Rational Expectations and Inflation: Third Edition [Third ed.] 9781400847648

Table of contents :
Contents
List of Figures
List of Tables
Acknowledgements
Preface to the Third Edition
Preface to the Second Edition
Preface to the First Edition
1. Rational Expectations and the Reconstruction of Macroeconomics
2. Reaganomics and Credibility
3. The Ends of Four Big Inflations
4. Stopping Moderate Inflations: The Methods of Poincaré and Thatcher
5. Some Unpleasant Monetarist Arithmetic
6. Interpreting the Reagan Deficits
7. Speculations about the Speculation against the Hong Kong Dollar
8. Six Essays in Persuasion
9. Macroeconomic Features of the French Revolution
10. United States Then, Europe Now
References
Author Index
Subject Index

Citation preview

Rational Expectations and Inflation

Rational Expectations and Inflation Third Edition

Thomas J. Sargent

Princeton University Press Princeton and Oxford

Copyright © 2013 by Princeton University Press Published by Princeton University Press 41 William Street, Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press 6 Oxford Street, Woodstock, Oxfordshire, OX20 1TW All Rights Reserved Library of Congress Control Number: 2013930076 ISBN 978-0-691-15870-9 British Library Cataloging-in-Publication Data is available This book has been composed in LATEX using Peter Wilson’s memoir class The publisher would like to acknowledge the author of this volume for providing the camera-ready copy from which this book was printed. Printed on acid-free paper. ∞ press.princeton.edu Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

For Anne, Addison, and Ari

Contents

L IST OF F IGURES

xi

L IST OF TABLES

xiii

A CKNOWLEDGEMENTS

xv

P REFACE TO THE T HIRD E DITION

xvii

P REFACE TO THE S ECOND E DITION

xix

P REFACE TO THE F IRST E DITION

xxi

1. R ATIONAL E XPECTATIONS AND TION OF M ACROECONOMICS

THE

R ECONSTRUC -

Behavior Changes with the Rules of the Game The Investment Decision — Are Government Deficits

1 1

Inflationary?

General Implications of the Examples New Econometric Methods Historical and Cross-Country Analysis Implications for Policymakers 2. R EAGANOMICS AND C REDIBILITY Dynamic Games An Economy as a Dynamic Game Are Government Deficits Inflationary? Reaganomics and Credibility Conclusion 3. T HE E NDS OF F OUR B IG I NFLATIONS Introduction The Gold Standard Austria

10 11 13 15 17 19 21 26 33 36 38 38 42 47

vii

viii

Contents

Hungary Poland Germany Czechoslovakia Conclusion Appendix 4. S TOPPING M ODERATE I NFLATIONS : T HE M ETHODS OF P OINCARÉ AND T HATCHER Introduction The “Poincaré Miracle” The British Experience Mrs. Thatcher’s Plan The Outcome of the Plan So Far Overshooting the £M3 Target The Government Deficit North Sea Oil and the Pound Sterling Conclusion Second Thoughts

53 58 60 64 66 67 111 111 116 123 124 129 130 132 136 142 143

Poincaré and Moreau

Methods of Thatcher and Reagan Where to Play Chicken 5. S OME U NPLEASANT M ONETARIST A RITHMETIC

145 146 162

with Neil Wallace

Introduction A Simple Monetarist Model The Cagan-Bresciani-Turroni Effect Conclusions and Qualifications Appendix 1 Appendix 2 Appendix 3 by Danny Quah

162 166 171 173 177 180 193

6. I NTERPRETING THE R EAGAN D EFICITS

197

Government Budget Balance Barro Tax Smoothing Wallace’s Game of Chicken Conclusion

197 199 204 207

ix

Contents

Appendix

208

Hall’s Model — Barro’s Model

7. S PECULATIONS ABOUT THE S PECULATION THE H ONG K ONG D OLLAR

AGAINST

211

with David T. Beers and Neil Wallace

A Look at Hong Kong’s Monetary System The Official Float Policy An Indeterminate Exchange Rate? A Timely Depreciation Hong Kong’s Other Options Epilogue 8. S IX E SSAYS IN P ERSUASION Back to Basics on Budgets Confrontations over Deficits An Open Letter to the Brazilian Finance Minister Letter to Another Brazilian Finance Minister The Economics Are Simple — Credibility — A Delicate Institutional Issue — Advice Reasonable Doubt about the Real Plan An American History Lesson for Europe

212 214 215 218 222 223 228 228 231 233 237

241 245

9. M ACROECONOMIC F EATURES OF THE F RENCH R EVO LUTION

248

with François R. Velde

Introduction

248

Chronology of Events — Macroeconomic Theories Coloring Our Observations

Before the Revolution Even Absolute Monarchies Have Budget Constraints — The British Experience — Across the Channel — French Fiscal Backwardness or Optimality? — Snapshot of the Old Regime Laws — Offices — Taxes — Apology for the Old Regime — Fiscal Pressures to Reform — Sentiment for Copying Britain — A Temptation Resisted — Sustainability or Bad Choices? — The Last Fiscal Crisis of the Old Regime

251

x

Contents

The Search for a New Order Annus Mirabilis (1789) — Some Adverse and Persistent Revenue Shocks — Designing a New Tax System — Birth of a Currency Rise and Fall of the Assignat A Real-Bills Regime (1790–92) — War — Terror (1793–94) — Legal Restrictions — Hyperinflation (1795–96) — The Government’s Response — Default The Legacy

267

274

292

Unintended Consequences for England through Specie Flows

Appendix: The Auction Mechanism

294

10. U NITED S TATES T HEN , E UROPE N OW

297

Introduction The Math

297 300

Need for More Economic Theory? — A Humbling Message? — Modeling Reforms

The United States

307

Victorious but in Default — Trade Policies — Crisis and a Second Revolution

Restructuring Fiscal Institutions

310

The Order of Fiscal and Monetary Reorganizations

— Fiscal Policy — Discrimination and Liquidity — Federal Bailout of States — Why Pay? — Monetary Arrangements — A National Bank? — A Mint — Outcomes

Following Through?

319

Federal and State Paper Monies? — What Kind of Currency Union?

What Kind of Fiscal Union?

323

Another Federal Bailout of the States?

Lessons for Now? Appendix: Outcomes in Graphs

328 331

Data Sources

R EFERENCES A UTHOR I NDEX S UBJECT I NDEX

339 357 361

List of Figures

3.1 3.2 3.3 3.4 3.5 3.6

Retail prices in Austria, 1921–24 Wholesale prices in Hungary, 1921–24 Wholesale prices in Poland, 1921–24 Wholesale prices in Germany, 1919–24 Inflation path with a fully anticipated decrease in money supply growth Inflation path with a previously unexpected decrease in money growth

41 42 43 44 69 70

4.1

Gross domestic fixed capital formation

136

5.1 5.2

The equilibrium law of motion of the inflation rate The equilibrium dynamics of inflation.

182 183

9.1

Ratio of debt service to taxes, Britain and France, 1688–1788 Revenues and spending in Britain, 1689–1790 Revenues and spending in France, 1759–88 Military spending in Britain and France, 1688–1789 Index of real per capita revenues in France, 1726–1845 Assignats used for debt reimbursement and total issue, 1790–92 Scatter plot of the logarithm of the inflation rate versus real balances of assignats Real balances of assignats Price level and price of gold Prices in Paris of foreign exchange futures, January 1789–July 1793 Spending and revenues, 1791–95 Seigniorage raised on the assignat Price in Paris of Hamburg currency delivered in Hamburg, March 1795–September 1796 Price of a 5 F perpetual, 1795–1802

9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9 9.10 9.11 9.12 9.13 9.14

xi

252 253 256 263 270 275 277 278 279 281 284 286 288 292

xii

List of Figures

9.15 Price of a 5 F perpetual, September–December 1799

293

10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8

334 334 335 335 336 336 337

Composition of federal revenues by source Composition of federal expenditures by type Per capita composition of federal revenues by source Per capita composition of federal expenditures by type Primary deficit-to-GDP ratio Par value debt-to-GDP ratio Annual inflation and real GDP growth Composition of the debt outstanding by type of obligation 10.9 Per capita real GDP (2005 dollars) 10.10Per capita nominal GDP

337 338 338

List of Tables

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9

3.26 3.27

Austrian Budgets, 1919–22 Total Note Circulation of Austrian Crowns Austrian Retail Prices, 1921–24 Exchange Rates, Austrian Crowns The Austrian Budget, 1923–25 Austrian National Bank Balance Sheet Number Of Austrian Unemployed in Receipt of Relief Hungarian Budget Estimates, 1920–24 Balance Sheet of Hungarian Central Bank or State Note Institute Hungarian Price and Exchange Rate Hungarian Budget, 1924–25 Number of Unemployed in Hungary Polish Receipts and Expenditures Balance Sheet of Bank of Poland, 1918–25 Polish Index Numbers of Wholesale Prices, 1921–25 Polish Exchange Rates, 1919–25 Polish Unemployed German Wholesale Prices, 1914–24 German Exchange Rates, 1914–25 Real German Revenues and Expenditures Balance Sheet of German Reichsbank, 1921–24 Ordinary Revenues and Expenditures of the German Federal Government Index of Physical Volume of Production per Capita in Germany Czechoslovakia, Receipts and Expenditures 1919–25 Note Issue of Banking Office of Czechoslovakia, 1919– 24 Czechoslovakian Exchange Rates, 1919–24 Czechoslovakian Wholesale Prices, 1922–24

4.1

French Wholesale Price Index

3.10 3.11 3.12 3.13 3.14 3.15 3.16 3.17 3.18 3.19 3.20 3.21 3.22 3.23 3.24 3.25

xiii

70 71 72 73 73 74 75 75 76 80 81 82 82 83 89 90 91 92 94 95 96 106 106 107 108 109 110 147

xiv

List of Tables

4.2 4.3

Dollar Exchange in Paris GDP at Factor Cost and Final Expenditures on Goods and Services at Market Prices 4.4 Exchange, Price, and Balance 4.5 Output and Unemployment 4.6 Gross Domestic Fixed Capital Formation by Sector 4.7 Money Supply in the United Kingdom 4.8 General Government Receipts and Expenditure 4.9 Financial Transactions of the Public Sector 4.10 Net Purchases or Sales of Government Debt, by Maturity 4.11 Calculated Redemption Yields of Government Bonds 4.12 Interest Rates, Security Prices, and Yields

147 148 150 151 152 153 155 157 159 160 161

5.1 5.2 5.3 5.4

Price Level and Real Balance Dynamics Behavior of the System Before and After T Two Economies Different θ s

174 188 190 192

7.1 7.2 7.3

The Exchange Fund’s Balance Sheet Balance Sheet Effects of Issuing HK$ 100 How an Exchange Rate Depreciation Can Help a Financial Firm

213 215 220

State of Debt in 1789

262

9.1

Acknowledgements I would like to thank the following publishers for granting permission to reprint articles included in this book: “Rational Expectations and the Reconstruction of Macroeconomics,” originally published in the Federal Reserve Bank of Minneapolis Quarterly Review, Summer 1980; “Reaganomics and Credibility,” originally published in Monetary Policy, MIT Press, edited by Albert Ando et al., 1985. Reprinted with permission of the MIT Press. “The Ends of Four Big Hyperinflation,” originally published in Inflation, edited by Robert E. Hall and published by the University of Chicago Press for the National Bureau of Economic Research in 1983; “Stopping Moderate Inflations: The Methods of Poincaré and Thatcher,” originally published in Inflation, Debt and Indexation, MIT Press, edited by Rudiger Dornbusch, 1983. Reprinted with permission of the MIT Press. “Unpleasant Monetarist Arithmetic,” written with Neil Wallace, and originally published in the Federal Reserve Bank of Minneapolis Quarterly Review, Fall 1981; “Speculating about the Speculation Against the Hong Kong Dollar,” written with David Beers and Neil Wallace, and originally published in the Federal Reserve Bank of Minneapolis Quarterly Review, Fall 1983; “Interpreting the Reagan Deficits,” originally published in the Federal Reserve Bank of San Francisco Review, Fall 1986; “Back to Basics on Budgets,” published in the New York Times on August 10, 1983; “Confrontations Over Deficits,” published in the New York Times on August 12, 1983; “An Open Letter to the Brazilian Finance Minister,” published in the Wall Street Journal on January 30, 1986; “Letter to Another Brazilian Finance Minister,” published in the Banco de Investimentos Garantia S.A. Newsletter, November 1993; “Reasonable Doubt about the Real Plan,” published in the Banco de Investimentos Garantia S.A. Newsletter, December 19, 1995; “An American History Lesson for Europe,” Wall Street Journal, February 3, 2012; “Macroeconomic Features of the French Revolution,” coauthored with François Velde, from the Journal of Political Economy in 1995. Chapter 10 is “US Then, Europe Now,” published earlier

xv

xvi

Acknowledgements

by the Nobel Foundation and the Journal of Political Economy in 2012. I would like to thank my friends David Beers, François Velde, and Neil Wallace for giving me permission to reprint papers that we co-authored. I thank François Velde for having crafted LATEX programs that typeset this book and for his generosity and patience in helping me solve many typesetting problems.

Preface to the Third Edition I have added new material to this edition. Chapter 9 is “Macroeconomic Features of the French Revolution,” co-authored with François Velde, from the Journal of Political Economy in 1995. Chapter 10 is “United States Then, Europe Now,” published earlier by the Nobel Foundation and the Journal of Political Economy in 2012. I have also added three essays in persuasion to chapter 8, two from the Banco de Investimentos Garantia S.A. Newsletter and one from the Wall Street Journal. Milton Friedman taught that governments confront unavoidable problems only after they become crises. “Macroeconomic Features of the French Revolution” and “United States Then, Europe Now” illustrate that principle. In the 1780s, governments in France and the United States had made promises about taxes and expenditures that couldn’t be honored so long as decisions continued to be made under existing constitutions. Political revolutions in France and the United States replaced old constitutions with new ones designed better to serve interests of particular groups of citizens. Those 1789 political revolutions confronted perennial questions about government debt: 1. Should a government pay its debts? When a government pays its debts, different groups of people reap benefits or suffer costs. Whether a government chooses to pay always depends on a balance of interests and a distribution of political power. 2. In a federal or confederal system, should a central government pay debts incurred by subordinate governments? When a central government bails out a subordinate government, it rewards taxpayers and creditors of the subordinate government. That fosters adverse incentives for a subordinate government to be profligate again. To prevent that, a central government that bails out a subordinate government often requires the subordinate government to surrender fiscal sovereignty. 3. Should government debt be risk-free or should it be ex ante xvii

xviii

Preface to the Third Edition

state-contingent and therefore risky? Theories of dynamic optimal taxation say that governments should issue statecontingent debt (e.g., Lucas and Stokey (1983)). Nevertheless, today most governments issue debt that promises to provide a risk-free nominal return. Creditors care about real returns that are heavily influenced by inflation rates. Inflation allows governments and other issuers of risk-free nominal debt de facto to issue risky real debt. Chapter 7 about Hong Kong in 1983 gives an example in which a government, purposefully or not, used inflation to repair balance sheets of imperiled financial intermediaries. That is one way a central bank can magically convert “bad loans into good ones.” Pressures exist for central banks to do that today both in Europe and the United States. Thomas J. Sargent New York, New York September 2012

Preface to the Second Edition I have used the occasion of a reprinting to add two chapters to this book. Chapter 6 reprints the article “Interpreting the Reagan Deficits,” which I published in the Federal Reserve Bank of San Francisco Review in 1986. Chapter 8 reprints three “essays in persuasion” that were published in the New York Times and the Wall Street Journal. Aside from correcting a few grammatical and typographical errors, I have not altered the original articles. However, in response to comments by John Taylor and Joseph Bisignano, I have added a few remarks at the end of Chapter 4, confessing errors that I made in comparing and interpreting the monetary and fiscal policies of Poincaré and Thatcher. Because of fundamental and enduring changes in fiscal policy that Thatcher engineered at about the time that I wrote chapter 4, the appropriate contrast is not between Thatcher’s policies and Poincaré’s, but between Thatcher’s and Reagan’s. In recent years, there has been much good research on subjects treated in this book. About the experiences of countries that have stopped or attempted to stop big inflations, systematic work has been done by Bomberger and Makinen (1980, 1983), and Siklos (1989, 1993), who have assembled interesting additional data bearing on the interpretation of inflation in terms of the “monetarist arithmetic” emphasized in this book. On the German hyperinflation, see Holtfrerich (1980), Webb (1989), and Garber (1982). Garber’s paper is noteworthy for the real distortions associated with the German hyperinflation that he documents, and on the basis of which he warns the reader against interpreting big inflations as purely “neutral” monetary phenomena. Also see the paper on the Hapsburg successor states by Wicker (1986). For useful critical summaries of the literature, see Dornbusch et al. (1990) and Bruno (1989). A number of papers have been written extending and modifying our “Unpleasant Monetarist Arithmetic.” Examples include papers by Drazen (1985) and Liviatan (1984).

xix

xx

Preface to the Second Edition

The articles in this book are nontechnical attempts to put rational expectations methods to work to interpret macroeconomic events at an informal level. Some of the arguments in this book now seem to me to be too simplistic, and would require modification if they were to be recast in modern technical terms. This is especially true of the passages in chapters 1, 2, and 3 in which I discuss “policy regimes.” I compare two ways of making and analyzing public policy choices: taking “isolated actions” versus setting “rules” to be applied recurrently in the future. An up-to-date treatment of this matter would focus on issues of commitment and the timing of decisions and how they depend upon the presence or absence of institutions or reputational mechanisms that might give a current government the ability to tie the hands of its successors. If a government is operating in an environment where some such commitment mechanism is in place, then it can, should, and (under some theories) will choose a “policy regime” once and for all, with future governments simply administering a set of rules they have been handed. However, where such a commitment mechanism is not in place, it is more natural to model the government as choosing sequentially, doing the best it can period-by-period. In the last five years, macroeconomists have made important strides in clarifying these issues by borrowing and adapting concepts from game theory. See Stokey (1989) and Chari and Kehoe (1990). This book uses historical evidence at many points. I am neither an historian nor an economic historian, and have relied entirely on secondary sources. In writing these articles, I am not pretending to be an historian, only a macroeconomist interested in giving my theories a shot at some interesting data sets that others have taken the time to assemble. Thomas J. Sargent Chicago, Illinois February 1992

Preface to the First Edition During the last decade, the idea of rational expectations transformed macroeconomic theory. A simple intuition motivates this idea; namely, that people do not systematically ignore readily available information that could be used to improve their decisions. However, the initial applications of this idea in macroeconomic theory and econometrics have been technically demanding. The highly technical nature of this early work is not surprising, since this work itself grew out of the technical literatures on optimally controlling Keynesian macroeconometric models and on deriving optimal decision rules for the Keynesian categories of consumption, investment, and money demand. One consequence of the highly technical orientation of early work on rational expectations in macroeconomics is that an appreciation has been slow to develop for the relevance of the ideas for the practice of day-to-day macroeconomics. Rational expectations macroeconomics reasons with a new set of concepts and a new language. It is not simply proposing new answers to old questions but a new view about which questions are useful to ask. During the long period of dominance of Keynesian macroeconomic theory, there developed a broad body of practical experience in applying the theory quickly in informal contexts. Such experience created ease of application and communication, which were important ingredients in the practical success of Keynesian economics as an applied theory. One of the (legitimate) reasons for resistance to rational expectations ideas in macroeconomics has been that they are perceived as destroying the old language and tools of applied macroeconomics while providing no workable substitute. My main reason for writing this book was to try my hand at putting rational expectations macroeconomics to work at an informal, noneconometric level in order to describe and interpret some recent and historical economic events. The events studied mainly involve periods of actual or threatened depreciation in the value of a country’s currency and in which a government was

xxi

xxii

Preface to the First Edition

interested in preventing that depreciation. To interpret these events, I have used a rational expectations theory of inflation. Besides being a phenomenon of great contemporary interest, the process of inflation has long been thought to be influenced by people’s methods of forming expectations about inflation. Thus, the study of inflation provides a natural context for trying to apply rational expectations theories. Two main ideas run throughout the text. The first is the principle of strategic interdependence, which holds that one person’s pattern of behavior depends on the behavior patterns of those forming his environment. When behavior patterns of those forming a person’s environment change, the individual can usually profit by adjusting his or her own behavior pattern. This principle, which is at the heart of rational expectations theory, was ignored under standard practice in Keynesian macroeconomics. One of the purposes of this text is to pursue the logical implications of this principle and to apply it repeatedly in interpreting a variety of episodes. The second idea is that monetary and fiscal policies must be coordinated because the government faces a budget constraint. It follows that there is no “purely monetary” cure for inflation, if purely monetary means a policy that can be imposed by the central bank without any assistance from the taxing and spending authorities. This idea will guide our interpretation of the causes of the beginnings and endings of inflationary episodes. The first chapter, “Rational Expectations and the Reconstruction of Macroeconomics,” is an overview that describes some key ideas of rational expectations macroeconomics. This chapter describes in greater detail the role of informal historical studies relative to formal econometric ones. The second chapter, “Reaganomics and Credibility,” elaborates the principle of strategic interdependence and applies it to interpret the struggle between monetary and fiscal policy that has occurred in the United States since 1981. The next two chapters, “The Ends of Four Big Inflations” and “Stopping Moderate Inflations: The Methods of Poincaré and Thatcher,” provide interpretations of economic policies and outcomes during several periods when rapid inflation was either occurring or thought to be threatening to occur. The chapter “Some

Preface to the First Edition

xxiii

Unpleasant Monetarist Arithmetic,” written with Neil Wallace, explores some aspects of the coordination problem facing the monetary and fiscal authority. The unpleasant arithmetic alluded to in the title prevents the central bank from successfully fighting inflation by itself if the fiscal authority persists in running a netof-interest budget deficit. The analytics of the monetary-fiscal coordination problem that are laid out in this chapter are used extensively in the other chapters. The chapter “Speculations about the Speculation against the Hong Kong Dollar” was written with David Beers and Neil Wallace. It describes some recent events in Hong Kong involving the depreciation of the Hong Kong dollar. We think of the situation as one in which Hong Kong banks, being confronted with a situation in which many of their loans were threatening to go bad, found these bad loans turned into good ones when the central bank acted to depreciate the currency in which the loans and the banks’ deposits were denominated. It requires only a little imagination to see the situation in Hong Kong as a microcosm of the international debt crisis. A depreciation of the US dollar via inflation would go a long way toward “making good loans out of bad ones” in the countries now on the verge of technical default. Thomas J. Sargent Minneapolis, Minnesota March 1985

1 Rational Expectations and the Reconstruction of Macroeconomics

The government has strategies. The people have counterstrategies. Ancient Chinese proverb

Behavior Changes with the Rules of the Game In order to provide quantitative advice about the effects of alternative economic policies, economists have constructed collections of equations known as econometric models.1 For the most part these models consist of equations that attempt to describe the behavior of economic agents—firms, consumers, and governments—in terms of variables that are assumed to be closely related to their situations. Such equations are often called decision rules because they describe the decisions people make about things like consumption rates, investment rates, and portfolios as functions of variables that summarize the information people use to make those decisions. For all of their mathematical sophistication, econometric models amount to statistical devices for organizing and detecting patterns in the past behavior of people’s decision making, patterns that can then be used as a basis for predicting their future behavior. As devices for extrapolating future behavior from the past under a given set of rules of the game, or government policies, these models appear to have performed well.2 They have not performed well, however, when the rules changed. In formulating advice for policymakers, economists have routinely used these models to predict the consequences of historically unprecedented, hypothetical 1 Lucas

and Sargent (1979) provide a brief explanation of econometric models and their uses in macroeconomics. 2 This evidence is cited by Litterman (1979) and his references.

1

2

Chapter 1: Rational Expectations

government interventions that can only be described as changes in the rules of the game. In effect, the models have been manipulated in a way that amounts to assuming that people’s patterns of behavior do not depend on those properties of the environment that government interventions would change. The assumption has been that people will act under the new rules just as they have under the old, so that even under new rules, past behavior is still a reliable guide to future behavior. Econometric models used in this way have not been able to predict accurately the consequences of historically unparalleled interventions.3 To take one recent example, standard Keynesian and monetarist econometric models built in the last 1960s failed to predict the effects on output, employment, and prices that were associated with the unprecedented large deficits and rates of money creation of the 1970s. Recent research has been directed at building econometric models that take into account that people’s behavior patterns will vary systematically with changes in government policies—the rules of the game.4 Most of this research has been conducted by adherents of the so-called hypothesis of rational expectations. They model people as making decisions in dynamic settings in the face of welldefined constraints. Included among these constraints are laws of motion over time that describe such things as the taxes people must pay and the prices of the goods they buy and sell. The hypothesis of rational expectations is that people understand these laws of motion. The aim of the research is to build models that can predict how people’s behavior will change when they are confronted with well-understood changes in ways of administering taxes, government purchases, aspects of monetary policy, and the like.

3 Sims

(1980) and Lucas (1976) describe why econometric models can perform well in extrapolating the future from the past, assuming no changes in rules of the game, while performing poorly in predicting the consequences of changes in the rules. 4 For an example of such research and extensive lists of further references, see Hansen and Sargent (1980) and Lucas and Sargent (1981a).

Behavior Changes with the Rules of the Game

3

T HE I NVESTMENT D ECISION A simple example will illustrate both the principle that decision rules depend on the laws of motion that agents face and the extent that standard macroeconomics models have violated this principle. Let k t be the capital stock of an industry and τt be a tax rate on capital. Let τt be the first element of zt , a vector of current and lagged variables, including those that the government considers when it sets the tax rate on capital. We have τt ≡ e T zt , where e is the unit vector with unity in the first place and zeros elsewhere.5 Let a firm’s optimal accumulation plan require that capital acquisitions obey6 ∞

k t = λk t−1 − α ∑ δ j Et τt+ j j =0

(1.1)

α > 0, 0 < λ < 1, 0 < δ < 1 where Et τt+ j is the tax rate at time t that is expected to prevail at time t + j . Equation 1.1 captures the notion that the demand for capital responds negatively to current and future tax rates. However, equation 1.1 does not become an operational investment schedule or decision rule until we specify how agents’ views about the future, Et τt+ j , are formed. Let us suppose that the actual law of motion for zt is zt+1 = Azt

(1.2)

5 Here

T denotes matrix transposition. investment schedule can be derived from the following dynamic model of a firm. A firm chooses sequences of capital to maximize

6 The

∞

E0

∑ βt

t =0



f1 kt −

 f2 2 d k − f 3 k t τt − (k t − k t−1 )2 2 t 2

where f 1 , f 2 , f 3 , d > 0; 0 > β > 1 ; and E0 is the mathematical expectation operator conditioned on information known at time 0 . The maximization is subject to k t−1 , τt being known at time t . Maximization problems of this kind are analyzed in Sargent (1979a). The parameters λ, α , and δ can be shown to be functions of f 1 , f 2 , f 3 , and d .

4

Chapter 1: Rational Expectations

where A is a matrix conformable with zt .7 If agents understand this law of motion for zt , the first element of which is τt , then their best forecast of τt+ j is e T A j zt . We impose rational expectations by equating agents’ expectations Et τt+ j to this best forecast. Upon imposing rational expectations, some algebraic manipulation implies the operational investment schedule k t = λk t−1 − αe T ( I − δA)−1 zt

(1.3)

In terms of the list of variables on the right-hand side, equation (1.3) resembles versions of investment schedules that were fit in the heyday of Keynesian macroeconomics in the 1960s. This is not unusual, for the innovation of rational expectations reasoning is much more in the ways equations are interpreted and manipulated to make statements about economic policy than in the look of the equations that are fit. Indeed, the similarity of standard and rational expectations equations suggests what can be shown to be true generally: The rational expectations reconstruction of macroeconomics is not mainly directed at improving the statistical fits of Keynesian or monetarist macroeconomics models over given historical periods and that its success or failure cannot be judged by comparing the R2 ’s of reconstructed macroeconomics models with those of models constructed and interpreted along earlier lines. Under the rational expectations assumption, the investment schedule (equation (1.3)) and the laws of motion for the tax rate and the variables that help predict it (equation (1.2)) have a common set of parameters, namely, those of the matrix A. These parameters appear in the investment schedule because they influence agents’ expectations of how future tax rates will affect capital. Further, notice that all of the variables in zt appear in the investment schedule, since via equation (1.2) all of these variables help agents forecast future tax rates. (Compare this with the common econometric practice of using only current and lagged values of the tax rate as proxies for expected future tax rates.) The fact that equations (1.2) and (1.3) share a common set of parameters (the A matrix) reflects the principle that firms’ optimal 7 The

eigenvalues of A are assumed to be less than δ−1 in absolute value.

Behavior Changes with the Rules of the Game

5

decision rule for accumulating capital, described as a function of current and lagged state and information variables, will depend on the constraints (or laws of motion) that firms face. That is, the firm’s pattern of investment behavior will respond systematically to the rules of the game for setting the tax rate τt . A widely understood change in the policy for administering the tax rate can be represented as a change in the first row of the A matrix. Any such change in the tax rate regime or policy will thus result in a change in the investment schedule (equation (1.3)). The dependence of the coefficients of the investment schedule on the environmental parameters in matrix A is reasonable and readily explicable as a reflection of the principle that agents’ rules of behavior change when they encounter changes in the environment in the form of new laws of motion for variables that constrain them. To illustrate this point, consider two specific tax rate policies. First, consider the policy of a constant tax rate τt+ j = τt for all j ≥ 0. Then zt = τ1 , A = 1, and the investment schedule is k t = λk t−1 + h0 τt

(1.4)

where h0 = −α/(1 − δ) . Now consider an investment tax credit on-again, off-again tax rate policy of the form τt = −τt−1 . In this case zt = τ1 , A = −1, and the investment schedule becomes k t = λk t−1 + h0 τt

(1.5)

where h0 = −α/(1 + δ) . Here the investment schedule itself changes as the policy for setting the tax rate changes. Standard econometric practice has not acknowledged that this sort of thing happens. Returning to the more general investment example, the usual econometric practice has been roughly as follows. First, a model is typically specified and estimated of the form k t = λk t−1 + hzt

(1.6)

where h is a vector of free parameters of dimension conformable with the vector zt . Second, holding the parameters h fixed, equation (1.6) is used to predict the implications of alternative paths for

6

Chapter 1: Rational Expectations

the tax rate τt . This procedure is equivalent to estimating equation (1.4) from historical data when τt = τt−1 and then using this same equation to predict the consequences for capital accumulation of instituting an on-again, off-again tax rate policy of the form τt = −τt−1 . Doing this assumes that a single investment schedule of the form of equation (1.6) can be found with a single parameter vector h that will remain fixed regardless of the rules for administering the tax rate.8 The fact that equations (1.2) and (1.3) share a common set of parameters implies that the search for such a regime-independent decision schedule is misdirected and bound to fail. This theoretical presumption is backed up by the distressing variety of instances in which estimated econometric models have failed tests for stability of coefficients when new data are added. This problem cannot be overcome by adopting more sophisticated and more general lag distributions for the vector h , as perhaps was hoped in the 1960s. A RE G OVERNMENT D EFICITS I NFLATIONARY ? A second example that well illustrates our general principles about the interdependence of the strategic behavior of private agents and the government concerns the inflationary effects of government deficits. We can discuss this matter with the aid of a demand function for base money of the specific form Mt p = α1 − α2 Et t+1 pt pt

α1 > α2 > 0

(1.7)

where Mt is the stock of base money at time t , pt is the price level at time t and Et (·) is the value of (·) expected to prevail at time t . Equation (1.7) is a version of the demand schedule for money that Phillip Cagan (1956) used to study hyperinflations. It depicts the demand for base money as decreasing with the expected gross rate of inflation, Et pt+1 /pt . A variety of theories imply a demand function for base money with this property. 8 This

is analogous to assuming that Houston’s propensity to punt on fourth down does not depend on the number of downs per series determined by the NFL rules.

7

Behavior Changes with the Rules of the Game

Equation (1.7) is a difference equation, a solution of which is pt =

1 α1

∞

∑

j =0

α2
j Et Mt+ j α1

(1.8)

which expresses the price level at t as a function of the supply of base money expected to prevail from now into the indefinite future. We shall use equation (1.8) as our theory of the price level. According to equation (1.8), if government deficits are to influence the price level path, it can only be through their effect on the expected path of base money. However, the government deficit and path of base money are not rigidly linked in any immutable way. The reason is that the government can, at least to a point, borrow by issuing interest-bearing government debt, and so need not necessarily issue base money to cover its deficit. More precisely, we can think of representing the government’s budget constraint in the form Gt − Tt =

Mt − Mt − 1 + Bt − (1 + rt−1 ) Bt−1 pt

(1.9)

r t −1 ≥ 0 where Gt is real government expenditures at t , Tt is real taxes net of transfers (except for interest payments on the government debt), Bt is the real value at t of one-period government bonds issued at t , to be paid off at t + 1 and to bear interest at the net rate rt . For simplicity, equation (1.9) assumes that all government interest-bearing debt is one period in maturity. Equation (1.9) must hold for all periods t . Again for simplicity, we shall also think of equations (1.8) and (1.9) as applying to an economy with no growth in population or technical change. Under the system formed by equations (1.8) and (1.9), the inflationary consequences of government deficits depend sensitively on the government’s strategy for servicing the debt that it issues. We consider first a strict Ricardian regime in which government deficits have no effects on the rate of inflation. In this regime, the government always finances its entire deficit or surplus by issuing or retiring interest-bearing government debt. This regime can be

8

Chapter 1: Rational Expectations

characterized by either of the following two equations, which are equivalent in view of equation (1.9): Mt − Mt − 1 = 0

(1.10)

∞

1 Bt = Et ∑ R− tj ( Tt+ j+1 − Gt+ j+1 )

(1.11)

j =0

j

where Rtj = ∏i=0 (1 + rt+i ) . Equation (1.10) states that the supply of base money is always constant, while equation (1.11) states that the real value of government debt equals the present value of prospective government surpluses. In this regime a positive value of interest-bearing government debt signals a stream of future government budgets that is in surplus in the present value sense of equation (1.11). Increases in government debt are temporary in a sense made precise in equation (1.11). In the Ricardian regime, government deficits have no effects on the price path because they are permitted to have no effects on the path of base money. For the path of base money to be unaffected by government deficits, it is necessary that government deficits be temporary and be accompanied by exactly offsetting future government surpluses. Since the Ricardian regime may seem remote as a description of recent behavior of the US federal government, it is worthwhile to recall that cities and states in the United States are constitutionally forced to operate under a Ricardian rule, since they have no right to issue base money. There are alternative debt-servicing strategies under which government deficits are inflationary. To take an example at the opposite pole from the Ricardian regime, consider the rule recommended by Milton Friedman (1948) under which Bt = 0 for all t

Gt − Tt =

Mt − Mt − 1 Pt

(1.12)

(1.13)

Behavior Changes with the Rules of the Game

9

According to this rule, deficits are always to be financed entirely by issuing additional base money, with interest-bearing government debt never being issued. In this regime, the time path of government deficits affects the time path of the price level in a rigid and immediate way that is described by equations (1.8) and (1.13). Under this regime it is possible for the government budget to be persistently in deficit, within limits imposed by equations (1.13) and (1.7). Deficits need not be temporary. Bryant and Wallace (1980) and Sargent and Wallace (1981) have described debt-servicing regimes that are intermediate between Ricardo’s and Friedman’s. In all versions of these regimes, interestbearing government debt is issued, but is eventually repaid partly by issuing additional base money. In the regime studied by Sargent and Wallace, the deficit path Gt − Tt is set in such a way, and the demand schedule for interest-bearing government debt is such that eventually the inflation tax must be resorted to, with increases in base money eventually having to be used to finance the budget. In all regimes of the Bryant-Wallace variety, increases in interestbearing government debt are typically inflationary, at least eventually because they signal prospective increases in base money. Sooner or later these eventual increases in base money will affect the price level, how soon depending on the coefficients α1 and α2 in equations (1.7) and (1.8). This discussion indicates that the observed correlation between the government deficits and the price level depends on the debtrepayment regime in place when the observations were made. It would be a mistake to estimate the relationship between the deficit and the price path from time-series observations drawn from a period under which a Ricardian regime was in place, and to assert that this same relationship will hold between the deficit and inflation under a regime like the one described by Bryant and Wallace. It would be a mistake because private agents’ interpretations of observed deficits, and consequently the impact of observed deficits on the price level, depend on the debt-servicing regime they imagine to be in place. This can thus be viewed as another example of our principle that private agents’ rules of behavior depend on their perceptions of the rules of the game they are playing.

10

Chapter 1: Rational Expectations

General Implications of the Examples

These two examples illustrate the general presumption that the systematic behavior of private agents and the random behavior of market outcomes both will change whenever agents’ constraints change, as when government policy or other parts of the environment change. To make reliable statements about policy interventions, we need dynamic models and econometric procedures that are consistent with this general presumption. Foremost, we need a new and stricter definition of the class of parameters that can be regarded as structural. The body of doctrine associated with the simultaneous equations model in econometrics properly directs the attention of the researcher beyond reduced-form parameters to the parameters of structural equations that are meant to describe those aspects of people’s behavior that remain constant across a range of hypothetical environments. Although such structural parameters are needed to analyze an interesting class of policy interventions, most often included among them have been parameters of equations describing the rules of choice for private agents. Consumption functions, investment schedules, and demand functions for assets are all examples of such rules of choice. In dynamic settings, regarding the parameters of these rules of choice as structural or invariant under policy interventions violates the principle that optimal decision rules depend on the environment in which agents believe they are operating. If parameters of decision rules cannot be regarded as structural or invariant under policy interventions, deeper objects that can must be sought. The best that can be hoped for is that parameters characterizing private agents’ preferences and technologies will not change when changes in economic policy change the environment. If dynamic econometric models were formulated explicitly in terms of the parameters of preferences, technologies, and constraints, in principle they could be used to predict the effects on observed behavior of changes in policy rules. In terms of our investment example with equations (1.2) and (1.3), the idea would be to estimate the free parameters of the model ( λ, α, δ, A ). With these estimates, economists could predict how the investment schedule

New Econometric Methods

11

would change if different A ’s occurred.9

New Econometric Methods Private Agents’ Strategies Reflect Government’s Choice of Rules of the Game A major research effort is currently under way by economists to develop theoretical and econometric methods capable of isolating parameters that are structural in the above sense, that is, they are invariant under government interventions in the form of changes in the rules of the game. This is a very ambitious undertaking, one that is in many ways more difficult and ambitious than was the impressive effort of the Cowles Commission in the late 1940s that created the econometric methods that made Keynesian econometric models possible. For what is required is a theoretical and statistical framework that permits the economist to estimate how private agents’ decision rules or strategies depend on the decision rules or strategies used by the government. Any successful version of this research effort will embody the principle that the parameters of private agents’ decision rules are not among the free parameters of the model, but are themselves functions of (among other things) parameters describing the rules used by the government. Achieving success in this endeavor requires that many new methods and results be achieved in technical aspects of econometrics and dynamic economic theory. Lucas and Sargent have formalized the ideas behind this research effort in the following way. They let h denote a collection of decision rules of private agents. Each element of h is itself a function that maps some private agent’s information about his state at a particular point in time into his decision at that point in time. Consumption, investment, and demand functions for money are all examples of elements of h . Lucas and Sargent let f denote a collection of elements that forms the “environment” facing private agents. Some elements of f represent rules of the game or decision 9 As

claimed in note 8, the parameters λ, α, δ can be shown to be functions of the parameters f 1 , f 2 , f 3 , d of the present value function being maximized in the equation.

12

Chapter 1: Rational Expectations

rules selected by the government, which map the government’s information at some date into its decisions at that date. For example, included among f might be decision rules for fiscal and monetary policy variables. The fundamental principle that we are concerned with can be summarized as stating that the elements of h are partly functions of f . Lucas and Sargent represent this mapping formally by h = T (f)

(1.14)

The mapping T represents “cross-equation restrictions,” since each element of h and of f is itself a decision rule or equation determining the choice of some variable under an agent’s control. The new econometric methods have been aimed principally at utilizing time series of observations on an economy that was operating for some period under a single set of government rules or strategies.10 The idea is to impose sufficient structure on the observations (i.e., sufficient structure on the mapping T in equation (1.14)) that by observing the decisions h of private agents and f for the government, we can isolate the free parameters of agents’ preferences and constraints that determine T and that will enable us to predict how private agents’ decision rules h would change if the government were to adopt some new and perhaps historically unprecedented rules f . The theoretical models that enable one to hope to carry off this task are characterized by “cross-equation restrictions,” across equations for the decision rules of private agents and the decision rules of the government. These restrictions summarize the dependence of private agents’ strategies on the government’s strategies. To understand the economic process, in the sense of being able to predict the consequences of changes in the rules of the game set by the government, is to understand these restrictions. The hope is that by utilizing these restrictions, observations on an economy operating under a single set of rules 10 If

sufficient data are available, these same econometric methods can be readily modified to pooling observations on hi , f i from several regimes i = 1, · · · , n in order to estimate T . The reader of Hansen and Sargent (1980) can immediately see how samples of data drawn from different hi , f i can be pooled using either the maximum likelihood or method of moments estimator.

Historical and Cross-Country Analysis

13

can be interpreted and used to make predictions about how the economy would behave under brand new rules. Possessing the ability to do this is a sine qua non for scientific and quantitative evaluation of alternative government rules. Impressive technical progress is being made in this research endeavor, and there are grounds for being reasonably optimistic that economists will eventually be able to deliver econometric methods that are useful for predicting the effects of changes in government policy rules. However, it is wise to keep in mind what an ambitious task this is. It requires several leaps of faith in economic theory to hope that this line of research will realize all of its aims, although the success of this research line will eventually have to be judged not in absolute terms but vis-à-vis alternative lines of research designed to achieve the same objectives. Though they are being developed rapidly, the new methods described above are still in their infancy, and as yet capable of handling only relatively small and simple dynamic systems. We now have several interesting empirical applications of these methods, which demonstrate their feasibility. But we are still some way from a model suitable for analyzing many interesting questions of macroeconomic policy. This is one reason that it is useful to pursue alternative styles of analysis that try to reflect the basic principle that private agents’ behavior depends on the rules of the game set by the government. This brings us to economic history.

Historical and Cross-Country Analysis History provides a number of examples of economies that have apparently operated for more or less extended periods of time under alternative rules of the game or government strategies for selecting fiscal, monetary, and regulatory actions. By studying records of these economies, one can hope to find direct evidence of the dependence of private agents’ decision rules h on those aspects of the environment f selected by the policy authorities. These historical records amount to distinct pairs of observations h1 , f 1 ; h2 , f 2 ; · · · ; hn , f n that can be viewed as observations at different points of the mapping T given by equation (1.14). In a concrete

14

Chapter 1: Rational Expectations

sense such a collection of cross-regime observations permits direct observations on at least parts of the T mapping. These are the evident advantages of the historical or crosscountry analysis. There are also evident disadvantages. The observations on different hi , f i pairs usually come in the form of observations from distant times and/or places. For this reason, the data are often so fragmentary and questionable in quality that they are incapable of supporting the kind of formal time-series econometric analysis described above. Less formal methods of analysis must be used for such data. To the extent that these less formal methods require more judgment, discretion, and cleverness from the analyst, they are in a sense less reproducible and automatic than are the formal methods. The weaving of plausible heuristic stories this style of analysis involves is one of the things that the formal time-series methods developed by rational expectations analysts and their econometric predecessors sought to avoid. Furthermore, it is often an arguable matter to identify alternative countries or historical periods as operating under distinct government policy regimes or rules of the game f . Despite these genuine disadvantages of the historical and crosscountry method of study, it is my belief that such studies are well worth the effort. The hope of getting some direct peeks at distinct observations, however fragmentary and noisy, on the h = T ( f ) mapping easily justifies appreciable efforts in this direction. It is this hope that motivated the three historical chapters that comprise the heart of this book. Besides, the practical success of the rational expectations approach to macroeconomics perhaps ultimately depends on whether it can become a routine device that can be used to think about macroeconomic problems informally and on the backs of envelopes. These chapters represent attempts to put the theory into practice in such an informal context.

Implications for Policymakers

15

Implications for Policymakers These ideas have implications not only for theoretical and econometric practices but also for the ways in which policymakers and their advisers think about the choices confronting them. In particular, the rational expectations approach directs attention away from particular isolated actions and toward choices among feasible rules of the game, or repeated strategies for choosing policy variables. While Keynesian and monetarists macroeconomic models have been used to try to analyze what the effects of isolated actions would be, it is now clear that the answers they have given have necessarily been bad, if only because such questions are ill-posed. In terms of our investment example, by selecting different values for the first row of A , we can analyze the effects on current and subsequent investment of switching from one well-understood policy for setting the tax rate to another—that is, we can analyze the effects of different strategies for setting the tax rate. However, we cannot analyze the effects on current and subsequent investment of alternative actions consisting of different possible settings for the tax rate τt at a particular point in time t = t . For in order to make predictions, we must specify agents’ views about the law of motion A , and this is not done when we simply consider actions consisting of alternative settings for τt at one isolated point in time. This idea is so widely accepted as to be uncontroversial among decision theorists (and football fans); but even today practicing macroeconomists often ignore it. To take a concrete example, in the United States there was recently interest in analyzing what would happen to the rate of domestic extraction of oil and gas if the tax on profits of oil producers increased a lot on a particular date. Would supply go up or down if the tax were raised to X percent on July 1? The only scientifically respectable answer to this question is “I don’t know.” Such a rise in the oil-profits tax rate could be interpreted as reflecting one of a variety of different tax strategies ( A matrices), each with different implications for current and prospective extraction of oil. For example, suppose that oil companies had reason to believe that the increase in the tax is temporary and will be repealed after the election. In that case, they would respond by decreasing their

16

Chapter 1: Rational Expectations

rate of supply now and increasing it later, thus reallocating their sales to periods in which their shareholders get a larger share of profits and the government a smaller share. Yet suppose that oil companies believed that the increase in the tax rate on July 1 is only the beginning and that further increases will follow. In that case the response to the tax rate increase would be the reverse: to increase supply now and decrease it later in order to benefit companies’ shareholders. This example illustrates that people’s views about the government’s strategy for setting the tax rate are decisive in determining their responses to any given actions and that the effects of actions cannot be reliably evaluated in isolation from the policy rule or strategy of which they are an element. Another version of this example can easily be constructed for our deficit example around the question “Are large current government deficits accompanied by tight monetary policy actions inflationary?” This question is also ill-posed because it fails to specify the debt repayment strategy to be used by the government. The way in which current deficits are correlated with inflation depends sensitively on what debt repayment regimes is in place. What policymakers (and econometricians) should recognize, then, is that societies face a meaningful set of choices about alternative economic policy regimes. For example, the proper question is not about the size of tax cut to impose now in response to a recession but about the proper strategy for repeatedly adjusting tax rates in response to the state of the economy, year in and year out. Strategic questions of this nature abound in fiscal, monetary, regulatory, and labor market matters. Private agents face the problem of determining the government regime under which they are operating, and they often devote considerable resources to doing so. Whether governments realize it or not, they do make decisions about these regimes. They would be wise to face these decisions deliberately rather than ignore them and pretend to be able to make good decisions by taking one seemingly unrelated action after another.

2 Reaganomics and Credibility

H OWARD: Dandy, the Vikings had the momentum throughout the first half. Let’s see if the momentum stays with the Vikes into the second half. Otherwise, it’s going to be a long night for the Bears. D ANDY D ON: Yes, Howard. It’ll also be interesting to see if the Bears continue to be confused by the new formation that Bud Grant has installed for this game. We haven’t seen the Vikings throw play-action passes as much as they have this evening, and this has surprised the Bears. If the Bears can figure out the Vikes’ new strategy and adjust to it, it will be a new ball game. Monday Night Football

An offensive football team is a collection of individuals with a common objective: to score a touchdown. This objective is attained by the cooperation of eleven players, each of whom is ultimately in control of his own actions. The effectiveness of any one player’s actions depends intricately on the actions of his teammates. If the quarterback decides to throw the ball 30 yards downfield to the right side on a count of three, it is necessary for success that a receiver run a pass pattern that will place him in a position to catch the ball. If the quarterback calls a “keeper” and runs around the end, it is important that the end not run a pass pattern but that he block. Thus, the quarterback and the end, and all the other players, face a problem of coordination. It will not do simply to announce vague objectives in the huddle, such as “Let’s score a touchdown.” Instead, somehow each player must reach a precise understanding of what each of his colleagues is planning to do on the next play and of the “contingency plans” that each player will use as the play 17

18

Chapter 2: Reaganomics and Credibility

develops or breaks down. All football teams (except apparently one that I root for) accomplish this coordination task by giving one player, either the quarterback or a player just sent in by the coach, the authority to direct the actions of all the others by calling the play. A football team is an example of a system for which complete decentralization or laisser-faire is not a good idea. The example from football contains important lessons about making macroeconomic policy. Within a single country, the authorities who are charged with responsibility for making monetary and fiscal policy are very much in the position of the end and the quarterback, for their activities must be coordinated, one way or the other, and their objectives are presumably identical. In the world as a whole, the monetary and fiscal authorities of different countries have to somehow coordinate their policies, since one country’s choices of monetary and fiscal strategies influence the options open to the others so long as there is some freedom to exchange goods and make loans across borders. However, despite the interrelated consequences of their actions upon a common system, fiscal and monetary authorities from different countries sometimes have differing and even opposing goals. (Sometimes the goals may even seem so opposed that the proper analogy is not to a quarterback and an end but to two opposing football teams— say the Cowboys and the Redskins). Presumably, the example of the quarterback and the end rings truest for the coordination of monetary and fiscal policy within one country, for the assumption that the authorities share common objectives is better here than in the international case. This chapter views the monetary and fiscal authorities of a single country as a “team” and judges their patterns of behavior against standards absorbed from the sports pages. This view provides a broad framework for summarizing classic doctrines and controversies in government finance, and also serves as a basis for criticizing the way in which monetary and fiscal policies have been coordinated de facto in the United States over the last several years. I shall begin with a few formal definitions of concepts that will help to clarify the analogy between the quarterback-end problem and the monetary-fiscal problem.

Dynamic Games

19

Dynamic Games A game consists of a collection of players and a set of rules specifying rewards and penalties. A dynamic game is one that requires time to complete and whose current score depends on past actions of the various players. In life, most games are dynamic. Each player in a game is supposed to have a goal or objective that depends on the rewards and penalties specified in the rules of the game. This goal may be idiosyncratic (such as personal glory or personal profit) or altruistic (such as the success of one’s team or country). A team game is one in which two or more players have a common objective. Football and soccer are team games. So, perhaps, is the game of managing a country’s monetary and fiscal affairs, at least if those in charge have in mind a common objective. Each player in a dynamic game tries to achieve his objective by choosing a strategy. A strategy is defined as a rule that describes how a player’s actions during the game depend on the information he receives during the course of the game. Another term for a player’s strategy is contingency plan. This term evokes the notion that each action taken by a player ought to depend on the situation as it is understood when that action is executed. A strategy relates a player’s actions over time into a sensible pattern. Since time elapses during a dynamic game, whether a single action (or move) is a good one cannot be judged in isolation from past and subsequently planned moves. In general, each player chooses his strategy given his perception of the strategies of other players and given his perception of the influence that his own choice of strategy has on the strategies chosen by other players. Patterns of influence of one player’s strategy upon the strategies of others are determined by the sizes of the various players and by the timing of moves. To complete a description of a dynamic game, it is necessary to specify the timing of moves precisely. An equilibrium or solution of a dynamic game is a collection of strategies of all the players in the game that maximizes their respective objective functions, subject to each player’s perception of the strategies of all the remaining players. Evidently, a solution of a dynamic game imposes that all players’ strategies be feasible,

20

Chapter 2: Reaganomics and Credibility

in the sense of being consistent with the physical technologies in place and with the strategies being employed by the other players (or agents). Alternative rules about the timing of moves give rise to different ways of playing a game, or really different games. One common assumption about timing in a dynamic game has all of the players moving simultaneously during each period that the game endures. This kind of assumption is used to model situations in which none of the agents has access to a “commitment technology” that would enable it to enter into binding precommitments, i.e., to make agreements now to execute particular actions in the future. However, sometimes we do want to study what happens when one or more players does have the capacity to precommit. We model this situation by permitting the precommitting player to move only once, at the beginning of the game. At the beginning of the game, this player is given the opportunity to choose a contingency plan describing what he will do during the entire course of the game, in all future periods and in all possible contingencies. Having made his choice, this player has to “walk away from the table,” and is given no opportunity subsequently to reconsider his decision. The remaining players are permitted to choose sequentially. These players are thus given multiple “moves”: they are given the opportunity to decide what to do during each time period that the game endures. Sometimes a game with this timing structure—one player moving once and for all at the beginning of the game, with his opponents moving sequentially—is called a Stackelberg game, and the player who moves once at the start of the game is called the Stackelberg leader. A principle that emerges from the study of dynamic games is that players’ strategies are interdependent. Interdependence of strategies generally holds regardless of the structure of dominance, though the exact forms of dependence will hinge on it. In football, the principle of strategic interdependence is reflected in the need for a quarterback and an end to coordinate their strategies. It is also reflected in the coordination problem facing monetary and fiscal authorities. The reader who is familiar with the game of football will be able

An Economy as a Dynamic Game

21

to recognize how the categories defined above apply to football. From the viewpoint of a single football team, football is a dynamic team game in which each player’s optimal strategy depends on the optimal strategies of the other players as well as on the strategy being used by the opposing team and the rules set by the league. The optimal strategy for a given team depends on the rules of the game and on the strategies chosen by the opposing team. Since my main purpose is to analyze the macroeconomy and not the National Football League, I will not pursue the analysis of football as a dynamic game any further here. Instead, I shall now describe aspects of the economy of a single country as a dynamic game.

An Economy as a Dynamic Game The economy or game is imagined to consist of three players: the public, the monetary authority, and the fiscal authority. The public consists of people, who are organized into households, agencies of the government, and corporations and who are the ultimate beneficiaries of all economic activity. The public makes decisions about consumption, investment, and private employment and pays the taxes imposed on it by the fiscal authority. The public also sets the terms on which it will accumulate government debts of various forms. The fiscal authority makes decisions about public expenditures and about the rates at which taxes are to be collected from the public. By making these decisions, the fiscal authority determines the rate of government deficit—the amount by which government expenditures exceed tax collections.1 The deficit is financed by the issuing of government debt, either in the interest-bearing form of government bonds or in the non-interest-bearing form of currency and bank reserves (often called “high-powered money”). The decision about the composition of the debt as between bonds of various maturities and currency or high-powered money is, at each point in time, under the control of the monetary authority. The monetary authority exercises this control through its authority to engage in open-market exchanges of one kind of public debt for another. Thus, while the fiscal authority influences the rate of 1 Technically,

this is the government deficit net of interest payments.

22

Chapter 2: Reaganomics and Credibility

addition to the public debt, the monetary authority determines its composition. Debt management is a term that is aptly used to describe what the monetary authority does.2 Macroeconomic analysis of the rational expectations variety aims to study the interactions of these three classes of agents as a dynamic team game. Abstracting from distributional effects across members of the public (according to a long tradition in macroeconomics), the monetary and fiscal authorities are imagined to share common objectives with the public and with each other. These common objectives make it a team situation. The aspect that all three players are making decisions that affect the future state of the system makes it a dynamic game. Thus, the public chooses investment rates in physical and human capital and the terms on which it is willing to accumulate various amounts and types of government debt; the fiscal authority determines the current and the prospective state of total government indebtedness; and the monetary authority determines the composition of the debt. I now put this structure of ideas to work by using it to analyze a classic issue of government finance that is important today: the inflationary consequences of government deficits and of alternative ways of financing them. Government expenditures can be financed by alternative combinations of levying taxes, borrowing in interest-bearing form, and printing high-powered money. The consequences for the price level path of alternative methods of financing a given stream of government expenditures can differ, and in ways that depend on how the strategies of the public and of the fiscal and monetary authorities are imagined to interact. To discuss these consequences, we need models of the decision strategies of each of our three groups of agents and of their interactions. We can describe some of the major issues with the aid of simple strategic models for each of our players. In the tradition of Keynes, the public is assumed to be willing to hold interest-bearing government debt on the same terms on 2 Since its decisions about the composition of the debt can influence the interest payments that the government must make, the monetary authority helps determine the government deficit gross of interest payments and thereby the rate at which total government debt changes.

An Economy as a Dynamic Game

23

which it holds private evidences of indebtedness. This means that public borrowing is assumed to pay the same interest rate as private borrowing and that the total amount of government and private borrowing must be consistent with the public’s limited capacity to accumulate wealth. Let us assume that all interestbearing government debt is one period in maturity, and let us denote the one-period real pretax net rate of return on private securities between t and t + 1 as r (t) . We assume that r (t) is an exogenous sequence and that r (t) is greater than 0. For simplicity, we assume an economy that is not growing over time. We also abstract from uncertainty. The public’s willingness to accumulate real interest-bearing government debt, B(t) , is assumed to be limited. In particular, we assume that B(t) is constrained by B(t) ≤ B¯

(2.1)

Equation (2.1) asserts that, like all private borrowers, the government is faced with an upper bound on the amount of debt that it can place. One upper bound on B(t) is total wealth in a country. When all savings of a country have been absorbed in government debt, no more government debt can be placed. In practice, the actual upper bound B¯ is far lower than the total wealth. In August 1982, B(t) appeared to have hit B¯ . In France, between 1924 and 1926, B(t) appeared to have been close to B¯ , precipitating a continuing financial crisis and the “waltz of the portfolios” of the finance ministers of France. The public’s willingness to accumulate base money is assumed to be described by a demand function of the specific form h p ( t + 1) i M(t) a (1) > a (2) ≥ 0 = a(1) − a(2) Et p(t) p(t)

(2.2)

where M (t) is the stock of base money at time t , p(t) is the price level at time t , and Et [·] is the value of [·] expected to prevail by the public as of time t . Equation (2.2) is a version of the demand function for money that Cagan (1956) used to study hyperinflation. It depicts the demand for real base money as a decreasing function of the expected

24

Chapter 2: Reaganomics and Credibility

gross rate of inflation Et [ p(t + 1)/p(t)] . A variety of theories imply a demand function for base money of this form. There is also ample empirical evidence that is consistent with the inverse dependence between real balances M (t)/p(t) and expected inflation Et [ p(t + 1)/p(t)] that is posited by equation (2.2). For example, in the year before August 1946, the price level in Hungary increased by a factor of about 4 × 1024 . It is reasonable to expect that people had caught on to the extraordinarily rapid ongoing inflation, so that Et [ p(t + 1)/p(t)] was large by the middle of 1946. In August 1946 the real value of high-powered money M(t)/p(t) in Hungary, measured in 1946 US dollars, was less than $25,000. The system that emerges from writing down the version of equation (2.2) that is appropriate for dates t, t + 1, t + 2, . . . can be solved to express p(t) solely in terms of expected future values of M(t) : p(t) =

1 a (1)

∞

∑

j =0

 a (2)  j a (1)

Et M(t + j)

(2.3)

Equation (2.3) expresses the price level at t as a function of the supply of base money expected to prevail from now into the indefinite future. The logic underlying this equation is simple: Equation (2.2) implies that the price level at t varies directly with the money supply at t and with the price level expected to prevail at t + 1. Equation (2.2) also implies that the price level at t + 1 varies directly with the money supply at t + 1 and with the price level expected to prevail at t + 2, and so on. Upon the elimination of future expected price level from this infinite sequence of relationships, equation (2.3) emerges. Notice that in the special case of a(2) = 0, equation (2.3) becomes a simple version of the quantity theory of money, stating that the price level at t is proportional to the supply of high-powered money at t . Equation (2.3) shows how the price level at t is determined by the interaction of the public’s preference for holding high-powered money, which is reflected in the parameters at a(1) and a(2) , with the expected path of high-powered money now and into the indefinite future. According to equation (2.3), if government deficits are to influence the price level, it can only be through their effects on

An Economy as a Dynamic Game

25

the expected path of high-powered money. In this sense, equation (2.3) embodies the monetarist presumption that inflation is always a monetary phenomenon. The government deficit and the level and rate of change of the stock of base money are not related in any necessary way at a particular point in time. The reason is that the government can (at least up to a point) borrow by issuing interest-bearing debt, and so need not issue base money to cover its deficit. More precisely, we can think of representing the government’s budget constraint in the form G (t) − T (t) =

M ( t ) − M ( t − 1) p(t)

(2.4)

+ B(t) − [1 + r (t − 1)] B(t − 1) where G (t) is real government expenditures at t, T (t) is real tax collections net of transfers (except for interest payments on the government debt), and B(t) is the real value of t of one-period bonds issued at t , to be paid off at t + 1 and to bear interest at net real rate r (t) . Equation (2.4) asserts that the real government deficit at t, G (t) − T (t) , can be financed by a combination of printing new high-powered money in the amount M(t) − M(t − 1) , which raises [ M(t) − M (t − 1)]/p(t) in real resources, and borrowing in interest-bearing form B(t) in excess of the principle and interest on the debt that is maturing, [1 + r (t − 1)] B(t − 1) . Equation (2.4) must hold for all t . For simplicity, equation (2.4) assumes that all government interest-bearing debt is one period in maturity. It is important to point out that the formulation of equation (2.4) assumes in effect that government debt is indexed and constitutes a sure claim on given amounts of future goods. Either the debt is regarded as explicitly indexed or else the bonds are nominal ones, with the nominal rate of interest imagined to adjust by the subsequently realized rate of inflation so that they turn out to bear real rate r (t) in equilibrium. In a rational expectations model in which there is no objective uncertainty—the kind of model we have in mind here—these two interpretations are equivalent. In either one of these interpretations, the government is imagined to honor its commitments to repay interest-bearing debt at the real interest rate that was anticipated at the time the debt was

26

Chapter 2: Reaganomics and Credibility

contracted. In reality, when part of the outstanding government debt is nominal, the government has the option of “defaulting” on part of it by acting so as to inflate at a higher rate than had been expected when the debt was contracted. In the subsequent presentation I shall begin by assuming that the government always abstains from defaulting on any of its interest-bearing debt.3 Imagine that there is a fiscal authority that selects a time stream of G (t) and T (t) . A consequence of the fiscal authority’s choice is a stream of government deficits net of interest payments, G (t) − T (t) . There is also a monetary authority that determines the composition of the government debt in the hands of the public through open-market operations. The monetary authority’s open-market operations at time t are subject to a constraint, which is derived by simply rearranging equation (2.4): M (t) + p(t) B(t) = M(t − 1) + [1 + r (t − 1)] p(t) B(t − 1)

+ p(t)[ G (t) − T (t)] The monetary authority is free to choose M(t) and p(t) B(t) subject to the constraint that they add up to the total on the right-hand side of the preceding equation. In other words, at a point in time the monetary authority can exchange base money for bonds of equal value.

Are Government Deficits Inflationary? Under the system formed by equations (2.3) and (2.4), the inflationary consequences of a government deficit at time t depend 3 Another

way to write equation (2.4) is G ( t ) − T ( t ) + r ( t − 1) B ( t − 1)

=

M ( t ) − M ( t − 1) + B ( t ) − B (1 − 1) p(t)

The term G (t) − T (t) + r (t − 1) B(t − 1) is often called the government deficit gross of interest payments, while G (t) − T (t) is termed the government deficit net of interest payments. The monetary authority is assumed to control the ratio of B(t) to M (t) at each point in time. It thereby influences the subsequent rate of growth of total government indebtedness by influencing the interest expenses r (t − 1) B(t − 1) that appear in equation (2.4).

Are Government Deficits Inflationary?

27

sensitively on the government’s strategy for servicing the debt that it issues. This dependence can be illustrated by considering two polar regimes for servicing the debt and for coordinating monetary and fiscal policy. Consider first a strict Ricardian regime in which government deficits have no effects on the rate of inflation. In this regime, the government always finances its entire deficit or surplus by issuing or retiring interest-bearing debt. Additional base money is never issued to finance a deficit. This regime can be characterized by either of the following two equations, which are equivalent in view of equations (2.4) and (2.1): M ( t ) − M ( t − 1) = 0

(2.5)

∞

B(t) = Et ∑ Rtj−1 [ T (t + j + 1) − G (t + j + 1)]

(2.6)

j =0

where j

Rtj =

∏[1 + r(t + i)],

(2.7)

i =0

and where (2.5) and (2.6) are required to hold for all t . Equation (2.5) states that the supply of base money is always a constant; equation (2.6) states that the real value of interest-bearing government debt at t equals the present value of prospective government surpluses. In this regime a positive value of interest-bearing government debt signals a stream of future government budgets that is in surplus in the present value sense of equation (2.6). Increases in government debt are necessarily temporary, in a sense made precise by equation (2.6).4 In the Ricardian regime, government deficits have no effects on the price path because they are permitted to have no effects on the 4 One

way to implement the regime given by equations (2.5) and (2.6) is simply to adjust current taxes T (t) by an amount equal to any variations in interest payments r (t − 1) B(t − 1) that are associated with variations in past government expenditures or taxes. This policy amounts always to levying current taxes sufficient to service the interest payments that are currently due. This is the way McCallum (1984) proceeds in one of his experiments.

28

Chapter 2: Reaganomics and Credibility

path of base money. For the path of base money to be unaffected by government deficits, it is necessary that the government deficits be temporary and be expected to be accompanied by offsetting future government surpluses. In the Ricardian regime, the government behaves like a firm with respect to financing its deficit. To finance a given deficit, the government competes for funds from lenders on an equal footing with private borrowers. To attract funds, the government must offer lenders a prospective stream of net revenues sufficient to support the value that it presently proposes to borrow. The government’s stream of net revenues is T (t) − G (t) . The present value of this stream forms the “backing” for the government’s borrowing, just as the present value of a stream of prospective net revenues from a new machine might form the backing for a private loan. Furthermore, like any private borrower, the government can borrow in interest-bearing form only a limited amount determined by the maximum present value of the prospective government surpluses that the economy can support. This is the limit B¯ embodied in equation (2.1). The Ricardian regime may seem remote as a description of recent behavior of the US government and some major US trading partners. It is worthwhile to recall that states and cities in the United States are constitutionally required to operate under a Ricardian rule, since they have no right to issue base money. In the nineteenth century the Ricardian rule was followed, with temporary lapses, by Great Britain, the United States, and the more advanced European countries. (It is no coincidence that the economically advanced countries all adopted such a rule and that they all abandoned it at about the same time, during and after World War I. There are irresistible forces impelling countries that trade with each other to coordinate their monetary and fiscal policies. These forces often cause countries to run similar fiscal policies. A country had to follow a Ricardian rule, or something close to it, in order to adhere to the international gold standard.) Alternatives to the Ricardian debt-servicing regime exist under which government deficits are inflationary. To take an example at the opposite pole from the Ricardian regime, let us consider a rule that was followed for a while during the great revolutions in France and Russia, was used during each of the great European

Are Government Deficits Inflationary?

29

hyperinflations of the twentieth century, and was advocated in one version by Milton Friedman in 1948. This rule can be characterized by either of the two following equations: B(t) = 0 G (t) − T (t) =

(2.8)

M ( t ) − M ( t − 1) p(t)

In view of the government budget constraint (equation (2.4)), these two equations are equivalent characterizations of a rule in which the entire deficit is always immediately financed by the printing of additional base money. Interest-bearing debt is never issued. In this regime, the time path of government deficits affects the time path of both base money and the price level in a rigid and immediate way that is described by equation (2.8) and by our theory of the price level, equation (2.3). Under this debt-servicing regime, it is possible for the government budget to be persistently in deficit, within limits imposed by equation (2.8) and by the demand function for base money (equation (2.2)) or its implied theory of the price level (equation (2.3)). Deficits need not be temporary. In this regime, the government finances a current deficit not by a promise to run surpluses in the future, as in the Ricardian regime, but instead by levying an immediate “inflation tax” on the present holders of base money. Whereas the Ricardian regime involves a commitment ultimately to abstain from any resort to an inflation tax, the polar alternative involves a promise that any government deficit will be immediately and fully monetized. I shall return later to the question of why Milton Friedman, who has never been an advocate of monetary regimes leading to rampant inflation, would at one time have advocated full monetization of government deficits—the regime that has accompanied the worst inflations in history. It is possible to imagine deficit-financing regimes that are intermediate between Ricardo’s and Friedman’s. Bryant and Wallace (1980) and Sargent and Wallace (1981) have described such regimes. In all versions of these regimes, interest-bearing government debt is issued, but it is eventually repaid at least partly by the issuing of additional base money. In the regime studied by Sargent

30

Chapter 2: Reaganomics and Credibility

and Wallace, the deficit path involves such a persistent stream of large deficits that eventually the inflation tax must be resorted to; increases in base money have to be used to finance the budget. In all these intermediate deficit-financing regimes of the BryantWallace variety, increases in interest-bearing government debt are typically inflationary, at least eventually, because they signal eventual increases in base money. Sooner or later, these perspective increases in base money will increase the price level; how soon depends on the coefficients a(1) and a(2) in equation (2.2). According to equation (2.3), the closer a(2)/a(1) is to unity, the bigger is the effect of a given future increase in base money on the price level today. This is true because, according to equation (2.2), the larger a(2) is relative to a(1) , the more sensitive the current price level is to the expected future price level and therefore also to expected future values of base money. The preceding discussion indicates that the observed correlation between government deficits and the price level depends on the debt-repayment regime that was in place when the observations were generated. On the one hand, under a Ricardian regime, deficits and the price level would be uncorrelated, because government deficits would not cause movements in the stock of base money. On the other hand, under Friedman’s regime, deficits would be highly correlated with the price level. It would therefore be a mistake to estimate the relationship between the deficit and the price path from time-series observations drawn from a period under which a Ricardian regime was in place and to assert that this same relationship will hold between the deficit and inflation under a regime like that described by Friedman or Bryant and Wallace. It would be a mistake because private agents’ interpretations of observed deficits, and consequently the impact of observed deficits on the price level, depend on the debt-servicing regime that they imagine to be in place. The Ricardian regime, Friedman’s 1948 regime, and the intermediate Bryant-Wallace regimes each involve solutions of one kind or another to the problem of coordinating the actions of the monetary and fiscal authorities. The government budget constraint (equation (2.4)) implies that coordination is necessary, for a monetary authority, by virtue of its control over the division of the government debt

Are Government Deficits Inflationary?

31

in the hands of the public between interest-bearing debt and base money, controls the flow of revenues from the inflation tax that can be used to cover current and future deficits. In principle, the monetary authority has the power to force the system into the Ricardian regime by simply refusing to monetize any interest-bearing government debt. The fiscal authority would thereby be compelled to place its debt with private lenders, presumably by competing on an equal footing with other borrowers. Under each of the debt-servicing regimes described so far, interest-bearing government debt has been assumed in effect to be indexed. As mentioned earlier, either the debt is regarded as explicitly indexed, or else the rationality of the public’s price level expectations and the fact that the government is imagined to adhere to policies or entire time paths of G (t) − T (t, M (t) , and B(t) mean that a system with nominal government debt behaves just like a system with indexed government bonds.5 Since we want to apply the results of our reasoning to recent US experience in which government borrowing is in nominal terms, it is important to stress the aspect of the preceding regimes that government plans are adhered to. Though in some of the above regimes the government may resort to an inflation tax, it is known in advance that the government plans to do so. There is no element of fraud or deception in the inflation generated under such regimes. However, when all or part of the government interest-bearing debt is in nominal form, at each point in time the government appears to have the option of defaulting on part of the debt by inflating at a rate greater than had initially been expected. In the context of the rational expectations assumption that we are working with here, inflation at a greater rate than had initially been expected by the public is brought about when the government departs from an initial plan for G (t) − T (t, M(t) , and B(t) that was thought by the public to be in place and embarks on a plan implying a higher price level for the present period than had originally been anticipated. Resorting to this option is a form of default because it gives holders of interest-bearing government debt and base money different real 5 This

is one of the findings of the theoretical literature on indexed government bonds: that, under general circumstances, they make no difference to a rational expectations equilibrium. See Liviatan (1983) or Peled (1980).

32

Chapter 2: Reaganomics and Credibility

rates of return than they initially had bargained for on the basis of the originally planned time paths of G (t) − T (t), M(t) , and B(t) . This default option can be represented by reformulating the government budget constraint (equation (2.4)) in terms of the nominal interest rate on government interest-bearing debt. Let rn (t) be the nominal interest rate on one-period debt from period t to t + 1. Then the real rate of interest r (t − 1) is related to rn (t − 1) follows: 1 + r (t − 1) = [1 + rn (t − 1)]

p ( t − 1) p(t)

This equation states that, with a previously fixed nominal rate of interest, the realized real rate of interest between t − 1 and t is lower the higher is the price level at t . We can use the above equation to write the government budget constraint as M ( t ) − M ( t − 1) + B(t) p(t) p ( t − 1) − B(t − 1)[1 + rn (t − 1)] p(t)

G (t) − T (t) =

(2.9)

This equation shows how generating a higher price level than had previously been expected helps to finance current deficit and to diminish the need to sell new government debt. There are serious questions about whether, and under what circumstances, a government should resort to the option of defaulting that is present when part of the debt is in nominal terms. There is also a serious question of the scope that a government actually has for repeatedly resorting to the default option.6 Presumably, a government that once reneges on its plans for G (t) − T (t), M(t) , and B(t) is less likely to be trusted the next time. The public can be expected to evaluate subsequent government plans and announcements against the background of the government’s reputation for executing previous plans. Prospective lenders to a government and holders of its base money thus have some latitude to punish a government with a history of defaulting on its plans. 6 These

questions are raised and discussed by Kydland and Prescott (1977) and Calvo (1978).

Reaganomics and Credibility

33

Reaganomics and Credibility I have argued that the government budget constraint requires that monetary and fiscal policies be coordinated and that a variety of coherent and default-free schemes for coordinating them can be imagined, the Ricardian regime and Friedman’s 1948 regime being polar examples. I have also indicated that when some of the government debt is in nominal form (and remember that base money itself is in such a nominal form) there lurks the possibility of defaulting on part of the debt by reneging on the original plan for time paths of monetary and fiscal variables. I shall now use these ideas as a basis for criticizing the program for coordinating monetary and fiscal policy that was implicit during the first year and a half or so of the first Reagan administration. The Reagan administration came into power encouraging a policy for the monetary authority that would be appropriate for the Ricardian regime but advocating plans for taxes and expenditures that could be feasible only, if at all, under some version of a Bryant-Wallace regime. The administration initially supported a commitment to a monetarist policy of M(t) − M(t − 1) = 0 forever. Simultaneously, however, in conjunction with the Congress, the administration adopted tax and expenditure plans that implied large positive values of G (t) − T (t) into the indefinite future. As we have seen above, such monetary and fiscal policies are incompatible; it is simply not feasible to carry out both of them. My colleague Neil Wallace has described the scheme for coordinating monetary and fiscal policies that was being utilized as the inception of the Reagan administration as coordination via resort to a “game of chicken.” The monetary authority had promised to stick to a tight-money policy of M(t) − M(t − 1) = 0 for all future t ’s, come hell or high water, but meanwhile the fiscal authority had set in place tax and expenditure plans that implied large values of G (t) − T (t) into the indefinite future. On the one hand, if the monetary authority could successfully stick to its guns and forever refuse to monetize any government debt, then eventually the arithmetic of the government’s budget constraint would compel the fiscal authority to back down and to swing its budget into balance. On the other hand, if the fiscal authority were to stick

34

Chapter 2: Reaganomics and Credibility

to its guns and simply refuse to reduce the stream of G (t) − T (t) , then eventually the arithmetic of the government budget constraint would compel the monetary authority to monetize large parts of the deficit. All that is clear is that in this situation, one of the two parties to the conflict eventually has to give in. (The party to capitulate is called a “chicken.”) This situation can be likened to one in which the quarterback of a football team (the fiscal authority) announces that he is going to run the ball and wants the tight end to block, while simultaneously the tight end (the monetary authority) announces that he wants to catch a pass and will run a pass route on the next play. The quarterback and the tight end point out to one another that the other had better capitulate or else the next play will go badly. About the only thing that is certain about this situation is that it cannot long endure. Coordinating monetary and fiscal policies by use of such a game of chicken necessarily confronts private agents with uncertainty about subsequent taxes, rates of inflation, and rates of interest on government securities. Unlike uncertainty about the weather or about the success of a new technology or machine, the uncertainty injected into the economy over the outcome of a struggle between monetary and fiscal policies, such as I have described, is entirely avoidable and unnecessary. Private agents are forced to form opinions about when and how the conflict between government agencies will be resolved. Some of the observed market reactions to that situation can be interpreted in terms of the preponderance of public opinion about how the conflict would eventually be resolved. For example, the high long-term nominal interest rates that prevailed in 1981 and 1982 can be interpreted as reflecting the market’s guess that large deficits would persist and eventually be monetized in large part, leading to high inflation rates in the future. In addition, the very injection of substantial extraneous uncertainty is capable of triggering contractions in output and expansions of unemployment due to the additional sheer confusion faced by agents. In this interpretation Reaganomics was not credible because it was not feasible. It was simply not feasible simultaneously to carry out both the fiscal and the monetary aspects of Reaganomics.

Reaganomics and Credibility

35

Therefore, to rational observers, Reaganomics was incredible. This was paradoxical because, more than in any other recent administration, spokesmen for the Reagan administration initially placed substantial stress on “announcement effects” and the immediate benefits that would flow from adhering to a credible long-run strategy. However, there may be another, more favorable interpretation of Reaganomics that involves a more complicated game of chicken against the background of the government budget constraint. This game of chicken involves not two but three players. Imagine that a first player sets a path for T (t) , a second player sets a path for G (t) , and a third player (via open-market operations) sets a path for M (t) − M(t − 1) . Suppose that the first and third players wish to reduce the size of the government, as measured by the stream of G (t) . Although these two players do not directly control G (t) , by acting together they can bring pressure upon it; for if the entire path of T (t) is somehow reduced and if the monetary authority maintains a policy of setting M (t) − M (t − 1) = 0 for all future t ’s, then the arithmetic of equation (2.4) and the implied need to finance current deficits by promising to run future surpluses will cause the second player to capitulate and to reduce G (t) . Though oversimplified, this three-player game captures the motivation of some advocates of the Reagan administration’s policies. The administration can be viewed as having implemented a strategy of moving quickly to reduce taxes before announcing or planning concrete expenditure reductions, while simultaneously encouraging tight monetary policy, and then opposing rescinding tax decreases in order to balance the large deficits that threatened to develop in the future. Viewed in this way, this game of chicken, fought against the backdrop of the arithmetic of equation (2.4), is a struggle over how large the government of the United States is to be. The particular strategy for reducing the size of the government that I have described is attractive, even for one who wants a smaller government, only if one is relatively confident that the uncertainties injected into monetary and fiscal arrangements by fighting the struggle in this way will not have an unduly adverse effect on the performance of national output and employment.

36

Chapter 2: Reaganomics and Credibility

Conclusion There are a variety of methods of coordinating monetary and fiscal policies that are superior to Wallace’s game of chicken. For example, a case can be made that either the Ricardian regime or Friedman’s 1948 regime dominates the game of chicken. The game of chicken that I have described amounts to a struggle for dominance between the fiscal and monetary authorities, in which each player promises to stick to a strategy that is feasible only if the other player accommodates its strategy appropriately. Under the Ricardian regime, the monetary authority in effect dominates the fiscal authority insofar as decisions about the present value of government deficit are concerned. Under Friedman’s 1948 regime, it is the fiscal authority that dominates the monetary authority insofar as decisions about the rate of growth of base money are concerned. Each of these polar regimes has a well defined structure of dominance, and each has relatively straightforward implications for the paths of government interest-bearing debt and base money, to which the public can be imagined to adjust readily. Furthermore, each of the polar regimes entails a relatively clear assignment of responsibility for inflation, insofar as government policy influences the rate of inflation. As I have portrayed the structure of the economy and characterized the conduct of policy under the Ricardian regime, inflation can emerge only if there occur changes in the preferences of the public, in the structure of legal regulations, or perhaps in the conduct of foreign governments that supply substitutes for base money and government debt—each of which would be reflected in a change in our parameters a(1) and a(2) . Under Friedman’s 1948 scheme, government deficits have direct and immediate inflationary consequences that everyone can see. How is the question of coordination of monetary and fiscal policies to be resolved? Current legislation in the United States leaves the method of resolution open, so that in practice it is resolved by the successive interactions of a succession of personalities and administrations within our fiscal and monetary institutions. It can be argued that superior outcomes would be achieved if the responsibilities of the monetary and fiscal authorities were to be legislatively or constitutionally restricted so as to determine in

Conclusion

advance which institutions are to lead and which are to follow.

37

3 The Ends of Four Big Inflations Introduction Since the middle 1960s, many Western economies have experienced persistent and growing rates of inflation. Some prominent economists and statesmen have been convinced that this inflation has a stubborn, self-sustaining momentum and that either it simply is not susceptible to cure by conventional measures of monetary and fiscal restraint or, in terms of the consequent widespread and sustained unemployment, the cost of eradicating inflation by monetary and fiscal measures would be prohibitively high. It is often claimed that there is an underlying rate of inflation that responds slowly, if at all, to restrictive monetary and fiscal measures.1 Evidently, this underlying rate of inflation is the rate of inflation that firms and workers have come to expect will prevail in the future. There is momentum in this process because firms and workers supposedly form their expectations by extrapolating past rates of inflation into the future. If this is true, the years from the middle 1960s to the early 1980s left firms and workers with a legacy of high expected rates of inflation that promise to respond only slowly, if at all, to restrictive monetary and fiscal policy actions. According to this view, restrictive monetary and fiscal actions in the first instance cause substantial reductions in output and employment but have little, if any, effects in reducing the rate of inflation. For the economy of the United States, a widely cited estimate is that for every one percentage point reduction in the annual inflation rate accomplished by restrictive monetary and fiscal measures, $220 billion of annual GNP would be lost. For the $2500 billion US economy, the cost of achieving zero percent inflation would be 1 “Most economists believe that the underlying inflation rate—roughly defined as wage costs less productivity gains—now stands at 9 to 10 percent, and that only a long period of restraint can reduce that rate significantly” (Newsweek, 19 May 1980, p. 59).

38

Introduction

39

great, indeed, according to this estimate. An alternative “rational expectations” view denies that there is any inherent momentum in the present process of inflation.2 This view maintains that firms and workers have now come to expect high rates of inflation in the future and that they strike inflationary bargains in light of these expectations.3 However, it is held that people expect high rates of inflation in the future precisely because the government’s current and prospective monetary and fiscal policies warrant these expectations. Further, the current rate of inflation and people’s expectations about future rates of inflation may well seem to respond slowly to isolated actions of restrictive monetary and fiscal policy that are viewed as temporary departures from what is perceived as a long-term government policy involving high average rates of government deficits and monetary expansion in the future. Thus inflation only seems to have a momentum of its own; it is actually the long-term government policy of persistently running large deficits and creating money at high rates that imparts 2 Paul

Samuelson has aptly summarized the rational expectations view: “I should report that there is a new school, the so-called “rational expectations.” They are optimistic that inflation can be wiped out with little pain if only the government makes credible its determination to do so. But neither history nor reason tempt one to bet their way” (Newsweek, 28 April 1980). The second sentence of this quote is probably as shrewd a summary of the rational expectations view as can be made in a single sentence. However, it is difficult to agree with the third sentence: As for “reason,” no one denies that logically coherent and well-reasoned models underlie the claims of the “rational expectationists”; as for history, the evidence summarized in this chapter is surely relevant. 3 There is actually no such thing as a “rational expectations school” in the sense of a collection of economists with an agreed-upon model of the economy and view about optimal monetary and fiscal policy. In fact, among economists who use the assumption of rational expectations there is wide disagreement about these matters. What characterizes adherents to the notion of rational expectations is their intention to build models by assuming that private agents understand the dynamic environment in which they operate approximately as well as do government policymakers. Adherence to this notion leaves ample room for substantial diversity about the many other details of a model. For some examples of rational expectations models with diverse implications, see Lucas (1972a, 1980), Barro (1974, 1976), Wallace (1980), Townsend (1980), and Sargent and Wallace (1982). Despite their diversity, it is true that all of these models impel us to think about optimal government policy in substantially different ways than were standard in macroeconomics before the advent of the doctrine of rational expectations in the early 1970s.

40

Chapter 3: The Ends of Four Big Inflations

the momentum to the inflation rate. An implication of this view is that inflation can be stopped much more quickly than advocates of the “momentum” view have indicated and that their estimates of the length of time and the costs of stopping inflation in terms of foregone output ($220 billion of GNP for one percentage point in the inflation rate) are erroneous. This is not to say that it would be easy to eradicate inflation. On the contrary, it would require far more than a few temporary restrictive fiscal and monetary actions. It would require a change in the policy regime: There must be an abrupt change in the continuing government policy, or strategy, for setting deficits now and in the future that is sufficiently binding as to be widely believed. Economists do not now possess reliable, empirically tried-and-true models that can enable them to predict precisely how rapidly and with what disruption in terms of lost output and employment such a regime change will work its effects. How costly such a move would be in terms of foregone output and how long it would be in taking effect would depend partly on how resolute and evident the government’s commitment was. This chapter describes several dramatic historical experiences that I believe to be consistent with the “rational expectations” view but that seem difficult to reconcile with the “momentum” model of inflation. The idea is to stand back from our current predicament and to examine the measures that successfully brought drastic inflations under control in several European countries in the 1920s. I shall describe and interpret events in Austria, Hungary, Germany, and Poland, countries which experienced a dramatic “hyperinflation” in which, after the passage of several months, price indexes assumed astronomical proportions. The basic data to be studied are the price indexes in Figures 3.1–3.4. These data are recorded in a logarithmic scale, so that they will fit on a page. For all four countries, and especially Germany, the rise in the price level was spectacular. The graphs also reveal that in each case inflation stopped abruptly rather than gradually. I shall also briefly describe events in Czechoslovakia, a country surrounded by neighbors experiencing hyperinflations, but which successfully achieved a stable currency itself. My reason for studying these episodes is that they are laboratories for the study of regime changes. Within each of Austria, Hungary, Poland, and Germany, there occurred a

41

Introduction

dramatic change in the fiscal policy regime, which in each instance was associated with the end of a hyperinflation. Further, though it shared some problems with its four neighbors, Czechoslovakia deliberately adopted a relatively restrictive fiscal policy regime, with the avowed aim of maintaining the value of its currency. 1E+05

1E+04

1E+03

1E+02

1E+01 1921

1922

1923

1924

Figure 3.1 Retail prices in Austria, 1921–24

While there are many differences in details among the Austrian, Hungarian, Polish, and German hyperinflations, there are some very important common features. These include the following: 1. The nature of the fiscal policy regime in effect during each of the hyperinflations. Each of the four countries persistently ran enormous budget deficits on current account. 2. The nature of the deliberate and drastic fiscal and monetary measures taken to end the hyperinflations. 3. The immediacy with which the price level and foreign exchanges suddenly stabilized.4 4 Bresciani-Turroni

wrote: “Whoever studies the recent economic history of Europe is struck by a most surprising fact: the rapid monetary restoration of some

42

Chapter 3: The Ends of Four Big Inflations

1E+07

1E+06

1E+05

1E+04

1E+03 1921

1922

1923

1924

Figure 3.2 Wholesale prices in Hungary, 1921–24

4. The rapid rise in the “high-powered” money supply in the months and years after the rapid inflation had ended. I shall assemble and interpret the facts in the light of a view about the forces that give money value and about the way the international monetary system worked in the 1920s. Before interpreting the historical facts, I now turn to a brief description of this view.

The Gold Standard After World War I, the United States was on the gold standard. The US government stood ready to convert a dollar into a specified countries where for several years paper money had continually depreciated. In some cases the stabilization of the exchange was not obtained by a continuous effort, prolonged over a period of years, whose effects would show themselves slowly in the progressive economic and financial restoration of the country, as occurred before the War in several well-known cases of monetary reform. Instead, the passing from a period of tempestuous depreciation of the currency to an almost complete stability of the exchange was very sudden” (BrescianiTurroni 1937, 334). Compare these remarks with the opinion of Samuelson cited in footnote 2.

43

The Gold Standard

1E+09

1E+08

1E+07

1E+06

1E+05

1E+04 1921

1922

1923

1924

Figure 3.3 Wholesale prices in Poland, 1921–24

amount of gold on demand. To understate things, immediately after the war, Hungary, Austria, Poland, and Germany were not on the gold standard. In practice, their currencies were largely “fiat,” or unbacked. The governments of these countries resorted to the printing of new unbacked money to finance government deficits.5 This was done on such a scale that it led to a depreciation of the currencies of spectacular proportions. In the end, the German mark stabilized at 1 trillion (1012 ) paper marks to the prewar gold mark, the Polish mark at 1.8 million paper marks to the gold zloty, the Austrian crown at 14,400 paper crowns to the prewar Austro-Hungarian crown, and the Hungarian krone at 14,500 paper crowns to the prewar Austro-Hungarian crown (League of Nations 1946, 101). This chapter focuses on the deliberate changes in policy that Hungary, Austria, Poland, and Germany made to end their hyper5 The

notes were “backed” mainly by treasury bills that, in those times, could not be expected to be paid off by levying taxes, but only by printing more notes or treasury bills.

44

Chapter 3: The Ends of Four Big Inflations

1E+15 1E+14 1E+13 1E+12 1E+11 1E+10 1E+09 1E+08 1E+07 1E+06 1E+05 1E+04 1E+03 1E+02 1919

1920

1921

1922

1923

1924

Figure 3.4 Wholesale prices in Germany, 1919–24

inflations, and the deliberate choice of policy that Czechoslovakia made to avoid inflation in the first place. The hyperinflations were each ended by restoring or virtually restoring convertibility to the dollar or equivalently to gold. For this reason it is good to keep in mind the nature of the restrictions that adherence to the gold standard imposed on a government. Under the gold standard, a government issued demand notes and longer-term debt it promised to convert into gold under certain specified conditions, that is, on demand, for notes. Presumably, people were willing to hold these claims at full value if the government’s promise to pay were judged to be good. The government’s promise to pay was “backed” only partially by its holding of gold reserves. More important in practice, since usually a government did not hold 100 percent reserves of gold, a government’s notes and debts were backed by the commitment of the government to levy taxes in sufficient amounts, given its expenditures, to make good on its debt. In effect, the notes were backed by the government’s pursuit of an appropriate budget policy. During the 1920s, John Maynard

The Gold Standard

45

Keynes emphasized that the size of a government’s gold reserve was not the determinant of whether it could successfully maintain convertibility with gold: its fiscal policy was.6 According to this view, what mattered was not the current government deficit but the present value of current and prospective future government deficits. The government was like a firm whose prospective receipts were its future tax collections. The value of the government’s debt was, to a first approximation, equal to the present value of current and future government surpluses. So under a gold standard, a government must honor its debts and could not engage in inflationary finance. In order to assign a value to the government’s debt, it was necessary to have a view about the fiscal policy regime in effect, that is, the rule determining the government deficit as a function of the state of the economy now and in the future. The public’s perception of the fiscal regime influenced the value of government debt through private agents’ expectations about the present value of the revenue streams backing that debt.7 It will be worthwhile to keep this view of the gold standard in mind as we turn to review the events surrounding the ends of the four hyperinflations.8 6 Keynes wrote:

“It is not lack of gold but the absence of other internal adjustments which prevents the leading European countries from returning to a pre-war gold standard. Most of them have plenty of gold for the purpose as soon as the other conditions favorable to the restoration of a gold standard have returned” (Keynes 1925, 132). Writing about Germany in 1923, Keynes said: “The government cannot introduce a sound money, because, in the absence of other revenue, the printing of an unsound money is the only way which it can live” (Keynes 1971, 67). 7 This view can be expressed more precisely by referring to the technical literature of optimum economic growth. I am recommending that a good first model of the gold standard or other commodity money is a real equilibrium growth model in which a government issues debt, makes expenditures, and collects taxes. Examples of these models were studied by Arrow and Kurz (1970). In such models government debt is valued according to the same economic considerations that give private debt value, namely, the prospective net revenue stream of the institution issuing the debt. A real equilibrium growth model of this kind can also be used to provide a formal rationalization of my claim below that open-market operations in private securities, foreign exchange, and gold should have no effect on the price level, i.e., the value of government demand debt. 8 It is relatively straightforward to produce a variety of workable theoretical models of a commodity money or gold standard, along the lines of note 7. It is considerably more difficult to produce a model of a fiat money, which is cost-

46

Chapter 3: The Ends of Four Big Inflations

However, it will be useful first to expand a little more generally on the distinction between the effects of isolated actions taken within the context of a given general strategy, on the one hand, and the effects of choosing among alternative general strategies or rules for repeatedly taking actions, on the other. The latter choice I refer to as a choice of regime. The values of government expenditures and tax rates for one particular quarter are examples of actions, while the rules, implicit or explicit, for repeatedly selecting government expenditures and tax rates as functions of the state of the economy are examples of regimes. Recent work in dynamic macroeconomics has discovered the following general principle: whenever there is a change in the government strategy or regime, private economic agents can be expected to change their strategies or rules for choosing consumption rates, investment rates, portfolios, and so on.9 The reason is that private agents’ behavior is selfish, or at least purposeful, so that when the government switches its strategy, private agents usually find it in their best interests to change theirs. One by-product of this principle is that most of the empirical relations captured in standard econometric models cannot be expected to remain constant across contemplated changes in government policy regimes. For this reason predictions made under the assumption that such relations will remain constant across regime changes ought not to be believed. The estimate that a 1 percent reduction in inflation would cost $220 billion GNP annually is one example of such a faulty prediction. When an important change in regime occurs, dynamic macroeconomics would less to produce, inconvertible, and of no utility except in exchange. Kareken and Wallace (1980), Wallace (1980), and Townsend (1980) describe some of the ramifications of this observation. The workable models of fiat money that we do have—for example, those of Townsend and Wallace—immediately raise the question of whether voluntarily held fiat money can continue to be valued at all in the face of substantial budget deficits of the order of magnitude studied in this chapter. Such models lead one to assign an important role to government restrictions, particularly on foreign exchange transactions, in maintaining a valued, if involuntarily held, fiat money. Keynes (1971) and Nichols (1974) also emphasized the role of such restrictions. 9 The sweeping implications of this principle for standard ways of formulating and using econometric models were first described by Lucas (1976). The principle itself has emerged in a variety of contexts involving economic dynamics. For some examples, see Lucas (1972a) and Sargent and Wallace (1975, 1976).

Austria

47

predict that the entire pattern of correlations among variables will change in quantitatively important ways. While the distinction between isolated actions and strategy regimes is clear in principle, it is an admittedly delicate task to interpret whether particular historical episodes reflect isolated actions within the same old rules of the game or whether they reflect a new set of rules or government strategies.10 All that we have to go on are the recorded actions actually taken, together with the pronouncements of public officials, laws, legislative votes, and sometimes constitutional provisions. Out of this material we are to fashion a view about the government strategy being used. Common sense suggests and technical econometric considerations confirm the difficulties in making such interpretations in general. Having said this, I believe that the examples discussed below are about as close to being laboratories for studying regime changes as history has provided.

Austria At the end of World War I, the Austro-Hungarian Empire dissolved into a number of successor states, of which Austria was one. From having been the center of an empire of 625,000 square kilometers and 50 million inhabitants, Austria was reduced to a mere 80,000 square kilometers and 6.5 million inhabitants. Having suffered food scarcities during the war that were produced by an effective Allied blockade, Austria found itself confronted with new national borders and trade barriers that cut it off from the food sources formerly within its empire. Further, the government of Austria reabsorbed a large number of Austrian imperial bureaucrats who were no longer welcome in the other successor states. Austrians also faced a large-scale unemployment problem stemming from the need to reconvert the economy to peaceful activities and to adjust to the new national borders. If this were not enough, as a loser of the war Austria owed the Reparations Commission sums 10 Sargent

and Wallace (1976) describe a sense in which it might be difficult to imagine that a regime change can occur. As they discovered, thinking about regime changes in the context of rational expectations models soon leads one to issues of free will.

48

Chapter 3: The Ends of Four Big Inflations

that for a long time were uncertain in amount but were presumed eventually to be substantial. The Reparation Commission, in effect, held a blanket mortgage against the assets and revenues of the Austrian government. Austria responded to these pressing problems by making large expenditures in the form of food relief and payments to the unemployed. In addition, the state railroads and monopolies ran deficits, as taxes and prices were kept relatively low. The government did not collect enough taxes to cover expenditures and so ran very substantial deficits during the years 1919 to 1922 (see Table 3.1). As Table 3.1 shows, in these years the deficit was typically over 50 percent of the total government expenditures. The government financed these deficits by selling treasury bills to the Austrian section of the Austro-Hungarian bank. The result was a very rapid increase in the volume of “high-powered” money, defined as the notes and demand deposit obligations of the central bank (see Table 3.2). As the figures in Table 3.2 indicate, between March 1919 and August 1922 the total note circulation in Austria11 of the Austro-Hungarian bank increased by a factor of 288. This expansion of central bank notes stemmed mainly from the bank’s policy of discounting treasury bills. However, it also resulted partly from the central bank’s practice of making loans and discounts to private agents at nominal interest rates of between 6 and 9 percent per annum, rates that by any standard were far too low in view of the inflation rate, which averaged 10,000 percent per annum from January 1921 to August 1922 (Table 3.3).12 In response to these government actions and what seemed like prospects for their indefinite continuation, the Austrian crown depreciated internationally and domestic prices rose rapidly (see Tables 3.3 and 3.4). While between January 1921 and August 1922 the note circulation of the central bank increased by a factor of 39, 11 The

Treaty of Saint-Germain, signed in September of 1919, required the successor states of the Austro-Hungarian empire to stamp their share of the notes of the Austro-Hungarian bank. The stamp converted those notes to the currency, i.e., debt, of the new states. The Austrian section of the old Austro-Hungarian bank functioned as the central bank of Austria for several years after the war. 12 Needless to say, the central bank encountered a strong demand for loans at this rate and had to ration credit.

Austria

49

the retail price index increased by a factor of 110 (see Tables 3.3 and 3.4) so that the real value of the note circulation diminished during the currency depreciation.13 The “flight from the crown” occurred as people chose to hold less of their wealth in the form of the rapidly depreciating crown, attempting instead to hold foreign currencies or real assets.14 From the viewpoint of financing its deficit, the government of Austria had an interest in resisting the flight from the crown because this had the effect of diminishing the resources that the government could command by printing money. Therefore, the government established a system of exchange controls administered by an agency called the Devisenzentrale. The essential function of this agency was to increase the amount of Austrian crowns held by Austrians, which it accomplished by adopting measures making it difficult or illegal for Austrians to hold foreign currencies and other substitutes for Austrian crowns.15 Despite these regulations, it is certain that Austrian citizens were holding large amounts of foreign currencies during 1921 and 1922. Table 3.4 reveals that the Austrian crown abruptly stabilized in August 1922, while Table 3.3 indicates that prices abruptly stabilized a month later. This occurred despite the fact that the central bank’s note circulation continued to increase rapidly, as Table 3.1 indicates. Furthermore, there occurred no change in currency units or “currency reform,” at least not for another year and a half. The depreciation of the Austrian crown was suddenly stopped 13 At

the time, some commentators argued that since the real value of currency had decreased and so in a sense currency was scarce, the increased note issue of the central bank was not the prime cause of the inflation. Some even argued that money was “tight” and that the central bank was valiantly struggling to meet the shortage of currency by adding printing presses and employees. This argument is now widely regarded as fallacious by macroeconomists. Disturbingly, however, one hears the very same argument in the contemporary United States. 14 “In Vienna, during the period of collapse, mushroom exchange banks sprang up at every street corner, where you could change your krone into Zurich francs within a few minutes of receiving them, and so avoid the risk of loss during the time it would take you to reach your usual bank. It became a reasonable criticism to allege that a prudent man at a cafe ordering a bock of beer should order a second bock at the same time, even at the expense of drinking it tepid, lest the price should rise meanwhile” (Keynes 1924, 51). 15 See Young (1925, 2:16). That a government might want to adopt such measures if it were using inflationary finance was pointed out by Nichols (1974).

50

Chapter 3: The Ends of Four Big Inflations

by the intervention of the Council of the League of Nations and the resulting binding commitment of the government of Austria to reorder Austrian fiscal and monetary strategies dramatically. After Austria’s increasingly desperate pleas to the Allied governments for international aid had repeatedly been rejected or only partially fulfilled, in late August 1922 the Council of the League of Nations undertook to enter into serious negotiations to reconstruct the financial system of Austria. These negotiations led to the signing of three protocols on 2 October 1922 that successfully guided the financial reconstruction of Austria. It is remarkable that even before the precise details of the protocols were publicly announced, the fact of the serious deliberations of the council brought relief to the situation. This can be seen in Tables 3.3 and 3.4, and was described by Pasvolsky (1928, 116) as follows: The moment the Council of the League decided to take up in earnest the question of Austrian reconstruction, there was immediately a widespread conviction that the solution of the problem was at hand. This conviction communicated itself first of all to that delicately adjusted mechanism, the international exchange market. Nearly two weeks before Chancellor Seipel officially laid the Austrian question before the Council of the League, in August 25, the foreign exchange rate ceased to soar and began to decline, the internal price level following suit three weeks later. The printing presses in Austria were still grinding out new currency; the various ministries were still dispersing this new currency through the country by means of continuing budgetary deficits. Yet the rate of exchange was slowly declining. The crisis was checked. The first protocol was a declaration signed by Great Britain, France, Italy, Czechoslovakia, and Austria that reaffirmed the political independence and sovereignty of Austria.16 The second 16 The

content of this protocol is highly sensible when it is remembered that the value of a state’s currency and other debt, at least under the gold standard, is determined by its ability to back that debt with an appropriate fiscal policy. In this respect its situation is no different from that of a firm. In 1922 there was widespread concern within and without Austria that its sovereignty was at risk.

Austria

51

protocol provided conditions for an international loan of 650 million gold crowns to Austria. The third protocol was signed by Austria alone and laid out a plan for reconstruction of its fiscal and monetary affairs. The Austrian government promised to establish a new independent central bank, to cease running large deficits, and to bind itself not to finance deficits with advances of notes from the central bank. Further, the government of Austria agreed to accept in Austria a commissioner general, appointed by the Council of the League, who was to be responsible for monitoring the fulfillment of Austria’s commitments. The government of Austria also agreed to furnish security to back the reconstruction loan. At the same time it was understood that the Reparations Commission would give up or modify its claim on the resources of the government of Austria. The government of Austria and the League both moved swiftly to execute the plan outlined in the protocols. In legislation of 14 November 1922, the Austrian National Bank was formed to replace the old Austrian section of the Austro-Hungarian bank; it was to take over the assets and functions of the Devisenzentrale. The new bank began operations on 1 January 1923 and was specifically forbidden from lending to the government except on the security of an equal amount of gold and foreign assets. The bank was also required to cover its note issues with certain minimal proportions of gold, foreign earning assets, and commercial bills. Further, once the government’s debt to the bank had been reduced to 30 million gold crowns, the bank was obligated to resume convertibility into gold. The government moved to balance its budget by taking concrete steps in several directions. Expenditures were reduced by discharging thousands of government employees. Under the reconstruction scheme, the government promised gradually to discharge a total of 100,000 state employees. Deficits in government enterprises were reduced by raising prices of government-sold goods (See the desperate note delivered by the Austrian minister to the Supreme Council of the Allied governments quoted by Pasvolsky (1928, 115)). The first protocol aimed to clarify the extent to which Austria remained a political and economic entity capable of backing its debts. A similar protocol was signed at the inception of Hungary’s financial reconstruction.

52

Chapter 3: The Ends of Four Big Inflations

and services. New taxes and more efficient means of collecting tax and custom revenues were instituted. The results of these measures can be seen by comparing the figures in Table 3.5 with those in Table 3.1. Within two years the government was able to balance the budget. The stabilization of the Austrian crown was not achieved via a currency reform. At the end of 1924 a new unit of currency was introduced, the schilling, equal to 10,000 paper crowns. The introduction of this new unit of currency occurred long after the exchange rate had been stabilized and was surely an incidental measure.17 Table 3.2 reveals that from August 1922, when the exchange rate suddenly stabilized, to December 1924, the circulating notes of the Austrian central bank increased by a factor of over 6. The achievement of price stability in the face of a sixfold increase in the stock of “high-powered” money was widely regarded by contemporaries as violating the quantity theory of money, and so it seems to do. However, these observations are not at all paradoxical when interpreted in the light of a view that distinguishes sharply between unbacked, or “outside,” money, on the one hand, and backed, or “inside,” money, on the other hand. In particular, the balance sheet of the central bank and the nature of its open-market operations changed dramatically after the carrying out of the League’s protocols, with the consequence that the proper interpretation of the figures on the total note obligations of the central bank changed substantially. Before the protocols, the liabilities of the central bank were backed mainly by government treasury bills; that is, they were not backed at all, since treasury bills signified no commitment to raise revenues through future tax collections. After the execution of the protocols, the liabilities of the central bank became backed by gold, foreign assets, and commercial paper, and ultimately by the power of the government to collect taxes. At the margin, central bank liabilities were backed 100 percent by gold, foreign assets, and commercial paper as notes and the deposits were created through open-market operations in those assets (see Table 3.6). The value 17 It

should be noted that for two years the new bank vigorously exercised its authority to control transactions in foreign currency. Only after March 1925 were restrictions on trading foreign exchange removed.

Hungary

53

of the crown was backed by the commitment of the government to run a fiscal policy compatible with maintaining the convertibility of its liabilities into dollars. Given such a fiscal regime, to a first approximation, the intermediating activities of the central bank did not affect the value of the crown so long as the assets purchased by the bank were sufficiently valuable. Thus the sixfold increase in the liabilities of the central bank after the protocols ought not to be regarded as inflationary. The willingness of Austrians to convert hoards of foreign exchange into crowns, which is reflected in Table 3.6, is not surprising since the stabilization of the crown made it a much more desirable asset to hold relative to foreign exchange.18 The available figures on unemployment indicate that the stabilization of the crown was attended by a substantial increase in the unemployment rate, though unemployment had begun to climb well before stabilization was achieved (see Table 3.7). The number of recipients of state unemployment benefits gradually climbed from a low of 8,700 in December 1921 to 83,000 in December 1922. It climbed to 167,000 by March 1923, and then receded to 76,000 in November 1923 (Young 1925, 161). How much of this unemployment was due to the achievement of currency stabilization and how much was due to the real dislocations affecting the Austrian economy cannot be determined. However, it is true that currency stabilization was achieved in Austria very suddenly, and with a cost in increased unemployment and foregone output that was minor compared with the $220 billion GNP that some current analysts estimate would be lost in the United States per one percentage point inflation reduction.

Hungary Like its old partner in the Hapsburg monarchy, Hungary emerged from World War I as a country much reduced in land, population, and power. It retained only 25 percent of its territory (down from 325,000 square kilometers to 92,000) and only 37 percent of its population (down from 21 million to about 8 million). Its financial 18 This

explanation is consistent with the argument advanced by Fama (1980). See the appendix for an alternative explanation.

54

Chapter 3: The Ends of Four Big Inflations

and economic life was disrupted by the newly drawn national borders separating it from peoples and economic institutions formerly within the domain of the Hapsburg monarchy. At the end of the war, Hungary experienced political turmoil as the Hapsburg King Charles was replaced by the government of Prince Karolyi. In March 1919, the Karolyi government was overthrown by the Bolsheviks under Bela Kun. The regime of Bela Kun lasted only four months, as Rumania invaded Hungary, occupied it for a few weeks, and then withdrew. A new repressive right-wing regime under Admiral Horthy then took power. The “white terror” against leftists carried out by supporters of Horthy took even more lives than the “red terror” that had occurred under Bela Kun. At the end of the war, the currency of Hungary consisted of the notes of the Austro-Hungarian bank. By the provisions of the peace treaties of Trianon and Saint-Germain, the successor states to the Austro-Hungarian empire were required to stamp the notes of the Austro-Hungarian bank that were held by their residents, in effect, thereby recognizing those notes as debts of the respective new states. Before Hungary executed this provision of the Treaty of Trianon, the currency situation grew more complicated, for the Bolshevik regime had access to the plates for printing one- and two-crown Austro-Hungarian bank notes, and it used them to print more notes. The Bolshevik government also issued new socalled white notes. Each of these Bolshevik-issued currencies was honored by the subsequent government. The Austro-Hungarian bank was liquidated at the end of 1919, and it was replaced by an Austrian section and a Hungarian section. The functions of the Hungarian section of the old bank were assumed in August 1921 by a State Note Institute, which was under the control of the minister of finance. In August 1921, the Note Institute issued its own notes, the Hungarian krone, in exchange for Hungarian stamped notes of the Austro-Hungarian bank and several other classes of notes, including those that had been issued by the Bolshevik regime. As a loser of the war, Hungary owed reparations according to the Treaty of Trianon. The Reparation Commission had a lien on the resources of the government of Hungary. However, neither

Hungary

55

the total amount owed nor a schedule of payments was fixed for many years after the war. This circumstance alone created serious obstacles in terms of achieving a stable value for Hungary’s currency and other debts, since the unclear reparations obligations made uncertain the nature of the resources that backed those debts. From 1919 until 1924 the government of Hungary ran substantial budget deficits. The government’s budget estimates in Table 3.8 are reported by (Pasvolsky 1928, 298) substantially to understate the size of the deficits. These deficits were financed by borrowing from the State Note Institute and were a major cause of a rapid increase in the note and deposit liabilities of the institute. An additional cause of the increase in liabilities of the institute was the increasing volume of loans and discounts that it made to private agents (see Table 3.9). These loans were made at a very low interest rate, in view of the rapid rate of price appreciation, and to a large extent amounted to simple gifts from the Note Institute to those lucky enough to receive loans on such generous terms. These private loans accounted for a much larger increase in high-powered money in the Hungarian than in the other three hyperinflations we shall study. As Table 3.10 shows, the Hungarian krone depreciated rapidly on foreign exchange markets, and domestic prices rose rapidly. Between January 1922 and April 1924, the price index increased by a factor of 263. In the same period, the total notes and deposit liabilities of the Note Institute increased by a factor of 85, so that the real value of its liabilities decreased substantially. As in the case of Austria, this decrease was symptomatic of a “flight from the krone,” as residents of Hungary attempted to economize on their holdings of krones and instead to hold assets denominated in more stable currencies. As in the case of Austria, the government of Hungary resisted this trend by establishing in August 1922 a Hungarian Devisenzentrale within the State Note Institute. Table 3.10 indicates that in March 1924, the rise in prices and the depreciation of the krone internationally both abruptly halted. The stabilization occurred in the face of continued expansion in the liabilities of the central bank, which increased by a factor of 3.15 between March 1924 and January 1925 (see Table 3.9). This pattern

56

Chapter 3: The Ends of Four Big Inflations

parallels what occurred in Austria and has a similar explanation. As in Austria, the financial reconstruction of Hungary was accomplished with the intervention of the League of Nations. Together with the Reparations Commission and the government of Hungary, the League devised a plan that reduced and clarified the reparations commitment of Hungary, arranged for an international loan that would help finance government expenditures, and committed Hungary to establish a balanced budget and a central bank legally bound to refuse any government demand for unbacked credit. On 21 February 1924, the Reparations Commission agreed to give up its lien on Hungary’s resources so that these could be used to secure a reconstruction loan. A variety of Western nations also agreed to give up their liens on Hungary so that the new loan could be floated. The League’s reconstruction play was embodied in two protocols. The first was signed by Great Britain, France, Italy, Czechoslovakia, Rumania, and Hungary and guaranteed the “political independence, territorial integrity, and sovereignty of Hungary.” The second protocol outlined the terms of the reconstruction plan and committed Hungary to balance its budget and form a central bank truly independent of the Finance Ministry. The government was also obligated to accept in Hungary a commissioner general, responsible to the League, to monitor and supervise the government’s fulfillment of its commitment to fiscal and monetary reform. A reconstruction loan of 250 million gold krones was successfully placed abroad in July 1924. The loan was secured by receipts from customs duties and sugar taxes and revenues from the salt and tobacco monopolies. The purpose of the loan was to give the government a concrete means of converting future promises to tax into current resources while avoiding the need to place its debt domestically. By a law of 26 April 1924, the Hungarian National Bank was established, and it began operations on 24 June. The bank assumed the assets and liabilities of the State Note Institute and took over the functions of the foreign exchange control office, the Devisenzentrale. The bank was prohibited from making any additional loans or advances to the government, except upon full security of gold or foreign bills. The bank was also required to hold gold

Hungary

57

reserves of certain specified percentages behind its liabilities. The government of Hungary also tried to establish a balanced budget. Both by cutting expenditures and raising tax collections, the government was successful in moving quickly to a balanced budget (see Table 3.11). Indeed, the proceeds of the reconstruction loan were used perceptibly more slowly than had been anticipated in the reconstruction plan. As Table 3.9 confirms, the stabilization of the krone was accompanied by a substantial increase in the total liabilities of the central bank. But as with Austria, the drastic shift in the fiscal policy regime that occasioned the stabilization also changed the appropriate interpretation of these figures. As Table 3.9 indicates and as the regulations governing the bank required, after the League’s intervention the note and deposit liabilities of the central bank became backed, 100 percent at the margin, by holdings of gold, foreign exchange, and commercial paper. In effect, the central bank’s liabilities represented “fiat money” before the League’s plan was in effect; after that plan was in effect, they represented more or less backed claims on British sterling.19 the foreign currency to which Hungary pegged its exchange as a condition for British participation in the reconstruction loan. Figures on unemployment in Hungary are reported in Table 3.12, and unfortunately begin only immediately after the price stabilization had already occurred. All that can be inferred from these figures is that immediately after the stabilization, unemployment was not any higher than it was one or two years later. This is consistent either with the hypothesis that the stabilization process had little adverse effect on unemployment or with the hypothesis that the adverse effect was so long-lasting that no recovery occurred within the time span of the figures recorded. The former hypothesis seems more plausible to me.

19 Within

a year and a half, these became a claim on gold as Britain returned to the gold standard.

58

Chapter 3: The Ends of Four Big Inflations

Poland The new nation of Poland came into existence at the end of World War I and was formed from territories formerly belonging to Germany, Austro-Hungary, and Russia. At the time of its formation, Poland possessed a varied currency consisting of Russian rubles, crowns of the Austro-Hungarian bank, German marks, and Polish marks issued by the Polish State Loan Bank, which had been established by Germany to control the currency in the part of Poland occupied by Germany during the war. For Poland, the armistice of 1918 did not bring peace, a costly war with Soviet Russia being waged until the fall of 1920. Poland was devastated by the fighting and by Germany’s practice of stripping it of its machinery and materials during World War I.20 The new government of Poland ran very large deficits up to 1924 (see Table 3.13). These deficits were financed by government borrowing from the Polish State Loan Bank, which the new government had taken over from the Germans. From January 1922 to December 1923, the outstanding notes of the Polish State Loan Bank increased by a factor of 523 (Table 3.14). Over the same period, the price index increased by a factor of 2402 while the dollar exchange rate decreased by a factor of 1397 (see Tables 3.15 and 3.16). As in the other inflations we have studied, the real value of the note circulation decreased as people engaged in a “flight from the mark.” Extensive government exchange controls were imposed to resist this trend. Tables 3.14 and 3.15 indicate that the rapid inflation and exchange depreciation both suddenly stopped in January 1924. Unlike the cases of Austria and Hungary, in Poland the initial stabilization was achieved without foreign loans or intervention, although later in 1927, after currency depreciation threatened to renew, a substantial foreign loan was arranged (League of Nations 1946, 111). But in terms of the substantial fiscal and monetary regime changes that accompanied the end of the inflation, there is much similarity to the Austrian and Hungarian experiences. The two interrelated changes were a dramatic move toward a balanced government budget and the establishment of an independent central 20 Unlike

Austria, Hungary, and Germany, Poland did not owe war reparations.

Poland

59

bank that was prohibited from making additional unsecured loans to the government. In January 1924, the minister of finance was granted broad powers to effect monetary and fiscal reform. The minister immediately initiated the establishment of the Bank of Poland, which was to assume the functions of the Polish State Loan Bank. The eventual goal was to restore convertibility with gold. The bank was required to hold a 30 percent reserve behind its notes, to consist of gold and foreign paper assets denominated in stable currencies. Beyond this reserve, the bank’s notes had to be secured by private bills of exchange and silver. A maximum credit to the government of 50 million zlotys was permitted. The government also moved swiftly to balance the budget (see Table 3.13). In January 1924, a new currency unit became effective, the gold zloty, worth 1.8 million paper marks. The zloty was equal in gold content to 19.29 cents. Table 3.14 reveals that from January 1924 to December 1924, the note circulation of the central bank increased by a factor of 3.2, in the face of relative stability of the price level and the exchange rate (see Tables 3.15 and 3.16). This phenomenon matches what occurred in Austria and Hungary and has a similar explanation. As Table 3.14 reveals, the increased note circulation during this period was effectively backed 100 percent by gold, foreign exchange, and private paper. The available figures on unemployment are summarized in Table 3.17. The stabilization of the price level in January 1924 is accompanied by an abrupt rise in the number of unemployed. Another rise occurs in July of 1924. While the figures indicate substantial unemployment in late 1924, unemployment is not an order of magnitude worse than before the stabilization, and certainly not anywhere nearly as bad as would be predicted by application of the same method of analysis that was used to fabricate the prediction for the contemporary United States that each percentage point reduction in inflation would require a reduction of $220 billion in real GNP. The Polish zloty depreciated internationally from late 1925 onward but stabilized in autumn of 1926 at around 72 percent of its level of January 1924. At the same time, the domestic price level stabilized at about 50 percent above its level of January 1924.

60

Chapter 3: The Ends of Four Big Inflations

The threatened renewal of inflation has been attributed to the government’s premature relaxation of exchange controls and the tendency of the central bank to make private loans at insufficient interest rates (League of Nations 1946, 108).

Germany After World War I, Germany owed staggering reparations to the Allied countries. This fact dominated Germany’s public finance from 1919 until 1923 and was a most important force for hyperinflation. At the conclusion of the war, Germany experienced a political revolution and established a republican government. The early postwar governments were dominated by moderate Socialists, who for a variety of reasons reached accommodations with centers of military and industrial power of the prewar regime (Paxton 1975, 146–50). These accommodations in effect undermined the willingness and capability of the government to meet its admittedly staggering revenue needs through explicit taxation. Of the four episodes that we have studied, Germany’s hyperinflation was the most spectacular, as the figures on wholesale prices and exchange rates in Tables 3.18 and 3.19 reveal. The inflation became most severe after the military occupation of the Ruhr by the French in January 1923. The German government was determined to fight the French occupation by a policy of passive resistance, making direct payments to striking workers that were financed by discounting treasury bills with the Reichsbank. Table 3.20 estimates the budget of Germany for 1920 to 1923 (Graham 1930, 40–41). The table reveals that except for 1923, the budget would not have been badly out of balance except for the massive reparations payments made. The disruption caused to Germany’s finances by the reparations situation is surely understated by the reparations figures given in Table 3.20. For one thing, considerably larger sums were initially expected of Germany than it ever was eventually able to pay. For another thing, the extent of Germany’s total obligation and the required schedule of payments were for a long time uncertain and under negotiation. From the viewpoint that the value of a state’s currency and other debt

Germany

61

depends intimately on the fiscal policy it intends to run, the uncertainty about the reparations owed by the German government necessarily cast a long shadow over its prospects for a stable currency. As Table 3.21 reveals, the note circulation of the Reichsbank increased dramatically from 1921 to 1923, especially in the several months before November 1923. As pointed out by Young (1925), at the end of October 1923, over 99 percent of outstanding Reichsbank notes had been placed in circulation within the previous 30 days.21 Table 3.21 reveals the extent to which the Reichsbank note circulation was backed by discounted treasury bills. During 1923, the Reichsbank also began discounting large volumes of commercial bills. Since these loans were made at nominal rates of interest far below the rate of inflation, they amounted virtually to government transfer payments to the recipients of the loans. Especially during the great inflation of 1923, a force came into play that was also present in the other hyperinflations we have studied. Given the method of assessing taxes in nominal terms, lags between the time when taxes were levied and the time when they were collected led to reduced revenues as the government evidently repeatedly underestimated the prospective rate of inflation and as the rapid inflation gave people a large incentive to delay paying their taxes. This effect probably partially accounts for the reduced tax revenues collected during the first nine months of 1923. The French occupation of the Ruhr also helps explain it. In response to the inflationary public finance and despite the efforts of the government to impose exchange controls, there occurred a “flight from the German mark” in which the real value of reichsmark notes decreased dramatically. The figures in Table 3.18 indicate that between January 1922 and July 1923, wholesale prices increased by a factor of 2038 while Reichsbank notes increased by a factor of 378. Between January 1922 and August 1923, wholesale prices increased by a factor of 25,723 while Reichsbank notes circulating increased by a factor of 5,748. The fact that prices increased proportionately many times more than did the Reichsbank note 21 Keynes

wrote: “A government can live for a long time, even the German government or the Russian government, by printing paper money.... A government can live by this means when it can live by no other” (Keynes 1971, 47).

62

Chapter 3: The Ends of Four Big Inflations

circulation is symptomatic of the efforts of Germans to economize on their holdings of rapidly depreciating German marks. Toward the end of the hyperinflation, Germans made every effort to avoid holding marks and held large quantities of foreign exchange for purposes of conducting transaction. By October 1923, it has been roughly estimated, the real value of foreign currencies circulating in Germany was at least equal to and perhaps several times the real value of Reichsbank notes circulating.22 The figures in Tables 3.18 and 3.19 show that prices suddenly stopped rising and the mark stopped depreciating in late November 1923. The event of stabilization was attended by a “monetary reform,” in which on 15 October 1923 a new currency unit called the rentenmark was declared equivalent to 1 trillion (1012 ) paper marks. While great psychological significance has sometimes been assigned to this unit change, it is difficult to attribute any substantial effects to what was in itself only a cosmetic measure.23 The substantive aspect of the decree of 15 October was the establishing of a Rentenbank to take over the note issue functions of the Reichsbank. The decree put binding limits upon both the total volume of rentenmarks that could be issued, 3.2 billion marks, and the maximum amount that could be issued to the government, 1.2 billion marks. This limitation on the amount of credit that could be extended to the government was announced at a time when the government was financing virtually 100 percent of its expenditures by means of note issue (Young 1925, 1:421). In December 1923, the management of the Rentenbank was tested by the government and effectively made clear its intent to meet its obligation to limit government borrowing to within the amount decreed. Simultaneously and abruptly three things happened: additional government borrowing from the central bank stopped, the government budget swung into balance, and inflation stopped. 22 See

Young (1925, 1:402) and Bresciani-Turroni (1937, 345). reading an earlier draft of this chapter, John Kennan directed me to the following passage in Constance Reid’s biography of the mathematician David Hilbert: “In 1923 the inflation ended abruptly through the creation of a new unit of currency called the Rentenmark. Although Hilbert remarked skeptically, ‘One cannot solve a problem by changing the name of the independent variable,’ the stability of conditions was gradually restored” (Reid 1979, 162–63). 23 After

Germany

63

Table 3.22 shows the dramatic progress toward a balanced budget that was made in the months after the Rentenbank decree. The government moved to balance the budget by taking a series of deliberate, permanent actions to raise taxes and eliminate expenditures. Young reports that “by the personnel decree of October 27, 1923, the number of government employees was cut by 25 percent; all temporary employees were to be discharged; all above the age of 65 years were to be retired. An additional 10 percent of the civil servants were to be discharged by January 1924. The railways, overstaffed as a result of post-war demobilization, discharged 120,000 men during 1923 and 60,000 more during 1924. The postal administration reduced its staff by 65,000 men; the Reichsbank itself which had increased the number of its employees from 13,316 at the close of 1922 to 22,909 at the close of 1923, began the discharge of its superfluous force in December, as soon as the effects of stabilization became manifest” (Young 1925, 1:422). Substantially aiding the fiscal situation, Germany also obtained relief from its reparation obligations. Reparations payments were temporarily suspended, and the Dawes plan assigned Germany a much more manageable schedule of payments. Table 3.21 documents a pattern that we have seen in the three other hyperinflations: The substantial growth of central bank note and demand deposit liabilities in the months after the currency was stabilized. As in the other cases that we have studied, the best explanation for this is that at the margin the postinflation increase in notes was no longer backed by government debt. Instead, in the German case, it was largely backed by discounted commercial bills. The nature of the system of promises and claims behind the central bank’s liabilities changed when after the Rentenbank decree the central bank no longer offered additional credit to the government. So once again the interpretation of the time series on central bank notes and deposits must undergo a very substantial change. By all available measures, the stabilization of the German mark was accompanied by increases in output and employment and decreases in unemployment (Graham 1930, Chapter 12). While 1924 was not a good year for German business, it was much better than 1923. Table 3.23 is representative of the figures assembled by Graham, and shows that 1924 suffers in comparison with 1922 but

64

Chapter 3: The Ends of Four Big Inflations

that 1925 was a good year. In these figures one cannot find much convincing evidence of a favorable trade-off between inflation and output, since the year of spectacular inflation, 1923, was a very bad year for employment and physical production. Certainly a large part of the poor performance of 1923 was due to the French occupation of the Ruhr and the policy of passive resistance. Despite the evident absence of a “Phillips curve” trade-off between inflation and real output in the figures in Tables 3.18 and 3.23, there is ample evidence that the German inflation was far from “neutral” and that there were important “real effects.” Graham (1930) gives evidence that the inflation and the associated reduction in real rates of return to high-powered money and other government debt were accompanied by real overinvestment in many kinds of capital goods.24 There is little doubt that the “irrational” structure of capital characterizing Germany after stabilization led to subsequent problems of adjustment in labor and other markets.

Czechoslovakia After World War I, the new nation of Czechoslovakia was formed out of territories formerly belonging to Austria and Hungary. Under the leadership of a distinguished minister of finance, Dr. Alois Rasin, immediately after the war Czechoslovakia adopted the conservative fiscal and monetary policies its neighbors adopted only after their currencies had depreciated radically. As a result, Czechoslovakia avoided the hyperinflation experienced by its neighbors. Under Rasin’s leadership, Czechoslovakia early on showed that it was serious about attaining a stable currency. Even before the peace treaties required it, Czechoslovakia stamped the Austro-Hungarian notes then circulating within its border with the Czechoslovakian stamp, thereby recognizing them as its own debt. There was considerable drama associated with this event, as the National Assembly passed the plans for stamping in secret sessions on 25 February 1919. From 26 February to 9 March, the frontiers 24 Theoretical models of money along the lines proposed by Samuelson (1958) predict that too much capital will be accumulated when the government fiscal policy is so profligate that money becomes valueless. See Samuelson (1958) and Wallace (1980).

Czechoslovakia

65

of the country were unexpectedly closed and foreign mail service was closed. Only Austro-Hungarian notes circulating within the country could be presented for stamping. As part of the stamping process, the government retained part of the notes in the form of a forced Treaty of Saint-Germain loan.25 About 8 billion crowns were stamped. A banking office in the Ministry of Finance took over the affairs of the old Austro-Hungarian bank. Czechoslovakia moved quickly to limit by statute the total government note circulation and to prevent inflationary government finance. A law of 10 April 1919 strictly limited the fiduciary or unbacked note circulation of the banking office to about 7 billion crowns. This law was obeyed, and forced the government to finance its expenditures by levying taxes or else issuing debt, which, because of the statutory restriction on government note issues, were interpreted as promises to tax in the future. From 1920 on, Czechoslovakia ran only modest deficits on current account (see Table 3.24). Among other taxes, Czechoslovakia imposed a progressive capital levy on property, which raised a cumulative amount of about 11 billion crowns by 1925. It also imposed an increment tax on the increased wealth individuals had obtained during the war. Table 3.25 shows the note and deposit liabilities of the banking office. The government’s abstention from inflationary finance shows up in these figures. Table 3.26 shows the path of exchange rates and how, after declining until November 1921, the Czechoslovakian crown rapidly gained to about 3 United States cents. Table 3.27 shows the price levels. From 1922 to 1923, Czechoslovakia actually experienced a deflation. Indeed, Rasin’s initial plan had been to restore the Czechoslovakian crown to the prewar gold par value of the old Austro-Hungarian crown. Following Rasin’s assassination, this plan was abandoned and the crown was stabilized at about 2.96 cents. 25 The

frontiers were closed to prevent notes from Austria and Hungary from entering the country. The Treaty of Saint-Germain, signed 10 September 1919, provided that the successor states should stamp the Austro-Hungarian notes, signifying their assumption of the debt.

66

Chapter 3: The Ends of Four Big Inflations

Conclusion The essential measures that ended hyperinflation in Germany, Austria, Hungary, and Poland were, first, the creation of an independent central bank that was legally committed to refuse the government’s demand for additional unsecured credit and, second, a simultaneous alteration in the fiscal policy regime.26 These measures were interrelated and coordinated. They had the effect of binding the government to place its debt with private parties and foreign governments that would value that debt according to whether it was backed by sufficiently large prospective taxes relative to public expenditures. In each case we have studied, once it became widely understood that the government would not rely on the central bank for its finances, the inflation terminated and the exchanges stabilized. We have further seen that it was not simply the increasing quantity of central bank notes that caused the hyperinflation, since in each case the note circulation continued to grow rapidly after the exchange rate and price level had been stabilized. Rather, it was the growth of fiat currency that was unbacked, or backed only by government bills, which there never was a prospect to retire through taxation. The changes that ended the hyperinflations were not isolated restrictive actions within a given set of rules of the game or general policy. Earlier attempts to stabilize the exchanges in Hungary under Hegedus (Pasvolsky 1928, 304–7), and also in Germany, failed precisely because they did not change the rules of the game under which fiscal policy had to be conducted.27 26 Of

inflationary finance, Keynes wrote: “It is common to speak as though, when a government pays its way by inflation, the people of the country avoid taxation. We have seen this is not so. What is raised by printing notes is just as much taken from the public as is beer-duty or an income-tax. What a government spends the public pay for. There is no such thing as an uncovered deficit. But in some countries it seems possible to please and content the public, for a time at least, by giving them, in return for the taxes they pay, finely engraved acknowledgments on water-marked paper. The income tax receipts, which we in England receive from the surveyor, we throw into the wastepaper basket; in Germany they call them bank-notes and put them into their pocketbooks; in France they are termed Rentes and are locked up in the family safe” (Keynes 1924, 68–69). 27 A deep objection to the interpretation in this paragraph can be constructed along the lines of Sargent and Wallace (1976), who argue that for a single economy it is

67

Appendix

In discussing this subject with various people, I have encountered the view that the events described here are so extreme and bizarre that they do not bear on the subject of inflation in the contemporary United States. On the contrary, it is precisely because the events were so extreme that they are relevant. The four incidents we have studied are akin to laboratory experiments in which the elemental forces that cause and can be used to stop inflation are easiest to spot. I believe that these incidents are full of lessons about our own, less dramatic predicament with inflation, if only we interpret them correctly.

Appendix The explanation mentioned in association with footnote 34 is consistent with the argument advanced by Fama (1980). There is an alternative explanation of these observations that neglects the distinction between inside and outside money, and that interprets the observations in terms of a demand function for the total quantity of “money.” For instance, Cagan (1956) posited the demand schedule for money to take the form Mt − Pt = α( Et Pt+1 − Pt )

α 0 from time 0 to time T − 1, and then at the rate zero from time T onward. In this case the inflation rate would follow the path depicted in Figure 3.1. For the second experiment, suppose that initially everyone expected the money supply to increase at the constant rate µ forever but that at time T it becomes known that henceforth the money supply will increase at the rate 0 forever. In this case the inflation rate takes a sudden drop at time T , as shown by the path in Figure 3.2. Now since the inflation and the expected inflation rate experience a sudden drop at T in this case, it follows from equation (3.1) that real balances must increase at T . This will require a sudden once and for all drop in the price level at T . This second example of a previously unexpected decrease in the inflation rate provides the material for an explanation of the growth of the money supplies after currency stabilization. In the face of a previously unexpected, sudden, and permanent drop in the rate of money creation, the only way to avoid a sudden drop in the price level would be to accompany the decrease in the rate of money creation with a once and for all increase in the money supply. In order to stabilize the price level in the face of a decreased rate of change of money, the level of the money supply must jump upward once and for all. What actually occurred in the four countries studied here was not a once and for all jump but a gradual increase in the money supply over many months. This could be reconciled with the observations within the model (3.1) if people were assumed only

Appendix

69

Pt+1 –Pt

μ

0

0

T

Figure 3.5 Inflation path with a fully anticipated decrease in money supply growth from µ to 0 at time T . From time 0 to T the money growth rate is µ , but people are assumed to know that it will drop to zero at time T .

gradually to catch on to the fact of stabilization and to decrease the rate of inflation that they expected as the currency stabilization continued to hold. I find this explanation hard to accept, but it is a possibility. An alternative way to reconcile the preceding explanation with the gradual upward movement of “high-powered” money after the stabilizations is to add adjustment lags to the portfolio balance schedule (1). For example, consider replacing (1) with

( Mt − Pt ) = α( Et Pt+1 − Pt ) + λ( Mt−1 − Pt−1 ) α < 0, 0 < λ < 1 (3.3) In this case, an abrupt stabilization of expected inflation induces only a gradual adjustment of real balances upward at the rate of 1 − λ per period. My own preference at this point is for an explanation that stresses the distinction between backed and unbacked money.

70

Chapter 3: The Ends of Four Big Inflations

P

t +1

P

t

μ

0

0

T

Figure 3.6 Inflation path with a previously unexpected decrease in money growth from µ to 0 at time T . From time 0 to T , people think that the growth rate of money will be µ forever.

Table 3.1 A USTRIAN B UDGETS , 1919–22 in millions of paper crowns

ExpenReceipts ditures 1 January–30 June 1919 1 July 1919–30 June 1920 1 July 1920–30 June 1921 1 January–31 December 1922 Source: Pasvolsky (1928, 102).

1,339 4,043 6,295 16,873 29,483 70,601 209,763 347,533

Deficit 2,704 10,578 41,118 137,770

Percentage of expenditures covered by new issues of paper money 67 63 58 40

71

Appendix

Table 3.2 T OTAL N OTE C IRCULATION OF A USTRIAN C ROWNS in thousands of crowns 1919

1920

1921

1922

January February March April May June July August September October November December January February March April May June July August September October November December January February March April May June July August September January February March April

— — 4,687,056 5,577,851 5,960,003 7,397,692 8,391,405 9,241,135 9,781,112 10,819,310 11,193,670 12,134,474 13,266,878 14,292,809 15,457,749 15,523,832 15,793,805 16,971,344 18,721,495 20,050,281 22,271,686 25,120,385 28,072,331 30,645,658 34,525,634 38,352,648 41,067,299 45,036,723 45,583,194 49,685,140 54,107,281 58,533,766 70,170,798 227,015,925 259,931,138 304,063,642 346,697,776

Source: Young (1925, 2:292).

1922

1923

1924

1925

May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March April

397,829,313 549,915,678 786,225,601 1,353,403,632 2,277,677,738 2,970,916,607 3,417,786,498 4,080,177,238 4,110,551,163 4,207,991,722 4,459,117,216 4,577,382,333 4,837,042,081 5,432,619,312 5,684,133,721 5,894,786,367 6,225,109,352 6,607,839,105 6,577,616,341 7,125,755,190 6,735,109,000 7,364,441,000 7,144,901,000 7,135,471,000 7,552,620,000 7,774,958,000 7,995,647,000 5,894,786,367 7,998,509,000 8,213,003,000 8,070,021,000 8,387,767,000 7,902,217,000 7,957,242,000 7,897,792,000 7,976,420,000

72

Chapter 3: The Ends of Four Big Inflations

Table 3.3 A USTRIAN R ETAIL P RICES , 1921–24 Retail price index, 52 commodities 1921

1922

1923

1924

January February March April May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March April May June

Source: Young (1925, 2:293).

100 114 122 116 121 150 143 167 215 333 566 942 1,142 1,428 1,457 1,619 2,028 3,421 4,830 11,046 20,090 18,567 17,681 17,409 17,526 17,851 18,205 19,428 20,450 20,482 19,368 18,511 20,955 21,166 21,479 21,849 22,941 23,336 23,336 23,361 23,797 24,267

73

Appendix

Table 3.4 E XCHANGE R ATES , A USTRIAN C ROWNS per US dollar, in New York market

January February March April May June July August September October November December

1919

1920

1921

1922

1923

1924

17.09 20.72 25.85 26.03 24.75 29.63 37.24 42.50 68.50 99.50 130.00 155.00

271.43 250.00 206.66 200.00 155.83 145.00 165.00 237.14 255.00 358.33 493.66 659.40

654.00 722.50 676.00 661.00 604.00 720.00 957.00 1,081.50 2,520.00 4,355.00 8,520.00 5,275.00

7,375.00 6,350.00 7,487.50 7,937.50 11,100.00 18,900.00 42,350.00 77,300.00 74,210.00 73,550.00 71,400.00 70,925.00

71,500.00 71,150.00 71,000.00 70,850.00 70,800.00 70,800.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00

70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00 70,760.00

Source: Young (1925, 2:294).

Table 3.5 T HE A USTRIAN B UDGET, 1923–25 in millions of schillings Closed accounts Item

1923

1924

1925

Total revenue Current expenditures Deficit (−) or surplus (+) Capital expenditures Total balance

697.4 779.6 −82.2 76.0 −158.2

900.6 810.0 +90.6 103.6 −13.0

908.5 741.4 +167.1 90.6 +76.5

Source: Pasvolsky (1928, 127). Note: 1 schilling = 10,000 paper crowns.

74

Chapter 3: The Ends of Four Big Inflations

Table 3.6 A USTRIAN N ATIONAL B ANK B ALANCE S HEET in millions of crowns

Gold 1923: January February March April May June July August September October November December 1924: January February March April May June July August September October November December 1925: January February March April

Foreign exchange and Loans and Treasury Notes in currency discounts bills circulation Deposits

49,304 83,438 86,097 73,270 73,391 73,391 73,391 73,391 73,391 62,117 62,117 83,177

1,058,244 1,029,134 1,336,385 1,439,999 1,682,209 2,532,316 2,947,216 3,050,085 3,126,599 3,356,232 3,504,652 3,832,132

731,046 728,884 821,397 741,858 875,942 730,848 658,966 647,936 863,317 1,069,340 1,094,620 1,325,380

2,556,848 2,552,682 2,550,159 2,550,159 2,550,159 2,547,212 2,539,777 2,538,719 2,537,661 2,536,604 2,535,547 2,534,490

4,110,551 4,207,992 4,459,117 4,577,382 4,837,042 5,432,619 5,684,134 5,894,786 6,225,109 6,607,839 6,577,616 7,125,755

279,092 178,752 329,109 226,273 343,339 362,237 535,121 413,383 373,673 414,882 617,321 649,424

91,274 105,536 106,663 107,059 107,443 107,762 108,342 108,256 108,950 109,327 110,643 110,890

3,811,148 3,921,594 3,953,872 3,669,333 3,344,337 3,178,339 3,254,477 3,453,177 3,724,916 4,032,485 4,312,355 4,770,548

1,253,110 1,737,334 1,733,400 2,131,984 2,660,449 3,092,470 3,304,876 3,226,962 2,852,688 2,379,700 1,945,627 1,881,593

2,533,434 2,532,379 2,295,428 2,294,471 2,259,839 2,237,794 2,231,173 2,219,459 2,210,527 2,202,106 2,196,181 2,178,185

6,735,109 7,364,441 7,144,901 7,315,471 7,554,620 7,774,958 7,995,647 8,002,142 7,998,509 8,213,003 8,072,021 8,387,767

536,982 558,800 752,814 696,141 641,001 741,400 896,032 997,677 890,537 502,579 484,750 533,450

111,314 111,474 111,649 112,168

3,337,911 3,310,032 3,202,802 3,474,672

1,545,295 1,285,158 1,047,719 1,059,069

2,172,491 2,150,151 2,107,949 2,088,777

7,902,217 7,957,242 7,897,792 7,976,420

438,390 315,771 295,498 236,957

Source: Young (1925, 2:291).

75

Appendix

Table 3.7 N UMBER OF A USTRIAN U NEMPLOYED IN R ECEIPT OF R ELIEF in thousands Beginning of January April July October

1922

1923

1924

1925

1926

17 42 33 38

117 153 93 79

98 107 64 78

154 176 118 119

208 202 151 148

Source: League of Nations (1926, 87).

Table 3.8 H UNGARIAN B UDGET E STIMATES , 1920–24 in millions of paper crowns

1920–21 1921–22 1922–23 1923–24

Revenue

Expenditures

10,520 20,296 152,802 2,168,140

20,210 26,764 193,455 3,307,099

Source: Pasvolsky (1928, 299).

Deficit

Percentage of expenditures covered by issues of paper money

9,690 6,468 40,653 1,138,959

47.9 24.1 21.0 34.4

76

Chapter 3: The Ends of Four Big Inflations

Table 3.9 B ALANCE S HEET OF H UNGARIAN C ENTRAL B ANK OR S TATE N OTE I NSTITUTE in millions of kronen

1921: January February March April May June July August September October November December 1922: January February March April May June July August September October November December 1923: January February March April May

Gold coin and bullion

Silver coin

Foreign currency and exchange

— — — — — — — 4 5 12 12 12

— — — — — — — 1 1 1 1 1

— — — — — — — — — — — —

10,924 13,202 12,862 12,178 11,847 11,693 11,787 17,799 20,994 22,403 23,650 23,859

13 13 13 13 13 13 13 13 14 14 15 16

1 1 1 1 1 1 1 1 1 1 1 1

— — — — — — — — — — — —

24,195 23,952 24,574 25,120 25,326 25,445 28,783 37,617 46,963 51,631 49,246 50,702

14 23 23 23 23

1 1 1 1 1

— — — — —

54,516 58,358 71,284 83,800 93,396

Bills discounted

77

Appendix

Table 3.9 B ALANCE S HEET OF H UNGARIAN C ENTRAL B ANK OR S TATE N OTE I NSTITUTE (continued) in millions of kronen

Advances on securities

Advances to treasury

Notes in circulation

Current accounts and deposits

195 162 160 110 111 108 107 1,199 1,194 1,185 1,176 1,158

— — — — — — — — — 900 1,000 900

15,206 15,571 15,650 13,114 13,686 18,096 15,799 17,326 20,845 23,643 24,742 25,175

3,851 5,531 5,246 6,802 5,760 1,162 3,532 2,975 2,407 2,154 2,353 2,240

1,147 1,504 1,565 1,565 1,560 1,556 1,546 1,773 1,848 1,728 1,861 2,016

1,300 1,900 3,000 4,100 5,500 6,900 7,200 7,600 8,900 12,000 12,500 16,500

25,680 26,758 29,327 30,580 31,930 33,600 38,357 46,242 58,458 70,005 72,016 75,887

2,488 2,354 2,224 2,901 3,289 3,741 3,929 5,417 5,929 5,189 6,408 4,761

2,007 2,013 2,584 2,817 1,763

20,000 24,000 29,000 37,000 47,200

73,717 75,135 82,205 100,101 119,285

5,888 6,600 11,152 9,793 10,609

78

Chapter 3: The Ends of Four Big Inflations

Table 3.9 B ALANCE S HEET OF H UNGARIAN C ENTRAL B ANK OR S TATE N OTE I NSTITUTE (continued) in millions of kronen

Gold coin and bullion June July August September October November December 1924: January February March April May Junea July August September October November December 1925: January February March April

Silver coin

Foreign currency and exchange — — — — — — —

Bills discounted

23 22 22 22 23 23 23

1 1 1 1 1 1 1

120,608 165,927 273,605 380,454 494,501 531,403 582,117

24 23 24 24 24 246,947 441,832 449,945 540,425 503,377 508,411 532,842

1 1 1 1 1 9,823 13,545 13,558 13,560 13,301 13,301 13,299

— — — — — 681,268 1,110,926 1,382,885 1,385,880 1,658,674 1,816,102 1,933,356

654,294 746,471 802,756 1,125,898 1,420,385 1,192,517 1,257,597 1,438,454 1,756,636 1,872,385 1,984,540 1,976,888

509,848 596,334 669,107 653,534

12,373 12,374 12,374 12,136

1,967,314 1,989,096 1,984,006 2,081,998

1,848,620 1,676,594 1,514,532 1,485,898

Source: Young (1925, 2:321). Note: Figures prior to June 1924 are those of the State Note Institute. The Hungarian National Bank opened 24 June 1924 and took over the affairs of the institute. a After

this date gold and silver holdings are shown in terms of paper crowns. Other changes were also made in the presentation of accounts after the opening of the new Hungarian National Bank in June.

79

Appendix

Table 3.9 B ALANCE S HEET OF H UNGARIAN C ENTRAL B ANK OR S TATE N OTE I NSTITUTE (continued) millions of kronen

Advances on securities

Advances to treasury

Notes in circulation

Current accounts and deposits

2,490 1,762 1,789 1,776 1,663 1,047 935

59,700 79,700 143,000 243,000 269,000 306,000 401,000

154,996 226,285 399,487 588,810 744,926 853,989 931,337

12,742 21,977 23,629 60,246 60,176 74,970 84,791

9,346 34,023 4,598 12,456 13,437 17,566 — — — — — —

526,000 699,000 824,000 944,000 1,054,000 1,980,000 1,980,000 1,978,130 1,977,306 1,976,455 1,975,631 1,974,781

1,084,677 1,278,437 1,606,875 2,098,091 2,486,257 2,893,719 3,277,943 3,659,757 4,115,925 4,635,090 4,442,644 4,513,990

105,481 164,838 253,935 308,121 527,137 1,135,710 1,424,578 1,473,231 1,416,400 1,465,356 1,929,754 2,069,468

— — — —

1,973,930 1,973,163 1,969,809 1,968,987

4,449,650 4,237,985 4,270,096 4,526,216

2,138,629 2,542,262 2,552,762 2,470,507

80

Chapter 3: The Ends of Four Big Inflations

Table 3.10 H UNGARIAN P RICE AND E XCHANGE R ATE Hungarian index of pricesa 1921: July August September October November December 1922: January February March April May June July August September October November December 1923: January February March April May June July August September October November December

Cents per crown in New York

4,200 5,400 6,250 6,750 8,300 8,250

0.3323 0.2629 0.1944 0.1432 0.1078 0.1512

8,100 8,500 9,900 10,750 11,000 12,900 17,400 21,400 26,600 32,900 32,600 33,400

0.1525 0.1497 0.1256 0.1258 0.1261 0.1079 0.0760 0.0595 0.0423 0.0402 0.0413 0.0430

38,500 41,800 66,000 83,500 94,000 144,500 286,000 462,500 554,000 587,000 635,000 714,000

0.0392 0.0395 0.0289 0.0217 0.0191 0.0140 0.0097 0.0056 0.0055 0.0054 0.0054 0.0052

81

Appendix

Table 3.10 H UNGARIAN P RICE AND E XCHANGE R ATE (continued)

1924: January February March April May June July August September October November December 1925: January February March

Hungarian index of pricesa

Cents per crown in New York

1,026,000 1,839,100 2,076,700 2,134,600 2,269,600 2,207,800 2,294,500 2,242,000 2,236,600 2,285,200 2,309,500 2,346,600

0.0039 0.0033 0.0015 0.0014 0.0012 0.0011 0.0012 0.0013 0.0013 0.0013 0.0013 0.0013

2,307,500 2,218,700 2,117,800

0.0014 0.0014 0.0014

Source: Young (1925, 2:323). a From

July 1921 through November 1923, the index numbers represent retail prices and are based on 60 commodities with July 1914 = 100. From December 1923 through March 1925, the figures are based on wholesale prices computed by the Hungarian Central Statistical Office. They refer to the prices of 52 commodities on the last day of the month with 1913 = 100.

Table 3.11 H UNGARIAN B UDGET, 1924–25 in millions of crowns

Period Jul.–Dec. 1924 Fiscal year 1924–25

Preliminary treasury accounts

Reconstruction scheme

ExpenReceipts ditures

Surplus Surplus (+) or Expen- (+) or deficit Receipts ditures deficit

208.0 453.1

Source: Pasvolsky (1928, 322).

205.9 422.8

+2.1 +30.3

143.8 293.8

186.3 393.9

−42.5 −100.1

82

Chapter 3: The Ends of Four Big Inflations

Table 3.12 N UMBER OF U NEMPLOYED IN H UNGARY a in thousands of workers End of January February March April May June July August September October November December

1924

1925

1926

— — — 22 23 25 31 30 20 30 31 33

37 37 37 36 30 34 32 27 25 23 26 27

28 29 29 26 28 26

Source: League of Nations (1926, 50). a Figures

relate only to members of Union of Socialist Workers.

Table 3.13 P OLISH R ECEIPTS AND E XPENDITURES in thousands of zloty

Receipts: Administration State enterprises Monopolies Total Expenditures: Administration State enterprises Monopolies Total Deficit Surplus

1921

1922

261,676 11,413 72,222 345,311

467,979 14,556 47,893 530,428

765,263 115,589 — 880,852 535,541 —

734,310 145,003 — 879,313 348,885 —

1923

1924

1925

— — — 426,000

— — — 1,703,000

1,491,743 133,530 356,611 1,981,884

— — — 1,119,800 692,000 —

— — — 1,629,000 — 74,000

1,830,231 106,343 45,019 1,981,593 — 251

Source: Young (1925, 2:183). Note: Conversion from marks to zloty was made on the following basis: 1921, 1 zloty = 303.75 marks. First quarter 1922, 1 zloty = 513.52 marks; second quarter, 691.49 marks; third quarter, 1,024.97 marks; and fourth quarter, 1,933.87 marks.

83

Appendix

Table 3.14 B ALANCE S HEET OF B ANK OF P OLAND , 1918–25 end of month figures (Polish State Loan Bank Prior to May 1924) in millions of marks

Month 1918: October November December 1919: January February March April May June July August September October November December 1920: January February March April May June July August September October November December

Golda

Silvera (including base coin)

Balances with foreign banks

— — —

— — —

— — —

— — 3.7 3.7 3.7 4.9 5.7 6.1 6.3 6.5 6.6 6.6

— — 4.2 4.4 8.9 14.8 20.1 20.5 21.6 24.3 24.6 25.5

— — 3.9 9.4 5.8 14.6 13.3 20.3 69.8 91.0 151.6 344.6

6.6 6.8 6.8 6.8 6.8 6.8 6.8 9.0 9.1 9.5 10.1 12.4

25.5 25.9 25.9 25.9 25.9 25.9 25.9 33.8 34.1 34.4 35.4 37.6

244.1 565.7 685.4 685.5 565.7 894.7 1,130.9 1,273.4 174.9 236.7 203.8 80.7

84

Chapter 3: The Ends of Four Big Inflations

Table 3.14 B ALANCE S HEET OF B ANK OF P OLAND , 1918–25 (continued) Advances Government

Note circulation

Discounts

Commercial

7.0 7.0 6.4

180.8 184.0 183.7

— 13.9 117.8

880.2 930.5 1,023,8

5.0 4.2 3.5 2.5 1.8 1.3 1.1 0.7 0.1 0.3 3.4 3.9

194.7 196.4 189.7 192.8 193.2 185.9 193.9 107.4 218.9 242.4 270.2 243.8

209.9 315.0 400.0 575.0 925.0 1,125.0 1,925.0 2,525.0 3,225.0 4,375.0 5,375.0 6,825.0

1,098,1 1,160.0 1,223.2 1,346.0 1,548.3 1,784.6 2,087.9 2,466.6 2,964.7 3,723.6 4,236.21 5,316.3

3.7 6.4 8.2 14.8 47.2 161.4 325.9 465.8 333.9 259.1 396.0 611.6

278.5 303.0 319.1 316.7 320.9 488.2 9,847,5 1,466.1 1,862.9 2,527.0 3,278.4 3,999.2

8,275.0 10,75.0 14,775.0 19,375.0 22,375.0 27,625.0 33,375.0 40,625.0 40,625.0 46,925.0 49,625.0 59,625.0

6,7198.9 8,300.3 10,690.6 16,027.9 17,934.7 21,730.1 26,311.4 31,085.8 33,203.5 38,456.8 43,236.2 49,361.5

85

Appendix

Table 3.14 B ALANCE S HEET OF B ANK OF P OLAND , 1918–25 (continued)

Month 1921: January February March April May June July August September October November December 1922: January February March April May June July August September October November December 1923: January February March April May

Golda

Silvera (including base coin)

Balances with foreign banks

12.7 12.8 13.1 13.4 13.5 14.3 19.1 19.2 19.4 20.2 22.6 24.9

39.2 39.2 39.8 40.3 40.1 41.1 41.5 42.0 42.5 42.9 43.5 43.9

205.8 476.0 908.5 870.7 536.5 493.6 601.3 368.7 1,217.5 2,341.3 7,040.1 12,707.9

26.3 28.3 29.0 29.5 30.1 30.9 31.5 31.6 32.4 33.5 33.8 41.0

44.2 44.4 44.7 45.2 45.3 45.3 45.4 45.4 45.4 45.4 45.4 45.4

13,614.2 14,207.7 1,156.4 7,388.0 23,073.4 20,521.4 21,741.0 51,747.2 67,384.1 64,060.9 78,959.0 48,580.4

41.1 41.4 41.7 41.9 41.9

44.1 44.1 44.2 44.2 44.3

34,721.8 71,883.7 29,868.7 50,851.9 43,900.7

86

Chapter 3: The Ends of Four Big Inflations

Table 3.14 B ALANCE S HEET OF B ANK OF P OLAND , 1918–25 (continued) Advances Discounts

Commercial

Government

Note circulation

1,040.2 955.1 781.0 927.0 1,395.2 1,557.3 2,504.2 3,885.4 6,237.3 9,529.5 14,347.2 15,324.4

4,100.2 4,143.5 4,745.7 4,994.4 4,979.0 5,306.5 6,291.5 7,776.9 9,878.6 12,022.3 15,144.3 19,300.0

65,625.0 8,777,125.0 93,625.0 106,625.0 117,625.0 130,625.0 140,625.0 158,000.0 178,000.0 198,500.0 214,000.0 221,000.0

55,079.5 62,560.4 74,087.4 86,755.3 94,575.8 102,697.3 115,242.3 133,734.2 152,792.1 182,777.3 207,029.0 229,537.6

15,951.6 19,555.0 25,451.1 28,688.8 34,555.0 46,629.8 47,661.2 56,366.6 64,093.0 81,781.9 107,320.1 133,400.8

21,776.9 22,327.7 25,473.3 29,063.7 26,067.0 24,499.5 24,054.4 21,079.9 22,239.4 26,576.5 41,278.1 47,904.1

227,350.0 230,600.0 232,100.0 220,000.0 217,000.0 235,000.0 260,000.0 285,000.0 342,000.0 453,500.0 519,500.0 675,600.0

239,615.3 247,209.5 250,665.5 260,553.8 276,001.1 300,101.1 335,426.6 385,787.5 463,706.0 579,972.7 661,092.4 793,437.5

174,950.1 219,610.7 274,657.8 304,725.4 449,440.7

51,899.9 61,037.1 85,323.2 156,815.4 217,162.3

799,500.0 1,085,000.0 1,752,000.0 2,161,500.0 2,377,000.0

909,160.3 1,177,300.8 1,841,205.6 2,332,396.8 2,733,794.1

87

Appendix

Table 3.14 B ALANCE S HEET OF B ANK OF P OLAND , 1918–25 (continued)

Month

Silvera (including base coin)

Golda

1923: June July August September October November December 1924: January February March April

Balances with foreign banks

43.9 46.9 48.0 53.2 54.2 54.3 54.9

39.8 34.8 32.9 20.7 19.1 19.5 19.6

276,506.3 384,375.1 340,354.4 857,084.5 1,510,794.3 6,499,791.5 57,499,741.7

66.2 66.7 68.0 55.7

19.8 19.8 20.3 21.2

91,533,085.2 172,626,128.8 220,658,210.7 277,340,925.7

After conversion of State Loan Bank into Bank of Poland, figures in gold zlotys; no ciphers omitted; 1 zloty = 19.3 cents May June July August September October November December 1925: January February March April

11,684,963b 83,392,914b 93,683,430b 98,288,324b 99,900,015b 100,686,634 16,521,223 102,809,285 21,951,828 103,362,870 27,543,698

214,191,336 256,972,386 272,137,898 266,390,583 233,646,562 241,894,738 247,034,974 269,045,551

104,249,258 107,032,735 116,619,825 117,428,697

242,115,258 206,317,320 259,392,902 216,114,621

27,658,749 27,481,871 28,158,597 28,358,000

Source: Young (1925, 2:348). a Gold

at par; silver coin at face value.

b Gold

and silver.

88

Chapter 3: The Ends of Four Big Inflations

Table 3.14 B ALANCE S HEET OF B ANK OF P OLAND , 1918–25 (continued) Advances Discounts

Commercial

Government

Note circulation

627,339.5 758,112.8 1,372,150.9 2,077,128.6 3,540,434.4 8,467,033.7 20,588,037.9

310,862.7 390,850.9 637,268.2 670,019.6 1,836,712.7 3,951,781.9 28,065,396.8

2,996,500.0 4,190,500.0 6,473,000.0 10,265,500.0 19,080,500.0 42,854,000.0 111,332,000.0

3,566,649.1 4,478,709.0 6,871,776.5 11,197,737.8 23,080,402.2 53,217,494.6 125,371,955.3

43,916.802.8 67,216,289.7 138,649,934.8 199,248,956.4

54,181,445.2 83,829,440.5 81,231,988.5 60,589,081.0

238,200,000.0 291,700,000.0 291,700,000.0 291,700,000.0

313,659,830.0 528,913,418.7 596,244,205.6 570,697,550.5

After conversion of State Loan Bank into Bank of Poland, figures in gold zlotys; no ciphers omitted; 1 zloty = 19.3 cents 126,522,906 138,862,243 166,713,469 199,710,736 233,788,177 245,054,984 249,560,999 256,954,853

1,801.936 5,826,971 8,236,693 8,224,610 9,230,850 12,374,342 12,371,166 23,897,766

— — — — — — — —

244,977,010 334,405,730 394,262,550 430,263,045 460,383,770 503,701,830 497,600,470 550,873,960

270,423,615 286,229,180 306,562,690 294,632,508

23,468,829 28,467,930 25,477,638 27,319,944

— 18,222,212 403,354 35,977,630

553,174,980 549,637,420 563,171,945 567,178,830

89

Appendix

Table 3.15 P OLISH I NDEX N UMBERS OF W HOLESALE P RICES , 1921–25 Wholesale price indexa

Year

Month

1921 January February March April May June July August September October November December 1922 January February March April May June July August September October November December 1923 January February March

Wholesale price indexa 25, 139 31, 827 32, 882 31, 710 32, 639 35, 392 45, 654 53, 100 60, 203 65, 539 58, 583 57, 046 59, 231 63, 445 73, 465 75, 106 78, 634 87, 694 101, 587 135, 786 152, 365 201, 326 275, 647 346, 353 544, 690 859, 110 988, 500

Source: Young (1925, 2:349). a 1914

= 100.

Year

Month

On paper currency basis

1923 April 1, 058, 920 May 1, 125, 350 June 1, 881, 410 July 3, 069, 970 August 5, 294, 680 September 7, 302, 200 October 27, 380, 680 November 67, 943, 700 December142, 300, 700 1924 January 242, 167, 700 February 248, 429, 600 March 245, 277, 900 April 242, 321, 800 May June July August September October November December 1925 January February March April May

On zloty basis

89 95 73 81 95 107 112 110 109 104 101 102 109 112 116 117 118 120 121 122 118 118

90

Chapter 3: The Ends of Four Big Inflations

Table 3.16 P OLISH E XCHANGE R ATES , 1919–25

Year

Month

Cents per Polish mark

1919

July August September October November December January February March April May June July August September October November December January February March April

6.88 5.63 3.88 3.08 1.88 1.29 0.70 0.68 0.67 0.60 0.51 0.59 0.61 0.47 0.45 0.37 0.26 0.16 0.145 0.130 0.132 0.130

May June

0.124 0.082

July August September October November December

0.0516 0.0489 0.0256 0.0212 0.0290 0.0313

January February March April May June July August

0.0327 0.0286 0.0236 0.0262 0.0249 0.0237 0.0185 0.0135

1920

1921

1922

Source: Young (1925, 2:350).

Year

Month

Cents per Polish mark

1922

September October November December January February March April May June July August September October November December January February March April May

0.0127 0.0095 0.0065 0.0057 0.0043 0.0025 0.0024 0.0023 0.0021 0.0013 0.0007 0.0004 0.00035 0.0001113 0.0000502 0.0000234 0.0000116 0.0000109 0.0000113 0.0000114 —

1923

1924

Cents per zloty

1925

June July August September October November December

10.29 19.25 19.23 19.22 19.22 19.21 19.20

January February March April May June

19.18 19.18 19.18 19.18 19.18 19.18

Appendix

Table 3.17 P OLISH U NEMPLOYED 1921: January February March April May June July August September October November December 1922: January February March April May June July August September October November December

74,000 90,000 80,000 88,000 130,000 115,000 95,000 65,000 70,000 78,000 120,000 173,000 221,444 206,442 170,125 148,625 128,916 98,581 85,240 69,692 68,000 61,000 62,000 75,000

1923: January February March April May June July August September October November December 1924: January February March April May June July August September October November December

81,184 106,729 114,576 112,755 93,731 76,397 64,563 56,515 — — — 67,581 100,580 110,737 112,583 109,000 84,000 97,870 149,097 159,820 155,245 147,065 150,180 159,060

Source: Statistiches Jahrbuch für das Deutsche Reich (1924/25).

91

92

Chapter 3: The Ends of Four Big Inflations

Table 3.18 G ERMAN W HOLESALE P RICES , 1914–24 Year

Month

1914

January February March April May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March

1915

1916

1917

Price index 96 96 96 95 97 99 99 109 111 118 123 125 126 133 139 142 139 139 150 146 145 147 147 148 150 151 148 149 151 152 161 159 154 153 151 151 156 158 159

Year

1918

1919

1920

Month

Price index

April May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March April May June

163 163 165 172 203 199 201 203 203 204 198 198 204 203 209 208 235 230 234 234 245 262 270 274 286 297 308 339 422 493 562 678 803 1, 260 1, 690 1, 710 1, 570 1, 510 1, 380

93

Appendix

Table 3.18 G ERMAN W HOLESALE P RICES , 1914–24 (continued) Year

1921

1922

Month

Price index

July August September October November December January February March April May June July August September October November December January February March April May June July August September

1, 370 1, 450 1, 500 1, 470 1, 510 1, 440 1, 440 1, 380 1, 340 1, 330 1, 310 1, 370 1, 430 1, 920 2, 070 2, 460 3, 420 3, 490 3, 670 4, 100 5, 430 6, 360 6, 460 7, 030 10, 160 19, 200 28, 700

Year

1923

1924

Month

Price index

October November December January February March April May June July August September October November December January February March April May June July August September October November December

56, 600 115, 100 147, 480 278, 500 588, 500 488, 800 521, 200 817, 000 1, 938, 500 7, 478, 700 94, 404, 100 2, 394, 889, 300 709, 480, 000, 000 72, 570, 000, 000, 000 126, 160, 000, 000, 000 117, 320, 000, 000, 000 116, 170, 000, 000, 000 120, 670, 000, 000, 000 124, 050, 000, 000, 000 122, 460, 000, 000, 000 115, 900, 000, 000, 000 115a 120a 127a 131a 129a 131a

Source: Young (1925, 1:503). a On basis of prices in reichsmarks. [1 reichsmark = 1 trillion (10 12 ) former marks.]

94

Chapter 3: The Ends of Four Big Inflations

Table 3.19 G ERMAN E XCHANGE R ATES , 1914–25 Year

Month

1920 January February March April May June July August September October November December 1921 January February March April May June July August September October November December 1922 January February March April May June July

Cents per mark Year 1.69 1.05 1.26 1.67 2.19 2.56 2.53 2.10 1.72 1.48 1.32 1.37 1.60 1.64 1.60 1.57 1.63 1.44 1.30 1.19 0.96 0.68 0.39 0.53 0.52 0.48 0.36 0.35 0.34 0.32 0.20

Month

1922 August September October November December 1923 January February March April May June July August September October November December 1924 January February March April May June July August September October November December 1925a January

Cents per mark 0.10 0.07 0.03 0.01 0.01 0.007 0.004 0.005 0.004 0.002 0.001 0.000, 3 0.000, 033, 9 0.000, 001, 88 0.000, 000, 068 0.000, 000, 000, 043 0.000, 000, 000, 022, 7 22.6 21.8 22.0 22.0 22.3 23.4 23.9 23.8 23.8 23.8 23.8 23.8 23.8

Source: Young (1925, 1:503). a Cents per rentenmark and (after October 1924) per reichsmark. 1 rentenmark is equivalent to 1 reichsmark or 1 billion former paper marks. The reichsmark is the equivalent of the gold mark worth 23.82 cents.

95

Appendix

Table 3.20 R EAL G ERMAN R EVENUES AND E XPENDITURES , C ALCULATED ON THE B ASIS OF THE C OST- OF -L IVING I NDEX in millions of gold marks Revenue

Taxes 1920–21 1921–22 1922–23 1923–24 (first 9 months)

Deficit covered by loan Sundries transactions

4, 090.8 5, 235.7 3, 529.1 1, 496.1

132.9 100.5 51.4 180.6

7, 041.9 6, 627.4 6, 384.5 11, 836.5

Total 11, 265.6 11, 963.6 9, 965.0 13, 513.2

Expenditures Repayment Interest on Execution of of floating floating Subsidies to Versailles debt debt railroads Treaty Sundries 821.7 1, 039.5 81.0 —

— 811.6 344.4 931.0

Source: Young (1925, 2:393).

— 1, 114.4 1, 685.5 3, 725.0

— 5, 110.6 3, 600.0 —

— 5, 738.4 4, 254.1 —

Total 11, 265.6 11, 963.4 9, 965.0 13, 513.2

96

Chapter 3: The Ends of Four Big Inflations

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 Gold Held abroad 1921: January February March April May June July August September October November December 1922: January February March April May June July August September October November December 1923: January February March April May June July

In vault

Total

Other metallic currency

1,091,635 1,091,616 1,091,604 1,091,579 1,091,574 1,091,563 1,091,557 1,023,708 1,023,704 993,631 993,698 995,392

6,377 7,974 9,012 8,857 9,482 11,205 14,258 16,848 16,064 14,237 12,995 11,612 13,493 15,590 19,286 19,269 18,768 19,743 19,990 22,656 28,884 63,143 64,782 69,195 69,528 69,424 6,050,558 13,597,431 15,026,230 19,185,800 23,815,964

16,144 47,207 50,032 50,032 50,032 50,032 50,032 50,032 50,032 50,032

980,733 953,660 952,832 953,827 954,828 954,827 954,824 954,821 954,816 954,811

995,690 996,388 996,877 1,000,867 1,002,864 1,003,859 1,004,860 1,004,859 1,004,856 1,004,853 1,004,848 1,004,843

50,032 50,032 164,952 217,282 181,032 140,032 90,032

954,800 954,797 839,878 702,627 576,883 576,880 506,319

1,004,832 1,004,829 1,004,830 919,909 757,915 716,912 596,351

Appendix

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued) Total coin and bullion

Treasury and loan bank notes

Notes of other banks

1,098,012 1,099,590 1,100,616 1,100,436 1,101,056 1,102,768 1,105,815 1,040,556 1,039,768 1,007,868 1,006,693 1,007,004

22,810,443 21,982,303 23,836,278 24,149,059 14,361,888 8,311,208 5,397,594 3,044,888 3,128,791 4,589,992 4,472,583 6,963,607

2,266 1,502 1,407 1,393 2,000 1,738 1,198 3,644 2,618 3,082 4,066 2,084

1,009,183 1,011,978 1,016,163 1,020,136 1,021,632 1,023,602 1,024,850 1,027,515 1,033,740 1,067,996 1,069,630 1,074,038

5,807,290 4,633,744 6,272,763 3,898,700 4,901,822 14,605,085 13,083,023 23,239,947 21,828,959 41,026,323 77,786,910 238,481,105

4,059 3,841 4,553 4,123 4,877 4,257 4,890 3,759 1,531 3,625 9,526 10,393

1,074,360 1,074,253 7,055,388 14,517,340 15,784,145 19,902,712 24,412,315

380,902,256 708,305,283 1,147,004,307 1,464,270,320 1,892,215,265 2,866,836,761 3,987,474,460

14,009 111,870 166,484 350,587 343,582 472,566 627,594

97

98

Chapter 3: The Ends of Four Big Inflations

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued) Gold Held abroad 1923: August September October 15 November 30 November December 1924: January February March April May June July August September 15 October 31 October November December

In vault

Total

Other metallic currency

11,312 11,312 11,312 11,312 11,312 21,312

499,174 432,615 455,713 455,713 455,713 445,718

510,486 443,927 467,025 467,025 467,025 467,030

20,935,658 20,410,169 10,404,528 5,103,187 4,723,004 1,887,375

21,312 21,312 21,247 16,138 16,138 30,318 44,948 77,083 136463 172143 240,033 240,033 207,114

445,719 443,553 443,593 425,689 425,775 431,869 439,374 439,401 441,456 441,482 454,191 455,454 552,495

467,031 464,865 464,840 441,827 441,913 462,187 484,322 516,484 577,919 613,625 694,224 695,487 759,609

782,774 327,305 416,602 273 6,187 23,498 26,107 26,066 21,729 59,359 55,437 49,525 46,379

99

Appendix

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued) Treasury and loan bank notes

Notes of other banks

21,446,144 20,854,098 10,871,553 5,570,212 5,190,029 2,354,405

14,176b 941,064b 114,874,046b 1,996,264,991b 178,540,780,992b 202,232,340,999b

32,195,193 133,223,352 40,698b 1,888,242b 82,686,780b 33,009,141b

1,249,805 792,170 881,442 442,100 448,100 485,685 510,429 542,550 599,648 672,984 749,661 745,012 805,988

9,256 9,539 4,124 1,252 982 698 473 415 340

7 1 1 1

Total coin and bullion

Foreign exchange

204,541 231,407 231,829 253,202

Rentenbank notes

3 6,788 7,625

177,780 15,591 225,389 242,836 321,836 278,188 267,997 237,470 222,469

100

Chapter 3: The Ends of Four Big Inflations

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued) Discounted bills

Treasury bills 1921: January February March April May June July August September October November December 1922: January February March April May June July August September October November December 1923: January February March April May June July

Commercial bills

Total discounted treasury and commercial bills

50,594,540 53,690,412 64,533,894 58,841,630 62,953,604 79,607,790 79,981,967 84,043,891 98,422,137 98,704,768 114,023,417 132,330,906

2,742,406 2,760,927 2,268,745 2,052,099 1,809,936 1,565,406 1,135,529 1,002,497 1,142,218 881,474 1,445,667 1,061,754

53,336,946 56,451,339 66,802,639 60,893,729 64,763,540 81,173,496 81,117,496 85,046,388 99,564,355 99,586,242 115,469,084 133,392,660

126,160,402 134,251,808 146,531,247 155,617,524 167,793,922 186,125,747 207,858,232 249,765,773 349,169,650 477,201,494 672,222,197 1,184,464,359

1,592,416 1,856,936 2,151,677 2,403,044 3,376,599 4,751,748 8,122,066 21,704,341 50,234,414 101,155,267 246,948,596 422,235,296

127,752,818 136,108,744 148,682,924 158,020,568 171,170,521 190,877,495 215,980,298 271,470,114 400,004,064 578,356,761 919,170,793 1,606,699,655

1,609,081,121 2,947,363,994 4,552,011,661 6,224,899,348 8,021,904,840 18,338b 53,752b

697,216,424 1,829,341,080 2,372,101,757 2,986,116,724 4,014,693,720 6,914,198,630 18,314b

2,306,297,545 4,776,705,074 6,924,113,418 9,211,016,072 12,036,598,560 25,252,198,630 72,066b

Appendix

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued)

Advances 8,881 11,522 2,805 9,238 16,624 6,079 10,686 7,704 3,289 47,775 90,370 8,476 20,548 62,305 20,688 134,314 54,361 58,994 141,276 172,966 61,516 624,368 51,425,030a 773,974 95,316,552 27,422,282 2,132,906 20,466,948 61,030,322 188,548,574 2,553,177,597

Securities

Notes in circulation

147,126 185,788 217,044 225,777 258,664 282,716 283,381 258,319 277,977 282,179 247,699 195,912

66,620,804 67,426,959 69,417,228 70,839,725 71,838,866 75,321,095 77,390,853 80,072,721 86,384,286 91,527,679 100,943,632 113,639,464

198,725 25,362 205,936 229,242 199,314 307,564 313,488 241,162 416,193 502,348 381,068 469,972

115,375,766 120,026,387 130,671,352 140,420,057 151,949,179 169,211,792 189,794,722 238,147,160 316,869,799 469,456,818 754,086,109 1,280,094,831

483,318 1,209,935 1,690,011 1,207,105 697,611 344,819 1,422,291

1,984,496,469 3,512,787,777 5,517,919,651 6,545,984,355 8,563,749,470 17,291b 43,595b

101

102

Chapter 3: The Ends of Four Big Inflations

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued) Discounted bills

Treasury bills 1923: August September October 15 November 30 November December 1924:d January February March April May June July August September 15 Octobere 31 October November December

987,219b 45,216,224b 6,578,650,939b 189,801,468,187b 96,874,330,250b c

— — — — — — — — — — — — —

Commercial bills

Total discounted treasury and commercial bills

164,644b 3,660,094b 1,058,129,855b 39,529,577,254b 347,301,037,776b 322,724,948,986b

1,151,863b 48,876,318b 7,636,780,794b 229,331,045,441b 444,175,368,026b 322,724,948,986b

— — — — — — — — — — — — —

755,866 1,165,649 1,767,443 1,916,969 1,954,930 1,897,959 1,798,097 1,860,843 2,169,684 2,153,943 2,339,616 2,290,166 2,064,094

Appendix

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued)

Advances

Securities

Notes in circulation

25,261b 15,539,853 663,200b b 98,522 1,801,579,570 28,228,815b 41,787,532b 9,536,953b 2,496,822,909b 535,714,637b 8,901,495b 92,844,720,743b 7,742,665,263b 336,495,629b 400,267,640,302b 268,325,819,530b 336,520 306,618 143,102 156,362 128,597 108,789 62,489 59,983 54,424 15,947 33,443 18,628 16,960

65,791,385b 496,507,424,772b 12 483,675 25 587,875 533 689,864 91,984 776,949 80,011 926,874 76,378 1,097,309 76,509 1,211,038 76,331 1,391,895 78,305 1,520,511 77,517 1,396,748 77,699 1,780,930 77,808 1,863,200 77,999 1,941,440

103

104

Chapter 3: The Ends of Four Big Inflations

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued) Demand deposits

Public 1921: January February March April May June July August September October November December 1922: January February March April May June July August September October November December 1923: January February March April May June July

Other

Total demand deposits

4,055,904 7,291,052 15,206,381 11,595,618 3,548,492 5,647,805 4,810,026 4,850,843 4,618,087 5,239,628 5,144,615 7,591,343

11,778,060 10,066,036 12,836,292 9,260,271 10,545,201 14,744,903 11,014,130 8,798,756 15,362,208 13,063,035 20,168,499 25,314,330

15,833,964 17,357,088 28,042,673 20,855,889 14,093,693 20,392,708 15,824,156 13,649,599 19,980,295 18,302,663 25,313,114 32,905,673

5,286,950 5,806,922 7,743,735 7,577,862 7,711,279 10,125,837 9,197,727 13,708,213 30,034,309 34,270,926 50,353,945 153,190,991

18,125,502 20,719,150 25,614,597 24,038,306 25,416,711 27,047,908 30,778,489 42,416,241 79,978,068 106,508,333 190,615,514 377,335,296

23,412,452 26,526,072 33,358,332 31,616,168 33,127,990 37,173,745 39,976,216 56,124,454 110,012,377 140,779,259 240,969,459 530,526,287

605,205,692 1,329,065,770 1,903,533,291 3,399,871,714 4,410,494,865 8,304,602,339 24,078b

763,264,229 1,582,981,036 2,272,083,584 3,854,274,793 5,063,070,231 9,952,716,666 27,857b

157,058,537 253,915,266 368,550,293 454,403,079 652,575,366 1,648,114,327 3,779,235,298

105

Appendix

Table 3.21 B ALANCE S HEET OF G ERMAN R EICHSBANK , 1921–24 (continued) Demand Deposits

Public

Other

Due to Total demand deposits

1923:d August 206,168 384,912 591,080 September 8,185,467 8,781,150 16,966,617 October 606,660,673 3,261,424,030 3,868,085,703 15 November 72,457,230,513 57,095,366,904 129,552,597,417 30 November 120,478,936,906 253,497,803,653 373,976,740,559 December 303,114,560,004 244,909,637,001 548,024,197,005 1924:d January 492,985 281,320 281,305 February 367,551 282,958 650,509 March 352,360 352,334 704,694 April 474,411 330,561 804,972 May 545,252 259,203 804,455 June 493,043 280,884 773,927 July 452,597 290,390 742,987 August 264,064 297,791 561,855 September 307,515 362,581 670,096 15 Octobere — — 828,511 31 October — — 708,728 November — — 703,938 December — — 820,865

the Rentenbank — — — — — — 200,000 400,000 800,000 800,000 800,000 800,000 800,000 800,000 800,000 800,000 800,000 800,000 456,508

Source: Young (1925, 1:524–25, 528–29). Note: End of month figures, in thousands of current marks; from January 1924 in thousands of rentenmarks or reichsmarks. 1 rentenmark is equivalent to 1 reichsmark or 1 trillion (10 12 ) former paper marks. The reichsmark is the equivalent of the gold mark worth 23.82 cents. a The large increase of advances at the close of November 1922 occurred because the Reichsbank had to take over temporarily the financing of food supplies from the loan bureaus (Darlehnskassen), as the latter were unable to extend the needed accommodation, their outstanding notes having reached the maximum amount permitted by law. b In

billions.

cA

decree of 15 November 1923 discontinued the discounting of treasury bills by the Reichsbank. d See

note above.

e Date

of first statement of reorganized Reichsbank.

106

Chapter 3: The Ends of Four Big Inflations

Table 3.22 O RDINARY R EVENUES AND E XPENDITURES OF THE G ERMAN F EDERAL G OVERNMENT From Wirtschift und Statistik, issued by the Statistisches Reichsamt, in millions of gold marks Excess of revenue (+)

Ordinary revenue

1923: November December 1924: January February March April May June July August September October

Total

Of which taxes yielded

Ordinary expenditures

68.1 333.9

63.2 312.3

— 668.7

— -334.8

520.6 445.0 632.4 579.5 566.7 529.7 622.2 618.2 665.6 714.3

503.5 418.0 595.3 523.8 518.7 472.3 583.1 592.0 609.2 686.7

396.5 462.8 498.6 523.5 459.1 504.5 535.1 597.6 581.6 693.0

+124.1 -17.8 +133.8 +56.0 +107.6 +25.2 +86.9 +20.6 +84.0 +21.3

Source: Young (1925, 1:422).

Table 3.23 I NDEX OF P HYSICAL V OLUME OF P RODUCTION PER C APITA IN G ERMANY Year

Index of production

Year

1920 1921 1922 1923

61 77 86 54

1924 1925 1926 1927

Source: Graham (1930, 287).

Index of production 77 90 86 111

or expenditure (−)

107

Appendix

Table 3.24 C ZECHOSLOVAKIA , R ECEIPTS AND E XPENDITURES 1919–25 Exclusive of expenditures for capital improvements covered by loans 1919

1920

1921

Estimated Actual Estimated Actual Estimated Actual Revenue Ordinary Extraordinary Total Expenditure Ordinary Extraordinary Total Deficit Surplus

2, 614 1, 096 3, 710

— — —

7, 950 2, 477 10, 427

— — 13, 455

15, 923 1, 376 17, 299

— — 21, 894

2, 610 6, 005 8, 615 4, 905 —

— — 7, 450 — —

7, 175 8, 103 15, 278 4, 851 —

— — 13, 931 476 —

10, 672 7, 354 18, 026 727 —

— — 18, 558 — 3, 336

1922

1923

1924

1925

Estimated Actual Estimated Actual Estimated Actual Estimated Actual 17, 291 1, 593 18, 884 13, 289 6, 524 19, 813 929 —

— — 17, 733 — — 18, 663 930 —

17, 961 851 18, 812 13, 605 5, 773 19, 378 565 —

Source: Young (1925, 2:71).

— — 15, 664 — — 16, 540 876 —

15, 987 404 16, 391 12, 200 4, 703 16, 993 603 —

— — — — — — — —

— — 15, 702 — — 15, 974 272 —

— — — — — — — —

108

Chapter 3: The Ends of Four Big Inflations

Table 3.25 N OTE I SSUE OF B ANKING O FFICE OF C ZECHOSLOVAKIA , 1919–24 in thousands of Czech crowns Year

Month

1919

April May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March April

1920

1921

1922

State notes in circulation — — — 161, 106 664, 997 1, 443, 570 2, 512, 199 3, 513, 405 4, 723, 303 5, 574, 688 6, 462, 825 7, 216, 438 7, 216, 438 8, 268, 695 9, 729, 233 9, 267, 874 9, 814, 920 10, 310, 228 10, 920, 514 10, 946, 653 11, 288, 512 10, 888, 319 10, 914, 786 10, 921, 956 10, 928, 560 10, 851, 403 11, 167, 515 11, 134, 327 11, 455, 175 11, 570, 881 12, 327, 159 11, 871, 647 12, 129, 573 11, 230, 065 10, 743, 958 10, 323, 069 10, 075, 757

Source: Young (1925, 2:305–6).

Year

Month

1922

May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March April

1923

1924

1925

State notes in circulation 9, 717, 750 9, 838, 205 9, 916, 077 10, 171, 383 10, 196, 880 10, 139, 366 9, 996, 550 10, 064, 049 9, 222, 434 8, 947, 988 9, 157, 407 9, 567, 369 9, 327, 676 9, 375, 991 9, 448, 086 9, 218, 475 9, 311, 378 9, 278, 999 9, 250, 688 9, 598, 903 8, 820, 093 8, 506, 467 8, 280, 390 8, 198, 653 9, 078, 418 8, 081, 106 8, 090, 034 9, 139, 792 8, 222, 658 8, 585, 847 8, 500, 942 8, 810, 357 7, 916, 540 7, 727, 880 7, 680, 867 7, 525, 934

109

Appendix

Table 3.26 C ZECHOSLOVAKIAN E XCHANGE R ATES , 1919–24 Year

Month

Cents per crown

1919

January February March April May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March

— — — 6.135 — — 5.625 4.575 4.575 3.100 1.950 1.900 1.425 0.975 1.275 1.530 2.195 2.335 2.195 1.810 1.535 1.245 1.165 1.190 1.300 1.290 1.307 1.365 1.460 1.420 1.312 1.225 1.160 1.049 1.038 1.249 1.732 1.855 1.733

1920

1921

1922

Source: Young (1925, 2:307).

Year

Month

Cents per crown

1922

April May June July August September October November December January February March April May June July August September October November December January February March April May June July August September October November December January February March April May June

1.960 1.921 1.924 2.185 2.902 3.231 3.285 3.176 3.097 2.856 2.958 2.969 2.978 2.979 2.993 2.997 2.934 2.995 2.971 2.906 2.925 2.898 2.902 2.902 2.957 2.939 2.936 2.953 2.979 2.993 2.981 2.989 3.018 3.000 2.960 2.970 2.960 2.960 2.960

1923

1924

1925

110

Chapter 3: The Ends of Four Big Inflations

Table 3.27 C ZECHOSLOVAKIAN W HOLESALE P RICES , 1922–24 Year

Month

1922

January February March April May June July August September October November December January February March April May June July August September

1923

Wholesale price index 1, 675 1, 520 1, 552 1, 491 1, 471 1, 471 1, 464 1, 386 1, 155 1, 059 1, 017 999 1, 003 1, 019 1, 028 1, 031 1, 030 1, 001 968 958 957

Source: Young (1925, 2:307). Note: July 1914 = 100.

Year

Month

1923

October November December January February March April May June July August September October November December January February March April May

1924

1925

Wholesale price index 973 965 984 974 999 1, 021 1, 008 1, 015 981 953 986 982 999 1, 013 1, 024 1, 045 1, 048 1, 034 1, 019 1, 006

4 Stopping Moderate Inflations: The Methods of Poincaré and Thatcher Introduction In June 1979 Margaret Thatcher’s administration began governing Great Britain. One of her primary goals was markedly to reduce the rate of inflation, an understandable goal in view of the experience of the past decade when Great Britain’s rate of inflation had on average exceeded the rate of inflation in other industrial countries. Advocates of the two main groups of contemporary theories about inflation dynamics could have told Mrs. Thatcher that achieving that goal would be difficult, although each group would have characterized the nature of the difficulties quite differently. The first group consists of the “momentum” or “core inflation” theories (e.g., see Eckstein (1981)). The second group comprises the rational expectations-equilibrium theories (e.g., see Barro (1976), Lucas (1972a), McCallum (1978)). The first group of theories posits that there is some inherent momentum in the process of inflation itself, and that this momentum or persistence is neither superficial nor merely a reflection of slowly moving deeper forces that themselves cause inflation to behave as it does. Two distinct possible sources of sluggishness in inflation have been proposed. One is the notion of adaptive or autoregressive expectations. According to this doctrine, workers and firms form expectations about future rates of inflation by computing a moving average of current and lagged rates of inflation. The moving average makes expected inflation a simple function of current and past rates of inflation. Further, the weights in the moving average are assumed to be fixed numbers that are independent of the economic environment, including government monetary and fiscal policy, and are taken to characterize the psychology of expectations. Since firms and workers set current and future nominal wages and prices partly as functions of their expected rates 111

112

Chapter 4: Stopping Moderate Inflations

of inflation, this model of inflationary expectations determines the actual rate of inflation partly as a long weighted average of past inflation rates. The other main determinant of inflation is the unemployment rate with which, by means of a Phillips curve mechanism, inflation varies inversely. According to this theory, the only way to eliminate inflation through conventional monetary and fiscal restraint is by moving along the short-run Phillips curve and suffering a period of high unemployment that is long enough to break the slowly moving inflationary expectations. In this model the momentum in the inflation process and the high cost in unemployment of ending inflation is caused by the irrational nature of agents’ expectations. Reductions in inflation are costly because it takes agents a long time to understand that they are in a less inflationary environment. If they learned faster, reducing inflation would be less costly. A second, more sophisticated mechanism that can lead to a notion of intrinsic momentum in inflation is the staggered wage contract model of John Taylor (1979) and Phelps and Taylor (1977). Taylor posits rational expectations, so that agents in his model form expectations of inflation as functions of all of the variables relevant for forecasting future inflation. As a result of positing rationality, the particular function that agents optimally use to forecast inflation responds systematically and predictably to the economic environment, including the monetary policy and fiscal policy regimes, contrary to the fixed-function forecasts assumed under adaptive expectations. The source of momentum or persistence in Taylor’s model comes from the overlapping structure of multiperiod wage contracts, and a particular nonstate contingent form that he imposes on contracts. In this class of models, in terms of unemployment it is costly to end inflation because firms and workers are now locked into long-term wage contracts that were negotiated on the basis of wage and price expectations that prevailed in the past. In Taylor’s model, as in all rational expectations models, the observed momentum or serial correlation in the inflation process partly reflects the serial correlation in the “first causes” of inflation, such as monetary and fiscal variables. In addition, however, the wage-contracting mechanism contributes some momentum of its own to the process, so that the resulting slug-

Introduction

113

gishness in inflation cannot be completely eliminated or overcome by appropriate changes in monetary and fiscal policies. The wagecontracting process gives rise to a non-trivial trade-off between the variance of inflation and the variance of unemployment. Although they both embody a measure of momentum in wage and price dynamics, the adaptive expectations and Taylor wage contract models have substantially different implications about the unemployment costs of deflationary policies. The adaptive expectations or “core inflation” theory implies that inflation can be reduced only through a more or less extended period of higher unemployment. On the other hand, the wage-contracting models of Taylor and others imply the existence of a variety of alternative policy strategies, each of which could successfully deflate the economy with no costs in terms of higher unemployment. The classes of strategies that eliminate inflation without imposing unemployment costs are all characterized by a policy of gradual tightening of monetary and fiscal policy. How gradual to make this process depends precisely on the dynamics of the wage contracts. To avoid increases in unemployment, the deflationary actions must respect the persistence in nominal wages built in by the old wage contracts. Thus an inflation can be eliminated costlessly, but only if it is done gradually. For such a policy to work, it is necessary that the strategy of gradual tightening of aggregate demand be precommitted by the demand management authorities in a way that is sufficiently forceful and binding that it assures private agents’ belief in the plan. Later in this chapter, I shall argue that there is a tension inherent in this kind of gradual policy, since promises to take strong contractionary actions in the more distant future in the face of only mild contractionary actions in the present and near future are open to skepticism. (Recall the campaign promises of Presidents Carter and Reagan to balance the federal budget in the last year of their respective administrations.) At this point it is sufficient to emphasize that adaptive expectations and Taylor-like contracting models with rational expectations have different practical implications about the feasibility of relatively painless disinflations. The reason for these important differences in implications resides in the difference between the assumptions of adaptive and rational expectations. The latter hypothesis has private agents’ decision

114

Chapter 4: Stopping Moderate Inflations

rules responding in a particular stabilizing way with respect to changes in the government’s strategy for demand management, while the former does not. We now turn to the second group of theories of inflation, which are the rational expectations, and equilibrium theories. These theories maintain that essentially all of the characteristics of the serial correlation of inflation are inherited from the random properties of the deeper causes of inflation, such as monetary and fiscal policy variables. These theories differ from Taylor’s kind of theory in viewing wage and price contracts, whether implicit or explicit, as more state contingent, and contracting procedures as more responsive to the economic environment.1 In order to explain observed Phillips curve trade-offs, these theories resort to the Phelps-Lucas device of information limitations and the temporary confusions that they cause. When measures of aggregate demand and/or variables that partially reflect them, such as prices and interest rates, are realized to differ from what they had previously been rationally expected to be, it sets in motion movements in real economic variables. On this view the first cause of business cycle fluctuations is uncertainty about the position of future and maybe even current relative prices and productivity disturbances. Although differing among themselves in many important substantive details, members of the second group of theories are united by their assertion that under the proper hypothetical conditions a government could eliminate inflation very rapidly and with virtually no Phillips curve costs in terms of foregone real output or increased unemployment. The “measure” that would accomplish this would be a once-and-for-all, widely understood, and widely agreed upon change in the monetary or fiscal policy regime. Here a regime is taken to be a function or rule for repeatedly selecting the economic policy variable or variables in question as a function of the state of the economy. Particular models within this class differ widely with respect to the particular policy variables (e.g., high-powered money, a wider monetary aggregate, or total gov1 Robert

Barro (1977) has pointed out that after a change in policy regime, it can happen that it is in the interests of neither party to enforce some long-term contracts of the Taylor-Fischer variety that had been agreed upon before the regime change. Presumably such contracts would never be enforced.

Introduction

115

ernment debt) that are focused upon. However, all the theories require that the change in the rule for the pertinent variable be widely understood and uncontroversial, and therefore unlikely to be reversed. These characteristics are essential in eliminating the costs in terms of foregone output that information limitations and confusions cause with the Phelps-Lucas version of the Phillips curve. According to each of these two theories, Mrs. Thatcher has faced a formidable task. The momentum view obviously implies that she could use monetary and fiscal variables to depress inflation only at the cost of also depressing real economic activity. The rational expectations equilibrium view suggests that it is not in the power of a prime minister or even a united political party to create the circumstances required to bring about a quick and costless end to inflation. Whether or not the stage is set for successfully implementing a significant new policy regime is the result of intellectual and historical forces that individual political figures influence only marginally. Mrs. Thatcher came to power against the background of over 20 years of stop-go or reversible government policy actions.2 Her economic policy actions are vigorously opposed both by members of the Labor Party and by a strong new party, the Social Democrats. Thus the economic spokesman for the Labor Party, Mr. Peter Shore, advocated an immediate 40 percent devaluation and a larger government deficit. Mrs. Thatcher’s party now runs third in the political opinion polls. In addition, throughout her administration, speculation has waxed and waned about whether Mrs. Thatcher herself would be driven to implement a U-turn in macroeconomic policy actions, and whether her stringent monetary policy actions would be reversed by the Con2 Leland

Yeager (1976, 472) summarizes British postwar macroeconomic policy as follows: “The rapid reversibility of British policy . . . has been almost comical at times. Balance-of-payments troubles have brought a variety of ad hoc responses, including two devaluations and one abandonment of exchange rate pegging, the selective Employment Tax of 1966, the import surcharge of 1964, the import deposit scheme of 1968, the tightening and loosening of various exchange controls on current and capital transactions, and various attempts at wage and price control, as well as turnarounds in domestic financial policy. Reliance on such expedients creates chances of improper timing, of anticipatory private actions, of overshooting the mark, and of intensified instability as a result.”

116

Chapter 4: Stopping Moderate Inflations

servative Party itself, by choosing a new party leader. Furthermore there is widespread dissent from Thatcher’s actions among British macroeconomic scholars, so that she cannot be regarded as implementing a widely agreed upon theory. For all of these reasons, it is difficult to interpret Thatcher’s policy actions in terms of the kind of once-and-for-all, widely believed, uncontroversial, and irreversible regime change that rational expectations equilibrium theories assert can cure inflation at little or no cost in terms of real output.3 This is not to render a negative judgment on Thatcher’s goal or her methods, but only to indicate that the preconditions for the applicability of rational expectations “neutrality” or “policy irrelevance” theorems don’t seem to exist in Margaret Thatcher’s England. Where these conditions are not met, rational expectations equilibrium models imply that contractionary monetary and fiscal policy actions are likely to be costly in terms of real output and unemployment.

The “Poincaré Miracle” We have seen that extensive preconditions must obtain before rational expectations, equilibrium theories can be taken to imply that there is a costless cure to inflation, or equivalently, that the neutrality theorems of the theory can be expected closely to approximate reality. It has been argued by some that these preconditions are so stringent that they have rarely if ever been satisfied in practice, so that the example of Thatcher’s England is the standard case. While this is a respectable argument, it is useful to point out that there are repeated historical episodes that seem to fit the rational expectations equilibrium model fairly well. I have described four such episodes, namely, the events surrounding the ends of hyperinflations in Poland, Germany, Austria, and Hungary from 1922 to 1924 (Sargent 1982). Each of those countries successfully stopped drastic inflations dead in their tracks by interrelated fiscal and monetary policy changes that can be interpreted as abrupt changes in regime. The costs in foregone output were much smaller than would be 3 It

goes without saying that the “credibility” that is essential under the rational expectations theory cannot be manipulated by promises or government announcements.

The “Poincaré Miracle”

117

suggested by modern estimates of Phillips curves and were in no sense proportional to the magnitudes of the inflations that were halted. Some readers’ responses to those examples have been that because those inflations were so spectacular, between 5,000 and 1,000,000 percent per year, the procedures undertaken to end them have few implications about the problem of ending more moderate inflations like the ones faced by industrialized countries today. The argument seems to rest on an appeal to a model in the style of Taylor (1979). It asserts that the hyperinflations had proceeded to the point where long-term nominal contracts had ceased to operate, thereby destroying the dominant source of momentum in the inflation process. The argument is then that for milder inflations, the existence of long-term nominal contracts still remains a source of momentum that will make it costly in terms of real output and unemployment to end inflation quickly by draconian changes in fiscal or monetary regime.4 However, in the 1920s other countries successfully used essentially the same monetary and fiscal reforms that worked in Austria, Poland, Germany, and Hungary to stabilize much milder currency depreciation.5 One dramatic example was the stabilization of 4 Rudiger

Dornbusch made this argument in oral comments on Chapter 3 of this book. 5 The reader is referred to the accounts of post–World War I stabilizations in Brown (1940) and Young (1925). For example, the Italian stabilization might as easily have served as our example as the French one. Brown (1940, 431) quotes Count Volpi’s account of the important aspects of the plan that the Italian government used to stabilize the lira: 1. Balancing of the national budget. 2. Consolidation of war debts. 3. Unification of the note issue and its concentration in the hands of the Bank of Italy. 4. Progressive and more efficient utilization of Italian resources and raw materials. 5. Gradual deflation in currency and in credit. 6. Consolidation of the floating debt and reorganization in the Treasury Department. 7. Regulation of the influx of foreign capital into Italian industry. 8. Reorganization in the whole field of production, and readjustment of taxes with a view to increased industrial efficiency. 9. Gradual amortization of the domestic debt. 10. Defense of the treasury surplus by the reduction of state expenditures.

118

Chapter 4: Stopping Moderate Inflations

the French franc that was achieved by the government formed by Raymond Poincaré in July 1926.6 (Tables 4.1 and 4.2 report the French wholesale price index and the dollar-franc exchange rate between 1923 and 1927). Poincaré formed his government at a time when it was universally recognized that “the country was in trouble again, and all political parties except the Socialists and Communists gathered behind Poincaré. Five former premiers joined his government. There was a political truce” (Shirer 1969, 163). For some time there had been broad consensus both about the principal economic factors that had caused the depreciation of the franc—persistent government deficits and the consequent pressure to monetize government debt—and the general features required to stabilize the franc—increased taxes and reduced government expenditures sufficient to balance the budget, together with firm limits on the amount of government debt monetized by the Bank of France. For several years a political struggle had been waged over whose taxes would be raised, with the monied interests in the country resisting efforts to raise taxes on them.7 The accession to power of Poincaré in 1926 settled that issue in a fashion acceptable to the country’s monied interests. France financed its effort in World War I by borrowing at home and abroad, mainly in the United States. After the war France continued to run substantial government deficits. That it did so was partly rationalized by the expectation that “Germany will pay” for the French deficits. Under the Treaty of Versailles, Germany was obligated to pay massive reparations, which the French used partly to finance the reconstruction of territories devastated during the war. However, neither the total amount to be paid, nor the payment schedule was fixed by the treaty. Instead, these were to be determined by the Reparations Commission, and in the event were subject to continuous revision and renegotiation. The uncertain character of these claims complicated the public finances of both Germany and the countries that were owed reparations by GerCount Volpi was the “architect of Italy’s return to gold.” accounts of the “Poincaré miracle” appear in Shirer (1969), Yeager (1976), Alpert (1951), Haig (1929), and Rogers (1929). 7 William Shirer (1969) describes this struggle. 6 Interesting

The “Poincaré Miracle”

119

many.8 With the collapse of the German mark during 1923 and the relief from reparations provided Germany under the Dawes plan in 1924, it became clear that France could not continue to expect that German reparations would be sufficient to redeem the French government’s debt. From that time on the franc depreciated, and the domestic price level rose, as Table 4.2 shows.9 The big financial question for French governments was how much of its outstanding debt would be paid off or honored by channeling increased tax revenues to bondholders, and how much would be defaulted on through depreciation of the franc. The period from 1924 through July 1926 was marked by political instability and a rapid succession of governments and finance ministers in “the waltz of the portfolios.” There were repeated and unsuccessful attempts to deal with the increasing difficulties associated with refinancing the massive government debt as it gradually became due. The controversy was tainted by scandal as it was revealed that the government under Herriot had cooperated in an accounting subterfuge that concealed the fact that the Bank of France had exceeded the legal limit on the amount of its note issue that could be used to purchase government bonds. The period was also characterized by a massive flight of French capital abroad, partly an anxiety reaction to some of the tax proposals under discussion, such as a capital levy, and partly a reaction to the deteriorating prospects for the returns of franc-denominated assets. Raymond Poincaré was a fiscal conservative, who had raised taxes while prime minister in 1924 and was known to advocate a balanced budget and France’s return to gold. In 1926 he served as his own finance minister. As soon as he assumed control of the government, and even before his program was enacted by the legislature, the franc recovered, and inflation stopped. Under Poincaré, taxes were raised with an eye toward assuring persistent balanced or surplus government budgets. Some direct tax rates 8 Another

element of uncertainty was injected by the substantial war debts owed by the United States, coupled with the French belief that the United States should not insist that these be repaid. 9 For accounts of the effects of war debts and reparations on the public finances and currencies in Europe after World War I, see Yeager (1976) and Alpert (1951).

120

Chapter 4: Stopping Moderate Inflations

were actually reduced, including the highest rate for the general income tax, from 60 to 30 percent, and the rates of inheritance and estate taxes. However, indirect taxes were raised markedly. The government was authorized to raise all specific taxes up to six times their prewar rates, and decrees were issued implementing this authority (Haig 1929, 163). Customs duties were raised, and postal rates increased, as were taxes on passenger and freight rail service and on autos. The basic income tax rates were also raised, for example, from 12 to 18 percent on income from land and securities and from 7.2 to 12 percent on labor income. A once-andfor-all tax of 7 percent on the first sale of real estate or a business, a kind of capital levy, was also imposed. There was also established an independent special fund to pay off outstanding government debt, administered by the Caisse d’Amortissement, a newly created agency independent of the treasury and with its own earmarked revenues from the tobacco monopoly, the total receipts from the inheritance and estate taxes, and the new 7 percent tax on first sale of real estate and businesses. As the figures in Tables 4.1 and 4.2 show, these measures resulted in a sudden recovery of the franc and a cessation of inflation. The franc was permitted by the French authorities to appreciate from July until December, at which time France de facto returned to the gold standard. The appreciation of the franc was accompanied by open-market purchases of foreign assets by the French monetary authority as French citizens repatriated capital in response to Poincaré’s policies. While Poincaré himself had wished to restore the franc to its prewar par, it was decided to halt the appreciation of the franc in December 1926 and de facto to return to gold at that rate. This amounted to an 80 percent depreciation of the franc from its prewar par. This magnitude indicates the substantial extent to which France had financed the war by issuing bonds to its citizens, on which it largely eventually defaulted. This is to be contrasted with the situation in England, which returned to the prewar par in 1925, thereby indicating an intention not to default on its long-term debt. However, the French did not default as thoroughly as did the Austrians or the Germans.10 10 The

strength and endurance of French politicians’ resolve not to repeat such a

The “Poincaré Miracle”

121

The stabilization of the French franc was followed by several years of high prosperity. The French stabilization thus seems to fit the predictions of the rational expectations equilibrium approach. To the extent that it does fit, one reason is probably the high degree of political and intellectual consensus that prevailed at the time. It should be remembered that the French stabilization occurred after a variety of neighboring countries had successfully stabilized by resorting to the same budgetary principles that France eventually applied. At the time there was widespread professional consensus about the general budgetary situation that would have to prevail in order for the franc to be stabilized in the absence of exchange controls. (Shirer 1969, 166) indicates the degree of political consensus when he reports that “Frenchmen became obsessed with the idea that the ‘Poincaré franc,’ shrunk though it was, must never again be devalued lest they be ruined anew.” The French stabilized the franc by de facto returning to the gold standard. This amounted to standing ready to convert the debt of the French government into gold on specified terms, such as on demand for currency. In order to make a domestic currency freely convertible into gold, or into any foreign money for that matter, it is necessary that the government run a fiscal policy capable of supporting its promise to convert its debt. What backs the promise is not only the valuable stocks of gold, physical assets, and private claims that the government holds but also the intention to set future taxes high enough relative to government expenditures. This method of stabilizing a currency remains available to a small country today, even though the world is no longer on the gold standard. One country, call it the domestic country, can obtain a domestic rate of inflation no greater than, and even less than, that of a large foreign country to whose currency it pegs its own currency.11 To support this policy requires that the domestic country default was indicated by the fact that France was the last of the major countries to devalue its currency in terms of gold in the 1930s: France devalued in 1936, while England did so in 1931 and the United States in 1933. 11 Stanley Fischer (1981) provides a more complete discussion of this issue and the other issues described in this paragraph. Bryant and Wallace (1980) discuss optimal seigniorage from the viewpoint of price discrimination. They describe setups in which a government can find it worthwhile to issue an array of debt

122

Chapter 4: Stopping Moderate Inflations

abstain from or at least much restrict the extent to which it resorts to inflationary finance. Indeed, the domestic government collects seigniorage only to the extent that it engages in clever devices such as holding its reserves of the foreign currency denominated assets in the form of interest-bearing assets, while at the same time adopting legal restrictions and fostering institutions that prompt its own residents to hold currency and other zero or low nominal interest assets. It is possible for a domestic government actually to experience a lower rate of inflation than the country to which it pegs its currency if it sets things up so that government and private institutions back their monetary liabilities with interest-bearing foreign-denominated assets and also pass the interest returns to their depositors. In so doing, the government completely abstains from using inflationary finance and provides domestic residents with a higher real rate of return on “money”—a lower rate of inflation than is experienced by residents of the foreign country who happen to be holding currency and other zero nominal rate of interest assets.12 As we turn out attention to Mrs. Thatcher’s actions, it is useful to keep in mind a number of characteristics of the French financial crisis of 1926 and the subsequent salvation of the franc by Poincaré: 1. The extent to which the large interest-bearing French government debt created during the war and the reconstruction period became more and more difficult to refinance, thereby generating increasing pressures for its eventual monetization. This pressure eventually led to fraudulent accounting practices by the Bank of France and a scandal that brought down a government. The forces underlying these events are pertinent in Britain and elsewhere today in estimating the with differing yields, tailored to segments of the market with differing interest elasticities of demand for government debt. Applying their idea to the issue in the present discussion, setups can be imagined where the domestic country arranges to hold high-yielding foreign government debt and where it is in the interests of both the foreign and the domestic country to permit the domestic country to back its monetary liabilities by the higher-yielding foreign government debt rather than the lower-yielding debt. 12 John Bilson (1980) describes a scheme of this sort that can lead to a positive real return on government-issued or privately issued “currency” through a process of deflation.

The British Experience

123

likely consequences and even the very feasibility of policies that propose to combat inflation with restrictive monetary policies alone, while at the same time permitting substantial government deficits to continue. 2. The manner in which France stabilized by pegging the franc to a foreign currency and adopting changes in tax and expenditure laws that delivered the prospective budget surpluses needed to support that peg without exchange controls. A similar course was available to Britain in 1979, but it did not choose to follow it. 3. The sense in which the preconditions for a successful and relatively costless stabilization along the rational expectations equilibrium model were met in France in 1926. Whether these preconditions are met is in large part a consequence of historical circumstance. However, it is also perhaps partly a function of the particular lines along which a stabilization is sought. For example, it is arguable that pegging to a foreign currency is a policy that is relatively easier to support and make credible by concrete actions, since it is possible to hook the domestic country’s price expectations virtually instantaneously on to the presumably exogenous price expectations process in the foreign country. 4. The fact that France chose to stabilize at a value that was widely believed to undervalue the franc. To this the French prosperity of the late 1920s has often been partly ascribed (e.g., see Alpert 1951 or Yeager 1976). This is to be contrasted with the situation in England today, where contemporary monetary and fiscal policies have permitted a substantial appreciation of the pound, with consequent depressing influences on export industries.

The British Experience Tables 4.3 through 4.12 report statistics that summarize the recent behavior of United Kingdom aggregates. Since Mrs. Thatcher took office in June 1979, much of the news has been bad. Real GNP has declined; industrial production, especially in manufacturing,

124

Chapter 4: Stopping Moderate Inflations

has fallen precipitously; the unemployment rate has climbed from about 5 percent in June 1979 to over 10 percent in March 1981, to attain its highest level since the 1930s. Meanwhile, inflation in the retail price index accelerated for the first year of Thatcher’s administration, though in the last nine months it has receded markedly so that the inflation rate in the United Kingdom during this more recent period was actually less than it was in the United States. The pound sterling rose vis-à-vis the US dollar, from 2.11 $/£ in June 1979 to 2.40 $/£ in January 1981, while the balance of payments in current account swung toward surplus. Interest rates rose to very high levels. Recent economic events in Britain have been well summarized in papers by Meltzer (1981) and in the Morgan Guaranty Trust Company (1981) Survey. I refer the reader to those sources for many interesting details and will devote most of my space to highlighting and interpreting a few of the facts from the viewpoint of rational expectations macroeconomics.

Mrs. Thatcher’s Plan A hallmark of Mrs. Thatcher’s publicly announced economic strategy is gradualism. For the most part her government did not propose to execute any abrupt or discontinuous changes in aggregate government variables such as tax collections, government expenditures, or the money supply. Instead, the Conservatives proposed to carry out a preannounced and gradual tightening of monetary and fiscal policy over a five-year period. These intended goals were embodied in the “medium-term financial strategy” (MTFS) the new government announced in 1979. The plan included the following elements: 1. A gradual reduction in the rate of growth of the money supply over a five-year period. The monetary aggregate that was chosen as the monetary instrument variable was “sterling M3” or “£M3,” which corresponds to currency plus sterlingdenominated demand and time deposits of United Kingdom commercial banks. The initial plan called for £M3 to grow annually at a 9 percent rate in 1980–81, with its rate of growth gradually to decline to 6 percent by 1983–84.

Mrs. Thatcher’s Plan

125

2. A reduction in the real value of government spending within four years to a level 5 percent less than the level in 1979–80. 3. A public sector borrowing requirement (PSBR) of £8.5 billion in 1979–80, and £7 billion in 1980–81 (both in 1978–79 prices). Even these borrowing requirements, reduced though they were from those projected under the previous Labor government’s policies, represent deficits that as a ratio to GNP are several times those experienced in the United States. Other elements of the government’s plan were executed immediately. These included reductions in marginal income tax rates ranging from 33 to 30 percent for the lowest brackets to between 83 to 60 percent for the highest brackets, and substantial increases in the taxes on consumption, most notably a substantial increase in the value-added tax (VAT).This change in tax structure was made with an eye toward increasing the rate of saving. In addition, in October 1979, exchange controls were removed, so that for the first time since World War II, residents of Britain were permitted freely to invest abroad. The government committed itself to flexible exchange rates with neither current nor capital account exchange controls, nor substantial government open-market operations in foreign assets designed to peg or influence the exchange rate. In line with the theme of deregulation, the government permitted the Special Depository Regulations, widely known as the “corset,” to expire. These regulations had directly limited the extent to which the banks could increase their interest-bearing deposits. The corset represented an attempt to influence directly the £M3 aggregate and was widely and correctly believe to distort the interpretation of the £M3 figures, as depositors moved into close substitutes for £M3 in response to the restrictions. In its conception, and even more so in its execution, the plan incorporates central aspects of monetarism. The key monetarist plank is embodied in the use of gradual reductions in a measure of the money supply, sterling M3, as the central vehicle for reducing the rate of inflation. These reductions in the money supply are recommended despite very large planned budget deficits, planned deficits that in the actual event have been overrun. A keystone of monetarist doctrine is that even in the face of persistent government deficits, by managing the money supply properly, the

126

Chapter 4: Stopping Moderate Inflations

government can avoid inflation (Friedman 1970, 1972). Referring to England’s experience, Allan Meltzer (1981, 24) put the case as follows: “Excess public spending, larger than expected budget deficits and the growth of money in excess of targets are related problems. The relation would disappear if the central bank changed its operating procedures and permitted market rates to fluctuate as much as is required to control money. The excess deficit would than be financed by domestic saving or by foreigners, but money growth and inflation would fall.” There are various possible interpretations of this argument, not all equally credible. In one rational expectations interpretation, by restricting itself now and forever to a binding “k-percent growth rule”13 for the monetary base, the government effectively limits the extent to which it will collect seigniorage by resorting to inflationary finance now and in the future. Under rational expectations, current government budget deficits—expenditures net of both explicit taxes and seigniorage—must be balanced by prospective government surpluses in the future. That is, additional government bonds will be valued according to the same principles that give bonds of private corporations value: their real prospective returns. Ultimately, these prospective returns are represented by the government’s willingness to tax highly enough in the future. On this view a k-percent rule for the monetary base plays a similar role as a gold standard rule, in the sense that it places a limit on the time path of real government deficits. Both the k-percent rule and the gold standard rule in effect require that if the government is to sell its debt, the expected present value of the current and prospective government surpluses must be positive. Each rule permits the government to run deficits, even a number of deficits in succession, but these deficits must be accompanied by prospects that eventually the government budget will turn to surplus in sufficient amount to outweigh the deficits. This interpretation of the k-percent rule is one that is compatible with the Barro result about the equivalence of bond and tax financing (Barro 1974). In this interpretation a k-percent rule is not compatible with an everlasting government deficit but only with a deficit that is temporary in the 13 Presumably,

a rule in which k is a small number.

Mrs. Thatcher’s Plan

127

appropriate sense. In my view the preceding interpretation of the relationship of a k-percent rule to the budget deficit is the correct one. As with most rational expectations lines of thought, that interpretation emphasizes the dynamic or intertemporal features of the process, and the constraints that a k-percent rule requires on the future time path of the government deficit. There is an alternative, and I believe defective, view that seems to assert that a k-percent rule is compatible with a more or less permanent deficit. This view is based on reasoning from standard Keynesian or monetarist models without rational expectations. Versions of those models exist in which the government can control inflation by sticking to a k-percent rule for the monetary base given an unrestricted path for the deficit.14 The preceding argument raises questions about the credibility of an announced plan to lower the monetary growth rate and to move to as k-percent rule, while simultaneously projecting substantial government budget deficits for the several years in the immediate future. The doubtful credibility of such a plan stems from the fact that a large permanent real government deficit is simply incompatible with a k-percent rule for the monetary base. A minimal requirement that a plan be credible is that it be feasible in the first place. As we have argued, a restrictive k-percent rule for the base and a permanent and large government deficit just are not feasible. On this view, in order that the current British plan be viewed as credible, it is necessary that the large prospective government deficits over the next several years be counterbalanced by prospective surpluses farther down the line. It is difficult to point to much either in current legislation or, equally important, in the general British political climate that could objectively support such an outlook. On this view, the large government deficits that have accompanied the government’s medium-term financial strategy raise serious questions about whether the plan has the logical 14 In

the literature it has been pointed out that such a k-percent rule implies an explosive path for the government interest-bearing debt. As Bennett McCallum (1978) has pointed out, depending on the precise specification of the model, that fact may or may not imply that other variables in the model that are of interest are unstable.

128

Chapter 4: Stopping Moderate Inflations

coherence that is necessary for the plan to be credible to the public. Samuel Brittan has drawn attention to a closely related issue. At the same time that the government has touted its determination to bring inflation permanently down through monetary restraint, the substantial government deficits have been financed by issuing large amounts of nonindexed long-term government debt at nominal yields to maturity ranging between 13 and 14 percent. Attention is directed to Tables 4.10 and 4.11. Table 4.10 indicates the substantial extent to which the government has been financing its deficit by selling additional long-term government debt. Thus in financial year 1979–80 most of the additional government debt was over 15 years in maturity. Now if the government were actually to deliver on its hope permanently to reduce the inflation rate, it would imply substantial increases in the real value of the long-term government debt and the real value of the interest payments on the debt. For example, investors who purchased debt at nominal rates of 14 percent while expecting average inflation of 12 percent and a real return of 2 percent would experience ex post real yields higher than 2 percent, precisely to the extent to which realized inflation falls short of the 12 percent inflation rate that they had expected.15 For the same reason, but in the other direction, governments in the past have sometimes given way to the temptation to default on part of their interest-bearing government debt by causing inflation to occur at a higher rate than was anticipated at the time that the debt was sold.16 This same incentive confronts the government now and raises suspicions about the current and future governments’ commitment permanently to lower the inflation rate. According to this argument, a government intent on eradicating inflation has a strong incentive to finance its deficit and refinance its outstanding debt by issuing indexed government bonds. This would isolate it from any increase in the real value of the burden of the debt once inflation is lowered. This the British government has not done to any significant extent. For advocates of “Irving Fisher’s effect,” Table 4.11 contains an important piece of evidence about the public credibility of Mrs. 15 The

recent issue of indexed bonds in England sold at a real rate of interest of about 2 percent. 16 This issue was central to the struggle for the post–World War I stabilizations.

The Outcome of the Plan So Far

129

Thatcher’s plan for reducing inflation over the longer run. The term structure of interest rates on government bonds is high and fairly flat, suggesting that the market expects the continuation of high inflation rates on a sustained basis. As emphasized earlier, in the rational expectations view, these matters of coherence and credibility are very important in determining the likely effects of a program on real variables such as output and employment. If a program is constructed in a fashion that makes private agents believe that its execution is uncertain, then, even if preannounced, restrictive monetary policy actions can easily produce substantial reductions in output and employment.

The Outcome of the Plan So Far Having described the government’s anti-inflation plan and some possible reservations about it, I shall now proceed briefly to describe how events have actually unfolded. First, sterling M3 has exceeded its target range, despite the fact of a restrictive minimum lending rate (MLR) and a basically tight open-market stance. For example, the fiscal year 1980–81 target range for £M3 of 7 to 11 percent per year is to be compared with the annual rate of increase in £M3 of 21 percent between February 1980 and February 1981. During the same time period, sterling M1—currency plus demand deposits—increased by only 8 percent. Despite the overshooting of £M3, British interest rates have been very high, making many commentators of Keynesian inclination believe that monetary policy is very tight. Second, the public sector borrowing requirement has overrun its target. The 1980–81 PSBR had been forecast in the government’s 1980 budget at £8.5 billion, or 4.5 percent of GDP, while it is now expected to be around £13.5 billion, or 6 percent of GDP. I shall comment in turn on the overshooting of each of these targets.

130

Chapter 4: Stopping Moderate Inflations

Overshooting the £M3 Target There have been several reactions to the overshooting of the £M3 target.17 One has been to argue that since the overshooting reflects mainly a response to removal of the distorting effect of the corset, it does not indicate a failure to pursue a tight monetary policy. As evidence in support of this position, the relatively slow growth of £M1 is often cited. Another response has been to criticize the Bank of England’s operating procedures for focusing too heavily on interest rates as an intermediate instrument. Allan Meltzer (1981) takes this line in arguing that by pegging interest rates, the British monetary authority necessarily gave up direct control over monetary aggregates and allowed them to be market determined. The analytics of using monetary aggregates as opposed to interest rates as the monetary instrument have been characterized in 17 From

the technical viewpoint of controlling monetary aggregates, the banking and financial intermediary systems in the United Kingdom differ in important respects from those in the United States. First, in the United Kingdom banking is more concentrated, there being five main “Clearing Banks.” Second, in the British assets eligible to meet the 12.5 percent reserve requirement include all of the following interest-bearing assets: money at call from discount houses, treasury bills and other short-term government securities, local authority paper, corporate tax anticipation certificates, and bills of exchange. Notice that some of these assets are evidences of government indebtedness, while others are private debts. Since demand deposits do not bear interest in the United Kingdom, vis-à-vis the US system, this system of reserve requirements tends to increase the banking system’s share of seigniorage revenues relative to that of the government. On the other side of this issue, currency is a higher proportion of £M1 in the United Kingdom than in the United States. Third, the building societies (the analogue of savings and loan institutions in the United States) have long issued mortgages with variable maturities and variable rates of interest both linked to the general level of market interest rates. Therefore in the United Kingdom high interest rates do not produce the disintermediation from saving institutions that is so troublesome for the conduct of monetary policy in the United States. Fourth, partly as a result of the third feature, there is no analogue of Regulation Q in the United Kingdom, and small savers have access to a variety of instruments yielding close to market rates, as for example, building society shares. This fact also explains the absence of money market funds in the United Kingdom. Fifth, the Bank of England does not lend directly to the clearing banks but instead operates a discount window for the discount houses that make markets and hold portfolios of short-term government and private securities. The minimum lending rate, formerly known as the bank rate, applies to the Bank of England’s loans to the discount houses.

Overshooting the £M3 Target

131

Martin Bailey’s (1971) book and in papers by William Poole (1970) and John Kareken (1970). The case for superiority of a particular monetary aggregate over a particular interest rate depends on the demand schedule for that monetary aggregate being less uncertain than is the aggregate demand schedule expressed as a function of that particular interest rate. Other things equal, factors that contribute to uncertainty about the demand for a given monetary aggregate diminish the relative merit of using that aggregate as the monetary instrument. This analytical argument is quite pertinent in evaluating the controversy about the overshooting of the £M3 target (and also about the appropriate monetary instrument for the United States in 1981). The removal of the corset and the dismantling of exchange controls at the outset of Mrs. Thatcher’s administration presumably shifted the demands for a whole host of assets in historically unprecedented and uncertain ways. Regardless of the possible merits of the case for relaxing these controls, it seems clear that for some time after they are relaxed the interpretation of a variety of monetary aggregates becomes more uncertain and difficult than it had been. During such periods, the case for using an interest rate rather than a monetary aggregate as the monetary instrument becomes substantially strengthened. It is ironic that both in the United Kingdom and in the United States the accession of monetarists to a dominant influence over policy has coincided with substantial revisions in the structure of financial regulations that at least temporarily cloud the meaning of the particular monetary aggregates they favor controlling.18 It seems to me that it is a defensible view that, despite their own problems of interpretation, the high nominal interest rates in Britain over the last year have more appropriately signaled the stance of monetary policy than any particular monetary aggregate.19

18 I

have in mind the Monetary Control Act.

19 As in the United States, in the United Kingdom there is a bewildering variety of

monetary aggregates. The main ones are £M1, £M3, PSL 1 (Public Sector Liquidity number 1), and PSL 2. The variety of aggregates is spawned by the vagueness of “means of payments” as a category setting off one class of assets as “money.” See Sargent and Wallace (1982).

132

Chapter 4: Stopping Moderate Inflations

The Government Deficit I now turn to discuss the behavior of the public sector borrowing requirement, which so far has exceeded the government’s target by so much that the government has moved to correct the situation by raising taxes. The 1980–81 PSBR which has been forecast by the government to be £8.5 billion or 4.5 percent of GDP, appears to be coming in at £13.5 billion of 6 percent of GDP. In the March 1981 budget the chancellor of the exchequer, Sir Geoffrey Howe, announced a number of tax increases designed to reduce the prospective PSBR for 1981–82 to about £10.5 billion. Without those additional tax measures the government estimated that the 1981– 82 PSBR would have been about £14.5 billion. The new revenueraising measures included increases in the excise taxes on drink, tobacco, gasoline, diesel road fuel, cigarette lighters, matches, and road vehicles. The extent to which income tax payments were indexed against inflation was reduced. A Supplementary Petroleum Duty on North Seal oil and gas was announced, which together with adjustments in the Petroleum Revenue Tax was expected to yield about £1 billion. The chancellor also announced a once-andfor-all tax on low-interest bank deposits that was expected to yield £4 billion in 1981–82.20 The government announced these tax increases because it became increasingly aware of the threat that a persistent and large government deficit sooner or later poses to an anti-inflation policy based on monetary restraint. Before considering the nature of the British deficit in more detail, it helps to remember a few analytical principles about government finance. In interpreting reported figures on the government’s budget deficit, it is useful to keep in mind the hypothetical distinction between current account and capital account budgets and their deficits. A pure current account expenditure is for a service or perfectly perishable good that gives rise to no government-owned asset that will produce things of value in the future. A pure capital account expenditure is a purchase of a durable asset that gives the government command of a prospective future stream of returns, 20 Recall

the remarks in note 17 about the way in which seigniorage is allocated between the banks and the government under the British system of reserve requirements.

The Government Deficit

133

collected, for example, through user charges, whose present value is greater than or equal to the present cost of acquiring the asset. A pure capital account budget would count as revenues the interest and other user charges collected on government-owned assets, while expenditures would be the purchases of capital assets. By these definitions government debt issued on capital account is selfliquidating and fully backed by the user charges that are earmarked to pay it off. Government debt issued to finance a pure capital account deficit is thus not a claim on the general tax revenues that the government collects through sales and income taxation. The principles of classical economic theory condone government deficits on capital account. The idea is that certain government capital projects are worthwhile on cost-benefit grounds and that it is reasonable to finance them by levying taxes on the people who receive the benefits throughout the time the benefits accrue. In short, so far as capital account deficits are concerned, a government is in a sense like a firm, it being wise to borrow to finance worthwhile long-lived projects with taxes and other user charges whose stream over time matches the time profile of the benefits. A deficit on current account is very different because it is not self-liquidating. The classical economic doctrine was, first, that the current account budget should always be balanced and, second, on those extraordinary occasions such as wars when it could not be balanced, that a current account budget deficit should be financed by long-term debt and a plan to run current account surpluses in the future sufficiently large to retire the debt. Thus a current account deficit, should it be unavoidable, was to be financed by earmarking some future general tax revenues for the purpose of retiring the debt. It is no coincidence that these classical doctrines about government finance were developed at a time when England and other leading economic powers were on the gold standard, each government promising to convert its currency and other government debt into gold on certain specified conditions. To make good on that promise, a government had to back its debt with sufficiently large and sufficiently probable prospective government surpluses denominated in gold. Deficits on capital account did not threaten a government’s adherence to a gold standard, while deficits on

134

Chapter 4: Stopping Moderate Inflations

current account did. The force of a gold standard was to cause the government to back its debt and to refrain from raising revenues from seigniorage. Under contemporary monetary institutions in which currency is inconvertible or fiat, governments have access to seigniorage as an additional means of raising government revenues. (Whether the additional freedom this gives government is helpful is very controversial both among theorists and practical people.) When a government finances its long-term debt without indexing repayment to the price level, the freedom to expand government demand debt and longer-term debt without the limits imposed by adherence to the gold standard gives the government a wide range of options about if, when, and to what extent to default on its long-term debt by monetizing it and depreciating its real value.21 Under a fiat money regime the extent to which a current account deficit is inflationary depends on the extent to which private agents believe that the government will ultimately finance it by monetization. For example, the Ricardo-Barro doctrine about the irrelevance of the current taxation-debt-issuing choice assumes that the government refrains from monetizing the debt and in effect binds itself to a classical financial policy. Under that policy current real government deficits are not inflationary because they are accompanied by expectations of future government surpluses. The additional real government debt is backed by prospective real tax revenues. However, as Bryant and Wallace (1980) have emphasized, in a policy regime where the current deficit is eventually monetized in some proportion, a current account deficit is inflationary. In some models it is more inflationary; the larger the proportion of it that is eventually monetized and the sooner the monetization occurs. In those models the precise dynamics by which the prospect of future monetization of the debt influences inflation rates depends on the detailed specification of the demand functions for assets, in particular, on how responsive they are to the expected rate of return on currency. It can readily happen, for example, as under a demand function for money like Phillip Ca21 This

was the choice that French politicians consciously faced and struggled with from 1919 to 1926 and that politicians also face today, although perhaps less consciously.

The Government Deficit

135

gan’s (1956), that current rates of inflation respond positively to the prospect of future increases in money brought about be eventual monetization of government debt; see Sargent and Wallace (1982). Although it seems not to have been something that Keynes himself would have advocated, the widespread adoption of Keynesian ideas about fiscal policy after World War II has been accompanied by abandonment of the classical public budgeting and accounting procedures at the level of national governments (although not at the level of state and local governments in the United States). For example, in the United Kingdom the nationalized industries do not float their own debt. Instead, they borrow from the National Loan Board, which in turn borrows by issuing government debt. This arrangement is one that departs from or at least obscures the earmarking of revenues from particular projects to back a given bond issue. Moreover the recent history of public finances in the United Kingdom displays little sensitivity to the distinction between capital and current account so, for example, as Table 4.6 and Figure 4.1 show, capital formation by the general government and nationalized industries has stagnated or actually fallen. Further, as Table 4.9 shows, while general government final consumption and current grants and subsidies have risen substantially in recent years, and have continued to rise under Mrs. Thatcher, capital expenditures have not. Under both Mrs. Thatcher’s government and the previous Labor government, belt tightening has fallen largely on public sector investment items. According to the classic canons of public finance that we alluded to earlier, this structure of expenditure cuts is perverse from the viewpoint of anti-inflationary policy. The failure of Mrs. Thatcher’s government to control public expenditures has been widely commented upon and will receive only brief mention here. Mrs. Thatcher has been criticized for a number of what are essentially tactical errors, for example, in her stance toward pay for public employees. Her early decision to stand by the Conservative Party’s campaign pledge to honor the recommendations of the Clegg Commission, which the Labor government instituted to establish pay standards for civil servants comparable to those in the private sector, resulted in earnings increases for civil servants of 24.5 percent between 1979III and

136

Chapter 4: Stopping Moderate Inflations

4000

3000 Private sector

2000

General government

1000 Public corporations

0 1975

1976

1977

1978

1979

1980

Figure 4.1 Gross domestic fixed capital formation (millions of pounds, 1975 prices seasonally adjusted).

1980III. Thatcher responded, albeit belatedly, to that criticism by eventually abolishing the Clegg Commission in the fall of 1980. Since that time, the government has announced the adoption of a cash limit system, which essentially creates a total wages fund with which the government intends to confront a given public sector union or collection of unions. The idea is to force the unions to take into account a trade-off between wage rates and the number of public sector employees. A problem is that such cash limits have been breached in the past and have already been breached by Thatcher in the coal miners’ settlement.

North Sea Oil and the Pound Sterling In the late 1970s the magnitude of Britain’s prospective revenues from North Sea oil became clearer and coincided with a simultaneous appreciation of the pound sterling from 1.65 US $/£ in 1975IV to 2.39$/£ in 1980IV, and a swing of the current account bal-

North Sea Oil and the Pound Sterling

137

ance of payments from a deficit toward surplus. The fact that Great Britain moved from being a net importer to being a net exporter of petroleum helped swing the current account into surplus. Some observers have attributed the strength of the pound to Britain’s claim on North Sea oil and its contribution in swinging the current account toward surplus. However, few macroeconomic models imply that there is any direct connection between possessing oil and having a strong currency. There is an indirect connection, namely, that North Sea oil is heavily taxed and thus contributes to prospective government revenues, thereby tending to diminish the government deficit. Revenues from the taxation of North Sea oil are substantial and are expected to grow over time. The 1981 budget forecasts revenues from all taxes on oil (in 1979–80 prices) of £3.25 billion in 1980–81, £4.50 billion in 1981–82, £4.75 billion in 1982–83, and £5.25 billion in 1983–84. These revenues are a substantial fraction of the current government deficit of £13.5 billion. However, most macroeconomic theories assert that only the total deficit and how it is financed influence inflation and the international value of the pound. So long as total prospective government deficits remain as large as they are, it is difficult to subscribe to the view that the United Kingdom’s possession of oil strengthens the pound. As a scrap of empirical evidence supporting this judgment, it has been observed that Norway is in a similar situation to England vis-à-vis North Sea oil and that neither has its currency appreciated internationally nor has it experienced the severe depression of industry that England has. One explanation for the difference appears to be that Norway has embarked on a looser monetary policy than has England. Another popular explanation for the strong pound is that OPEC countries have begun to diversify their overseas investments by holding assets denominated in a variety of foreign currencies and that this has resulted in an increased preference for pound-denominated assets. This factor is probably part of the explanation but is not the dominant force leading to a strong pound. Probably the most plausible explanation for the emergence of a strong pound builds on the “overshooting” idea of Dornbusch and has the advantage that is simultaneously explains other aspects of the current situation including high British interest rates and

138

Chapter 4: Stopping Moderate Inflations

depressed British output and employment.22 Dornbusch modeled a small country under the following assumptions. 1. The domestic price level has some sort of stickiness. Either domestic prices are exogenous, as in one of Dornbusch’s original formulations (Dornbusch 1976), or there is a Taylor-like long-term contracting mechanism, as in a later contribution by Dornbusch (1980, ch. 9), or there are information discrepancies à la Lucas and Phelps that prevent domestic prices from adjusting instantaneously to certain classes of random events, as in the setup of Nasser Saidi (1980). Which of these devices is resorted to makes an important difference, as we shall see. 2. There is assumed to be perfect international capital mobility in high-yielding assets. This implies that the interest parity condition must hold. The interest parity condition states that the domestic interest rate must equal the foreign rate of interest minus the expected rate of appreciation of the domestic currency. Thus, let r jt be the domestic nominal interest rate on j -period bonds, r ∗jt be the foreign nominal interest rate on j -period bonds, et be the exchange rate or foreign price of domestic money, measured in foreign currency per unit of domestic currency, and Et ln et+ j be the logarithm of the exchange rate expected as of time t to prevail at time t + j , the interest parity condition is 1 r jt = r ∗jt − ( Et ln et+ j − ln et ) j The interest parity condition ensures that foreigners can attain the same nominal rate of return, in terms of their own currency, by investing in the domestic country as by investing elsewhere. 3. The market for domestic currency and other “money” is assumed to be isolated internationally in the sense that the real rate of return on domestic money is permitted to be strictly dominated by other assets, including domestic and foreign bonds and equities and maybe also foreign currencies. What is crucial for the results is that there be some restriction on the scope of international currency substitution, most formulations ruling out any currency 22 See

Dornbusch (1976, 1980). Buiter and Miller (1981) argue that Dornbusch’s idea explains contemporary observations in the United Kingdom.

North Sea Oil and the Pound Sterling

139

substitution at all. Notice the asymmetry between the assumption of integrated world bond and equities markets but nationalistic markets for currencies. The demand for domestic real balances is assumed to vary inversely with the domestic nominal interest rate, and directly with domestic real output, in the standard way. 4. The demand for domestic real output depends, among other things, inversely on the domestic real rate of interest and on the “real exchange rate” or terms of trade. Thus, letting p be the domestic price level measured in pounds per British good, p∗ the foreign price level measured in dollars per US good, and e the exchange rate measured in dollars per pound, the “real exchange rate” is defined as ep/p∗ . Thus, given p and p∗ , an increase in e decreases the demand for British output, since it raises the relative price of British goods in terms of US goods. 5. While the domestic price level is to some extent sticky, the exchange rate and domestic interest rate are perfectly flexible instantaneously. 6. The foreign price level and interest rate are exogenous to events in the domestic country, the operational meaning of the small-country assumption. 7. Expectations are rational. Given these assumptions, consider a situation where the British monetary authority undertakes a restrictive monetary action. Because of price level stickiness, the initial effect is to drive the domestic interest rate upward. But the upward tendency in the domestic interest rate threatens to disturb the interest parity condition and to create a capital inflow. To maintain interest parity in the face of less than perfectly flexible prices, the entire expected exchange rate path must adjust to generate an expected path of subsequent depreciation of the pound sufficient to offset the higher British interest rate. For this to happen, the exchange rate e must first jump upward to a higher level than before the restrictive monetary action, from which higher level it gradually falls in order to generate the rational expectations of a depreciating pound needed to maintain interest parity. Thus the immediate effect of the restrictive monetary action is to cause the exchange rate initially to appreciate suddenly and subsequently to depreciate gradually.

140

Chapter 4: Stopping Moderate Inflations

However, since the domestic price level is somewhat sticky, the initial effect of the appreciated pound is to raise the real exchange rate ep/p∗ , and so to reduce the demand for British goods. This effect reinforces the effect on demand of the higher real domestic interest rate and leads to a recession along standard Keynesian lines of insufficient aggregate demand. This sequence of events depends on there being some source of price stickiness that prevents the domestic labor market from clearing. Had domestic prices and wages been assumed perfectly flexible, the response to a downward movement in the domestic money supply would have been very different than that described here. In particular, under flexible prices equilibrium is restored by a drop in the domestic price p proportional to the drop in the money supply, together with an offsetting increase in the exchange rate just sufficient to leave the real exchange rate ep/p∗ unaltered. No changes in the domestic interest rate or output are needed to restore equilibrium. A version of classic neutrality occurs in these models under flexible domestic prices. Versions of the model such as Saidi’s that rest on limited information and temporary confusion to provide price stickiness, or a Phillips curve, exhibit an interesting mixture of the responses under sticky prices and under perfectly flexible prices. In particular, with respect to monetary disturbances that were perfectly predictable given private agents’ information and understanding in the past, the system responds exactly as if prices were perfectly flexible: There are no real effects, the domestic price level and the exchange rate adjusting just enough to offset the disturbance while leaving domestic real output and employment unaltered. However, with respect to monetary disturbances that are not predictable, given agents’ information and understanding, the system responds qualitatively in the same fashion as described when prices are sticky. Each of these variants of the Dornbusch model works in explaining the broad features of recent British experience, including high nominal interest rates, a strong pound sterling, and depressed industry. However, the different versions of the model support different interpretations and perhaps also policy recommendations.23 23 One

popular policy recommendation stemming from the momentum version

North Sea Oil and the Pound Sterling

141

On the one hand, according to the models that rely on momentum or long-term contracts to generate domestic price inflexibility, the response to restrictive monetary actions will be qualitatively similar whether or not those actions were foreseen by private agents. Such versions of the model could explain events even on the interpretation that Mrs. Thatcher’s restrictive actions represent execution of a once-and-for-all regime change that is widely believed and irreversible. On the other hand, according to versions of the model like Saidi’s that rely solely on information limitations to induce a Phillips curve, the events must be interpreted as reflecting the perceived temporary and reversible nature of the restrictive monetary actions that the government has undertaken. Explanations along Dornbusch’s line seem to be the best ones available for simultaneously explaining the strong pound, depressed British industry, and persistent British inflation. However, the literature on currency substitution makes it clear that this argument is delicate in that it depends on a demand function for domestic currency that permits domestic currency to be dominated in rate of return by large and variable amounts by foreign currencies and other assets. As several researchers have emphasized, there are incentives for international currency substitution that threaten the temporal stability of the demand schedule for domestic currency and the durability of the preceding class of explanations.24 The literature on currency substitution points toward a problem that may loom on the horizon for British policy. That literature predicts that a country that runs much larger persistent deficits than its neighbors and that monetizes a large fraction of them will require the imposition of international currency controls if it is to support its currency internationally (Kareken and Wallace 1978b). The models analyzed in the currency substitution literature thus indicate that high and persistent government deficits are over the long haul incompatible with permanent abstention from of the model is to impose inward capital controls, for example, an interest equalization tax on the yields of British securities held by foreigners. Such a tax is presumed to weaken the pound and stimulate aggregate demand and real domestic output. See Buiter and Miller (1981). 24 Kareken and Wallace (1981) propound a model with an extreme amount of currency substitution.

142

Chapter 4: Stopping Moderate Inflations

exchange controls. While it might take some time for these forces to break through various frictions, they will acquire strength and create problems precisely to the extent that large budget deficits loom in the future. It is certainly arguable that only temporarily can a tight monetary policy delay the operation of these forces, à la Dornbusch, in the face of large, persistent government deficits.

Conclusion The theoretical doctrines and the historical evidence described in this chapter provide little reason for being optimistic about the efficacy of a plan for gradual monetary restraint that is simultaneously soft on the government deficit. Gradualism invites speculation about future reversals, or U-turns, in policy. Large contemporary government deficits unaccompanied by concrete prospects for future government surpluses promote realistic doubts about whether monetary restraint must be abandoned sooner or later to help finance the deficit. Such doubts not only call into question the likelihood that the plan can succeed in reducing inflation permanently but also can induce high real costs in terms of depressed industry and lengthened unemployment in response to what may be viewed as only temporary downward movements in nominal aggregate demand due to the monetary restraint. These considerations are pertinent in assessing the state of the United Kingdom’s economy today and the situation facing the French in the 1920s. They are also pertinent in evaluating the wisdom of passing Kemp-Roth in the United States while simultaneously planning to implement a tight k -percent monetary rule. If we are bent on reducing inflation, then by consulting both our theoretical imaginations and history, we can find methods that improve on gradualist money restraint in the face of large government deficits. That is why it behooves us to recall Poincaré and his contemporaries even as we think about Thatcher.25 25 Economists

have begun devoting more attention to devising ways of reducing the cost of winding down inflation. For example, Jeffrey Shaefer and Axel Leijonhufvud have recently described a kind of dynamic currency reform scheme

Second Thoughts

143

Second Thoughts I now believe that this chapter is misleading in the methods, motives, and understandings that it attributes to important public figures.26 I take this opportunity to correct some misimpressions that I was under. These corrections do not diminish the role that I assigned to the arithmetic of the government budget constraint which set the circumstances and opportunities that confronted these people.

that aims to eradicate the costs of eliminating inflation that are due to long-term contracts. To illustrate their scheme, suppose that up to date t , the monetary and fiscal policy regime and the other random processes that influence inflation have been such as to make it rational for private agents to expect that future prices will follow some given path p¯ (t + j), j ≥ 0 , where the expected price level p¯ (t + j) is measured as usual in units of green dollars at time t + j per good at time (t + j) . For example, if a constant rate of inflation of π is expected, then p¯ (t + j) = (1 + π ) j p(t) , where p(t) is the actual price level at t . If these price expectations are built into long-term contract that have been entered into t and earlier, and so form a legacy that influences actual prices and quantities at time t + j , then the act of bringing inflation to a sudden halt will cause substantial redistributions across traders. To the extent that actual prices turn out to be less than those expected at the time that the contracts were negotiated, real output and unemployment will be adversely affected. The idea of Shaefer and Leijonhufvud is to circumvent these costs by carrying out an imaginative kind of currency reform. The government passes a law at date t that states that all contracts that call for payment of y dollars at date t + j can be discharged by paying only yp(t)/ p¯ (t + j) dollars. Thus, in the constant expected inflation case, dollars due at t + j are paid off at only 1/(1 + π ) j on the dollar. More important, the government successfully commits itself to run a fiscal and monetary policy that implies a stable price level so that the actual price p(t + j) = p(t) for all j ≥ 1 . With a constant actual price path of p(t + j) = p(t) , and the new debt conversion law, both sides of all contracts end up being just as well off as if the debt conversion law had not been enacted and prices had risen as expected, p(t + j) = p¯ (t + j) . Thus the debt conversion law is crafted to neutralize the real effects of the monetary and fiscal policies needed to support a zero inflation price path. It is as if the government announces that it is calling in all the green-colored currency and issuing new blue-colored currency on the following terms: green dollars will be converted into blue dollars at par at time t , and subsequently the green dollar price of a blue dollar is p¯ (t + j)/p(t) . 26 These remarks were added in April 1992. I thank Joseph Bisignano and John Taylor for pointing out some oversights in my treatment of the Poincaré miracle.

144

Chapter 4: Stopping Moderate Inflations

P OINCARÉ AND M OREAU There is a chance that my treatment assigned to Henri Poincaré too heroic a role in the creation of the “Poincaré miracle” and that it failed adequately to sing the praises of lesser known but more genuine heroes of that episode.27 First and foremost, I neglected to mention the important role played by the Report prepared by the “Committee of Experts” (the “Sergent committee”), which had been appointed by the Briand government on May 31, 1926, and which issued its report on July 3, 1926, twenty days before Poincaré formed his government. The quality of that committee’s recommendations and the thoroughness and promptness with which democratic institutions implemented them offer testimony to the value of practical macroeconomics. The committee consisted of a group of men widely believed to include some of the best academic and practical financial experts in France. It recommended a coordinated system of fiscal, monetary, and foreign-exchange reforms and operating rules that included virtually of all the measures eventually implemented to produce the “Poincaré miracle.” Poincaré’s government promptly implemented those measures, but as we shall see, the romantic vision that I offered of Poincaré as a principled monetary and fiscal conservative who had been waiting in the wings only to be called upon to implement his beliefs, is probably wrong. At least it is wrong if we believe Émile Moreau. In his memoirs, Émile Moreau, who was governor of the Bank of France when Poincaré took office, asserts that Poincaré initially attempted to induce the Bank of France to monetize additional debt issued by the Treasury, and threatened to remove Moreau if he would not comply.28 Moreau paints Poincaré as a bully, one willing for reasons of political expediency temporarily to monetize even if doing so risked inflating away government debt.29 Moreau 27 On

more general issues associated with the “Poincaré miracle,” I recommend the paper by John Taylor (1982) and the first chapter of R. G. Hawtrey (1933). Both Hawtrey and Taylor draw attention to the contraction in output that followed the stabilization of 1926, and mention that the committee of experts had been worried about such an occurrence. 28 See Moreau (1991, 45–56). 29 Moreau also describes how two years later he opposed Poincaré’s wishes to

Methods of Thatcher and Reagan

145

tells how he stood firm in support of the measures recommended by the experts, and how after several stormy meetings, Poincaré finally came around. Moreau’s story is no less consistent with “monetarist arithmetic” than was my original story, but it makes Moreau not Poincaré the hero. In effect, Moreau claims to have bested Poincaré in a version of Wallace’s game of chicken played out during the first few days of Poincaré’s ministry.

Methods of Thatcher and Reagan I was also wrong about Margaret Thatcher. I wrote this chapter the very month the analogy between Thatcher’s policies and Reagan’s broke down. For the first two years of her administration, Thatcher combined what threatened to be sustained fiscal deficits with a promise of sustained monetary stringency. However, in March 1981 Mrs. Thatcher’s government implemented an increase in revenue collections so substantial and sustained that it moved Britain’s net-of-interest government deficit to balance or surplus for the rest of the decade. After March 1981, Mrs. Thatcher’s policies with respect to the coordination of monetary and fiscal policy are interesting for their contrast to those of Reagan, not their similarity. Interestingly enough, had the Reagan administration raised taxes by the same proportion of GNP as did Mrs. Thatcher, the US federal budget would not have displayed a net-of-interest deficit during the Reagan years. Evidently Mrs. Thatcher raised taxes in March 1981 against the objections of many of her economic and political advisors, and she did so in the midst of a recession. According to what Alan Walters has told me, Mrs. Thatcher acted against the advice of most of her political advisors because she understood that doing so was needed to render her preferred monetary policies sustainable. It is especially remarkable for an observer from the United States that Thatcher made her corrections in the midst of a deep recession. So perhaps Mrs. Thatcher’s monetary and fiscal policies should be criticized, but not for the reasons that I gave. In the end, her move in an opposite direction, when Poincaré wanted to set the value of the franc against gold to a level that Moreau thought too high.

146

Chapter 4: Stopping Moderate Inflations

policies form a contrast, not with those of Poincaré, but with those of Reagan.

Where to Play Chicken It is interesting to speculate why Thatcher quickly desisted from presiding over a government whose monetary and fiscal branches seemed to be playing “chicken,” while Reagan did not. I suspect that the answer resides in the fact that British political institutions are much more centralized than are those of the United States. In Britain, both the chancellor of the Exchequer and the governor of the Bank of England serve at the pleasure of the prime minister. Further, in heading both the executive and legislative branches of government, the prime minister is more immediately and authoritatively in control of tax and expenditure policy. If a British prime minister, especially a forceful one, were to preside over a game of chicken, it would be really a game of chicken with herself. Institutions in the United States are far more decentralized, and the president much less powerful in setting government policy. The Federal Reserve is much more independent of the president and the Treasury than the corresponding institutions in Britain, and the president has to negotiate with two houses of Congress, sometimes led by an opposition party, to legislate taxes and expenditures. Decentralized situations like this are the fields on which games of chicken might better be played.

147

Where to Play Chicken

Table 4.1 F RENCH W HOLESALE P RICE I NDEX Base 1913 = 100, 1913 from 1901–10 index = 115.6 Year

Jan.

Feb.

March

April

May

June

1923 1924 1925 1926 1927

386.9 494.0 514.4 633.5 621.8

421.8 543.7 515.0 635.6 631.6

424.0 499.3 513.5 631.8 641.4

414.7 450.0 512.8 650.1 636.5

406.5 458.5 519.8 687.9 628.6

408.7 465.3 542.6 738.4 622.6

July

Aug.

Sept.

Oct.

Nov.

Dec.

406.7 481.0 556.8 836.2 619.9

413.1 476.6 557.2 769.5 617.9

423.6 485.6 555.7 786.9 600.3

420.5 497.1 572.3 751.5 587.5

442.9 503.5 605.5 683.8

458.6 507.2 632.4 626.5

Source: League of Nations Bulletin, Haig (1929, p. 448).

Table 4.2 D OLLAR E XCHANGE IN PARIS

January February March April May June July August September October November December

1923

1924

1925

1926

1927

15.57 16.45 15.23 14.84 15.15 16.32 16.84 17.65 16.15 17.19 18.51 19.59

21.74 25.57 18.32 15.45 18.80 18.88 19.86 18.48 18.96 19.13 18.82 18.56

18.49 19.38 19.06 19.20 19.83 21.56 21.11 21.30 21.12 23.92 26.09 26.90

26.77 27.49 28.65 30.15 30.60 34.93 41.15 35.12 35.66 32.52 28.11 25.25

25.32 25.55 25.54 25.53 25.53 25.54 25.56 25.51 25.48 25.47 25.43 25.40

Note: Averages of daily rates for final weeks of each month, francs per dollar.

148

Chapter 4: Stopping Moderate Inflations

Table 4.3 GDP AT FACTOR C OST AND F INAL E XPENDITURES ON G OODS AND S ERVICES AT M ARKET P RICES in £ million, current prices

(1) GDP at (2) GDP at current prices 1975 prices based on based on expenditure expenditure data data

Implicit price deflator 1975 = 100 (column 1 divided by column 2)× 100

GDP at market prices

Consumers’ expenditure

1970 1971 1972 1973

43,532 49,442 55,276 64,258

85,402 87,572 88,719 95,506

50.97 56.46 62.30 67.28

51,065 57,291 63,390 72,936

31,778 35,599 40,183 47,759

1974 1975 1976 1977

74,414 93,954 111,245 126,111

94,527 93,954 97,971 98,993

78.72 100.00 113.55 127.39

82,879 104,413 124,330 143,064

52,849 64,424 74,751 85,474

1978 1979

144,442 163,647

101,929 102,563

141.71 159.56

164,034 189,702

98,395 114,805

149

Where to Play Chicken

Table 4.3 GDP AT FACTOR C OST AND F INAL E XPENDITURES ON G OODS AND S ERVICES AT M ARKET P RICES (continued) Value of physical Gross General increase in government domestic stocks and Exports of Imports of final fixed capital work in goods and goods and Taxes on consumption formation progress services services expenditure Subsidies 8, 991 10, 250 11, 675 13, 380 16, 609 23, 074 26, 779 26, 209 32, 934 38, 316

9, 470 10, 517 11, 606 14, 238 16, 867 20, 417 23, 599 25, 739 26, 695 33, 646

421 158 44 1, 448 1, 304 1, 534 864 1, 860 1, 070 2, 760

11, 551 12, 960 13, 653 17, 124 22, 985 27, 011 35, 211 43, 352 47, 442 54, 676

11, 146 12, 193 13, 771 19, 013 27, 375 28, 979 36, 874 42, 570 45, 502 54, 501

8, 417 8, 788 9, 627 10, 121 11, 469 14, 162 16, 553 20, 252 23, 253 30, 361

884 939 1, 153 1, 443 3, 004 3, 703 3, 468 3, 299 3, 661 4, 306

150

Chapter 4: Stopping Moderate Inflations

Table 4.4 E XCHANGE , P RICE , AND B ALANCE Retail price (1975 = 100) Current balance (million £) 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1976

1977

1978

1979

1980

1981

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 Jan. Feb.

+799 +1, 076 +189 -1, 056 -3, 380 -1, 674 -1, 060 -206 +707 -1, 630 -2, 737 +106 -352 -436 -378 -362 -431 +307 +280 -194 +417 +87 +397 -692 -192 -189 -557 +70 -88 +870 +1, 885 +1, 042 +614

Exchange rate ($/£)

Index

Percentage increase on one year earlier

2.396 2.444 2.502 2.453 2.340 2.220 1.805 1.746 1.920 2.122 2.328 1.998 1.805 1.766 1.651 1.714 1.719 1.735 1.813 1.928 1.835 1.932 1.984 2.016 2.081 2.234 2.157 2.254 2.286 2.382 2.387 2.405 2.294

54.2 59.3 63.6 69.4 80.5 100.0 116.5 135.0 146.2 165.8 195.6 110.9 114.9 117.6 123.0 129.2 134.9 137.0 139.0 141.4 145.3 147.8 150.3 155.0 160.7 171.4 176.2 184.6 195.3 199.4 203.2 205.7 207.6

6.4 9.4 7.1 9.2 16.1 24.2 16.5 15.8 8.3 13.4 18.0 22.5 16.0 13.7 15.0 16.5 17.4 16.5 13.0 9.5 7.6 7.9 8.1 9.6 10.6 16.0 17.3 19.1 21.5 16.4 15.3 13.0 12.5

151

Where to Play Chicken

Table 4.5 O UTPUT AND U NEMPLOYMENT GDP (at 1975 factor cost, £ million seasonally adjusted) 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1976 1 2 3 4 1977 1 2 3 4 1978 1 2 3 4 1979 1 2 3 4 1980 1 2 3 4

85,402 87,572 88,719 95,506 94,527 93,954 97,971 98,993 101,929 102,973 24,486 24,156 24,518 24,811 24,397 24,660 24,677 25,259 25,156 25,602 25,507 25,664 25,175 26,287 25,655 25,856 25,596 25,445 24,991

Industrial production (1975 = 100, seasonally adjusted) All industries Manufacturing 99.7 99.8 102.0 109.5 105.1 100.0 102.0 105.9 109.8 112.6 100.4 101.8 101.4 104.4 106.2 105.5 105.9 106.1 107.6 110.6 111.0 110.0 110.1 115.0 112.7 112.6 110.0 106.6 102.9 100.2

98.0 97.5 100.0 108.4 106.6 100.0 101.5 103.0 103.9 104.2 99.4 101.7 101.6 103.2 104.5 102.5 102.7 102.4 102.9 104.5 104.9 103.1 102.3 107.3 103.2 104.2 100.0 96.8 93.3 89.1

Unemployment rate 2.4 3.1 3.4 2.4 2.3 3.6 4.9 5.2 5.2 5.0 5.1 5.3 5.4 n.a. 5.5 5.6 5.8 5.9 5.8 5.8 5.7 5.4 5.6 5.4 5.2 5.3 5.7 6.2 7.0 8.4

152

Chapter 4: Stopping Moderate Inflations

Table 4.6 G ROSS D OMESTIC F IXED C APITAL F ORMATION BY S ECTOR

1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1975

1976

1977

1978

1979

1980

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

Total

Private sector

18,954 19,460 19,743 19,823 21,195 20,616 20,417 20,636 20,089 20,802 20,506 5,112 5,086 5,178 5,041 5,226 5,158 5,203 5,049 4,883 5,065 4,997 5,144 5,287 5,282 5,136 5,097 4,998 5,052 5,182 5,274 5,169 5,058 4,923

10,390 10,685 11,099 11,776 12,267 11,641 11,530 11,811 12,438 13,793 13,761 2,916 2,846 2,986 2,782 2,844 2,920 3,097 2,950 2,892 3,149 3,119 3,278 3,493 3,499 3,401 3,400 3,318 3,401 3,436 3,606 3,547 3,429 3,357

General government 5, 385 5, 475 5, 297 5, 076 5, 793 5, 418 4, 974 4, 786 3, 964 3, 520 3, 352 1, 239 1, 306 1, 165 1, 264 1, 280 1, 226 1, 156 1, 124 1, 079 988 953 944 938 899 868 825 818 820 882 832 765 741 719

Public corporations 3, 201 3, 316 3, 334 2, 932 3, 135 3, 557 3, 913 4, 039 3, 687 3, 489 3, 393 957 934 1, 027 995 1, 102 1, 012 950 975 912 928 925 922 856 894 867 872 862 831 864 836 857 888 847

Where to Play Chicken

Table 4.7 M ONEY S UPPLY IN THE U NITED K INGDOM in £ million: amounts outstanding Seasonally adjusted At end period 1970

1971

1972

1973

1974

1975

1976

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

M1

Sterling M3

M3

8,640 8,920 9,020 9,420 9,820 9,900 10,210 10,310 11,200 11,680 11,750 12,240 12,280 13,130 12,660 13,040 12,870 13,370 13,510 14,330 14,880 16,080 16,770 17,070 17,940 18,530 19,100 18,980

16,000 16,460 16,830 17,300 18,020 18,270 18,670 19,530 21,140 22,480 23,320 24,720 26,290 27,650 29,620 31,450 32,730 32,810 33,490 34,610 35,560 35,840 37,030 36,920 37,960 33,790 40,300 40,380

16,450 16,980 17,350 17,810 18,510 18,780 19,180 19,960 21,670 23,090 23,970 25,520 27,390 28,720 30,940 32,880 34,520 34,940 35,940 37,100 38,120 38,100 39,780 40,010 41,160 42,210 44,310 44,470

153

154

Chapter 4: Stopping Moderate Inflations

Table 4.7 M ONEY S UPPLY IN THE U NITED K INGDOM in £ million: amounts outstanding (continued) Seasonally adjusted At end period 1977

1978

1979

1980

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

M1

Sterling M3

M3

19,540 20,530 22,020 23,180 24,350 25,090 26,010 27,020 27,580 28,250 28,950 29,460 29,370 30,110 29,780 30,520

40,720 41,740 42,990 44,540 46,880 48,230 49,560 51,310 52,370 54,380 56,210 57,830 59,250 62,570 65,340 68,350

45,070 46,220 47,390 48,950 51,480 53,260 54,480 56,350 57,150 59,290 61,040 63,270 65,110 68,140 71,200 74,870

155

Where to Play Chicken

Table 4.8 G ENERAL G OVERNMENT R ECEIPTS AND E XPENDITURE in £ million Receipts

1975 1976 1977 1978 1979 1977

1978

1979

1980

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

Expenditure

Taxes national insurance, etc., contributions

Trading income rent, interest, etc.

Final consumption

38,547 44,724 51,008 56,704 68,053 12,700 12,371 12,822 13,115 14,116 13,539 13,897 15,152 16,032 16,387 17,250 18,384 20,845 18,713 21,811

4,439 5,223 5,909 6,488 7,353 1,527 1,371 1,697 1,314 1,720 1,512 1,702 1,554 1,917 1,722 1,962 1,752 2,045 2,136 2,157

23,074 26,779 29,209 32,934 38,316 7,040 7,257 7,368 7,544 7,972 8,073 8,252 8,637 8,875 9,316 9,896 10,229 10,872 11,656 12,386

Gross domestic capital formation 5, 064 5, 483 4, 935 4, 741 5, 239 1, 565 1, 007 1, 171 1, 192 1, 483 962 1, 145 1, 151 1, 500 1, 023 1, 356 1, 360 1, 700 1, 144 1, 415

Note: An article describing the new presentation of government income and expenditure was published in the March 1977 issue of Economic Trends.

156

Chapter 4: Stopping Moderate Inflations

Table 4.8 G ENERAL G OVERNMENT R ECEIPTS AND E XPENDITURE (continued) Expenditure Current and capital transfers Current grants and subsidies 14,353 17,015 19,502 23,239 27,348 4,742 4,833 4,847 5,080 5,658 5,725 5,783 6,073 6,522 6,855 6,617 7,354 7,747 8,145 8,075 a

Capital transfers 1, 196 1, 435 1, 537 2, 027 1, 901 470 363 328 376 693 413 442 479 504 454 443 500 576 521 618

Debt interest 4,211 5,394 6,373 7,224 8,829 1,832 1,314 1,747 1,480 1,995 1,455 1,951 1,823 2,330 1,877 2,523 2,099 3,149 2,209 3,299

Net lending, etc.a

Total

3, 755 2, 365 251 1, 687 3, 273 433 -178 88 -92 74 439 729 445 554 670 1, 054 995 152 1, 444 1, 185

51,653 58,471 61,807 71,852 84,906 16,082 14,596 15,549 15,580 17,875 17,067 18,302 18,608 20,285 20,195 21,889 22,537 24,196 25,119 26,978

Net lending to public corporations, private sector and overseas; cash expenditure on company securities, etc. (net).

157

Where to Play Chicken

Table 4.9 F INANCIAL T RANSACTIONS OF THE P UBLIC S ECTOR in £ million Financial deficit

Total 1971 1972 1973 1974 1975 1976 1977 1978 1979 1979

1980

1 2 3 4 1 2

300 1, 547 2, 764 4, 695 7, 705 8, 413 5, 868 8, 048 8, 344 2, 329 1, 823 2, 636 1, 556 1, 700 3, 543

General government -786 804 1, 997 3, 165 4, 912 6, 159 4, 639 6, 973 6, 227 1, 782 1, 416 1, 623 1, 406 1, 226 2, 646

Public corporations 1, 086 743 767 1, 530 2, 793 2, 254 1, 229 1, 075 2, 117 547 407 1, 013 150 474 897

Net lending, etc., to private sector and overseas 620 558 880 1, 697 1, 833 1, 286 126 467 432 167 224 104 -63 -128 490

Total 920 2, 105 3, 644 6, 392 9, 538 9, 699 5, 994 8, 515 8, 776 2, 496 2, 047 2, 740 1, 493 1, 572 4, 033

158

Chapter 4: Stopping Moderate Inflations

Table 4.9 F INANCIAL T RANSACTIONS OF THE P UBLIC S ECTOR (continued) Receipts Public sector borrowing requirement Contributions Financial by central transactions (net receipts) Total government -483 55 -547 -41 -946 572 -1 184 -3, 788 1, 031 -1, 298 -1, 085 -2, 436 2, 771 -802

1, 403 2, 050 4, 191 6, 433 10, 484 9, 127 5, 995 8, 331 12, 564 1, 465 3, 345 3, 825 3, 929 -1, 199 4, 835

637 1, 600 2, 331 3, 523 8, 345 6, 786 4, 469 8, 371 10, 396 247 3, 797 2, 842 3, 510 -1, 950 4, 587

ContriContriSeasonally butions butions adjusted by public by local total authorities corporations 676 514 1, 348 2, 161 1, 629 1, 103 183 659 1, 732 1, 003 -267 666 330 1, 397 574

90 -64 512 749 510 1, 238 1, 343 -699 436 215 -185 317 89 -646 -326

1, 403 2, 050 4, 191 6, 433 10, 484 9, 127 5, 995 8, 331 12, 564 2, 117 3, 006 3, 893 3, 548 131 4, 122

159

Where to Play Chicken

Table 4.10 N ET P URCHASES ( + ) OR S ALES ( − ) OF G OVERNMENT D EBT, BY M ATURITY Classification by maturity Redemptions Over 1 Over 5 Total and Up to and up to and up to stocks conversions 1 year 5 years 15 years

Over 15 years and dated

Financial years 1975/76 +4159 1976/77 +6290 1977/78 +6684 1978/79 +6256 1979/80 +8977

-735 -703 -672 -404 -1133

-1120 -1402 -2259 +1098 -2068

+2196 +2600 +2931 +1994 -2333

+1008 +817 +2826 +1441 +2905

+2810 +4978 +3858 +4323 +6940

Quarter ended 1978 Sept. +793 Dec. +1288

-151 -16

-364 -57

+257 +2

+154 +802

+897 +557

1979 Mar. June Sept. Dec.

+2254 +2732 +2648 +2511

-234 -1 -403 -431

-324 -314 -932 -178

+824 +358 +1062 +159

+486 +1159 +496 +1317

+1502 +1530 +2425 +1644

1980 Mar. June Sept. Dec.

+1086 +3377 +3186 +3055

-298 -544 -19 -263

-644 -574 -136 -734

+754 +1358 -261 +1186

-67 +943 +3130 +1425

+1341 +2194 +472 +1441

160

Chapter 4: Stopping Moderate Inflations

Table 4.11 C ALCULATED R EDEMPTION Y IELDS OF G OVERNMENT B ONDS Percent per annum Short-dated (5 years)

Medium-dated (10 years)

Long-dated (20 years)

Last working days 1980 Oct. Nov. Dec.

13.15 12.97 13.30

13.29 13.43 13.89

13.15 13.35 13.80

1981 Jan. Feb.

13.21 13.00

13.86 13.84

13.86 13.94

Mondays 1980 Oct. 20 Oct. 27

13.07 12.93

13.33 13.07

13.23 12.95

Nov. Nov. Nov. Nov.

3 10 17 24

13.29 13.38 13.11 13.03

13.40 13.53 13.26 13.22

13.24 13.35 13.10 13.08

Dec. Dec. Dec. Dec. Dec.

1 8 15 22 29

12.95 13.04 13.62 13.18 13.28

13.34 13.44 14.08 13.79 13.88

13.26 13.34 13.94 13.70 13.80

1981 Jan. 5 Jan. 12 Jan. 19 Jan. 26

13.19 13.38 13.30 13.29

13.89 14.06 14.05 13.95

13.85 14.02 14.06 13.95

Feb. Feb. Feb. Feb.

2 9 16 23

13.19 13.29 13.24 12.97

13.83 13.89 13.92 13.83

13.83 13.90 13.94 13.93

Mar. 2 Mar. 9 Mar. 16

13.03 13.15 12.85

13.87 13.95 13.59

13.97 14.03 13.63

161

Where to Play Chicken

Table 4.12 I NTEREST R ATES , S ECURITY P RICES , AND Y IELDS percentage rate

Last Friday

1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

Last working day

Bank of England’s minimum lending rate to the market

Treasury bill yield

Eurodollar 3-month rate

Building Societies Association recommended rate on shares

8 7 5 9 13 11.50 11.25 14.25 7 12.50 17 14

7.80 6.93 4.46 8.48 12.82 11.30 10.93 13.98 6.39 11.91 16.49 13.45

10.07 6.57 5.75 5.91 10.19 10.07 5.88 5.07 7.19 11.69 14.50 17.75

5.00 5.00 5.00 5.25 7.50 7.50 7.00 7.80 6.00 8.00 10.50 10.50

Average of working days

British government securities: long-dated (20 years) 9.05 9.25 8.90 8.97 10.78 14.77 14.39 14.43 12.73 12.47 12.99

5 Some Unpleasant Monetarist Arithmetic with Neil Wallace

Introduction This chapter is a variation on the theme that monetary and fiscal policies are interrelated and must necessarily be coordinated. They must be coordinated because the monetary authority controls the rate of seigniorage from money creation, which is a revenue source in the government’s budget. The issue of coordination arises when one seeks to answer the following question: Is it possible for monetary policy permanently to influence an economy’s inflation rate? The answer to this question hinges on how monetary and fiscal policies are imagined to be coordinated. On the one hand, one can imagine a monetary authority sufficiently powerful vis-à-vis the fiscal authority that by the imposition of slower rates of growth of base money, both now and into the indefinite future, it can successfully constrain fiscal policy by telling the fiscal authority how much seigniorage it can expect now and in the future. In this setting, monetary and fiscal policies are coordinated by having the monetary authority discipline the fiscal authority. Under this coordination scheme, the decisions of the monetary authority about the stream of seigniorage it will supply to the fiscal authority can permanently influence the inflation rate. On the other hand, one can imagine that the monetary authority is not in a position to influence the government’s deficit path but is limited simply to managing the debt that is implied by the deficit path chosen by the fiscal authorities. Under this second coordination scheme, the monetary authority is much less powerful than under the first scheme. Under the second coordination scheme, it is the monetary authority that is constrained by the fiscal authority by having ultimately to deliver a seigniorage stream sufficient to finance the deficit stream that is chosen by the fiscal authority. 162

Introduction

163

This chapter characterizes some of the dimensions of the limits of the monetary authority and discloses some of the choices open to it under the second coordination scheme. We analyze that scheme partly because we suspect that over the last decade the second coordination scheme has more nearly been the one in place in the United States and other major industrial countries; but even apart from that empirical judgment, we believe that analyzing some of the implications of the second scheme is a useful vehicle for directing attention to the issue of coordinating monetary and fiscal policies. Our attention was attracted to this issue by the high pretax real rates of return on government securities that prevailed in the United States during the first half of 1981. The prevalence of before-tax rates of return on government securities that exceed the economy’s rate of growth confronts the monetary authority with a difficult choice about inflation. With the government budget persistently in deficit and real rates of interest exceeding the economy’s growth rate, the Fed must choose between fighting present inflation with “tight” monetary policy now or fighting future inflation with “easy” monetary policy now. Put differently, without help from the fiscal authorities, fighting current inflation with tight monetary policy must eventually lead to higher future inflation.1 That a real rate of return on government securities in excess of the economy’s growth rate n puts the monetary authority in this bind is a consequence of a simple accounting identity. If the real rate of interest on government securities exceeds, n , then simply to “roll over” a fixed amount of interest-bearing government debt in private hands requires raising revenue from other sources to pay the interest on the debt. In particular, the government must either levy taxes or reduce purchases or print currency to pay the interest. Another seeming alternative is to issue additional interest-bearing debt to pay the interest on the existing debt. However, this last 1 The messages of this chapter are very similar to those of Preston Miller (1981) and

Lucas and Sargent (1981a). The papers of Bryant and Wallace (1980), McCallum (1978, 1981) and Scarth (1980) are also related to this chapter. McCallum and Scarth analyze some of the implications of an explosive difference equation in bond holdings like ours, while Bryant and Wallace discuss debt management as an optimal seigniorage problem.

164

Chapter 5: Some Unpleasant Monetarist Arithmetic

alternative implies a growing stock of real government interestbearing debt per capita. While such growth can occur for a while, it surely cannot go on forever. A limit to such growth would certainly be encountered when real interest-bearing government debt per capita equaled total savings per capita, and probably far before then. Thus, eventually either taxes must be levied or currency must be issued to service the debt. Moreover, the larger the value at which the real interest-bearing debt per capita is ultimately stabilized, the larger the amount of revenue that eventually has to be raised either via tighter fiscal policy or via greater monetary expansion. In the United States today, the responsibility for determining fiscal policy (the level of federal government expenditures and explicit taxes) rests with Congress and the executive branch. Subject to congressional oversight, the responsibility for determining the division of total government debt between interest-bearing debt and non-interest-bearing debt (base money) rests with the Federal Reserve Board. Given the growth in the total debt implied by fiscal policy, the Federal Reserve uses open-market exchanges of interest-bearing government bonds for base money to determine how fast base money and interest-bearing debt grow. If the federal budget is in deficit, then for example, the Fed can make the base money component grow more slowly than the growth rate of total debt implied by the deficit only by making the growth rate of the interest-bearing component exceed the growth of the total government debt. However, once the limit on the federal debt per capita that can be marketed with the public has been reached, the Fed has no choice: It must increase base money. That is, it must “monetize” all of the additional government borrowing by purchasing all real additions to the stock of interest-bearing debt that the treasury issues. More generally, given the time path of fiscal policy and given that government interest-bearing debt can be sold only at a real interest rate exceeding the growth rate n , the tighter current monetary policy is, the higher the inflation rate must be eventually. This chapter describes the choices open to the monetary authority within a model that is extremely monetarist. The model goes as far as anyone would go in assigning monetary policy influence over the time path of the price level. It is also monetarist in imput-

Introduction

165

ing to monetary policy influence over almost no real variables. In his 1968 presidential address to the American Economic Association, Milton Friedman (1968) warned not to expect too much from monetary policy. In particular, Friedman argued that monetary policy could not permanently influence the levels of real output, unemployment, and real rates of return on securities. However, Friedman did assert that the monetary authority could exert substantial control over the inflation rate, especially in the long run. The purpose of this chapter is to argue that if monetary policy is interpreted as open-market operations of the Federal Reserve under our second coordination scheme, then Friedman’s list of the limits of monetary policy needs to be expanded to include severe limits on the ability of monetary policy to affect inflation permanently. The results in the monetarist model occur because, given our specifications about fiscal policy and the real rate of return on government securities, tighter money now necessarily implies looser money later. This explains why in the monetarist model tighter money today attains a lower inflation rate today at the cost of enduring a higher inflation rate in the future. Next we produce an example of an economy in which there are even more severe limits on the ability of tight money to fight inflation. By amending the monetarist model to include a more sophisticated and dynamic description of the demand for base money, we describe an example in which tighter money today leads to a higher inflation rate and price level not only eventually but starting today. This example is spectacular because it describes a situation in which tighter money today lacks even the temporary ability to fight inflation. While the example is an extreme one, and probably overstates the limits on the achievements of tight money in the real world, it also has the virtue of isolating a force limiting the ability of tight money to fight inflation that is omitted in the monetarist model and that is probably present in the real world.

166

Chapter 5: Some Unpleasant Monetarist Arithmetic

A Simple Monetarist Model We describe a simple model that embodies unadulterated monetarism. The model incorporates the following features: (1) a common constant growth rate of n for real income and population, (2) a constant real return on government securities that exceeds, n , and (3) a quantity theory demand schedule for base or highpowered money, on that exhibited constant income velocity.2 A model with these features exhibits the two limitations on monetary policy stressed by Milton Friedman in his AEA presidential address. First, there is a natural, or equilibrium, rate of real income that monetary policy is powerless to affect. Second, the real rate of interest on government bonds is beyond the influence of monetary policy. Our purpose in focusing on this very simple model is to describe yet one more serious limitation on the power of monetary policy, a limitation on its ability to influence the economy’s inflation rate, over which Friedman and monetarists have asserted that monetary policy exerts considerable influence. In particular, we shall show how in this most monetarist of models, monetary policy by itself is only capable of postponing inflation. We have purposefully chosen to argue this proposition in the context of a model that embraces as unqualified a set of monetarist assumptions as we can imagine. We do this to show that the argument about the limitations of monetary policy is not based on abandoning any of the key assumptions that have been made by those monetarists who have stressed the potency of monetary policy for controlling inflation. Instead, the argument hinges entirely on taking into account the future budgetary consequences of alternative current monetary policies when the real rate of return on government bonds exceeds n. We describe fiscal policy by a time path or sequence D (1), D (2), . . . , D (t), . . . , where D (t) is measured in real terms (time t goods) and is defined as real expenditures minus real tax collections net 2 In

Appendix 1 we analyze a simple general equilibrium model that can be used to rationalize the demand relationships posited below in the text. The model of the appendix has the virtue that since individual agents are identified, policies can be compared in terms of the welfare of the individuals in the model.

A Simple Monetarist Model

167

of all transfers except interest on the government debt and where, for convenience, we label the current date t = 1. We describe monetary policy by a time path H (1), H (2), . . . , H (t), . . . , where H (t) is the stock of base or high-powered money at time t . If, for simplicity, we assume that the entire government debt consists of one-period debt, then we can write the consolidated government budget constraint (consolidating the treasury and the Federal Reserve) as3

D (t) =

H ( t ) − H ( t − 1) + B(t) − B(t − 1)[1 + R(t − 1)] p(t)

(5.1)

t = 1, 2, . . . We are letting p(t) be the price level at time t , while R(t − 1) is the real rate of interest on one-period government bonds between time t − 1 and time t; B(t − 1)[1 + R(t − 1)] is the real par value of one-period government bonds that were issued at time t − 1 3 The

government collects income taxes on the interest payments on government debt, so that the after-tax real yield on short-term government debt might be negative in the United States today. However, the pretax yield is what belongs in equation (5.1), so long as private securities and government securities are taxed at a common rate. The reason is that any change in B(t − 1) is offset by an equal change of K (t − 1) in the opposite direction, where K (t − 1) is private investment measured in time t − 1 goods. Therefore, the taxes collected by taxing interest on government bonds only just replace the taxes lost because of the private investment that is crowded out by a larger B(t − 1) . Thus, define g(t) as government purchases minus all taxes except taxes on private and government securities, and let τ be the marginal tax rate. Then we have D (t) = g(t) − τ (1 + R)K (t − 1) − τ (1 + R) B(t − 1) , and equation 5.1 becomes g ( t ) − τ (1 + R ) K ( t − 1)

=

H ( t ) − 1 + ( t − 1) + B(t) − B(t − 1)(1 + R)(1 − τ ) p(t)

(5.1’)

which is the form of equation (5.1) in which the after-tax yield appears. We are assuming that complete crowding out occurs, which can be expressed as a constant level of total investment, B(t − 1) + K (t − 1) = B¯ . Substituting this equation into equation (5.1)’ gives H ( t ) − H ( t − 1) g(t) − (1 + R) B¯ = + B(t) − B(t − 1)(1 + R) p(t) This is equivalent with equation (5.1) in the text.

168

Chapter 5: Some Unpleasant Monetarist Arithmetic

and fall due in period t , where B(t − 1) is measured in units of time t − 1 goods and 1 + R(t − 1) is measured in time t goods per unit of time t − 1 goods. In equation (5.1) B(t) is government borrowing between periods t and t + 1, measured in units of time t goods. Equation (5.1) states that the excess of government expenditures over tax receipts must be financed by some combination of issuing additional currency and interest-bearing debt. Finally, we let N (t) be the population at time t . We assume that N (t) obeys the difference equation

N ( t + 1) = (1 + n ) N ( t )

t = 1, 2, . . .

(5.2)

with N (1) > 0 being given and n a constant exceeding −1. Dividing both sides of equation (5.1) by N (t) and rearranging gives the following form of the government’s budget constraint:  1 + R ( t − 1)  B ( t − 1) B(t) = N (t) 1+n N ( t − 1) h D (t)  H (t) − H (t − 1) i + − N (t) N (t) p(t)

(5.3)

We shall now use equation (5.3) and our monetarist model—assumptions (1)–(3)—to illustrate a version of the following proposition: If fiscal policy in the form of the D (t) sequence is taken as given, then tighter current monetary policy implies higher inflation in the future. To illustrate tighter current monetary policy, we specify alternative time paths for monetary policy in the following way. We take H (1) as predetermined and let alternative monetary policies be alternative constant growth rates of H (t) for t = 2, 3, . . . , T where T is some date greater than or equal to 2. For t > T , we suppose that the path of H (t) is determined by the condition that the outstanding stock of interest-bearing real government debt per capita be held constant at whatever level it attains at t = T . The restriction on monetary policy from time T onward is one way of taking into account that the real debt per capita cannot grow

A Simple Monetarist Model

169

without limits.4 Thus, with H (1) taken as given, we suppose that H ( t ) = (1 + θ ) H ( t − 1)

t = 2, 3, . . . , T

(5.4)

and examine the consequence of various choices of θ and T . Notice that we have written equation (5.1) in terms of real debt and real rate of return. If we want to apply this to a setting in which government bonds are not indexed, which is the situation in the United States today, then we must ensure that anticipated inflation is the same as actual inflation. We impose that condition, in part, by supposing that both the path of fiscal policy, the D (t) sequence, and the path of monetary policy, θ and T , are announced at t = 1 and known by private agents. Once we assume that, it does not matter whether nominal or indexed debt is issued from t = 1 onward. Now, to proceed, note that assumptions (1) and (3) imply that the price level at any time t is proportional to the time t stock of base money per capita, H (t)/N (t) , namely, p(t) =

1 H (t) h N (t)

(5.5)

for some positive constant h . From equation (5.5) it follows that for t − 2, . . . , T one plus the inflation rate is given by p(t)/p(t − 1) = (1 + θ )/(1 + n) . Thus, when we specify monetary policy, a θ and a T , we are simultaneously choosing the inflation rate for periods t = 2, 3, . . . T . We are interested in determining how the inflation rate for periods subsequent to T depends on the inflation rate chosen for the periods prior to T . We do this in two simple steps. We first determine how the inflation rate subsequent to T depends on the stock of real government debt per capita attained at T and to be held constant thereafter. We denote that per capita stock by bθ ( T ) . We then show how bθ ( T ) depends on θ . To find the dependence of the inflation rate for t > T on bθ ( T ) , we use equation (5.3) for any date t > T , substituting into it 4 Note

that equation (5.5) is a special case of equation (5.10) with h = γ1 /2 , γ2 = 0 . See Appendix 1 for an underlying model that implies equation (5.10) and all of our other assumptions.

170

Chapter 5: Some Unpleasant Monetarist Arithmetic

B(t/N (t) = B(t − 1)/N (t − 1) = bθ ( T ) and H (t) = hN (t) p(t) as implied by equation (5.5). The result can be written 1  p ( t − 1) = 1+n p(t) i 1 h D ( t )  R ( t − 1) − n  + bθ ( T ) h N (t) 1+n

1−



(5.6)

In order that equation (5.6) be satisfied by positive price levels p(t) and p(t − 1) , it is necessary that the right-hand side be less than unity. This is one condition that limits the size of the stock of real interest-bearing government debt per capita. If R(t − 1) − n is a positive constant, as stated by assumption (2), then the right-hand side of equation (5.6) increases when bθ ( T ) increases. This in turn implies that the inflation rate is higher the higher is bθ ( T ) , a conclusion that holds for all t > T . To complete the argument that a tighter monetary policy now implies higher future inflation, we must show that the smaller is θ , the higher is bθ ( T ) . To find bθ ( T ) and its dependence on θ , we first find B(1)/N (1) ≡ b(1) and then show how to find the entire path b(1), bθ (2), bθ (3), . . . , bθ ( T ) . We solve for b(1) from the t = 1 version of equation (5.3), namely b (1) =

D (1) H (1) − H (0) B˜ (0) + − p (1) N (1) N (1) N (1) p (1)

(5.7)

Here, in place of B(0)[1 + R(0)]/(1 + n) we have inserted B˜ (0)/p (1) N (1), B˜ (0) being the nominal par value of the debt issued at t = 0. By making this substitution, we avoid assuming anything about the relationship between actual inflation between time t = 0 and t = 1 and what it had been anticipated to be. In conjunction with equation (5.5), equation (5.7) permits us to solve for b(1) in terms of D (1), N (1), H (1), H (0) , and B˜ (0) . Note that b(1) does not depend on θ . We now proceed to find bθ (2), bθ (3), . . . , bθ ( T ) . Using equations (5.4) and (5.5) and the definition b(t) = B(t)/N (t) , we can

171

The Cagan-Bresciani-Turroni Effect

write equation (5.3) as  1 + R ( t − 1) 

D (t) b ( t − 1) + 1+n N (t)  θ  −h t = 2, 3, . . . , T 1+θ

b(t) =

(5.8)

By repeated substitution it follows for any t > 2 and t ≤ T that t

bθ (t) = φ(t, 1)b(1) +

φ(t, s) D (s) hθ − N (s) 1+θ s =2

∑

t

∑ φ(t, s)

(5.9)

s =2

where φ(t, s) =

[1 + R(s)][1 + R(s + 1)] · · · [1 + R(t − 1)] t>s (1 + n ) t − s

It follows from equation (5.9) that bθ ( T ) is larger the smaller is θ . This completes our demonstration of a version of the proposition that less inflation now achieved through monetary policy on its own implies higher future inflation. It is crucial for such a result that the real rate of return on government securities exceeds n from T onward (see equation (5.6)) and that the path of fiscal policy given by D (1), D (2), . . . , D (t) . . . does not depend on θ .

The Cagan-Bresciani-Turroni Effect In the analysis of the preceding section, circumstances have been delineated in which tighter monetary policy lowers the current rate of inflation, but at the cost of increasing inflation in the future. Our having assumed a demand schedule for money of the simplest quantity theory from (equation (5.5)) not only much simplified the analysis but had the substantive aspect of ignoring any dependence of the demand for base money on the expected rate of inflation. This dependence is widely believed to be important and acts to complicate the dynamics of the influence of monetary policy on the price level. Bresciani-Turroni (1937) and Cagan (1956) adduced substantial evidence for the existence of this effect by studying countries that had undergone rapid inflation. If the demand for

172

Chapter 5: Some Unpleasant Monetarist Arithmetic

money depends on the expected rate of inflation, then it turns out (see Sargent and Wallace (1973)) that the current price level depends on current and all anticipated future levels of the money supply. This sets up a force whereby high rates of money creation anticipated in the future tend to raise the current rate of inflation. As we shall show, this force can limit the power of tighter monetary policy to deliver even a temporarily lower inflation rate. We maintain all of the assumptions of the monetarist model except that we replace equation (5.5) by5 H (t) γ γ2 p(t + 1) = 1− p(t) N (t) 2 2 p(t)

t ≥ 1, γ1 > γ2 > 0

(5.10)

Equation (5.10) is a version of the demand schedule for money that Phillip Cagan used in studying hyperinflations. The equation is shown in Appendix 2 to imply the following equation for the price level at t : 2 ∞  γ2  j H (t + j) p(t) = . γ1 j∑ γ1 N (t + j) =0 This equation expresses the current price level in terms of current and all future value of the per capita supply of base money. Thus, the current price level and inflation rate depend not only on how tight money is today but also on how tight it is for all “tomorrows.” If the situation is, as in the monetarist model, that tighter money now causes looser money in the future, then this equation for p(t) suggests the possibility that tighter money today might fail to bring about a lower inflation rate and price level even today. We shall now discuss an example in which this possibility occurs. As above, policy consists of a deficit sequence D (t) , a date T at which monetary policy from then on is determined by the condition that the real per capita interest-bearing government debt be held constant, and θ , the growth rate of the monetary base for periods prior to T . In the model of this section, the path of the price level prior to T depends on all of these aspects of policy, and not just on θ as was the case for the monetarist model. 5 See

Appendix 3 for a discussion of how to find parameter values that imply this “paradoxical” price level behavior.

Conclusions and Qualifications

173

Appendix 2 describes a way of solving for the paths of the endogenous variables. Here we simply present an example in which a tighter monetary policy in the form of a lower θ implies a uniformly higher price level and inflation rate. The economy of the example is characterized by γ1 = 3.0, γ2 = 2.50, R = 0.05, and n = 0.02. The common features of policy are a per capita deficit sequence d(t) with d(t) = 0.05 for t = 1, 2, . . . , 10 and d(t) = 0 for t > 10, T = 10, and [ H (0) + B˜ (0)]/H (1) = 200/164.65. Two different θ ’s are studied: θ = 0.106 and θ = 0.120. It turns our that the price level at t = 1 is 1.04 percent higher in the economy with the smaller θ . Table 5.1 compares the inflation rates, price levels, per capita bond holdings, and per capita real money balances for the two economies. This example is spectacular in that the easier monetary policy is uniformly better than the tighter policy. (In terms of the model of Appendix 1, the equilibrium for the easier monetary policy is Pareto superior to that for the tighter monetary policy.) In this example, the tighter current monetary policy fails to buy even a temporary improvement in inflation vis-à-vis the looser current policy.6

Conclusions and Qualifications It bears emphasizing that our argument uses the assumption of a rational expectations equilibrium, which in this case is equivalent with perfect foresight because there is no randomness in the model. Thus, our statements involve comparing alternative paths for monetary and fiscal variables that are known in advance. The authorities are assumed to stick with the plans that they announce, and not to default, in real terms, on the interest-bearing debt issued from time 1 onward, so that it is as if all interest-bearing debt were indexed. We believe that such an assumption is appropriate for analyzing the available options with regard to choosing time sequences or strategies for monetary policy variables. We would 6 By

pursuing the CBT example and other examples comparing welfare of agents across stationary states, the model can be used to support Milton Friedman’s (1948) prescription that the entire government deficit be financed by creating base money.

174

Chapter 5: Some Unpleasant Monetarist Arithmetic

Table 5.1 P RICE L EVEL AND R EAL B ALANCE D YNAMICS Price level Date

θ = 0.106

1 2 3 4 5 6 7 8 9 t ≥ 10

Inflation rate

θ = 0.120

θ = 0.106

θ = 0.120

1.08415 1.08413 1.08411 1.08409 1.08406 1.08404 1.08401 1.08397 1.08394 1.08390

1.08250 1.08082 1.07892 1.07677 1.07434 1.07156 1.06839 1.06474 1.06051 1.05557

1.13695 1.12105 1.23262 1.21354 1.33632 1.31161 1.44871 1.41512 1.57053 1.52376 1.70255 1.63703 1.84562 1.75418 2.00067 1.87416 2.16867 1.99549 p(10)π (10)t−10

(continued) Per capita bond holdings

Per capita real balances

θ = 0.106

θ = 0.120

θ = 0.106

θ = 0.120

0.08109 0.11960 0.15923 0.20003 0.24202 0.28525 0.32974 0.37554 0.42268 0.47120

0.08153 0.11797 0.15522 0.19328 0.23214 0.27177 0.31213 0.35320 0.39490 0.43718

0.12023 0.14484 0.14486 0.14489 0.14492 0.14496 0.14499 0.14504 0.14508 0.14513

0.14687 0.14898 0.15136 0.15404 0.15708 0.16055 0.16405 0.16907 0.17436 0.18053

be prepared to argue for the appropriateness of this assumption despite historical examples in which governments have defaulted on substantial fractions of their interest-bearing debt by inflating it away. Our reason is that such a default option is not available as a recurrent policy to which the government can persistently plan to resort. We should also note that our argument is actually a version of a line of reasoning that exists in the literature on the macroeconomics of the government budget constraint, which is discussed, for example, by McCallum (1978, 1981) and Scarth (1980). Various authors have pointed out that in the presence of real rates of interest ex-

Conclusions and Qualifications

175

ceeding the growth rate of the economy, the government budget constraint can be interpreted as an unstable difference equation in real debt and that it implies an explosive path for real government debt per capita under some “k-percent” rules for monetary growth. What we have done in this chapter is simply to point out the implications for anti-inflation policy and that it is not feasible to follow such explosive paths forever. We have made two crucial assumptions to obtain our conclusions. The first is that the real rate of interest exceeds the growth rate of the economy. We have used that assumption because it seems to be maintained by many of those who argue for a low rate of growth of money no matter what is the current deficit. If we were to replace that assumption, we would instead assume that the public’s demand for government bonds is an increasing function of their real rate of return, with an initial range over which that demand is positive at rates of return that are negative or less than the growth rate of the economy. But to induce larger demands for government bonds, we would assume that their real rate of return would have to exceed the rate of growth of the economy. We would also continue to assume that there is an upper bound on the quantity of bonds demanded per capita. Such a demand function for government bonds would imply that monetary policy helps determine the real rate of interest on government bonds, and that for some monetary policies entailing low enough bond supplies, seigniorage can be earned on bonds as well as on base money. However, an analysis that includes such a demand schedule for bonds would share with ours the implication that a sufficiently tight monetary policy can imply growth in government interestbearing indebtedness so rapid that inflation in the future results in being higher than it would be with an easier current monetary policy. The second crucial assumption that we have made is that the path of D (t) is given and does not depend on current or future monetary policies. This assumption is not about the preferences, opportunities, or behavior of private agents, as is our first crucial assumption, but is about the behavior of the monetary and fiscal authorities and the “game” that they are playing. Since the monetary authorities affect the extent to which seigniorage is ex-

176

Chapter 5: Some Unpleasant Monetarist Arithmetic

ploited as a revenue source, monetary and fiscal policies simply have to be coordinated. The question is, which authority “moves first”—the monetary authority or the fiscal authority—or to put it differently, who imposes discipline on whom? The assumption made in this chapter is that the fiscal authorities move first, their “move” consisting of an entire D (t) sequence. Given that sequence, monetary policy is to be determined in a way consistent with that D (t) sequence, if that is possible (as we have seen, it may not be possible if the D (t) sequence is “too big” for “too long”). Given this assumption about the game played by the authorities, and given our first crucial assumption, the monetary authorities can make money tighter now only by making it looser later. One can interpret the call for monetary restraint differently than we have in this chapter, and as a call to let the monetary authority move first and thereby impose discipline on the fiscal authority. On this interpretation, the monetary authority moves first by announcing a fixed θ rule like equation (5.4) not just for t = 2, 3, . . . , T but for all t ≥ 1. By doing this in a binding way, the monetary authority forces the fiscal authority to choose a D (t) sequence consistent with the announced monetary policy. This form of permanent monetary restraint is a mechanism that effectively imposes fiscal discipline. Alternative monetary mechanisms that impose fiscal discipline that have been suggested are fixed exchange rates or a commodity money standard. It is evident that nothing in our analysis denies the possibility that monetary policies can permanently affect the inflation rate under such a monetary regime that effectively permanently disciplines the fiscal authority. We wrote this chapter not because we think that our assumption about the game played by the monetary and fiscal authorities describes the way monetary and fiscal policies should be coordinated but out of a fear that it may describe the way the game is now being played.

Appendix 1

177

Appendix 1

This appendix describes a simple formal model that implies the assumptions used in the text. It is a version of Paul Samuelson’s (1958) model of overlapping generations in which people use money to smooth out their consumption streams over time. We describe the evolution of the economy from time t = 1 onward. The economy is populated by two-period lived agents. At each date t ≥ 1 there are born N1 (t) identical poor people who are endowed after taxes with α1 units of the single good when young and α2 units when old. At each date t ≥ 1 there are also born N2 (t) identical rich people who are endowed after taxes with β of the good when young and zero when old. We assume that N1 (t) = (1 + n) N1 (t − 1), N2 (t) = (1 + n) N2 (t − 1) for t ≥ 1, with N1 (0) and N2 (0) given and n > −1. The total population is N (t) = N1 (t) + N2 (t) . There is available a physical technology for converting time t goods into time t + 1 goods. In particular, if k (t) ≥ k goods are stored at time t ≥ 1, then (1 + R)k (t) goods become available at time t + 1. This is a constant returns-to-scale technology with a constant real rate of return on investment R > 0. We assume that there is a minimum scale of k at which this investment can be undertaken and that β/2 > k > α1 , the first period endowment of the poor. We also assume that a legal restriction on intermediation prevents two or more of the poor from sharing investments, thereby preventing the poor from holding the real investment. The government issues currency, which does not bear interest, and interest-bearing bonds. The currency is held by the poor because government bonds are issued in such large denominations that only the rich can hold them. Again, a legal restriction on intermediation is relied on to prevent two or more people from sharing a government bond. There is no uncertainty in the model, so that the rich will hold government bonds only if the real interest rate on bonds at least equals that on private investment, which must be at least as large as the yield on currency. As in the text, the government finances a real deficit D (t) via some combination of currency creation and bond creation. The

178

Chapter 5: Some Unpleasant Monetarist Arithmetic

government’s budget constraint is D (t) =

H ( t ) − H ( t − 1) + B(t) − B(t − 1)(1 + R) p(t)

t ≥ 1 (5.11)

where H (t) is measured in dollars, p(t) in dollars per time t goods, D (t) in time t goods, and B(t) in time t goods. In addition, at time t = 1 there are N1 (0) + N2 (0) old poor and rich people, respectively, who hold H (0) units of currency and maturing bonds of par nominal value B˜ (0) . The old people alive at time t = 1 simply offer all of their currency inelastically in exchange for goods to those who are young at time 1. The young of each generation t ≥ 1 are assumed to maximize the utility function cth (t)cth (t + 1) where cth (s) is consumption of an agent of type h of the s -period good of an agent born at time t . Letting wth (s) be the endowment of agent type h of the s -period good of an agent born at t and assuming that each agent faces a single rate of return Rh , a young agent h at generation t chooses a lifetime consumption bundle to maximize utility subject to the present value constraint: cth (t) +

wth (t + 1) cth (t + 1) h ( t ) + = w t 1 + Rh 1 + Rh

The solution to this problem is the saving function: wth (t) − cth (t) =

wth (t) − wth (t + 1)/(1 + Rh ) 2

(5.12)

Since all saving of poor people is in the form of currency, if h is poor, 1 + Rh = p(t)/p(t + 1) . Moreover, in the range where p(t)/p(t + 1) < 1 + R , only the poor hold currency. Thus, in this range the money market equilibrium condition is that H (t)/p(t) equals the total real saving of the poor, which by equation (5.12) is N1 (t)[α1 − α2 p(t + 1)/p(t)]/2. Dividing by N (t) , we can write this condition as H (t) [α − α2 p(t + 1)/p(t)] N1 (t) = 1 p(t) N (t) 2N (t)

(5.13)

This is equation (5.10) if we let α1 /2 = α1 N1 (t)/2N (t) and α2 /2 = α2 N1 (t)/2N (t) . We get equation (5.5) if α2 = 0.

Appendix 1

179

According to equation (5.12), each rich person saves a constant amount β/2 per period. So long as government bonds bear the real rate of return R , each rich person is indifferent between holding government bonds or private capital. However, in the aggregate, the rich only wish to save N2 (t) β/2 per period. The number β/2 determines the upper bound on per capita holdings of interestbearing government debt that was alluded to in the text. We let K (t) denote the total amount of real investment (“storage”), measured in goods, undertaken by the young members of generation, t , all of them rich. We then have K (t) + B(t) =

N2 (t) β = B¯ (t) 2

(5.14)

where B(t) is the amount of loans to the government. Equation (5.14) expresses the result that additional government borrowing merely crowds out private investment on a one-for-one basis. The national income identity can be written: N1 (t)c1t (t) + N2 (t − 1)c1t−1 (t) + N2 (t)c2t (t)

+ N2 (t − 1)c2t−1 (t) + K (t) + G (t) = N1 (t)α1 + N1 (t − 1)α2 + N2 (t) β + T (t) + (1 + R)K (t − 1) (5.15) Here G (t) denotes government purchases and T (t) denotes total direct taxes. The government deficit as defined in the text is related to G (t) and T (t) by D (t) = G (t) − T (t) . Thus, so long as solutions satisfy p(t)/p(t + 1) < 1 + R and total real bond supply is less than B¯ (t) , the model just described implies all the assumptions made in the text. This particular model also has implications for how different agents fare under different policies. It is evident from the present value budget constraint set out above that each poor person is better off the lower the inflation rate, that each rich person is unaffected by the inflation rate, and that those who at t = 1 are in the second period of their lives and who are holding currency or maturing bonds are better off the lower is the initial price level p(1) . These observations lie behind

180

Chapter 5: Some Unpleasant Monetarist Arithmetic

our claim that for the Cagan-Bresciani-Turroni (CBT) example, the tight-money policy is Pareto inferior to the loose-money policy.7

Appendix 2 In this appendix, we analyze the CBT model that results from generalizing the analysis of the monetarist model by assuming a demand schedule for base money that depends on the expected rate of inflation. The particular demand schedule that we use resembles Phillip Cagan’s (1956) famous demand schedule and can be deduced formally from the model in Appendix 1 by assuming that the “poor” of each generation are endowed with γ1 N (0)/N1 (0) > 0 units of the assumption good when young and γ2 N (0)/N1 (0) > 0 units when they are old. (The monetarist model emerges when we set γ2 = 0.) Except for this generalization, all other features of the model remain as discussed. We now assume a demand schedule for base money of the form H (t) γ γ2 p(t + 1) = 1− p(t) N (t) 2 2 p(t)

t≥1

(5.16)

where γ1 > γ2 > 0. Except for replacing equation (5.5) with (5.16), we retain all the other features of the monetarist model, including the budget restraint (5.1) and the law of motion of total population (5.2). We describe experiments similar to the monetarist model: We hold the per capita real government debt b(t) constant for t > T and examine the choice of alternative rates of growth of base money θ for t = 2, . . . , T . The step of replacing equation (5.5) with (5.16) substantially complicates the dynamics of the system, as we shall see. We begin by examining the behavior of the system for t > T . For t > T + 1 we specify as before that monetary policy is determined so that b(t) = b(t − 1) = b( T ) . Using the budget constraint (equation (5.1)) together with this condition implies H ( t ) − H ( t − 1) R ( t − 1) − n D (t) = b( T ) + , t > T + 1 (5.17) N (t) p(t) 1+n N (t) 7 Where

necessary, the time index is indicated by a subscript. This departs from the practice in the text but should cause no confusion.

181

Appendix 2

We now assume that D (t) = d, N (t)

t≥T

where d is a constant, which is a computationally convenient assumption, although the general flavor of our results does not depend on making it. We now define per capita real balances as m(t) = H (t)/N (t) p(t) and the one-period gross inflation rate π (t) = p(t)/p(t − 1) . In terms of these variables, equations (5.16) and (5.17) become m(t) =

m(t) −

γ1 γ2 − π ( t + 1) 2 2

m ( t − 1) =ξ π (t)(1 + n)

t≥1

(5.18)

t ≥ T+1

(5.19)

where ξ = [( R − n)/(1 + n)] b( T ) + d . The variable ξ has the interpretation of the per capita deficit that must be financed by seigniorage from time T + 1 onward. Eliminating m(t) and m(t − 1) from these equations by substituting equation (5.18) into (5.19) leads to the following nonlinear difference equation in π (t) : γ1 1 2ξ + − − γ2 1 + n γ2 γ1  1
1 , t ≥ T+1 γ2 1 + n π (t)

π ( t + 1) =

(5.20)

This equation is graphed in Figure 5.2. It is readily verified that if γ

1

γ2

+

1 2ξ 2 4γ1 − >0 − 1 + n γ2 γ2 (1 + n)

(5.21)

then equation (5.20) has two stationary points, their values given

182

Chapter 5: Some Unpleasant Monetarist Arithmetic

π ( t +1)

π(t+1)=

γ

1

+

γ2

γ1 γ

2

1 1+ n

2ξ γ2

1 1 n + 1 π( t )

45°

π1

π

π2

γ1 γ

π( t )

2

Figure 5.1 The equilibrium law of motion of the inflation rate. The curved line shows the equilibrium inflation rate in period t + 1, πt+1 , as a function of the inflation rate in period t . The dotted line is the 45 degree line. The intersections of the curved line with the 45 degree line are stationary equilibrium inflation rates.

by 1 1 n γ1 2ξ + − 2 γ2 1 + n γ2 1 h γ 1 2ξ 2 4γ1 i 2 o 1 − + − − γ2 1 + n γ2 γ2 (1 + n) n 1 γ1 1 2ξ π2 = + − + 2 γ2 1 + n γ2 1 h γ 1 2ξ 2 4γ1 i 2 o 1 + − − γ2 1 + n γ2 γ2 (1 + n) π1 =

(5.22)

We let ξ¯ be the value of ξ for which the left side of equation (5.21) equals zero. Evidently, ξ¯ is a function of γ1 , γ2 , and n and represents the maximum stationary per capita deficit that can be financed by seigniorage. From equation (5.21) it follows that if

183

Appendix 2

π ( t +1) γ

1

γ2

+

1 1+ n

2ξ γ2

γ1

+

γ2

1 1+ n

π1

π2

γ

1

γ

1

γ

2

1 1+ n

2ξ γ2

1 1 n + 1 π( t )

π( t )

γ2

Figure 5.2 The equilibrium dynamics of inflation.

ξ = 0, then π1 = 1(1 + n) and π2 = γ1 /γ2 . From Figure 5.1 it immediately follows that for ξ¯ > ξ > 0, π1 > 1/(1 + n), π2 < γ1 /γ2 and that raising ξ causes π1 to raise and π2 to fall. Inequality (5.21) is a necessary and sufficient condition for it to be possible to finance the per capita deficit ξ by seigniorage. Assuming equation (5.21) is satisfied, there exists a multiplicity of inflation-real balance paths that finance the deficit. Any setting of π ( T ) satisfying π1 < π ( T ) < γ1 /γ2 , with π (t) for t > T + 1 being given by equation (5.20), results in the deficit being financed. (Later we shall describe the money supply paths needed to accomplish these paths for inflation.) Figure 5.3 again graphs equation (5.20) and shows that for any π ( T ) > π1 , π (t) → π2 as t → ∞ . Thus, three classes of inflation paths finance the deficit: (a) the stationary path with π (t) = π1 , t > T; (b) the stationary path with π (t) = π2 , t > T ; and (c) the class of nonstationary paths with γ1 /γ2 > π ( T ) > π1 and lim π (t) = π2 as t → ∞ . We assume that the government selects the money supply path so that π ( T ) = π1 and so that the deficit is financed by the uni-

184

Chapter 5: Some Unpleasant Monetarist Arithmetic

formly lowest inflation rate path, and therefore, in view of equation (5.18), the lowest price level path. This assumption is reasonable since this selection leaves the government with the same resources as any other selection, while leaving holders of money better off. Having determined the inflation rate p(t)/p(t − 1) = π1 for t > T + 1 from equation (5.21), we can determine the time T real balances and price level by setting t = T in equation (5.18): H (T ) γ γ2 = 1 − π1 p( T ) N ( T ) 2 2 or p( T ) =

 H (T ) 2 1 γ1 1 − (γ2 /γ1 )π1 N ( T )

(5.23)

Since H ( T ) and N ( T ) are given at T , this equation determines p( T ) as an inverse function of π1 . Also, since π (t) is constant for t > T + 1, we have from equation (5.18) and the definition of m(t) ≡ H (t)/N (t) p(t) that H (t) H ( t − 1) = π1 N (t) N ( t − 1)

t ≥ T+1

so that per capita nominal balances grow at the constant gross rate π1 , which is the rate of inflation for t > T + 1. It is instructive briefly to describe the following alternative way to solve the system for t > T + 1 by obtaining a pair of linear difference equations. Define h(t) = H (t)/N (t) , and write the budget constraint (equation (5.17)) as  1  h ( t − 1) + ξ p ( t ) t ≥ T + 1 h(t) = 1+n and the demand function for base money (equation (5.16)) as p(t) =

γ2 2 p ( t + 1) + h ( t ) γ2 γ1

t≥1

Using the lag operator L , we write these two equations as h  1  i 1− L h(t) = ξ p(t) t ≥ T + 1 1+n (5.24)  γ2 −1  2 L p(t) = h(t) t ≥ 1 1− γ1 γ1

185

Appendix 2

Solving the second equation in terms of h(t) gives p(t) =

 γ t γ2 −1 −1 2 1 1− L h(t) + c γ1 γ1 γ2

(5.25)

or p(t) =

2 γ1

∞

∑

j =0

 γ j 2

γ1

h(t + j) + c

 γ t 1

γ2

t≥1

(5.26)

where c is any nonnegative constant. Substituting equation (5.25) into (5.24) and operating on both sides of the result with [1 − (γ2 /γ1 ) L−1 ] gives the following homogeneous difference equation in h(t) : n γ 1 2ξ  1 + − L L −1 − 1 + γ2 1 + n γ2  1 γ o 1 2 − L h(t) = 0 1 + n γ2

(5.27)

The characteristic polynomial in L can be factored in the usual way, so that L−1 [(1 − π1 L)(1 − π2 L)]h(t) = 0

(5.28)

where π1 and π2 are the same roots given in equation (5.22). Since for ξ > 0 we have π1 < π2 < γ1 /γ2 , it follows that the geometric sum in current and future h(t) that appears in equation (5.25) converges for any h(t) paths that satisfy equation (5.28) or, equivalently, h ( t ) = ( π1 + π2 ) h ( t − 1) − π1 π2 h ( t − 2) t ≥ T + 1,

(5.29)

where h( T ) is given, and h(t + 1) is free. To ensure that the deficit is financed each period, we have to add two side conditions to those listed under equation (5.29): it is necessary to set c = 0 in equation (5.25) and to set h( T + 1) so that (5.25) implies that π ( T + 1) < γ1 /γ2 . All of the price level paths with c > 0 have lim t→∞ π ( t ) = γ1 /γ2 , which in view of equations (5.18) and (5.19)

186

Chapter 5: Some Unpleasant Monetarist Arithmetic

imply that lim m(t) = 0 and that a positive deficit cannot be financed. Any path with π ( T ) > γ1 /γ2 implies nonpositive real balances at T . Since we are assuming that the government selects h(t) = π1 h(t − 1), t > T + 1, h( T ) given, equation (5.26) with t = T becomes equivalent to equation (5.3). We note that the admissible paths given by equation (5.29) with h( T + 1) 6= π1 h( T ) have limt→∞ h(t)/h(t − 1) = π2 and so constitute the per capita nominal money supply paths that correspond to the inflation paths with π ( T ) > π1 in Figures 5.1 and 5.2. In summary, we have that for t > T the price level and the stock of base money per capita evolve according to p(t) =

 2 1 h(t) γ1 1 − γ2 /γ1 π1

t>T

h ( t + 1) = π1 h ( t )

(5.30)

(5.31)

subject to h( T ) given, where from equation (5.22) we have π1 =

1 n γ1 1 2ξ + − − 2 γ2 1 + n γ2 1 h γ 1 2ξ 2 4γ1 i 2 o 1 + − − γ2 1 + n γ2 γ2 (1 + n)

and ξ=

R − n

b( T ) + d 1+n We now describe the behavior of the price level, supply of base money, and stock of real government debt per capita for t < T . As in the monetarist model, we assume a constant growth rate of base money. H ( t ) = (1 + θ ) H ( t − 1)

t = 2, 3, . . . , T

(5.32)

Equation (5.25) with c = 0 implies that for all t > 1, p(t) =

2 γ1

∞

 γ j

j =0

γ1

∑

2

h(t + j)

(5.33)

187

Appendix 2

Further, we know from equations (5.32) and (5.31) that h(t) = µh(t − 1)

t = 2, 3, . . . , T

(5.34)

t = T + 1, T + 2, . . .

(5.35)

1+θ 1+n

(5.36)

h ( t ) = π1 h ( t − 1) where µ=

Let us define the parameter φ by φ=

γ2 γ1

(5.37)

and write (5.33) for t < T as p(t) =

2 γ1

T −t

2

∞

∑ φ j h(t + j) + γ1 ∑

φ j h(t + j)

(5.38)

j = T − t +1

j =0

Substituting equations (5.34) and (5.35) into (5.38) and using some algebra implies p(t) =

2 1 − φπ1 + (π1 − µ)φ T −t+1 µ T −1 h(t) γ1 (1 − φπ1 )(1 − φµ)

t2

(5.40)

Next, we define s(t) as per capita seigniorage, s(t) ≡

H ( t ) − H ( t − 1) p(t) N (t)

For t < T , we have that s(t) =

1 θ h ( t − 1) p(t) 1 + n

or s(t) =

 θ  h(t) 1 + θ p(t)

Using equation (5.34) in the above equation gives s(t) =

θ γ1 (1 − φµ)(1 − φπ1 ) 1 + θ 2 1 − φπ1 + (π1 − µ)φ T −t+1 µ T −t

188

Chapter 5: Some Unpleasant Monetarist Arithmetic

Table 5.2 B EHAVIOR OF THE S YSTEM B EFORE AND A FTER T 1 2)

B˜ (0) H (1) − H (0) + d (1) − p (1) N (1) N (1) p (1)

(1 − φµ)(1 − φπ1 ) θ γ1 1 + θ 2 (1 − φπ1 + (π1 − µ)φ T −t+1 µ T −t )

(5.41)

(5.42)

( t ≥ 2)

(5.40)

R − n

1+θ γ b( T ) + d φ = 2 µ = 1+n γ1 1+n H (t) h(t) ≡ N ( t ) = (1 + n ) N ( t − 1) N (t)  1 γ1 1 2ξ π1 = + − − 2 γ2 1 + n γ2   12  γ1 1 2ξ 2 4γ1 + − − γ2 1+n γ2 γ2 (1 + n)

ξ=

(5.39)

(5.22)

t>T h ( t + 1) = π1 h ( t ) p(t) =

i −1 2 h γ 1 − 2 π1 h(t) γ1 γ2

b(t) = b( T ) i  γ1 h 1 γ s(t) = 1− 1 − 2 π1 2 π1 (1 + n ) γ1 d(t) = d

(5.31) (5.30)

Appendix 2

189

Using (5.1), the definition of s(t) and the definition d(t) = D (t)/N (t) , we have the law of motion for per capita real interest-bearing government debt 1 + R b ( t − 1) + d ( t ) − s ( t ) t > 2 (5.41) b(t) = 1+n Finally, we repeat equation (5.7), the special version of equation (5.41) for t = 1, b (1) =

H (1) − H (0) B¯ (0) + d (1) − p (1) N (1) p (1)

(5.42)

where B˜ (0) is the nominal par value of the one-period interestbearing debt that was issued at time t = 0. We have collected the equations describing the equilibrium before and after T in Table 5.2. Starting at t = 1, the system works as follows. We take as exogenous a time path of the per capita deficit net of interest payments, {d(t), t > 1} , with d(t) = d, t > T . We further take as exogenous B˜ (0) and H (0) , which give the nominal par value of government debt inherited from the past. The date T is also taken as exogenous. The monetary authority chooses settings for H (1) and θ . Then equations (5.34), (5.36), and (5.39) to (5.42) simultaneously determine p(t), b(t), t = 1, . . . , T , while equation (5.30) determines p(t) for t > T . The equations of the model are linear in the endogenous variables, given a value for π1 . However, from equation (5.22) and the fact that ξ = [( R − n)/(1 + n)]b( T ) + d , we see that π1 is itself a function of b( T ) , which in turn depends on the value of π1 through its effect on the behavior of p(t) and s(t) for 1 < t < T , via equation (5.39). Thus, determining the equilibrium of the system involves solving a nonlinear system of equations. While the system can be solved in a variety of ways, we have found it convenient to use the following procedure based on backward recursions. We begin by taking θ as given, but not H (1) . We choose a value for b( T ) and solve equation (5.22) for π1 . Then we recursively solve equations (5.40) and (5.41) backward for values of {b(t), s(t + 1); t = T − 1, T − 2, . . . , 1} . Also, from equation (5.39) we can determine per capita real balances h(t)/p(t) for t = 1, . . . , T . Finally, given the values of b(1) and h(1)/p(1) thus

190

Chapter 5: Some Unpleasant Monetarist Arithmetic

Table 5.3 T WO E CONOMIES Price level

Inflation rate

Date

θ = 0.106

θ = 0.120

θ = 0.106

θ = 0.120

1 2 3 4 5 6 7 8 9

1.13699 1.23267 1.33637 1.44878 1.57061 1.70264 1.84573 2.00079 2.16881

1.12300 1.21553 1.31362 1.41710 1.52568 1.63883 1.75578 1.87546 1.99637

1.08415 1.08413 1.08411 1.08409 1.08407 1.08404 1.08401 1.08398 1.08394

1.08240 1.08070 1.07878 1.07662 1.07416 1.07136 1.06816 1.06447 1.06020

(continued) Per capita bond holdings

Per capita real balances

θ = 0.106

θ = 0.120

θ = 0.106

θ = 0.120

0.08109 0.11959 0.15923 0.20002 0.24202 0.28524 0.32974 0.37554 0.42268 0.47120

0.08109 0.11750 0.15472 0.19274 0.23156 0.27114 0.31146 0.35247 0.39411 0.43631

0.14481 0.14483 0.14486 0.14489 0.14492 0.14495 0.14499 0.14503 0.14507 0.14512

0.14701 0.14913 0.15152 0.15423 0.15730 0.16079 0.16480 0.16941 0.17475 0.18099

determined, we solve equation (5.42) for the value of H (1) (or equivalently of p(1) ). This procedure produces a choice of H (1) and θ , and associated sequences for b(t) , p(t) , h(t) , and s(t) that solve the system. By employing iterations on this procedure, it is possible to solve the model taking b(1) as given. The method is simply to search over solutions of the type described in the previous paragraph, varying b( T ) until the specified initial value of b(1) is found. In this way, a set of equilibria with different θ ’s can be calculated, each one of which starts from the same value of b(1) . In a similar fashion, it is possible to generate equilibria with different θ ’s, each one of which starts from the same value of H (1) (of course b(1)

Appendix 2

191

will then differ across the different θ ’s). This last procedure was the one used to generate the examples in the text, each of which started with H (1) = 164.65. We now describe the results of using this solution procedure to compute equilibria of economies with identical values of the parameters {γ1 , γ2 , N (0), d(t), B˜ (0), H (0), T, b(1)} but different values of θ . Since their values of θ are different, the values of the endogenous variables { p(t), t ≥ 1} and {b(t), t ≥ 2} will in general be different. Table 5.3 compares two economies, each with {γ1 = 3,γ2 = 2.5,N (0) = 1000, n = 0.02, d(t) = 0.05, 1 ≤ t ≤ T; d(t) = d = 0, t > T; B˜ (0) = 100, H (0) = 100, T = 10, b(1) = 0.08109, R = 0.05} . The tight-money economy has θ = 0.106, while the loosemoney economy has θ = 0.120. As can be seen from Table 5.3, the θ = 0.120 economy has a uniformly higher price level and inflation rate for all t ≥ 1 than does the θ = 0.106 economy. We note that as expected, the θ = 0.120 economy is associated with a slower rate of bond creation from t = 1 to t = 10 and that therefore the looser-money θ = 0.120 economy ends up permitting slower growth in base money T than does the alternative economy with tighter money initially. Thus, tighter money now eventually implies looser money later, as in the economy described in the monetarist model. In the present example, however, the effect of expected future rates of money creation on the current rate of inflation is sufficiently strong that the economy with tighter money initially is the one with higher inflation both now and in the future. This happens because, via equation (5.33), the higher eventual rate of money creation associated with the lower path more than offsets the downward effects on the initial inflation rates that are directly associated with the lower initial rate of money creation. Like the closely related example in the text, this comparison provides a spectacular example in which tighter money now fails to buy even a temporarily lower inflation than does looser money now. Table 5.4 compares different θ ’s in an economy that provides an example intermediate between the economy of the monetarist model and the “spectacular” examples. Both economies of Table 5.4 have the parameters {γ1 = 2, γ2 = 1.5, N (0) = 1000, n = 0.02; d(t) = 0.05, 1 ≤ t ≤ T; d(t) = d = 0, t > T; B˜ (0) =

192

Chapter 5: Some Unpleasant Monetarist Arithmetic

Table 5.4 D IFFERENT θ S Price level

Inflation rate

Date

θ = 0.010

θ = 0.030

θ = 0.010

θ = 0.030

1 2 3 4 5 6 7 8 9

0.57675 0.57924 0.58441 0.59315 0.60662 0.62640 0.65455 0.69386 0.74802

0.59590 0.60732 0.62071 0.63671 0.65617 0.68023 0.71040 0.74870 0.79782

1.00431 1.00893 1.01495 1.02272 1.03259 1.04494 1.06005 1.07806 1.09887

1.01916 1.02205 1.02578 1.03057 1.03666 1.04435 1.05391 1.06561 1.07962

(continued) Per capita bond holdings

Per capita real balances

θ = 0.010

θ = 0.030

θ = 0.010

θ = 0.030

0.15000 0.20200 0.25558 0.31079 0.36770 0.42637 0.48688 0.54930 0.61372 0.68020

0.15000 0.19761 0.24670 0.29734 0.34961 0.40358 0.45935 0.51701 0.57667 0.63845

0.24678 0.24330 0.23879 0.23296 0.22556 0.21629 0.20295 0.19145 0.17585 0.15846

0.23563 0.23347 0.23067 0.22487 0.22250 0.21674 0.20957 0.20080 0.19028 0.17798

100, H (0) = 100, T = 10, b(1) = 1.4999, R = 0.05} . The tightmoney economy has θ = 0.01, while the loose-money economy has θ = 0.03. The tight-money economy experiences a lower inflation rate for 1 ≤ t ≤ 5 but a higher rate for t ≥ 5. (Here the gross inflation rate at t is defined as the “right-hand” rate p(t + 1)/p(t) .) In this case the effect of the higher eventual rate of money creation that is associated with the initially tighter θ = 0.01 economy causes inflation to be higher even before T when money actually becomes looser. But this effect is not strong enough to eliminate completely the temporary benefits of tight money on the current inflation. Notice, however, that as compared to the monetarist example, the effect of the initial tight money on the

Appendix 3 by Danny Quah

193

initial inflation rate is considerably weakened. For with all other parameters the same but γ2 = 0 (the monetarist case), we would have had p(t + 1)/p(t) = (1 + θ )/(1 + n) = 0.9902 for 1 ≤ t ≤ T . In the following appendix, Danny Quah establishes conditions on the parameters of the model that are sufficient to deliver the spectacular example of the Table 5.2 variety with b(1) fixed across the economies.

Appendix 3 by Danny Quah This appendix establishes sufficient conditions for the case where tighter monetary policy (lower θ ) leads to uniformly higher price level and inflation for all t ≥ 1. The method is by construction: A pair of inequalities will be reduced to a single relation by the correct choice of certain parameter values. We satisfy the inequalities by making the implicit discount rate (1 − γ2 /γ1 ) sufficiently low, while maintaining convergence of the relevant infinite sum. Let θh > θl . Then we want8 pt (θl ) > pt (θh )

(a)

p t +1 ( θ l ) p (θ ) > t +1 h pt (θl ) pt (θh )

(b)

for all t . By equations (5.30) and (5.31) in Table 5.1, for t ≥ T , pt+1 (θ )/pt (θ ) = π1 (θ ) . For policy experiments that fix b(1) , it is clear that (over the relevant range) a lower θ leads to a higher b( T ) and hence to a higher ξ . This is exactly the statement that a tighter monetary policy now implies a higher deficit to be financed by seigniorage from time T + 1. From Figure 5.1 an increase in ξ increases the value of the root π1 . Therefore, π1 (θl ) > π1 (θh ) . Hence (b) is satisfied for t ≥ T . Condition (a) follows, at most, T 0 periods after T (where T 0 is finite) given (b) for t ≥ T . 8 This

procedure almost always obtains the desired example. I say “almost” because strictly speaking ( a000 ) and (b000 ) imply (c) but not conversely.

194

Chapter 5: Some Unpleasant Monetarist Arithmetic

Hence we restrict attention to t < T . It is clear that if (b) holds for t < T, p1 (θl ) > p1 (θh ) implies (a) for t < T and therefore for all t . From equation (5.7) p1 ( θ ) =

[ B˜ (0) + H (0)]/N (1) b(1) − d(1) + H1 (θ )/N (1) p1 (θ )

But by equation (5.39) H1 (θ ) [1 − φπ1 (θ )][1 − φµ(θ )] h (θ ) γ = 1 = 1 N (1) p1 ( θ ) p1 ( θ ) 2 1 − φπ1 (θ ) + [π1 (θ ) − µ(θ )]φ T µ(θ )T −1 Calling this m1 (θ ), p1 (θ ) = k1 /[k2 + m1 (θ )] where k1 > 0, k2 + m1 (θ ) > 0. Then p1 (θl ) > p1 (θh ) > p1 (θh ) if and only if m1 (θh ) > m1 ( θ l ) . Define the function χ(φ, θ, t) = 1 − φπ1 (θ ) + [π1 (θ ) − µ(θ )] φ T −t+1 µ(θ )T −t . Then using equations (5.34), (5.36), and (5.39) to write out explicitly pt+1 (θ )/pt (θ ) and the above characterization of the price level condition, (a) and (b) for t < T are equivalent to

[1 − φπ1 (θl )][1 − φµ(θl )] [1 − φπ1 (θh )][1 − φµ(θh )] > χ(φ, θh , 1) χ(φ, θl , 1)

(a’)

χ(φ, θl , t + 1) χ(φ, θh , t + 1) > (1 + θ h ) , t 1 − φµ(θl ) χ(φ, θh , 1) (1 + θ l ) > (1 + θ h )

χ(φ, θh , t + 1) χ(φ, θh , t)

t ≤ T−1

(a”)

(b”)

It will be shown below that we want to set φ ∼ = π1 (θh ) . Then χ(φ, θh , t + 1) ∼ = [φµ(θh )]−1 χ(φ, θh , t) so that the right-hand side of (b00 ) is approximately t independent. Therefore, consider (b00 ) for t = 1, and rewrite ( a00 ) .

[1 − φµ(θl )]−1 >

χ(φ, θh , 1) [1 − φπ1 (θh )][1 − φµ(θh )]

(1 + θ l ) > (1 + θ h )

χ(φ, θh , 2) χ(φ, θh , 1)

(a”’)

(b”’)

Maintain (1 − φµ(θh ) and 1 − φπ1 (θh ) positive; multiply the leftand right-hand sides of ( a000 ) and (b000 ) to get, after some manipulation  1 + θ  1 − φµ(θ )  χ(φ, θh , 2) l h > (c) 1 + θh 1 − φµ(θl ) 1 − φπ1 (θh ) The left-hand side of (c) is the product of two terms, each of which is easily seen to be slightly less than unity for small θh − θl > 0. Therefore, the left-hand side of (c) is 1 − ε for small ε > 0. Write the right-hand side as 1+

[π1 (θh ) − µ(θh )]φT −t+1 µ(θh )T −t = 1+δ 1 − φπ1 (θh )

196

Chapter 5: Some Unpleasant Monetarist Arithmetic

By the choice of θl , π1 (θl ) = µ(θl ) . Therefore, θh > θl implies π1 (θh ) < µ(θh ) . Hence δ < 0 and can be made arbitrarily large in magnitude when φ approaches arbitrarily close to π1 (θh )−1 from below. This will satisfy condition (c). The condition for real and positive π1 given b( T ) (for d = 0) is b( T ) ≤ γ2/2 (1 + n)/( R − n){1/φ + 1/(1 + n) − 2[φ(1 + n)]−1 } . Values for b( T ) that are “too low” will imply negative b1 . To guarantee strictly positive b(1) , set (b( T ) as high as desired by increasing both γ2 and γ1 , keeping φ = γ2 /γ1 at the chosen value. In recapitulation, the method involves carrying out the steps above in reverse order. Choose γ2 /γ1 sufficiently close to 1; set γ2 so that maximum b( T ) appears “high enough.” Calculate θl and work backward from b( T ) to b(1) . Then, using this value of b(1) , set θh so that θh − θl is small and positive.

6 Interpreting the Reagan Deficits Government Budget Balance Figures on government deficits are difficult to interpret because the economically relevant budget constraint is an intertemporal one. As such, it restricts the present value of a sequence of government deficits but not the size of deficits for particular years or even for long strings of years. For any observed string of government deficits, there always exists a string of prospective future surpluses that renders the budget in balance in the present value sense. By alluding to prospects for future government surpluses, anyone can therefore assert that a record of observed deficits is consistent with maintaining sound government credit and a stable government currency. Several years of big deficits by themselves therefore fail to indicate that the entire sequence of government budgets is out of balance. This fact opens recent deficit figures for the United States to alternative interpretations, some hopeful, others foretelling doom. This chapter tries to rationalize the large net-of-interest deficits in the federal budget of the United States that have marked the Reagan administration. I take for granted that the recent deficits are temporary and that they foretell future government surpluses. I spend no time discussing the view that the deficits are simply a mistake, a failure of policy, or the result of shortsightedness or ignorance of the intertemporal government budget constraint. Instead I focus on alternative interpretations of recent events that are consistent with George Stigler’s vision that all agents in a social system are rational and purposeful. I seek to explain the fiscal and monetary actions observed during the Reagan administration as reflecting the optimal decisions of government policymakers. There will be one equation in the background of my discussion, one whose validity is granted by all competing theories of macroeconomics. This equation is the intertemporal government budget 197

198

Chapter 6: Interpreting the Reagan Deficits

constraint. It states that, at any moment, the value of interestbearing government debt is equal to the sum of two terms: the present value of future government surpluses net of interest, and the present value of future government revenues from printing currency (seigniorage revenues). Presumably, the government deficit net of interest and the revenues from currency creation are controlled by separate and independent agencies of the US government. However, in a recurrent and strategic sense, independence is not feasible. Because revenues from printing currency are one component of the government budget constraint, the notion that there can be truly independent monetary and fiscal authorities is a myth. Arithmetic makes the strategies of the monetary and fiscal authorities interdependent.1 Classic recommendations for the conduct of monetary policy, such as Friedman’s (1960) k-percent growth rule for currency or the gold standard, are well understood as coordination rules for monetary and fiscal policy. For these coordination rules to be feasible, the intertemporal government budget restraint must be respected. Throughout this chapter, I will always assume that a version of Friedman’s k-percent coordination rule (one with a small value of k) is followed. I seek to interpret the following observations about monetary and fiscal policy during the Reagan years: a string of large annual net-of-interest government deficits accompanied by a monetary policy stance that has been tight, especially before February 1985, and even more so before August 1982. I take as indicators of tight monetary policy high real rates of interest on US government debt and pretax yields that exceed the growth rate of the economy. (Real rates of this magnitude imply that the interest-bearing government debt is growing relative to the size of the economy unless the net-of-interest government budget is in sufficient surplus.) I take for granted that the string of net-of-interest government deficits and tight monetary policies (low rates of seigniorage production) cannot both continue forever, simply because they would violate the intertemporal government budget constraint.2 1 Sargent

and Wallace (1981) describe some of the implications of the interdependence between monetary and fiscal policy. 2 This is Sargent and Wallace’s unpleasant monetarist arithmetic.

Barro Tax Smoothing

199

I shall describe two rationalizations of recent observations on government policy, each of which is consistent with the government budget constraint, under the hypothesis of “rational expectations” and the presumption that the government as a whole is committed to a monetary regime with low inflation rates over the long haul. The last stipulation is equivalent to an assumption that the present value of seigniorage in the government budget restraint is taken for granted to be small.

Barro Tax Smoothing The first rationalization is constructed by applying the optimal tax smoothing model of Robert Barro (1979). I assume that the monetary authorities are committed to supplying little or no seigniorage, and that this is beyond dispute. It follows therefore that the present value of seigniorage is small. Because of the government budget constraint, the net-of-interest government budget must be in surplus in present value by an amount equal to the current value of interest-bearing government debt. How can this implication be reconciled with the string of large net-of-interest deficits observed during the Reagan administration? Barro’s model supplies a possible answer. Barro’s model of tax smoothing can be thought of as a reinterpretation of Milton Friedman’s (1956) model of permanent income as developed by Robert E. Hall (1978).3 The permanent income model of consumption confronts a consumer with an exogenous process for labor income and a constant real rate of return on savings. The consumer has preferences over a long horizon that can be represented as a discounted sum of a current period utility function that depends on current consumption alone. That is, preferences are additively time-separable, and the utility function is concave in current consumption. Hall shows that for a discount factor equalling the reciprocal of the gross interest rate on assets, the marginal utility of consumption follows a random walk. To the extent that the marginal utility 3 See

Sargent (1987, chapters XII and XIII) for a formal technical comparison of the features of Hall’s and Barro’s models.

200

Chapter 6: Interpreting the Reagan Deficits

of consumption is approximately linear in consumption, consumption itself may approximately follow a random walk. As Hall has stressed, for any income process, no matter how unsmooth, the model predicts that consumption is approximately a random walk. This means that at every point in time, future consumption is expected to be approximately constant. Hall’s model precisely represents the consumption smoothing idea present in Friedman’s original work on the consumption function. A possibly very unsmooth labor income process is used to support a consumption process whose future is expected at each point in time to be perfectly smooth. Borrowing and lending are used to convert an unsmooth income path into a smooth consumption path. At any time, the mean of the consumption path is set so that the present value of consumption equals the present value of labor income plus initial nonhuman assets. Barro can be regarded as having changed the names of the variables in Hall’s model and applied them to the government. In place of the household budget constraint, Barro uses the government budget constraint. What was the exogenous labor income process in the Friedman model becomes an exogenous process for government purchases. What was consumption in the household budget constraint becomes total tax collections in Barro’s model. What were household assets become the stock of interest-bearing government debt. The interest rate confronting the household in Hall’s model becomes the interest rate at which the government can borrow and lend in Barro’s model. The intertemporal version of the re-interpreted budget constraint is precisely the intertemporal government budget constraint described above, with seigniorage assumed to have a present value of zero. In place of the preference function used by Hall, Barro uses an additively time-separable loss function measuring distortions from taxing. The current period loss function is convex in total tax collections. Barro poses the problem of a government that faces an exogenous and given stochastic process for government purchases and that chooses a tax strategy to minimize the expected discounted value of losses from tax distortions. In mathematical terms, this model is equivalent to Hall’s consumption model, with the change

Barro Tax Smoothing

201

of variables described above. It follows that the model gives the result that optimally, total tax collections should follow a random walk. That is, in the face of an unsmooth government expenditure stream, tax collections should be smoothed. In this way, distortions are allocated over time in a way to minimize the present value of the distortion. We note that this result depends critically on the feature of the loss function that the distortion at time t is assumed to depend only on total tax collections at t , and not on future tax collections, as would occur in a model in which private agents are speculating about future government tax collections. In Barro’s model, expected future tax collections are set equal to current tax collections, with current tax collections set to satisfy the intertemporal government budget constraint. The Barro model can be used to rationalize the observed deficits of the Reagan administration as part of an optimal tax smoothing response to an “innovation” about the present value of government expenditures that arrived coincidentally with Reagan’s election. Assume that the election of Reagan signaled a downward revision in the size of the US government, as measured by the expected present value of federal expenditures. Assume further that the path of reductions, compared to the path that could have been expected prior to Reagan, was skewed toward the future or “back-loaded.” That is, the election of Reagan meant reductions in the government expenditures could be expected to take place gradually over time, with larger reductions in the future than in the present. Given such a change in the path of expected government expenditures at the start of the Reagan administration, Barro’s tax smoothing model predicts that the (optimal) response of the government would be an immediate permanent reduction of tax collections, relative to the pre-Reagan path. The consequence of these immediate reductions would be a string of deficits while expenditures remained high, to be followed by a string of net-of-interest government surpluses after the reductions in expected government expenditures had been realized. According to this scenario, there is nothing pathological about the large deficits we have observed. Instead, they are to be interpreted as the result of optimal tax smoothing by the federal

202

Chapter 6: Interpreting the Reagan Deficits

government. Note that Barro’s argument implies that the Reagan administration should have tried for a 25 percent reduction in tax rates at one shot, rather than the 5-10-10 phasing in over three years embodied in the Kemp-Roth tax legislation. Barro’s model implies that the large deficits observed pose no inflationary threat because they pose no danger of being monetized subsequently. The fact that the interest-bearing US government debt has grown under Reagan is merely a signal that the budget will swing into surplus at some time in the future, and that government expenditures are destined to fall relative to their pre-Reagan path. The scenario described depends critically on a controversial aspect of Barro’s specification of the function measuring the current loss from tax distortions in the government’s objective function. In particular, Barro specifies that the current distortion at time t depends only on current tax collections, and is not a function of the public’s expectation of future taxes set by the government. This feature is critical in giving rise to the random walk characterization of taxes, which is at the heart of our interpretation of the Reagan deficits. It is also crucial in rendering Barro’s solution of the optimal tax problem time-consistent.4 However, in models in which there is capital, either physical or human, the current distortion from taxation at time t typically depends in part on people’s expectations about future taxes. In making investment decisions, people look and respond to the government’s strategy for taxing in the future. Expectations about future taxes therefore distort private decisions.5 Such distortions would alter Barro’s loss function in a way that would make it suboptimal if tax collections followed a random walk. It would also render the solution of the optimal tax problem time-inconsistent. A sequence of administrations differing over time would therefore 4 See

Kydland and Prescott (1977) for a discussion of the time-inconsistency problem in macroeconomics. See Lucas and Stokey (1983) for a study of dynamic inconsistency in the context of an optimal tax smoothing model that shares many features with Barro’s model. 5 See Sargent’s (1987) chapter on dynamic optimal taxation for an extended example exploring the time-inconsistency phenomenon created by the responsiveness of investment to anticipations of future taxes.

Barro Tax Smoothing

203

be unable to carry out any solution. As it turns out, when optimal tax problems are solved for systems with physical or human capital, the optimal tax strategy usually is far from a random walk prescription. Usually high taxes are called for in the present, to be followed by lower taxes in the future. Since high taxes now are imposed on existing capital and existing capital is perfectly inelastic in supply, the taxes take on a lump sum character. As a result, such current taxes should be imposed heavily to minimize the present value of distortions. Anticipated future taxes, in contrast, do distort investment decisions and therefore future values of capital. As a result, they should be used sparingly. The asymmetry in attitude toward current and future taxes on capital is at the heart of the time inconsistency of the solution, as well as of the suboptimality of tax smoothing. In summary, by restricting the nature of the function that is assumed to measure the losses from the distortions that taxes impose, Barro was able to create a model calling for “tax-smoothing.” By tax-smoothing, he meant that, even if government expenditures were expected to vary in the future, it would be optimal for consumers to expect taxes to remain unchanged. Applied to the current situation in the United States (supposing that the election of Ronald Reagan signaled that government expenditures would fall relative to their pre-Reagan path), the model rationalizes a string of deficits like the one we have experienced. Not only does the model “explain” those deficits, but it also implies that they are not signs of a “problem.” Rather, the current deficits are simply a “signal” of future reductions in the path of government expenditures. This application of Barro’s model is attractive because it explains many aspects of the current situation and supports a sanguine interpretation of recent US deficits. However, such an application is not beyond criticism for reasons alluded to above. In particular, the restrictions on the loss function measuring distortions in Barro’s model are very strong ones. Indeed, the restrictions suppress any “supply-side” effects flowing from expectations about future taxes to current decisions.

204

Chapter 6: Interpreting the Reagan Deficits

Wallace’s Game of Chicken I now turn to an alternative interpretation, one due to Neil Wallace. Wallace’s interpretation hinges on the observation that economic policymaking in the United States is decentralized over a variety of agencies, and that government expenditures cannot be reduced without a struggle among those agencies. Wallace’s explanation makes the deficit an instrument in that struggle. Wallace’s interpretation assumes that the “game” played by government policy authorities has a different structure from that assumed by Barro. In particular, Wallace has interpreted monetary and fiscal policy during the Reagan administration as unfolding like a game of chicken among distinct branches of government with different preferences about the size of the US federal government.6 In this game of chicken, reducing the present value of government expenditures is not a given, but instead is the objective of one of the participants in the game. This objective, in turn, is actually opposed by another player. The players’ weapons consist of their separate authorities to set paths for government expenditures, tax collections, and currency creation. Using Wallace’s analogy, the Reagan administration plays the game for the purpose of reducing the present value of government expenditures—an objective whose attainment Barro’s explanation took for granted. The game of chicken is played among decentralized branches of government that control separate elements of the government budget constraint. There is a tax authority, whose role I shall assign to the Reagan administration, and whose responsibility is to select a stochastic process for tax collections. There is a government expenditure authority, here assigned to Congress, that determines the stream of government expenditures. Finally, there is a central bank (the Federal Reserve System) that determines a time stream of emissions of currency and thereby controls the present value of seigniorage that appears in the government budget constraint. While these three players, the tax authority, the expenditure authority, and the monetary authority, must coordinate their strategies because of the arithmetic of the intertemporal government 6 Wallace

advanced his ideas orally in March 1981 during discussions that later led to our co-authoring “Unpleasant Monetarist Arithmetic” (1981).

Wallace’s Game of Chicken

205

budget constraint, they are not forced to do so on a day-to-day basis by any formal legal or constitutional mechanism. The coordination of monetary and fiscal policy in the United States is not governed by a set of well-understood, recurrently applied, or explicit rules. Instead, policy actions seem to emerge from a process that is decentralized across institutions (Congress, President, and Federal Reserve) and spread over time through a succession of administrations and personalities. This decentralization opens the way to the playing of what Neil Wallace has characterized as a game of chicken. In the game of chicken being played under the Reagan administration, the tax and monetary authorities jointly desire a reduction in the present value of government expenditures (something they do not control) as well as a stable price level. The expenditure authority is assumed to desire a larger government in the sense of a larger expected present value of government expenditures than does the tax authority. To achieve its objective, suppose that the tax authority plays the game as follows. It achieves a once-and-for-all reduction in tax collections that reduces the present value of tax collections relative to its initial value. The tax authority (the president) then encourages the central bank to adhere to a k-percent rule for the monetary base for the indefinite future. Such a monetary policy implies that the central bank withholds seigniorage revenues from the government. Given these “plays” by the president and the Federal Reserve, the only plays open to the government expenditure authority are ones that capitulate to the president’s objective and that reduce the present value of government expenditures by an amount commensurate with the reduction in the present value of tax collections. As long as the president and the Federal Reserve adhere to their strategies, the stream of government expenditures must be reduced because of the arithmetic of the government budget restraint. Congress may, however, reason as follows. It can simply refuse to reduce the present value of government expenditures despite the tax reduction engineered by the tax authority. Then, as long as the monetary authority refuses to monetize interest-bearing government debt, the arithmetic of the government budget constraint requires that the tax authority eventually reverse itself and raise

206

Chapter 6: Interpreting the Reagan Deficits

taxes by an amount that makes the present value of taxes equal to the present value of expenditures plus whatever debt has accumulated. If the monetary authority and Congress both refuse to chicken out, then the arithmetic of the budget constraint asserts that the only feasible thing for the tax authority to do is to raise taxes. Of course, it is feasible that neither the tax authority nor the expenditure authority will chicken out. In that case, the central bank would be forced to chicken out by departing from its k-percent rule and generating substantial seigniorage. By monetizing the debt, the central bank would permit government expenditures to exceed tax collections in present value terms, albeit at the cost of generating inflation. While the authorities are playing this game of chicken, we would observe large net-of-interest government deficits, low rates of monetization of government debt (low growth rates for the monetary base), and maybe also high real interest rates on government debt. The result of high real interest rates on government debt and the net-of-interest government deficit is a growing real value of the stock of interest-bearing government debt. The rising stock of this debt would be a signal that the game is not yet over, in the sense that there has been insufficient capitulation. In the United States today, the real stock of interest-bearing federal debt continues to grow in relation to GNP. The game of chicken interpretation has a number of merits as an explanation of these events. While it is tempting to criticize resorting to a game of chicken as an inferior way to run a government, such criticism ignores the extensive decentralization across time and institutions that exists under US government. Given the limited power assigned to the presidency for economic policy in general and government expenditures in particular, resorting to the game of chicken may be the best method available for achieving the preference, reflected in Reagan’s policies, for reducing the size of the US government. Several important macroeconomic policy events during the Reagan years bear interpretations in terms of one party or another in our game of chickening out. The Federal Reserve partly chickened out on two occasions, one in August 1982, and another at

Conclusion

207

the start of 1985. Each time, the Fed was responding to outside pressures that were partly consequences of, and which in turn fed back upon, the original game of chicken. In August 1982, the Fed substantially eased monetary policy, increasing the growth of narrow monetary aggregates and driving real interest rates downward. These actions were in large part responses to the international debt crisis that coincided with the high real interest rates associated with the game of chicken that dominated US macropolicy. The Fed eased its monetary policy specifically in response to the Mexican crisis and the threat it posed for US financial stability. Such concerns limit the Fed’s ability to continue to play a tight monetary policy in the face of continued net-of-interest US government budget deficits. The second partial capitulation by the Fed was associated with a move starting in early 1985 to lower real interest rates in the United States as a device to drive down the value of the dollar. The Fed was responding to the increasing strength of protectionist pressures in the United States that were themselves responses to the US trade deficit, which was, in turn, one consequence of the string of government deficits associated with the game of chicken.

Conclusion While they differ in a number of respects, our two alternative rationalizations of the Reagan deficits share the premise that, compared to the pre-Reagan path, US federal expenditures are destined to fall. In Barro’s model, the fall in the path of expenditures occurs exogenously, and precipitates the Reagan deficits via optimal tax smoothing. In Wallace’s view, the fall in the path of federal expenditures relative to the pre-Reagan path is an outcome of (or “reward to”) the game of “chicken,” with an endless string of prospective budget deficits being the stick by which the president and Federal Reserve persuade a reluctant Congress to reduce federal expenditures. According to both explanations, large net-of-interest deficits are signals of prospective surpluses to be achieved via reductions in expenditure. Each rationalization relies on the looseness of the intertemporal

208

Chapter 6: Interpreting the Reagan Deficits

government budget constraint to which I referred at the beginning of the paper. A long string of large deficits is consistent with budget balance provided that sufficient surpluses occur later. We have rationalized the large Reagan deficits by appealing to the idea that they are temporary and bound to be replaced by surpluses long before they damage the economy. Some readers may find the entire endeavor of rationalizing the large Reagan deficits to be misplaced. Perhaps it is farfetched to rationalize deficits in the ways that we have, and better to regard them simply as reflecting shortsighted mistakes that the United States is bound to pay for in the future via more inflation, increased financial fragility, or higher taxes. Nevertheless, to reach the conclusion that the US deficits of the last decade were mistakes, one must first understand the arguments that could rationalize them.

Appendix We describe linear-quadratic versions of the models of Hall (1978) and Barro (1979). The exposition is designed to highlight parallels between the two models. For further details and implications of the models, see Sargent (1987). H ALL’ S M ODEL A representative consumer chooses a contingency plan for (ct )∞ t =0 to maximize ∞

E0 ∑ βt u(ct ), 0 < β < 1

(H1)

t =0

where ct is consumption at t, Et is the mathematical expectation operator conditioned on information known to the consumer at t , and u is a one-period utility function given by u ( c t ) = u0 + u1 c t −

u2 2 c 2 t

(H2)

where u0 , u1 , u2 > 0. The objective ( H1) is maximized with respect to (ct , At+1 )∞ t=0 subject to the sequence of budget constraints At+1 = R[ At + yt − ct ], t = 0, 1, . . .

(H3)

Appendix

209

where A0 is given and where R ≥ 1 is the gross rate of return on savings between periods t and t + 1, At is assets (indebtedness, if negative), and yt is noncapital or labor income at t . We assume that yt is a given stochastic process, outside the agent’s control. We assume that βR = 1. We assume that (yt ) is a stochastic process that satisfies lim j→∞ Et β j yt+ j = 0 for all t . We impose upon assets the condition E0 At ≥ M > −∞

(H4)

for all t , which rules out a strategy of larger and larger borrowing to support bliss consumption. Subject to ( H4) , the solution of the difference equation ( H3) is the following intertemporal version of the budget constraint:  j ∞  j 1 1 ∑ R Et ct+ j = At + ∑ R Et yt+ j j =0 j =0 ∞

(H5)

Equation ( H5) states that the expected present value of consumption equals the expected present value of labor income plus the value of initial assets At . Hall shows that a first order necessary condition associated with the problem of maximizing ( H1) subject to ( H5) is Et u0 (ct+1 ) = ( βR)−1 u0 (ct ) . Since we have set βR = 1, this becomes Et u0 (ct+1 ) = u0 (ct ) . With utility given by ( H2) , this in turn implies Et ct+1 = ct , so that consumption is a random walk. B ARRO ’ S M ODEL A government chooses a tax collection sequence (τt )∞ t=0 to maximize ∞

− E ∑ βt L(τ1 ), 0 < β < 1

(B1)

t =0

where τ1 are total tax revenues, and L is the loss function v2 L(τt ) = v1 τt + τt2 2

(B2)

The maximization is carried out subject to the sequence of government budget constraints

210

Chapter 6: Interpreting the Reagan Deficits

Bt+1 = R[ Bt + gt − τt ],

t = 0, 1, . . .

(B3)

B0 given, where Bt is the initial level of government interestbearing debt, R ≥ 1 is the gross real rate of return on government debt, and gt is government expenditures at t . Government expenditures are taken to be an exogenous stochastic process that satisfies lim j→∞ β j Et gt+ j = 0 for all t . We impose the boundary condition Bt ≤ M < +∞

(B4)

which rules out a strategy of “never tax, always borrow more.” Subject to ( B4) , the solution of the difference equation ( B3) is the intertemporal government budget constraint ∞

∞

j =0

j =0

Et ∑ R− j τt+ j = Bt + Et ∑ R− j gt+ j

(B5)

Equation ( B5) states that the expected present value of tax collections equals the sum of the current value of debt plus the expected present value of government expenditures. The government is assumed to maximize ( B1) subject to ( B5) by choosing a strategy for setting (τt )∞ t=0 , taking B0 and the stochastic process for gt as given. Assume as in Hall’s model that βR = 1. Mathematically, Barro’s model is equivalent to Hall’s. Simply replace ct in Hall’s model with τt in Barro’s, yt in Hall’s model with gt in Barro’s, At in Hall’s with Bt in Barro’s, (u0 , u1 , u2 ) in Hall’s model with (0, −v1 , v2 ) in Barro’s. It immediately follows from Hall’s results that the optimal tax collection strategy in Barro’s model satisfies Et τt+1 = τ1 , so that tax collections are a random walk.

7 Speculations about the Speculation against the Hong Kong Dollar with David T. Beers and Neil Wallace

In this chapter, we attempt to explain the Hong Kong dollar’s recent substantial depreciation, or its loss in value in terms of foreign exchange. This task is a challenge mainly because of the unusual monetary policy recently espoused by the government of Hong Kong. From 1974 until October 1983, the officially professed policy allowed both the quantity and the exchange value of Hong Kong currency to be determined by market forces (or to float). The subject is also interesting because current events in Hong Kong are so much affected by expectations, expectations about what will happen after 1997, the year that Britain’s lease on most of the colony’s territory expires. The recent depreciation was indeed substantial. Between mid1980 and October 1983, when the government announced a new policy, the US dollar value of the Hong Kong dollar fell about 33 percent. During that period, there also occurred substantial declines (even in terms of local currency) in all the value of assets located in Hong Kong—Hong Kong real estate and common stock in Hong Kong companies. All of these declines coincided with a generally unfavorable reassessment of the prospects for maintaining the political status quo beyond 1997. The unfavorable reassessment easily accounts for a decline in real estate and stock market values. Prospective events after 1997 can reasonably be expected to reduce the stream of returns flowing to the owners of real estate and shares in Hong Kong so that their value now would fall. The fall in the Hong Kong dollar has also been widely attributed to unfavorable speculation about 1997, apparently on the basis of a vague notion that the Hong Kong dollar is somehow a claim against the wealth of the current

211

212

Chapter 7: Speculations about the Speculation

residents of Hong Kong. Closer scrutiny shows that movements in the value of the Hong Kong dollar are not directly linked with speculation about 1997. This is because the events of 1997 do not adversely affect interest-bearing foreign assets held as backing for the currency by the Hong Kong government. Thus, distinct lines of reasoning seem to be needed to account for the fall of Hong Kong real estate and stock prices, on the one hand, and the fall of the foreign exchange value of the Hong Kong dollar, on the other. We explain the depreciation of the Hong Kong dollar as the result of a chain of influences involving the Hong Kong government and speculation about the events of 1997. The existence of this chain of influences is supported by indirect evidence that the Hong Kong government actually departed from its pre–October 1983 official float policy. In particular, we interpret the depreciation of the Hong Kong dollar as having been welcomed and supported by the government in order to help private financial intermediaries in Hong Kong face the substantial declines in the real value of Hong Kong real estate and other real assets. Our case for this interpretation rests heavily on how we think the Hong Kong monetary system would have functioned if the government had actually adhered to its float policy. We argue that under that policy market fundamentals alone could not have determined an equilibrium exchange rate for the Hong Kong dollar and, in particular, did not propel that rate downward. Our circumstantial evidence supporting the case that the government departed from its official policy in order to promote the depreciation of the Hong Kong dollar is supported by an analysis of who stood to gain and who stood to lose from such a depreciation.1

A Look at Hong Kong’s Monetary System Monetary and fiscal affairs in Hong Kong are now managed by a single institution, known as the Exchange Fund. (Though originally 1 Except

for minor editorial changes, most of this chapter was written during the summer of 1983. The last section, an epilogue, was written after October 15, 1983, when the Hong Kong authorities announced and implemented a shift to a policy of pegging the value of the Hong Kong dollar to that of the US dollar.

A Look at Hong Kong’s Monetary System

213

Table 7.1 T HE E XCHANGE F UND ’ S B ALANCE S HEET Branch Monetary

Fiscal

Assets

Liabilities

Foreign exchange

Certificates of indebtedness

Hong Kong bank deposits denominated in Hong Kong dollars

Debt certificates

Foreign exchange

Net worth

Hong Kong bank deposits denominated in Hong Kong dollars Debt certificates

just Hong Kong’s monetary institution, it took on most of the treasury’s functions in 1977.) The types of items on the Exchange Fund’s balance sheet, divided into parts attributable to the monetary branch and the fiscal branch, are shown in Table 7.1. The certificates of indebtedness are non-interest-bearing instruments held by private banks as backing for currency. The debt certificates are an internal bookkeeping device between the monetary and fiscal branches of the Exchange Fund, which completely net out in the consolidated balance sheet. They seem to be a device for crediting the revenues accumulated from currency issue (the seigniorage) to the fiscal branch. The Exchange Fund issues all of Hong Kong’s coins, but two large private banks issue most of the paper currency, known as the Hong Kong dollar (HK$). These banks are the Hong Kong and Shanghai Bank, which issues from 80 to 90 percent of the paper currency, and the Chartered Bank, which issues most of the rest. These private banks issue the currency entirely at their own discretion, but they do not profit greatly from this privilege. Since the banks’ names appear prominently on the notes, they get some prestige and free advertising from the privilege. However, as already noted, the government collects all of the seigniorage. It does this by requiring the banks to hold, behind all currency issues, 100 percent reserves of the non-interest-bearing certificates of indebtedness

214

Chapter 7: Speculations about the Speculation

mentioned above as liabilities of the Exchange Fund.2 The banks must buy these certificates of indebtedness from the Exchange Fund, which holds as assets behind these certificates a portfolio of interest-bearing assets denominated in foreign currencies and interest-bearing deposits in accounts in Hong Kong banks. These assets can thus be thought of as ultimately backing the Hong Kong dollar. The regulation governing certificates of indebtedness states that they are redeemable at the option of the financial secretary. The asset effects of currency issues on the Exchange Fund and the banks are illustrated in Table 7.2. Although only two banks issue notes, there is relatively free entry into other aspects of banking in Hong Kong. Furthermore, there are no reserve requirements against deposits and no government-supplied deposit insurance. There is a cartel, the Hong Kong Association of Banks, which sets interest rates.

The Official Float Policy The system just described contains vestiges of Hong Kong’s pre– 1972 system, which was a British Currency Board System. In that system, the Exchange Fund pegged the Hong Kong dollar to the British pound. Behind certificates of indebtedness the Exchange Fund held 100 percent reserves in interest-bearing British pound instruments. After 1972, for about two years, the Hong Kong dollar was pegged to the US dollar. Thereafter (until October 1983) it was allowed to float. During the float period, the Exchange Fund held a portfolio of interest-bearing instruments denominated in a variety of foreign currencies and of interest-bearing deposits in Hong Kong banks. According to the official policy, the Exchange Fund passively supplied any amount of certificates of indebtedness that the private banks requested in exchange for foreign currencies at market rates of exchange. In effect, then, the Exchange Fund was willing to buy and sell Hong Kong dollars for foreign currencies at market rates of exchange. During the float period, the Exchange Fund mostly kept secret 2 The

government compensates the banks for the cost of issuing and maintaining the stock of currency.

215

An Indeterminate Exchange Rate?

Table 7.2 B ALANCE S HEET E FFECTS OF I SSUING HK$ 100 Assets Exchange Fund

Private Hong Kong banks

Liabilities

Foreign exchange

85

Hong Kong bank deposits denominated in Hong Kong dollars

15

Certificates of indebtedness

100

Certificates of indebtedness

100

Hong Kong dollar notes

100

its portfolio composition and its strategy for shifting the composition of the portfolio between foreign currencies and domestic deposits. (The exception is that the amount held in British pound sterling was revealed.) However, the guiding principle of the Exchange Fund’s open-market strategy seems to have been to hold all or most of its portfolio in foreign currencies.3 When the Hong Kong dollar was experiencing temporary weakness, the Exchange Fund might move temporarily into deposits denominated in Hong Kong dollars. Such moves were exceptional and temporary, though. The Exchange Fund professed no intention and acknowledged no ability to influence the basic direction of exchange rates. According to government officials, the exchange rate was determined by fundamental market forces that the Exchange Fund should not, and could not, oppose for long.

An Indeterminate Exchange Rate? We strongly suspect that the Hong Kong government did not actually follow its official float policy. A major reason for this suspicion is that if the policy had been followed, then the exchange rate for the Hong Kong dollar would have been indeterminate. By indeterminate, we mean that there would have been a wide range of values of foreign exchange rates for the Hong Kong dollar that would have cleared markets. This indeterminacy is a consequence of the 3 Here

and elsewhere in this chapter, our descriptions of the Hong Kong government’s strategy rely heavily on Sargent’s discussions with various Hong Kong economists in June 1983.

216

Chapter 7: Speculations about the Speculation

lack of any anchor in a system in which both the exchange rate and the quantity of Hong Kong dollars float. It results under any theory of the demand for Hong Kong dollars. The indeterminacy is a product of the responses required of the Exchange Fund under the official policy. To establish that result, let us first examine the options available to the Hong Kong government if it were to have pegged the exchange rate for its currency to one or several foreign currencies. A government doing that simply announces that it is willing to buy and sell unlimited amounts of its currency at the chosen exchange rate. Then it must be prepared to cope with a run on its currency, or an attempt by holders of the domestic currency to convert their holdings into foreign currency. The government need not be concerned about high demands for its own currency, since it can always provide more if needed. This asymmetry produces a range of values at which it is feasible to peg the exchange rate. To simplify the exposition, suppose that there is only one foreign currency and that at some time a monetary authority has a balance sheet consisting of f units of foreign currency, which it possesses as assets, and h units of its own (home) currency outstanding, which are its liabilities. Starting in this situation, it is feasible for the authority to fix the exchange rate, the value of its own currency in terms of foreign currency, at any value between zero and the ratio f /h . That is because, with the exchange rate in this range, the foreign exchange value of its h units of home currency remains less than or equal to f (since (h)( f /h) = f ). In other words, with the rate in this range, no exchange that the authority could be called on to make would more than exhaust its holdings of foreign currency.4 (Notice, by the way, that the 4 In

this pure system, the monetary authority can be regarded as acting as a warehouse for foreign currency, always having at least enough foreign-denominated assets to convert its total outstanding stock of notes. In practice, however, many countries’ monetary authorities have managed to peg a rate of exchange of domestic currency for foreign currency while holding foreign exchange reserves that are smaller in value than the stock of outstanding domestic notes. Such a system can be administered if the monetary authority has a commitment from the fiscal authority to run prospective government surpluses sufficient to cover the obligations of the monetary authority in the event of a run. We assume that the Hong Kong system is a pure one, which is to say that the monetary branch of

An Indeterminate Exchange Rate?

217

net worth of the monetary authority [assets less liabilities] varies inversely with the exchange rate, being equal to f when the exchange rate is zero [ f − (h)(0) = f ] and being equal to zero when the exchange rate is zero f /h[ f − (h)( f /h) = 0] .) Evidently, any such fixed-rate policy is effective in the sense that the market rate must be the fixed rate. Under the fixed-rate policy, the monetary authority would supply the market with any amount of home currency that is demanded at the fixed rate; in other words, at that rate, supply would be perfectly elastic. Regardless of the demand for home currency, then, the exchange rate would be the one announced. Now consider matters under a strict version of Hong Kong’s official float policy. Suppose again that the authority starts out holding f units of foreign currency with h units of home currency outstanding. The monetary authority’s policy is now to buy and sell any amount of its currency at the exchange rate that the market determines. However, what rate would the market determine? Evidently, under this policy, it could be any rate between zero and f /h . To establish that any such rate could be an equilibrium, we only need to verify that the home currency’s supply and demand would be equal at that rate. Consider any particular rate in the interval [0, f /h] . Under the float policy, supply at that rate would be perfectly elastic as well because, according to the policy, the authority is willing to buy and sell unlimited amounts at the market rate. At the particular rate, then, regardless of the nature of the demand for home currency, demand would be satisfied. Therefore, the posited rate—any rate between zero and f /h —would be an equilibrium, or ruling, exchange rate under the official policy. To demonstrate in this way that under Hong Kong’s official float policy the exchange rate had to be indeterminate cannot be the end of things. For somehow the exchange rate was being determined. Therefore, our analysis is, at best, incomplete. In principle, our analysis could be regarded as defective either because we have mischaracterized how things are supposed to have worked under the Hong Kong Exchange Fund operates independently of the fiscal branch and cannot expect future government surpluses or deficits either to augment or to impair the funds available to back the currency.

218

Chapter 7: Speculations about the Speculation

the official policy or else because the Exchange Fund actually pursued a policy different from the officially professed one. We favor the latter possibility. We think that the Hong Kong dollar’s exchange rate was determined throughout the float period by a series of discrete deviations from the official policy. In practice, the policy is likely to have more closely resembled a fixed-rate policy than a strict version of the official policy. In our view, the observed exchange rate was determined by signals given by the Exchange Fund, signals that in effect revealed the exchange rate at which the authority was willing to buy and sell Hong Kong dollars. This was feasible since the Exchange Fund was able to supply unlimited quantities of Hong Kong dollars and since it was continually engaging in transactions with the dealers who announced exchange rates.

A Timely Depreciation We have no direct evidence that the Exchange Fund ever actually deviated from its official float policy. Our evidence is circumstantial and consists partly of the fact that in practice the exchange rate was somehow tied down even though, under the official policy, it should not have been. Our view is also based on the fact that the recent depreciation in the Hong Kong dollar was quite timely for the Hong Kong government. It came, that is, just when the government would have welcomed it. Seeing why that is so requires a look at who gains and who loses from the depreciation of this currency. Changes in the exchange rate of the Hong Kong dollar affect the distribution of wealth among issuers and holders of Hong Kong dollar-denominated liabilities. As mentioned above, since the Exchange Fund’s portfolio consists primarily of assets denominated in foreign currency while its liabilities are in Hong Kong dollars, its net worth varies inversely with the exchange value of the Hong Kong dollar (the lower the exchange rate, the greater the government’s net worth). Offsetting this, in terms of a consolidated balance sheet, everyone else has a net worth that varies directly with the exchange value of the Hong Kong dollar. However, the effects on any particular

A Timely Depreciation

219

asset holder depend on the holder’s portfolio composition, namely, on the extent to which the holder is unhedged (exposed to risk) with respect to the exchange rate and the direction in which the holder is unhedged. It is primarily in terms related to such effects that we will explain why the authority might have welcomed a depreciation of the Hong Kong dollar. We will do this by examining the balance sheet of a hypothetical Hong Kong financial firm as it experiences what some real Hong Kong firms may actually have experienced: first a substantial decline in real asset prices and then that decline joined by a substantial decline in the value of the Hong Kong dollar. Our financial firm has some assets in the form of loans, denominated in Hong Kong dollars, that have as collateral real Hong Kong assets and some liabilities, denominated in Hong Kong dollars, in the form of deposits. We will assume, as seems plausible, that the value of the real assets in terms of foreign currency is determined in a world capital market. That is, their value measured in terms of foreign currencies does not depend on the exchange value of the Hong Kong dollar. Our hypothetical financial institution begins with the balance sheet in the first column of Table 7.3. We measure all entries in foreign currency so that entries denominated in Hong Kong dollars will vary with the exchange rate but those denominated in foreign currency will not. Note that this institution has initial net worth equal to 10 percent of total assets and that it is unhedged with respect to the exchange rate in only a minor way. That is, an exchange rate depreciation would lower net worth in an amount proportional to the initial net worth denominated in Hong Kong dollars, namely, 10 percent of x . We now describe the effect on net worth of a substantial decline in the value, in terms of foreign currency, of the assets that serve as collateral for the loans denominated in Hong Kong dollars. We will assume for the sake of argument a 50 percent decline in the value of these assets, measured in terms of foreign currency, and we will assume that initially the value of the assets was equal to 43 of the loans. (Put another way, the loans were equal to 75 percent of the value of the collateral.) In describing the firm’s portfolio after the decline, we take the value of such loans to be the value of

220

Chapter 7: Speculations about the Speculation

Table 7.3 H OW

AN

NANCIAL

E XCHANGE R ATE D EPRECIATION C AN H ELP F IRM

FI-

After asset value decline and 33 13 % depreciation

After 50% decline in HD asset values

Initially

A

Assets Loans backed by HK assets denominated in HK$

x

1 4 2 3x

1 4 2 3x

Foreign assets denominated in foreign currency

y

y

y

x+y

2 3x + y

2 3x + y

Deposits denominated in HK$

9 10 x

9 10 x

2 9 3 10 x

Other liabilities denominated in foreign currency

9 10 y

9 10 y

9 10 y

Total Liabilities

9 10

Total

( x + y)

9 10

( x + y)

9 2 10 [ 3

x + y]

Net Worth Assets less liabilities

1 10

[ 23 −

( x + y) = 1 1 10 x + 10 y

9 10 ] x + 1 10 y = 1 7 y − 30 x + 10

1 2 10 [ 3 x + y ] = 1 1 15 x + 10 y

the collateral if the collateral is worth less than the initial value of the loans. If the value of the collateral assets falls 50 percent and the exchange rate does not change, then these assumptions imply that loans backed by Hong Kong dollars are defaulted on and the firm’s balance sheet changes to that in the second column of Table 7.3. A comparison of this column and the first shows that because of the asset price decline, net worth declines. In fact, it becomes negative, implying bankruptcy, if x is sufficiently larger than y .

A Timely Depreciation

221

If the value of the Hong Kong dollar falls at the same time as asset prices, however, the results is not as bad for the financial firm. Suppose that, along with the 50 percent decline in the value of real Hong Kong assets in terms of foreign currency, there occurs a decline of 13 in the exchange value of the Hong Kong dollar. (It takes 1 12 times as many Hong Kong dollars to buy one unit of foreign currency as originally.) Then, per Hong Kong dollar of loans originally granted, there is collateral worth ( 43 )( 12 )( 32 ) , or one Hong Kong dollar. Thus, there need be no default on the loans. As the new balance sheet, the last column of Table 7.3, indicates, the firm’s net worth still declines somewhat from its initial value, but not as much as with just a decline in real asset values. When a depreciation accompanies the asset price decline, therefore, the firm can stay in business. Individual depositors of the financial firm, of course, would not prefer this situation to that last, for a depreciation reduces the value of their deposits. At the same time, the value of assets in terms of foreign currency is the same whether or not a depreciation occurs. This is true in general. The extent of a depreciation of the Hong Kong dollar determines how the value of total assets is divided among categories of liabilities: the greater the depreciation of the Hong Kong dollar, the smaller the share that goes to deposits denominated in Hong Kong dollars. Over the last two years, many Hong Kong financial institutions seem to have been in positions like the one depicted here. If they were, then the Exchange Fund could easily have viewed a depreciation of the Hong Kong dollar as a way to avoid the major disruption to the economy that numerous financial firm bankruptcies would cause. A depreciation could smooth the Hong Kong economy’s adjustment to lower real property values in terms of foreign currency. The adjustment would be smoother not because financial firms’ capital losses would be averted, but because they would be allocated partly to depositors, thereby helping to preserve existing financial institutions.

222

Chapter 7: Speculations about the Speculation

Hong Kong’s Other Options With the exchange rate indeterminate under the official float policy, it is not surprising that the Hong Kong government should have departed from that policy in practice and reverted to an unofficial policy of pegging the exchange rate. It did, however, have other alternatives. One feasible alternative is to make its unofficial policy official: to adopt a fixed-rate policy in which the Exchange Fund acts as a warehouse for foreign exchange. Then, if government budgets are balanced or in surplus (in a present value sense), it could not only peg but gradually appreciate the Hong Kong dollar. Appreciation could be engineered as a way of paying out all or part of the interest that is earned on the foreign securities that the Exchange Fund holds as reserves. (Singapore evidently manages its currency this way.) Such a policy would be feasible regardless of the fate of the wealth of Hong Kong, as determined by news about 1997 or other factors, as long as the government balanced its budget so that the portfolio available to back the Hong Kong dollar remained intact.5 Another alternative that has been suggested is for the Hong Kong government to fix the quantity rather than the exchange rate of Hong Kong dollars and allow the exchange rate to float (Asian Monetary Monitor, 1982). However, this proposal is not complete without specific rules governing the asset side of the monetary authority’s balance sheet. Whether the proposal would work depends on whether the foreign exchange backing of the currency stock is to be eliminated or maintained. In one version of this proposal, fixing the quantity of Hong Kong dollars is viewed as a substitute for maintaining backing. 5 This

argument rests on the same considerations that led Keynes to write this of France in March 1924 (Keynes 1971, xvii): “It is often argued that the franc cannot fall in value because France is a wealthy, thrifty and industrious country, or because her balance of trade is prima facie satisfactory. This . . . springs from confusion as to the causes which ultimately govern the value of money. A very rich country can have a very bad currency, and a very poor country a very good one. The wealth of France and her balance of trade may render it easier for her authorities to pursue a sound monetary policy. But they are not the same thing. The value of a country’s monetary unit is not a function of its wealth or even its trade balance.”

Epilogue

223

Advocates of a fiat money (or unbacked paper money) regime seem to have this view. They see the ability to dispense with backing as a one-time dividend for the monetary authority to spend. The size of this dividend is sometimes said to be a measure of the inefficiency of a backed system compared to a fiat system. This sort of fiat system would probably not work in Hong Kong. For a fixed-quantity, no-backing regime to determine the exchange rate, there must be a stable demand function for unbacked Hong Kong dollars; in other words, a preference among Hong Kong residents for Hong Kong dollars over other currencies. However, it is questionable that such a preference could exist for a fiat currency issued by the government of an economy as open to trade and capital transactions as is Hong Kong’s economy. Foreign currencies circulate in Hong Kong now, and international transactions are not restricted. This suggests that extreme currency substitution potentially prevails in Hong Kong. If it does, then the exchange rate would not be determinate under a regime of a fixed quantity of unbacked Hong Kong dollars (Kareken and Wallace 1981). The other version of the proposal to fix the quantity of Hong Kong dollars would work, but it does not appear to be optimal. In this version, while the quantity of dollars is fixed, its backing is maintained, thus preventing Hong Kong dollars from becoming worthless. However, then there is no reason to fix the quantity. Indeed, when backing of the currency is maintained, it is desirable to allow the demand for Hong Kong dollars to determine their quantity for the same reason that it is desirable to let demand determine the quantities of other kinds of intermediary liabilities (Sargent and Wallace 1982).

Epilogue On October 15, 1983, the government abandoned the official float policy and announced a policy of pegging the Hong Kong dollar at 7.80 Hong Kong dollars per US dollar. Short-term interest rates denominated in Hong Kong dollars reacted by climbing sharply, to as high as 40 percent per annum for overnight money. Such high nominal rates can be interpreted as reflecting the market’s expecta-

224

Chapter 7: Speculations about the Speculation

tion that the government would not long maintain the current peg and that it would soon devalue. This widespread public pessimism is likely due to a variety of political forces currently operating in Hong Kong that are likely to affect and restrict the government’s actual choices in ways we have so far abstracted from. We have analyzed the valuation of the Hong Kong currency by viewing the monetary affairs branch of the Exchange Fund as though it were managing a mutual fund.6 Given an initial stock, f , of assets that are potentially available to redeem Hong Kong currency in terms of foreign exchange, the managers of the mutual fund face a variety of feasible payout policies. The policy that is adopted determines the net worth left to the managers of the fund. For example, among fixed exchange rate policies, the lower is the value of Hong Kong currency, the higher is the net worth. The Exchange Fund is likely to be managed by the British government until 1997 and by the Chinese government thereafter. In this situation, the Chinese government cares about how the Exchange Fund is administered before 1997, for this will influence the new worth of the fund at the time that the Chinese take over. To see this, suppose, for example, that the British managers of the Exchange Fund were to peg the Hong Kong dollar at the upper bound of the feasible range we have delineated, namely, f /h . Suppose further that holders of Hong Kong dollars (the fund’s shareholders) were uncertain about how the new Chinese managers of the fund would manage it. Inevitably, a run on Hong Kong dollars 6 Consider

a mutual fund with h shares outstanding and with initial assets worth f invested in foreign exchange. There are at least two feasible ways that the mutual fund could operate. First, the mutual fund could close, with no new shares to be issued. On the understanding that shareholders are entitled to a pro rata share of the assets, the market could be permitted to determine a price of shares in the mutual fund. Abstracting from fees, the equilibrium price would be f /h . Here the mutual fund would be priced as though it were operating a warehouse for foreign securities. Second, the mutual fund could remain open by offering to sell new shares or redeem existing ones at any fixed price between zero and f /h . Thus, either the mutual fund could fix its supply of outstanding shares h and allow the market to price them or it could set a share price at which it would be willing to issue new shares and invest the proceeds. However, a policy of offering to sell or redeem new shares at a price to be set by the market, with the proceeds to be invested and available for redeeming shares, would not pin down enough about the mutual fund’s policy.

Epilogue

225

would occur before 1997, a run the Exchange Fund could meet by drawing down to zero the foreign exchange backing, f . Then, the Hong Kong dollar would simply disappear too as the Exchange Fund paid out the backing to the owners of Hong Kong dollars. A fixed-rate policy such as this, which values the Hong Kong dollar at (or near) the upper bound f /h , could easily be viewed as unfriendly by the current Chinese government. Concern by the current government of Hong Kong to avoid such unfriendly acts could thus constrain the way the Exchange Fund is now managed. Additional doubts about the likely management of the Exchange Fund between now and 1997 stem from what threaten to be new and growing departures from Hong Kong’s traditionally conservative fiscal policy. These departures are partly induced by the fall in value of land and other assets in Hong Kong, the sale and taxing of which the government depends on for revenues. During the 1970s, the Hong Kong government had budget surpluses totaling 8.6 billion Hong Kong dollars (usually expressed HK$ 8.6 billion). Government investment was financed largely by current revenues and the proceeds of public land sales. As a result, even now Hong Kong’s public sector has large net claims on the domestic banking system and on the rest of the world. However, recent declines in government revenues have outpaced attempts to curb public expenditures, with the result that, in 1982–83 and 1983–84, the Hong Kong government had budget deficits for the first time in nearly a decade. These deficits are estimated at HK$ 3.9 billion and HK$ 3.2 billion, respectively. By themselves, two consecutive annual deficits of such magnitude do not threaten responsible management of the Exchange Fund, because the government has accumulated substantial fiscal reserves. At the start of the 1983–84 financial year, fiscal reserves amounted to HK$ 18.7 billion, of which HK$ 11.5 billion constituted free reserves, that is, reserves left after HK$ 7.2 billion were allocated to cover possible losses on capital projects. Deficits of the current magnitude would not wipe out these free reserves for about three years. However, continued deficits—and devaluations—may be seen as the only politically feasible ways for the British government to deal with the government’s capital losses resulting from the 1997

226

Chapter 7: Speculations about the Speculation

change in management. The 1997 situation creates special problems with regard to the pricing of services of long-lived, publicly owned capital—like the roads, the subway system, and the public housing. After 1997, the public capital of Hong Kong is likely to belong to the government of China, so that the tail of the stream of taxes and tolls from long-lived capital currently in place or under construction may not be available as backing for debt of the current government. To deal with this situation, the current government could increase charges on existing projects so that the revenues stream collected through 1997 would be sufficient to balance the budget in a present value sense, and it could undertake only those new public investment projects that would pay for themselves by 1997. Such policies would be drastic and amount to managing the economy as though it were going to end in 1997, a stance that could be expected to be opposed both by some current residents of Hong Kong and by the government of China. Such drastic policies could to some extent be avoided by accepting deficits and devaluations. Finally, one other event raises questions about the Hong Kong government’s commitment and ability to adhere to its tradition of conservative fiscal policy. As mentioned earlier, there is no government-supplied deposit insurance in Hong Kong. In the past, some banks have failed, but government intervention has been officially limited to a policy of accepting or encouraging the takeover or merger of threatened banks by or with stronger ones. However, on September 28, 1983, the Hong Kong government departed from this long-standing policy and took over the Hang Lung Bank. The financial secretary, Sir John Bremridge, explained this action by saying that the Hang Lung Bank’s failure and the resulting losses to depositors were unacceptable to the government. This departure from long-standing policy can be viewed as a threat to fiscal conservatism and a threat of further devaluations because it represents a major change in the categories of public and private liabilities that are potentially claims against the monetary and fiscal reserves of the government. Under the long-standing policy, only Hong Kong currency was a claim against the government’s monetary reserves, while bank deposits represented (risky) claims backed by the (risky) loans and investment of the banks. If the government has instituted a new policy of protecting depos-

Epilogue

227

itors against potential losses, the stream of implicit government obligations has increased substantially. Without tax increases, these extra obligations threaten to impair the funds available for backing the Hong Kong currency. While it might be desirable for other reasons, therefore, the takeover of the Hang Lung Bank is a signal that market participants could readily interpret as portending a weak Hong Kong dollar.

8 Six Essays in Persuasion This chapter consists of six essays that use “unpleasant monetarist arithmetic” to interpret events during the 1980s and 1990s in the United States and Brazil. During the 1980s, the United States took steps along a path upon which Brazil had travelled much further, a path along which interest-bearing government debt is growing as a percentage of GNP. The US government was able readily to borrow large amounts, and had far to go before the government’s budget constraint threatened to impose painful choices among the options of raising taxes, lowering government expenditures, or printing currency. Brazil found its ability to borrow very limited, and therefore had to confront those painful choices immediately.1

Back to Basics on Budgets The collapse of faith in the Phillips curve—the purported trade-off between inflation and unemployment that was generally accepted in the 1960s—has led to a renaissance of the classical doctrine about government finance, both among economists and statesmen. But evidently, the doctrine has not yet decisively influenced the US government. In classical economic doctrine, the case for balancing a government’s budget is based on the notion that a government should finance itself like a private firm or individual. The idea is that in a competitive economy, good results are promoted when the government commits itself to compete for loans on an equal footing with 1 “Back to Basics on Budgets” appeared in the New York Times on August 10, 1983. “Confrontations over Deficits” appeared in the New York Times on August 12, 1983. “An Open Letter to the Brazilian Finance Minister” appeared in the Wall Street Journal on January 30, 1986. “Letter to Another Brazilian Finance Minister” was published in Banco de Investimentos Garantia S.A. Newsletter, November, 1993. “Reasonable Doubt about the Real Plan” was published in Banco de Investimentos Garantia S.A. Economic Letter, December 19, 1995. “An American History Lesson for Europe” was published in the Wall Street Journal, February 3, 2012.

228

Back to Basics on Budgets

229

private parties. “On an equal footing” means that the government’s budget should be balanced in a present value sense, its prospective tax revenues being sufficient both to cover its prospective expenditures and to service any outstanding debt. Capital markets force private parties to have balanced budgets in a present value sense, but a government is usually sufficiently powerful that it can evade this requirement if it chooses. The good results that emerge from a balanced budget policy for the government include the financing of a given stream of expenditures in a way that is both most efficient and that adds no inflationary pressure to an existing monetary system. By balancing its budget in a present value sense, a government does not resort to the printing press and finance part of its expenditures via inflation. Even in its purest form, this doctrine permits the government to run temporary deficits, but only according to the following wellspecified principle: Borrowing is thought to be justified when a government undertakes expenditures that can be expected to generate prospective revenues to the government sufficient to recover the principal and any associated interest expenses. Long-lived or capital projects should be financed by a stream of taxes or other charges that match in time the stream of benefits that the project yields. Toll bridges, dams, and school buildings are often financed in this way. According to classical doctrine, deficits on “capital account” are justified because they are by definition temporary, and the loans undertaken to finance them are “self-liquidating.” Classical doctrine distinguishes sharply between such capital account deficits, and deficits that are incurred on “current account.” A current account expenditure is one that gives rise to no future stream of prospective tax revenues or other charges. Current account expenditures are ordinarily supposed to be financed out of current tax collections. As an exception to the general rule, borrowing on current account is justified to finance exceptionally large temporary expenditures, most notably those associated with wars. The justification for this exception comes from the fact that taxes distort incentives and private decisions about allocating resources. Generally, the amount of distortion over time can be minimized

230

Chapter 8: Six Essays in Persuasion

if tax rates do not change too much over time, but vary slowly and smoothly. For example, by smoothing tax rates over time in this way, the government can minimize the extent to which private citizens withhold resources during periods of temporarily high tax rates in anticipation of lower tax rates in the future. This is the classical justification for financing a war by borrowing and raising tax rates somewhat both during and after the war to repay the war debt. In ordinary times, however, when government expenditures on current account are varying smoothly, they are supposed to be financed out of current tax collections. For centuries, this was the accepted doctrine about government finance. And although this was probably not Keynes’s intention, it was acceptance of Keynesian economics that caused the balanced budget doctrine to be abandoned widely. (It is easy to read Keynes as advocating only that capital account deficits and surpluses be timed so as to counter the business cycle. Clearly, this advice can be heeded while respecting the balanced budget doctrine.) The 1960s notion of a Phillips curve, which depicted a purported trade-off between unemployment and inflation, promised that lower average unemployment rates could be experienced over the business cycle provided that the government engineer a higher average inflation rate. One good way to engineer this inflation was to run a persistent government deficit and to finance it by issuing additional unbacked paper money. This policy was tried in the 1960s and 1970s in a variety of countries. The result was not as promised, but rather stagflation, a combination of high inflation rates and unemployment rates averaging no lower than had been experienced by adhering to a balanced government budget policy. This failure diminished faith in Keynesian economics and prompted a renewal of classical principles. It is paradoxical that an administration that came to office rejecting the whole apparatus of Keynesian economics finds itself presiding over a stream of what threaten to be permanent deficits.

Confrontations over Deficits

231

Confrontations over Deficits The Reagan administration presides over the threat of large budget deficits into the indefinite future. Such deficits are not consistent with classical governmental financial policy and in themselves must be viewed as undesirable by sophisticated supporters of the administration. Yet, the administration is willing to tolerate these deficits as a weapon in a struggle to reduce the size of the government. The administration began office committed to support these goals: a path of reduced government spending, a budget that eventually will be balanced, and a program of slow and steady growth in the nation’s stock of money. The budget and money supply goals are inextricably interrelated. Unless the budget is balanced in a present value sense, it is impossible to adhere for long to slow and steady growth in the money supply. It is not feasible for a government permanently to run large deficits without eventually printing money at rapid rates to “repay” the government debt. The desire to reduce spending is not necessarily related to the last two goals, but it has become intimately linked to them by the way the administration has chosen to go about achieving it. Moreover, without keeping the spending goal in mind, it is hard to make sense of the apparent conflict between monetary and fiscal policies over the past two and a half years. The administration tried to reduce the prospective time path of government spending indirectly by first achieving a reduced stream of prospective tax collections. So long as the prospective path of government spending remains unchanged, the consequence of such a tax reduction must be a stream of larger federal deficits. One possible response to this situation is simply to leave the path of government spending unchanged and eventually to print more money to help finance the deficit. The cost of this option would be eventually abandoning the money supply goal and enduring more inflation. A second option is for the Federal Reserve to stick to a policy of slow and steady growth for money into the indefinite future,

232

Chapter 8: Six Essays in Persuasion

thereby foreclosing the possibility of ever monetizing the deficit. Under this option, sustained government deficits are simply infeasible. Denied the printing press, the government would have to promise higher and higher real interest rates on its ever-increasing debt. This force would eventually, and probably quite soon, create irresistible pressures to balance the budget, either by raising taxes or by reducing spending. The original hope of the Reagan administration was that the Federal Reserve would stick to monetary restraint and that reduced spending would be the outcome. That this indirect (and risky) method of trying to reduce the size of the government was resorted to is partly a product of the decentralization of responsibility for monetary and fiscal decisions that exists in the United States. (In Britain, responsibility for monetary and fiscal decision is centralized in the hands of the prime minister, which may help to explain why monetary and fiscal actions under Margaret Thatcher’s administration have been more synchronized with one another than under Mr. Reagan.) In the United States, decentralization exists both across institutions and over time. Despite this, it is the prospective eventual outcome of the entire policy-making process, much more than today’s actions alone, that influences economic events. Market participants care about long-term, repeatedly executed strategies of the monetary and fiscal authorities because it is prospects for future monetary and fiscal policy actions that influence the rates of return on government debt and currency that are issued today. Policy strategies in this sense are not set by a single group at a single moment but are determined by a succession of personalities within a succession of administrations and across a variety of institutions, including the Federal Reserve Board, the administration, and various committees of Congress. It is not possible for office holders to commit their successors to any policy, but the private sector’s conjecture about what those successors will do influences outcomes that are attainable today. Despite such extensive decentralization, monetary and fiscal policies have to be coordinated, one way or the other. The reason is that the Federal Reserve is a potential contributor to financing the government, and somehow its role as a source of revenue must

An Open Letter to the Brazilian Finance Minister

233

be determined. For example, one classical coordination plan is to have a truly independent central bank that is constitutionally instructed to refuse ever to purchase any government debt. This cuts off the government from the printing press and forces budget balancing in the present value sense. (A monetarist rule of low and steady growth in currency is an attempt to achieve such a result.) A coordination plan at the other extreme is to have a totally passive central bank that is instructed to monetize all of the debt that is issued by the Treasury. This coordination scheme, in use during each of the hyperinflations, permits a government to run persistent deficits. Neither of these coordination plans is now in effect in the United States. Monetary and fiscal authorities in the United States seem to be locked in struggle. Neil Wallace has observed that monetary and fiscal authorities seem to have been playing chicken over the past two and a half years. The Federal Reserve resolved to stick to a policy that is feasible only if the budget is approximately balanced, while Congress and the executive branch together have determined prospects for taxes and spending that are feasible only if the central bank eventually becomes passive and accommodating. With such mutually infeasible prospects, all that is certain is that one side or the other must eventually give in. Outside parties are thrust into the uncertain position of betting on which side in the game of chicken will eventually capitulate and on how and when. The Reagan administration must hope that the game of chicken will end with Congress capitulating by reducing outlays.

An Open Letter to the Brazilian Finance Minister Senhor, Your job is unenviable because you seem to have so much responsibility but have so little power.2 Inevitably, either you 2I

traveled to Brazil in December 1985, a time when inflation was rampant and there was high public interest in proposals for stabilization, in particular for so-called “heterodox” programs that typically emphasized a need to eliminate wage indexation and also assigned at least a temporary role to wage and price controls. At the time, it seemed to me that insufficient recognition was being

234

Chapter 8: Six Essays in Persuasion

or one of your successors will soon have to take some unpleasant fiscal measures. These unpleasantries can be delayed for a time, but only by accepting increasingly large costs in terms of the severity of the adjustments that ultimately will have to be made. Your job is so difficult now because your predecessors chose to delay. Arithmetic and compound interest make your options unpleasant. Either you or one of your successors will have to undertake at least one (and probably more than one) of the following three actions: reduce government purchases drastically, raise tax collections drastically, or default on substantial portions of your foreign and domestic government debt. You manage the fiscal affairs of a government that is currently spending much more than it is taxing, both explicitly and implicitly via the inflation tax. The key indication of the deficiency of tax collections relative to government expenditures is that you are borrowing more and more domestically. Arithmetic implies that your current policies can be continued only so long as you are able to borrow increasingly large amounts in real terms in the foreign and domestic markets. You will not be able to do so for long. There are limits to how much any country can borrow, even a country with resources as great as yours. The limit is imposed by lenders’ views of your capacity to run future government surpluses sufficiently large to service the debt. You have already reached the limit on your international borrowing. Limits on the volume of your domestic borrowing, indicated by the high real rates of interest, of 15% a year that you now pay, loom before you. When you have exhausted all of your opportunities to borrow, domestically as well as internationally, you will have to make given to the fiscal causes of inflation. Thus, Rudiger Dornbusch, commented as follows on one widely cited proposal for reform: “The Larida proposal starts from the premise that indexation, not monetized deficits, is the cause of inflation, and I share that view completely. Money creation, and more importantly, rising velocity because of monetary deregulation, are at best the air in the tires; indexation is decidedly the engine of inflation.” (See Dornbusch 1985.) Because of the weight placed on such opinions in Brazil, it seemed to me timely to remind people of the options described in 1926 by John Maynard Keynes in “An open letter to the French Finance Minister.” I wrote my “Open Letter to the Brazilian Finance Minister” as a reminder and application of the principles described in Keynes’s letter. Keynes’s letter is reprinted in his Essays in Persuasion.

An Open Letter to the Brazilian Finance Minister

235

one or more of these unpleasant adjustments: raise taxes, lower government expenditures, or default on some of your debt. The arithmetic that impels some combination of these adjustments is the heart of your problem, and it cannot be avoided. The most evident aspect of your difficulties is the rapid inflation over which you find yourself presiding. This is a symptom of your problems, which reflects deeper causes and which cannot be stopped without addressing those deeper causes. By rapidly printing currency to help finance your government, your central bank is now engineering an inflation rate of 250% a year. In doing so, the central bank is operating openly as an effective agent of your internal revenue service. Inflation is very useful to you because it raises an “inflation tax” of substantial proportions, on the order of 2% of your gross national product annually. Inflation is indispensable to you so long as no adjustments are made in other elements of your tax and expenditure policies. Technically, you could arrest inflation (but only temporarily) simply by ordering the central bank to stop printing additional currency. However, you would have to make up for the lost revenues that the inflation tax is bringing you. In your current situation, you would have to make up for those lost revenues by borrowing in the form of indexed government debt. But you are paying very high real rates of interest on that debt, meaning that increased borrowings only add to the interest and principal payments that you or your successors will have to cover later on. The high real rate of interest that you are paying is causing your domestic debt to explode even now. The real interest rate on your indexed domestic debt far exceeds the sustainable growth rate of your national product, meaning that your indexed debt is growing as a proportion of your national product. This it cannot do indefinitely. Reducing your inflation tax by borrowing more in your domestic market is counterproductive because it means passing on even larger debt to your successors than your current policies threaten. So my first piece of advice is that you had better retain the inflation tax, and make heavy use of it, until other adjustments have been made in your finances. Your budgeting problems are created by the need to meet large

236

Chapter 8: Six Essays in Persuasion

interest payments on your externally and internally held government debt. Economically, the large interest payments on your external debt act like the reparations payments that drove Germany to hyperinflation in 1922 and 1923. For you, the unexpected increase in the real interest rates on your external debt that occurred after 1980 acted like an increase in reparations payments. Those increases affect you as they did the finance ministers of Germany in the early days of the Weimar Republic, creating an irresistible pressure to resort to inflationary finance. (Germany’s hyperinflation ended only when the country was granted relief by the Reparation Commission in the form of a promise of substantial downward revision in reparations debts. Your inflation may not end until your debts are written down by your creditors. Maybe you, too, will feel compelled to run a hyperinflation to impress upon creditors your need for relief.) But there are differences between Germany’s situation after World War I and yours today. Germany’s reparations debts were imposed upon it by the Allies; your debts were freely negotiated by your government (though in contracting those loans your government and the foreign lenders probably counted on a continuation of world real interest rates at the lower levels experienced in the 1970s). At the end of the war, Germany had a large interest-bearing internal government debt that was denominated in German marks and whose real financial burden on the government was eliminated by inflation. You and your predecessors have chosen to respond to the difficulty of acquiring new foreign loans by borrowing large amounts domestically in indexed form at high real rates of interest. You cannot default on this debt gracefully as did the German government by inflating it away. Your indexed domestic debt is a time bomb. I admit that it is a bit impertinent of me to bring this unpleasant arithmetic to your attention. After all, for the past four years the government of my own country has been following the very same fiscal policies whose arithmetic implications I am telling you about. (We haven’t yet resorted to the inflation tax, but the real value of our interest-bearing debt has been increasing as a result of unprecedented peacetime government deficits and high real interest rates on government debt.)

Letter to Another Brazilian Finance Minister

237

I also admit that my government’s pursuit of those policies probably contributed to the higher real interest rates that you have to had to pay on your foreign debt. But it doesn’t diminish the relevance of this arithmetic for you that it applies also to the Treasury of my country. Besides, the fiscal authorities of my country are more powerful than yours are because they control the value of the dollar, the currency in which your foreign debt and my government’s foreign and domestic debts are denominated. By engineering an unexpected inflation, the fiscal authorities in my country have the power to effect a graceful default on our government debt, and some of yours as well, a power that you do not possess. Your obedient servant, Thomas J. Sargent

Letter to Another Brazilian Finance Minister Dear Sir: Events in Brazil since 1986 have proved again the time-honored arithmetical principles of government finance that I summarized for one of your many predecessors in January 1986.3 A country’s inflation rate at any moment emerges out of the sustained monetary and fiscal policies that it chooses, now and in the future. The phrase “in the future” occurs in this formula because the value of a country’s currency today is vitally affected by what people think its value will be tomorrow. This "in the future" caveat has frequently played the villain in formulations of halfway doomed plans that wished indefinitely to postpone the unspecified. painful adjustments that would have been necessary to make the plans fit together. I prefer not to catalogue the halfway plans that your government has started and then dropped. Your government has tried many tricks in vain attempts to avoid confronting the fact that inflation originates in monetary and fiscal policies. You have tried price controls, credit controls, and tight money unaccompanied by 3I

confess that the title of my earlier letter was my indirect way of urging you to read Keynes’s 1923 letter with a similar title.

238

Chapter 8: Six Essays in Persuasion

fiscal constraint. These experiments were interesting for macroeconomists (they produced results that our models had led us to expect), but they were cost1y for your country. T HE E CONOMICS A RE S IMPLE Persistent high inflation is always and everywhere a fiscal phenomenon, in which the central bank is a monetary accomplice. A government administering a fiat paper currency has an intertemporal budget constraint that forces its central bank and fiscal authorities to cooperate, sooner or later, somehow. Persistent inflation is caused by a coordinated monetary and fiscal policy regime that calls for the central bank persistently to print currency to supply the fiscal authority with revenues via an implicit “inflation tax.” A central bank cannot by itself stop an ongoing inflation against the will of a fiscal policy authority determined to run persistent budget deficits. Indeed, a central bank determined to "go it alone" and to fight inflation with tight money in the face of persistent deficits can achieve only temporary gains in the battle against inflation, and at the cost of making inflation worse in the future. This outcome results because, in the face of a persistent fiscal deficit, a central bank can achieve go-it-alone tight money only by forcing the fiscal authority to issue increasing amounts of interest-bearing debt: without an enduring fiscal adjustment, the monetary authority will eventually be forced to generate more inflation and to raise the inflation tax. This arithmetic is unpleasant, but it is also true. C REDIBILITY Your economic advisors have told you what it would require to arrest inflation in Brazil: a credible switch to a sustained fiscal regime that, by administering a government budget balanced in the present value sense, would permanently relieve the monetary authority of any temptation to print too much Brazilian currency. A credible policy is a course of announced future government policy actions that the public can rely on the government actually to execute. How is the public to go about deciphering a credible policy? Government policies emerge from political and administrative

Letter to Another Brazilian Finance Minister

239

processes, and are implemented by many public servants and statesmen. A credible public policy has the property that it is in the self-interest of each responsible public servant to implement his part of the plan: this is what makes it believable by the public, and “sustainable” as an ongoing public policy. Credibility is not something that a small number of people, even presidents and ministers, can manipulate. In a democracy, a fiscal policy credible for supporting a stable currency must be arranged so that “the votes are there" for levying enough explicit taxes to cover whatever expenditures have been voted. Our “in the future” caveat is all about this reading of the political process. This description is about as far as your economic advisors can take you. We economists can explain the arithmetic of the government budget constraint (according to which curing inflation is an easy problem, as economic problems go); and we can tell you how important the acquisition of credibility is, and how limited in a democracy your role is in lending credibility to government policy. Your country has a technically competent central bank, and has had some knowledgeable finance ministers. Those two things help, but neither the finance minister nor the central bank can “deliver the votes” at the times and on the measures that will be required to stabilize your currency. A D ELICATE I NSTITUTIONAL I SSUE Is good credible public policy easier to attain under democratic or authoritarian form of government? Bismarck argued for the authoritarian side. Bismarck said he disliked dealing with the British because an election could overturn an understanding he had reached with a prime minister. He preferred to deal with an autocrat who could be counted on to have the authority to follow through with his commitments. The distinguished French finance minister Jacques Necker in 1784 argued for the other side. Necker pointed out that the same authority that made it easy for the king of France (an “absolute monarch”) to default on government debt made it difficult for the finance minister to sell the government’s debt in the market. Necker pointed out the king’s vast power to default implied that

240

Chapter 8: Six Essays in Persuasion

he could sustain public trust only by giving “reassurances on the sovereign’s intentions, and by proving that no motive can incite him to fail his obligations.” As one of the last finance ministers before the Revolution, Necker had good reason to worry about the credibility of the king’s government. Necker was one of a large number of statesmen who at the end of the Ancien Régime sought to make it easier for the French government to acquire credibility for good public policies by separating and decentralizing government authority along the lines embraced by the British in the Glorious Revolution of 1688. There is ample historical evidence on Necker’s side of this argument, some of it summarized by North and Weingast (1989). North and Weingast document how the institutional changes implemented during Britain’s Glorious Revolution—changes that inc1uded founding the Bank of England and establishing parliamentary primacy in matters of borrowing and raising taxes to service government debt—coincided with the British government’s acquiring the easy access to domestic and world capital markets, and that set the stage for an astounding period of British political and economic preeminence. A DVICE Despite your limited influence on forming a credible anti-inflation plan, you can help. First, because a plan can be credible only if it is feasible, if you do compose a plan it should fit together and transparently respect the intertemporal budget constraint. Second, it might help if you endorse measures like privatization schemes that promise to alter the constellation of private interests away from favoring government deficits. The public institutions that you help design and administer will affect the credibility of your anti-inflationary intentions. Countries as diverse as contemporary New Zealand and Revolutionary France have consciously designed institutions to alter subsequent political outcomes. In this vein, you can push for arranging your social security and social insurance institutions to be “fully funded” and less of an open ended “put” on general revenues. Finally, you can beware of economists bearing magical schemes for painless solutions of monetary problems that forget the arith-

Reasonable Doubt about the Real Plan

241

metic of your budget constraint, and that ignore how a credible public policy must emerge from Brazil’s own political process. Your obedient servant, Thomas J. Sargent

Reasonable Doubt about the Real Plan The mathematician David Hilbert lived through the German hyperinflation, but doubted the currency reform used in November 1923 to attain an abrupt and permanent stabilization of prices.4 “One cannot solve a problem by changing the name of the independent variable,” said Hilbert. Though he was not an economist, Hilbert was stating the prediction of monetary theory that, by themselves, pure units changes are irrelevant. Unless they are accompanied by changes in the fundamentals—the fiscal forces that cause base money to grow —changes in the units in which prices and the money supply are measured cannot affect the rate of inflation. Nevertheless, the Real Plan originated with a change in the units of measurement. The Real (R$) was born in July 1994 as the physical manifestation of an earlier abstract unit of account, called the URV (real unit of value), and in terms of which Brazilians expressed prices and wages. From March to June 1994, one URV equaled one US dollar. At the mandate of the government, prices and wages were stated in URV’s, but payments continued to be made in Cruzeiros Reals. Using the URV as a unit of account reduced the costs of posting price changes, because continuously rising prices in Cruzeiros Reals translated into stable URV prices. Stating prices in terms of URV’s also eliminated a variety of mechanisms for indexing payments in Cruzeiros Reals in what had been imperfect measures to isolate real values from variations in nominal ones. The URV implemented a government-coaxed dollarization of accounts, but not payments. The URV became currency at the beginning of July 1994, when the entire currency stock in Cruzeiros Reals was redeemed for col4 The

author is grateful to Rodrigo T. da Rocha Azevedo, Claudio Haddad, and Alexandre Ferraz de Marinis for discussing the subject of this essay with him, but assumes responsibility for all opinions expressed.

242

Chapter 8: Six Essays in Persuasion

ored pieces of paper, called Reals, at the prevailing URV-Cruzeiro Real (really, dollar-Cruzeiro Real) exchange rate. Thus, in July 1994, the unit of account and the physical medium of exchange —which had been divorced by the introduction of the URV—were remarried in the Real. The Real Plan suddenly arrested inflation, mysteriously, because Hilbert’s doubts make us search for changes in the fundamental determinants of inflation, and these are difficult to find. Inflation equates the demand for money with the rate of money creation, itself ultimately determined by fiscal decisions of the government. Here is where stabilizations are made and broken. The currency reform that replaced the Cruzeiro Real with the Real was a superficial technical measure designed to ease the burdens of the monetary authorities by eliminating adverse nominal price and wage dynamics that had emerged in the highly indexed, high inflation Brazilian economy. Those dynamics were not the first causes of high inflation, but adaptations to cope with it. They threatened to inhibit a smooth transition to a low inflation economy, and that motivated the monetary authority to eradicate them. A paradox (or excuse) arose with the end of the extensive but imperfect systems of indexation in place with the Cruzeiro Real, because the inadequate indexation of government appropriations seemed to reduce real government expenditures, thereby limiting the government deficit. The story circulates widely that the end of this mechanism has made control of government spending more difficult, and is partly to blame for the burst in expenditures that has broken budget balance despite record tax collections as a percent of GDP. It is easy to admire the technical skill and creativity of the Brazilian monetary authorities, both during the last years of the inflation and the first 18 months of the Real Plan. But without more help from the fiscal authorities, monetary policy can only do so much. Except for its adverse effects on government expenditures, the de-indexation of the economy accomplished by the Real Plan is a technical detail, a sideshow that hasn’t touched the fundamental causes of inflation. The monetary authorities know this, and have accompanied the Real Plan with transient measures to jolt the fundamentals, hopefully to buy time for the fiscal authorities to

Reasonable Doubt about the Real Plan

243

make the adjustments necessary for the stabilization to endure. Some of these temporary measures prop up the demand for. base money, others arrest the supply. The monetary authorities have used powerful expedients, some of which counter the eventual goal of complete economic liberalization, including substantial increases in reserve requirements and new controls on quantities and durations of credit—measures that boost the demand for base money by in effect taxing alterative assets. These onetime measures to press the price level down were reinforced by the acceleration in 1994 of the process of opening Brazil to import competition, and by the sustained high real domestic interest rates which propped up the international value of the Real. This battery of ad hoc measures combined with a one-time “velocity bonus” to arrest inflation for 18 months. The velocity bonus, an increase in people’s demand for currency as a fraction of GNP from 1.5% to over 4%, partly measures the credibility the stabilization program has acquired, but it also partly reflects the efficacy of those ad hoc measures. Persistent inflation is always and everywhere a fiscal phenomenon. Its cause is persistent government deficits that must be financed by printing money, its cure is government budgets that are on average balanced over long periods of time, and which therefore will not ultimately have to be financed by printing money. A government that commits itself forever to abstain from using inflation as a tax lets the market evaluate its debt according to the present value of its prospective surpluses, and cannot run a persistent deficit. It backs its debt with firm prospects of surpluses, just like a private firm. This is the first principle of inflation stabilization. This fiscal force is mediated through a mechanism that helps to determine the short-run dynamics of inflation, the public’s expectations about near-term inflation. People economize on money holdings more when they expect higher rates of inflation, which puts upward pressure on the actual rate of inflation. The inverse effect of expected inflation on actual inflation is the source of the “velocity bonus” that attends successful stabilizations. If the public’s expectations about inflation were “independent variables”—and were susceptible to manipulation by the government and other public opinion makers—then inflation could be “talked down” by

244

Chapter 8: Six Essays in Persuasion

convincing people to expect lower inflation. But expectations influence actual inflation in the same way expectations about future asset prices influence their current market prices. The only enduring way to manipulate the public’ s expectations about inflation is to control the “fundamentals” that determine inflation—which are fiscal. Manipulating people’s expectations is a short term tactic of doubtful efficacy and longevity. The interplay between these two principles makes stabilizing inflation a challenging problem. Inflationary expectations might seem to have free play in the short run, but inevitably there occurs a reconciliation between inflation and the fiscal fundamentals. Inflationary expectations could have an enduring independent influence only if people were persistently wrong in their forecasts of inflation, and that is not possible under Abraham Lincoln’s proposition that you cannot fool all of the people all of the time. Expectations about inflation ultimately stalk expectations about the fiscal fundamentals. A government cannot for long attain a stable currency by borrowing large amounts at high interest rates and investing the proceeds at low rates, even if those investments are labeled “foreign reserves.” But that is part of Brazil’s policy today. Brazil’s reserves of 50 billion dollars “back” the Real, but it is expensive backing. (It is a strange world in which the directors of a private business would be impeached for borrowing at 30% and investing at 5%, but the directors of the central bank are praised for their prudence.) When he was still a classical economist in 1923 and the world monetary system was still attached to gold, John Maynard Keynes asserted that a country’s gold reserves were irrelevant to whether it could sustain a strong currency, but that its fiscal conditions were everything. This is Keynesian advice worth hearing. The high interest rates that Brazil’s government pays signal that an enduring fiscal solution is not yet in place. They impose costs on private firms and banks now, and on future governments, who will be pressed to bail out banks and other debtors put in distress by those high interest rates, and who will ultimately pay or default on the debt that the arithmetic of compound interest and current policies hand them. The authors of the Real Plan hope that the fiscal authorities will

An American History Lesson for Europe

245

make good use of the time that their costly temporary expedients have purchased. The high technical competence of the monetary authorities has lent the government’s plan a degree of credibility that past stabilization plans lacked, as reflected in the substantial velocity bonus that has been attained without resort to a price freeze. The Real Plan so far is the boxing shorts awaiting the boxer, the opera orchestra awaiting arrival of the singers. Until now, the audience has been satisfied with these signs that there will be a performance. It has not yet begun.

An American History Lesson for Europe In 1789, the political price for our federal constitution included a bailout of the thirteen indebted states. But it was by refusing to bail out the states a second time in the 1840s that the United States preserved its federal system, with substantial fiscal independence for state governments. Facing a similar moment, Europe might learn from our experience. The 1789 bailout was part of a grand bargain designed by Alexander Hamilton to convert the creditors of the thirteen states into advocates of a stronger federal government—one having the ability to raise all revenues required to service the large debts that the Continental Congress and the thirteen states had both accumulated to finance that "Glorious Cause," our War of Independence. Hamilton and George Washington wanted those debts to be paid. They had to engineer institutional changes to achieve that goal. Under our first constitution, the Articles of Confederation, the continental government had virtually no power to tax. For revenues it depended on voluntary contributions from the thirteen states. About two-thirds of our total debts were owed by the continental government, the other third by the thirteen states. If they had been valued at par, federal and state debts together would have constituted about 40% of gross domestic product. But because tax revenues were not big enough to service them, both federal and state debts traded at very deep discounts, deeper than those we

246

Chapter 8: Six Essays in Persuasion

see in Europe today. From the point of view of the creditors of the states and the United States, if not our taxpayers, there was a fiscal crisis in the 1780s. Fiscal crises often end in rearrangements of political institutions designed to sort out which old promises will be broken and which sustained. Hamilton’s Report on Public Credit from 1790 describes the grand bargain and his reasons for advocating it. The Articles were replaced by the new US Constitution, which shifted exclusive authority to levy tariffs from the states to the federal government. In return, the Congress assumed the states’ debts in August 1790. The federal government immediately imposed a tariff, and it used about half of the ample revenues that soon rolled in to service its debts. State and federal debts went from trading at deep discounts to par in the early 1790s. Why did Hamilton and Washington want to honor our debts? Because they wanted the US government to build a good reputation vis-à-vis its creditors. Hamilton reasoned that honoring the existing debts would allow the government to borrow on good terms in the future. That ability to borrow would generate fiscal flexibility by creating a prospective source of revenues beyond current taxes, one that might be used to finance surges in government expenditures associated with wars and other future difficulties and opportunities. Why did Hamilton and Washington want the federal government to bail out the thirteen states? Because they wanted to realign interests in a way that would induce voters to support a federal government with ample ability to tax. They thus built a good reputation for the US by creating another, potentially troublesome, reputation with the states—the perception that the federal government would bail them out. Although the trade-off was made mainly for political reasons, one of Hamilton’s stated reasons for bailing out the states was that most of their debts had been incurred for their contributions to a national public good: acquiring independence from Great Britain. Was that reason authentic or was it just a convenient cover, a rationalization for Hamilton’s real goal, which was to convert the creditors of the state governments into advocates of the central government’s ability to tax?

An American History Lesson for Europe

247

Now fast-forward into the next century. To finance canals and railroads, many state governments incurred large debts in the 1820s and 1830s. A financial crisis in the late 1830s pushed many of those state debts into default. Appealing to the precedent set by the 1789 bailout, state creditors asked the federal government to bail out the states once again. After an enlightening debate, in the early 1840s Congress declined, so many states repudiated their debts. In the aftermath of those repudiations, many states rewrote their constitutions to require year-by-year balanced budgets, something they had never done before. As noted, fiscal crises, like the one in Europe today, often produce political rearrangements—at best peaceful ones like these. Did the federal government do the right thing in refusing to bail out the states in the early 1840s? By doing so, the federal government reset its reputation vis-à-vis the states, telling them in effect not to expect it to underwrite their profligacy. In the short run, that cost the federal government substantially in terms of its reputation with its own creditors. Federal credit abroad suffered along with state credit. But in the long run, the decision exposed state governments to continuing market discipline, making future crises and requests for federal bailouts less likely. If the federal government had chosen to bail out the states a second time, it probably would have taken greater control over state taxes and revenues in order to prevent yet another bailout situation. Refusal to bail out the states was thus a pivot point in sustaining a federal system in the United States. It led the states to discipline themselves by rearranging their constitutions in ways designed to allow them to retain freedom and responsibility for taxing and spending within their borders. Europeans today might be tempted to say “yes” to bailouts. Or they might also recall a time when Americans preserved their own federal system by saying “no.”

9 Macroeconomic Features of the French Revolution with François R. Velde

Introduction C HRONOLOGY OF E VENTS This chapter interprets the French Revolution from the vantage point of macroeconomic theories about government budget constraints. From 1688 to 1788, Britain won and France lost three of four wars. France recurrently defaulted on its debt, Britain did not. After 1688, Britain had reformed its institutions to allow it to raise enough taxes during peacetime to finance debts incurred in times of war, while France sustained institutions designed to constrain the king’s revenues. Modernizing forces in France wanted to eliminate government defaults and to smooth taxes and to strengthen France’s government by enhancing its access to capital markets. When King Louis XVI acceded to the throne in 1774, he pledged to honor the Crown’s debt commitments, and successive governments initiated fiscal reforms to facilitate raising and smoothing taxes. Yet reform was uneven, and the pressure not to default outran the government’s ability to raise taxes. This led to a run-up of debt and interest payments in the years preceding 1789. Rather than default again, the king called the Estates General to accelerate the reform process.1 The National Assembly aggravated the debt problem by immediately reducing taxes and by repurchasing many offices previously sold by the Old Regime. The government nationalized Church properties and transformed the debt problem into a “privatization problem.” It used the confiscated National Estates to administer a tax-backed money scheme by issuing notes acceptable at auctions 1 Conklin

(1993) asked: to those who trusted in prior arrangements, what is the difference between a "reform" and a "default"? Not much as far as we know.

248

Introduction

249

of specific Church properties.2 The government spent the notes for goods, services, and debt payments, auctioned some land on credit, then burned the notes that were returned at the auctions. However, many notes remained in circulation, especially ones of small denomination. Specie left France for England and the rest of Europe. After two years of large note issues, the supply of new currency rose to 70% of the pre–1789 stock of specie, and the price level rose about 30%. The tax-backed money scheme functioned adequately until a war broke out in 1792, which initially went badly for France. The government wanted more resources, so it divorced note issues from the land sales. The tax- backed money plan devolved into a fiat money scheme, causing real balances to drop and prices to rise quickly in early 1793 and threatening the base of the tax (the inflation tax) that was the government’s lifeline. The Girondins bequeathed this situation to the Jacobins, who responded by imposing wage and price controls and a set of legal restrictions designed to boost the demand for assignats (a “guillotine-backed currency” scheme). As a result, real balances rose and the price level fell in the face of extraordinary issues of assignats, and the government raised immense revenues. In 1794, the war turned in France’s favor, and the sense of emergency diminished, making unenforceable the legal restrictions that had propped up the currency. The Jacobins fell in the summer of 1794, and the legal restrictions supporting the demand for assignats disappeared along with the Terror. Then France experienced a hyperinflation: real balances fell and prices exploded. The Directory administered the first classic hyperinflation in modern Europe until 1797 when it defaulted on 2/3 of the government’s debt. Specie returned from England to France. In 1797, France returned to a specie standard and remained on it throughout the Napoleonic Wars. In 1797, Britain suspended convertibility with specie and did not reinstate it until six years after Napoleon had permanently left France.3

2 “Tax-backed”

because the land underlying this “land-backed” scheme was acquired by confiscation. 3 This summary is based on our reading of the historical record. Other readings include Stourm (1885), Marion (1914–31), Harris (1930), Aftalion (1987), White (1989), Bordo and White (1991), and Brezis and Crouzet (1994).

250

Chapter 9: Macroeconomic Features of the French Revolution

M ACROECONOMIC T HEORIES C OLORING O UR O BSERVATIONS Two macroeconomic ideas and three models of money inform our chronicle of events. Unpleasant arithmetic.— Government budget constraints and the arithmetic of compound interest impose restrictions on government deficits and debt. We use this arithmetic despite two difficulties. First, we have to assume that some commitment mechanism is available to support any sovereign borrowing.4 Second, when it is assumed that the government can borrow because it can commit to repay its debt, the government budget constraint alone imposes virtually no restrictions on tax and expenditure processes (see Hansen et al. (1991)). We obtain restrictions like those in Sargent and Wallace’s (1981) “unpleasant arithmetic” only by arbitrarily putting an upper bound on the amount of government borrowing. Assertions that a fiscal crisis sparked the French Revolution hinge on positing a bound, and on asserting that the French government was nearly hitting it. Sustainable plans.— Chari and Kehoe (1990) define a “sustainable” government plan to be one which enlists the self-interest of each group to implement its part when the time or the contingency comes for it to act. “Reform” of the Old Regime was difficult precisely because its institutions were largely sustainable, as a sequence of French finance ministers from 1775 to 1789 discovered. Tax-backed or asset-backed models of the demand for currency.— These models describe accompanying fiscal arrangements through which new issues of paper money cause little or no inflation. Legal restrictions models of the demand for currency.— These models study how a government can tax its citizens by forcing them to hold paper money it is issuing to finance a deficit; they describe circumstances in which large new issues of paper currency cause less inflation than would be expected if people were voluntarily holding the currency. Classical hyperinflation models along lines described by Phillip Cagan (1956).— These models describe circumstances in which rapidly 4 See

Bulow and Rogoff’s (1989) and Chari and Kehoe’s (1993) analyses of the severe limits on sovereign borrowing caused by the market’s limited ability to punish sovereign defaults. Also see Prescott (1977) and Manuelli (1988).

Before the Revolution

251

issuing a paper currency causes prices to rise even faster than the quantity of currency. We use these theories first to shape our descriptions and second to interpret how the Revolutionaries explained their actions. We document how the Revolutionaries used elements of these theories in the debates that shaped the Revolution. Our double use of theories reflects the rational expectations hypothesis that parts of a time series model are used by the people within the model to guide their forecasts and decisions. We interpret inflation during the Revolution in terms of a procession of regimes in which the “if” parts of the three types of monetary models are approximately fulfilled.

Before the Revolution E VEN A BSOLUTE M ONARCHIES H AVE B UDGET C ONSTRAINTS The immediate cause of the French Revolution was the fiscal crisis of 1788. For seventy years, France had confronted a sequence of similar crises, all stemming from France’s incomplete efforts to adopt fiscal policies that Britain had used since 1688. Figure 9.1 displays the ratio of debt service to fiscal revenues for France and Britain during the 18th century.5 The two series display the same pattern, increasing from very low values in 1689 to about 60% in 1789. The upward movements correspond to the major episodes of the “Second Hundred Years War” between Britain and France: the Nine Years War of 1688–97, Spanish Succession War of 1701–14, Austrian Succession War of 1741–48, Seven Years War of 1756–63, and the American War of 1776–83. During these conflicts, alliances changed often, but France and England were never on the same side. The similarities in the movements of this ratio in the two countries conceal differences in policies: while Britain would modify the denominator (fiscal revenues), France acted upon the numerator (debt service).

5 The

sources for France are those cited for Fig. 9.3, as well as Weir (1989).

252

Chapter 9: Macroeconomic Features of the French Revolution

70

60

% of Taxes

50

40

30

20

10

0 1690

1700

1710

1720

1730

1740

1750

1760

1770

1780

Figure 9.1 Ratio of debt service to taxes, Britain and France, 1688–1788. Sources: for Britain: Mitchell (1988); for France: Weir (1989) and the references listed in n. 10.

T HE B RITISH E XPERIENCE The British pattern was for this ratio to increase during wars, when debt was incurred to pay for large military expenditures. During war, taxes were raised to assure adequate funds to service the loans. After a war, the floating debt6 was consolidated into perpetual annuities and taxes were further increased to generate a sufficient net-of-interest surplus to service the debt. Figure 9.2 shows the components of Britain’s budget constraint during the same period. Total fiscal revenues are set against total spending decomposed into military, civil, and debt service spending. The net-of-interest civil government spending is roughly constant. Military spending surges during wars. In contrast to the volatility of total government expenditures, revenues are smooth. 6 By

floating debt we mean the following: in Britain, the unfunded debt (notes issued by the Exchequer or various departments, without an act of Parliament); in France, the anticipations, notes issued by the departments or financial officers on behalf of the government, typically maturing within two years.

253

Before the Revolution

30

Millions of pounds

25

20

total govt spending

15

10 revenues

5

civil plus debt service civil

0 1690

1700

1710

1720

1730

1740

1750

1760

1770

1780

1790

Figure 9.2 Revenues and spending in Britain, 1689–1790. Total spending is decomposed into three components: civil, debt service, and military expenditures. The three lines recorded for expenditures are for civil expenditures, civil plus debt service, then total expenditures, so that the vertical distances between these lines represent, respectively, civil expenses, debt service, and military expenditures. Total revenues are depicted with small circles. Source: Mitchell (1988).

The British government incurred large deficits in wartime and generated small but sufficient surpluses in peacetime. We observe a cycle of debt service rising during each war and then slowly declining with the onset of peace. Britain did not default on its debt during the one hundred years following the Glorious Revolution of 1688, which reflected the existence of mechanisms intended to make the state credit worthy.7 The British king retained executive power, while the Parliament gained the powers to examine and censor the budget and to vote taxes. By 1715, the system had been refined into a method of funding by which each loan was accompanied by a parliamentary vote for a specific tax to service the loan. Established in 1694, the Bank of England became an important 7 See

North and Weingast (1989) for a modern account.

254

Chapter 9: Macroeconomic Features of the French Revolution

element of a mechanism committing the government to pay its debts. By the mid-1720s, after the South Sea Bubble, the bank had acquired virtual monopolies of serving government debt and issuing notes. The bank was designed to prevent the government from playing one lender against another.8 The bank’s charter made it more difficult for the government to default, and the prominence of principal owners of the bank ensured that any attempt to default would be well-publicized. In exchange for abstaining from opportunistic behavior, the government acquired a flexible, guaranteed credit line at reasonable rates.9 A CROSS THE C HANNEL For France, the three sharp falls in the ratio in Figure 9.1 each correspond to an episode of reimbursement suspension and default in the form of cuts in interest payments. The Spanish Succession War in 1713 marks the first episode. By 1715, the debt service had been reduced through defaults on significant parts of the floating debt. The Regency (1715–23) witnessed the “System” of John Law, a vast operation that first reimbursed the debt with bank notes which devalued quickly. Next, the debt was reconverted into perpetuals and life annuities, with sharply reduced capital value. Finally, because these measures proved insufficient, an interest rate cut was imposed in 1726, by which time the debt service ratio stabilized at 30%. The second episode occurred during the Seven Years War in 1759 when the government converted the floating debt into perpetuals and halted scheduled reimbursements of fixed-term loans. These reductions continued until the Peace of Paris forced the government to resume its obligations for a short time. Lack of funds soon curtailed compliance. Meanwhile, the floating debt 8 For interpretations of the Bank of England as a commitment mechanism, see Thomas Babbington Macaulay (1831) and John R. Hicks (1969, 93–95). See Greif et al. (1994) for a related analysis. 9 In 1797, the British government was to discover another major advantage to the institution, namely the suspension of convertibility of the notes, which allowed the financing of the French Wars. Significantly, the government did not default on the notes, and resumed the gold standard at par in 1819, in effect redeeming the notes at face value and giving a handsome return to those who had accepted the notes during the Bank Restriction.

Before the Revolution

255

had again bloated to unmanageable proportions, so a third episode started in 1770. The ministry of Terray suspended reimbursements, converted the floating debt to perpetuals, and imposed coupon reductions on bonds of between 7% and 50%. Taxes were also increased in 1771–73. When Louis XVI succeeded his grandfather in 1774, the floating debt was negligible, debt service stood at less than 40% of revenues, and the budget was nearly balanced: 375 million in revenues offset 415 million in expenditures, of which 40 million was reimbursements of outstanding debt. These recurrent French defaults reveal different patterns of government revenues and spending between France and England. For this period, budget data for France are not as available as for Britain. To provide a counterpart to Figure 9.2, we constructed estimates of revenues and the components of spending for the period 1759–88 (see Figure 9.3).10 Revenues increased sharply between 1770 and 1773, and grew steadily from 1776 on. The ends of the Seven Years War and the American War triggered reductions in spending. The contrast with Figure 9.2 is instructive. In 1763 and 1783 there was no equivalent to the British debt funding, namely a tax increase sufficient to fund the interest on the debt accumulated during the previous war. In the 1760s, tax revenues remained constant. In the 1780s, they grew too slowly, causing debt service to increase. By 1788, as in 1770 and during the Regency, the inexorable compounding of interest brought France to a fiscal crisis.11 Figure 9.3 shows France in the grips of some “unpleasant arithmetic.”12

10

France did not publish the government’s accounts before the 19th century, and the financial archives burned in 1871. The remaining information is scarce, sparse and contradictory. We have constructed the estimates shown here using various published and archival sources (among which Necker (1820), Mathon de La Cour (1788), Marion (1914–31), Clamageran (1876), Compte rendu au Roi (1788), Compte général (1789) for the published sources; Paris, Bibliothèque Nationale, Mss. Joly de Fleury 1437 and 1442, MFF 7749, MFF 4580; Archives Nationales, F4 1082 for the manuscript sources; see also James Riley (1987). 11 Over the whole period, France was at war only one year out of three. The gross-of-interest budget was in balance only briefly in the 1770s. 12 See Sargent and Wallace (1981).

256

Chapter 9: Macroeconomic Features of the French Revolution

700

600

total spending

Millions of livres

500

revenues

400 military

300

civil

200

100 debt service

0 1760

1765

1770

1775

1780

1785

Figure 9.3 Revenues and spending in France, 1759–88. Revenues are depicted with small circles. Expenditures are decomposed as in Figure 9.2, with successive strings of points denoting civil expenditures, civil plus debt service, and total spending respectively. So the vertical distances between successive strings of points for expenditures represent civil, debt service, and military expenditures. Source: see n. 10.

F RENCH F ISCAL B ACKWARDNESS OR O PTIMALITY ? In Figure 9.2, Britain looks like a simulation from a Barro (1979) tax smoothing model, while in Figure 9.3, France does not. It is tempting to compare France’s financial arrangements unfavorably to Britain’s, but theories of dynamic Ramsey taxation instruct us to be cautious about condemning France’s recurrent defaults or praising Britain’s abstinence. These theories have governments offering their creditors state contingent after-tax returns that respond to news about the government’s prospective net-of-interest fiscal surplus. Realization of a positive shock to government expenditures or a negative shock to “exogenous” government revenues results in a lower payoff, and good news in a higher payoff.13 We have not 13 This

is the message of Lucas and Stokey’s (1983) Examples 7 and 8, and Chari et al.’s (1994) Figure 6.

Before the Revolution

257

attempted to match these theories to our 18th century observations on French and British finance. We would have to struggle to reconcile the timing of the French defaults with these theories, in which low after-tax returns are paid on government debt at the starts of “wars.” Perhaps it could be argued that the French defaults were in response to reckonings that the prospective revenues expected to accrue with French War victory had evaporated with defeat. It would require much more work to coax from these theories an understanding of why France refused to default in 1789, despite its earlier intermittent defaults.14 In 1789 modernizing elements in France did not regard past and prospective government defaults as part of an optimal fiscal arrangement. They hoped to reform fiscal institutions to rid France of those defaults, and this is one of the reasons they welcomed the king’s call to the Estates General. S NAPSHOT OF THE O LD R EGIME L AWS France had the appearance of absolute monarchy. The king created law by Edicts and took executive actions by Arrêts du Conseil, although not all laws emanated from the king. There were also legal traditions, with forms of common law and written law varying from province to province which pre-existed royal edicts. The king confronted the power of a dozen Parlements, with the Paris court preeminent, which verified that new edicts were consistent with the existing body of laws in its jurisdiction. After a law passed muster, it was registered by the Parlements and enforced in court. Parlements also functioned as courts of law and courts of appeal. The Parlements had the right to remonstrate, that is, to send a written exposition of their misgivings about a law, and to request that the law be reconsidered. If the king persisted, the Parlement could issue “iterative remonstrances.” The next step was for the 14 Although

Britain did not default, it did issue securities callable at the government’s option. Such securities expand the range of contingencies against which a government can insure and tax smooth. Equilibrium tax-smoothing models with a government constrained to issue only callable securities have not been worked out, so we don’t know how closely government callable debt would approximate a setting with a complete set of state contingent securities. Also see Grossman and Huyck (1988).

258

Chapter 9: Macroeconomic Features of the French Revolution

king to summon and preside over a special Parlement session called a lit de justice. When the king sat in Parlement, he could compel registration of a law. Further resistance from the Parlement was met by force, exile or prison. There was no written constitution, yet legal scholars spoke of one and disputed it. An uneasy sense of balance existed between the king’s divine right to rule and the secular necessities of legal continuity. France presented a paradox. The absolute power of the monarchy was acknowledged while institutions were reinforced that effectively constrained centralized authority. O FFICES Positions in the Parlement were sold by the king and were inherited or re-sold by the “owner” with tacit royal consent and payment of a transaction tax. Holders of parliamentary offices, as well as other venal offices of justice, were known as the noblesse de robe. They formed a cohesive class united by their common arrangements with the Crown. Beyond the noblesse de robe there was a larger class of officers, more than 50,000 by the time of the Revolution, who also owned their offices. These positions in the judicial, police, administrative, and fiscal systems were properties owned and bequeathed by the officers. An officer’s wages represented the interest on the sum, called the finance of the office, which he or his predecessor had paid to acquire the office.15 The finances of all offices formed a component of the public debt which could not be altered or diminished without major institutional changes.16

15 Office

owners did not welcome increases in wages because they forced the owners to increase the finance or forfeit the office. 16 An officer’s compensation often went beyond the wages themselves to include all forms of pecuniary and non-pecuniary rewards: feudal-like dues, legal or illegal bribes, prestige, tax exemption, and that most coveted prize, nobility itself (see Bien 1987.

Before the Revolution

259

TAXES Royal taxes under Louis XIV consisted of the taille, a direct tax of medieval origin from which the nobility was exempt, and a vast array of indirect taxes on consumption and movements of commodities across France’s numerous internal borders (droits, traites, aides, gabelles). The king collected various feudal dues and revenues from the monopolies that he owned including coin minting, salt, postal service, tobacco, parts of foreign trade, gunpowder manufacture, and the lottery. The taille was collected differently in various regions. Within the kingdom’s historic core, a system of direct allocation of tax liabilities by royal officers had evolved.17 The monies were collected by the receveurs, venal officers who pledged to pay the Treasury the amounts assessed according to a specific timetable. They regularly gave the Treasury rescriptions or claims on themselves due a year hence, which the Treasury resold at discount. The receveurs earned interest on the funds collected and also received a commission on the taxes collected.18 Notes issued by the receveurs and billets issued by the Fermes Générales were called anticipation notes. These notes formed the floating debt system, and also allowed private parties to transfer funds to and from Paris in the absence of a developed banking system. More “recently” absorbed territories such as Brittany, Burgundy, Provence, Artois (15th century), Béarn, Foix, Bigorre (16th century), Franche-Comté, and Flanders (17th century) negotiated their global tax dues with the king through the provincial Estates which met regularly and were charged with collecting tax monies. Other direct taxes (the capitation in 1695) followed the same mode of assessment as the taille. Repeatedly throughout the 18th century, attempts were made to impose income taxes without respect to status or privilege. These attempts failed both in terms of the modest amounts collected in 1789 (12% of revenues) and the persisting irregularities in assessment. Indirect taxes and monopolies were leased to syndicates of individuals who contracted with the king. Tax farmers collected 17 These 18

provinces were called pays d’élection. See Bosher (1970) for details.

260

Chapter 9: Macroeconomic Features of the French Revolution

taxes within the limits of the law in exchange for a specified amount of money for each year of the lease. The lease stated a minimum payment to the farmers and a rule for sharing the excess over that minimum between the king and the farmers. After 1726, the Fermes Générales collected all consumption taxes and internal duties, and managed the salt monopoly.19 A last form of tax collection called the régie was collected by government employees who received a salary and a commission. Louis XIV attempted but failed to use the régie. It was reintroduced first in the 1750s as a tax on leather, and later to manage the king’s feudal rights and the royal demesne. By 1789, 25% of fiscal revenues were collected through régies, and 35% through farms. The remaining direct taxes were collected through the receveurs and provincial officers. Taxation was not uniform across classes, occupations, or provinces. Specific groups often had the opportunity to pay an abonnement or waiver of a new tax or duty. The clergy paid taxes by recurrently conceding a don gratuit (free gift).20 By 1789, the annual average of the clergy’s tax bill was ten million, or less than 2% of the state’s fiscal revenues, even though the Church owned 10% to 15% of the property in France. Cities, provinces, professional corps, the royal princes, and religious associations also benefited from the waiver system.21 The state’s liabilities and commitments included favors, gifts, and pensions bestowed by the king. Favors often took a symbolic form. Historians marvel at how Louis XIV had the nobility competing for the honor of holding his shirt while he dressed. Other favors were more costly. Spending on pensions grew at an even 19 As

a variant of tax-farming, the financiers paid a capital sum corresponding to the present value of a tax. 20 Individuals or communities could negotiate remises, décharges, modérations, indemnités: exemptions or special rebates. 21 A map of the prices at which salt was sold by the Fermes Générales, which managed the monopoly, showed a mosaic of exemptions. Some were due to the fact that a region produced salt itself (Brittany and the Atlantic Coast). Others had historical explanations. Prices varied from 1.5 livres for a hundred pounds of salt to 62 in Burgundy (but within Burgundy 20 villages enjoyed prices lower by a half or more). The increase in price when crossing from Brittany to Maine was thirty-fold. This created opportunities for smuggling.

Before the Revolution

261

faster rate than on defense during the reign of Louis XVI. The Revolutionaries were astonished by the thousands of names on the pension lists, as well as by the multiple secret favors consigned in the Red Book, and indignantly published by the Assembly. Other favors took the form of exchanges of lands, purchases of fiefdoms, and so on. A POLOGY FOR THE O LD R EGIME The Revolutionaries considered this structure to be antiquated and inefficient. But it served a purpose. The French nation had been formed over the centuries by granting concessions as well as by conquest, inheritance and acquisition. Local customs and traditions, privileges and hierarchies could not easily be overturned or ignored. Incorporating the local power structures into the French realm made it easier to win acceptance of the new rule. F ISCAL P RESSURES TO R EFORM As wars expanded across continents and oceans, costs rose, and they were longer funded in the traditional medieval way of raising sufficient taxes during wartime to cover expenditures. Figure 9.4 compares the history of military expenditures in France and in England. The wars of Louis XIV (between 1688 and 1714) occasioned the first large-scale recourse to borrowing in the form of publicly sold annuities. This began 75 years of increasing reliance on marketable public debt, bringing the debt/GNP ratio to 63% in 1789. By 1789, debt issued on the market represented close to 80% of the state’s liabilities. Table 9.1 shows that the three categories of perpetual debt, life annuities, and securities (fixed-term loans and floating debt) each represented 26% of the state’s liabilities, while the finance of all offices represented 735 million or 17% of the total debt, or 21% if we add bond monies. Credit was useful to France’s foreign policy, but difficult for a sovereign monarch to obtain. An obstacle to getting credit was the sovereign’s monopoly of force and his power over contract enforcement. This problem was analyzed by Bulow and Rogoff (1989) and Chari and Kehoe (1993). Chari and Kehoe’s analysis is driven by the calculation that so long as a government does not

262

Chapter 9: Macroeconomic Features of the French Revolution

Table 9.1 S TATE OF D EBT IN 1789

Capital outstanding 1. Rents: Perpetuals Life annuities 1,154.0 2. Fixed-term loans 3. Anticipations 4. Bond monies 5. Financed offices Total (May 1789) Ratio to GNP Ratio to revenues 6. Offices 7. Feudal rights 8. Debts of clergy 9. Debts of abolished institutions Total (November 1789) Ratio to GNP Ratio to revenues

1, 136.0 105.0 830.3 322.3 201.8 119.2 3, 763.6

Debt service 56.8 111.5* 15.8 10.2 5.0 304.3

63.3% 64.4% 616.7 130.0 95.0 228.4 4, 816.9

9.7 ... 4.75 11.4 325.4

81.1% 68.9%

Source: Compte général (1789, 44–48, 50–54, 82, 98–99); Archives parlementaires (1862–, 10:96, 72:199, 41:145–52, 30:329). * of which 34.7 interest payments.

lose access to a means of saving, the punishment of not being able to borrow in the future is not sufficient to deter a default: if it can lend the government can tax smooth almost as well as it can by borrowing. From the standpoint of the analysis of Chari and Kehoe (1993), the mystery is that the king could borrow as much as he did. Nevertheless, the debt sustainability problem preoccupied the king’s ministers. Necker (1784, 3:155) wrote in 1784: The absolute power of a monarch and full public trust are two notions which need intermediaries to be perfectly conciliated. This authority is in France subject to certain restraints when it comes to an increase in the sovereign’s revenues, since the laws which levy

263

Before the Revolution

450 400

Millions of livres

350 300 250 200 150 100 50 0 1690

1700

1710

1720

1730

1740

1750

1760

1770

1780

1790

*: France

Figure 9.4 Military spending in Britain and France, 1688–1789. Sources: for Britain: Mitchell (1988); for France: Mallet (1789), Forbonnais (1758), and the references listed in n. 10.

new taxes must be registered in the Parlements, and these courts can then enlighten the monarch’s justice by their remarks; but a simple arrêt du Conseil or a ministerial order authorized by the sovereign are enough to suspend reimbursements or impose a reduction in interest.... Therefore one can rekindle or sustain public trust only by giving reassurances on the sovereign’s intentions, and by proving that no motive can incite him to fail his obligations. The last sentence is a succinct definition of “credible policy.”22 However, several pieces of evidence indicate that France had somehow solved the sovereign debt problem. First, in the late 1780s, the king had accumulated large debts and devoted over 50% of the state’s revenues to servicing them. Second, in the late 1780s the 22 James

Conklin (1993) studies this problem for the finances of the Spanish monarchy in the 16th century, and adduces the presence of punishments that the king’s creditors could administer to supplement the act of refusing further lending.

264

Chapter 9: Macroeconomic Features of the French Revolution

king adopted extraordinary measures to refrain from defaulting and to acquire more flexibility in servicing his debt, most significantly by convening the Estates General. Something impelled him not to default.23 Third, bond prices in the late 1780s were not especially depressed.24 Fourth, the Revolution would strain to honor the king’s debts, and would default only after eight years of trying and failing with a variety of debt management schemes. Up to 1789, France’s fiscal arrangements had evolved unevenly, and the ability to adjust taxes did not match the king’s plans to service his debts. As we have seen, France lived with an array of institutions whose purpose it was not only to provide funds to the king, but also to constrain him. These constraints impinged particularly when it became desirable to raise taxes to service debts. The history of repeated defaults prior to Louis XVI is a symptom of France’s failure fully to implement the commitment structures and tax system required to support a British-like fiscal policy. Necker (1784, 3:157) observed, “suspension of payments is much less a cause than a result of the lack of trust.” Institutional constraints shaped policies, much as they were designed to. When finance ministers pushed against the constraints, the constraints held firm. S ENTIMENT FOR C OPYING B RITAIN Figure 9.4 shows that relative to its population and national output, wars cost Britain more than France. The average annual cost of these wars typically amounted to 1 to 1.5 years of revenues in Britain, compared to 0.5 to 0.8 in France. The Seven Years War, which proved so costly and disastrous to the French, cost 40% more to the British, although England’s population was a third of 23 The

government seemed to care about its creditors more than it should according to representative agent models of sustainable government debt. But relative to models with agents who differ in their holdings of government debt, representative agent models overstate how difficult it is to sustain government debt. See Rogers (1986) for a study of how heterogeneity impinges on credibility of capital taxation. 24 See Velde and Weir (1992). The period from 1764 to 1770 shows a steady increase in the market discount on government loans, up to yields of 10%; in contrast, up to August 1788, only the most recent loans were discounted, only up to yields of 8% while perpetuals remained under 6.5%. The suspension of reimbursements of August 1788 affected fixed-term loans but perpetuals remained stable.

Before the Revolution

265

France’s, and its income half. As Necker noted in 1781, “England still today [in the midst of a war] can find 300 million to borrow at 3% each year, and exerts amounts of efforts and power out of proportion with its wealth and population.” It is known that French officials understood the case for smoothing taxes. In the 1770s Montyon, a senior civil servant in the French Finance Ministry, wrote:25 Great Britain finances by taxation neither all nor part of the costs of war, it finances them by loans and increases the annual tax burden only by the amount necessary to face the interest and redemption of the loan. That is the regime that France must adopt, and will adopt sooner or later because its value is only too obvious, and our own mistakes will force us to return to this policy. In war-time it is our habit to increase taxes, at a time when perhaps they should be decreased. Indeed in wartime the country suffers enough from the labor withdrawn from agriculture and manufactures to be sent into the army, the navy, and into the production activities necessitated by war. French policy makers understood the advantages of Britain’s constitution. They also realized that establishing a permanent legislative body26 would not only diminish the power of the central government but also that of a multitude of provincial bodies, officers, and single- interest groups and corporations. As Montyon put it: “Which European states now enjoy the soundest credit? Those where the authority of a single man is less prominent, and we cannot disguise the fact that Holland and England have a great advantage over France by their constitution.... If we faced the sad alternative of sacrificing the [French] constitution to finances or finances to the constitution, we should not hesitate to choose the former.” 25 These

citations come from the manuscript of a book on the public debt, kept at the Archives de l’Assistance Publique, Paris (Fond Montyon, Fossoyeux 101, carton 22). 26 The only institution that came close was the Estates General, which had last met in 1614.

266

Chapter 9: Macroeconomic Features of the French Revolution

A T EMPTATION R ESISTED A French leader had once contemplated trying to install a British financial “technology.” After the disastrous last wars of Louis XIV, the Duke of Orleans acted as regent for the young Louis XV from 1715 to 1723. Orleans was tempted to initiate a process of change. Plans circulated in Versailles to establish a permanent version of the Estates General. The regent decided against it because he was advised that to do so would unleash uncontrollable forces. He then compromised with the Parlement and presided over the defaults associated with John Law.27 Until the start of the Revolution, France remained under a nominal absolute monarchy encumbered by a tax system in need of reform. S USTAINABILITY OR B AD C HOICES ? It has been argued, most recently by Riley (1986), Simon Schama (1989), and Eugene White (1989), that the fiscal problems of the Old Regime in France were not the consequence of fatally wounded institutions, but of particular choices made by the king’s ministers; that different and wiser policy actions had been feasible; and that failure to take wiser actions indicated the incompetence or competing interests of the policy-makers. Such arguments forget that the French monarchy was not really absolute, and that social institutions put a host of participation and incentive constraints on people. French finance ministers understood arithmetic and compound interest.28 They tried to emulate the fiscal policies of Britain, but faced different constraints on their actions. Their failure to achieve outcomes that would have been attainable had they more authority to make commitments reflects a deficient “commitment technology,” or the existence of mechanisms designed to protect other commitments. 27 Louis

XIV, whose first years of reign were troubled by the Paris Parlement’s unrest, had imposed his will on the Parlements, and deprived them of the right to remonstrate. When he died and left the regent’s powers severely curtailed in his will, the regent bartered with the Parlement: the court would declare the will null and void, and he would return the right of remonstrance. 28 See also Velde and Weir (1992) on the financial competence of French governments.

The Search for a New Order

267

T HE L AST F ISCAL C RISIS OF THE O LD R EGIME The reign of Louis XVI had started auspiciously in 1774 because the defaults and tax increases imposed by Terray in 1770–72 had left the new finance minister Turgot with close to a balanced budget. The 20-year old monarch pledged never to default on the public debt. However, Turgot’s aversion to military adventures and efforts to implement far-reaching reforms placed him at odds with vested interests which eventually led to his downfall. Rather than follow Turgot’s advice, France persisted in its old pattern of engaging in wars without adequate provisions for financing them. Turgot’s successors Necker and Calonne financed the war primarily through borrowing and inconspicuous tax increases. The treaty of Paris in 1783 left the government with a large peacetime gross-of-interest deficit. Calonne tried to delay the inevitable with more borrowing, but in 1786 he announced to the king that the current fiscal course was not sustainable and proposed a major tax reform. Calonne attempted to bypass the Parlements and failed. He was dismissed and had to flee to England to avoid the Parlement’s indictment in 1787. He was replaced by Loménie de Brienne, who unsuccessfully proposed another package of loans and tax increases. As the summer of 1788 began, the government faced stark alternatives: either repeat the operations of 1770–72 by abolishing the Parlements, increasing taxes, and defaulting on the debt, or foster wide national consensus to undertake reform on a larger scale. Unpleasant arithmetic and a reluctance to default prompted King Louis XVI to awaken a dormant institution, the Estates General, which 60 years earlier the regent had let sleep.

The Search for a New Order A NNUS M IRABILIS (1789) The French Revolution sought a combination of institutions that would allow France to make the fiscal conversion achieved by Britain in 1688. Necker explained the alternative chosen by the king when he spoke to the Estates General on May 5, 1789 (Archives Parlementaires, 8:14–15).

268

Chapter 9: Macroeconomic Features of the French Revolution

And please note, gentlemen, so that you may feel even more love for your gracious monarch, you do not owe the precious benefit of being assembled here in Estates General by H.M. to an overwhelming financial distress. Indeed, for the greater part, the measures we have presented as ways of eliminating the deficit were always available to the king ... and supposing, if you will, implausible impediments, what resources would authority not have, had the king only been concerned with his finances, had he followed the ways several predecessors have shown him, and absolved himself from all or part of the various obligations whose extinction would have considerably alleviated the Treasury. Therefore, gentlemen, it is to H.M.’s virtues that you owe his enduring determination in planning and wishing for the Estates General. He would have solved his financial difficulties without their help, had he not taken a great interest in upholding property rights, preserving rewards earned in his service, respecting the claims of misfortune, and hallowing all the commitments made by the rulers of a nation true to its honor and its word, ... H.M. has seen that the people, concerned with the financial difficulties and the state of credit, were aspiring to a return of order and trust that would not be short-lived, nor tied to the contingencies of our current ordeal. ... He has reflected that, in order to reach such a worthwhile goal, it was necessary to summon new keepers of public peace, and put so to speak the order of finances under the whole nation’s supervision. These statements were true. The government could have followed Terray’s example of 1770 and balanced the budget with an equivalent combination of default and tax increases. It did not and instead sought consensus to change fiscal arrangements. The market for government bonds at the time, even before the call of the Estates General, indicated that expectations were nothing like those in 1769–70, when anticipation of Terray’s measures sharply drove prices down. It is tempting to interpret the bond market data

The Search for a New Order

269

and the tone of Necker’s remarks as reflecting an emerging understanding between government and governed, and debtor and creditors, that France could not prevail in international struggles, nor could enterprise and commerce prosper, with a government prone to defaults. S OME A DVERSE AND P ERSISTENT R EVENUE S HOCKS Soon after Necker’s speech, the Court understood the depth of the aspirations aroused by the Estates General, estimated the costs of reform to be unacceptable, and tried to retreat. The Court failed to arrest the momentum gathered behind the forces of reform. Popular pressure and violence prevailed, and in July 1789 the Estates were transformed into a National Assembly empowered to “regenerate” France. The collapse of central authority had fiscal consequences. In eager anticipation of the new order yet to come, castles of feudal lords, tollhouses and residences of tax collectors were burned across the country. On June 22, 1789, the Assembly decreed that all current taxes would become illegal.29 Partly to contain the breakdown of the old order, the Assembly took drastic steps. On August 4, 1789, the Revolutionaries terminated the feudal order by abolishing all privileges, including financial ones, and suppressing all offices. However, the Assembly did not intend to default on the state’s obligations to the officers. It regarded offices as properties, and intended to repurchase them. Table 9.1 shows how reimbursements of offices appeared on the balance sheet of the government. The Assembly transformed the implicit interest component of the officers’ salaries into an explicit interest payment due on the capital owed to the former officer. On August 4, the Assembly eliminated many of the institutions which the Old Regime used for tax collection and short-term credit. Figure 9.5 shows the consequent, unprecedented drop in fiscal revenues. The Assembly embraced a paradox. Although publicly committed to honoring the debt, political circumstances forced it to eradicate revenues without diminishing its obligations.

29 Many

people ignored the provision in the decree stating that those taxes would remain in force until the promulgation of a Constitution.

270

Chapter 9: Macroeconomic Features of the French Revolution

2.5

1726 = 1

2

1.5

1

0.5

0 1740

1760

1780

1800

1820

1840

Figure 9.5 Index of real per capita revenues (excluding revenues from currency creation) in France, 1726–1845. Sources: the sources listed in n. 10; Nicolas (1882) for the nineteenth century; and Dupâquier (1988) for population estimates; the population for 1795–1815 is adjusted for current boundaries.

D ESIGNING A N EW TAX S YSTEM Simple rules, easy collection, transparent accounts —The deputy La Rochefoucauld, chairman of the Taxation Committee, Aug. 18, 1790 (AP 18:145).

The Assembly discussed tax reform in terms of Physiocratic theories and was willing to take the time required to implement a neutral tax system designed to distort relative prices minimally.30 The Assembly relied heavily on the contribution foncière and abolished consumption taxes because it wanted to tax pure rents. The administrative instructions on how to implement the foncière specified that the piece of land itself was really being taxed, and that the owner was but a proxy for the purposes of that payment. 30

Reforming tax laws and tax administration was the preeminent financial tasks mandated by the cahiers de doléances.

The Search for a New Order

271

Administering the foncière would require compiling and continually updating a complete register of all plots in the kingdom. The failure of the Revolution’s tax system traces to the fact that rather than establish a new central administration, the Constituants relied on the newly created local administrations which had scant incentives to be helpful. The shortage of tax revenues later determined the destiny of the assignat. Figure 9.5 shows that revenues took a long time to return to pre-revolutionary levels. Per capita taxation in the French Empire did not reach pre-Revolution levels until 1810. Thus tax reform did not take effect rapidly enough to solve the debt problem. B IRTH OF A C URRENCY In response to a motion by the least ecclesiastical of bishops, Talleyrand, the National Assembly placed the Church’s assets “at the Nation’s disposal,” giving the state a “privatization problem.” Necker and the Constituants planned to solve the privatization problem and the debt problem by creating a new currency. They devised a scheme to raise revenues by auctioning the confiscated lands, thereby withdrawing paper notes issued on the security of the lands sold by the government. This “tax-backed money” scheme propelled the State into the domain of monetary experimentation. Records of their debates show how members of the Assembly marshaled theory and evidence to assess the likely effects of their innovation. They quoted David Hume and Adam Smith and cited John Law’s System of 1720 and the American experiences with paper money fifteen years earlier as examples of how paper money schemes can go awry. Necker’s original plan embodied two components: a national bank and a new financial instrument, the assignat. Necker’s national bank was patterned after the Bank of England. He proposed to transform the Caisse d’Escompte into a national bank by granting it a monopoly on issuing notes and marketing government debt. The Caisse was a discount bank founded in 1776 whose main function was to discount commercial bills and issue convertible notes. Although independent of the government in principle, it had occasionally been used as a source of loans. Its notes had been

272

Chapter 9: Macroeconomic Features of the French Revolution

declared inconvertible in August 1788, and by the time of Necker’s proposal, its reserves were exhausted. Necker’s plan placed the National Estates (as the Church lands became known after the addition of the royal demesne) at the center of the financial picture: a “Bank of France” would issue a 5% security mortgaged on the prospective receipts from the modest sale of some 400 millions’ worth of National Estates in the years 1791 to 1793.31

Must we hand the State over to a syndicate of publicans? —The deputy Rewbell, Dec. 19, 1789.

The deputies split Necker’s plan by accepting the assignat but not the bank. They refused to establish a bank mainly because they understood it to be a commitment mechanism. A masterful lecture by Dupont de Nemours on the real bills doctrine explained how a bank would operate. Bankers such as Necker, Lecouteulx, and Laborde backed the proposal,32 but the deputy La Rochefoucauld attacked the idea of a bank on the grounds that “the funds it would lend to the State, and the services it would seem to render, would force the government to show regards it would not have to display toward a multitude of individual enterprises” (emphasis added; AP 10:673) The chairman of the Finance Committee Anson said: “Let us never overlook the fact that the various forced paper-monies issued in a kingdom by the sole authority of a monarch or his cabinet, after having taken the nature of an injustice from their origin, will encounter resistance in their use, and their easy multiplication offers countless abuses; but all these flaws disappear when a papermoney is an emanation of the general will. Who among us would dare question its value? That would be mistrusting ourselves” (AP 12:606). 31 Only

170 million was to be used initially to cover the deficits of 1789 and 1790. The British funding system was also based on the word of the Parliament; but the Parliament entered into a balance of power with the executive, while the Constituent Assembly did all it could to diminish the executive power (the king and his cabinet) which it did not trust. 32

The Search for a New Order

273

To end in a day the work of a half-century —The deputy Montesquiou, chairman of the Finance Committee, Aug. 27, 1790.

By mid-1790, members of the National Assembly had agreed to sell the National Estates and to use the proceeds to service the debt33 in a “tax-backed money” scheme.34 The government would issue securities with which it would reimburse debt. The securities were acceptable as payment for National Estates purchased at auctions; once received in payment, they were to be burned. (The Appendix describes the mechanics of the auctions.) The Estates available for sale were thought to be worth about 2,400 million, while the exactable debt (essentially fixed-term loans, unpaid arrears, and liquidated offices) stood at about 2,000 million. The value of the land was sufficient to let the Assembly retire all of the exactable debt and thereby eliminate the interest payments on it. After lengthy debates, in August 1790, the Assembly set the denomination and interest rate structure of the debt. These debates foreshadowed many issues that were later to be studied by Milton Friedman (1948) and Bryant and Wallace (1979). Proponents of “bonds” (large-denomination, high-interest liabilities) contended against advocates of issuing “money” (low-denomination, lowinterest liabilities).35 Bond advocates asserted that it would be 33 The

current weight of the debt on government revenues was over 60% when due reimbursements were included. 34 Two distinct aspects of monetary theory help in thinking about the assignat plan. First, a system beginning with a commodity standard typically has room for a once-and-for-all emission of (an unbacked) paper currency that can replace the commodity money without generating inflation. Sargent and Wallace (1983) describe models with this property. That commodity money systems are wasteful underlies Milton Friedman’s (1960) preference for a fiat money regime over a commodity money. Second, in a small country on a commodity money system that starts with restrictions on intermediation, those restrictions can be relaxed by letting the government issue bank notes on the security of safe private indebtedness, while leaving bank notes convertible into gold at par. See Adam Smith (1776) and Sargent and Wallace (1982) for expressions of this idea. 35 The National Assembly debated many now classic questions in monetary economics. Under what conditions would money creation generate inflation, with what consequences for business conditions? Distinctions were made between issue of money to pay off debt, on one hand, and monetization of deficits, on the

274

Chapter 9: Macroeconomic Features of the French Revolution

wise to isolate these “credit” operations from monetary arrangements, and cited it as an advantage that, by issuing bonds to pay off the national debt, the operation exchanging the National Estates for the debt would involve only financially sophisticated people.36 Advocates of money asserted that lower denominations would bring about lower interest costs. They also recommended low denominations as a means of involving many people in the swap and binding them to the Revolution. They asserted that a low-denomination assignat would provide France with a new monetary instrument, and do for its depressed economy what bank notes seemed to do in Britain: ease credit, lower interest rates, and facilitate trade. Initially, the arguments for large denominations prevailed. The initial 400 million issued in April 1790 was in denominations of 200 to 1,000 livres, as were the Caisse d’Escompte notes. In the plan adopted in September 1790, the lowest denomination was 50 livres, and 35% of the notes were ultimately issued in denominations between 50 and 100 livres.37 Within a few months, the denomination would be lowered to 5 livres and again to 0.5 livres in 1792.38 The debates did not anticipate the speed with which private intermediaries would emerge to undo the government’s choice of large denominations.

Rise and Fall of the Assignat Figures 9.6, 9.7, 9.8, and 9.9 are a biography of the assignat. Figure 9.6 shows how the assignat was ultimately to be used in its other. Would assignats be akin to notes emitted under a real bills regime, and cause loss of specie, or would they circulate alongside specie, thus increasing the money stock? Would inflation affect real wages? How would it impact foreign trade, competitiveness of French industry and agriculture, balance of trade, foreign exchange? 36 It had been conventional to issue bonds in denominations of 500 or 1,000 livres. 37 As a point of comparison, per capita GNP was around 200 livres; a day’s wage for unskilled labor was 1 livre. Adam Smith advised against letting banks issue notes smaller than £ 5, or 115 livres. 38 Since gold coins were of 24 and 48 livres, and silver coins mostly of 3 and 6 livres, the assignat was to substitute for gold specie at first, and then silver; in the end it also replaced copper and billon.

275

Rise and Fall of the Assignat

3500 3000

Millions of livres

2500 2000 Total issue

1500 1000

Debt liquidation

500 0 1790

1790.5

1791

1791.5

1792

1792.5

1793

Figure 9.6 Assignats used for debt reimbursement and total issue, 1790–92. Source: Archives Nationales, Paris, AD IX 497bis and 586.

two capacities: debt redemption and deficit financing.39 Figure 9.7 plots logarithm of the inflation rate against real balances. Figure 9.8 records real balances of assignats from 1790 to 1796. Figure 9.9 records measures of the price level.40 We have partitioned Figures 9.7, 9.8, and 9.9 into three periods, corresponding to different monetary regimes or episodes. The three clouds of points in Figure 9.7 depict different real balanceinflation relationships. Only the cloud for the third period has the inverse relationship familiar to us now from twentieth-century 39 The

assignat was adopted on September 29, 1790. The first debts to be reimbursed were the abolished offices and the financiers’ short-term notes, which reflects the Assembly’s insistence on respecting past commitments. 40 Two measures of the value of the assignat are used: one is its price against gold, as quoted on the market; the other is an index of its value against land, gold and commodities, called the Caron index. In spite of the Terror, trade in specie continued in Paris, and prices are available for the entire period. Another measure relies on depreciation tables published after the demise of the assignat, to implement a law on debts contracted during the inflation (see below). These tables, one for each département, in principle measure depreciation against most consumption goods and land, as well as gold. The index used here, referred to as the Caron index from the name of the editor of the tables (Caron 1909), is an average across the 83 départements within the 1791 borders.

276

Chapter 9: Macroeconomic Features of the French Revolution

hyperinflations. The first period ends in the late summer of 1793, and is characterized by growing real balances and moderate inflation. The second period begins and ends with the Terror. It is marked by high real balances, around 2,500 millions, and roughly stable prices. The fall of Robespierre in late July 1794 begins the third of our episodes, in which real balances decline and prices rise rapidly. We interpret these three episodes in terms of three separate theories about money: a “backing” or “real bills” theory (the text is Adam Smith [1776], a legal restrictions theory,41 and a classical hyperinflation theory.42 We view these theories not as competitors but as alternative collections of “if-then” statements about government note issues, each of which finds its conditions more nearly met in one of these episodes than in the other two. The price level rose gently through December 1792 (in this month the assignat was discounted 30% vis-à-vis gold), to the accompaniment of rapidly growing real balances of assignats. The assignat displaced specie, causing it to leave the country. From January 1793 to the summer, however, the growth of paper money accelerated to almost 9% per month, and the price level rose. The Terror stopped this incipient inflation, and caused the price level to fall sharply and money growth to slow down to about 3% per month. The last period began in August 1794. The inflation rate suddenly accelerated to reach 60% per month. The last period was a forerunner of the twentieth-century hyperinflations studied by Cagan (1956), Sargent and Wallace (1973), and others. A R EAL -B ILLS R EGIME (1790–92) Bryant and Wallace (1979, 1984) analyzed fiscal implications of alternative denomination structures of government debt, and the incentives that interest rate spreads would give private intermediaries to unravel a government’s decision to restrict its issues to large denominations. Large positive nominal interest rates on 41 Keynes

(1940) urged a forced saving program for war finance. Bryant and Wallace (1984) and Villamil (1988) have formalized aspects of Keynes’s analysis. 42 According to Cagan’s (1956) definition of hyperinflation, beginning in the month that inflation exceeds 50 percent per month and ending in the month before inflation drops below 50 percent per month for at least a year, the assignat experienced a hyperinflation from May to December 1795.

277

Rise and Fall of the Assignat

0.6 0.5 0.4

Inflation

0.3 0.2 0.1 0 −0.1 −0.2 0

50

100

150

200

250

300

Real balances

Figure 9.7 Scatter plot of the logarithm of the inflation rate (log( pt )/pt−1 ) on the vertical axis versus real balances of assignats, on the horizontal axis; the circles depict points from the “real bills” period from January 1791 to July 1793; the crosses are for the Terror from August 1793 to July 1794; the stars trace the hyperinflation from August 1794 to March 1796. Sources: Braesch (1934–36) for 1790–91, reports of the Treasury and the Caisse de l’Extraordinaire for 1791–95 (Paris, Archives Nationales, AD IX 497bis and 586), and Ramel de Nogaret (1801) for 1795–96.

government debt create an opportunity for intermediaries to earn riskless profits by issuing low denomination notes to intermediate government debt. During the first months of the assignat experiment, the government confronted and gradually learned to manage the forces that Bryant and Wallace wrote about. It appears from available data on circulation by denomination that the large denominations were redeemed quickly. By July 1792, 75% of the stock of notes of 500 to 2,000 livres had been redeemed in payments on Estates or exchanged for lower denominations.43 By contrast, almost 98% of the 50 livres notes were still circulating. The interest 43 The

proportion of redeemed notes of all denominations rises abruptly in April 1792, when war was declared.

278

Chapter 9: Macroeconomic Features of the French Revolution

3000

real value

2500

Millions of livres

2000

1500

1000 gold value 500 Terror 0 1790

1791

1792

1793

1794

1795

1796

Figure 9.8 Real balances of assignats (in gold and real terms). Sources: Braesch (1934–36) for 1790–91, reports of the Treasury and the Caisse de l’Extraordinaire for 1791–95 (Paris, Archives Nationales, AD IX 497bis and 586), and Ramel de Nogaret (1801) for 1795–96.

coupons on the first batch of assignats, worth between three and fifteen livres, circulated at a premium until the government issued its five-livre notes in late 1791. More evidence of strong demand for small denomination currency comes from the fact that within weeks of the September 1790 vote, private as well as public banks appeared across the country to provide intermediation services. The National Assembly abstained from regulating the industry, and soon there were hundreds of banks. In exchange for assignats, these banks issued their own currency in denominations ranging from 0.025 to 25 livres. Some banks were privately owned and operated for profit; others were set up by local authorities and cities. Public banks tended to operate with 100% reserves (see White 1990). Some private coinage also occurred, particularly in Paris and Lyon, where merchants minted copper coins redeemable in assignats. The Assembly responded by eliminating its competitors in the

279

Rise and Fall of the Assignat

10 3

Terror

10 2

gold

10 1

price level

10 0 1790

1791

1792

1793

1794

1795

1796

Figure 9.9 Price level and price of gold (log scale). Source: Caron (1909).

business of issuing currency. In June 1791, it approved a plan to issue five-livre notes, thereby placing the assignat squarely in competition with silver, the main specie in France.44 The notes were quickly issued in exchange for higher denominations, and by August 1792 they represented over a third of the total outstanding stock of assignats. The private banks then began issuing smaller denominations, and the billets de confiance (as this private money was collectively known) multiplied.45 Again the Assembly moved to meet the competition from the private banks. In December 1791, notes of 0.5 to 2.5 livres were approved without much debate. But there were delays and difficulties in manufacturing them. For the high denominations of previous issues, production costs had been a very low percentage of face value. The low face value of the new notes caused the government to simplify and cheapen the production process, at 44 Two 45 See

thirds of the pre-Revolution coinage was in silver. White (1990).

280

Chapter 9: Macroeconomic Features of the French Revolution

the risk of making counterfeiting easier.46 In August 1792, the first “small assignats” appeared. The government could now impose a monopoly: within a few weeks, all private banks were forbidden to issue their notes, and private coins were outlawed. The various Caisses Patriotiques, as the money-issuing banks were most often called, were conveniently accused of fraud, and liquidated. For the small copper coinage, the government melted down church bells to mint new coins. Soon inflation made its contribution to lowering the average denomination of the currency. Small denomination assignats tended to remain in circulation and to drive specie out of the country, just as Adam Smith (1776) had predicted. Figure 9.10 shows the price of foreign exchange futures in Paris in the early years of the assignat. The prices remained above specie parity most of the time, providing incentives for exporting specie from France. Buying a few choice church lands is still the best way to invest one’s money. —Marie-Antoinette, Queen of France, to the count of Fersen, June 5, 1792.

Marie-Antoinette’s own assessment of investment opportunities in France two months before the overthrow of the monarchy (on August 10, 1792) attests to the success of the land sales. The sales, which opened in Paris in October 1790, were the pivotal element of the new currency system. In the early months of 1791, sales were rapid and prices high. By November 1791, over 1,500 million worth of Estates had been adjudicated across France. Prices paid for the lands reflected a capitalization rate of 3% to 3.5% applied to the yields on land, which matched rates then being applied to exchanges of rental properties in France. Overbidding in expectation of inflation was thus limited. In May 1790, when the specific arrangements for the sales were voted, it had been decided to offer purchasers of National Estates generous payment terms. A down payment of only 12% to 30% was 46 These

notes were nevertheless the most expensive assignats made, at a cost close to 1% of face value.

281

Rise and Fall of the Assignat

1.3

1.2 London

1.1 Hamburg

1 Amsterdam

0.9

0.8

0.7 1789

1790

1791

1792

1793

Figure 9.10 Prices in Paris of foreign exchange futures, January 1789–July 1793 (end of month). Prices are normalized so that parity is 1 (gold parity for London, silver otherwise). The contracts are 30 days for Amsterdam, 60 days for London and Hamburg. A normalized price of 1.1 means that 1.1 ounce of specie in Paris buys 1 ounce of specie abroad. Source: Bouchary (1937), Bailleul (1796). The gold content of the French livre given by Bouchary (1937, 21) is incorrect after the recoinage of 1785.

required, with the balance payable in assignats and due in 12 yearly installments. The government accepted large claims on private debtors, payable in assignats. This arrangement is interpretable as a version of a financing scheme described by Persson et al. (1987), by which a government eliminates the time consistency problem of Auernheimer (1974) and Calvo (1978)—its temptation to inflate away outstanding nominal claims on it—by acquiring nominal claims on the public. In November 1791 the government tried to shorten the terms of payment to four years for all subsequent sales. This immediately had a negative effect on sales, which forced the government to reverse its decision within weeks. The 12-year schedule of payment (shortened to ten years in 1794) was maintained until 1796. In June 1793, after its policy had evolved

282

Chapter 9: Macroeconomic Features of the French Revolution

from “real bills” to “inflationary finance” under the pressures of war, the Convention voted to sell these claims. There were few buyers. WAR The declaration of war on the German Empire and Austria on April 20, 1792 proved to be a turning point in the Revolution. Most people supported the conflict although isolated individuals such as Robespierre voiced their opposition in vain. War forced changes in the monetary-fiscal experiment. Debt payments were suspended indefinitely, and the assignat was converted from its initial purpose to become the main means of financing the war. When the war started, the assignat had already depreciated 25% in real terms and 40% against gold. As Laffon-Ladébat explained on May 14: The 1,500 millions of assignat balances are virtually the only monetary instruments (signes représentatifs); this is close to 1,000 millions less than the specie in France before the Revolution. Compared to gold, they represent no more than 750 or 800 millions; they are clearly insufficient. Thus it is not the quantity emitted which occasions the assignat’s depreciation, but its nature, and the distrust shared even by men who should reassure the Nation. The high price of gold is not caused by an excessive mass of assignats in circulation, and a larger issue presents no risks as long as you take all measures necessary to restore public confidence and the balance between the various standards of value. The Convention would soon shore up “public confidence” with drastic measures. War started in the summer of 1792, the monarchy was abolished in September 1792, and the king was executed in January 1793. After the king’s execution, France fought against Britain, Holland, Piedmont, and Spain. In Spring 1793, defeats followed one another, officers deserted, armies collapsed, and the assignat’s value fell, threatening the base of the inflation tax. Summer 1793 brought more military defeats and civil war. With the country experiencing chaos, internal rebellion, and external war, the fiscal

Rise and Fall of the Assignat

283

resources of the central government were depleted (see Figure 9.11). The months preceding the imposition of the Terror in Figures 9.9 and 9.10 display signs of an incipient hyperinflation: falling real balances and rising inflation. To arrest the flight from currency, the government created an artificial demand by imposing severe legal restrictions. The government closed the stock market, abolished joint stock companies, and imposed harsh penalties for refusing to take the assignat at par. In September, the Terror began.47 T ERROR (1793–94) Proponents and agents of the Terror have been portrayed as preferring authoritarian methods, and the defeated Girondins have been cast in the role of liberals overwhelmed by the totalitarian Montagne. Leninist writers saw the Terror as a necessary phase of revolution, and took state control of the economy as a model of economic policy independent of the specific circumstances of the French Revolution. Our view is that the Convention embraced extreme measures reluctantly because it faced extreme fiscal exigencies.48 Two of the Convention’s Committees composed of fewer than a dozen men administered the Terror. Legislation suspended most civil liberties and gave broad powers to representatives of the Convention throughout France. The Committees repressed rebellious acts and subjugated all other interests to the imperative of survival of the government in the face of general invasion and widespread rebellion. Economic and political dictatorship merged. L EGAL R ESTRICTIONS Threat of death for offenders enforced parity of assignats with specie. It was illegal to hold such assets as commodities, private financial securities, precious metals, specie, jewelry, or foreign 47 Brezis

and Crouzet (1994) emphasize the distortions produced by the price controls. 48 When the National Assembly adopted a Penal Code in 1791 incorporating the death penalty, Robespierre argued against it. Circumstances, not ideology, propelled Robespierre from being an opponent of capital punishment to being leader of a bloodthirsty Montagne. Likewise circumstances impelled the economic measures instituted during the Terror.

284

Chapter 9: Macroeconomic Features of the French Revolution

200 180 160 Spending

Millions of livres

140 120 100 80 60 40

Revenues

20 0 1791

1792

1793

1794

Figure 9.11 Spending and revenues, 1791–95 (real value). Archives Nationales, Paris, AD IX 497bis and 586.

1795

Sources:

exchange, and markets in them were closed. Holding land was permitted. All assets and bonds comprising the public debt were converted into a single nontransferable perpetual rent title. In response to inflation, grain prices, then consumer prices and wages, and later producer prices were controlled with the so-called laws on the Maximum.49 Currency demand was supported by the laws on the Maximum and restrictions against hoarding. Under the Terror, any citizen accused of violating these laws could expect swift and arbitrary proceedings. The law on parity of the assignat called for arraignment and trial within 48 hours of the offense. The law encouraged denunciations from informants and gave extravagant powers to local authorities to enforce the restrictions. In a few dozen instances, the death penalty was imposed for

49 This

was a massive undertaking: the law of March 1794, called the Second General Maximum, was published in three volumes running 1,100 pages; prices and wages were set for each of the 550-some districts making up France. The law was never applied in much detail.

Rise and Fall of the Assignat

285

crimes against the assignat or for hoarding.50 The debt remained intact. The conversion of outstanding bonds into perpetuals in August 1793 protected the value of these claims against reimbursement in depreciated paper, leaving future governments with a still-overwhelming burden of debt service.51 Market prices for government bonds, available until June 1793, remained high, suggesting that a major default was not deemed likely, even while the political situation deteriorated. The May 1794 conversion of life annuities into perpetual annuities is an example of the Terrorists’ commitment to property rights.52 Cambon faulted the last ministers of Louis XVI for selling life annuities at a single rate, independent of age, which offered higher-than-market returns for most annuitants. Genevan bankers had noticed the loophole and exploited it by buying annuities on groups of young girls, then selling the pooled annuities to investors. The scheme, although legal, was considered unacceptable. One remedy would have been outright default or at least reneging on the Genevan annuities. Instead, Cambon thoroughly investigated the available international literature on life annuities (including impressive mathematical treatments) to compute fair rates of conversion for annuities on all ages, as well as annuities on multiple nominees of various ages. Such care contrasts with the usual portrayal of the Terror as a period of government-sponsored theft and murder, and reflects the French government’s continuing commitment to honor debts from the time of Louis XVI.53 50 The

death penalty for hoarding was suspended in January 1794 after an innocent man was rescued at the last minute by his son’s plea to the Convention. Proclaiming it an honor to save an innocent person, Danton himself carried the decree to the tribunal. 51 This contrasts with the previous episode of inflation in France. In 1719, all government debt was reimbursed in notes of John Law’s Bank. The subsequent inflation, and re-conversion to annuities, performed a 40–50% reduction in debt service. 52 Implementing optimal fiscal and debt management policies requires recourse to long-term bonds with the possibility of conversion at the debtor’s option. France was forced into the use of short-term loans and self-amortizing life annuities, while Britain successfully converted to perpetuals. 53 The Jacobins consistently refused to alter the terms under which the sales had been made; even in the summer of 1795, they remained opposed to the abandonment of the assignat’s legal equivalence to gold in payments for National

286

Chapter 9: Macroeconomic Features of the French Revolution

Figure 9.12 shows seigniorage raised by assignat creation, Mt − Mt − 1 ( pt + pt−1 )/2

(9.1)

where pt is the price level, and Mt is the nominal stock of assignats. Seignorage measures the real revenues raised by printing currency at near-zero cost and exchanging it for goods and services. The average monthly revenue in 1788 is shown for comparison. From April 1792 to August 1795 (when the only remaining belligerents were Austria and Britain), spending amounted to some 4,500 million in real terms, of which over 3,700 million was raised through money creation. Reinforced by legal restrictions, the assignat financed the war. 180 160 140

Millions of livres

120 100 80 60 40 Revenues in 1788

20 0 1791

1792

1793

1794

1795

Figure 9.12 Seigniorage raised on the assignat, in real terms.

Estates and taxes.

1796

Rise and Fall of the Assignat

287

H YPERINFLATION (1795–96) We are crushed under the weight of our victories —The deputy Merlin de Thionville to the Convention, on May 12, 1795.

By the summer of 1794, the war turned in France’s favor. Victory at Fleurus on June 26 led to a swift six-month conquest of Belgium and Holland, and enemies were expelled from all French territory. By May 1795, peace treaties had been signed with all powers except Austria and Britain. The war no longer justified the repressive economic and political apparatus of the Terror, and the Jacobin party was overthrown. Robespierre was executed on July 28, 1794. The last significant remnants of the party (including Cambon) were purged in April 1795. On his way to the guillotine, the fallen dictator was jeered with the cry, ”Foutu Maximum!”54 Within two months, it was evident that efforts to enforce the Maximum were futile. Legal restrictions were ignored, although the laws were not repealed until January 1795. A police report in early October 1794 notes : “The majority of the people wants complete freedom of trade again; it thinks that supply will then be plentiful, that the price of commodities would rise at first but then fall with competition.” A few weeks after Cambon’s arrest, trade in specie became legal and the commodities, foreign exchange, and financial markets reopened. In June 1795, the government ceased to value the assignat at par for receipts, payments on National Estates, or expenditures. Instead, the government specified a rule according to which currency was valued as a function of the current money stock (each increase of the stock by 500 million triggered a 20% devaluation). A month later, creditors were relieved of the legal obligation to accept reimbursements in paper. The assignat had become a full-fledged fiat currency, severed from specie. As the apparatus of legal restrictions disappeared, the demand for the currency fell, and the assignat became known as “Parisian money.” Police reports attest that by the fall of 1795, even Parisians 54 Rough

translation: There goes the f——g Maximum.

288

Chapter 9: Macroeconomic Features of the French Revolution

230 220 210 200 190

silver parity

180 170 end of the assignat 160 150 140 130 Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep 1795 1796

Figure 9.13 Price in Paris of Hamburg currency delivered in Hamburg, March 1795–September 1796 (daily); silver par=186.7. A rate of 170 means that 1 ounce of silver in Paris buys 1.1 ounce in Hamburg, affording a 10% profit (inclusive of transport costs) from shipping the silver from Hamburg to Paris. Source: Bouchary (1937).

were deserting the assignat in favor of hoardable commodities, a portfolio adjustment labeled “speculation”(agiotage) by an irate government. People joked that if the walls of Parisian houses were to tumble, the starving population would be smothered in food and drowned in wine. Specie was also in great demand, and there is evidence that gold appreciated relative to assignats, particularly in exchange for manufactured goods. In terms of gold, prices were lower than in 1790, creating trading opportunities for the savvy. A Swiss visiting Paris hastened to change his gold for paper, bought hundreds of shoes, stockings, and hats, shipped them off to Switzerland and lived in Paris like a king for a month. Foreign exchange prices corroborate these anecdotes by showing that in the fall of 1795, silver was 20% cheaper in Paris than in Hamburg, a signal prompting imports of specie into France. Figure 9.13 shows

Rise and Fall of the Assignat

289

the price on the Paris market of a futures contract on German marks, a claim to a fixed amount of specie in Hamburg 60 days hence. T HE G OVERNMENT ’ S R ESPONSE The Thermidorian Convention bequeathed a hyperinflation to the Directory in October 1795. Faced with a dwindling inflation tax base, war, and limited fiscal resources, the new regime experimented. It briefly reenacted Jacobin measures against hoarding. In December 1795, financial markets were again closed and denounced as a “den of thieves and conspirators, who plot to starve us back to monarchy.” The government reopened the markets three weeks later when it wanted to know the market price of its money. The government implemented a mandatory open-market operation, called the Forced Loan (December 10, 1795). Taxpayers were required to turn over a specific quantity of assignats in exchange for interest-bearing coupons admissible in payment of future tax liabilities. The assignats were accepted at a 100:1 ratio, and the government expected to collect 600 millions (in real terms) within ten weeks. The Directory’s inability to collect taxes efficiently doomed the plan: it took five months to collect 50 millions. The assignat’s price lifted briefly from 0.5% to 0.7% of face value, then fell even further. Nothing could save the assignat. Ramel-Nogaret, the new finance minister, said “it was impossible to maintain the assignat’s initial credit, if it is true, as experience has taught us, that any currency (un signe quelconque) may lose its value not only proportionately to increases in quantity, but even at a progressive rate.” The government made a final effort to support the assignat. On December 23, 1795, the legislature reckoned the current circulation to be 24 billion, and set a final ceiling of 40 billion. When that limit was reached, all the instruments, fonts, and forms were carried to the Place Vendôme in Paris and disposed of in a solemn ceremony (February 19, 1796). Ramel-Nogaret eulogized the assignat:55 “The assignats have made the Revolution; they have brought about the 55 Moniteur,

14 Pluv. 4.

290

Chapter 9: Macroeconomic Features of the French Revolution

destruction of orders and privileges; they have overturned the throne and founded the Republic; they armed and equipped those formidable armies which have carried the tricolor flag beyond the Alps and the Pyrenees.... We owe them our freedom.” Meanwhile, the Directory’s need for funds continued. Two abortive attempts were made to create a successor currency to the assignat. The first took place outside the Legislature in February 1796. A private group of bankers, including shareholders of the defunct Caisse d’Escompte and future organizers of the Bank of France (which would be created in 1800), proposed to take over the remaining National Estates, and to manage or sell them as it saw fit. In exchange, the bankers would provide the government every month with 25 million in bank notes, convertible on demand into specie. In spite of the executive’s interest, the project was defeated in the legislature where opposition to “putting the Constitution in the safe deposit of the bankers”56 was still very strong. The legislature decided to retry the real-bills method of issuing money. A specific set of National Estates (including Church lands in annexed territories) was earmarked for sale: 2,400 million of a new currency, the mandat, were issued to retire the assignats at 30:1 and to finance spending for 1796. The sales were to be made at fixed prices57 rather than through auctions, and payments were scheduled over a six month period. Starting at 35% of face value in March, the mandat fell to 6% by early summer 1796. The mandat failed even though the government did not issue one note beyond the self-imposed 2,400 million ceiling. By August the government ceased to accept the mandat at par for taxes or payments for the Estates. Land sales were suspended in October 1796, and the mandat and assignat were demonetized on February 4, 1797, thus ending seven years away from a metallic standard. One last monetary experiment accompanied the demise of paper money. In June 1797, a law established a depreciation table for reimbursing nominal contracts made between January 1791 and July 1796. The amount of specie to be repaid was adjusted to reflect the value of the assignat at the time of the contract. In each dé56 In

the words of an opponent. mechanically by multiplying the property’s income in 1790 by a capitalization factor. 57 Computed

Rise and Fall of the Assignat

291

partement, local administrations computed an index of the value of the assignat against gold and also against commodities and land.58 By December 1797 these indexes were available to settle claims between individuals. D EFAULT The debt problem, fomenter of Revolution, became acute in 1796. The government, with a larger perpetual debt to service after the conversion of August 1793, remained unable to raise revenues from the fiscal system inherited from the Constituante. Once papermoney became unavailable, the government started paying the interest on the debt with IOU’s acceptable in payment for national lands. The market discounted these 77% to 85%. The price of government debt remained accordingly low (see Figure 9.14) since the market did not expect major improvements in interest payments.59 The conservative majority elected in the Spring 1797, and the resounding defeat of the former members of the Convention, raised hopes which were quickly dashed by the 18 Fructidor coup. The executive purged the legislature and cut the debt by two-thirds. The “Two-Thirds Bankruptcy” applied to both perpetual and life annuities. It consisted of a “mobilization” of two-thirds of the capital60 which was reimbursed to debt holders in the form of bons du 2/3, or vouchers, admissible in payment for National Estates. The remaining third of the capital was “consolidated” into annuities, and exact payments were promised in the future. The 2/3 vouchers were nearly a total loss since the market priced them at 1% to 2.5% of face value. As Figure 9.14 shows, the market anticipated the bankruptcy.61 58 This

scheme was to be proposed again in the 1970s by Jeffrey Shaefer and Axel Leijonhufvud (1984). See Sargent (1986, 153). A version of this scheme was part of the ill-fated Austral plan in Argentina in 1986. 59 The price of a perpetual rent of 5 Francs was priced around 9 F in February– April 1797, shot up to 33 after the elections, and fell between 7 and 9 when the bankruptcy was announced. 60 Life annuities were capitalized at 10 times. 61 After the bankruptcy, the government still paid interest in bons d’arrérages admissible in payment of taxes; the market valued them at a 20% to 35% discount. The ratio of the cash value of the bons to the price of bonds remained between 0.33 and 0.40, until November 1799.

292

Chapter 9: Macroeconomic Features of the French Revolution

70

60

Francs

50

40

30

20

10

0 1796

1797

1798

1799

1800

1801

1802

*: Before the 2/3 bankruptcy

Figure 9.14 Price of a 5 F perpetual, 1795–1802. Source: Moniteur universel.

The Legacy U NINTENDED C ONSEQUENCES FOR E NGLAND THROUGH S PECIE F LOWS According to R. G. Hawtrey (1919), much of the specie that left France, in response to the market signals depicted in Figure 9.10, went to England, fueling an expansion of credit and inflation there. When the hyperinflation began in the Spring of 1795, market prices signaled for importing metal into France (Figure 9.13). The expansion of credit in England came to a halt, and deflation began. Hawtrey claimed that flows of gold back to the Continent caused the drain on the Bank of England and prompted the Bank Restriction Act of 1797.62 It is ironic that Britain should have stumbled upon a temporarily inconvertible currency as a successful means of finance while France deliberately created a new system of currency 62 Britain

passed the Bank Restriction Act on February 27, 1797, three weeks after France returned to specie. See Fetter (1965), chapter I, for an alternative to Hawtrey’s account.

293

The Legacy

but failed to sustain it except by the Terror.63

20

15 Francs

19 brumaire

17 brumaire 10

5

0 Vendémiaire 8

Brumaire

Frimaire

Figure 9.15 Price of a 5 F perpetual, September–December 1799 (daily). Source: Moniteur Universel.

The Revolution is over. —Proclamation of the Consuls Bonaparte, Cambacérès and Ducos on December 15, 1799.

The Constitution of the Year 3 survived until 18 Brumaire 8 (November 1799), when General Bonaparte established a military dictatorship. Much later, when an irritated Emperor Napoleon asked Talleyrand how he had become so rich, the expert flatterer answered: “Why, Sire, I simply bought a lot of government bonds on 17 Brumaire and sold them three days later.” Figure 9.15 confirms that Talleyrand earned a handsome return. This anecdote suggests that some people expected a strong government to end 63 Bordo

and White (1991) do not emphasize the accidental nature of the suspension, and prefer to interpret it as the application of a well-understood “contingent gold-standard rule.”

294

Chapter 9: Macroeconomic Features of the French Revolution

financial disorder. France returned to an absolute monarchy under another name. Paradoxes accompanied the Napoleonic stabilization. Numerous practices of the Old Regime were reestablished for tax collection. Bond monies were required from officials who were granted a commission on amounts collected.64 A strong central control over allocation, assessment, and collection of taxes was asserted, reversing the Revolution’s ineffectual reliance on local authorities. Indirect taxation, abandoned by the Constituante and reluctantly brought in on a small scale by the Directory, was considerably extended with custom tariffs and excise taxes. Figure 9.5 shows how tax revenues increased. The residual debts of the Directory were liquidated, and interest payments on the debt in specie resumed for the first time in 12 years. The budget of 1802 was balanced 25 years after Turgot. Napoleon disliked capital markets and did not borrow or use his creation, the Bank of France, for financing. Budgets remained balanced until the Russian campaign of 1812. The low price of government bonds (i.e., the rentes that had survived the Two-Thirds Bankruptcy) suggested that the costs of borrowing on the market would have been even more expensive than it ever was for 18th century governments (see Velde and Weir (1992)). The lowest level to which interest rates fell was to 6% in 1809, only briefly. Napoleon’s policy of financing wars through taxation was partly shaped by the French government’s recent bankruptcy record, but partly also by his own unchecked authority. The dictatorial regime had advantages for some aspects of tax collection, but left intact the same “sovereign borrowing” problem that Necker had so sharply characterized in 1784.65

Appendix: The Auction Mechanism The term National Estates covered all the property of the Church (chattels and real estate), the Royal demesne (i.e., the king’s per64 Thereby

reversing the Revolution’s reforms, which Bosher (1970) sees as the “triumph of the general interest over selfish private ones.” 65 See Bordo and White (1991).

Appendix: The Auction Mechanism

295

sonal possessions), the estates of the émigrés,66 and those of individuals executed under the laws of the Terror.67 From May 1790 to July 1793, the estates sold were primarily the Church lands and buildings. Some feudal dues and rights were also put on sale, until they were abolished outright in 1792. These properties, called “First Origin Estates,” were sold as follows. Either individuals or townships could ask to purchase an estate. The value of the estate was then estimated by the authorities (usually by multiplying its income by 22). The authorities then published a list of selected estates with these estimates. As soon as at least one offer was made, an auction date was set and published. An “English auction” was held, and once the highest bid had been offered the auction was adjourned. News posters were published announcing the highest bid and setting the date and place of a second auction, to be held within a month of the first. At the second auction, one-minute candles were lit in sequence and bidding started at the highest bid established during the first auction. If two candles expired without any new bids, the estate was awarded to the highest bidder from the first auction. Within a fortnight, the buyer was to pay a down payment (between 12 and 30 percent), and sign 12 promissory notes for the rest, in 12 equal payments over 12 years, at 5% interest. The buyer retained the option to prepay part or all of the remaining balance.68 Acceptable means of payment were gold and silver, assignats, and some debt certificates used by the Liquidation agency to liquidate small offices. The latter were few in quantity (less than 2% of the value of the assignats issued in March 1792). When payment was made 66

The Assembly had obtained estates of émigrés, French nationals residing outside the territory since 1789 who would not return or could not justify their absence before a deadline. On February 9, 1792, these estates were confiscated and assigned to “compensate for the costs that their activities have occasioned to the Nation.” Although some properties were heavily mortgaged, the émigrés’ estates represented a vast resources of approximately the same order of magnitude as the first lands sold in 1791. The émigrés’ lands were added to the National Estates in July 1792 and sold in late 1793. 67 Under the Old Regime auctions were used to sell government-owned farmlands, logging rights, and assets that the government had acquired through judicial proceedings (inheritance, bankruptcy). 68 Later, premiums were offered to those who paid early.

296

Chapter 9: Macroeconomic Features of the French Revolution

in assignats, the assignats were canceled immediately, then sent to Paris for burning. When payment was made in coin, the coins were sent to Paris and then exchanged at the Treasury for assignats held by the Treasury, which were then canceled and burned. From July 1793 to February 1796, the “Second Origin Estates” from the estates of the émigrés were auctioned. These were sold under similar rules to the earlier ones, but with ten-year terms instead of twelve. By January 1794, both categories of Estates were placed under the same rules.69 From March 1796 to November 1796, a similar sales system was implemented as part of the mandat experiment. Estates were estimated at 22 times income (18 for houses) and awarded to the first bidder who made a 25% deposit, the balance of which was due within three months, payable in mandats at face value. In late July, the balances due were to be paid in mandats at market value. In September, auctions replaced the single-bid system. Sales of National Lands continued until 1815.70

69 In

June 1795 the sale procedure changed: any estate was estimated at 75 times the income. The first individual to submit a bid for that price was immediately awarded the estate. The ensuing buying frenzy quickly convinced the government that it was making a big mistake, and the change in procedure was rescinded within three weeks. 70 See Bournisien (1908, 1909) and Marion (1908) for details.

10 United States Then, Europe Now Introduction I work in a macroeconomic tradition developed by John Muth, Robert E. Lucas, Jr., Edward C. Prescott, Finn Kydland, Nancy Stokey, and Neil Wallace. I use macroeconometric methods championed by Lars Peter Hansen and Christopher A. Sims. I interpret macroeconomic history in ways advanced by Irving Fisher, Milton Friedman, Anna Schwartz, and François Velde.1, 2 To illustrate how these research traditions have shaped me, I tell how predicaments facing the European Union today remind me of constitutional decisions the United States faced not once, but twice. I begin with a simple expected present value model for government debt and explain how Hansen and Sargent (1980) used rational expectations econometrics to render this model operational by deducing cross-equation restrictions that characterize how the value of a government’s debt depends on statistical properties of the government’s net-of-interest surplus. This econometric specification isolates essential determinants of the value of a country’s debt or currency. The econometric theory leaves open who chooses the all important statistical process for the government netof-interest surplus. In democracies, voters choose. To understand more, we think about outcomes that emerge under alternative 1 See Muth (1960, 1961), Lucas (1972b, 1976), Lucas and Stokey (1983), Lucas and Prescott (1971, 1974), Kydland and Prescott (1977), Hansen and Sargent (1980), Hansen (1982), Sims (1972, 1980), Fisher (1926, chs. XI, XII), Friedman and Schwartz (1963), and Velde (2009). Fisher (1926, chs. XI, XII), entitled “Statistical Verification,” set out a road map for Friedman and Schwartz (1963). Velde and Weber (2000) beautifully formalize and extend an enlightening model of bimetallism created by Fisher (1926). The issues described in this chapter have been with us for a very long time. See Conklin (1998) for a description and analysis of sovereign debt issues faced by Spain under Phillip II. 2 For an exquisite example of how theory imitates life, see Velde (2009) for an account of an actual pure change-of-units monetary experiment that is a key ingredient of the mental experiment analyzed by Lucas (1972b).

297

298

Chapter 10: United States Then, Europe Now

democratic political arrangements. A case study illustrates how democracies have balanced conflicting interests. My case study is how the constitutions of the United States have influenced the government net-of-interest surplus process and therefore the value of government debt. I say constitutions, plural, because we Americans have tried two of them, first the Articles of Confederation that were ratified in 1781, and then the US Constitution that was ratified in 1788. Those constitutions embraced two very different visions of a good federal union. The first constitution was designed to please people who preferred a central government that would find it difficult to tax, spend, borrow, and regulate foreign trade. The second served opposite interests. The US framers abandoned a first constitution in favor of a second because they wanted to break the prevailing statistical process for the net-of-interest government surplus and replace it with another one that could service a bigger government debt.3 Exactly how and why they did that is enlightening: starting in 1789, they rearranged fiscal affairs first and then approached monetary arrangements as an afterthought. The fiscal institutions of the European Union today remind me of those in the United States under the Articles of Confederation. The power to tax lies with member states. Unanimous consent by member states is required for many important EU-wide fiscal actions. Some lessons from US history are these: 1. The ability to borrow today depends on expectations about future revenues. Without institutions that provide adequate revenue sources, governments may have neither the current revenue nor the ability, by issuing debt, to pledge future revenues when occasions demanding especially large public expenditures arise. The inability to issue debt comes from the fact that prospective debt holders rationally anticipate that the government will be constrained in its ability to raise enough revenues to service the debt. To provide public goods, even 3 The

term “framers” rather than “founders” or “founding fathers” is more descriptive of how they thought of themselves, namely, as creators of an institutional framework within which their successors would act. See Rakove (1997).

Introduction

2.

3.

4.

5.

299

rare ones like surges of defense spending during wars, governments require the flexibility to tap adequate sources of revenue. Free-rider problems exist for subordinate governments vis-à-vis a central government. Because there is a classic free-rider problem in paying for public goods, subordinate governments, like states in the United States or nations in the European Union, cannot be relied on voluntarily to provide revenue to the central government to pay for public goods. Each state has an incentive to refuse, hoping that other states will accept the burden. Good reputations can be costly to acquire. In deciding whether or not to pay preexisting debts, governments have strong incentives to default. Their anticipations of default make prospective creditors reluctant to purchase debts in the first place. Governments therefore have incentives to earn reputations that they will pay off their debts in the future. Acquiring such a reputation can be costly because it might well require making apparently unnecessary payments to debts incurred before the current government took office. Compensating such historical debt holders can seem unjust to current taxpayers, but it may be necessary for the long-run health of a republic. It can help to sustain distinct reputations with different parties. It is challenging for a government simultaneously to sustain distinct reputations with disparate parties. This challenge manifested itself when the US federal government struggled to confront British trade restrictions from 1790 to 1812 and in the early 1840s when it wanted its actions to send separate nuanced messages to foreign and domestic creditors as well as various state governments. Confused monetary-fiscal coordination creates costly uncertainties. Fiscal and monetary policies are always coordinated and are always sustainable, even though they may be obscure. In the beginning, the United States coordinated them by adopting a commodity money standard and restricting states and banks’ ability to create fiduciary monies. Other arrangements are possible. You can have a monetary union without having

300

Chapter 10: United States Then, Europe Now

a fiscal union. You may want a fiscal union even though you don’t want a monetary union. Obscure coordination arrangements increase uncertainty in markets and among ordinary citizens.

The Math A basic theory about how creditors value a government’s debt starts with a sequence of one-period budget constraints gt + bt = Tt + R−1 bt+1 , or bt = st + R−1 bt+1 , t ≥ 0, where R > 1 is the gross return on one-period inflation-indexed government debt, bt is the stock of one-period pure discount (zero coupon) inflation indexed bonds issued at t − 1 and falling due in period t , and gt , Tt , st = Tt − gt are government expenditures net of interest payments on the debt, total tax collections, and the government net-of-interest surplus, respectively. Iterate the government budget constraints for t ≥ 0 backwards to get bt = − R[st−1 + Rst−2 + · · · + Rt−1 s0 ] + Rt b0 , t ≥ 1, which states that large government debts come from accumulating big government deficits −st− j , j = 1, . . . , t, as well as rolling over any initial debt b0 . But to sustain large government debts requires prospects of big government surpluses in the future. To appreciate this, iterate the budget constraints for t ≥ 0 forward to get ∞

bt =

∑ R− j st+ j ,

j =0

which states that the value of government debt equals the discounted present value of current and future government surpluses. Recognizing that future surpluses can be forecast only imperfectly induces us to replace st+ j with Et−1 st+ j , where Et−1 (·) temporarily denotes the public’s forecast based on time t − 1 information known by prospective bond holders at time t − 1 to be pertinent for forecasting future surpluses. (Remember that these one-period bonds are purchased at time t − 1 and redeemed at time t , so it is

The Math

301

information at time t − 1 that is pertinent for valuing bonds that mature at t .) Then the value of government debt becomes ∞

bt =

∑ R− j Et−1 st+ j .

(10.1)

j =0

To get practical implications from the bond pricing equation (10.1) requires a theory about how people forecast the present discounted value of the government surpluses dedicated to servicing its debt. In situations like this, Hansen and Sargent (1980) joined Muth (1960, 1961), Lucas (1972b, 1976), and Lucas and Prescott (1971, 1974) in applying the economist’s venerable device of modeling decisions as optimization problems.4 When the decision is to choose a sequence of forecasts, this approach is said to impose rational expectations.5 Evidently, optimal forecasts depend on the statistical properties of the object to be forecast. Suppose that the actual process for the government surplus is the first component of an m × 1 vector stochastic process yt that is governed by a moving average representation, so that st = es yt , where es is a selection vector and ∞

yt =

∑ C j wt− j ,

(10.2)

j =0

where {wt } is an m -dimensional martingale difference sequence and the information set Jt known to prospective bond holders at t is generated by wt , wt−1 , . . . . Assume that Ewt wt0 = I . Here wt constitutes “news” that arrives at time t . Following Hansen et al. (1991), it is convenient to write the first equation of (10.2) as ∞

st =

∑ σj wt− j = σ( L)wt ,

(10.3)

j =0 4 Also

see Sargent (1971, 1977, 1979b). (1960, 1961) began this approach. Situations in which those people who most influence prices forecast optimally can themselves be the outcomes either of long experiences from individuals’ statistical learning processes (see Bray and Kreps (1987) and Marcet and Sargent (1989)) or else a competitive process that somehow encourages the survival of the fittest (Blume and Easley (2006)). See Sargent (2008) for implications for macroeconomics. 5 Muth

302

Chapter 10: United States Then, Europe Now

where L is the lag operator meaning L j wt = wt− j and σ( L) = j −iω ) ∑∞ j=0 σj L . Assume that the spectral density matrix Sy ( ω ) = C ( e C (eiω )T has full rank m for almost all ω ∈ (−π, π ) , a condition equivalent with y being stochastically nonsingular.6 It is revealing and convenient to compute the value of bonds under rational expectations in two steps by applying an argument that invokes the law of iterated expectations. First, temporarily give bond purchasers “too much” information by replacing the subjective expectation Et−1 (st+ j ) ≡ E(st+ j | Jt−1 ) in equation (10.1) with E(st+ j | Jt ) , the mathematical expectation of st+ j conditional on the history of shocks wt , wt−1 , . . . in equation (10.2). Under this expanded information assumption, Hansen and Sargent (1980) showed in another context that7 ∞

bt =

∑ κ j wt− j ,

(10.4)

j =0

or bt = κ ( L ) w t , where κ (z) =

zσ(z) − R−1 σ( R−1 ) , z − R −1

(10.5)

where z is a scalar complex variable and κ (z) is the z -transform of the {κ j } sequence.8 Next, to reduce information to the set wt−1 , wt−2 , . . . actually available to prospective bond holders when they purchase the bonds at time t − 1, we follow Hansen et al. (1991) who establish that the requirement that bt be measurable 6 Stochastic

nonsingularity means that no component of y can be expressed exactly as a linear combination of past, present, and future values of other components of y . 7 Hansen et al. (1991) extend this formula to handle the interesting case in which the first difference of st is a linear combination of a stationary vector process yt like (10.2). See Hansen (2011) and Hansen and Sargent (2013) for further generalizations. 8 The numerator of κ ( z ) is designed to contain a zero that cancels the pole at R−1 , i.e., the zero in the denominator at R−1 . This makes the Taylor series and Laurent series expansions of κ (z) coincide.

The Math

303

with respect to time t − 1 information Jt−1 information implies that κ j = κ (0) = 0, which in light of equation (10.5) requires that9 σ( R−1 ) = 0.

(10.6)

Equation (10.6) has a natural economic interpretation: it states that the present value of the moving average coefficients for the net-ofinterest surplus must equal zero. This condition renders the value of debt maturing at t measurable with respect to Jt−1 . Equations (10.2), (10.3), (10.4), (10.5), and (10.6) encode crossequation restrictions that are hallmarks of rational expectations econometrics: the coefficients κ j that tells the response of debt bt to past shocks wt− j are nonlinear functions of the discount factor R−1 and the coefficients σj in the moving average representation for the net-of-interest surplus st .10 Equations (10.2), (10.3), (10.4), (10.5), and (10.6) illustrate much of the logical structure and empirical power of rational expectations econometrics:11 • Current and lagged values of all components of the shock vector wt that impinge on future surpluses st+ j appear in the debt valuation equation (10.4). • The shock response coefficients κ j in equation (10.4) for the value of debt would change if government policy were permanently to alter the σj ’s in (10.3) that characterize the stochastic process for the government surplus. This technical finding is the heart of the influential critique of pre-rational expectations econometric evaluation procedures forcefully stated by Lucas (1976). I argue on page 307 that George Washington and Alexander Hamilton understood that to increase the value of US government debt they would have to break the stochastic process (10.3) for {st } that had prevailed in the United States in the 1780s. 9 Related

measurability requirements play a key role in Aiyagari et al. (2002). Sargent (1981) for the role of those cross-equation restrictions in other contexts. 11 This is the theme of the papers in the volume about rational expectations econometrics edited by Lucas and Sargent (1981b), especially the introductory essay. Hansen (1982) and Hansen and Sargent (1991) extended and refined rational expectations econometrics. 10 See

304

Chapter 10: United States Then, Europe Now

• The same basic theory applies when there are prospects for default. For example, in each period suppose that there is a probability π ∈ (0, 1) that the government will write off a fraction φ ∈ (0, 1) of its debt.12 Let R˜ −1 be the discount factor applying to default-free debt. Then a “certainty equivalent” discount factor R−1 that compensates a risk-neutral creditor for holding default-prone debt is   R−1 = R˜ −1 (1 − π ) + π (1 − φ) .

(10.7)

With this adjustment to the discount factor, the preceding theory applies. Bigger haircuts φ and higher probabilities of default π lower the discount factor R−1 and thereby reduce the value of the debt.13 • Hansen et al. (2007) opened the way to extending the theory to incorporate variable discount factors that can absorb some of effects of the news shocks wt . • Important technicalities impede linking our theory to vector autoregressions. Shocks in vector autoregressions for yt must be in the Hilbert space spanned by yt , yt−1 , . . . (see Sims (1980)). These so-called “fundamental” shocks emerge from constructing the Wold moving average representation for yt that is associated with the limit of a sequence of finite order vector autoregressions as the lag length is driven to +∞ . Hansen et al. (1991) show that the internal logic of the present value equation (10.1) and the associated restriction σ( R−1 ) = 0 imply that the moving average (10.2) is not a Wold representation because the shocks wt , wt−1 , . . . span a larger space than the linear space spanned by yt , yt−1 , . . . , and so the wt shocks are not what would be recovered by running a vector autoregression. Hansen et al. (1991) discuss substantial implications of this fact for extracting econometrically testable implications from the theory.14 12 I assume that

φ and π are constant and do not depend on the stochastic process for the net-of-interest surplus st . 13 Arellano (2008) used related ideas to model sovereign risk. 14 Thus, there is a subtle relationship between the present value theory described in this section and causality in the sense of Granger (1969) and Sims (1972).

The Math

305

N EED FOR M ORE E CONOMIC T HEORY ? This piece of economics-plus-statistical forecasting theory forms the essence of the pricing model used by prospective buyers and sellers of government debt.15 For the purposes of those buyers and sellers, it is enough to have a good fitting statistical model of the stochastic process (10.3) governing the government surplus. But for other purposes, a statistical model alone is inadequate. The model formed by equations (10.2), (10.3), (10.4), (10.5), and (10.6) is superficial because the government surplus process {st } is itself the outcome of a political decision process.16 The model summarizes but does not purport to explain the statistical properties of the surplus process (10.2)–(10.3) in terms of the balance of conflicting interests that actually created it. Economic theory goes deeper by analyzing contending economic and political forces that actually produce a statistical regime. In economic theory, an agent is a constrained optimization problem. A model consists of a collection of constrained optimization problems. Theories of general equilibrium, games, and macroeconomics acquire power by deploying an equilibrium concept whose role is to organize disparate choice problems by casting them within a coherent environment.17 In the presence of one or more large player—governments in this case—decisions of some agents typically impinge on the constraint sets of others, and therefore on their incentives to take subsequent decisions. In such cases, the statistical process that represents an equilibrium outcome emerges jointly with agents’ beliefs about what would happen in situations that they never face. Beliefs about those events have 15 It

is highly simplified relative to papers that embody standard practice today. In particular, the assumption that the interest rate is risk-free and constant is a big oversimplification. See Lucas (1978), Harrison and Kreps (1979), Hansen and Singleton (1983), Hansen and Richard (1987), Eaton and Gersovitz (1981), Arellano (2008), and Hansen and Jagannathan (1991) and references that they cite and that cite them for extensions of the basic model that relax that assumption about the interest rate. 16 The adjective “superficial” is descriptive, not critical. 17 Kreps (1997) describes common features of the equilibrium concepts used in theories of games and general equilibrium. To understand the empirical observations in the US case study presented later in this chapter might require going beyond this equilibrium concept to incorporate improvisation and adaptation in new ways that Kreps indicates at the end of his chapter.

306

Chapter 10: United States Then, Europe Now

important influences on outcomes that do happen.18 Chari and Kehoe (1990), Stokey (1991), and Bassetto (2005) have explored and applied notions of equilibrium appropriate to situations where a large government interacts with many atomistic private agents. I won’t formally use a single such model in the rest of this chapter. But broad insights from this class of models shape virtually everything I detect in the fiscal and monetary history of the United States. A H UMBLING M ESSAGE ? Macro models use the standard equilibrium concept to produce statistical processes for things like the government surplus as outcomes. This is a powerful method for “explaining” objects like {st } . But the equilibrium concept can disable someone who proposes to improve outcomes. Why? Because the equilibrium already contains the best responses of all decision makers, including any government agents who inhabit the model.19 Assuming that an equilibrium that explained the historical data can also be expected to “work” in the future puts a model builder in the position of not being able to recommend changes in policy precisely because he has understood the forces that have led policy makers to do what they do. The model builder’s way of understanding them is to say that they were optimizing. And giving advice would imply that he thinks that they weren’t optimizing or weren’t well informed.20, 21 18 Fudenberg

and Levine (1993) and Sargent (2008) and the references there describe and apply notions of self-confirming equilibrium, a type of rational expectations equilibrium in which possibly erroneous beliefs about events that don’t happen in equilibrium still have big effects on observed equilibrium outcomes. 19 Goethe said it this way: “So divinely is the world organized that every one of us, in our place and time, is in balance with everything else.” 20 The issue of whether equilibrium models are normative or positive was raised at a general level by Sargent and Wallace (1976) and more specifically in the context of interpreting vector autoregressions by Sargent (1984). 21 The only time I saw Milton Friedman speechless was at a dinner party at Stanford in the mid-1980s. His close friend George Stigler trapped Friedman by asking him two questions. First, Stigler asked whether Friedman consulted for private businesses. Friedman said no, that because businessmen had more information and had already optimized, he had nothing useful to tell them. Then Stigler said, “Well that makes sense to me Milton, but then why are you always telling governments what to do?”

The United States

307

M ODELING R EFORMS By an environment, economic theorists mean a list of agents, a specification of actions available to every agent, a timing protocol telling who acts when, and an information flow telling what is known, and when and by whom it is known. Some changes in an environment can amount to changes in institutions, e.g., reassigning particular decisions to an independent central bank or assigning particular taxes exclusively to states or exclusively to a central government within a federal system. This concept of equilibrium ties our hands by asserting that if you want to change outcomes, like the government surplus process mentioned above, then you have to reform institutions, which can mean agreeing on a new constitution. This is subversive. Nevertheless, that is what economic theory teaches. George Washington and Alexander Hamilton knew it and that is why they led a second political revolution, this one against the Articles of Confederation. They redesigned American institutions partly because they did not like the (equilibrium) {st } process and the implied value of government debt that the Articles of Confederation regime had fostered.

The United States Acknowledging that I lack anything approaching a complete model, but highly prejudiced by a class of equilibrium models, I now pursue an informal pattern recognition exercise to organize historical events that occurred in the United States and that remind me of choices being faced now as Europe struggles to manage a common currency.22 I see the authors of the constitution in 1787 and the architects of our federal government’s institutions and policies in 1790–92 to be wrestling with the implications of the government budget constraint (10.1), an equation that preoccupies both the 22 Maybe

it is a pattern imposition exercise. I did not select facts out of the blue. You can’t get anywhere accepting a complete “democracy of facts,” as Borges (1962) illustrated in his story about Funes the Memorious, who refused to impose patterns because he wanted to account for everything. My exercise amounts to pattern recognition with strong preconceptions. Prejudices help because data are limited.

308

Chapter 10: United States Then, Europe Now

United States and some European states today.23 V ICTORIOUS BUT IN D EFAULT The United States emerged from the US war for independence in 1783 with big debts and a constitution that disabled the US central government. The Articles of Confederation established a Continental Congress and an executive weak beyond the sweetest dreams of a contemporary American advocate of small government. The Articles worked as intended to restrain the central government from taxing and spending. That outcome served the interests of some US citizens, but not of others. It was not good for the Continental Congress’s creditors. The Continental Congress lacked powers adequate to service its substantial foreign and domestic debts. To levy taxes, the central government required unanimous consent of thirteen sovereign states.24 To finance the war, the Continental Congress had printed IOU’s in the forms of non-interest-bearing paper money (“bills of credit”) as well as interest-bearing debt.25 So had each of the thirteen states. After the war, the states could levy taxes to service at least parts of their interest-bearing debts.26 The central government could not. It regularly pleaded for contributions from the states, with at most limited success.27 An outcome was that Continental debts traded at deep discounts and so did debts of many states. Paper currencies depreciated markedly.28 23 The

remainder of this chapter relies on empirical evidence assembled for Hall and Sargent (Forthcoming). 24 Cournot (1897, ch. 9) constructed a model of a monopolist that buys complementary inputs from n monopolists. That model can be reinterpreted to explain how decision making by consensus leads to very inferior outcomes. 25 Bills of credit were small denomination circulating paper notes. They were not legal tender. Before the revolution, American colonies had issued paper notes declared to be legal tender, but the British government had prohibited them from being legal tender in an act of 1764. 26 See Wood (2009) for an account of differing states’ debt positions and how this fed into the politics. Also see Elkins and McKitrick (1993) for a comprehensive account of the political struggles associated with creating and running US institutions during the Washington administration. 27 Mailath and Postlewaite (1990) and Chari and Jones (2000) explain why decentralized systems with voluntary participation cannot be relied upon to provide public goods. 28 The Continental currency eventually declined to 1/40 or 1/100 of its initial value, but that inflation in the paper currency is not revealed by aggregate price

The United States

309

Deprived of tax revenues, the Continental Congress tried to roll over its maturing debt and to pay interest falling due by borrowing more.29 This became increasingly difficult as the 1780s unfolded. Ultimately, the Continental Congress stopped paying its creditors and watched interest payments in arrears grow in the form of new IOU’s called indents. Authority to levy tariffs, the most remunerative potential source of tax revenues, resided in the states. In 1781 and 1783, the Continental Congress asked the thirteen states to ratify amendments to the Articles of Confederation that would have allowed it to impose a Continental import duty whose proceeds were to be devoted entirely to servicing the Continental debt. Each time, twelve states approved, but one state did not (Rhode Island the first time, New York the second), killing the amendments.30 T RADE P OLICIES In the 1780s under the Articles of Confederation, the US had thirteen tariff policies and thirteen trade policies. The states’ main trading partner, Great Britain, discriminated against American shipping and American goods. Britain had done less of that before the revolution, but a foreseeable consequence of victory in the American Revolution was that the thirteen American states would be excluded from the British imperial trading system. Occasionally individual American states sought to retaliate against British discrimination, but their efforts were always undermined by neighboring states.31 The British could play one US state against another. indexes. David and Solar (1977) report an authoritative price index for the United States during this period. An interesting thing about their series (David and Solar (1977, 17)) is that because the unit of account was in specie, the depreciation of the paper Continental currency does not show up. It is an interesting contrast that during the US Civil War, the paper greenback displaced specie as the unit of account in most states that remained in the union. California and Oregon were exceptions. Their courts refused to enforce the federal legal tender law and they stayed on a specie standard. 29 This ignites the dynamics that underlie the unpleasant arithmetic of Sargent and Wallace (1981). 30 See McDonald (1985, 170–71). 31 See Irwin (2009) and Rakove (1997, ch. II) for the history and Cournot (1897, ch. 9) for the theory.

310

Chapter 10: United States Then, Europe Now

C RISIS AND A S ECOND R EVOLUTION Milton Friedman said that countries confront problems only after they have become crises. In the 1780s, the huge interest-bearing debts and currencies that had been issued to finance the war set the stage for a prolonged fiscal crisis from the point of view of the government’s creditors, if not its taxpayers. Measured at par (but not at the deeply discounted values then prevailing in the market), the ratio of Continental plus state debt to GDP stood at about 40%, a massive debt at a time when the government could raise at most only a small percentage of GDP in taxes. About 2/3 of this debt had been incurred by the Continental Congress, the rest by the thirteen states.32 Sometimes fiscal crises have provoked political revolutions that renegotiate past promises and resettle accounts among taxpayers and government creditors, as they did in France in 1789 and the United States in 1787–88.33

Restructuring Fiscal Institutions To rearrange powers and incentives, the framers scrapped the original constitution, the Articles of Confederation, and wrote an entirely new one better designed to protect US government creditors.34 The US Constitution realigned incentives and authorities in ways that (a) let the central government devote enough tax revenues to service debts that the Continental Congress and the states had issued to pay for the war, and (b) gave the central government exclusive authority to tax and regulate US international trade. That gave the federal government the tools to implement a national trade policy that could deter British discrimination against US citizens. In the early days of the United States, the government budget constraint linked debt service capacity very closely to trade policy. 32 Hamilton

(1790) estimated that at the beginning of 1790, the total debt at par stood at 79 million dollars, of which 25 million was owed by the states, and 12 million was owed to foreigners. 33 Sargent and Velde (1995) see the French Revolution through the lens of the government budget constraint. 34 There is a grain of truth in a controversial interpretation of the framers’ motives authored by Beard (1913).

Restructuring Fiscal Institutions

311

That tariffs were the main source of federal revenues confronted the country with a choice that framed US politics from 1789 to 1815. Britain was the main potential trading partner of the United States. Raising revenues to fund US debt required sizeable and reliable trade volumes with Britain, even if that meant restraining US reactions to British discrimination against US goods and ships. But because they put a high priority on faithfully servicing the US government’s debt and thereby earning the United States a reputation for paying its bills, Washington and Hamilton and the Federalists made preserving a difficult peace with Britain a cornerstone of their policy. So they refrained from retaliating against British trade restrictions. Later, because they wanted to retaliate against British trade restrictions, Jefferson and Madison and the Republicans were willing to imperil trade volumes with Britain and to sacrifice federal tariff revenues. They were willing to do that even if it affected US creditors adversely. Irwin (2009) describes how choices about these trade-offs can explain political outcomes in the United States both in the 1790s when the Federalists protected trade and peace and also after 1805 when the Republicans jeopardized trade and peace first with an embargo and then with the War of 1812 against Britain.35 T HE O RDER OF F ISCAL AND M ONETARY R EORGANIZATIONS Hamilton and the first Congress reorganized fiscal affairs first.36 F ISCAL P OLICY Dates reflect priorities. Congress created the Treasury department on September 2, 1789, a Bank of the United States on February 25, 1791, and a US mint in the Coinage Act of April 2 1792. On September 21, 1789, the Congress directed newly appointed Secretary of Treasury Alexander Hamilton to prepare a plan for “an adequate provision for the public credit.” Hamilton delivered his 35 A

theme of Wills (2002) is that James Madison overestimated the damage that an embargo could inflict on Britain and that he underestimated the damage that it would do to American commerce and the ties that bound New England to the union. 36 See Sylla (2009) for a comprehensive account and interesting interpretation of Hamilton’s plans. Also see Wright (2008).

312

Chapter 10: United States Then, Europe Now

Report on Public Credit to Congress on January 14, 1790. Congress accepted Hamilton’s recommendations, including his proposal to nationalize the states’ debts, in the acts of August 4 and August 5, 1790.37 Those acts set out a detailed plan for rescheduling Continental debt by selling a set of securities that Congress designed with Hamilton’s advice. These new debts promised to pay specific sequences of payoffs denominated in a unit of currency called a “dollar,” which in August 1790 was a silver coin issued by Spain. Hamilton (1790) told Congress that honoring the Continental Congress’s original promises to pay would drive down prospective returns on government debt by raising ex post returns relative to what had been expected during the 1780s when Continental debt had traded at deep discounts.38 He also argued that prospective returns could be lowered if private traders would come to regard government debt as a fully trusted obligation to the bearer, increasing its liquidity. Confirming Hamilton’s expectations, discounts on Continental and state bonds evaporated when news about the pro-administration outcome of the debate spread.39 D ISCRIMINATION AND L IQUIDITY An especially fascinating part of Hamilton’s report is his response to James Madison’s proposal to discriminate among current owners of Continental bonds according to when they had purchased 37 Acts

of the First Congress, Second Session included the act of August 4, 1790 making provision for the debt of the United States, the act of August 5, 1790 to provide more effectually for the settlement of accounts between the United States and the individual states, and the act of August 19, 1790 making further provision for the payment of the debts of the United States. 38 Remember formula (10.7) for the discount factor. 39 Hamilton had altered creditors’ views about the government’s “type.” The situation of the new government in United States in 1789 reminds me of an example about sovereign default in Bassetto (2005, sect. 4). Assume that a government with a dubious fiscal record leaves office and is replaced by a new government that is perfectly credible and dedicated to repay the debt. Despite the best intentions, whether or not the new government defaults is still influenced decisively by the private agents’ beliefs. If they persist with beliefs that the new government will default, they will demand prohibitive interest rates, rendering even a wellmeaning government eventually unable to meet its obligations at those rates. So to succeed the new government will have to implement good economic policies and also benefit from good (or lucky?) “expectations management,” whatever that means. See Bassetto (2006).

Restructuring Fiscal Institutions

313

them.40 Motivated by concerns about fairness, Madison wanted to take away inordinate capital gains from people who had purchased Continental bonds at discount; he also wanted to compensate former owners who had sold them at discount. Hamilton convinced Congress that such ex post discrimination would adversely affect the beliefs of prospective purchasers of government debt and would thereby damage liquidity and trust in the market for bearer government bonds.41, 42 F EDERAL B AILOUT OF S TATES The United States began with a comprehensive bailout of the individual states when on August 4, 1790 the US Congress accepted Alexander Hamilton’s proposal to nationalize (or “assume”) states’ debts. That completed a negotiation begun at the Constitutional convention when authority to tax imports had been transferred from the states to the federal government. In exchange for acquiring that most important revenue source, the federal government agreed to bail out the states, a decision that realigned creditors’ interests away from states and toward the federal government.43 By converting creditors of the states into creditors of the central government, Hamilton converted those bond holders into advocates of a federal fiscal policy that devoted a substantial share of the proceeds of a revenue-raising tariff to servicing those bonds. A stated justification for nationalizing the states’ debts was that most of them had been incurred to finance states’ contributions to the national war for independence. The US Treasury set up a system designed to account for each state’s contributions to the Glorious 40 See

Hamilton (1790). the Congress defeated Madison’s proposal for discrimination, a related idea returned to affect the Madison administration two decades later during the War of 1812. Dewey (1912, 134) describes an act of March 24, 1814 that required the government retroactively to offer more favorable terms to previous creditors if subsequent issues garnered lower market prices. 42 Proposals to discriminate among creditors often surface during negotiations to reschedule debts. For example, there are proposals for private holders of Greek government debt to take substantial voluntary haircuts while non-private creditors are to be paid in full. 43 McDonald (1985, 166–67) describes how in the early 1780s Superintendent of Finance Robert Morris tried but failed to organize the Continental Congress’s domestic creditors as a nationalizing force. 41 Although

314

Chapter 10: United States Then, Europe Now

Cause and to compensate them accordingly. It would have been prudent for subsequent lenders to appreciate that Congress had reasoned that it was states’ contribution to that national enterprise that justified the 1790 bailout. Investors should not have interpreted it as a promise to bail out states in the future no matter what, but apparently some of them did, to their eventual regret (please see the section “Another Federal Bailout of the States?” below). W HY PAY ? The government institutions that they designed and the decisions that Congress and the president took in 1790 and 1791 confirm that the framers intended fully to honor the debts that they had inherited from the Continental Congress. Making good on the promises originally made to Continental and state debt holders to finance the US War of Independence meant disappointing other expectations and breaking promises at least implicitly made about other dimensions of fiscal policy, for example, to keep taxes low. The deep discounts at which Continental debts traded in the mid-1780s reflected traders’ anticipations of those low-tax policies. Why, then, did the framers choose to keep some promises (ones to its creditors that had apparently already been substantially discounted) by breaking other promises (those to Continental taxpayers) that had been protected by the Articles of Confederation?44 If, as seems appropriate, we regard 1787 or 1789 as a new beginning —“time 0” in models of Ramsey plans and recursive mechanism design— then Ramsey models in the representative agent tradition of Lucas and Stokey (1983), Chari et al. (1994), and Jones et al. (1997) won’t help us to answer that question. Those models typically advise a government to default on all initial public debts45 and thereby impose that least distorting of taxes, an unforeseen capital levy. Other revolutionaries have done that,46 but not the US framers. 44 The

“why pay?” question has been sharply posed by Bulow and Rogoff (1989) and Kletzer and Wright (2000). 45 Sometimes they have also done whatever they could to acquire net claims on the private sector in order to finance future expenditures efficiently. Paal (2000) describes how the Hungarian communists deliberately reset “time 0 ” after World War II and acquired claims on the public by restarting the monetary system. 46 Lenin and Trotsky and their admirers in Eastern Europe did that. The leaders of the French Revolution in 1789 did not, instead struggling valiantly for years to

Restructuring Fiscal Institutions

315

Their purpose in realigning authorities and interests was to affirm that a “deal is a deal,” at least so far as concerned obligations to the government’s creditors, if not to taxpayers. To understand why Hamilton and Washington and other framers wanted to pay, we have to take into account heterogeneities of economic situations and consequent conflicting interests47 as well as reputational considerations that are absent from these Ramsey models. The purposes for which those initial debt were incurred, the identities of the individual creditors, and the perceived adverse consequences of default all mattered in ways neglected at least by the three representative agent Ramsey models cited above.48 Such Ramsey models help explain government policies after some political revolutions, but not those of the United States in 1789. In paying those Continental and state obligations, Secretary of Treasury Hamilton wanted the federal government to gain enduring access to domestic and international credit markets. That would expand options for financing temporary surges in government expenditures by borrowing, thereby allowing his successors service the pre-revolution debt until circumstances eventually forced them into a substantial default in 1797. See Sargent and Velde (1995). 47 See Meltzer and Richard (1981). 48 American politics and policies toward debt management in the aftermath of the US war for independence differed strikingly, for example, from those in Germany after World War I. Domestic creditors owned most of a very large government debt that Germany had accumulated during World War I, but then the Versailles treaty imposed big further debts on the German government in the form of huge and uncertain reparations payments to some of the victors. Politics in the United States after the war for independence differed from those in Germany after World War I because the US foreign debt had come from the benevolence and trust of friends in France and Holland who had sent us resources during the war, not the vengeance of foreign powers that had defeated us, as was true in Germany. A hyperinflation produced consequences that allowed Germany to escape most of those reparations payments, albeit at the cost of tremendous collateral damage in the form of a massive redistribution away from German nominal creditors to German nominal debtors as the value of German mark depreciated from its pre–World War I value by a factor of 1012 by November 1923. Sargent (1982) describes how Germany abruptly ended its hyperinflation by using a version of the simple theory (10.1) for valuing government debt. Before November 1923, the most important component of Germany’s government surplus process st was an inflation tax. The hyperinflation was arrested by adopting policies that adjusted government expenditures and taxes, along with fortifying a central bank that would refuse to levy the inflation tax.

316

Chapter 10: United States Then, Europe Now

to moderate the contemporary tax increases needed to finance those surges.49 He also asserted that an outstanding stock of government debt earning a relatively risk-free return would foster the development of domestic credit markets, which he thought would be a boon to commerce and industry.50, 51 M ONETARY A RRANGEMENTS Only after fiscal policy had been set on course in the acts of August 4 and August 5, 1790 did Hamilton and the Congress then turn to monetary policy. Hamilton presented his Report proposing a Bank of the United States on December 14, 1790 and his Report proposing that the United States mint US silver and gold coins only on January 28, 1791. It was widely presumed that the United States would follow leading European countries in embracing a commodity money standard. So the remaining monetary policy decisions for the framers simply involved choices of coin sizes and of a seigniorage rate for the mint.52 A N ATIONAL B ANK ? After a tense debate during which James Madison argued that a federally chartered monopoly bank would be unconstitutional, the Congress awarded an exclusive 20-year federal charter to a Bank of the United States.53 The bank was mostly privately owned 49 That

is, he wanted the option to issue debt in the fashion made explicit by Secretary of the Treasury Albert Gallatin in his 1807 report to Congress (see Dewey (1912, 128)), a policy later formalized in the tax-smoothing models of Barro (1979) and Aiyagari et al. (2002). 50 See Krishnamurthy and Vissing-Jorgensen (2010) and references cited there for modern arguments about good effects fostered by a stock of safe government debt. 51 See Brewer (1989) and North and Weingast (1989) for accounts of the flexibility that the government of Britain had achieved by successfully implementing fiscal institutions that Hamilton admired. An implication of Bassetto (2005, 2006) is that even with good institutions and well-intentioned policy makers, sometimes there are multiple equilibria, and we need luck or skill to select among them. 52 They set the seigniorage rate to zero, a decision called “free coinage.” 53 Madison changed his mind, when, serving as president twenty years later, the bank’s charter came up for renewal and opponents of the bank brought up Congressman Madison’s 1791 arguments to use against his administration’s request to renew the bank’s charter. Although he changed sides, Madison was

Restructuring Fiscal Institutions

317

and mostly operated in the interests of its private shareholders, though it did serve as fiscal agent of the federal government and as a depository for federal revenues. It also issued bank notes that circulated as currency and were convertible into specie on demand. It issued notes only in exchange for short term loans to the federal government or very short term commercial loans promising low risk. It avoided real estate and other long term and risky loans. In these ways, it could be said to implement the “real bills” regime of Adam Smith (1806), whose writings on the subject very probably influenced Hamilton.54 A M INT The framers seem to have regarded monetary policy as a side show to be tidied up only after a sound fiscal policy had been secured. The act of August 4, 1790 (1 Statutes, 138) had prescribed detailed procedures for funding US and states’ debts. New federal IOU’s were to be denominated in “dollars,” which on August 4, 1790 meant Spanish dollars because at that time there were no US dollars. In a report on coinage delivered in May 1791, Hamilton proposed that the United States manufacture a silver dollar defined to have the same silver content as a Spanish dollar.55 The Mint Act of April 2, 1792 accepted Hamilton’s recommendations virtually intact by creating a US dollar. In terms of the fundamental determinant of on the losing side both times, as Congress refused to renew the Bank’s charter in 1811, causing the United States to finance the War of 1812 with its longstanding fiscal agent having just been abolished and scrambling to improvise alternative arrangements for acquiring short-term credit. Whether to have a national bank serving as fiscal agent of the federal government is something that statesmen like James Madison and Henry Clay changed their minds about, and so did the country. The charter of the first Bank of the United States was not renewed in 1811, and neither was the charter of the second Bank in 1836. 54 Smith’s real bills doctrine stresses benefits from permitting a government owned or private financial intermediary to issue circulating notes that are backed by safe evidences of private indebtedness. To Smith, “real” meant relatively risk free. Smith pointed to efficiency gains that could be gathered by allowing paper notes backed by safe private evidences of indebtedness to circulate and displace precious metals that would otherwise serve as media of exchange. See Sargent and Wallace (1982) for an analysis of pros and cons of the real bills doctrine. 55 Section 9 of the act of April 2, 1792 states that each dollar is “to be of the value of a Spanish milled dollar as the same is now current.”

318

Chapter 10: United States Then, Europe Now

its value, namely, its metal content, the US dollar was a copy of the Spanish dollar, the only difference being that it had American and not Spanish “advertisements” stamped on its sides. In terms of essential economic forces, whether or not the United States actually issued these dollars was incidental.56 O UTCOMES The Appendix to this chapter displays important outcomes in graphs of data taken mostly from early reports of the United States Treasury. Deep discounts on the Continental debt evaporated and the federal government successfully rescheduled its debt (again see equation (10.7) for the discount factor). Tariffs comprised virtually all federal revenues. About 2% of GDP was collected in federal taxes annually during the 1790s. About 40% of those revenues were used to service the debt. Under Hamilton and his Federalist successors, the debt was serviced and the principal rolled over, but substantial economic growth allowed the debt/GDP ratio to decline more or less continuously until the War of 1812, except for an increment used to finance some of the 15 million dollars paid to Napoleon Bonaparte for Louisiana.57 In 1790, a big “fiscal space” (see Ghosh et al. (2011)) for the United States was provided by prospects for rapid population and economic growth, prospects that were realized in the 25 years after 1790.

56

The US mint functioned as European mints typically did in those days. The mint stood ready to sell on demand at a fixed price, but did not purchase, gold or silver coins in exchange for gold or silver bullion, respectively. If you wanted to purchase coins from the mint, you took your bullion to the mint. The mint assayed the metal, then forged and stamped coins that they returned to you. If you wanted to melt the coins to retrieve the bullion, you could melt them yourself, or you could export or sell the coins to private parties for specie. 57 To put the magnitudes in perspective, at par value, the total Continental and state debt that Hamilton rescheduled in 1790 was about 79 million dollars, which at that time was about 40% of GDP, an estimate subject to substantial uncertainty. The Louisiana Purchase was a good bargain for the United States.

Following Through?

319

Following Through? Timing protocols that prevail in a democratic society open enduring issues about the roles of commitments, precedents, and reputations. Expectations about future governments’ decisions influence prices and quantities today, but today’s citizens and policy makers cannot bind future citizens to prescribed courses of action.58 Decisions made in 1790 and 1791 were just the beginning of the great American fiscal and monetary adventure. Conjectures about how their successors would complete or modify their plans vitally concerned the framers.59 They had sought to create institutions (timing protocols?) and precedents (reputations?) that they hoped would limit subsequent choices in ways that would induce their successors to choose good public policies. Subsequent US history witnessed tax revolts (an armed rebellion against the federal government in 1794 western Pennsylvania when farmers protested a federal excise tax on whisky) and tariff and trade regulation revolts (in 1814 when New England states threatened to dissolve the union, and in the early 1830s when President Jackson faced down John C. Calhoun and South Carolina during the nullification crisis). Struggles over how much the federal government should tax and spend and regulate continued until the US Civil War and beyond. It is useful at this point to mention examples of how an administration’s decisions interacted with those of its predecessors and those of its successors.

58 Kydland

and Prescott (1977) delineated this tension. See Klein et al. (2011), Klein and Ríos-Rull (2003), and Debortoli and Nunes (2011) for a small sample of an important literature in macroeconomics that uses Markov perfect equilibria to study quantitatively how outcomes under a sequential timing protocol differ from those under a timing protocol that awards a government the ability to choose once and for all. See Battaglini and Coate (2008) for a political-economic equilibrium under a sequential voting protocol. 59 At the Convention on June 26, 1787, James Madison (1956) said, “In framing a system which we wish to last for ages, we shd. not lose sight of the changes which ages will produce.” In 1811 Secretary of the Treasury Albert Gallatin told Congress “To meet these loans in the future we must depend on coming prosperity and the wisdom of successors; that is, favorable circumstances and rigid economy.”

320

Chapter 10: United States Then, Europe Now

F EDERAL AND S TATE PAPER M ONIES ? The authors of the Constitution and their supporters abhorred paper money and the sorry state to which American domestic, if not foreign, credit had been reduced. That attitude set the stage for a debate at the constitutional convention about which powers over monetary standards to assign to state and federal governments, and which to deny them. Delegates to the convention agreed to prohibit state governments from issuing bills of credit or otherwise make a paper currency a legal tender.60 What about the federal government? Preliminary drafts of the constitution had given the federal Congress the right to issue bills of credit. Thus, even though the Convention had already agreed explicitly to forbid states from issuing paper money, on the morning of August 16, 1787, the eighth clause of the seventh article in the draft of the constitution said that “The legislature of the United States shall have the power to borrow money and emit bills on the credit of the United States.” Madison’s notes of the convention’s proceedings on August 16, 1787 record a debate about a motion to strike out the clause authorizing congress to emit bills of credit. The motion carried 9 states to 2. Three contributions to the August 16 debate especially fascinate me. (1) James Wilson’s clear statements stressing the ex ante advantages in terms of promoting credit to be reaped by denying future government decision makers the authority to take actions that would occasionally tempt them ex post; (2) George Mason’s and Edmund Randolph’s statements urging the convention to appreciate the advantages of reserving for future decision makers enough flexibility to deal with contingencies of a kind that could not be foreseen in 1787; and (3) Madison’s remark that withholding the authority to make government bills of credit legal tender would 60 Article

1, section 10 includes the restrictions

No State shall enter into any Treaty, Alliance, or Confederation; grant Letters of Marque and Reprisal; coin Money; emit Bills of Credit; make any Thing but gold and silver Coin a Tender in Payment of Debts; . . .

Following Through?

321

be sufficient to restrain potential abuses.61, 62 Partly influenced by their understanding of that August 16, 1787 debate, during the first three quarters of the nineteenth century, many Americans believed that the framers had intended to shut the door on the federal government’s issuing a paper legal tender, and that the fact that the majority of the delegates did not go further and explicitly prohibit the federal government from issuing bills of credit simply reflected the constitutional convention delegates’ presumption that powers not explicitly awarded should be understood to be denied to Congress.63 An extensive review of the documentary record convinced Bancroft (1886) that the framers’ intent was clearly not to allow Congress to make a paper currency a legal tender.64 W HAT K IND OF C URRENCY U NION ? Before 1789, the thirteen states already had joined a currency union. All used the Spanish dollar. Article 1, section 8 of the US Constitution gives the federal Congress the exclusive power To coin Money, regulate the Value thereof, and of foreign Coin, and fix the Standard of Weights and Measures; . . . As we saw in the preceding section, the Constitution expressly prohibited states from issuing paper currency, and most believed that prohibition extended to the federal government. The federal 61 See

Bancroft (1886) for histories of legal tender acts in colonial America and of the framers’ aversion to making paper monies legal tender. 62 Madison stood true on this matter. As president from 1809 to 1817, Madison presided over an administration that issued federal bills of credit to finance most expenditures for the War of 1812, but that did not make them legal tender. 63 Sustaining this tradition, the confederacy did not make its paper currency a legal tender. 64 Bancroft’s review of the evidence was prompted by what he regarded as the Supreme Court majority’s flagrant disregard of the historical record in deciding the 1884 legal tender case Juilliard v. Greenman. The court reasoned that because Congress had the power to pay debts, it could do so by any means not expressly prohibited by the constitution; and that little attention needed to be paid the debates and votes at the constitutional convention because it was difficult to glean a consensus from them; that Congress’s power to borrow money included the power to issue obligations in any appropriate form, including hand-to-hand currency; and that the authority to issue legal tender notes accompanied the right of coinage (see Dewey 1912, 366–67).

322

Chapter 10: United States Then, Europe Now

government only modestly and temporarily65 circumvented that implicit limitation by allowing the Bank of the United States to issue circulating notes in exchange for short-term government debts. It took longer for the states to circumvent the restriction.66 In January 1837, in Briscoe v. Bank of Kentucky, the majority of the US Supreme Court, including newly appointed Chief Justice Taney, decided that state chartered and state owned banks had the right to issue paper bank notes (see Howe (2007, ch. 11)). The real bills reasoning of Adam Smith (1806) and Sargent and Wallace (1982) or the Modigliani-Miller reasoning of Wallace (1981) indicates how this decision effectively disarmed the Article 1, section 10 prohibition against states’ issuing bills of credit by allowing state banks to purchase state bonds with their circulating bank notes. After that and until Congress taxed them out of existence during the Civil War, a multitude of currencies circulated within and across states during what has been mislabeled a “free banking era.”67 Many such currencies circulated simultaneously with fluctuating rates of exchange that reflected probabilities that state chartered bank notes could be converted on demand into specie. So before the US Civil War from 1861 to 1865, we had a currency union in one sense—the precious metals were the unit of account throughout the union; but in another sense we did not—we had multiple currencies that presented citizens with choices about holding currencies bearing different risks and returns. There was no lender of last resort, no deposit insurance, and no presumption of federal bailouts of banks’ depositors. All that stood behind those notes was the prudence of bank managers promoted by what Bagehot (1920) called the

65 Congress

refused to renew the Bank’s charter in 1811. some state-chartered banks were issuing notes before Congress chartered the first Bank of the United States. 67 Free banking—in the sense of free entry—did not prevail. Most banks had to have state charters. Many of those state bank charters contained explicit provisions requiring the bank to make loans to the state or to buy bonds issued to fund canals, railroads, or turnpikes. Most of the assets that these banks purchased with notes were loans and discounts. However, banks that operated under socalled “free banking laws” were required to purchase state bonds to back their notes. 66 Actually,

What Kind of Fiscal Union?

323

“preservative apprehension” of owners of bank notes.68 So if the framers intended to establish a currency union, they had at best mixed success, at least before the Civil War. And if they had wanted a currency union, it apparently would have been based on a commodity money, not a managed fiat currency like the one we have in the United States today. We now turn to continuing controversies about the scope of the fiscal union that the framers established in August 1790.

What Kind of Fiscal Union? From the start of the republic in 1789 until the Civil War, Americans continued to dispute the proper scope and magnitude of federal tax, spend, transfer, and regulation policies. Interests that coalesced around the great Whig statesman Henry Clay’s American System in the 1830s advocated federal expenditures on infrastructure projects—roads, canals, railroads, universities—public goods that they argued merited national fiscal support. A coalition of interests with strong support in the southern states blocked most such measures.69 McPherson (1988, Sec. III, ch. 14) documents how the 37th Congress (1861–62) seized the occasion of the secession of most slave states to reorder the federal union along lines that fulfilled many of Clay’s goals. On July 1, 1862 the Congress passed the Internal Revenue Act, which among other things imposed the first federal income tax. On that same day, the Congress passed the Pacific Railroad Act awarding public lands and federal loans to companies that would construct intercontinental railroads. On 68 See

Rolnick and Weber (1983, 1984). With multiple private media of exchange bearing different and fluctuating rates of return, issuers usually accepted (but did not redeem) the demand liabilities of others. An outcome was that issuers typically wanted to redeem and clear notes issued by other banks in order to augment their holdings of specie (or “lawful money”). From the 1820s to the 1850s, the Suffolk Bank of Boston successfully administered a private note clearing operation for banks from all over New England. The Suffolk Bank managed a private “currency union” in the sense that notes of New England banks circulated at par throughout the region. See Weber (2009). 69 Those southern interests were enthusiastic about using federal resources to pursue military adventures, like the war in Mexico opposed by Abraham Lincoln and other Whigs, through which the United States acquired territories for building additional slave states and senators.

324

Chapter 10: United States Then, Europe Now

July 2, 1862, Congress passed the Morrill Act awarding grants of federal land for establishing what came to be known as land grant colleges.70 Earlier, similar legislation had been defeated by a Democratic Party, now decimated by the loss of its core to the Confederacy, that had wanted a weaker federal union than Clay and Lincoln. The seceding states expressed those preferences when they wrote a Confederate Constitution that in important ways more closely resembled the Articles of Confederation than the US Constitution. It took four years of awful civil war to force rebels to accept not only Abraham Lincoln’s interpretation of what it meant for all men to have been “created equal” but also the type of federal union that Hamilton and Washington had begun and that Abraham Lincoln preserved and extended. A NOTHER F EDERAL B AILOUT OF THE S TATES ? A sequel to Hamilton’s 1790 bailout of the states’ debts provides another example of how fiscal crises can provoke enduring institutional changes, this time at the level of individual states.71 Today, many US state constitutions require state governments to balance their budgets annually. Before the 1840s, state constitutions of US states did not impose year-by-year balanced budgets. Adams (1887) tells how, in response to adverse fiscal occurrences in the late 1830s and early 1840s, many states rewrote their constitutions to require balanced budgets annually.72 Here is the story. During the 30 years after 1789, citizens debated whether the federal government should or could finance public infrastructure projects. Before the Civil War, they decided that it couldn’t. In response to a string of presidential vetoes of public works appropriations, state governments assumed responsibility for public works projects. After 1829, many state governments ran large government deficits, substantial parts of which were justified at the time because they were said to be deficits on capital account, not current account. The logic was that those state bonds had been 70 The

Congress also passed a law granting federal land to settlers (“homesteaders”). 71 One of Milton Friedman’s favorites was a “law of unintended consequences.” 72 Those new constitutions thereby mandated that states forego the efficiency gains of tax-smoothing delineated by Barro (1979) and Aiyagari et al. (2002).

What Kind of Fiscal Union?

325

issued to help finance public or private infrastructure projects. People advanced the theory that those bonds would be self-financing because ultimately they would promote growth and larger state government tax receipts in the forms of fees or taxes on increased land values. Belief in that theory allowed state bonds to be sold widely. Some were purchased by Europeans who were partly convinced by the self-finance theory and who also apparently mistakenly understood them to carry as much investor protection as federal bonds, which had earned a good reputation through a sustained record of having been honored after the Wars of Independence and 1812. And investors in state bonds knew that the federal government had comprehensively bailed out state debts at the beginning of the republic. Also, Article 4, section 1 of the US Constitution mandates strong protection for owners of state debts: Full Faith and Credit shall be given in each State to the public Acts, Records, and judicial Proceedings of every other State. And the Congress may by general Laws prescribe the Manner in which such Acts, Records and Proceedings shall be proved, and the Effect thereof. But foreign investors in state bonds may not have noticed weakened investor protection created by the Eleventh Amendment to the Constitution, passed in 1793 after a citizen of one state had taken a grievance against another state into a federal court. The Eleventh Amendment disarms the investor protection originally guaranteed by Article 1, section 1 by stating: The Judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States by Citizens of another State, or by Citizens or Subjects of any Foreign State. For European and other bond holders, the story did not end happily. During a recession at the end of the 1830s, many states defaulted.73 European bond holders then learned that the Eleventh Amendment deprived them and other creditors of American states of protection in federal courts. During the 1840s, Congress debated but ultimately rejected proposals for the federal government to 73 See

Scott (1893) and Ratchford (1941).

326

Chapter 10: United States Then, Europe Now

pay those state debts. During the congressional debates, advocates of a bailout recited the precedent set by Hamilton’s 1790 bailout of the states. But opponents successfully argued that Hamilton had bailed out state debts incurred for a glorious national purpose, while the debts of the early 1840s had been incurred for disparate causes to finance local projects. That and other arguments led Congress to refuse to bail out the state debts. This episode cost the United States a hard-earned high-quality reputation for all US government debt, federal as well as state, and cast long reputational shadows in two directions. It seems that the international bond markets’ response to these state bond failures did not immediately include an inclination to adopt a nuanced view that discriminated finely between the credit worthiness of federal and state authorities. For years, the reputation of federal credit in Europe suffered along with that of the states. But the Congress’s decision not to bail out the states had other, arguably more beneficial consequences for the country. A legacy of the Congress’s decision was that in the 1840s more than half of the US states rewrote their state constitutions to require year-by-year balanced budgets. This is yet another example of fiscal crises that have produced the lasting institutional changes that we sometimes call revolutions.74, 75 74 See

Wallis and Weingast (2005). As noted, the Eleventh Amendment to the US Constitution stated that state debts can’t be enforced in federal courts. However, debts of municipal corporations and counties are enforceable in state and federal courts. Adams (1887) claimed that this system of arrangements for protecting investors and the balanced budget restrictions in state constitutions explain the marked shift in expenditures and debts from states to local and municipal and county governments during the 19th century. Wallis (2000, 2001) has effectively taken up this theme. 75 The story does not end here. Section 4 of the Fourteenth Amendment to the US Constitution says: The validity of the public debt of the United States, authorized by law, including debts incurred for payment of pensions and bounties for services in suppressing insurrection or rebellion, shall not be questioned. But neither the United States nor any State shall assume or pay any debt or obligation incurred in aid of insurrection or rebellion against the United States, or any claim for the loss or emancipation of any slave; but all such debts, obligations and claims shall be held illegal and void.

What Kind of Fiscal Union?

327

Did the Congress do the right thing in refusing to assume those state debts? There is a strong case to be made that it did: at the cost of temporarily sacrificing the federal government’s hard earned good reputation with international creditors who were unable or unwilling to distinguish between the repayment records of federal and state governments, that decision succeeded in establishing a strong reputation of the federal government vis a vis the states. The Congress told the states not to expect the federal government to backstop their profligacy.76 To put the point bluntly, if by bailing out those state debts the federal government had set up expectations that they would back up state loans in the future, that would have exposed the United States to adverse consequences like ones that Kareken and Wallace (1978a) warned about in another context, namely, the insurance of financial institutions. Kareken and Wallace taught that under-priced government insurance of deposits of inadequately regulated financial intermediaries provides incentives for those intermediaries to become as big as possible and as risky as possible. That will almost surely put the government into the position of eventually having to bail them out. Therefore, Kareken and Wallace said that if you want to extend deposit insurance, you had better regulate financial intermediaries’ portfolios. Extending and applying the Kareken and Wallace logic to federal bailout of states, in exchange for offering such insurance, a federal bailout of the states would have set the United States on the road to extended federal control of states’ fiscal policies. And where would that have ended? With federal control of cities too?77 Without Congress’s 1840s refusal to bail out the states, it is probable The Fourteenth Amendment strives simultaneously to protect the reputation of federal debt and to eradicate the reputation of state debts issued by Confederate states. 76 See Fudenberg and Kreps (1987) for how difficult it can be to sustain distinct reputations with multiple parties. Another example of this difficulty might be that in the arrangements and decisions that it has set up to pay federal and state debts in the 1790s, the United States led by the Federalists in the 1790s had set precedents that inadvertently created expectations on the part of state creditors that it would backstop their profligacy. 77 Related issues may return to the United States soon: Will the federal government bail out high-debt states? Should state income tax be deductible on federal tax returns, thereby administering a transfer from the frugal states to the profligate states?

328

Chapter 10: United States Then, Europe Now

that those state constitutions would never have been rewritten to mandate year-by-year balanced budgets.

Lessons for Now? For the type of government we had under the Articles of Confederation in the 1780s—a weak fiscal union unlikely to pay its creditors what they had been promised—those deeply discounted Continental bonds had been fairly priced in the 1780s. Hamilton and Washington had set out to change the government’s “type” by realigning interests in ways that would induce the United States to pay what it had promised earlier and would promise later. And Hamilton wanted the market to price the bonds accordingly (via formula (10.7) for the discount factor again). Hamilton set out to manipulate current and prospective public creditors’ expectations about whether the government would honor its bonds the only way he knew: by creating a fiscal union with institutions and interests aligned in ways that would increase the actual probability that the federal government would pay. The framers’ purpose in creating that fiscal union was not primarily to facilitate a monetary union, a distinct project about which they revealed substantial ambivalence in their subsequent indecision about whether to charter a national bank or whether instead to foster competition among private currencies issued by state chartered banks. In terms of fiscal arrangements, the European Union today has features reminiscent of the United States under the Articles of Confederation. The power to tax lies with the member states. Unanimous consent by member states is required for many important EU-wide fiscal actions. Reformers in Europe today seek to redesign these aspects of European institutions, but so far the temporal order in which they have sought to rearrange institutions has evidently differed from early US experience in key respects. The United States nationalized fiscal policy first, and for the US framers, monetary policy did not mean managing a common fiat currency, or maybe even having a common currency at all. The European Union has first sought to centralize arrangements for managing a common fiat currency and until now has not wanted

Lessons for Now?

329

a fiscal union. And to begin its fiscal union, the United States carried out a comprehensive bailout of the government debts of the individual states. So far, at least, the European Union does not have a fiscal union, and few statesmen now openly call for a comprehensive bailout by the European Union of the debts owed by governments of the member states. Especially because of the contentious and obscure state of politics influencing monetary and fiscal policy in the United States today, an American is certainly not qualified to advise European citizens about what lessons, if any, to draw from the story about how the United States created a fiscal union. To ferret out useful lessons, it would be important to identify circumstances in Europe now that match those of the United States then, and circumstances that differ. The United States created its fiscal union at a time when the vast majority of people worked and lived on farms and when a substantial minority were slaves. People were much poorer then than now. Life expectancies were so very much shorter then than now that few working people lived long enough or ever earned enough to be able to stop working much before they died. Doctors and medicine often did more harm than good, so it was probably better that most people could not afford them. Deferred compensations, mostly for military service (pensions) but also some for land confiscated from Native Americans, were the only legal entitlements to government-financed transfer payments. Most people could not vote. The federal government was small and it redistributed only a small fraction of GDP. In peacetime in the first two decades of the United States , federal expenditures averaged 1 or 2 percent of GDP and in the beginning in the 1790s the federal government allocated 40% of its tax revenues to servicing the federal debt. The government debt that the Congress and president nationalized in 1790 had been incurred for a widely endorsed national cause.78 And fifty years later when Congress refused another massive federal bailout of state debts, its actions proved that the purpose for which those state debts had been incurred mattered. Many of these circumstances differ in Europe today. Unlike the central government of the United States then, the European 78 The

Tories had either left or remained quiet.

330

Chapter 10: United States Then, Europe Now

Union itself does not have a large debt; instead, the troublesome debts that the market discounts are all obligations incurred by subordinate governments. People live longer, and most do not work on farms. They retire for substantial periods of their lives, and many do not start working until much later in their lives than those early Americans did. There are large public expenditures on education. Medicines and doctors make people healthier and older. Families are weaker. Government-financed safety nets and retirement and medical systems are pervasive and absorb substantial fractions of national budgets. Government regulations of labor markets have changed —slavery is gone; there are minimum wages, unemployment and disability compensation arrangements, and employment protection laws. These differ in their generosity and strength across EU states.79 Are there greater differences in these institutions and people’s skills and preferences across EU member states today than there were in the United States then? In some ways, US member states were much more diverse, for example, in attitudes toward slavery. But in terms of the fraction of GDP that citizens in different states wanted the federal government to consume or redistribute, there was probably much more agreement across member states then than there is in the United States today. Then, beyond redistributing from taxpayers to government creditors, the federal government’s redistributional activities were minimal. Some proponents of a fiscal union in Europe today may want more redistribution and some opponents may want less. There are lessons for the United States now. The government budget constraint and a pricing equation for government debt always prevail. The message of the unpleasant arithmetic of Sargent and Wallace (1981) is that with a responsible fiscal policy—one that sustains present value government budget balance with zero revenues from the inflation tax—it is easy for a monetary authority to sustain low inflation; but that with a profligate fiscal policy, it is impossible for a monetary authority to sustain low inflation because the intertemporal government budget then implies that 79 Ljungqvist

and Sargent (2008) study how differences in these features of social safety nets across countries and continents can account for different outcomes for unemployment in the face of common changes in the microeconomic environment.

Appendix: Outcomes in Graphs

331

the monetary authority must sooner or later impose a sufficiently large inflation tax to finance the budget. In this sense, monetary and fiscal policies cannot be independent. They must be coordinated. There are simple and transparent devices for coordinating fiscal and monetary policies.80, 81 Other more obscure ways are also possible, like one that seems to prevail in the United States today.

Appendix: Outcomes in Graphs Figures 10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, 10.9, and 10.10 show some of the fiscal outcomes of the policies that Washington and Hamilton designed. Figures 10.1 and 10.3 show federal revenues by source from 1790 to 1820, both relative to GDP and per capita, respectively. These figures confirm that customs duties were the dominant source of federal revenues. Notice how much those revenues suffered when, during Madison and Jefferson’s embargo in 1808 and 1809, the United States did eventually use trade policy to retaliate against the British. Today, Hamilton is sometimes characterized as someone who advocated a big state, but that has to be put in the context of the 1790s when, as figure 10.1 shows, a “big state” advocate wanted to raise about 2 percent of GDP in federal revenues and to use much of those revenues to service federal debt. Hamilton and Washington’s policy of forbearance toward the British during the 1790s was designed to protect federal revenues and to avoid the outcomes that Madison and Jefferson’s policy 80 Milton

Friedman may have appeared abruptly to have changed his mind about how to coordinate monetary and fiscal policy, but if you look at it more deeply, he really didn’t. Friedman recommended two apparently diametrically opposed ways to coordinate monetary and fiscal policy. In Friedman (1953), he recommended that the monetary authority use open market operations to purchase 100% of all government debt. That put responsibility for money growth squarely on the shoulders of the fiscal authorities. To control money growth, he recommended that fiscal authorities balance the budget over the business cycle. In Friedman (1960), Friedman reversed himself and instead recommended a version of the famous rule that the monetary authority commit itself to print government issued fiat money at k percent per period no matter what, thereby committing itself to finance at most a small fraction of any government deficit. 81 See Sims (2001) for some pros and cons of “dollarization” as a coordinating device.

332

Chapter 10: United States Then, Europe Now

eventually temporarily brought about. Figures 10.2 and 10.4 show the composition of federal expenditures, both relative to GDP and per capita, respectively. Evidently, throughout the period, a large fraction of expenditures went to servicing the federal debt. Figure 10.5 shows the ratio of the net of interest federal deficit to GDP, while figure 10.6 shows the debt to GDP ratio, where debt is being valued at par. Figure 10.7 shows the growth rate of GDP and the inflation rate. Both of these figures should be viewed as subject to substantial measurement errors. Figure 10.8 shows the composition of the federal debt. The figure shows how the domestic unfunded debt was converted into instruments described in the Act of August 4, 1790 and how rapidly the Treasury managed to carry out that successful debt restructuring. Notice the debt that was issued to help purchase the Louisiana Territory in 1803. Figures 10.9 and 10.10 show per capita real GDP and per capita nominal GDP, again both probably subject to substantial measurement errors. Evidently, the debt to GDP ratio shrank over the period mainly through growth in GDP. D ATA S OURCES The data in these graphs come from the following sources. Figure 10.1: Revenue: Annual Report of the Secretary of the Treasury on the State of the Finances for the Year 1870 (Government Printing Office, Washington, 1870), Table K. Statement of the receipts of the United States from March 4, 1789 to June 30 1870, by calendar years to 1843, and by fiscal years (ending June 30) from that time, pages XXVI-XXVIX. Nominal GDP: Louis Johnston and Samuel H. Williamson, “What Was the U.S. GDP Then?” Measuring Worth 2010. URL: < http://www.measuringworth.org/usgdp/ > . Figure 10.2: Expenditures: Annual Report of the Secretary of the Treasury on the State of the Finances for the Year 1870 (Government Printing Office, Washington, 1870), Table L. Statement of the expenditures of the United States from March 4, 1789 to June 30 1870, by calendar years to 1843, and by fiscal years (ending June 30) from that time, pages XXX-XXXI. Nominal GDP: same as figure 10.1. Figures 10.3 and 10.4: Same revenue and expenditure data as in Figures 10.1 and 10.2. Population: HSUS Table Aa7 (the numbers

Appendix: Outcomes in Graphs

333

include slaves). Also from Measuring Worth. Figure 10.5: Hall and Sargent calculated the primary deficit using the revenue and expenditure data cited above. Nominal GDP same as in Figures 10.1 and 10.2. 1 and 2. Figure 10.6: Debt: Annual Report of the Secretary of the Treasury on the State of the Finances for the Year 1870 (Government Printing Office, Washington, 1870), Table H. Statement of the outstanding principal of the public debt of the United States on the 1st of January of each year, from 1790 to 1842, inclusive; page XXV. Figure 10.7: GDP deflator and real GDP growth from Measuring Worth. Figure 10.8: Reports of the Secretary of the Treasury of the United States: Report on the Finances, December 1815, Table C. Statement of the Public Debt on the 1st day of January, in each of the years, from 1791 to 1815 inclusive; pages 47–50. < http://fraser.stlouisfed.org/docs/publications/treasar/AR _ TREASURY _1815.pdf > . Figures 10.9 and 10.10: Measuring Worth.

334

Chapter 10: United States Then, Europe Now

6 Internal revenue Customs Direct tax Public lands Misc.

5

% of GDP

4

3

2

1

0

1795

1800

1805

1810

1815

1820

Figure 10.1 Composition of federal revenues by source

4 Indians Misc. Interest & Premiums Pensions War Navy

3.5 3

% of GDP

2.5 2 1.5 1 0.5 0

1795

1800

1805

1810

1815

Figure 10.2 Composition of federal expenditures by type

1820

335

Appendix: Outcomes in Graphs

6

Per capita (nominal dollars)

5

Internal revenue Customs Direct tax Public lands Misc.

4

3

2

1

0

1795

1800

1805

1810

1815

1820

Figure 10.3 Per capita composition of federal revenues by source

4.5 4

Per capita (nomial dollars)

3.5 3

Indians Misc. Interest & Premiums Pensions War Navy

2.5 2 1.5 1 0.5 0 1795

1800

1805

1810

1815

1820

Figure 10.4 Per capita composition of federal expenditures by type

336

Chapter 10: United States Then, Europe Now

2

% of GDP

1

0

−1

−2

−3

1790

1795

1800

1805

1810

1815

1820

1810

1815

1820

Figure 10.5 Primary deficit-to-GDP ratio

40 35 30

% of GDP

25 20 15 10 5 0 1790

1795

1800

1805

Figure 10.6 Par value debt-to-GDP ratio

337

Appendix: Outcomes in Graphs

20 15 Inflation

Annual % change

10 5 0 −5

Real GDP

−10 −15 −20

1790

1795

1800

1805

1810

1815

1820

Figure 10.7 Annual inflation and real GDP growth

Par value (millions of nominal dollars)

120

100

Foreign Unfunded Act of Aug 1970 Temp Loans Other Loans 6% Stock of 1812−14

80

60

40

20

0 1795

1800

1805

1810

1815

Figure 10.8 Composition of the debt outstanding by type of obligation

338

Chapter 10: United States Then, Europe Now

1600

1500

2005 dollars

1400

1300

1200

1100

1000 1790

1795

1800

1805

1810

1815

1820

1810

1815

1820

Figure 10.9 Per capita real GDP (2005 dollars)

1600

1500

2005 dollars

1400

1300

1200

1100

1000 1790

1795

1800

Figure 10.10 Per capita nominal GDP

1805

References Adams, Henry C. 1887. Public Debts: An Essay in the Science of Finance. New York: D. Appleton and Company. Aftalion, Florin. 1987. L’Économie de la Revolution française. Paris: Hachette. Aiyagari, S. Rao, Albert Marcet, Thomas J. Sargent, and Juha Seppala. 2002. Optimal Taxation without State-Contingent Debt. Journal of Political Economy 110 (6):1220–1254. Alpert, Paul. 1951. Twentieth Century Economic History of Europe. New York: Henry Schuman. Archives parlementaires. 1862–. Archives parlementaires, 1re série: 1787– 1794. Paris: P. Dupont (Vols. 1–82), CNRS (Vols. 83–95). Arellano, Cristina. 2008. Default Risk and Income Fluctuations in Emerging Economies. American Economic Review 98 (3):690–712. Arrow, Kenneth J. and Mordecai Kurz. 1970. Public Investment, the Rate of Return, and Optimal Fiscal Policy. Baltimore and London: Johns Hopkins University Press. Auernheimer, Leonardo. 1974. The Honest Government’s Guide to the Revenue from the Creation of Money. Journal of Political Economy 82 (3):598–606. Bagehot, Walter. 1920. Lombard Street: A Description of the Money Market. New York: E.P. Dutton and Company. Originally published in 1873. Bailey, Martin. 1971. National Income and the Price Level. 2nd ed. New York: McGraw-Hill. 175–186. Bailleul, Jean-Antoine-Guillaume. 1796. Tableau complet de la valeur des assignats, des rescriptions et des mandats, jour par jour depuis leur émission. Paris: au bureau du Journal du commerce. Bancroft, George. 1886. A Plea for the Constitution of the U.S. of America: Wounded in the House of Its Guardians. New York: Harper and Brothers. Barro, Robert J. 1974. Are Government Bonds Net Wealth? Journal of Political Economy 82 (6):1095–1117.

339

340

References

———. 1976. Rational Expectations and the Role of Monetary Policy. Journal of Monetary Economics 2 (1):1–32. ———. 1977. Long Term Contracting, Sticky Prices, and Monetary Policy. Journal of Monetary Economics 3 (1):305–16. ———. 1979. On the Determination of the Public Debt. Journal of Political Economy 87 (5):940–71. Bassetto, Marco. 2005. Equilibrium and Government Commitment. Journal of Economic Theory 124 (1):79–105. ———. 2006. Fiscal Policy and Price Stability: The Case of Italy, 1992–98. Chicago Fed Letter . Battaglini, Marco and Stephen Coate. 2008. A Dynamic Theory of Public Spending, Taxation, and Debt. American Economic Review 98 (1):201–36. Beard, Charles A. 1913. An Economic Interpretation of the Constitution of the United States. History of Economic Thought Books. Hamilton, Canada: McMaster University Archive for the History of Economic Thought. Bien, David. 1987. Offices, Corps and a System of State Credit: The Uses of Privilege under the Ancien Regime. In Keith Baker, ed., The French Revolution and the Creation of a Modern Political Culture. Oxford: Pergamon Press. Bilson, John. 1980. A Proposal for Monetary Reform. Domestic Studies Program working paper 80–07, Hoover Institution, Stanford, CA. Blume, Lawrence and David Easley. 2006. If You’re so Smart, Why Aren’t You Rich? Belief Selection in Complete and Incomplete Markets. Econometrica 74 (4):929–966. Bomberger, William A. and Gail E. Makinen. 1980. Indexation, Inflationary Finance, and Hyperinflation: The 1945–1946 Hungarian Experience. Journal of Political Economy 88 (3):550–60. ———. 1983. The Hungarian Hyperinflation and Stabilization of 1945– 1946. Journal of Political Economy 91 (5):801–24. Bordo, Michael D. and Eugene N. White. 1991. A Tale of Two Currencies: British and French Finance during the Napoleonic Wars. Journal of Economic History 51 (2):303–316. Borges, Jorge Luis. 1962. Labyrinths. New York: New Directions Publishing Company.

References

341

Bosher, John Francis. 1970. French Finances 1770–1795: From Business to Bureaucracy. Cambridge: Cambridge University Press. Bouchary, Jean. 1937. Le Marché des changes à Paris à la fin du XVIIIe siecle (1778–1800). Paris: P. Hartmann. Bournisien, Charles. 1908. La Vente des Biens nationaux (1). Revue historique 99:244–66. ———. 1909. La Vente des Biens nationaux (2). Revue historique 100:15–46. Braesch, Frédéric. 1934–36. Finances et monnaie révolutionnaires. Nancy: Roumegoux. Bray, Margaret and David M. Kreps. 1987. Rational Learning and Rational Expectations. In George Feiwel, ed., Arrow and the Ascent of Modern Economic Theory, 597–625. New York: New York University Press. Bresciani-Turroni, Costantino. 1937. The Economics of Inflation. London: Allen and Unwin. Brewer, John. 1989. The Sinews of Power: War, money and the English State, 1688–1783. London: Unwin Hyman Ltd. Brezis, Elise S. and François M. Crouzet. 1994. The Role of the Assignats during the French Revolution: Evil or Rescuer? Unpublished. Brown, William Adams, Jr. 1940. The International Gold Standard Reinterpreted. 1914–1934, Vol. 1. New York: National Bureau of Economic Research. Bruno, Michael. 1989. Econometrics and the Design of Economic Reform. Econometrica 57 (2):275–306. Bryant, John and Neil Wallace. 1979. The Inefficiency of Interest-Bearing National Debt. Journal of Political Economy 87 (2):365–381. ———. 1980. A Suggestion for Further Simplifying the Theory of Money. Staff Report 62, Federal Reserve Bank of Minneapolis, Minneapolis. ———. 1984. A Price Discrimination Analysis of Monetary Policy. Review of Economic Studies 51 (2):279–288. Buiter, Willem and Marcus Miller. 1981. Monetary Policy and International Competitiveness. Oxford Economic Papers 33 (supp):143–75.

342

References

Bulow, Jeremy and Kenneth Rogoff. 1989. Sovereign Debt: Is to Forgive to Forget? American Economic Review 79 (1):43–50. Cagan, Phillip. 1956. The Monetary Dynamics of Hyperinflation. In Milton Friedman, ed., Studies in the Quantity Theory of Money, 25–117. Chicago: University of Chicago Press. Calvo, Guillermo A. 1978. On the Time Consistency of Optimal Policy in a Monetary Economy. Econometrica 46 (6):1411–1428. Caron, Pierre. 1909. Tableaux de dépréciation du papier-monnaie. Paris: Imprimerie Nationale. Chari, V. V., Lawrence J. Christiano, and Patrick J. Kehoe. 1994. Optimal Fiscal Policy in a Business Cycle Model. Journal of Political Economy 102 (4):617–52. Chari, V. V. and Larry E. Jones. 2000. A Reconsideration of the Problem of Social Cost: Free Riders and Monopolists. Economic Theory 16 (1):1–22. Chari, V. V. and Patrick J. Kehoe. 1990. Sustainable Plans. Journal of Political Economy 98 (4):783–802. ———. 1993. Sustainable Plans and Mutual Default. Review of Economic Studies 60 (1):175–195. Clamageran, Jean-Jules. 1876. Histoire de l’impôt en France. Vol. 3. Paris: Guillaumin. Compte général. 1789. Compte général des recettes et des dépenses fixes au 1er de mai 1789. Paris: Imprimerie royale. Compte rendu au Roi. 1788. Compte rendu au Roi au mois de mars 1788. Paris: Imprimerie royale. Conklin, James. 1993. Mechanisms of Commitment that Backed Crown Debt during the Reign of Philip II. Working paper, Universidad Carlos III, Madrid. ———. 1998. The Theory of Sovereign Debt and Spain under Philip II. Journal of Political Economy 106 (3):483–513. Cournot, Augustin. 1897. Researches into the Mathematical Principles of the Theory of Wealth. New York: Macmillan Company.

References

343

David, Paul A. and Peter Solar. 1977. A Bicentenary Contribution to the History of the Cost of Living in America. Research in Economic History 2:pp.1–80. Debortoli, Davide and Ricardo Nunes. 2011. Lack of Commitment and the Level of Debt. International Finance Discussion Papers 938, Board of Governors of the Federal Reserve System (U.S.). Dewey, Davis R. 1912. Financial History of the United States, 4th ed. New York: Longmans, Green, and Company. Dornbusch, Rudiger. 1976. Expectations and Exchange Rate Dynamics. Journal of Political Economy 84 (6):1161–1176. ———. 1980. Open Economy Macroeconomics. New York: Basic Books. ———. 1985. Comments, Brazil. In John Williamson, ed., Inflation and Indexation: Argentina, Brazil and Israel. Cambridge, MA: MIT Press. Dornbusch, Rudiger, Federico Sturzenegger, and Holger Wolf. 1990. Extreme Inflation: Dynamics and Stabilization. In Brookings Papers on Economic Activity, 2:1–64. Drazen, Allan. 1985. Tight Money and Inflation: Further Results. Journal of Monetary Economics 15 (1):113–20. Dupâquier, Jacques. 1988. Histoire de la population française. Paris: Presses Universitaires de France. Eaton, Jonathan and Mark Gersovitz. 1981. Debt with Potential Repudiation: Theoretical and Empirical Analysis. Review of Economic Studies 48 (2):289–309. Eckstein, Otto. 1981. Core Inflation. Englewood Cliffs, NJ: Prentice Hall. Elkins, Stanley and Eric McKitrick. 1993. The Age of Federalism. New York: Oxford University Press. Fama, Eugene F. 1980. Banking in the Theory of Finance. Journal of Monetary Economics 6 (1):39–58. Fetter, Frank W. 1965. Development of British Monetary Orthodoxy, 1797– 1875. Cambridge, MA: Harvard University Press. Fischer, Stanley. 1981. Seigniorage and the Case for a National Money. Working paper, Massachusetts Institute of Technology, Cambridge, MA.

344

References

Fisher, Irving. 1926. The Purchasing Power of Money: Its Determination and Relation to Credit, Interest, and Crises. New York: Macmillan Company. Forbonnais, François Véron de. 1758. Recherches et considérations sur les finances de la France. Basel: Kramer. Friedman, Milton. 1948. A Monetary and Fiscal Framework for Economic Stability. American Economic Review 48 (3):245–264. ———. 1953. A Monetary and Fiscal Framework for Economic Stability. In Milton Friedman, ed., Essays in Positive Economics, 133–156. Chicago: University of Chicago Press. ———. 1956. A Theory of the Consumption Function. Princeton: Princeton University Press. ———. 1960. A Program for Monetary Stability. New York: Fordham University Press. ———. 1968. The Role of Monetary Policy. American Economic Review 58 (1):1–17. ———. 1970. A Theoretical Framework for Monetary Analysis. Journal of Political Economy 78 (2):193–238. ———. 1972. Comments on the Critics. Journal of Political Economy 80 (5):906–950. Friedman, Milton and Anna Jacobson Schwartz. 1963. A Monetary History of the United States, 1867–1960. Princeton, NJ: Princeton University Press. Fudenberg, Drew and David M. Kreps. 1987. Reputation in the Simultaneous Play of Multiple Opponents. Review of Economic Studies 54 (4):541–68. Fudenberg, Drew and David K. Levine. 1993. Self-Confirming Equilibrium. Econometrica 61 (3):523–45. Garber, Peter M. 1982. Transition from Inflation to Price Stability. CarnegieRochester Conference Series on Public Policy 16:11–42. Ghosh, Atish R., Jun I. Kim, Enrique G. Mendoza, Jonathan D. Ostry, and Mahvash S. Qureshi. 2011. Fiscal Fatigue, Fiscal Space and Debt Sustainability in Advanced Economies. NBER Working Paper 16782, National Bureau of Economic Research, Inc.

References

345

Graham, Frank D. 1930. Exchange, Prices, and Production in Hyperinflation: Germany 1920–23. New York: Russell & Russell. Granger, C. W. J. 1969. Investigating Causal Relations by Econometric Models and Cross-Spectral Methods. Econometrica 37 (3):424–38. Greif, Avner, Paul Milgrom, and Barry R. Weingast. 1994. Coordination, Commitment, and Enforcement: The Case of the Merchant Guild. Journal of Political Economy 102 (4):745–76. Grossman, Herschel and John Van Huyck. 1988. Sovereign Debt as a Contingent Claim: Excusable Default, Repudiation and Reputation. American Economic Review 78 (5):1088–1097. Haig, Robert Murray. 1929. The Public Finances of Post-War France. New York: Columbia University Press. Hall, George J. and Thomas J. Sargent. Forthcoming. A Fiscal History of the United States. Monograph in preparation. Hall, Robert E. 1978. Stochastic implications of the life cycle – permanent income hypothesis: Theory and Evidence. Journal of Political Economy 86 (6):971–87. Hamilton, Alexander. 1790. Report on Public Credit. Presented to Congress on January 9, 1790. Hansen, Lars Peter. 1982. Large Sample Properties of Generalized Method of Moments Estimators. Econometrica 50 (4):1029–54. ———. 2011. Dynamic Valuation Decomposition with Stochastic Economies. Fisher-Schulz lecture, University of Chicago manuscript. Hansen, Lars Peter, John Heaton, Junghoon Lee, and Nikolai Roussanov. 2007. Intertemporal Substitution and Risk Aversion. In James J. Heckman and Edward E. Leamer, eds., Handbook of Econometrics, Volume 6, Part 1, 3967–4056. Amsterdam: Elsevier. Hansen, Lars Peter and Ravi Jagannathan. 1991. Implications of Security Market Data for Models of Dynamic Economies. Journal of Political Economy 99 (2):225–62. Hansen, Lars Peter and Scott F. Richard. 1987. The Role of Conditioning Information in Deducing Testable Restrictions Implied by Dynamic Asset Pricing Models. Econometrica 55 (3):587–613.

346

References

Hansen, Lars Peter, William Roberds, and Thomas J. Sargent. 1991. Time Series Implications of Present Value Budget Balance and of Martingale Models of Consumption and Taxes. In Rational Expectations Econometrics. Boulder, CO: Westview Press. Hansen, Lars Peter and Thomas J. Sargent. 1980. Formulating and Estimating Dynamic Linear Rational Expectations Models. Journal of Economic Dynamics and Control 2 (1):7–46. ———. 1991. Rational Expectations Econometrics. Boulder, CO: Westview Press. ———. 2013. Risk, Uncertainty, and Value. Princeton, NJ: Princeton. Hansen, Lars Peter and Kenneth J. Singleton. 1983. Stochastic Consumption, Risk Aversion, and the Temporal Behavior of Asset Returns. Journal of Political Economy 91 (2):249–65. Harris, Seymour E. 1930. The Assignats. Cambridge, MA: Harvard University Press. Harrison, J. Michael and David M. Kreps. 1979. Martingales and Arbitrage in Multiperiod Securities Markets. Journal of Economic Theory 20 (3):381–408. Hawtrey, Ralph G. 1919. Currency and Credit. London: Longmans, Green and Co. ———. 1933. The Art of Central Banking. New York: Longmans, Green and Co. Hicks, John R. 1969. A Theory of Economic History. Oxford: Oxford University Press. Holtfrerich, Carl-Ludwig. 1980. The German Inflation, 1914–23. Berlin and New York: Walter de Gruyter. Howe, Daniel Walker. 2007. What Hath God Wrought: The Transformation of America, 1815–1848. New York: Oxford University Press. Irwin, Douglas A. 2009. Revenue or Reciprocity? Founding Feuds over Early U.S. Trade Policy. In Founding Choices: American Economic Policy in the 1790s, NBER Chapters, 89–120. Cambridge, MA: National Bureau of Economic Research, Inc.

References

347

Jones, Larry E., Rodolfo E. Manuelli, and Peter E. Rossi. 1997. On the Optimal Taxation of Capital Income. Journal of Economic Theory 73 (1):93– 117. Kareken, John H. 1970. The Optimum Monetary Instrument Variable. Journal of Money, Credit, and Banking 2 (3):385–390. Kareken, John H. and Neil Wallace. 1978a. Deposit Insurance and Bank Regulation: A Partial-Equilibrium Exposition. Journal of Business 51 (3):413–38. ———. 1978b. Samuelson’s Consumption Loan Model with CountrySpecific Fiat Monies. Staff Report 24, Federal Reserve Bank of Minneapolis, Minneapolis. Kareken, John H. and Neil Wallace, eds. 1980. Models of Monetary Economies. Minneapolis: Federal Reserve Bank of Minneapolis. Kareken, John H. and Neil Wallace. 1981. On the Indeterminacy of Equilibrium Exchange Rates. Quarterly Journal of Economics 96 (2):207– 222. Keynes, John Maynard. 1924. Monetary Reform. New York: Harcourt Brace Jovanovich. ———. 1925. The United States and Gold. In John Parke Young, ed., European Currency and Finance. Washington: Government Printing Office. ———. 1940. How to Pay for the War. London: Macmillan and Co. ———. 1971. A Tract on Monetary Reform, Preface to the French edition, in The Collected Writings, 4:xvi–xxii. London: Macmillan, St. Martin’s Press, for the Royal Economic Society. Klein, Paul, Per Krusell, and José-Víctor Ríos-Rull. 2011. Time-Consistent Public Policy. University of Minnesota manuscript. Klein, Paul and José-Víctor Ríos-Rull. 2003. Time-Consistent Optimal Fiscal Policy. International Economic Review 44 (4):1217–1245. Kletzer, Kenneth M. and Brian D. Wright. 2000. Sovereign Debt as Intertemporal Barter. American Economic Review 90 (3):621–639. Kreps, David M. 1997. Economics – The Current Position. Daedalus 126 (1):59–85. Proceedings of the American Academy of Arts and Sciences.

348

References

Krishnamurthy, Arvind and Annette Vissing-Jorgensen. 2010. The Aggregate Demand for Treasury Debt. Working paper, National Bureau of Economic Research. Kydland, Finn E. and Edward C. Prescott. 1977. Rules Rather Than Discretion: The Inconsistency of Optimal Plans. Journal of Political Economy 85 (3):473–91. League of Nations. 1926. Memorandum on Currency and Central Banks: 1913–1925. Geneva: League of Nations. ———. 1946. The Course and Control of Inflation. Geneva: League of Nations. Leijonhufvud, Axel. 1984. Inflation and Economic Performance. In Barry N. Siegel, ed., Money in Crisis: The Federal Reserve, the Economy and Monetary Reform, 19–36. San Francisco: Pacific Institute for Public Policy Research. Litterman, Robert B. 1979. Techniques of Forecasting Using Vector Autoregressions. Working paper 115, Federal Reserve Bank of Minneapolis, Minneapolis. Liviatan, Nissan. 1983. On Equilibrium Wage Indexation and Neutrality of Indexation Policy. In P. Aspe-Armella, R. Dornbusch, and M. Obstfeld, eds., Financial Policies and the World Capital Market: The Problem of Latin American Countries. Chicago: University of Chicago Press. ———. 1984. Tight Money and Inflation. Journal of Monetary Economics 13 (1):5–15. Ljungqvist, Lars and Thomas J. Sargent. 2008. Two Questions about European Unemployment. Econometrica 76 (1):1–29. Lucas, Robert E., Jr. 1972a. Econometric Testing of the Natural Rate Hypothesis. In O. Eckstein, ed., The Econometrics of Price Determination. Washington, DC: Board of Governors, Federal Reserve System. ———. 1972b. Expectations and the Neutrality of Money. Journal of Economic Theory 4 (2):103–124. ———. 1976. Econometric Policy Evaluation: A Critique. CarnegieRochester Conference Series on Public Policy 1 (1):19–46. ———. 1978. Asset Prices in an Exchange Economy. Econometrica 46 (6):1429–45.

References

349

———. 1980. Equilibrium in a Pure Currency Economy. In John H. Kareken and Neil Wallace, eds., Models of Monetary Economies. Minneapolis, MN: Federal Reserve Bank of Minneapolis. Lucas, Robert E., Jr. and Edward C. Prescott. 1971. Investment Under Uncertainty. Econometrica 39 (5):659–81. ———. 1974. Equilibrium Search and Unemployment. Journal of Economic Theory 7 (2):188–209. Lucas, Robert E., Jr. and Thomas J. Sargent. 1979. After Keynesian Macroeconomics. Federal Reserve Bank of Minneapolis Quarterly Review 3:1–16. ———. 1981a. Introductory Essay. In Robert E. Lucas, Jr. and Thomas J. Sargent, eds., Rational Expectations and Econometric Practice. Minneapolis: University of Minnesota Press. Lucas, Robert E., Jr. and Thomas J. Sargent, eds. 1981b. Rational Expectations and Econometric Practice. Minneapolis: University of Minnesota Press. Lucas, Robert E., Jr. and Nancy L. Stokey. 1983. Optimal Fiscal and Monetary Policy in an Economy Without Capital. Journal of Monetary Economics 12 (1):55–93. Macaulay, Thomas B. 1831. The History of England from the Accession of James II. Vol. 3. Boston: DeWolfe. Madison, James. 1956. Notes of Debates in the Federal Convention of 1787. New York: Norton. Mailath, George and Andrew Postlewaite. 1990. Asymmetric Information Bargaining Problems with Many Agents. Review of Economic Studies 57 (3):351—-367. Mallet, Jean Roland. 1789. Comptes rendus de l’administration des finances du royaume de France. Paris: Buisson. Manuelli, Rodolfo E. 1988. A General Equilibrium Model of International Credit Markets. Working paper, Stanford University, Stanford. Marcet, Albert and Thomas J. Sargent. 1989. Convergence of Least Squares Learning Mechanisms in Self-Referential Linear Stochastic Models. Journal of Economic Theory 48 (2):337–368. Marion, Marcel. 1908. La Vente des Biens nationaux pendant la Revolution. Paris: H. Champion.

350

References

———. 1914–31. Histoire financière de la France. Paris: Arthur Rousseau & Cie. Mathon de La Cour, Charles J. 1788. Collection de comptes-rendus. Paris: Cuchet & Gattey. McCallum, Bennett T. 1978. On Macroeconomic Instability from a Monetarist Policy Rule. Economics Letters 1 (2):121–124. ———. 1981. Monetarist Principles and the Money Stock Growth Rule. American Economic Review 71 (2):134–38. ———. 1984. Are Bond-Financed Deficits Inflationary? A Ricardian Analysis. Journal of Political Economy 92 (1):123–135. McDonald, Forrest. 1985. Novus Ordo Seclorum: the Intellectual Origins of the Constitution. Lawrence, Kansas: University of Kansas. McPherson, James. 1988. Battle Cry of Freedom. New York: Oxford University Press. Meltzer, Allan H. 1981. Test of Inflation Theories from the British Laboratory. The Banker 131:21–27. Meltzer, Allan H. and Scott F. Richard. 1981. A Rational Theory of the Size of Government. Journal of Political Economy 89 (5):914–27. Miller, Preston. 1981. Fiscal Policy in a Monetarist Model. Staff Report 67, Federal Reserve Bank of Minneapolis, Minneapolis. Mitchell, Brian R. 1988. British Historical Statistics. Cambridge: Cambridge University Press. Moreau, Émile. 1991. The Golden Franc. Boulder, CO: Westview Press. Morgan Guaranty Trust Company. 1981. Thatcherism: A Mid-Term Review. Muth, John F. 1960. Optimal Properties of Exponentially Weighted Forecasts. Journal of the American Statistical Association 55 (290):299–306. ———. 1961. Rational Expectations and the Theory of Price Movements. Econometrica 29 (3):315–335. Necker, Jacques. 1784. De l’administration des finances de la France. Paris. ———. 1820. Œuvres complètes. Paris: Treuttel et Wurtz.

References

351

Nichols, Donald. 1974. Some Principles of Inflationary Finance. Journal of Political Economy 82 (2–1):423–30. Nicolas, Charles. 1882. Les Budgets de la France depuis le commencement du 19e siecle. Paris: Lahure. North, Douglass C. and Barry R. Weingast. 1989. Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England. The Journal of Economic History 49 (4):803–32. Paal, Beatrix. 2000. Destabilizing Effects of a Successful Stabilization: A Forward-Looking Explanation of the Second Hungarian Hyperinflation. Economic Theory 15 (3):599–630. Pasvolsky, Leo. 1928. Economic Nationalism of the Danubian States. New York: Macmillan. Paxton, Robert O. 1975. Europe in the Twentieth Century. New York: Harcourt Brace Jovanovich. Peled, Dan. 1980. Government Index Bonds—Do they Improve Matters? Ph.D. thesis, University of Minnesota. Persson, Mats, Torsten Persson, and Lars E. O. Svensson. 1987. Time Consistency of Fiscal and Monetary Policy. Econometrica 55 (6):1419– 1431. Phelps, Edmund S. and John B. Taylor. 1977. Stabilizing Powers of Monetary Policy under Rational Expectations. Journal of Political Economy 85 (1):163–190. Poole, William. 1970. Optimal Choice of Monetary Policy Instruments in a Simple Stochastic Macro Model. Quarterly Journal of Economics 84 (2):197–216. Prescott, Edward C. 1977. Should Control Theory Be Used for Economic Stabilization. Carnegie-Rochester Conference Series on Public Policy 7:13– 38. Rakove, Jack M. 1997. Original Meanings: Politics and Ideas in the Making of the Constitution. New York: Vintage. Ramel de Nogaret, Dominique-Vincent. 1801. Des Finances de la Republique française en l’an IX. Paris: H. Agasse.

352

References

Ratchford, Benjamin Ulysses. 1941. American State Debts. Durham, NC: Duke University Press. Reid, Constance. 1979. Hilbert. New York: Springer-Verlag. Riley, James C. 1986. The Seven Years War and the Old Regime in France. Princeton: Princeton University Press. ———. 1987. French Finances, 1727–68. Journal of Modern History 59:209– 243. Rogers, Carol Ann. 1986. The Effect of Distributive Goals on the Time Consistency of Optimal Taxes. Journal of Monetary Economics 17 (2):251– 270. Rogers, James Harvey. 1929. The Process of Inflation in France 1914–1927. New York: Columbia University Press. Rolnick, Arthur J. and Warren E. Weber. 1983. New Evidence on the Free Banking Era. American Economic Review 73 (5):1080–91. ———. 1984. The Causes of Free Bank Failures: A Detailed Examination. Journal of Monetary Economics 14 (3):267–291. Saidi, Nasser H. 1980. Fluctuating Exchange Rates and the International Transmission of Economic Disturbances. Journal of Money, Credit, and Banking 12 (4–1):575–91. Samuelson, Paul A. 1958. An Exact Consumption-Loan Model of Interest with or without the Social Contrivance of Money. Journal of Political Economy 66 (6):467–82. Sargent, Thomas J. 1971. A Note on the “Accelerationist” Controversy. Journal of Money, Credit and Banking 3 (3):721–25. ———. 1977. The Demand for Money During Hyperinflations under Rational Expectations: I. International Economic Review 18 (1):59–82. ———. 1979a. Macroeconomic Theory. New York: Academic Press. ———. 1979b. A Note on Maximum Likelihood Estimation of the Rational Expectations Model of the Term Structure. Journal of Monetary Economics 5 (1):133–143. ———. 1981. Interpreting Economic Time Series. Journal of Political Economy 89 (2):213–48.

References

353

———. 1982. The Ends of Four Big Inflations. In Inflation: Causes and Effects, NBER Chapters, 41–98. Cambridge, MA: National Bureau of Economic Research, Inc. ———. 1984. Autoregressions, Expectations, and Advice. American Economic Review 74 (2):408–15. ———. 1986. Rational Expectations and Inflation. New York: Harper and Row. ———. 1987. Macroeconomic Theory, 2d ed. New York: Academic Press. ———. 2008. Evolution and Intelligent Design. American Economic Review 98 (1):5–37. Sargent, Thomas J. and François R. Velde. 1995. Macroeconomic Features of the French Revolution. Journal of Political Economy 103 (3):474–518. Sargent, Thomas J. and Neil Wallace. 1973. Rational Expectations and the Dynamics of Hyperinflation. International Economic Review 14 (2):328– 50. ———. 1975. Rational Expectations, the Optimal Monetary Instrument, and the Optimal Money Supply Rule. Journal of Political Economy 83 (2):241–54. ———. 1976. Rational Expectations and the Theory of Economic Policy. Journal of Monetary Economics 2 (2):169–83. ———. 1981. Some Unpleasant Monetarist Arithmetic. Federal Reserve Bank of Minneapolis Quarterly Review 5:1–17. ———. 1982. The Real-Bills Doctrine versus the Quantity Theory: A Reconsideration. Journal of Political Economy 90 (6):1212–36. ———. 1983. A Model of Commodity Money. Journal of Monetary Economics 12 (1):163–87. Scarth, William M. 1980. Rational Expectations and the Instability of Bond-Financing. Economics Letters 6 (4):321–27. Schama, Simon. 1989. Citizens. New York: Knopf. Scott, William A. 1893. The Repudiation of State Debts: A Study in the Financial History of Mississippi, Florida, Alabama, North Carolina, South Carolina, Georgia, Louisiana, Arkansas, Tennessee, Minnesota, Michigan, and Virginia. New York and Boston: Thomas W. Crowell & Company.

354

References

Shirer, William L. 1969. The Collapse of the Third Republic: An Inquiry into the Fall of France in 1940. New York: Simon & Schuster. Siklos, Pierre L. 1989. The End of the Hungarian Hyperinflation of 1945– 1946. Journal of Money, Credit and Banking 21 (2):135–147. ———. 1993. Interpreting a Change in Monetary Policy Regimes: a Reappraisal of the First Hungarian Hyperinflation and Stabilization, 1921—28. In Michael D. Bordo, ed., Monetary Regimes in Transition, 274–311. Cambridge: Cambridge University Press. Sims, Christopher A. 1972. Money, Income, and Causality. American Economic Review 62 (4):540–52. ———. 1980. Macroeconomics and Reality. Econometrica 48 (1):1–48. ———. 2001. Fiscal Consequences for Mexico of Adopting the Dollar. Journal of Money, Credit and Banking 33 (2):597–616. Smith, Adam. 1776. An Inquiry into the Wealth of Nations. London: Strahan and Cadell. ———. 1806. An Inquiry into the Nature and Causes of the Wealth of Nations, in three volumes, Vol. II. Edinburgh: William Greech. Stokey, Nancy L. 1989. Reputation and Time Consistency. American Economic Review 79 (2):134–39. ———. 1991. Credible Public Policy. Journal of Economic Dynamics and Control 15 (4):627–656. Stourm, Robert. 1885. Les Finances de l’Ancien régime et de la Révolution. Paris: F. Alcan. Sylla, Richard. 2009. Financial Foundations: Public Credit, the National Bank, and Securities Markets. In Founding Choices: American Economic Policy in the 1790s, NBER Chapters, 59–88. Cambridge, MA: National Bureau of Economic Research, Inc. Taylor, John B. 1979. Estimation and Control of a Macroeconomic Model with Rational Expectations. Econometrica 47 (5):1267–1286. ———. 1982. The Role of Expectations in the Choice of Monetary Policy. In Monetary Policy Issues in the 1980s, Economic Symposium Conference Proceedings August 9–10, 1982, 47–76. Kansas City: Federal Reserve Bank of Kansas City.

References

355

Townsend, Robert M. 1980. Models of Money with Spatially Separated Agents. In John H. Kareken and Neil Wallace, eds., Models of Monetary Economies. Minneapolis: Federal Reserve Bank of Minneapolis. Velde, François R. 2009. Chronicle of a Deflation Unforetold. Journal of Political Economy 117 (4):591–634. Velde, François R. and Warren E. Weber. 2000. A Model of Bimetallism. Journal of Political Economy 108 (6):1210–1234. Velde, François R. and David R. Weir. 1992. The Financial Market and Government Debt Policy in France, 1746–1793. Journal of Economic History 52 (1):1–39. Villamil, Anne. 1988. Price Discriminatory Monetary Policy: A Nonuniform Pricing Approach. Journal of Public Economics 35:385–392. Wallace, Neil. 1980. The Overlapping Generations Model of Fiat Money. In John H. Kareken and Neil Wallace, eds., Models of Monetary Economies. Minneapolis: Federal Reserve Bank of Minneapolis. ———. 1981. A Modigliani-Miller Theorem for Open-Market Operations. American Economic Review 71 (3):267–74. Wallis, John Joseph. 2000. American Government Finance in the Long Run: 1790 to 1990. Journal of Economic Perspectives 14 (1):61–82. ———. 2001. A History of the Property Tax in America. In Wallace Oates and Wallace E. Oates, eds., Property Taxation and Local Government Finance, 123–147. Cambridge, MA: Lincoln Institute of Land Policy. Wallis, John Joseph and Barry R. Weingast. 2005. Equilibrium Impotence: Why the States and Not the American National Government Financed Economic Development in the Antebellum Era. Working paper 11397, National Bureau of Economic Research. Webb, Steven B. 1989. Hyperinflation and Stabilization in Weimar Germany. Oxford: Oxford University Press. Weber, Warren E. 2009. Clearing Arrangements in the United States before the Fed. Federal Reserve Bank of Minneapolis, manuscript. Weir, David R. 1989. Tontines, Public Finance, and Revolution in France and England, 1688–1789. Journal of Economic History 49 (1):95–124. White, Eugene N. 1989. Was There a Solution to the Ancien Régime’s Financial Dilemma? Journal of Economic History 49 (3):545–568.

356

References

———. 1990. Free Banking during the French Revolution. Explorations in Economic History 27 (3):251–76. Wicker, Elmus R. 1986. Terminating Hyperinflation in the Dismembered Habsburg Monarchy. American Economic Review 76 (3):350–64. Wills, Gary. 2002. James Madison. New York: Times books, Henry Holt and Company. Wood, Gordon. 2009. Empire of Liberty: A History of the Early Republic, 1789– 1815 (Oxford History of the United States). Oxford: Oxford University Press. Wright, Robert E. 2008. One Nation Under Debt: Hamilton, Jefferson, and the History of What We Owe. New York: McGraw Hill. Yeager, Leland B. 1976. International Monetary Relations: Theory, History, and Policy. 2nd ed. New York: Harper & Row. Young, John Parke. 1925. European Currency and Finance. Commission of Gold and Silver Inquiry, United States Senate, serial 9. Washington, DC: Government Printing Office.

Author Index

Adams, Henry C. 324, 326 Aftalion, Florin 249 Aiyagari, S. Rao 303, 316, 324 Alpert, Paul 118, 119, 123 Arellano, Cristina 304, 305 Arrow, Kenneth J. 45 Auernheimer, Leonardo 281 Bagehot, Walter 322 Bailey, Martin 131 Bailleul, Jean-Antoine-Guillaume 281 Bancroft, George 321 Barro, Robert J. 39, 111, 114, 126, 199, 208, 256, 316, 324 Bassetto, Marco 306, 312, 316 Battaglini, Marco 319 Beard, Charles A. 310 Bien, David 258 Bilson, John 122 Blume, Lawrence 301 Bomberger, William A. xix Bordo, Michael D. 249, 293, 294 Borges, Jorge Luis 307 Bosher, John Francis 259, 294 Bouchary, Jean 281, 288 Bournisien, Charles 296 Braesch, Frédéric 277, 278 Bray, Margaret 301 Bresciani-Turroni, Costantino 42, 62, 171 Brewer, John 316 Brezis, Elise S. 283 Brittan, Samuel 128 Brown, William Adams, Jr. 117 Bruno, Michael xix

Bryant, John 9, 29, 121, 134, 163, 276 Buiter, Willem 138, 141 Bulow, Jeremy 250, 261, 314 Cagan, Phillip 6, 23, 67, 135, 171, 180, 250, 276 Calvo, Guillermo A. 32, 281 Caron, Pierre 279 Chari, V. V. xx, 250, 256, 261, 262, 306, 308, 314 Christiano, Lawrence J. 256, 314 Clamageran, Jean-Jules 255 Coate, Stephen 319 Conklin, James 248, 263, 297 Cosell, Howard 17 Cournot, Augustin 308, 309 Crouzet, François M. 283 David, Paul A. 309 Debortoli, Davide 319 Dewey, Davis R. 313, 316, 321 Dornbusch, Rudiger xix, 117, 138, 233, 234 Drazen, Allan xix Dupâquier, Jacques 270 Easley, David 301 Eaton, Jonathan 305 Eckstein, Otto 111 Elkins, Stanley 308 Fama, Eugene F. 53, 67 Fetter, Frank W. 292 Fischer, Stanley 114, 121 Fisher, Irving 297 Forbonnais, François Véron de 263

357

358

Author Index

Friedman, Milton 8, 29, 126, 165, 166, 173, 198, 199, 273, 297, 331 Fudenberg, Drew 306, 327 Garber, Peter M. xix Gersovitz, Mark 305 Ghosh, Atish R. 318 Graham, Frank D. 60, 63, 64, 106 Granger, C. W. J. 304 Greif, Avner 254 Grossman, Herschel 257 Haig, Robert Murray 118, 120 Hall, George J. 308 Hall, Robert E. 199, 208 Hamilton, Alexander 310, 312, 313 Hansen, Lars Peter 2, 12, 250, 297, 301–305 Harris, Seymour E. 249 Harrison, J. Michael 305 Hawtrey, Ralph G. 144, 292 Heaton, John 304 Hicks, John R. 254 Holtfrerich, Carl-Ludwig xix Howe, Daniel Walker 322 Huyck, John Van 257 Irwin, Douglas A. 309, 311 Jagannathan, Ravi 305 Jones, Larry E. 308, 314 Kareken, John H. 46, 131, 141, 223, 327 Kehoe, Patrick J. xx, 250, 256, 261, 262, 306, 314 Keynes, John Maynard 22, 45, 46, 49, 61, 66, 222, 233, 237, 276 Kim, Jun I. 318 Klein, Paul 319 Kletzer, Kenneth M. 314 Kreps, David M. 301, 305, 327 Krishnamurthy, Arvind 316 Krusell, Per 319

Kurz, Mordecai 45 Kydland, Finn E. 32, 202, 297, 319 League of Nations 43, 58, 60, 75, 82 Lee, Junghoon 304 Leijonhufvud, Axel 142, 291 Levine, David K. 306 Litterman, Robert B. 1 Liviatan, Nissan xix, 31 Ljungqvist, Lars 330 Lucas, Robert E., Jr. xviii, 1, 2, 11, 39, 46, 111, 114, 163, 202, 256, 297, 301, 303, 305, 314 Macaulay, Thomas B 254 Madison, James 319 Mailath, George 308 Makinen, Gail E. xix Mallet, Jean Roland 263 Manuelli, Rodolfo E. 250, 314 Marcet, Albert 301, 303, 316, 324 Marion, Marcel 249, 255, 296 Mathon de La Cour, Charles J. 255 McCallum, Bennett T. 27, 111, 127, 163, 174 McDonald, Forrest 309, 313 McKitrick, Eric 308 McPherson, James 323 Meltzer, Allan H. 124, 126, 130, 315 Mendoza, Enrique G. 318 Meredith, Don 17 Milgrom, Paul 254 Miller, Marcus 138, 141 Miller, Preston 163 Mitchell, Brian R 252, 253, 263 Moreau, Émile 144 Morgan Guaranty Trust Company 124 Muth, John F. 297, 301 Necker, Jacques 255, 262, 264 Nichols, Donald 46, 49 Nicolas, Charles 270

Author Index

North, Douglass C. 240, 253, 316 Nunes, Ricardo 319

359

273, 276, 291, 297, 301–304, 306, 308–310, 315–317, 322, 324, 330 Scarth, William M. 163, 174 Ostry, Jonathan D. 318 Schama, Simon 266 Schwartz, Anna Jacobson 297 Paal, Beatrix 314 Scott, William A. 325 Pasvolsky, Leo 50, 51, 55, 66, 70, 73, Seppala, Juha 303, 316, 324 75, 81 Shaefer, Jeffrey 142 Paxton, Robert O. 60 Shirer, William L. 118, 121 Peled, Dan 31 Siklos, Pierre L. xix Persson, Mats 281 Sims, Christopher A. 2, 297, 304, Persson, Torsten 281 331 Phelps, Edmund S. 112, 114 Singleton, Kenneth J. 305 Poole, William 131 Smith, Adam 273, 280, 317, 322 Postlewaite, Andrew 308 Solar, Peter 309 Prescott, Edward C. 32, 202, 250, Stokey, Nancy L. xviii, xx, 202, 256, 297, 301, 319 297, 306, 314 Stourm, Robert 249 Quah, Danny 193 Sturzenegger, Federico xix Qureshi, Mahvash S. 318 Svensson, Lars E. O. 281 Sylla, Richard 311 Rakove, Jack M. 298, 309 Ramel de Nogaret, Taylor, John B. 112, 117, 144 Dominique-Vincent 277, 278 Townsend, Robert M. 39, 46 Ratchford, Benjamin Ulysses 325 Reid, Constance 62 Velde, François R. 264, 266, 294, Richard, Scott F. 305, 315 297, 310, 315 Riley, James C. 255, 266 Villamil, Anne 276 Ríos-Rull, José-Víctor 319 Vissing-Jorgensen, Annette 316 Roberds, William 250, 301, 302, 304 Wallace, Neil 9, 29, 33, 39, 46, 47, Rogers, Carol Ann 264 64, 66, 121, 131, 134, 135, 141, Rogers, James Harvey 118 163, 172, 198, 223, 250, 255, 273, Rogoff, Kenneth 250, 261, 314 276, 306, 309, 317, 322, 327, 330 Rolnick, Arthur J. 323 Wallis, John Joseph 326 Rossi, Peter E. 314 Webb, Steven B. xix Roussanov, Nikolai 304 Weber, Warren E. 297, 323 Saidi, Nasser H. 138 Weingast, Barry R. 240, 253, 254, Samuelson, Paul 39 316, 326 Samuelson, Paul A. 64, 177 Weir, David R. 251, 252, 264, 266, 294 Sargent, Thomas J. 1–3, 9, 12, 29, 39, 46, 47, 66, 116, 131, 135, 163, 172, White, Eugene N. 249, 266, 279, 198, 199, 202, 208, 223, 250, 255, 293, 294

360

Wicker, Elmus R. xix Wills, Gary 311 Wolf, Holger xix Wood, Gordon 308 Wright, Brian D. 314

Author Index

Wright, Robert E. 311 Yeager, Leland B. 115, 118, 119, 123 Young, John Parke 49, 53, 61–63, 71–74, 78, 81, 82, 87, 89, 90, 93–95, 105–110, 117

Subject Index

actions isolated versus recurrent, 46 agent as a constrained optimization problem, 305 Allied blockade, 47 armistice of 1918, 58 assignat, 249 Austro-Hungarian Empire, 47 backing of currency guillotine, 248 land, 248, 250 legal restrictions, 250 taxes, 248, 250 backing of government debt seigniorage, 45 taxes, 28, 45 balanced budget, 57 Bismarck, Otto von, 239 Brazil, 228 Bremridge, Sir John, 226 Briand, Aristide, 144

US two of them, 297, 298 consumption function, 10 contingency, 19, 208 contingency plan, 17, 20 contracts overlapping, 112 Taylor model, 112 coordination monetary and fiscal policy, 18, 162 core inflation, 111 Cowles Commission, 11 credible policy, 239 diciphering, 239 Necker’s explanation, 262 Reaganomics, 33 Currency Board, 214 currency reform, 49

Dawes, Charles, 62, 63 debt management, 22, 163 decision rules, 1, 10–13, 113 default, 248, see also reform deficit Caisse d’Amortissement, 120 capital account, 133 capital levy, 119 current account, 133 Carter, Jimmy, 113 Devisenzentrale, 49 Charles, King of Hungary, 54 Directory, 249 Chartered Bank, 213 dominance, 20, 36 chicken fiscal versus monetary authorigame of, 33, 35, 145, 146, 204, ties, 36 205, 233 dynamic game, 19–22 Neil Wallace, 204 China, 226 econometric model, 1, 2, 6, 10, 46 Clegg Commission, 135, 136 European Union, 298 commitment technology, 20 Exchange Fund, 212–218, 221, 222, Constitutions 224, 225

361

362

Subject Index

exchange rate depreciation, 218 determinacy, 215 Federal Reserve Board, 164, 232 Fisher effect, 128 football team, 17 France 1926, 23 hyperinflation in, 287 revolution, 248 free-rider problem, 299 game dynamic, 19 of chicken, 33, 35, 145, 146, 204, 205, 233 team, 19 timing protocol, 19, 307 Girondins, 249, 283 Gold standard, 42 government budget constraint intertemporal, 197 government deficit inflationary consequences, 6, 22 guillotine backed currency scheme, 249 wage and price controls, 249 Hamilton, Alexander, 245 Hapsburg monarchy, 53 Hegedus, Lorant finance minister in Hungary, 66 Herriot, Edouard Marie French prime minister, 119 Hilbert, David and units of measurement, 62, 241 Hong Kong 1997, 211 dollar, 213

Exchange Fund, 212 float policy, 214 Hong Kong and Shanghai Bank, 213 Horthy, Admiral Miklos, 54 Howe, Sir Geoffrey, 132 hyperinflation in Austria, 47 in France, 249, 276, 287 in Germany, 60, 236 in Hungary, 24, 53 in Poland, 58 not in Czechoslovakia, 64 investment decision, 3 schedule, 3, 10 tax credit, 5 investment decision, 203 irrelevance of some policies, 116 Jacobins, 249 k-percent rule, 127 Karolyi, Prince, 54 Kemp-Roth tax legislation, 202 Keynes, John Maynard, 45, 222 King Louis XVI, 248 Kun, Bela, 54 League of Nations, 43, 50, 56 Lucas Critique, 11 Maximum laws of, 284, see also wage and price controls Mexico 1982, 23 model as collection of constrained optimization problems, 305 momentum, 39, 111 monetarist arithmetic

Subject Index

363

Rentenmark, 62 Reparation Commission, 236 claims on Hungary, 55 Reparation payments Germany and Brazil, 236 Reparations Commission, 47, 51, 56, Necker, Jacques, 239 118 neutrality, 116, see also irrelevance reputations North Sea Oil, 136 costly to acquire, 299 sustaining distinct ones with Pareto inferior, 180 different people, 299 pattern imposition, 307, see also patrules of the game, 1, 5 tern recognition Rumania, 54, 56 pattern recognition, 250, see also patRussia, 28, 58, 61 tern imposition permanent income, 199 seigniorage, 162 Phillips curve, 64, 112, 114, 115, 117, self-confirming equilibrium, 306 140, 141, 228, 230 simultaneous equations Poincaré, Raymond, 118 econometrics, 10 Poincaré miracle, 116, 143 speculation, 211 Polish State Loan Bank, 58 stabilization in Austria, 52 quantity theory, 24, 52 in France, 119, 291 Rasin, Dr. Alois, 64 in Germany, 62 rational expectations, 2, 4, 15, 22, 25, in Hungary, 56 31, 39, 40, 47, 67, 111–116, in Italy, 117 121, 123, 124, 126, 127, 129, in Poland, 58 139, 173, 199 Stackelberg Reagan timing protocol, 20 monetary and fiscal policies, Stigler, George, 197 197 strategic interdependence, xxii, 20 Reaganomics, 33 structural parameter, 10 real interest rates tax smoothing, 199 high, 198 terror Real plan, 241 in France, 283 reconstruction loan, 56 red, 54 reform, 248, 307, see also default white, 54 regime, 46 Thatcher, Margaret, 111, 232 Bryant-Wallace, 29 time consistency, 32, 202 debt servicing, 9, 31 Treaty Ricardian, 8, 27, 33, 36 of Saint-Germain, 48, 54, 65 Reichsbank, 60 Rentenbank, 62 of Trianon, 54, 55 unpleasant, 198 monetarist models, 127, 166 monetary policy tight or easy, 163 Moreau, Émile, 144

364

Subject Index

of Versailles, 118 value added tax (VAT), 125 Volpi, Count, 117 wage and price controls and guillotine, 249

in Brazil, 233 in the UK, 115 Walters, Alan, 145 waltz of the portfolios, 23, 119 Washington, George, 245