia, 151–157 map, 116–117 monetary transmission channels, 124–129

566   Index policy rate and monetary operations, 120–123 in United States, 143–145 Monetary policy trilemma, 90 index in Indonesia, 407 of open economy, 110, 402–405 Monetary shocks, 63 Monetary stability, 35–36, 400 empirical studies on effect of capital flow volatility, 400–402 foreign capital flows and monetary policy dynamics in Indonesia, 406–409 monetary policy trilemma in open economy, 402–405 Monetary transmission channels, 124 credit view, 129–140 money view, 124–129 Monetary transparency, 351 Money, 11, 36, 49, 70–71 center, 449 consensus regarding role of, 65–66 in economy, 49–50, 63 money–output correlation against imperfect market competition and price rigidity, 59–61 neutrality, 50–51, 55, 59 relationship between money– output and imperfect information, 57–59 role in general equilibrium analysis framework, 52–57 role in monetarist analysis framework, 61–63 substitution, 85 supply, 51–52, 75, 194–196, 351, 468 theoretical review, 50 Money market equilibrium, 88 instrument, 190 Money market account (MMA), 70

Money view of monetary transmission (see also Credit view of monetary transmission), 124 asset price channel, 126–127 exchange rate channel, 127–128 expectations channel, 128–129 interest rate channel, 124–126 Mundell–Fleming model, 11, 82, 85–87, 101, 402 National Development Program, 268 National Inflation Task Force, 279, 317 Negotiable certificates of deposit (NCD), 190 Neoclassical economics, 59 investment theory, 124–125 model, 55 theory, 388–391 Net Foreign Assets (NFA), 406–407 Net interest margin (NIM), 132 Net International Reserves (NIR), 218 Net open position (NOP), 412, 454 Neutrality of money, 50–51 hypothesis, 163 Normalization of monetary policy, 30 monetary policy, 508–509 O/N interbank rate, 220 Oil-based economic growth, era of, 188 Open economy, 110, 119, 161 monetary policy trilemma in, 402–405 Open global economic dynamics, 208 Open market operations (OMO), 92, 112, 208, 216, 246 SUN purchases on primary market for, 285 Optimal monetary policy, 199–200 aspects of optimal monetary policy formulation, 201–204 strategy, 196

Index    567 Optimal policy mix, 461 rule, 232–234, 329 Organization for Economic Co-operation and Development (OECD), 443 Output gap, 287 output-prices trade-off in economic cycle, 173–174 stability, 162–166 stabilization, 366–368 volatility, 376–377 Parameterization approach, 307–308 Payment instruments, 5, 21 development, 6 Payment system, 5, 426, 467, 470 policy, 4–5, 461 stability, 36–39 People’s Bank of China (PBoC), 456 Philips curve, 88, 171, 201 behavior, 302 equation, 480 formation of expectations in, 172–173 linearity in, 172 presence of, 171–172 Pigou effect analysis, 51 Policy accountability, 319 authorities, 294 formulation process, 355, 372 framework, 469 implementation, 246–247, 311n6 independence, 319, 342–344 innovations, 8 mix targets and instruments, 504–507 Policy credibility, 293, 319, 344 effort to improving credibility, 295–298 in Indonesia during ITF implementation, 311–314 problem, 293–295, 298–306

Policy instruments, 194 standard Poole model with demand-side shocks, 194–195 Policy rate, 120, 499 á la Taylor rule, 403 Policy response, 197 feedback rule and optimal monetary policy, 199–200 simple feedback rule, 200–201 Policy rule, 203, 212–215, 295, 340 Policy transparency, 14, 352 conceptual dimensions, 350 economic transparency, 368–369 empirical studies of transparency and communication strategies, 370–380 implications of imperfect transparency, 370 monetary policy communication strategy, 358–365 public uncertainty concerning preferred weight, 366–368 theoretical models on, 365 transparency, monetary policy regime, and democratic accountability, 355–358 transparency and communication strategy in Indonesia, 380–384 viewpoints concerning transparency, 351–355 Policy-making characteristics, 308 credibility, 295 process, 325 Political economy, 6–9 imperfect transparency, 365 independence, 320 support for rule of law, 297 transition process in Indonesia, 272 transparency, 325, 332, 352 uncertainty, 295

568   Index Portfolio investment (PI), 388 theory, 388–391 Price equilibrium, 89 formation, 351 price-setting equation, 61 stability, 7, 12, 19–20, 29, 32–33, 49, 76, 162–166, 235, 402, 462–467, 476–477 stabilization strategy, 195 Price index, 237 selection, 236–237 Price rigidity, money–output correlation against, 59–61 Price-based approach, 217 quantity-based approach to, 217–220 Private capital flows, 443–444 Property crisis, 442 prices, 464–465 survey, 267 Prudential Regulation Authority (PRA), 43, 513 Prudential regulations, 412–414 on foreign loans, 416–417 Public uncertainty to changes in economic factors, 368–369 concerning preferred weight, 366–368 Push and pull factors, 391–395 “Push-me, pull-you” factor, 474–475

Reserve Bank of New Zealand (RBNZ), 323, 360 Retail bank, 494 deposit bank branch, 496 loan bank branch, 495 payment system, 38 sales survey, 267 Risk premium, 392 risk-averse behavior of banks, 71 Risk-management behavior, 490 techniques, 138 Risk-taking behavior, 400, 474–475, 477, 486–487 Risk-taking channel, 130, 137 liquidity, 139–140 risk-taking dynamics, 138–139 Risk-weighted assets, 28 Rogoff model, 13 Rules vs. discretion, 197–201, 432–433 Rupiah devaluation policy, 110

Say’s law, 49–50 Seigniorage, 57n1 Short-term cyclical factors, 471 exchange rate volatility, 88 foreign loans, 415–416 goals, 359 interest rates, 353 macromodel equation, 62 output-prices trade-off, 171–173 Quantitative easing (QE), 28 Phillips curve, 287 Quantitative modeling approach, Short-Term Forecast Model for the 309–311 Indonesian Economy Quantity-based approach, 216–217 (SOFIE), 267, 287–288 to price-based approach, 217–220 South Korea, macroprudential Quarterly monetary base targets, regulations on capital flows 268n9 in, 415–416 Standard & Poor index (S&P index), Ramsey model, 55, 57 393n1 Real exchange rate, 409 Stocks, 463 gap, 480n3, 482 of credit, 50 Reserve Bank of India (RBI), 209, 447 market, 118

Index    569 prices, 153 yields, 394 Structural macroeconomic model, 462, 473, 479 central bank policy mix formulation, 481–486 coordination and losses, 487 flexible to integrated inflation targeting, 474–479 model structure and analysis framework, 479–481 Superneutrality, 51–52 of money, 55–57 Supply-side shocks, 164, 166 Poole model variations with, 195–197 Swiss National Bank (SNB), 185 “SysMo”, 466 Systemic risk, 455 accumulation, 504–505 Systemically important banks (SIB), 453, 500 T-Bills, 69n6, 208 Taylor rule, 13, 40, 200–202, 212, 214–215, 228, 234, 247, 287, 469, 473–474, 499 equation, 312n7 policy response, 479 Taylor’s aggregate demand equation, 61 The Federal Reserve (the Fed), 321 Time inconsistency, 294, 298, 306–311 Barro-Gordon model, 301–304 credibility problem analysis with supply shocks, 304–306 in Indonesia, 314–318 Kydland and Prescott’s model, 298–301 theory, 13, 225 Tinbergen rule, 506 for closed economy, 87

Transparency and communication strategy, 349, 370 assessing transparency by IMF, 370–372 central banks, 372 in crisis period, 377–380 in Indonesia, 380–384 monetary and financial policy transparency, 373 monetary policy transparency and economic performance, 374–377 Transparency assessment by IMF, 370–372 Unemployment, 178 United States (US) crisis, 462 expansionary monetary policy, 419 monetary policy in, 178, 418–421 monetary policy transmission in, 143–145 subprime mortgage debacle, 451 US Federal Reserve, 349, 358, 360 US Monetary Policy Reaction Function, 213 Vector autoregression approach (VAR approach), 64–65, 174, 176 Vector autoregressive model (VAR model), 141–143, 152, 154, 244 Volatility Index (VIX), 393 Wage behavior, 61 Wage-price mechanism, 60 Wage-setting model, 60 Walras’ law, 49n1, 54 World Bank (1997), 392 Zero interest-rate policy (ZIRP), 215