Economic Policy Reforms and the Indian Economy 9780226454542

Citation preview

Economic Policy Reforms and the Indian Economy

Center for Research on Economic Development and Policy Reform

Economic Policy Reforms and the Indian Economy

Edited by

Anne O. Krueger

The University of Chicago Press Chicago and London

A O. K is currently the first deputy managing director at the International Monetary Fund. At the time of editing this volume, she was the Herald L. and Caroline L. Ritch Professor of Economics and director of the Center for Research on Economic Development and Policy Reform at Stanford University.

The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London © 2002 by The University of Chicago All rights reserved. Published 2002 Printed in the United States of America 11 10 09 08 07 06 05 04 03 02 1 2 3 4 5 ISBN: 0-226-45452-5 (cloth)

Library of Congress Cataloging-in-Publication Data Economic policy reforms and the Indian economy / edited by Anne O. Krueger. p. cm. Papers presented at a conference held in May of 2000 at Stanford’s Center for Research on Economic Development and Policy Reform. Includes bibliographical references and index. ISBN 0-226-45452-5 (cloth : alk. paper) 1. India—Economic policy—1980—Congresses. 2. Fiscal policy—India—Congresses. 3. Industrial policy—India— Congresses. 4. Education and state—India—Congresses. I. Krueger, Anne O. II. Stanford University. Center for Research on Economic Development and Policy Reform. HC435.2 .E31363 2002 338.954—dc21 2002018129

o The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI Z39.48-1992.

Contents

Foreword George P. Shultz

vii

Acknowledgments

xi

Abbreviations Chronology of Major Political and Economic Events Introduction Anne O. Krueger

xiii xv 1

I. C S   E 1. The Indian Economy in Global Context Anne O. Krueger and Sajjid Chinoy 2. India’s Fiscal Situation: Is a Crisis Ahead? T. N. Srinivasan Comment: Shankar Acharya Comment: Kenneth Kletzer Comment: N. K. Singh 3. State-Level Performance under Economic Reforms in India Montek S. Ahluwalia Comment: Shankar Acharya

9 47

91

v

vi

Contents

II. P E A 4. Doing Business in India: What has Liberalization Changed? Naushad Forbes 5. Bangalore: The Silicon Valley of Asia? Annalee Saxenian Comment: N. R. Narayana Murthy and Sandeep Raju Comment: Ashok Desai

129 169

III. G A 6. Small-Scale Industry Policy in India: A Critical Evaluation Rakesh Mohan Comment: Roger G. Noll

213

7. Emerging Challenges for Indian Education Policy 303 Anjini Kochar 8. Does Economic Growth Increase the Demand for Schools? Evidence from Rural India, 1960–99 Andrew D. Foster and Mark R. Rosenzweig 9. Priorities for Further Reforms Anne O. Krueger Conference Participants Contributors Author Index Subject Index

329 355 363 367 369 373

Foreword George P. Shultz

India has been very much in my consciousness for at least half a century. I remember that during my days as a graduate student at Massachusetts Institute of Technology just after World War II many of my fellow students were from India, perhaps more than from any other country outside the United States. MIT had a joint program having to do with economic development in India, and India was a favored country. Mahatma Gandhi was a hero and had a great following among students in the United States. I was reminded of Mahatma Gandhi’s use of silence right after I was nominated to be secretary of state. I knew that, between nomination and confirmation, the name of the game was to keep your public mouth shut, so I invoked Mahatma Gandhi’s use of silence as a way of avoiding unwise exposure in the press. I remember the first major visitor to Washington after I became secretary of state: Indira Gandhi. She came determined to put our relationship on a stronger and more constructive footing. I particularly remember her composure in the face of rude questions from rude reporters. She said, “I did not come all the way to Washington to criticize the United States. I have come here to develop a more constructive relationship between us.” She then walked serenely away from the gaggle of reporters. I was impressed, among other things, with her bearing, her dignity. Two years later, I led the delegation of former U.S. ambassadors to her funeral in India. What a distinguished group: Senators Pat Moynihan and John Cooper; Bob Goheen, the president of my university, Princeton; John Kenneth Galbraith; and the Senate majority leader, Howard Baker. The distinction of the delegation showed a tradition of respect. George P. Shultz is the Thomas W. and Susan B. Ford Distinguished Fellow at the Hoover Institution, Stanford University.

vii

viii

George P. Shultz

And now I live at Stanford, so to speak in Silicon Valley, where so many talented Indians are working in its booming economy. They personify the new opportunities in both countries that can be based on mutual respect and progress. But the post–World War II history of India was a frustration. In a country with the second-largest population in the world, and the largest democracy in the world as measured by numbers of people, the economy was becoming less and less important to the world. In the decades after India’s independence, India’s rate of growth in real per capita income remained stubbornly low, at around 1.5 to 2.0 percent per annum, while its share of world trade fell from 1.8 percent in the early 1950s (a share larger than Japan’s) to 0.6 percent by the late 1970s (a share smaller than Algeria’s). Although progress was made in reducing poverty, 330 million people, more than half the population, were still living in desperate poverty in 1974. By 1999, the record was better but, even so, estimates are that about 270 million people, over one-fourth of the population, live in poverty. Thus, the pressing need to bring better living standards, increased literacy, and improved health to this enormous population makes this economic performance very disappointing. Moreover, the contrast between Indian economic performance and that of many other developing countries is striking. In South Korea, a country almost equally poor in 1960, real per capita income was doubling every seven years. What could explain the difference in economic performance? Certainly the people of India have outstanding capability. The answer has to lie in economic policies. South Korea and many other rapidly growing countries began liberalizing trade as early as 1960 and developed an export-oriented growth strategy with related reforms designed to improve the performance of their domestic economies. By contrast, Indian economic policy, mired in a time warp of British socialist teaching, was predicated on protecting the domestic economy from international forces, relying on “import substitution” and government controls over economic activity. Even with all these contrasts apparent, Indian economic growth remained slow and its economic policies unchanged, until something important happened nearby. China began growing rapidly, having dramatically changed its economic policies. The contrast was clear, and it became evident that even the Indian growth rate attained in the decade of the 1980s was unsustainable. When crisis struck in the early 1990s, the Indian government began to reform its economic policies. Protection was reduced; quantitative restrictions on trade were dismantled; the financial sector was partially liberalized. Private economic activity was permitted in a number of areas previously reserved to the public sector, and the extent of controls was reduced. To students of economic policy, the resulting acceleration of economic

Foreword

ix

growth in India during the 1990s, reaching an average level of almost 7 percent per annum in the latter part of the decade, was not surprising, but it was surely gratifying. All friends of India, students of India, and potential investors in India are watching the current pace of reforms with cautious hope. What should we expect in the future? The essays in this volume, developed for a conference at Stanford’s Center for Research on Economic Development and Policy Reform, try to answer some of these questions. Policy makers, businessmen, and academics participated in three days of intense discussion in May 2000. All the conference participants agreed that, although much progress had been made in improving India’s economic policy framework, a great deal remains to be done. Thus the essays provide an overview of the reforms, an assessment of the impact of those reforms on education and business, and an evaluation of critical problems in some key areas, particularly fiscal policy and policy toward small-scale industry and regional disparities. Millions of Indians who live in poverty have a tremendous stake in the improvement of India’s economic performance. So does the immense middle class in India, let alone the entrepreneurs and investors. The entire world economy will benefit from an open, successful Indian economy. India’s traditional reverence for education and its readiness to encourage and invest in excellence are increasingly recognized and valued throughout the world. I remember all those students I knew in my earlier days at MIT— how good they were but also how hard they worked to figure out how to stay in the United States rather than go back home. We benefit in the United States from all this brainpower that comes our way, but we will benefit even more when conditions in India cause the flow to move in the other direction, as is increasingly the case. If the essays in this volume can better inform the international community of scholars and policy makers of the progress India has made, as well as the problems still to be confronted, they will have made an immense contribution. Even more, if they can help improve the basis on which future policy reforms are made and strengthen economic performance, then India and all its partners and friends around the world will benefit. Then cautious hope can give way to confident optimism.

Acknowledgments

This volume owes much to the efforts of several key people. Thanks are due to those who enhanced the intellectual content of the conference, those who made all the arrangements, and those whose financial support made the entire venture possible. Taking intellectual content first, the greatest debt is to the conference participants. Sessions were lively, and many complex issues were greatly clarified by discussion. Special thanks are due to the Honorable Montek Ahluwalia, whose early ideas helped shape the conference format and topics, and to the Honorable Yashwant Sinha, whose participation, prepared remarks, and thoughtful comments were an invaluable guide to policy priorities. Kanwal Rekhi’s sage advice improved both the organization and participation in ways that added greatly to the outcome. Sajjid Chinoy made valuable contributions both as an author and conference participant, and in assisting in the editing of this volume. In its final stages, Irena Asmundson also ably contributed to preparation of the volume for press. Logistical support was more difficult than is the case for most conferences, given its numerous overseas participants. Nicholas Hope, deputy director of the Center for Research on Economic Development and Policy Reform, cheerfully supervised the effort. Deborah Carvalho and Dafna Baldwin were indispensable in making conference arrangements. Rina Rosenberg helped coordinate conference invitations and correspondence. During the conference, Chonira Aturupane and Mu Yang helped Sajjid Chinoy ensure that papers and other conference materials reached participants, and that participants’ several needs were met. Last, but by no means least, the conference and this volume are the product of generous financial support. My colleagues and I are deeply grateful

xi

xii

Acknowledgments

to Kanwal Rekhi, K. B. Chandrasekhar, Ed Shea, and Suhas Patil for supplying the great majority of financing for the conference. We are grateful too for supplementary funding from the President’s Fund of Stanford University, and from the own resource of the Center for Research on Economic Development and Policy Reform.

Abbreviations

ARIS ASI BJP CBDT C-DOT CII CMIE CSO DOE DOT FCI FERA FICCI FIPB FPC GDP GOI IDA IIR IMF MOU MRTP MTNL NASSCOM NPA NSS NTP

Additional Rural Incomes Survey Annual Survey of Industries Bharatiya Janata Party Central Board of Direct Taxes Centre for Development of Telematics Confederation of Indian Industry Centre for Monitoring Indian Economy Central Statistical Organization Department of Electronics Department of Telecommunications Food Corporation of India Foreign Exchange Regulation Act Federation of Indian Chambers of Commerce and Industry Foreign Investment Promotion Board Fifth Pay Commission gross domestic product Government of India International Development Association Indian Infrastructure Report International Monetary Fund Memorandum of Understanding Monopoly and Restrictive Trade Practices Mahanagar Telephone Nigam Ltd. National Association of Software and Service Companies non-performing asset National Sample Survey New Telecom Policy

xiii

xiv

Abbreviations

NTPC OECD PDS PSE RBI REDS SDP SEB SEBI SIDC SIDO SSI STPI TRAI VAT VSNL WTO

National Thermal Power Cooperation Organization for Economic Cooperation and Development public distribution system public sector enterprise Reserve Bank of India Rural Economic and Demography Survey state domestic product State Electricity Board Securities and Exchange Board of India State Industrial Development Corporation Small Industries Development Organization small-scale industry Software Technology Parks of India Telecom Regulatory Authority of India value added tax Videsh Sanchar Nigam Ltd. World Trade Organization

Chronology of Major Political and Economic Events

1947 1948

1950

1951 1956 1957

1961 1962 1964 1965

India becomes independent. Jawaharlal Nehru is chosen prime minister. Mahatma Gandhi is assassinated. The government issues the First Industrial Policy Resolution, reserving certain industries for the public sector. The Reserve Bank of India is nationalized and becomes the central bank. The Republic of India is declared with the promulgation of the constitution. Dr. Rajendra Prasad is elected the first president. The Indian Planning Commission is constituted with Prime Minister Nehru as its chairman. A draft of the First Five-Year Plan is published. It is primarily a public expenditure plan. The Second Five-Year Plan is presented to parliament and approved. It shifts emphasis to government-led industrialization. Stringent import and foreign exchange controls are imposed in response to the growing fiscal and balance-of-payments deficits arising from the implementation of the Second Five-Year Plan. The Congress party retains power in the second general election. The Third Five-Year Plan continues the investment patterns of the second plan, focusing on government-led industrialization. The Congress party retains power in the third general election. India and China engage in a month-long border war. Prime Minister Jawaharlal Nehru dies. Lal Bahadur Shastri is chosen as the next prime minister. India and Pakistan go to war over Kashmir. A cease-fire is declared in the same year.

xv

xvi

1966

1967 1969

1970 1971

1972 1973 1975 1977

1979

1980

1984 1985

Chronology of Major Political and Economic Events

Lal Bahadur Shastri dies in Tashkent. Nehru’s daughter, Indira Gandhi, becomes prime minister. Poor harvests trigger a food shortage and a balance-of-payments crisis. India devalues the rupee by 36 percent and receives food and monetary aid under an IMF-World Bank program. Indira Gandhi remains prime minister after the Congress party retains power in the fourth general election. Parliament passes the Bank Nationalization Bill nationalizing all domestically owned commercial banks. The Fourth Five-Year Plan, placing greater priority on the agricultural sector than earlier plans, is adopted. The Monopoly and Restrictive Trade Practices Act, regulating the activities of business houses, comes into effect. Indira Gandhi continues as prime minister after the Congress party wins the fifth general election. Pakistan declares war on India but surrenders in the same month, and the state of Bangladesh is formally recognized. The government nationalizes all insurance companies. The Foreign Exchange Regulation Act, controlling foreign investment in India, comes into effect. Indira Gandhi declares a state of emergency, arresting opposition leaders and rescinding some civil liberties. The state of emergency is lifted, and parliament is dissolved. The Janata Party coalition wins the sixth general election, and Moraji Desai is appointed prime minister. The targets laid out in the Fifth Five-Year Plan are abrogated by the new government, which resorts to an annual planning mechanism. Moraji Desai resigns. Charan Singh leads a new coalition as prime minister but soon loses support, and parliament is dissolved. India suffers her worst drought since independence. The Congress party returns to power in the seventh general election, and Indira Gandhi is appointed prime minister. Facing a significant balance-of-payments problem, India negotiates a loan from the IMF under its Extended Fund Facility. The Sixth Five-Year Plan sets out a target annual growth rate of about 5 percent, which is achieved over the next five years. Indira Gandhi is assassinated, and her son, Rajiv Gandhi, is chosen prime minister after the Congress party wins the ensuing elections. The Rajiv Gandhi government initiates modest liberalization measures as regards industrial licensing and import and export regulation. The Seventh Five-Year Plan sets out a target annual growth rate of about 5 percent, which is achieved over the next five years at the cost of increasing fiscal imbalance.

Chronology of Major Political and Economic Events

1989 1990

1991

1992 1993 1996

1997

1998

1999

2000

xvii

A coalition of non-Congress parties comes to power after winning the ninth general election. V. P. Singh is appointed prime minister. V. P. Singh resigns as prime minister after the Bharatiya Janata party (BJP) withdraws support. Chandra Shekhar leads a minority government supported from the outside by the Congress party. The minority government loses support, and new elections are ordered. Rajiv Gandhi is assassinated during the election campaign. The Congress party wins the tenth general election. The new prime minister, P. V. Narasimha Rao, appoints Manmohan Singh as finance minister. India faces a severe macroeconomic and balance-of-payments crisis, and the government responds with stabilization measures and structural reforms (see chapter 1). The Eighth Five-Year Plan, placing greater emphasis on private initiative in industrial development than any previous plan, is adopted. Financial sector reforms, based on the recommendations of the Narasimham Committee, are initiated. After the eleventh general election, the National Front coalition forms the government supported by the Congress party. H. D. Deve Gowda is chosen prime minister. H. D. Deve Gowda resigns as prime minister and is replaced by I. K. Gujral. I. K. Gujral resigns, and fresh elections are ordered after the government falls. The Ninth Five-Year Plan, placing priority on agricultural and rural development, is adopted. The BJP secures a plurality in the twelfth general election, and heads a coalition government with A. B. Vajpayee as prime minister. The BJP-led coalition government falls, and in the ensuing elections, the National Democratic Alliance—headed by the BJP— wins a majority in parliament. A. B. Vajpayee is chosen prime minister. Liberalization measures are initiated in the areas of insurance, consumer good imports, and domestic telephony.

Introduction Anne O. Krueger

It is well known that India is the second most populous country in the world, and has a low per capita income. Moreover, in contrast to some other countries that were very poor in the immediate aftermath of the Second World War, the rate of growth of per capita income in India was low, averaging under 2 percent annually from the late 1940s until 1980. Indeed, total growth of real gross national product (GNP) was consistently around 3.8 percent annually—which came to be dubbed the “Hindu rate of economic growth”! Some politicians, policy makers, and journalists believed, or seemed to believe, that this growth rate was inevitable. This low growth rate persisted into the 1970s despite the Indian government’s continuing rhetoric that a central goal of policy was to achieve improved living standards for the majority of the people. To that end, a succession of Five Year Plans set forth a large number of economic policies controlling and directing private economic activity and guiding publicsector enterprises that absorbed a high and increasing fraction of investment. During that period, Indian monetary and fiscal policies remained very conservative by standards of other developing countries. By the 1980s, however, fiscal and monetary policy became significantly more expansionary. While that policy stance was unsustainable in the long run, the real rate of economic growth accelerated to average around 5.8 percent annually during the 1980s, leaving the “Hindu rate of economic growth” well behind. In 1991, however, the inevitable consequences of expansionary policy were felt in the form of a balance-of-payments crisis, as Anne O. Krueger is currently the first deputy managing director at the International Monetary Fund. At the time of writing this chapter, she was the Herald L. and Caroline L. Ritch Professor of Economics, and director of the Center for Research on Economic Development and Policy Reform at Stanford University.

1

2

Anne O. Krueger

foreign debt had accumulated during the 1980s and inflation rose to a peak of 17 percent, much above the politically acceptable rate in unindexed India. In earlier periods, the government of India had met crises by borrowing short-term from the International Monetary Fund (IMF) and undertaking a series of measures that addressed the short-term issues, but left the longterm underlying economic policies relatively unchanged—especially those policies regarding governmental regulation and control of economic activity and insulation of the economy from foreign producers and investors with high walls of protection. In 1991, however, the immediate responses to the emergency were accompanied by announcements and actions signifying that underlying economic policies would be altered. The following several years witnessed a variety of significant changes, many of which are discussed later in this volume. Foreign trade and investment barriers were substantially reduced, and some were eliminated. Financial markets were significantly liberalized. Some controls on private economic activity were removed, and some others relaxed. Economic growth accelerated to almost 7 percent per year. But by the mid-1990s, India met a period of frequent changes of government, as no single party could command a majority in Parliament and coalitions were short-lived. Because of that, and for other reasons, the momentum for reform slowed significantly. Indeed, what had begun as “reform by stealth” became almost imperceptible reform, although some further measures were taken. The rate of economic growth slowed (but was still above its levels of the 1980s), and analysts questioned whether the halfway house, in which some regulations had been removed, or at least been mitigated in their impact, while incentives and a framework for private activity were not fully developed, could sustain the 5-6 percent rate of growth achieved at the end of the decade. It seemed appropriate, therefore, to take stock of and analyze Indian economic policy reform early in the year 2000. To that end, academic specialists on India and on policy reform, Indian administrators, and present and past policy makers were invited to explore a number of topics of particular concern at a conference held at the Center for Research on Economic Development and Policy Reform at Stanford University in May 2000. The papers presented at the conference were revised on the basis of (usually very lively) discussion that followed each paper. They constitute the core of this volume, together with an initial chapter that places the Indian reforms in the context of the history of Indian economic policy and the Indian economy. Some of the discussants’ comments, which shed additional light on the subject at hand, are also included in the volume. In one conference, it was too much to hope that all critical reform topics could be discussed. Instead, it was decided to focus in more depth on some key issues: the fiscal situation, the environment for private economic activity, education, the reservation of certain activities for small-scale industry,

Introduction

3

and determinants of differentials in rates of growth across the different Indian states. As the center’s mission is to bring high-quality academic research to bear on pressing policy issues, the papers ranged from analyses of policies in urgent need of change (such as fiscal adjustment and small-scale industry reservation) to research “at the frontier” in educational policy reform that can build a basis for sounder measures in the future as understanding of the underlying behavioral patterns increases. To provide perspective and place these topics in the context of other areas where reform is clearly desirable, in the first chapter Anne Krueger and Sajjid Chinoy present a brief overview of Indian economic policies and development over the period since independence. They pay particular attention to policy reforms deemed critical for Indian economic growth that are not the focus of later chapters. These include especially the prospects for infrastructure (notably telecommunications, transport, and power, where evidence of striking deficiencies and of their detrimental effect on growth and living standards is overwhelming), reforms in regulations governing labor markets, and the need for attention to public-sector enterprises and their transformation into more productive entities. In chapter 2, T. N. Srinivasan describes the current state of Indian fiscal affairs: After the crisis of 1991, there was a period during which the fiscal deficit was somewhat reduced, but it has since increased, and prospects for its reduction without further policy measures are quite bleak. Srinivasan first shows the causes for concern about the size of the prospective deficits and then assesses the current structure of expenditures and revenues in order to analyze the policy alternatives confronting the government if a future fiscal crisis is to be avoided. In his comments, Shankar Acharya reinforces Srinivasan’s concerns with respect to the influence of state fiscal balances on the center’s overall fiscal position. He notes that the recent increase in civil servant salaries earlier in 2000 (without the commensurate reduction in force advocated by the pay commission) had several ill effects: Not only did it worsen the fiscal situation of both center and states; it also introduced further distortions into the labor market. Kenneth Kletzer, in his comments, focuses on elaborating the concerns that arise from the Indian fiscal deficit. He notes that the banking sector is required to hold a considerable portion of its assets in government liabilities, and that this requirement hobbles the financial sector and results in financial repression. That, in turn, prevents international financial integration. He also comments on Srinivasan’s concern that state governments’ fiscal behavior is a major factor contributing to the Indian fiscal problem, and supports the proposition that the states must increase their own fiscal performance along a number of lines. In reflecting upon the current status of the reform agenda, N. K. Singh comments on the lack of political consensus over reforms as a key factor in their “incomplete” nature. He notes the widespread belief that the reforms

4

Anne O. Krueger

have benefited the elite, primarily the corporate sector. He then focuses on the proposition that there have been some successful reforms with respect to the tax structure and administration, but that effective controls on government expenditures are still in need of significant improvement. Like Acharya, Singh believes there has been significant reform of the tax structure during the 1990s, although he agrees that there is considerably more reform required. He provides important data on the shift in tax structure from reliance on indirect taxes to a higher proportion of direct taxes for the economy as a whole, although his data also reflect the relatively low Indian ratio of tax to gross domestic product. Singh examines the determinants of this performance in some depth. The third chapter, by Montek Ahluwalia, presents new evidence on differentials in growth rates among the Indian states, and undertakes a preliminary analysis of the reasons for these differentials. Differences between states are indeed large; it is interesting that these have not been the subject of considerable economic analysis in prior years. But Ahluwalia provides an initial assessment of some of the plausible reasons for these differences among states. Ahluwalia presents data on differences among the fourteen large states in their growth rates before and after reforms. He finds that the dispersion in state growth rates has increased, but that states whose growth accelerated were relatively evenly spread throughout the country, both geographically and in terms of their initial income levels. Both northern and southern, coastal and interior, and some poor and some rich states improved upon their earlier performance relative to other states in the same categories. He then provides a preliminary examination of the various factors that have contributed to differences in growth rates. This is one of many areas in which it is clear that policy makers could benefit from a great deal of additional economic analysis. In his comments on Ahluwalia’s chapter, Shankar Acharya seeks to understand further the determinants of state growth rates, focusing as far as the data permit on the breakdown between agricultural and nonagricultural growth rates. However, he recognizes, as does Ahluwalia, that a great deal remains to be learned about the determinants of differentials in state economic performance. The next two chapters focus on aspects of government policy toward private economic activity in India. In a first chapter, Naushad Forbes assesses the changes in the environment in which private firms operated between the period just prior to reforms and the late 1990s. Himself a businessman, Forbes documents the extent to which the number of permits, regulations, and steps through which businesses have to go has diminished, and regards the changes in policy as a distinct improvement over the earlier situation. He notes, however, that delays in getting permissions and clearances can be considerably longer than in other countries, often constituting a deterrent to exporters when contrasted with their foreign competitors.

Introduction

5

In the following chapter, Annalee Saxenian considers the growth of the Indian software industry, and provides some contrasts between Bangalore—the center of software in India—and Silicon Valley. The software industry in India has been highly visible as a rapidly growing, modern, industry in the midst of much more slowly growing traditional industries. As is well known, many CEOs in Silicon Valley, and a much larger number of high-tech professionals, are of Indian origin, and it is natural to inquire as to the role of the industry for India’s future economic growth. Saxenian documents the rapid growth of the industry and its prospects for future expansion both in the domestic market and for export. She also notes, however, that there are many Indians without access to the necessary education and training to be able to become productive in these industries, and worries about the development of a society bifurcated between those who have the skills and those who do not. Comments by N. R. Narayana Murthy, from his perspective as the leader of one of the most successful Indian software firms, and by Ashok Desai, who participated as a policy maker in the initial period of reforms and is now a prominent journalist and economist, shed additional light on the circumstances confronting private producers in India. While conditions for private economic activity have improved markedly, there are a number of issues they believe must be addressed in order to support future sustained rapid growth. Increasing educational attainments, improving the functioning of the legal system, and attention to infrastructure are among the key issues that they highlight. They also note that the software industry’s performance has not been replicated, and that this may, in the longer run, constitute a negative factor even for the growth of the software industry itself, as well as limiting India’s overall rate of economic growth. The government of India has, since independence, sought to achieve increases in living standards, especially for the poor. For that purpose, several policies are particularly important. As in most countries, educational policy is crucial. But in India, in addition, policies have been maintained, even past the period of reforms, to “reserve” designated industries for “smallscale” firms. As Rakesh Mohan explains in his chapter, these policies were designed to encourage new entrants and to protect small-scale firms from what was believed would be unfair competition from established firms. But, as Mohan demonstrates, the policy has, from the beginning, been counterproductive. Small-scale firms are discouraged from expanding. The reserved industries are often labor-intensive and might have been key exporters had incentives been more aligned with economic efficiency; instead, employment growth—especially among unskilled workers—has been extremely sluggish. From Mohan’s chapter, and the discussion at the conference, it would appear that repeal of the small-scale industries reservation policies would improve the efficiency and equity of the Indian economy considerably. After the conference, the garment industry was removed from

6

Anne O. Krueger

the industries reserved for small-scale enterprises by parliament. It will be instructive to learn the effects of that change on the industry’s performance. If conditions of the majority are to be significantly improved, it is evident that increased educational opportunities and attainment for young people will be crucial. Evidence that increased educational attainments are a necessary condition for higher wage incomes for much of the labor force is by now widely accepted. In these areas, the basic facts are known, and it is widely recognized that the quality and quantity of Indian educational opportunities, especially at the primary and secondary level, must be improved. However, there are significant questions as to what are the most effective measures to improve the educational attainments of young people in India. The last two chapters in the volume, one by Anjini Kochar and the other by Andrew D. Foster and Mark R. Rosenzweig, present research results that shed some light on these issues. Kochar examines the supply-side factors that influence the decision to extend children’s school attendance to longer periods. She uses available variables to indicate the quality of Indian schooling, and finds that, especially for the more disadvantaged parts of the population, better quality of schooling induces students to remain in schools longer. Economists and educators have long since sought better-quality schooling as a mechanism for improving the effectiveness of former students later in life, but Kochar’s analysis suggests that, in addition, better quality of schools can increase the quantity of schooling that children receive. The benefits for increasing quality, based on this analysis, arise not only for those already in schools, but for those who will stay in school longer when quality increases. Foster and Rosenzweig examine another avenue affecting the demand for schooling. Using data from 240 Indian villages, they analyze the degree to which low rates of economic growth reduce the demand for additional years of schooling at the secondary level. This can happen, according to their analysis, because of the consequent low return to investment in schooling because of lack of opportunity (due to low growth) to use the skills that come with education. Their data strongly indicate that a more rapid rate of economic growth shifts the demand curve outward for additional years of schooling. Hence, when economic policies are not conducive to rapid economic growth, one consequence is that the educational attainments of the young diminish. Consequently, one cannot regard the low educational attainments of the labor force in a slowly growing country as being independence of that slow growth. The Foster-Rosenzweig evidence indicates that low levels of educational attainment are, at least partially, a result of slow growth. A final chapter provides an overview of the current state of economic policy reform, discusses the key areas where further reform is urgently needed if India is to sustain a 6–7 percent annual rate of economic growth, and suggests some topics for further research.

1 The Indian Economy in Global Context Anne O. Krueger and Sajjid Chinoy

Since Indian independence, a great deal of economic progress has taken place. By any measures—life expectancies, infant mortality rates, nutritional standards, literacy and educational attainments, or real per capita incomes—the Indian people are considerably better off than they were a halfcentury ago. Yet there is a certain impatience and uneasiness: For all of the many achievements, there is a strong sense that more rapid improvements in these and other indicators of well-being are quite possible. Moreover, if they are possible through realignment of existing policies, per capita incomes for many Indians are still so low that more rapid growth is manifestly desirable. It is the purpose of this chapter to provide an overview of the strengths and weaknesses of Indian economic policy and performance to date, in order to set the stage for an analysis of policies and policy changes that can enable even better performance in the future. Such an assessment necessarily entails an examination of Indian economic policies and performance in comparative perspective. In-depth analysis of particular policies then follows in later chapters. The towering international economist of the 1960s and 1970s, Harry G. Johnson, once heard an economist start a comment with the phrase “In a country like India. . . .” He did not let him finish his thought, but interrupted to ask, “What other country is there like India?” He was, of course, Anne O. Krueger is currently the first deputy managing director at the International Monetary Fund. At the time of writing this chapter, she was the Herald L. and Caroline L. Ritch Professor of Economics, and director of the Center for Research on Economic Development and Policy Reform at Stanford University. At the time of writing this chapter, Sajjid Chinoy was a doctoral student in the department of economics at Stanford University.

9

10

Anne O. Krueger and Sajjid Chinoy

right, but while there is no other country like India, it is nonetheless the case that inferences relevant to India can be drawn from experiences, achievements, and effects of reforms in other countries. Thus, with due recognition that growth rates among Indian states and regions differ significantly (see Ahluwalia, chapter 3 in this volume), and that comparisons of growth rates and other indicators economy-wide may be misleading given India’s size and diversity, we nonetheless proceed with an assessment of performance to date, of bottlenecks to further improvements in performance, and of the potential for improved performance were these bottlenecks addressed. Section 1.1 provides a brief survey of the evolution and structure of the economy prior to the 1990s. Section 1.2 then analyzes the reforms of the 1990s and economic performance during the decade. Section 1.3 traces the response of the economy in the 1990s to the reforms. Section 1.4 then analyzes some of the key areas not discussed in depth in later chapters in this volume that still offer great potential for accelerating growth if reforms are carried out. Section 1.5 concludes. 1.1 The Evolution and Structure of the Indian Economy There are many good economic histories of India’s economic evolution since independence, and that economic history is well known (see Ahluwalia 1985, Chakravarty 1987, Joshi and Little 1994, and Srinivasan 2000). Nonetheless, any effort to assess priorities to improve economic performance must be made in light of an understanding of the evolution of policy and the economy at least since independence. Here, we do that with a very broad brush, and the interested reader may consult any of the references above. The natural starting point is that Indian per capita incomes were very low in the late 1940s and early 1950s, and there had been little improvement in living standards over the previous century. Real income per capita is estimated to have grown at an average annual rate of only 0.7 percent between 1870 and 1913, and of only 0.2 percent over 1913 to 1950 (Morawetz 1977, 14). Table 1.1 gives estimates of per capita incomes of selected Asian countries for 1950 and 1974. Indian per capita income was estimated at US$95 (in 1974 prices) in 1950. While some countries probably had lower per capita incomes (such as several African countries and Burma—now Myanmar—in Asia), India was poor even by standards of developing countries. However, differentials in estimated per capita incomes were generally not high—the Philippines was estimated to have a per capita income about 70 percent higher than India, while Malaysian per capita income was thought to be almost four times as high. Certainly, achieving higher living standards for the Indian people was seen to be a major goal after independence. A great deal of thought and dis-

The Indian Economy in Global Context Table 1.1

11

Comparative Per Capita Incomes and Growth Rates, Selected Asian Countries, First Quarter-Century of Development (1974 constant prices)

Country Burma (Myanmar) China, Peoples’ Republic India Indonesia Malaysia The Philippines South Korea Taiwan Thailand

Per Capita Income 1950

Growth Rate, 1950–74

Per Capita Income 1974

57 113 95 103 350 168 146 224 132

2.3 4.2 1.5 2.0 2.6 2.8 5.1 5.3 3.6

100 320 138 169 665 340 504 817 319

Source: Morawetz (1977), table 5, taken from World Bank (1977).

cussion in planning for independence focused on the need for rapid economic growth and rising living standards. Nehru and Gandhi had, indeed, differed on what economic policy should be, but the two leaders agreed on the centrality of economic developmental goals as a top priority after independence.1 At independence, India was a predominantly agricultural economy, with more than 70 percent of the population deriving its livelihood from agriculture, and just under 50 percent of gross domestic product (GDP) originating in agriculture.2 The Nehruvian view—derived predominantly from Fabian Socialism—endorsed the need for rapid development led by state economic activity and planning. The first few years after independence were naturally focused predominantly on establishing institutions. Even then, a First Five Year Plan was formulated, although it did little more than bring together various projects that were already under way or in the advanced planning stage. Interestingly, the early planning documents regarded the chief barrier to accelerated growth as the then-low Indian savings rate, and set out a twenty-five-year perspective. The Planning Commission documents stated that a major challenge was to raise the Indian savings rate to 20 percent and concluded that, if that could be attained, Indian economic growth could achieve a satisfactory rate of 5 percent annually.3 1. See for example, Nehru’s (1958, 402) address to Congress in 1936: “I believe in the rapid industrialization of the country; and only thus, I think, will the standards of the people rise substantially and poverty be combated.” 2. Data are from Organization for Economic Cooperation and Development (1967) and Central Statistical Organization (1967), and refer to 1950–51. 3. See the discussion by Chakravarty (1987). Sir Arthur Lewis had first articulated the view that increasing the rate of savings, and thus presumably of capital formation, was the central challenge for raising the rate of growth from earlier very low levels to satisfactory levels in all developing countries.

12

Anne O. Krueger and Sajjid Chinoy

The Second Five Year plan (1957–1962) articulated a philosophy of state responsibility for development, and laid down much of the strategy that was to be followed in Indian economic policy until the 1990s. The goal of increasing savings remained, although the plan itself provided for more expenditures than could be financed by foreseeable revenue sources. The Second Plan recognized the importance of education, infrastructure, and so on, but it focused on a strategy for rapid industrialization through development of heavy industry, especially for production of capital goods. By that time, it had been decided that certain industries (the “commanding heights” of the economy) were to be reserved to the state, some industries were to be jointly developed by the state and private industry, and some activities were to be reserved to private industry.4 Within the plan, targets were established for production levels. These were given to state enterprises. For private sector production, targets were to be implemented through investment licensing, under which the authorities would not grant licenses for more additional capacity than was sanctioned under the plan.5 Indeed, investment licenses specified the permitted output of factories, and owners in the private sector were not permitted to produce more than their licensed amount without amending the license.6 The underlying philosophy of the second and subsequent plans was that Indian development would have industrialization as the major “engine of growth” and that import-substitution policies would be used to stimulate growth, especially in the manufacturing industries. Influenced heavily by P. C. Mahalanobis, whose economic model ignored foreign trade and assumed that domestic investment was limited by the domestic ability to produce capital goods, the general philosophy was to push heavily to “make machines to make machines,” while simultaneously encouraging cottage industry (as had been advocated by Gandhi) to provide employment. Both because the policy makers (and Indian public) were suspicious of foreign trade—after years of colonial rule and in the aftermath of The Great Depression—and because export growth lagged while import demand accelerated in response to the expenditure patterns generated by the plans with consequent “foreign exchange shortage,” domestic production of importcompeting goods was encouraged by imposing a regime of import licensing 4. See Bhagwati and Desai (1970), especially chapter 6. 5. Officials issuing investment licenses therefore kept track of the amount of additional capacity that was to be built under the plan. Once licenses had been issued in that amount, no more were given. This led some to obtain licenses in order to preclude their competitors’ increases in capacity, although they had no intention of using the entire permitted capacity increase themselves. 6. In later years, these restrictions were somewhat relaxed. At first, production of up to 125 percent of the licensed amount was permitted. Thereafter, further liberalization occurred. For example, starting in 1975, fifteen engineering industries were granted automatic approval for increases in licensed capacity at the rate of 5 percent per year, over and above the 125-percent rule. In the 1980s, the 125-percent rule was amended to 133 percent of best past production. See Ahluwalia (1985) and Government of India, Economic Survey, various issues.

The Indian Economy in Global Context

13

and preventing imports of goods when domestic producers were deemed able to meet domestic demand. Although India’s monetary and fiscal policies were relatively conservative contrasted with those in a number of developing countries, the upward shift in demand for imports arising from the plan pattern of expenditures and a rate of inflation above that in the rest of the world combined to lead to increasing rupee overvaluation. That overvaluation, plus the pull of resources into import-substitution induced by the high levels of protection (both from tariffs and from the unavailability of import licenses), discouraged export production and exporting. The “export pessimism” of the early years, which was one rationale for adopting import-substitution policies, became a self-fulfilling prophecy. Consequently, Indian export growth was sluggish during the period when the world economy was expanding rapidly in the 1950s and 1960s, and India’s share of world markets fell. Table 1.2 gives some pertinent data. As can be seen, exports grew at an average annual rate of less than 1 percent in the 1950s, and at 4.60 percent in the 1960s (when the average annual rate of growth of world exports was well above 10 percent). Even in the 1970s and 1980s, Indian export growth rates were below world rates, and India’s share fell continuously until the end of the 1980s: India’s average share of world trade for that decade averaged only one half of one percent. Only in the 1990s—after policy reforms to be discussed later—did India’s share of world markets begin to recover, as exports grew more rapidly than world trade as a whole. This sluggish growth of exports was accompanied by a falling share of exports in GDP during the 1950s and 1960s. For the decade of the 1960s as a whole, India’s exports averaged only 4.25 percent of GDP. Reflecting the “foreign exchange shortage,” even imports averaged only 5.83 percent of Table 1.2

Indian Export Performance, 1950–97 Average Annual Growth Rate over Period

Period 1951–60 1961–70 1971–80 1981–90 1991–97

Exports

Imports

Exports

Imports

Share of India’s Exports in World Exports (%)

0.7 4.6 6.8 6.1 11.4

8.6 0.3 8.7 3.9 14.4

6.3 4.2 5.8 6.5 9.9

8.0 5.8 6.7 8.4 10.6

1.4 0.9 0.5 0.5 0.6

Percent of GDP

Sources: International Monetary Fund (various issues). Notes: Growth rate of exports and imports was computed using a volume index based on data for export and import values and unit value indices given by the International Monetary Fund in various issues of International Financial Statistics. Exports and imports as a percentage of GDP represent exports of goods and services. Other data are for merchandise trade alone. The percentage constituted by Indian exports of world exports was calculated as an average over the period. In 1997, Indian merchandise exports were 0.63 percent of world exports.

14

Anne O. Krueger and Sajjid Chinoy

GDP, as their growth rate also plummeted (see column 2 of table 1.2) in response to the strict import licensing regime at that time. Indeed, overall economic growth was sharply reduced during the balance-of-payments crisis of 1966–67. The crisis, which was manifest in the foreign exchange position of the government, was greatly intensified by poor harvests. The devaluation of 1966, however, was a failure in several regards: (a) the poor harvest itself resulted in worsened economic conditions; (b) the additional foreign exchange that had been committed by foreign governments and multilateral institutions (to support the devaluation and accompanying policy measures) was not forthcoming; and (c) the devaluation raised the rupee reward for exporting very little, if at all, as export subsidies were simultaneously removed.7 In some instances, the rupee subsidy per dollar had been greater than the increase in the rupees per dollar accompanying devaluation.8 Despite initial liberalization moves accompanying devaluation, including shifts of some commodities to open general licenses, the overall trade regime was probably more restrictive by the early 1970s than it had been in 1964–65. Moreover, other measures (such as the restrictions on large industrial houses with gross assets exceeding rupees (Rs) 20 crores under the Monopoly and Restrictive Trade Practices Act that came into effect in 1970) also increased the degree of detailed regulation by the government of the Indian economy. It should also be noted that an increasing number of economic activities took place in the public sector. State-owned enterprises had been built in industries such as steel, fertilizer, heavy chemicals, machine tools, and so on. Even hotels were owned by the public sector. In some of these cases, stateowned enterprises operated alongside private sector firms,9 whereas in others the state enterprise was a monopoly. Public corporations were established and given monopoly positions in activities as diverse as insurance, importation of bulk consumer goods (canalization) under which only the government entity could import items such as petroleum and export goods such as sugar. Naturally, telecommunications, (telecoms), railroads, and other infrastructure activities were also in public hands. In the late 1960s, the banks were all nationalized. Hence, in addition to the government’s controls over the private sector, publicly owned entities themselves undertook a great deal of economic activity. By the late 1970s and early 1980s, it was obvious to many that the perva7. See Bhagwati and Srinivasan (1975), part III, for a full discussion of the failure of the 1966 devaluation. 8. It was also widely believed that there had been a considerable amount of “false” exports where nonexistent exports were reported in order to receive the export subsidy. When the subsidy was eliminated, false recording stopped and so recorded exports dropped. Whether actual exports dropped is more problematic. 9. In the case of steel, private steel companies were even taxed to help finance public-sector steel production.

The Indian Economy in Global Context

15

sive regulation and controls over private economic activity by the government had had effects opposite to those intended and had inhibited economic efficiency and economic growth. Indeed, when Rajiv Gandhi became prime minister, he declared that his primary objective was to “rationalize” controls. The intent was clearly to reduce the number of overlapping and sometimes even inconsistent regulations. Despite that campaign, which raised the ceiling below which licenses were not needed and simplified procedures in some instances, most analysts did not regard the results as having significantly reduced the regulatory burden. Joshi and Little (1996, 4) concluded that “Rajiv Gandhi had embarked upon some liberalization in 1985. Although he seems to have quickly lost interest, this helped to put such reform on the political agenda.” Meanwhile, by the 1980s, other structural problems were becoming more and more evident and acute. Two in particular deserve mention. First, India’s infrastructure—which was widely recognized to be of inadequate quality and quantity even during the 1960s and 1970s—became increasingly stretched, as economic activity (despite the low rate of growth) led to increased demands on infrastructure at a rate greater than that at which supply was increasing. Second, it became increasingly apparent that public sector enterprises, which had been established in order to accelerate economic growth, were not achieving this goal. Rates of return were low or even negative, and, instead of fostering economic growth, these enterprises became a drain on public resources (and one of the factors therefore accounting for the inability to increase infrastructure availability at the same pace as increases in real GDP). However, during that same period, the government’s fiscal stance became significantly more expansionary. Table 1.3 gives an indication of this beTable 1.3

Macroeconomic Imbalances during the 1980s

Year

Government Expenditures
GDP

Government Revenues
GDP

Fiscal Deficit
GDP

Current Account Deficit
GDP

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990

18.3 17.8 18.6 18.7 20.3 22.3 23.7 22.8 22.2 22.7 22.6

11.8 12.3 12.6 12.3 12.7 13.8 14.4 14.4 14.1 14.8 13.5

6.5 5.6 6.0 6.4 7.6 8.5 9.3 8.4 8.1 7.9 8.1

–1.3 –1.0 –1.2 –2.0 –2.0 –2.0 –2.5 –2.4 –2.3

Sources: International Monetary Fund (1999, India country pages). Notes: All numbers are percentages. Government Expenditures include center-lending to states.

16

Anne O. Krueger and Sajjid Chinoy

havior over the 1980s. As can be seen, the share of government spending in GDP (including “lending”—much of which is not repaid—to the states) rose rapidly during the decade, while revenues grew much more slowly. By the end of the decade, India’s fiscal deficit (financed primarily from domestic borrowing) was averaging well above 8 percent of GDP—an unsustainable number. In addition, the current account deficit had risen significantly and was above 2 percent of GDP in each year in the second half of the 1980s. In the short run, the effect of this expansion was to accelerate the rate of economic growth, as can be seen from table 1.4. The average annual rate of growth of real GDP was 5.4 percent in the first half of the 1980s, and 6.3 percent in the second half. In the short term, that apparently improved economic performance considerably reduced any political pressure or impetus to tackle underlying structural problems, and per capita incomes rose at a rate well above that experienced in India over any previous period. At first, the current account deficit absorbed some of the excess demand created by the fiscal expansion, but inflation nonetheless accelerated, reaching an average annual rate of over 7 percent in the second half of the 1980s, and 13.5 percent by 1991. In addition, the current account balance was deteriorating during the latter part of the 1980s and 1990, and debt was building up. By 1990, imports had to be cut back, as financing was simply not available. The underlying problem was that growth spurred by excess aggregate demand resulting from fiscal deficits was unsustainable. Not only were the current account deficit and the inflation rate rising, but the Iraqi invasion of Kuwait and subsequent events had resulted in a sharp increase in the price of oil, and a drop in workers’ remittances as workers in the Gulf were repatriated. These events were the trigger leading to a crisis in 1991, although it is clear that the existing mix of policies was unsustainable and Table 1.4

Year 1981–85 1986–90 1991 1992 1993 1994 1995 1996 1997 1998

Macroeconomic Indicators, 1980s and 1990s Rate of Growth of Real GDP

Rate of Inflation

Investment
GDP

Current Account Deficit
GDP

5.4 6.3 0.4 5.5 4.8 7.4 7.6 7.0 4.5 6.8

6.8 7.4 13.5 11.9 7.5 10.5 9.3 5.9 5.2 6.9

22.8 23.9 22.7 24.0 21.3 23.5 26.5 21.9 23.4 21.8

–2.2 –1.6 –1.6 –0.7 –0.5 –1.5 –1.5 –0.7 –1.6

Sources: International Monetary Fund, International Financial Statistics Yearbook, 2000. Inflation data are for producer prices. Current account balance data from p. 162; investment data from p. 173.

The Indian Economy in Global Context

17

would have resulted in a crisis at some point, perhaps a little later in the 1990s. Before recounting the measures taken during the crisis, and the ways in which those began the reform process, it is useful to pause and assess the state of the Indian economy as of the early 1990s. Despite the problems just enumerated, the Indian economy of the late 1980s was markedly different from that of the 1950s, and progress had been made on a number of fronts. Per capita income was significantly higher and had grown more rapidly than in pre-independence years, although the rate of growth of per capita income was well below that of many other developing countries, and India remained a very poor country. Life expectancy at birth, which was 32 years based on data from the 1950–51 census, rose to 41.3 in 1960–61, 45.6 a decade later, 50.4 in 1980–91, and was 59.2 by 1990–91 (Dreze and Sen 1995, table A4). Infant mortality rates, which were still 151 per thousand in 1965, had fallen to 70 per thousand by 1994. While still far in excess of comparable rates in industrial countries, which are typically below ten per thousand, this represented considerable progress in delivery of health care. Data on school enrollment rates in India are highly suspect. Official data show enrollment rates in primary school to have risen to 74 percent by 1965, reached an estimated 98 percent by 1983, and become almost universal by the early 1990s, while secondary school enrollments rose from 27 to 34 percent of the eligible age group over the same time period.10 However, estimates based on data from the National Sample Survey and the National Council of Applied Economic Research show a much lower percentage of children in school, especially primary school, so these numbers should be interpreted with caution. Even so, there was a significant improvement in access to schooling, which is reflected in literacy statistics. Whereas the literacy rate among the adult population was estimated to be only 18.3 percent in 1951, it increased to 28.3 percent by 1960, 34.4 percent by 1971, 43.6 percent by 1981, and 52.0 percent by 1991. The latest reported National Sample Survey estimate for 1996–97 put literacy at 62 percent (Ahluwalia 1999, 2). While that still leaves considerable room for improvement, poor educational attainments of the bulk of the population cannot be the drag on growth that they were in earlier decades. Moreover, the combination of longer life expectancy and improved access to schooling itself signifies an increase in well-being for much of the population. There is ample room for improvement in education, however, as is discussed in the chapters by Kochar and by Foster and Rosenzweig in this volume. Another significant change from the 1950s to 1990 was that the rate of population growth had fallen from almost 3 percent per annum to about 2 percent. That drop significantly reduced the challenges associated with pro10. From World Bank, World Development Reports, various years.

18

Anne O. Krueger and Sajjid Chinoy

vision of basic education and health to the growing population and gave rise to hope for further declines in the population growth rate.11 Estimates of poverty rates, while always fraught with difficulties, indicated a considerable reduction in the proportion of those below various poverty lines: For example, the headcount ratio (the proportion of the population living below the poverty line) was estimated to have fallen from 57.33 in 1970 to 37.48 in 1990 in rural areas and from 45.89 in 1970 to 34.76 in 1990 in urban areas.12 Most other poverty indicators showed similar declines.13 Moreover, agricultural production had grown considerably, at an average annual rate of 2.8 percent per year from 1965 to 1980, and agricultural value added at constant prices had increased at an average annual rate of 3.1 percent from 1980 to 1990 (World Bank 1988, 226, and 1998/9, 210). Food production per capita was estimated to have increased by 3 percent over the decade of the 1970s, and by 12 percent over the decade of the 1980s. At the same time, agriculture’s share in national income had fallen from around 50 percent in 1960 to 32 percent by the mid 1980s, as industrial and service output and value added had grown at a more rapid rate (World Bank 1988, 234 and 226, and 1998/9, 204 and 210). Industrial growth, by contrast, was regarded as having been less satisfactory. This was the economic activity on which the Five Year Plans had focused, and into which resources had been poured. Industrial economic activity grew at an average annual rate of 5.4 percent from 1960 to 1970, 4.4 percent during the next decade, and at an average annual rate of 7.1 percent from 1980 to 1990.14 These rates were relatively low both in contrast to rates achieved in other countries and also given the resources devoted to investment in industry. There was also considerable evidence that total factor productivity in Indian manufacturing had been falling over the years. Ahluwalia (1991) estimates that the total factor productivity growth rate (which takes into account increases in both labor and capital inputs) in 11. The estimated rate of population growth between 1965 and 1980 was 2.3 percent, and between 1980 and 1990 it was 2.1 percent; it has subsequently fallen further and is put at 1.8 percent for the 1990–97 period. Estimates are from World Bank, World Development Reports, various years. 12. Data are from Tendulkar (1998), 288–290. These numbers are based on the Planning Commission’s criterion that monthly total per capita expenditures of less than Rs 49 (rural) and Rs 57 (urban) in 1973–74 prices constituted poverty. 13. There is evidence that the various poverty indicators all rose somewhat during the crisis and its immediate aftermath. However, preliminary data from the National Council of Applied Economic Research suggest that the number in poverty dropped sharply once growth accelerated in the mid–1990s. The poverty gap indicator for the rural areas is estimated to have risen from 0.032 in 1990–91 to 0.042 in 1992 before falling back to 0.031 in 1993–94 (the latest year for which data are presented), while the urban poverty gap is estimated to have stood at 0.033 in 1990–91, risen to a peak of 0.036 in 1993, and then fallen to 0.026 in 1994. See Indira Gandhi Institute of Development Research (1997), 64. 14. World Bank 1983, 204, and 1998/9, 210. Data for earlier years refer to industrial production, whereas data for 1980–90 are reported in units of value added.

The Indian Economy in Global Context

19

manufacturing had been only 0.2 percent in the 1959–66 period, fallen to minus 0.3 percent in the period 1966–67 to 1980, and then risen to 3.4 percent annually in the years 1980–91 to 1985–86. In light of India’s abundance of unskilled labor, it is noteworthy that the manufacturing sector as a whole had increased employment at annual rates of less than 4 percent prior to 1980, and in fact shed labor (at a rate of 0.7 percent annually) during the years of positive productivity growth in the first half of the 1980s (Ahluwalia 1991, tables 3.1 and 3.2). As already discussed, the public sector’s economic activities had risen sharply. By the late 1970s, it is estimated that 62.1 percent of total “productive capital” was in state-owned enterprises, as was 26.7 percent of employment. By contrast, industrial value added originating in public sector enterprises was only 29.5 percent of the total (see Bardhan 1984, 102). Data for the 1980s suggest that the imbalance between private-sector and publicsector productivity increased further: The estimated overall real rate of return on investment in state-owned manufacturing enterprises is estimated to have been no more than 2 percent. T. N. Srinivasan summarized the situation well: “Thus, by and large, the public sector has acted as a brake on private sector development. Choice of location, technology, employment and pricing policies of the public sector had become . . . politicized so that efficient development was precluded. Far from generating resources, the public sector had become a monumental waste and liability for taxpayers. It is true that the industrialization strategy did generate a diversified industrial base, and a capability for designing and fabricating industrial plants and machinery. But the strategy virtually ignored considerations of scale economies, vastly restricted domestic and import competition, constrained technological upgrading through licensing and purchase of foreign technologies, encouraged capital-intensive production and discouraged employment generation that was further constraints by the high costs of hiring and firing imposed by our restrictive labour laws. The consequence was a high cost and globally uncompetitive industrial sector which was also out of tune with India’s capital scarcity and labour abundance” (Srinivasan 2000, 6). As of 1991, the public sector still dominated economic activity. Not only were public-sector enterprises overmanned and inefficient, but controls over private-sector economic activity had intensified. Import licenses and investment licenses were still required for all but small-scale businesses in the private sector. Considerable red tape and delay were associated with the receipt of an import license, and there were prohibitions against imports of many goods, including an almost-blanket prohibition of imports of consumer goods and of goods where indigenous productive capacity was deemed available. Rates of protection were extremely high even for those goods whose importation was permitted. India averaged among the highest average rates of protection against imports in the world, even without tak-

20

Anne O. Krueger and Sajjid Chinoy

ing into account the effect of quantitative restrictions and import prohibitions. Some manufacturing activities were reserved for small-scale businesses;15 others were still reserved for the public sector; and some could be undertaken by both private and public entities. Virtually the entire financial sector was publicly owned, including insurance. In most regards India remained heavily controlled and provided a difficult environment in which to do business. In addition, business infrastructure provision was poor. Telephone service was difficult to obtain, at least legally, and was of poor quality. Other communications were little better. Ports were high-cost, with slow turnaround times. Roads were highly congested, resulting in heavy reliance on railway transport for goods. Even domestic air travel was in excess demand and often subject to major uncertainties given the domestic monopoly of Indian Airlines. Government-owned banks rationed credit slowly to established enterprises, while insurance companies, also government owned, were high cost and provided only a small range of coverage. Government expenditures themselves were not allocated in ways conducive to rapid growth. Much of the rapid increase in expenditures in the late 1980s had been the result of changes in the provision of subsidies—for fertilizer, electricity, water, food consumption by the poor, and exports. By 1990, subsidies as a proportion of central government expenditures had reached 11.6 percent of the entire central government expenditure and 2.3 percent of GDP. Total plan “assistance to the states” was 12.1 percent of total central government expenditure, while non-plan “loans to the states” constituted an additional 7.1 percent of central government expenditures and 1.4 percent of GDP. Many of these expenditures, including loans to states, were intended to help the poor (such as fertilizer and electricity subsidies to farmers). But the evidence indicated that a very large fraction of the expenditures ended up in the pockets of the well-to-do among the eligible groups, and simultaneously encouraged waste of scarce water, power, fertilizer, and so forth, while leading at the same time to poor financial results for public-sector enterprises in these areas and thus reducing the capacity for further investments. Thus, as of 1990–91, India had not corrected the underlying structural problems of the relationship between the public and the private sector, and was allocating government resources to activities that in many instances were detrimental to growth. The government of India had maintained growth during the late 1980s only by increasing fiscal deficits. By 1990, debt 15. Small-scale reservation continues to this day. It has a number of significant negative consequences, in part because small-scale industry is normally labor-intensive and could potentially successfully export, but the necessary expansion is not possible. The effects of reservation are not analyzed in detail in this chapter because they are the focus of the chapter by Rakesh Mohan in this volume.

The Indian Economy in Global Context

21

ratios were high, and foreign exchange reserves were falling. In January 1991, the government of India reached agreement with the International Monetary Fund (IMF) for a large loan. Despite that, foreign exchange reserves continued to dwindle, workers’ remittances plummeted (due both to the impact of the Gulf War and to anticipation of an exchange rate change), and the deterioration in the economic situation continued. 1.2 The Crisis of 1991 and the Reforms of the 1990s A new government, with P. V. Narasimha Rao as prime minister and Manmohan Singh as finance minister, came to power in mid-1991. It was apparent that a crisis was at hand: The current account deficit was running at an annual rate of about US$10 billion, while reserves were down to about two weeks of imports. This situation arose in spite an IMF loan of $1.8 billion in January 1991 and sharp cuts in imports starting earlier in the year (imports had been $23.4 billion in 1990 and were $21.1 billion in 1991). Exports, discouraged in part by excess demand in the domestic economy, and in part by the appreciation of the real exchange rate, were falling in dollar terms. Inflation, as already mentioned, had reached 13 percent16 in 1991, and workers’ remittances from abroad, which had been flowing into the country, had dried up. Previous foreign exchange crises had also been triggered by inability to continue voluntary debt service as foreign exchange reserves diminished, and the response had been a traditional stabilization program: The most notable had been those of 1966–67 and 1981. Finance Minister Singh addressed the stabilization measures rapidly. But he went beyond the traditional stabilization package, announcing a program of economic reforms that constituted a significant reversal of some of the most egregious aspects of earlier policies of regulation and government intervention in the economy. The stabilization itself entailed several measures. Underpinning the program was a reduction of the fiscal deficit, initially from 8.1 percent of GDP in 1990–91 to 5.7 percent of GDP in 1992/93.17 It was announced that the government intended to reduce the deficit still further over coming years, but that intention was not realized. A second important measure was a devaluation of the rupee. At the same time, export subsidies were abolished and an import entitlement scheme for exporters was announced. Between 1991 and 1993, a dual exchange rate 16. The political reaction in India to inflation seems stronger than that in many countries. One hallmark of the Indian economy until the 1980s had been the low rate of inflation and the relatively conservative fiscal policy. It is likely that there would have been a strong political imperative to reduce excess demand and bring down the rate of inflation in 1991 even if there had not been balance-of-payments difficulties. 17. The initial level of the primary deficit (not including debt service) was lower but the proportionate reduction was about the same. See Joshi and Little (1996), chapter 2 and Srinivasan (2001, this volume) for particulars.

22

Anne O. Krueger and Sajjid Chinoy

(with a free market rate for exporters) in fact prevailed, but the exchange rate was finally unified in 1993. Joshi and Little (1996, 50) estimate the net real devaluation between 1991 and 1993 to have been about 25 percent. Within a year, the combined impact of the fiscal tightening and the exchange rate change resulted in a reduction in the current account deficit from 2.3 percent of GDP in 1990 (despite import controls) to 0.7 percent of GDP in 1993. Inflation fell, but not as much as had been targeted. Even with recession, the wholesale price index rose 11.9 percent in 1992, 7.5 percent in 1993, 10.5 percent in 1994, and 9.3 percent in 1995. Overall, the stabilization aspects of policy in 1991 bore short-term success. But longer-term issues regarding fiscal policy remained. These are addressed in Srinivasan’s chapter in this volume on the fiscal problem and are not dealt with further here. Suffice it here to say that, as of 2000, a sustainable budgetary situation of the center and the states had not been achieved, and the pattern of expenditures (and especially subsidies) was not conducive to rapid growth.18 What was unusual about the government’s response to the 1991 crisis was that the prime minister and finance minister began addressing the underlying structural issues that had hampered earlier growth. In addition to the traditional (and necessary) stabilization measures, a series of other policy measures were enacted in rapid succession in the first two years after reforms, quickly and significantly reduced the negative effects of controls on domestic economic activity. These “structural” reforms represented a surprising departure from the general direction of policy that had been in place since the 1950s. The most significant structural reforms, reviewed briefly below, were focused on the trade and payments regime, on the one hand, and on the domestic financial sector, on the other. However, there were a number of other important measures that effectively reduced the extent to which the public sector both undertook economic activity and controlled and influenced the profitability of alternative activities in the private sector.19 Before turning to some of the key measures, we should note that the reforms that were implemented were, by and large, implemented slowly. While the 1991–93 measures certainly signaled that change was afoot, change was far less marked than in many other countries’ reform episodes.20 18. One of the many reasons for concern with the fiscal situation is the infrastructure deficit. Although the government of India intends to attract private foreign investment to support infrastructure development, it is clear that much infrastructure expenditure must be financed by the government. Since there is obviously an upper bound on government expenditures, expenditures on subsidies will obviously reduce the resources available for financing the expansion of infrastructure capacity and the improvement of services from it. 19. See the chapter by Forbes in this volume for an analysis of how the private sector was affected by the elimination and/or relaxation of some controls. 20. Policy makers pointed out that the pace of reforms was such that there was no year in which real income actually fell.

The Indian Economy in Global Context

23

Indeed, while momentum for further controls and regulation of the economy was reversed, change was gradual and piecemeal. Some have even referred to the reforms of this early period as “reform by stealth.” For that reason, even where significant reforms were achieved, there remains a great deal to be done—as is discussed in section 1.4. The Indian trade and payments regime as of 1991 was highly restrictive. Joshi and Little (1996) concluded that “In June 1991, India was the most autarkic non-communist country in the world. Despite a little liberalization in the 1980s all imports were subject to licensing or were prohibited. Licenses were in general granted only on proof that there was no source of indigenous supply . . . All bulk items (e.g. cereals, petroleum, ores, metals, fertilizers) . . . could be imported only by a government monopoly (Joshi and Little 1996, 63). In the first two years of the reforms, measures liberalizing the trade regime included: (a) the removal of import licensing requirements for most imports (although prohibitions on the import of consumer goods remained); (b) the beginning of a program of tariff reductions; (c) restrictions on inflows of foreign direct and portfolio investment were significantly eased; (d) a number of export restrictions were removed or relaxed (although some remained). As already mentioned, at the beginning of the crisis, the exchange rate was devalued. Subsequently, small changes were made, but the major change took place in 1991, with the 19 percent real devaluation.21 That in itself, of course, liberalized the trade regime, as the quantity of imports demanded fell in response to higher import prices, thus making licensing less restrictive, while the quantity of goods available for export increased in response to their higher price. Tariffs, which stood at an average rate of 125 percent (with the highest rate at 355 percent) in 1991, were successively lowered until, by 1995, the peak rate was 50 percent, with the average probably under 40 percent. The import licensing regime was replaced by a “negative list,” which listed all those goods (including consumer goods) which could not be imported. Those items not so listed were eligible for importation without license. Thus, not only were tariffs lowered, but the protection domestic producers received from quantitative restrictions on imports (and prohibitions when indigenous supply was available) was removed as well. Measures were also taken to encourage exports. Exchange rate depreciation was one measure; reduction of protection through tariffs and the removal of quantitative restrictions on imports also made exporting more attractive. In addition, some goods whose export had been permitted only through government trading companies were decanalized. 21. In the initial stage of reforms, a dual exchange rate system was introduced under which a free market was used for most transactions and a controlled rate for certain categories of imports. See IMF (1992).

24

Anne O. Krueger and Sajjid Chinoy

Similarly, the government of India for the first time attempted to attract foreign direct and portfolio investment by reducing and removing restrictions on it. Until 1991, foreign investment was permitted only in cases in which it provided technology transfer, and equity participation above 40 percent was generally not permitted. In July 1991, foreign technology agreements, foreign direct investments, and industrial licensing were all liberalized.22 In some circumstances, the Reserve Bank of India (RBI) was allowed to give automatic approval to foreign acquisition of equity, and in many other instances, procedures were greatly simplified. The percentage shares of domestic firms that could be owned by foreigners were increased, and efforts were made to simplify the approvals process. Finally, export controls were removed in some cases and relaxed in others. Nonetheless, restrictions remained, especially when agricultural exports were involved. Thus, between 1991 and 1995 (with liberalization concentrated in 1993 and 1994), the Indian trade and payments regime was substantially liberalized. However, despite repeated tariff reductions, Indian tariffs still remain high by world standards, as other countries have also been lowering their tariffs. One of the areas in which further reform is called for and that is not covered in one of the background chapters is the trade regime, a topic to which we therefore return in section 1.4. The financial sector was likewise significantly liberalized, but again from a very illiberal base and with gradual reforms. Over two-thirds of financial assets in India were held in banks by 1991. The banks themselves had been nationalized since 1969 and were subject to two sets of reserve requirements that effectively meant they held more than half their assets in government paper. The banks were also subject to “directed lending,” whereby some of the remaining portion of their assets was to be extended in loans to specified activities, such as agriculture. Moreover, it was generally agreed that the quality of the banks’ portfolios was poor,23 and that the quality of the banks’ portfolios had deteriorated significantly during the 1980s (Joshi and Little 1996, 113). Bank supervision and accounting rules were lax: For example, income was booked when it accrued (even if it was not received), so that it was difficult for anyone to ascertain the true state of the banks’ assets and liabilities. The real rate of return to the banks on their assets in the latter half of the 1980s was estimated to be about 0.15 percent, a very low figure by any standard; this reflected the low quality of their loan portfolios. While “resource mobilization,” in the form of higher savings, had been achieved since independence, the quality of resource allocation achieved by 22. It was at this time that 100 percent foreign ownership of power projects came to be permitted. 23. Joshi and Little (1996), 113, tell us that in 1991, 24 percent of the advances of publicsector banks were non-performing.

The Indian Economy in Global Context

25

the banks was poor. It was even then recognized24 that this was a significant constraint on satisfactory macroeconomic performance and growth. In the structural reforms of the early 1990s, therefore, a number of measures were taken to attempt to improve that performance. Recognizing that the poor state of the financial system was a significant drain on economic growth, the government of Prime Minister Narasimha Rao appointed a Committee on the Financial System under the chairmanship of M. Narasimham (known as the First Narasimham Committee). The conclusions of this committee formed the basis for significant financial reforms. In 1992, the RBI issued revised standards for income recognition, asset classification, and provisioning, and established new capital adequacy standards meeting the Basle Accord. These were in full force by 1996. The government also established a Board of Financial Supervision within the RBI. In addition, the RBI strengthened its own monitoring procedures for banks. At the same time, appropriate revenue recognition rules and other accounting changes exposed the true weakened state of the banking system: Nonperforming assets (NPAs) constituted about 24 percent of the total loan portfolio of the public sector banks (virtually the entire banking system). In the strongest banks, the proportion was 8–10 percent, while in the weakest, it was 25–40 percent! While they are still high, NPAs had fallen to 16 percent of gross advances in 1997–98 (Reserve Bank of India 1999). Restructuring of bank portfolios was therefore of vital importance, and the banks were mostly recapitalized by 1995. A second Narasimham Committee also advocated the adoption of capital adequacy standards commensurate with the riskiness of loans and advocated a level of 10 percent for most Indian loans to the private sector; this was incorporated into law in the budget for 1998–99, to be phased in gradually. Although problems remained and remain in the banking sector, especially with regard to the efficiency of bank management and the extent of directed credit, there is no doubt that the first round of reforms was “just in time.” In hindsight, it seems evident that a financial collapse may have been imminent. At the same time as bank restructuring was taking place, interest rate ceilings were being removed, and interest rates were freed to a considerable degree.25 Banks may now set lending and deposit rates on most accounts ex24. The vital importance of the quality of bank lending for economic growth has been better appreciated since the Asian financial crisis of 1997–98. Especially in countries such as India, in which banks account for a very large proportion of finance for economic activity, the quality of their lending directly and importantly affects the rate of growth. 25. One purpose of interest rate ceilings was to make it low cost to finance the government’s borrowing requirements. Changing these procedures after 1991 was of great importance, and constituted a major reform. It is not further dwelt on here because the crucial changes have already been effected.

26

Anne O. Krueger and Sajjid Chinoy

cept for export credits, savings accounts, and loans of less than Rs 2 lakh (Reserve Bank of India website). Directed lending, however, remains important. As of March 1999, it still constituted about 33 percent of gross bank lending (Reserve Bank of India 1999). In addition, banks are still considered to be very high cost, and inefficiently run, with overmanning and high staff costs. Their operating expenses average between 2.5 and 3 percent of total assets, which is very high by international standards. This is a significant factor in maintaining high lending rates from banks.26 To the extent that labor market regulations and failure to privatize the banks account for these high costs, the reforms that would support improvement in this area are discussed further in section 1.4. Another aspect of financial reform, which has not proceeded as far as did financial deregulation, has been the introduction of competition into the banking system. Banks are still primarily government-owned, although the percentage of permitted private equity has been increased. Anecdotal evidence suggests that banks are still relatively high-cost, and that efforts to reduce costs and increase efficiency have to date been limited. Entry into banking is still relatively difficult, and certainly not encouraged. It was mentioned above that about two-thirds of Indian financial assets are in the banking system. For the non-banking components of the financial system, a Securities and Exchange Board of India (SEBI) was established in 1988, but given statutory powers to oversee securities markets and so forth only in 1992, with strengthened powers in 1995. For the primary equity market, SEBI has considerably improved oversight, insisting on adequate reporting in prospectuses and requiring restitution when false information is provided. Although problems remain, and there is considerable scope for further improvements, there is little doubt that SEBI’s authority and use of it have significantly improved the efficiency of the primary equity market in India. SEBI was also assigned the task of improving the working of the Indian stock exchange. Here, changes came about in part in response to the opening of the National Stock Exchange as the Bombay Stock Exchange responded to competition. Screen-based trading has begun, depositories have been established, and the market is considerably more transparent than it was previously. Less was accomplished in the early 1990s with respect to development finance institutions and to insurance. Nonetheless, the overall impression is, and should be, that the early years of reform, from 1991 to 1994 or 1995, were years of significant changes in regulation, in institutional arrangements, and in other factors affecting the financial environment in which businesses undertook their activities. 26. Of course, the major factor is the high demand for bank loans relative to supply, given that banks are constrained to place such a high fraction of their resources in government paper.

The Indian Economy in Global Context

27

In addition to reforms in the financial sector and in the trade and payments regime, the Rao government tackled a variety of other problem areas. Efforts were made to improve telecoms by changing the regulatory framework and other means; these efforts are described in Section 4. Likewise, the monopoly of Indian airlines was broken up, and entry was permitted into the airline, and a number of other, industries. Efforts were also made to reform the power sector in light of the urgent need for additional power, but to date these have not had as much of an impact as they were expected to due to remaining bottlenecks (see section 1.4). To be sure, there were areas where reforms were either nonexistent or very, very tentative. Chief among these, to which attention returns in section 1.4, are labor markets, operation of state-owned enterprises, and efforts to privatize existing national entities (beyond selling minority equity, as mentioned above). There were also a number of other areas in which the government was fully cognizant of difficulties but unable to address the issues, as with the state of infrastructure, as discussed in section 1.4. Nonetheless, the period starting in 1991, and especially the first few years, stands in marked contrast to earlier years in Indian economic policy making. For the first time, not only was it acknowledged that reforms were highly desirable, and indeed essential if growth was to accelerate, but reforms were actually begun. Despite the agenda discussed in section 1.4, one should not lose sight of the significance of that change. 1.3 Economic Growth over the 1993–2000 Period The momentum for reform was greatest in the years immediately following the crisis of 1991. By the mid-1990s, reforms had largely stalled. While there were some further changes in the control and regulatory environment and the role of the public sector, by and large the major changes had taken place in the first half of the decade. In part because governments were not secure and in any event consisted of coalitions, no major new initiatives were undertaken. Nonetheless, economic growth accelerated markedly during the 1990s, due in large part to the stimulus provided by the reforms themselves. While there is some evidence that in 1998 and 1999 the stimulus to growth had diminished, it nonetheless seems clear that what was done in the early 1990s was sufficient to enable the country to attain a more satisfactory economic performance than at any earlier point. Growth rates reached 7 and 8 percent during the late 1990s. Moreover, analysts and policy makers alike became convinced that growth rates of 7 percent and more were achievable, something that would have been vigorously contested only a decade earlier. Nonetheless, all observers agree that reforms have lost their momentum and that, if the economy is to be able to attain and sustain a growth rate of 7 percent or more, the momentum for reform will once again have to be regained.

28 Table 1.5

Anne O. Krueger and Sajjid Chinoy Fiscal Stance after the Reforms

Year

Government Expenditures
GDP

Government Revenues
GDP

Fiscal Deficit
GDP

Current Account Deficit
GDP

1991 1992 1993 1994 1995 1996 1997 1998

20.4 20.0 18.9 18.4 16.8 16.9 18.4 17.6

14.5 14.2 11.8 12.7 12.6 12.6 12.6 12.3

5.8 5.7 7.0 5.6 5.1 4.9 5.8 5.3

–1.6 –1.6 –0.7 –0.5 –1.5 –1.5 –0.7 –1.6

Sources: International Monetary Fund, International Financial Statistics Yearbook 2000, pp. 162–63 for Current Account Deficit; India pages for Government Expenditures, Revenues, and Deficit. Notes: Government Expenditures are central government expenditures including loans (net of repayments) to the states. Government Revenues, Expenditures, and Deficit figures for 1997 and 1998 are preliminary.

In this section, we assess economic growth during the 1990s. In section 1.4, we then examine the key areas in which reforms are urgently needed. To assess economic growth, it is useful to start with data on fiscal and current account behavior over the period since the crisis. These data are given in table 1.5. As can be seen, central government expenditures as a percentage of GDP have been reduced, falling from 20.4 percent of GDP in 1991 to 18.4 percent by 1997 and 17.6 percent in 1998. Government revenues also declined, although by a lesser proportion, so that the fiscal deficit fell from around 7 percent of GDP in 1993 to 5.3 percent in 1998.27 Reflecting this trend, the current account deficit fell from its very high levels of the late 1980s to more moderate and sustainable levels of around 1 to 1.5 percent by the mid 1990s.28 Table 1.4 gives the key macroeconomic indicators for the 1990s. As can be seen, after 1991 the growth rate of real GDP rose markedly, reaching 7.4 percent in 1994 and 7.6 percent in 1995, but falling somewhat thereafter. At the same time, the rate of inflation fell from its high of 13.5 percent in 1991 to 7.5 percent by 1993, but then rose to 10.5 percent in 1994. Only after 1995 did it drop significantly to the 5–7 percent range, and the current account, as previously mentioned, remained at very manageable levels. Thus, the key 27. See the chapter by Srinivasan in this volume for a thorough exploration of the fiscal position and the relationship between the budget of the center and total government expenditures and revenues. 28. This reduction in the fiscal deficit undoubtedly had a number of effects. Among them, it permitted the unfreezing of interest rates without undue increases in them. Had the fiscal deficit remained at earlier levels, either monetary policy would have had to be tightened, with (probably unacceptable) pressure for higher interest rates, or the rate of inflation would have accelerated rapidly.

The Indian Economy in Global Context

29

macroeconomic indicators have performed well, and the average rate of economic growth reached unprecedented highs. There were disquieting signs on the fiscal front, however. First, as has already been seen and as is explored in Srinivasan’s chapter in this volume, the fiscal deficit was not reduced as much as had been announced at the outset of the reforms. Second, the composition of government expenditures has not been altered to increase their efficiency. The need for infrastructure and other public sector investments is very great, and yet prospects for increased resources in the public sector do not appear good: Reallocation of resources is clearly called for. Third, investment as a percentage of GDP has remained fairly stagnant, despite the pressing need for new investments in a number of areas. If one were to assess fiscal performance since the crisis, the conclusion would have to be that the initial stabilization package met with some success, but that reforms faltered before the underlying structural problems of inefficient expenditures and inelastic sources of revenue could be tackled. Success was greater in some other areas. Export performance improved notably. Table 1.2 gives data on exports in the decades since independence. As can be seen, export volume grew at an average annual rate of 11.45 percent from 1991–97—almost double the highest rate achieved in any prior decade. This was reflected in an increased share of exports of goods and services in GDP by the latter part of the 1990s—9.94 percent, compared to 5.83 percent in the 1971–80 decade and 6.52 percent in the 1980s. The growth of export earnings was rapid enough to reverse India’s declining share in world trade: Having started at 1.42 percent in the 1950s, it had fallen to a low of 0.49 percent in the 1980s. Then, in the 1991–97 period, it rebounded to 0.58 percent. Reflecting import liberalization, imports also grew rapidly over the 1990s, rising at an average annual rate of 14.36 percent, to a 10.61 percent share of GDP. Import liberalization no doubt served to enable Indian businesses to compete more effectively abroad, as well as to obtain needed capital and intermediate goods at more competitive rates in a timely way. Even then, there had been little liberalization of consumer goods imports by the late 1990s. 1.4 An Overview of Priorities for Reform as of 2000 As of 2000, it was clear that the Indian economy had changed markedly over the post-crisis years. Growth rates did accelerate (although there are questions as to the sustainability of these higher rates without a new wave of reform), the economy was opened considerably, and the extent to which controls stifled economic activity had been reduced. As in all growing economies, however, the degree to which regulations

30

Anne O. Krueger and Sajjid Chinoy

and controls are harmful to growth and living standards changes as economies grow. The set of import substitution policies under which growth in many developing countries reached 6–7 percent in the 1950s and 1960s, for example, constrained growth to rates of at most 3–4 percent by the 1980s. The same is probably true for India: Removal of some of the inefficiencies resulting from the trade regime, the structure of the financial sector, and industrial licensing clearly permitted a spurt of growth. Whether growth at the rates of the late 1990s is sustainable without further reforms is questionable; what is unarguable is that, with further reforms, it is likely (in the absence of adverse developments internationally or other major setbacks) that growth rates could rise to the 7–8 percent range over the next decade.29 Some essential changes were already in motion in 2000. A decade earlier, most analysts would have regarded the low levels of educational attainments of much of the Indian labor force as constituting a potentially serious constraint on the rate of growth. Improvements in access to education have probably reduced the severity of that bottleneck, although much needs to be done to improve access to secondary education and quality of primary education. Since those issues are discussed by Kochar and by Foster and Rosenzweig in their chapters in this volume, they are not further addressed here. Another much-needed step has been the removal of barriers to imports of consumer goods and a significant reduction of protection to domestic producers of consumer goods. Not only will there be gains to Indian consumers and an allocation of resources to more efficient uses, but in addition, competition is likely to spur increased efficiency of consumer goods producers in India. Relaxation of restrictions on the imports of consumer goods were negotiated under the World Trade Organization (WTO), and licensing and quota restrictions for 714 consumer goods imports were relaxed on April 1, 2000, although they were still going to be subject to 44 percent import duty. It was announced that another 715 items were to be removed from licensing and quota restrictions a year later.30 Further gradual reforms are taking place in the financial sector as well. The insurance industry is being opened up, and improvements in banking regulation, supervision, and incentives, continue. In the fall of 1999, after a coalition government was reelected, it fought through and won a new insurance law, which permitted foreign investment in insurance companies (up to 25 percent of equity) and otherwise began liberalizing insurance law. 29. Of course, even if adverse international circumstances do occur, those countries whose economies are most flexible will best be able to cope. 30. The Times of India, “Doors Now Wide Open For Consumer Goods Imports,” 19 April 2000. It was also announced that special economic zones (SEZs) would be established outside of Indian customs jurisdiction in an effort to spur exports. These zones, however, were to be subject to Indian labor laws.

The Indian Economy in Global Context

31

Other issues require additional attention. Most analysts believe that there remain considerable bottlenecks, red tape, and bureaucratic delays associated with doing business in India. These are, of course, difficult issues to quantify, but their importance emerges repeatedly in the chapters by Forbes, Murthy and Raju, and Saxenian in this volume. Further fiscal measures are urgently needed. Public sector fiscal deficits, as discussed in depth by Srinivasan, with the higher market-determined interest rates at which the government is currently borrowing, mean that the government is perilously close to an “internal debt trap,” having to borrow simply to continue debt-servicing obligations at an increasing rate. Moreover, the government’s continuing borrowing has kept interest rates relatively high. To the extent that inflation has been avoided, it has been by restraining private investment through these interest rates. Addressing the fiscal issue is essential if only to avoid destabilizing the entire macroeconomy; but, in addition, it will release investment resources for productive uses in other sectors of the economy. As in most developing countries, issues of governance are also important. Not only is there an urgent need for increased efficiency and timeliness of the court and legal system, but issues of corruption and evasion will need to be addressed. Whether they can be better dealt with head on, or whether a gradual and piecemeal approach will be more conducive to strengthening the commercial code and governance institutions is an open question, and one not addressed in this volume. While action on these fronts is clearly desirable, achieving a 7–8 percent rate of economic growth over the coming decade will clearly require other measures. Perhaps most obvious, but not necessarily most important, is the need to address the inadequacy of existing infrastructure capacity and quality, and present prospects for expansion and improvement. During the 1990s, it is widely believed, rapid growth was possible because there was slack in the system (Ahluwalia 1998, 89). By the end of the decade, however, the growth of the 1990s had already pushed infrastructure utilization well above the optimum. It seems unquestionable that there will be strong diminishing returns to economic activity unless means are found for improving infrastructure quality and quantity. Since this is not the subject of a separate chapter in this volume, the situation and problems confronting the delivery of infrastructure are discussed first. However, while infrastructure is the most obvious bottleneck, there are others. To date, little has been done with regard to privatization or exit of uneconomic state-owned enterprises. Similarly, it is increasingly recognized that existing labor laws will require significant changes if growth is to be maintained and accelerated. Each of these areas is briefly discussed following the discussion of infrastructure; failure to cover them in depth is simply a reflection of capacity limits as to what can be accomplished at one conference.

32

Anne O. Krueger and Sajjid Chinoy

1.4.1 Infrastructure That infrastructure is a bottleneck to Indian growth is widely recognized.31 What is not so often recognized is the extent to which even the existing stock of infrastructure serves as a growth barrier. In this section, we first briefly address the issue of the linkages between infrastructure and growth, and then consider some of the indicators of the extent to which India’s infrastructure lags behind that of most other countries. We then turn to the potential for addressing this critical issue for individual infrastructure sectors. Negative Effects of Infrastructure Inadequacy on Economic Performance That infrastructure is essential to support economic activity is selfevident. If there is no road, or if telecommunications facilities do not exist, it is not possible for producers to enter into market activity with other areas. But the fact of physical availability does not tell even half of the story. When infrastructure facilities are inadequate, they impinge heavily on producers’ costs, and hence on the productivity and efficiency of other resources. If roads are badly maintained and full of potholes, for example, trucks’ economic lives are significantly shortened, and the time goods and drivers are in transit increases. Both of these effects represent real costs that must be deducted from the price received by producers when goods are sold in the market at their point of delivery. For many products in today’s world, delays in transport and communications serve as an even more serious bottleneck. Many producers themselves use just-in-time inventory, and will not deal with suppliers whose deliveries are tardy or unreliable.32 Especially for products that are intensive in the use of unskilled labor, buyers are seeking low-cost sources. When poor infrastructure adds to costs and simultaneously creates uncertainty in delivery times (or inability to be assured of instantaneous communications), buyers are likely to seek alternative sources in this competitive world environment. In India’s case, high costs imposed on producers by infrastructure are no doubt a significant factor in raising costs to producers, both domestic and potential exporters. In addition, however, perceptions of India’s infrastructure are so negative that they probably constitute a significant barrier to foreigners’ willingness to undertake direct foreign investment in India. In 1999, the World Economic Forum (1999) asked businessmen to rank countries according to the adequacy of their overall infrastructure and of its various segments. Altogether, fifty-nine countries were ranked. Of those fifty-nine, India was ranked fifty-fifth in terms of the “adequacy of overall infrastructure,” and fiftieth of fifty-nine in the importance accorded to in31. See, for example, Planning Commission (1997), vol. I, ch. 1, and Ministry of Finance (2000), ch. 9. 32. Much of Hong Kong’s early success as an exporter is attributed to the ability of buyers in other countries to obtain reliable delivery on short notice. See Morawetz (1981).

The Indian Economy in Global Context

33

frastructure by the government. For individual components of infrastructure, only railroads fared significantly better, where India was ranked twenty-eighth of fifty-nine. For adequacy and efficiency of road infrastructure (fifty-fifth), port facilities (fifty-third), telecommunications (fifty-first), and air transport (forty-fifth), India was ranked well down the scale. Quantitative rankings tend to confirm these perceptions. In telephone penetration (measured by main lines per 100 of population), India was fiftyseventh, in rate of filling orders for new connections, India was fiftieth, in kilometers of road per million persons India was forty-first; and in investment in telecommunications per inhabitant India was fifty-second. There are some estimates as to the costs of this poor infrastructure.33 It is estimated that India lost between $5 and $6 billion annually because of the slow speeds and other inadequacies of the roads on which commercial vehicles have to operate. Likewise, it is estimated that at least $70 million was lost annually due to container delays in ports. While it is more difficult to estimate losses for what is not there, estimates are that as much as $23 billion information technology export revenues and 650,000 jobs failed to materialize over an eight-year period because of the state of telecommunications infrastructure (NASSCOM-McKinsey 1999). These estimates include only fairly tangible items. Electricity shortages and voltage fluctuations impose considerable costs on firms (and even households that buy electric generators). And much cargo is loaded on small ships at Indian ports only to be transferred to larger vessels in Singapore, Colombia, and Hong Kong. India Today put the cost of poor infrastructure at about 3 percent of GDP annually, which is about two-thirds of what is spent annually on infrastructure (India Today, 31 January 2000). In addition to these losses, there are heavy penalties, especially for the poor, who cannot compensate with electric generators, bottled water, or other substitutes for existing public infrastructure.34 Not only are there costs in terms of health and well-being, but opportunities to engage in productive activities are also limited by the scope of the market. Power It is difficult to estimate the relative needs in different infrastructure sectors. Many see the shortage of power as the most critical constraint. During 33. Data in the next several paragraphs are from Government of India (1996). 34. The World Bank estimates that of the twenty-seven cities in Asia with populations over 1 million, India’s four largest cities are among the five worst in terms of water availability, and that 1.5 million children under the age of 5 die annually from water-borne diseases (as reported in Financial Times, 17 March 2000). One hundred fifty million households are still without electricity, and 2.8 million people are still awaiting telephone connections. Half of all villages are not yet connected by all-weather roads, and the traffic fatality rate is estimated to be 25 times that in the United States. (Electricity information is from India Today, 31 January 2000. Telephone information is from International Telecommunications Union, 1998. Road information is from Indira Gandhi Institute of Development Research [1997].)

34

Anne O. Krueger and Sajjid Chinoy

the 1990s, power use grew at about 1.5 times the rate of growth of output, but supply was not increased commensurately, despite the fact that power supply per capita had tripled over the 1975–95 period. Relative to power supplies elsewhere, India’s lag is huge. Korea, Malaysia, and Argentina, for example, have per capita capacities between five and eight times higher than India’s. China’s power supply per capita was judged to be about the same as India’s and is now 1.9 times as high.35 While there are opportunities for sizable increases in operating efficiencies,36 increased capacity will be essential if power capacity is not to serve as a severe bottleneck to growth. Problems in increasing capacity are manifold: Obviously the desirable amount of investment is large, but the return may be high if investments are wisely done. In most countries in these circumstances, profits from existing companies would provide a sizable share of financing for new plants. However, the Indian domestic power producers are generally loss-making entities. The State Electricity Boards (SEBs) consistently made losses in the 1990s, rising from 12.7 percent on invested capital in 1991–92 to 21.1 percent in 1998–99 (Government of India, Ministry of Finance 2000). Part of this loss resulted from the large subsidies implicit in the low price of electricity to farmers and other consumers: for example, compared with an average cost of generation and distribution of power of Rs 1.86 per kilowatt in 1996–67, the average price to farmers was only Rs 0.21 and that to all domestic consumers was only Rs 0.90. Industrial users paid more than average cost, but from all users of power, the average revenue for that year was Rs 1.49, which covered only 80 percent of average cost. Hence, the electric power producers require financing simply to cover operating costs, much less to finance new capacity. While measures have been taken to encourage private investors in the power sector, those investors too are likely to be deterred by the ability of SEBs to price electricity at uneconomically low rates until the problem of pricing is realistically addressed.37 Meanwhile, those industrial users paying high prices are either locating inefficient (but lower-cost) captive power suppliers than the SEBs, or they are at a competitive disadvantage. 35. Data are from United Nations, Energy Statistics Yearbook, various years. In 1975, India generated 143 kilowatt hours per capita, contrasted with China’s 173, Egypt’s 279 and Korea’s 1,514; in 1995 the comparable figures were 448 for India, 839 for China, 787 for Egypt, and 4,567 for Korea. 36. It is estimated that the State Electricity Boards (SEBs) have plant load factors of between 55 and 60 percent, contrasted with private and central power sector plants, which run close to 70 percent. Simply raising the load factors by 10 percentage points would increase the supply from SEBs by about 18 percent. In addition, it is estimated that India suffered transmission and distribution losses of 18 percent in 1990, well above the international standard of 10 percent, and greater than those of any other countries reported except Nigeria and Bangladesh. 37. For instance, the tenuous financial position of the Maharashtra State Electricity Board (MSEB) has meant that it has, on occasion, defaulted on its payments to the local subsidiary of the U.S. power company, Enron, and has had to be bailed out by the state and central government.

The Indian Economy in Global Context

35

Both because of the critical importance of increasing power capacity and production, and because changing the pricing of electricity would itself do much to make usage more efficient and generate some of the needed additional funding (directly through increased profits of the electricity producers and through the increased attractiveness of power investments to foreign investors), reform of the relationships of center and states with regard to electricity pricing is clearly a high priority for reform. Even with such reforms, additional financing from the center would probably in any case have a high rate of return. However, given the fiscal difficulties confronting the center, failure to reform pricing policies would inevitably result in smaller investments in power capacity than appear economically warranted. Telecommunications For telecoms, significant efforts at reforms were under way in the 1990s, and there have been significant improvements in both quality and availability of service.38 However, the starting base was so low that much remains to be done. We have already mentioned that as of the mid-1990s, international data show India to have been well down the list in terms of the number of telephone lines per 100 inhabitants and investment in telecommunications per 100 inhabitants. Indeed, the country also had the lowest telecommunications revenue of the fifty-three countries covered in the 1996 survey, despite having the fourth largest telecommunications staff (behind the United States, China, and Russia), with 421,060 employees. Interestingly, India had 18.6 telephone main lines per 1000 people in 1997, contrasted with 444 in Korea, 107 in Brazil, 250 in Turkey, and 96 in Mexico. There were 33.7 main lines per employee, compared to 294 in Korea, 224 in Argentina, and 114 in China. Average telephone faults, usually chosen as the best available indicator of the quality of service, were 196 per hundred main lines in 1995, compared to 3 in Brazil, 18 in Korea, and 82 in Cote d’Ivoire. Waiting time for a phone (in months) in 1995 was 12.17, whereas it stood at 4.2 months in Turkey, 2.84 months in Morocco, and 0.68 months in China. By contrast, it stood at 79.68 months in Bangladesh, the country with the longest reported waiting time among the twenty-one countries covered. Accelerating growth will certainly increase the demand for quality and timely telecommunications services rapidly. The importance of both quality and availability of reliable telecommunications services cannot be exaggerated, especially in the context of India’s need to integrate further with the global economy. Historically, the telecom sector in India has been dominated by the Department of Telecommunications (DoT) and two government-run companies: Videsh Sanchar Nigam Limited (VSNL), which still 38. Ahluwalia (1998), 103, for example, cites telecommunications reforms as a relative success.

36

Anne O. Krueger and Sajjid Chinoy

controls a large part of international communications, and Mahanagar Telephone Nigam Limited (MTNL), which operates local telephone services in major cities. The DoT first attempted to introduce competition in the area of basic telecom services in 1994, but private operators were burdened with licensing fees that the government companies did not have to pay. Moreover, auctioning of licenses took place with restrictions on the number of entrants, evidently for purposes of maximizing revenue: Srinivasan appears to believe that the auction charges essentially were high enough so that monopoly, rather than competition, would prevail (Srinivasan 2000, 62). The resulting process resulted in a vicious bidding war between companies to acquire the license in each calling circle. As was later discovered, many overly optimistic bidders had bid well beyond their capacities and subsequently were unable to keep their commitments, and so were forced to default. Six years later, the basic telecom service sector is stagnant: Private operations have been rolled out in a limited way in only two of the thirteen basic circles for which service licenses were auctioned (Economic Times, 31 August 2000). Five years after this fiasco, the government attempted to jump-start the liberalization process with the New Telecom Policy (NTP) of 1999. Under this policy, the government has recently decided to introduce muchanticipated competition into the national long distance sector on a revenuesharing basis. Unlike its earlier experiment at liberalization, the government has not limited the number of potential entrants this time. Instead, the sector is open to anyone, subject to an entry fee and a net-worth requirement. Furthermore, the government has also decided to introduce competition—and eliminate VSNL’s monopoly—in the international long distance market much earlier than pledged in the commitment given to the WTO. Finally, efforts are on to resuscitate the competition into the basic telecom services through easier entry norms. Under current proposals, fifteen vacant circles will be opened to unrestricted entry, on a revenuesharing basis, subject to an entry fee and a net-worth requirement analogous to the framework governing the national long-distance sector. Apart from improving service quality, the impending competition in the national long-distance market promises to reduce prices for households and businesses significantly and to provided a boost to the information technology sector. All this while, the DoT, in its bid to cross-subsidize local calls, has been keeping long-distance charges exorbitantly high. Longdistance rates within India are between four and fourteen times as high as similar calls within the U.S. (Economic Times, 30 August 2000). And this markup is reflected in the accounts of the DoT. In 1998–99, for instance, the fixed and operational cost of national long-distance services constituted only fourteen percent of its revenues from this sector. To address this im-

The Indian Economy in Global Context

37

balance, the Telecom Regulatory Authority of India (TRAI), established as a regulator of the industry in 1999, twice reduced long-distance rates in 1999–2000.39 However, competition in this sector promises to bring further price relief to consumers and households. Railroads As indicated above, railroads are in relatively better condition than other forms of transport and communications. The Indian railway system is the second largest in the world, and modernization is occurring, but much remains to be done. It is estimated that 63 percent of the 40,000 route miles were still under broad gauge as of 1996, while electrified networks covered only 22 percent of the total route length. Moreover, the railroads are certainly operating at, if not above, full capacity, and expansion will have to occur to accommodate rapid economic growth. As in other infrastructure sectors, pricing is a major problem. The railroads are required to charge low prices for “social obligations,” including the transport of mass consumption goods such as sugar cane, salt, and edible oils. These goods are carried below cost, as are passengers on short distance second class and season tickets. For these reasons, the railroads themselves have had low profitability, which restricts the degree to which they can self-finance maintenance, modernization, and capacity expansion, and at the same time deters other investors. Ports Whereas railroads are overburdened, port capacity and organization is an even more pressing issue. The 1996 India Infrastructure Report (IIR) indicated that in 1994–95 seven of the eleven major ports were operating at capacity levels above 100 percent. Indian ports are also inefficient when compared with their competitors in nearby countries. Only 7 percent of traffic in Indian ports in 1994 was containerized, despite the great potential for efficiency gains. India had an average ship turnaround of seven days in 1993–94, contrasted with six to eight hours in Singapore. The World Bank estimates that the cost burden to importers and exporters because of these and other inefficiencies is about $250 million per year.40 39. Efforts at rebalancing tariffs on the part of TRAI, thus far, have inevitably led to conflicts between TRAI and DoT, but, as concluded by Srinivasan, “For Indian telephony to grow and be efficient, someone has to set the right prices in uncompetitive markets, and manage the process by which companies interconnect with each other so that multiple systems can still operate together as a single national telecommunications system. A respected, independent regulator is necessary to absorb and deflect all of the short-term political heat that naturally is created . . .” (1999, 63). 40. These estimates are reported in the IIR. It should be noted that these estimates are costs to existing importers and exporters and do not include the exports and imports that do not occur because of the port situation.

38

Anne O. Krueger and Sajjid Chinoy

Not only are there long delays in Indian ports, but also low labor and capital productivity. The IIR reports that the number of units handled per crane per hour is seven in Mumbai fifteen in Chennai, twenty-six in Colombo, and thirty-two in Singapore. While it is evident that increases in port capacity will require investment in modernization, there is likely ample scope for efficiency improvements within the existing infrastructure, or with relatively low investments. Reforms in the organization of ports may be an area in which the capital costs of improvements are significantly less than in some other areas of infrastructure. 1.4.2 Relatively Untouched Areas While it is clear that infrastructure needs will have to be addressed and that the fiscal gap must be closed, there are other areas where reform has barely started, and where appropriate measures can significantly accelerate economic growth. These include reform of labor markets, increased efficiency and effectiveness of the commercial code and judicial system, and the restructuring of state-owned enterprises.41 Of course, there are close linkages between various parts of the economy, and improvements in the functioning of any market has spillovers to other parts. One of the key factors affecting the demand for labor in India, for example, is the reservation of a number of industries to small scale firms. These reservations, intended to protect small-scale producers from competition they presumably cannot handle from larger firms, have severe effects on many aspects of Indian economic activity: Reservation penalizes successful small firms and creates a barrier to their expansion. Since smallscale firms are normally labor-intensive, the inability to expand probably represents a significant deterrent to exporting in just those labor-intensive areas where Indian industry might have a considerable competitive advantage if allowed to expand (and small-scale firms suffer a number of disadvantages in export markets). Rakesh Mohan covers these and other detrimental effects of small-scale reservation in his chapter in this volume, so the impact of reservation will not receive further consideration here, although it is perhaps one of the politically easier reforms to undertake. Likewise, improvements in the flexibility and functioning of financial markets can have positive spillover effects on exports, on the ability of competitive firms to expand, and thus on the demand for labor. Nonetheless, if India is to achieve anything approaching her growth potential, it seems evident that reforms in labor laws will be essential. We turn our attention next to the importance of these reforms. 41. Further reforms in the financial and other sectors are also highly desirable, but probably not of such pressing importance as beginning to change the labor market and state-ownedenterprise structures.

The Indian Economy in Global Context

39

Labor Markets At an aggregate level, the average real wage in an economy cannot for long increase at any rate significantly different from that at which the average productivity of labor grows. Employers cannot be induced to employ workers when the wage is above the (marginal) productivity of those workers, and when real wages are high, employers hire fewer workers at high wages at the expense of other workers who might find employment. There are many determinants of labor productivity. Among them, the quality of labor is important, and is heavily influenced by the quality of education and the level of educational attainment of the labor force. Improving the “human capital” of India’s labor force (including access to health care as well as education) is clearly an important task in the years ahead. Since educational reforms are covered in this volume, they are not considered further here. At any given time, the educational attainments and work experience of the labor force are largely a given, and thus it is conditions of employment (the amount and quality of machinery and equipment that workers use, the efficiency of the organization in which they are employed, and so on) that influence the productivity of labor. Labor market regulations affecting those variables affect productivity and, hence, for given wages, the degree to which employers will demand labor or, instead, either use machinery or produce less than they otherwise would have. When efforts are made to regulate the labor market, policy makers face several trade-offs. They can attempt to enforce higher wages only at the cost of lower employment (at the least) and they may drive economic activity “underground” or to the “informal sector” at even lower wages. They can attempt to protect the job rights of existing workers only at the cost of encouraging employers to adopt capital-using techniques (as the advantage to hiring labor diminishes when employers must accept that they have no future flexibility), hence reducing the number of new jobs. Additionally, when benefits for workers (such as housing, education, and health services) are mandated, those benefits raise employers’ costs in much the same way as wage increases do. Indian labor regulation, which was intended to protect workers, has clearly had unintended side effects. These have included locking workers and employers into jobs (and thus both reducing labor mobility to places where it earns the highest return and discouraging employers from hiring additional workers), giving workers’ unions sufficient power so that productivity has fallen (with random and unpredictable strikes),42 and discour42. Any 10 percent of employees have a right to form a union in India. This ruling is a reform from the rule prior to 1991, which permitted any seven employees to form a union. Even the 10 percent rule permits a large number of unions, rivalries between them, unpredictability of strikes (called by various unions), and an inability of management to establish meaningful

40

Anne O. Krueger and Sajjid Chinoy

aging the development of labor-intensive industries that might otherwise have a competitive advantage in export markets. Contrasting the evolution of the labor markets in India and in East Asian countries is instructive.43 In East Asia, where labor laws are considerably less restrictive and flexibility in the labor market is much greater, employment in manufacturing grew at an average annual rate of 6.43 percent over the 1972–92 period, while real wages in manufacturing grew at an average annual rate of 5.3 percent. By contrast, in India, the growth rate of employment was 1.83 percent, and that of real wages 1.12 percent. Flexibility in East Asia was therefore accompanied by a more rapid growth of both employment and real wages. Not only has regulation of the labor market discouraged the growth of employment and productivity in those activities subject to the law, but it has also pushed many activities into the unorganized sector. It is estimated that the annual rate of growth of employment in the organized sector over the 1981–91 decade was 1.58 percent, contrasted with 2.73 percent in the unorganized sector. Job security in India also appears to be so rigid as to provide disincentives for workers, to remove employers’ ability to discipline workers, and to be a major deterrent to employing new workers. In the public sector, workers “have enjoyed almost complete job security since independence” (Agrawal 1997, 161), and large private sector firms employing over 100 workers may not retrench any worker who has been employed for over 240 days without permission from the government. Even with permission, the firm must give the employee three months’ notice, as well as fifteen days’ wages for each year of service with the employer. It is reported, however, that permission is “almost never granted” (Agrawal 1997, 161). Promotions are based almost entirely on seniority, especially in the public sector, which accounts for 70 percent of employment in the organized sector. As aptly put by Agrawal, “Firms are not able to rationalize their operations and labour force in response to changing market conditions. Even loss making firms (referred to as ‘sick’ firms) are not allowed to close down but given subsidized credit and other facilities to continue operation. The subsidies given to the loss making firms sometimes help to keep the product prices artificially low which in turn may spread sickness to other firms in the industry as well. In addition, these subsidies give perverse incentives to become ‘sick’ and to remain ‘sick’, resulting in a huge and ever increasing drain on public

dialogues with union leaders (who change frequently). Agrawal (1997) reports that, based on ILO statistics, over the periods 1972–81 and 1982–92 India lost an average of 4.07 and 5.74 workdays per employee due to strikes, contrasted with 0.108 and 0.010 days in Thailand for the same periods, and 0.333 and 0.168 for France, although he notes that the number of days lost to strikes declined somewhat in the mid-1990s. 43. The material in this section is drawn from Agrawal (1997), 156 ff.

The Indian Economy in Global Context

41

resources. . . . It also leads to valuable capital and labour resources getting locked-up in inefficient production” (1997, 162) Agrawal further notes the tendency of Indian firms to use casual and contract workers who are not protected by these laws and regulations. While low living standards and the absence of a safety net may make total reform of the labor market extremely difficult, it is evident that a number of measures could increase flexibility: The length of time required before employees cannot be retrenched could be increased to two to three years; more use could be made of bonuses instead of wage increases, which would permit reduction of bonuses when firms have difficult years; a mechanism could be devised to permit employers to discipline workers for excessive absences or malfeasance in their jobs; rules governing union formation, conditions for strikes, and union responsibilities could be gradually changed to reduce the incidence of rivalry-related difficulties; and “sick” firms should be permitted to exit even if some public assistance for laid-off workers is substituted. It is difficult to estimate how much benefit would come from these reforms, especially without specifying in advance the degree of relaxation of rules that might occur. Nonetheless, it is evident that the present state of labor market regulation in India discourages the growth of employment, retards the development of export activities reliant upon the use of unskilled labor, and in fact creates a growing wedge between the privileged workers in the organized sector and the others in the informal sector. Privatization and ‘Sick’ Enterprises The fact that “sick” enterprises cannot simply be closed down has already been mentioned. That has the effect of keeping resources in lowproductivity industries or firms that could otherwise be employed elsewhere in the economy. Likewise, when public-sector enterprises are loss-making (or making unacceptably low rates of return on invested capital), keeping them functioning by covering their losses from government revenues not only causes fiscal problems (as discussed by Srinivasan) but also has deleterious effects on other producers. When state-owned enterprises are effectively operating under a “soft” budget constraint, any would-be entrants to the industry are confronted with the knowledge that, despite any possibility that they might be able to produce more efficiently, they may have to compete with loss-making firms that do not pay a penalty for cutting prices. Moreover, the fact that there is no strong penalty to managers of state-owned enterprises or to firms that do poorly in the private sector undoubtedly serves as a disincentive. Given the large size of India’s public-sector enterprises, and the extent to which this size has been swelled by acquisition of “sick” enterprises, it is probable that there is room for a large improvement in the productivity and efficiency of the Indian economy if means can be found for improving the

42

Anne O. Krueger and Sajjid Chinoy

efficiency of the activities carried out by these enterprises. To be sure, some enterprises will simply have to be closed down, as they may simply be uneconomic. But for many others, there are doubtless significant opportunities for efficiency improvement with appropriate managers and incentives and with a relaxation of some of the restrictions (such as those on the employment of labor discussed above). In most discussions of privatization, analysis is divided into two or three parts: (a) what is involved in privatization of small shops and other smallscale enterprises; (b) the means for privatizing state-owned enterprises that compete either with other private-sector producers or with imports; and (c) the problems associated with privatization of entities that have a significant degree of monopoly power. The first category, small shops and enterprises, hardly exists in India. At any event, privatization is fairly easy and can be achieved in a reasonably short time. This has been the experience of Eastern European countries as well as those developing countries that had small-scale state-owned firms. The second category should also be fairly easy. With competition either from other domestic firms or from imports, the chief problem is that of appropriate valuation of enterprises. Experience suggests that the gains from privatization can come about both because the new owners can achieve efficiencies the public sector cannot (as, for example, shedding workers or uneconomic locations) or by reorganizing production in a more efficient manner. Especially in the latter case, it is difficult for would-be acquirers of public-sector enterprises to know what the new rate of profitability would be. Experience suggests that it is desirable to privatize fairly rapidly, paying more attention to providing conditions under which the newly privatized entities can function effectively than to maximizing the amount of revenue the government gets.44 Of course, processes for privatizing must be transparent if the process is to be politically acceptable, but it is not possible to ascertain ex ante how much individual producers will be able to achieve in productivity gains. The same considerations apply to “sick” enterprises. To the extent that restructuring can enable them to survive (as for example by shedding labor), that should of course be encouraged. Clearly, though, means must be determined by which truly sick enterprises can be permitted to cease their activities. As of 1999, Srinivasan pointed out that the Disinvestment Commission had achieved little. Sixty-four public-sector enterprises had been referred to the commission, which had made recommendations on fifty-eight. The recommendations in only thirteen cases were or are being implemented. As Srinivasan (1999, 10) notes, “Reducing the commission’s power and delay44. Srinivasan believes that some industries facing competition from imports should receive moderate tariff protection for a period of time after privatization, although he also notes the desirability of rapid reduction in the rates of import duty (1999, 8).

The Indian Economy in Global Context

43

ing actions on its recommendation send the most unfortunate signals about the disinvestment process to foreign and domestic investors who are the potential purchasers of divested equity.” More generally, over the last decade the government has consistently fallen short of disinvestment targets by significant proportions. The average target achievement rate every year has been an abysmal 10 percent, barring 1998–99, when the target was met, largely due to the creation of crossholdings among various PSUs. It is the third category of privatization—where there may be natural monopoly—that is the most difficult. Ironically, it is this area in which India appears to have concentrated most effort to date. A number of questions can be raised as to the reasons why these (at least partial) “natural monopolies” should have been subject to privatize first, or, more to the point, why privatization of the enterprises with built-in competition did not happen sooner. At present, however, partial private ownership has been established, and in some instances new entrants have been encouraged to enter and to compete with former monopolies. But a clear definition of the framework for these activities—such as which parts will be competitors and which natural monopolists, the criteria for regulation of natural monopolists and for setting their prices, the criteria for permitting specified percentages of foreign ownership, and so on—will certainly accelerate the rate at which infrastructure investments can increase. 1.5 Conclusions The Indian economy in the year 2000 is vastly different from what it was prior to the beginning of reforms in the early 1990s. Quite clearly, the reforms to date have improved the functioning of the economy and permitted a higher rate of economic growth than was regarded as attainable on a sustainable basis in earlier decades. However, economic growth at rapid rates requires continuing reform efforts: A policy framework that was adequate to permit 4–5 percent growth twenty years ago would no longer sustain that growth rate, much less a higher one. And the more progress is made with reform in some sectors, the greater the payoffs to reforms elsewhere. Thus, achieving further reforms will increase the payoffs from those reforms already undertaken, but simultaneously will increase the payoffs to be achieved by still further liberalization. This chapter has provided a brief overview of India’s development and economic policies relating to economic growth to date. It has also focused on some aspects of economic policy that are not covered by other chapters in this volume. However, the remaining chapters in this volume also stress crucial aspects and issues to be faced by policy makers. Reducing fiscal deficits, as is discussed by Srinivasan, is desirable in order to avoid another

44

Anne O. Krueger and Sajjid Chinoy

macroeconomic crisis along the lines of the crisis of 1991; but it is also desirable in order to free resources for more productive investments. Additional resources for investment, in turn, can make the challenges involved in further financial restructuring somewhat more tractable. Likewise, significant changes to the Reservation of Small-Scale Industries policies, as discussed by Rakesh Mohan, could have a significant impact on the efficiency with which resources are used throughout the economy: Some small firms would grow large enough to become significant exporters and, since small firms are often labor-intensive, their expansion would provide for greater employment growth. But more rapid growth of output and employment will also increase the returns to improving the quality and availability of education, as is discussed by Kochar and by Foster and Rosenzweig. All of these reforms will improve Indian economic performance and growth. In so doing, they will make investing in India, both by Indians and by foreigners, more attractive. However, as discussed in the chapter by Annalee Saxenian and the contributions of Naushad Forbes, Narayan Murthy, and Sandeep Raju and others, there are also many bureaucratic processes and red tape whose removal could improve economic efficiency, providing scope for yet further efficiency gains. Hence, all of the discussion in this volume focuses on areas of Indian economic policy reforms in which policy changes could have potentially sizable effects on growth and living standards. Even if all of these reforms, and more, were carried out and the Indian economy were to achieve an 8–10 percent growth rate over the next decade, there would still be room ten years hence for another conference considering further reforms in order to sustain and accelerate growth. The question is not whether further reforms will be on the agenda: It is whether reforms will take place quickly enough to accommodate an 8–10 percent annual rate of growth, or whether, instead, they will proceed at a sufficiently relaxed pace so that 5–6 percent growth will be the norm.

References Agrawal, Pradeep. 1997. Labour policy: Striking a balance. In India Development Report 1997, ed. Kirit Parikh, 155–66. Indira Gandhi Institute of Development Research, Bombay: Oxford University Press. Ahluwalia, Isher. 1985. Industrial growth in India. New Delhi: Oxford University Press. ———. 1991. Productivity and growth in Indian manufacturing. Oxford University Press, New Delhi. ———, and I. M. D. Little. 1998. India’s economic reforms and development: Essays for Manmohan Singh. Calcutta: Oxford University Press.

The Indian Economy in Global Context

45

Ahluwalia, Montek S. 1998. Infrastructure Development in India’s Reforms. In India’s economic reforms and development: Essays for Manmohan Singh, ed. Isher Ahluwalia and I. M. D. Little, 87–121. Calcutta: Oxford University Press. ———. 1999. Priorities for Economic Reforms. Text of the Thirteenth Jawaharlal Nehru Memorial IFFCO Lecture. New Delhi. Mimeograph. Bardhan, Pranab. 1984. The political economy of development in India. Oxford: Basil Blackwell. Bhagwati, Jagdish, and Padma Desai. 1970. India: Planning for industrialization. London: Oxford University Press for the Organization for Economic Cooperation and Development. Bhagwati, Jagdish, and T. N. Srinivasan. 1975. Foreign trade regimes and economic development: India. New York: Columbia University Press. Central Statistical Organization. 1967. Estimates of National Income. New Delhi: Government of India. Chakravarty, Sukhamoy. 1987. Development planning: The Indian experience. Oxford: Oxford University Press. Dreze, Jean and Amartya Sen. 1995. India: Economic development and social opportunity. New York: Oxford University Press. Government of India. 1996. India infrastructure report: Policy imperatives for growth and welfare. Report of expert group on the commercialization of infrastructure projects for the Ministry of Finance. New Delhi: Thomson Press. Indira Gandhi Institute of Development Research. 1997. India development report. New Delhi: Oxford University Press. International Monetary Fund. 1992. Exchange arrangements and exchange restrictions: Annual report. Washington, D.C.: International Monetary Fund. ———. Various issues. International financial statistics. Washington, D.C.: IMF. ———. 1999. International financial statistics yearbook. Washington, D.C.: IMF. Joshi, Vijay, and I. M. D. Little. 1994. India: Macroeconomics and Political Economy 1964–1991. Washington, D.C.: World Bank. ———. 1996. India’s Economic Reforms 1990–2001. Oxford: Clarendon Press. Ministry of Finance. 2000. Economic Survey 1999–2000. New Delhi: Government of India. Morawetz, David. 1977. Twenty-five years of economic development: 1950 to 1975. Washington, D.C.: World Bank. ———. 1981. Why the emperor’s new clothes are not made in Columbia. Washington, D.C.: Oxford University Press. NASSCOM-McKinsey. 1999. NASSCOM-McKinsey study: Indian I. T. strategies. New Delhi. Nehru, Jawarhalal. 1958. Toward freedom. Boston: Beacon Press. Organization for Economic Cooperation and Development (OECD) Development Center. 1967. Population of less developed countries. Paris: OECD. Planning Commission. 1997. Ninth Five Year Plan. New Delhi: Government of India. Reserve Bank of India. 1999. Annual report 1998–1999. Bombay. Srinivasan, T. N. 2000. Eight lectures on India’s economic reforms. New Delhi: Oxford University Press. Tendulkar, Suresh. 1998. Indian economic policy reforms and poverty: An assessment. In India’s economic reforms and development: Essays for Manmohan Singh, ed. Isher Ahluwalia and I. M. D. Little, 280–309. Calcutta: Oxford University Press. World Bank. 1977. World Bank atlas. World Bank. ———. 1983. World Development Report. World Bank. ———. 1998/9. World development report. World Bank. World Economic Forum. 1999. Global competitiveness report 1999. Geneva.

2 India’s Fiscal Situation Is a Crisis Ahead? T. N. Srinivasan

2.1 Introduction India has been among the fastest-growing economies of the world in the last two decades. According to the World Bank (1999, table 11), during the 1980s, prior to the severe balance-of-payments and macroeconomic crisis of 1991, India’s gross domestic product (GDP) growth rate accelerated to an average of 5.8 percent per year, a rate that was exceeded by only 9 out of 123 countries. Growth in GDP resumed after 1991–92, the year of stabilization, and has averaged 6.1 percent per year in the 1990s. Again, this growth rate was exceeded in relatively few (specifically 19 out of 137) countries.1 The economy weathered the Asian crisis admirably, the monetary authorities having handled the situation very well. The macroeconomic picture is comforting, with growth expected to be around 6 percent in 1999–2000 and likely to accelerate in 2000–01. Inflation has moderated, current account deficit is around 1 percent of GDP, private capital inflows have recovered since the Asian crisis, and industrial growth has accelerated. T. N. Srinivasan is the Samuel C. Park, Jr. Professor of Economics at Yale University and nonresident senior fellow at the Center for Research on Economic Development and Policy Reform at Stanford University. The author would like to thank Anne O. Krueger; discussants N. K. Singh; Shankar Acharya, and Kenneth Kletzer; Ronald McKinnon; Gian Sahota; Patricia Reynolds; and the participants of the conference on Indian Economic Prospects: Advancing Reform for their comments on an earlier version of this chapter. I alone am responsible for any errors that remain. 1. Among countries that grew faster than India, many were in East and Southeast Asia. Thus, India’s growth performance relative to its neighbours in Asia is not that impressive. It is also the case that the growth rate has to be still higher if poverty is to be eradicated within a not–too–distant time horizon.

47

48

T. N. Srinivasan

Short-term external debt, which accounted for slightly more than 10 percent of total debt in 1990–91 and was four times the stock of external reserves, was only around 4.4 percent of total debt and, even more remarkably, less than a sixth of the stock of external reserves in 1998–99 (World Bank 2000, table A3.1a). Debt service as a proportion of current account receipts has dropped from more than one-third to less than one-fifth during the same period. The major disquieting element in this otherwise bright macroeconomic picture is the fiscal imbalance. Until the early 1980s, India’s macroeconomic policies were conservative. Current revenues of the central government exceeded current expenditures, so that a surplus was available to finance in part the deficit on capital account, a deficit that is normal for a developing country. In the early 1980s, fiscal prudence was abandoned—with the consequence that current revenue surpluses turned into deficits. This meant that the government had to borrow at home and abroad to finance not only its investment, as would normally be the case in a developing country, but also its current consumption. Fiscal deficits, as published in government budget documents, have tended in the past to understate the real imbalances.2 The reason is that the rates of interest, at which the government appropriated a large share of the loanable resources of the banking system through statutory liquidity ratio (38.5 percent maximum) and cash reserve ratio (15 percent maximum), were administratively set below what would have been market clearing levels. Also, at least in the early years, external borrowing was largely on concessional terms for multilateral lending institutions and from bilateral government to government external aid transactions. As the 1980s wore on, the government also resorted to borrowing from abroad on commercial terms, from both the capital market and nonresident Indians. In 1980–81, out of $18.3 billion of public and publicly guaranteed external debt, $2.0 billion was owed to private creditors (World Bank 1990, table 4.1). On the eve of the macroeconomic crisis in 1990–91, external debt had nearly quadrupled to $71.1 billion, of which $23.0 billion was owed to private creditors (World Bank 2000, table 3.1a). Debt to external private creditors grew elevenfold in ten years. What was left of the gross fiscal deficit—after domestic and external borrowing, small saving, and provi2. The overall fiscal deficit of the central and state governments is an indicator of borrowing by the government. It is not necessarily a good indicator of a broader measure of fiscal balance based on the difference between what resources the government is able to generate through means that cause the least distortion (i.e., welfare loss to the rest of the economy) and what are needed to sustain its appropriately defined role in the economy. For example, Sahota (2000) argues that by not taking into account tax expenditures (e.g., revenue loss from tax waivers and concessions) and transfers that cannot be deemed socially desirable, conventionally measured fiscal deficits understate the fiscal imbalance. Rao and Nath (2000) point out that reducing fiscal deficits by reducing socially desirable capital expenditure would weaken broader fiscal balance.

India’s Fiscal Situation: Is a Crisis Ahead?

49

dent funds—was monetized through the ad hoc sale of treasury bills to the Reserve Bank of India. For example, in 1988–89 and 1989–90, before the crisis year of 1991, the gross fiscal deficits of the center and states together were 356.7 and 452.0 billion ruppes (Rs.), respectively, and nearly Rs. 62.4 billion and 109.1 billion, respectively, of the deficits, were financed by treasury bills (World Bank 2000, table A4.4). Domestic debt held outside the banking system grew sixfold, from Rs. 290 billion to Rs. 1,752 billion in the decade 1980–81 to 1990–91. Debt held by the banking system grew roughly at the same rate, from Rs. 257 billion to Rs. 1,419 billion during the same period (World Bank 2000, table A4.14). Clearly, the reckless fiscal expansionism of the 1980s was unsustainable. Accompanied by certain liberalizing changes, however, it did generate growth. For a few industries these changes simply involved delicensing; others were allowed more flexibility in the use of their own capacities, thanks to the relaxing of some import restrictions and the permitting of changes in their product mixes within the licensed capacity (under so-called broad banding). Indeed, an industrial mini-boom took place from 1985 to 1988. The average annual growth rates of real GDP under the sixth and seventh plans (which covered the 1980s) were 5.5 and 5.8 percent, respectively— much higher than the Hindu growth rate of 3.5 percent during the previous three decades. The 1980s also covered the period of a steep reduction in the proportion of poor in India’s population—from 48.36 percent in 1977–78 to 34.07 percent in 1989–90 (World Bank 2000, annex table 1.1). Needless to say, reduction in poverty achieved during a period of unsustainable debt-led growth could not have lasted. When the macroeconomic crisis hit in 1990–91, the gross fiscal deficit of the central and state governments had grown to 9 percent of GDP at market prices. If one includes the losses of the nonfinancial public-sector enterprises and the oil pool balance, the consolidated public-sector deficit stood at around 10.9 percent of GDP in 1990–91. Nearly two-fifths of this deficit, or 4.3 percent of GDP, was for interest payments on domestic and external debt (World Bank 2000, annex table 8.6). An analysis by Buiter and Patel (1992) showed that unless corrective steps were taken, India faced fiscal insolvency.3 It is no surprise, therefore, that one of the major objectives of (then) finance minister Manmohan Singh’s reforms of 1991 was to reduce the central government’s fiscal deficit from 7.7 percent of GDP in 1990–91 to 3. The conventional solvency criterion requires that the outstanding debt at any time not exceed the expected present value of future primary surpluses, using a discount rate that is the difference between the rate of interest on public debate and the growth rate of real GGDP. Although in a growing economy debt-GDP ratio could grow indefinitely without violating the conventional solvency criterion, this would require that the ratio of primary surplus to GDP grow as well. Kletzer (see comment section at end of chapter) notes that if the primary surplus is based on distortionary taxation, this may not be feasible.

50

T. N. Srinivasan

around 4.0 percent or lower in three years or so. In fact, he did achieve a significant reduction to 5.9 percent in 1991–92, his first full year as finance minister, and to 5.5 percent in 1992–93. However, it ballooned to 6.9 percent in 1993–94. The deficit was estimated at 6.5 percent in 1998–99 (World Bank 2000, figure 1 and annex table 8.6).4 A new definition of the center’s fiscal deficit was introduced on 1 April 1999, which excluded the state’s share of small saving from the center’s expenditure. Under the new definition, there was a peak fiscal deficit of 6.4 percent in 1993–94 and a deficit of 5.0 percent in 1998–99 (Government of India 2000, table 2.1). With the unexpected conflict in Kargil and the associated increase in spending, the hope of reducing the budget figure to 4.1 percent for 1999–2000 receded. The realized figure is likely to be 5.6 percent, according to the budget data for 2000–01.5 The budgeted figure for 2000–01 is 5.1 percent of GDP.6 Neither the government’s annual Economic Survey, nor the budget document, estimates the deficit of the nonfinancial public sector. If we adjust the World Bank’s (old-definition) estimate of the budgeted deficit for 1999–2000 by the 1.6 percent slippage in the center’s fiscal deficit, the overall deficit for the nonfinancial public sector in 1999–2000 is likely to have been 10.8 percent of GDP, or about the same as in 1990–91 just prior to the crisis and reforms! Among countries with over 20 million in population, the Indian central government’s average fiscal deficit during the 1987–97 decade was exceeded 4. The figures of fiscal deficit as a percentage of GDP in the publications of Government of India, International Monetary Fund, and the World Bank differ for at least three reasons: (1) whether the government’s new definition of the deficit is adopted (this affects only the central government’s deficit and not the total deficit of Center and states together); (2) whether the GDP for years prior to 1993–94 are changed to account for the difference between the old series GDP with 1980–81 as base and the new series with 1993–94 as base; (3) whether proceeds from disinvestment are included in current receipts; and (4) whether or not the entire profits of the Reserve Bank of India are transferred to the central budget. Rao and Nath (2000, Table 1) provide a comparable series of fiscal deficits for the central government and find that their adjusted fiscal deficit was 6.8 percent of GDP in 1990–91 and 5.71 percent in 1998–99. Although for these reasons the levels of the fiscal deficit as published by various sources naturally differ, their time trends appear to be very similar. 5. Reserve Bank of India (2000) reports a figure of 9.9 percent of GDP for the gross fiscal deficits of central and state governments. 6. Acharya (see comment) notes that, despite occasional slippage, the fiscal situation was in fact improving until 1996–97. It has been steadily deteriorating since then. He attributes the deterioration primarily to the increase in central and state government expenditures on wages and salaries following the implementation of the recommendations of the Fifth Pay Commission. Although these increases undoubtedly contributed to the worsening situation, Acharya is wrong in viewing them as unanticipated and exogenous shocks. Even if the size of the recommended increase could not have been fully anticipated, there was no doubt that some increase would be recommended. In fact, the actual increase granted by the government was even more generous than that recommended by the commission. A government concerned about the likely impact of a pay increase on the fiscal deficit would have planned for reductions in other expenditures in order to absorb any anticipated pay increase, and, in any case, would not have been more generous than the commission itself.

India’s Fiscal Situation: Is a Crisis Ahead?

51

by the deficits of only three countries: Nigeria, Pakistan, and Brazil (World Bank 2000, figure 8.3). There is no doubt that most economists would deem India’s fiscal deficit unsustainable. In fact, the Reserve Bank of India has warned that “while the high level of government sector deficit is attributable to some unavoidable expenditure commitments as well as unanticipated shocks, any further erosion of the fiscal position could turn out to be unsustainable” (Reserve Bank of India 2000, part 7). Studies of the experience of a cross-section of countries suggest that large public-sector deficits reduce growth by crowding out productive private investment. Monetization of the deficit could be inflationary. Inflation tax is not obviously legislated, and it is also the most regressive in that it most hurts the poor, who have no way of hedging against inflation. On the other hand, attempts to reduce monetization of the deficit through domestic borrowing will raise interest rates and increase the cost of investment. Some studies (e.g., World Bank 2000, 114) suggest that in India, during the eleven-year period since 1986–87, an increase in the central government’s fiscal deficit (inclusive of oil pool deficit) by one percent of GDP was associated with a reduction in private corporate investment by one percent of GDP.7 Although correlation is by no means causation, and other factors could also have induced a reduction in corporate investment, one cannot rule out the possibility that fiscal deficits crowded out private investment. There are activities that are socially important and for which there are no viable private-sector substitutes for the government’s involvement. A strong case can be made for increasing expenditures on such activities by redirecting public expenditures away from other activities that are better left to the private sector, even in the absence of a fiscal deficit. When deficits are financed by increasing public debt, a potential debt trap can arise if the nominal interest rate on debt exceeds the rate of growth of nominal GDP, because debt service payments would outstrip output growth. The Reserve Bank of India has been drawing attention to this possibility, most recently in its annual report for 1999–2000, wherein it said, “the debt growth has generally exceeded the nominal GDP growth since 1997–98 with an exception to 1998–99. As high levels of public debt have deleterious effects on macroeconomic stability, the need for reducing the fiscal deficit and debt to sustainable levels is widely felt” (Reserve Bank of India 2000, part four). In a recent econometric analysis of this issue, Olekalns and Cashin (2000) come to the sobering conclusion that current fiscal policies are unlikely to be sustainable in the long run. Although the average interest rate on gov7. Sahota, in a private communication, suggests that to the extent reduction in deficits releases resources for investment in infrastructure, private investment may go up even further; i.e., the elasticity of private investment with respect to fiscal deficit would rise above one.

52

T. N. Srinivasan

ernment debt has been below the rate of growth of nominal GDP, so that new borrowing undertaken to service existing debt has not raised the debtGDP ratio, this favorable differential between interest rate and GDP growth is unlikely to last forever. Because the large fiscal deficit of the last decade has apparently neither raised inflationary pressures nor increased current account deficits, some argue that the fiscal deficit has not adversely affected growth or the reform process.8 In their view, the concern about fiscal deficit has been excessive and has prevented an expansion, as in the 1980s, that would have triggered faster growth and poverty reduction. For example, Chandrasekhar (2000) argues that, first, the deficit is not now being monetized as it was in the past; and second, even if it had been, it would be Inflationary only if the system is at full employment or is characterised by supply bottlenecks in certain sectors. The fact of the matter is, not only is the industrial sector burdened with excess capacity at present, but the government is burdened with excess foodstocks and foreign exchange reserves. . . . Since inflation is already at an all time low, this provides a strong basis for an expansionary fiscal stance, financed if necessary with borrowing from the central bank. To summarise, in the current context a monetised deficit is not only non-inflationary, but virtuous from the point of view of growth. (1141) In his view, reforms that resulted in the government’s borrowing at market rates and providing tax concessions for private saving and investment explain the failure of the government to rein in fiscal deficits. He concludes, “In sum, the evidence suggests that the fiscal effects of economic reform have put the government in a state of paralysis with respect to triggering growth and reducing poverty, even though current circumstances offer an opportunity for major advances in these areas” (1142). This line of reasoning implies that (a) financial repression is beneficial; (b) the reason for excess capacity is lack of domestic demand; (c) expansionary fiscal policy would result in its utilization and generate growth; and (d) the most productive use of stocks of food grains and foreign exchange reserves is to support such fiscal expansion. Each of these implications is questionable. First, even Joseph Stiglitz, who is sympathetic to mild financial repression, has not argued in favor of enabling the government to borrow at administered low interest rates is appropriate, regardless of the use 8. Acharya is intrigued by the fact that GDP grew at 6 percent to 7 percent per year in the 1990s despite large fiscal deficits. There is nothing surprising or intriguing about this. Clearly, to the extent that there is unutilized capacity, fiscal expansion can generate growth as it did in the 1980s without a step-up in investment and without major reforms. Equally, efficiency gains from reforms can also generate growth without a substantial increase in the investment rate, as I believe was the case in the 1990s. However, in both instances the spurt in growth is most unlikely to be sustainable: Sooner or later the excess capacity would disappear, further costless efficiency gains would be exhausted, and investment rates would have to be increased to sustain growth. Then fiscal deficits would begin to bind, as they would crowd out investment.

India’s Fiscal Situation: Is a Crisis Ahead?

53

to which the borrowed funds are put.9 Second, viewing domestic demand as a constraint on capacity utilization reflects a closed-economy mindset of the past. Third, there are options other than supporting fiscal expansion for the use of stocks of food grains (e.g., exporting them) and of foreign exchange reserves (e.g., reducing the stock of external debt). Without examining what the appropriate level of food or foreign exchange stocks would be in order to reduce volatility in food prices and contain any sudden outflow of short-term capital, the assertion that such stocks could have been used to support fiscal expansion has no analytical basis. The most critical issue facing the government at all levels is the unsustainability of the fiscal situation. In what follows, I will explore some of its contributory causes and possible approaches to restoring sustainability.10 These include state finances (section 2.2), tax and expenditure reform (section 2.3), reduction of subsidies (section 2.4), and disinvestment (section 2.5). section 2.6 concludes. 2.2 State Finances The fiscal deficit of the states fell from 3.2 percent of GDP in 1990–91 (just before reforms) to a low of 2.3 percent of GDP in 1993–94, and has since risen to more than 4.3 percent of GDP in 1998–99 (World Bank 2000, table 3.5 and annex table 3.1; Business India, 3–16 April 2000). The revised estimates for 1999–2000 show a deficit of 4.8 percent of GDP (Reserve Bank of India 2000, part four). There is little doubt that the fiscal deficits of several states have become unsustainable. The average fiscal deficit during 1991–97, as a percentage of state domestic product (SDP) among fourteen major states, varied from 2.5 percent in the high-income state of Maharashtra to 5.9 percent in the poor state of Orissa. Correspondingly, average outstanding debt varied from 13.4 percent of SDP in Maharashtra to 41.2 percent in Orissa. In fact, in all four of the low-income states of Bihar, Orissa, Rajasthan, and Uttar Pradesh, average fiscal deficit and outstanding debt during 1991–97 exceeded 4.1 percent and 29.2 percent of GDP, respectively. The emerging fiscal crisis in the states has caught media attention—the magazine India Today, in its 14 February 2000 issue, headlined 9. Kletzer suggests that financial repression is, in part, a consequence of the Indian government’s reliance on a closed domestic capital market for financing public expenditures. He notes that the large spread between lending and deposit rates is a tax on financial intermediation. Although the differential between the interest rates on bank loans and public debt has shrunk in the 1990s, and the banks hold no more government paper than they are required to hold, Kletzer is right in rejecting the conclusion that these facts reflect a convergence of riskadjusted returns on private and public debt, for the reason that most banks are publicly owned, and thus the portfolio decisions of their managers are not necessarily market driven. 10. There have been several analyses of India’s fiscal situation, notably by Joshi and Little (1994, 1996a, b); Buiter and Patel (1992, 1996, 1997); Burgess, Howes, and Stern (1993, 1994); Olekalns and Cashin (2000); and, above all, by the Chelliah Committee (Ministry of Finance 1992) on tax reforms.

54

T. N. Srinivasan

the crisis on its cover page with these ominous words: “States Going Broke: Bankruptcy Stalking States, Threatening a Collapse of Public Services.” Although sensationalism is to be expected from the news media, the seriousness of the situation has dawned even on policy makers, leading to eleven states’ signing Memoranda of Understanding (MOUs) with the central government in 1999–2000 promising fiscal reforms in return for bailouts from the center in the form of ways and means advances from tax transfers and grants due to them. These are the domestic counterparts of loans from multilateral agencies conditioned on structural reforms, that is, policy-based lending. Although one year is too short a time in which to evaluate the MOUs, the experience is not very encouraging—transfer payments had to be withheld for noncompliance on occasion, and part of the advances had to be converted into three-year loans. The states have been allowed to increase their market borrowing as well. It is unlikely that there will be any further advances under the MOU process in 2000–01. The current precarious fiscal situation of states is the result of increasing implicit and explicit subsidies on goods and services supplied by the government (electricity, irrigation water, transport, education, health) with virtually no attempt to raise revenues to finance them. States also competed with each other in offering tax and other fiscally costly incentives in an attempt to attract investment. Overstaffing of administration and public enterprises was the norm in all states. Reserve Bank of India (2000, part 7) rightly emphasizes that state public enterprises, like the state electricity boards and state road transport corporations, have been reporting losses and absorbing scarce funds through budgetary support, and thus contributed to the growth in the implicit or contingent liabilities of state governments.11 Salary increases awarded to the staff of all state and state-supported institutions, following similar awards by the central government to its employees, have increased the already high share of employee compensation in state revenues. For example, in the high-income state of Maharashtra, the wage bill accounted for more than 70 percent of tax revenue (India Today, 14 February 2000). Business India (3–16 April) reports that in states such as Uttar Pradesh, salaries make up almost 89 percent of total non-plan expenditures. In Madhya Pradesh, nearly 75 percent of the increase in non-plan current expenditures between 1999 and 2000 is meant for “dearness” allowances (compensation for price inflation) for the state bureaucracy. Debt service is another major expenditure— 11. Acharya is right in emphasizing the seriousness of the situation. He points out that politicians believe that the role of the state is to provide jobs, and not services. As a consequence, the state governments spend on little other than wages and salaries, and thus the already inadequate provision of public goods and services such as education, health, and roads is further eroded. Moreover, the labour market is severely distorted by non-market-determined high wages in the public sector. Such high wages make growth less employment-intensive than it would otherwise have been.

India’s Fiscal Situation: Is a Crisis Ahead?

55

interest payments alone, at 2 percent of GDP, amounted to 35 percent of states’ own tax revenue and 15 percent of states’ revenue receipts from all sources. The constitution has assigned exclusive responsibility to the states for most matters concerned with the life and welfare of the population, including public health, infrastructure, agriculture, and land and water management.12 However, with a large part of the states’ fiscal resources increasingly being spent on nondevelopmental activities, expenditure (revenue and capital) on developmental activities has remained static with significant yearto-year fluctuations. What is worse, developmental capital expenditure has been declining, and there is no evidence of any improvement in the efficiency of capital use, which, had it occurred, could have offset—at least in part—any decline in investment. One cannot underestimate the deleterious consequences of the precarious condition of the states’ fiscal health, both for poverty alleviation and for reaping the benefits of economic reforms already instituted, let alone implementing further reforms. The constitution empowers states to tax land, agricultural income, and sales, and imposes excise taxes on alcohol; it empowers the center to collect taxes on personal and corporate income, wealth, and foreign trade, as well as excise taxes. It also requires the center to share the revenue from certain taxes in proportions recommended by the Finance Commission, which the president is constitutionally required to appoint every five years. In addition to the transfers recommended by the Finance Commission, the Planning Commission, a nonstatutory body established by a resolution of Parliament, makes transfers for financing approved outlays of the states on their annual development plans. Although some legal scholars have questioned the constitutionality of transfers made by the Planning Commission, I am not aware of any supreme court decision on the issue. The sum of the states’ share in central taxes and grants from the center has accounted for around 40 percent of the total revenue receipts of the states during 1990–91 to 1997–98 (World Bank 2000, annex table 8.8). The center has the responsibility to provide crucial public goods, such as defense and a common currency. In addition, given the diversity in per capita incomes and other indicators of development of the states, the center has a redistributive role in making transfers to the poorer states. The states have been assigned large responsibilities in crucial sectors such as education, health, irrigation, and other investment in agricultural development. Naturally, they cannot discharge these responsibilities adequately if they cannot generate the required resources. The transfers from the center to the states are meant in part to bridge the gap between the resources states 12. There are several excellent studies on state finances and fiscal federalism of the Indian Constitution (see in particular Rao and Singh [1999a, b, c, d], Bajpai and Sachs [1999]). The report of the World Bank (2000) also discusses the issues in some detail.

56

T. N. Srinivasan

require to meet their assigned responsibilities and the resources they can raise themselves. In practice, the tax transfer system has had some disincentive and efficiency-reducing effects. The center’s efforts to collect taxes whose revenues have to be shared with the states, such as income tax, are likely to have been less vigorous than their efforts to collect taxes that accrue entirely to the center, such as customs. However, the recently enacted amendment to the Constitution requires that 29 percent of the net proceeds of all taxes be transferred to the states. This removes the incentive for the center to raise and vigorously collect those taxes whose entire proceeds accrue to the center. Similarly, if the transfers received from the center form a large part of its expenditure, and these transfers have no relationship to its own efforts to raise resources, clearly a state is unlikely to be diligent in raising resources. The approach of the first (1951–56) to eighth (1985–90) finance commissions was to fill the gap between the states’ revenues and expenditures, and to completely ignore this incentive effect. The eleventh finance commission for 2000–05, appointed in July 1998, has been given broad terms of reference with which to review the state of finances of the union and the states, and to suggest ways and means to restore budgetary balance and macroeconomic stability. These terms were augmented in April 2000 to ask the commission to draw a monitorable fiscal reforms program and recommend ways to link grants to states to their progress in implementing the program. Several states have apparently objected to this linkage, with one state chief minister comparing any link between devolution and fiscal performance with IMF conditionalities! The commission submitted its report in June 2000 on its initial term of reference, and a report on its augmented terms of reference is expected at the end of August 2000. The report of June 2000 goes part of the way in adopting a normative approach, but until its final report is received, it is hard to judge how firm and enforceable its recommendations for linking performance on implementing fiscal reform with transfers would be.13 The states have the right to impose sales taxes, a right that extends to sales to buyers in other states. Obviously, an interstate sales tax is a tax on the exports of one state to another. Such taxes prevent India from being a common market and consumers from reaping the static and dynamic efficiency gains of a large market. Unless a state has a national monopoly over a com13. Kletzer argues that linking policy reforms of states and central transfers to them runs the risk of time inconsistency. Since the transfers in large part finance mandated social expenditures by states, the center’s commitment to withhold transfers in case a state fails to institute policy changes in return for transfers may not be credible. Moreover, the lack of credibility of the threat not to bail out states if they run excessive deficits will lead the states to accumulate more and more deficits (which become contingent liabilities of the central government) in a timeconsistent equilibrium. However, it is possible, if the center’s commitment is the result of an agreement between all the states and the center, the credibility problem could be mitigated, because other states will have an incentive to side with the center if any individual state deviates.

India’s Fiscal Situation: Is a Crisis Ahead?

57

modity that is competitively produced within its border, a case that is extremely unlikely to arise, it hurts itself by taxing its exports to other states. Even when a state has monopoly power, its exercise will be at the expense of consumers in other states, restricting commerce and inviting retaliation by other states. In the end, consumers in all states are likely to lose. Fortunately, the states have come to a unanimous decision to impose a uniform sales tax throughout the country from 1 January 2000. This is a step in the right direction. Recent reforms induced states to compete with each other to attract private investment by offering tax concessions. Such competition might end up transferring resources from taxpayers to investors without affecting their location decision. Fortunately, the states have decided not to compete with tax concessions. The states are constitutionally barred from borrowing in international financial markets and need the center’s consent for any borrowing in the domestic market if they are indebted to the center. Since all states are indebted to the center, this constraint is binding on all of them. The states share in the central government’s borrowing from captive sources of finance such as banks, insurance companies, and non-government pension and provident funds, which are required to be invested in designated government securities. Of course, the states cannot directly monetize any part of their deficits. It would seem that these restrictions impose a hard budget constraint on states. Moreover, if “hard” budget constraints on individual states in the United States are the reason that government expenditure is restrained there, as is often argued, and if looser budget constraints on Canadian provinces explain poorer fiscal discipline in Canada, then in India one should see greater fiscal discipline in the states. However, this is obviously not the case. In fact, the states have succeeded in circumventing ostensibly hard budget constraints by diverting resources meant for investments to current expenditure, and through indirect borrowing by running up arrears with central public sector enterprise (PSEs). For example, state-owned electricity boards (SEBs) have been tardy in paying their dues to Coal India and National Thermal Power Corporation (NTP), which are owned by the center. This, in turn, has led the center and central PSEs to impose restrictions— Coal India has instituted a cash and carry policy, and central government has chosen to retain up to 15 percent of the transfers due to the states to clear the SEBs’ arrears with NTP. Prior to 1994–95, each state PSE was set separate borrowing limits for each year. With the removal of these limits, state guarantees given to borrowing by state PSEs have become an easy way for states to circumvent the ceiling on borrowing imposed by the central government. The volume of state-guaranteed loans to state PSEs has been increasing at 12 percent a year recently (World Bank 2000, box 3.1, and 33). The perverse incentives created by the planning commission in financing

58

T. N. Srinivasan

the capital and operating costs of new (centrally sponsored) state projects in each five-year plan for the duration of that plan should be noted. States are thus encouraged to attract such projects, but once the center ceases to pay for the operating costs, their incentives to run the projects wane. Also, since all states pay the same rate of interest on loans from the central government, they feel no market pressure to maintain credit worthiness. Under the present system, the states have established a large infrastructure and spent on social services, including many populist programs without adequate financing through taxes or recovery of costs from users of state-provided goods and services. In the context of electoral politics as practiced in India, populism is attractive to politicians of all shades. Supply of electricity free of charge or at a heavily subsidized price to farmers is one of many populist policies. The gross subsidy on account of sale of electricity in 1999–2000 was estimated at Rs. 338 billion, of which three-fourths was from sales to the agricultural sector. The rate of return on invested capital in state electricity enterprises was –31 percent! If the state electricity boards were to increase prices just enough to raise revenues to the equivalent of a 3 percent rate of return on capital that Section 59 of the Electricity Supply Act of 1949 enjoined them to do, 70 percent of the gross subsidy would not be needed. If they were to increase just Rs. 0.50 per kilowatt hour to agricultural users, as the chief ministers of states had once agreed to do, gross subsidy would go down by 10 percent (Government of India 2000, table 9.4). Other state pricing policies are also damaging. Charging reasonable tuition fees at state-supported institutions of higher learning would help. Many states have not revised water rates for irrigation in over three decades. The current charges do not even cover the costs of operation and maintenance. These problems of inadequate cost recovery are massive by any reckoning. As noted earlier, state payrolls have expanded to the point that a substantial part of their revenues are spent on wages and salaries alone. The states had little choice but to follow the center’s implementation of the recommendations of the fifth pay commission. This has accentuated their already high wage and salary expenditures. The economic reforms of 1991 have radically altered the options open to states: They can now compete for private investment in infrastructure sectors that were formerly financed entirely by the government. However, states will be unable to attract private capital without good quality infrastructure, particularly power and roads, an educated labour force, and efficient, business-friendly bureaucracy. Improvement in the quality and an increase in the quantity of infrastructures services depend on reforms of the operation, pricing, and regulation of the PSEs, as well as on additional investment that would require substantial resources. Both for sending the appropriate signals to users and for raising resources for investment, states will have to address pricing issues with respect to electricity, irrigation, water, education, and other goods and services provided by the government.

India’s Fiscal Situation: Is a Crisis Ahead?

59

There is a lot to be said for the view that the center’s roles should be confined to national defense, external relations, the maintenance of a common currency and of national networks of communications and transport, and the assurance that there is a single—that is, national—common market. To these I would also add a relatively circumscribed role of redistribution across states. All other activities could be left to the states. Such a setup would approximate the market preserving federalism (MPF) of Weingast (1995). Under MPF, the central government would ensure that goods and factors are mobile across states, limited revenue sharing exists among levels of government, and access to capital markets is limited for all levels of government. Under these assumptions, the ability of any level of government to tax and spend unwisely would be severely limited, since factors would flee any state that did so. However, adhering to other requirements of MPF in the absence of mobility of factors across states might accentuate the existing inter-state disparities in income and poverty as well as lower public expenditures on rural development and poverty alleviation. Alternatively, it might encourage states to compete to provide an environment that is most conducive to economic and social development. I am optimistic that the latter situation would occur and that interstate competition in tax rates and in the provision and pricing of infrastructures services will in fact be a race to the top rather than a race to the bottom. My optimism is based on the following, admittedly scanty, evidence.14 First, there is now much greater mobility of factors, even of labor, and it is increasing. Second, thanks to the spread of communications, awareness among the general public of the failure of some states and, the success of other states in the performance of their social sectors is also increasing. Consequently, politicians in failing states would have much explaining to do. In any case, any move in this direction would require a major overhaul of the constitution. The recently appointed Constitutional Review Committee should look carefully into whether the provisions adopted five decades ago relating to the roles, powers and resources assigned to the centre and states need revision in the contemporary context. The committee ought to consider whether the planning commission should be abolished and replaced by two institutions analogous to International Bank For Reconstruction and Development (World Bank) and International Development Association (IDA). The former will provide “hard” loans to relatively richer states after a thorough appraisal of the projects to be financed by loans, and the latter “soft” loans to relatively poorer ones, on the basis of criteria similar to those used by the two Washington institutions. Let me hasten to add that although such a change would help to improve efficiency in the use of scarce resources at the disposal of governments at all levels, it 14. Kletzer, who expresses some skepticism about my optimism, might not find these arguments persuasive. He might point to the reported statement of the inimitable Laloo Prasad Yadav, the former chief minister of the abysmally poor and poorly governed state of Bihar: “Development does not buy votes: Empowerment does!”

60

T. N. Srinivasan

is not necessarily the most efficient strategy. After all, it shares the fundamental fault of IDA: That is, it would alleviate the resource scarcity of poorer states through the provision of subsidized loans. As is well known, resource transfers that are “tied,” in this case to investing in projects financed by subsidized loans, are less efficient than untied transfers. 2.3 Tax and Expenditure Reform In the decade of the 1990s gross (i.e., gross of states’ share) tax revenues of the central government have declined from about 10.1 percent of GDP in 1991–92 to 8.7 percent in the budget estimates of 1999–2000 (World Bank 2000, annex table 8.5). However, the ratio of direct taxes to GDP has increased. As noted earlier, the situation of the states is not much better— revenues from their own taxes have stagnated at about 7.8 percent of GDP (World Bank 2000, annex table 8.8). Substantial reductions in tariff rates, and some reduction and rationalization of excise rates, explain much of the reduction in the sum of revenues from customs duties and excise taxes from 7.5 percent of GDP in 1991–92 to 5.7 percent of GDP in 1999–2000. It is also the case that there were reductions in personal income tax rates. There is, however, some evidence that buoyancy of gross tax revenue has gone down since the mid-1990s (World Bank 2000, annex table 4.12). The revenue losses from tax concessions and exemptions are also significant (World Bank 2000, annex table 4.11). With fiscal deficits continuing to be large, the need for raising tax revenue as a proportion of GDP is evident. The task of reforming the complex set of taxes, direct and indirect, at the center and in the states is a challenging and daunting task. The contours of needed reforms are well known and have been discussed in several studies, in particular by the Chelliah Committee (Ministry of Finance 1992). Some measures were announced in the central budget for 1998–99 to expand the tax base by identifying potential taxpayers through several presumptive criteria. At the same time, the budget raised income tax exemption limits, which are already high relative to per capita income. While there may be good administrative reasons for this step, nonetheless it has the undesirable consequence of reducing the tax base. The 1999–2000 budget reduced the multiplicity of tax rates, particularly tariffs, rationalized the rate structure, and reduced the number of excise duty rates from eleven to three so that the system is much closer to that of value added tax (VAT) at a single rate.15 That budget also phased out the tax exemption of export income. The policy statement of March 2000 on exports and imports replaced half of India’s 15. Acharya comments that although in his two budgets, Finance Minister Yashwant Sinha has moved the excise tax structure closer to a single-rate central value-added tax (CENVAT), problems remain because of continued exemptions for small-scale industry and many products. In addition, CENVAT is confined to the manufacturing end because of the constitutional constraints on taxes that the center can levy on domestic trade.

India’s Fiscal Situation: Is a Crisis Ahead?

61

remaining quantitative restrictions (QRs) on imports by tariffs, with the remaining QRs to be eliminated by April 2001. The budget for 2000–01 cut subsidies on food and fertilizers. However, the increased revenues from changes in taxes and subsidies are offset by a significant increase in defense expenditure. Moreover, the implementation of the recommendation of the eleventh finance commission will involve increases in states’ share of central tax collections and central grants to states. The three important sectors that are yet to be fully integrated into the tax base are agriculture, small-scale industry, and services. Land tax as a source of revenue has virtually disappeared, and agricultural incomes remain largely outside the tax net. Additionally, replacing the existing set of indirect taxes through a system of value added taxes (VATs) has not made rapid progress. Issues such as whether, by amending the constitution if necessary, the center should levy the VAT and share the revenues with the states, or whether the VAT should be completely in state hands with each state choosing its own rates, coverage and exemptions, are yet to be discussed extensively, let alone resolved. However, the states have agreed to implement a VAT from 1 April 2001, with the central government pledging to compensate states for any revenue losses during initial years of implementation. Almost half of the central government’s expenditures are accounted for by interest payments on public debt, defense outlays and transfers to state governments. A significant part of the fiscal correction in the early 1990s was through reductions in development spending, particularly capital spending, which could jeopardize growth. Unfortunately, the government chose not to accept the fifth pay commission’s recommendation to reduce government employment by 30 percent over ten years, while being even more generous than the commission’s recommendations in increasing the wages and salaries of its employees. Because what the central government pays its employees inevitably extends to institutions financed by the central government and central public-sector enterprises, as well as to corresponding categories in the states, the central government’s decisions on the pay commission recommendations have been costly. Tax expenditures (i.e., tax exemptions and concessions) are ubiquitous at the central and state levels. The revenue loss from these exemptions and concessions is not negligible. To the best of my knowledge, however, there has been no serious analysis of the social rationale for such tax expenditures. The cost of explicit and implicit subsidies is taken up in the next section. The decision to establish an expenditure reforms commission, the introduction of zero-based budgeting, and the possibility that a fiscal responsibility act will be passed by parliament are hopeful signs. It is clear that greater and faster progress toward needed fiscal correction can be made if possibilities on both revenue and expenditure sides of the budget are explored. These naturally include, besides an increase in the tax-GDP ratio, a

62

T. N. Srinivasan

substantial reduction in implicit and explicit subsidies, as well as more aggressive privatization. 2.4 Reduction of Subsidies Explicit subsidies of the central government accounted for around 1.1 percent to 2.1 percent of GDP during the 1990s (World Bank 2000, annex table 8.5). Major subsidies of the central government are budgeted to absorb 22 percent of non-interest-non-plan revenue expenditure and 12 percent of net revenue receipts of the central government (Government of India 2000, paragraph 2.17). State governments also subsidize, and both levels of government offer implicit subsidies (e.g., tax concessions, pricing of public enterprise output such as electricity, irrigation water at below cost, concessions on interest rates). It is not easy to define precisely whether or not some good or service is being subsidized, to measure accurately the extent of such a subsidy once defined, and to analyze its ultimate incidence. The National Institute of Public Finance and Policy (NIPFP) estimated (Government of India 1999) that the explicit and implicit subsidies were as high as 14.4 percent of GDP in 1994–95, of which three-fourths were on non-merit goods (defined as goods that neither generate significant positive externalities nor were of importance in poverty alleviation).16 The extent of subsidies in 1994–95 was indeed staggering, and further, there is no reason to believe that the situation has changed much since. The two politically sensitive subsidies relate to food and fertilizers. Until recently, when a distinction was made between those below the poverty line who were to pay only 50 percent of the cost to the government of food supplied through the public distribution system (PDS), food subsidies were not targeted at the poor. In his budget for 2000–01, the finance minister has withdrawn subsidized sugar allocation for those above the poverty line and raised the sale price of rice and wheat supplied through the PDS, linking it to their open market price. The opposition parties have condemned these proposals, and the one to raise subsidized fertilizer prices, on the grounds that they adversely affect the poor. However, the available evidence does not support the belief that the poor are benefiting from the PDS. The study of Radhakrishna and Subbarao (1997) is very revealing. They find that in 1986–87, the PDS and other consumer subsidy programs accounted for less than 2.7 percent of the per capita expenditure of the poor in rural areas and 3.2 percent in the urban areas. The impact on poverty and nutritional status of the population was 16. Gian Sahota (2000), in an as yet unpublished paper, has strongly criticized the methodology of the NIFP study for its conceptual weakness; its omission of tax expenditures, implicit interest subsidies on concessional loans, subsidy in the development plan outlays, and extra budget subsidies; and, finally, its underestimation of cost recoveries in publicly provided services.

India’s Fiscal Situation: Is a Crisis Ahead?

63

minimal. The PDS had at most brought down the poverty ratio to 38 percent from 40 percent that year, a small reduction indeed. What is even more disturbing, the PDS had a negligible impact in the rural areas where more than three-fourths of the poor live. The cost of the transfer through PDS and other subsidies was very high. The World Bank (1998, table 4.2, 39) reports that the participation of the poorest quintile in the population in PDS as compared to the average is lower at 92 percent and the next two quintiles participate at a slightly higher rate than the average. Even the marginal odds of participation—that is, the gain for each quintile following a rupee increase in aggregate spending on PDS—is only 1.06 for the poorest quintile, as compared to the richest quintile’s 0.81 (World Bank 1998 table 4.3, 41). Thus, PDS subsidies are not particularly pro-poor in their incidence. The central government alone spent more than Rs. 4.25 to transfer one rupee to the poor. Combining central and state government expenditures, in Andhra Pradesh it took Rs. 6.35 to transfer one rupee to the poor. It is generally recognized that a significant part of what is included under the heading “food subsidies” in the budget reflects the inefficiency of the public sector enterprise Food Corporation of India (FCI). In other words, the cost of procuring, transporting, storing, and delivering the amounts supplied by the PDS of various commodities is believed to be much higher than what an efficient enterprise would have incurred. Balakrishnan and Ramaswami (2000) state that the definition of subsidy in government accounting implies that the higher the stocks held by the PDS, the higher is the recorded subsidy, due to carrying costs.17 Clearly, if the FCI is as inefficient as it is believed to be, “using the ‘economic cost’ of FCI as the benchmark is fixing the PDS [issue] price would be to institutionalize the poor quality of the system as a whole” (Balakrishnan and Ramaswami 2000, 1138; Gulati 2000, 1145). These authors and others have proposed replacing the existing PDS with a system of food stamps. Fertilizer subsidies also reflect in part the relatively high cost of domestic fertilizer production, which in turn is a legacy of the autarkic industrial policy of the prereform era. Indeed, the relatively high cost of domestic production arises from the fact that the industry consists of plants of various vintages, less than efficient sizes, and different technologies, using a plethora of feed stocks. Some plants are owned by the government and others by cooperatives and the private sector. The Fertilizer Pricing Committee (Department of Fertilizers 1998) has pointed out that the retention pricing scheme (RPS), which in effect is based on ensuring that even the most costly plant breaks even, is a sure prescription for inefficiency and has recommended a flat rate subsidy. A respected elder statesman and former fi17. In an attempt to reduce the mounting stocks, the government recently announced a reduction in the price of wheat issued to distribution outlets. Such a reduction, if it increases the amount of wheat distributed, would also increase the distribution subsidies, while having only a limited effect on reducing the costs of carrying stocks.

64

T. N. Srinivasan

nance and food and agriculture minister, C. Subramaniam (The Hindu, 1 March 2000), has recently argued that inefficiency is being subsidized. The current finance minister himself drew attention to this in his budget speech to the parliament on 29 February 2000. Gulati (2000) points out that part of the high cost of domestic fertilizer production is due to the higher cost of feed stock, such as natural gas in India, compared to the price of gas in West Asia. He goes on: And on top of that there are certain plants, some of them new ones, which have taken full advantage of the RPS and gold-plated themselves, while some others have understated their capacity. All this is pretty well known in fertiliser industry circles. The problem with this RPS is that it encourages gold plating and does not provide extra incentive to those plants that are on the lower part of the cost curve (economically more efficient) to expand. There is always a tendency to put up another new plant rather than expand the more efficient ones. Lately this has been realised and some expansions have taken place. The industry has to face this reality of open imports very soon. Earlier they prepare themselves, the better it is for them. (1145–1146) Although India as a large importer would face a rising supply price of imports, Gulati is surely right in suggesting that Thus, it is better to dismantle RPS, align domestic price of urea with long-term import parity price to ensure healthy growth of domestic industry. This would be a better system than the existing one, and it can cut down the fertiliser subsidy by a much greater amount than has been achieved by the finance minister in this budget. (1146) The issue of explicit and implicit subsidies is not simply one of raising or lowering this price or that price at the margin, but a much deeper one of domestic political economy. Whatever the true ultimate incidence of any of the subsidies and the factors that brought them about, vested interests have formed in ensuring that they are not touched. The pressure on the finance minister, both from some of the constituents of his own coalition, the National Democratic Alliance, and from the opposition, to roll back the modest reductions in subsidies he has proposed in the budget for 2000–01 has been unrelenting. The leader of the opposition, Sonia Gandhi, has led a “protest march” of members of parliament from her party to the prime minister’s house (The Hindu, 17 May 2000). Fortunately, there is some faint hope that wiser counsels may prevail. Dr. Manmohan Singh, the architect of reforms, and a senior leader of Mrs. Gandhi’s party, has himself argued for the abolition at least of “non-merit” subsidies. Although a spokesman of his party took pains to say, not very convincingly, that Singh’s views do not contradict the party president’s, he did not enthusiastically endorse them either. However, the prime minister has embraced Dr. Singh’s views

India’s Fiscal Situation: Is a Crisis Ahead?

65

and suggested that they could become the basis of a consensus (The Hindu, 18 May 2000).18 2.5 Disinvestment From the fiscal perspective, disinvestment—that is, the sale to private investors of the government’s equity in public-sector enterprises (PSE)—is relevant for at least two reasons. First, with disinvestment, only a part of the surpluses and deficits of PSEs will enter the overall deficit of the nonfinancial public sector. Second, the proceeds from the sale of government equity can be used to reduce the fiscal deficit. Of course, the improvement, if any, in the efficiency of resource allocation in the economy and the more rapid growth that might be brought about by the shift of ownership to private hands both indirectly affect government revenues. Until now, the government has chosen to retain more than 50 percent of the equity in most PSEs put up for disinvestment. Such disinvestment, as contrasted with outright sale of a PSE through full privatization, does not transfer control and operation of the PSE to private hands, and no significant efficiency gain or more rapid growth could be expected from disinvestment. The government’s recent commitment to reduce its equity in nonstrategic PSEs to 26 percent or less, and in public-sector banks to 33 percent, is a step in the right direction. Also, the statement of the then minister for information and broadcasting that “in the last four months we have changed the emphasis from the selective sale of minority shares of successful PSEs to the strategic sale of a significant holding to a strategic partner” (India Today, 15 May 2000, 4) is also encouraging. I will not touch on the broader issues of disinvestment and privatization, such as modalities of sale of equity; the need for ensuring that a PSE, once privatized, will face adequate competition, and the need for setting up an appropriate regulatory framework in sectors where only a few enterprises can be expected to operate. I will focus only on the fiscal impact of disinvestment. A disinvestment commission was set up in 1996 to advise the government on the extent, mode, or timing of disinvestment. At the beginning of 1998, the commission was divested of its powers to monitor and supervise the disinvestment process. The term of the commission expired in 1999 and was not renewed. Instead, a new department for disinvestment has been created. During its existence, the commission issued eleven reports containing 18. As this paper was being presented on May 31, 2000, the minister of telecommunications announced that all employees of the Department of Telecommunications (DoT) would be provided free telephone connections. There was absolutely no rationale for this other than a desire to pander to employees who were threatening to go on strike against the corporatization of service-providing units of the DoT. It is a pity that the prime minister and the government chose to approve this bonanza.

66

T. N. Srinivasan

recommendations on fifty-eight of the sixty-four PSEs referred to it. The recommendations in only thirteen cases were or are being implemented. The “Disinvestment Fund,” set up in 1996 on the advice of the commission to use the sale of proceeds for restructuring PSEs and to finance voluntary retirements of excess staff, is not yet operational. Proceeds of disinvestment, after reaching a peak of Rs. 50.18 billion in 1994–95, fell to a paltry Rs. 3.62 billion in 1995–96. They recovered subsequently to Rs. 4.55 billion in 1996–97, and Rs. 9.0 billion in 1997–98. In 1998–99 they reached another peak of Rs. 53.71 billion, versus a target of Rs. 50 billion. However, this peak is somewhat illusory: The purchases by some PSEs of the equity divested in other PSEs, rather than sale of equity to private investors, was largely responsible for it. The budgeted figure for 1999–2000 was Rs. 100 billion (World Bank 2000, annex 8.3); however, only Rs. 26 billion is expected to be realized (Budget Speech of the Finance Minister, paragraph 62). A target of Rs. 100 billion is again set for 2000–01. It is often suggested that since the sale of public equity in a PSE is in fact a sale of government assets, the proceeds from such a sale should be used to reduce the government’s liabilities by the same amount through reduction of public debt. Likewise, it is further argued that such proceeds should not be used to reduce current fiscal deficits. This argument may have some merit as a way of putting pressure on the government to undertake the politically more difficult tasks of raising revenues or cutting expenditures in order to reduce the fiscal deficit. Logically, however, if the government efficiently uses its resources, it should not make any difference whether a rupee of proceeds is used to reduce the stock of debt or the deficit, since in either case the relevant cost is the same, namely the marginal cost of new debt issue. Moving from sale of equity to the surpluses and deficits of PSEs, according to the economic survey (Government of India 2000, table 7.8), the ratio of pre-tax profit to capital employed of central public sector undertakings rose from an average of 3.5 percent during 1990–93 to 8.0 percent during 1995–98. This aggregate profitability measure is misleading, since the enterprises whose profits are aggregated included state oil and petroleum monopolies. In addition, at least for those enterprises that are competing with private enterprises, their post-tax return has to be compared with similar returns for their private counterparts. Be that as it may, if we measure PSE performance by the deficit in their plan expenditure (mainly on investment) relative to their net internal resources, the deficit of central government PSEs (CPSEs) as a ratio of GDP has halved from 3.0 percent in 1990–91 to a budget estimate of 1.5 percent in 1999–2000 (World Bank 2000, table 8.6). The factors contributing to the decline are the fall in PSE investment from 4.8 percent of GDP in 1990–91 to 3.4 percent in 1999–2000 and the growing importance of petroleum and telecom enterprises, which now account for nearly half the investment and generate more than two-thirds of the net internal resources of CPSs. In other words, government support (through

India’s Fiscal Situation: Is a Crisis Ahead?

67

loans and financing of losses) of CPSE has declined sharply as a proportion of GDP. This is welcome. The government’s economic survey for 1999–2000 does not provide any information on the so-called “sick enterprises” (public and private). The economic survey of 1998–1999 reported that between its inception in May 1987 and the end of November 1998, India’s Bureau for Industrial and Financial Restructuring (BIFR) received 3441 references, of which 2404 were registered and 452 were dismissed as non-maintainable under the Sick Industries Company Act of 1985. It recommended winding up 606 and rehabilitating another 637 and declared 214 as no longer sick. Out of the 225 PSEs that were referred, it registered 157, recommended closing down 29 and rehabilitating 50, and declared 6 to be no longer sick (Government of India 1999, 111). Unfortunately, most of the recommendations are yet to be implemented. With failing enterprises having to obtain rarely granted government permission to close down, and in the absence of well crafted bankruptcy legislation and reform of labour laws that would make it easier to reduce employment, restructuring the viable and closing the non-viable among failing enterprises is difficult to bring about. The continued operation of sick PSEs is a fiscal drain. The industrial relations bill that would have amended India’s restrictive labour laws has yet to be approved by parliament. The politicians and the judiciary do not seem to appreciate that laws protect only the labour aristocracy employed in organized manufacturing and the public sector, and at the expense of the vast majority of workers in other sectors. Recently two Supreme Court judges even castigated Steel Authority of India, a PSE, for not giving employment on compassionate grounds to the family members of employees who died while in service, even after the families had received gratuity and pension benefits! 2.6 Conclusions The clearest statement on the seriousness of the fiscal situation and the need for fiscal correction can be found in the Economic Survey, 1999–2000 (Government of India 2000, 185): More effective management of public finances continues to be the central challenge facing all levels of government of India. In some ways the challenge has become more daunting in recent years pursuant to the sharp increase in government wage bills resulting from the Fifth Pay Commission . . . The adverse effects of large fiscal and revenue deficits on virtually every important dimension of macro-economic performance are well known. They range from low savings and investment, high real interest rates and reduced growth, to adverse pressure on inflation, financial markets and the external sector. Furthermore, the continuous series of large deficits lead to inexorably mounting interest payments, leaving a declin-

68

T. N. Srinivasan

ing share of government expenditure available for essential functions such as defence, law and order, social services and public investment in infrastructure. The recently published estimate of a significant decline in domestic savings and investment in 1998–99 is primarily traceable to burgeoning revenue deficits of Central and State Governments. Similarly, the high level of real interest rates prevailing in recent times is largely due to high fiscal deficits . . . Quite clearly, the prospects for accelerating economic growth depend crucially on the success in managing the fiscal challenge confronting the economy. The same document is explicit on the need for “hard decisions and many fronts . . . They include: a redefinition and narrowing of government responsibilities to those functions that only government can discharge effectively, with a view to down-sizing government; systematic efforts to reduce subsidies by targeting them to the poorest segments of society; a vigorous drive to divest commercial undertakings such as power utilities and transport undertakings; a concerted programme to deploy user charges for economic services rendered by government; systematic induction of information technology tools and modern management practices to enhance efficiency of government; resource generation through transparent sale of under-utilised public properties such as land; and, above all, a determined political commitment to truly effective expenditure management. (186) Although there are some who deny that fiscal deficits are serious enough to matter, fortunately they are a minority and, in any case, I have argued that they have not made a plausible, let alone convincing, case. The majority view, which I share, is eloquently expressed by the economic survey. Indeed, it is not an exaggeration to call the situation a crisis that needs to be attended to immediately. Otherwise, the overarching objective of Indian development that was articulated before independence, namely the eradication of mass poverty through rapid and well-distributed growth, is unlikely to be realized in the foreseeable future. When the economy seems to be at last on the verge of achieving sustained and rapid growth, jeopardizing it is unconscionable. The catalogue of “hard decisions” (Government of India 2000, 186) required is unexceptionable. Whether some or all of these decisions will be taken depends on “determined political commitment,” not only for an “effective expenditure management,” but also for radically reshaping the economic, political, and social institutions. Such reshaping is essential if the Indian economy is to be well integrated, as it must be, with the global economy that has changed and is changing rapidly. Unfortunately, there are not many visible signs that the required political consensus and commitment are emerging. On the contrary, there are several discouraging signs but few encouraging ones.

India’s Fiscal Situation: Is a Crisis Ahead?

69

Let me start with the most encouraging sign.19 In a forthright speech to the Inter-State Council on 20 May 2000, the prime minister pointed out that an unsustainable fiscal deficit has serious negative effects on the economy. It preempts resources that would otherwise be available for investment by the nongovernment sector. This is reflected in the high cost and limited availability of credit; high fiscal deficit also leads to rising debt burdens and a continuous growth in interest payments. This, in turn, squeezes outlays for essential social and economic infrastructure. What is of even greater concern is that these high deficits are not being used to finance investment. They are increasingly being used to finance rising levels of non-Plan expenditure. He continued, “if we do not reverse the trend, we will not be able to achieve the desired GDP growth. We will neither generate employment opportunities nor achieve a reduction in poverty,” and went onto list several critical tasks: the reform of the power sector; rationalizing of user charges on utilities and services; the evolution of a more rational policy in respect of state PSEs; and a much faster rate of growth in government employment in state governments than the center, although this policy is unsustainable. The prime minister rightly emphasized that the Inter-State Council, in which the chief ministers of all the states are members, is an intergovernmental forum that could be used for evaluating policy, ensuring its implementation, and, more generally, for strengthening India’s plural democracy, polity, and society. Although he did not say so himself, the council could be used to evolve a consensus on right sizing of government. Among the many discouraging signs is the tendency of some private sector entrepreneurs to be nostalgic for the good old days when they did not face serious external competition: The prime minister’s advisory committees on industry and trade, headed by leaders of the private sector industry, have apparently recommended further restrictions on foreign investment and the reimposition of quantitative restrictions and tariffs on textile imports. To be fair, we must note that other entrepreneurs have been forwardlooking and are putting pressure on the government to accelerate reforms. However, the preference for insulating the Indian economy from external competition is not confined to a few entrepreneurs. It appears to be shared by supporters of the dominant Bharatiya Janata Party (BJP) in the ruling coalition at the center, such as the Swadeshi Jagaran Manch and the supreme leader of the Rashtriya Swayam Sevak Sangh. A member of the coalition, the Telugu Desam Party, has cautioned the government against “hasty” privatization of some PSEs, raising the specter of unemployment. The trade union arm of BJP has opposed the amendment to the Industrial 19. Acharya adds a few others: The shocks to expenditures from the implementation of the Fifth Pay Commission’s recommendations and the increase in defense needs following Pakistani incursion in Kargil are already mostly absorbed; tax-GDP ratio will improve with better compliance and the bracket creep associated with growing incomes; finally, the pressure for cost recovery is growing and will result in reduction in subsidies.

70

T. N. Srinivasan

Disputes Act that would have raised to 1000 the lower limit of employment in an enterprise that needs to obtain government permission for any retrenchment. The opposition Congress Party, which initiated the process of reforms, has recently appointed a committee to “re-examine” reforms for their bias against the poor. The Indian communist parties continue to be Stalinist and against most reforms, particularly of the public sector. Organized labor is against privatization, regardless of whether the enterprise to be privatized serves any social purpose: For example, workers of a public-sector bakery have recently moved the courts to block its sales to a private enterprise. One hundred twenty members of parliament ranging across party lines have asked for a discussion in Parliament of each and every PSE to be sold to private parties. What is worse, the powers of the recently created regulatory agencies in telecommunications and electricity sectors are being challenged by government departments and enterprises in those sectors. Of course, one could view all these signs optimistically as manifestations of a vibrant democratic polity. Unfortunately, the views expressed and positions taken seem to be those of some powerful and organized special interest groups and against the interests of the public at large, particularly the poor among them. I sincerely hope my pessimism is exaggerated, if not unwarranted.20

References Bajpai, N., and J. Sachs. 1999. The state of state government finances in India. Discussion Paper no. 719. Cambridge: Harvard Institute for International Development. Balakrishnan, P., and B. Ramaswami. 2000. Vision and illusion in fiscal correction. Economic and Political Weekly (Bombay), 1 April: 1137–39. Buiter, W., and U. Patel. 1992. Debt, deficits and inflation: An application to the public finances of India. Journal of Public Economics 47:171–205. ———. 1996. Solvency and fiscal correction in India: An analytical discussion. In Fiscal Policy in India, ed. S. Mundle, New Delhi, India: Oxford University Press. ———. 1997. Budgetary aspects of stabilization and structural adjustment. In Macroeconomic dimensions of public finance: Essays in honour of Vito Tanzi, ed. M. Blejer and T. Ter-Minassian, 363–412. London: Routledge. Burgess, R., S. Howes, and N. Stern. 1993. Tax reform in India. Suntory-Toyota In20. Acharya is less pessimistic than I am in viewing the fiscal situation as heading toward a crisis, although he concedes that the fiscal pressure could reduce potential growth and that this drag could worsen if corrective action is not taken soon. He sees signs of the drag in declining investment, the high real interest rate, and the rekindling of inflation. He deems it unlikely that simultaneous fiscal and balance of payment (BOP) crises, as occurred in 1991, will happen again, primarily because the reforms of the external sector have made a BOP crisis less likely. As I argue in the paper, the fact that fiscal deficits have not thus far resulted in deleterious consequences does not imply that they are sustainable.

India’s Fiscal Situation: Is a Crisis Ahead?

71

ternational Centre for Economics and Related Disciplines, London School of Economics. Discussion Paper no. 45 (processed). ———. 1994. Reform of domestic trade taxes in India: Issues and options. New Delhi: National Institute of Public Finance and Policy (processed). Chandrasekhar, C. P. 2000. Economic Reform and the Budget. Economic and Political Weekly (Bombay), 1 April: 1140–42. Department of Fertilisers. 1998. Fertiliser pricing policy. New Delhi: Government of India, Department of Fertilisers and Chemicals. Government of India. 1999. Economic survey 1998–1999. New Delhi: Government of India Press. ———. 2000. Economic survey 1999–2000. New Delhi: Government of India Press. Gulati, Ashok. 2000. Millennium budget. Economic and Political Weekly (Bombay), 1 April: 1143–46. Joshi, V., and I. Little. 1994. India: Macroeconomics and political economy 1964– 1991. Washington, D.C.: World Bank. ———. 1996a. India’s economic reforms 1991–2001. Oxford, U.K.: Clarendon Press. ———. 1996b. Macroeconomic management in India, 1964–1994. In Trade and development: Essays in honour of J. N. Bhagwati, ed. V. Balasubramaniam and D. Greenaway. London: Macmillan. Ministry of Finance. 1992. Final report of the Tax Reforms Committee. New Delhi: Ministry of Finance. Olekalns, N., and P. Cashin. 2000. An examination of the sustainability of Indian fiscal policy. University of Melbourne, Working Paper. Radhakrishna, R., and K. Subbarao. 1997. India’s public distribution system. Discussion Paper no. 380. Washington, D.C.: World Bank. Rao, M. G., and H. K. A. Nath. 2000. Fiscal correction: Illusion and reality. Economic and Political Weekly (Mumbai). 5 August: 2806–09. Rao, M. G., and N. Singh. 1999a. The assignment of taxes and expenditures in India. Stanford University, Center for Economic Policy Research, Working Paper no. 30a. ———. 1999b. Fiscal overlapping, concurrency and competition in Indian Federalism. Stanford University, Center for Economic Policy Research, Working Paper no. 30b. ———. 1996c. Intergovernmental transfers: Rationale, design and Indian experience. Stanford University, Centre for Economic Policy Research, Working Paper no. 30c. ———. 1999d. Analysis of explicit and implicit intergovernmental transfers in India. Stanford University, Center for Economic Policy Research, Working Paper 30d. Reserve Bank of India. 2000. Annual report, 1999–2000. Mumbai: Reserve Bank of India. Sahota, G. 2000. Flawed methodology employed for measuring subsidies in India. Vanderbilt University, Department of Economics, Manuscript. Weingast, B. 1995. The economic role of political institutions: Market-preserving federalism and economic development. Journal of Law and Economics and Organization 11 (1): 1–31. World Bank. 1996. India Country Economic Memorandum, Report no. 15882-IN. Washington, D.C.: World Bank. ———. 1998. India: Reducing Poverty, Report no. 17881-IN. Washington, D.C.: World Bank. ———. 1999. World Development Report. Washington, D.C.: World Bank. ———. 2000. India: Policies to Reduce Poverty and Accelerate Sustainable Development, Report no. 19471-IN. Washington, D.C.: World Bank.

72

T. N. Srinivasan

Comment

Shankar Acharya

T. N. Srinivasan’s paper provides a useful survey of some current fiscal problems facing India. It is unusually difficult for me to comment critically on this paper since, in his conclusions, Srinivasan states that “The clearest statement on the seriousness of the fiscal situation and the need for fiscal correction can be found in the Economic Survey, 1999–2000 (Government of India 2000, 1.85),” which he then goes on to cite at some length. As many at this conference may know, I had something to do with drafting the paragraphs he cites! Nevertheless, I will essay a few comments under three broad headings: diagnosis of problems, prescription of solutions, and prognosis for the future. Diagnosis of Problems To begin with, the paper does not give the reader any sense of the evolution of the overall fiscal problem, for example, as measured by the general government deficit (center and states combined) over the decade of the 1990s. This is a pity, because there is an interesting and important story to be told here. Perhaps the key point here is the resurgence of fiscal pressure in the last three years, stemming largely from the governmental decisions on the recommendations of the Fifth Pay Commission (FPC) in 1997. Between 1990–91 and 1996–97 there was tangible progress in reducing the fiscal deficit (barring significant slippage in 1993–94) from 9.2 percent of GDP in 1990–91 to 6.2 percent in 1996–97.1 Since then the deficit has worsened markedly, mainly (but by no means solely) due to the wage-bill (including pensions) consequences of the FPC. This ratcheted up central government expenditure by nearly 1 percent of GDP in 1997–98 and beyond, with a knock-on effect of similar magnitude on state government expenditure in 1998–99 and subsequently.2 In sum, the FPC decisions led to a direct worsening of the general government deficit by about 2 percent of GDP by 1998–99. It was a clear case of a significant exogenous fiscal shock to which the corrective fiscal response was less than adequate. As a result, the gen-

Shankar Acharya is currently Honorary Professor at the Indian Council for Research on International Economic Relations. At the time of writing this article, he was a Visiting Research Fellow at Merton College, Oxford University, on leave from his regular assignment as Chief Economic Adviser in the Ministry of Finance, Government of India. The views expressed in this Comment are those of the author and should not be attributed to the government of India. 1. See table 4.2, p. 94, Reserve Bank of India Annual Report 1998–99. The fiscal consolidation is slightly exaggerated by change in GDP estimates to the new series for the years 1993–94 and beyond. 2. Although the FPC decisions relate strictly to the central government employees, the follow-on impact on state government wages has been blessed by past precedents and was largely predictable.

India’s Fiscal Situation: Is a Crisis Ahead?

73

eral government deficit was back up above 9 percent of GDP by 1999–2000, that is, comparable to the pre-crisis level of 1990–91. The FPC wage increases not only worsened central and state fiscal deficits, thus eroding aggregate saving, investment, and growth performance, but also weakened the economy’s dynamism through other channels: 1) States now spend on little else than wages, salaries and pensions. Hardly any money is left over for the complementary inputs that, together with employee services, go into the effective provision of public and quasi-public goods, such as primary education, healthcare, agricultural services, roads, and so forth. In an important sense, state government expenditures have become more like transfers to the protected segment of state government employees than expenditure on public service provision! Thus, the FPC decisions have significantly impaired the provision of these key social and economic services, which are so important to the economy’s long-run growth potential. 2) The associated high public wage syndrome has further distorted the labor market in India and has thus eroded competitiveness and growth potential, while rendering growth even less employment-intensive. The problem of pervasively inadequate cost recovery (and associated high explicit and implicit government subsidies), mentioned by Srinivasan, is certainly serious. I would note that the problem has undoubtedly been aggravated by the FPC wage hikes. Srinivasan’s discussion of trends in the tax-GDP ratio notes a decline since the early 1990s, which is certainly indisputable. However, his discussion remains at an aggregate level and ignores some noteworthy elements: 1. The ratio of direct taxes to GDP improved markedly during the 1990s, from about 2 percent of GDP in 1990–91 to around 3 percent by 1996–97. 2. Furthermore, this improvement occurred in both personal and corporate taxes, and despite significant reductions in tax rates during the period. In other words, there was successful base-broadening. 3. The drop in the customs revenues to GDP ratio observable over the decade was predictable in view of the very substantial tariff rate reductions implemented (especially in the first half of the decade) as part of the government’s deliberate strategy of reforming trade and tariff policies.3 4. The real problem with the unexpected decline in the tax-GDP ratio has been with central excise, revenues from which should ideally have risen to compensate for the predictable decline in customs collections. Quite apart from administrative weaknesses, there are at least three major reasons for the weak performance of excise collections: a) the generous exemptions for small scale industrial units, which also encourage splitting of enterprises; 3. For further details, see Shankar Acharya, “Managing External Economic Challenges in the Nineties: Lessons for the Future,” The 18th Anniversary Lecture of the Center for Banking Studies, printed as Occasional Papers Number 33–1999, Central Bank of Sri Lanka.

74

T. N. Srinivasan

b) the continuation of too many product-specific exemptions, although the 1990s did see several serious attempts to reduce these; c) most importantly, the constitutional limitation that confines excise taxation to the manufacturing level and thus places value-added in services (the fastest growing sector of the Indian economy) outside the pale of central domestic trade taxes. This limitation also encourages manufacturing units to price and account in a manner that pushes as much of value to the post-manufacturing stage as possible. Srinivasan’s treatment of tax reform in the 1990s is somewhat patchy and fails to credit adequately the major advances that have taken place, notably: 1. the apparent acceptance, across different governments, of a strategy of modest rates and broader bases as an approach to direct taxes. This is important given the earlier history of extortionary tax rates, which imparted massive incentives for evasion and avoidance and imposed serious allocative costs; 2. the remarkable reduction in peak customs tariff rates, from around 300 percent (!) in 1990–91 to around 40 percent by the end of the decade; 3. the difficult but largely successful transformation of a multiple rate (more than a dozen rates as recently as 1997) excise structure to a nearly single-rate manufacturing level value-added tax, with some luxury items bearing the additional burden of special excises. The transition to this fairly common pattern of modern taxation has been largely completed by the present government. Srinivasan is also wrong in criticizing the recently passed constitutional amendment, which unifies the proportion of major centrally collected taxes devolved to the states. Recommended by the Chelliah Tax Reform Committee—which Srinivasan commends highly—and many public finance specialists, this measure eliminates the distortionary incentives embedded in the earlier non-uniform structure, which typically encouraged central tax policy to load exemptions and tax sops on personal income and excise taxes, the bulk of whose collections was devolved to the states. Where Srinivasan is correct is in observing the lack of serious progress in integrating the VAT-type Central Excise (or MANVAT) and the state sales tax systems into some form of integrated VAT taxation. However, as other federations have discovered, this is no easy task and typically demands years of effort to persuade constituent states and provinces. In India it would almost certainly also require amendment of the constitution. Some progress has been made recently in encouraging states to reform their sales tax systems along VAT lines. Finally, the paper is surprisingly silent on the issues of fiscal sustainability and debt dynamics, although these issues are clearly important and have been fairly seriously analyzed by some of the authors Srinivasan cites, in-

India’s Fiscal Situation: Is a Crisis Ahead?

75

cluding Buiter, Joshi, Little, and Patel. More broadly, Srinivasan does not seriously explore the interactions between fiscal deficits and economic growth. In particular, he does not address the intriguing question of how India sustained fairly high rates of growth in both the 1980s and 1990s despite high levels of fiscal deficit in both decades. Prescription of Solutions I generally agree with the thrust of Srinivasan’s prescriptions for improving the fiscal situation. These include reduction of explicit and implicit subsidies through far more effective deployment of cost recovery policies across a wide range of goods and services provided by the government and public sector; increase in the tax-GDP ratio; a more aggressive privatization program; and a greater role for conditional finance in center-state fiscal relations. However, I have brief comments on each of these. First, reduction of large subsidies in power, water, road transport, fertilizers, and so forth, is clearly called for. But no one should underestimate the political obstacles to more sensible cost recovery policies. Similarly, increasing the tax-GDP ratio is easier said than done, especially when one recognizes that the share of customs revenues is likely to continue to decline as tariffs are brought down closer to East Asian levels. The main hopes here must lie with closer approximation to VAT structures encompassing services in the base of indirect taxes and with the continued broadening of the base for income tax (coupled with rising incomes), including the phasing out of revenue-costly concessions, such as that for income from export. Moreover, as regards more aggressive privatization, the efficiency gains may well be more important than the short-term fiscal rewards. In either case, acceleration of the program is desirable, as the sale value of many public enterprises is likely to diminish quite swiftly. On fiscal-federalism, Srinivasan advocates a wholesale reengineering of the structure. This is not likely to happen. More realistically, it may be feasible to build in a larger component of conditional finance in center-state flows. Steps along these lines have already been taken in the context of the scheme for extended ways and means financing and the associated Memoranda of Understanding for fiscal reforms in states. It is possible that the ongoing Eleventh Finance Commission might also make recommendations along these lines. Prognosis The title of Srinivasan’s paper suggests an impending fiscal crisis, and the concluding section of his paper does not dispel this suggestion. Interestingly, Srinivasan does not moot the possibility of a confluence of fiscal crisis with financial sector pressures arising from a growing need to recapitalize weak banks. In part this may be because Srinivasan’s paper ignores contingent fiscal liabilities, a potentially serious lacuna.

76

T. N. Srinivasan

Nor does Srinivasan elaborate on the likely trajectory and emerging pressure points of a possible fiscal crisis. Ideally, of course, a crisis should be forestalled by anticipatory corrective actions of the type discussed above, as well as by other measures such as the fiscal responsibility legislation currently being drafted. Even if the fiscal pressures were allowed to go unchecked, it is quite possible that, instead of bringing about a replay of the 1991 crisis (when fiscal pressures spilt into a balance of payments crisis), this time the costs might manifest themselves through high real interest rates, crowding out, low savings and investment, and reduced growth. Some would argue that this process has already been in train over the last few years. In either case, there are very strong arguments for accelerating corrective fiscal action.

Comment

Kenneth Kletzer

T. N. Srinivasan has written an insightful and careful analysis of the fiscal policy challenges facing India at the beginning of a new decade. His discussion is organized around five topics: the overall public sector budget deficit, state finances, tax and expenditure reform, subsidies, and disinvestment. The most important contemporary issues for Indian fiscal reform are covered by these topics, and this survey provides a broad and comprehensive overview of the current fiscal situation. Although Srinivasan has been guided in his selection and emphasis by pressing current problems and recent policy reforms, the issues he raises are perennial ones for Indian fiscal policy. He makes several provocative and thoughtful suggestions for future reform efforts. I limit my discussion to only a few of the points made in the paper. This paper begins with a most appropriate question: Are the fiscal and financial policies of the public sector sustainable? Srinivasan shows us that the deficit reductions of the 1990s were short-lived. In the past year, the operational (interest-inclusive) deficit of the consolidated public sector has returned to roughly 11 percent of gross domestic product (GDP), its level in 1990–91 just before the macroeconomic crisis. It is argued here that such deficits are unsustainable given the current stock of outstanding government debt. I concur with Srinivasan’s conclusion that the current fiscal deficits are unsustainable and will lead to another macroeconomic crisis without a significant correction; the consolidated finances of the Indian public sector present a case in which common sense and econometric tests reach the same conclusion. Time series analyses reveal that the debt-toGDP ratio for India followed a non-stationary growth path in the 1990s, just as it did in the 1980s. Kenneth M. Kletzer is professor of economics at the University of California, Santa Cruz.

India’s Fiscal Situation: Is a Crisis Ahead?

77

It is useful to recall the assumptions underlying this approach to testing public sector solvency. The conventional solvency condition requires that the outstanding debt-to-GDP ratio be no larger than the expected present value of future primary public sector budget surpluses (as ratios of GDP). This implies that the debt-to-GDP ratio cannot grow asymptotically at a rate faster than the difference between the real interest rate and the growth rate of real GDP. The conventional criterion allows debt to grow indefinitely as a proportion of output at any rate less than the long-run difference between the real interest rate and the growth rate, implying that the deficit cannot be positive in the long run if the government has outstanding public debt. The stationarity tests adopted by Buiter and Patel (1992), Olekalns and Cashin (2000), and others test for whether the debt-to-GDP ratio converges. This is a much more restrictive condition, but one that is implied by the conventional solvency criterion when all tax instruments are distortionary. In the case of India, however, primary deficits are positive and rising, so that even the most generous solvency condition should lead us to conclude that the current fiscal and financial policies of the government cannot be sustained. The persistence of deficit financing by the public sector is particularly troubling in the case of India. The government of India has taken advantage of a captive capital market to subsidize its borrowing. The banking sector is presently required to hold 35 percent of its assets in government liabilities through a combination of currency and interest-bearing debt. Even though these requirements have been reduced in recent years from even higher levels, they result in a sizable tax on financial intermediation. The negative interest rate differential paid to financial institutions on government debt relative to industrial borrowers has also decreased significantly over the last several years as a policy initiative to reduce the degree of financial repression. It is interesting to note that the banking sector holds government debt in excess of its requirement under the statutory liquidity ratio at current interest rates. This should not necessarily be interpreted to mean that risk corrected rates of return have converged when banks are state-owned and managed by risk-averse public employees. Further, in a closed capital market, a rise in the ratio of government debt to output tends to increase the real rate of interest. Taxing intermediation remains an important part of public finance and appears to be a preferred alternative to the monetization of government deficits in the absence of fiscal reforms. The resulting financial repression distorts the pattern of investment and production and can be quite costly in terms of real economic growth. Running persistent public-sector budget deficits and borrowing on subsidized terms from the financial sector is a bad habit. The addiction of the government to deficit financing and financial repression creates a barrier to further economic liberalization. Despite the

78

T. N. Srinivasan

convergence of interest rates on loans to the government and to the enterprise sector, capital account liberalization would result in a fiscal crisis without serious deficit reduction. One of the costs of Indian fiscal policies is the sacrifice of the potential gains from international financial market integration. It is also possible that a fiscal crisis could lead to the reversal of recent capital market reforms and to increased financial repression as devaluation and domestic inflation are resisted. Budgetary imbalances for state governments are a major factor in the persistence and growth of India’s public sector deficits and occupy a central place in Srinivasan’s analysis. The finances of the states pose a serious problem for fiscal management and a major challenge for fiscal reform. The states bear responsibility for social welfare spending under the constitution but receive a large share of their resources in the form of transfers from the center. The central government has tended to finance state government deficits over the past several years. Srinivasan and others suggest that the center should condition transfers on the past fiscal policies of each recipient state. That is, the central government could attempt to discourage state governments from incurring debt by imposing a form of conditionality. We should worry that threats to reduce future transfers may not be credible when these transfers finance the political and social goals of the national government. In a recent incident, the state of Minas Gerais defaulted on debt repayments to the Brazilian government, calling the bluff of national political leaders. The inability of governments to refuse to take over the debt obligations of private banks and corporations, let alone state governments, has been a frequent and troublesome event. In writing the existing fiscal arrangements between the center and the states, India recognized the incentive problems of allowing the states to borrow on capital markets when a significant portion of state government resources is collected and distributed by the center. The problem has become one of the states continuing to run deficits and borrowing directly from the center, which in turn provides the resources for repayment. A formal policy of making transfers contingent on past fiscal performance will probably be time inconsistent, given the disincentives of the national government to reduce regional development expenditures. This paper also considers alternative reforms that may mitigate against the time consistency problem. These involve reassigning tax authority and expenditure responsibility within the fiscal federal system. There are both theoretical reasons and historical evidence to justify rewriting the fiscal arrangement between the center and the states. Much could be done to impose more discipline on state fiscal authorities at the same time that they achieve the fiscal tools to provide state-level public goods. Increasing the capacities of the states to raise revenue should reduce the adverse fiscal incentives facing the states and encourage more prudent public-sector financing decisions.

India’s Fiscal Situation: Is a Crisis Ahead?

79

Srinivasan suggests just such an approach following the notion of market-preserving federalism outlined by Weingast (1995). The differences in per capita income across the states of India are large, so that the elimination of all interjurisdictional transfers may be socially undesirable. However, significant increases in assignment of tax authority to the states can dramatically reduce the magnitude of center-state transfers and bias toward deficitfinancing of state expenditures. It is difficult to impose credible restrictions on state-level borrowing as long as the states are not given sufficient means to raise revenues but remain reliant on transfers from the center to finance their mandates. Srinivasan is optimistic that fiscal competition between the states will lead to a race to the top. It is well understood that allocative efficiency in local public goods provision relies on somewhat strict assumptions, but this is a minor quibble. More importantly, I believe that Srinivasan makes a valid argument that increasing the tax capacity of the states and reducing transfers from the central government to the states as much as possible will improve fiscal policy-making in India. References Buiter, W., and U. Patel. 1992. Debt, deficits and inflation: an application to the public finances of India. Journal of Public Economics 47:171–205. Olekalns, N., and P. Cashin. 2000. An examination of the sustainability of Indian fiscal policy. University of Melbourne, Department of Economics, Research Paper 748. Weingast, B. 1995. The economic role of political institutions: Market-preserving federalism and economic development. Journal of Law and Economics and Organization 11 (1): 1–31.

Comment

N. K. Singh

I would like to make a few observations at this stage, followed by more detailed comment on the direct tax reforms strategy. First, I believe that the national consensus on the direction and content of the economic reform strategy has been somewhat erratic. While it is broadly true that the direction of the policies initiated in 1991 remained unaltered by the coalition governments that replaced the Congress Government following the elections in 1996, the commitment to pursue key elements of the reform did not remain constant, and emphasis kept changing. Following the 1991 crisis, we entered into a medium-term arrangement with the World Bank and the International Monetary Fund (IMF). The reform strategy during the two sucN. K. Singh is currently a member of the planning commission in the government of India. At the time of writing this article, he was Secretary to the Prime Minister of India.

80

T. N. Srinivasan

ceeding years, when we had a program with the IMF, was clear, because it involved compliance with the conditionalities and structural benchmarks laid down by the multilateral lending institutions. Thereafter, following electoral reverses in some states, the commitment to the reform had already began to sag. While it would be fair to say that the election manifestoes of most political parties in the general elections that followed the defeat of the congress in 1996 had a broad measure of similarities, it would be an exaggeration to suggest that the commitment of all political parties to a “common economic program” had remained unaltered. An illustration of this approach is the hesitation of the United Front Government to adjust the issue price of food grains consequent on the increase of minimum support price for rice or wheat, or to adjust diesel and kerosene prices in line with the movement of the international prices resulting in an unsustainable deficit in the Oil Pool Account. In fact, this conclusion is only strengthened by the present ongoing debate within the ruling party in India and the recent statements of four former prime ministers who had headed coalition governments and even of the prime minister who was in office in 1991, under whose leadership the present reform process was initiated. While the logic of incomplete action in certain key areas and the reform strategy based on their implementation may suggest consensus, it would be naïve to conclude that the last ten years have seen a continued national consensus on a predetermined reform strategy. Second, the reforms have been perceived in India to be elitist-driven and designed primarily to benefit the corporate sector. This is because reform of agriculture and the social sector has not even commenced. To carry the reform momentum forward, it will be necessary to bridge this “disconnect” between reform and what is perceived to be elitist-driven strategy. This would require the initiation of serious reforms of the agriculture sector so that the constituency for economic reforms can be significantly enlarged. It would also require a serious change in focus to include the agriculture sector and social policy issues like education, health, demographic management, rural connectivity, and shelter. Third, in the area of fiscal policies, some useful reforms have been initiated with respect to direct and corporate taxes, while action in the area of improved expenditure management has continued to be weak. Expenditure management policy has been nearly absent in spite of the commitments made from time to time in parliament and elsewhere for adopting a performance-based or zero-based budgeting to subject public expenditure portfolios to qualitative review. Action has been even weaker in downsizing the government and rationalizing manpower in public sector undertakings. The expenditure budgets of some state governments have been compromised further, not only by tardy action to improve the finances of state electricity boards and of the transport sector, and by delays in imposition of user charges for a wide variety of inputs in the agriculture sector, but also

India’s Fiscal Situation: Is a Crisis Ahead?

81

by the implementation of the recommendations of the Fifth Pay Commission, which significantly enhanced the salary and perks of government employees. Tax Policy Reforms In the area of tax policy, the process of reform was initiated as early as 1985, and then accelerated in the post-1991 period. Significant changes were introduced in 1997–98, when the personal income tax rates were simplified and reduced to only three rates, namely, 10, 20, and 30 percent. Simultaneously, corporate tax rates were also rationalized, and the corporate tax rate for companies in which the public was substantially interested was sharply reduced to 35 percent. These sharp reductions in tax rates were effected under the assumption that the revenues forgone would lead to higher disposable income for individuals and corporates who are likely to utilize the resources more efficiently, leading to higher growth rate in the medium term. There is also a presumption that the lowering of tax rates will not affect revenues because the more moderate rates are likely to induce greater compliance, resulting in increased revenues. Table 2C.1 indicates the behavior of the tax-GDP ratio in India since 1984–85. Even though the overall tax-GDP ratio in the post-reform period has shown a downward trend, the direct tax-GDP ratio has shown a distinct increase over the same period. Therefore, one of the underlying assumptions for reduction in tax rates appears to be validated. However, it is necessary Table 2C.1 Fiscal Year 1984–1985 1985–1986 1986–1987 1987–1988 1988–1989 1989–1990 1990–1991 1991–1992 1992–1993 1993–1994 1994–1995 1995–1996 1996–1997 1997–1998 1998–1999 1999–2000

Tax-GDP Ratio in India (1984–85 to 1999–2000) Ratio of Direct Taxes to GDP1

Ratio of Indirect Taxes to GDP

Ratio of Tax to GDP

2.14 2.23 2.22 2.13 2.32 2.28 2.16 2.60 2.69 2.60 2.95 3.15 3.15 3.492 2.89 3.24

8.12 8.93 9.24 9.50 9.11 9.20 8.84 8.55 8.12 6.90 7.02 7.12 7.10 6.37 5.82 6.21

10.26 11.16 11.47 11.63 11.43 11.48 10.99 11.15 10.82 9.49 9.97 10.26 10.25 9.85 8.71 9.45

The GDP figures for the various years are under the new series. This includes collections under the amnesty scheme.

1 2

82

T. N. Srinivasan

to test whether the assumptions on which tax rates were reduced are validated by time series data and to determine the impact of aggregate tax rates and enforcement level on the overall tax realization. Typically, the personal tax collections in any given year are a function (f) of the size of the aggregate tax base, the tax rates, and the level of enforcement. Therefore, (1)

PIT
f (Tb, Tr, E)

where PIT refers to personal income tax, Tb to tax base, Tr to the tax rate, and E to the level of enforcement. In principle, the variable representing the aggregate tax base should be representative of the real tax base. However, in the absence of accurate data relating to the real tax base and agricultural income remaining outside the purview of personal income tax, we use the nonagricultural gross domestic product at factor cost (NAGDP) as a proxy for the size of the aggregate tax base. Similarly, we use the maximum marginal rate (MXR) of personal income tax inclusive of union surcharge, if any, as a proxy for the tax rate, Tr, because any change in MXR is also accompanied by appropriate changes in tax rates at all other lower income levels. Similarly, we use the number of tax returns filed in any year (RET) as a proxy for E, because any change in enforcement level must be reflected in a corresponding change in the number of non-filers and stop-filers, which would show up as a change in the number of tax returns filed. Therefore, the relationship in equation (1) is reduced to (2)

PIT
f (NAGDP, MXR, RET).

Empirical analysis of income tax collection requires specification of equation (2) in terms of an empirically testable equation, and quantification of its arguments using aggregate measures. Since the relation of PIT is nonlinear in the dependent variables, and to overcome problems associated with time series data, we begin by postulating the following equation: (3)

PITGt
  1 · MXRt  2 · RETGt  3 · NAGDPGt · RETGt  Ut

where , 1, 2, and 3 are coefficients required to be estimated, U is the random disturbance term, t is the annual time index, PITG is annual growth in the personal income tax, RETG is the annual growth in the number of tax returns filed, NAGDPG is the annual growth in nonagricultural gross domestic product, and all other notations are as defined previously. The interactive term, NAGDPG • RETG, represents the effective tax base. The historical data series for each of the variables is generated after adjusting for inflation by using the index of consumer prices for non-manual urban workers (base year 1973–74). Using the historical data series so generated for the dependent and the explanatory variables, equation (3) above is estimated. From the figures reported in Table 2C.2 below, it is evident that

India’s Fiscal Situation: Is a Crisis Ahead? Table 2C.2

83

Estimated Equation for Personal Income Tax Collections in India (1984–1985 to 1999–2000) Independent Variables MXR RETG NAGDPG  RETG Constant R2 D.W N

Coefficients

t-ratios

–0.619615 –2.061772 35.85569 0.378980 0.722 2.364 16

–2.472 –6.243 6.276 3.052

Note: See text for definitions of variables. Table 2C.3

Personal Income Tax Elasticities

Fiscal Year

Maximum Marginal Rate of Personal Income Tax

Number of Tax Returns Filed

Nonagricultural GDP at Factor Cost

1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99 1999–20000

–0.3834 –0.3098 –0.3098 –0.3253 –0.3253 –0.3346 –0.3470 –0.3470 –0.2776 –0.2776 –0.2478 –0.2478 –0.2478 –0.1859 –0.1859 –0.2045

0.4261 0.7192 –0.2537 –0.4976 1.3441 1.8543 –0.4963 –1.7212 –0.5846 1.1040 1.4449 0.7072 –0.3739 0.3894 –1.1048 –0.0657

0.0415 11.5651 –3.9399 –7.2451 4.0147 1.3600 5.2494 1.1752 3.2134 3.0188 0.7296 0.2810 5.7328 7.5097 14.9948 6.9790

the model captures the aggregate tax collection relationship. All relevant coefficients have the expected signs and are significant at the 0.05 level. The equation explains 72 percent of the variation in the income tax collections. In other words, the three explanatory variables, namely, maximum marginal rate of personal income tax, growth in the number of returns filed, and growth in the nonagricultural gross domestic product, account for 72 percent of the variation in the income-tax collections. The 28 percent of the unexplained variation is due to some year-to-year random factors. Based on the estimated functional relationship above, the tax elasticity with respect to tax rates and levels of enforcement, and the tax buoyancy with respect to the nonagricultural GDP at factor cost, are indicated in Table 2C.3.

84

T. N. Srinivasan

As expected, the tax elasticity with respect to tax rates is negative. Further, it also increases with reduction in the maximum marginal rate of personal income tax inclusive of surcharge, if any. Therefore, there is evidence to support that reduction in the personal income tax rates over the years has induced greater compliance, resulting in higher revenue collections. Figure 2C.1 shows the estimated trend of personal income tax collections at 1984–85 prices in response to changes in the maximum marginal rate of personal income tax controlling for the effects of the level of enforcement and the nonagricultural gross domestic product at 1984–85 levels. The assumption underlying the tax rate cuts in 1997–98 appears to be validated; the tax collections increased by 5.5 percent. Similarly, the tax elasticity with respect to levels of enforcement has varied over the years, the extremes being a low of –1.7212 in 1991–92 and a high of 1.8543 in 1989–90. Further, we observe a negative elasticity not only in years when the nonagricultural GDP registered a negative growth rate, but also in other years. The effectiveness of the tax administration is determined by its ability to deal with non-filers, stop-filers, tax evaders, and delinquent accounts. The negative tax elasticity with respect to levels of enforcement in some years implies that deterrence due to enforcement effort had a positive impact on non-filers and stop-filers. However, it had a negative impact on tax evaders and delinquent accounts. The net effect was positive (negative) in years in which the positive impact on non-filers and stop-filers was greater (lower) than the negative effect on tax evaders and delinquent accounts. Further, controlling for the effects of the maximum marginal rate of personal income tax and the nonagricultural GDP at 1984–85 levels, the enforcement effect over the period 1984–1985 to 1999–2000 has shown a declining trend in terms of its effect on revenue collections (fig. 2C.2). This is not surprising, given the fact that the capacity of the tax administration to deal with noncompliance has remained unchanged during this period. Noncompliance has shifted according to the change in the focus of the tax administration. While the administration has embarked on a program of large-scale induction of information technology, employees continue to resist computerization. Hence, its success leaves much to be desired. There continues to be a serious mismatch in the capacity of the tax administration and the requirements of the ever-increasing workload, exacerbated by new and sophisticated methods of tax evasion. Do the aforementioned results hold true even for corporate tax revenues? We use a similar methodology to investigate the evidence in support of the underlying assumption that corporate tax revenues respond positively to reduction in corporate tax rates. As in the case of personal tax collections, corporate tax collections in any given year are determined by the size of the aggregate tax base, the tax rates, and the level of enforcement. Therefore, (4)

CT
f (Tb, Tr, E),

Fig. 2C.1

Relationship between personal tax rates and personal tax revenues in India

Fig. 2C.2

Trend of enforcement effectiveness in India

India’s Fiscal Situation: Is a Crisis Ahead?

87

where CT refers to corporate tax, Tb to tax base, Tr to the tax rate, and E to the level of enforcement. Again, in the absence of accurate data relating to the real tax base and with agricultural income remaining outside the purview of corporate tax, we use the nonagricultural GDP at factor cost (NAGDP) as a proxy for the size of the aggregate tax base. Because corporate profits are closely linked to the value of output produced by a company, and NAGDP represents the value of goods and services produced in the nonagricultural sector, the NAGDP is a good proxy for the aggregate tax base. Similarly, we use the corporate tax rate applicable to companies in which the public is substantially interested (CTR) as a proxy for the tax rate, Tr. Collections from other forms of corporate bodies are negligible. In the absence of any satisfactory measure of E for corporate tax, we assume it to be constant over time for the purpose of this analysis, and therefore the relationship in equation (1) is reduced to a simple relationship between CT and NAGDP and CTR: CT
(NAGDP, CTR)

(5)

Empirical analysis of corporate tax collection also requires specification of equation (5) in terms of an empirically testable equation, and quantification of its arguments using aggregate measures. Since the relation of corporate tax collection (CT) is nonlinear in the tax base variable (NAGDP), we begin by postulating the following equation: (6)

Log(CTt )
  1 · CTRt  2 · LOG(NAGDPt ) · CTRt  Ut ,

where , 1, and 2 are coefficients required to be estimated, U is the random disturbance term, t is the annual time index, and all other notations are as defined previously. As in the case of the personal income tax model, the historical data series for each of the variables is generated after adjusting for inflation by using the index of consumer prices for non-manual urban workers (base year 1973–74). Using the historical data series generated for the dependent and the explanatory variables, equation (6) above is estimated. From the figures reported in Table 2C.4, it is evident that the regression diagnostic results are Table 2C.4

Estimated Equation for Corporate Tax Collections in India (1984–85 to 1999–2000) Independent Variables

Coefficients

t-ratios

CTR Log(NAGDPG)  CTR Constant R2 D.W N

–30.67999 2.608713 7.895044 0.926390 1.146249 16

–8.507674 7.185879 23.23205

88

T. N. Srinivasan

robust and that the model successfully captures the aggregate corporate tax collection relationship. All relevant coefficients have the expected signs and are significant at the 0.05 levels. The equation explains 92 percent of the variation in the corporate tax collections. The 7.4 percent of the unexplained variation is due to some year-to-year random factors. Based on the estimated functional relationship above, the corporate tax elasticity with respect to corporate tax rates and the tax buoyancy with respect to the nonagricultural GDP at factor cost are indicated in Table 2C.5. The corporate tax elasticity with respect to the corporate tax rate is negative for most years, and positive and greater than one with respect to nonagricultural GDP. In other words, the corporate tax elasticity with respect to the rate is not independent of the level of the NAGDP. This is unlike the case of personal income tax. Therefore, given the level of NAGDP in 1996–97, there was a case for increasing the corporate tax rate without fear of losing revenues. Any broad generalization from the available data may be somewhat premature because the habit—of and attitude toward—tax compliance changes over the medium term, and the major tax reform initiatives since 1997 have been in place for too short a period to draw any long-term conclusion. It would also be reasonable to suggest that moderate rates of taxes induce a culture of voluntary compliance; this may be a necessary but not a sufficient condition to sustain tax buoyancy in the long run. The initiatives toward moderation of tax rates have to be supported by a major reform of the tax administration system. The key elements would be to redesign busiTable 2C.5

Corporate Tax Elasticities

Fiscal Year

Corporate Tax Rate

Nonagricultural GDP at Factor Cost

1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99 1999–2000

–1.1287 –0.9242 –0.8157 –0.7974 –0.6742 –0.5489 –0.4158 –0.4533 –0.3981 –0.2847 –0.1421 –0.0532 0.0022 0.0621 0.0869 0.1501

1.5065 1.3696 1.3044 1.3696 1.3696 1.4087 1.2000 1.3500 1.3500 1.3500 1.2000 1.2000 1.1217 0.9130 0.9130 1.0044

India’s Fiscal Situation: Is a Crisis Ahead?

89

ness processes consistent with the state-of-the-art information technology, and to retrain and re-deploy existing manpower in order to effectively enable tax administration to detect and penalize noncompliance more effectively while rendering service to taxpayers for facilitating voluntary compliance.

3 State-Level Performance under Economic Reforms in India Montek S. Ahluwalia

The impact of India’s economic reforms on economic performance has been the subject of much academic study and public debate in India, but the focus has been largely on the performance of the economy as a whole or of individual sectors. The performance of individual states in the postreform period has not received comparable attention, and yet there are very good reasons why such an analysis should be of special interest. First, balanced regional development has always been one of the declared objectives of national policy in India, and it is relevant to ask whether economic reforms have promoted this objective. Second, India’s federal democracy is increasingly characterized by regionalization of politics, with politics at the state level being driven by state rather than national issues, and this makes the economic performance of individual states an issue of potential electoral importance. This is particularly so because liberalization has eliminated many of the controls earlier exercised by the central government and thereby increased the role of state governments in many areas that are critical for economic development. Finally, since state-level performance shows considerable variation across states, with many states recording strong growth in the postreform period, it is important to identify the reasons for their success in order to replicate it in other states. This paper attempts to document the performance of the major states in the postreform period 1991–92 to 1998–99 and to compare it with perforMontek Singh Ahluwalia is currently Director of the International Monetary Fund’s Independent Evaluation Office. At the time of writing this article, he was a member of the Planning Commission in the Government of India. The views expressed in the paper are those of the author and do not necessarily reflect the views of the government of India. Thanks are due to Shankar Acharya, Anne O. Krueger, Jairam Ramesh, and T. N. Srinivasan for helpful comments on an earlier draft.

91

92

Montek S. Ahluwalia

mance in the previous decade. It also seeks to explore the reasons for the differences in growth across states and to identify the critical policy issues that need to be addressed if the slow-growing states are to achieve more respectable growth rates in future. We note at the outset that there are severe data limitations that limit our ability to explain inter-state variations in performance. Nevertheless, we attempt to explore these issues to the extent possible, recognizing that in many cases we will raise more questions than we can answer. 3.1 Performance of the States: A Review The growth performance of the fourteen major states in the pre- and postreform period can be studied on the basis of the available data on the gross state domestic product (GSDP) for each state.1 A comment on data problems is appropriate at this stage. Ideally, the GSDP data series for individual states should be fully consistent with the national accounts estimates of gross domestic product (GDP), so that the disaggregated picture of economic performance at the state level corresponds with the picture for the country as a whole that emerges from the national accounts. This type of consistency is not possible at present. Time series data on the GSDP in each state are prepared by the statistics department of state governments, but these estimates do not add up to the GDP presented in the national accounts. The GSDP data prepared by the statistical departments of the states are used by the CSO as an input into national accounts estimation, but there are differences in methods of estimating the GSDP in different states, and the state GSDP series are not modified to make them consistent with each other and with the national accounts. The data problems associated with the GSDP series are important, but they should not deter us from using these data for analyzing state performance. Most Indian states are much larger than most developing countries, and the national accounts data of developing countries have similar problems, but this has not deterred development economists from comparing performance across developing countries and drawing lessons from intercountry variations. Following established academic tradition, we therefore acknowledge the problem, but proceed undeterred. 3.1.1 Growth Performance Table 3.1 presents the estimated growth rates of GSDP in the fourteen major states in the prereform period 1980–81 to 1990–91 and in the post1. Because of the special features of the Northeastern and other special category states and some gaps in the data for some of these states, they have been excluded from the analysis. The small states of Goa and Delhi have also been excluded, the latter having the additional special feature of being the capital. This section of the paper draws upon Ahluwalia (2000); However, the coverage has been extended to include data for 1998–99.

State-Level Performance under Economic Reforms Table 3.1

93

Rate of Growth of Gross State Domestic Product (percent per year)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Bihar Rajasthan Uttar Pradesh Orissa Madhya Pradesh Andhra Pradesh Tamil Nadu Kerala Karnataka West Bengal Gujarat Haryana Maharashtra Punjab

Combined GSDP of 14 states GDP (national accounts)

1980–81 to 1990–91

1991–92 to 1998–99

4.66 6.60 4.95 4.29 4.56 5.65 5.38 3.57 5.29 4.71 5.08 6.43 6.02 5.32

2.88 5.85 3.58 3.56 5.89 5.20 6.02 5.61 5.87 6.97 8.15 5.13 8.01 4.77

5.24 5.47

5.90 6.50

Source: Planning Commission.

reform period 1991–92 to 1998–99.2 The following conclusions are worth noting: 1. The growth rate of the combined GSDP of all the fourteen states taken together increased from 5.2 percent in the prereform period to 5.9 percent in the postreform period. This acceleration in the combined GSDP is similar to the picture that emerges from the national accounts, except that the postreform acceleration of GDP in the national accounts is much sharper. GDP grew at 5.5 percent per year in the first period, which was only marginally faster than the 5.2 percent growth recorded by the combined GSDP of the fourteen states. However, GDP growth accelerated to 6.5 percent in the second period, which was much faster than the 5.9 percent growth in the combined GSDP. The faster growth recorded in the national accounts probably reflects the impact of the revision in the national accounts GDP series introduced from 1993–94 onward. It is possible that if the GSDP data were revised similarly, the growth rates of GSDP of the different states in the second period would be correspondingly higher. We note that if such a revision were to affect some states more than others, it could also alter our assessment of the relative performance of states, but in the absence of specific 2. The growth rate for each state in each period is estimated based on a log-linear trend. Assuming that the underlying relationship is Y
A(1  r)t, we estimate the regression equation log Y
a  bt where b
log(1  r). The growth rate is then calculated as r
(antilog b) –1, where b is the regression estimate. Y is the current output; A is the initial output; r is the growth rate; and t is the time period between initial and current output.

94

Montek S. Ahluwalia

information that might have guided us on this issue, we assume that the adjustment would adjust growth rates upward across all states, leaving relative performance unaffected. 2. There is variation in growth performance across states in both periods, with some states growing faster than others, but the degree of dispersion in growth rates increased very significantly in the 1990s. The coefficient of variation of the growth rates increased from 0.15 in the first period to 0.27 in the second. The range of variation in the first period was from a low of 3.6 percent per year for Kerala to a high of 6.6 percent in Rajasthan, which gives a ratio of 1.8 between the highest and the lowest. In the second period, the range increased from a low of 2.9 percent per year for Bihar to a high of 8.2 percent for Gujarat, increasing the ratio to 2.8. 3. The increased variation in growth performance in the 1990s reflects very different behavior at different ends of the spectrum of per capita GSDP. Growth accelerated sharply for two states at the upper end of the spectrum, namely Gujarat and Maharashtra, but it actually decelerated in Bihar, Uttar Pradesh, and Orissa, all three of which were not only at the lower end of the per capita GSDP spectrum but also had relatively low rates of growth to begin with. The growth pattern in the 1990s therefore increased regional inequality, an aspect discussed at greater length in the next subsection. 4. Only four states achieved relatively strong growth with growth rates of GSDP in the 1990s above 6.0 percent in the second period. It is interesting to note that these states are fairly well distributed regionally, with Gujarat (8.2 percent) and Maharashtra (8.0 percent) in the west, West Bengal (7.0 percent) in the east, and Tamil Nadu (6.0 percent) in the south. In addition, Madhya Pradesh and Rajasthan in the north and Karnataka in the south all grew at 5.9 percent, which is almost at the 6 percent level. It is very likely that if the GSDP series were revised to reflect the changes made in the national accounts, the growth rates of all seven states would exceed 6 percent. Except for Rajasthan, all these states also show an acceleration in growth compared with the prereform period. An interesting feature of the performance in the 1990s is that the popular characterization of the so-called BIMARU states (Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh) as a homogeneous group of poor performers, a grouping originally proposed in the context of observed commonalities in demographic behavior, does not hold as far as economic performance in the postreform period is concerned.3 Bihar and Uttar Pradesh performed very poorly, growing much more slowly than the average, but the 3. The acronym BIMARU, taken from the initial letters for Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh, was a pun on the Hindi word Bimar, meaning sick, and was first used by Ashish Bose in the context of demographic analysis as these states displayed much higher fertility rates than other states in the country.

State-Level Performance under Economic Reforms

95

other two members of this group, Rajasthan and Madhya Pradesh, have performed reasonably well. Rajasthan shows a deceleration in growth of GSDP compared with the 1980s, but it remained a good performer in the 1990s, growing at about the average for all states. Madhya Pradesh, on the other hand, which had grown more slowly than the average in the 1980s, accelerated significantly in the 1990s. Simplistic perceptions about the role of geography in determining performance, such as the view that only the coastal states or the southern states have done well in the period of liberalization, are also not universally valid. Orissa is a coastal state, but its growth performance is very poor, while Madhya Pradesh and Rajasthan are both heartland states and have performed reasonably well. Only two of the southern states, Tamil Nadu and Karnataka, made it to the top six in terms of growth of GSDP in the 1990s. The southern states as a group have done well, but they were by no means the only beneficiaries of the growth acceleration witnessed in the 1990s. The remarkable performance of Gujarat and Maharashtra, both of which grew at over 8 percent per annum in the 1990s, a rate normally associated with “miracle growth” economies, deserves careful study. These states clearly benefited the most in the postreform period, but it is important to note that their superior performance was not the result of any conscious policy that limited the benefits of liberalization to these states, as was the case for example in China, where initially liberalization was deliberately limited to designated coastal zones. Their superior performance must be attributed primarily to the ability of these two states to provide an environment most conducive to benefiting from the new policies. Their experience, together with the experience of the other strong performers, should provide the basis for identifying the critical ingredients of success in accelerating growth, which should be emulated by others. The performance of Kerala in the 1990s also deserves special mention. Kerala has long been commended for its achievements in human development, especially education and health, but it has also been criticized for underperformance in economic growth.4 However, Kerala’s economic performance, which was relatively lackluster in the 1980s, appears to have improved markedly in the postreform period. From a GSDP growth rate of 3.6 percent in the 1980s, much below the average for the fourteen states, it accelerated to 5.6 percent in the 1990s. This was still below the average, but 4. For example, Dreze and Sen (1995), commenting on the experience of Kerala in the 1980s, have stated, “Kerala has been very successful in developing the social opportunities (related to widespread education, health care, land reforms, social security, etc.) that constitute the centrally important social conditions for having participatory growth. And, yet, Kerala has had, in fact, little participatory economic growth at home. The failure in this case has arisen not from any lack of participation but the slow growth of Kerala’s domestic economy. The roots of this failure include the continuation of over-regulated economic governance that has blighted the prospects of economic expansion all over India for many decades, the removal of which has met more resistance in Kerala than in most other Indian States” (197–98).

96

Montek S. Ahluwalia

Table 3.2

Annual Rates of Growth of Per Capita Gross State Domestic Product (percent per year)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

1980–81 to 1990–91

1991–92 to 1998–99

2.45 3.96 2.60 2.38 2.08 3.34 3.87 2.19 3.28 2.39 3.08 3.86 3.58 3.33

1.27 3.48 1.28 2.08 3.67 3.67 4.78 4.35 4.08 5.14 6.73 2.85 6.19 2.93

3.03

4.02

Bihar Rajasthan Uttar Pradesh Orissa Madhya Pradesh Andhra Pradesh Tamil Nadu Kerala Karnataka West Bengal Gujarat Haryana Maharashtra Punjab

Combined GSDP of 14 states

because of Kerala’s low population growth, its performance in terms of growth of per capita GSDP in the 1990s was actually better than the average (table 3.2). 3.1.2 Implications for Interstate Inequality The deceleration of growth in the poorer states witnessed in the 1990s has important implications for regional balance. Regional differences in per capita income levels have long been a matter of concern in India, and for good reason. The per capita GSDP of Punjab, the richest state, is five times that of Bihar at the other end of the spectrum. Balanced regional development has always been stated as an objective in India’s plans, and although this objective has never been quantified in terms of rates of convergence of per capita GSDP, or a reduction in regional inequality to some specified target in terms of one of the inequality measures, the objective surely implies that regional differences in per capita incomes should narrow with development, and in any case not widen. There are several studies that have sought to determine whether India’s growth process shows convergence in per capita GSDP over time.5 These studies deal with long-term trends, and the general conclusion (different studies have used different periods between 1960 and 1995) seems to be that there is no evidence of unconditional convergence but that there is evidence of conditional convergence. In other words, the long-term time paths of per 5. See, for example, Cashin and Sahay (1996); Nagraj, Varondakis, and Veganzones (1998); and Bajpai and Sachs (1996).

State-Level Performance under Economic Reforms Table 3.3

97

Trend in Inter-State Inequality Fiscal Year 1980–81 1981–82 1982–83 1983–84 1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99

Gini Coefficient 0.152 0.152 0.152 0.151 0.154 0.159 0.157 0.161 0.158 0.175 0.171 0.175 0.199 0.207 0.206 0.230 0.222 0.235 0.233

capita GSDP across states show convergence after allowance is made for differences across states in some of the initial conditions that affect growth rates, such as the share of agriculture and some measure of infrastructure development. Conditional convergence is of course quite consistent with divergence in per capita GSDP over certain periods. In this paper we are concerned not with convergence in the sense of underlying long term trends, but the actual behavior of per capita GSDP in the postreform period, compared with prereform behavior. From this perspective, the impact of the growth process on regional inequality in the 1990s is best seen by constructing a Gini coefficient for the total population of the fourteen states, assuming that all individuals within a state have a gross income equal to the per capita GSDP.6 This provides a measure of inequality in the total population of the fourteen states that ignores the inequality arising out of the unequal distribution within each state, and focuses only on the inequality that arises because of inter-state differences in per capita GSDP. As shown in table 3.3, the inter-state Gini coefficient was fairly stable up to about 1986–87 but began to increase in the late 1980s, and this trend continued through the 1990s. The increase in the Gini coefficient 6. Ideally, we should use per capita state income and not per capita state domestic product for constructing the Gini coefficient. An income concept would take into account net factor income from outside the state accruing to residents in the state. Unfortunately, data on net factor income accruing to each state are unavailable, making it impossible to construct an inequality measure based on state per capita income.

98

Montek S. Ahluwalia

from about 0.16 in 1986–87 to 0.23 in 1997–98 is very substantial, and fitting a time trend to the series shows a statistically significant positive slope. While inter-state inequality as measured by the Gini coefficient has clearly increased, the common perception that “the rich states got richer and the poor states got poorer” is misleading. Table 3.2, which presents growth rates of per capita GSDP of the different states in the two periods, suggests that the pattern is somewhat more nuanced. 1. It is not true that all the richest states got richer relative to the poorest states. Punjab and Haryana were the two richest states in 1990–91, but their growth rates of per capita GSDP not only were lower in the 1990s than in the 1980s, but in both cases actually fell below the national average. Except for Bihar, Uttar Pradesh, and Orissa, which grew even more slowly, all other states narrowed the distance between themselves and Punjab and Haryana. The deceleration in growth in Punjab and Haryana in the 1990s deserves closer study to understand the reasons for the loss of growth momentum in these states, which were among the good performers in the 1980s. 2. Maharashtra and Gujarat, which are in the high income group and were ranked just below Punjab and Haryana at the start of the 1990s, accelerated very significantly and achieved the fastest rates of growth in per capita GSDP. These states clearly pulled ahead of all other states. 3. Three of the poorest states, Bihar, Uttar Pradesh, and Orissa, which together account for about one-third of the population of the fourteen states, fared very poorly in the 1990s. It is important to emphasize that they did not actually become poorer, as they also had positive growth rates of per capita GSDP, but the growth rates were very low. In the case of Bihar and Uttar Pradesh, per capita GSDP growth was a little less than 1.3 percent per year, which was less than one-third of the national average. It is possible that growth rates of income may be higher because of remittances from migrant labor. It is also important to note that not all the poorer states performed badly. Rajasthan, for example, experienced fairly good growth in per capita GSDP, more than double that of the other poor states, though still below the national average. 4. Performance of four middle-income states (West Bengal, Tamil Nadu, Kerala, and Karnataka) was above the average, with West Bengal showing very strong growth. These states not only improved their position relative to the average, but also grew faster in terms of per capita GSDP than they did in the 1980s. 3.1.3 Implications for Poverty The difference in growth performance across states, with relatively low rates of growth for Uttar Pradesh, Bihar, and Orissa, have important implications for poverty reduction in India. India’s past experience at the national level shows that as long as GDP growth was modest, that is, between

State-Level Performance under Economic Reforms

99

3.5 and 4 percent up to the late 1970s, there was no significant reduction in poverty; the percentage of the population below the poverty line fluctuated, falling in good agricultural years and rising in bad, but with no trend decline. It was only after GDP growth accelerated in the 1980s that a trend reduction in poverty began to be noticed. Drawing on this experience, India’s poverty reduction strategy consisted of a two-pronged approach relying on an acceleration in growth to bring about a general improvement in living standards, supplemented by poverty alleviation programs directed at identified poverty groups that may not benefit sufficiently from the growth process. Extending this approach to the state level implies that poverty reduction in the major states requires rapid growth of GSDP, capable of generating a broad-based expansion in employment and income levels. This assertion needs to be modified to the extent that migration of labor from slowgrowing to faster-growing states allows the benefits of growth to “trickle down” across states through the flow of workers’ remittances. However, while labor migration is important, it can have only a limited impact, especially in the larger states. It certainly cannot substitute for acceleration of growth of the domestic economy in these states. The trends in poverty in individual states in the pre- and postreforms period can be seen from the official estimates of poverty at the state level and the all-India level presented in table 3.4. The planning commission bases these estimates on the so-called “large sample” surveys covering about 120,000 households, conducted periodically by the National Sample SurTable 3.4

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Bihar Rajasthan Uttar Pradesh Orissa Madhya Pradesh Andhra Pradesh Tamil Nadu Kerala Karnataka West Bengal Gujarat Haryana Maharashtra Punjab

All 14 states All India

Percentage of Population in Poverty 1983

1987–88

1993–94

1999–2000

52.22 34.46 47.07 65.29 49.78 28.91 51.66 40.42 38.24 54.85 32.79 21.37 43.44 16.18

52.13 35.15 41.46 55.58 43.07 25.86 43.39 31.79 37.53 44.72 31.54 16.64 40.41 13.20

54.96 27.41 40.85 48.56 42.52 22.19 35.03 25.43 33.16 35.66 24.21 25.05 36.86 11.77

42.60 15.28 31.15 47.15 37.43 15.77 21.12 12.72 20.04 27.02 14.07 8.74 25.02 6.16

43.80 44.48

39.92 38.86

36.25 35.97

26.43 26.10

Source: Planning Commission.

100

Montek S. Ahluwalia

vey Organisation (NSSO).7 Table 3.4 shows that the percentage of the population below the poverty line in the fourteen major states has declined steadily from 43.8 percent in 1983 to 26.4 percent in 1999–2000. The tenyear period from 1983–84 to 1993–94 saw a relatively modest reduction of about 7.5 percentage points in the percentage of the population below the poverty line, followed by a much larger decline of almost ten percentage points in the subsequent six-year period 1993–94 to 1999–2000. It is tempting to conclude that the faster growth in the postreform period led to a faster pace of reduction in poverty, but this conclusion needs to be qualified: The 1999–2000 survey may not be fully comparable with the earlier survey because of certain modifications in the method of collecting information on consumption.8 However, while the noncomparability may exaggerate the extent of the decline, the direction of movement is not in doubt. It is also consistent with the findings of household income surveys conducted by the National Council of Applied Economic Research and reported in Lal, Mohan, and Natarajan (2001). Table 3.4 shows a significant decline in poverty in the postreform period in all states except Orissa. Poverty in Punjab and Haryana, which was low to begin with, has become marginal at 6 percent and 9 percent, respectively. In Kerala, too, it is down to 12 percent. Significant gains have also been made in Tamil Nadu, Karnataka, Andhra Pradesh, and Rajasthan. Table 3.4 also shows a significant decline in poverty in Uttar Pradesh and Bihar despite relatively poor growth of GSDP in both states. This could reflect the impact of migrants’ remittances, although we do not have reliable data on the extent of migration from these states from the relevant income classes. It is also possible that the extent of the decline between 1993–94 and 1999–2000 is exaggerated because of the noncomparability problem. How7. The NSSO also conducts annual surveys—the so-called “thin sample” covering about 25,000 households—but the sample size is too small to provide reliable estimates of poverty for individual states. However, there has been an active debate on poverty estimates for the country as a whole in the postreform period, based on the thin sample surveys. Some scholars (e.g., Datt 1999; Gupta 1999) have commented that these surveys show a very marginal decline in poverty despite rapid growth. These conclusions have been challenged by other studies (e.g., Bhalla 2000; Natarajan 1998). For a review of these issues see Ahluwalia (2000). The results of the 1999–2000 survey, however, indicate that poverty has fallen over the period. 8. Earlier, information on household expenditure for all items was collected on the basis of a 30-day recall period. In the 1999–2000 survey, the recall period for durable goods was changed to 365 days. In the case of food, the survey has adopted two alternatives. In earlier experiments conducted in the annual thin sample surveys, the NSSO had experimented with 7day and 30-day recall periods for food, applying these periods to two different subsamples. The response on the basis of 7-day recall consistently showed higher food consumption, which was also more consistent with the national accounts. In 1999–2000, large-sample survey information based on the two different recall periods was sought from the same set of households. The 7-day recall still yields higher consumption, but the difference has narrowed. The estimates presented in table 3.4 are based on the 30-day recall period for food, which is comparable with the earlier surveys. However, it could be argued that questioning the same household about both the 7-day and 30-day recall has led to the 30-day estimate’s being adjusted upward, making it noncomparable with earlier surveys.

State-Level Performance under Economic Reforms

101

ever, it is important to note that even after allowing for the decline, the level of poverty in these states is still high and, indeed, India’s poverty problem is becoming increasingly concentrated in this area. In 1980, the three states of Bihar, Uttar Pradesh, and Orissa accounted for 37.5 percent of the total population below the poverty line in India, but by 1999–2000 this had increased to 46.0 percent. Continuation of the growth pattern observed in the 1990s, in which a region accounting for one-third of the population and the largest concentration of poverty derives very little benefit, while the rest of the country enjoys robust growth, presents obvious problems. It will exacerbate regional inequality with further concentration of poverty in a particular region, which is surely a recipe for political instability. The development strategy for the future must therefore ensure that the slow-growing states accelerate to a respectable growth of GSDP of, say, 6 percent per year. This would ensure per capita GSDP growth of around 4 percent, which is certainly needed if there is to be a significant reduction in poverty in these states over the next ten years. 3.2 The Determinants of Growth in the States In this section, we attempt to explain the variation in growth across states, especially in the postreform period. First, we consider a question which has been the subject of much discussion in India, that is, whether the economic reforms are in some ways directly responsible for the divergent pattern of growth witnessed in the 1990s. Thereafter, we seek to explain growth of GSDP in individual states in terms of the familiar explanatory variables conventionally used in such analyses, namely the level of investment in states, the quality of human resources, and infrastructure endowments. The reader is warned that data limitations prevent us from proceeding very far with this approach, but it is worth exploring the limits of what is possible with the data available. 3.2.1 Have Economic Reforms Caused Regional Inequality? The rationale of the various economic reforms initiatives at the national level, such as abolition of industrial licensing and other types of control over private investment, liberalization of trade policy, financial-sector reforms, and so forth, was that they would increase efficiency and lead to higher factor productivity. Since these policies are generally applicable to all states, there is a natural presumption that they would provide efficiency gains for all these states, which should increase the growth potential of each state. In this view, the reforms generate potential gains for each state, and while some states may benefit more than others, the reforms do not hurt any states. If some states have decelerated in the 1990s, this must arise from other factors, especially differences in policies followed by individual states.

102

Montek S. Ahluwalia

However, it is important to recognize that even though the reforms themselves are nondiscriminatory, they will affect states differently because of differences in state-specific characteristics, and this could lead to a deceleration in some states. For example, opening the economy to foreign trade can be viewed as improving the efficiency of resource use in the economy as a whole and thus potentially beneficial to all states, but if some states have a greater comparative advantage in exports, while others have developed a production structure excessively dependent on uncompetitive importsubstituting industries, the process of opening up could well lead to an acceleration in growth in the former in the short run while slowing it down in the latter, as investment is likely to move from the latter to the former, at least in the short run. This implies, of course, that some of the factors that make for greater competitive advantage are immobile in the short run. However, over a period of time, production structures, including factors that account for comparative advantage in particular states, can change and states initially excluded from acceleration can catch up. The dismantling of industrial licensing provides another example in which economic reform could generate differential outcomes, leading to a deceleration in some states. The abolition of licensing eliminated the central government’s ability to spread investment evenly across the country, which was a common practice earlier, leading to fragmented capacities, which not only were suboptimally located but also could not benefit from economies of scale. With liberalization of investment control and much stronger pressure of competition, especially competition from imports, investment size began to be determined on economic grounds, and location was also determined to a much greater extent by economic considerations. It is very likely that in practice this led to a reallocation of investment in favor of states perceived as having better infrastructure facilities, better labor skills and work culture, and a more investor-friendly environment. The resulting reallocation of investment in the postreform period could lead to a substantial increase in investment in the better-performing states, and a consequent increase in their growth rate, with a corresponding reduction in investment in less well endowed or well governed states and a deceleration in their growth. The impact of economic reforms at the national level on the growth rate of individual states therefore depends on the net effect of two sets of forces. There are the positive efficiency effects of reforms, which are potentially available to all states and which by themselves should improve factor productivity and growth in all states. However, there is also a potential reallocation of resources across states in search of efficiency. This reallocation may be driven by natural comparative advantage, such as a coastal location for a petrochemical complex dependent on imported feedstock, or by initially favorable conditions that are not immutable, such as better infrastructure or a more favorable state policy environment. One must recognize

State-Level Performance under Economic Reforms

103

that such reallocation is necessary if the efficiency benefits of the reforms are to be realized for the country as a whole, but it can lead to negative effects in particular states. In certain circumstances these negative effects can swamp the positive effects, in which case economic growth could actually fall. The solution to this problem does not lie in backtracking from reforms, or even slowing them down. On the contrary, the compulsions of globalization are such that India must look to every possible means of enhancing efficiency in resource use in order to increase competitiveness. Unless this is done, it will certainly not be possible to sustain the growth achieved in the postreform period, let alone accelerate it further. The better positioned states must therefore be allowed, and indeed even encouraged, to perform up to their full potential, and the lessons learned from their success should be spread elsewhere. However, the states that have not benefited from the reforms, and indeed may even have suffered because of a reallocation of investment resources toward other better endowed states, must be assisted by addressing the specific deficiencies that are holding them back. To do this, we need to have some idea of the critical determinants of growth at the state level, given the existing framework of national policy, and how these determinants can be influenced through policy. 3.2.2 Investment Ratios at the State Level The rate of investment is generally regarded as one of the most important factors explaining growth in any economy, and it is therefore appropriate to consider whether inter-state differences in growth are associated with differences in the rate of investment in individual states. This is particularly so in view of the possibility, discussed above, that economic reforms in certain circumstances could lead to a reallocation of investment away from some states and toward others. If this is indeed the explanation for the divergence in growth rates observed in the 1990s, with some states accelerating while others decelerated, it should be reflected in divergent movements in the investment ratio. Unfortunately, data on the level of investment in individual states, comparable with the investment data at the national level obtained from the national accounts, are simply not available. The only information on investment expenditures at the state level available at present is the capex database compiled by the Centre for Monitoring the Indian Economy (CMIE). These data suffer from a number of problems, which are worth listing at the outset. First, the data exclude investment in the unorganized or household sector, which is about 33 percent of total investment in the economy. The capex data refer only to expenditure on government projects (center, state, and local) and on private-corporate sector projects that are currently being implemented. Second, the investment expenditure reported is not the expenditure by each project in a year, but the total expenditure for complet-

104

Montek S. Ahluwalia

Table 3.5

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Bihar Rajasthan Uttar Pradesh Orissa Madhya Pradesh Andhra Pradesh Tamil Nadu Kerala Karnataka West Bengal Gujarat Haryana Maharashtra Punjab

All 14 states

Investment Activity in the States in 1995–96 (as percentage of GSDP) Government Projects

Private Projects

All Projects

17.02 19.86 20.65 48.55 36.56 21.78 7.18 17.25 18.13 17.23 27.40 17.25 10.95 12.28

2.68 9.27 12.22 15.17 6.51 15.87 17.84 1.77 23.93 12.21 57.68 4.81 17.80 6.42

19.70 29.14 32.87 63.72 43.07 37.65 25.02 19.02 42.06 29.45 85.08 22.06 28.76 18.70

19.06

16.45

35.51

Source: Capex database, Centre for Monitoring the Indian Economy.

ing each project. The investment expenditure therefore captures investment made in previous years in all ongoing projects and also the expected investment in future years on these projects. Finally, the data are collected from multiple sources, including newspaper accounts and press releases, and the investment values are therefore not based on comparable prices. These are important limitations, but since the capex database provides the only information available on investment expenditures at the state level, it is worth examining in some detail. Table 3.5 presents the investment expenditure reported in the capex database for government projects and private corporate sector projects in each state at the end of the fiscal year 1995–96, expressed as a ratio of GSDP in 1995–96 in current prices. These investment ratios are much higher than the investment ratio derived from the national accounts because the investment refers not to the annual expenditure but the total expenditure on all ongoing projects.9 However, if the exaggeration is uniform for all states, the investment ratio de9. Total investment in government-sector projects reported in table 3.5 for all the fourteen states is more than 19 percent of combined GSDP, whereas the ratio of public investment to GDP for the country as a whole in the national accounts is only 6.5 percent. Similarly, the ratio of private-sector investment expenditure for all states is almost 16.5 percent of their combined GSDP, whereas private-sector investment in the national accounts is only 8.6 percent of GDP. The larger discrepancy in the case of public-sector projects is probably due, in part, to the fact that these projects are typically of longer gestation, making the use of cumulative figures more distorting; and in part to the fact that because public-sector projects are typically underfunded, leading to a proliferation of projects that take much longer to complete than they should. This creates a situation in which a large number of projects are in progress at any given time, magnifying the degree of exaggeration caused by using total investment expenditure over the life of the project.

State-Level Performance under Economic Reforms

105

rived from the capex data could be used as a proxy for the underlying investment ratio. Since the data series is available only from 1995 onward, it cannot be used to ascertain whether the rate of investment in different states behaved asymmetrically in the postreform period, as speculated above. However, the available data can be used to test whether variations in investment across states are correlated with variations in growth. We note that the many limitations listed above warrant more than the usual caveats for the results reported. Three separate regression equations were estimated in which the dependent variable in each case was g
growth of GSDP in 1991–92 to 1998–99, while the independent variables were IPUB (cumulative expenditure in public-sector projects as a ratio of GSDP), IPVT (cumulative expenditure in private-sector projects as a ratio of GSDP), and ITOT (the sum of IPUB and IPVT). The results are reported below (t
ratios in parentheses):10 (1)

g
6.263 – 0.0349 · IPUB R2
0.06 (0.86)

(2)

g
4.609  0.0635 · IPVT R2
0.32 (2.42)

(3)

g
4.686  0.024 (1.05)

ITOT R2
0.08

There is no significant relationship between the variation in growth across states and the variation in the capex public investment ratio. On the other hand, the private investment ratio proves to be highly significant and has the expected positive sign. This variable alone explains almost one-third of the variation in growth across states. It would be wrong to conclude from the lack of a significant relationship between growth and public investment that public investment is not important. It is entirely possible, as we shall argue, that acceleration of growth in future requires increased public investment in critical areas, especially economic and social infrastructure. However, the regression equation does not reveal any significant relationship between growth and the measure of public investment available to us. This may be due to the fact that the capex public investment data are subject to large errors because of factors mentioned earlier. The inclusion of future investment in unfinished projects in particular is likely to introduce a larger error the more poorly managed the investment program. Not only is there a data error in such cases, but those situations, in which there is underfunding of public-sector investment proj10. The investment data pertain to the year 1995, but because they include the total investment over the life of all projects under implementation in that year, they probably come close to measuring investment activity over most of the postreform period.

106

Montek S. Ahluwalia

ects leading to a proliferation of incomplete projects, are precisely those in which such public investment expenditure as does take place is also infructuous, and not reflected in a commensurate increase in growth. The positive and significant coefficient on the private investment variable, on the other hand, conforms with the expectation that private investment matters and that states with a higher ratio of private investment to GSDP are likely to experience faster growth. It should be noted that the capex private investment data are subject to the same data infirmities that affect public investment data, and yet the variable turns out to be highly significant. Since private investment is subject to greater financial discipline, the data error arising from a large number of unfinished and underfunded projects is likely to be much smaller. Besides, private investment may also be more directly correlated with growth because of greater efficiency of resource use. These results certainly suggest that private investment is one of the principal drivers of growth, and slow-growing states must therefore place special importance on identifying the factors that would stimulate private investment. 3.2.3 Plan Expenditure In the absence of reliable data on public investment, the only substitute available is the size of plan expenditure. Plan expenditure is not identical to public investment, but it has the advantage that data are available on an annual basis.11 Plan expenditure is undertaken by both the central government and the state government, and what is relevant for the development of a state is the volume of plan expenditure in the state by both the center and the state. Unfortunately, while data on total plan expenditure by the central government are readily available, they cannot be disaggregated according to the state in which the expenditure was incurred. The only information available on plan expenditure in a state therefore relates to the state plan. Since a great deal of attention is focused on the size of state plan expenditure in public discussion in the performance of individual states, it is worth exploring the relationship between state plan expenditure and growth of GSDP. Table 3.6 presents the average ratio of plan expenditure to GSDP in each state in the 1980s and compares it with the average ratio in the 1990s. The following features are worth noting: 1. State plan expenditure as a percentage of GSDP has declined in almost all the states (Rajasthan and Karnataka are the only exceptions). The percentage for the fourteen states taken together declined from 5.7 percent 11. Plan expenditure in Indian parlance refers to expenditure on new projects taken up during a particular five-year period. However, it includes both capital expenditure and current expenditure on the project. Investment expenditure is therefore typically less than plan expenditure.

State-Level Performance under Economic Reforms Table 3.6

107

Plan Expenditure as Percentage of Gross State Domestic Product Average

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Bihar Rajasthan Uttar Pradesh Orissa Madhya Pradesh Andhra Pradesh Tamil Nadu Kerala Karnataka West Bengal Gujarat Haryana Maharashtra Punjab

All 14 states

1980–81 to 1990–91

1991–92 to 1997–98

6.20 5.89 6.33 7.41 7.39 5.70 6.19 5.22 5.61 3.56 6.52 6.41 5.68 5.63

2.87 6.54 4.56 7.10 4.97 4.28 4.60 4.99 6.49 2.70 4.51 3.94 3.97 3.94

5.69

4.50

Source: Planning Commission. Data on plan expenditure at the state level for 1998–99 were not readily available and therefore have not been included in computing the average for the 1990s.

in the 1980s to 4.5 percent in the 1990s. Since the current expenditure component of the plan has increased over this period, the decline of 1.2 percentage points in state plan expenditures as a percentage of GSDP indicates an even larger decline in public investment by state governments over the period. 2. The decline in plan expenditures as a percentage of GSDP is not a phenomenon unique to the slower-growing states. The drop is the largest in Bihar, but Gujarat and Maharashtra, two of the best performers, also show a significant decline, as do other good performers such as Madhya Pradesh, Tamil Nadu, and West Bengal. 3. There is no obvious relationship between the ratio of state plan expenditure as a percentage of GSDP and growth performance across states in either decade. Orissa, which had the highest ratio of state plan expenditure to GSDP at 7.10 percent in the 1990s, had a GSDP growth rate of only 3.25 percent. West Bengal, with the lowest plan ratio of 2.70 percent, had a relatively robust growth of 6.9 percent. Maharashtra, which was the second fastest growing state, had an average plan ratio of only 3.97 percent, well below the average. Gujarat, which was the fastest-growing state, had a plan ratio about equal to the average. The lack of any significant relationship between the size of the state plan in relation to GSDP and growth of GSDP is borne out by the following re-

108

Montek S. Ahluwalia

gression equations, in which P
average ratio of plan expenditure to GSDP in the relevant period. 1980–81 to 1990–91: (4)

g
5.2084 – 0.0049 · P (0.02)

R2
0.01

1991–92 to 1997–98: (5)

g
6.0724 – 0.1149 · P (0.33)

R2
0.01

The absence of any significant impact of plan size on growth is sobering, considering the attention focused on the size of state plans as instruments of development. One reason for this could be that it is the investment component of state plans that is potentially relevant for growth, and this component has declined steadily over time until it now accounts for only about half of state plan expenditure. Since state plan expenditures amount to about 4.5 of percent of the total GSDP of all fourteen states, this means that investment in the state plans is only about 2.25 percent of GSDP, or only about 10 percent of the total investment in the economy. In other words, state plan expenditures can be very important for certain sectors, but they are a small part of total investment in the state, and this explains the lack of any significant relationship with growth. It is also true that many plan programs are ill-designed and indifferently executed. There is an accumulation of evidence that many public expenditure projects at the state level are ineffective in promoting their stated economic and social objectives, which makes their contribution to growth and development highly questionable. 3.2.4 Human Resources The quality of human resources, broadly defined to mean the educational attainment and skill level of the labor force, is another factor generally regarded as a critical determinant of growth. We should expect that states with superior availability of human skills, and more rapid growth in these skills, are more likely to have higher per capita GSDP and to experience faster growth. However, since data on the educational and skill characteristics of the labor force are simply not available, the literacy rate of the population is commonly used as a proxy for the quality of human resources. The data on literacy are summarized in table 3.7. Table 3.7 confirms that literacy in the slow-growing states of Uttar Pradesh, Bihar, and Orissa is indeed very low. However, the poor growth performance of these states cannot be explained solely by the low levels of literacy. The situation in Madhya Pradesh, Rajasthan, and Andhra Pradesh at the start of the decade was only marginally better, and yet these states showed a much better performance in the 1990s. Estimating a regression

State-Level Performance under Economic Reforms Table 3.7

109

Total Literacy Rate 1981

1991

1997

Bihar Rajasthan Uttar Pradesh Orissa Madhya Pradesh Andhra Pradesh Tamil Nadu Kerala Karnataka West Bengal Gujarat Haryana Maharashtra Punjab

26 24 27 34 28 30 47 70 38 41 44 36 47 41

38 39 42 49 44 44 63 90 56 58 61 56 65 59

49 55 56 51 56 54 70 93 58 72 68 65 74 67

All India

36

52

62

Source: Planning Commission.

equation relating growth in each period to the percentage of literacy L in the base year of the period yields the following results. 1980–81 to 1990–91: (6)

g
6.218 – 0.0272 · L

R2
0.16

(1.49) 1991–92 to 1997–98: (7)

g
2.735157 + 0.0513 · L

R2
21

(1.76) The literacy variable has the wrong sign and is not significant in the first period. It has the right sign in the second period, but the level of significance, although greatly improved, is still low. Using literacy in the base year of each period to explain variations in growth amounts to explaining growth in terms of a stock variable. We have also used the change in literacy in each period as an explanatory variable, but this does not yield a significant relationship. One could argue that the role of human skills in promoting growth is not independent of the level of investment and that the two interact with each other to generate positive responses. With fourteen observations, we have too few degrees of freedom to use additional explanatory variables. We therefore estimated a regression equation relating growth to a composite variable obtained by multiplying each of the capex investment ratios with the literacy rate in the base year of the postreform period. The multiplicative form implies that the response of growth to a higher investment rate is

110

Montek S. Ahluwalia

greater the larger the literacy variable, thus building in a positive interaction effect. The results obtained are presented below. (8)

g
3.945 – 0.1686 ·(IPUB L) (1.79)

R2
0.21

(9)

g
4.513 0.1918 ·(IPVT L) (2.80)

R2
0.40

(10)

g
3.89  0.0887 (ITOT L) (2.97)

R2
0.42

The introduction of a composite variable that allows for an interactive effect clearly improves the explanatory power of each of the investment variables. Interestingly, the R2 obtained from equation (10), which uses the total investment ratio to construct the composite variable, is higher than the R2 in equation (9), which uses only the private investment ratio. 3.2.5 Quality of Infrastructure The quality of infrastructure is widely regarded as an essential determinant of growth in the states. Infrastructure in this context is clearly a multidimensional feature. Agricultural growth depends upon rural infrastructure such as the spread and quality of irrigation, development of land, extent of rural electrification, and the spread of rural roads. Nonagricultural growth depends critically upon sectors such as electric power, road and rail transportation, ports and airports, and, increasingly, telecommunications. Good infrastructure not only increases the productivity of existing resources going into production and therefore helps growth, but also helps to attract more investment, which can be expected to increase growth further. The CMIE has produced a composite index of the relative infrastructure capacity of different states based on thirteen separate components.12 The values of composite index for different years are summarized in table 3.8 (the individual components are listed in the footnote to the table). The relative index values for individual states conform with some expectations but also contain some surprises. Bihar fits the pattern of expectations and scores lowest on infrastructure. Its relative position has also deteriorated over time. Somewhat surprisingly, however, Uttar Pradesh has a higher value for the index than the average for the country and scores higher than Andhra Pradesh, Karnataka, and West Bengal, all states that have grown markedly faster. 12. The thirteen variables are per capita electric power; percent of villages electrified; railway route length per 000 square kilometers; surfaced road length per 000 square kilometers; unsurfaced road length; handling capacity of major ports; gross irrigated area as a percentage of cropped area; and tele-density plus the following per lakh of population: bank branches, post offices, primary schools, hospital beds, and primary health centers. Each indicator is computed for each state relative to the all India average of 100. The composite index is the weighted sum of individual indices. For details see CMIE (1997).

State-Level Performance under Economic Reforms Table 3.8

111

Relative Infrastructure Development Index

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Bihar Rajasthan Uttar Pradesh Orissa Madhya Pradesh Andhra Pradesh Tamil Nadu Kerala Karnataka West Bengal Gujarat Haryana Maharashtra Punjab

All India

1980–81

1991–92

1996–97

83.5 74.4 97.7 81.5 62.1 98.1 158.6 158.1 94.8 110.6 123.0 145.0 120.1 207.3

81.7 82.6 102.3 95.0 71.5 96.8 145.9 158.0 96.5 92.1 122.9 143.0 109.6 193.4

77.8 83.9 103.8 98.9 74.1 93.1 138.9 155.4 94.3 90.8 121.8 137.2 111.3 185.6

100.0

100.0

100.0

Source: Centre for Monitoring the Indian Economy.

Testing for a statistically significant correlation between growth across states and the base year value of the composite infrastructure index INF, we have estimated the following equations. 1980–81 to 1990–91: (11)

g
5.118  0.0005 · INF (0.08)

R2
0.00

1991–92 to 1997–98: (12)

g
5.183  0.0021 · INF (0.24)

R2
0.00

In order to test for possible interaction effects between private investment and infrastructure, we also estimated equations using a composite variable that is the product of the investment ratio in the 1990s and the infrastructure variable. The composite explanatory variable is significant but does not add to the explanatory power of the investment variable. The absence of any significant relationship between growth and infrastructure is somewhat disappointing, although it can be explained by the fact that some of the thirteen variables included in the CMIE infrastructure index are not very relevant in explaining growth, for example, the number of post offices, hospital beds, or primary health centers per hundred thousand of the population. The quality of road connectivity is also poorly captured by the road density within a state, since hinterland states may suffer from disadvantages because of poor connectivity through other states. We have also tested for the impact of each of the individual components

112

Montek S. Ahluwalia

of the infrastructure index on growth in the states by estimating separate regression equations, using each of the thirteen individual components as explanatory variables. We find a significant relationship between growth and three of the individual indices in the second period, specifically the percentage of villages electrified in the base year V, per capita energy consumption E, and telecommunications density (or tele-density [T]). (13)

g
0.225 – 0.0596 · V (2.16)

R2
0.28

(14)

g
3.896  0.018 · E (2.05)

R2
0.26

(15)

g
3.54  3.3013 T (3.09)

R2
0.44

The positive significant relationship between growth and the two electricity-related indices and the tele-density index are broadly in line with expectations. The equation with tele-density has the highest R2, which will no doubt cheer telecommunications enthusiasts, but we hasten to caution that although telecommunication is undoubtedly important (as an efficiencyenhancing and therefore growth-promoting factor), this result should not be misread to imply that telecommunication is all that matters! We note that the absence of a positive relationship in the first period between growth and any of the infrastructure variables (including the three discussed above) remains a puzzle. The statistical results presented in this section are clearly mixed. They provide welcome confirmation that variations in the private investment ratio are positively and significantly correlated with variations in growth. They also provide some confirmation that certain elements of infrastructure, and to some extent also literacy, are associated with variations in growth. They also suggest that public investment and state plan expenditure are not nearly as obviously correlated with growth as many would have expected. While this may reflect data limitations, it also suggests the need for some soul-searching on the effectiveness of these expenditures. Needless to say, all these conclusions, including the lack of a significant relationship in some cases, are subject to the general qualification that the data available are far from ideal. Much more work needs to be done in improving the data available on possible factors that may help explain the variations in growth across states. 3.3 Toward a Strategy for Slow-Growing States In this section, we consider some of the policy issues that need to be addressed if the growth rate of the slow growing states is to be raised to a minimum of, say, 6 percent. Doubling the rate of growth in a group of states

State-Level Performance under Economic Reforms

113

with a population exceeding 300 million is obviously not an easy task. Much of the responsibility for such acceleration lies with the state governments and will require a major reorientation of policy in these states. This reorientation will have to take place within a framework defined by the evolving economic reforms, which have been under way for the past decade and which are expected to be continued and further strengthened. The slower-growing states must therefore devise a strategy that recognizes that the economy will become more competitive internally and also more open to both foreign trade and foreign investment. The critical drivers of growth in this environment will be private investment and improvements in factor productivity. State policies must therefore focus on how private investment can be stimulated and supported. Since private investment is potentially mobile across states, all states must compete with each other to attract private investment. This is true not just of foreign investment but also of domestic corporate investment. Noncorporate investors and small business are seen to be less mobile than corporate investors, but it must be recognized that concentrations of large corporate investment become a hub around which smaller noncorporate investment also flourishes. 3.3.1 Development of Economic and Social Infrastructure The most important instrument through which government policy can help to accelerate development in a liberalized economy that relies upon private investment to achieve growth is the provision of basic economic and social infrastructure. The poorer and slower-growing states generally lag behind the better-performing states in this area, and in a competitive environment this puts them in a disadvantaged position relative to the more advanced states. Because of the constitutional division of powers between the center and the states, some of the infrastructure needs fall exclusively in the area of the central government, for example, railways, national highways, telecommunications, major ports, and airports. Infrastructure needs in these sectors must be met either directly through increased central public investment or, when private investment is also feasible, by a combination of public and private investments. However, a large part of what is needed by way of infrastructure in individual states either falls in the exclusive area of responsibility of the state government (i.e., irrigation) or in what is described in the constitution as the concurrent list, as is the case for education and electric power. Both the center and the states can legislate in these areas, and state laws must be consistent with central laws, but the delivery system in practice is generally in the hands of the state government. Agriculture is important for all states, but it is especially so for the poorer states, which are more dependent on agriculture. All the relevant infrastructure needs of agricultural development—irrigation, land development and water management programs, rural road connectivity, rural electrification, and so forth—fall in the area of the state governments. This is also the

114

Montek S. Ahluwalia

case with social infrastructure, that is, the provision of health and education services, both areas in which the poorer states have exceptionally large gaps and improvements are needed to stimulate growth. State governments are also responsible for many of the critical infrastructure requirements of industrial and commercial development. The availability of power at an appropriate price and of acceptable quality is a critical requirement for industrial and commercial development, and this is also a state government responsibility. The generation transmission and distribution of power in all the major states is a state monopoly operated by the state electricity boards (SEBs). The financial position of the SEBs has deteriorated massively over time because of a combination of operational inefficiency and irrational electricity pricing, with very low electricity tariffs for farmers and household consumers cross-subsidized by very high electricity tariffs on industrial and commercial users. Operational inefficiencies are particularly marked in distribution where corruption is widespread, leading to under-billing for electricity consumed. The resulting financial difficulties of SEBs have led to inadequate investment in both generation and distribution, leading in turn to power shortages, erratic voltage, and unreliable supply. Major reforms in the power sector are desperately needed in all states to bring about rational tariff fixation and to create stronger incentives to improve efficiency at all levels. Fortunately, the need for reforms in this area is now well recognized and two of the slower-growing states— Uttar Pradesh and Orissa—have actually commenced the process, but progress as yet has been slow. This is clearly an area that must have very high priority. Urban infrastructure is also an important precondition for attracting private investment, especially foreign investment. This, too, is entirely a state government responsibility, and the slow-growing states suffer from a severe competitive handicap in this area. Improvements in urban infrastructure must, therefore, be an area of priority attention for state governments wishing to attract private investment. In practice, infrastructure development calls for additional financial resources as well as improvement in governance, which would ensure that the resources are well spent. Some suggestions in this regard are offered in the next two sections. 3.3.2 The Problem of State Finances As recently as 1990–91, several states had a positive balance from current revenues (BCR), which contributed at least a modest surplus, supplemented by borrowings to finance state plan expenditures.13 This balance has now turned negative for all states, which means that state governments have 13.The BCR is the surplus of current revenues over nonplan current expenditure. When this is positive, it contributes (along with borrowed resources) to finance the plan.

State-Level Performance under Economic Reforms

115

to borrow to finance the negative BCR and then borrow even more to finance the plan. The extent of fiscal stress in the states is perhaps best reflected in the fact that the states are resorting to larger and larger volumes of borrowing: The gross fiscal deficit of the states has increased from 3.2 percent of GDP in 1990–91 to 4.3 percent in 1998–99. Yet, as we have seen, plan expenditure as a percentage of GDP has declined in almost all states. In the process, there has been a steady buildup of debt, which in turn has generated a rising interest burden. These problems are not unique to the poorer states, but they are almost certainly more severe in these states, and corrective action is therefore more urgent. The steps that need to be taken to restore financial viability in the states are well known, though that does not make them any easier. 1. Direct and indirect subsidies provided by state governments, most of which are not well targeted, have become unsustainable. For example, the average tariff rate for electricity supplied to agriculture is around 25 paise per kilowatt-hour (kwh) for all states (some states have actually made it free), whereas the average cost of supplying power is Rs. 2.81 per unit. Irrigation charges at present cover only around 20 percent of the maintenance costs of the system, to say nothing of capital charges. Fees in higher education have not been raised for several decades, with the result that income from this source has declined from 20 percent of total costs in the 1960s to less than 6 percent. Public-sector road transport services incur large losses. Health services, including hospital care in public hospitals, are very heavily subsidized. Since the state budgets are unable to provide sufficient funds to the departments providing these services, the result is that the quality of services provided has deteriorated. An increase in user charges in all these areas is urgently needed to reduce the financial burden of providing these services. 2. The SEBs are clearly the largest drain on the system. In 1992–93, SEB losses were Rs. 2,725 crores, or 10 percent of total state plan expenditure in that year. They have increased to around Rs. 25,000 crores in 1999–2000, or 30 percent of total state plan expenditure! The need for reforms in this area has been mentioned earlier. The inefficiency in the distribution segment is a major problem, especially the theft of power through tampering with meters with the connivance of the distribution staff. While standards of governance in the public-sector distribution system can be improved significantly, privatization of distribution is probably the best way of minimizing such losses. Understandably, privatization is strongly resisted by vested interests, and many states are therefore reluctant to accept this as an objective, preferring instead to focus, at least initially, on improving the efficiency of the public sector system. However, many state governments such as Karnataka, Andhra Pradesh, Gujarat, Uttar Pradesh, and Orissa have initiated the first steps toward privatization.

116

Montek S. Ahluwalia

3. Public-sector enterprises (PSEs) have proliferated at the state government level and many of them are little more than vehicles for creating jobs at all levels.14 There are 1,071 public sector enterprises in the various states, and of these only 247 are profit-making. Most state PSEs are unlikely to yield significant resources from privatization proceeds, but privatization could at least help avoid recurring losses that are otherwise a burden on the budget. There are some enterprises, such as tourism corporations running hotels, or cement factories and sugar plants, that could be privatized with some budgetary gain. 4. The tax administration needs to be massively modernized in all states to create simple systems with transparent administrative procedures and an honest tax administration. States often complain that they do not have sufficient taxing power, but in fact they have not used the powers available in many areas. The taxation of agriculture, for example, is constitutionally a state subject, but states have left this tax base untouched, with no state seeing fit to levy an agricultural income tax even on large farmers. Land revenue, which is a form of agricultural taxation, and could be a substitute for agricultural income tax, has been reduced over the years to negligible levels. 5. Most states suffer from deteriorating urban infrastructure because municipal tax revenues are inadequate to finance infrastructure, and user charges for most services are very low. Urban property taxation is an important source of municipal revenue in most countries, but the system in most Indian states is hopelessly outdated, with poor valuation practices leading to very low revenues. A major modernization in the system of property taxation is urgently needed. 6. Bureaucracy in the states has proliferated to a much greater extent than in the center. This not only imposes a financial strain, but also perpetuates inefficiency and sluggishness in the system. Agencies responsible for delivering economic and social services face a situation in which their budgets are almost exhausted after paying the salaries of a bloated bureaucracy, leaving little or nothing to meet the minimum non-salary cost of delivering the services that the programs are meant to provide. It is necessary to downsize government as a whole and use the resources thus released to increase expenditures in critical areas such as health and education that are currently underfunded. Several states have recognized this problem and declared the intention of reducing the scale of the government, primarily by reducing recruitment to fill vacancies caused by normal retirement. The seriousness of this commitment and its impact on the size of the bureaucracy can only be evaluated over time. As pointed out earlier, these problems affect all states to varying degrees, but they are much more severe in the poorer states. Paradoxically, it is diffi14. In many states, political personalities are appointed as chairmen of state corporations and given the rank of cabinet minister in the state government.

State-Level Performance under Economic Reforms

117

cult to believe that the poorest states will actually take the lead in this area. They are more likely to follow the lead of the more advanced states, but this only means their performance will continue to lag behind the others. 3.3.3 The Policy Environment and Governance There is general agreement that growth depends heavily upon the efficiency of resource use, which in turn is determined by the overall policy environment and the quality of governance. There are no objective measures of the quality of governance, but impressionistic evidence suggests that the slower-growing states clearly lag behind the others. Good governance affects growth in several ways. First, it has a direct impact on the effectiveness with which public-sector developmental programs in the state are implemented. Poor administration and corruption (the two are in fact intimately linked) are now widely recognized as major problems that reduce the effectiveness of many government programs. Since additional public investment in the infrastructure and social sectors is an important part of the growth acceleration strategy for poorer states, it follows that parallel improvements in governance at the state level are needed to ensure that the resources provided for this purpose are well spent. In many cases, improving the effectiveness of public expenditure requires decentralized control over the programs with much greater people’s participation. There are many successful examples of decentralization in states such as Kerala, Karnataka, Andhra Pradesh, and Madhya Pradesh. These experiments need to be replicated in the slower-growing states, such as Uttar Pradesh, Bihar, and Orissa. Another channel through which the quality of governance at the state level can stimulate growth is by making the policy environment more business-friendly. While the economic reforms have reduced the burden of central government controls on investment activity, there is a need to introduce similar liberalization at the state level. Entrepreneurs setting up an industrial unit typically need as many as thirty separate permissions from various state government departments responsible for state level clearances, such as those related to environment regulations, labor welfare regulation, utilities, health, sanitary and safety inspection, sales tax, and so on. Each interface with a separate part of the bureaucracy subjects the entrepreneur to the triple vicissitudes of harassment, delay, and corruption. The high transaction costs are particularly onerous for small business, which is precisely the group most state governments are otherwise keen to promote. One of the positive developments in recent times is that many states have taken initiatives in this area and have introduced simplified procedures and one-window arrangements to improve the business climate. However, these experiments are relatively recent, and the lead has been taken by the betterperforming states. The poorer-performing states have generally lagged far behind the others in this dimension. Sweeping reform of these regulatory systems at the state level is needed.

118

Montek S. Ahluwalia

The general condition of law and order is another aspect of governance at the state level that is relevant for creating an environment conducive to investment. There are no objective measure with which to assess performance in this dimension, but impressionistic evaluations suggest that the slowergrowing states suffer from more than the usual problems in this dimension. Tensions associated with economic and caste stratification in parts of the country, especially in rural areas, have created disturbed conditions in some of the slow-growing states, a situation that is bound to have an impact on developmental activity. There are reports of urban mafias engaged in extortion, various types of protection rackets, and even kidnapping in parts of some states. It is difficult to imagine any significant acceleration in economic growth without a significant improvement in this aspect of governance. An area where the slower-growing states could improve the investment climate at relatively low cost is the flexibility with which labor laws are administered. India’s labor laws are often criticized because retrenchment of labor and closure of units both require the permission of the state government, which is almost never given. A comprehensive solution requires amending the relevant central legislation to remove the need for permission, and this is one of the items on the agenda of second generation reforms that are currently being discussed in India. However, this is also an area in which state governments could act on their own by prescribing more flexible guidelines within which the relevant departments would act on these matters. States suffering from low levels of investment could reduce their competitive disadvantage vis-à-vis more industrialized states by allowing greater flexibility with regard to labor. No state government has experimented with this possibility thus far. 3.3.4 The Role of the Central Government The resources problem of the poorer states and their consequent inability to develop economic infrastructure and provide essential social services raises the issue of what the central government can do to assist these states to achieve these objectives. The total resources devolved from the center to the states in the form of the statutory devolution of the states’ share of central taxes and special grants recommended by the finance commission, together with the flow of central assistance in support of state plans through the planning commission, already add up to a substantial amount, and the central government’s fiscal position does not allow any significant expansion in these flows. The central government’s fiscal deficit was 5.6 percent of GDP in 1999–2000, and the consolidated deficit for the center and states together was almost 10 percent of GDP. Given the central government’s evident compulsion to reduce its own fiscal deficit, there is obviously little scope for increasing the total flow of resources to the states. However, there is room to reorient the expenditure undertaken by the center in a manner that provides greater developmental

State-Level Performance under Economic Reforms

119

support to the states, especially the poorer states. At present, a very large volume of resources under the direct or indirect control of the central government is devoted to various types of poverty alleviation programs. For example, the budget directly provides Rs. 7000 crores for poverty alleviation schemes in rural development, Rs. 12,000 crores for food subsidy, and Rs. 14,000 crores fertilizer subsidy. Other programs involve underpricing of certain goods and services based on cross-subsidization from other parts of the system—for example, a subsidy of Rs. 10,000 crores on kerosene (financed by overpricing gasoline) and about Rs. 3,000 crores on railway passengers (financed by overcharging freight). Although these amounts are crosssubsidized by other parts of the system, they can become available as additional resources if user charges are raised to eliminate the need for crosssubsidy, and the resources thus released are mopped up through taxation.15 The total amount involved in these subsidies comes to Rs. 46,000 crores, which exceeds the total central assistance provided by the central government to the states in support of their plans. If these programs could be reduced in scale by 50 percent, the resources so released could be used to expand central assistance to the states for infrastructure development. Eliminating subsidy programs is not easy, but all the available evidence suggests that the effectiveness of the existing programs is extremely limited and that the same resources would be much better spent in building infrastructure. There is little doubt that such a reallocation would strengthen the development prospect of the poorer states, and make a much bigger contribution to poverty reduction in the country. A related issue, which has not received the attention it deserves, is the scope for improving the development effectiveness of central assistance to the states by linking it to performance. At present, most of the central assistance provided to support state plans is not subject to specific performance criteria or conditionality. It can be argued that such assistance would be more effective if it were linked to policy reforms and other specific performance criteria that would be designed to address the factors constraining the growth performance of the states. Advocates of decentralization will no doubt object to the suggestion on the grounds that resources to the states should be provided on the basis of an entitlement criterion and that accountability for the use of these resources should be left to the normal political process at state level. However, this approach also implies that the center can have no particular responsibility for ensuring that the specific constraints to growth at the state level are effectively addressed. A new window of concessional assistance was introduced in the year 2000–01 in the form of an accelerated power development program, under which states will 15. For example, the underpricing of kerosene is covered by the overpricing of gasoline. If the extent of underpricing kerosene could be reduced, it would be possible to levy an excise tax on gasoline and mop up the corresponding component of the gasoline price without increasing the price of gasoline to the consumer.

120

Montek S. Ahluwalia

receive central assistance to support a program of power-sector reforms. Such mechanisms should be expanded in future. In recent years, the progressivity built into the distribution of various types of central transfers to the states under which poorer states receive additional resources has been criticized by some of the better-performing states on the grounds that such mechanisms reward inefficiency and mismanagement. This type of argument can be made against any progressive distribution scheme and does not provide a sufficient justification for abandoning progressivity. However, it does suggest that if transfers are to be progressive, they should at least be made conditional so that they are seen to contribute to a solution of the problem in the longer run. For example, if growth in the poorer states is held back by gaps in infrastructure and social development, then central assistance should perhaps be made available to these states and linked to specific performance requirements, which may be project-specific or linked to implementing broader policy reforms in critical sectors. This may appear intrusive, but as long as the design of the programs, and the identification of milestones for implementation, has the full involvement of the state, there can be no objection to the arrangement. The central government has a major role in developing infrastructure in the poorer states by shaping its own expenditure on infrastructure to help overcome infrastructure bottlenecks in the poorer states as quickly as possible. As pointed out earlier, many infrastructure areas are the specific responsibility of the central government, especially the national highways, railways, telecommunications, airports, and major ports. Expanding central government expenditures to improve services in these areas, with special concern for meeting the needs of the slower-growing states, can make a major contribution to accelerating growth in these states. The recently launched National Highways Development Project, which is being funded by a tax on gasoline and diesel, is an example of a central government program that could help to overcome transport bottlenecks affecting hinterland states. The project aims at four-laning about 6,000 kilometers of the national highways on the so-called “Golden Quadrilateral” linking Delhi, Mumbai, Chennai, and Calcutta, which carries a very large portion of the country’s road traffic. The project would greatly improve connectivity for Uttar Pradesh, Bihar, and Orissa. A similar large-scale effort is needed to modernize the railway system, with particular emphasis on its freight-carrying capacity. Hinterland states would benefit the most from an efficient railway system capable of transporting freight over long distances at attractive rates. Paradoxically, the pricing policy of the railways, which subsidizes passenger traffic by overcharging for freight, has precisely the opposite effect because it increases the cost of rail freight, a price distortion that has made it cheaper for coastal power plants in the south to import coal from Australia rather than transport it from the coalfields of Bihar! Ironically, although it is the freight-

State-Level Performance under Economic Reforms

121

carrying function of the railways that is most likely to spur development in hinterland states, the demands upon the railways, even from these states, is usually for the addition of new railway lines and the introduction of new passenger trains, which only worsen the financial position of the railways. The scale on which the center can undertake infrastructure development aimed at helping the poorer states is obviously constrained by the total availability of resources. One way of overcoming the resources problem would be to divert resources from existing subsidy-oriented programs toward infrastructure development, as discussed earlier. Another is to accelerate the privatization of central PSEs and to earmark the proceeds from these sales specifically for the development of much-needed economic and social infrastructure in the backward states. This would be a much more effective way of helping the poorer states than the traditional approach of pushing existing PSEs to make commercial investments in the less developed states. Such initiatives have done little in the past for the economic development of the area and have often increased the probability of driving the PSEs into sickness. On the other hand, privatizing existing central PSEs and using the proceeds to build social and economic infrastructure in the less developed states will increase the efficiency with which existing PSE assets are used, while simultaneously helping to improve efficiency of resource use in the poorer states and hopefully leveraging a greater flow of private investment. As pointed out earlier, doubling the rate of growth for the slow-growing states, which have a combined population exceeding 300 million, is not a simple task, and the policy directions outlined above are certainly not a comprehensive blueprint of what is needed. Detailed strategies have to be evolved for individual states to address constraints and circumstances specific to each state. However, the agenda we have outlined identifies some of the major areas of policy that must be addressed if the slow-growing states are to achieve a growth rate of 6 percent in the future. Determined action in these areas should make it possible to achieve this target. We hasten to add that even if this growth objective is achieved, regional inequality would continue to increase since the rest of the country is projected to grow at a faster rate. However, the rate of divergence in per capita incomes would be considerably moderated, compared to the trends observed in the 1990s, and the slower-growing states would at least experience strong growth in per capita GSDP and sharply falling poverty levels.

References Ahluwalia, M. 2000. Economic performance of states in the post-reform period. Economic and Political Weekly 5 May: 1637–48.

122

Montek S. Ahluwalia

Bajpai, N., and J. Sachs. 1996. Trends in inter-state inequalities of income in India. Harvard Institute for International Development. Development Discussion Paper no. 528. Bhalla, S. 2000. Growth and poverty in India: myth or reality? Paper prepared for conference in honor of Raja Chelliah. 17 January, Institute of Economic and Social Change, Bangalore. Cashin, P., and R. Sahay. 1996. Regional economic growth and convergence in India. Finance & Development 33 (1): 49–52. Datt, G. 1999. Has poverty declined since economic reforms? Economic & Political Weekly 34 (50): 3516–18. Deaton, A., and A. Tarozzi. 1999. Prices and poverty in India. Princeton University, Department of Economics. Mimeograph. Dreze, J., and S. Amartya. 1995. India: economic development and social opportunity. Delhi: Oxford University Press. Gupta, S. P. 1999. Trickle down theory revisited: The role of employment and poverty. V. B. Singh Memorial Lecture. Mimeograph. Lal, D., R. Mohan, and I. Natarajan. 2001. Economic reforms and poverty alleviation: A tale of two surveys. New Delhi: National Council of Applied Economic Research. Mimeograph. Nagraj, R., A. Varoudakis, and M. A. Veganzones. 1998. Long-run growth trends and convergence across Indian states. OECD Technical Paper no. 131. Cambridge, Mass.: Organization for Economic Cooperation and Development. Natarajan, I. 1998. India market demographics report. Delhi: National Council of Applied Economics Research. Nosbusch, Y. 1999. Convergence across regions: Evidence from India. London School of Economics and Political Science, Quantitative Economics Project.

Comment

Shankar Acharya

I found this a very stimulating paper. It really made me think—and that’s not a characteristic of most papers I read. I do hope my comments will have a reciprocal effect on the author! Scope In Montek’s words, the paper is a “modest foray into the economic performance of individual states” and “the variation in performance across states.” Actually, the focus of the paper appears to be much more on the variation in growth performance across states, with regrettably little analysis or discussion of performance of individual states. Shankar Acharya is currently Honorary Professor at the Indian Council for Research on International Economic Relations. At the time of writing this article, he was a Visiting Research Fellow at Merton College, Oxford University, on leave from his regular assignment as Chief Economic Adviser in the Ministry of Finance, Government of India. The views expressed in this Comment are those of the author and should not be attributed to the government of India.

State-Level Performance under Economic Reforms

123

Second, the concept of performance does seem limited, by and large, to aggregate growth; there is no discussion of other dimensions such as literacy, life expectancy, and other indicators of living standards (there is some discussion of trends in headcount ratios of poverty). Even the discussion of growth performance is largely at the aggregate level, with no discussion of performance of major sectors. This is a pity since table 3.1 of the paper provides a disaggregation of growth performance into agriculture and non-agriculture. Even a cursory inspection of this table suggests opportunities for interesting analysis and assessments. Just to take a few examples: • The deceleration in aggregate growth between the two periods for Uttar Pradesh and Bihar (the two most populous states) is attributable largely to non-agricultural performance, which declines sharply, by about 50 percent for Bihar and 33 percent for Uttar Pradesh. • Rajasthan also suffered a sharp drop in growth of non-agriculture, but this was almost wholly offset by a remarkable acceleration of agricultural growth to 7 percent. Indeed, this extraordinarily high growth of agriculture (and a similarly high rate in West Bengal) surely bears investigation and explanation. • Conversely, the deceleration in overall growth in Orissa is almost entirely explained by the apparently sustained implosion of the agriculture sector at –2 percent per year. • The deceleration of growth of Punjab and Haryana is attributable largely to the decline in agricultural growth. I could go on with other examples to illustrate the general point that a look at the two-sector data raises a lot of interesting questions and issues (including about the reliability of the data), especially when one recalls Montek’s comment that many Indian states are more populous than most developing countries—and therefore merit serious analysis on their own. Divergence in Growth Performance across States The paper provides an interesting account of trends in inter-state inequality. Somewhat surprisingly, it makes no reference to the substantial literature on convergence of growth rates, or what is termed betaconvergence. A very recent Oxford master’s thesis, by Kamakshysa Trivedi, carefully analyzes the data on Indian states for 1960–90 and arrives at the following conclusions (Trivedi 2000): 1. There is no evidence of unconditional beta-convergence for Indian states; 2. There is strong evidence for conditional convergence, once one controls for steady state income determinants of education, infrastructure, and government investment.

124

Montek S. Ahluwalia

It would be helpful if Montek wove this literature and its findings into the body of his paper. Impact on Poverty I won’t get involved in this evergreen debate on poverty trends except to mention some recent work by Deepak Lal, which uses National Council of Applied Economic Research large sample household panel data to support the view that there have been substantial declines in headcount ratios of poverty incidence in India during the 1990s. This is at variance with the oftcited National Sample Survey household data, and Lal casts doubt on the quality of that data and the trends based on them. Explanations of Growth Differentials I have several reservations about Montek’s explorations of possible explanations of inter-state variations in growth. First, there seems to be excessive preoccupation with investment as an explanatory variable. In a period of major policy reform, would we not expect to find higher productivity from existing capital stock (both physical and human) and other factors of production? It is not at all clear that investment is a good proxy for this. Second, and more generally, the search for explanations appears to underplay the role of policy reforms. After all, at the national level it is generally believed (including by Montek) that reforms in trade and exchange rate regimes, in industrial policies, in fiscal and financial policies, and so on have boosted factor productivity and growth. Surely one should then seek these links at the state level, since national performance is ultimately an aggregation of statewise performance? Finally, testing for the role of plan expenditure seems a little forced, given that more than half of such expenditure consists of revenue (or current) expenditure. Furthermore, we know from the national level that the acceleration in growth to above 7 percent in the mid-1990s occurred during periods of relatively low plan expenditure. Why then should we expect this variable to be a powerful explanatory variable at the state level? Policy Conclusions In this section the paper says a little about infrastructure and a lot about state finances—and there isn’t much else. This seems a little unbalanced, when we think of all the major factors (and policies) that might impinge on the objective of “a minimum growth rate of around 6 percent per year in the poorer, slow growing States.” I have already mentioned the range of trade, payment, industrial, fiscal, and financial policies that presumably affect growth of states. There is also the entire gamut of policies that sustain or boost agricultural growth, which we know has varied enormously across states. There is little mention of these in the paper, except for a few elements

State-Level Performance under Economic Reforms

125

that are subsumed in the discussion of “state finances.” Finally, what is happening to services sector growth across states and why—and what is the scope for policy intervention? This last, in particular, may be a somewhat unfair question since there is grossly inadequate understanding of the issue even at the national level. However, there is no harm in posing the question to the author! Reference Trivedi, Kamakshya. 2000. Economic growth, convergence and levels of income: Evidence from states in India, 1960–1990. M. Phil. Thesis. Oxford University, Oxford, England.

4 Doing Business in India What Has Liberalization Changed? Naushad Forbes

What is the reason that our nation Is not well known for innovation? We spend so little on R & D, And the results are poor, obviously. And I, for one, would lay the charge Quite squarely on our Licence Raj. A licence then was a sinecure, A perfect method to ensure A steady stream of easy dough And this went on and on, you know. The dinosaurs that roamed the land, I’m sure that you will understand, Had no need to innovate For, after all, they were doing great. But in ’91 the meteor hit. And stirred things up quite a bit. As dinosaurs were now laid low, The nimbler mammals start to grow. In other lands it is a fact The smaller firms are quick to act. Entrepreneurs don’t hesitate; It is their task to innovate. But smaller firms faced disruption: They’re harder hit by corruption And more entangled in red tape, Which in India you can’t escape. And because of this sorry state Both big and small didn’t innovate. One strategy then was to steal, Pretend to reinvent the wheel,

So ideas known in other nations Were passed off here as innovations. The picture that I paint is bleak, But now is the time for me to speak Of areas where it can be said That Indians are a bit ahead. In low-cost goods we specialize, Simple products in a tiny size, Is all that many can afford And in this field we have scored. There is no doubt we have the brains, And therefore some have taken pains To organize our talent pools, As software firms, computer schools That clearly pass every test And are considered the very best. Sometimes it is our many flaws Such as the lack of patent laws, Our many years of price controls And endless number of poor souls, That enabled us to take the lead In processes that succeed, In developing generic drugs And pesticides for all the bugs. And that is all I have to say; I wish you all a good day. For those of you who swear by prose, You must admit, a tiny dose Of early morning rhyming verse, As long as it is crisp and terse,

Naushad Forbes is a consulting professor in the department of industrial engineering, Stanford University, and director of Forbes Marshall, India.

129

130

Naushad Forbes

Can serve to wake, if not the dead, Those who still think they are in bed.

The bell has rung, get, up, stand straight: It’s not too late, let’s innovate. —Innovation,1 Nadir Godrej

Industry in India is characterized by contrast. The British economist Joan Robinson once said about India that whatever could be said about it, so could the opposite (quoted in Bhagwati 1985). Even a casual visitor to India would experience this contrast: A typical street scene would involve a bullock-cart being overtaken by a Hindustan Motors Ambassador (a 1960smodel Morris Oxford), which in turn would be overtaken by a 1990s Suzuki or Opel Astra. What makes the scene distinctively Indian is that both the bullock-cart and Ambassador could have been built anything from thirty years ago to thirty days ago. Software firms in Bangalore and Hyderabad have much more in common with firms in Silicon Valley than with the state fertilizer factory in Talcher, which celebrated its silver jubilee before producing its first ton of fertilizer. A visit to the finance ministry convinces one that the Indian bureaucrat is being cast in a new mold, that of a market-oriented individual with an international outlook. One remains convinced until one visits certain other ministries and meets bureaucrats who, one suspects, have not yet had the end of the cold war brought to their attention. 4.1 Introduction: Much Change, but Not Everywhere Let me begin with a series of random facts: 1. The Indian car market in 1999 at 600,000 vehicles passed China, turning India into the third-largest car market in Asia (after Japan and South Korea). In 1991, the Indian car market was sixth, after Japan, South Korea, China, Taiwan, and Malaysia. 2. India has one of the world’s most efficient cable TV systems, with 30 million connections and a daily cable audience estimated at over 100 million people (“The Wiring of India,” The Economist, 27 May 2000, 73). An average rental fee of 150 rupees (Rs.) per month provides around forty channels, through over 30,000 independent cable television operators. This is from a standing start: In 1991, India had one television channel, the stateowned Doordarshan. 3. At country number 177 of 209 countries in 1997, India remains one of the world’s poorest countries, with 16 percent of the world’s population but 1. Nadir Godrej is managing director of Godrej Soaps, Ltd. He read out this poem at a breakfast meeting of the World Economic Forum/Confederation of Indian Industry’s annual India Economic Summit in December 1999.

Doing Business in India: What Has Liberalization Changed?

131

35 percent of the world’s population below the poverty line (World Bank 1999). 4. Azim Premji, who owns 75 percent of the Indian firm Wipro (which deals mainly in IT, but which has its origins as a family business in soap and edible oil), is one of the world’s ten richest men. Briefly, in February 2000, he was reputed to be the world’s third richest, based on a stock market valuation for Wipro that had risen 800 times in the last eight years. 5. Successful Indian software firms like Infosys, Wipro, and NIIT, all of which have seen skyrocketing stock market valuations and price-earnings (PE) ratios of 100 or more, are lobbying the Indian government to permit them to purchase foreign firms for $1 billion with no prior approval. 6. In May 2000, the Indian opposition organized a national strike against liberalization, demanding in particular that the government roll back the modest subsidy cuts that it had announced in the February 2000 budget. Sonia Gandhi, as leader of the Congress opposition, rose in parliament to support the strike. The general tone was one of “just politics.” 7. A survey by the Centre for the Study of Developing Societies (CSDS) asked people what they thought of the economic reforms. Eighty percent of the population had not noticed any change in economic policies in the last ten years.2 8. Hindustan Motors, which for forty years has made essentially the same 1960 Morris Oxford, the Ambassador, at its factory near Calcutta, continued to be decently profitable even in 1995. It is only in the last three years that making a forty-year-old model with indifferent quality finally seems to have become an unprofitable activity—sales of the Ambassador have fallen to under 1 percent of the Indian car market. 9. The Talcher unit of Fertilizer Corporation of India, the state-owned fertilizer factory, recently celebrated its silver jubilee. It has yet to produce its first kilogram of fertilizer. A few years ago, its workers went on strike (one wonders from what), demanding higher wages. The same government that put through the 1991–93 reforms gave in to the demand. 10. The Uttar Pradesh government, one of the most corrupt and inefficient in the country, put through a major deregulation of the state electricity board in early 2000, beginning by corporatizing the board into three companies, for generation, distribution, and transmission. The state employees went on strike, but the state and central government stood firm, taking out full-page ads in the papers that showed how the reform was to provide the public with a better service. This list illustrates the complex picture that always emerges when one studies India. Change has been dramatic in some sectors, where Indian con2. Swaminathan Aiyar, Lack of growth trickles down. The Times of India, 9 April 2000.

132

Naushad Forbes

sumers can buy world-class products at internationally comparable prices. In other sectors, though, change has been halting to the point of being difficult to perceive. The list also illustrates that the change is dramatic relative to India’s own past. When compared with other countries, India still has a long reform agenda pending. What is clear, though, is that the reforms since 1991 have unleashed change that is increasingly self-perpetuating. This chapter reflects on a few areas of this change: Section 4.2 begins by looking at what has changed for Indian firms, how their own perspectives on reform evolved, how the structure of Indian industry is different in 2000 from 1991, and which firms mattered then and now. The next section (section 4.3) looks at foreign firms and their evolving relationship with Indian firms since 1991. Section 4.4 raises some points regarding the way ahead—pending areas of reform that are comparatively neglected in public debate. In the context of the long-term success of firms, nothing is more important than technology: Thus, section 4.5 takes a look at how technology and innovation have changed in Indian industry since 1991, examining what has happened to efficiency, to technology import, and to investment in R&D. Finally, section 4.6 includes a few comments on winners and losers in Indian industry. 4.2 The Changing Structure of Indian Industry: Becoming Normal The nine-year period since liberalization3 began in 1991 is best considered as two periods, the boom of 1991 to 1996 and the slowdown from 1996 to 1999.4 The five years from 1991 to 1996 saw a boom, as the lifting of controls on licensing in particular, but also on technology import and foreign investment, led to a big increase in industrial investment. A major liberalization of the capital market freed firms to price their own issue, instead of being constrained by pricing determined by a government regulator, and this freedom combined with a strong feel-good factor of operating in a new era that could only be good for industry. The result was an investment boom and a major increase in foreign investment, all driven by the private sector. Industrial investment rose dramatically, with investment intentions, as indicated by filing of industrial entrepreneur’s memoranda with the government, peaking at 6,900 proposals worth US$42 billion in 1995, and then falling by two-thirds by 1998. As an insightful recent study of Indian industry by the National Council of Applied Economic Research (NCAER) says, the data [suggest] a remarkable boom in industry in the first half of the 1990s, accompanied by a considerable acceleration of growth in both employment 3. For a comprehensive background, see Ahluwalia et al. (1998) and Bhagwati (1993). 4. There are signs the slowdown is ending—business confidence indicators are increasingly positive as of mid-2000.

Doing Business in India: What Has Liberalization Changed?

133

and productivity . . . The growth between 1991–2 and 1995–6 came partly from the growth of old establishments and partly from the emergence of new establishments. . . . [For the old establishments] their average capital-output ratio rose from 0.5 to 0.63. Worker productivity rose by almost a third; wages per worker rose almost as much. . . . Profit margins increased significantly. Thus the older establishments invested heavily, and both worker productivity and profitability increased” (NCAER 1999, 9) The slowdown since 1996 has been attributed to many causes. The NCAER study identifies poor investment decisions of the boom years as a major cause. The investment boom of the early 1990s was dominated by Indian firms, mainly operating in commodity industries. Used to operating in a shortage economy in which profits were limited by how much one could make, firms invested heavily in sectors like steel, fertilizers, cement, petrochemicals, and aluminium (NCAER 1999, 12). Capacity increased much faster than the market, and by 1996 overcapacity combined with a liquidity crunch and a psychological swing from feel-good to feel-bad (driven by the political uncertainty that came from four governments in less than two years) and led to the slowdown. Foreign firms—as the NCAER study shows—stayed out of the commodity businesses that Indian firms dominated, entering fields like pharmaceuticals, banking, and consumer goods, where they encountered less competition. The slowdown has been particularly important as a driver of change in industry. Many industrialists reflect that the period from 1991 to 1996 was one in which the sudden burst of freedom meant one could simply do more. In spite of record growth in sales and profits for many firms, change in what firms did, and how they did it, was relatively modest. It was only as growth fell and profit margins were squeezed that firms started to seriously restructure. The picture that emerges of Indian industry after nine years of liberalization, then, is roughly as follows: 1. The public sector plays a declining role, but it is important to recognize that the role is still major. Public sector enterprise (PSE) sales may have declined from 45 percent to 37 percent of total sales, but 37 percent is still a large number (NCAER 1999, 11–2).5 Seven of the top ten firms by sales in 1998, seven years after liberalization began, are public sector firms, six being oil companies (22 of the top 100 firms are PSEs). The last two years have finally seen talk of privatization, but progress has been glacial. There was much publicity when the first PSE was privatized in December 1999: Modern Foods, a bread company, was sold to Hindustan Levers (HLL). The 5. The source quoted is the CMIE (Centre for Monitoring the Indian economy) Prowess database, a balanced sample of 1,172 publicly listed manufacturing companies.

134

Naushad Forbes

deal was important more as symbol—the national bread company being sold to a multinational—than as reality: The value was $20 million, well under 0.01 percent of the total PSE stock. Although privatization is now being discussed, and the report of the disinvestment commission has been accepted by the government “in principle,” there has been little progress in practice. Take the comments of Madhavrao Scindia of the Congress Party, normally considered to be a more progressive politician: “If the public sector undertaking is efficient, why should the government be in a hurry to throw away a family jewel? . . . Secondly, would you dispose of a large source of livelihood? In the case of Modern Foods, its 4500 employees have been told that only 700 will be retained.”6 2. Small industry continues to be an area no government has had the courage to tackle. Over 700 products, including, for example, toilet paper and household utensils, remain reserved for the small-scale sector. Reservations have particularly benefited the few large firms that made the item before reservation: Under a grandfather clause, they could continue manufacture and are protected from competition from large new entrants. The lack of political will is best illustrated by the government’s having agreed with the United States to open the market to consumer goods imports by 2001, two years before being required to do so under the World Trade Organization (WTO). This means that if a large Indian firm or multinational wishes to enter the Indian market for toilet paper or soap, it can set up a factory in Sri Lanka and import the product into India, but cannot set up its own factory in India! This is clearly ridiculous, but there is a striking lack of public awareness of this issue. Rakesh Mohan’s chapter in this volume discusses this issue in detail. 3. The most striking feature is the growth of competition. In sector after sector, new, often foreign entrants now compete against old, and the choice for the Indian consumer has changed dramatically, as table 4.1 on changes in ten industry segments shows. 4. Foreign investment today is essentially free. Automatic approval for investment upto 74 percent is permitted in all except four industries, and 100 percent subsidiaries are permitted by specific approval. Foreign direct investment accounted for 2.7 percent of all investment in India in 1998, up from 0.1 percent in 1991. However, this 2.7 percent still compares poorly with China’s 12.7 percent or Malaysia’s 16.5 percent. 5. Indian firms now compete against imports in several industries; this is particularly true of industrial products like machine tools and instrumentation, but imported television sets, pagers, mobile phones, and refrigerators are increasingly common. 6. As the introductory chapter in this volume documented, India’s share in world trade has finally shown an increase in the 1990s after four decades 6. Interview in Business Standard, 19 May 2000.

Doing Business in India: What Has Liberalization Changed? Table 4.1

135

Impressions of Change in Ten Industries Leading Players

Industry Sector

Major Drivers of Changes

1991

2000

Pharmaceuticals

Signing WTO; changes to Indian Patent Law of 1970.

Glaxo, HCG, Ranbaxy

Glaxo, Cipla, Ranbaxy, HMR

Automobiles

Delicensing and new foreign investment permitted. Telco’s development of own SUV and car; M&M’s development of new SUV. Entry of Hyundai, Daewoo, Ford, GM, Toyota, Honda, Daimler Benz, Mitsubishi into market.

Maruti Suzuki, Hindustan Motors, Premier Automobiles

Maruti Suzuki, Telco, Hyundai, Daewoo

Two-wheelers

New foreign investment, delicensing, removal of maximum capacity restrictions.

Bajaj, LML, Kinetic Honda

Bajaj, TVS Suzuki, Hero Honda

Oil

Removal of PSU monopoly.

All PSUs

Reliance at 25 percent of Indian refining capacity, PSUs, MRPL

Synthetic fibers

Removal of capacity restrictions. Imports and tariff reduction.

Reliance, JCT, JK Synthetics, Bombay Dyeing, Raymonds, Indo Rama, Baroda Rayon, Grasim

Reliance, Indo Rama, SRF in niche, Mitsubishi, Grasim

Color televisions

Freer component imports, foreign-firm entry

BPL, Videocon, many others

BPL, Videocon, foreign brands

Soaps and detergents

Fewer limits on HLL and large Indian firms; sale of Tomco to HLL

HLL, Tomco, Godrej

HLL, Nirma, Godrej

Television broadcasts

No policy change, but satellite technology bypasses state

Doordarshan

Zee, Star, Doordarshan, other private

Airlines

Opening to private firms

Indian Airlines

IA, Jet

Beverages

Opening to foreign investment

Parle, Pepsi, Dukes

Coca Cola, Pepsi

Source: Author’s analysis. Notes: Abbreviations are explained in chapter text.

of decline. Exports are serious business for a growing number of Indian firms, and account for the major share of sales in software and pharmaceuticals. 7. The emergence of new firms, which have caught the fancy of the stock market, dominated the business press last year. The information technology

136

Naushad Forbes

(IT) industry is today the single largest industry by stock market valuations, accounting for over one-fourth of total stock market capitalization. India’s most valuable companies are Wipro, Infosys, and HLL, with HLL twice as valuable as the next company for the last five years.7 8. Foreign institutional investors are now active players on the Indian stock markets. Although they account for under 10 percent of total market capitalization, they account for a disproportionate share of daily trading, and their buying preferences have led to a penalty on stock valuations of firms that have group cross-holdings or inadequate disclosures or that are closely held. This in turn has fed interest in corporate governance. 9. There has been much restructuring of several industry sectors: The cement industry has seen particular consolidation, with the entry through acquisition of two major foreign groups (Lafarge and Blue Circle), and acquisitions by Indian firms of smaller players to grow market share. In particular, cement firms that were secondary businesses of major industrial houses are being sold off. Given that industrial licensing led to fragmented capacity in many industries with a dozen or more small players, restructuring was inevitable but is only now gathering pace. Reliance has been buying up synthetic fiber competitors (Raymond Synthetics, DCL Polyester, ICI Fibres, JK Corp, and India Polyfibres) to the point that it now controls 60 percent of the market. Increasingly, the criterion for success in more and more industries is to make a product that more people wish to buy more efficiently than others. Through 1991, on the other hand, success for many industrial segments was measured more by the licenses that could be captured. In other words, Indian industry is becoming normal. 4.3 The Changing Relationship with Foreign Firms8 Perhaps the greatest driver of change in the relationship with foreign firms has been the change in perceptions of India as an attractive market. Before 1991, Indian policy saw foreign investment as a necessary evil, the price for desirable technology. Foreign firms saw India as a place that was more trouble than it was worth. They entered India by ceding the lead role to an Indian firm, content to establish a presence in a market of future importance, but leaving management control firmly in the hands of the Indian partner. In any case, the Indian partner controlled the key success factor of obtaining an industrial license. Since 1991, three things changed: First, foreign firms could now invest much more freely in India, including setting up 7. It is always rash to write about stock-market valuations. When I began writing this chapter, Wipro was first, Infosys second, and HLL third. Today, Wipro and Infosys have fallen below HLL! 8. This section tells a story. It is not based on a statistical sample of the manner in which foreign firms and Indian firms have interacted.

Doing Business in India: What Has Liberalization Changed?

137

100 percent subsidiaries. Second, the change in competition meant that there was a demand for new technology, to introduce new products or more efficient processes. The Indian firm often sought this technology from its foreign partner. Third, the foreign firm’s perception of the Indian market had changed dramatically—there was a sudden discovery of a large (and usually grossly overestimated) middle class, and what was clearly going to be one of the world’s top few markets in a few years. The first few years after 1991 were a honeymoon: The sudden freedom to enter new fields, license new technologies, and expand capacity meant that everyone was happy, Indians and foreigners. As competition increased and foreign firms began to put together more active India strategies, however, tensions emerged. At just the time that Indian firms began to look for more technology and had the freedom to license it, foreign firms began to seek more control over the technology they provided and over their Indian operations. The result has been an ambivalent attitude to foreign investment by Indian firms. There has been much public talk by Indian industrialists of “level playing fields” as an argument for continued protection. The ambivalence is between foreign firms with the freedom to invest in India, with whom one will have to compete, and having the freedom to sell one’s own firm to foreigners. This ambivalence is reflected in statements by the leading industry group Confederation of Indian Industry (CII),9 which has changed stance from being unqualified liberalizers in 1991 to qualifying their welcome to foreign firms today. In December 1998, the government, led by the Bharatiya Janata Party (BJP), passed a notification requiring that foreign firms that wished to set up a new Indian operation get a no objection certificate from any existing Indian partners or licenses. Industry associations have since argued against any attempt to relax this regulation. It seemed that this ambivalent attitude from some sections of industry toward foreign investment struck a chord with the BJP government, whose dominant nationalist rhetoric was Swadeshi when it first formed the government in 1998. In its second incarnation, though, the word Swadeshi has been conspicuous by its absence from the BJP platform: Even the rhetoric was quietly dropped from every public statement, whether the national budget or policy announcements. The reality is that decisions by the BJP have not been hostile to foreign firms—witness the sale of Modern Foods to HLL, or the telecom reform, which mainly met the representations of foreign firms, or indeed the opening-up of the insurance sector to limited foreign investment.10 9. The other leading industry associations are the Federation of Indian Chambers of Commerce and Industry (FICCI), which has always been more parochial, and the Associated Chambers of Commerce (Assocham), in which MNCs are strongly represented. 10. This came after the BJP itself successfully voted against the United Front’s Insurance Reform Bill on precisely the same grounds two years earlier.

138

Naushad Forbes

Again, the changing relationship between Indian and foreign firms should be seen as the normalization of Indian industry. The old reasons that foreign firms needed Indian firms—obtaining licences, dealing with the bureaucracy—have not completely gone, but are dramatically reduced. Indian firms still need foreign firms as a source of technology: This has not changed. That is why tensions have emerged: Foreign firms have become more assertive, seeing less that their Indian partners can do for them. The key to a new relationship built around true partnership lies in technology, which is discussed in section 4.5. A true partnership requires proprietary competencies on the Indian side, and the success in international alliances of Indian software and pharmaceutical firms points the way for the rest of industry. Next, though, let us look at some of the pending reform areas that can indeed help Indian firms compete internationally. 4.4 The Pending Agenda for Reform Many others in this volume deal with the need to move on all the bigger pending reforms—of infrastructure, of subsidy reduction and the reform of government finances, of changing the system of center-state finance, of public sector privatization, and of labor reform. These are all areas whose problems and solutions are well known; every major political party agrees on these solutions, yet no government has had the courage and political will to push anything serious through in these areas when in power. So, instead, this chapter will deal with relatively neglected areas of pending reform: the need to take reform to the local level, and the need to build a public lobby for reform. 4.4.1 The Need for Reform at the Local Level In 1991, much changed at the central level. A common industry cocktailparty conversation topic in 1991 was that one no longer needed to catch a weekly flight to Delhi or maintain a liaison office there to get permissions through the government. Previously, any technology license, foreign investment, major expansion, investment in a new factory, manufacture of a new item, or even expansion of capacity required a visit to Delhi and often months of form-filling and –pushing, often with bribes to grease the way. In 1991, this need was largely eliminated at a stroke; the importance of this change is difficult to appreciate for those who have not experienced the old system. The reforms that took place at a stroke, though, were largely confined to the finance, commerce and industry Ministries. What did not change was the local inspector Raj, where several government inspectors could drop in on firms at will, each with absolute power to make firms cease operations first and challenge the order later. Table 4.2 lists the inspectors that one medium-sized engineering firm (and an engineering firm experiences far

Doing Business in India: What Has Liberalization Changed? Table 4.2 Inspector

139

The Inspector Raj in May 2000 Powers

Change Since 1991

Factory

Can halt work at a particular work area where there is a fault, and can issue penalties

License fee, formerly paid every year, can now be paid in a five-year lump sum.

Electrical

Can issue a notice and levy penalties

None

ESIS

Can levy penalties

None

Octroi

In case of default, can collect the difference in amount

None

Food

Can close down canteen

None

Income tax

Can levy penalties

None

Sales tax

Can levy penalties

None

Excise

Can stop dispatch of material

Excise records are simplified. For merly, use of preprinted, serially num bered invoices was compulsory; now can have own serial numbers and computerized records. Verification/ inspection of documents, formerly semi-annual, is now annual.

Municipal corporation

Can serve notice and demolish unauthorized construction

None

Lift

Can stop functioning of lift

None

Customsa

Does not clear goods; can impose penalties (pay first, argue later)

Discretionary power greatly reduced by reducing the number of different classifications and different rates. Clearances still take a minimum of two days.

Source: Author. Note: These are the inspectors/inspections to which Forbes Marshall, Pune was subject in 1991 and the changes affected by 2000. a Not an inspector but a major regulatory mechanism

fewer regulations than, say, a chemical firm) experiences in May 2000, and what has changed since 1991. What emerges is that, with the exception of much simplification and removal of discretion of customs regulations, and some limited simplification of excise regulations, the remaining changes only add up to minor tinkering with the system. The result is that Indian firms, which now have to compete with firms operating in countries like Singapore, are tied into processes that have not changed. Consider the case of exports, which all in government claim to be a priority and something to be encouraged. If the firm I work for in India wishes to export something from our base in Pune, the elapsed time

140

Naushad Forbes

Table 4.3

Steps in Exporting Goods for a Medium-Sized Engineering Firm in Pune

Step

Actions

Time Required

1. Excise clearance

Exports are exempt from excise duty. Every export dispatch must be personally authorized by an excise inspector.a

1/2 to 1 working day

2. Octroi clearance

Goods entering Bombay for shipment out are exempt from octroi. Clearance requires a few hours but can be done before goods leave factory.

1/2 working day

3. Customs clearance

Shipping bill filed in customs. Octroi clearance takes place based on shipping bill.

1/2 day

Goods dispatched to port or airport the next day.

1 day (from Pune to Bombay)

Customs carries out “valuation and clarification” of every consignment. If no query raised, clearance given to bring goods to docks.

1 day if no query raised

Customs agent physically examines goods.b If no problem, a “let export” is issued.

1/2 day

Goods handed over to airline or shipping line.

Airline: 24-hour “cooling-off” period (for security reasons) Ship: 2 to 7 days from time ship is available on docksc

4. Payment Total

Nationalized bank pays against a sight letter of credit.

15 to 30 days 5 daysd

a In the past, excise inspector made personal visits for domestic shipments as well, but a change in the procedure for domestic despatches now permits self-removal. b Today, most consignments are not physically checked, but the process must be followed nonetheless. c Ship-loading times in India are among the worst in the world. d From the time goods are ready at factory to the time goods are on aircraft or ship. Shipment to Bombay is 1 day, airline cooling-off period is 1 day, minimum procedural time is 3 days. Walking the goods through (with one person devoted exclusively to the job) can reduce the total from 5 days to 2 or 3 days.

from the day the item is ready to the day it is on a flight out of Bombay is, on average, four days (see table 4.3). Two years ago, a customer in Sri Lanka wanted to buy a valve we had in stock. Ours was half the price of our Singapore associate. However, by physically walking the product through every process personally, we would take two days to have it on a flight to Sri Lanka. Singapore had it on a flight that same afternoon and we lost the order. Take another example, in April 2000. A start-up in Pune imported a biodigesting chemical. The invoice listed kilograms; the bottle listed liters. The customs refused to clear the consignment because of this discrepancy, in spite of the firm offering to file a bond. The reagent ultimately went dead on

Doing Business in India: What Has Liberalization Changed?

141

the shelf in customs and had to be scrapped. When such stories make the rounds at meetings between industry associations and bureaucrats, there is always complete agreement on the need for change. What is missing is the nitty-gritty of reform, the commitment to drive detailed change through at the local level. This is a considerable missed opportunity for reform, as the changes are far too detailed to be politically contentious. 4.4.2 Public Articulation of the Case for Reform Apart from these specific measures, the key qualitative issue, in my opinion, is an inadequate focus on economic growth per se, and too little public articulation of the case for reform and of how reform enhances the wellbeing of the ordinary citizen. Ashok Desai, Swaminathan Aiyar, and T. N. Ninan write hard-hitting editorials offering insight, but they are mainly read by the converted. The front page headlines of their own newspapers repeat populist liberalization-bashing by every party when liberals are not in power, and often even when they are in government. India’s economic performance from 1991 to 1997 was excellent relative to India’s standards, and decent even compared with the world’s fastest-growing economies. One would, however, never know it from reading India’s very vibrant popular press, which reflects public opinion reasonably accurately. Economic growth is simply not an issue at the top of the public policy agenda in India. After five years of government with unheard-of growth for India, the Congress Party did not even run on its growth record in the 1996 election. Where economic reform did come up at all, it came up apologetically in a “we’re sorry we had to do this” tone. Economic reform was simply not a mass issue, and no political party chose to make it one. Writing about a poll conducted just before the 1996 election, Swaminathan Aiyar had this to say: • Only 19 percent of people had heard of the new economic policy, of whom 10 percent approve. • In rural areas, only 14 percent have heard of the reforms, of whom half approve. • Among graduates, 66 per cent are aware of the issue, and 44 per cent approve. • Only 7 per cent of the very poor have heard of the reforms, of whom 3 per cent approve. Some enthusiasts in the finance ministry may interpret these figures as majority support for the reforms. Left ideologues may interpret them as meaning most poor people are against reform. Both contentions are ludicrous. What the survey really shows is that economic reform is largely a non-issue, especially for the poor (“Swaminomics,” Sunday Times of India, 25 August 1996). Reading a description by Ronald Dore of Japan in the 1950s and 1960s, when economic growth and Japan’s per capita GDP ranking merited inclu-

142

Naushad Forbes

sion in no more than a little table in a side column of the daily newspaper (rather like the weather report), one sees that the contrast with India could not be greater.11 In 1995–96, when the Indian economy grew at 6.5 percent instead of the 5.8 percent estimated by the finance ministry, this merited an article on the inside page of the financial press. When the economy grew at a Tiger-like 7.5 percent in 1996–97, versus the 6.8 percent that was estimated through January 1998, the story again appeared on an inside page in the financial press (in an article titled “Government Figures of 7.5 percent GDP Growth in 1996–97 Belie Advance Estimates”—one would never know the economy grew 10 percent faster than expected without reading the small print!).12 It is my estimate that the average Indian college graduate has no idea that India ranks in the world’s poorest 20 percent of countries, or that China, growing at twice India’s rate for the past twenty years, is now twice as rich.13 All of this adds up to the need to build a strong and articulate public lobby for reform. When Sonia Gandhi stands up in parliament and supports a strike against liberalization, as she did in May 2000, why does everyone accept this as “just politics”? Over the last month, our papers have been full of reports on a recent national sample survey that, most headlines claimed, shows that poverty has not fallen significantly in the 1990s as growth has taken off as it did in the 1980s. The clear message is that liberalization is good for growth but not for poverty reduction. The only concrete response to this has been from the journalist Swaminathan Aiyar, presently sitting in Washington! There have been no pro-reform editorials, no statements by our highly qualified and eloquent present or past finance ministers or senior finance ministry officials. The need for an articulate and public lobby for liberalization is best seen by this illustration of its absence. 4.5 Technology and the National Innovation System:14 Pre- and Post-1991 The success of liberalization for Indian industry will ultimately lie in the emergence of internationally competitive firms. Technical capability is at the heart of competing in the long term.15 Technical capability comes from 11. See Ronald Dore’s many wonderful pieces comparing Japan’s early development after Meiji with that of developing countries, in particular, Dore (1964) and Dore (1971). 12. Business Standard, 7 February 1998. Never mind that it took ten months after the financial year ended to discover that the economy grew 10 percent faster than thought earlier! Moreover, two years later the growth rates were revised upward to an even more tigerlike 7.9 percent in 1994, 8 percent in 1995, and 7.5 percent in 1996. 13. Parochialism is not limited to India, of course. I am reminded of the poll of American undergraduates, taken some years back, at the height of the Cold War, reporting that over half the students thought the Soviet Union was a founding member of NATO. 14. For a conceptual background on this, see Nelson (1993). 15. A distinction is drawn between competitiveness in the short term and technical capability, which is the ability to compete in the long term. Short-term competitiveness can come, for example, from low labor cost. However, as labor costs rise, firms need to build the technical capability for productivity and higher-value added activities. See Forbes et al. (2000) for more detail.

Doing Business in India: What Has Liberalization Changed? Table 4.4

How India Compares in Spending on Research and Development

R&D Spending (US$M)

India South Korea United States Brazil Mexico Malaysia Thailand Indonesia

143

R&D % of GDP

Gross Domestic Expenditure on R&D by Source of Funds (%)a

1990

1996

1990

1996

Government

Universities

Business Enterprises

2,495 3,209 150,765 899 427 36 104 154

2,188 13,522 184,665 4,070 886 226 208 187

0.9 1.8 2.8 0.4 0.2 0.1 0.2 0.2

0.8 2.8 2.4 0.6 0.3 0.3 0.1 0.1

75 16 36 57 66 14 61 16

1 — 5 3 8 — 7 1

24 84 59 40 18 8 — 76

Sources: [http://unescostat.unesco.org/en/stats/stats0.htm]; Lall (1996); NSF (1993) a Data for India, South Korea, and Indonesia for 1994; for United States and Mexico for 1995; and for Brazil, Malaysia, and Thailand for 1996.

learning, from technical effort. Four decades of protection and inwardlooking policies fostered much technical effort. No developing country had a leader as interested in science as Nehru;16 or invested as much, as consciously, and as early in science and technology as India;17 or protected local technology as much. How much of this technical effort has been useful, though, in building the technical capabilities necessary for internationally competitive firms? 4.5.1 Pre-1991 India’s industrial policy through 1991 can best be summed up in a quotation from Nehru, speaking in the 1950s: “I believe as a practical proposition that it is better to have a second-rate thing made in one’s own country than a first-rate thing one has to import.” Policies affecting technology developed hand-in-hand with this inward-looking industrial policy regime, with an ideology of self-reliance prevailing for over four decades. The Indian Patent Act of 1970 was the most conscious attempt among developing countries (together with Brazil) to improve terms for accessing international intellectual property (Bagchi 1984). A motif of self-reliance as an end in itself was given form through major investments by the state in civilian R&D and by a policy of protecting local technical effort by strictly regulating and restricting the import of technology. The state has dominated R&D in India since independence as both funder and doer (table 4.4). 16. Nehru’s interest in science is well documented. Any collection of his speeches contains several gems on the value of science—as a way of thinking and of life generally, and for scientific research particularly. Nehru made a point of attending every Indian Science Congress and, as prime minister, of meeting leading international scientists when they visited India. 17. See Nayar (1983) for a background on India’s approach to science.

144

Naushad Forbes

The Indian state invested more in science, and did it earlier, than any other developing country. By 1980, India invested 0.6 percent of GDP in R&D, more than any other newly industrialized country, or NIC, at that time, about the same level as South Korea and Taiwan, two countries whose industrial progress had been much more rapid.18 As of 1991, India spent US$1.8 billion on R&D, 86 percent financed by the state, and 74 percent performed in state laboratories. Twelve percent of national R&D was financed by the state but carried out in the in-house R&D laboratories of public sector firms, and this should more properly be combined with private sector in-house R&D. Adhering to the simultaneous protection of industry, the imperative for Indian firms was to produce everything locally. This technical effort aimed at indigenization: Productivity and product improvement were not part of the imperative. Much of Indian industry was characterized by widespread inefficiency and product obsolescence. As the nature of competition changed with liberalization, much of this technical effort turned out to be useless. 4.5.2 The Content of R&D through 1991 More important than quantitative indicators is the qualitative issue of just what we mean by R&D: What was the content of Indian R&D? Very few serious studies of R&D in India have examined this question, the best being a pioneering effort by Ashok Desai in 1969 with a follow-up in 1980. In a study on private industry in India, he concluded that research absorbed approximately 2–3 percent of corporate R&D, development around 30–40 percent, with operational investigations—problems of raw material supply, manufacturing problems, and customer’s problems—absorbing the rest of corporate R&D effort (Desai 1980, 81). It can be expected that the proportion of R&D time spent on “operational investigations” would have dropped by 1991, largely as a result of a more formal separation of the R&D activity both functionally and geographically, which provided insulation from the demands of day-to-day skirmishing. What can be termed development would then constitute the great bulk of R&D. Development, though, meant something quite different in India: indigenization. If a product was imported, one must instead produce it locally. The content of R&D in industry, then, became primarily one of developing local suppliers of raw materials and components, developing substitutes where the exact item was unavailable, and developing a local manufacturing process. The main objective was to achieve a product that could be made locally almost in its entirety (Desai 1988). Not only was indigenization almost the only objective of R&D, but it was seen as the only proper objective of R&D. One need only 18. This continues to be the case: Only Korea, Taiwan, and Singapore—whose average R&D investment is at the OECD average—today invest proportionately more in R&D than India.

Doing Business in India: What Has Liberalization Changed?

145

observe the awards introduced “to recognize R&D effort” by the government (the Department of Science and Technology [DST]) and by various chambers of commerce (for example, the Federation of Indian Chambers of Commerce and Industry). All, including those presented at the annual R&D summit jointly held by the DST and the chambers, rewarded import substitution.19 Whether the Indian product was comparable to what was internationally available or was sold at a price that reflected international competitiveness was not a secondary concern: It was no concern at all. Most private firms were quite happy with this state of affairs, content to make obsolete products in a stable economic environment, deliver them late, and charge high prices for them. The true miracle is that, in spite of the incentive structure, some firms did worry about the development of new products. New product development and R&D in general, however, became something good to do, not an imperative for survival. Largely absent in this picture is the kind of firm that led to the emergence of East Asia as a model of industrial development: the firm that began by importing technology from the developed world and ended up being an internationally competitive, technologically independent firm. Firms such as Posco, the four leading Korean chaebols Samsung, Hyundai, Daewoo, and LG, the Taiwanese firms Acer, Taiwan Semiconductor Manufacturing Corporation (TSMC), and others20 are the source of industrial growth in their countries. The contrast with India is striking, and reflects the difference with which the government pursued a policy of self-reliance. The difference was not wideranging infant-industry protection or the objective of self-reliance, common features in both Korea and India. In India, infant-industry protection combined with an inward-looking trade regime to protect permanent infants, while in Korea and Taiwan, the infants had to grow up and compete internationally. The government intervened directly to ensure that firms that did grow up got more subsidies, while those that did not were merged with the more successful firms.21 Thus, R&D in Indian industry meant one word: indigenization. If the finished product, intermediate or raw material was imported, Indian companies were to source it locally. Indigenization took place at any cost and with compromises in quality. A general lack of access to imports forced much 19. These awards for import substitution are still presented today, nine years after liberalization began. Forty years of import substitution means there is still an automatic, if slowly weakening, association between technical effort and indigenization. 20. See Kim (1997) for a series of such case studies from Korea. See also Hobday (1995) for cases from Taiwan, Singapore, and Hong Kong in addition to Korea. The criticism that the chaebols have faced in the Korean financial crisis of the late 1990s may be well placed, but their problems should not obscure the great technical capability that they have built up in a wide range of industries. 21. This is why the whole states versus markets debate has focused on the wrong question. The issue was not state or market, but what mix of state and market led to what kinds of learning in firms.

146

Naushad Forbes

technical effort through indigenization, and this technical effort involved much learning. There is no question that Indian industry built substantial technical capability by learning to make almost everything locally. The key question is how much of this technical effort was worthwhile and would turn out to be useful in building internationally competitive products. Waste was rife: indigenization resulted, for example, in locally manufactured printed circuit boards (PCBs) with runs of twenty boards per year (imports were banned through 1991 with no exceptions) and in a personal computer industry that ten years ago was shown to have added negative value locally.22 But indigenization also resulted in forcing the local development of a range of intermediate components of some sophistication. India in 1991 had a large and fairly competitive casting, forging, and PCB industry.23 While many suppliers produced what could charitably be called substandard products, there were also several who made products of a high enough quality to be exported to Germany. The same applied for many other industries. Indeed, even profitable firms in technology-intensive industries in India were characterized by contrasts—bad-quality manufacturers coexisted with decent-quality ones, and firms that imported everything in knocked-down form thrived alongside those who added significant value locally. In the process of indigenization as job number one, number two and number three, capabilities both useful and useless developed simultaneously. Overall, one can sum up India’s technological achievements by 1991 by recalling Nehru’s “practical proposition” of making second-rate goods in the country instead of importing first-rate ones. That is exactly what India achieved. 4.5.3 What has Changed for Technology after Nine Years of Liberalization? Competition from New Firms and New Products The change in the last nine years in the availability of products that are at or close to the international frontier is dramatic. The visibility of foreign brands has increased dramatically: Small towns boast Coca-Cola and Akai signs, while in the larger cities cellular phone services from international players vie for billboard space with signs proclaiming Citibank’s sleep disorders.24 More importantly, there has been a dramatic increase in competi22. Local value added was negative because the import of complete PCs was not allowed. As a result, firms imported components, boxes, screens, and keyboards—all separately. The total cost in foreign exchange of importing such PC components exceeded the cost of buying it whole! See Mahalingam (1989). 23. My estimate is that there are on the order of 10,000 foundries, forgers, or PCB manufacturers all over the country, ranging hugely in size and capability. 24. The Citi, as we are told, never sleeps.

Doing Business in India: What Has Liberalization Changed?

147

tion across the economy. Anyone who doubts the impact of competition on service levels and innovation (a point that will recur later in a specific discussion of the effect of the changes on technology) has not been a frequent—and recent—visitor to India. The quality of air service improved dramatically with the entry of private airlines—it used to be said that Indian Airlines published a timetable for the sole purpose of allowing passengers to calculate how late they were—and the improvement has continued even as most private carriers subsequently disappeared.25 The sudden availability of a range of international quality consumer goods has led to a boom in demand for consumer durables like refrigerators, two-wheelers, and television sets—all advertised on Star TV, and all growing in volume at 20–30 percent annually. Refrigerators by Samsung and GE and televisions by Sony and Akai (and even Konka of China) compete head-on with those from Godrej, Voltas, BPL, Videocon, and Philips (seen as an “Indian” firm, since it has been in India so long). The change in industrial products has been even more dramatic: Essentially all capital goods can be imported at a nominal tariff of 20 percent (and an effective tariff that is negative in some sectors), and every major international player now competes with Indian firms. It is this increase in competition, both foreign and domestic, that is driving changes in technology. It is important to recognize that it is the nature of competition that changed, not just the amount. Most Indian markets have always been competitive, in terms of the number of firms that make up the market. However, each firm did much the same thing. As Madhur Bajaj, the operating head of India’s largest two-wheeler manufacturer, Bajaj Auto, puts it: With a waiting list of five to eight years for a scooter (or car, for that matter, all involving twenty-year-old technology) and a moratorium on capacity expansion, “the job of Marketing was one of Allocation, not Selling” (personal communication, December 1999). Since 1991, most markets have seen several new players enter, primarily foreign firms, but also Indian firms that were formerly prevented from entering a particular market. They have entered the market by doing different things, providing new products, or using more efficient processes. Coupled with freer imports, the new firms offer today’s Indian consumer the choice of an internationally available product at an internationally comparable price. With this change, firms that manufacture locally have had to become efficient and to introduce new products, 25. The changes in the airline industry characterize what has and has not changed in India: The change from the Indian Airlines monopoly of ten years ago is dramatic. Today, the one thriving private sector carrier, Jet Airways, provides excellent service—and competition has led to a dramatic improvement in Indian Airlines, which has discovered “the passenger.” This is an indication of how much has changed. However, the fact that a joint venture between Tatas and Singapore Airlines has been simultaneously forbidden to proceed by the Congress, United Front, and BJP governments, and on the flimsiest of pretexts, indicates how much still needs to change. The real reason is said to be that Jet Airways has influenced each government to refuse Tatas permission.

148

Naushad Forbes

both practices that involve much technological change. This in turn has driven the import of more technology and, in some cases, greater attention to in-house R&D. The overall impression is one of significantly greater technological dynamism, driven by the demand for innovation. 4.5.4 The Macro-View of Technology in Industry since 1991 A macro-view of technology in India after nine years of liberalization shows some change albeit not a dramatic one. The percentage of total spending on R&D by the state (including PSEs) is slightly lower (78 percent in 1996–97 versus 86 percent in 1991),26 more from the inherent inertia in state R&D budgets than from explicit policy. A very similar pattern continues in the number of private firms performing R&D (about 1,200) and their concentration in industrial sectors: Transportation, pharmaceuticals, chemicals, and electricals and electronics account for about 60 percent of the total for private industry. Technology imports continue to provide the bulk of new manufacturing processes and new product technologies, and have significantly increased. Coupled with the major increase in direct foreign investment we discussed earlier, total investment in technology by Indian industry would seem to have increased, with a rise in spending on both technology import and in-house development.27 4.5.5 The Micro-View The level of the individual firm points, however, to more radical change of technology in Indian industry. This evidence is based on a small pilot study of the changes liberalization has brought to technology in Indian industry performed at the Asia Pacific Research Center (APARC) at Stanford University in 1998. What evidence I can piece together from a combination of newspaper and magazine reports, personal visits to firms, and direct management of a firm in India shows much more technological change than the macro-statistics would indicate. Firms have responded to liberalization in four ways. First, some industries have restructured, and the import of technology has shown a significant increase, in the form of both increased foreign investment and technology licensing. Second, many firms have improved manufacturing efficiency. Third, it seems as if some (perhaps only a few) firms have chosen 26. The Department of Science and Technology publishes an annual in R&D statistics and industry, the most authoritative source of data on Indian R&D. The most recent volume available is for 1996–97, published in June 1999, with a two- to three-year delay. However, less systematic reports indicate that the share of industrial spending in the total (including PSEs) has continued to rise, and for the year ending March 2000 should be around 35 percent of total national R&D spending, up from 25 percent in 1991. 27. Although the rise is significant, too much should not be made of it—both imports and in-house investment have risen over a small base. As of 1991, technology imports both in absolute terms and relative to sales were the lowest of all the major NICs—Korea, Taiwan, China, Brazil, and Mexico.

Doing Business in India: What Has Liberalization Changed?

149

to substantially increase investment in in-house R&D. Finally, some firms have not responded to liberalization with any substantive change; they are increasingly having change thrust upon them. Two caveats accompany the analysis that follows. First, these technological responses have indeed come from competition: Liberalization has driven competition, and competition has driven technical change. But they have not come only from competition; the greater freedom of a liberalized environment and the ambition of Indian firms are being translated into technological initiatives. Second, these responses are not exclusive. Firms have often changed in some combination of these three responses. Indeed, the changes are cumulative: The firms raising spending on in-house R&D are to some degree a subset of those that have invested in higher technology imports, who are in turn to some degree a subset of the firms that have improved manufacturing efficiency. Industry Restructuring There are several visible examples of industry trying to restructure. Past government policy that parceled out industry capacity to different groups means that India’s larger firms have a huge range of business activities. There has been some attempt by some groups to focus on fewer activities: SRF, for example, bought a tire-cord plant—an industry in which it is the leading player—and sold its finance company to its joint venture partner. The Tata group sold its personal products company Tomco to HLL, which has emerged as the country’s leading personal products and foodprocessing firm, and the firm with the highest market capitalization on the Bombay Stock Exchange. Tata Tea, in turn, bought Tetley Tea of the United Kingdom, in the largest overseas acquisition by any Indian firm. However, group diversity still remains more similar to the structure one would find with the Korean chaebol than to the Anglo-Saxon model of focused firms. It seems as if every Indian group of any size invested in power, telecommunications, and finance ventures in the 1993–95 period, all sectors in which they usually had no background. It is only in the last four years, as industrial growth has fallen and industry has come under the dual pressure of competition from imports and falling margins, that firms have been forced to look at which activities they really wish to retain. Ready buyers for the businesses being sold by Indian groups as they restructure are foreign firms (one sees the same thing beginning to happen with the chaebols as a result of the financial crisis in Korea). Foreign firms are increasing their share of or buying out their joint venture partners, such as Cummins with Kirloskars or the HLL-Tomco buyout mentioned earlier. The soft-drink market has seen Coca-Cola buy out Parle, owner of the leading Indian brand Thums Up, and now compete head-on with Pepsi. However, the story has by no means only been one of foreign firms buying out or eclipsing Indian firms. The U.S. cosmetics firm Revlon entered the market

150

Naushad Forbes

with a sub-standard product and awoke the Indian firm Lakme into action, which went on to trounce Revlon—which withdrew from the market and re-entered more cautiously.28 One also sees firms in many industries investing in new process plants—including the previously dormant textile industry, where firms have invested in more new plants in the last five years than they did in the previous twenty. Response 1: Become more efficient. Most firms have seen significant improvements in productivity, reflected in a much improved capital-output ratio for the economy as a whole. A quick look at the Super 100 index for 1998–99 in table 4.5 shows that profits have grown faster than sales. Profitability is an imperfect indicator of efficiency, but the overall impression is of much efficiency gain when combined with countless stories in the business press of firms reducing employment through voluntary retirement schemes. In the last year, ending March 2000, the first 1,036 results declared show sales rising 15 percent and net profits 27 percent.29 Some of this rise has come from a fall in interest payments, mainly through retiring debt as working capital has been better managed. Response 2: Increase imports of technology. Both the number and value of technology collaborations have risen sharply since 1991. The number of technology collaborations approved averaged 830 from 1986 to 1990, rising to 1,630 from 1991 to 1995 (DSIR 1995). Cases involving foreign investment rose even faster—from an average of 240 per year to an average of 840. Approval of payment for technology also rose, from US$300 million a year to US$1 billion a year. Combined with the increase in DFI (from an average of US$150 million per year in the 1980s to US$3 billion per year by the mid-1990s) that we spoke of earlier, this adds up to a significantly higher level of technology imports. Qualitative evidence of growing technology imports complements this picture. Six of the nine engineering firms we visited in our APARC study have significantly increased technology imports. Several firms cited the freedom to license technology as a major factor, but were particularly appreciative of smaller operational changes that greatly eased the ability to acquire foreign technology. For example, before 1991 if an Indian firm wished to hire a foreign consultant and pay him even US$1000, it required a separate application, with much chasing with the Reserve Bank (RBI) in Bombay and perhaps even ministries in Delhi. Approval would almost 28. In a further indication of industry restructuring, the Tata group has since sold Lakme to Hindustan Levers. 29. Business Standard, 22 May 2000. It is worth adding that excellent results from Reliance dominate the picture, accounting for 10 percent of the total net profits of the 1,000
firms and 25 percent of the profit increase over the previous year.

Tata Chemicals Tata Power Company Videocon International Nestle (continued)

Nirma

Reliance Industries Larsen and Toubro TISCO Hindustan Lever Hindalco Industries ITC Grasim Industries BSES Bajaj Auto Mahindra and Mahindra Sterlite Industries India ACC Indo Gulf Corp Gujarat Ambuja Cement Ranbaxy Laboratories Wipro

Company

Table 4.5

Tata Birla New New New

Commodities Commodities Commodities

Pharmaceuticals

Information Technologies Consumer Products Engineering Utilities

Consumer Products

22,840 15,450

Multinational

13,950 12,690

12,750

18,040

14,310

23,990 14,900 10,890

18,480

132,310 69,930 57,420 97,270 19,120 36,370 38,670 23,490 33,420 35,710

Sales

New

Tata Tata

New

New

Commodities

Commodities Utilities Engineering Engineering

Commodities

Ambani Professional Tata Multinational Birla Multinational Birla Professional Bajaj Mahindra

Ownership

Commodities Engineering Commodities

Industry

1999

862

1,509

1,898 1,658

1,709

1,702

1,560

568 1,640 1,505

1,608

17,036 3,896 3,976 8,374 5,668 6,234 1,638 2,702 5,405 2,286

Profits

Top 100 Private Firms from Business India Super 100 (1998–99)

6,704

10,917

8,315 10,892

3,488

7,633

6,869

18323 5,162 4,095

5,307

56,491 31,150 42,091 28,869 10,830 25,397 21,739 12,144 19,300 18,075

Sales

405

881

2,867 1,182

416

322

1,104

1,443 1,686 1,005

847

10,649 2,774 2,811 1,900 2,920 2,616 3,086 1,467 3,052 1,170

Profits

1995

3,148

5,173

3,447 4,765

2,603

11,697

21,382 14,335 23,308 12,071 6,965 23,164 13,234 4,668 12,146 10,379

Sales

151

262

445 346

121

1,213

1,256 913 1,601 659 648 775 1,005 332 510 123

Profits

1991

5

4

4 3

5

2

6 5 2 8 3 2 3 5 3 3

SalesGrowth Multiple (1999/1991)

6

6

4 5

13

0

14 4 2 13 9 8 2 8 11 19

Profits Growth Multiple (1999/1991)

Ashok Leyland SPIC Indian Hotels ISPAT Industries

Chambal Fertilizers TVS Suzuki Indian Aluminum Motor Industries MRPL Castrol India Tata Tea

Indian Rayon BPL TELCO Raymond

Nagarjuna Fertilizers Hero Honda Motors Great Eastern Shipping MRF

Company

Table 4.5

New

Consumer Products Commodities Engineering Engineering ConsumerTextiles Commodities Engineering

Consumer Products Engineering Commodities Services Commodities

Commodities

Munjal Sheth

Engineering Services

New Chidambaram Tata New

New Iyengar Multinational Multinational Birla Multinational Tata

Birla New Tata Singhania

New

Ownership

Commodities

Industry

(continued)

18,540 24,330 6,230 13,010

8,020 13,290 10,340 12,250 26,060 9,480 8,740

13,930 17,860 57,180 12,970

17,860

15,500 9,810

12,410

Sales

1999

204 508 1,197 251

1,455 821 788 529 141 1,784 1,288

1,060 1,025 –49 866

1,022

1,213 1,264

1,437

Profits

13,611 12,612 3,820 8,120

5,224 4,527

6,646 4,119 9,097

9,956 7,826 50,442 8,026

8,758

4,822 8,247

8,630

Sales

706 675 821 790

730 593

1,329 320 918

1,329 484 3,190 705

208

194 1,716

1,929

Profits

1995

263 63 110

3,560

516 9,274 8,536

3,071

517 149

1,421 299

25,960 4,377

5,842 3,324

150

191

158 371

Profits

5,085

6,866

2,155 2,891

Sales

1991

4

2 3

3

2 4

2 3

3

3

7 3

SalesGrowth Multiple (1999/1991)

2

1 8

2

2 4

0 3

7

5

8 3

Profits Growth Multiple (1999/1991)

Phillips India (continued)

Cummins India Duncans Industries Essar Steel Global Telesystems Infosys Technologies

Jindal Strips Exide Industries Brittania Industries

Colgate Palmolive CIPLA Ahmedabad Electricity Oswal Chemicals Usha Beltron Adani Export Asea Brown Boveri Apollo Tyres

Escorts EIH Dabur Jaiprakash Industries Pentafour Software

Asian Paints

Commodities Commodities Services Information Technologies

Consumer Products Commodities Commodities Consumer Products

Commodities Other Other

Pharmaceuticals Utilities

Consumer Products Engineering Services Pharmaceuticals Other Information Technologies

Multinational

Multinational Goenka New New New

Jindal Shroff Wadia

New Multinational Raunaq Singh

Oswal

Multinational New Professional

New

Nanda Oberoi Burman

New

16,410

6,620 10,560 21,340 5,400 5,130

10,370 6,460 10,150

7,330 7,810 21,890 8,930 9,030

9,720 6,160 8,820

11,370 4,760 9,010 9,370 5,310

9,100

427

748 769 –4,965 632 1,372

374 412 396

291 426 673 377 400

457 1,150 451

321 964 497 327 1,191

698

11,191

442

510

184

5,143 5,510

727

508 259

170

712 248 139

509 581 239 955

435

7,162

6,435 5,791

3,070

6,540 2,839 5,387

12,693 2,878 4,050 6,805

5,220

5,693

2,780

3,680

3,434

2,013

3,010

2,699

292

190

101

205

144

420

79

345

331

5,070

3,285

334

158

9,759

2,635

3

2

3

3

4

2

3

3

2

1

3

1

4

4

2

3

1

6

1

1

1

4

Lupin Laboratories

Smithkline Beecham Usha (India) HCL Infosystems

Finolex Industries Crompton Greaves Eveready Industries ICI India EID Parry India

Satyam Computer

South India Corp NIIT

Novartis India Glaxo Madras Cements Arvind Mills

Century Textiles

Company

Table 4.5

Engineering Information Technologies Pharmaceuticals

Consumer Products and Commodities

Commodities Consumer– Textiles Services Information Technologies Information Technologies Commodities Engineering Engineering

Consumer– Textiles

Industry

(continued)

New

Multinational New New

New Thapar New Multinational Maruguppa

New

New New

Lalbhai

Multinational Multinational

Birla

Ownership

6,710

5,740 13,900 8,970

5,250 15,580 7,360 7,690 9,310

3,780

17,110 4,590

7,340 7,610 5,240 9,830

19,770

Sales

1999

253

808 441 585

480 231 348 349 399

728

239 1,084

747 671 399 145

–913

Profits

3,942

2,506 3,696 5,877

3,256 4,755 3,070 5,186 5,263

4,690 5,652 2,989 6,389

Sales

352

272 312 408

333 190 245 331.5 232.1

200 190 534 1,059

8,839

Profits

1995

211 65

7,568.2 2,618

61

212

140 106

2

Profits

6,244

3,020 3,703

824

Sales

1991

1.2 4

2

2 2

–1

SalesGrowth Multiple (1999/1991)

10

6

1

5 6

Profits Growth Multiple (1999/1991)

Commodities Commodities Consumer Products Commodities

Commodities Commodities Commodities Commodities

Engineering

11,660 108,405

1,598,540

5,220 6,770 10,100

4,400 9,900 6,060 8,750 7,740 5,380

New

New New Goenka

New Multinational Birla Birla New New

4,090

Multinational

Pharmaceuticals

5,110 10,480 3,850

Tata Multinational New

Utilities

10,670 17,010 4,080 4,400 7,120 9,130

Singhania Birla New New Multinational Multinational

Commodities Utilities Services Services

Source: Business India, 29 November 1999. Note: Sales and profits in millions of rupees (Rs M).

Indo Rama Synthetics Total

JK Industries CESC HFCL Essar Shipping Bata India Hind Lever Chemicals Tata Hydro-electric Siemens Dr Reddys Laboratories Proctor & Gamble India Bharat Forge Coats Viyella India Century Enka Zuari Industries Malavika Steel Jindai Vijayanagar Steel Deepak Fertilizers Bhushan Steel Ceat

787,940

–1,571

727 405 173

373 263 313 130 115 –290

559

696 –227 518

221 –1,281 374 480 383 424

77,078

2,647 8,062

3,171 7,579 2,865 6,917

2,524

393,820

292 212

258 420 83 576

140

452 351

907 10

4,856 4,841

4,318 8,981

181 685

5,335 9,686

19,850

5,153

5,234 2,872

2,471

3,368 5,430

4

224

325 110

144

166 94

5

2

1 3

2

3 3

1

1 1

3

1 –14

156

Naushad Forbes

always eventually be granted, especially from the 1980s onwards, but the sheer hassle of following up with several trips to Bombay and Delhi acted as a considerable deterrent even to applying. Since 1991, approval is still required but is “automatic”—one simply needs to register the payment with the RBI. Three firms we visited, Telco, Bajaj Auto, and Mahindra & Mahindra (all in the automotive industry), gave us examples of licensing new product designs or acquiring engine technology. They recounted descriptions of sending teams of engineers to Italy, Japan, the United States, and Australia to work alongside the foreign design consultants hired and to absorb the tacit knowledge that is so essential to building technical competence. All three firms are today spending an order of magnitude more on technology import than they did before 1991, but all have chosen to unpackage technology. For example, the technology for engines was licensed from firms in Japan and Australia, the styling from Italy, and the manufacturing and tool-making capability from Japan, with the Indian firm thus retaining proprietary control over the complete product. The descriptions read very much like descriptions of Korean firms such as Hyundai that, ten or fifteen years earlier, sent teams similarly to the United States and Japan to learn from different technology sources.30 Response 3: Re-think in-house R&D and invest more in it. The data show some rise in R&D spending by Indian firms, to around 0.64 percent of sales in 199631 for 990 private firms with R&D units registered with the department of science and technology, still a very small number in international terms (DST 1999). The aggregate figures conceal, as in every country, major inter-firm differences: Even more, the small aggregate change conceals major change in some industry sectors and firms. Table 4.6 shows R&D spending for the top twenty firms performing R&D in Indian industry in 1998–99 versus 1992–93. Note the jump in R&D spending for several private sector firms—up two to twenty times in six years. Note also, though, that the absolute quantum of spending is low. The total spending of the top ten R&D firms in India is under US$150 million. Firms argue that R&D is cheap in India, based on an abundance of low-cost skilled labor. Several foreign firms, following the software lead, have now set up R&D labs in India. General Electric, for instance, has set up its second largest R&D centre in the world (and largest outside the United States) in Bangalore, recruiting around 3,000 people with research facilities for at least three GE companies—GE plastics, GE India Technology Center, and GE aircraft engines. A few firms have fundamentally rethought their approach to in-house 30. See Kim (1997) for several such descriptions. 31. It is probably higher today, but will still be below 1 percent of sales.

Doing Business in India: What Has Liberalization Changed? Table 4.6

157

Research and Development Expenditures

Firm Reliance Industries Ltd. Mahindra & Mahindra Ltd. Ranbaxy Laboratories Ltd. Eicher Ltd. Wockhardt Ltd. Indian Oil Corporation Ltd. Crompton Greaves Ltd. Hindustan Lever Ltd. TELCO Ashok Leyland Ltd. Bajaj Auto Indian Telephone Industries Ltd. Bharat Heavy Electricals Ltd. Steel Authority of India Ltd. Oil & Natural Gas Corporation Ltd. Bharat Electronics Ltd. DRL

1998–99a

1992–93a

Growth Multiple

751 414 523 222 156 772 217 373 1,000 217 315 338 527 483 250 661 212b

24 33 84 40 33 185 54 113 308 94 144 212 430 395 221 705 31

31 12 6 5 5 4 4 3 3 2 2 2 1 1 1 1 7

Source: Research and Development in Industry 1992–93 and 1998–99. In millions of rupees (Rs M) b Figure for 1996–97. a

product development, involving changing the entire role of in-house R&D. The task for R&D has moved from indigenization to developing products with technology that is distinctive and proprietary to the firm. Each of the three auto-industry firms mentioned above has sharply increased R&D spending in the last ten years. Each employs from two to four times as many R&D engineers as ten years ago, with R&D spending as a percent of sales more than doubling. Expatriate managers have been hired (for example, Mahindra & Mahindra have hired a head of R&D from GM and a president from Ford) and key individuals in R&D have been trained at leading firms overseas. The change in R&D investments is particularly dramatic in the pharmaceutical sector. Many Indian pharmaceutical firms benefited from the 1970 Indian Patent Act (which made product patents illegal and only permitted process patents for pharmaceuticals), and started by producing a series of reverse-engineered drugs. Change is being driven both by liberalization in general and more importantly, by India’s signing of the General Agreement on Tariffs and Trade/World Trade Organization agreement, which means pharmaceuticals within the country will be protected by product patents by the year 2005.32 Research and development spending by Indian pharma32. See Smith (2000) for more detail on the impact of the GATT/WTO agreement on Indian pharmaceutical firms.

158

Naushad Forbes

ceutical firms has doubled in the last five years (to US$70 million) as exports have tripled (to US$1.5 billion).33 The qualitative picture again complements and accentuates this change: Several Indian pharmaceutical firms, in particular Ranbaxy, Dr. Reddy’s Laboratories (DRL), and Lupin, have launched their own drug-discovery programs, with the objective of patenting their own molecules. Both Ranbaxy and DRL have recently licensed patented molecules to MNCs, and both have bought foreign firms as a way of entering international markets. Most recently, Ranbaxy bought the generics business of Bayer as a way of entering the German market. The Indian firm Nicholas Piramal (NPIL) bought the entire R&D laboratory of Hoechst Marion Roussel as a way of jump-starting its own R&D effort. Individually Wockhardt, Cipla, NPIL, Ranbaxy, DRL, and Lupin are today spending an order of magnitude more on R&D than ten years ago. Again, foreign scientists have been hired, and Ranbaxy has even started a modest R&D effort in the United States as a way of “monitoring” new developments—shades of Japan and South Korea again. The software sector has attracted the greatest publicity, with an annual growth rate of 50 percent over the last decade, exports growing even faster, and firm valuations that have skyrocketed to the point that it is cheap for some Indian software firms to acquire American software firms. The success has been built around an abundance of low cost English-speaking programmers and not product development skills. The software sector cannot properly be seen as a beneficiary of liberalization, however, except in the most general sense of easing the import and cost of computer hardware, making foreign travel easier, and freeing up the capital market. The industry was never subject to licensing, and could essentially bypass import and export controls. The fear now, as the Delhi wits have it, is that the success of the software sector could finally be in jeopardy because the government has finally recognized its importance and created a ministry for it. 4.6 Winners and Losers: 1991 to 1999 So far, it has been argued that change in Indian industry since 1991 has been dramatic. This change shows up in the numbers, and it shows up even more in the qualitative picture of change in individual firms. What is clear is that the criteria for success in Indian industry have changed from capturing industrial licenses through acquaintance with Rajiv Gandhi, bribes for bureaucrats, or liaison men in Delhi. Today, success increasingly comes from making a better product more efficiently than anyone else, as it does in most of the world. Indian industry is becoming normal. But which firms have succeeded in this new economic environment? Who 33. Numbers from Business India Intelligence, February 2000.

Doing Business in India: What Has Liberalization Changed?

159

are the winners and who are the losers? To approach this question somewhat systematically, let us return to the Business India Super 100 indices for 1991 and 1999. What changes are visible? Table 4.7 lists the winners, defined as firms that are in the 1999 Super 100 index but were not in 1991 (section A) and as firms that have increased their profitability by more than four times since 1991 (section B). Table 4.8 lists the losers, firms that were in the Super 100 in 1991 but are either not even in the Super 250 in 1999 (section A; call these the Super-losers), or have dropped more than fifty places since 1991 (section B). The winners: Of the Super 100 firms for 1999, forty-eight are new entrants from 1991 (table 4.7, section A). Seventeen firms are in the traditional commodity industries, of which six are in fertilizers (where success is built around a government subsidy).34 Seven firms are consumer product and engineering firms, in consumer electronics, two-wheelers, textiles, and ready-made garments, and detergents. Six are software firms and four pharmaceuticals.35 Multinational corporations total up to six, with ABB, Castrol, Hind Lever Chemicals, and Smithkline Beecham all representing the liberalized environment in which they can operate.36 Six are in services such as hotels and logistics. The firms that have shown above average growth in profits (table 4.7, section B), are concentrated in the non-commodity industries37—consumer goods, engineering, and services: five of the twenty-three firms are in commodities. Five are multinationals, operating in a much less restrictive environment—HLL, Nestle, ITC, Glaxo, and Novartis. Hindustan Levers is a case in point, illustrating how the opening-up after 1991 allowed an excellently managed MNC to expand and enter new businesses. The losers: Of the 100 firms in the Super 100 of 1991, 17 are not even in the Super 250 in 1999, as shown in table 4.8, section A. These super-losers see the old predominate, which comes as no surprise. Eleven of the seventeen firms are in traditional commodity businesses, two are in textiles, two are in consumer products (Dunlop and Shaw Wallace),38 one in engineering (Premier Automobiles, manufacturer of the 1960 Fiat), and one is a multinational (VST). 34. In 1991, commodity firms like those in cement, metal products, and fertilizer would have dominated any such Indian industry list. In 1991, almost half the Super 100 consisted of firms in traditional commodity industries or textiles. 35. There are more pharmaceutical firms as new entrants in the 100 to 250 listing, indicating that many are still relatively small. 36. Two MNCs may seem very small, but remember that this sample is limited to publicly listed firms in India. The new 100 percent subsidiaries like Hyundai or IBM, or the joint ventures like DCM Daewoo and Birla AT&T, would be left out of such a list. 37. The major exception is Reliance, which has followed a strategy of building the world’s largest-capacity plants using the latest licensed technology in world-record construction times—demonstrating that Korea can happen in India. 38. Both were old British firms eventually taken over by Chhabria, an Indian NRI, who spectacularly mismanaged them.

Adani Export

BPL

Asea Brown Boveri Castrol India Chambal Fertilizers Duncans Industries Essar Steel Exide Industries Finolex Industries Gujarat Ambuja Cement Indo Gulf Corporation Madras Cements MRPL Nagarjuna Fertilizers Sterlite Industries India Nirma

Company

Table 4.7

New

Consumer Products Consumer Products Consumer Products New

21,890

17,860

12,750

18,480

673

1,025

1,709

1,608

7,826

3,488

5,307

8,630

2,989

5,162

New–Agarwal

399 141 1,437

1,640

Commodities

5,240 26,060 12,410

14,900

Birla

Multinational Multinational

Commodities Commodities Commodities

Sales

Birla

Profits

Commodities

Sales

484

416

847

1,929

534

1,686

333 1,005

508 730 1,329

Profits

1995

Commodities Commodities Commodities Commodities Commodities Commodities

Ownership

1999

A. New entrants to the Super 100 8,930 377 6,435 9,480 1,784 5,224 8,020 1,455 6,646 Goenka 10,560 769 New–Ruia 21,340 –4,965 Shroff 6,460 412 New–Chhabria 5,250 480 3,256 New 10,890 1,505 4,095

Industry

“Winners”

Sales

Profits

1991

SalesGrowth Multiple (1999/1991)

Profits Growth Multiple (1999/1991)

Lupin Laboratories HFCL (continued)

Usha Beltron CIPLA Dabur Indian Hotels EIH Global Telesystems South India Corporation Eveready Industries EID Parry India Smithkline Beecham Usha (India) HCL Infosystems

Wipro

Satyam Computer

Pentafour Software

TVS Suzuki Infosys Technologies NIIT

Arvind Mills

Engineering Information Technologies Pharmaceuticals Services

Engineering Commodities

Consumer– Textiles Engineering Information Technologies Information Technologies Information Technologies Information Technologies Information Technologies Other Pharmaceuticals Pharmaceuticals Services Services Services Services

New New

Multinational New New

New

Rai New Burman Tata Oberoi New New

New–Premji

New

6,710 4,080

7,360 9,310 5,740 13,900 8,970

7,810 6,160 9,010 6,230 4,760 5,400 17,110

18,040

3,780

5,310

4,590

New New

13,290 5,130

9,830

Iyengar New

Lalbhai

253 374

348 399 808 441 585

426 1,150 497 1,197 964 632 239

1,702

728

1,191

1,084

821 1,372

145

3,942

3,070 5,263 2,506 3,696 5,877

2,839 4,050 3,820 2,878

7,633

4,119

6,389

352

245 232 272 312 408

248 239 821 581

322

320

1,059

2618

61

65

Hindustan Lever ITC Glaxo Nestle Novartis India

Essar Shipping Hind Lever Chemicals Tata Hydro-Electric Dr Reddys Laboratories Proctor and Gamble India Coats Viyella India Malavika Steel Jindal Vijayanagar Steel Deepak Fertilisers Bhushan Steel Indo Rama Synthetics India

Company

Table 4.7

4,090 9,900 7,740 5,380

Multinational Multinational New New New New New

Commodities Commodities

Commodities Commodities Commodities

727 405 –1,571

263 115 –290

559

696 518

480 424

Profits

2,647

7,579

2,524

4,318

4,856

Sales

292

420

140

452

907

Profits

1995 Sales

659 775 105 151 140

Profits

1991

B. Super 100 firms that increased profits 400% or more since 1991 Multinational 97,270 8,374 28,869 1,900 12,071 Multinational 36,370 6,234 25,397 2,616 23,164 Multinational 7,610 671 5,652 190 3,703 Multinational 15,450 862 6,704 405 3,148 Multinational 7,340 747 4,690 200 3,020

5,220 6,770 11,660

5,110 3,850

Tata New

Utilities Pharmaceuticals

4,400 9,130

Sales

New Multinational

Ownership

1999

Services

Industry

(continued)

8.1 1.6 2.1 4.9 2.4

SalesGrowth Multiple (1999/1991)

12.7 8.0 6.3 5.7 5.3

Profits Growth Multiple (1999/1991)

Mahindra Bajaj Munjal Tata Professional New–Singh Professional Professional Tata

New

Engineering Engineering Engineering Engineering Pharmaceuticals

Utilities Utilities

Utilities

Commodities Information Technologies

9,310 8,970

12,690

23,490 8,820

33,420 15,500 13,950 69,930 14,310

35,710

9,100

17,860

New–Mapillai New–Choksey

132,310 19,120 24,330 13,930 22,840

Ambani Birla Chidambaram Birla New–Dhoot

Commodities Commodities Commodities Commodities Consumer Products Consumer Products Consumer Products Engineering

399 585

1,658

2,702 451

5,405 1,213 1,898 3,896 1,560

2,286

698

1,022

17,036 5,668 508 1,060 1,509

5,263 5,877

10,892

12,144 5,387

19,300 4,822 8,315 31,150 6,869

18,075

5,220

8,758

56,491 10,830 12,612 9,956 10,917

232 408

1,182

1,467 139

3,052 194 2,867 2,774 1,104

1,170

435

208

10,649 2,920 675 1,329 881

Sources: Author’s analysis drawn from Business India, 30 September 1991 and 29 November 1999. Note: All figures except Growth Multiples in millions of rupees (Rs.M).

Mahindra & Mahindra Bajaj Auto Hero Honda Motors Tata Chemicals Larsen & Toubro Ranbaxy Laboratories BSES Ahmedabad Electricity Tata Power Company EID Parry India HCL Infosystems

Asian Paints

Reliance Industries Hindalco Industries SPIC Indian Rayon Videocon International MRF

2,618

4,765

4,668 2,699

12,146 2,155 3,447 14,335 2,603

10,379

2,635

6,866

21,382 6,965 8,536 5,086 5,173

65 61

346

332 79

510 158 445 913 121

123

158

191

1,256 648 63 150 262

3.6 9.5

2.7

5.0 3.3

2.8 7.2 4.0 4.9 5.5

3.4

3.5

2.6

6.2 2.7 2.9 2.7 4.4

6.2

4.8

8.1 5.7

10.6 7.7 4.3 4.3 12.9

18.6

4.4

5.4

13.6 8.7 8.1 7.0 5.8

164 Table 4.8

Naushad Forbes Super Losers and Losers 1991

Company

Industry

Ownership

Sales Turnover

Profitability

A. Super losers (1991 Super 100 firms not in the 1999 Super 250) Ballarpur Commodity Thapar 7,135 Baroda Rayon Commodity Gaekwad 2,574 Birla Jute Commodity Birla 5,489 India Cements Commodity Sanmar 2,594 Indian Dyestuff Commodity Mafatlal 2,645 Indian Organic Chemical Commodity Ghai 2,510 J K Synthetics Commodity Singhania 8,392 Modipon Commodity Modi 2,408 Mysore Cements Commodity Birla 2,427 National Organic Commodity Mafatlal 5,159 Special Steels Commodity Tata 2,267 Shaw Wallace Consumer Chhabria 4,024 Dunlop Consumer Chhabria 6,090 Bombay Dyeing Consumer Textiles Wadia 4,572 JCT Consumer Textiles Thapar 3,646 Premier Auto Engineering Walchand 5,591 VST Multinational 4,461

423 37 392 191 89 47 70 220 173 358 127 91 118 359 231 145 234

B. Losers (1991 Super 100 firms that dropped more than 50 places by 1999) Century Enka Commodity Birla 5,234 Coromandel Fert Commodity Professional 2,607 Mukand Commodity Shah 5,817 Orient Paper Commodity Birla 3,215 T. N. Petro Commodity Chidambaram 2,294 Ceat Consumer Goenka 5,154 Godfrey Phillips Consumer Modi 5,359 McDowell Consumer Mallya 3,013 Modi Rubber Consumer Modi 5,256 Voltas Consumer Tata 5,496 Century Text Consumer-Textiles Birla 8,839 Kesoram Consumer-Textiles Birla 3,017 Standard Consumer-Textiles Mafatlal 2,667 Bharat Forge Engineering New-Kalyani 2,471 Hindustan Motors Engineering Birla 6,444 Lakshmi Mach Engineering Lakshmi 2,695 Hindustan Develop Services Modi 2,776 CESC Utilities Goenka 5,430 Hoechst Multinational 2,807 ICI India Multinational 7,568 Coats Viyella India Multinational 4,039 Phillips India Multinational 5,693 SKF Bearings Multinational 2,017 Siemens Multinational 3,906

325 316 117 192 203 224 118 65 71 183 824 97 160 144 3 108 162 94 67 211 327 292 177 113

Sources: Author’s analysis drawn from Business India, 30 September 1991 and 29 November 1999 Note: All figures in millions of rupees (Rs M).

Doing Business in India: What Has Liberalization Changed?

165

It is noteworthy that there are as many MNC losers as there are winners. section B of table 4.8 lists six of them. One might think that MNCs would be prime beneficiaries of liberalization. This is simply not true: For every HLL that has thrived, there is a Siemens or Philips that has struggled to restructure just as many Indian firms have. Finally, a comment on ownership: Old business families dominated Indian industry. Thirty-seven firms of the Super 100 in 1991 were owned by six business families, all dating from well before independence.39 By 1999, Tatas dropped from ten to eight firms, Birlas from twelve to nine (six of the nine remaining Birla firms belong to the Aditya Birla group), Thapar from four to one, Singhania from four to two, and Mafatlal and Modi from four to zero. Qualitative impressions over the last ten years are entirely consistent with this quantitative picture: The Tatas have long been considerably more professionally managed than any of the other groups and avoided the license route to success. The Singhanias, Modis, Mafatlals, and Thapars have been in the public eye more for family feuds than for what they have been doing to restructure their firms. The Aditya Birla group has a reputation for being much better managed than the other Birla groups. In other words, as we all know, it is not just a matter of being in the right business: It is even more a matter of management, of taking the right decisions and responding to change as it comes. If it is good management that determines the success or failure of firms, as everywhere in the world, then that indicates the success of liberalization. 4.7 Conclusion What do we know about how Indian industry has responded to liberalization? While the Indian example demonstrates so clearly the dangers of autarchy and import-substitution, these very policies forced much technical effort by industry and built a wide-ranging manufacturing base. Certainly, much of this was wasteful, creating firms that remained infants for decades until eliminated by competition in the last decade. The losers in table 4.8 show that this process is well under way. But protection did also force effort and learning, which is turning out to be useful in building a proprietary technological base. Would Ranbaxy and Dr. Reddy’s Labs and Mahindra & Mahindra and Tata Tea exist as potential world-beaters without forty years of protection? On the other hand, in a more competitive environment would they and others have emerged a decade and more ago as firms that made “Made in India” mean something? That some firms used protection to build a technical base while others were quite content to re39. Several other old business families appeared in the 1991 Super 100 and also in 1999— Bajaj, Mahindra, Sheth, and Goenka all accounting for one or two firms.

166

Naushad Forbes

main technologically inactive behind protectionist barriers is as clear an indication as any of the difficulty of building technical capability. What is without dispute is that the Indian consumer in particular, and the Indian economy in general, has paid a heavy price for inward-looking protection, a price that adds up to among the worst development records in Asia. The changes since 1991 have unleashed a new dynamic in Indian industry, a dynamic that is forcing change in every sector as firms are finally compelled by new firms and by the availability of imported products to provide consumers with products, prices, and service that begin to approach internationally competent levels. Efficient firms have benefited, even thrived, in the new environment. Inefficient ones have improved, merged, or have finally begun to disappear. Four decades of protection is long enough for any infant to mature. This paper has argued that some Indian firms have built up the technical capabilities that can make a tiger of the Indian economy. It might be a jungle out there, but that is where tigers—not pussycats—are found. It is time Indian industry found out which it is.

References Ahluwalia, Isher, and I. M. D. Little, eds. 1998. India’s economic reforms and development: essays for Manmohan Singh. Delhi: Oxford University Press. Bagchi, A. K. 1984. The Indian Patents Act and its relation to technological development in India. Economic and Political Weekly (Bombay), 18 February. Bhagwati, Jagdish. 1985. Wealth and poverty: Essays in development economics, vol. I. Bombay: Oxford University Press. ———. 1993. India in transition: Freeing the economy. Oxford, England: Oxford University Press. DSIR-Department of Scientific and Industrial Research (1995). A Compilation of Foreign Collaboration Approvals (Government of India Publications, Delhi). DST-Department of Science and Technology (DST). 1999. Research and Development Statistics, 1996–97. (Government of India Publications, Delhi). ———. Research and Development in Industry, 1992–93 and 1998–99 (Government of India Publications, Delhi). Desai, Ashok. 1980. The origin and direction of industrial R&D in India. Research Policy 9. ———. 1988. Technology absorption in Indian industry. New Delhi: Wiley Eastern. Dore, Ronald. 1964. Latin America and Japan compared. In Continuity and change in Latin America, ed. John Johnson, Stanford: Stanford University Press. ———. 1971. Japanese industrialisation and the developing countries: model, warning, or source of healthy doubts? Singapore: Institute of Southeast Asian Studies. Forbes, Naushad, and David Wield. 2000. Innovation in NICs: Managing R&D in technology-followers. Research Policy, December. Hobday, Michael. 1995. Innovation in East Asia: the challenge to Japan. Aldershot, England: Edward Elgar. Kim, Linsu. 1997. Imitation to innovation. Cambridge, Mass.: Harvard Business School Press.

Doing Business in India: What Has Liberalization Changed?

167

Lall, Sanjaya. 1996. Learning from the East Asian tigers: Studies in technology and industrial policy. London: Macmillan. Mahalingam, Sudha. 1989. The computer industry in India: strategies for latecomer entry. Economic and Political Weekly (Bombay), 21 October. NCAER (1999). Industry. Paper presented at conference, ADB-NCAER Project on Economic and Policy Reforms in India. 9–10 December, Delhi. NSF—National Science Foundation (1993). Human resources for science and technology: The Asian region. Washington: NSF. Nayar, Baldev Raj (1983). India’s quest for technological independence: Policy foundation and policy change and India’s quest for technological independence: The results of policy. Delhi: Lancers International. Nelson, Richard, ed. 1993. National innovation systems. Oxford: Oxford University Press. Smith, Sean Eric. 2000. Opening up to the world: India’s pharmaceutical companies prepare for 2005. Stanford University, Institute for International Studies, Asia/Pacific Research Center, Occasional Paper. World Bank. 1999. World development indicators. Washington, D.C.: World Bank.

5 Bangalore The Silicon Valley of Asia? AnnaLee Saxenian

Information technology (IT) has become the mantra of Indian politicians and policy makers. Soon after taking office in 1998, Prime Minister A. B. Vajpayee announced the widely quoted goal of making India a “global information technology superpower” and a “forerunner in the age of the information revolution.” India’s software industry has grown so rapidly that it evokes frequent comparisons between Bangalore, one of India’s leading software producing regions, and Silicon Valley. Moreover, the achievements of Indian professionals in leading-edge technology industries abroad have contributed to a growing sense of confidence in India, confidence that did not exist before—largely because India and Indians have participated in the information technology revolution. The performance of India’s IT industry during the 1990s has been impressive, particularly in contrast to other sectors of the Indian economy. The sector’s compound annual growth rate (CAGR) for 1994–99 exceeded 40 percent, compared to only 7 percent for the economy as a whole (NASSCOM 2000; Ministry of Finance 2001). This strong growth was led by the software industry, which in 1999 accounted for 65 percent of India’s total IT revenues and employed more than 200,000 workers. Total software revenues of $3.9 billion in 1999 were close to four times those of IT hardware manufacturing and grew more than 55 percent per year in the late 1990s. MoreAnnaLee Saxenian is professor of regional development at the University of California at Berkeley and a visiting senior fellow at Stanford Institute of Economic Policy Research. The author is especially indebted to Balaji Parthasarathy, now of the Indian Institute of Information Technology-Bangalore, and to the participants of the workshop on Equity, Diversity, and Information Technology held at the National Institute of Advanced Study, Bangalore in December 1999. Their research and wisdom are reflected throughout this document. Any errors are the author’s alone. The author devotes special thanks as well to Sajjid Chinoy and Suraj Jacob, who provided outstanding research assistance on very short notice.

169

170

AnnaLee Saxenian

Table 5.1

The Information Technology Industry in India, 1993–99 (US$ millions) 1993–94

Software Domestic Exports Total Hardware Domestic Exports Total Grand total

1994–95

1995–96

1996–97

1997–98

1998–99

230 330 560

350 485 835

490 734 1,224

670 1,083 1,753

950 1,750 2,700

1,250 2,650 3,900

490 93 583 1,143

590 177 767 1,602

1,037 35 1,027 2,296

1,050 286 1,336 3,089

1,205 201 1,406 4,106

1,026 4 1,030 4,930

Source: NASSCOM (1999, 2000).

over, the software industry’s growth was driven primarily by exports. While the domestic market for software has grown in absolute terms, software exports account for a large and increasing share of total industry revenue (table 5.1). This export success is particularly striking for an industry that remained peripheral to world markets throughout most of the 1980s. This chapter examines the growth and performance of India’s IT industries, with particular attention to the role of policy in this process. It first reviews the evolution of the software industry, highlighting the policy departures that have contributed to its rapid emergence and growth over the past two decades. It then turns to the formation of the National Information Technology and Software Development Task Force in 1998 and its policy recommendations that aimed at making India the number one provider of IT products in the world. The concluding section steps back to address the role of the IT industry in India more broadly. It suggests that comparisons between regions like Bangalore and Silicon Valley not only mislead, but also distract attention from the deeper challenges and opportunities that IT offers for the Indian economy. In particular, the range of actors and the scope of policy debates need to be expanded significantly to realize the full potential of the IT revolution in India. It is important to begin by putting the performance of Indian IT in a global perspective. India’s $4 billion in software revenues in 1998–99 represented a very small fraction of an estimated world software market of some $300–500 billion (Arora et al. 2000). In spite of the sector’s rapid export expansion between 1985 and 1995, from $28 million to $481 million, India’s share of total world IT exports remained stable at 0.5 percent (Organization for Economic Cooperation and Development [OECD] 1997, 50). Moreover, India’s 0.5 percent share of world IT exports in 1995 was less than the country’s 0.6 percent share of world aggregate exports in the same year (Ministry of Finance 2001). These figures suggest the need for skepticism about the more inflated claims currently circulating concerning India’s IT industries. The weakest link in the IT sector, hardware development and manufac-

Bangalore: The Silicon Valley of Asia?

171

turing, remains small and barely viable. The industry suffered tremendously from the protection provided by the high import duties as well as from very limited access to foreign technology after IBM left the country in 1978. Personal computers, components, and other IT products manufactured in India in the 1980s were both costly and technically backwards, which limited demand and hence the volume of production. The inability to gain scale economies proved fatal in a highly capital-intensive industry and, over time, contributed to the industry’s decline. By the 1990s, most of India’s IT hardware companies were transformed into direct or indirect dealerships for foreign brand computers and related products (Jhungjhunwala 1999). India’s total hardware revenues amounted to $1.03 billion in 1998–99 (table 5.1). The sector has not grown in the past five years, and its exports are negligible. The reduction of import duties to zero by 2002, as recommended by the World Trade Organization (WTO) will put severe pressure on India’s IT manufacturers, and it seems likely that only those that are able to develop higher value–added products are likely to survive. The remainder of this paper focuses on the software industry, as it is the sector in which an innovative policy regime has stimulated a different developmental dynamic. 5.1 The Evolution of the Indian Software Industry, 1984–98 Prior to 1984, the Indian software industry operated within the framework of a highly regulated, autarkic model of import-substitution–led industrialization (ISI) and the ideology of self-reliance that guided the Indian economy. This policy stifled entrepreneurs and isolated India from the global economy. As a result, efforts to promote software exports during the period never took off. Policies that permitted the import of state-of-the-art computers in exchange for a guarantee to export a certain amount of software were not enthusiastically received (Subramanian 1992). Import procedures were cumbersome, duties were high, and obtaining foreign exchange for business expenses was difficult.1 5.1.1 Policy Reform in Software The election of Rajiv Gandhi as prime minister marked the turning point for policy reform in India’s software and computer industries. Gandhi’s administration was the first to emphasize new policies for electronics, software, telecommunications, and other emerging industries.2 A computer 1. IBM’s 1978 departure from India, after its refusal to comply with the requirements of the Foreign Exchange and Regulation Act, is indicative. For details, see Grieco (1984). 2. Rajiv Gandhi’s administration also initiated the computerization of the railway reservation system and several government processes. One of his most innovative contributions was the creation of the Center for the Development of Telematics, which pioneered indigenous digital switching technology to facilitate India’s shift from electromechanical to digital switching and transmission.

172

AnnaLee Saxenian

policy announced in November 1984 recognized software as an “industry,” making it eligible for an investment allowance and other incentives. The policy also lowered import duties on software and personal computers (PCs) and permitted the import of computers in exchange for software exports at a special low duty. The passage two years later of the 1986 Computer Software Export, Development, and Training Policy marked an explicit rejection of Indian ISI and the idea of self-reliance in software. The policy was designed to promote the domestic software industry and facilitate a “quantum jump” in software exports by providing Indian firms with liberal access to the latest technologies and software tools to enhance their global competitiveness and to encourage higher value-added exports. To that end, the import of software in any form was permitted and various procedures simplified. The policy also invited foreign investment and promised to make venture capital available to encourage new firm formation and export growth. The 1984 and 1986 policies were championed by N. Seshagiri, additional secretary at the Department of Electronics (DoE), who had long argued that India’s policies were too restrictive, its procedures too cumbersome, and its idea of self-reliance self-defeating (Sridharan 1996). He also argued that for India to become a major software exporter, it would have to begin with high volume, low value-added exports and move up the value chain. He believed that India’s failure to follow such a strategy had left it far behind the East Asia NICs in hardware exports. Thus, the 1984 policy explicitly recognized bodyshopping—the provision of labor-intensive, low valueadded programming services, such as coding and testing, at client sites overseas—as valid exports. In spite of some ambivalence within the government about promoting bodyshopping, which a few technologically conversant policy makers regarded as “intellectual coolieism,”3 Seshagiri was able to push through his policy, but only because of misconceptions that prevailed about software among most policy makers: [I]f the administrators and some of the bureaucrats had too deep knowledge, they might have prevented bodyshopping or on site services. Software was seen as a glamorous high tech industry. So they said, alright, do it. (quoted in Parthasarathy 2000a) If limited understanding of the software industry allowed Indian firms to begin bodyshopping, it also prevented policymakers from taking decisive steps to actively promote the software industry. The 1984 and 1986 policies merely removed barriers to its growth. Sen writes that “until 1991–92, there was virtually no policy support at all for the software sector. Even the term ‘benign neglect’ would be too positive a phrase to use in this connection” (Sen 1994, 55). 3. Interview with N. Vittal, former secretary of the department of electronics, New Delhi, 25 June 1996; cited in Parthasarathy (2000a).

Bangalore: The Silicon Valley of Asia?

173

The greatest challenge for Indian companies in the 1980s was the lack of the international telecommunication links that are the necessary infrastructure for software exports. While the export of data via satellite links was permitted, establishing an earth station was a protracted procedure requiring permission from multiple government departments. When Texas Instruments set up the first earth station in Bangalore in 1986, for example, the process involved removing or breaking twenty-five different government rules (Parthasarathy 2000a). Without reliable telecommunications links, Indian firms had no alternative to providing contract programming on-site (at the customer’s facilities), typically in the United States.4 The Software Technology Parks (STP) scheme introduced by the DoE in the early 1990s insured that the infrastructure and administrative support for exporting were available in India. An STP is like an export processing zone for software: It gives export-oriented software firms in designated zones tax exemptions for five years and guaranteed access to high-speed satellite links and reliable electricity. The DoE also provides basic infrastructure, including core computer facilities, reliable power, ready-to-use office space, and communications facilities including internet access and sixty-four kilobits per second data-lines. As in the predecessor programs for export-oriented units, firms in the STP are allowed to import all equipment without duty or import licenses, and 100 percent foreign ownership is permitted in exchange for a sizable export obligation.5 STP firms are also allowed to freely repatriate capital investment, royalties, and dividends after paying the necessary taxes. Administratively, the STPs provide a decentralized, single window clearance mechanism for applications from potential investors. While STPs can be established by anybody, anywhere in the country, the DoE announced the first three in 1990 in Bangalore, Pune, and Bhubaneshwar, and another four the following year. In June 1991, the Software Technology Parks of India (STPI) was registered as an autonomous agency, reflecting the desire of the DoE to avoid direct government involvement in the industry. The local directors of individual STPs have wide-ranging powers and are intended to serve as “friend, philosopher, and guide” to the industry while also functioning as the eyes and ears of the DoE.6 Inclusion of industry representatives on the boards and councils of the STPs was also meant to emphasize the industry-friendly approach of the scheme. By 1998 there were twenty-five STPs under various 4. The terminology used in the Indian software industry can be confusing: Onsite services are those in which programmers work at the customer’s facilities, while offshore services are performed in a remote location, in this case in India. 5. Firms have to earn a net amount equal to 150 percent of the hardware imported within four years. They also have to earn a net amount equal to 150 percent of their wage bill on an annual basis. Though the STP scheme was meant for 100 percent export units, in January 1995 STP firms were allowed to sell 25 percent of their output to the domestic tariff area. The figure was revised to 50 percent in 1999 (Parthasarathy 2000b). 6. Interview with S. K. Agarwal, director of STPI, New Delhi, 20 June 1996. Cited in Parthasaraty (2000a).

174

AnnaLee Saxenian

stages of planning and development in different parts of the country (in addition to those sponsored by the DoE). The Information Technology Park, Ltd. in Bangalore, for example, is a partnership between the Karnataka government, Tata Industries, and a consortium of Singapore firms. The introduction of the STPs coincided with the initiation in 1991 of the economic liberalization process in India.7 Software producers benefited from general policy changes such as the devaluation of the rupee and the growing openness to foreign direct investment. They also benefited from the exemption from income tax of profits on software and other service exports and, most importantly, the 1992 removal of import licensing on equipment and industrial imports. This allowed Indian companies to import the computers that its clients used and to produce or modify software for them directly. To summarize, the policy reforms of the 1980s facilitated the emergence of an export-oriented software industry in India. However, export growth in this period was based exclusively on bodyshopping on site (with Indian programmers working at the client site, typically in the United States). The shift to offshore production, allowing the programmers to work at facilities in India, was only possible following the reforms of the early 1990s, particularly the removal of licenses on imports of industrial equipment and the establishment of the STPs. Even after the pace of liberalization slowed in the rest of the economy in the mid-1990s, the software industry continued to benefit from a series of sector-specific policy reforms. This was largely due to aggressive lobbying by the industry association, the National Association of Software and Service Companies (NASSCOM).8 In 1997, for example, all import duties on software were eliminated, and software firms were allowed to invest in foreign joint ventures and wholly owned subsidiaries to a limited extent.9 And in 1998, software firms were permitted to offer ADR/GDR-linked stock options to employees.10 The active role of the industry association, NASSCOM, in shaping policy distinguishes the software industry from the computer hardware and other older Indian industries. NASSCOM has been influential in shaping 7. For further details on policy changes in 1991 and since, see Krueger and Chinoy (this volume). 8. NASSCOM was founded in 1988 with thirty-eight members. By 1999 it had 464 members and accounted for 95 percent of software industry revenues. 9. Firms were allowed to invest up to 50 percent of their foreign exchange earnings in the previous three years, subject to a maximum of $25 million. 10. ADR/GDR: American/global depository receipts. The ADR is a certificate issued by a U.S. bank that trades like a share on NASDAQ, allowing the U.S. investor to invest in a foreign market without having to deal with the risk of currency transactions. ADRs represent a certain number of domestic shares of the firm deposited with the bank. GDRs are similar to ADRs, except that they are traded on international stock exchanges such as London’s. The Reserve Bank of India permits Indian employees to remit up to $50,000 in a block of five years to ADRs/GDRs.

Bangalore: The Silicon Valley of Asia?

175

the DoE strategy of working with software companies to provide critical infrastructure, while explicitly avoiding more detailed regulation or intervention. This is evident, for example, in the decision to organize the STPI program as an autonomous unit (and eventually to privatize it). The DoE thus represents a very different model for India from an older generation of “strategic” ministries that sought to specify, develop, and directly regulate technology and industry structure.11 NASSCOM’s leaders interact continually with politicians and policy makers, and the association is represented on many influential committees of the government of India. It also sponsors high-profile conferences and studies, consults for state governments, and promotes the Indian software industry around the world through a very effective web site as well as through attendance at international trade shows and foreign visits (see [http://www.nasscom.org/]). NASSCOM is also the sole source of IT industry data in India. Its annual strategic review provides the only detailed and up-to-date figures on employment, revenues, exports, and market share for the software and other IT industries. This provides leverage for the association, but is not an optimal situation for policy makers or scholars.12 5.1.2 Software Industry Growth and Transformation, 1984–98 The post-1984 policy changes were crucial to the growth of the Indian software industry because they allowed domestic producers to exploit domestic resources in global markets. India’s greatest asset is a large, educated, English-speaking workforce that is willing to work at relatively low wages. In spite of widespread illiteracy, India boasts thousands of educated engineers who have remained either underemployed or unemployed for decades. Few countries can match India’s combination of low-wage, highly skilled workers. In 1994, wages for software programmers and systems analysts in India were less than one-tenth of those for their U.S. counterparts, and lower even than those in other developing countries, such as Mexico (table 5.2). Indian programmers also had the unanticipated advantage of familiarity with the Unix operating system in the 1990s. The failure to develop a commercially viable computer following IBM’s departure from the country meant that Indian users relied on imports of a wide range of models and vintages from different manufacturers. Indian programmers thus learned to work on a variety of platforms, which proved helpful in acquiring contracts for the maintenance of various legacy systems. More important, computer 11. See Parthasarathy (2000b) for a detailed analysis of the telecommunications case. 12. NASSCOM’s data includes only the numbers provided by its members and thus overlooks large numbers of smaller software and IT companies that are not part of the association. NASSCOM’s goal of promoting software industry may tend to bias the data as well. Future policy reform would ideally include creation of a reliable, independent source of detailed industry data.

176

AnnaLee Saxenian

Table 5.2

Country India United States Japan Germany France Britain Hong Kong Mexico

International Wage Rates, Software Industry, 1994 Programmer (US$)

Programmer Index

Systems Analyst (US$)

Systems Analyst Index

4,002 46,600 51,731 54,075 45,431 31,247 34,615 26,078

100 1,164 1,293 1,351 1,135 781 865 652

5,444 61,200 64,519 65,107 71,163 51,488 63,462 35,851

100 1,124 1,185 1,196 1,307 1,287 1,166 658

Source: Business India (1995, 199), as cited in Parthasarathy (2000a).

manufacturers in the 1980s had no alternative but to rely on Unix (the first portable, machine-independent, multi-user operating system) even though foreign companies were developing proprietary systems at the time. By the 1990s, however, when Unix became the system of choice for PCs and workstations, India’s Unix programmers had a skill that was extremely scarce elsewhere in the world. Indian producers entered the world market in the 1980s by exploiting their cost advantage in the most routine, low-value-added segments of software production, such as coding, testing, and maintenance. The vast majority of these exports derived from bodyshopping, which has been referred to as an “input-less” export because it requires only an overseas contract, a minimal amount of finance, and names of local programmers (Heeks 1996). In these contracts, the amount of software code is specified in advance, and revenue is earned per line of code. Indian engineers who work overseas are paid their salaries in rupees and provided with minimal allowances for housing and expenses. Some refer to this business as “resumé selling” because it is, in essence, a lucrative form of labor cost arbitrage—and India boasts ample resumés. Indian educational institutions and polytechnics train more than 67,000 computer science professionals annually. Another 200,000 individuals enroll annually in the private software training institutes that have mushroomed in India in the 1990s (NASSCOM 1999). The policy changes of the 1980s are typically credited with stimulating the accelerated growth of Indian software exports. However, it is worth noting that this growth, at least initially, was more impressive in rupee terms than in dollar terms. Using quarterly data, Sen (1994) shows that between 1987 and 1993 a significant portion of export growth was accounted for by the falling value of the rupee (table 5.3.) The devaluation meant that the growth due to a lower exchange rate was almost as great as the real growth rate in dollars. The introduction of the STPs facilitated a gradual shift away from on-site

Bangalore: The Silicon Valley of Asia? Table 5.3 Period 1987–93 1987–90 1990–93

177

Decomposing the Annual Growth of Indian Software Exports, 1987–93 Total Growth (%)

Real Growth (%)

Exchange Rate (%)

46 41 52

28 29 28

18 12 24

Source: Sen, as cited in Parthasarathy (2000b).

to offshore (in India) service provision during the 1990s.13 While on-site production accounted for 90 percent of Indian software exports in 1990, the share had fallen to 58 percent by 1998 (Parthasarathy 2000a, 28). One advantage of offshore production soon became apparent. The twelve and onehalf-hour difference between Indian standard time and Pacific standard time allowed Indian firms to perform maintenance and reengineering tasks for U.S. customers by accessing their computers after regular users had finished for the day. Combined with growing shortages of skilled labor in the West, this helps to explain why hundreds of U.S. and European corporations increasingly outsourced routine, labor-intensive projects, such as coding, maintenance, and Y2K solutions, to Indian software houses in the 1990s. A growing number of foreign companies followed the earlier model of Texas Instruments and Hewlett-Packard and located offshore development centers (ODCs) in India in the 1990s. They were motivated by the labor cost difference, to be sure, but the availability of high quality skill was essential to these decisions as well. According to one U.S. employer in Bangalore, the low wages matter because they provide an attractive trade-off to working in an environment plagued by chronic infrastructure problems: [W]e came here because of the skills. We expanded because of the skills. We were able to come to India because the risk of being 10,000 miles away, the risk of the satellite link and the telephones and the flights were offset by the costs.14 There is anecdotal evidence that in the late 1990s the ODCs began to take on more sophisticated design and programming projects, either jointly or independently, and often as equal partners with their parent organizations.15 This underscores the potential for upgrading in India. The chief executive of one ODC explains why his company waited until the 1990s to move to Bangalore: 13. By 1995, 435 companies were registered under the STP scheme, accounting for more than 16 percent of exports (Software Technology Parks of India 1995.) 14. Interview with industry representative, Bangalore, 30 July 1996 as cited by Parthasarathy (2000a). 15. The engineers at the Texas Instruments development center, for example, developed a new digital signal processing chip that has become a standard.

178

AnnaLee Saxenian

Now . . . the feeling is that high-tech, leading edge quality, timely, delivered, supported software will come out. That was not a risk you could have taken five years ago. . . . [I]f they had asked me then, I would have said no. I would have said, “do anything, bodyshopping, subcontracting, modular work, but don’t give full dependability here [in India] because nobody’s ever done it. It’s not proven.” (Parthasarathy 2000a) He compared the work done at his center with that at the headquarters in Silicon Valley: New products come from here, new versions of old products come from here. . . . [P]roducts on a particular hardware platform come from here as opposed to an application for a customer. . . . It’s the same thing that they do over there. Technologically, there’s zero difference. (Parthasarathy 2000a) The growth of offshore facilities also allowed some established Indian companies to begin building a base of in-house knowledge and to develop internal training programs, quality processes, and productivity tools. This has facilitated the upgrading of their capabilities. In December 1999, 137 Indian companies had obtained either ISO 9000 or SEI-CMM level two certification, and ten companies were certified at level five (the highest level, at which only six U.S. companies are certified). Quality certification serves as an important marketing device for Indian companies while also improving their ability to manage time and resources involved in large projects (Arora 1999). Today, some of the largest Indian software houses, such as Wipro and Infosys, have track records that allow them to win bigger consulting contracts, often on a turnkey basis.16 This allows them to take on a greater range of software development processes and managerial tasks (such as overall project scheduling, quality, and productivity) than are required in bodyshopping and to begin charging higher rates for their work. Rather than compete on the basis of hourly productivity, they aim to accumulate intellectual property by converting the knowledge gained during a series of consulting projects into broadly applicable software components that can in turn be customized for clients with similar needs. Industry observers have suggested that India’s share of the world market for customized, as opposed to packaged, software is significantly higher than the aggregate data suggest (Arora et al. 2000). In spite of the evidence that a handful of Indian software companies are gradually moving up the value chain and gaining international recognition for their quality and performance, the industry as a whole remains significantly less productive than its global competitors. While the Indian software industry employed some 180,000 workers in 1998, the annual revenue 16. Five companies account for close to 50 percent of software industry exports: Tata Consultancy Services, Infosys, Pentafour, Tata Infotech, and Wipro.

Bangalore: The Silicon Valley of Asia?

179

per employee in India was $15,000–20,000. This compared to $100,000 per employee in other software-producing countries such as Israel and Ireland (Arora et al. 2000). Moreover, it appears that the software boom has exacerbated the “brain drain.” Programmers in India are increasingly aware of the global demand for their skills and the substantially higher compensation available in more developed economies. Many thus aspire to work for foreign companies not only for the relatively high wages but also for the opportunity to be transferred overseas. The United States has been a major beneficiary of this trend. A recent study of the H-1B visa, which grants temporary work authorization to highly skilled foreign persons, reports that the number of Indian H-1Bs grew steadily from 1989, clearly becoming the largest category in 1994, doubling in size by 1996, and quintupling by 1999. Indians accounted for nearly half of all visas issued in 1999 (forty-seven percent). This amounted to 55,047 Indian workers in 1999 alone, and a total of 195,083 between 1989 and 1999. The next largest groups of H-1B visa holders were from the United Kingdom and China, but each accounted for 6 percent or less of the total visas granted (Lowell 2000). There is growing recognition among Indian policy makers and software producers of the need to accelerate the industry’s shift into higher valueadded activities for two different reasons. On one hand, the developmental potential of the current trajectory is quite limited. The provision of routine software services for export may be highly profitable for individual companies, but it provides few opportunities for longer-term technological learning and upgrading (Arora 1999). Meanwhile, India’s labor cost advantage is eroding, in spite of its sizable labor pool. The software industry association estimates that wages in the software industry rose 21 percent per year in the late 1990s, albeit from a low base (NASSCOM 1999). Some analysts report that shortages of IT professionals are constraining the industry’s growth (“India Hit by Shortage of IT Professionals,” siliconindia.com 21 May 2000). As a result, India’s producers face increasing competition from other low-wage, humancapital–rich countries like the Philippines and China. In the words of Desai (1999a), chairman and managing director of Mastek Ltd., a leading Indian software company: Indian [software] corporations are at a crossroads, faced with growing globalization and competition. . . . [I]t is becoming difficult for them to compete only on one differential—the cost advantage; and this forces them to move to higher value addition in their offerings. 5.2 The IT Action Plan While the central government initiated India’s economic liberalization in the early 1990s, the state governments have pioneered some of the most

180

AnnaLee Saxenian

far-reaching policy innovations in the IT sector. The chief minister of Andhra Pradesh, Chandrababu Naidu, has drawn attention both in India and around the world for his entrepreneurial, high-profile attempts to attract technology investment to the state and to promote the use of IT in his administration. Naidu has effectively promoted the concept of “e-governance” (the use of IT in delivering public services) as a way to insure greater accountability, transparency, and efficiency in the government of Andhra Pradesh. Many of the state’s departments—such as treasury, employment, commercial taxes, rural development, registration, irrigation, excise, and police—are being computerized in order to both reduce corruption and improve service delivery.17 Naidu’s efforts have triggered escalating competition from neighboring states. The recently elected government of Karnataka, for example, has laid out an ambitious plan for upgrading its overburdened roads and other infrastructure. The governments of Tamil Nadu and Kerala are also developing IT policies that include investing in infrastructure, computerization of government offices, single window clearance for IT ventures, and IT-related education. Today Andhra Pradesh and the other southern states where the software industry is concentrated are well ahead of the government of India in their implementation of e-governance and other IT reforms.18 Naidu has thus initiated a bottom-up process of policy reform in this historically centralized polity. He has been a vocal proponent of national policy reform as well. His recent book, Plainly Speaking, lays out his views on many issues related to governance and information technology in India. He calls, for example, for greater devolution of central tax revenues to state governments and greater flexibility in fiscal management. He also argues that there is an urgent need for administrative reform and for the removal of discretionary powers in a country where “bloated governments have bled their exchequers dry.” And he claims that the IT-related initiatives undertaken by his government have helped to redefine the meaning and content of governance within the country. Naidu was also instrumental in raising the issue of IT at the national level.19 The prime minister’s office responded in 1998 with the formation of the National Task Force on Information Technology and Software Development. The task force was a high-powered group that included senior representatives from the private sector, government, and universities. It included 17. One project will facilitate integrated delivery of eighteen services (such as payments for water, electricity, property taxes, etc.) from six different departments. Another, the MultiPurpose Household Study, will develop records of all individuals in the state and provide uniform data for all of the departments. See [http://www.andhrapradesh.com]. 18. The IT industry is concentrated in Andhra Pradesh, Karnataka, and Tamil Nadu, largely because of their concentration of engineering manpower. Sixty percent of India’s computer science graduates come from these three states (Dossani 2000). 19. This summary is drawn from a review of Naidu’s book on the official website of Andhra Pradesh [http://www.andhrapradesh.com].

Bangalore: The Silicon Valley of Asia?

181

Naidu and Sheshagiri (currently director of the National Informatics Centre) as well as the executive director of NASSCOM, senior executives from Infosys and Wipro, and a range of other scientists, professionals, educators, and military officials. It also seconded the secretaries of the departments including electronics, finance, commerce, and telecommunications. The task force moved extremely quickly—far more so than is the norm in India—and released its “Information Technology Action Plan” a year after the group was convened. The task force also developed an unusually open and transparent process for collecting information and formulating recommendations, a process that involved consultation with an unusually wide variety of public- and private-sector actors.20 All of the task force documents are available on the internet, and, in the words of one of the background reports: This is the first time in India that representatives of so many ministries, departments, industry associations, business houses, educational institutions and State Governments have interacted so intensively and in such a short period of time to cover so many bottleneck and promotional areas. . . . (National Taskforce on Information Technology and Software Development 2000) As a result, the IT Action Plan is the most ambitious IT-related policy proposal in India since the Computer Policy of 1984 and the Software Policy of 1986. The plan lists 108 recommendations of “revisions and additions to the existing policy and procedures for removing bottlenecks and achieving a pre-eminent status for India.” Additionally, it sets $50 billion in software exports and “IT penetration for all” as targets for 2008. The report is wide-ranging in coverage and sober in its assessments of the current constraints on IT development, in spite of often hyperbolic (if laudable) goals. It reflects a clear understanding of the needs of the industry and of the limitations of the Indian business environment—an understanding that could only have grown through consultation with the private sector and other industry specialists. This collaborative process in itself reflects an important step forward in policy making in India. However, the wide-ranging nature of the report raises concerns about the implementation process and what, if any, processes are in place to insure that the more politically difficult or longer-term reforms are carried through.21 The action plan addresses the two concerns that are most frequently articulated by software and other IT producers in India: (a) the inadequacy of 20. The task force set up four working groups, on IT research, design, and development; IT human resources development; citizen-IT interface; and content creation and content industry. These each had twelve to sixteen members and drew in a still wider range of perspectives. The task force secretariat also reportedly received some 8,000 e-mail messages providing policy suggestions. See [http://it-taskforce.pic.in/]. 21. Sheshagri claims that 80 percent of the recommendations have been implemented already, but there is no way to confirm this figure (Prabhakar 2000b).

182

AnnaLee Saxenian

the infrastructure: telecommunications in particular, but also roads, airports, and power supply; and (b) the cumbersome bureaucratic hurdles and regulatory red tape involved in doing business (see, for example, Desai 1999b and Saxenian 1999). As we have seen, India’s telecommunications, roads, and air transport infrastructures rank extremely poorly, in the bottom 10 percent of ranked countries, on a global scale (Krueger and Chinoy, this volume). Note, for example, telecommunications: In 1997 there were only 18.6 telephone main lines per 1000 people in India, compared to 55.7 in China, and the wait for a new connection was 12.17 months, compared to China’s 0.68 months (International Telecommunications Union 1998). The state of the infrastructure imposes significant direct and indirect costs on producers and undoubtedly constitutes a barrier to foreign investment.22 The processes of starting and running an IT business in India have been simplified and streamlined since 1984. However, the complex rules and lengthy procedures for transacting business remain a source of tremendous cost and frustration. The costs are especially severe for companies in globally competitive industries like software, where success depends critically on speed, or “time to market.” In the recent words of Infosys chairman N. R. Narayana Murthy: “If you want to be the first mover in India, please expect a lot of delays and trying times. . . . [Al]though in absolute terms the country may have made substantial progress, on a relative scale we have slowed down.”23 The infrastructure section of the IT Action Plan calls for the liberalization of the telecommunications market, particularly in the area of data communications, and expanded access to the internet. It recognizes the bottleneck created by the power of the department of telecommunications, Mahanagar Telephone Nigam Ltd. and Videsh Sanchar Nigam Ltd. (VSNL) in this sector. The report’s recommends: (a) elimination of the license fee for internet service providers, (b) termination of the VSNL monopoly as international gateway for the internet, (c) removal of the DoT monopoly on the long-distance backbone to allow railways, state electricity boards, and others to host fiber-optic backbones, (d) provision of free permits for last mile access, and (e) opening of a specified radio frequency band for public wireless usage. The action plan also provides thirty-nine recommendations calling for systematic rationalization of Indian duty structure and of the companies act. It proposes the exemption of public and private infrastructure providers from all import duties. It proposes phasing in the zero-duty regime earlier than was agreed to at the WTO Information Technology Agreement 22. It is worth repeating the McKinsey-NASSCOM estimate that as much as $23 billion in IT export revenues and 650,000 jobs fail to materialize over an eight-year period because of limitations of the telecommunications infrastructure (NASSCOM 1999). 23. Cited in C. Chitti Pantulu, “Entrepreneurs Should Be Prepared for Delays, Says Narayana Murthy,” The Financial Express, 6 February 2000.

Bangalore: The Silicon Valley of Asia?

183

in 1996. This section also recommends an overhaul of financial regulations to enable the accelerated expansion of IT. It designates IT as a priority sector in order to insure a greater flow of funds from banks into the industry, it calls on banks to establish venture capital (VC) funds, and it recommends the removal of regulatory constraints limiting the availability of venture capital. It also proposes expansion of the software industry definition to include IT-enabled service exports such as data entry, call centers, and other back-office operations, to insure that these businesses benefit from the tax exemptions currently granted to software exports. The section on “IT for all by 2008” calls for development of e-commerce or cyber law; a campaign for universal computer literacy; schemes to insure provision of computers and the internet in all schools, colleges, and public hospitals by 2003; and a variety of IT programs in universities. This section also calls for IT in rural India, the use of Indian languages for computers, and the development of indigenous technologies. The final section recommends bringing IT into government by allocating 1–3 percent of the budget of each ministry and department for IT applications. This IT Action Plan has provided an impetus for change as well as an ambitious roadmap for India in the IT sphere. Prime Minister Vajpayee signaled his political support for the plan in late 1999 by creating a new ministry of information technology to oversee its implementation. The newly appointed IT minister has in turn promised that all of the recommendations will be implemented by 2001. Many, such as the development of cyber law and regulations concerning overseas investment in venture capital, have been acted on already.24 Other sections of the plan will be significantly more difficult to implement for political or institutional reasons (e.g., because of resistance from bureaucrats who fear the loss of control) or because they are far too ambitious—at least in the short run. A telling example of the challenges involved in achieving regulatory reform in India today is the recent efforts to facilitate the growth of the VC industry. The IT Action Plan recommends the promotion of VC, and most industry representatives and analysts agree that a dynamic venture capital industry will be critical to the long-term development of Indian IT. They argue that a healthy VC industry will stimulate new entrepreneurial entry, broaden the range of activities in the field, and accelerate the country’s move into higher value-added activities. However, the supply of venture capital in India remains very small by international standards, largely because the industry is governed by a multiplicity of conflicting and often cumbersome regulations and discriminated against in a variety of ways. The industry is currently regulated by three different regulatory bodies: the Securities and Exchange Board of India 24. Task force convenor Sheshagri claims that roughly 80 percent of the recommendations have been implemented, but it is very difficult to assess this claim.

184

AnnaLee Saxenian

(SEBI), the Ministry of Finance, and the Central Board of Direct Taxes (CBDT). In addition, foreign VC firms are also governed by the Foreign Investment Promotion Board (FIPB) and the Reserve Bank of India (RBI). As a result, there are now three different, and inconsistent, sets of regulations governing the industry. For example, each prescribes different investment criteria for VC funds. These statutes in turn compete with existing corporation, tax, and currency laws—many of which are extremely anachronistic (including some that predate India’s independence).25 This helps account for the modest size of the VC industry in a financial system that boasts substantial domestic and foreign investment. In 1998 there were only twenty-one companies registered with the Indian Venture Capital Industry Association, with approximately $700 million available for investment. This compares to Israel’s 100 firms with $4 billion investible funds (in 1999) and Taiwan’s 110 funds with $1.32 billion investments. Moreover, most of India’s VC firms are funded either by the public sector or by multilateral funding agencies (Dossani and Saez 2000). These firms typically lack the expertise or contacts in the IT field, or the willingness to take risks, that would be essential to the sort of value-added financing that is associated with VC in places like Silicon Valley. In an attempt to address these limitations, in 1999 SEBI convened a committee on venture capital, led by a successful nonresident Indian entrepreneur from Silicon Valley, K. B. Chandrasekhar, with the task of recommending steps to promote VC in India. The committee’s report develops a comprehensive vision for the growth of India’s VC industry, based on a survey of the global experience, and it proposes a series of regulatory and institutional reforms to achieve this goal (SEBI 2000). The SEBI board adopted the report in January 2000, signaling the seriousness of the government’s intentions to pursue its recommendations. However, many of the committees’ proposals require changes that go beyond SEBI’s jurisdiction, so that the final outcome will depend on the report’s acceptance by other parts of the government, particularly the CBDT, the FIPB, and the RBI. Thus the pace of change remains difficult to predict, in spite of the ongoing efforts at reform by the Ministry of Finance (which, for example, recently proposed exempting venture capital funds from direct taxation) and SEBI, as well as the support of the IT ministry for these reforms. 5.3 IT in Indian Development: The Need for a Larger Vision Comparisons between Indian regions like Bangalore, where future IT growth depends either on continued supplies of low-cost skill or on shifting into higher value-added activities, and the world center of technological in25. The discussion of India’s venture capital industry draws from Dossani and Saez (2000).

Bangalore: The Silicon Valley of Asia?

185

novation, Silicon Valley, remain premature at best. This is not to discount either India’s achievements or its potential. India’s large skill base is an important competitive asset in the knowledge-based economy. And the successes of Indian engineers in the United States demonstrate their technical and entrepreneurial capabilities when working in a supportive environment. However, comparisons with Silicon Valley are misleading because they imply that India could, or should, seek to replicate the U.S. model in information technology. There are compelling reasons that India will need to define its own pathway in the IT era. Silicon Valley emerged in the postwar U.S. economy with the advantage of a large domestic market, a widely educated population, and well-functioning infrastructure and regulatory institutions. The same factors have facilitated the swift diffusion of information technology into the U.S. economy and society—and supported a virtuous cycle of technological innovation to meet the needs of local producers and consumers. In India, by contrast, a vast rural, as well as urban, population lives in poverty, lacking even minimal levels of education (Kochar, this volume.) The nation’s transportation and communications infrastructures remain woefully inadequate (Krueger and Chinoy, this volume). Furthermore, substantial bureaucratic and regulatory constraints continue to hinder the modernization of the private sector. In fact, a recent survey ranked India’s bureaucracy as the worst in Asia in terms of efficiency and integrity (Mukherji 2000). The current approach to IT policy in India addresses the immediate obstacles to growth identified by a small number of established, exportoriented software producers. This has proven successful: IT policy reforms, business confidence, and investment have become mutually reinforcing. This is reflected in the escalating valuations of technology companies on the Indian stock exchanges over the past year.26 It is also reflected in the growing number of multinationals locating overseas development centers in the established IT regions. And this process of policy reform continues. The IT bill passed in mid-2000, for example, sets up a framework for electronic commerce in India. The growing political influence of the software industry means that much-needed regulatory reform, particularly in the telecommunications sector, is being initiated by the government of India. Moreover, competition between state governments for IT investments should insure improvements in transportation and communications infrastructure, at least in select urban areas. However, there is need for a substantially broader perspective on policy than that determined by the immediate needs of the software industry. The 26. Software and related IT services companies now comprise 20–25 percent of India’s total stock market capitalization.

186

AnnaLee Saxenian

current policy approach risks accelerating the growth of IT as a small, modern enclave in a poor and backward economy. The export earnings from IT are important to India’s GDP growth and foreign exchange reserves, but they could be detrimental to the rest of the economy. Several observers have cautioned against the dangers of the “Dutch Disease,” in which dollars earned from a narrow sector like IT (which remains under 1 percent of gross domestic product) sustain an increasingly strong Indian rupee and hurt the competitiveness of other less productive sectors of the economy (Mukherjee 2000). The concentration of the software industry in a small number of cities in the south has the potential to exacerbate the already disparate rates of growth across states and regions in India (Ahluwalia, this volume). Software employment and investments are overwhelmingly concentrated in urban areas in the southern states of Karnataka (Bangalore), Andhra Pradesh (Hyderabad), and Tamil Nadu (Chennai), along with the western state of Maharashtra (Mumbai) and areas surrounding Delhi. The evidence from the United States suggests that such spatial agglomerations of IT production resist decentralization due to powerful supply-side externalities in the provision of skill, inputs, and technology. Moreover, as incomes in the software sector increase, they will likely continue to diverge from those in other sectors. Professionals in IT enclaves could become better connected, both economically and socially, to distant regions than to the rest of the Indian economy. Already most IT development occurs in STPs that are insulated from the day-to-day challenges of doing business in India by dedicated communications links, private power sources, and liberal rules for investment and taxation. While the growing traffic of managers and policy makers between India and Silicon Valley has obvious benefits for India, it risks creating an international technical community with diminishing ties to (or beneficial impacts on) the rest of the country.27 The task for policy makers who aspire for IT to become more than an enclave in an otherwise backward economy is to develop a wider range of industries and institutions to support the economic and spatial diffusion of IT. This will require more far-reaching attention to development of the infrastructure and to education in rural as well as urban India. It will also require a sustained attack on the political and bureaucratic obstacles to the adoption of IT in both public and private sectors. Indian workers today are contributing to the development of software to modernize foreign governments and corporations while their counterparts at home remain woefully backward. 27. Mukherji (2000) describes historic examples of Indian breakthroughs in mathematics and metallurgy that largely bypassed the general population and economy, often providing benefits to outsiders.

Bangalore: The Silicon Valley of Asia?

187

While investment in IT grew rapidly in India during the 1990s, the country’s use of IT remains extremely low by international standards. In 1996, spending on IT was only 0.5 percent of GDP in India, compared to 2.8 percent in the United States and 1.3 percent in Malaysia. In 1997, India had only 2.1 PCs per thousand people, compared to 406.7 in the United States and 46.1 in Malaysia. Even China and the Philippines boasted higher rates of PC penetration, with 6.0 and 13.6 per thousand people, respectively. Finally, in January 1999, India had only 0.13 internet hosts per 10,000 people, while Malaysia had 21.36 and the Philippines 1.21 (table 5.4). This is by no means to suggest that India should pursue a policy of “computers at all costs.” Indeed, we have seen that there are many other (often more pressing) needs in India, ranging from investments in basic infrastructure to improvements in the quality and accessibility of education. And it is possible that a place like Malaysia has actually over-invested in IT given its level of development. The disappointing performance of the Multimedia Super Corridor, which has failed to attract private investments, suggests that the substantial resources devoted to its high-speed communications network and other twenty-first-century infrastructure might have been more wisely invested. However, judicious investments in IT offer the opportunity to improve the productivity of many other sectors of the Indian economy. Applications could be developed to meet many of India’s domestic needs, from revamping the education and health care systems to modernizing the retail and agricultural sectors. The state of Andhra Pradesh has been a leader and a model in public-sector adoption of IT. Naidu’s administration has pioneered the computerization of land records, for example, which improves the efficiency of public service while it also reduces opportunities for corruption (formerly ample). The challenge for India is to overcome the buTable 5.4

Country United States Singapore South Korea Ireland Malaysia Mexico Brazil Thailand The Philippines China India

International Reliance on Information Technology Ratio of IT Spending to GDP, 1994 (%)

PCs per 1,000 People, 1997

Internet Hosts per 10,000 People, 1999

2.8 1.9 1.6 1.3 1.3 0.9 0.9 0.6 0.5 0.5 0.5

406.7 399.5 150.7 241.3 46.1 37.3 26.3 19.8 13.6 6.0 2.1

1,131.52 210.02 40.00 148.70 21.36 11.64 12.88 3.35 1.21 0.14 0.13

Sources: IT/GDP: OECD (1997); PCs: World Development Indicators (1999–2000).

188

AnnaLee Saxenian

reaucratic resistance motivated by fear of the loss of jobs—or of opportunities for graft. IT also offers potential efficiencies in a wide range of private-sector activities, from distribution and marketing to banking to agriculture. Farmers, for example, can use IT for managing their timetables, crop scheduling, soil testing, controlling insects and rodents, and for marketing and water management. Similarly innovative applications of IT can help develop local language software and local content that will allow the entire population of India to access the benefits of the internet. This, too, will require both the continuation of liberalization and regulatory reforms as well as incentives for investments in innovation to meet domestic needs. 5.3.1 Looking Ahead: Strategies for IT The new communications technologies have generated important new opportunities for India that should not be overlooked, opportunities to expand remote services such as medical transcription or call centers. These IT-enabled services involve tasks that are too routine for western workers, but not so repetitive that they can be automated. Such remote services could ultimately provide more employment in India than the software services sector, because they depend not on engineers but on large numbers of people with English language skills and the willingness to work for very low wages.28 In addition, most scenarios for the IT sector envision Indian software companies’ starting to develop innovative products and applications as well as continuing to provide low value-added services for export (NASSCOM 1999). Yet they typically overlook the opportunities in India for a localized process of innovation. IT producers typically must work closely with customers to develop expertise and to define and test new products. However, as long as Indian software houses continue to rely primarily on exporting, they forgo the opportunity to test and perfect products through interaction with end-users. India has the technical skill needed to experiment with developing new products and services for the domestic market. In this scenario, IT products would be developed as a means to enable creative solutions to local problems. The prerequisites for such a strategy include continued deregulation in telecommunications, support for entrepreneurship, and the leveling of the playing field so that IT products and services sold domestically enjoy the same tax benefits as those currently enjoyed by software export units. This should allow the private sector to find economically viable ways to serve the domestic market. This strategy involves a commitment to experimentation with technology appropriate to the Indian environment. Products developed in the West are 28. See “Indian Business: Spice Up Your Services,” The Economist, 16 January 1999.

Bangalore: The Silicon Valley of Asia?

189

typically too costly and provide more features than are needed by the vast majority of the Indian population. If products and services were developed that were affordable and reliable, they could transform what is now a potential market into a very sizeable customer base. Consider a product like an electronic pager. The pagers available in India today are produced in the West and sell for approximately Rs. 2,2000, well beyond the means of most of the Indian population. However, the technology is so simple that a pager could be developed and manufactured locally for only Rs. 100. At this price it would be affordable to 20 percent of the Indian population (which is a very large market, equal to the size of the West). Such products would in turn be likely to have substantial export potential elsewhere Asia, in Latin America, and in the rest of the developing world (Jhungjhunwala 1999). The Simputer represents a model of innovation to meet domestic needs. The Simputer is a very low-cost mobile personal computer (priced at under Rs. 9,000, or approximately US$200) that was developed by a Bangalorebased team of engineers. The team—which is drawn from the Department of Computer Science and Automation at the Indian Institute of Science (IISc) and a local design company, Encore Software—designed the product explicitly for the Indian market. While the Simputer is extremely low cost, it applies leading-edge technologies. It is based on free software (the Linux operating system), designed to be open and modular, and offers multiple connectivity options. It also includes a SmartCard reader/writer, which provides a delivery vehicle for financial transactions on the internet and for ecommerce. The Simputer project offers a model of collaboration for India, as well as an innovative product. The collaboration between the IISc and Encore, a public-sector university and a private-sector company, is rare in India, but offers a way to efficiently leverage local capabilities. Similarly, a team at the Indian Institute of Management, Bangalore, is conducting a study of the likely applications for such a device in rural and semi-urban areas. Some of the potential applications include using the Simputer as a platform for microbanking, for data collection, for internet access, for dissemination of agricultural information, and as a laboratory for experiments in rural schools. India already has the design capabilities for developing such products. These capabilities are also evident in the growth of very large-scale integrated circuit design in the ODCs, as well as in the activities of some small indigenous firms like Bangalore-based Silicon Automation Systems and Encore, which are currently developing the sophisticated intellectual property components for semiconductor design. However, India lacks the environment needed to support experimentation with the application of these skills to new markets. Policies to support innovation should facilitate new firm formation. While existing producers typically have resources and experience, as well as

190

AnnaLee Saxenian

established reputations, entrepreneurial start-ups offer flexibility and focus without vested interests. This is why they are frequently the first movers in defining innovative products and services. In India, however, the ten largest firms account for more than 50 percent of software exports at the same time that they represent a small proportion of the total firms in the sector.29 A vibrant entrepreneurial sector could ideally generate innovative small firms that, over time, collaborate with established IT producers to take advantage of their respective strengths. Venture capital is the first step toward encouraging entrepreneurship. If widely available, venture capital can support multiple experiments with new products, new services, and new applications. But venture capital alone is not sufficient. The greater challenge for India will be to create the social and institutional environments that support a decentralized process of experimentation and innovation. The lesson of Silicon Valley is clear: Entrepreneurship is a collective, not an individual process. It depends upon a wider process of collective learning, typically within a localized community (Saxenian 1994). Such a technical community is built through collaborations of the sort that are rarely practiced in India today: collaborations between firms of all sizes, ages, and specializations, between firms and universities or research institutes, and between firms and financial institutions (especially venture capital). The Simputer project represents an important model that should be replicated across India. Collaboration between IT start-ups and established producers with knowledge of particular domains could be especially important in the Indian context. A Bangalore company, Innomedia Technologies, for example, developed a low-cost technology for interactive television that uses existing cable-TV infrastructure to provide video on demand, interactive media, and online shopping. Once the technology was defined, the firm built an alliance with the large, established manufacturer, Reliance Industries, to undertake volume production and distribution. Such collaborations can, of course, involve partners from other regions in India and even elsewhere in the world. The large NRI community in Silicon Valley could become an invaluable resource in identifying and coordinating such long distance partnerships. However, the precondition is the creation in India of the local networks that support the recombination of capital, skill, and technology into new ventures. Such an environment is emerging from an R&D group led by the Telecommunications & Computer Networks Group at the IIT Madras. This group includes university faculty and several small R&D companies formed by alumni, as well as distant collaborators. The group’s mission is to make possible 25 million internet con29. NASSCOM (1999) reports that there are 826 companies in India engaged in the business of software exports.

Bangalore: The Silicon Valley of Asia?

191

nections in India in less than ten years. It has strategic alliances with IC manufacturers abroad to develop wireless access, fibre access, and internet access systems specific to the needs of developing countries (Jhungjhunwala, Ramamurthi, and Gonsalves 1998). Comparable networks can be created in other regions. Policy makers (ideally state governments, since they are typically closer and more responsive to local needs) might provide incentives for collaborations between companies, or between companies and local universities or other research institutions. Or they might facilitate associational activities that bring together local producers, researchers, and service providers to seek solutions to shared problems such as the shortage of skilled labor or the need for better infrastructure. This process should facilitate the creation of crosscutting social and technical networks that, over time, support information sharing and collective learning. The independent, outwardly oriented companies and institutions that currently characterize the Indian scene have the potential to become localized technical communities with differing specializations related to their institutional and resource endowments. India’s secretive public-sector units, such as the aerospace and defense research outfits in Bangalore, for example, could provide a rich source of technological opportunities if their boundaries were opened up and skill and know-how were allowed to flow more freely within the region. Similarly, venture capitalists and other service providers could, with time, become more knowledgeable about local capabilities, opportunities, and resources in order to play a growing role in coordinating and facilitating local experiments across India. Finally, while the Indian Institutes of Technology produce among the best engineers in the world, their graduates still leave the country in large numbers. This group (or even a subset of them) could play a technological leadership role in India in the coming decades if more were to return to or stay in the country. As it stands now, however, too few remain or return to make an impact. By accelerating the deregulation of telecommunications and other key sectors, upgrading the physical infrastructure, and enhancing conditions for entrepreneurship, the government could create conditions under which more NRI’s would be willing to invest in the Indian economy. It is even possible that young Indian engineers would return in far greater numbers than in earlier generations if they saw viable economic opportunities at home. This could make a substantial difference to India’s future. 5.3.2 Concluding Comments The IT industry has brought a wide range of important and tangible improvements to India. It has provided the confidence that India has a future in the new economy. And it has generated jobs, wealth and exports. More-

192

AnnaLee Saxenian

over, the pace of policy reform in the IT industry has been unprecedented. This reflects, at least in part, the opening up of the policy debates to include new actors. The industry association, NASSCOM, has accelerated the policy reform process through its aggressive lobbying while helping to define a minimally interventionist model of industrial promotion. Meanwhile, entrepreneurial state governments, spurred by the example of Andhra Pradesh, have pioneered a potentially far-reaching, bottom-up process of policy reform. However, there are also substantial dangers in the current fascination with IT in India. The challenge today is twofold. First, there is a need to be very realistic about the limits of software as a development strategy for India. Bangalore is not Silicon Valley, and IT is not going to solve all of India’s problems. IT is still a very small piece of the Indian output and exports, and even if it grows rapidly it will remain only one among many sectors that contribute to Indian development in coming decades. This suggests the second challenge, the need to widen the range of participants in the policy debates and to broaden their scope still further. The alliance between the large software industry and the government has restricted the debate over IT policy. The goal should not be to simply meet the needs of a handful of producers, but rather to use IT as a means to strengthen the fabric of the entire economy and to enhance opportunities and living conditions for the whole Indian population.

References Arora, Ashish. 1999. Quality certification and the economics of contract software development: A study of the Indian software industry. NBER Working Paper no. W7260. Cambridge, Mass.: National Bureau of Economic Research, July. Arora, Ashish, V. S. Arunchalam, Jai Asundi, and Ronald Fernandes. 2000. The Indian software services industry. Carnegie-Mellon University, Heinz School of Public Policy and Management, Working Paper. Desai, Ashank. 1999a. The domestic software industry in perspective. Times Computing Online, 6 January. Available at [http://www.timescomputing.com/19990106/ spk1.html]. ———. 1999b. Problems confronting the software entrepreneur. Times Computing Online, 17 March. Available at [http://www.timescomputing.com/19990317/spk1. html]. Dossani, Rafiq. 2000. Reforming venture capital in India: Creating the enabling environment. Stanford University, Asia/Pacific Research Center, Working Paper. Dossani, Rafiq, and Lawrence Saez. 2000. Venture capital in India. Stanford University, Asia/Pacific Research Center, Working Paper. Grieco, Joseph M. 1984. Between dependency and autonomy: India’s experience with the international computer industry. Berkeley: University of California Press. Heeks, Richard. 1996. India’s software industry: State policy, liberalisation and industrial development. New Delhi: Sage Publications.

Bangalore: The Silicon Valley of Asia?

193

International Telecommunications Union (ITU). 1998. World telecommunications indicators database. Geneva: ITU. Jhunjhunwala, Ashok. 1999. Remarks given at conference on Equity, Diversity, and Information Technology, National Institute of Advanced Study. 3–4 December, Bangalore, India. ———. 2000. Can information technology help transform India? Madras: Indian Institute of Technology, Department of Electrical Engineering. Jhunjhunwala, Ashok, Bhaskar Ramamurthi, and Timothy A. Gonsalves. 1998. The role of technology in telecom expansion in India. IEEE Communications Magazine 36 (11): 88–94. Lowell, B. Lindsay. 2000. H-1B temporary workers: Estimating the population. Georgetown University, Institute for the Study of International Migration. Unpublished Manuscript. Ministry of Finance. 2001. Economic Survey 1999–2000. New Delhi: National Informatics Center, government of India. Mukherji, Joydeep. 2000. Information technology in India: Yet another missed opportunity? Standard & Poor’s CreditWeek, 12 July: 18–25. National Association of Software and Service Companies (NASSCOM). 1999. Indian IT strategies. Report prepared by McKinsey & Co. New Delhi: NASSCOM. ———. 2000. The software industry in India: A strategic review. New Delhi: NASSCOM [http://www.nasscom.org]. National Taskforce on Information Technology and Software Development. 2000. IT action plan. New Delhi: National Informatics Centre. Available at [http:// it-taskforce.nic.in]. Organization for Economic Cooperation and Development (OECD). 1997. Information technology outlook. Paris: OECD. Parthasarathy, Balaji. 2000a. Globalization and agglomeration in newly industrializing countries: The state and the information technology industry in Bangalore, India. Ph.D. diss. University of California at Berkeley. ———. 2000b. An Asian Silicon Valley in Bangalore? Evidence for the changing organization of production in the Indian computer software industry. University of California at Berkeley, Department of City and Regional Planning. Unpublished Manuscript. Prabhakar, Mohana. 2000. Cyber laws will be in place: Dr. N. Seshagiri. Itspace. com. Available at [http://www.itspace.com/ItspaceAlpha/features/print/Itpolicy/ seshagiri.asp]. Saxenian, AnnaLee. 1994. Regional advantage: Culture and competition in Silicon Valley and Route 128. Cambridge, Mass.: Harvard University Press. ———. 1999. Silicon Valley’s new immigrant entrepreneurs. San Francisco: Public Policy Institute of California. Available at [http://www.ppic.org/publications/ PPIC120/ppic120.abstract.html]. Securities and Exchange Board of India (SEBI). 2000. Report of K. B. Chandrasekhar committee on venture capital. New Delhi: SEBI. Sen, Pronab. 1994. Exports from India: A systemic analysis. Electronics Information and Planning 22 (2): 55–63. Software Technology Parks of India (STPI). 1997. Directory of STP Units. New Delhi: STPI. Sridharan, Eswaran. 1996. The political economy of industrial promotion: Indian, Brazilian, and Korean electronics in comparative perspective 1969–1994. Westport, Connecticut: Praeger. Subramanian, C. R. 1992. India and the computer: A study of planned development. New Delhi: Oxford University Press.

194

AnnaLee Saxenian

Comment on Chapters 4 and 5

N. R. Narayana Murthy and

Sandeep Raju The Indian IT industry underwent an unprecedented transformation in the 1990s. Industry revenues grew in excess of 50 percent annually after 1991 and aggregated US$5.7 billion in the fiscal year 2000. A 1999 McKinsey study projected that revenues would grow to $87 billion by 2008. Of this, it was anticipated that $50 billion would be from exports. As of 2000, India has an 18.5 percent share in the global cross-country customized software market: Over 200 Fortune 1000 corporations now outsource their IT requirements to India. Moreover, in the past five years, around 25 percent of the technology ventures in Silicon Valley have had Indian minds behind them—over 750 technology companies in the Valley are now Indian-managed. Clearly, India has become a leading force in the global IT arena. The economic impact of this phenomenon is tremendous and will become more so. By 2008, the IT sector is expected to be the largest Indian exporter, accounting for nearly 35 percent of exports from India. It will account for $4 to 5 billion of foreign direct investment into India and will create 2.2 million jobs across various skill classes. The hi-tech boom has the potential to bring unprecedented economic prosperity to the Indian populace and to help raise living standards for people across all classes of society. We wish to add our own perspective to the insights offered by Saxenian and Forbes. We focus first on the domestic development of the IT industry as related to Forbes’s discussion, and then turn to contrasts between Silicon Valley and Bangalore and give our own perspective on the contrasts. Looking at the development of the domestic software industry, we first discuss factors that have fueled this phenomenon, and then analyze areas where significant improvements could occur. Turning first to the factors leading to the success of Indian IT, we can identify eight. First, by far the greatest advantage of India is its pool of well-educated, high-quality, English-speaking people. In the software services industry, where scalability is of paramount importance and analytical reasoning is critical, India’s talent pool is a unique asset. There are 300,000 professionals employed today in the Indian IT industry, a number second only to the United States, and there are 75,000 new professionals every year. India also has over 4.3 million technical workers and around 2,000 institutions of higher education, which train about 74,000 software professionals every year. In addition, the Indian Institutes of Technology and the Indian Institute of Science offer technical training comparable to the very best in the world. N. R. Narayana Murthy is chairman and chief executive officer of Infosys Technologies Limited. Sandeep Raju heads the Life Sciences practice at Infosys Technologies Limited. This comment was written in October 2000.

Bangalore: The Silicon Valley of Asia?

195

A second advantage is the nine- to twelve-hour difference in time zones between the United States and India. The bulk of Indian software exports, around 61 percent in 1999, are to North America. The difference in time zones facilitates efficient project execution, with compressed time frames, through seamlessly integrated cross-border teams. This twenty-four-hour virtual workday enables IT services companies and their clients to fully leverage the benefits of globalization and is a significant comparative advantage for India. Yet a third source of strength for the Indian IT industry is the Indian software industry’s focus on quality. Of the 21 companies worldwide that have a Level 5 certification on the Capability Maturity Model (CMM) of the Software Engineering Institute (SEI) at Carnegie-Mellon University, the process quality benchmark for the software industry, 15 are located in India. Further, 148 more companies from India have achieved ISO 9000 or SEI Level 2, or other equivalent certification, and 136 more are in the process of acquiring the same. These statistics point to a substantial difference between the quality focus of Indian companies and their counterparts in other countries. To date, however, Indian companies have not been very successful in leveraging this quality focus to command higher price points for their work. Related to quality is the fact that the latest hardware and software technologies are sold and supported in India. Consequently, Indian IT professionals have easy access to the latest developments in their domains. Further, the vibrant education and training market quickly assimilates new technologies and thereby offers ample opportunities for prospective IT engineers to be proficient in cutting-edge technologies. Moreover, technical consultancy services in software development methodologies and qualityrelated areas are readily available to Indian companies. While India may be lacking in some other aspects of communications infrastructure, an advantage for the IT industry is that, in India, satellite links are readily available up to a capacity of 512 kilobytes per second (although the costs are relatively high contrasted with other countries with which Indian firms compete).1 These links are a key factor in enabling teams that span the globe to accomplish software development tasks. Yet another advantage is well known: The low cost of living in India helps companies maintain wage levels that are far lower than their counterparts in developed nations. Personnel costs are a key determinant of profitability, and India has a great advantage on this front. The average Indian software professional with two to three years of experience would command a salary of around $10,000 a year, while a comparable figure in Silicon Valley would be at least six times as much! Further, real estate is far less expensive in India than in the technology habitats of the West. The per-square-foot cost of 1. See Saxenian’s discussion of Indian offshore development (this volume).

196

AnnaLee Saxenian

a world-class facility in India is around $60, far cheaper than any of the hitech regions in the West. Since 1991, India has had liberalization-friendly regimes that have acted as a catalyst for economic growth. The early phase of economic reforms saw the abolition of widespread industrial licensing, rationalization of taxes, a strong thrust on exports, a reduction in import tariffs, reforms on foreign exchange regulation, free pricing of initial public offerings, foreign participation in Indian capital markets, and a regulatory framework that has permitted employee stock option plans. The mid-1990s witnessed further significant structural reforms, especially in the capital markets. The government of India has also enacted entrepreneurship-friendly policies and has been proactive in supporting the IT sector. The IT industry currently has a single window for all operations-related transactions, the Software Technology Parks of India (STPI). There are no licensing-related inefficiencies. Unlike in earlier years, boardroom decisions are taken without government interference. Fiscal incentives to the IT industry were one of the cornerstones of government policy in the early years of the high-tech boom. Import duties were withdrawn on imports of capital goods by IT exporters, and profits from software services were made tax-free. Many of these incentives continue to be available to the IT industry, but the policy regime as a whole has also switched to being positively supportive in recent years. A key shift has been the move to involve industry representatives actively in the policy formulation process. In 1998, the central government established the national IT task force, consisting of leaders of the software and hardware industries. Further, the establishment of the Prime Minister’s Council on Trade and Industry, which included a group on knowledgebased industries, helped bridge the disconnect between government policy and industry aspirations. More recent initiatives have also attempted to actively tap the expertise of successful Indian technopreneurs in the United States. Apart from the emergence of IT as an issue of economic interest at the federal level, there is tremendous competition among various state governments to attract investments in this sector. For IT companies, this translates to land and infrastructure at very low prices, accelerated decision-making within the state government, and additional fiscal incentives. This intense competition is manifested in state-level policy initiatives: Today nearly twenty states in India have announced IT policies, many in collaboration with industry. Today India is at a stage where, for the first time in history, its corporate leaders are listened to and their suggestions acted upon. This, coupled with the government’s sense of urgency to change things for the better, is indeed a remarkable achievement. A final advantage is attractive financing opportunities. Even after the drop in technology stock prices in the first half of 2000, the valuations of IT

Bangalore: The Silicon Valley of Asia?

197

stocks in India are far higher than their traditional economy counterparts. The presence of an attractive capital market, with high-quality regulation, has been a boon to Indian entrepreneurs. For early-stage companies, venture capital is readily available. As in the West, money is abundant; it is the business ideas and execution capabilities that are at a premium. In 1998 and 1999, the Indian venture capital industry not only saw a massive increase in the quantum of funds available, but it also witnessed increased sophistication and differentiation among VCs. In contrast with even two years earlier, there emerged specialized funds focusing on seed funding, early-stage investments, midstage investments, and so on. Further, again unlike the situation of earlier decades, debt finance is readily available for companies without a large base of physical assets both from banks and from stateowned institutions. Despite all of these advantages, there remain a number of disadvantages confronting the Indian IT industry. Perhaps the chief disadvantage remains physical infrastructure. Roads, airports, and power supplies all need improvement. Core infrastructure areas are largely state-owned monopolies; this has resulted in extreme inefficiency in these sectors with large wastage of public funds. A second and related disadvantage, especially for the IT industry, is that while high-quality datacom links are available to corporate users, bandwidth is still a constraint, especially for individual users. While the developed world has enthusiastically embraced the e-business paradigm, India runs the risk of choking its nascent digital revolution due to inadequate bandwidth and appalling last-mile connectivity. Third, the lack of convertibility on capital account has often had an adverse impact on the day-to-day operations of global Indian corporations. While macroeconomic imperatives may warrant prudent foreign currency management, removal of procedural bottlenecks and a concerted effort toward full convertibility as soon as feasible is an urgent need. Likewise, despite improvement in acquisition guidelines, these are still inadequate by global standards and put Indian companies at a disadvantage when negotiating with their foreign counterparts. Negotiating from a position of strength and ability to implement high-speed execution are of paramount importance in such transactions. Further, although Indian laws have evolved to make them IT-friendly, the judiciary and the legal process definitely need to be overhauled. The Indian legal system is notoriously slow—disputes often take several years to be resolved. Another disadvantage is that Indian companies have not yet been able to achieve brand recognition. The Indian software development paradigm has been acknowledged the world over as a powerful model for delivering highquality IT solutions at competitive costs. Leading companies from across the globe have set up operations in India. However, despite isolated efforts

198

AnnaLee Saxenian

to build brand equity, Indian companies are still far behind their global counterparts in positioning, visibility, and access to the top echelons of their clients. It is regrettable that, in an industry whose demand levels are incredible and whose only constraints are on the supply side, Indian companies have not been able to improve their brand images and increase their gross margins. Yet another disadvantage is that India’s performance on core indicators of development is abysmal, as is indicated in chapter 1 of this volume. The indifference to the poignant circumstances of the urban and rural poor is appalling. There is no doubt that the information revolution has created numerous middle-class millionaires, but there is still a chasm between the haves and have-nots in India. Poverty, literacy, primary education, health care, caste-driven oppression, and related issues remain to be addressed both for social reasons and to increase the supply of individuals who can contribute productively. Finally, while India’s performance on the software front is creditable, the country is still plagued by archaic practices in core industries and by inadequate focus on hardware and fabrication industries, as Saxenian’s data indicate. While Taiwan, Korea, and Japan have managed to attract billions of dollars of investments in silicon chips and electronic components, India has yet to see a single large investment by any of the global leaders in this area. Fabrication facilities are highly capital-intensive, requiring upfront investments in excess of a billion dollars, and there is tremendous economic potential that could be tapped through proactive initiatives by policy makers to attract investment in this area. Summing up the positives and negatives together, we see that India’s performance on software has been outstanding, and the future is promising. The pace of change in the business environment has been remarkable, as documented by Forbes’s account; high-quality governance in recent years has been a tremendous asset for the country. Consequently, India has gained immense respect in the world community and has emerged as a leading destination for investment and outsourcing in the IT industry. Of course, a lot remains to be done—the next ten years are crucial. However, the momentum of economic reforms, the increased focus on human development, a stable and proactive governance, and the perceptible trickle-down of economic prosperity are compelling reasons for continued optimism. Turning now to the contrasts between Silicon Valley and Bangalore, we see that over the last decade Bangalore has emerged as a focal point in the landscape of the global IT industry. This metropolis of around 5 million people, in south India’s Deccan Plateau, has emerged as a major center of the country’s IT industry, with Indian firms and branches of leading global IT players as well as fine academic institutions, IT-focused venture capitalists, and innumerable small-scale outfits that have mushroomed to tap the ever-expanding opportunities offered by the global and Indian market-

Bangalore: The Silicon Valley of Asia?

199

place. Today, Bangalore has gained international recognition as being among the hottest tech places in the world—a distinction that places it in the same league as Silicon Valley, Cambridge, Tel Aviv, Singapore, and other techno-hubs. The reasons for this are not far to seek. As already indicated, an incredible talent pool, extraordinary cost competitiveness, a strong entrepreneurial culture, and the advantages for the industry I have already listed are a potent mix, especially in the context of a third-world economy otherwise saddled with poverty, illiteracy, and social inequality. Naturally, the meteoric rise of the city invites comparisons with Silicon Valley, the nucleus of the information revolution and historically the most popular destination for high-technology companies. While the socioeconomic contexts of these two hi-tech habitats are vastly different, there are indeed similarities that have facilitated the emergence of knowledge economy companies in both places. There are contrasts in the entrepreneurial culture. Although both are dream places for hi-tech entrepreneurs, the Bangalore professional is much more easily satisfied than his or her counterpart in the Valley. Given the “contentment” imperative that guides many an Indian’s ambitions, it is still not glamorous to get rich in Bangalore, especially for the typical technically trained, middle-class IT professional. Further, middle-class India still places great stress on a conservative approach to career planning and is not comfortable with the singleminded pursuit of material gains. Further, other key aspects of an entrepreneur-friendly environment, outlined below, are not yet advanced in India. For these reasons, the motivation for entrepreneurship is not as high in Bangalore as in the Valley. Going forward, as the climate for entrepreneurship improves and as more and more role models emerge, there is likely to be a significant shift in this mind-set. There are also contrasts in business domains. Silicon Valley is the seat of innovation in several technology domains: The region has attracted product companies, dot-coms, cutting-edge technological infrastructure players, and the like. The region thrives on the commercialization of leadingedge research and plays host to many R&D-intensive, development-stage firms. By contrast, Bangalore is largely a software services powerhouse. India lacks the precision engineering capabilities required to emerge as a leading player in hardware-intensive domains. Naturally, software is the most attractive opportunity for Bangalore-centric firms. Within software, a majority of Bangalore-based companies are IT service providers, as noted by Saxenian. They span the entire spectrum from being world-class IT solutions companies to providing staffing solutions to offering manpower-intensive-IT-enabled services using local employees. While the past few years have witnessed an occasional high-quality product from Bangalore, we have not as yet witnessed an outstanding worldbeating package emerging from the city. Of course, several U.S.-based com-

200

AnnaLee Saxenian

panies have based their R&D and product development teams in the city. However, the thrust of the IT revolution in the city is heavily tilted toward services. Among the reasons for this, the first is that Bangalore has not possessed a vibrant VC industry. A successful product requires heavy upfront investments in development and branding, which are only feasible if there is such a set of VCs. Moreover, the appetite for risk among financial institutions, capital markets, and even individual entrepreneurs is not as high in Bangalore as it is in the Valley. A second reason relates to the necessity for marketing and branding efforts for a successful product. Given that the largest market for any potential world-beating software packages lies in the West, the ability to penetrate the U.S. and European markets is of paramount importance. Traditionally, Indian managers have not been very successful at this. Third, creating a world-class product is far more innovation-intensive than providing world-class services; this points to the criticality of lateral thinking and the need for early-stage IT education. Today, the average IT professional in Bangalore is exposed to the field only toward the end of his or her education—PC penetration is still low, and schools often do not provide quality IT education. Historically, this may have been a cause for India’s limited success on the products front. Over time, this should become less of an issue. The availability of talent as a major factor giving India an advantage was already mentioned. IT is the aspirational career for many of the over 250,000 technical professionals that India’s more than 2000 institutions produce each year. And Bangalore attracts more IT professionals than any other city in India. Likewise, the cost advantages cited for India pertain especially to Bangalore, which was historically relatively low cost even in India. Similarly, Bangalore has several world-class institutions of research and learning, as mentioned earlier. I have already mentioned that the VC industry is springing up fast in India, and especially in Bangalore. But Silicon Valley has the advantage of having competent venture capitalists. In an age in which money is a commodity, the ability to add business value to a venture is the key differentiator between a good VC and an also-ran. Though Bangalore has attracted several high-quality technology-focused venture capitalists, it has a long way to go before matching the hands-on approach, commitment, relationships, and risk appetites of some of the leading VC funds in the Valley. By far the most important driver of Silicon Valley’s success has been the proliferation of linkages between business, academia, and venture capitalists on the west coast. Historically, the benefits derived from networking among, and within, these constituencies have been on a scale far greater than in Bangalore. In the words of Saxenian, the culture of the Valley has improved both intra- and inter-firm porosity.

Bangalore: The Silicon Valley of Asia?

201

Efficient commercialization of cutting-edge output from research labs, entrepreneurship forums at universities, highly efficient alumni networks, close links between leaders in academia and business, risk appetites of venture capitalists, synergies between science/engineering schools and business schools, collaborative research among universities, keiretsus bringing together businesses and venture capitalists, angels with the willingness to nurture talent, the abundance of forums where youngsters may put forth their ideas and interact with industry leaders, opportunities for collective learning—all these are differentiators that put the Valley several notches above other hi-tech habitats. In sum, Silicon Valley operates in a vibrant market economy that reveres innovation. The Bangalore entrepreneur, on the other hand, does not have easy access to all these resources. However, it must be borne in mind that the information revolution is a fairly recent phenomenon in Bangalore: The next few years will definitely see increased networking among entrepreneurs, universities, and venture capitalists in the region. Another contrast relates to the use of stock options. Employee ownership in ventures and the use of stock as a currency for professional services are common in the Valley and enable a reduction in cash costs in early stages of firms’ development. In contrast, Bangalore has yet to see high acceptance of stock as a currency in lieu of cash payments. Recently, employee stock option plans have become fairly commonplace, although certain regulatory and tax issues remain to be addressed. The smaller acceptability of stock options is even more significant when considered in the context of the maturity of financial markets in India. In contrast to those in the Valley, India’s capital markets are not yet efficient or sophisticated enough to command the value they deserve. The regulatory framework is still in the process of being adapted to the needs of technology companies. Moreover, investors in the Indian market have not yet had adequate exposure to the technology sector and occasionally fail to grasp its nuances. The fact that there are restrictions on capital flows to and from abroad, as mentioned above, further limits the financing opportunities of Bangalore-based firms. In contrast to Silicon Valley’s vibrant economy and the American government’s laissez-faire approach, Bangalore’s IT industry still faces some challenges from various aspects of governmental regulations, including labor legislation, limited government-industry interaction, and the lack of efficient economic diplomacy. As mentioned earlier, however, recent times have seen IT-friendly governments at the central and state levels, and the next few years may see even more progress in this dimension. In a related issue, smaller enterprises and individuals are nowhere close to tapping the potential efficiency gains from e-commerce. The reasons for this include severe constraints on bandwidth, limited PC penetration, concerns about the security of e-transactions, lack of local language content,

202

AnnaLee Saxenian

regulatory bottlenecks, low credit card penetration, fulfillment capabilities of e-tailers, a lack of upfront investments to establish a credible presence on the web, and the like. All these factors, in addition to the fact that traditional distribution networks are fairly well developed in India, imply that it will take a while for e-business to witness explosive growth in this country— possibly another half decade or so. As a result, in sharp contrast to the Valley, Bangalore firms are largely export-oriented, as the domestic market is relatively less attractive for high-tech companies. Finally, both the physical and the technological infrastructure limitations discussed for all-India also apply to Bangalore. This includes not only the bandwidth and physical infrastructure already discussed, but the availability of legal and accounting services. Perhaps the biggest contrast between Bangalore and Silicon Valley, however, lies in the socioeconomic context of the two cities. In Bangalore the benefits of IT have yet to percolate down to the masses. Further, the leaders and professionals in Bangalore have to transition twice a day from a thirdworld environment to the demands of first-world corporations. This annealing process toughens people but can be mentally exhausting. The key to narrowing this economic divide is reform at the grassroots level—increasing literacy, access to health care, land reform, efficient agricultural practices, and so on. IT has the potential to play an important role here, with cities like Bangalore best poised to take the lead on this front. There are a number of ways IT can help. First, the Internet has the potential to enable large scale information dissemination and to facilitate exchange of ideas at the grass roots. This would be especially useful in the area of education and would also help voluntary organizations and village communities exchange best practices. Although these changes are not imminent, the next few years should witness an increase in grassroots involvement in the information revolution. Second, advances in IT will increase efficiency in utilizing natural resources and in predicting weather patterns and natural disasters; these in turn have far-reaching impacts on the well-being of the rural population. Third, e-governance initiatives will increase efficiency of governance: This will increase the percentage trickle-down of public funds to grassroots projects and will also free up funds otherwise spent on overly elaborate administrative structures. Fourth, efficiencies from IT-enabled disintermediation will help rural producers get a larger share of the economic pie. Fifth and finally, the trickle-down effect of the growth in the Indian IT industry will help raise the standard of living among the less privileged sections of society, at least in high-tech cities like Bangalore. In sum, Bangalore and Silicon Valley are similar in many respects—aspirational leadership, high-quality people with high mobility, world-class research and educational institutions, great climate, availability of venture

Bangalore: The Silicon Valley of Asia?

203

capital, and so on. Both are located away from the capitals of their respective countries—which is probably a plus for entrepreneurship! However, there are wide differences between the two because of their respective histories and socioeconomic contexts. While Silicon Valley’s track record provides very useful pointers to the direction Bangalore should take, the two regions are also complementary in many respects. Given that Indians have already made a mark in the Valley, the future should see increased symbiotic linkages between these two high-tech habitats.

Comment

Ashok V. Desai

The ascent of the Indian software industry is recent, and hence underresearched; AnnaLee Saxenian is one of a handful of pioneers in this field. Her early work was more on the Silicon Valley end—on the growing presence of Indians in it. In this paper she turns to the Indian end and she takes up two tasks. First, she updates the history of the industry, on which the two authoritative sources, Greico (1984) and Heeks (1996), have become outdated. Next, she evaluates the development trends in the industry, in particular its heavy export orientation, and proposes an alternative that might serve India’s national interests better. The paper by N. R. Narayana Murthy and Sandeep Raju is more factual and provides confirmatory underpinning for Saxenian’s paper. The History Murthy and Raju confine themselves to the current state of the industry and have little to say on the history. Saxenian’s account chronicles the rapid growth of the software industry, but also takes in the failure of the hardware industry. She rightly points out the harm done to the hardware industry by heavy protection in the 1970s and 1980s, to which she attributes its failure to achieve economies of scale and its stagnation in the 1990s. It may be added that the dismantling of that protection in the 1990s was responsible for the stagnation; finding itself unable to compete with imports, the industry became an assembler of imports with little value addition. Although India has largely removed quantitative restrictions and scaled down tariffs, its general tariff level is still high. Tariffs on computer hardware were brought down early and more sharply than the rest; this was to serve the interests of the software industry. Thus, the success of the software industry and the failure of the hardware industry are not unconnected. Saxenian cites some interesting material from interviews. N. Seshagiri, Ashtok V. Desai is consulting editor of the Business Standard, India.

204

AnnaLee Saxenian

who as chief of the government’s National Informatics Centre long played a critical role in IT policies, is a prime source. His and other contributions are the basis for Saxenian’s interpretation of policies and the role they played in the emergence of software. His version of events, which holds that software flourished because bureaucrats did not understand it enough to tie it into knots, is basically correct. But there is also the fact that the instruments available to the government were incapable of arresting the growth of service exports, especially those delivered abroad; controls on industrial production, trade, and investment could not curb them. A couple of million Indians went to work in the Middle East after the oil boom; the Indian government could do nothing to control them. It did try to introduce controls on the issue of passports, but was prevented by the Supreme Court, which ruled that all nationals had a right to a passport. It also tried to control the remittances of Indian workers abroad, but failed completely. Informal channels developed to deliver their savings to their families in India, and their exchange earnings were used to smuggle in gold, which also came in despite official prohibition; in fact, India became the world’s biggest market for gold in the late 1980s despite a ban on imports. Bodyshopping, the form that early software exports took, was also the export of labor, and evaded government controls in the same way as other labor exports. Thus it might have continued into the 1990s; software exports could have contributed to an even faster growth of smuggling. What the liberalization of the 1990s has done is to bring them—and indeed all labour exports—into the mainstream of the economy, and especially to permit the emergence of offshore development centers, which produce software in India and export it through satellites. This illustrates and suggests that the macroeconomic side of the story may need to be told in greater detail. The availability of engineers trained in Unix was important; so were India’s engineering colleges, themselves a carryover from the heyday of socialism and heavy industry in the 1960s. But even more fateful was P. Chidambaram’s trade policy of 1992, which abolished quantitative import restrictions on computers and peripherals, and the drastic reduction in tariffs over the following three years. The domestic software industry—as distinct from bodyshopping—could not have become internationally competitive if India’s computers had continued to cost 50–100 percent more than its competitors’. Its export success is a measure of the damage India has done to its other exports by its continuing policy of high tariffs. Similarly, India’s restrictions on foreign direct investment have been virtually suspended in the software industry. In respect of every other industry, foreign investors have to run a gauntlet: After applying to the Secretariat for Investment Approvals in the Ministry of Industry, they must lobby bureaucrats and ministers, then wait around for months, sometimes forever. For software companies, however, none of these restrictions apply; their ap-

Bangalore: The Silicon Valley of Asia?

205

plications sail straight through. If the software tycoon is as big as Bill Gates, chief ministers of states line up to beg him to invest in their states. Virtually every major software multinational has a base in India today. Nor has this destroyed Indian enterprises; in fact, there is no industry in which Indian enterprise is more active than software. What has happened is a certain degree of functional specialization: The presence of foreign companies has given Indian workers an entry into the U.S. market without leaving India, while Indian companies have gone about developing niche products and services for large clients abroad. Thus, the success of foreign investment in the software industry is a measure of the failure of India’s restrictions on foreign investment elsewhere. The Product Market This brings me to the most important part of Saxenian’s paper. Saxenian rightly deprecates the comparisons between Bangalore and Silicon Valley, and points to the features of the latter that cannot be replicated in India— the high level of education, the excellent infrastructure, and the quality of regulation. All these features are stressed by Murthy and Raju. She notes that the software boom is concentrated in the south, and that the vast population of northern India is barely touched by it: It is confined to a few urban centers and has left the rural areas untouched. She fears that the engineers newly enriched by the industry would emulate the lifestyles of their peers abroad, and that vast disparities would emerge between them and the rest of the population. These are legitimate concerns, and it would be folly to ignore them. Her remedy to avoid these ill-effects is threefold: 1. Innovations to serve the domestic market, and the development of local networks which support innovation and bring them to fruition, such as are common in Silicon Valley, 2. A venture capital industry to fund small entrepreneurs who would take up the innovations, and 3. The setting up of two-year and four-year colleges to train mid-level IT workers. These proposals are important, and I would like to discuss them in some detail. But before I begin, I would like to say a word about what I think Saxenian does not mean. The government’s IT Task Force, which she refers to, shared her concern about the domestic market, and made a large number of recommendations to expand it. Its basic approach was to throw money at the market: to give subsidized computers to schools, students, teachers, and others, to finance software in local languages, to spend money on finding uses for IT in the governments, and so on. This type of hijacking of government resources is not unique; it has happened many times before. There was once a campaign to bring water to every village, then one to bring electricity to every village, then one to con-

206

AnnaLee Saxenian

vert all meter-gauge railway tracks to broad gauge, and right now there is one to bring a telephone line to every village. Politicians are extremely fond of such campaigns, and often a beneficiary industry encourages them. They all involve considerable government expenditure, which everyone—politicians, bureaucrats, industrialists—can intercept with profit; and they all have two basic characteristics. First, the criterion of evaluation is deliberately kept very simple: It is the number of villages with water, electricity, telephone or access to internet, and never involves the number of people or the quality of supply. The result is predictable: The majority of the villages are connected or supplied at enormous expense, but the people’s access remains poor. Second, there is never any discussion, let alone evaluation, of the cost of achieving the goal, never an attempt to minimize cost, and never a study of the efficiency with which it was achieved. Usually such plans meander on for years, waste enormous amounts of money, and are then forgotten. I am sure, although it is not clear from her paper, that Saxenian does not mean this kind of development of a local market. Murthy and Raju, although they run a company which would benefit from this type of government action, do not mention it among the government measures they feel are required. Instead, they focus on the disadvantages of outdated labor legislation, limited government-industry interaction, and the lack of efficient economic diplomacy. Additionally, although this is less clear from her paper, I am also fairly confident that Saxenian does not share the export phobia from which many Indian leftists suffer. She associates the export boom with enclave development of an elite community. For one thing, the Indian society is not one of the most unequal among developing countries. For another, inequality does not arise from the rise of a single export industry, be it textiles in the 1970s, gemcutting in the 1980s, or software in the 1990s. Inequality is endemic in Indian society; it was as sharp before the software boom as after, and is as sharp in cities untouched by this boom as in those that are its centers. It is a function of the property distribution, of production technology, and of the way the labor markets work; none of these is going to be significantly affected by the software boom in the short run, and its long-term effects, if any, can be only beneficial. One should not underestimate the multiplier effects of prosperity; the poor of India’s more prosperous states, such as Punjab, Gujarat, and Maharashtra, live better than the poor in other states, and so will the poor of the states that export software. The choice is not between serving the domestic and the international markets. The latter will be served whether we like it or not, and Indians will be employed in serving it, since they have the skills, are available at lower wages, and love the fabulous earnings that software production brings. The question is whether they will serve it from within India or from somewhere in the United States, Britain, Germany, or Singapore. It is better that they should be in India, for the multiplier effects on the Indian economy, and on its poor, will be considerably greater if they do.

Bangalore: The Silicon Valley of Asia?

207

Finally, one should not project their present high level of wages into the future. It is due to a shortage of labor in an industry with high gestation lags for training, especially of managers and architects. Over a longer period, supply is bound to catch up with demand, and wages are likely to come down to more realistic levels. As supply increases and relative wages decline, employment will rise more quickly, with a greater multiplier effect. Innovation and Networks Saxenian would like to see in India something that is so evident in Silicon Valley and so fundamental to its economic success—namely, cooperative relationships between innovators, financiers, producers, universities, nongovernmental organizations, and others that lead to a rapid adoption and diffusion of innovations. She notes the contrasting absence of such relationships in India, and the stubborn estrangement of different classes in Indian society, and wishes India were a bit more like Silicon Valley. She is absolutely right to do so; the question is, how can it be made so? What policies would “encourage producers to experiment with developing new products and services for the domestic market”? The examples she gives suggest that she has in mind products and services not very different from those serving industrial markets, but much cheaper, and if necessary simpler, to develop a mass market at India’s lower income levels. Such products apparently already exist, and more could be innovated—if only there were a market. All the products she mentions require electricity, and some, telecommunications. It seems to me that to diffuse IT, India needs to reform its power and telecommunications industries rather than have an innovative IT policy. Further, the products she cites, such as pagers and computers, hardly require fresh software. Software is an ingredient of widespread process innovation in industrial economies, not so much of product innovation; especially in the case of web-embodied software, it is shortening value chains and cutting into distribution costs. It can do the same in India. In order for it to do so, however, transport needs to be considerably improved, and barriers to inter-state movement of goods arising from state taxes and controls need to be removed. Here, too, the crucial reforms lie outside IT. The distance in India between science, production, and consumption is remarkable. Scientists in universities and laboratories think that industrialists are philistines who run abroad whenever they need technology; industrialists think that scientists are unreliable and unworldly, so that when one needs an innovation, he does not know to whom to go. This is a general problem, and not one confined to IT. Some parts of the remedy are obvious. If, for instance, universities and public laboratories had to support themselves by selling their services, and if their salaries were flexible, their scientists would be more inclined to sell technology. If the tax structure did not confine the R&D laboratories of companies to doing research for the com-

208

AnnaLee Saxenian

panies alone, and did not penalize them for selling their services outside, they would be prepared to look for outside problems to solve. In general, the government needs to get away from funding institutions and to begin to fund solutions. Thus, while Saxenian’s concern with the dysfunctional qualities of Indian infrastructure for innovation is legitimate, the solutions for the problem are quite general and lie far outside the IT industry. Venture Capital Saxenian’s concern about venture capital has been widely shared in India; she describes the steps taken by the government to facilitate its emergence. However, the authorities in India want a venture capital industry because there is one in Silicon Valley. Saxenian’s interest is quite different. She notes the high level of concentration in the Indian software industry, contrasts it with the proliferation of start-ups in Silicon valley, and sees venture capital primarily as a means to encourage the emergence of small, innovative firms. Here, she rightly stresses the bureaucratic and legal obstacles to the emergence of a venture capital industry in India, but we should also ask why this industry has flourished in the United States and is so unimportant, for instance, in Europe and Japan. I would suggest that this is due to two primary factors. First, there are portfolio investors in the United States who are so wealthy that they want to gamble on the fortunes of small, high-risk, highreturn companies. Venture capital investment is akin to horse racing; it has become as much an American industry as horse racing is Australian. There simply is not so much capital belonging to rich Indians who are prepared to make a throw. And second, the United States is peculiar in having venture capitalists who understand the innovation game, who keep innovators on a tight leash, and who ruthlessly back success and abandon failures. Murthy and Raju, who must have hands-on experience of the problem, stress the quality of venture capitalists rather than the supply of venture capital; they note “the hands-on approach, commitment, relationships and risk appetites of some of the leading funds in the Valley.” This quality makes all the difference. Rather than replicate the American venture capital industry in India, it would be simpler to import it. A number of American venture capitalists have already made forays into India. Further, Indian innovators who have made a fortune in the United States are coming increasingly to India to set up and fund start-ups. Investment funds all over the world have an eye on the Indian software industry; as the more reputable firms get bought into and become overvalued, capital will flow to smaller and lesser-known firms. Thus, in my view, the lack of a venture capital industry is not an important handicap, as long as capital mobility is not hindered by the Indian government. The Indian government’s heavy-handed efforts to create a domestic venture capital industry are really an effort to indigenize an industry that already exists

Bangalore: The Silicon Valley of Asia?

209

elsewhere and will readily come to India if it is provided with a conducive legal environment. Education Saxenian notes a bias in Indian IT education: The four Indian Institutes of Education provide first-class education, but they are too few, and most of their graduates emigrate. A large number of private training centers have also emerged in response to the booming demand, but their standards are uneven, and they train only technicians. She proposes the setting up of twoyear and four-year colleges to train mid-level personnel. There are a number of skill levels in the industry. At the top are designers who conceive and plan software projects; they are complemented by translators who convert customers’ requirements into programming projects. Then there are managers who coordinate the work of programmers and implement projects, then the people who actually write the software. In addition, there are people employed by users of software for day-to-day maintenance and to assist other employees who have to use IT without understanding it. The bulk of the demand is for the last category—support workers—and as IT applications in the economy grow, their proportion grows. The estimates of IT workers in the United States vary between 2 million and 10 million; the difference is accounted for by support workers. The demand for support workers is even greater in India, with the universal computer illiteracy of its white-collar workers, and it is this demand that private training centers are largely meeting. There are also thousands of three-year science colleges and five-year engineering colleges attached to universities, many of which are introducing courses in information technology. But their salaries are pitiful, and they are incapable of attracting and retaining competent teachers. Hence, their students go to private training centers for supplementary training and practice. Teachers in the formal university system are contemptuous and jealous of these training institutes, which they think make much money for poor teaching. It is true that their standards are uneven. But they are meeting a felt demand. Their success is proof of this, and they hardly deserve to be penalized for it. If any penalty is to be exacted, uniform open examinations should be developed so that their training can be standardized. But Saxenian’s dissatisfaction is not on the same grounds as that of Indian professors. She would like Indian IT professionals to work in India and solve Indian problems instead of gravitating to Silicon Valley, and is looking for a degree that would train them to work in India but not abroad. Even now, however, the system trains thousands who would neither qualify for a work visa abroad nor meet the skill requirements abroad; they are already working in India. But they are not innovating for India’s needs. The implicit proposition that people can be taught in order to spread out into the coun-

210

AnnaLee Saxenian

tryside and solve people’s problems, whether they are paid for it or not, and whether they have institutional support or not, is naïve. The obstacles to innovations, as discussed above, lie elsewhere, and they are not confined to the IT industry. Hence, I am not convinced that changes in IT education are necessary or would be sufficient to generate indigenous innovative solutions to local problems. Before such solutions can emerge, a domestic market for them, as distinct from a need, is necessary; it can emerge only when telecommunications have spread across the country and the use of the internet has expanded by a few magnitudes. Additionally, competition has to emerge in the technology market, which is currently dominated by a small number of government laboratories that do not have to earn their living and by corporate R&D centers that serve only their parent companies. Murthy and Raju also note the lack of innovativeness amongst Indian software professionals, but their explanation is very different. They think it is due to the fact that they are exposed to computers much later in life than are Americans. It is an intriguing hypothesis, and one that needs to be explored. Conclusion Saxenian’s paper is an excellent first cut into an industry that has suddenly gained prominence and may soon become so important as to affect India macroeconomically; Murthy and Raju provide a first-hand factual backdrop. The rapidity of its growth indicates the urgency of understanding its economics. It already poses a number of conundrums to which we have no answers: For instance, what its market structure is, what the large number of small firms that have recently entered it do, where it gets its manpower, how qualified it is, what it is employed to do, what the substantial domestic sales claimed for it consist of, what the import intensity of its exports is, and so on. I trust that Saxenian’s pathbreaking studies will lead to a more systematic data collection from which we can gain a reliable picture of this industry. Once we do so, informed discussion of alternative visions for it, such as Saxenian has presented, will become possible. References Greico, Joseph M. 1984. Between dependency and autonomy: India’s experience with the international computer industry. Berkeley: University of California Press. Heeks, Richard. 1996. India’s software industry: State policy, liberalisation and industrial development. New Delhi: Sage Publications.

6 Small-Scale Industry Policy in India A Critical Evaluation Rakesh Mohan

6.1 The Existing Framework of Support of Small-Scale Industries in India1 Most countries in the world, both developed and developing, are seen to evolve a set of policy measures and incentives to provide special support for the encouragement of small-scale enterprises. However, India is probably unique in the wide spread of its policies and in its choice of policies to protect and support small-scale enterprises. As this chapter will argue, it is also remarkable how harmful these policies have been for the growth of manufacturing in India over a long period of time. Small-scale enterprises are believed to deserve special support because Rakesh Mohan is currently the adviser to the finance minister and chief economic adviser in the Ministry of Finance, Government of India. At the time of writing this article, he was the director-general of the National Council of Applied Economic Research in New Delhi. The author has drawn liberally in this paper from the works of his colleagues at NCAER and from the Report of the Expert Committee on Small Enterprises submitted to the government of India in 1997. As the author was the member secretary of that committee, it should surprise no one that the views propagated in this chapter are almost identical to those contained in that report. The author would like to specifically acknowledge the help of his colleagues: Rajesh Chadha for sharing his work on exports; Saurabh Bandopadhyaya for further work on export data; K. R. Pandit for his expertise and assistance on the analysis of SSI data; and Praveen Sachdeva for structuring the myriad tables in this chapter. The author also owes a debt of gratitude to Dipak Mazumdar, who provided an analytical structure for the quantitative work embedded in the Expert Committee work. This chapter employs the same structure and extends the analysis. Anne Krueger provided detailed comments on the first draft of this chapter, and her comments helped greatly in sharpening the focus of the chapter and in attempting to make a forest out of the trees. Finally, as always, none of this would have been possible without the devoted assistance of the author’s longtime secretary, Ajay Gupta. Readers will find this chapter quite data intensive and hence difficult to digest; there was no other way to present the evidence at hand! 1. For greater detail on support programs see Abid Hussain (1997).

213

214

Rakesh Mohan

they are hampered in their growth by imperfections in factor markets: capital, land, and labor. Typically, the factor market most focused on is the capital market, distortions in which are seen to especially discriminate against small-scale enterprises (SSEs). Capital costs faced by SSEs are typically higher because of market imperfections in the availability of information for investors and lenders. Transaction costs in bank lending exhibit pronounced economies of scale with respect to loan size. Thus, the unit transaction costs for SSEs are higher than those for large firms. The information costs inherent in appraisals of SSE projects decrease per unit of loan as loan size increases. Provision of collateral or other risk-reducing securities is often difficult for SSEs. Thus, many SSE promotional efforts in most countries focus on different forms of credit support. Similarly, the per unit cost of regulatory hurdles usually present in achieving appropriate access to land also makes the unit land cost higher for SSEs. Urban land markets are typically subject to a plethora of zoning and other regulations. The cost of obtaining all such regulatory approvals may not differ greatly between large and small parcels of land. Finally, in the case of labor, labor market regulations typically distort the price of labor, making it higher for larger firms and lower for small firms in the unorganized sector. Larger firms compensate for such higher wages by using higher capital intensity in production and employing higher productivity labor. Smaller firms, facing a higher cost of capital and lower cost of labor than the optimal ratio, would exhibit overall higher cost of unit output through suboptimal production efficiency. The economic argument would then be that, in the face of factor market distortions, special support policies for small-scale industries would remove the various factor market distortions at their source. In practice, it is difficult to remove such factor market distortions through direct interventions. The result is that a whole plethora of other supportive policies for SSEs emerge. India has differed from other countries in its degree of concern for supporting small-scale industries. In fact, among developing countries, India was the first to display such special concern before it became fashionable to do so. A basic focus of Indian government thinking has been that employment generation is of paramount importance in a labor surplus economy. Small enterprises manufacturing labor-intensive products, it was argued, make economical use of capital and absorb the abundant labor supply that characterizes an underdeveloped economy. The belief has been that large enterprises are capital intensive and reward only a small minority of labor, which is skilled and urban. Indian concern and support for SSEs has focused excessively on small scale industry (SSI) as distinguished from smallscale enterprises in general. This can perhaps be traced back to Mahatma Gandhi’s special concern for handicrafts and village-based industries. In the nineteenth century, there was a widespread perception in India that the

Small-Scale Industry Policy in India: A Critical Evaluation

215

import of mass-manufactured products had affected millions of handloom textile workers and other craftspeople, and this experience also contributed to the special concern for protecting SSIs. It is therefore ironic, as is documented in this chapter, that a particular failure in economic policy in India has been the sluggish growth in manufacturing employment. Moreover, the growth in value added in SSI has also been consistently lower than that of large-scale industry. It is also argued that small-scale policies have contributed to the slow growth of Indian manufactured exports. These outcomes are perhaps the result of the specific policies that have been practiced in India, which are more protective and regulatory then promotional. The various measures used for the development of SSI have included product reservations, fiscal concessions, preferential allocation of credit and interest subsidy in a credit-rationing framework, extension of business and technical services by the government, and preferential procurement by the government. Thus reservations as well as fiscal concessions have sought to protect SSI from the competition of large companies. Few other countries have put in place such a wide array of support instruments. Most countries, including developed countries, have different kinds of credit support mechanisms for SSI. Capital market distortions are widely recognized and attempts made to compensate for them, particularly to assist entrepreneurs in making entry into manufacturing and other activities. What sets Indian policies apart is the exclusive concern with SSI and its unique policy of small-scale reservations, which is seldom found in other countries. The eligibility of SSI firms to take advantage of the various incentives offered is dependent on the definition of SSI used. Whereas most countries define SSEs in terms of employment levels, usually taken to be firms that employ more than 5 but less than 50 or 100 workers, the Indian definition has been based largely on cumulative amount of investment in plant and machinery.2 Consequently, with inflation, these limits need to be periodically adjusted upward (see table 6.1 for the evolution of these limits since 1950). However, a curious development took place in 1999 when the limit was revised downward. This took place due to political pressure from the smaller SSI firms, 2. This is not the only definition used. The Factories Act defines a factory as one which employs ten workers or more if the unit uses power, or twenty workers or more if it does not use power. All such units have to be registered under the Factories Act and are subject to various labor laws, including the provision of medical insurance and some social security. This is known as the registered sector. Within this definition, those units that employ more than 50 workers (if using power) or more than 100 (if not using power) must register themselves with the state governments in order to operate. The main source of data for manufacturing is the Annual Survey of Industries (ASI) (see appendix A). The coverage of this survey is limited to those factories registered under the Factories Act. The third definition is that used for giving fiscal concessions. At present, units that have less than Rs. 10 million turnover are fully exempt from excise taxes, and there is a sliding scale of concessions available for small scale enterprises with a turnover of up to Rs. 30 million.

216

Rakesh Mohan

Table 6.1

Evolution of Investment Limits for Small-Scale Industries

Year

Investment Limits

Additional Condition

1950

Up to Rs. 0.5 million in fixed assets

1960 1966 1975 1980 1985 1991 1997 1999

Up to Rs. 0.5 million in fixed assets Up to Rs. 0.75 million in plant and machinery Up to Rs. 1 million in plant and machinery Up to Rs. 2 million in plant and machinery Up to Rs. 3.5 million in plant and machinery Up to Rs. 6 million in plant and machinery Up to Rs. 30 million in plant and machinery Up to Rs. 10 million in plant and machinery

Table 6.2

Less than 50/100 persons with or without power. No condition No condition No condition No condition No condition No condition No condition No condition

Investment Ceilings for Small-Scale Industries (December 1999)

Type of SmallScale Industry

Investment Limit

Small-scale industry Ancillary

Rs. 10 million Rs. 10 million

Export oriented Tiny enterprise Service and business enterprise Women enterprise

Rs. 10 million Rs. 2.5 million Rs. 0.5 million

Historical cost of plant and machinery At least 50 percent of its output should go to other industrial undertakings Obligation to export 30 percent of production No location limits No location limit

Rs. 10 million

51 percent equity holding by women

Remarks

Notes: Small-scale industry cannot be owned or controlled by (or be a subsidiary of) another industrial undertaking. The policy framework for all segments is the same except for some incentives. The limits have been periodically revised upward, but in December 1999, it has been revised downward from the Rs. 30 million set in December 1997.

which feared greater competitive pressure as the larger among them modernized and expanded after the substantial increase in investment limit decreased in 1997. Thus, the real value of the current limit is lower than it was in 1991 (see table 6.2 for details of current definition). These definitional investment limits are of great importance to SSI in India since, as a promotional and protective measure, the government of India has “reserved” a large number of industries for exclusive production in the small-scale sector. Only those firms that have cumulative investment lower than the limit are allowed to produce items covered in the reserved list. 6.1.1 Fiscal Incentives The fiscal incentives are of relatively recent vintage, having started in 1986. SSI units as defined by the investment limits and subject to specified turnover limits are either totally exempt from paying excise duties on their products or permitted to pay lower excise duties on the same items than do

Small-Scale Industry Policy in India: A Critical Evaluation

217

large industries. Such a fiscal regime provides further incentives to SSI units not to grow above the thresholds in order to continue benefiting from these tax concessions. It distorts the whole fiscal structure faced by large manufacturing units, also thereby affecting the industrial structure as a whole. This added complexity in administration of the indirect tax regime has also acted as a further hurdle in bringing about indirect tax reform in the movement towards a full scale value added tax (VAT) system. 6.1.2 Credit Support Until the economic reforms in the 1990s, bank credit in India channeled through the nationalized commercial banking system was governed by a credit allocation system in a credit rationing framework. Interest rates were fixed and amounts allocated for different purposes by the Reserve Bank of India under instructions from the government of India. In this system, SSIs were given preferential treatment through the provision of lower interest rates as well as the requirement for a minimum credit allocation from each commercial bank. In this system, 40 percent of all bank credit was to be allocated to the “priority” sector, which includes SSI and other areas such as agriculture and exports, with a minimum of at least 15 percent to be allocated to SSI. The evidence suggests that banks seem to lend to SSI only because of the administrative requirements. This has also militated against the development of specialized appraisal skills within banks for lending to SSI firms. The government has also run a system of state government-owned term lending institutions present in every state of the country. These institutions received concessional refinance and therefore lent at subsidized interest rates. The irresponsibility in lending practiced by these institutions is visible in their recovery record: an average of 37 percent over the past thirty to forty years. The initiation of financial sector reforms in the 1990s has resulted in better supervision of banks, enforcement of prudential guidelines, interest rate deregulation, and substantial (if incomplete) elimination of interest rate subsidies. This has put credit support to the SSI units under great pressure: Most loans are now given at commercial interest rates. With higher unit transaction costs and the absence of good credit assessment capabilities engendered by the earlier regime, the consequence has been a significant jump in interest rates applicable to SSI units. Thus, the Indian attempt at correcting for capital market distortions through administrative means has met with perverse results, and a whole new approach is called for. 6.1.3 Promotion Program The central and state governments have each put in place a plethora of SSI support programs. In the central government, 409 manufactured products are reserved for their procurement program. Furthermore, 15 percent

218

Rakesh Mohan

price preference is given to SSI products in procurement tenders. Each state government also has its own program of product reservations for procurement purposes and price preferences. This is another support program for SSI that provides incentives for small-scale units to remain within the smallscale definition. Growth beyond small-scale limits would render them ineligible to benefit from such preferential treatment. The central government directly operates a remarkably large system for assisting SSIs in various business and technical aspects throughout the country: tool rooms, product-cum-process development centers, small industry service institutes, and the like. Whereas it may have been necessary in the 1950s to set up such a government-run system to provide what are essentially business support services, it is now difficult to understand the rationale of running such a system under the supervision of the central government in a country as large and diverse as India. That this program is very thinly spread can be gauged from the fact that the total budget of the Department of Small Scale and Agro Related Industries (SSARI) is in the region of Rs. 2000 to 2500 million (about US$50 to 60 million). Various studies attempting to assess the efficacy of these institutions have found that only a small proportion of SSI enterprises are even aware of their existence (NCAER 1993). Given the relatively small amount of resources devoted to the support of these institutions, this is not surprising. Clearly, the availability of business support services, technical consultancy agencies, and the like is now such that there is little rationale for the central government to be operating such business services directly. As might be expected, state governments operate further programs for the support of small-scale industries in their respective states. Most states have state industrial development corporations (SIDCs), which are often supplemented by small-scale industries development corporations. These agencies provide support for setting up government-run industrial estates that provide infrastructure and industrial plots, often at concessional rates for SSI units. In principle, such programs are designed to remove the land market distortions faced by the SSI firms. Each state also runs district industry centers in each district, which are supposed to provide technical and business support to SSI units in their areas of operation. Most states have complex programs for providing different kinds of subsidies for SSI units. These include subsidies on power consumption, capital subsidies, exemption from sales tax, subsidies for location in backward areas, and subsidies for technical and feasibility studies for SSI units. As can be imagined, administration of such a large number of schemes providing for tax concessions and cash subsidies requires a complex administration mechanism. The consequence is that the district industry centers operate mainly as government offices for the purpose of registration of SSI units for different schemes. This system also lets loose an army of government inspectors who have to verify the antecedents of each of the SSI units wanting any of these

Small-Scale Industry Policy in India: A Critical Evaluation

219

incentives. Paradoxically, a key complaint of SSI entrepreneurs is the harassment caused to them by such inspectors. 6.1.4 Small-Scale Reservation Policy The policy of small-scale reservations was initiated in 1967 as a promotional and protective measure for the small-scale sector vis-à-vis the large scale sector. Under this policy, selected products are identified for exclusive production in the small-scale sector. The overwhelming considerations for reservation are whether it is technically feasible to produce that item in the small-scale sector, whether the manufacturing process is of a simple nature (i.e. is essentially labor-intensive), and whether the small-scale units can meet the requirements of consumers, both in terms of quantity and quality. The rationale for reservation was based on the advantages of the small-scale sector, like labor intensity and adaptability to a semi-urban and rural environment. Another objective was to make SSI products competitive with those of the large scale by offsetting the disadvantage of mass scale production, economies of scale, wider marketing network, better credit availability, and publicity through mass media and advertisements. The basic features of reservation policy are as follows. 1. The policy is applicable only to the manufacturing sector. It does not take into account the service sector, including product repair. 2. No new unit in the medium- or large-scale sector is allowed to be set up after the date of reservation, nor is any further capacity expansion in the existing medium- or large-scale units permitted. All further expansion or capacity creation is reserved for the small-scale sector only. 3. Existing large scale units that were manufacturing these reserved items at the time of reservation were allowed to continue their activities indefinitely, but their capacity was frozen at the existing levels—that is, they were prohibited from expanding further. 4. Creation of new capacity in the reserved areas is permitted among medium- or large-scale units if they undertake to export a minimum of 75 percent of their production (50 percent in the case of ready-made garments). 5. There is no restriction on the marketing by large units or big companies of products reserved for manufacture in the SSI sector. 6. A statutory Advisory Committee on Reservation was established to undertake the review of items from time to time for de-reservation of items which are already reserved, reservation of new or additional items, and change of the nomenclature of the items. A curious feature of this reservation policy, which was introduced in 1967, was that it had no legal backing until 1984. It is a remarkable comment on the heavily regulated economy of the time that a draconian regulation such as this was not legally challenged. It was only in 1984 that gov-

220

Rakesh Mohan

Table 6.3

Date of Notification Phase I 1 April 1967 19 Feb 1970 24 Feb 1971 11 Nov 1971 26 Feb 1974 5 June 1976 26 April 1978 Phase II 26 April 1978 30 Dec 1978 12 May 1980 19 Feb 1981 3 Aug 1981 23 Dec 1981 14 Oct 1982 19 Oct 1982 3 Sept 1983 18 Oct 1984 30 May 1984 30 Oct 1986 13 Feb 1987 20 July 1987 18 Mar 1988 3 Mar 1989 31 July 1989

Progressive Reservation of Items for Exclusive Manufacture in the Small-Scale Sector No. of Items Reserved

No. of Items Dereserved

47 8 73 4 53 3 324 807a 1 27 1 9 2 9 35 1 7 1

3 1

1 13 3

1 14b 7 13 3 1 14

Cumulative Net No. of Items Reserved

47 55 128 124 177 180 504 807 806 833 833 842 831 282 837 872 873 869 863 850 847 846 835 836

Source: Expert Committee on Small-Scale Enterprises. a In 1978 it was decided to recast the reserved list by following codes adopted in the NIC; in this process, the list of reserved items expanded from 504 to 807 items. b Since it included three subitems, the effective number comes to only 11.

ernment plugged this legal inadequacy and put this policy on a statutory footing in the IDR Act. Initially, only forty-seven items were reserved, but the number has grown considerably since 1967 (see table 6.3). The rationale used for the selection of items to be reserved is not available in any official documents. The only criterion mentioned in such documents is the ability of the small-scale sector to manufacture such an item. This is perhaps not surprising, since it would actually be difficult to find standards with which to determine the optimal capital labor ratio for the production of different items through the examination of the whole range of production techniques that can be used to produce any item. Hence, the choice of products for reservation was necessarily arbitrary. Out of almost 100 three-digit

Small-Scale Industry Policy in India: A Critical Evaluation

221

categories of industrial production, 80 percent of the reserved items are concentrated in only eleven three-digit National Industrial Classification (NIC) categories.3 Among these categories are included a large number of products that are labor-intensive and in which India would characteristically have comparative advantage. Such products include all kinds of clothing, knitted textiles, shoes and leather products, most sporting goods, toys, stationery, office products, furniture, simple electrical appliances, simple extruded plastic products, and the like. As will become evident in the next section, this policy has probably constrainedthe growth of labor-using production of such goods and has therefore stunted the growth of manufacturing employment in India. Similarly, as will also be shown, this policy has had a large effect on the ability of Indian manufactured exports to grow as quickly as demonstrated by East Asian countries. As mentioned, large-scale industries that were producing reserved items at the time of reservation were permitted to continue producing those items after reservation. This policy, though perhaps unavoidable, provided unintended great protection to these existing units from the possibility of increased competition from new large units. Studies such as Guhathakurta’s (1993), on office furniture, have shown that large-scale units protected in this fashion continue to dominate the market, whereas the small-scale units served the lower end of the market. A particular example is that of toothpaste, which was also reserved. Thus Colgate has had an effective monopoly in the market, resulting in healthy profit growth on a continuous basis. SSI reservation policy has been set without reorganizing the significance of quality differentials within the product groups. Most items of production can be produced at differing levels of quality and with different sets of attributes. Each kind of product can be produced through different techniques and with different sizes of plant. For example, high quality is often achieved through labor-intensive hand crafting techniques (as in the case of carpets), whereas in other cases capital-intensive techniques yield higher quality and higher consistency in quality. Failure to take this into account has had negative impacts on the quality of outputs in a number of sectors. Another key issue has to do with technical upgradation and expansion of production. A successful small-scale enterprise would expect to continue expanding production to attain higher economies of scale and also to upgrade production technology to achieve higher quality and higher unit value levels. This opportunity has also been denied to Indian producers producing reserved items, thus stunting entrepreneurial growth, manufactur3. These are: knitting in mills (260); manufacture of plastic products (303); manufacture of basic and industrial organic and inorganic chemicals (310); paints, varnishes and lacquers (312); photochemicals and sensitized fibers (319); fabricated metal products, metal boxes, cans, safes, and vaults (340); hand tools and general hardware (343); electrical appliances, domestic appliances, switches, and sockets (363); auto parts (374); bicycles, rickshaws, and parts (376); mathematical and miscellaneous instruments (380).

222

Rakesh Mohan

ing production, and employment growth. This particular handicap has probably affected the growth of Indian exports the most. Expansion in exports typically requires the accumulation of experience in the home market before expansion into the export market. Success in one leads to the other. This, however, cannot take place if production is limited by capacity constraints caused by SSI reservations. It can be argued that policy allows export production at higher scales with the commitment of a minimum of 75 percent of production for exports. But this is difficult to achieve without the availability of the home market to accumulate experience. It is also too risky to commit export levels in excess of 75 percent of production. Furthermore, the reservation policy disallows equity of greater than 24 percent in SSEs producing reserved items either from large domestic industry or through foreign direct investment. Thus another source of financial support, technology support, and marketing support is lost to Indian SSEs. In view of these considerations, it is difficult to understand the continued persistence of this policy of SSI reservation. Its deleterious effect on Indian industry and on economic growth, as documented in this chapter, should be obvious to all. Furthermore, in recent years, with the opening of Indian trade, almost 75 percent of all reserved items are now already importable with the removal of quantitative restrictions (QRs.) in the last few years. India has also committed to remove the remaining QRs. by April 2001. We therefore now have a curious situation, in that the reserved items can be produced by large foreign enterprises and imported into India, whereas Indian large enterprises are not allowed to produce the same items! Even this change in the external environment has so far not persuaded the authorities to change this policy of SSI reservation. 6.1.5 Summary This brief review of the framework of Indian policies protecting and supporting SSI has illustrated the broad scope of such policies. It has also shown how these policies and programs are thinly spread, thereby leading to relative ineffectiveness. Many of the policies are such that they discourage growth of small-scale units into larger ones and have a stunting effect on manufacturing employment and output growth. I have placed particular emphasis on the peculiar policy of product reservation for production by SSI units, arguing that this policy has militated against labor-intensive growth in manufacturing, particularly of exports. The next section examines the performance of the Indian manufacturing sector and of SSIs in particular to substantiate this central proposition of this chapter. Some of the policies reviewed may have been useful in the earlier stage of Indian industrialization and in the context of a highly controlled and closed economy. With all the changes in economic environment that have taken place in the 1990s, the indication is that future policies for the promotion of SSIs must be more growth- and outward-oriented and more general, rather

Small-Scale Industry Policy in India: A Critical Evaluation

223

than sector-oriented. It would be more useful if such policies were designed to promote entrepreneurial entry, growth of enterprises, technology upgradation, and labor productivity in a pervasive manner regardless of specific sectors. Specific suggestions on the content of a new approach are provided in the concluding section. 6.2 Performance of the Small-Scale Manufacturing Sector Evaluation of the effectiveness of policies concerned with SSIs in India requires a quantitative assessment of their performance. In this section I have pieced together data from different sources to document and evaluate the performance of the small-scale sector and the Indian manufacturing sector as a whole in terms of three key aspects of performance. Since the key rationale behind SSI policies in India has been the promotion of higher growth in manufacturing employment, the evaluation of such growth forms a basis for assessing the effectiveness of SSI policy. Second, the record of SSI growth in value-added compared with that of large-scale industries provides further evidence of policy effectiveness. An assessment of the performance of the reserved items within SSI is of particular interest here. Because India is an industrializing country, one would expect the growth in Indian manufacture exports to be fueled by high growth of exports of labor-intensive products. To the extent that a large proportion of SSI reserved products consist of typically labor-intensive products, the third key component of SSI performance lies in an evaluation of growth in Indian manufacturing exports. As will be evident from the analysis put forward, a great deal of difficulty is encountered in conducting such an assessment because of the lack of reliable data indicating the real size of SSIs in India and their trends in growth over time.4 Different sources of data have been used and a number of inconsistencies are found across the different data sources. An additional problem that emerges from the analysis is that the zeal of the administrative authorities seeking to promote SSIs may have itself contributed to the low quality of data available in the system. This is typical of bureaucratic behavior of any governmental system. The fight for the allocation of scarce budgetary resources and empire building inevitably results in agencies’ tending to exaggerate the importance of their function. Despite the data limitations and inconsistencies encountered, the conclusions reached are robust enough to be credible. 6.2.1 Employment Growth in Small-Scale Sector Manufacturing The most remarkable feature of the record of manufacturing employment in India is that its share does not appear to have increased over the 4. See appendix A for a description of the Indian Statistical System as related to manufacturing as a whole and SSI in particular.

224

Rakesh Mohan

thirty years from 1961 to 1991, when the last census was conducted (see table 6.4).5 It was only during the 1970s that there seems to have been an employment surge. This is rather unusual for a country in its industrializing stage. Total manufacturing employment comprises just a little over 10 percent of all employment in the country, and it has remained stagnant at this level. Manufacturing employment in China now accounts for about 16 percent of total employment. Within manufacturing the structure has changed significantly, with the share of household industry reducing consistently and significantly over the whole period, as would be expected in a modernizing economy. The share of nonhousehold industry has increased significantly over the same period, as should be expected. What is most notable, however, is the slowdown in employment growth in this sector during the 1980s, and reduction in share in total employment. This record can be compared with that of East Asian countries in the 1980s and 1990s (see table 6.5). Significant growth took place in manufacturing employment in each country up to a certain stage. Most annual employment growth rates were in excess of 5 percent. Even in China, annual growth in manufacturing employment exceeded 4 percent in the 1980s. Among the more industrialized countries in this group, the share of manufacturing employment has increased to over 20 percent, whereas in the less industrialized countries like Thailand and Indonesia it is in the range of 12 to 14 percent. No other country shown has had manufacturing employment growth rates as low as those exhibited by India in the last thirty to forty years. It does seem, however, that after initial rapid growth in laborusing manufacturing activity, there is a plateauing out of such growth. This probably happens as a result of productivity enhancements leading to wage growth, which in turn gives rise to technology upgradation and greater capital intensity in choice of technique. What is interesting about the Indian record is that there seems to have been some acceleration in manufacturing employment growth in the 1970s and a severe slowing down in the 1980s. The plateauing seems to have taken place rather early at low wage levels. On the policy front, two measures put 5. Different data sources classify the manufacturing sector in different ways. The decennial population censuses of the country provide information on employment characteristics of the labor force. It divides manufacturing employment into “household enterprises” and “nonhousehold enterprises.” The former category covers “those establishments which carry out their operations from their own residence.” Nonhousehold industries cover all other employment in the manufacturing sector, including that in the “Registered Sector.” The main labor legislation concerned with the protection of workers in the organized sector, the Factories Act, defines factories as those establishments employing more than ten workers when using power, or more than twenty workers when not using power. It therefore includes both large-scale industry and small-scale industry. Whereas there is no internationally accepted definition of small scale industries, a definition commonly used covers manufacturing enterprises employing 5 or more workers, but fewer than 50 or 100. Thus, to obtain a comparable estimate for India involves the use of different data sources.

Cultivators Agricultural Labourers Livestock, etc. Mining and quarry Total manufacturing Household industries Other Construction Trade and Communication Transport, etc. Other services Total (continued)

–1.57 –6.19 3.02 0.78 2.77 3.90 –2.12 –0.43

–2.37 4.20 –1.87

1.69 1.57 1.51 3.19 3.95 1.96 4.99 4.85 3.33 3.26 2.16 2.12

180,486

314,604

143,668

164,019

199,027

B. Total employment growth rates (in thousands) 1.81 0.36 9.38 1.19 1.32 3.00 2.92 5.24 0.92 2.89 1.92 0.51 –1.29 1.70 1.26 3.32 3.26 3.39 1.32 1.21 0.22 3.75 1.10 –1.25 –1.88 –3.76 1.18 –2.13 2.29 3.43 3.23 4.88 2.06 4.51 3.36 1.07 4.75 4.80 4.34 3.48 3.35 3.21 4.32 2.82 3.33 3.78 3.32 2.85 4.14 1.36 –1.14 1.91 3.90 2.54 1.40 1.46 1.76 2.24

222,520

1.68 –1.60 3.38 3.52 3.63 3.31 1.54 1.82

0.96 3.32 0.55

221,659

–8.70 –11.79 0.92 –1.69 –3.73 8.43 –6.49 –6.19

–11.92 1.09 –4.08

64,856

4.84 2.78 0.62 2.77 5.25 4.49 6.33 5.79 5.11 1.53 3.58 3.68

33,535

188,416

1991

Total

1981

9,304 15,795 783 124 2,196 1,331 865 204 557 146 2,230

1971

A. Main workers classified by industrial category for all India, 1961–91 (in thousands) 99,510 78,267 92,523 110,702 66,487 68,963 77,591 88,481 33,103 31,482 47,493 55,500 74,598 17,312 31,698 34,732 46,165 14,170 5,190 4,297 4,993 6,041 4,003 3,514 4,160 4,716 1,187 923 1,264 1,752 799 1,101 1,537 — 19,988 17,068 25,145 28,671 14,553 14,872 21,482 23,969 5,455 12,031 6,352 7,713 6,804 7,385 5,021 5,648 4,555 4,666 7,957 10,716 17,432 21,867 7,168 9,851 15,834 19,414 789 2,055 2,220 3,565 5,543 1,813 2,016 3,207 5,123 242 7,640 10,042 13,930 21,296 6,825 9,485 13,013 19,863 815 3,003 4,403 6,069 8,018 2,938 4,257 5,899 7,810 65 19,548 15,773 19,531 29,312 15,184 13,543 16,360 23,995 4,364

1961

Cultivators Agricultural Laborers Livestock, etc. Mining and quarry Total manufacturing Household industries Other Construction Trade and Communication Transport, etc. Other services

1991

1971

1981

4.06 3.19 4.75 2.76 2.53 0.86 4.38 1.63 4.57 2.04 5.30 3.63

48,638

14,932 20,768 833 163 3,663 2,065 1,598 358 917 170 3,171

1981

Females 1961

1971

Males

1961

Persons

Employment Classifications, Growth Rate, and Structure, 1961–91

Category

Table 6.4

–0.49 –2.40 3.85 1.86 1.90 3.95 0.66 0.26

–1.32 2.35 0.37

64,274

22,221 28,433 1,325 214 4,702 2,249 2,453 421 1,434 208 5,316

1991

52.81 16.71 2.75 0.00 10.61 6.39 4.22 1.09 4.05 1.59 10.37

100.00

Cultivators Agricultural Labourers Livestock, etc. Mining and Quarry Total Manufacturing Household industries Other Construction Trade and Communication Transport, etc. Other services

Total

Source: Census of India.

1961

(continued)

Category

Table 6.4

100.00

43.36 26.31 2.38 0.51 9.46 3.52 5.94 1.23 5.56 2.44 8.74

1971

100.00

100.00

100.00

100.00

100.00

100.00

39.92 20.83 2.13 0.69 10.81 2.05 8.76 2.31 8.96 3.52 10.83

1971

C. Structure of total employment (in percentages) 41.58 38.72 51.47 46.24 43.70 24.94 26.09 13.41 21.25 19.56 2.24 2.11 3.10 2.36 2.34 0.57 0.61 0.00 0.54 0.62 11.30 10.03 11.27 9.97 12.10 3.47 2.38 5.71 3.37 3.18 7.83 7.65 5.56 6.60 8.92 1.60 1.94 1.41 1.35 1.81 6.26 7.45 5.29 6.36 7.33 2.73 2.8 2.28 2.85 3.32 8.78 10.25 11.76 9.08 9.21

1961

1991

1991

Males 1981

1981

Persons

100.00

55.73 23.85 2.00 0.00 9.18 7.86 1.33 0.41 1.37 0.11 7.35

1961

100.00

29.69 50.40 2.50 0.40 7.01 4.25 2.76 0.65 1.78 0.47 7.12

1971

100.00

33.20 46.18 1.85 0.36 8.14 4.59 3.55 0.80 2.04 0.38 7.05

1981

Females

100.00

34.57 44.24 2.06 0.33 7.32 3.5 3.82 0.66 2.23 0.32 8.27

1991

Small-Scale Industry Policy in India: A Critical Evaluation Table 6.5

227

Growth in Employment in the Manufacturing Sector in Selected Countries (in millions)

Year and Labor Force 1980 Total employed Employed in manufacturing sector Percentage in manufacturing 1990 Total employed Employed in manufacturing sector Percentage in manufacturing 1997 Total employeda Employed in manufacturing sector Percentage in manufacturing

China

Indonesia

Korea

Malaysia

Taipei, China

Thailand

423.61 67.14 15.85

51.55 4.68 9.08

13.68 2.96 21.64

4.84 0.75 15.50

6.55 2.15 32.82

22.52 1.79 7.95

639.09 96.98 15.17

75.85 7.69 10.14

18.09 4.91 27.14

6.69 1.33 19.88

8.28 2.65 32.00

30.84 3.13 10.15

688.50 109.38 15.89

87.05 11.22 12.89

21.05 4.47 21.24

8.39 2.31 27.53

9.18 2.57 28.00

32.23 4.33 13.43

4.17 2.03 3.31

5.67 6.50 5.62

5.78 –1.55 2.61

6.57 9.64 7.28

2.35 –0.51 1.12

6.41 5.56 6.07

Manufacturing employment growth rates (%) 1980–90 1990–97a 1980–97 Source: Asian Development Bank (1998). a 1996 for China and Thailand.

into effect in the late 1970s could have contributed to this slowdown: major expansion in the limit of SSI reservations in 1977, and amendments to the Industrial Disputes Act in 1975 and 1980 making it impossible to fire organized sector industrial workers without government permission. The first measure has already been discussed in some detail. It restrained the rapid growth of labor-using factory employment in the industries in which India possesses comparative advantage. The second measure has been equally influential in slowing down Indian manufacturing employment growth. It is virtually impossible for an Indian employer to retrench workers as a consequence of business downturns, or for any other reason. The Industrial Disputes Act requires the enterprise to obtain government permission for so doing, and this is seldom granted. Labor cost in industry has therefore become like a fixed cost, thus providing every incentive to Indian industry not to use labor in this labor-abundant country (see Fallon 1986). At the same time, these overall data also contribute to the understanding of the concerns of the Indian government related to manufacturing employment and its espousal of SSIs as employment generators. The number of household industry workers almost halved between 1961 and 1991, from over 12 million to just under 7 million. As would be expected, most of this loss was in rural areas, hence the concern for “village industries” and smallscale industries. This is to be expected in a modernizing economy. However,

228

Rakesh Mohan

the aim should have been to generate higher productivity employment as these low productivity jobs fall, rather than to preserve low productivity jobs, as Indian policy has attempted to do. In obtaining a fix on the size of the SSI sector, there are other data that can also be utilized. The Central Statistical Organisation (CSO) conducts an economic census through a survey of nonagricultural establishments in the unorganized sector every five years. I examine data related to manufacturing enterprises. Three economic censuses are deemed reliable enough to use.6 These were conducted in 1984–85, in 1989–90, and in 1994–95. The establishments are classified into three groups: 1. Directory Manufacturing Establishments (DME) (six or more workers) 2. Non-Directory Manufacturing Establishments (NDME) (one to five workers) 3. Own Account Enterprises (OAE) (no hired workers)7 The statistics provided by the economic census add to our information because it is possible to classify the small-scale sector by further size categories, in particular to distinguish the small sector (six or more workers) from the tiny sector, consisting of the NDME (one to five workers) and the own account establishments. The statistics provided by the more recent surveys of DME and NDME & OAE8 are reproduced in table 6.6. The trends revealed by the data suggest that there were some differences in coverage between these censuses. Nonetheless they corroborate the basic thrust revealed by the population census data: that there has been significant decline in the most unorganized 6. The first economic census, conducted in 1978–79, is not regarded as reliable. 7. Unfortunately, the OAE sector does not coincide with the household sector as conceived of in the population census, although there is considerable overlap between the two. The population census considers the household sector to be those establishments that carry out their operations from their own residences. 8. The absolute numbers employed in manufacturing are not comparable between the population and economic censuses. There are two conceptual reasons for this, apart from errors in recording. First, considerable uncertainty and volatility exist in the population census because of the problems in identifying “main” and “subsidiary” workers employed in manufacturing. This is why in the figures given in table 6.4 only the census figure of main or principal workers is used. Here also, the coverage of female workers has varied across the censuses because of varying definition. It is not known if the economic census confronted the problem of distinguishing main and subsidiary workers, and if so how they solved it. At least, on this point, it is reassuring to find that the economic census figure of the total employed in the nonfactory sector is considerably higher than the figure given in table 6.4: 40.25 million in the 1989–90 CSO survey results as against 28.67 million in the 1991 Census of Population, suggesting that the former did include some secondary workers. A second reason that the absolute figures of employment in manufacturing might differ in the two censuses is that the boundary between manufacturing and repair/services is very hazy in the SSI sector, and particularly in the household sector. In any event, more interest attaches to the proportions of employment in the different sectors as recorded in the two censuses.

Small-Scale Industry Policy in India: A Critical Evaluation Table 6.6

229

Manufacturing Employment in Different Segments

Sector

1984–85

1989–90

1994–95

In millions I Factory sector (of which SSI) II DME (6 + workers) III NDME (1–5 workers) IV OAE (own account)

7.866 (2.669) 4.682 4.682 29.590

8.130 (3.059) 5.836 2.576 23.707

9.116 (3.392) 5.639 5.003 22.642

Total Sectors I
II
III

46.820 17.230

40.250 16.542

42.200 19.758

In percentages I Factory sector (of which SSI) II DME III NME (1–5 workers) IV OAE (own account)

16.8 (5.7) 10.0 10.0 63.2

20.2 (7.6) 14.5 6.4 58.9

21.5 (8.0) 13.3 11.8 53.4

Total Sectors I
II
III Total in number (in millions)

100.0 36.8

100.0 41.1

100.0 46.6

46.8

40.3

42.4

Sources: ASI, 1984–85 and 1985–90 Summary Results for Factory Sector (CSO); Directory Manufacturing Establishments Survey Results of 1994–85, 1989–90 (CSO); Non-Directory Manufacturing Establishments Survey Results of 1984–85 and 1989–90 (NISSO). Notes: DME  Directory Manufacturing Establishments. NME  Non-Directory Manufacturing Establishments. OAE  Own Account Enterprises.

portion of manufacturing employment and that growth in larger or more modern manufacturing employment has been slow. In terms of size, comparable with international definitions, the modern small-scale sector may be taken to mean the DME
SSI portion of the factory sector. The share of employment in these segments has varied between 15.7 percent and 22.1 percent of total manufacturing employment between 1984–85 and 1994–95. Very broadly, it may be seen to account for about 20 percent of total manufacturing employment or about half of “modern” manufacturing employment. Clearly, it is a segment of some importance to Indian manufacturing industry. These data draw attention to the small size of factory employment in India relative to that in other countries. As documented by Adrian Wood (2000), almost half of all Chinese manufacturing employment is in factories with an average size of employment of about 100 workers. With almost 112 million workers (11 percent of total employment) in manufacturing, large factories in China therefore employ about 55–60 million workers. In contrast, Indian factories employ less than 10 million workers (including SSI), which comprise less than 30 percent of all manufacturing workers (includ-

230

Rakesh Mohan

ing SSI). It is difficult to avoid the conclusion that there has been something significantly wrong in Indian policies.9 Gary Fields (1985) has provided carefully researched evidence on the structural changes in employment that took place in the 1960s and 1970s in the “Asian Tigers.” What is most noteworthy is the rapidly rising share of manufacturing employment in all these countries. Manufacturing employment in the Republic of Korea increased from about 8 percent in the early 1960s to about 22 percent by the late 1970s. Similarly, in Taiwan this proportion increased from about 16 percent in the early 1960s to about 32 percent. The annual rate of manufacturing employment growth in Korea was about 11 percent throughout the 1960s and 1970s, and that in Taiwan was about 8.5 percent. Labor-intensive sectors like textiles, garments, leather and footwear, wood and furniture, and metal products either increased their share in manufacturing employment or retained their original levels. This demonstrates that labor-intensive manufacturing growth can indeed take place with the right mix of policies. It is also useful to examine the sectoral pattern of evolution of manufacturing employment (see tables 6.7 and 6.8). Only data for males are reported for the sake of simplicity, and also because of better consistency of definition. Employment data for female workers are strictly not comparable over censuses because of definitional changes. For household industries as a whole, there has been a fall in employment across almost all sectors in rural areas. There seems to have been some increase in a few sectors in the urban areas, though the increase was more marked until 1981. In nonhousehold industries, the rate of growth has been marginally higher in rural areas. The key findings are: 1. Manufacturing employment growth clearly slowed down between 1981 and 1991. 2. The sectors that had consistently high employment growth were food products; wool, silk, and synthetics; wood products; rubber, plastics, and petroleum products; metal products; electrical machinery; and other manufacturing. 3. Household industry employment fell consistently. The conclusion is inescapable. The wholesale reservation of small scale industry products in the 1970s did not promote employment growth in manufacturing, nor did it arrest the decline of employment in household industries. Although it is not possible to adduce causation, the growth in manufacturing employment certainly slowed down in the 1980s—particularly the period after expansions of SSI reservations and tightening of labor laws. 9. In making comparisons with China, some allowance must be made for excessive employment in its public-sector enterprises.

Total manufacturing (continued)

2.16 1.11 9.47 17.69 0.22 8.84 0.03 0.42 9.49 7.18 n.a. 0.05 0.03 0.81

6.63 n.a.

127 65 555 1,037 13 519 2 25 556 421

3 2

47

389

100.00

5.43 21.42

319 1,256

5,864

9.01

%

528

No. (in thousands)

1961

5,021

392 121

11

60 3

97 19 681 912 15 301 6 12 508 11 335

335 879

325

No. (in thousands)

1971

7.8 2.4

0.23

1.19 0.06

1.94 0.38 13.56 18.16 0.3 5.99 0.12 0.24 10.11 0.22 6.66

6.68 17.5

6.47

%

100.00

Employment in Household Industry, 1961–91

All India, males only 20–21 Food products 22 Beverage and tobacco 23 Cotton textiles 24 Wool, silk, and synthetic 25 Jute textiles 26 Textile products 27 Wood products 28 Paper 29 Leather and fur 30 Rubber, plastics 31 Other chemicals 32 Nonmetallic miner 33 Basic metals 34 Metal products 35 Nonelectric machinery 36 Electric machinery 37 Transport equipment 38 Other manufacturing 39, 97 Repair

NIC Codes and Divisions

Table 6.7

5,647

288 693

11

50 5

98 13 702 952 22 197 11 21 509 12 324

401 999

341

No. (in thousands)

1981

100.00

5.1 12.26

0.2

0.88 0.09

1.73 0.24 12.42 16.86 0.4 3.49 0.19 0.37 9.01 0.21 5.74

7.1 17.68

6.03

%

4,553

1,006 112

7

51 4

155 23 262 788 18 122 17 13 402 23 211

346 742

251

No. (in thousands)

1991

100.00

22.10 2.46

0.15

1.12 0.09

3.40 0.51 5.75 17.31 0.40 2.68 0.37 0.29 8.83 0.51 4.63

7.60 16.30

5.51

%

–1.64

0.07

–13.26

34.46 5.2

–2.62 –11.57 2.06 –1.28 1.5 –5.3 12.8 –6.85 –0.91 –30.52

0.51 –3.51

–4.75

1.18

8.02 19.11

–0.26

–1.32 4.68

0.06 –3.39 0.3 0.43 4.08 –4.14 5.42 5.47 0.02 0.91 0.31

1.8 1.29

0.48

–2.13

13.32 –16.66

–4.42

0.20 –2.21

4.69 5.87 –9.39 –1.87 –1.99 –4.68 4.45 –4.68 –2.33 6.72 –4.20

–1.46 –2.93

–3.02

–0.84

3.22

–6.15

9.90 2.34

0.67 –3.40 –2.47 –.091 1.09 –4.71 7.39 –2.16 –1.08 –9.24

0.27 –1.74

–2.45

1961–71 1971–81 1981–91 1961–91

Rates of Growth (%)

Total manufacturing

1961

4.67 0.46 9.89 10.02 0.79 6.85 0.09 0.9 5.26 5.76 n.a. 0.09 0.11 1.13

8.7 n.a.

57 6 121 123 10 84 1 11 64 71 n.a.

1 1

14

107 n.a.

100.00

6.35 31.51

78 386

1,225

7.43

%

91

No. (in thousands)

(continued)

Urban areas, males only 20–21 Food products 22 Beverage and tobacco 23 Cotton textiles 24 Wool, silk, and synthetic 25 Jute textiles 26 Textile products 27 Wood products 28 Paper 29 Leather and fur 30 Rubber, plastics 31 Other chemicals 32 Nonmetallic miner 33 Basic metals 34 Metal products 35 Nonelectric machinery 36 Electric machinery 37 Transport equipment 38 Other manufacturing 39, 97 Repair

NIC Codes and Divisions

Table 6.7

1,257

121 40

3

6 1

47 2 197 143 11 68 3 7 63 2 65

84 319

73

No. (in thousands)

1971

100.00

9.62 3.2

0.26

0.47 0.1

3.77 0.15 15.69 11.34 0.86 5.41 0.27 0.55 5.03 0.18 5.18

6.7 25.37

5.84

%

1,716

146 254

5

10 3

51 2 213 181 17 54 8 13 86 7 71

106 367

121

No. (in thousands)

1981

100.00

8.53 14.8

0.27

0.61 0.19

2.96 0.13 12.41 10.55 1.02 3.13 0.47 0.77 5.02 0.38 4.15

6.16 21.42

7.02

%

1,372

367 40

2

6 3

77 6 74 136 14 47 11 8 71 9 49

94 287

71

No. (in thousands)

1991

0.8 –1.89

–2.14

100.00

26.75 2.92

0.15

0.44 0.22

0.26

1.27 n.a.

–13.26

18.68 –0.19

3.16

1.93 20.21

3.45

5.75 9.9

0.7 1.56 0.77 2.41 4.94 –2.32 8.9 6.79 3.14 11.46 0.9

2.29 1.43

5.09

–2.21

9.66 –16.88

–8.76

–4.98 0.00

4.21 11.61 –10.03 –2.82 –1.92 –1.38 3.24 –4.74 –1.90 2.54 –3.64

–1.19 –2.43

–5.19

0.38

4.19

–6.28

6.15 3.73

1.01 0.00 –1.63 0.34 1.13 –1.92 8.32 –1.06 0.35 –6.65

0.62 –0.98

–0.82

1961–71 1971–81 1981–91 1961–91

5.61 –1.86 0.44 –10.2 5.39 5 9.91 1.51 1.02 1.09 3.43 –2.08 0.80 11.99 0.58 –4.61 5.17 –0.19 0.66 –29.26 3.57 n.a.

6.85 20.92

5.17

%

Rates of Growth (%)

1.5 1.28 9.36 19.71 0.07 9.37 0.02 0.3 10.5 7.55

0.04 0.01 0.72 6.08

70 59 434 914 3 435 1 14 487 350 0

2 1

33

282 0

4,633

Total manufacturing

Source: Census of India.

5.19 18.75

241 870

100.00

9.42

437

Rural areas, males only 20–21 Food products 22 Beverage and tobacco 23 Cotton textiles 24 Wool, silk, and synthetic 25 Jute textiles 26 Textile products 27 Wood products 28 Paper 29 Leather and fur 30 Rubber, plastics 31 Other chemicals 32 Nonmetallic miner 33 Basic metals 34 Metal products 35 Nonelectric machinery 36 Electric machinery 37 Transport equipment 38 Other manufacturing 39, 97 Repair 3,764

271 80

8

54 2

50 17 484 769 4 233 3 5 444 9 269

251 560

251

100.00

7.19 2.13

0.21

1.43 0.05

1.32 0.46 12.84 20.43 0.11 6.18 0.07 0.14 11.81 0.23 1.16

6.67 14.87

6.68

3,931

142 439

6

39 2

47 11 489 771 5 143 3 8 422 6 253

295 631

220

100.00

3.61 11.16

0.16

1 0.04

1.19 0.29 12.43 19.61 0.13 3.65 0.07 0.19 10.74 0.14 6.43

7.51 16.05

5.6

3,181

639 72

5

45 1

78 17 188 652 4 75 6 5 331 14 162

252 455

180

100.00

20.09 2.26

0.16

1.41 0.03

2.45 0.53 5.91 20.50 0.13 2.36 0.19 0.16 10.41 0.44 5.09

7.92 14.30

5.66

–2.07

2.22

–5.55

18.17 8.59

–3.29 –11.71 1.08 –1.72 2.58 –6.06 13.89 –9.06 –1.01 –30.8

0.41 –4.32

–5.38

0.44

–6.26 18.51

–2.23

3.1 –1.38

–0.59 –4.12 0.11 0.02 1.6 –4.73 –0.92 3.52 –0.51 –4.57 –0.63

1.63 1.21

–1.32

0.15 –2.14

–2.91

–2.09

16.23 –16.54

–1.81

1.44 –6.70

–1.25

2.76

–6.10

10.94 0.00

5.20 0.36 4.45 –4.06 –9.12 –2.75 –1.66 –1.12 –2.21 0.96 –6.25 –5.69 7.18 6.15 –4.59 –3.37 –2.40 –1.28 8.84 –10.17 –4.36

–1.56 –3.22

–1.99

Total manufacturing

1.29 3.48 8.62 7.74 3.82 3.03 0.94 3.3 7.39 10.01 n.a. 1.29 1.68 7.32 6.75 n.a.

92 250 619 556 275 217 68 237 531 719

93 120

526

485

100.00

6.34 16.46

455 1,183

7,185

10.55

%

758

No. (in thousands)

1961

9,852

477 734

359

483 239

131 313 879 697 392 229 158 381 612 342 497

527 1,436

967

No. (in thousands)

1971

100.00

4.84 7.45

3.65

4.9 2.43

1.33 3.18 8.93 7.07 3.98 2.33 1.6 3.87 6.21 3.47 5.04

5.35 14.57

9.82

%

Employment in Nonhousehold Industry, 1961–91

All India, males only 20–21 Food products 22 Beverage and tobacco 23 Cotton textiles 24 Wool, silk, and synthetic 25 Jute textiles 26 Textile products 27 Wood products 28 Paper 29 Leather and fur 30 Rubber, plastics 31 Other chemicals 32 Nonmetallic miner 33 Basic metals 34 Metal products 35 Nonelectric machinery 36 Electric machinery 37 Transport equipment 38 Other manufacturing 39, 97 Repair

NIC Codes and Divisions

Table 6.8

15,834

844 1,437

507

704 424

225 387 1,773 1,101 586 276 299 584 961 652 778

496 2,110

1,691

No. (in thousands)

1981

100.00

5.33 9.08

3.2

4.45 2.67

1.42 2.44 11.2 6.95 3.7 1.75 1.89 3.69 6.07 4.11 4.92

3.13 13.33

10.68

%

19,993

1,626 2,371

656

703 565

397 361 870 1,610 712 490 840 522 1,349 841 1,275

527 2,090

2,188

No. (in thousands)

1991

100.00

8.13 11.86

3.28

3.52 2.83

1.99 1.81 4.35 8.05 3.56 2.45 4.20 2.61 6.75 4.21 6.38

2.64 10.45

10.94

%

3.21

–0.18

–3.74

17.95 7.11

3.53 2.3 3.57 2.28 3.62 0.54 8.8 4.87 1.43 –7.17

1.47 1.96

2.47

4.86

5.88 6.95

3.51

3.85 5.87

5.59 2.13 7.26 4.68 4.1 1.88 6.63 4.35 4.61 6.56 4.59

–0.6 3.93

5.74

2.36

6.78 5.14

2.61

–0.01 2.91

5.84 –0.69 –6.87 3.87 1.97 5.91 10.88 –1.12 3.45 2.58 5.06

0.61 –0.10

2.61

3.47

4.11

0.74

6.98 5.30

4.99 1.23 1.14 3.61 3.22 2.75 8.74 2.67 3.16 0.52

0.49 1.92

3.60

1961–71 1971–81 1981–91 1961–91

Rates of Growth (%)

Total manufacturing (continued)

Urban areas, males only 20–21 Food products 22 Beverage and tobacco 23 Cotton textiles 24 Wool, silk, and synthetic 25 Jute textiles 26 Textile products 27 Wood products 28 Paper 29 Leather and fur 30 Rubber, plastics 31 Other chemicals 32 Nonmetallic miner 33 Basic metals 34 Metal products 35 Nonelectric machinery 36 Electric machinery 37 Transport equipment 38 Other manufacturing 39, 97 Repair

11.46 4 8.37 6.31 4.62 3.17 1.16 3.47 4.65 11.23

1.46 2.07 8.54 7.46

75 206 431 325 237 163 60 179 239 578 n.a.

75 106

439

384 n.a.

100.00

4.42 19.3

227 993

5,144

8.31

427

6,885

350 570

310

389 207

103 258 571 396 329 152 133 297 274 268 378

218 1,154

529

100.00

5.09 8.27

4.51

5.65 3.01

1.49 3.75 8.29 5.76 4.78 2.21 1.93 4.31 3.98 3.9 5.48

3.16 16.76

7.68

10,605

608 1,044

440

570 361

175 315 1,064 567 475 186 236 453 362 505 538

238 1,599

869

100.00

5.73 9.85

4.15

5.38 3.4

1.65 2.97 10.03 5.34 4.48 1.76 2.22 4.27 3.41 4.76 5.08

2.24 15.08

8.19

12,949

1,085 1,646

532

539 468

296 287 562 827 562 337 621 389 514 632 924

256 1,426

1,046

100.00

8.38 12.71

4.11

4.16 3.61

2.29 2.22 4.34 6.39 4.34 2.60 4.80 3.00 3.97 4.88 7.14

1.98 11.01

8.08

2.96

–0.91

–3.42

17.89 6.9

3.15 2.31 2.85 2.02 3.31 –0.66 8.29 5.2 1.37 –7.38

–0.44 1.52

2.16

4.41

5.67 6.25

3.55

3.9 5.71

5.48 2 6.43 3.64 3.73 2.03 5.91 4.32 2.82 6.52 3.61

0.91 3.32

5.08

2.02

5.96 4.66

1.92

–0.56 2.63

5.40 –0.93 –6.18 3.85 1.70 6.12 10.16 –1.51 3.57 2.27 5.56

0.73 –1.14

1.87

3.13

3.52

0.64

6.79 5.07

4.68 1.11 0.89 3.16 2.92 2.45 8.10 2.62 2.59 0.30

0.40 1.21

3.03

Source: Census of India.

Total manufacturing

1961

0.83 2.16 9.24 11.34 1.82 2.67 0.39 2.85 14.3 6.94

0.87 0.7 4.24 4.96

17 44 189 232 37 55 8 58 292 142

18 14

87

101

100.00

11.16 9.32

228 190

2,041

16.19

%

331

No. (in thousands)

(continued)

Rural areas, males only 20–21 Food products 22 Beverage and tobacco 23 Cotton textiles 24 Wool, silk, and synthetic 25 Jute textiles 26 Textile products 27 Wood products 28 Paper 29 Leather and fur 30 Rubber, plastics 31 Other chemicals 32 Nonmetallic miner 33 Basic metals 34 Metal products 35 Nonelectric machinery 36 Electric machinery 37 Transport equipment 38 Other manufacturing 39, 97 Repair

NIC Codes and Divisions

Table 6.8

2,966

126 164

49

94 32

28 55 309 300 63 77 25 84 338 73 119

309 282

438

No. (in thousands)

1971

100.00

4.26 5.53

1.65

3.16 1.09

0.94 1.86 10.41 10.12 2.12 2.6 0.84 2.85 11.39 2.48 4.02

10.42 9.5

14.77

%

5,229

236 393

65

134 63

50 72 709 534 111 90 64 131 599 147 240

258 511

822

No. (in thousands)

1981

100.00

4.51 7.52

1.28

2.56 1.2

0.95 1.38 13.55 10.21 2.12 1.72 1.22 2.5 11.45 2.8 4.59

4.93 9.77

15.72

%

7,044

541 725

124

164 97

101 74 308 783 150 153 219 133 835 209 351

271 664

1,142

No. (in thousands)

1991

100.00

7.68 10.29

1.76

2.33 1.38

1.43 1.05 4.37 11.12 2.13 2.17 3.11 1.89 11.85 2.97 4.98

3.85 9.43

16.21

%

3.81

2.22

–5.55

18.17 8.59

5.07 2.27 5.05 2.64 5.4 3.51 12.06 3.79 1.47 –6.36

3.1 4.01

2.86

5.83

6.45 9.13

3.23

3.63 6.81

5.97 2.73 8.66 5.93 5.86 1.57 9.88 4.47 5.89 7.15 7.24

–1.8 6.13

6.5

3.02

8.65 6.32

6.67

2.04 4.41

7.28 0.27 –8.00 3.90 3.06 5.45 13.09 0.15 3.38 3.58 3.87

0.49 2.65

3.34

4.22

5.75

1.19

7.64 6.66

6.12 1.75 1.64 4.14 4.78 3.47 11.66 2.80 3.56 1.30

0.58 4.26

4.21

1961–71 1971–81 1981–91 1961–91

Rates of Growth (%)

Small-Scale Industry Policy in India: A Critical Evaluation

237

Overall, one has to express concern over the rather slow expansion of manufacturing employment in India. It has now become fashionable to argue that with changing technology, it is unrealistic to expect high growth in manufacturing employment. The fast-growing countries in East Asia showed much higher rates of growth in manufacturing employment in the 1960s, 1970s, and 1980s. A notable feature of this growth was a significant increase in the share of metal products. In the case of South Korea, the share of metal products (sectors 33, 34, 35, and 37 in tables 6.7 and 6.8) in total manufacturing employment increased from 12.5 percent in 1961 to 28 percent in 1980, and in Taiwan it increased from 20 to 33 percent. In India this share has not increased as much. Clearly, if our engineering industry, which is significantly labor-using, had expanded more rapidly, we would have seen higher employment growth. Similarly, if areas such as clothing, footwear, plastic products and the like had not been reserved for SSIs, it is possible that manufacturing employment would have grown more rapidly in these sectors as well. SSI reservations have been particularly applicable in these sectors. In summary, the unchanging share of manufacturing employment that has been observed over three decades in India is indeed atypical, demanding explanation and meriting concern. 6.2.2 The Size and Growth of Value Added in Small-Scale Industries The National Accounts Statistics provide the most consistent series of value added for the manufacturing sector (see table 6.9). The share of the unregistered sector has been falling consistently throughout the last two decades, from about 43 percent in 1980–81 to about 32 percent in 1997–98. The total SSI sector consists of the unregistered sector, as enumerated in the national accounts, and the small-scale portion of the registered sector. The latter can be estimated from the data available on the “factory” sector enumerated in the Annual Survey of Industries. Computed in this fashion, the share of the total SSI sector is seen to fall from about 47.3 percent of manufacturing value added in 1984–85 to about 44 percent ten years later in 1994–95. The trend is consistent with the data that can be pieced together from the Annual Survey of Industries (ASI) and the economic census (see table 6.10), though the proportions come out to be different. The fall in share of the unregistered sector is much more drastic according to this source. Having obtained broad estimates of value added in each of the segments of the manufacturing sector by size, it is now possible to obtain an idea of the relative gross value added per worker in each segment (see table 6.11). This has been done merely by dividing the proportion of value added in table 6.10 by those of employment in table 6.6. We now get an idea of the vast productivity differences between the different manufacturing sector segments. The workers in the non-SSI factory sector are, on average, 2.5 to 3 times as productive as the SSI factory sector, five to ten times as the

0

8.79

8.25

43.04

240 137 103

2.70

9.82

41.40

256 150 106

12.96

Growth rate registered (%)

14.37

15.3

32.84

7.01

7.98

32.64

4.31

7.96

31.89

1852 1261 590

3.72

8.32

37.92

300 186 114

6.92

2.67

38.88

313 191 122

Sources: National Accounts Statistics for 1990, 1994, 1996, and 1999 (CSO).

6.04

33.02

Share unregistered (%)

(Rs billion) at 1993–94 prices Total manufacturing 1401 1611 1734 Registered 938 1082 1168 Unregistered 463 529 566

Growth rate unregistered (%)

3.32

14.45

38.94

282 172 110

1994–95 1995–96 1996–97 1997–98

0

Growth rate unregistered (%)

42.92

Share unregistered (%)

Growth rate registered (%)

221 126 95

8.60

6.38

39.37

336 203 132

7.92

7.58

39.44

361 219 143

(Rs billion) at 1980–81 prices

5.89

10.37

38.46

393 242 151

8.20

14.62

37.10

440 277 163

7.37

7.08

37.17

472 296 175

–4.79

–0.30

36.10

463 296 167

5.26

2.24

36.77

478 302 176

4.47

5.03

36.65

501 317 184

1980–81 1981–82 1982–83 1983–84 1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94

Gross Value Added in Indian Manufacturing

Total manufacturing Registered Unregistered

Table 6.9

Small-Scale Industry Policy in India: A Critical Evaluation Table 6.10

239

Gross Value Added (GVA) in Manufacturing in Different Sectors (in percentages)

Sector

1984–85

1989–90

I Factory Sector (of which SSI) II DME III NDME IV OAE

66.2 (10.4) 9.2 8.8 15.9

73.3 (11.5) 8.2 6.4 11.7

Total Total GVA (Rs. billion)

100.0 377

100.0 710

1994–95 81.2 (10.9) 6.9 3.1 8.8 100.0 1,567

Source: Annual Survey of Industries (sector I); Economic Census (sectors II–IV). Table 6.11

Indices of Relative Gross Value Added per Worker

Sector

1984–85

1989–90

1994–95

IA Factory Sector (non-SSI) IB Factory Sector (SSI) II DME III NDME IV UAE

503 182 92 88 25

490 151 57 100 20

521 136 52 26 16

Total

100

100

100

Sources: See tables 6.2 and 6.8.

DMEs, and five to twenty-five times as the NDMEs and OAEs. These productivity differences between different segments of the manufacturing sector are much higher than what is typically found in other countries. Thus, preservation of low-level manufacturing activities essentially prolongs low income employment activities. Whereas this may be desirable as an alternative to unemployment, it would be better to design policies to wean people away from such low-productivity activities to higher-level activities. Preservation of SSIs is therefore not a virtue. Promotion of SSIs and entrepreneurship, however, could be a desirable objective if it serves as a preparation for higher productivity activities. The particularly low productivity of OAE and NDME activities suggests that these must increasingly be part-time activities providing supplementary income. I will now derive the growth rate of value added of SSIs. The main annual series available are those from national accounts (NAS) data, the ASI data, and the publications of the Small Industries Development Organisation (SIDO). As already noted, the national accounts provide information on the unregistered sector. The factory sector SSI data can be added to this to provide an approximation to the growth of the SSI sector. The data (see table 6.12) show that the SSI sector has not grown as rapidly as the large scale sector throughout the 1980s and 1990s in terms of value added. The

95,030 103,380 106,170 109,690 113,770 121,640 132,100 142,560 150,960 163,340 175,370 166,970 175,750 183,610 194,702 222,677 238,276 248,539

2

6.79 5.27 6.07

22.60 –6.44 22.12 –3.82 6.77 10.03 7.06 –8.88 32.03 –8.18 26.70 –6.25 –0.14 5.11 9.47

1.90

4 119,730 128,550 132,500 141,970 143,970 158,520 167,570 180,430 192,630 207,950 216,020 220,640 225,030 246,050 253,242 281,137 299,726 315,807

5

GVA of SSI (unregistered NAS
ASI)

6.33 5.36 5.87

7.37 3.07 7.15 1.41 10.11 5.71 7.67 6.76 7.95 3.88 2.14 1.99 9.34 2.92 11.02 6.61 5.37

6

Annual Growth Rate of SSI (%)

12.11 8.65 10.47

3.10 5.60 7.10 10.10 11.40 11.32 8.43

8.54 8.14 10.31 11.97 12.84 13.16 12.66 12.88 11.76

7

Annual Growth Rate of SSI Productiona Given by SIDO (%)

113,760 127,896 141,273 160,270 162,397 176,191 173,211 186,560 205,649 234,894 265,883 260,877 312,770 338,765 378,293 457,194 451,639 497,647

8

GVA of Large Scale Industry (%)

8.39 9.84 9.07

12.43 10.46 13.45 1.33 8.49 –1.69 7.71 10.23 14.22 13.19 –1.88 19.89 8.31 11.67 20.86 –1.22 10.19

9

Annual Growth Rate of LSI (%)

Sources: National Accounts Statistics for 1990, 1993, 1996, and 1999 (CSO); Annual Survey of Industries for 1980–81 to 1997–98 (CSO). Notes: For constant prices, index of WPI (manufacturing) has been used as deflator. The investment limit for SSI was increased in 1985 from Rs. 2 million to Rs. 3.5 million, and in 1991 from Rs. 3.5 million to Rs. 6.0 million. Hence, the high rates of growth in these years. a Value of product (not value added).

24,700 25,170 26,330 32,280 30,200 36,880 35,470 37,870 41,670 44,610 40,650 53,670 49,280 62,440 58,540 58,460 61,450 67,268

3

GVA (SSI in ASI)

Annual Growth Rate of SSI in ASI

Growth Rates of Small-Scale (unregistered NAS + ASI) and Large-Scale Sectors (Rs. million at 1980–81 prices)

GVA (unregistered NAS)

Growth rate (%) 1980–90 1990–98 1980–98

1980–81 1981–82 1982–83 1983–84 1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98

1

Year

Table 6.12

Small-Scale Industry Policy in India: A Critical Evaluation

241

data published by the SIDO have, however, consistently provided a picture of the SSI sector’s growing much more rapidly than the non-SSI manufacturing sector. Hence, the common erroneous perception is that SSIs have been doing extremely well. The exaggeration is not minor. Whereas the SIDO data show that the production in the small scale sector grew at a rate of 10.5 percent throughout the 1980s and 1990s, the NAS data show that value added in SSI grew at less than 6 percent per year (see columns 6 and 7 in table 6.12). This discrepancy is probably caused by the deficient methodology used by SIDO in estimating the growth rate of SSIs. It appears that SIDO arrives at its figures on a gross basis from an estimated number of SSI units without allowing for mortality of previously counted units. If an allowance is made for mortality of around 4 percent per annum, the cumulative effect over time is quite significant. That their data are grossly inflated is also illustrated by the employment growth shown by their data. According to SIDO, employment in SSI grew from about 9 million in 1984–85 to 14.7 million in 1994–95. Their estimates do not include various sectors like handlooms, power looms, coir, and the like. Yet the total employment shown in table 6.6a in DME, NDME, and SSI (factory sector) amounts to about 12 million in 1984–85 and 13.9 million in 1994–95. These data include all sectors not otherwise covered by SIDO. There is a general perception that large scale industries (LSI) have not been growing as quickly as small scale industries. As table 6.13 shows, this is clearly incorrect. Value added in LSI has grown at a healthy rate of over 9 percent throughout the 1980s and 1990s, this growth rate being at least 1.5 times as large as the SSI growth rate. It is possible that, had SSI reservations not constrained LSI growth, Indian industry might have grown at doubledigit growth rates throughout the 1980s and 1990s. Because SIDO is a government agency, the data they publish must be repeated by all government and government-associated agencies, including the Planning Commission. Thus a distorted picture on the progress of SSI has been consistently provided to policy makers and observers alike. It is possible that, had the correct picture been available to policy makers, there might have been less complacency with regard to SSI policy. If, however, those data are correct, then the National Accounts data on manufacturing must be significantly understated. 6.2.3 Performance of the Reserved Items in Production Throughout this whole period of reservation there has been little analysis of the effects of this policy. Data are hard to come by. It is not possible to obtain a regular series of the production of reserved items within the small scale sector because there are no regular annual surveys that could provide this information. The only two sources available are the two censuses of small scale industries carried out by SIDO in 1972 and 1987–88. The first

Table 6.13

Share of Reserved Products in Total Output 1987–88

NIC Codes and Divisions 20-21 Food products 22 Beverages, tobacco, and tobacco products 23 Cotton textiles 24 Wool, silk, and synthetic fiber textiles 25 Jute, hemp, and mesta textiles 26 Hosiery and garments 27 Wood products 28 Paper products and printing 29 Leather products 30 Rubber and plastic products 31 Chemicals and chemical products 32 Nonmetallic mineral products 33 Basic metal products 34 Metal products 35 Machinery and parts, except electrical 36 Electrical machinery and parts 37 Transport equipment and parts 38 Miscellaneous manufacturing 97 Repair services 99 Services not elsewhere classified OT Other services and products All industries

1972

No. of Reserved Products

Percentage Share in Output of Group

No. of Reserved Products

Percentage Share in Output of Group

17

35.85

0

0

1 0

0.20 0

0 0

0 0

0

0

0

0

0

0

0

0

31 14

80.11 56.85

0 2

0 20.58

30 17

24.79 46.86

0 2

0 12.06

99

30.92

7

32.43

166

29.74

19

26.55

39

14.47

8

28.75

14 131

4.18 42.62

0 62

0 49.06

55

8.83

2

32.76

59

8.57

22

37.55

102

23.80

48

8.58

68 0

35.22 0

5 0

64.31 0

0 0 843

0 0 29.36

0 0 177

0 0 23.71

Sources: Reports on Census of SSI Units, 1972 and 1987–88, respectively; reproduced from Ramaswamy (1994), table A-10.

Small-Scale Industry Policy in India: A Critical Evaluation

243

census was carried out before the large-scale expansion of the number of reserved products, from 177 in 1972 to 843 by 1987–88. Despite this large expansion in the list of reserved items, the share in production increased only marginally, from 24 to 29 percent (as computed by Ramaswamy 1994, table A-10, reproduced here as table 6.13). This small increase in the share was surprising, given the large-scale expansion of products under reservation that took place in 1977. In both cases, capacity utilization in units producing reserved items was 47 to 48 percent on average, whereas the average level of capacity utilization was over 50 percent in units producing unreserved items. It was also found in 1987–88 that a large number of reserved items were not produced at all in any unit. The experience with reservation tells a story of love’s labors lost. The government’s good intentions have manifestly not registered in the business calculations of small enterprises. The second census of SSIs provides persuasive evidence of the misplaced importance given the policy of reservation. Out of a total of 200 products leading in value of output produced by the small-scale sector, it was found that reserved products accounted for only 21 percent. Only 210,000 small-scale units, less than half of a total of 582,000 units, manufactured the reserved products at all. No less than 233 reserved items out of a total of 1,076 (when expanded at a lower level of aggregation in the NIC code) were found not to be manufactured at all according to the census. Although further inquiries have revealed that many of these products are indeed manufactured by some units, the fact remains that their production is in negligible quantities. Conversely, very few of the reserved products attracted significant levels of participation from small-scale units. As many as ninety products were found to be manufactured by just one company each. The sum total of the value of production of all small scale companies in as many as 692 items was a low of Rs. 100 million or less. Just sixty-eight reserved items accounted for 81 percent of the total value of production of reserved products and 83 percent of the units. Clearly, substantial growth in SSI took place outside of the reserved categories. At a more detailed level, the 1987–88 census found a total of 7,500 products manufactured in the small-scale sector. Out of these, 1,075 products were in the reserved category. Of 200 products classified as “leading products” (with total production of each exceeding Rs. 400 million in 1987–88), only 48 were in the reserved category. Looking at the product composition of reserved categories, one observes that a high proportion of metal products (NIC codes 33, 34, 35, 37) have been reserved for the small-scale sector (see table 6.14). The relatively slow growth in employment in the engineering sector as compared with fast industrializing countries is consistent with this observation. The lack of evidence showing higher growth in the reserved categories demonstrates the weak rationale behind the policy of reservation. It would

244 Table 6.14

Rakesh Mohan Growth and Structure of Employment by Major Product Groups

Major Product Groups

1972

1987–88

Annual Average Growth Rate (%)

Food and beverages Textiles Metals and electricals Other manufacturing Services

136,000 75,000 705,000 698,000 39,000

555,000 238,000 1,092,000 1,524,000 256,000

9.87 8.00 2.96 5.34 13.36

1,653,000

3,665,000

5.45

Employment

Total

Share of Total Employment (%) 1972

1987–88

8.2 4.5 42.6 42.2 2.5

15.1 6.5 29.8 41.6 7.0

100

100

Source: Census reports for 1972 and 1987–88.

be useful to test the relative efficiency of reserved sector units, but the data do not allow such an analysis. The 1987–88 census did, however, reveal that the capacity utilization of the SSI units in manufacturing reserved items was lower than others, both at the aggregate level and in individual industries (Sandesara 1993). Thus, the available quantitative evidence does not provide evidence of the efficiency of the policy of SSI reservations in promoting SSI growth. 6.2.4 Impact of SSI Reservation on Exports10 A key proposition in this chapter is that the policy of SSI reservations has had a very deleterious effect on the growth of both manufacturing employment and exports. An important feature of industrialization of the fastgrowing East Asian countries during the last three decades or so has been high growth in manufactured exports accompanied by high growth in manufacturing employment. The Indian experience has been different. As already demonstrated, the growth in Indian manufacturing employment has been poor, and so has its export growth (see table 6.15). The share of exports in Indian GDP has barely reached 10 percent now. Although this is a significant improvement over the 3 percent share of 1970 and 5 percent in 1980, the Indian economy remains the least open among major countries in Asia, including China. The volume of Indian exports in 1970 was the third highest among the ten Asian countries listed in table 6.15. Today it is the second lowest. While Chinese exports grew from about US$18 billion in 1980 to about US$120 billion in 1994, Indian exports during that period grew from US$8.6 billion to US$25 billion. Given that the composition of exports of industrializing countries is largely labor-intensive, one of the reasons behind slow employment growth in manufacturing in India is clearly related to the slow growth in exports. 10. All the data in this section have been taken from Rajesh Chadha’s (1998) work at NCAER.

22,500 21,650 11,595 8,260 3,480 2,410 950 880 530 320

Singapore Hong Konga Taiwanb South Korea Malaysia Thailand The Philippines Indonesia Chinac India

21,900 n.a. n.a. 10,330 8,440 6,970 2,740 3,600 2,510 1,280

PPP 1,896 2,951 5,670 8,580 4,003 6,543 6,999 9,151 138,813 53,947

1970 11,718 28,496 41,402 63,661 24,488 32,354 32,500 78,013 201,696 172,321

1980

GDP ($ millions)

68,949 131,881 241,014 376,505 70,626 143,209 64,162 174,640 522,172 293,606

1994 1,447 2,546 1,469 882 1,640 686 1,064 1,173 5,680 1,879

1970 19,400 19,800 19,811 17,500 13,000 8,510 5,740 21,900 18,100 8,590

1980

1994 96,800 151,395 93,049 98,000 58,756 45,262 13,304 40,054 121,047 25,000

Exports ($ millions)

76 86 26 10 41 10 15 13 4 3

1970

166 69 48 27 53 20 18 28 9 5

1980

140 115 39 25 83 32 21 23 23 9

1994

Exports/GDP (%)

Source: Reproduced from Chadha (1998). Notes: All 1970 data have been generated from World Tables (1980). GDP in local currency has been adjusted by the exchange rate to obtain $GDR Exports have been taken as merchandise exports only. All other data are from 1995 and 1996. PPP  purchasing power parity. a Hong Kong 1970 data are taken from World Tables (1976) because World Tables (1980) does not give the export figure. b Taiwan figures for 1980 and 1994 from Key Indicators (1996). Taiwan figures for 1970 taken from World Tables (1980). c Before China denotes that the data for 1973 are used for 1970 column. China 1973 data are taken from World Tables (1995).

Actual

GNP per Capita, 1994 ($)

Exports, GDP, and Degree of Openness

Country

Table 6.15

246

Rakesh Mohan

In order to substantiate this assertion, I will now examine the commodity composition of exports of key newly industrializing Asian countries over the last three decades or so. It is useful to classify the export commodity groups into five broad categories on the basis of skill, technology, and capital. Group 1 consists of primary commodities, including processed foods; group 2 consists of labor-intensive and resource-based industries requiring low levels of technology (textiles, clothing and footwear, toys and sporting equipment, wood and paper products, and nonmetallic mineral products). Group 3 consists of sectors that require low to medium levels of technology sophistication (iron and steel, fabricated metal products, transport equipment other than motor vehicles and aircraft, and sanitary and plumbing equipment). Group 4 consists of industries with medium to high levels of technology sophistication (rubber and plastic products, nonelectrical machinery, and motor vehicles). Finally, the sectors with the highest levels of technology sophistication are contained in group 5 (chemical and pharmaceutical products; computer and office equipment; semiconductors; aircraft and aeronautical equipment; and scientific instruments, watches and photographic equipment). This classification helps us to observe the changing export product composition over time in order to examine whether technology upgradation has been taking place or not. What is expected is that as countries industrialized, they would gain in technology sophistication, and the export share of groups 4 and 5 would rise. The pattern of change in export composition of the so-called Asian Tigers is remarkably uniform (see table 6.16). In the first stage of change, the share of group 1 exports, or primary products, falls drastically and the share of group 2 products tends to increase. Second, the share of group 3 starts rising at the expense of groups 2 and 1, and finally in the last stage the share of group 4 starts rising at the expense of groups 1, 2, and 3. Similar patterns can be seen in the changing export composition of countries such as Indonesia, Malaysia, and Thailand (see table 6.17). In all these cases, the initial kick in export growth is provided by group 2 products, particularly the categories of textiles, clothing and footwear, and to some extent toys and sporting equipment. Among group 4 products, it is the export of relatively simple kinds of electrical machinery (e.g., radios and other domestic electrical appliances) that are exported by industrializing countries. Finally, sophisticated items such as electronic equipment, computers, and telecommunications equipment start to get exported. It is the initial kick given by the group 2 items that provides the reputation for quality that then enables the export of other more sophisticated goods. It is within these groups also that consistent quality upgradation take place and unit values rise as better quality clothing, footwear, and such begin to get exported. The profits from these exports provide the surpluses necessary for companies to diversify into more sophisticated areas of exports. In order to examine the changing composition of Indian exports, a similar disaggregation is provided in tables 6.18 and 6.19. What is most

Group IV Rubber and plastic products (continued)

Group III Iron and steel Fabricated metal products Ships and boats Other a

3.6

53.3

5.6

43.9

2.3

1.5

11.0 4.9

2.6 3.0 0.6

10.9

3.7

25.3

43.8

11.1

30.9

1.7

0.1

9.1 7.7

1.3 0.0 0.1

3.0

0.7

17.7 14.1

42.8 17.5

Group I Food Other primary commodities

Group II Wood and paper products Textiles, cloth, and footwear Nonmetallic mineral products Toys and sports equipment

1975

1985

2.0

13.4

5.2 17.9 1.3

30.8 6.4

2.3

1.2

32.1

0.7

36.3

1.5

5.9 4.4

1985

Republic of Korea

2.2

35.3

2.8 5.6 1.0

14.7 5.4

0.7

0.7

22.7

1.1

25.2

2.5

5.3 2.8

1994

0.3

3.1

1.2 0.0 0.4

4.2 2.6

0.6

2.6

15.8

7.3

26.3

7.0

60.0 53.0

1965

3.1

11.7

2.7 0.4 1.1

6.1 1.9

3.5

1.1

38.9

5.2

48.7

2.4

19.0 16.6

1975

4.1

19.0

5.4 0.6 3.0

11.1 2.1

6.6

2.3

32.6

2.9

44.4

2.4

8.6 6.2

1985

3.9

29.2

6.1 0.5 1.0

9.6 1.9

3.4

1.2

19.4

1.7

25.7

3.0

7.0 4.0

1994

Taiwan Province of China

0.8

13.7

2.5 0.1 0.5

5.4 2.3

0.3

1.9

9.1

1.3

12.6

39.9

61.1 21.2

1965

0.6

24.8

2.0 3.8 0.2

8.6 2.6

0.4

1.1

7.8

3.1

12.4

25.1

36.9 11.8

1975

0.8

29.4

1.7 1.4 0.3

4.8 1.4

0.8

0.9

6.5

2.4

10.6

15.7

23.3 7.6

1985

Singapore

Commodity Structure of Exports from Asian Tigers, 1965–94 (percentage of total non-oil exports)

Commodity Group

Table 6.16

1.1

32.7

1.3 1.1 0.6

3.9 0.9

0.4

7.0

4.0

1.0

6.1

3.8

8.6 4.8

1994

1.0

4.7

3.6 0.5 2.6

7.7 0.9

8.8

0.9

64.2

0.5

74.4

2.8

7.5 4.7

1965

2.8

9.5

2.9 0.3 1.2

4.4 0.1

7.5

0.8

60.4

0.3

69.0

1.3

3.2 1.9

1975

2.3

12.9

2.5 0.1 0.9

3.6 0.1

8.6

0.5

46.7

0.6

56.4

1.9

4.0 2.1

1985

Hong Kong

1.8

18.2

2.3 0.0 0.2

2.7 0.2

1.4

0.7

44.8

1.5

48.4

2.1

5.0 2.9

1994

0.7

6.4 0.1

7.2

1.6

1.0

3.0 1.7

0.3 0.6

1.4

0.2

0.0

0.9 0.3

1975

1.5

1985

5.7 2.1

2.1

3.6

13.5

7.2 2.2

2.0

1985

Republic of Korea

6.7 1.7

4.0

7.1

19.5

20.8 6.6

5.7

1994

13.0 0.1

0.0

4.9

6.4

1.4 0.0

1.4

1965

9.0 1.8

1.6

2.0

14.4

5.1 0.7

2.8

1975

7.7 1.9

4.5

2.9

17.0

9.1 1.3

4.5

1985

6.6 2.3

13.5

6.1

28.5

15.1 2.0

8.2

1994

Taiwan Province of China

0.5 0.9

0.3

5.7

7.4

1.8 6.5

4.6

1965

4.9 4.0

2.6

6.0

17.4

13.2 2.4

8.6

1975

8.6 5.3

9.3

8.7

31.9

19.0 1.0

8.6

1985

Singapore

Commodity Structure of Exports from Asian Tigers, 1965–94 (percentage of total non-oil exports)

Source: Reproduced from Chadha (1998). a Transport equipment other than road motor vehicles, ships, aircraft, and sanitary and plumbing products. b Telecommunications, sound recording and reproducing apparatus and equipment, and semiconductors. c Aircraft and associate equipment; and scientific instruments, including watches and photo equipment.

Group V Chemicals and pharmaceuticals Computers and office equipment Communication equipment Other c

Nonelectrical machinery Electrical machinery Road motor vehicles

Commodity Group

Table 6.16

10.4 3.8

27.6

6.9

48.7

23.4 0.7

7.5

1994

3.6 0.9

0.0

1.3

5.8

3.1 0.0

0.6

1965

7.0 4.3

1.7

0.9

13.9

6.1 0.0

0.6

1975

6.9 9.4

5.7

1.1

23.1

9.0 0.0

1.6

1985

Hong Kong

4.5 10.2

7.1

4.0

25.8

13.1 0.0

3.3

1994

95.80 22.70 73.10 0.40 0.10 0.30 0.00 0.00 0.30 0.00 0.20 0.10 0.00 1.40 0.00 0.80 0.60 0.00 2.10 1.40 0.00 0.10 0.60

96.70 27.00 69.70

0.20 0.00 0.20 0.00 0.00

0.10 0.00 0.10 0.00 0.00

2.50 0.00 2.50 0.00 0.00

0.50 0.50 0.00 0.00 0.00

Group I Food Other primary commodities

Group II Wood and paper products Textiles, cloth, and footwear Nonmetallic mineral products Toys and sports equipment

Group III Iron and steel Fabricated metal products Ships and boats Other a

Group IV Rubber and plastic products Nonelectrical machinery Electrical machinery Road motor vehicles

Group V Chemicals and pharmaceuticals Computers and office equipment Communication equipment Other c

6.30 5.90 0.00 0.00 0.40

1.00 0.10 0.20 0.80 0.00

0.40 0.30 0.00 0.00 0.00

16.40 10.00 6.10 0.40 0.00

75.90 14.60 61.30

1985

7.50 3.20 0.90 2.50 0.90

3.90 0.90 0.60 2.10 0.40

3.00 1.00 1.10 0.20 0.70

43.60 17.30 24.70 1.00 0.60

42.00 11.70 30.30

1994

1.20 1.10 0.00 0.00 0.10

2.30 0.50 0.70 0.20 1.00

0.30 0.10 0.20 0.00 0.00

1.50 0.70 0.50 0.30 0.00

94.80 6.90 87.90

1965

6.90 1.00 0.90 0.60 4.40

5.70 0.70 1.60 3.10 0.40

0.80 0.20 0.40 0.10 0.10

5.60 2.60 2.70 0.30 0.00

81.00 7.70 73.30

1975

6.50 1.60 0.20 3.30 1.40

20.70 0.60 2.10 17.90 0.20

2.20 0.50 0.50 1.20 0.10

7.00 1.50 4.50 0.50 0.40

63.60 6.10 57.50

1985

Malaysia

a

Source: Reproduced from Chadha (1998). Transport equipment other than road motor vehicles, ships, aircraft, and sanitary and plumbing products. b Telecommunications, sound recording and reproducing apparatus and equipment, and semiconductors. c Aircraft and associate equipment; and scientific instruments, including watches and photo equipment.

1975

1965

Indonesia

31.60 3.10 10.00 13.80 4.70

29.80 1.30 3.60 24.50 0.40

3.00 0.80 1.10 0.60 0.50

12.00 4.10 6.20 1.10 0.70

23.60 3.60 20.00

1994

Commodity Structure of Exports from Asian Cubs, 1965–94 (percentage of total non-oil exports)

Commodity Group

Table 6.17

0.10 0.10 0.00 0.00 0.00

0.10 0.00 0.00 0.10 0.00

0.10 0.00 0.10 0.00 0.00

1.60 0.10 0.50 1.0 0.00

98.00 55.20 42.80

1965

0.90 0.60 0.00 0.10 0.20

1.60 0.40 0.20 1.00 0.00

0.80 0.30 0.50 0.00 0.00

11.10 1.30 6.60 3.20 0.00

85.70 64.00 21.70

1975

3.00 1.40 0.80 0.10 0.70

9.60 1.30 1.80 6.30 0.20

1.70 1.00 0.60 0.00 0.10

22.50 1.30 16.70 4.20 0.30

63.30 47.40 15.90

1985

Thailand

20.20 3.00 9.50 4.20 3.50

20.70 2.80 3.70 12.70 1.50

3.30 0.70 1.50 0.20 0.90

27.10 1.10 20.40 4.00 1.60

28.70 22.70 6.00

1994

Hong Kong 1970 1980 1990 1993 Korea 1970 1980 1990 1993 Singapore 1970 1980 1990 1993 Taiwan, China 1970 1980 1990 1993 2.0 2.8 4.1 2.5 9.6 7.4 3.3 2.6 16.4 8.1 5.2 5.0 18.6 8.6 4.0 3.9

830 17,451 64,837 819,416

1,554 19,376 52,627 73,876

1,429 19,838 67,040 81,338

0+1+22+4

All Food Items

2,037 19,704 823,903 135,248

Total Value ($ millions)

3.3 1.5 1.5 1.5

28.3 10.3 2.6 1.4

7.1 1.4 1.3 1.2

0.7 1.5 0.6 0.7

23.2 28.9 17.9 12.2

1.1 0.3 1.0 2.2

0.0 0.4 0.6 9.0

3

2 less (22+27+28)

0.6 2.3 1.4 0.2

Fuels

Agricultural Raw Materials

1.3 0.5 1.2 1.0

1.6 2.4 1.6 1.5

5.7 1.0 0.8 0.8

1.5 2.0 1.1 1.1

Manufactured Products

75.8 87.9 92.5 92.9

27.5 43.1 71.7 78.5

76.5 89.5 93.5 93.1

95.7 91.1 91.8 93.5

2.4 2.5 4.0 4.7

2.7 3.4 6.6 6.4

1.4 4.3 4.0 8.0

0.9 3.4 5.3 3.4

5

56.7 60.6 50.4 47.2

13.9 12.9 17.6 17.3

67.9 64.9 52.6 44.4

83.0 68.2 60.8 64.3

(6+8) less 68

16.7 24.7 38.0 41.0

11.0 26.8 47.5 54.8

7.2 20.3 36.9 42.7

11.8 19.5 25.7 25.8

7

0.3 0.1 0.1 0.1

3.0 7.2 1.1 1.5

0.0 0.4 0.1 0.1

0.2 1.3 0.9 1.8

Manufactured Chemical Machinery Products Products Others and Transport Unallocated

27+28+68 5 to 8 less 68

Ores and Metals

Export Structure by Main Categories, India and East Asian Countries (in percentages)

SITC Code Revision 1

Country and Year

Table 6.18

19.6 7.6 11.2 10.8 12.6 15.0 11.7 9.4 44.0 35.9 24.9 15.6 52.3 47.0 28.7 21.9 n.a. 15.1 12.7 10.8 29.7 28.2 15.6 16.8

1,055 21,909 25,553 36,643

1,687 12,945 29,419 47,055

1,060 5,751 8,186 11,212

685 6,369 23,002 37,080

n.a. 18,270 62,091 91,744

20,126 7,511 17,859 22,207

Source: Reproduced from Chadha (1998).

Indonesia 1970 1980 1990 1993 Malaysia 1970 1980 1990 1993 The Philippines 1970 1980 1990 1993 Thailand 1970 1980 1990 1993 China 1970 1980 1990 1993 India 1970 1980 1990 1993 5.6 5.0 4.1 2.0

n.a. 17.3 3.5 2.0

24.7 11.2 5.1 4.0

25.8 6.1 4.8 1.5

50.0 30.9 13.8 8.7

34.8 14.1 5.0 4.2

0.8 0.4 2.9 2.2

n.a. 16.3 8.4 4.5

0.3 0.1 0.8 1.1

1.6 0.7 4.9 2.0

7.3 24.7 17.8 10.3

32.8 71.9 44.0 28.4

11.9 7.5 5.8 3.7

n.a. 3.4 2.1 1.7

14.6 13.6 1.0 0.5

21.0 20.8 6.1 4.5

22.6 10.2 2.1 1.2

11.4 3.9 4.4 3.5

51.7 58.6 70.1 73.8

n.a. 47.5 71.4 80.6

4.7 25.2 63.1 71.1

7.5 21.1 39.0 41.8

6.5 18.8 54.2 89.7

1.2 2.3 35.5 53.1

2.3 4.2 7.4 7.4

n.a. 6.0 6.0 5.0

0.2 0.7 2.0 2.8

0.5 1.5 3.4 2.3

0.7 0.6 1.7 2.1

0.5 0.4 2.4 2.2

44.6 46.2 55.2 59.6

n.a. 38.7 47.9 59.8

4.4 18.6 41.4 40.6

6.9 17.4 22.1 20.7

4.2 6.7 16.4 23.3

0.3 1.4 31.6 45.9

4.7 8.3 7.4 6.8

n.a. 2.9 17.4 15.8

0.1 5.9 19.6 27.7

0.1 2.1 13.5 18.7

1.6 11.5 36.1 44.2

0.3 0.5 1.4 4.9

0.3 0.3 1.6 1.7

n.a. 0.4 1.9 0.4

3.3 2.0 1.2 1.5

0.1 15.7 20.4 34.8

0.9 3.0 0.4 0.6

0.3 0.1 0.0 0.0

Hong Kong 1970 1980 1990 1993 Korea 1970 1980 1990 1993 Singapore 1970 1980 1990 1993 Taiwan, China 1970 1980 1990 1993

0.10 0.00 0.00 0.00

0.80 2.00 0.30 0.20

0.40 0.50 0.10 0.00

0.30 0.00 0.00 0.00

0.20 0.00 0.10 0.10

1.10 0.90 0.20 0.20

0.20 0.40 0.10 0.10

271
56

0.20 0.10 0.10 0.20

4

0.00 0 0.00 0.00

0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00

331

0.60 1.40 0.60 0.70

23.10 24.80 17.70 12.10

0.60 0.20 1.00 2.20

0.00 0.00 0.00 0.00

332

0.30 0.10 0.10 0.10

0.50 0.80 0.40 0.50

0.20 0.10 0.20 0.20

0.30 0.10 0.00 0.20

54

29.00 210 16.00 15.10

5.60 4.30 4.80 3.80

41.10 29.90 22.20 19.20

44.30 28.20 13.80 8.40

26
65
84

6.00 8.10 7.50 7.70

2.40 3.80 3.20 2.90

3.30 14.30 8.90 10.00

3.40 2.40 0.90 2.90

67
68
69

3.40 4.70 16.30 19.00

4.00 6.00 23.40 28.30

1.00 2.10 7.70 8.80

0.80 2.60 3.00 9.80

71

12.40 15.80 17.00 16.50

4.00 16.10 21.50 23.80

5.30 11.00 19.40 21.70

10.50 9.40 5.10 15.90

72

0.90 3.10 4.80 4.80

3.00 4.30 2.60 2.60

1.00 6.80 9.90 12.30

0.60 0.10 0.00 0.10

73

Crude and Medical and Textile Fibers, Metals and Machinery Manufactured Crude Petroleum Pharmaceutical Yarn, and Metal Transport Cereal Fertilizers Petroleum Products Products Clothing Manufacturers Nonelectrical Electrical Equipment

Export Structure by Selected Commodity Groups, India and East Asian Countries (in percentages)

SITC Code Revision 1

Country and Year

Table 6.19

0.00 0.20 0.80 0.40

0.00 0.00 0.20 0.20

0.00 0.00 0.90 0.80

0.00 0.00 0.00 0.00

n.a. n.a. 0.10 0.10

0.00 0.00 0.00 0.00

0.30 0.00 0.10 0.30

0.10 0.10 0.30 0.30

0.00 1.40 0.10 0.20

32.40 21.40 5.80 4.00

n.a. n.a. 1.00 1.80

0.50 2.70 1.60 2.00

Source: Reproduced from Chadha (1998).

Indonesia 1970 1980 1990 1993 Malaysia 1970 1980 1990 1993 The Philippines 1970 1980 1990 1993 Thailand 1970 1980 1990 1993 China 1970 1980 1990 1993 India 1970 1980 1990 1993 0.00 0.00 0.00 0.00

n.a. n.a. 5.50 2.60

0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.30

3.90 23.80 13.40 6.70

29.20 53.30 24.30 13.00

0.10 0.60 0.40 2.90

n.a. n.a. 1.60 0.90

0.30 0.10 0.40 0.80

1.60 0.60 1.60 0.90

3.40 0.80 1.50 1.10

3.60 5.40 4.60 2.50

0.60 1.50 2.50 2.20

n.a. n.a. 1.00 1.00

0.10 0.30 0.10 0.30

0.10 0.10 0.10 0.10

0.20 0.10 0.10 0.10

0.30 .010 0.10 0.10

28.70 25.50 29.20 27.60

n.a. n.a. 28.80 30.50

4.50 10.00 18.70 14.70

2.20 6.70 9.70 9.10

0.70 2.90 5.90 5.80

0.20 0.70 11.50 17.00

9.00 4.30 4.00 6.30

n.a. n.a. 5.30 5.00

11.80 12.10 2.30 2.10

1.20 3.60 4.70 3.80

20.00 9.50 3.00 2.90

0.80 2.10 3.10 2.50

1.80 3.30 3.50 2.80

n.a. n.a. 4.40 4.10

0.00 0.40 9.00 11.20

0.10 0.20 1.10 2.30

0.70 0.80 4.50 10.10

0.30 0.00 0.20 0.80

1.10 1.80 1.60 1.40

n.a. n.a. 6.40 9.80

0.10 5.20 9.80 14.20

0.00 1.30 10.30 15.40

0.30 9.90 26.60 31.10

0.00 0.40 0.80 2.80

1.80 3.20 2.20 2.60

n.a. n.a. 6.50 1.80

0.00 0.20 1.00 2.30

0.00 0.60 0.70 1.00

0.60 0.80 2.40 3.00

0.00 0.00 0.40 1.30

254

Rakesh Mohan

noteworthy in the evidence given is the relatively unchanging structure of Indian exports since the early 1980s. Like other countries, India diversified away from primary products quite successfully. But after this initial diversification, its export structure remained stagnant throughout the period, relative to other Asian countries. In particular, the shares of textiles, yarn, and clothing have remained stationary, as has the share of machinery and transport equipment. No other country exhibits such a stagnant picture of export product composition. Detailed examination would also show the stagnation in unit values of Indian exports in these low-technology categories (see Rakesh Mohan and Somnath Chatterjee 1993 on low and stagnant unit values in Indian garment exports). One may observe that a very high proportion of items in group 2 is reserved for SSIs in India. In addition, some of the simpler items (like radios and domestic electrical appliances, some motor vehicle and bicycle parts, extruded plastic products, and the like) in groups 3 and 4 are also reserved. It is therefore likely that India’s policies on small-scale reservation have had a particularly strong impact on Indian export performance. The impact of reserved items in exports can be seen in tables 6A.1 and 6A.2, where all the main products exported by developing countries are listed, first by descending order in terms of value in 1994–95, and then by growth rate between 1980–81 and 1994–95. For convenience, the three-digit categories within which a substantial proportion of items are found to be reserved for SSIs in India are given a dark shading. A glance at the list demonstrates the large number of items that developing countries typically export are subjected to small-scale reservations in India. As many as 65–70 three-digit SITC categories (out of a total of 190, covering almost all developing countries’ exports) are found to have a substantial presence of reserved items. In terms of total value of developing countries’ exports in 1994–95, about 25–30 percent of that value was from items substantially reserved for SSIs. What is also noteworthy is that the top seven items among developing countries’ exports consist of petroleum, petroleum products, or high technology electronic and telecommunication equipments. These items amount to almost 30 percent of total exports. As shown above, India has almost no presence among these product groups, nor do other developing countries until they reach a higher level of technology sophistication. Thus, among the remaining manufactured items that can realistically be seen to be available for export at India’s current level of industrialization and competitiveness, almost half are seen to be covered by small-scale sector reservations. Using similar data to analyze the export behavior of developing countries, Sanjaya Lall has classified export items into low technology, medium technology, and high technology categories. A particularly telling piece of information that emerges from that analysis is that between 1985 and 1995, Chinese “low technology” exports grew from a level of about US$3 billion

Small-Scale Industry Policy in India: A Critical Evaluation

255

to US$72 billion. During that period, Indian exports in similar categories grew from US$2.5 billion to about US$13 billion only. Whereas it is difficult to disaggregate such data into those items reserved for SSIs, it is quite likely that such a classification contains a very high proportion of items that are indeed reserved. These data suggest strongly that Indian export growth, and hence employment growth, has been constrained heavily by the small scale reservation policy. The ascendancy of China and other Asian countries in all the SITC categories in which items are reserved for SSIs is particularly evident in table 6A.3, which ranks the presence of developing country exporters for some key items of developing countries’ exports. Once again, this shows the basic competitive advantage of India in the categories that suffer from SSI product reservations. As is evident from tables 6.15 and 6.16, most countries graduate successively from low-tech labor-intensive items (group 1, group 2) to higher-tech items over time. Of India’s top ten export items (at the three-digit level), nine fall in groups 1 and 2—the maximum for any of the countries listed (see table 6.20). The composition of Indian exports has been the most stagnant among Asian countries. In the absence of SSI reservations, it is likely that large Indian enterprises would have entered the areas reserved for SSI (e.g., clothing, knitted fabrics, hosiery, footwear, toys, extruded plastic products, electrical appliances) and achieved much higher export growth rates. In the process, they would have gained greater experience and then upgraded technology gradually. With the greater knowledge of export markets so achieved they would also have diversified into newer areas, leading to successive change toward group 3 and group 4 items over time. Because of SSI Table 6.20

Country Hong Kong Korea Singapore Taiwan Indonesia Malaysia Thailand The Philippines India China

Grouping of Top Ten 3-Digit Exports into Five Groups Refined Group Group Group Group Group Petroleum Petroleum Special I II III IV V Products Products Transactionsa Total 6 3

2 2 5 3 3

3 5 2 3 3 6 7

2 1

1 3 2

4 4 6 4

1

4 2 2

1

Source: Reproduced from Chadha (1998). Note: The five groups are described in the text. a Special Transactions refers to un-classified exports.

1

1 1 1

2 1 1 1 1

10 10 10 10 10 10 10 10 10 10

256

Rakesh Mohan

reservations, foreign direct investment has also not been permitted in any of these areas (without onerous export obligation). Thus an important source of technology and marketing support has also been lost by Indian enterprises. Another aspect of Indian exports is worth noting. Within group 2 exports, Indian exports are typically at the low unit value end. This was demonstrated for clothing exports by Mohan and Chatterjee (1993). This is consistent with the fact that a good proportion of these exports is produced by SSI units that are constrained from making larger investments to achieve higher quality and hence higher unit value. Indian exports in the constrained categories compete mainly in price. Thus, substantive export growth occurs whenever significant real devaluation takes place (as in the late 1980s and mid-1990s) and then slows down as those benefits are exhausted. Indian export growth has therefore been found to take place in spurts. Sustained double-digit growth over one or two decades has so far not occurred. When we look at the record of Indian exports in comparison with East Asian countries, it is difficult to avoid the conclusion that Indian exports have been heavily constrained by the policy of SSI product reservation. It is possible that the damage caused by such policies was not very high in the 1970s, when competition in exports of low technology products was not as high as it is now. Furthermore, changing industry structure and demand patterns in the developed world now place a much higher premium on both product quality and service quality, in accordance with the inexorably rising incomes in the developed world. The average quality demanded for products such as clothing, footwear, toys, sporting equipment, and the like is getting higher and higher. Furthermore, the integration of information technology in even these industries also requires greater investment and greater labor sophistication. Such products are no longer regarded as freestanding products but are increasingly becoming parts of long value chains, with the share of value added in plain manufacturing perhaps falling. Higher quality requires high-level designing upstream, even for simple products. Downstream marketing involves linkages with large organizations that buy such products in bulk. Small enterprises sandwiched between such high-level organizations find it increasingly difficult to operate and be competitive. Thus, apart from the loss that India has suffered over the last two to three decades, it is likely that in the future scenario it will become even more difficult for Indian small enterprises to survive, particularly in the reserved sector. Another issue of note is the prospective dismantling of the Multi-Fibre Agreement. Paradoxically, although it may have seemed that textile quotas were inhibiting Indian exports, it is likely that they were actually protected through the MFA mechanism. This is also shown by Mohan and Chatterjee (1993), who documented the fact that Indian clothing exports went primarily to quota countries and were almost absent in the

Small-Scale Industry Policy in India: A Critical Evaluation

257

markets of nonquota countries. Thus, the removal of small-scale reservation is especially necessary in the items affected by the removal of MFA. 6.2.5 Conclusions This section has surveyed a large number of sources to throw light on different aspects of the structure, growth, and economic characteristics of the SSI sector in India. It might be useful to highlight the more important of the findings. All the evidence suggests that the Indian manufacturing sector is likely to have been constrained by the various antigrowth policies promoting the small-scale sector, particularly that of product reservation. 1. Manufacturing employment growth has been particularly slow in India, and its share in total employment has been stagnant for thirty years. This is in sharp contrast to the rapidly industrializing East Asian countries, where employment growth in manufacturing exceeded that in other sectors and increased its share. 2. On the size of SSI, in spite of the different measures obtained from the different sources, the orders of magnitude of the share of SSI in total manufacturing and of its various components are reasonably clear. We may distinguish the “small-scale modern” sector, consisting of units employing six or more workers, and the “tiny” sector, which includes household enterprises. In terms of employment around 1990, the former accounted for roughly 20 percent of all manufacturing employment, but nearly one-half of employment in the modern manufacturing sector. The tiny and household sector is 2.5 to 3 times as large, depending on whether we include secondary workers in the labor force. The contribution of SSI in terms of value added is, of course, much smaller—only one-third as far as the modern manufacturing subsector is concerned, and no more than 40 percent of all manufacturing value added. The last point emphasizes the enormous difference in labor productivity between the different subsectors of manufacturing. 3. The size of the large-scale manufacturing sector is atypically small in India: It employs less than 10 million workers, who constitute less than a third of all manufacturing workers. In China this share is nearer one-half, and the number employed is 55–60 million. 4. As in other countries, the household subsector has declined slowly over the last two or three decades. A more surprising finding is that in spite of the vigorous policy of protecting the small scale, this decline has not been fully compensated for by the growth of nonhousehold production in the small-scale sector. 5. The large-scale sector has grown much more rapidly than SSI over the 1980s and 1990s, recording growth in value added at about 8.4 percent annually in the 1980s and 9.8 percent in the 1990s. In contrast, SSI value added grew at an annual rate of about 6.8 percent in the 1980s. Probably due

258

Rakesh Mohan

to deficient methodology, official data from the Indian Ministry of Industry have consistently exaggerated the growth of SSI production by about 12.1 percent annually in the 1980s and 8.7 percent in the 1990s. 6. In spite of the vast increase in the number of reserved items, much of the growth in the SSI sector seems to have been in product lines outside the reserved list. It is possible that the policy of reservation might be merely protecting inefficient units in stagnant industries. 7. A particular casualty of SSI product reservations has been growth in Indian manufacturing exports. A large number of categories in which India exhibits comparative advantage have been reserved for SSI. Consequently, Indian industry is unable to upgrade quality on a continuous basis and is also unable to diversify to higher technology and higher value added items, thereby stunting export growth. 6.3 New Policy Directions This chapter has documented the key features of Indian policy for the promotion of small-scale industry. Within the limitations of data availability, it has also attempted to document the actual performance of Indian SSI in its various aspects. It is clear that there has been continuous concern in the Indian government to promote SSI in order to enhance the growth of manufacturing employment in the country. This review suggests that the performance of Indian manufacturing as a whole and of SSIs in particular cannot be described as successful. In fact, comparison with other Asian countries, including China, suggests that Indian manufacturing growth has been slower than it should have been. In particular, manufacturing employment growth has been especially sluggish. Over a period of thirty years or more, the share of manufacturing employment in the total has remained stagnant; this is clearly unusual among countries at this stage of industrialization. SSI policy seems to have failed in its particular objective of promoting manufacturing employment growth. This chapter has also documented failure on the export front and has suggested that the policy of small-scale product reservations may have had a particularly important role in constraining growth in Indian manufactured exports, and in the stagnation of the composition of exports relative to other comparable countries. In fact, the slow growth in manufacturing employment is also likely to be related to slow export growth. It therefore seems inescapable that Indian policy toward SSIs should undergo a drastic revision, and a new approach must be put into effect. Indian industrial policy has had a pronounced penchant for licensing and regulation as instruments of policy. This approach has also pervaded the design of policies for the promotion of SSIs. In an economic environment characterized by a pervasive system of commands and controls, entry into industry and success subsequent to entry were heavily dependent on

Small-Scale Industry Policy in India: A Critical Evaluation

259

the entrepreneur’s ability to obtain licenses, quotas, and permissions from the central government. The system was heavily biased against small- and medium-scale entrepreneurs, who could not participate successfully in this competition for licenses and quotas. Consequently, much of the policy regime for providing support for SSIs was protectionist in nature. It provided small enterprises with special access to raw materials allocations, imports, and credit in that control-oriented regime. This naturally expanded the direct role of government. Similarly, the policy of reservations specifically addressed the difficulty of small enterprises in competing with the large in obtaining industrial licenses and the like, and set aside specific areas for their exclusive operation. Indian policy on SSI attempted not only to promote the entry of new entrepreneurs but also to nurture enterprises during their lives through various protectionist policies. In the past, state policy for SSI stressed protection of small enterprises against predatory competition from large companies. There was excessive emphasis on small enterprises’ being in competition with the large. The experience in other countries, however, suggests that small and medium enterprises are as often complementary to the operation of large enterprises as they are in competition with them. A wide range of relationships exists between the small, medium, and large segments of industry: Small firms can act as ancillaries, suppliers, subcontractors, and the like. Each of these relationships implies interdependence rather than dependence of one on another. Consequently, new policies designed to support SSEs must encourage such interdependence. Policy must relinquish its focus on different-sized enterprise groups’ competing against one another as groups. Where market failure was pervasive, state-owned institutions substituted for equivalent private sector services. As detailed earlier, over the years both central and state governments have set up a host of institutions that directly provide support services to SSI, at a time when such support was not seen to be forthcoming from within the private sector. However, government institutions can only be one source of information and expertise for smallscale companies. The new approach must focus on facilitation of such services rather than their direct supply by government institutions. Small- and medium-scale companies prefer and can afford services offered by the private sector. However, the government can serve a useful function by sharing the costs of common facilities such as training, technological upgradation, infrastructure development, effluent treatment, and market promotion. The regime of fiscal concessions granted to SSI units is such that they have a large incentive not to graduate into larger size units, yet another “promotional” measure discouraging industrial growth. Credit support policies have also relied on command and control mechanisms by specifying minimum allocation of credit to SSI. The supply of credit and investment capital to small enterprises from state-level institu-

260

Rakesh Mohan

tions has shrunk due to poor recovery of loans arising from poor credit assessment and monitoring practices. So far, monitoring mechanisms to increase the rate of recovery have not been instituted. Similarly, the venture capital industry plays a marginal role due to its small fund base in the absence of private pension funds and insurance companies. Past policies have also concentrated excessively on SSI to the exclusion of other activities. It must be recognized that enterprise exists in a whole host of activities, and in many areas such as software development and other industrial support services, the distinction between industry and service sector activities is increasingly blurred. Consequently, the new approach is vital for the support of SSEs as a whole. The focus on new policies for promotion of SSIs must be on the facilitation of new entrepreneurial entry into SSI and of their growth in terms of output employment, productivity, and exports. The new approach to SSI support that needs to be designed must use market-related mechanisms in order to promote the growth of enterprises in the SSI sector. Unlike the previous approach of viewing SSI as being in competition with large-scale industries, and hence protecting them from this competition, the new approach must recognize the many different kinds of relationship that emphasize complementarity between large and small enterprises. Avenues for partnership and cooperation through outsourcing, subcontracting, crossholdings, and the like must be actively encouraged to the mutual benefit of both industry segments. Thus, regulations such as the 24 percent limit on equity holding by both foreign and domestic large enterprises must be abolished. The new approach must focus on improving the competitive ability of SSI units through forging improvements in their efficiency and quality levels, rather than protecting them from competition. In view of the externalities involved in training assistance and research and development, some element of government subsidy is essential in this area. The new approach must look for public-private partnerships through tie-ups with business associations, research and development associations, technical colleges, and the like. Common mechanisms for forging such partnerships include that of providing matching funds. Similarly, the new approach to credit support must eschew the quota allocation route and attempt to attack the problem at its root through the reduction of transaction costs in lending to SSI units and the training of staff for better appraisal, evaluation, and monitoring of projects. Clearly, the current policies are far too thinly spread. It would be much more advisable to concentrate on a few instruments that are pervasive in nature. The first necessary policy action is the immediate abolition of the product reservation of SSIs; second, specific focus on the use of clusters for the facilitation of services to SSI enterprises; and third, a new approach for improving the credit flow to SSI enterprises. It is to these three issues that I give special attention in this concluding section.

Small-Scale Industry Policy in India: A Critical Evaluation

261

6.3.1 Abolition of Small-Scale Industry Reservations SSI product reservations have been seen as a very important component of SSI policy. In reality, reservations may have played only a limited role in promoting SSIs, while restricting the entry of large companies into these industries. The existence of reservation policy has also provided an illusion to the government and the country at large that adequate protection/promotion was being provided to SSIs. This may have militated against more realistic policies for providing support to SSIs for entry and growth. The issue of investment limit is also of greater relevance for the items reserved for SSIs. In the case of items that are not reserved, SSIs are free to grow: The only incentive not to grow is the loss of the various facilities and incentives that the units may have partaken of as small-scale units. In the case of industrial units manufacturing reserved items, they are not permitted to cross the small-scale investment limits and are therefore unable to grow. This chapter has documented the slow growth in Indian manufacturing employment and manufactured exports over the past two to four decades. The evidence points to particularly slow growth in the labor-intensive manufacturing subsectors, particularly as compared with the fast-growing East Asian countries. The record of export performance has been particularly poor as compared with these countries. It is also clear that reserved products are particularly found among those products seen to dominate export growth of developing countries. Finally, within SSIs the production of reserved products as a group appears to have grown more slowly the others. The policy of reservation prevents the successful units from growing and therefore acts as a dampener on entrepreneurship. The biggest psychological incentive for new entrepreneurs is the example of small enterprises that have grown large. The company which, until recently had the highest market capitalization in the world, Cisco Systems, was a small company just fifteen years ago. The policy for promoting SSEs should induce new entry by small entrepreneurs and then aid them to prosper and grow. The SSI reservations policy prevents this. Since the changes in the trade policy instituted since 1991, most items are now freely importable. A careful examination of the import policy shows that more than 550 items on the list of over 800 reserved products have been freely importable since at least 1996 (see Hussain 1997 and appendix A). This means that whereas foreign companies producing these products can sell such items freely in India, large domestic companies are not free to manufacture them. However, some protection is provided by the applicable customs tariff. All quantitative restrictions on remaining imports were removed in April 2001. Thus all items reserved for SSIs have now become freely importable. This is the most powerful argument for the immediate abolition of SSI reservations. The existing small-scale units as well as new entrants in these

262

Rakesh Mohan

industries must be provided an opportunity for investment in appropriate size and technologies in order to compete with imports in the coming years. I have documented that a number of the labor-intensive industries that have flourished in East Asian economies, such as garments, shoes, leather goods, sporting goods, hand tools, and the like, have not done so in India due to the exclusion of large companies. The toy industry has prospered in other Asian countries, for example, but its growth in India has been unremarkable. Similarly, the light engineering industry, which produces items such as hand tools, is among the larger industries in East Asia but has not achieved significant progress in India. The food processing industry in India is a potential giant, but its growth has been arrested by the reservation policy. The reservation of products for the small-scale sector has proved to be a barrier to growth of exports in a number of these industries. It is true that despite severe odds, some of these industries, such as garments and shoes, have contributed tremendously to the country’s export effort, as is documented elsewhere. These are the sectors in which the country exhibits significant comparative advantage. However, various studies show that our exports of items such as garments and shoes are limited to a narrow range of goods and exhibit low unit values. They are also not able to supply branded goods in large quantities with consistent quality to large buyers, as are trading houses and department stores. In the case of garments, for example, our record of exporting to nonquota countries is poor because we are unable to compete. The dismantling of the Multi-Fibre Agreement will mean the existing quota markets will also become increasingly competitive over the next eight years. It is therefore imperative for future export growth to remove such SSI reservations so that adequate new investment and technology upgradation take place in these industries and so that existing units are allowed to upgrade. Existing units are unlikely to be hurt, as they could continue exporting their current composition of products: Instead, they would be enabled to grow. The experience in southeastern and East Asian countries suggests that these industries would grow quite spectacularly if they were freed in such a manner. Whereas many large units could be set up in sectors such as garments, toys, and shoes, a great amount of outsourcing and subcontracting would take place so that, in fact, even greater growth would take place in the small-scale sector with the attendant employment growth. Reserved items also have an impact on the large industries that supply intermediate goods to the final manufactures of the finished goods reserved for SSIs. For example, in the case of the textile industry, where fiber or fabric producers supply to garment producers in the small-scale sectors, domestic industry is handicapped because they do not have large-scale buyers. Large industries would become more competitive if they had more demanding buyers. They would then upgrade in quality as well.

Small-Scale Industry Policy in India: A Critical Evaluation

263

The removal of reservation will also pave the way for greater equity participation from large Indian companies and foreign investors along with greater subcontracting. Large companies will then have an incentive to establish long-term relationships and transfer proprietary technology for improving the quality of products supplied by small-scale companies. It would then be much easier to establish interdependent relationships between large, medium, and small industries as subcontractors, ancillaries, and suppliers of parts and components. The establishment of mother units would then spawn large numbers of small units in clusters. Due to all these considerations, the policy of reservation must be entirely abolished. 6.3.2 Clusters: A New Approach to Small-Scale Industry The accent on nurturing new enterprises has also resulted in inadequate attention to improving the competitiveness of small- and medium-scale enterprises. An overriding influence on the collective fortunes of small- and medium-scale enterprises is their overall business environment. The extraordinary success of Southern Italy (popularly termed as “Third Italy”) in fostering competitive small- and medium-scale firms, confirmed by similar experiments in several developing countries, has established that such companies need linkages with numerous organizations and individuals for their sustenance (see Humphrey and Schmitz 1995). Individually, smaller enterprises find it difficult to source new knowledge and skills with which to upgrade their production techniques and labor skills. Thus cluster-level institutions that can assist SSI enterprises are clearly necessary if the latter are to expand their skills and improve their technology and quality levels. It seem that the domination of government in all aspects of small industry development in India has discouraged other agencies that might otherwise have emerged to contribute to the strengthening of SSI enterprises. One would have expected a greater emergence of business associations, local government institutions, and other support organizations for helping SSI enterprises. Cluster development in India has taken place spontaneously in most cases, and without the help of government policy. It is quite remarkable to observe that although there has long been recognition within the government of the need to invest in growth centers for the promotion of SSEs, the tendency has always been to make such government investment in entirely new locations. There has been little attempt to recognize the existing clusters and make similar investments there. Thus the industries in these clusters have suffered from great inefficiencies because of the primitive state of the infrastructure services existing there (see Gulati 1996). Worldwide, a common characteristic of the growth of small- and medium-scale enterprises is their agglomeration in clusters. India is not an exception to this pattern of growth and Gulati’s survey for the United Na-

264

Rakesh Mohan

tions Industrial Data Organization has identified 138 such clusters (see table 6B.1). A fact of far-reaching significance is that the formation of only 13 of the 138 clusters has been induced by policy. The remaining 125 clusters have grown spontaneously on the initiative of entrepreneurs. No less than 99 of the clusters have grown in response to market opportunities in the region, only 6 have been attracted by the availability of infrastructure, and the remaining 33 were drawn by raw materials or skilled labor. Worldwide experience suggests that clusters can stagnate in their later stages of development unless they are supported by institutions. These clusters are appropriate units for focused development of a large number of enterprises in a way that lowers the administrative cost of development programs. Since these clusters usually specialize in a single area of activity, it is possible to design composite programs meeting interrelated needs. It will also be possible to forge government-private sector partnerships for the effective implementation of regional development programs. A key feature of cluster formation is the cross-flow of information. This is doubly true for science parks, where knowledge is translated into innovative products. An uninhibited flow of information can take place in an environment of mutual trust. The free flow of information is important not just to the knowledge of workers but also for the assessment of the viability of projects where tangibles like plant and equipment have not been installed. Venture capital financiers cannot evaluate the value of intangible innovative concepts unless their details are disclosed to them. However, entrepreneurs would be unwilling to share information about their inventions if they feared their appropriation by their competitors. Providing support through public private partnership is a very practical approach to SSE promotion in India because there already exists a large range of SSI clusters across the country. Many existing clusters of industries in India suffer from extremely inadequate infrastructure support, since most of them have emerged naturally and without the benefit of government support. The government policy should therefore be oriented much more toward recognizing natural existing clusters and promoting infrastructure support wherever possible. 6.3.3 Credit Support As noted at the start of this chapter, the key market failure facing SSEs is in the capital market. SSI enterprises face difficulties in accessing financial support for startups: They typically face higher cost of credit relative to large enterprises, and working capital support from commercial banks is often not timely and usually inadequate in volume. All of these problems are not specific to India but are faced by SSI enterprises around the world. The key problem at the root of these difficulties is related to high unit transaction costs inherent in delivering credit to SSI units. The Indian policy approach to solving this problem has been to devise schemes forcing com-

Small-Scale Industry Policy in India: A Critical Evaluation

265

mercial banks and other financial institutions to allocate credit to SSI enterprises above some specific minimum levels. As in the case of many other policies, there has been little attempt to address the problem at its root. The consequence has been that in the face of credit quotas and allocations, banks and financial institutions have had little incentive to improve their credit assessment methods or to reduce the transaction costs inherent in these activities. A new approach to promoting credit availability for SSI must be to strike at the root and improve the availability of information on SSI entrepreneurs. It would be very helpful if credit histories of budding or existing entrepreneurs were available to banks and financial institutions: This would reduce the transaction costs involved in making credit assessments of potential borrowers. One would have expected such services to have emerged by themselves in response to such demand from banks, and that this has not happened is probably due to two reasons. One, which has already been mentioned, is the absence of any incentive with commercial banks to look for suchservices. The second relates to the legal problem that currently exists in the Indian banking secrecy laws restraining sharing of credit information. The demand for such services should now rise with the more stringent supervision of banks. The way forward is clearly to remove these legal obstacles and to generate the emergence of an active credit information industry. However, this cannot be expected to solve all problems. One of the problems behind the continued existence of high transaction costs in the process of credit assessment of SSI entrepreneurs is the existing bureaucratic functioning of government-owned commercial banks, which continue to dominate the scene. Because of the system of regular transfer of personnel between bank branches, there is little opportunity for bank managers to internalize informal information on the credit history of potential and current borrowers. Although this is a large issue concerned with financial sector reform and the internal workings of Indian commercial banks, this problem can be partially solved by focusing on specialized branches that serve SSIs. In particular, bank branches could be particularly focused on existing clusters of industry. Presumably transaction costs involved in assessing borrowers are much less when borrowers are concentrated in the same or similar kinds of industries. The venture capital industry is in its infancy in India. Given the large size of the country, it would be useful if venture capital funds could be induced to exist in different parts of the country in order to utilize all the informal knowledge available in each area. This can perhaps be done by the banks themselves by participating in such funds designed to serve specific areas of industry groups. This activity will probably increase much more once the enterprises seeking reforms in India take root and new enterprises and pension funds support functioning. In the United States and the United King-

266

Rakesh Mohan

dom, 65 to 70 percent of venture capital funding comes from pension funds and insurance companies. The successful management of venture capital projects is ensured by partners with long experience in the kinds of businesses in which they invest. It also helps greatly when venture capital projects are located in science parks or other forms of clusters, which contributes significantly to the reduction in transaction costs. Location of projects in industrial or technology parks also contributes significantly to the probability of their success. Finally, SSI enterprises also face great difficulty in obtaining timely payments from their buyers. This also constraints their ability to manage their finances effectively and adds to their costs. Administrative measures providing for appropriate legal recourse for payment delays are useful but often ineffective. In this area also the approach adopted must be a marketrelated one, in which factoring services, bill discounting, and the like are promoted. These are only a few illustrations of all the kinds of new approaches that can be taken for the facilitation of better credit support to SSIs. The key point is that there has to be a new approach addressing the problems underlying the difficulties faced by SSI enterprises in achieving adequate access to credit. The solutions to these problems must be sought in improving the market mechanism itself, wherever necessary by government facilitation, rather than by specification of credit quotas and the like through commercial mechanisms.

References Asian Development Bank. 1998. Key indicators of developing Asian and Pacific countries, vol. 39 Manila: Oxford University Press. Chadha, Rajesh. 1998. India’s export performance: A comparison with East Asian countries. In Indian economy in transition: Environmental and development issues. Ed. Manmohan Agarwal, Alokesh Barua, Sandwip Kumar Das, and Manoj Pant, 141–210. New Delhi: Har Anand Publications. Fallon, Peter. 1986. The effects of Labour Regulation upon industrial employment in India. Washington D.C.: World Bank. Fields, Gary. 1985. Industrialisation and employment in Hong Kong, Korea, Singapore and Taiwan. In Foreign trade and investment in the newly industrialising countries, ed. Walter Galenson Madison: University of Wisconsin Press. Gulati, Mukesh. 1996. Restructuring and modernization of SME clusters in India. New Delhi: United Nations Industrial Development Organisation. Guhathakurta, Subhrajit. 1992. Regulation of firm size in industrial development: The experience of two manufacturing sectors in India. Berkeley Planning Journal 7:76–97. Humphrey, John, and Hubert Schmitz. 1995. Principles for promoting clusters and networks of SMEs. United Nations Industrial Development Organization, Vienna, October.

Small-Scale Industry Policy in India: A Critical Evaluation

267

Hussain, Abid. 1997. Report of the expert committee on small enterprises. New Delhi: National Council of Applied Economic Research. Ministry of Industry and Company Affairs. 1985. Small-scale industries in India: Handbook of statistics. New Delhi: Ministry of Industry and Company Affairs. Mohan, Rakesh, and Somnath Chatterjee. 1993. India’s garments exports. Economic and Political Weekly 28 (35): M-95–M-119. National Council of Applied Economic Research (NCAER). 1993. Structure and promotion of small scale industries in India. New Delhi: NCAER. ———. 1996. Impact of economic reforms on the large, small and medium enterprises in the organised and unorganised sectors. New Delhi: NCAER. Ramaswamy, K. V. 1994. Small-scale manufacturing industries in India: Some aspects of size, growth and structure. Bombay: Indira Gandhi Institute of Development Research, Discussion Paper no 105. Sandesara, J. C. 1993. Modern small-scale industry, 1972 and 1987–88: Aspects of growth and structural change. Economic and Political Weekly 28 (6): 223–29. Small Industries Development Bank of India (SIDBI). (1999). SIDBI report on small scale industries sector. Lucknow. United Nations Conference on Trade and Development (UNCTAD). 2000. Handbook of international trade and development studies, 1996/1997. Geneva: UNCTAD. Wood, Adrian. 2000. Awakening the other giant: Trade and human resources in India. Processed. University of Sussex, Institute of Development Studies.

Appendix A Note on Sources of Statistics on Small-Scale Industry in India There are two major sources of information on small-scale industries (SSI) in India: the Small Scale Industry Development Organisation (SIDO) and the Central Statistical Organisation (CSO): information on these and other sources in this appendix is from Hussain 1997). Although representing a substantial share of the total SSI population, SIDO does not cover all industries. More importantly, its data suffer from sampling problems and are highly aggregated. On the other hand, CSO provides better coverage and is less aggregated, but its information is scattered among different surveys and is not readily available. Administratively, India’s SSI is divided into seven industry groups: (1) handicrafts, (2) handlooms, (3) khadi, village, and cottage industries, (4) coir, (5) sericulture, (6) powerlooms, and (7) small-scale industries that are residual. The first five subsectors are collectively called the “traditional” sector, whereas the last two—powerlooms and residual small-scale industries—are known as the “modern” sector. Each of the subsectors has its own supervisory body or board, such as Khadi and Village Industries Commission, Development Commissioner for Handlooms, Development Commissioner for Handicrafts Board, Central Silk Board, Coir Board, and SIDO. The residual “small-scale industries” subsector is overseen by SIDO.

268

Rakesh Mohan

For the purposes of administering tax and promotional benefits, there is a unified official definition of SSI that cuts across the different supervisory lines. Since 1954, when the Small Industries Development Program was introduced, the definition has undergone continual change. In 1955, SSI was defined as establishments with fixed investment of less than Rs. 500,000, which employed less than 50 workers when working with power or less than 100 workers when not working with power. But the employment criterion was dropped in 1960, and SSI has since been officially defined solely in terms of investment in plant and machinery at original value. From time to time, the investment ceiling is revised upward. SIDO Data First Census The first attempt by SIDO to establish a database was made in 1972, when it undertook an All-India census of the registered small-scale industrial units. Detailed information was collected from around 140,000 units (out of 258,000 expected) on number of units, employment, fixed investment, borrowings, inputs, output, and exports at the four-digit level of the National Industrial Classification. Some of the results were presented in the Handbook of Statistics 1985. For the purpose of updating the census data on a year-to-year basis, a revised registration procedure was introduced in 1975. As before, registration with SIDO remained voluntary, but new provision was made in the application itself to collect all basic information, including product manufactured, employment, capacity, investment, and so on. Also, all registrants were requested to supply annual production returns to the Directories of Industries in each state. To keep track of production trends at the industry-group level, production data have been collected on a quarterly basis since 1976 from establishments registered with SIDO, covering 356 products based on a 2 percent sample (2,400 units) from the 1972 census. From this, a weighted production index for the registered small-scale sector is routinely computed using 1970 as the base year. Given the small sample size, the results are neither precise nor representative, while the corresponding index for unregistered units is nothing but a very rough approximation. Second Census Although the data collected during the first census continued to be periodically updated, it was found that the system was inadequate because it did not cover all the 2,400 products and that almost half the units sampled were either untraceable or had been closed down. A second All-India census of registered SSI units was conducted to obtain a more reliable frame of SSI

Small-Scale Industry Policy in India: A Critical Evaluation

269

units registered with the State/UT Industries Directorates up to 31 March 1988. The census was to cover around one million registered SSI units, but in the event only around 800,000 were enumerated. Annual Survey of Industries (ASI) The ASI is the principal source of industrial statistics in India. It is conducted annually under the statutory provisions of the Collection of Statistics Act, 1953. The field work of the survey is carried out by National Sample Survey Organization (NSSO) through its network of regional and subregional offices located in different parts of the country. The ASI covers the entire country except some of the smallest nonindustrialized states. All the electricity undertakings engaged in the generation, transmission, and distribution of electricity registered with the Central Electricity Authority are covered under ASI irrespective of employment size. ASI covers all factories registered under the Factories Act of 1948, which employed ten or more workers and using power or twenty or more workers and not using power on any day of the preceding twelve months. Certain services and activities like cold storage, water supply, and repair of motor vehicles and other consumer durables like watches are covered under the survey. The primary unit of enumeration is a factory in the case of manufacturing, a workshop in the case of repair services, an undertaking or a license in the case of electricity, gas, and water supply undertakings, and an establishment in the case of bidi and cigar industries. The owner of two or more establishments located in the same state and belonging to the same industry group is, however, permitted to furnish a single consolidated return, a common practice among bidi and cigar establishments, electricity, and certain public sector undertakings. The ASI frame is revised every four years by the Regional Offices of the NSSO, who liaise with the chief inspector of factories in the states. While names of the deregistered factories are removed from the ASI frame only once every four years, newly registered units are added every year. In enumeration, the units (often small-sized) that have been selected for the survey but are found to be nonexistent are excluded from calculation. For the purpose of the ASI, the factories in the frame are classified into two sectors, the census and the noncensus (or sample) sectors. Until 1987–88, the census sector was operationally defined as factories employing 50 or more workers and using power or those employing 100 or more workers but not using power, while the remaining factories constitute the noncensus sector. From 1987–88, the census sector is redefined as units with 100 or more workers, regardless of power use. Once a factory is classified into census/noncensus sector, its status is not altered for a period of four

270

Rakesh Mohan

years, that is, until the frame is revised, even though a change in employment might warrant it. All the census units are enumerated, while only half (one-third since 1987–88) of the sample sector are enumerated each year, with the exceptions of: (1) industries that, at the three- or four-digit level of NIC, do not total more than fifty factories in the whole country; and (2) factories located in relatively less industrialized states and union territories. Otherwise, individual units in the sample sector were enumerated every other year, based on a stratified unistage sample design. The reference period for ASI 1984–85 was the accounting year of the factory ending on any day during the fiscal year 1984–85. ASI provides results by employment size and investment value at the aggregate level, whereby SSI units may be readily distinguished. However, at detailed levels of two- or three-digit NIC, no such breakdowns are available, which by virtue of their relatively large employment size are registered with SIDO. However, these represent only the larger units of SSI. For information on smaller units, that is, those with less than ten employees and using power or less than twenty without power, one has to consult two other complementary surveys also by CSO, namely the Directory Manufacturing Establishment (DME) and Non-Directory Manufacturing Establishment (NDME) surveys, the latter including Own-Account Establishments (OAE). Directory and Nondirectory Manufacturing Establishment Surveys To obtain information on the “unorganized” nonagricultural sectors of the economy, the first economic census of nonagricultural establishments was carried out in 1977 by the Central Statistical Organisation. In the census, establishments were divided into three groups, namely (1) Directory Manufacturing Establishments (DME), which had six or more employees, at least one of whom was hired; (2) Non-Directory Manufacturing Establishments (NDME), which had one to five employees, at least one of whom was hired; and (3) Own Account Enterprises (OAE), if they had hired no workers at all. Based on the frame produced from the census, two rounds of sample surveys were subsequently conducted involving establishments that were not covered by ASI, that is, DME, NDME, and OAE. The first round took place in 1978–79 and covered units in manufacturing and repair services, while the second round, which followed six months later, was concerned with trade, restaurants and hotels, transport, and services. Conducted at the three-digit level of National Industrial Classification (NIC)11, the surveys covered the entire country excluding the rural areas of 11. Except for handicrafts, handloom, and khaki, which were classified at the four-digit level for identification purposes.

Small-Scale Industry Policy in India: A Critical Evaluation

271

Nagaland and Chandigarh, Sikkim, Lakshadweep, Dadra and Nagar Haveli, some part of Jammu and Kashmir, part of Madhya Pradesh and Maharashtra, Manipur, and Andaman and Nicobar Islands. In NDME and OAE, a combined total of 178,664 were enumerated, or about 2.2 percent of the estimated total of 8.13 million. In the DME, a total of 34,878 establishments were enumerated, or 10.4 percent of the estimated total of 334,896. Similar surveys were conducted five years later in 1984–85, based on the frame obtained from the Second Economic Census (1980). The second NDME survey, along with OAE, was undertaken during the period from July 1984 to June 1985 as part of the fortieth round of the NSS and was followed three months later by the second DME during the period from October 1984 to September 1985. In these surveys, the output/turnover/receipts criterion that was used in the first round was dropped, and distinction between directory and nondirectory is now based on the number of employees only. That is, units with six or more employees, at least one of whom was hired on a fairly regular basis, were classified as DME, including establishments manufacturing bidis and cigars other than those covered in the ASI establishments, irrespective of whether they were registered or not under the Bidi and Cigar Act. Likewise, NDME is defined as those with five or less employees, at least one of whom is a hired worker; establishments with no hired worker at all are classified as OAE. In the 1984–85 (along with OAE) survey, a total of 135,998 enterprises were canvased, or 0.7 percent of the estimated total of 19.2 million. Of this, 75 percent were in rural areas and 25 percent in urban areas. Own-account enterprises accounted for 89 percent of the total. In the corresponding DME, a total of 31,739 units were interviewed, or 6.7 percent of the estimated total of 474,882. Of this, 37.7 percent were in rural areas and 62.3 percent in urban areas. The geographic coverage of the second surveys was essentially the same as in the first. Further surveys were conducted in 1989–90 and 1994–95. Data Limitations and Alternatives None of the above data sets separately provide complete coverage of India’s SSI. To recapitulate, SIDO’s data and censuses only cover units under its purview, that is, the “residual” SSI in the modern sector, and are thus incomplete in their coverage. Of those units under its purview, only about two-thirds are reportedly registered; information on the rest is just a rough estimate. ASI is also partial in its coverage, as it only covers the larger spectrum of the SSI units that are registered factories. A further complication is that in the publicly available results these units cannot be distinguished from the rest of the sample except in the aggregates, where the results are broken

272

Rakesh Mohan

down by employment and investment size. No such distinction is made in industry-wise data. Another commonly used proxy for SSI’s output is the value added of “unregistered/unorganised sector” at the two-digit level of NIC as reported in the National Accounts Statistics. Based on DME, NDME, and OAE, these represent net output of the unregistered units that are not covered by ASI. However, as these establishments cover mainly those with ten or less employees, they represent the smaller units of the SSI sector. It should also be noted that because these surveys are conducted at five-year intervals, the national accounts figures for the unorganized sector for years other than 1978–79 and 1984–85 are essentially extrapolations based in some cases (e.g., chemicals) on growth rates of medium and large-scale enterprises.

Value ($ thousands)

575,748,864

285,230,336 4,012,587 37,776,352 371,713 3,765,063 686,182 3,945,947 3,698,567 1,425,380 2,152,654 3,869,325 1,371,593 3,657,623 3,079,432 3,216,150 3,907,784 902,200 1,426,735 12,171,689 1,865,332 654,860 2,357,641

Total

333 Crude petroleum 776 Transistors, valves, etc 334 Petroleum products, refined 752 Automatic data proc equip 764 Telecom equip, pts, access nes 759 Office, adp machine pts, access 843 Women’s outerwear, nonknit 845 Outerwear, knit, nonelastic 778 Electrical machinery nes 653 Wovn manmade fiber fabric 851 Footwear 781 Passeng motor veh, exc buses 842 Men’s outerwear, not knit 651 Textile yarn 894 Toys, sporting goods, etc 931 Special transactions 761 Television receivers 772 Switchgear, etc, parts nes 341 Gas, natural and mfd 846 Undergarments, knitted 583 Polymerization prods, etc 036 Shellfish, fresh, frzn (continued)

49.53 0.7 6.56 0.06 0.65 0.12 0.69 0.64 0.25 0.37 0.67 0.24 0.64 0.53 0.56 0.68 0.16 0.25 2.12 0.32 0.11 0.41

100

Percentage of the Grouping Total

1980–81

88.83 27.72 43.66 2.82 17.86 7.14 45.8 41.93 9.73 21.83 33.93 2.31 46.47 22.83 37.57 12.92 16.54 10.65 36.07 46 3.88 60.73

28.95

Percentage of World

149,143,280 58,661,360 37,833,760 35,440,656 28,139,024 26,408,624 20,145,184 17,564,448 16,931,696 16,058,171 15,710,801 15,312,109 14,867,337 13,111,746 12,890,361 12,727,996 12,590,018 12,112,674 11,690,024 11,689,295 11,486,216 11,335,360

115,990,4256

Value ($ thousands)

12.91 5 3.28 3.03 2.43 2.27 1.75 1.52 1.45 1.39 1.36 1.31 1.29 1.13 1.11 1.08 1.08 1.04 1.01 1.01 0.98 0.98

100

Percentage of the Grouping Total

1994–95

76.87 39.57 44.41 31.11 27.37 31.82 55.09 56.97 24.39 52.38 48.31 6.98 56.83 42.52 45.12 11.61 58.18 20.86 33.04 56.47 17.34 66.85

25.85

Percentage of World

–4.1 21 0 38.2 15.3 29.2 12.5 11.5 19.1 15.4 10.3 18.7 10 11 10.7 10.2 20.3 16.4 0.7 14.4 22.8 11.5

5.4

Growth Rate (%)

Developing Countries and Territories Export Structure at the SITC Revision 2, Group (3-digit) Level (ranked in descending order by average 1994–95 values)

SITC Codes and Descriptions

Table 6A.1

(continued)

071 Coffee and substitutes 682 Copper, exc cement copper 762 Radio broadcast receivrs 893 Articles of plastic nes 057 Fruit, nuts, fresh, dried 793 Ships and boats, etc 763 Sound records, phonogrph 652 Cotton fabrics, woven 667 Pearl, prec and semiprec stone 821 Furniture, pts thereof 844 Undergarments, not knit 784 Motor veh parts, access nes 287 Base metal ores, conc nes 848 Headgear, nontextl clothing 773 Electr distributing equip 775 Household type equip nes 699 Base metal mfrs nes 081 Feedstuff for animls 684 Aluminum 771 Electric power machry nes 684 Veneers, plywood, etc 897 Gold, silverware, jewelry 674 Iron, steel univ, plate, sheet 034 Fish, fresh, chilled, frzn 054 Veg, etc, fresh, simply prsvd

SITC Codes and Descriptions

Table 6A.1

10,515,505 4,131,725 1,843,512 1,192,562 4,731,375 2,196,406 576,988 2,213,210 1,668,200 1,292,578 1,738,328 1,380,387 5,983,019 1,652,724 586,709 1,391,450 1,096,704 3,719,198 1,462,284 433,592 1,971,863 692,995 1,069,506 1,841,890 2,651,670

Value ($ thousands) 1.83 0.72 0.32 0.21 0.82 0.38 0.1 0.38 0.29 0.22 0.3 0.24 1.04 0.29 0.1 0.24 0.19 0.65 0.25 0.08 0.34 0.12 0.19 0.32 0.46

Percentage of the Grouping Total

1980–81

90.93 34.21 30.85 14.83 43.58 12.83 8.71 32.47 11.82 12.15 70.82 4.06 46.82 47.27 10.53 15.6 9.78 34.71 12.23 11.85 41.74 12.52 5.4 33.95 33.46

Percentage of World 11,285,549 11,001,802 10,928,070 10,595,913 10,020,302 9,928,208 9,554,841 9,387,974 9,031,229 8,799,326 8,718,471 8,440,205 8,285,372 7,691,108 7,623,860 7,305,412 7,278,092 7,135,965 7,112,237 7,064,500 7,031,476 6,990,635 6,879,186 6,503,596 6,499,577

Value ($ thousands) 0.98 0.94 0.94 0.91 0.87 0.86 0.83 0.81 0.78 0.76 0.76 0.73 0.71 0.67 0.66 0.63 0.63 0.62 0.61 0.61 0.61 0.61 0.59 0.56 0.56

Percentage of the Grouping Total

1994–95

79.23 35.24 64.79 23.72 40.16 28.27 48.66 49.95 24.73 20.61 71.28 7.51 48.53 68.33 31.3 25.59 20.96 36.37 17.17 36.66 50.14 36.54 13.92 33.46 29.16

Percentage of World

0 7.6 12.7 16.3 5.7 12.1 21.2 10.7 13 15.5 11.5 13.3 3.2 11.4 19.7 13.1 13.5 5 12.6 21.5 9.3 17.2 14.6 9.9 7.1

Growth Rate (%)

658 Textile articles nes 061 Sugar and honey 263 Cotton 672 Iron, steel, primary forms 424 Fixed veg oil, nonsoft 232 Natural rubber, gums 611 Leather 899 Other manufactured goods 885 Watches and clocks 749 Nonelec mach pts, access nes 655 Knitted, etc, fabrics 011 Meat, fresh, chilled, frzn 541 Medicinal, pharm prods 785 Cycles etc, motorized or not 641 Paper and paperboard 831 Travel goods, handbags 713 Intrnl combus piston eng 657 Spl txtl fabric, prods 971 Gold, nonmonetary, nes 898 Musical instruments, pts 741 Heating, cooling, equip 248 Wood shaped, sleepers 716 Rotating electric plant 728 Other machry for spl indus 037 Fish, etc prepd, prsvd, nes 673 Iron, steel shapes, etc 042 Rice 625 Rubber tires, tubes, etc 058 Fruit, prsvd, prepd 671 Pig iron, etc 792 Aircraft, etc. (continued)

1,346,814 9,768,251 3,204,878 564,603 2,959,280 5,141,667 1,152,544 1,415,035 1,921,774 649,218 338,892 2,006,898 965,465 546,843 650,913 1,518,851 858,375 644,682 656,370 394,018 279,191 2,404,259 650,378 480,055 916,962 992,056 2,426,554 946,462 1,502,403 1,049,079 721,941

0.23 1.7 0.56 0.1 0.51 0.89 0.2 0.25 0.33 0.11 0.06 0.35 0.17 0.09 0.11 0.26 0.15 0.11 0.11 0.07 0.05 0.42 0.11 0.08 0.16 0.17 0.42 0.16 0.26 0.18 0.13

35.26 63.69 41.72 8.14 84.41 99.01 34.47 31.69 26.52 4.06 13.74 12.71 6.61 8.56 3.07 53.72 5.6 13.9 8.03 8.65 2.41 21.27 8.79 3.01 33.93 7.07 50.4 11.55 36.98 23.66 2.68

6,496,731 6,430,261 6,424,559 6,406,057 6,254,190 6,132,523 6,112,653 5,808,721 5,768,540 5,720,690 5,520,498 5,519,433 5,465,878 5,432,229 5,336,272 5,321,568 5,288,497 5,226,684 5,175,320 5,111,873 5,038,026 4,967,907 4,827,566 4,789,714 4,633,022 4,489,973 4,371,533 4,254,191 4,094,560 4,085,164 3,918,174

0.56 0.55 0.54 0.54 0.54 0.52 0.52 0.5 0.5 0.49 0.48 0.48 0.47 0.47 0.46 0.46 0.45 0.44 0.44 0.43 0.43 0.42 0.41 0.4 0.39 0.38 0.37 0.35 0.35 0.34 0.33

56.16 49.54 53.81 23.64 85.73 97.52 44.71 37.06 34.59 10.99 55.88 14.87 8.39 33.54 8.1 55.23 10.39 30.65 25.11 18.62 14.95 19.11 21.73 8.9 55.44 17.6 65.85 20.25 35.98 40.18 5.67

11.6 –2.2 5.9 20 5.2 1.5 12.3 10.3 8.2 16.9 21.3 7.8 13.3 17.1 17.1 9.6 13.7 16.1 16.6 19.5 22.9 4.8 16 17.7 12.1 10.4 5.6 11.1 8.5 11.3 12.3

(continued)

423 Fixed veg oils, soft 522 Inorg elemnts, oxides etc 582 Prod of condensation, etc 635 Wood manufactures, nes 782 Lorries, spl motor veh nes 281 Iron ore, concentrates 292 Crude veg materials nes 659 Floor covering, etc. 743 Pumps nes, centrifuges, etc 122 Tobacco, mfd 72 Cocoa 222 Seeds for soft fixed oil 751 Office machines 678 Iron, steel, tubes, pipes, etc 562 Fertilizers, mfd 247 Other wood, rough, squared 874 Measuring, contrling instr 511 Hydrocarbons nes, derivs 697 Base metal househld equip 642 Paper, etc, precut, arts of 512 Alcohols, phenols, etc 598 Misc chem prods nes 251 Pulp and waste paper 724 Textile, leather machry 513 Carboxylic acids, etc 812 Plumbg, heatg, lightg equip

SITC Codes and Descriptions

Table 6A.1

1,091,707 1,034,542 118,819 634,913 685,682 3,494,197 1,171,834 1,487,393 307,669 416,540 3,121,661 1,339,761 588,870 1,245,923 1,313,832 3,419,678 449,795 412,339 785,557 508,615 423,330 306,067 769,389 427,222 226,793 350,999

Value ($ thousands) 0.19 0.18 0.02 0.14 0.12 0.61 0.2 0.26 0.05 0.07 0.54 0.23 0.1 0.22 0.23 0.59 0.08 0.07 0.14 0.09 0.07 0.05 0.13 0.07 0.04 0.06

Percentage of the Grouping Total

1980–81

27.93 13.95 1.72 23.38 2.95 44.32 24.24 30.69 3.53 10.83 77.7 14.84 9.19 6.91 14.58 60.68 2.92 4.37 20.2 9.54 10.48 3.19 7.57 4.3 5 10.07

Percentage of World 2,870,679 3,861,291 3,722,827 3,711,674 3,635,339 3,514,411 3,440,320 3,424,798 3,419,564 3,401,150 3,360,657 3,316,877 3,316,332 3,294,789 3,285,219 3,246,329 3,190,247 3,188,686 3,179,630 3,166,519 3,159,209 3,071,135 3,016,890 3,011,346 2,949,828 2,871,994

Value ($ thousands) 0.33 0.32 0.32 0.31 0.3 0.3 0.3 0.29 0.29 0.29 0.29 0.29 0.28 0.28 0.28 0.28 0.27 0.27 0.27 0.27 0.26 0.26 0.26 0.25 0.25 0.25

Percentage of the Grouping Total

1994–95

38.24 25.04 14.81 29.68 8.33 42.57 25.14 35.71 11.76 20.79 69.2 28.38 23.88 15.51 19.06 34.65 6.61 16.63 36.9 15.96 24.52 9.66 13.4 13.13 18.76 21.98

Percentage of World

9.3 9.8 28.3 10.6 13.2 1 7.9 5.4 18.3 16.5 –0.3 5.4 12.6 8.4 7.5 –1.2 14.6 16 10.8 14.4 15 18 11.9 15 21.1 15.5

Growth Rate (%)

695 Tools 612 Leather, etc, mfts 661 Lime, cement, bldg prods 786 Trailers, nonmotor veh, nes 847 Textile clthng access nes 892 Printed matter 744 Mechanical handling equip 322 Coal, lignite, and peat 121 Tobacco, unmfd, refuse 881 Photo appar, equip nes 736 Metalworking, mach tools 654 Other woven textile fabric 694 Steel, copper nails, nuts, etc 514 Nitrogen-function compounds 266 Synthetic fibers to spin 056 Vegetables, etc, prsvd, prepd 112 Alcoholic beverages 723 Civil engineering equip, etc 691 Structures and pts nes 533a Pigments, paints, etc 431 Processed anml veg oil, etc 523 Other inorg chemicals, etc 048 Cereal, etc, preparations 664 Glass 278 Other crude minerals 098 Edible products, preps nes 872 Medical instruments nes 074 Tea and mate 515 Org and inorg compounds, etc 666 Pottery 001 Live animals for food 665 Glassware (continued )

583,501 371,299 1,003,836 604,274 524,169 581,647 327,737 116,631 1,808,947 357,467 480,396 713062 313,777 246,587 221,483 865,584 566,328 740,166 677,228 272,608 240,181 342,832 303,181 268,820 853,327 303,906 124,254 1,385,984 170,900 365,571 719,997 342,275

0.06 0.1 0.06 0.17 0.11 0.09 0.1 0.06 0.02 0.31 0.06 0.08 0.12 0.05 0.04 0.15 0.1 0.13 0.12 0.05 0.04 0.06 0.05 0.05 0.15 0.05 0.02 0.24 0.03 0.06 0.13 0.06

9.13 34.75 23.18 15.77 31.16 7.32 2.94 0.91 48.01 8.16 3.99 21.3 10.95 4.44 7.46 30.5 6.25 4.52 8.08 5.8 19.75 7.34 9.23 7.36 19.74 10.16 4.01 85.96 3.19 16.06 14.32 10.6

2,860,601 2,843,260 2,816,067 2,797,539 2,779,095 2,705,037 2,702,610 2,662,471 2,643,243 2,556,647 2,503,262 2,430,167 2,397,861 2,386,018 2,301,374 2,293,188 2,289,481 2,215,026 2,206,431 2,206,271 2,020,281 1,990,241 1,956,606 1,904,830 1,889,295 1,871,474 1,813,180 1,786,527 1,781,827 1,758,769 1,749,904 1,662,308

0.25 0.24 0.24 0.24 0.23 0.23 0.23 0.23 0.23 0.22 0.21 0.21 0.21 0.2 0.2 20 0.2 0.19 0.19 0.17 0.17 0.17 0.16 0.16 0.16 0.16 0.16 0.16 0.15 0.15 0.15 0.14

18.25 51.24 28.25 31.91 43.86 11.54 8.95 14.68 52.21 24.27 9.98 25.25 25.34 9.7 38.65 37.36 8.93 10.16 18.29 12.6 52.66 19.81 13.19 13.64 27.49 13.2 11.93 80.04 9.89 35.38 18.4 18.72

12 15.7 8.2 12.5 12.3 11.5 16.6 23.9 3.5 15.1 13.1 8.4 16 17.3 18.1 7.5 10.5 8.9 11.8 15.7 16.6 13.4 14.5 15.6 6.2 13.7 20 1.5 17.5 11.3 8.1 11.7

(continued)

531 Synth dye, nat indigo, lakes 554a Soap, cleansing, etc preps 553a Perfumery, cosmetics, etc 895 Office supplies nes 516 Other organic chemicals 714 Engines and motors nes 871 Optical instruments 656 Lace, ribbons, tulle, etc 288 Nonferrous metal, scrap nes 291 Crude animals matls nes 663 Mineral manufactures nes 044 Maize, unmilled 745 Nonelec machry, tools nes 884 Optical goods nes 014 Meat, prepd, prsvd, etc, nes 75 Spices 041 Wheat, etc, unmilled 882 Photo, cinema supplies 268 Wool (exc tops), animal hair 696 Cutlery 681 Silver, platinum, etc 335 Residual petroleum prod nes 628 Rubber articles nes 686 Zinc 687 Tin 783 Road motor veh nes

SITC Codes and Descriptions

Table 6A.1

101,904 319,443 234,354 143,202 154,217 238,355 37,517 184,864 523,365 385,197 254,470 1,490,796 161,200 249,041 706,613 808,729 878,630 168,701 713,431 292,561 859,213 2,141,242 170,061 245,292 2,570,002 258,267

Value ($ thousands) 0.02 0.06 0.04 0.02 0.03 0.04 0.01 0.03 0.09 0.07 0.04 0.26 0.03 0.04 0.12 0.14 0.15 0.03 0.12 0.05 0.15 0.37 0.03 0.04 0.45 0.04

Percentage of the Grouping Total

1980–81

3.34 10.06 8.03 8.64 4.63 3.02 3.23 13.91 15.08 29.14 6.7 12.64 1.97 10.58 30.38 84.72 5.4 2.73 15.83 19.92 12.9 36.33 8.44 14.44 87.19 6.41

Percentage of World 1,584,382 1,564,797 1,554,073 1,530,550 1,490,867 1,468,624 1,443,901 1,422,866 1,403,757 1,367,982 1,367,414 1,352,530 1,318,300 1,306,402 1,295,876 1,284,644 1,248,248 1,240,592 1,218,410 1,217,991 1,206,707 1,206,643 1,124,395 1,089,786 1,069,963 1,061,320

Value ($ thousands) 0.14 0.13 0.13 0.13 0.13 0.13 0.12 0.12 0.12 0.12 0.12 0.12 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.1 0.1 0.1 0.09 0.09 0.09

Percentage of the Grouping Total

1994–95

16.26 15.13 9.13 24.82 14.84 5.66 20.68 37.42 17.43 40.7 11.82 14.6 6.57 20.34 23.58 75.36 8.09 7.78 22.21 35.26 15.8 21.81 13.76 27.4 79.55 7.72

Percentage of World

20.1 12 14.4 20.8 18.1 12.8 29.6 15.8 7.6 10.9 12.9 –0.8 15.5 13.1 5 4.1 3.5 15 2.9 10.7 3.3 –2.6 14.6 12.7 –5.6 9.5

Growth Rate (%)

551,846,791

239,265 251,273 1,498,512 199,721 279,795 167,763 98,213 171,278 369,831 137,281 168,636 55,517 336,926 55,316 79,302 208,289 61,798 93,695 62,271 147,326 69,340 105,828 19,000 141,250 125,778 41,767 595,745

Source: UNCTAD (2000, table 4.3, 170–73). Note: Products in boldface are reserved for small-scale sector. a Some products in the group are reserved for small-scale sector.

SITC total

591 Pesticides, disinfectants 662 Clay, refactory bldg prod 271 Fertilizers, crude 692 Metal tanks, boxes, etc 693 Wire prods, nonelectric 742 Pumps for liquids, etc 582 Starch, insulin, gluten, etc 273 Stone, sand, and gravel 211 Hides, skins, exc furs, raw 62 Sugar preps, non-chocolate 111 Nonalcoholic beverages nes 679 Iron, steel castings, unworked 689 Nonferrous base metals nes 524 Radioactive, etc material 677 Iron, silver wire (excl wire rods) 551 Essential oils, perfume, etc 351 Electric current 621 Materials of rubber 737 Metalworking machry nes 282 Iron and steel scrap 233 Rubber synthtic, reclaimed 022 Milk and cream 323 Briquets, coke, semi-coke 35 Fish, salted, dried, smoked 721 Agric machry, exc tractors 261 Silk 896 Works of art, etc

0.04 0.04 0.26 0.03 0.05 0.03 0.02 0.03 0.06 0.02 0.03 0.01 0.06 0.01 0.01 0.04 0.01 0.02 0.01 0.03 0.01 0.02 0 0.02 0.02 0.01 0.1

5.58 5.79 60.76 8.15 12.66 2.91 5.66 14.66 14.35 15.45 19.01 5.27 21.55 1.61 4.38 13 2.98 5.6 1.76 5.86 2.31 2.14 0.79 10.64 2.22 75.33 18.99 1,151,810,554

1,025,506 1,010,590 9,76,848 969,024 941,926 933,497 924,132 902,733 881,877 869,510 863,920 827,767 806,492 789,635 745,547 703,708 698,457 680,498 671,779 660,738 634,177 623,335 588,140 540,533 449,812 443,348 430,721

0.09 0.09 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.07 0.07 0.07 0.07 0.06 0.06 0.06 0.06 0.06 0.06 0.05 0.05 0.05 0.05 0.04 0.04 0.04

10.51 10.14 75.95 15.36 22.18 5.81 12.95 25.88 14.5 23.58 20.61 23.72 25.83 13.54 19.03 13.6 9.83 11.64 7.93 9.05 11.24 5.02 26.99 20.71 4.62 87.64 6.96

11.4 10.4 –2.2 12.1 10 13.5 17.2 13.2 6.9 14.8 13 22.6 4.2 22.7 16.7 8.7 19.1 14.6 18.6 12.1 18.2 13.6 27.9 9.5 9.5 18.3 –1.4

371,713 37,517 686,182 118,819 19,000 116,631 279,191 654,860 55,316 555,517 433,592 338,892 576,988 226,793 4,012,587 143,202 902,200 101,904 564,603 124,254 586,709 394,018 1,425,380

575,748,864

Total

752 Automatic data proc equip 871 Optical instruments 759 Office, adp machine pts, access 582 Prods of condensation, etc 323 Briquets, coke, semi-coke 322 Coal, lignite, and peat 741 Heating, cooling equip 583 Polymerization, prods, etc 524 Radioactive etc material 679 Iron, steel castings, unworked 771 Electric power machry nes 655 Knitted, etc, fabrics 763 Sound records, phonograph 513 Carboxylic acids, etc 776 Transistors, valves, etc 895 Office supplies nes 761 Television receivrs 531 Synth dye, nat indigo, lakes 672 Iron, steel primary forms 872 Medical instruments nes 773 Electr distributing equip 898 Musical instruments, pts 778 Electrical machry nes

Value ($ thousands)

0.06 0.01 0.12 0.02 0 0.06 0.05 0.11 0.01 0.01 0.08 0.06 0.1 0.04 0.7 0.02 0.16 0.02 0.1 0.02 0.1 0.07 0.25

100

Percentage of the Grouping Total

1980–81

2.82 3.23 7.14 1.72 0.79 0.91 2.41 3.88 1.61 5.27 11.85 13.74 8.71 5 27.72 8.84 16.54 3.34 8.14 4.01 10.53 8.65 9.73

28.95

Percentage of World

35,440,656 1,443,901 26,408,624 3,722,827 588,140 2,662,471 5,038,026 11,486,216 789,635 827,767 7,064,600 5,520,498 9,554,841 2,949,828 58,661,360 1,530,550 12,590,018 1,584,882 6,406,057 1,813,180 7,623,860 5,111,873 16,931,696

1,159,904,256

Value ($ thousands)

3.03 0.12 2.27 0.32 0.05 0.23 0.43 0.98 0.07 0.07 0.61 0.48 0.83 0.25 5 0.13 1.08 0.14 0.54 0.16 0.66 0.43 1.45

100

Percentage of the Grouping Total

1994–95

31.11 20.68 31.82 14.81 26.99 14.68 14.95 17.34 13.54 23.72 36.66 55.88 48.66 18.76 39.57 24.82 58.18 16.26 23.64 11.93 31.3 18.62 24.39

25.85

Percentage of World

3.82 29.6 29.2 28.3 27.9 23.9 22.9 22.8 22.7 22.6 21.5 21.3 21.2 21.1 21 20.8 20.3 20.1 20 20 19.7 19.5 19.1

5.4

Growth Rate (%)

Developing Countries and Territories Export Structure at the SITC Revision 2, Group (3-digit) Level (ranked by average 1980–95 growth rates in descending order)

SITC Codes and Descriptions

Table 6A.2

351 Electric current 781 Passeng motor veh, exc buses 737 Metalworking machry nes 743 Pumps nes, centrifuges, etc 261 Silk 233 Rubber synthtic, reclaimd 266 Synthetic fibres to spin 516 Other organic chemicals 598 Miscel chem products nes 728 Oth machy for spcl indus 515 Org and inorg compounds, etc 514 Nitrogen-function compounds 897 Gold, silverware, jewelry 592 Starch, inulin, gluten, etc 785 Cycles etc, motrzd or not 611 Paper and paperboard 749 Nonelec mach pts, acc nes 677 Iron, steel wire (excl wire rod) 971 Gold, nonmonetary nes 744 Mechanical handling equip 431 Procesed anml veg oil, etc 122 Tobacco, mfd 772 Switchgear etc, parts nes 893 Articles of plastic nes 657 Special txtl fabrc, prods 716 Rotating electric plant 511 Hydrocarbons nes, derivs 694 Steel, copper nails, nuts, etc 656 Lace, ribbons, tulle, etc 612 Leather etc manufactures 533** Pigments, paints, etc (continued)

61,798 1,371,593 62,271 307,669 41,767 69,340 221,483 154,217 306,067 480,055 170,900 246,587 692,995 98,213 546,843 650,913 649,218 79,503 656,370 327,737 240,181 416,530 1,426,735 1,192,562 644,682 650,378 412,339 313,777 184,464 371,299 272,608

0.01 0.24 0.01 0.05 0.01 0.01 0.04 0.03 0.05 0.08 0.03 0.05 0.12 0.02 0.09 0.11 0.11 0.01 0.11 0.1 0.04 0.07 0.25 0.21 0.11 0.11 0.07 0.12 0.03 0.1 0.05

2.98 2.31 1.76 3.53 75.33 2.31 7.46 4.63 3.19 3.01 3.19 4.44 12.52 5.66 8.56 3.07 4.06 4.38 8.03 2.94 19.75 10.83 10.65 14.83 13.9 8.79 4.37 10.95 13.91 34.75 5.8

698,457 15,312,109 671,779 3,419,564 443,348 634,177 2,301,374 1,490,867 3,071,135 4,789,714 1,781,827 2,386,018 6,990,635 924,132 5,432,229 5,336,272 5,720,690 745,547 5,175,320 2,702,610 2,020,281 3,401,150 12,112,674 10,595,913 5,226,684 4,827,566 3,188,686 2,397,861 1,422,866 2,843,260 2,206,271

0.06 1.31 0.06 0.29 0.04 0.05 0.2 0.13 0.26 0.4 0.15 0.2 0.61 0.08 0.47 0.46 0.49 0.06 0.44 0.23 0.17 0.29 1.04 0.91 0.44 0.41 0.27 0.21 0.12 0.24 0.17

9.83 6.98 7.93 11.76 87.64 11.24 38.65 14.84 9.66 8.9 9.89 9.7 36.54 12.95 33.54 8.1 10.99 19.03 25.11 8.95 52.66 20.79 20.86 23.72 30.65 21.73 16.63 25.34 37.42 51.24 12.6

19.1 18.7 18.6 18.5 18.3 18.2 18.1 18.1 18 17.7 17.5 17.3 17.2 17.2 17.1 17.1 16.9 16.7 16.6 16.6 16.6 16.5 16.5 16.3 16.1 16 16 16 15.8 15.7 15.7

(continued)

664 Glass 821 Furniture, parts thereof 912 Plumbg, heatng, lghtng equip 745 Nonelec machry, tools nes 653 Woven man-made fiber fabric 764 Telecom equipt, pts, acc nes 881 Photo apparat, equip nes 512 Alcohols, phenols, etc 724 Textile, leather machnry 882 Photo, cinema supplies 62 Sugar preps non-chocolate 674 Iron, steel univ, plate, sheet 874 Measurng, controlng instr 628 Rubber articles nes 621 Materials of rubber 048 Cereal, etc preparations 846 Undergarments, knitted 642 Paper, etc, precut, arts of 553** Perfumery, cosmetics, etc 713 Intrnl combus piston engin 098 Edible prodcts, preps nes 022 Milk and cream 699 Base metal mfrs nes 742 Pumps for liquids, etc 523 Othr inorg chemicals, etc 784 Motor veh parts, acces nes

SITC Codes and Descriptions

Table 6A.2

268820 1,292,578 350,999 161,200 2,152,654 3,765,063 357,467 423,330 427,222 168,701 137,281 1,069,506 449,795 170,061 93,695 303,181 1,865,332 508,615 234,354 858,375 303,906 105,828 1,096,704 167,763 342,832 1,380,387

Value ($ thousands) 0.05 0.22 0.06 0.08 0.37 0.65 0.31 0.07 0.07 0.03 0.02 0.19 0.08 0.03 0.02 0.05 0.32 0.09 0.04 0.15 0.05 0.02 0.19 0.03 0.06 0.24

Percentage of the Grouping Total

1980–81

7.36 12.15 10.07 1.97 21.83 17.86 8.16 10.48 4.3 2.73 15.45 5.4 2.92 8.44 5.6 9.23 46 9.54 8.03 5.6 10.16 2.14 9.78 2.91 7.34 4.06

Percentage of World 1,904,830 8,799,326 2,871,994 1,318,300 16,058,171 28,139,024 2,556,647 3,159,209 3,011,346 1,240,592 869,510 6,879,186 3,190,247 1,124,395 680,498 1,956,606 11,689,295 3,166,519 1,554,073 5,288,497 1,871,474 623,335 7,278,092 933,497 1,990,241 8,440,205

Value ($ thousands) 0.16 0.76 0.25 0.11 1.39 2.43 0.22 0.26 0.25 0.11 0.07 0.59 0.27 0.1 0.06 0.16 1.01 0.27 0.13 0.45 0.16 0.05 0.63 0.08 0.17 0.73

Percentage of the Grouping Total

1994–95

13.64 20.61 21.98 6.57 52.38 27.37 24.27 24.52 13.13 7.78 23.58 13.92 6.61 13.76 11.64 13.19 56.47 15.96 9.13 10.39 13.2 5.02 20.96 5.81 19.81 7.51

Percentage of World

15.6 15.5 15.5 15.5 15.4 15.3 15.1 15 15 15 14.8 14.6 14.6 14.6 14.6 14.5 14.4 14.4 14.4 13.7 13.7 13.6 13.5 13.5 13.4 13.3

Growth Rate (%)

541 Medicinal, pharm products 782 Lorries, spcl mtr veh nes 273 Stone, sand and gravel 775 Household type equip nes 736 Metalworking mach-tools 884 Optical goods nes 667 Pearl, prec, semiprec stone 111 Non-alcoholic beverages nes 663 Mineral manufactures nes 714 Engines and motors nes 762** Radio broadcast receivrs 686 Zinc 684 Aluminum 751 Office machine 843 Women’s outerwear, nonknit 786 Trailers, nonmotr veh, nes 611 Leather 792 Aircraft, etc 847 Textile clthng acces nes 793 Ships and boats, etc 037 Fish, etc, prepd, prsvd nes 692 Metal tanks boxes, etc 282 Iron and steel scrap 695 Tools 554** Soap, cleansing etc preps 251 Pulp and waste paper 691 Structures and parts nes 665 Glassware 658 Textile articles nes 845 Outerwear knit, nonelastc 036 Shellfish, fresh, frzn (continued)

965,465 685,682 171,278 1,391,450 480,396 249,041 1,668,200 168,636 254,470 238,355 1,843,512 245,292 1,462,284 588,870 3,945,947 604,274 1,152,544 721,941 524,169 2,196,406 916,962 199,721 147,326 583,501 319,443 769389 677,228 342,375 1,346,814 3,698,567 2,357,641

0.17 0.12 0.03 0.24 0.06 0.04 0.29 0.03 0.04 0.04 0.32 0.04 0.25 0.1 0.69 0.17 0.2 0.13 0.11 0.38 0.16 0.03 0.03 0.06 0.06 0.13 0.12 0.06 0.23 0.64 0.41

6.61 2.95 14.66 15.6 3.99 10.58 11.82 19.01 6.7 3.02 30.85 14.44 12.23 9.19 45.8 15.77 34.47 2.68 31.16 12.83 33.93 8.15 5.86 9.13 10.06 7.57 8.08 10.6 35.26 41.93 60.73

5,465,878 3,635,339 902,733 7,305,412 2,503,262 1,306,402 9,031,229 863,920 1,367,414 1,468,624 10,928,070 1,089,786 7,112,237 3,316,332 20,145,184 2,797,539 6,112,653 3,918,174 2,779,095 9,928,208 4,633,022 969,024 660,738 2,860,601 1,564,797 3,016,890 2,206,431 1,662,308 6,496,731 17,564,448 11,335,360

0.47 0.3 0.08 0.63 0.21 0.11 0.78 0.07 0.12 0.13 0.94 0.09 0.61 0.28 1.75 0.24 0.52 0.33 0.23 0.86 0.39 0.08 0.06 0.25 0.13 0.26 0.19 0.14 0.56 1.52 0.98

8.39 8.33 25.88 25.59 9.98 20.34 24.73 20.61 11.82 5.66 64.79 27.4 17.17 23.88 55.09 31.91 44.71 5.67 43.86 28.27 55.44 15.36 9.05 18.25 15.13 13.4 18.29 18.72 56.16 56.97 66.85

13.3 13.2 13.2 13.1 13.1 13.1 13 13 12.9 12.8 12.7 12.7 12.6 12.6 12.5 12.5 12.3 12.3 12.3 12.1 12.1 12.1 12.1 12 12 11.9 11.8 11.7 11.6 11.5 11.5

(continued)

844 Undergarments, not knit 892 Printed matter 848 Headgear, nontxtl clothng 591 Pesticides, disinfectants 671 Pig iron, etc 666 Pottery 625 Rubber tyres, tubes, etc 651 Textile yarn 291 Crude animal mtrials nes 697 Base mtl household equip 894 Toys, sporting goods, etc 652 Cotton fabrics, woven 696 Cutlery 635 Wood manufactures nes 112 Alcoholic beverages 673 Iron, steel shapes, etc 662 Clay, refactory bldg prod 851 Footwear 899 Other manufactured goods 931 Special transactions 842 Mens outerwear, not knit 693 Wire products non electr 034 Fish, fresh, chilled, frzn 522 Inorg elemts, oxides, etc 831 Travel goods, handbags 783 Road motor vehicles nes 35 Fish, salted, dried, smoked

SITC Codes and Descriptions

Table 6A.2

1,738,328 581,647 1,652,724 239,265 1,049,079 365,571 946,462 3,079,432 385,197 785,557 3,216,150 2,213,210 292,561 834,913 566,328 992,056 251,273 3,869,325 1,415,035 3,907,784 3,657,623 279,795 1,841,890 1,034,542 1,518,851 258,267 141,250

Value ($ thousands) 0.3 0.09 0.29 0.04 0.18 0.06 0.16 0.53 0.07 0.14 0.56 0.38 0.05 0.14 0.1 0.17 0.04 0.67 0.25 0.68 0.64 0.05 0.32 0.18 0.26 0.04 0.02

Percentage of the Grouping Total

1980–81

70.82 7.32 47.27 5.58 23.66 16.06 11.55 22.83 29.14 20.2 37.57 32.47 19.92 23.38 6.25 7.07 5.79 33.93 31.69 12.92 46.47 12.66 33.95 13.95 53.72 6.41 10.64

Percentage of World 8,718,471 2,705,037 7,691,108 1,025,506 4,085,164 1,758,769 4,254,191 13,111,746 1,367,982 3,179,630 12,890,361 9,387,974 1,217,991 3,711,674 2,289,481 4,489,973 1,010,590 15,710,801 5,808,721 12,727,996 14,867,337 941,926 6,503,596 3,861,291 5,321,568 1,061,320 540,533

Value ($ thousands) 0.76 0.23 0.67 0.09 0.34 0.15 0.35 1.13 0.12 0.27 1.11 0.81 0.11 0.31 0.2 0.38 0.09 1.36 0.5 1.08 1.29 0.08 0.56 0.32 0.46 0.09 0.05

Percentage of the Grouping Total

1994–95

71.28 11.54 68.33 10.51 40.18 35.38 20.25 42.52 40.7 36.9 45.12 49.95 35.26 29.68 8.93 17.6 10.14 48.31 37.06 11.61 56.83 22.18 33.46 25.04 55.23 7.72 20.71

Percentage of World

11.5 11.5 11.4 11.4 11.3 11.3 11.1 11 10.9 10.8 10.7 10.7 10.7 10.6 10.5 10.4 10.4 10.3 10.3 10.2 10 10 9.9 9.8 9.6 9.5 9.5

Growth Rate (%)

721 Agric machy, exc tractors 634 Veneers, plywood, etc 423 Fixed veg oils, soft 723 Civil engneerg equip, etc 551 Essentl oils, perfume, etc 058 Fruit prsvd, prepd 678 Iron, stl, tubes, pipes, etc 654 Oth woven textile fabric 885 Watches and clocks 661 Lime, cement, bldg prods 001 Live animals for food 292 Crude veg materials nes 011 Meat, fresh, chilled, frzn 682 Copper exc cement copper 288 Nonferr metal scrap nes 562 Fertilizers, manufactured 056 Vegetbles, etc, prsvd, prepd 054 Veg, etc, fresh, simply prsvd 211 Hides, skins, exc furs, raw 278 Other crude minerals 263 Cotton 057 Fruit, nuts, fresh, dried 042 Rice 659 Floor covering, etc 222 Seeds for soft fixed oil 424 Fixed veg oil, nonsoft 081 Feeding stuff for animals 014 Meat, prepd, prsvd, nes, etc 248 Wood shaped, sleepers 689 Non-ferrous base metals nes 75 Spices 121 Tobacco unmfd, refuse 041 Wheat, etc, unmilled

125,778 1,971,863 1,091,707 740,166 208,289 1,502,403 1,245,923 713,062 1,921,774 1,003,836 719,997 1,171,834 2,006,898 4,131,725 523,365 1,313,832 865,584 2,651,670 369,831 853,327 3,204,878 4,731,375 2,426,554 1,487,393 1,339,761 2,959,280 3,719,198 706,613 2,404,259 336,926 808,729 1,808,947 878,630

0.02 0.34 0.19 0.13 0.04 0.26 0.22 0.08 0.33 0.06 0.13 0.2 0.35 0.72 0.09 0.23 0.15 0.46 0.06 0.15 0.56 0.82 0.42 0.26 0.23 0.51 0.65 0.12 0.42 0.06 0.14 0.02 0.15

2.22 41.74 27.93 4.52 13 36.98 6.91 21.3 26.52 23.18 14.32 24.24 12.71 34.21 15.08 14.58 30.5 33.46 14.35 19.74 41.72 43.58 50.4 30.69 14.84 84.41 34.71 30.38 21.27 21.55 84.72 48.01 5.4

449,812 7,031,476 3,870,679 2,215,026 703,708 4,094,560 3,294,789 2,430,167 5,768,540 2,816,067 1,749,904 3,440,320 5,519,433 11,001,802 1,403,757 3,285,219 2,293,188 6,499,577 881,877 1,889,295 6,424,559 10,020,302 4,371,533 3,424,798 3,316,877 6,254,190 7,135,965 1,295,876 4,967,907 806,492 1,284,644 2,643,243 1,248,248

0.04 0.61 0.33 0.19 0.06 0.35 0.28 0.21 0.5 0.24 0.15 0.3 0.48 0.94 0.12 0.28 20 0.56 0.08 0.16 0.54 0.87 0.37 0.29 0.29 0.54 0.62 0.11 0.42 0.07 0.11 0.23 0.11

4.62 50.14 38.24 10.16 13.6 35.98 15.51 25.25 34.59 28.25 18.4 25.14 14.87 35.24 17.43 19.06 37.36 29.16 14.5 27.49 53.81 40.16 65.85 35.71 28.38 85.73 36.37 23.58 19.11 25.83 75.36 52.21 8.09

9.5 9.3 9.3 8.9 8.7 8.5 8.4 8.4 8.2 8.2 8.1 7.9 7.8 7.6 7.6 7.5 7.5 7.1 6.9 6.2 5.9 5.7 5.6 5.4 5.4 5.2 5 5 4.8 4.2 4.1 3.5 3.5

859,213 5,983,019 713,431 5,141,667 1,385,984 3,494,197 12,171,689 37,766,352 10,515,505 3,121,661 1,490,796 3419678 595,745 9,768,251 1,498,512 2,141,242 285,230,336 2,570,002

551,846,791

64,989,890

681 Silver, patinum, etc 287 Base metal ores, conc nes 268 Wool (exc tops), anml hair 232 Natural rubber, gums 74 Tea and mate 281 Iron ore, concentrates 341 Gas, natural and mfd 334 Petroleum products, refin 071 Coffee and substitutes 72 Cocoa 044 Maize unmilled 247 Oth wood rough, squared 896 Works of art, etc 061 Sugar and honey 271 Fertilizers, crude 335 Residual petroleum prod nes 333 Crude petroleum 687 Tin

SITC total

SSI reserved total

0.15 1.04 0.12 0.89 0.24 0.61 2.12 6.56 1.83 0.54 0.26 0.59 0.1 1.7 0.26 0.37 49.53 0.45

Percentage of the Grouping Total

1980–81

12.9 46.82 15.83 99.01 85.96 44.32 36.07 43.66 90.93 77.7 12.64 60.68 18.99 63.69 60.76 36.33 88.83 87.19

Percentage of World

329,108,461

1,151,810,554*

1,206,707 8,285,372 1,218,410 6,132,523 1,786,527 3,514,411 11,690,024 37,833,760 11,285,549 3,360,657 1,352,530 3,246,329 430,721 6,430,261 976,858 1,206,643 149,143,280 1,069,963

Value ($ thousands) 0.1 0.71 0.11 0.52 0.16 0.3 1.01 3.28 0.98 0.29 0.12 0.28 0.04 0.55 0.08 0.1 12.91 0.09

Percentage of the Grouping Total

1994–95

15.8 48.53 22.21 97.52 80.04 42.57 33.04 44.41 79.23 69.2 14.6 34.85 6.96 49.54 75.95 21.81 76.87 79.55

Percentage of World

Source: UNCTAD (2000, table 4.3, 170–73). Notes: Total product groups  189. Product groups under SSI reserved category  68. Products in boldface are reserved for small-scale sector. a 99.3 percent of world export. b Some products in the group are reserved for small-scale sector.

Reserved group exports as % of World exports  28.37

Value ($ thousands)

(continued)

SITC Codes and Descriptions

Table 6A.2

3.3 3.2 2.9 1.5 1.5 1 0.7 0 0 –0.3 –0.8 –1.2 –1.4 –2.2 –2.2 –2.6 –4.1 –5.6

Growth Rate (%)

Fixed veg oil, nonsoft

Medicinal, pharm products Leather

424

541

(continued)

611

81

71

57

54

36

34

Meat, fresh, chilled, frzn Fish, fresh, chilled, frzn Shellfish, fresh,, frzn Veg, etc, fresh, simply prsvd Fruit, nuts fresh, dried Coffee and substitutes Feeding stuff for animals

Description

India (11.98)

Indonesia (17.19)

Brazil (17.85) China (12.05) Indonesia (9.40) China (20.04) Chile (8.62) Colombia (17.97) Argentina (18.24)

Rank II

Korea, Argentina Republic of (13.31) (23.13)

China (25.31)

Malaysia (55.48)

Taiwan (26.20) Taiwan (15.0) Thailand (20.90) Mexico (24.09) Turkey (11.43) Brazil (22.36) Brazil (30.38)

Rank I

Taiwan Province of China (13.28)

The Philippines (10.41) Singapore (10.03)

China (14.98) Korea (11.09) China (8.94) Thailand (8.90) Ecuador (7.80) Indonesia (6.06) Peru (10.55)

Rank III

Brazil (8.32)

Mexico (6.38)

China (4.07)

Argentina (13.37) Chile (9.90) India (7.66) Turkey (7.10) Costa Rica (7.24) Mexico (5.56) India (9.03)

Rank IV

India (6.15)

Slovenia (5.51)

India (2.63)

Thailand (7.44) Argentina (6.67) Ecuador (5.45) Argentina (4.07) Iran (5.62) Guatemala (3.80) Chile (7.92)

Rank V

Papua New Guinea (1.81) Korea, Republic of (4.37) China (5.12)

Uruguay (4.45) Indonesia (6.05) Mexico (4.03) Chile (2.67) Mexico (5.24) India (3.47) China (5.73)

Rank VI

Pakistan (4.21)

Bahamas (3.32)

Singapore (1.45)

India (2.80) Thailand (5.83) Morocco (3.90) Morocco (2.45) Colombia (4.73) Viet Nam (3.25) Thailand (3.09)

Rank VII

Thailand (4.06)

Cuba (2.97)

Core d’Ivoire (1.24)

Korea (1.68) Singapore (4.18) Korea (3.20) Taiwan (2.17) India (4.49) Costa Rica (3.22) Indonesia (2.10)

Rank VIII

Bangladesh (2.68)

Brazil (2.75)

Bahrain (0.67)

Nicaragua (1.09) India (2.83) Vietnam (3.14) India (2.09) China (4.20) Uganda (3.09) Malaysia (1.78)

Rank IX

Ten Leading Developing Exporting Countries Ranked for Commodity Groups at the SITC Revision 2, Group (3-digit) Level (ranked by average 1994–95 values)

11

SITC Codes

Table 6A.3

Uruguay (2.30)

Croatia (2.47)

Paraguay (0.94) Micronesia (1.49) Bahamas (3.13) Egypt (1.76) Argentina (3.60) Ecuador (2.91) American Samoa (1.58) Turkey (0.54)

Rank X

Description

Textile Yarn

Cotton fabric, woven

Woven manmade fiber fabric

Knitted or crocheted fabric

Textile articles nes

Pearl, prec, semiprec stone

Iron, steel univ, plate, sheet

Base metal mfrs nes

651

652

653

655

658

667

674

699

Taiwan Province of China (33.89)

Korea, republic of (34.71)

India (48.76)

Taiwan Province of China (39.66) China (36.53)

Korea republic of (38.43)

Taiwan Province of China (18.06) China (32.85)

Rank I

(continued)

SITC Codes

Table 6A.3

China (17.92)

Brazil (14.44)

Thailand (12.50)

Pakistan (10.29)

Taiwan Province of China (17.59) Korea republic of (21.27)

Pakistan (10.84)

China (14.83)

Rank II

Taiwan Province of China (13.25) Mexico (11.04)

China (4.99)

India (9.15)

China (14.97)

China (14.39)

India (9.98)

Pakistan (11.71)

Rank III

Taiwan Province of China (5.69) Pakistan (3.60)

Korea republic of (9.13)

Rank V

Singapore (6.63)

China (6.43)

Colombia (4.89)

Korea, Republic of (6.08)

Mexico (5.98)

(5.78) Ghana (3.36)

China, Singapore Hong Kong (3.76) SAR (8.05) Turkey Korea (6.89) republic of

China, Hong Kong SAR (9.08) Indonesia (7.15)

India (9.15)

Rank IV

Malaysia (3.89)

Kazakhstan (4.03)

Liberia (2.62)

Mexico (4.90) of China

Malaysia (2.32)

Thailand (3.54)

Korea, Republic of (5.27)

Indonesia (5.69)

Rank VI

China, Hong Kong SAR (3.55)

Singapore (3.92)

Sri Lanka (2.50)

Taiwan Province

Turkey (2.04)

Singapore (3.32)

Turkey (3.80)

Turkey (4.21)

Rank VII

Thailand (2.92)

Democratic Republic of the Congo (2.33) Argentina (2.70)

Brazil (3.80) (4.85)

Pakistan (1.68)

Turkey (2.72)

Indonesia (3.42)

Thailand (4.09)

Rank VIII

India (2.25)

India (2.54)

Congo (1.62)

Thailand (2.73)

India (1.28)

India (2.41)

Thailand (2.65)

Malaysia (3.21)

Rank IX

Brazil (1.96)

Venezuela (2.37)

French Polynesia (1.59)

Indonesia (2.71)

Philippines (0.85)

Malaysia (1.57)

Brazil (2.33)

Egypt (2.71)

Rank X

Other machry for spcl indus

Passeng motor veh, exc buses Motor veh prts, acce nes

728

781

Travel goods, handbags

Women outerwear, nonknit

Undergarments, not knit

831

843

844

(continued)

Cycles, etc, motrzd or not

785

784

Intrnl combus piston engine

713

China (25.76)

China (32.07)

Taiwan Province of China (53.18) China (49.18)

Mexico (28.49)

Taiwan Province of China (33.08) Mexico (41.09)

Mexico (47.65)

China, Hong Kong SAR (12.95) China, Hong Kong SAR (11.87)

Korea, Republic of (13.04)

China (14.92)

Korea republic of (38.25) Brazil (17.15)

Singapore (19.49)

Brazil (16.80)

India (8.90)

India (7.39)

Taiwan Province of China (8.53)

Taiwan Province of China (14.84) Singapore (7.30)

Slovenia (3.96)

Korea republic of (16.20)

Singapore (8.00)

China (3.85)

Taiwan Province of China (3.47) Malaysia (3.56)

Bangladesh (6.51)

Thailand (5.66)

Thailand (8.19)

Thailand (6.28)

Korea, republic of (5.64)

Turkey (5.13)

India (5.29)

India (5.08)

Thailand (5.36)

Korea, Republic of (4.77)

China, Hong Kong SAR (1.93)

Indonesia (4.51)

China, Hong Kong SAR (5.28) Brazil Oman United Arab (3.23) (2.17) Emirates (1.46) Korea, Argentina Singapore Republic of (5.94) (4.03) (6.92)

China (7.50)

Korea Republic of (4.45)

Indonesia (4.29)

Indonesia (4.34)

Mexico (1.76)

Korea, Republic of (3.13)

China (3.72)

Venezuela (1.42)

Mexico (3.25)

Argentina (3.18)

Turkey (3.69)

Mexico (3.06)

Philippines (1.75)

Malaysia (2.43)

India (3.13)

Argentina (1.37)

Brazil (2.19)

Malaysia (2.97)

Taiwan Province of China (3.50)

Sri Lanka (2.76)

Viet Nam (1.65)

Brazil (0.68)

Thailand (3.09)

Singapore (1.07)

Thailand (1.37)

India (2.13)

Malaysia (2.45)

Tunisia (2.54)

Indonesia (1.31)

Mexico (0.65)

Philippines (2.22)

India (1.07)

India (1.36)

Turkey (1.86)

Headgear, nontxtl clothing

Footwear

Articles of plastic nes

Gold, silverware, jewelry

Musical instruments, pts

848

851

893

897

898

China (21.65)

Indonesia (12.24)

Malaysia (11.11)

Turkey (9.65)

Rank II

China, Hong Kong SAR (12.62) Korea, Singapore Republic of (15.11) (25.27)

Taiwan Province of China (26.14) China (18.98)

China (38.13)

China (34.41)

China (22.44)

Rank I

(continued)

China (14.45)

Thailand (12.56)

Thailand (11.02)

Thailand (11.37)

China, Hong Kong SAR (8.19) Korea, republic of (8.67)

Rank III

Taiwan Province of China (11.61)

Indonesia (7.59)

Mexico (7.62)

Brazil (9.39)

Thailand (6.59)

Thailand (6.47)

Rank IV

Mexico (11.52)

Singapore (6.97)

Korea, Republic of (6.51)

Korea, Republic of (8.92)

Turkey (6.15)

Korea, Republic of (5.96)

Rank V

Source: UNCTAD (2000, table 4.4, 192–200). Note: Numbers in parentheses indicate percentage share of export among developing countries.

Undergarments, knitted

Description

846

SITC Codes

Table 6A.3

Malaysia (5.03)

India (6.85)

Taiwan Province of China (6.09) Taiwan Province of China (5.00) Singapore (5.94)

India (4.66)

Rank VI

Thailand (4.04)

China, Hong Kong SAR (4.02) Malaysia (6.70)

India (2.20)

India (5.78)

Indonesia (3.92)

Rank VII

India (2.86)

Korea, Republic of (4.77)

Malaysia (3.76)

Viet Nam (1.89)

Pakistan (5.02)

Mexico (3.63)

Rank VIII

Taiwan Province of China (2.78) China, Hong Kong SAR (2.46)

India (2.13)

China, Hong Kong SAR (2.73) Mexico (1.09)

Singapore (3.04)

Rank IX

Indonesia (1.91)

Argentina (2.48)

Indonesia (1.54)

Philippines (1.05)

Mexico (2.69)

Pakistan (2.93)

Rank X

Electronics Paper

Petrochemical industry Sports goods Agricultural implement Automobile components Automobile components Automobile components Automobile components Bed spread Brass & bell metal Brass goods Brass parts Brassware

3 4

5

(continued)

14 15 16

12 13

11

10

9

8

6 7

2

Artificial diamonds Automobile components

Cluster

1

Cluster Serial No.

Table 6B.1

Appendix B

Gujarat

Uttar Pradesh Himachal

Haryana

Tamilnadu

State

Bhandara Jamnagar Moradabad

Sholapur Hajo

Pune

Luthiana

Faridabad

Aurangabad

Maharashtra Gujarat Uttar Pradesh

Maharashtra Assam

Maharashtra

Punjab

Haryana

Maharashtra

Meerut Uttar Pradesh Patiala Bhadsor Punjab

Noida Solan Pradesh Vadodra

Gurgaon

Tiruchirapalli

Location

T T T

C TR

C

C

C

C

C T

C

T T

T

TR

N N N

N N

N

N

N

N

I N

I

I I

I

I

R R R

U R

R

R

R

R

R R

U

U U

R

U

Y Y Y

Y Y

N

N

N

N

Y N

N

N N

N

Y

Y Y N

Y Y

N

N

N

N

N N

N

N N

N

N

Y Y N

Y N

Y

Y

Y

Y

N Y

Y

Y Y

Y

Y

H B H

H H

B

B

B

H

H H

B

B H

V

V

Y Y Y

Y Y

Y

Y

Y

Y

Y Y

Y

Y Y

Y

Y

H H H

H H

H

H

H

H

H H

H

H H

H

H

Vertical (V), Metro (M), Traditional Large Unit Scope for City (C), Natural (N) Reserved (R) Traditional Consumer Modern Centered (L), Technology Scope for Town (T), or or Art/Craft Goods SSI or Horizontal Upgrade for Export Rural (TR) Induced (I) Unreserved (U) (Yes or No) (Yes or No) (Yes or No) Cluster (H) (Yes or No) (High or Low)

List of Clusters of Small Scale Industries in India

Handtools

35

32 33 34

31

Handmade silver jewelry Hand made paper Handicrafts Handlooms Handtools

Carpets Diamonds Electric fans Electric fans Electronics Electronics Electronics Fishing hooks Food products Food products Food products Ganesh statues Glass works

Cluster

30

29

28

27

26

25

17 18 19 20 21 22 23 24

Cluster Serial No.

Table 6B.1

Maharashtra

Himachal Pradesh Karnataka

Maharashtra

Uttar Pradesh Gujarat Punjab W. Bengal Karnataka Maharashtra Maharashtra W. Bengal

State

Pipli Panipat Agar Purulia (Dn.) Jallandhar

Jaipur

Punjab

Orissa Haryana West Bengal

Rajasthan

Shikhohabad/. Uttar Pradesh Frz Cuttack Orissa

Pen Panvel

Kullu & Sirmaur Karnataka

Pune

Bhadol Surat Ludhiana Calcutta Bangalore Pune Bombay Bankura

Location

(continued)

C

TR T T

C

C

T

T

T

T

C

TR C C M M C M TR

N

N N N

N

N

N

N

N

N

N

N N N N N N N N

R

R R R

R

R

R

U

U

U

R

R U U U U U U U

N

Y Y N

Y

Y

Y

Y

Y

N

N

Y Y N N N N N Y

N

Y Y N

Y

N

N

Y

Y

Y

Y

N N N N N N N Y

Y

N N Y

Y

N

N

N

N

Y

Y

Y Y Y Y Y Y Y N

H

H H H

H

H

H

H

H

B

B

B V L L B B B H

Y

Y Y Y

Y

N

N

Y

Y

Y

Y

N Y Y Y Y Y Y Y

H

H H H

H

H

H

H

H

H

H

H H H H H H H H

Vertical (V), Metro (M), Traditional Large Unit Scope for City (C), Natural (N) Reserved (R) Traditional Consumer Modern Centered (L), Technology Scope for Town (T), or or Art/Craft Goods SSI or Horizontal Upgrade for Export Rural (TR) Induced (I) Unreserved (U) (Yes or No) (Yes or No) (Yes or No) Cluster (H) (Yes or No) (High or Low)

Lacquar craft Lakh Leather Leather Leather Leather footwear Machine tools Machine tools Marble cutting Match box/ fire works Ready made garments Ready made garments Ready made garments Ready made garments Ready made garments Rubber goods

43

(continued)

58

57

56

55

54

53

52

51

50

49

44 45 46 47 48

Handtools Hosiery Hosiery Hosiery Hosiery Hosiery Kumkum

36 37 38 39 40 41 42

Jallandhar

Indore

Ahmedabad

Bombay

Bangalore

Delhi

Sivakasi

Kishangarh

Rajkot

Batala

Bhandara Madras Kanpur Howrah Agra

Nagaur Ludhiana Calcutta Delhi Tirupur Kanpur Kem Sholapur (Dn.) Jaipur

Punjab

Madhya Pradesh

Gujarat

Maharashtra

Karnataka

Delhi

Tamilnadu

Rajasthan

Gajarat

Punjab

Maharashtra Tamilnadu Uttar Pradesh West Bengal Uttar Pradesh

Rajasthan

Rajasthan Punjab West Bengal Delhi Tamilnadu Uttar Pradesh Maharashtra

C

C

C

M

M

M

TR

T

T

T

T M C M C

C

T C M M C C T

N

N

N

N

N

N

N

N

N

N

N N N N N

N

N N N N N N N

U

R

R

R

R

R

R

U

R

R

R R R R R

U

R R R R R R U

N

Y

Y

Y

Y

Y

Y

Y

N

N

Y Y Y Y Y

Y

N N N N N N Y

Y

Y

Y

Y

Y

Y

Y

Y

N

N

Y N N N N

Y

N N N N N N Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

Y

N Y Y Y Y

N

Y Y Y Y Y Y N

B

H

H

H

H

H

B

H

H

H

H B B H B

H

B V V V V V H

Y

Y

Y

Y

Y

Y

Y

N

Y

Y

Y Y Y Y Y

N

Y Y Y Y Y Y N

H

H

H

H

H

H

H

H

H

H

H H H H H

H

H H H H H H H

Ceramic industry Chuna Bhatti Rajura

Hyderabad Pradesh Bhandarg/ Shinna Khurja

Kullu

Amritsar

Jaipur

Sambhalpur Bankura

Sanganer

Jallandhar

Amritsar Panipat Mysore Hupparl

75

76

Rajasthan

Haryana

State

Mahrashtra

Uttar Pradesh

Maharashtra

Himachal Pradesh Arunachal

Punjab

Rajasthan

Orissa West Bengal

Rajasthan

Punjab

Punjab Haryana Karnataka Maharashtra

Coastal Kerala Kerala

Balhotra/Pali

Ambala

Location

Beedi

Scientific instruments Screen printing Seafood processing Shoddy yarn Shoddy yarn Silk Silver ornaments Sports goods Textile hand printing Tusser silk Wooden handicrafts Woolen carpets Woolen shawls Woolen shawls Electronics

Cluster

(continued)

74

73

72

71

70

68 69

67

66

62 63 64 65

61

60

9

Cluster Serial No.

Table 6B.1

T

T

TR

C

T

C

TR

TR TR

T

TR

C C C T

T

T

T

N

N

N

I

N

N

N

N N

N

N

N N N N

N

N

N

U

U

R

U

R

U

R

R R

R

R

U U R U

U

U

R

Y

Y

Y

N

Y

N

Y

Y Y

Y

Y

Y N Y Y

N

Y

N

Y

Y

Y

N

Y

Y

Y

Y Y

Y

N

Y N Y Y

N

Y

N

Y

Y

N

Y

N

Y

Y

N N

N

N

Y Y Y N

Y

Y

Y

H

H

H

L

H

L

H

H H

V

H

H B H H

B

V

B

N

Y

N

Y

N

N

N

Y N

Y

Y

Y Y Y N

Y

Y

Y

L

L

L

H

H

H

H

H H

H

H

H H H H

H

H

H

Vertical (V), Metro (M), Traditional Large Unit Scope for City (C), Natural (N) Reserved (R) Traditional Consumer Modern Centered (L), Technology Scope for Town (T), or or Art/Craft Goods SSI or Horizontal Upgrade for Export Rural (TR) Induced (I) Unreserved (U) (Yes or No) (Yes or No) (Yes or No) Cluster (H) (Yes or No) (High or Low)

Agricultural pumps Battery units Bicycle parts

98

(continued)

99 100

97

96

Chemicals Chemicals Engineering industry Engineering industry Textiles

93 94 95

92

91

90

89

88

85 86 87

Petha (sweets) Salt & salt based Stone crushers Wheat threshers Chemicals

Leather footwear (Juti) Light engineering Locks Nuts/bolts Oil paints

83

84

Essential oils Foundary Foundary Foundry Foundry Kaju

77 78 79 80 81 82

Himachal Pradesh West Bengal Haryana Gujarat

Haryana

Uttar Pradesh Mahrashtra Haryana West Bengal Uttar Pradesh Maharashtra

Howrah Luthiana

Baddi Barotiwal Coimbatore

Tiruchirapalli

Vapi/ Ankieshwar Nandesari Vashi Bhopal

Mogra

Rajoukri

Bhavnagar

W. Bengal Punjab

Himachal Pradesh Tamilnadu

Tamilnadu

Gujarat Maharashtra Madhya Pradesh

Gujarat

Punjab

Delhi

Gujarat

Howrah Rohtak Vallab Vidhyanagar Agra Uttar Pradesh

Parwanoo

Kannauj Kolhapur Samalkha Howrah Agra Vaugurla Ratnagi. Rewari

M C

C

T

C

C M C

T

T

T

T

C

M T T

T

T

T C C M C C

N N

N

I

I

I I I

I

N

N

N

N

N N N

N

N

N N N N N N

U R

R

U

U

U U U

U

R

R

U

U

U R U

U

R

R R R R R U

N N

N

N

N

N N N

N

N

N

Y

Y

Y N N

N

Y

Y N N N N Y

N N

N

N

N

N N N

N

N

N

Y

Y

Y N N

N

Y

N N N N N Y

Y Y

Y

Y

Y

Y Y Y

Y

Y

Y

Y

Y

Y Y Y

Y

N

Y Y Y Y Y Y

B L

H

H

B

L V B

B

H

H

H

H

B H B

H

H

B H B B B H

Y Y

Y

Y

Y

Y Y Y

Y

Y

Y

Y

N

Y Y Y

Y

Y

N Y Y Y Y N

M M

M

M

M

M M M

M

L

L

L

L

L L L

L

L

L L L L L L

Locks Mixies & grinders

115 116

114

113

112

111

110

109

108

107

106

Chappals Diesel engines Diesel engines Diesel engines Diesel engines Engineering industry Engineering industry Flooring tiles Gems & jewelry Induction furnace Isabgol

Cane products Cement Cement

Cluster

104 105

102 103

101

Cluster Serial No.

Table 6B.1

State

Solan & Sirmaur Unga (Mehsana) Aligarh Ambala

Jaipur

Morbi

Pinjore

Makar pura

Rajkot

Phagwara

Colmbatore

Sirohi Kala Amb (Sirmur) Kolhapur Kolhapur

Uttar Pradesh Haryana

Gujarat

Himachal

Rajasthan

Gujarat

Haryana

Gujarat

Gujarat

Punjab

Tamilnadu

Rajasthan Himachal Pradesh Maharashtra Maharashtra

Mohindergarh Haryana

Location

(continued)

T C

T Pradesh T

C

T

C

T

C

T

C

C C

T T

TR

N N

N

N

N

N

N

N

N

N

N

N N

N N

N

R U

U

U

U

U

U

U

R

R

R

R R

U U

R

Y N

Y

N

Y

N

N

N

N

N

N

Y N

N N

Y

N Y

Y

N

Y

N

N

N

N

N

N

Y N

N N

Y

N Y

N

Y

Y

Y

Y

Y

Y

Y

Y

Y Y

Y Y

N

H B

H

H

H

H

B

B

V

V

V

B V

H H

H

Y N

N

N

Y

Y

Y

Y

Y

Y

Y

N Y

N N

N

M M

M

M

M

M

M

M

M

M

M

M M

M M

M

Vertical (V), Metro (M), Traditional Large Unit Scope for City (C), Natural (N) Reserved (R) Traditional Consumer Modern Centered (L), Technology Scope for Town (T), or or Art/Craft Goods SSI or Horizontal Upgrade for Export Rural (TR) Induced (I) Unreserved (U) (Yes or No) (Yes or No) (Yes or No) Cluster (H) (Yes or No) (High or Low)

Mushroom cultivation Oil mills

Oil mills manufacture 120 Papad Mangodi, Namkeen 121 Powerlooms 122 Pwlooms texl procesng 123 Re rolling steel mills 124 Rigs 125 Rubber 126 Saree 127 Sarees 128 Sarees P.looms 129 Sewing mchn components 130 Shipping industry 131 Spinning & processing 132 Statue murticar 133 Textile processing 134 Textile processing 135 Textiles 136 Tuhar Dal 137 Well clocks 138 Weights & measures Source: Gulati (1996).

119

118

117

T T T T T C

Punjab

Tamilnadu Kerala Gujarat Maharashtra Maharahstra

Punjab

Maharashtra

Rajasthan

Rajasthan

Rajasthan

Gujarat

Gujarat Gujarat Gujarat Maharashtra

Luthiana

Malagaon

Bhilwara

Jaipur

Jodhpur

Jetpur

Surat Vasad (Kheda) Morbi Savarkundla

T T T T

C

C

C

T

T

C

T T

C

Rajasthan

Maharashtra Maharashtra

T

T

TR

Gujarat

Rajasthan

Haryana

Bhivandi Bewandi Malagarh Mandi Govindgarh Tiruchengod Kottayam Pattola Patan Paithan Nagpur

Alvar, S, Madhor Amrelli Juna Garh Bikaner

Sonipat

N N N N

N

N

N

N

N

N

N N N N N

N

N N

N

N

N

N

U U U R

U

U

U

U

U

U

R R U R U

U

R U

R

U

R

R

Y Y Y Y

Y

N

Y

N

Y

N

N Y Y Y Y

Y

Y Y

Y

Y

Y

N

N Y N N

Y

N

Y

N

Y

N

N Y Y N Y

N

Y Y

Y

N

Y

Y

Y Y N N

Y

Y

Y

Y

N

Y

Y Y Y N Y

Y

N Y

Y

Y

Y

N

B H H H

V

V

H

V

H

L

H H H H H

B

H V

H

H

H

H

Y N Y Y

Y

Y

N

Y

Y

Y

Y Y N N Y

Y

Y Y

Y

Y

Y

Y

M M M M

M

M

M

M

M

M

M M M M M

M

M M

M

M

M

M

298

Rakesh Mohan

Comment

Roger G. Noll

Rakesh Mohan has performed a valuable service by carefully documenting the economic foolishness of India’s long-standing policies to subsidize and protect small manufacturing firms. His research demonstrates that these policies have depressed India’s economic growth and, contrary to their reputed purpose, sentenced millions of entrepreneurs and employees to inescapably low incomes. Talented entrepreneurs can not expand successful businesses to an efficient scale. Workers are denied the opportunity to find the better jobs that would become available if efficient companies with high labor productivity were allowed to grow more rapidly. The most distressing feature of small-scale-industry (SSI) policies is their implicit pessimism about India’s long-run growth prospects. If the justification for these policies is taken at face value, their purpose is to cope with disguised unemployment by encouraging labor-intensive manufacturing. The most labor-intensive part of manufacturing is small-scale enterprise, which has appallingly low labor productivity. The implicit assumption behind limiting the growth of large firms is a version of the wages fund theory: Indian gross domestic product (GDP) is doomed forever to be very low, so that the relevant policy choice is how to distribute that income. Will a given low GDP be concentrated among a lucky few in high-productivity manufacturing, or spread among more workers in low-productivity jobs? Interestingly, Mohan’s devastating facts about output and incomes in manufacturing may serve only to encourage the advocates of this pessimistic view of India’s growth prospects. The pessimists may interpret India’s poor performance in comparison with its Asian neighbors as proof that India is bound to be poor, and so its industrial policy, by preserving a large number of lowwage jobs, is actually working. Mohan undoubtedly is correct in his main conclusions: Policies to promote small-scale manufacturing firms have severely restrained Indian economic development and so ought to be abolished. Moreover, his conclusion that the primary target ought to be the policy to reserve certain products exclusively for small-scale enterprises (subject to a grandfather clause for existing large firms) is also probably correct. In this comment, I will discuss how to implement this reform. Reforming India’s SSI policy has related economic and political elements. The economic component arises from the inefficiency of the status quo. As Mohan proves, SSI policies have caused major distortions in the distribution of output by firm size, industry, and region. This core fact implies that in the short run many entrepreneurs and workers could be severely Roger G. Noll is the Morris M. Doyle Professor of Public Policy in the department of economics at Stanford University and a senior fellow at the Center for Research on Economic Development and Policy Reform.

Small-Scale Industry Policy in India: A Critical Evaluation

299

hurt by the transition to a larger, more vibrant manufacturing sector. In purely economic terms, the transition is worth the cost because of the enormous differences in productivity between reserved and unreserved industries as well as between large and small firms. However, comparison of the costs and benefits of reform is the element missing from Mohan’s work, and is the next logical step in his research. Most likely, the benefits will be shown to exceed the costs by an overwhelming amount, especially given that much of the effect of SSI policies will be undone by the movement to an open economy. Documenting this fact will help overcome the implicit pessimism about India’s economic prospects. The political dimension is related to the economic dimension in that those who stand to lose by eliminating SSI policies are likely to generate considerable political opposition to reform. The problem is the classic asymmetric politics of mobilization bias: The losers are likely to know that they will suffer a substantial, if temporary, loss, but the workers and entrepreneurs who will benefit if high-productivity manufacturing is allowed to grow are less likely to recognize that they will be winners. As Mohan points out, even existing large firms in reserved industries benefit from existing policies because they do not face either entry or expansion from efficient competitors. Estimating transition costs will contribute to overcoming the political resistence to reform, and it will provide essential information for devising reform tactics that further minimize targeted transition costs (and hence organized political resistence). Notwithstanding the advance in understanding the cost of SSI policies that Mohan has made, the fact that these policies are very costly to the Indian economy is hardly news, yet reform has not happened. By implication, reform has not been politically feasible despite the high net benefits that it would create. To overcome resistence to reform, one needs to take advantage of the fact that the benefits of reform will far exceed the costs, which implies that the winners could easily bribe the losers to accept reform. One winner from reform will be the government, which will escape the cost of the many subsidies flowing to small-scale enterprise and will also collect more taxes if economic growth improves and inefficient firms are replaced by efficient ones. Hence, the government should be willing temporarily to increase the annual subsidy for small-scale firms if the point of this increase is to buy off resistence to reform. In the absence of estimates of the transition costs and fiscal effects of eliminating SSI policies, one can not confidently propose transition policies that would both ease the economic pain of transition and overcome enough political resistence to make policy reform feasible. The following is not offered as a definitive reform proposal, but as ideas about how to speed an efficient reform. These ideas provide a little something for all three sources of political resistance: owners of protected small firms, employees of these

300

Rakesh Mohan

firms, and owners of grandfathered large-scale firms in reserved industries. If all of these interests need not be bribed to make reform feasible, some of these ideas may be unnecessary. Reform Element #1: Ending Subsidy Programs The purpose of a transition program is to find a way to eliminate the array of subsidies that are now available to small scale manufacturing firms that is attractive enough to reduce their political resistence but also fiscally feasible. The basic elements of a transition policy for subsidies are as follows. Existing subsidies, including tax benefits and in-kind services, will not be available to new firms. For existing firms, the amount of subsidy would diminish over a few years (say, twenty percent per year for five years). For in-kind programs, these services would be available, but at ever-increasing prices until they no longer are subsidized or in demand. Finally, small businesses could cash out their subsidy by receiving a cash payment in return for forgoing further support in the future. The cash-out payments would be calculated to exceed the costs of the subsidies that firms would receive if reform did not take place. For example, the government could calculate the ratio of the total cost of all subsidies to the total sales of all subsidized small-scale firms, and offer each firm a cash-out payment that is a multiple (three to five times) of the product of this ratio and its average sales during the past three years. The cash-out would be available in diminishing amounts for a few years, then would disappear when the subsidy programs came to an end. Reform Element #2: Eliminating Reservation Policies As above, the objective of this reform is to eliminate controls on firm size and restrictions on entry for the 1,000-plus products that are now reserved for small-scale manufacturers. To reduce political resistence, this reform would be phased so as to eliminate restrictions on the beneficiaries of the existing program before entry is permitted. The removal of restrictions could proceed in three phases that are designed to delay expansion by de novo entry, and to privilege existing small firms over existing large ones. Phase One (a year or two) Restrictions on expansion of existing small-scale firms would be eliminated so that initially all of the benefits of ending the restrictions would be targeted toward existing firms. Small manufacturing firms could expand capacity by making new investments or by merging, but without access to credit subsidies beyond the existing limits on the size of small firms. Grandfathered large manufacturing firms could expand only by investing in exist-

Small-Scale Industry Policy in India: A Critical Evaluation

301

ing small firms, but would be required to operate acquired firms at their current site. Thus, the best small firms, which could grow on their own, would expand, and others could merge with larger companies, but without causing relocation and hence local job reduction. Phase Two (another year or two) Restrictions on the expansion of existing large manufacturing firms would be eliminated, but these firms would be required to buy existing small firms that account for some proportion of their increase in capacity. Large firms could close the firms that they bought and expand their own capacity by a small multiple (say, two) of the capacity of the closed firm, or continue to operate the closed firm and be permitted to expand their own capacity by a larger multiple of the output of the acquired firm (say, four). Thus, large firms could expand by investing in their own plants, rather than investing in the plants of small enterprises, but would have to acquire small firms to do so, thereby creating additional value for the owners of small firms. Once a small firm is acquired, it would lose all of its subsidy rights. Phase Three (about three years down the road) All capacity and entry restrictions are removed. Note that the restrictions end before the subsidies (Reform #1), so that small firms seeking to survive through expansion and efficiency improvements are given a small advantage for a year or two. Reform Element #3: Ending Guaranteed Employment The ultimate policy objective is to eliminate restrictions against terminating employees, but for several years the transition policy would be to allow termination but to give severance pay to terminated workers. The transition program could include a program whereby employees of small-scale enterprises are paid severance pay if they lose their jobs during the transition period. The severance payment would be set at a fewmonths’ pay. During the transition period, firms could terminate workers by paying them, say, two months’ salary. If a large firm buys and closes a small firm, the acquiring firm would be required to pay a little more, say three months’ pay. Finally, if a firm cashes out its subsidy and closes down, the owner would be required to contribute up to half of the cash-out to cover three months of severance pay, with the government making up any shortfall. The preceding ideas depart from the first-best, efficiency-maximizing reform because they maintain some restrictions on expansion and some subsidies for several years. The argument for a transition plan is that it reduces the losses experienced by those who are losers due to reform. Such a transition program has two attractions: It eliminates some real economic pain

302

Rakesh Mohan

that inevitably comes with removing significant policy-induced distortions in the economy, and it reduces political resistence to reform. The difficult task is to provide realistic estimates of the costs and likely effects of variants of these transition policies, and to decide which will be most effective in achieving both the economic and political objectives of the transition program.

7 Emerging Challenges for Indian Education Policy Anjini Kochar

7.1 Introduction The Indian economy has witnessed significant improvements in the level of schooling over the past fifty years. For example, literacy rates have increased from 27 percent for males (aged five and above) and only 9 percent for females in 1950, to 64 percent for males and 39 percent for females in 1991. Despite such improvements, the quality of elementary schooling in India remains very low (World Bank 1997). Drèze and Sen (1995) document the deficiencies in “unobservable” aspects of school quality, such as teacher attendance and performance. However, Indian schools fall short not only in such unobservable aspects of quality: They are also characterized by very low levels of “observable” measures of school quality, such as studentteacher ratios. The average student-teacher ratio in primary schools is 37 in urban India and 41 in rural India (NCERT 1993). This ratio varies considerably across India, averaging 31 in rural Kerala and 50 in rural areas of Andhra Pradesh and Bihar. This chapter provides evidence of the importance of observable measures of school quality for the schooling decisions made by Indian households. It also provides evidence that school quality has a greater effect on the schooling decisions of less advantaged households, that is, those with relatively low levels of parental schooling. While a number of studies have examined the dependence of schooling outcomes in India on student-teacher ratios Anjini Kochar is assistant professor of economics at Stanford University. This paper uses data from the National Sample Survey Organization, India, as well as data from the National Council of Applied Economic Research (NCAER) in New Delhi. NCAER data were provided as part of a collaborative research program on human development in India. The author thanks Abusaleh Shariff and other researchers at NCAER for their help in providing the data and other supporting material and for many useful discussions.

303

304

Anjini Kochar

and other measures of school quality, most of these studies have been based on relatively small samples of data from selected regions of the economy. This is because national household surveys providing information on the schooling of Indian children as well as on the quality of the schools they attend were, until now, not available. This study uses one of the few Indian data sets that provide this information, the 1993 HDI (Human Development of India) data set collected by the National Council of Applied Economic Research (NCAER). The differential effect of school quality on households from different socioeconomic backgrounds suggests that low school quality may increase inequality in schooling attainment. While increased schooling inequality is of direct concern, it is of additional concern because of recent research that links inequality to growth. Using data from recently released national sample surveys spanning the decade 1986 to 1996, this chapter provides evidence that schooling inequality in India has, indeed, increased significantly over the past decade. It also provides evidence of the correlation between schooling inequality and governmental investments in the schooling sector. Such evidence has not previously been documented. A rigorous empirical analysis of the factors that underlie the differential effects of school quality on household schooling investments is beyond the scope of this chapter. However, preliminary data analysis suggests that part of the explanation lies in households’ increasing use of privately financed schooling inputs. Data on enrollments in private schools and on the use of home tutors reveal that households have significantly increased their use of both these inputs in recent years. This increase has been particularly notable among better-off households, suggesting that the growth of private schooling inputs may underlie the observed growth in schooling inequality. However, the available data also suggest that the negative correlation between governmental schooling expenditures and inequality reflects the rise in home tutoring rather than the growth of private schools. This may reflect the “public” nature of many of India’s “private” schools: the majority of private schools are financed by state governments. It may be the case that the growth of the private schooling sector and, correspondingly, the extent to which it offers households an alternative to low-quality public schools, is limited by government regulation. The rest of this chapter is organized as follows. Section 7.2 presents a brief overview of schooling in India, describing data on both the level of attained schooling and its quality. The substantive empirical analysis of this chapter is in section 7.3, which establishes the effects of school quality on enrollments. This section also examines whether these effects differ across households differentiated by levels of parental schooling, and provides evidence of a correlation between public expenditures on schooling and schooling inequality. Section 7.4 undertakes a preliminary analysis of the hypothesis that the use of private schools and home tutors explains

Emerging Challenges for Indian Education Policy

305

the differential effects of school quality on enrollments. Section 7.5 concludes. 7.2 Overview of Schooling Attainments in India India has made substantial strides in improving educational attainment over the past fifty years. Gross enrollment rates, for example, have increased from 82 percent for boys and 33 percent for girls in 1951 to 116 percent for boys and 88 percent for girls in 1992.1 Despite relatively high dropout rates, currently 65 percent of all boys and 60 percent of all girls complete the first five-year cycle of primary schooling. However, much remains to be done. There are significant regional differences; primary school completion rates vary from 100 percent in Kerala to 40 percent in Bihar. Moreover, there are significant gender differences: In Rajasthan, 1993 data showed that only 38 percent of rural girls between the ages of six and eleven are enrolled in schools. Nevertheless, it appears that the margin of choice for many households in the country is slowly shifting away from a decision on whether to send their children to primary school to a decision on whether to enroll in the next level of schooling, that is, middle schools. These improvements in the level of schooling are partly the consequence of increased governmental expenditures on schooling. Table 7.1 provides data on schooling expenditures in constant 1980–81 rupees for India and for the sixteen major states. For the country as a whole, schooling expenditures per student have increased from just Rs. 190 in 1980–81 to Rs. 341 in 1995–96. State governments have considerable autonomy in educational decision making, and this is reflected in the significant variation in both the level of expenditure and in its growth across the states.2 Thus, in 1995–96 the state of Kerala spent Rs. 777 per student, an increase of 89 percent from Rs. 411 in 1985–86. In contrast, West Bengal increased its schooling expenditures per student by only 11 percent in this period, from Rs. 178 in 1985–86 to Rs. 198 per student in 1995–96.3 Government investments have financed a tremendous growth in the number of schools and in schooling infrastructure. The number of primary schools has increased from 209,671 in 1950 to 572,923 in 1993, so that currently 95 percent of the rural population has access to a school within a 1. Gross enrollment rates are defined as the percentage of children of school-going age enrolled in school. Delayed entry and the consequent presence of overage students result in rates exceeding 100 percent. 2. Education is on the “concurrent” list, which means that even though the broad guidelines and structure of education may be laid down by the central government, the states are free to evolve and frame their own policies and structures of education within a broad framework. Currently, approximately 89 percent of funding for education is state-provided, with the remaining 11 percent being provided by the central government. 3. All figures are in constant 1980–81 rupees.

306

Anjini Kochar

Table 7.1

Per-Pupil Expenditure on Elementary Education at Constant Prices (1980–81
100) Elementary Education

States

1980–81

1985–86

1990–91

1995–96

Andhra Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal All states Center

189 176 171 179 296 365 188 340 134 205 132 215 253 164 150 172 185 1

232 209 216 275 412 371 231 411 166 284 153 273 257 219 204 178 230 2

225 227 349 300 561 539 266 505 221 344 268 366 377 323 367 228 311 10

222 329 326 352 530 612 272 777 231 349 283 349 377 332 355 198 305 31

All India (centerstatesunion territories)

190

239

322

341

Sources: Ministry of Human Resource Development (n.d.[a,b]), Analysis of Budgeted Expenditure on Education and Selected Educational Statistics.

walking distance of one kilometer. Public schooling expenditures have also financed a number of initiatives intended to improve other aspects of schooling infrastructure and hence school quality. These include “Operation Blackboard,” a scheme introduced in 1987–88 with the objective of improving schooling infrastructure, including teachers, in all primary schools. More recently, the government has implemented an ambitious District Primary Education Programme (DPEP), which decentralizes decision making so as to improve the delivery of educational services. The government has simultaneously taken steps to improve the quality of teaching. It has set up a national council for teacher education to maintain standards for teacher education programs. It also offers pre-service and in-service training facilities for schoolteachers and those involved in adult education, and nonformal educational programs at the district level through District Institutes for Education and Training (DIETs). Despite these measures, the growth in infrastructure has barely kept pace with the growth in the school-age population. Data from the All-India Educational Surveys conducted by the National Council of Educational Research and Training (NCERT) reveal that the ratio of the number of children aged six to eleven to the number of primary school teachers actually

Emerging Challenges for Indian Education Policy

307

increased between 1978 and 1993, in both rural and urban India. The 1978 ratios were 70.8 and 63.8 in rural and urban areas respectively, while the 1993 ratios were 88.6 in rural India and 68.8 in urban India. Studentteacher ratios in rural middle schools showed little change; the ratio of the number of rural children aged eleven to fourteen to the number of rural middle school teachers fell from 58.9 in 1978 to 58.5 in 1993. Significant improvements occurred only in urban middle schools, where the studentteacher ratio fell from 55.9 in 1973 to 48.9 in 1993. 7.3 Does School Quality Affect Schooling Attainment? 7.3.1 Regression Estimates of the Effect of Student-Teacher Ratios on Enrollment The appendix to this chapter sketches a theoretical framework that explains the determinants of schooling decisions. This framework suggests that the demand for schooling will depend on the quality of schools, as well as on traditional factors such as household income and the direct and opportunity costs of schooling. It also suggests that the effect of school quality on schooling may differ across households distinguished by wealth. This section takes this framework to the data to test whether school quality affects enrollment decisions. The available empirical evidence on this issue is mixed. While much of the literature suggests that school quality has little or no effect on schooling,4 Lazear (1999) and others have suggested that estimates of the effect of school quality based on least squares regressions are biased. This is because student-teacher ratios may be correlated with unobservable determinants of schooling attainment. If so, it would be wrong to interpret the effect of the student-teacher ratio on schooling attainment as a pure “quality” effect. The correlation between student-teacher ratios and unobservable components of schooling attainment may reflect the endogenous determination of the number of students, but may also result if the number of teachers is endogenously chosen on the basis of local socioeconomic variables that also affect schooling investments, such as the return to schooling. Empirical studies that allow for the endogeneity of schooling investments do, indeed, find that school quality matters (Angrist and Lavy 1999, Card and Krueger 1992). Relatively few studies have examined the effect of school quality on schooling outcomes in India. Sipahimalani (1998), using data from a large 1993 cross-sectional survey of approximately 30,000 rural Indian households,5 4. This literature is reviewed in Hanushek (1995). 5. This is the NCAER-UNDP survey of human development in rural India. The same data set is used for the empirical analysis reported in this section.

308

Anjini Kochar

finds that variables such as the distance to a school, the proportion of male teachers, and the provision of midday meals do affect schooling attainment. Her estimates, however, do not allow for the possible endogeneity of these variables. Drèze and Kingdon (1999) also provide evidence of the importance of student-teacher ratios in India, based on a smaller survey of approximately 4,000 children in the Indian states of Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh.6 They find that the child-teacher ratio has a negative effect on schooling attainment. This negative effect persists, even in regressions that instrument the child-teacher ratio with the distance separating the village from the nearest road. The empirical work of this section provides further evidence on the effects of school quality on India, based on the larger NCAER data set, and using an alternative set of instruments. Empirical tests of whether school quality affects enrollment requires defining some quantifiable measure of school quality. While there are many dimensions to school quality, data availability frequently determines this choice; data on important attributes such as teacher attendance and the quality of teaching are rarely available. As in much of the empirical literature (Card and Krueger 1992, Drèze and Kingdon 1999), I proxy school quality by student-teacher ratios. Future research may find other dimensions of quality to be of greater quantitative importance. Nevertheless, evidence that any one measure of schooling quality affects schooling attainment does suggest the importance of schooling quality, even though it may provide little guidance as to which types of investments are likely to have the largest incremental effect on schooling attainment. The NCAER 1993 data set used for the empirical work of this section provides information on schooling as well as on standard household socioeconomic variables for approximately 30,000 rural Indian households in fifteen of India’s major states. The household questionnaire was supplemented by a village survey, which provides information on local schools. This data on school enrollments and the number of teachers facilitates the calculation of student-teacher ratios for the school attended by the student in question. I consider the effect of student-teacher ratios on the probability that a child between the ages of six and eleven in any given household will attend school. These regressions are run separately for boys and for girls. In addition to student-teacher ratios, the set of regressors includes the following variables: the number of children in the household in three different age groups (0–5, 6–11, and 12–19); the number of household adult males and females in two age groups; the proportion of primary-schooled males and literate women in the village; the standard deviation in these measures of adult schooling in the village; the mean and standard deviation in farm size in the village; and a set of dummy variables for the state of residence. 6. They use the PROBE data set. Findings of the PROBE team are documented in Probe Team (1999).

Emerging Challenges for Indian Education Policy Table 7.2

Effect of School Quality on Household Probabilities of Current Enrollment, Children Aged 6–11 Boys

Variable Student-teacher ratio Ratio ∗ head primary Ratio ∗ wife literate Head primary Wife literate Land size N

309

Girls

OLS

IV

IV

–0.0002 (0.00) —

–0.007** (0.002) —

—

—

–0.009** (0.002) 0.007** (0.002) 0.004** (0.002) –0.19** (0.08) –0.11 (0.08) 0.001** (0.0006) 7,306

0.14** (0.01) 0.07** (0.01) 0.002** (0.001) 7,306

0.14** (0.01) 0.06** (0.01) 0.001** (0.0005) 7,306

OLS

IV

–0.0002** –0.006** (0.0002) (0.002) — — —

—

0.17** (0.01) 0.13** (0.02) 0.001** (0.0005) 6,521

0.16** (0.01) 0.12** (0.02) 0.001** (0.0005) 6,521

IV –0.009** (0.002) 0.003* (0.002) 0.010** (0.002) 0.001 (0.08) –0.32** (0.10) 0.001** (0.0006) 6,521

Notes: All regressions include the following additional regressors: dummy variables for the state of residence; the number of boys and girls in the household in 3 age groups (0–5 years, 6–11 years, and 12–19 years); the number of household males and females in two age groups; child, male, and female village wages; the proportion of primary schooled males and literate females in the village; the standard deviation in these measures of adult schooling in the village; and the mean and standard deviation in land size in the village. OLS = ordinary least squares. IV = instrumental variables. Figures in parentheses are standard errors. **Significant at the 5 percent level. *Significant at the 10 percent level.

Table 7.2 provides results from both ordinary least squares (OLS) regressions and from instrumental variables (IV) regressions, which instrument student-teacher ratios so as to correct for the potential endogeneity of this variables.7 Since the government of India determines the number of teachers assigned to any given village school primarily on the basis of the childaged population in the village, I use the village child population between the ages of six and eleven, and the square of this variable, as instruments. The child population at the level of the village is calculated on the basis of the household survey data, appropriately weighted to correct for the nonrandom nature of the sample.8 While the OLS estimates, which treat the student-teacher ratio as an exogenous variable, yield no statistically significant effect of school quality on schooling attainment, the instrumental variable regressions find that school quality does matter, both for boys and for girls. An increase in the number of teachers per student, or a decline in the student-teacher ratio, significantly increases the probability that both boys and girls will attend school. 7. All standard errors reported in the tables are corrected for the grouped nature of the data. 8. Sampling weights are provided in the data.

310

Anjini Kochar

7.3.2 Testing for Differential Effects of School Quality Across Households The third regression in table 7.2 allows the effect of school quality to vary with the socioeconomic characteristics of the household. Specifically, it allows the student-teacher ratio to differentially affect the enrollment decisions of households distinguished by levels of parental education (dummy variables for whether the mother is literate and whether the father has completed primary school). The instrumental variable estimates accordingly treat the student-teacher ratio as well as its interaction with parental schooling as endogenous variables, using interactions of the school-aged population with the full set of exogenous variables, as well as higher order terms in these variables, as instruments. The results indicate that the effect of school quality on attendance is smaller in households in which parents are relatively better schooled. Thus, in households in which the father has completed primary schooling, a 10 percent improvement in the student-teacher ratio increases the probability that his son will attend school by only 1 percent, in contrast to the 6 percent increase observed in households in which the father has not completed primary school. Similarly, while school quality significantly affects the schooling outcomes of children of illiterate mothers, it has a statistically insignificant effect on the children of literate mothers. This means that school quality matters more in states such as Bihar, where only 38 percent of rural fathers have primary or higher levels of schooling and only 13 percent of rural mothers are literate, than in educationally advanced states such as Kerala, where the corresponding figures are 59 percent and 64 percent respectively. Similarly, within any state, the enrollment decisions made by poorer households are more affected by school quality than are the schooling decisions of richer households. Thus, educationally backward households and regions pay the costs of poor school quality to a far greater extent than do better-schooled households. 7.3.3 Public Schooling Expenditures and Schooling Inequality The differential effects of publicly financed school quality on the enrollment decisions of households distinguished by socioeconomic status suggest a correlation between the level of government schooling expenditures and schooling inequality, defined as the difference in the schooling attainment of children from different socioeconomic backgrounds. This section examines the evidence on schooling inequality in India and presents evidence of a link between schooling investments and governmental expenditures on schooling. The extent of inequality in a society is, of course, of direct policy concern, particularly in countries such as India that have always had a strong commitment to reducing socioeconomic inequalities. Recent literature suggests that high levels of income inequality also adversely affect

Emerging Challenges for Indian Education Policy

311

economic growth (Galor and Zeira 1993; Aghion and Bolton 1997; Alesina and Rodrik 1994), providing one more reason for concern about trends in schooling inequalities. Available data reveal that schooling inequality has increased over recent decades. Figure 7.1 compares schooling attainments for urban and rural households in the lowest expenditure quartile relative to the highest, graphing the proportion of fifteen- to eighteen-year-olds from each of these quartiles who completed the eight years of middle schooling. For urban households in the lowest income quartile, this proportion increased from 34 percent to 49 percent between the years 1986–87 and 1995–96, an increase of 44 percent. However, the proportion of fifteen- to eighteen-year-olds completing middle school from households in the highest expenditure quartiles increased by 73 percent in this period, from 52 percent to 90 percent. In rural India, too, the improvements in schooling witnessed in rich households far exceeded the improvements registered in poor households. The proportion of rural fifteen- to eighteen-year-olds completing middle school increased from 24 percent to 32 percent for households in the lowest expenditure quartiles, while it increased from 40 percent to 67 percent for households in the highest expenditure quartile. Figure 7.2 examines the correlation between schooling inequality and government expenditures on schooling. This is done by graphing levels of schooling inequality across states ranked by their 1995–96 levels of per capita educational expenditure on students in elementary schools. Schooling inequality is measured as the ratio of fifteen- to eighteen-year-olds completing middle school from households in the top expenditure quartile relative to those from the lowest expenditure quartile. This figure suggests that government schooling expenditures are also negatively correlated with schooling inequality. Inequality is highest in West Bengal, the state that spent the least on schooling per student, and lowest in Himachal Pradesh and Kerala, states that spent Rs. 612 and Rs. 777, respectively, on education per student. 7.4 Initial Evidence on Factors Underlying the Differential Effect of School Quality The regression estimates of the previous section do not provide insights into why student-teacher ratios matter less for more educated households. It may be that parents who are themselves well schooled place a high value on schooling, regardless of quality, and this difference in the utility value of schooling yields corresponding differences in the effect of school quality on schooling enrollments across households. Alternatively, as suggested in the theoretical framework, better-schooled parents may have access to substitutes for high student-teacher ratios, substitutes that are privately financed and hence less affordable for low-income households. Any negative correlation between the consumption of such private alternatives and school

Fig. 7.1 Proportion of 15–18-year-olds completing middle school in upper and lower expenditure quartile households, urban and rural India, 1985 and 1995

Fig. 7.2 Ratio of urban 15–18-year-olds completing middle school in top expenditure quartiles relative to lowest, by states ranked by per pupil educational expenditures, 1995–96

314

Anjini Kochar

quality could also generate the differential effects of student-teacher ratios on enrollments revealed in the previous section. An empirical investigation of the role of privately financed inputs in explaining the effect of student-teacher ratios on enrollments would have to allow for the endogeneity of investments in such inputs. Such an analysis is rendered difficult by the lack of appropriate instruments, that is, variables correlated with private schooling inputs but uncorrelated with enrollment decisions. Lacking such instruments, I instead confine myself to a descriptive analysis of the importance of two such inputs, private schools and home tutoring. Section 7.4.1 considers trends in the use of private schools and home tutors, and provides evidence that households from different expenditure quartiles do, indeed, differ in their use of these inputs. This suggests that the increased use of private schooling inputs may explain the increase in schooling inequality documented in the previous section. However, such evidence is not sufficient to conclude that government investments in schooling explain the growth of the private schooling sector. Further evidence on this point is provided in section 7.4.2, which examines the correlation between the growth of private schools and home tutors, and levels of governmental schooling expenditures. The growth in private schools appears to be higher in states that spend more on schooling, and section 7.4.3 suggests that this may reflect governmental regulation of the private schooling sector. Section 7.4.4 presents available data on observable dimensions of quality in private and public schools in order to evaluate the factors that explain the growth of private schools. 7.4.1 The Growth in Private Schools and Home Tutors Urban India has witnessed a staggering increase in private schooling at the primary and middle levels over the past decade.9 This is revealed in table 7.3, which provides data on the percentage of urban children enrolled in private schools for the four different levels of schooling in 1986–87 and in 1995–96. These data reveal that the percentage of urban students enrolled in private primary schools increased from 37 percent in 1986–87 to 59 percent in 1995–96, while the percentage enrolled in private middle schools increased from 35 percent to 57 percent in the same period. Rural India, too, has witnessed substantial increases in the importance of private schools (table 7.4). Thus, the percentage of rural children enrolled in private primary schools increased from 9 percent in 1986–87 to 21 percent in 1995–96, while the percentage enrolled in private middle schools increased from 17 percent to 26 percent in the same period. The percentage importance of private school enrollment at the second9. As discussed in section 7.4.3 below, private schools are all schools that are privately managed. These include “private aided” schools, which receive substantial funds from state governments.

Table 7.3

Percentage of Urban Children Enrolled in Private Schools Primary

State

Middle

Secondary

Higher Secondary

1986–87 1995–96 1986–87 1995–96 1986–87 1995–96 1986–87 1995–96

Andhra Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Jammu Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal

41 7 36 22 56 24 37 27 40 23 38 16 54 40 43 61 30

66 23 48 63 82 32 65 46 57 41 78 24 68 59 55 72 49

27 8 20 29 40 9 24 32 15 17 48 17 47 23 38 53 55

61 18 31 63 69 26 53 51 49 36 82 22 59 44 51 68 64

39 10 17 51 77 26 36 61 51 27 77 28 59 35 55 72 68

56 16 18 60 62 32 35 65 54 28 87 13 60 31 48 64 68

49 14 17 48 37 24 5 75 74 28 75 51 66 16 56 61 61

51 9 16 65 33 11 13 64 73 25 84 36 50 24 51 59 69

India

37

59

35

57

52

54

50

50

Source: Author’s tabulations based on the household data in National Sample Surveys, rounds 42 and 52. Table 7.4

Percentage of Rural Children Enrolled in Private Schools Primary

State

Middle

Secondary

Higher Secondary

1986–87 1995–96 1986–87 1995–96 1986–87 1995–96 1986–87 1995–96

Andhra Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Jammu Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal India Source: See table 7.3.

7 8 5 1 5 2 6 3 41 1 3 7 10 2 16 16 11

39 6 10 12 24 5 23 4 50 5 45 4 18 8 26 35 22

4 5 4 7 5 1 3 4 55 2 30 29 8 2 17 42 45

13 3 10 13 18 6 11 7 44 5 60 20 13 5 22 51 55

7 10 6 44 91 5 1 48 67 8 77 64 14 3 26 69 53

34 8 6 25 9 7 7 35 46 9 79 32 13 3 17 59 56

24 12 12 35 28 3 3 57 84 13 80 69 51 3 29 61 47

27 14 11 25 14 13 2 45 78 8 90 80 47 12 28 56 62

9

21

17

26

38

31

40

40

316

Anjini Kochar

ary and upper secondary level during this period did not change significantly. This may be a “cohort” effect; the increase in private schooling at the primary and middle levels between the years of 1986 and 1996 may very well result in an increase in private schooling at higher levels in the years to come. The lack of growth in private school enrollments at higher levels should not, however, mask the importance of private schools at this level. In urban India, 50 percent of the students who attend secondary and upper secondary schools are enrolled in private schools. The data also reveal that households from India’s upper expenditure groups increased their use of private schools to a far greater extent than households from the lower expenditure quartile. Figure 7.3 documents this increase in private schooling inequality at the middle school level for urban and rural households. In urban India, the percentage of children in households in the lowest expenditure quartile enrolled in private schools marginally declined from 30 percent in 1986–87 to 29 percent in 1995–96. In contrast, the percentage of children from households in the upper expenditure quartile in private schools increased from 42 percent to 70 percent in the same period. The Indian economy has also witnessed a significant increase in the use of the purchased time of home tutors. Tables 7.5 and 7.6 provide data on purchased home tutoring in urban and rural India, respectively. These data show that the percentage of urban households reporting expenditures on private tutoring has increased significantly at all levels of schooling. For example, at the upper secondary level, as many as 40 percent of households report home tutoring for sons, an increase from 32 percent a decade ago.10 Rural India has not witnessed the same increase in the percentage of households reporting home tutoring. Nevertheless, a quarter of rural households report expenditures on home tutoring for sons enrolled in secondary and upper secondary schools. As with private schools, it is rural and urban children from the upper expenditure quartile who have significantly increased their use of home tutors. This increase in home tutoring inequality is revealed in figure 7.4, which compares the importance of home tutoring for different expenditure quartiles between the years 1986–87 and 1995–96 in urban and rural India respectively. The greater use of home tutoring and private schools among better-off households suggests that the use of these inputs could explain why such households are less affected by student-teacher ratios in public schools. 7.4.2 Are Private Schools and Home Tutoring Substitutes for Government Schooling Expenditures? The increase in the use of private schooling inputs suggests that this increase may partly explain the rise in schooling inequality documented in the 10. The data for tuition for daughters is not significantly different from that for sons.

Fig. 7.3

Proportion of middle school students enrolled in private schools, by expenditure quartiles, urban and rural India, 1985 and 1995.

Table 7.5

Percentage of Boys Reporting Private Tuition, Urban India Primary

State

Middle

Secondary

Higher Secondary

1986–87 1995–96 1986–87 1995–96 1986–87 1995–96 1986–87 1995–96

Andhra Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Jammu Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal

24 18 20 10 13 3 10 15 25 13 15 30 11 10 17 14 45

24 23 28 17 18 6 33 16 17 10 21 49 15 6 17 18 59

31 21 23 15 15 5 25 16 29 22 24 29 18 13 23 18 54

35 33 36 25 22 11 43 19 26 19 32 56 23 8 27 30 76

35 38 33 28 30 11 42 19 43 28 37 47 21 27 31 27 68

40 46 44 36 32 23 58 22 27 33 49 74 36 19 29 43 88

38 22 33 27 19 — 39 9 42 26 36 12 29 33 29 23 66

37 45 36 38 31 22 82 19 33 36 41 36 29 25 32 40 90

India

18

22

24

32

34

42

32

40

Source: See table 7.3. Note: Long dashes indicate a negligible amount. Table 7.6

Percentage of Boys Reporting Private Tuition, Rural India Primary

State Andhra Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Jammu Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal India

Middle

Secondary

Higher Secondary

1986–87 1995–96 1986–87 1995–96 1986–87 1995–96 1986–87 1995–96 10 7 12 3 2 — 9 2 9 6 4 16 6 2 8 3 31

8 4 10 1 8 1 11 1 14 2 1 23 4 1 8 4 37

21 10 15 4 4 2 12 6 15 9 10 20 9 6 14 7 53

17 7 23 1 18 3 35 4 26 8 6 28 9 2 10 11 67

27 24 24 11 21 4 26 9 25 13 24 27 28 19 29 20 61

18 17 30 10 25 4 56 6 30 13 15 38 20 15 24 26 75

34 18 21 16 21 12 32 9 30 10 22 6 25 18 30 15 62

26 12 16 16 21 12 67 24 28 12 15 22 17 23 33 22 73

8

8

13

17

25

26

24

25

Source: See table 7.3. Note: Long dashes indicate a negligible amount.

Fig. 7.4

Proportion of middle school students receiving private coaching, by expenditure quartiles, urban and rural India, 1985 and 1995

320

Anjini Kochar

previous section. However, this need not imply that households use private schooling inputs to substitute for the poor quality of public schools. The growth in private schooling may be unrelated to governmental investments in schools; it may reflect an increase in the price of those dimensions of school quality that are exclusively offered by private schools. Similarly, home tutoring could have increased because it complements, rather than substitutes for, school quality. Yet, if the use of private schooling inputs is to explain the differential effects of school quality on enrollments (table 7.2), such inputs must be negatively correlated with observable dimensions of school quality. This section provides evidence on the nature of the correlation between government schooling expenditures and the growth in both home tutoring and private schools. Figure 7.5 documents the growth in home tutoring for middle school students in urban India between the years 1986–87 and 1995–96 across states ranked by their level of spending for elementary schools in 1985–86. The negative correlation between the growth of home tutoring and government expenditures on schooling is evident. Thus, Orissa, which spent only Rs. 153 per student on elementary education (in constant 1980–81 rupees) reports the highest growth in the number of middle school students reporting expenditures on home tutors, in both rural and urban areas. Orissa is followed closely by West Bengal, a state that has long been characterized by its low spending on schooling. In urban West Bengal, 90 percent of the students enrolled in upper secondary schools also receive home tutoring, up from 66 percent a decade ago. The negative correlation between the growth of private tutoring and state expenditures on schooling suggests that home tutoring does indeed substitute for low school quality. A similar examination of the correlation between the growth of private schools in urban India and state-level expenditures on schooling reveals little evidence of a negative correlation (figure 7.6). If anything, figure 7.6 suggests a positive relationship between schooling expenditures and the growth of private middle schools. Thus, states such as Andhra Pradesh, Gujarat, Maharashtra, and Haryana, all of which recorded above-average per capita expenditures on elementary schooling, report the greatest growth in the number of students enrolled in private middle schools in urban India. Conversely, the growth of private schools has been relatively low in states such as Orissa, West Bengal, and Assam, states characterized by low levels of per capita expenditure by state governments on schooling. 7.4.3 The “Public” Nature of Indian “Private” Schools The evidence suggests that states spending relatively more on schooling have experienced more rapid growth in private school enrollments. This may very well reflect underlying factors such as the return to schooling. Relatively high rates of return to schooling in states such as Maharashtra and Gujarat may cause state governments to respond by increasing expendi-

Fig. 7.5

Growth in private coaching by middle schools, urban India, 1985–95, by states ranked by spending on elementary education, 1985

Fig. 7.6

Growth in private middle schools, urban India, 1985–95, by states ranked by spending on elementary education, 1985

Emerging Challenges for Indian Education Policy Table 7.7

Types of Private Schools, 1995–96 Private School Enrollment (% of total enrollment)

Urban Primary Middle Secondary Higher Secondary Rural Primary Middle Secondary Higher Secondary

323

Enrollment in Recognized Schools (% of total in private schools)

Aided Private Schools

Unaided Private Schools

Aided Private Schools

Unaided Private Schools

25 32 36 38

25 15 11 9

96 98 99 99

73 82 82 77

6 14 21 28

7 5 4 8

95 98 98 97

47 59 67 57

Source: See table 7.3.

tures on schooling, and may also fuel the growth of the private sector. However, the nature of the private-schooling sector in India also suggests a direct link between state spending on schooling and the growth of the private sector. Private schools fall into two categories, aided and unaided. Table 7.7 provides information on enrollments in aided and unaided private schools. These data reveal that enrollments in aided private schools exceed those in unaided schools at all levels beyond the primary level. Aided private schools are privately managed, but are financed almost exclusively by state governments. Indeed, Tilak (1990) notes that 95 percent or more of the total expenses of aided private schools are borne by state governments. The dependence of aided private schools on state financing, and the relative importance of such schools, suggests that the level of governmental expenditure on schooling directly affects the quality not just of government schools, but also that of aided private schools. This in turn may explain the slow growth of private schools in regions characterized by low governmental schooling expenditures. The dependence of aided private schools on public financing reflects state regulation of the fees these schools can charge. In states such as Uttar Pradesh, aided private schools, like government schools, are prohibited from charging any tuition fees, even in secondary schools (Kingdon 1996). In 1995–96, only 53 percent of students enrolled in aided private schools in urban India reported paying tuition fees, and the average amount of such fees was only Rs. 366 per student per year. State regulation of aided private schools extends beyond the regulation of school fees. For example, aided private schools in Uttar Pradesh cannot

324

Anjini Kochar

recruit or dismiss their own staff (Kingdon 1996). Government regulation also extends to unaided private schools. This is because all private schools are further divided into “recognized” and “unrecognized” schools. “Recognition” is required, since only recognized schools can issue valid “transfer certificates” to students leaving the school, certificates that are in turn required for admission to other schools, including higher-level schools. Table 7.7 reveals that the overwhelming majority (approximately 80 percent) of unaided private schools in urban India are recognized schools. In order to be recognized, the school must abide by certain conditions. Again, these vary across states, but frequently extend to regulation of teacher salaries and tuition fees (Kingdon 1996). These regulations undoubtedly reduce the competitiveness and efficiency of private schools. 7.4.4 What Explains the Growth of Private Schools? The above section describes the extensive regulation of private schools, particularly aided private schools. Despite such regulation, there has been a significant increase in enrollments in private schools, an increase documented in section 7.4. What explains this growth? Table 7.8 compares student-teacher ratios across public and private schools in urban and rural India, further distinguishing between aided private schools and unaided schools. It also provides data on other observable components of school quality, namely the percentage of trained teachers and the percentage of teachers with graduate degrees. The data reveal that student-teacher ratios in private schools are not significantly higher than the ratios observed in government schools. Indeed, student-teacher ratios in secondary and upper secondary schools are marginally lower in government schools than in aided private schools, in both urban and rural India. The lack of any significant difference in studentteacher ratios and other observable measures of quality across private and public schools implies that such factors cannot explain the growth in private schools. This suggests that if private schools do produce superior schooling, it must be because they differ from public schools in “unobservable” components of quality, such as teacher performance and management practices (Bashir 1994). However, the importance of schooling practices and inputs relative to other variables, such as the superior quality of the student body in private schools, has not yet been established for India. Research from several other countries, however, has consistently found that the average quality of the school body plays an important role in explaining both completed years of schooling and performance in standardized tests (Evans, Oates, and Schwab 1992; Datcher 1982; Henderson, Mieszkowski, and Sauvageau 1978). The data of this section, which documents the trend toward even greater “sorting” of children from higher-income Indian households into private schools, suggest that the superior quality of the school body in

Emerging Challenges for Indian Education Policy Table 7.8

325

School Quality in Government and Private Schools, Urban and Rural India, 1993

Urban India Trained teachers (%) Primary Upper primary Secondary Higher secondary Teachers with college degrees (%) Primary Upper primary Secondary Higher secondary Student-teacher ratiosa Primary Upper primary Secondary Higher secondary Rural India Trained teachers (%) Primary Upper primary Secondary Higher secondary Teachers with college degrees (%) Primary Upper primary Secondary Higher secondary Student-teacher ratiosa Primary Upper primary Secondary Higher secondary

Government/ Local Body

Aided Private Schools

Unaided Private Schools

All Schools

92 93 92 87

97 93 97 89

86 73 86 73

93 88 93 86

40 49 40 99

35 49 35 99

42 73 42 99

38 55 38 99

35 33 29 31

43 42 35 39

30 33 29 30

37 37 32 35

88 87 88 77

94 88 94 88

77 69 77 53

89 86 89 80

32 43 32 99

27 51 27 98

27 69 37 99

29 46 29 99

40 36 29 31

37 34 30 36

33 31 23 32

41 36 29 34

Source: NCERT (1997). For student-teacher ratios, data in the first column are for government schools only.

a

private schools relative to public schools may also explain part of their advantage. 7.5 Conclusion The average quality of schools in India remains very low, despite the investments made by the central and state governments in schooling over the past fifty years. This chapter provides evidence that low school quality has a cost: It significantly affects households’ enrollment decisions. It particularly affects poorer households, characterized by low levels of parental

326

Anjini Kochar

schooling. This suggests a link between governmental expenditures on schooling and schooling inequality. The preliminary data analysis of this chapter does, indeed, find evidence of such a relationship. Given the low quality of government schools, one would have expected considerable growth in the private schooling sector. The data reveal that such growth has occurred, but also that private school enrollments have increased the most in states that spend relatively more on elementary schooling; private-sector growth has been lower in states where the quality of public schools is low. As this chapter suggests, this may reflect governmental regulation of private schools. A reduction in regulation may foster the growth of private schools. This is likely to improve the quality of schools, particularly in regions where school quality is currently low. Many argue against the growth of the private sector on the grounds that it would increase schooling inequality. However, there is little evidence on this issue; an increase in the supply of private schools may very well reduce school fees, making such schools more affordable to poorer households. The evidence suggests that even poor households are currently incurring significant schooling expenses, frequently in the form of fees for private tutoring. The willingness to pay for quality teaching suggests that poor households would take advantage of good quality private schools, should they become available. There has been surprisingly little discussion about the justification for and extent of regulation of India’s schooling sector, despite the fact that the structural reforms of recent years have opened up debate on the regulation of almost every other sector in the Indian economy. I believe that such a debate on school reform is long overdue, and is essential for the significant improvements in schooling that will be required to take India through the twenty-first century.

Appendix This appendix briefly sketches a theoretical framework for the determinants of schooling decisions. This framework helps interpret the regression and data analysis of this paper. Assume that households gain utility from some composite consumption good, c, and from the human capital of their children, h. The household’s utility function is therefore U(h,c). Human capital is “produced” by time spent in schools, s, the quality of schools, q, and by household expenditures on private schooling inputs, x, such as the purchased time of private home tutors. The vector x could also include aspects of school quality not readily available in public schools, such as teacher quality, but which parents can purchase by enrolling their children in private schools.

Emerging Challenges for Indian Education Policy

327

Let ui reflect the child’s innate ability. Then, the human capital production function for child i in region j is: (A1)

hij
h(sij , qij , xij , uij )

Levels of schooling, s, and expenditures on private schooling inputs, x, are chosen to maximize the household’s utility function, subject to both the human capital production function and a budget constraint that constrains expenditures on consumption and on schooling inputs to equal income. This simple framework yields a demand for schooling time that depends on the quality of schools, q. It also varies with household income and with the price of consumption goods ( pc ), the price of private inputs ( px ), and the price of schooling time ( ps ). The latter price could include both direct costs associated with schooling and also the opportunity cost of a child’s time. This framework also provides several explanations why the effect of school quality on schooling time may vary across households. One possible explanation is differences in the degree to which households can substitute private inputs for school quality. To see this, consider the conditional demand function for schooling time, s, which conditions on optimal investments in private schooling inputs, x*: (A2)

si
s(q, ps , pc , I, x*[q, px , pc , I ])

If some households are constrained in their consumption of private schooling inputs, x, the effect of q on s will vary across constrained and unconstrained households. Labor market imperfections may yield such constraints. For example, there may be no market for parental time spent on tutoring children. If better-educated parents are superior home teachers, households with less-educated parents will be constrained in their ability to substitute home schooling for parental time. In regions where a market for private tutors does exist, such as urban areas, any fixed costs of purchasing home tutors may render such tutoring unaffordable to low-income households. Such fixed costs will exist if home tutors stipulate a minimum number of tutoring hours that parents must purchase.

References Aghion, P., and P. Bolton. 1997. A theory of trickle-down growth and development. Review of Economic Studies 64:151–72. Alesina, A., and D. Rodrik. 1994. Distributive politics and economic growth. Quarterly Journal of Economics 109:465–89. Angrist, J. D., and V. Lavy. 1999. Using Maimonides’ Rule to estimate the effect of class size on scholastic achievement. Quarterly Journal of Economics 114:533–76. Bashir, S. 1994. Public versus private in primary education: A comparison of school

328

Anjini Kochar

effectiveness and costs in Tamil Nadu. Ph.D. diss. London School of Economics, London, England. Card, D., and A. B. Krueger. 1992. Does school quality matter? Returns to education and the characteristics of public schools in the United States. Journal of Political Economy 100 (1): 1–40. Datcher, L. 1982. Effects of community and family background on achievement. Review of Economics and Statistics 64:32–41. Drèze, J., and A. Sen. 1995. India: Economic development and social opportunity. Delhi: Oxford University Press. Drèze, J., and G. G. Kingdon. 2001. School participation in rural India. Review of Development Economics 5:1–24. Evans, W. N., W. E. Oates, and R. M. Schwab. 1992. Measuring peer group effects: A study of teenage behavior. Journal of Political Economy 100 (5): 966–91. Galor, O., and J. Zeira. 1993. Income distribution and macroeconomics. Review of Economic Studies 60:35–52. Hanushek, E. 1995. Interpreting recent research on schooling in developing countries. World Bank Research Observer 10 (2): 227–46. Henderson, V., P. Mieszkowski, and Y. Sauvageau. 1978. Peer group effects and educational production functions. Journal of Public Economics 10:97–106. Kingdon, G. G. 1996. Private schooling in India: Size, nature and equity effects. Economic and Political Weekly, 21 (December): 3306–14. Lazear, E. P. 1999. Educational production. Stanford University, Graduate School of Business. Unpublished manuscript. Ministry of Human Resources Development. n.d.[a]. Analysis of budgeted expenditure on education. New Delhi: Ministry of Human Resources Development. ———. n.d.[b]. Selected educational statistics. New Delhi: Ministry of Human Resources Development. National Council of Educational Research and Training (NCERT). 1997. Sixth allIndia educational survey. New Delhi: NCERT. The PROBE Team. 1999. Public report on basic education in India. New Delhi: Oxford University Press. Tilak, J. B. G. 1990. The political economy of education in India. Special Studies in Comparative Education no. 24, Comparative Education Center, Graduate School of Education, State University of New York, Buffalo. Sipahimalani, V. 1998. Education in the rural Indian household: A gender based perspective. Yale University, Department of Economics. Manuscript, October. World Bank. 1997. Primary education in India. Washington, D.C.: World Bank.

8 Does Economic Growth Increase the Demand for Schools? Evidence from Rural India, 1960–99 Andrew D. Foster and Mark R. Rosenzweig

8.1 Introduction Many social scientists have argued that a benchmark for assessing the performance of a country is the human capital of its citizens, and there is no doubt that India has lagged behind most of the countries of the world in terms of the educational attainment of its population. In recent years, most of the research on the determinants of human capital investment in low-income countries has focused on educational policies. Much of this literature addressing the effects of school policies emphasizes the supply of schooling services, and some of this research has indeed demonstrated that lack of access to schools and poor school quality do contribute to low schooling attainment (e.g., Duflo 1999; Kochar, chapter 7 in this volume). However, this research approach neglects factors that affect schooling demand, namely schooling returns.1,2 Andrew D. Foster is professor in the Departments of Economics and Community Health at Brown University. Mark R. Rosenzweig is a professor in the Department of Economics at the University of Pennsylvania. The research for this chapter was supported in part by grants NIH HD30907 and NSF, SBR93–08405 and by the Center for Research on Economic Development and Policy Reform, Stanford University. 1. The emphasis on educational policy has been buttressed by research showing that wealth differences across areas of India have little to do with schooling investments (Drèze and Sen 1998). However, cross-sectional wealth differences are not a good proxy for economic growth, which has important effects on the returns to schooling. It is not obvious that the returns to schooling are higher among wealthy families, and we show below that among farm households, high wealth and schooling costs are positively related. 2. A possible reason for this neglect is that the existence of high returns to schooling is taken for granted. Evidence is often cited of high returns in low-income countries based on wage regressions (Psacharopoulos 1998). But these estimates are from highly selective samples, usually of wage workers, which have little relevance to rural populations where many individuals are self-employed.

329

330

Andrew D. Foster and Mark R. Rosenzweig

Many economists (Welch 1970; Schultz 1975; Nelson and Phelps 1966) have argued that an important determinant of the returns to schooling is growth itself: The returns to general skills are enhanced in a dynamic environment in which skill in decoding information and in decision-making under changing circumstances has high payoffs. Empirical studies based on farm-level, county-level, and cross-country data have shown that returns to schooling are indeed greater in dynamic settings (Welch 1970; Foster and Rosenzweig 1996; Rosenzweig 1995; Lau et al. 1993; Benhabib and Spiegel 1994). Foster and Rosenzweig (1996), in particular, have shown that returns to schooling rose more in areas of India in which the green-revolution seed technology advanced most rapidly. There has been little study, however, of how the returns to schooling affect educational choices and how schooling returns are related to economic growth. In this chapter we examine whether low levels of schooling infrastructure—in particular, access to secondary schools—and low schooling investment are solely the product of failed educational polices or whether they reflect instead, or at least in addition, inadequate economic policies and consequently low school demand. Do poor economic growth performance and the lack of opportunity to exploit the skills that schooling produces importantly affect investments in school infrastructure, or does lack of school access solely reflect mismanagement of the economy? We use a newly available and newly assembled time series data on 240 rural villages spread throughout almost all of India covering the last forty years to test the hypothesis that an important determinant of school building and of schooling investment has been the expectations of future economic growth by local populations. We begin in section 8.2 with a simple model that incorporates the basic idea that investments in schools are based on expectations of schooling returns, which are augmented when productivity is expected to rise more rapidly. We show that in the rural sector the appreciation of land prices will reflect changes in farmers’ expectations of future productivity growth, and use the variation across villages and over time in yield growth rates, land prices, and school construction to estimate the effects of expected yield growth, wealth levels, and the stock of existing schools on secondary-school construction and on school enrollment rates. Section 8.3 documents the increases in rural secondary schools that have occurred in India over the past forty years, as well as the rise in the productivity of high-yielding-variety crops. The new data indicate that there has been considerable progress over this period in the availability of rural secondary schools. In 1960, less than one-third of villages were located within ten kilometers of a secondary school; now almost 90 percent of the villages have access to at least one secondary school. Rates of secondary-school construction have been particularly rapid in the poorer villages, and the distribution in the availability of secondary schools across Indian states is far

Does Economic Growth Increase the Demand for Schools?

331

less unequal in 1999 than forty years ago, although there are still large disparities. The progress in rural secondary-school building has been accompanied over the same period by a rapid increase in crop productivity. The data show, however, that rates of increases in yields have slowed in recent decades, and our analysis of the data suggests that farmers expect that yield growth will slow down further in the future. This, of course, will not necessarily lead to reduced returns to schooling to the extent that there is a growth in opportunities to employ skills in the nonagricultural sector. In section 8.4 we show that land prices and current land productivity predict well future growth rates, and in section 8.5 we exploit this to estimate the effects of expected future productivity increases on the building of secondary schools. The estimates suggest that expected growth rates in productivity have had significant effects for given wealth levels, both on school construction and on enrollment rates that exceed wealth effects. The results are thus consistent with the hypotheses that the returns to schooling are higher in a dynamic environment and that the Indian population responds to expectations of higher returns to schooling with respect to both school construction and schooling investment. Augmenting and sustaining the rate of economic growth is thus a powerful policy tool for increasing the human capital of the population. 8.2 Framework We wish to test the proposition that economic growth is an important determinant of the demand for schools because it raises the returns to schooling. Any analysis of investment in schooling, however, must take into account that plans for rural investments in schools in any period are driven not by current returns but by the expected returns to schooling in that period. In an agricultural context, based on the idea that schooling has higher payoffs in a dynamic environment, expected schooling returns will in turn depend positively on the expected rate of agricultural technical change. This variable should thus play a role in influencing school building, along with current wealth, preferences for schooling unrelated to schooling returns, population size, and the existing stock of schools. Of course, at any time t the future rate of productivity growth is unknown. The challenge in identifying empirically whether schooling decisions are influenced by expected growth and thus by the prospects of returns to schooling arises from the fact that, unlike actual income or yield growth, expected productivity cannot be readily measured using household survey data. Nor is having information on actual future productivity growth sufficient. This is because actual productivity growth is a noisy estimate of expected productivity growth. Estimates of the effect of expected growth on school investment based on actual future productivity growth would be biased due to this measurement or forecast error.

332

Andrew D. Foster and Mark R. Rosenzweig

A standard treatment for measurement error is instrumental variables. What is needed is a variable observed at time t that predicts expected productivity growth but does not otherwise affect current investment plans. We make use in our empirical analysis of the fact that the price of any plot of land, given an efficient market for land, reflects expectations about the stream of future revenue on that land, appropriately discounted. The price of agricultural land, however, for given wealth, should have no direct effect on school building.3 A second problem in identifying empirically whether schooling responds to changes in expectations of future growth is the presence of persistent, culturally determined factors that influence preferences for schooling. For example, as shown below, the Indian state of Kerala has for many decades been marked by higher school investments than other states, although there is no evidence that rates of returns to schooling were ever higher there. Persistent preferences for schooling will in general be correlated with the existing stock of schools. Moreover, land prices will be higher in areas with higher tastes for schooling as long as more schooling results in greater returns to technical change. Thus, land prices could not be used as an instrument for identifying the effects of expectations about future growth using cross-sectional data. In the appendix we show that the standard remedy for controlling for time invariant unobservables by differencing and applying instruments is, conversely, complicated by the nonobservability of expected productivity, and we indicate a method for dealing with this problem that relies on using the economics of land price determination. A key assumption is that we have the correct equation describing expectations. While this assumption is not directly testable due to the unobservability of expected incomes, we establish an indirect specification test in which one regresses future output growth on current income, wealth, the land price, and a measure of lagged yields. A significant coefficient on lagged yields would suggest that measurement error is present and thus would invalidate the identification strategy suggested above. 8.3 Data To carry out our analysis of the effects of expectations of growth on school building, we need data for multiple time periods that provide information not only on schools, but on land prices, wealth, population size, and yields. The data used in this study are constructed from data files produced by the National Council of Applied Economic Research (NCAER) from six rural surveys carried out in the crop years 1968–69, 1969–70, 1970–71, 3. We assume that in an agricultural context, the price of land is a negligible direct cost component in the building of a school.

Does Economic Growth Increase the Demand for Schools?

333

1981–82, and 1999–2000. The first set of three survey rounds from the Additional Rural Incomes Survey (ARIS) provides information on over 4,500 households located in 261 villages in 100 districts. These sample households are meant to be representative of all households residing in rural areas of India in the initial year of the survey, excluding households residing in Andaman, Nicobar, and Lakshadwip Islands. The most detailed information from the initial set of three surveys is available for the 1970–71 crop year and covers 4,527 households in 259 villages. The 1981–82 survey, the Rural Economic and Demographic Survey (REDS), was of a subset of the households in the 1970–71 ARIS survey plus a randomly chosen set of households in the same set of villages, excluding the state of Assam, providing information on 4,596 households in 250 villages, 248 of which are the same villages as in the ARIS. Finally, in 1999, households in the same set of original ARIS villages, this time excluding villages in the states of Jammu and Kashmir, were included in a new survey, the 1999 REDS. This survey was in the field at the time the research for this chapter was carried out. However, information on the characteristics of the sample villages was complete. The combined data sets provide information over a period exceeding thirty years for 240 villages spread throughout India on village infrastructure, household assets, demography, schooling investment and attainment, wealth, asset prices, and crop productivity. The timing of the surveys is particularly fortunate for the study of the consequences of economic growth. The initial survey occurred at the onset of the “green” revolution in India, which began in the late 1960s with the importation of new, high-yielding seeds, principally for wheat, corn, and rice. Since that time, Indian agriculture has experienced continuous improvements in crop productivity, including as well improvements in such crops as sorghum and cotton. Productivity in agriculture since the late 1960s, however, also grew very unevenly across areas of India, in large part due to persistent differences in agroclimate suitability for the new seeds.4 The 1999 REDS provides information on the history of school building in all of the surveyed villages, providing the dates of establishment for schools located within ten kilometers of the villages classified by schooling level—primary, middle, secondary, and upper secondary—and by whether they were public, private, aided, or parochial. It is thus possible to examine the determinants of school building over the entire span of the sample periods, relating intervals of school investment to initial village conditions. Before looking at this time series of schooling investments, however, it is useful to assess the accuracy of the data based on recollection. We can do this only indirectly: Neither the 1971 ARIS nor the 1982 REDS provides a 4. Foster and Rosenzweig (1996) show that net of differential investments in schooling and irrigation, among other endogenous changes, rates of increase in profitability over the period 1971–82 varied substantially across Indian states, districts, and villages.

334

Andrew D. Foster and Mark R. Rosenzweig

history of school building with which to compare that obtained in 1999 for overlapping years. However, both the 1982 and 1999 REDS provide a retrospectively ascertained history of village electrification. We compared the overlapping years of these two histories, from 1925 through 1982, and found them to be quite close. Regressing the dates of electrification obtained in 1999 on those obtained in 1982 for the overlapping 1925–1982 interval resulted in a coefficient of 1.02 (t
87.6), indicating a very small degree of “telescoping.” The correlation between the two series of dates is 0.989. Thus we believe the school-building histories accurately reflect the true changes in school availability over the last forty years in the 240 villages, with one caveat—that there are few schools that have been destroyed over the period, which would not be reflected in a school-building history based on schools in existence in the villages in 1999. For the analyses here, we will look at secondary, inclusive of upper secondary, schools. We do this because even in the 1960s primary schools were nearly universal—located within 90 percent of the sample villages by 1971. The relevant margin is at the secondary-school level. Figure 8.1 displays the growth in the average number of secondary schools located within ten kilometers of the 240 villages at ten-year intervals from 1960 through 1999. As can be seen, there has been remarkable growth—from an average of .31 schools per village in 1960 to almost 0.9 in 1999. Figure 8.2 shows, moreover, that the decadal rate of building secondary schools increased over the 1960–89 period, but has slowed in the last decade. The school establishment histories also indicate that there were large interstate disparities in the presence of rural secondary schools in 1960, but they show as well that there have been substantial variations in statewide school investments since then. Figure 8.3 provides the average number of proximate secondary schools per village by state in 1960 and 1999. The 1960 figures indicate the well-known and commented-upon fact that the state of Kerala was the most advanced state in terms of educational attainment. Indeed, the graph shows that Kerala had the largest number of secondary schools per village by a large amount, almost 40 percent more secondary schools per village than the next highest state (West Bengal)—a lead that is still in evidence now. However, rural rates of secondary-school building have been substantially higher in many of the other states since 1960, and the distribution of secondary schools across states is far less unequal than it was in 1960, although there are still substantial disparities in secondary-school availability. Another way of looking at the change in the distribution of secondaryschool availability over time is to divide up the villages by average perhousehold wealth levels. Figure 8.4 depicts the change in the stock of secondary schools proximate to households for four classes of villages based on average per-household wealth in 1971. The figures show that in 1960 the wealthiest quartile of villages had the smallest number of nearby secondary schools, and the upper-middle–quartile villages had the most secondary

Fig. 8.1

Decadal growth in the average number of secondary schools proximate to villages, 1960–2000

Fig. 8.2

Rates of new secondary school building per year, over three survey periods, 1960–99

Fig. 8.3

Average number of secondary schools proximate to villages, by state, in 1960 and 2000

Fig. 8.4

Decadal growth in the number of secondary schools proximate to villages, by average village, per household wealth in 1971: 1960–99

Does Economic Growth Increase the Demand for Schools?

339

schools on average, while the lowest-quartile villages had the second highest number of proximate schools. By 1999, however, there is a perfect inverse correlation between the ranks of villages in the 1971 wealth distribution and their rank in terms of secondary-school proximity. This is evidently because the lowest two quartile groups of villages based on wealth levels in 1971 experienced the highest rates of increase in secondary school building over the last forty years. The combined 1970–71, 1981–82, and 1999 surveys also enable the construction of measures of crop productivity and thus productivity changes by village over the past forty years. For the first two of these surveys, we obtained for each village median crop productivity for each of five crops— wheat, rice, corn, sorghum, and cotton—for both high-yielding (HYV) and traditional seed varieties based on individual farm household information on crop- and variety-specific acres planted and output. For each of these two survey years and for each village, we then weighted these crop- and variety-specific yields by the proportions of area planted that year in the village and by national 1971 crop-specific prices.5 We carried out a slightly different procedure with the 1999 data because the yields by crop and variety were only available at the village level. We constructed a comparable yield measure for that year, weighting the crop- and variety-specific yields in 1999 by their respective planted areas, again using 1971 crop prices. To construct a time series of yields for the villages, we used the resulting Laspeyres-weighted yields for HYV crops, substituting traditional-variety yields only for those villages not growing any HYV crops. The time series of yields is thus the “max-median” of yields for the villages. The yield index increases over time as (a) HYV seed productivity increases due to improvements in seed technology and to increases in irrigation and (b) as any farmers in a village are able to adopt HYV crops.6 Figure 8.5 displays the Laspeyres-weighted average per-acre yield indices based on the survey data for the years 1971, 1982, and 1999, and an estimate of yields for 1961, based on the assumption that no farmers could have been using HYV seeds in that year and on the proportion of farm land in 1971 devoted to HYV (5 percent). As can be seen, there has been remarkable growth in yields over the last forty years, with yields rising from 230 rupees (Rs.) per acre in 1961 to over Rs.1,000 per acre in 1999. The annual per-acre increase in yields over the three periods 1961–70, 1971–81, and 1982–99, shown in figure 8.6, exhibits an inverted U shape: Yield increases have slowed since 1982. Interestingly, the annual rates of increase in secondaryschool building exhibit the same inverted U pattern. Figure 8.7 also indi5. The prices are: 75 rupees (Rs.) per quintal for wheat, Rs. 60 per quintal for rice, Rs. 53 per quintal for corn, Rs. 75 per quintal for sorghum, and Rs. 225 per quintal for cotton. 6. If more farmers adopt HYV crops within a village, and crop-specific shares remain constant, there is no measured increase in crop productivity—the potential for productivity growth for any farmer is not affected.

Fig. 8.5

Laspeyres-weighted per-acre yields (1971 rupees) for five major crops (wheat, rice, maize, sorghum, and cotton): 1960–99

Fig. 8.6 Annual rates of growth in Laspeyres-weighted per-acre yields for five major crops (wheat, rice, maize, sorghum, and cotton) over three survey periods: 1960–2010

Fig. 8.7

Average annual rates of increase in crop yields by average village per-household wealth in 1971: 1960–99

Does Economic Growth Increase the Demand for Schools?

343

cates that rates of increase in yields over the period 1971 through 1999 were inversely related to the villages’ location in the 1971 (average perhousehold) wealth distribution, with the lowest-wealth villages experiencing the highest rates of increase in yields. The green revolution evidently did not most favor the wealthiest areas.7 It is tempting to conclude from the figures tracking aggregate yield growth and school building and the experiences of the different wealth classes of villages that school building and growth are importantly related. However, the decline in rates of increase in crop yields may be due to a slowdown in the spread of irrigation between the decades of the 1970s, the 1980s, and 1990s, and may also be influenced by differences in weather across the survey periods. Additionally, the overall decline in rates of school building may be due to the increased stock of schools that had accumulated by the early 1990s. School building may also have contributed to growth rates in yields. Comparisons of these rates are consequently not necessarily informative about the influence of yield increases and thus the returns to schooling on the demand for schooling. To ascertain whether school building is responsive to expectations about returns and thus expected growth in productivity, we implement the procedure outlined in the previous section, using data on yields from the 1970–71, 1981–82, and 1999 surveys; variables indicating land prices, village population size, shares of cultivated land irrigated in 1971 and 1982, the histories of school building, and whether yields in those years were adversely affected by weather. We also constructed mean household wealth in 1971 and 1982 based on the information provided in the household surveys on the value of landholdings, farm equipment, animals, and irrigation assets. Distinguishing between wealth effects and expected return effects is critical to the analysis, and the detail on asset holding in the surveys is an important feature of the data. Finally, we constructed school enrollment rates for children aged ten to fourteen, by gender, for the 1970–71 and 1981–82 years to see whether wealth and expected returns influenced not only investments in school buildings but also the actual schooling of the village residents. Table 8.1 reports the means and standard deviations of the variables used in the analysis for the three survey years, where available. The figures indicate that most of the elevated rate of secondary-school building in the 1980s was due to the addition of public schools, and that while the overall rate of secondary-school building dropped in the 1990s, the building of private schools increased in that decade compared to the previous one. The data also indicate that not only did the number of schools increase between 1971 and 1982, but so did the school enrollment rates of children, particularly for 7. Within a geographic area, larger landowners benefitted more than small landowners and the landless from crop productivity augmentation (Foster and Rosenzweig 1996).

344

Andrew D. Foster and Mark R. Rosenzweig

Table 8.1

Means and Standard Deviations: NCAER-ARIS, REDS82, and REDS99

Variable/Survey Year Per-village secondary schools built in subsequent decade Total Public Private Enrollment rates, children aged 10–14 years Total Boys Girls Total (output/acre in 1971 rupees) Proportion of farmers whose crop outputs are adversely affected by weather Price per acre of irrigated land (1971 rupees) Share of cultivated land irrigated Village population size Wealth per household (1971 rupees)

1971

1982

1999

.157 (.375) .0594 (.251) .0977 (.297)

.170 (.378) .117 (.328) .0528 (.224)

.137a (.410) .0612a (.268) .0759a (.265)

.386 (.289) .511 (.336) .239 (.302) 307.5 (270.8) .262 (.441) 4,629.7 (1558) .416 (.401) 2,147 (3,050) 13,628 (13,128)

.468 (.335) .552 (.372) .392 (.391) 559.5 (317.8) .239 (.427) 4,876.1 (4774) .602 (.411) 3,102 (4,703) 14,842 (9,850)

n.a. n.a. n.a. 1,009 (473.4) .357 (.480) 14,320 (10774) .634 (.385) 8,987 (17,076) n.a.

Notes: Sampling weights used in computing statistics. Number of villages
240. Standard deviations are in parenthesis. n.a.
not available. a Refers to decade 1990–99.

girls. Enrollment rates for boys increased by 8 percent over the eleven–year period, but rates for girls increased by over 64 percent. The relative gains for girls reduced but did not erase the disparities in sex–specific enrollment rates that existed in the prior period—the enrollment rate for girls is still 71 percent that of boys in 1982. This figure was, however, 47 percent in 1971. The data also indicate that some of the slowdown in the gains in yields between the 1970s and the 1982–99 period was due to both a slightly higher incidence of adverse weather in 1999 compared to earlier years and a slower pace in the rise in the share of irrigated land. The share of cultivated land that was irrigated rose from 42 percent in 1971 to over 60 percent in 1982. Between 1982 and 1999, however, irrigation coverage only increased from 60 percent to 63 percent. The prices of irrigated land, in 1971 rupees, rose by only a small amount between 1971 and 1982, but almost tripled in real terms between 1982 and 1999. Average wealth per household also increased

Does Economic Growth Increase the Demand for Schools?

345

in real terms between 1971 and 1982 by 8 percent, reflecting in large part the modest rise in land prices. 8.4 Land Prices and Expected Yield Growth We first assess whether current land prices reflect expectations of future productivity growth by estimating an equation8 in which the actual future rate of growth in yields in a village is a function of the current land price, current yields corrected for weather shocks and forecast error. In addition, we include in the equation village population growth and the share of land that is irrigated. Population growth, for example, would increase the price of land without any change in yield expectations. Similarly, to the extent that current yields reflect improvements in land through investment, they will overpredict future yield growth. The dependent variable is the village-specific annual rate of growth in yields. Because we have three observations on yields, for 1971, 1982, and 1999, we have two sets of growth rates for each village and can stack the two earliest sets of village-level observations. Theory does not provide guidance on the functional form for the forecast equation. Because we are interested in obtaining the best prediction for future productivity, we estimate the growth equation using two functional forms, employing alternatively as right-hand side variables either the linear form or the log of land prices, yields, and village population size. We also include in both specifications the changes in the share of land that is irrigated between the period 1971–82 and 1982–99 to take out from future yield growth that part due to investments. Table 8.2 reports the estimates of the future yield growth equation for the two functional forms. All t-statistics are corrected for the fact that we have nonindependent observations within villages. The signs of all of the parameters are as expected—in particular, net of current yields and the effects of population pressure, land prices have a positive and statistically significant relationship with future yield growth. Moreover, net of land prices, the higher are current yields and population size, the lower is future productivity growth, reflecting the fact that both are factors that push up the current price of land independent of expectations. In terms of explanatory power, the equation employing logs of the price, yield, and population variables performs substantially better, explaining more than half of the variation in future yield growth, compared with 34 percent for the linear specification.9 The log-form also more easily passes the test that the residuals from the forecast equation, which are supposed to reflect only forecast error, are un8. This is equation (A12) in the appendix. 9. Only a small part of this is due to the inclusion of future irrigation changes. Exclusion of this variable reduces the R2 to 0.498.

346

Andrew D. Foster and Mark R. Rosenzweig

Table 8.2

Forecasting Equations: Land Prices and Expected Future Annual-Yield Growth Rates

Variable/Specification

Log-Linear

Loga

Current price per acre of irrigated land

.00000119 (3.06) –.000160 (14.5) .0168 (2.00) –.000837 (1.36) –.00713 (0.79) .000337 (0.06) .107 (14.8)

.0220 (5.12) –.0714 (23.9) .0287 (3.53) –2.63 (0.96) –.00161 (0.20) .0137 (1.82) .273 (9.08)

Current yield Current share of cultivated land that is irrigated Village population size in current period (10–3) Adverse weather in current year Change in share of irrigated land in future period Constant R2

.343

.501

Notes: N
473. Absolute value of t-statistics, corrected for village cluster, in parentheses. In this specification, the land price, yield, and population size variables are included in log form. a

correlated with lagged yields. Regressing the residuals from the 1982 prediction on, respectively, the log and level of the yields in 1971 from the log and linear equations resulted in coefficients with t-values of, respectively, 0.65 and 1.34. The point estimate of the log of the land price suggests that, for given current yields and population size, an expected one percentage point increase in sustained annual yield growth rates would be associated with a 50 percent increase in current land prices. Land prices thus appear to be a moderately sensitive and statistically significant predictor of yield growth. We can thus use the estimates from table 8.2, along with the actual 1999 values of land prices and yields, to infer what Indian farmers currently expect the future rate of growth in yields to be. Based on the 1999 values, the estimates suggest that farmers expect to experience a yield growth rate of 3.5 percent per year, in the absence of increased irrigation coverage. It is notable that this figure is lower than the average annual increase in yields experienced in the 1982–99 period, even corrected for the moderate increase in irrigation coverage over that period. Farmers appear to believe that the benefits of the green revolution are diminishing. 8.5 Expected Growth, Wealth, and School Investment Table 8.3 reports separate estimates of the effects of expected growth rates, wealth levels, and the existing stocks of secondary schools on new sec-

Does Economic Growth Increase the Demand for Schools? Table 8.3

Expected Future Annual-Yield Growth, Current Wealth, and New Secondary School Building

Variable Expected annual yield growth

a

Secondary school in villagea Log of wealth per household Adverse weather in 1982 Log of village population size in current period Log of price per acre of irrigated land in 1971 Adverse weather in 1971 Yield in 1971

347

All Schools

Private Schools

Public Schools

3.25 (2.02) –1.53 (1.84) .0860 (2.63) .0491 (0.44) .0514 (1.03) .0000137 (1.26) –.00784 (0.06) –.000097 (0.43)

1.65 (1.49) –.579 (1.01) .0380 (1.69) .178 (2.29) .0341 (1.00) –.000005 (0.66) .147 (1.74) –.00010 (0.66)

1.59 (1.16) –.952 (1.34) .0480 (1.71) –.129 (1.33) .0173 (0.40) .000019 (2.00) –.154 (1.47) –.0000057 (0.03)

Notes: N
412. Fixed effects–instrumental variables (FE-IV) estimates. Absolute value of t-statistics in parentheses. a Endogenous variable: Instruments include the log of the irrigated land price, log wealth in 1971, village population size in 1971, household wealth in 1971, and number of households in 1971.

ondary-school building for all secondary schools, public schools, and nonpublic schools. The estimates are obtained using fixed effects with instrumental variables to eliminate the influences of both fixed areal preferences for schooling and errors in forecasts that will bias the estimates. The variables included in addition to wealth, weather, land price, and school stocks, as noted, serve to minimize bias in the key variable coefficients in the presence of forecast errors. The estimates are consistent with the hypothesis that the building of new secondary schools responds positively to expectations of future yield growth, for given levels of wealth and given the existing stock of schools. Additionally, increases in wealth, net of changes in expectations of future growth, also appear to have a positive effect on subsequent school construction. The point estimates for total secondary schools suggest that a rise in the annual rate of growth in yields from, say, the expected 3.5 percent to 4.5 percent would increase by 24 percent the number of secondary schools built over a ten-year period for the average village. Given expectations about future yield growth, this effect is equivalent to an increase in average wealth of 30 percent, which the point estimates suggest would increase the number of secondary schools in a decade by 21 percent on average. Wealth increases thus appear to have a significant but less powerful effect on rural secondary school construction compared with raising the expected growth rate and thus the returns to schooling.

348 Table 8.4

Andrew D. Foster and Mark R. Rosenzweig Expected Future Annual-Yield Growth, Current Wealth, and School Enrollment Rates, Children Aged 10–14 Years

Variable Expected annual-yield growth

a

Log of wealth per household Adverse weather in 1982 Log of village population size in current period Log of price per acre of irrigated land in 1971 Adverse weather in 1971 Yield in 1971

All Children

Boys

Girls

2.28 (2.21) –.00814 (0.33) –.0734 (1.07) –.0484 (1.34) .0000163 (2.09) –.0930 (1.09) .0000238 (0.23)

3.79 (3.00) –.109 (3.62) –.0216 (0.26) –.0127 (0.29) .000002 (0.21) .0113 (0.11) –.000001 (0.01)

1.28 (1.10) .0538 (1.95) –.0976 (1.26) –.0640 (1.59) .000028 (3.20) –.124 (1.29) .000089 (0.75)

Notes: N
410. Fixed effects–instrumental variables (FE-IV) estimates. Absolute values of t-statistics in parentheses. a Endogenous variable: Instruments include the log of the irrigated-land price, log wealth in 1971, village population size in 1971, household wealth in 1971, and number of households in 1971.

The sets of nonpublic and public school estimates indicate nonrejection of the hypothesis that each type of school responds equally to changes in expectations and wealth. The estimates of the determinants of decisions about new school building thus suggest that public and private secondary schools are viewed as close substitutes by the villagers. Do the effects of expected returns and wealth on secondary school accessibility show up as well in schooling investments? The household survey data for 1971 and 1982 provide information, elicited from the households, on the numbers of children enrolled in school. These data thus do not suffer from the problems associated with enrollment information provided by school officials, who have an incentive to inflate enrollment figures.10 Table 8.4 reports estimates of the determinants of school enrollment rates for all children aged ten to fourteen and for boys and girls separately in the same age category. These estimates indicate that both school construction and investments in schooling respond positively to expected yield growth. The point estimates suggest that a 10 percent rise in the expected annual growth rate in yields, for given initial wealth levels, would raise overall rural school enrollment rates by 5 percent, that for boys by almost 7 percent and that for girls by 3.3 percent. However, wealth appears to have an insignificant but negative effect on overall enrollment rates. 10. However, the data on enrollment, even if accurate, may overstate the amount of investment in schooling to the extent that there is variation in school attendance. Only the 1982 survey provides information on children’s time devoted to “study.”

Does Economic Growth Increase the Demand for Schools?

349

Does the lack of a positive wealth effect on schooling call into question the results for school building? The facts that enrollment rates reflect the allocation of time, while school construction does not, taken together with the separate sets of estimates for boys and girls, provide an answer. The enrollment rate estimates indicate that enrollment rates for boys respond strongly and positively to changes in expected future increases in yields, but boys’ enrollment in school declines with increases in household wealth, for given growth expectations. In contrast, the wealth effect on enrollment for girls is positive, and the enrollment response of girls to expected future yield growth is substantially smaller than that for boys. These results together are consistent with the division of labor between boys and girls: Boys are much more likely than girls to engage in activities that contribute to the household’s farm income than are girls. Wealth in the rural villages is dominated by land wealth, and greater land wealth reflects the higher productivity of inputs, including labor inputs, on the land. The negative wealth effect for boys thus reflects the opportunity cost of the time of boys who attend school. The contrast between the enrollment rate equations for boys and girls thus makes visible the roles of the two components of the returns to schooling—opportunity costs and profitability. Increases in the productivity of land increase the opportunity cost of attending school for those children used in farm production, but expected increases in yield rates, expected growth, raise the gross returns to investments in skills, skills that have enhanced payoffs in a dynamic environment. A one-shot increase in productivity thus may reduce schooling investment, while the change to a regime of sustained, rapid growth in productivity will raise investments in schooling. 8.6 Conclusion The newly constructed time series data on 240 villages across most of rural India suggest that there has been considerable progress in the last forty years in the availability of rural secondary schools. In 1960, less than onethird of villages were located within ten kilometers of a secondary school; now almost 90 percent of the villages have access to at least one secondary school. Rates of secondary-school construction have been particularly rapid in the poorer villages, and the distribution in the availability of secondary schools across Indian states is far less unequal now than forty years ago, although there are still large disparities. The progress in rural secondary school building has been accompanied over the same period by a rapid increase in crop productivity. The data show, however, that rates of increases in yields have slowed in recent decades, and our analysis of the data suggests that farmers expect that yield growth will slow down further in the future. Our analysis also suggests that

350

Andrew D. Foster and Mark R. Rosenzweig

expected growth rates in agricultural productivity have had significant effects for given wealth levels on school construction and on enrollment rates, while wealth effects, for given productivity increases, have on average had negligible effects on enrollment. These results are consistent with the hypothesis, advanced by many economists, that returns to schooling are higher in a dynamic environment and suggest that the Indian population responds to expectations of higher returns to schooling with respect to both school construction and schooling investment. Our findings, based on the large spatial variation in productivity growth experienced in India over the last forty years, suggest that augmenting and sustaining the rate of economic growth is a powerful policy tool for increasing the human capital of the population. The results should not be interpreted, however, as implying that augmenting agricultural growth rates leads to increased employment in agriculture. As farmers’ wealth levels increase, so does the demand for nonagricultural products, creating new opportunities for the use of skills outside of agriculture. When the new 1999–2000 household survey data are completed, it will be possible to assess by schooling level the extent to which agricultural productivity growth was associated with rural, nonagricultural employment. Nor should these findings be interpreted as suggesting that the only route to raising human capital levels is via investments in agricultural productivity. There are clearly areas of India in which the potential for raising crop productivity is low. To the extent that opportunities are expanded for the exploitation of skills outside of agriculture that arise, say, from the freeing-up of market restrictions, we would expect similar responses with respect to schooling investment. The results are also consistent with the idea that just as increased schooling facilitates the exploitation of new agricultural technologies by farmers, it also enhances abilities to exploit opportunities outside of agriculture and even outside the rural sector.

Appendix Using the Land Price to Identify the Effects of Expected Growth on School Building We posit a model in which plans for rural investments in schools in any period t over the subsequent interval t to t  1 is driven by the expected returns to schooling at time t, which we assume to be positively related to the expected rate of agricultural technical change. Investment plans are also influenced by current wealth, At, population size, Nt , and the existing stock of schools, St , at time t. At time t the rate of productivity growth between periods t and t  1, rt , is unknown, so that schooling decisions must be made

Does Economic Growth Increase the Demand for Schools?

351

based on expectations as of time t. Let st denote new school investment and Et rt denote the time t expected rate of growth in agricultural productivity between periods t and t  1. If et is the deviation of actual from expected growth, then the true rate of increase in productivity is (A1)

rt
Et rt  et ,

with et uncorrelated with anything known by the school investors at time t. The schooling decision rule may thus be written as (A2)

st
r Ert  A At  N Nt  S St    ut ,

where  captures time-invariant determinants of schooling decisions that are unmeasured and may vary across decision-making units. The parameter r reflects the influence of expected productivity growth on current schooling decisions. Identifying r is thus the key to assessing whether schools are built in response to growth prospects, which raise the returns to schooling. Using actual future productivity growth to substitute for expected growth in equation (A2) results, however, in inconsistent estimates of r. To see this, substitute equations (A1) into (A2) to get (A3)

st
r rt  A At  N Nt  S St – r et  ut .

Inconsistent parameter estimates arise due to the correlation of rt with et , the forecast error. The problem is essentially one of measurement error, in that actual productivity growth is a noisy estimate of expected productivity growth. The estimate of r is biased, along with the other parameters in equation (A3), due to this measurement or forecast error. A standard treatment for measurement error is instrumental variables. What is needed is a variable observed at time t that predicts expected productivity growth but does not otherwise appear in equation (A3). We make use of the fact that the price of any plot of land, pt , given an efficient market for land, reflects expectations about the stream of future revenue on that land, appropriately discounted. Thus, 

(A4)

pt
Et
 s yts , s
0

where  is the discount factor and yt is revenue from the land in period t. Assuming for notational simplicity a constant rate of growth rt from period t onward, equation (A4) can be rewritten as (A5)

 Et yt pt
Et yt
 s (1  r) s
 , 1 – [1  rt ] s
0

where income in period t, given by (A6)

yt
Et yt  wt ,

352

Andrew D. Foster and Mark R. Rosenzweig

is stochastic due to weather shocks wt . Thus, given equation (A3), the price of land at time t can serve as an instrument for expected future productivity because the land price, conditional on actual land income at time t, given by equation (A6), reflects expectations about the future productivity increase rt but has no direct effect on investment plans, given current wealth. A second problem, however, is the presence of persistent, culturally determined factors, impounded in  in equation (A3), that influence preferences for schooling. Such factors will in general be correlated with the existing stock of schools, thus inducing a correlation between the stock variables, St , and the residual. In the presence of the fixed effect, the instrumental-variables approach would not work, because, for example, prices will be higher in areas with higher tastes for schooling as long as more schooling results in greater returns to technical change, resulting in a correlation between the instrument and the residuals in equation (A3) which contain preferences . The standard remedy for controlling for time-invariant unobservables by differencing and applying instruments is, conversely, complicated by the nonobservability of expected productivity. Substituting equation (A2) and differencing across time yields (A7)

st
r rt  A At  N Nt  S St – r et  ut ,

where st
st1 – st and so forth. The standard problem that differenced state variables such as, in this case, St , are directly influenced by first-period shocks ut and can be addressed by instrumenting with initial-period stock levels. However, the approach suggested above of using land prices as instruments to remove the measurement error arising from the nonobservability of expected productivity must also be modified. In particular, differenced prices are not appropriate instruments because forecast errors in period t to t  1, et , will, definitionally, be correlated with actual incomes in period t  1 and will also likely influence the forecast of growth at t  1, and thus will be correlated with price in period t  1. While in principle it is possible to use initial prices pt to instrument the change in expected productivity growth, there is reason to expect that this instrument would be too weak to be of value in practice. In order to address this problem, we therefore make use of a linear approximation to the price equation (A4): (A8)

pt
y Et yt  r Et rt .

Substituting for the expectations variables and solving for et yields (A9)

y rt – 1 et
  ,

r(yt – wt )

t pt

which may, in turn, be substituted into equation (A7) to obtain

Does Economic Growth Increase the Demand for Schools?

(A10)

353

st
r rt  A At  N Nt  S St

y rt – 1  r    – ret1  u t ,

r (yt – wt )

r pt





which may be simplified to (A11)

st
rrt1  A At  N Nt  S St r r  

r ( yt – wt )

r –  – ret1  ut .

r pt

Note that in equation (A11) growth no longer appears in differenced form, and that since et no longer appears in the residual, differenced prices may be used to instrument growth in period t  1, rt1. As noted, first-period school stocks serve as instruments for the subsequent change in school stocks St . A key assumption in the above analysis is that equation (A9) holds exactly. While this proposition is not directly testable due to the unobservability of expected incomes Et yt , an indirect specification test may be constructed by solving the price equation (A5), given equation (A6), for rt : (A12)

– y 1 rt
  .

r (yt – wt )

r pt – et

If the forecast equation (A12) is correctly specified, the residual of this equation contains only the forecast error for the period between t and t  1, which must be uncorrelated with any variables known to the village at time t. By contrast, if one or more of the observables in equation (A12) are measured with error, a correlation is likely to be observed. A simple specification test is thus to regress rt on yt , wt , pt , and a measure of lagged yields yt–1. A significant coefficient on lagged yields would suggest that measurement error is present and thus would invalidate the identification strategy suggested above.

References Behrman, Jere R., Andrew D. Foster, Mark R. Rosenzweig, and P. Vashishtha. 1999. Female schooling, home teaching, and economic growth. Journal of Political Economy 107 (4): 682–714. Benhabib, Jess, and Mark Spiegel. 1994. The role of human capital in economic development: Evidence from aggregate cross-country data. Journal of Monetary Economics 34 (May): 143–73. Drèze, Jean, and Amartya Sen. 1998. India economic development and social opportunity. Delhi: Oxford University Press. Duflo, Esther. 1999. Schooling and labor market consequences of school construc-

354

Andrew D. Foster and Mark R. Rosenzweig

tion in Indonesia: Evidence from an unusual policy experiment. MIT, Department of Economics. Mimeograph. Foster, Andrew D., and Mark R. Rosenzweig. 1995. Learning by doing and learning from others: Human capital and technical change in agriculture. Journal of Political Economy 104 (December): 1176–1209. ———. 1996. Technical change and human capital returns and investments: Evidence from the green revolution. American Economic Review 86 (September): 931–53. Lau, Lawrence J., Dean T. Jamison, Shu-Cheng Liu, and Steven Rivkin. 1993. Education and economic growth: Some cross-sectional evidence from Brazil. Journal of Development Economics 41 (June): 45–70. Nelson, Richard, and Edmund Phelps. 1966. Investment in humans, technological diffusion, and economic growth. American Economic Review Papers and Proceedings 61 (May): 69–75. Psacharopoulos, George. 1988. Education and development: A review. The World Bank Research Observer 3 (1): 99–116. Rosenzweig, Mark R. 1995. Why are there returns to schooling? American Economic Review, Papers and Proceedings 85 (2): 153–58. Schultz, Theodore W. 1975. The value of the ability to deal with disequilibria. Journal of Economic Literature 13 (December): 827–46. Welch, Finis. 1970. Education in Production. Journal of Political Economy 78 (1): 35–59.

9 Priorities for Further Reforms Anne O. Krueger

Indian economic policy reforms of the 1990s are best characterized using the good news/bad news format. The good news is that they have achieved a great deal in liberalizing the state-controlled economy of the prereform era and integrated it to the world economy to a significant extent. But the bad news is that, while other reformers have gone farther, India’s reforms appear to have stalled after the politically relatively easy steps were taken. Both viewpoints obviously have a significant element of truth. Relative to the unchanging economic policy stance of the period from 1950 to 1990, the reforms in the 1990s were remarkable. However, contrasted with the economic policy framework that would permit much more rapid alleviation of poverty, rising living standards, and faster economic growth, much remains to be done. This chapter outlines priorities for a future reform agenda. They do not by any means exhaust the policy measures that could improve economic performance: As in almost all countries, to do so would require an encyclopedia, as regulations are individually analyzed with a view to either their elimination or their alteration. And, as has been evident from the papers and discussion in this volume, as well as from the experience of other policy reforms, details make a huge difference.1 Instead, the question here is what the key areas are that urgently need addressing in order for India to

Anne O. Krueger is currently the first deputy managing director at the International Monetary Fund. At the time of writing this chapter, she was the Herald L. and Caroline L. Ritch Professor of Economics, and director of the Center for Research on Economic Development and Policy Reform at Stanford University. 1. See Krueger (2000), especially chapter 16.

355

356

Anne O. Krueger

have reasonable prospects for achieving an average annual growth rate of 7 percent or more for the next decade.2 As of 2000, the growth rate was around 6 percent, but there was some evidence that it was decelerating somewhat from the late 1990s. Moreover, there had been ten consecutive good monsoons, and given past history, it cannot be expected that the periodic droughts that have afflicted agriculture in past years will have ceased entirely (although irrigation and other investments have somewhat reduced vulnerability to them, as has the reduction in the proportion of gross domestic product originating in agriculture). Many of India’s current economic ills would be greatly alleviated by a sustained 7 percent growth rate. Such a rate would enable more people to achieve incomes above the poverty line. Combined with needed reforms, nonagricultural employment opportunities would increase at a significantly faster rate than has historically been the case. In fact, there is evidence from the National Sample Survey data on employment and unemployment that employment in organized manufacturing industries in the private sector has grown in the reform period as compared to its stagnation in the 1980s. Additionally, with an increase in the rate of growth of employment and incomes, other barriers to more rapid growth might be expected to diminish: As documented by Foster and Rosenzweig, more rapid growth would likely result in decisions of a higher proportion of the eligible schoolage population to undertake more years of schooling. A higher growth rate, again with appropriate incentives, would also likely result in a higher savings rate, thus enabling, among other things, more investment in infrastructure, clearly a critical bottleneck as discussed below. If the tax reforms discussed by Srinivasan and Singh (or other reforms) were adopted and government revenues were more elastic with respect to income growth, the fiscal situation could itself be eased by accelerated growth. Achievement of a 7 percent (or more) annual rate of growth, however, will require both immediate measures to avoid short-run crisis and longerrun measures to enable the provision of a greater quantity and improved quality of infrastructure and human capital, both to increase the efficiency of the private sector and to generate political support for these measures. Clearly, the first short-run threat to the sustainability of Indian economic growth—even at the 6 percent level—is the fiscal situation. As spelled out in Srinivasan’s paper, and emphasized by virtually all (including the Ministry of Finance and the Reserve Bank of India), achieving a significant reduction in the present and prospective fiscal balances is a prerequisite for avoiding another fiscal crisis of the sort that occurred in 1991. Such a crisis 2. It is this author’s conviction that an even higher growth rate—8 or 9 percent—would be feasible were reforms undertaken single-mindedly in order to achieve growth. In the current Indian political environment, however, that degree of focus on a single objective does not seem politically feasible. Whether the reform process can be reinvigorated at all is the more important question.

Priorities for Further Reforms

357

would force the government of India (GOI) to take the necessary measures, but they would then come at the cost of significantly retarding economic growth for at least a few years. Such a retardation would in itself entail high costs, given the urgency of poverty alleviation. At present, the fiscal deficit is financed in considerable part through the forced acquisition of government obligations by the banks. While that may have the effect of reducing the inflationary pressures emanating from the fiscal deficit, it also has the effect of keeping real interest rates higher than they would otherwise be. If the fiscal deficit were greatly reduced, the resulting drop in real interest rates would enable greater private investment and hence stimulate economic growth. Addressing the fiscal situation is all the more urgent because it currently stands as a barrier to other needed reforms: Some of the prospective bottlenecks in infrastructure can be attributed to the states’ allocation of expenditures almost entirely to wages and salaries, with few resources to undertake investments. Moreover, there is need for more expenditures in areas such as education, health care services, and infrastructure, which cannot be provided until the fiscal gap is closed and new sources of revenue identified and tapped. That makes the fiscal challenge all the greater, and the urgency of finding new revenue sources and reducing wasteful expenditures—such as the very high fraction of subsidies (including the below-cost user charges of some infrastructure services) that go to the better-off sections of society rather than to the poor—all the more pressing. Perhaps equally important, particularly for mobilizing political support for further needed and more difficult reforms, is to convince the majority of the Indian people that the reforms are essential and in their own interest and that, even in the few years of reform, benefits have accrued, not only to the urban middle class and the rich, but to the rural poor as well. There is considerable evidence, as stressed by many conference participants, that the majority of Indians are not even aware that there have been reforms, let alone that they have benefited by them. Some of those aware of the reforms being undertaken nonetheless seem to believe that reforms were designed to benefit only the rich. While there is considerable evidence from surveys undertaken by the National Council of Applied Economic Research that the living standards of the poor have in fact risen with more rapid economic growth, the connection between reforms, more rapid growth, and rising living standards does not appear to be widely understood or recognized. Obviously, in a democratic society, reforms can proceed only with the consent, if not the support, of the majority of the population. Finding means to achieve greater support will undoubtedly entail a number of measures. Better explanations of reforms and their benefits by politicians and policy makers may make a significant difference. So, too, could more attention to education and health policy. Combining the reduction of subsidies—essential for fiscal reasons, as spelled out by Srinivasan—with the in-

358

Anne O. Krueger

troduction of much more highly focused elements of a social safety net for the poor would also make a difference. In addition, attention to agriculture, including provision of research and extension services, better delivery and pricing of water, and other measures, is clearly desirable both because it may generate needed support for the altered policies and because it will directly contribute to the end objectives of reforms: to achieve more rapid growth and higher living standards. In the intermediate run, there is universal agreement that Indian infrastructure in communications, transportation, and power will constitute a significant bottleneck even to the maintenance of a 6 percent growth rate, much less the achievement of an even higher 7–8 percent, unless investments in these activities accelerate considerably. Although the government of India has recognized the urgency of improving infrastructure quantity and quality, the measures taken to date both to attract private financing and to increase public expenditures do not appear likely to be able to support a 7 percent or higher growth rate. Indeed, some private foreign firms that had agreed to invest in urgently needed power projects have now withdrawn in frustration at the delays in obtaining clearances and reaching financial closure. Clearly, the environment for private investment in power and other infrastructure must be made more attractive, or the public sector must somehow raise additional resources. Indeed, it is probable that both private- and public-sector investments will be needed, and in greater quantity than is currently contemplated. Achieving an acceleration in the rate of increase in infrastructure capacity is complex: It involves center-state relations, especially in power and telecommunications. Removing subsidies and achieving reasonable user charges is highly desirable, not only for the contribution this would make to a sounder fiscal situation, but also because it would enable a more efficient use of existing infrastructure and help to finance additional investment in infrastructure. Unlike other areas that are high-priority candidates for reform, there is little, if any, political opposition to increasing the rate of infrastructure investment and capacity. The disagreements and bottlenecks lie in the arrangements for doing so: the extent to which private foreign investment in infrastructure facilities is encouraged; the degree to which states have the political will to raise tariffs on power and other infrastructure services; the sources of domestic funds to finance additional infrastructure investments; and the allocation of new resources across states. Given the long lags that can occur between decisions to build power plants or expand road or port capacities, failure to address these issues in the near term will have farreaching consequences for growth.3 To date, telecoms reform seems to have gone furthest. The Telecom Reg3. See chapter 1 of this volume, and the Government of India (1996).

Priorities for Further Reforms

359

ulatory Authority of India has been established, and issues surrounding the respective roles of the regulator and the provider are being sorted out. As this chapter was being written (September 2000), it was announced that the government’s monopoly of overseas telephone service would end in 2002.4 For many of the same reasons, accelerating the momentum for privatization of public-sector enterprises (PSEs) is highly desirable. Simultaneously, it needs to become clear to private firms that the government will no longer acquire “sick” firms and that unprofitable firms will go bankrupt. For large and visible private enterprises, the government of India has in the past acquired “sick” firms in order to maintain employment; the result has been that an increasing fraction of PSEs has been loss-making. Achieving privatization in ways that permit these firms to become more efficient, and closing down enterprises in which there is no hope of profitability, would do much both directly and indirectly to enhance the efficiency of the Indian economy and to improve growth prospects. Directly, the elimination of losses of unprofitable PSEs and permitting the closure of unprofitable private enterprises would raise the average productivity of Indian enterprises, while simultaneously freeing resources for more productive uses and improving the fiscal situation. Indirectly, the incentive effect for other firms could also contribute to increased output and productivity, as managers perceived that they had more to lose with poor performance of their enterprises. As of the time of the conference in May 2000, only one firm (a bakery) had been privatized. However, in the summer of 2000, there were indications that the government of India might be moving forward with respect to privatization efforts, prodded in part by the fiscal situation. Accelerating privatization activities and altering exit policy (either with formal declarations as to an altered policy or through the failure to acquire or support losing activities) could significantly improve growth prospects and contribute to resolving the fiscal situation. An area where further reform is urgently needed for accelerated growth is Indian government regulation of private-sector activity. While, as Forbes, Saxenian, and Murthy all discussed, regulation has become less onerous since the start of reforms, it is still there in abundance. Indeed, there is a risk, as noted by Saxenian, that the government of India will not liberalize generally, but instead undertake measures to foster the IT sector, ignoring the great potential for growth in traditional labor-intensive industries. As the Indian economy grows, the importance of uniform treatment for economic activities of different kinds will increase, and sector-specific regulations and incentives need to be phased out as rapidly as possible. 4. Three unions, with 380,000 members, immediately went on strike seeking guarantees of job security and pensions after the Department of Telecommunications was turned into a company on 1 October as a first step toward privatization. See Financial Times, 7 September 2000, 4.

360

Anne O. Krueger

As Rakesh Mohan’s chapter makes clear, the policy of reservation of certain industries to the small-scale sector has almost certainly had major deleterious effects on the Indian economy. It has not promoted the small-scale sector, but has instead given monopoly positions to large-scale firms that were already engaged in these activities before regulation was enacted. It has certainly not promoted employment, as the failure of employment in manufacturing to grow has been a vivid indicator of the fact that Indian development was not reaching an increasing percentage of the labor force. It has certainly deterred exports, which would likely have originated in laborintensive industries, precisely the ones that have been reserved for smallscale firms. Dereservation of small-scale industry (SSI) would certainly improve the functioning of not only the labor market but also the entrepreneurial market. Small firms with good managers and growth potential would be encouraged to expand, instead of remaining small (as is required under current policy). It is impossible to ascertain how much benefit would come from removal of SSI reservation, but it would surely be considerable. While the issue is probably politically contentious, there are ways in which the transition could be handled to reduce opposition. As suggested by Roger Noll, dereservation might be phased in over a period of several years, and support to existing enterprise might be extended during that time to enable them to make the adjustment. During the first year or so, existing SSI would be permitted to expand, but new entrants would be deterred to give time for adjustment of existing activities. Other transition solutions are also possible, but the key point is that there are continuing losses through the SSI policy, and there would be only a one-off transition to achieve the removal of these losses. A considerable amount of deregulation also remains to be done. Naushad Forbes’s chapter vividly demonstrates that the heavy hand of regulation has been at least somewhat lightened, but equally it shows that there is still a considerable amount of government intervention, whose only effect is often merely to slow down the rate at which entrepreneurs can respond to opportunities. Especially when these opportunities are for exports to foreign countries, the delays and other impediments presented by these regulations can give an overwhelming advantage to foreign competitors. One discussant noted that the reforms that had taken place so far had at least begun to reduce or mitigate government regulation of the private sector, but that government itself is still in need of economic reform. Issues of staffing of state enterprises, of the ways in which officials issue licenses and enforce regulations (even when there is a legitimate case for regulation), and of the very slow functioning of the legal system are all areas in which there is ample scope for increased efficiency within the public sector. At the same time, government delivery of education, health, agricultural research and extension, the judicial system, and other services can be

Priorities for Further Reforms

361

greatly streamlined. Not only could the government improve productivity by removing regulations that are counterproductive and more efficaciously enforcing needed regulations, but also its delivery of services itself could be greatly streamlined. Another vital area for reform is the Indian labor market. In a country as poor as India, it is understandable that the loss of a job is nothing short of a disaster in most instances. But the failure of productive employment opportunities to grow more rapidly is also a matter for great concern. As Mohan demonstrates in his chapter, employment opportunities have grown much more slowly in India than they did in other poor countries at their early stages of development. However, the rigidity of current labor laws in India is probably a major factor in deterring employers from taking on new workers, and also in preventing India from realizing the full benefits from the international economy that could be realized with a more flexible labor market. Devising mechanisms to provide some degree of social safety net for hardships would in the long run be far cheaper than maintaining the existing law prohibiting layoffs. That law obviously encourages employers to regard labor, like capital, as a fixed cost, and to substitute capital for labor to a much greater degree than would take place in a more flexible labor market. Another area, as yet virtually untouched, where there is ample room for considerable gain is in the area of center-state relations. On issue after issue at the conference, the complexities of center-state relations were seen to be a major contributing factor to difficulties. This was the case, for example, with efforts to attract private direct foreign investment into the construction of power-generating facilities. Under the constitution, the states are entitled to set the tariffs for power, while the central government must provide whatever guarantees need to be given regarding rights of profit repatriation and the like. The divided jurisdiction has led to a number of power project cancellations, and has slowed down even those that have proceeded. But the center-state relationship affects much else. Many of the subsidies, which were discussed in the Srinivasan chapter, are set at the state level. The fact that farmers often get free power, or get it at very low cost, has been mentioned. But water charges for irrigation and a number of other subsidies are also state-determined. While the states are enabled to borrow from the center (even if not entirely in accordance with the law) or to fail to repay as scheduled, their “soft” budget constraint encourages this type of behavior and misunderstanding. At the same time, the center, which cannot raise revenues through tax mechanisms the constitution reserves for the states, relies on less efficacious taxes that have much stronger disincentive effects throughout the economy. A final area in which there is ample scope for reform is the financial sector. It is widely agreed that this is a field in which liberalization has made a major difference already. But there is still much that needs doing. The In-

362

Anne O. Krueger

dian banks are still subject to guidance with regard to their holding of government paper (the requirement has been reduced but is still substantial), with regard to provision of credit for particular groups (such as SSI, small farmers, etc.), and for the allocation of credit more generally. Strengthening the Indian banks in and of itself, and enabling them to allocate credit to the most productive users, rather than by government allocation, would make a considerable contribution to the Indian economy’s growth potential. So, too, would removing the requirement that the banks finance the central government. This latter course of action, of course, is related to the solution to the government’s fiscal problems, mentioned above and analyzed by Srinivasan. Finally, it must be noted that, as is befitting an academic conference, there are a large number of areas in which additional research could greatly inform the policy reform process. The fact that there has been little detailed analysis of the effects of SSI at the microeconomic level is a striking example. So, too, is the difficulty that confronted Kochar and Rosenzweig in attempting to analyze Indian education: Data were available in reliable form only for a few years, and in many instances the variables that would have been most valuable for the analysis were not available at all. Studies of the price differentials between domestic consumer goods (whose importation has not yet been permitted) and imported goods, of the relative labor staffing of Indian public-sector enterprises and comparable producers in other countries, and of a variety of other matters would give policy makers a much better sense of the relative costs of different economic policies and of the potential benefits to be had from their alteration. While it is easy to point to the many reforms that could make a difference in India, it is not possible to pinpoint those that would achieve the greatest gains, and since policy makers have only limited capacity for implementing reforms at any given time, knowledge of the potential gains could guide the choice of next targets for reform. While it is to be hoped that the papers in this volume have contributed significantly to that knowledge, it is clear that much more research needs to be undertaken.

References Government of India. 1996. India infrastructure report: Policy imperatives for growth and welfare. Report of Expert Group on the Commercialization of Infrastructure Projects for the Ministry of Finance. New Delhi: Thomson Press. Krueger, Anne O. 2000. Economic policy reform: The second stage. Chicago: University of Chicago Press.

Conference Participants

Rajendra Abhyankar Consulate General of India Shankar Acharya Indian Council for Research on International Economic Relations Isher Judge Ahluwalia Director, Indian Council for Research on International Economic Relations Montek S. Ahluwalia Director, Independent Evaluation Office, International Monetary Fund Chonira Aturupane International Monetary Fund Abhijit Banerjee Department of Economics, Massachusetts Institute of Technology Marshall Bouton Executive Vice President, Asia Society Naresh Chandra Indian Ambassador to the United States

P. Chidambaram Former Minister of Finance, Government of India Sajjid Chinoy Doctoral Student, Department of Economics, Stanford University Michael Clark Executive Director, U.S.-India Business Council A. W. Clausen Chairman and CEO (Retired), Bank of America Tarun Das Economic Adviser, Ministry of Finance, Government of India Ashok Desai Consulting Editor, Business Standard Rafiq Dossani Senior Research Scholar, Asia/Pacific Research Center, Stanford University William Draper, III Draper International

K. B. Chandrasekhar Chairman, Exodus Communications

363

364

Conference Participants

Naushad Forbes Director, Forbes Marshall Peter Geddes Greystone Capital, Ltd. Stephen H. Haber Department of History, Stanford University Arnold Harberger Department of Economics, University of California at Los Angeles Nicholas Hope Deputy Director, Center for Research on Economic Development and Policy Reform, Stanford University K. V. Kamath Managing Director and CEO, Industrial Credit and Investment Corporation of India Narinder Kapany Chairman, K2 Optronics Kenneth Kletzer Department of Economics, University of California at Santa Cruz Anjini Kochar Department of Economics, Stanford University Stephen Krasner Department of Political Science, Stanford University Michael Kremer Department of Economics, Harvard University Anne O. Krueger First Deputy Managing Director, International Monetary Fund Rolf J. Luders Pontifical Catholic University of Chile Ronald McKinnon Department of Economics, Stanford University

Rakesh Mohan Chief Economic Adviser, Ministry of Finance, Government of India N. R. Narayana Murthy Chairman and CEO, Infosys Technologies, Ltd. Roger Noll Department of Economics, Stanford University Kiran Pasricha Director, Confederation of Indian Industry Suhas Patil President and CEO, Tufan, Inc. Kanwal Rekhi Co-Founder and past president, The Indus Entrepreneur Mark Rosenzweig Chair, Department of Economics, University of Pennsylvania Henry S. Rowen Director Emeritus, Asia/Pacific Research Center, Stanford University Il SaKong President, Institute for Global Economics AnnaLee Saxenian Professor of Regional Development, University of California at Berkeley Parth Shah President, Centre for Civil Society Raj Sharma New India Technology Ventures Edmund Shea Vice President, J. F. Shea & Co., Inc. George P. Shultz Distinguished Fellow, Hoover Institution, Stanford University

Conference Participants N. K. Singh Member, Planning Commission, Government of India Nirvikar Singh Department of Economics, University of California at Santa Cruz Yashwant Sinha Minister of Finance, Government of India

365

Neeraja Sivaramayya Center for Research on Economic Development and Policy Reform, Stanford University T. N. Srinivasan Department of Economics, Yale University, and nonresident Senior Fellow, Center for Research on Economic Development and Policy Reform, Stanford University

Contributors

Shankar Acharya Indian Council for Research on International Economic Relations India Habitat Center Core 6A, Lodhi Road New Delhi 110 003 India Montek S. Ahluwalia International Monetary Fund 700 19th Street NW Washington, DC 20431 Sajjid Chinoy Department of Economics Stanford University Stanford, CA 94305 Ashok Desai Business Standard Bahadurshah Zafar Marg New Delhi 110 002 India Naushad Forbes Forbes Marshall PB No. 29, Mumbai-Pune Road Kasarwadi, Poona 441 034 India Andrew D. Foster Department of Economics, Box B Brown University Providence, RI 02912

Kenneth Kletzer Department of Economics University of California at Santa Cruz Santa Cruz, CA 95064 Anjini Kochar Department of Economics Stanford University Stanford, CA 94305 Anne O. Krueger International Monetary Fund 700 19th Street NW Washington, DC 20431 Rakesh Mohan Ministry of Finance, Government of India North Block New Delhi 110 001 India N. R. Narayana Murthy Infosys Technologies, Ltd. Electronics City, Hosur Road Bangalore 561 229 India Roger Noll Department of Economics Stanford University Stanford, CA 94305

367

368

Contributors

Mark Rosenzweig Department of Economics University of Pennsylvania 160 McNeil Building 3718 Locust Walk Philadelphia, PA 19104-6297 AnnaLee Saxenian University of California at Berkeley Berkeley, CA 94720-1850

N. K. Singh Planning Commission, Government of India Yojana Bhavan, Sansad Marg New Delhi 110 001 India T. N. Srinivasan Department of Economics, Yale University and Center for Research on Economic Development and Policy Reform Stanford University Stanford, CA 94305

Author Index

Acharya, Shankar, 50n. 6, 52n. 8, 54n. 11, 73n. 3 Aghion, P., 311 Agrawal, Pradeep, 40–41 Ahluwalia, Isher, 10, 12n. 6, 17, 18, 19, 31, 35n. 38, 92n. 1, 100n. 7, 132n. 3, 186 Aiyar, Swaminathan, 141, 142 Alesina, A., 311 Angrist, J. D., 307 Arora, Ashish, 170, 178–79 Asian Development Bank (ADB), 227t Bagchi, A. K., 143 Bajaj, Madhur, 147 Bajpai, N., 55n. 12, 96n. 5 Balakrishnan, P., 63 Bardhan, Pranab, 19 Bashir, S., 324 Benhabib, Jess, 330 Bhagwati, Jagdis, 12n. 4, 14n. 7, 130, 132n. 3 Bhalla, S., 100n. 7 Bolton, P., 311 Buiter, W., 49, 53n. 10, 77 Burgess, R., 53n. 10 Business India, 53, 54, 151–55, 176t Business India Intelligence, 158n. 33 Business Standard, 150n. 29 Card, D., 307, 308 Cashin, P., 51, 53n. 10, 77, 96n. 5 Central Statistical Organization, 11n. 2

Chadha, Rajesh, 244n. 10, 245t, 247–53t, 255t Chakravarty, Sukhamoy, 10, 11n. 3 Chandrasekar, K. B., 184 Chandresekar, C. P., 52 Chatterjee, Somnath, 254, 256 Chidambaram, P., 204 Chinoy, Sajjid, 182, 185 Datcher, L., 324 Datt, G., 100n. 7 Department of Fertilisers, 63 Desai, Ashok, 141, 179, 182; 144 Desai, Padma, 12n. 4 Dore, Ronald, 141–42 Dossani, Rafiq, 180n. 18, 184 Drèze, Jean, 117, 95n. 4, 303, 308, 329n. 1 Duflo, Esther, 329 Economic Times, 36 The Economist, 130, 188n. 28 Evans, W. N., 324 Fallon, Peter, 227 Fields, Gary, 230 Forbes, Naushad, 142n. 15 Foster, Andrew D., 330, 33n. 4, 343n. 7 Galor, O., 311 Gandhi, Mahatma, 11, 214 Gandhi, Rajiv, 15, 158, 171

369

370

Author Index

Gandhi, Sonia, 64, 131, 142 Godrej, Nadir, 129–30 Government of India, 50, 58, 62, 66, 67–68, 358n. 3 Greico, Joseph M., 171n. 1, 203 Guhathakurta, Subhrajit, 221 Gulati, 63, 64 Gulati, Mukesh, 263, 291–97t Gupta, S. P., 100n. 7 Hanushek, E., 307n. 4 Heeks, Richard, 176, 203 Henderson, V., 324 Howes, S., 53n. 10 Humphrey, John, 263 Hussain, Abid, 213n. 1, 261, 267 India Today, 33, 54 Indira Gandhi Institute of Development Research, 18n. 13, 33n. 34 International Monetary Fund (IMF), 15t, 16t, 23n. 21 International Telecommunications Union, 182 Jhungihunwala, Ashok, 171, 189 Johnson, Harry G., 9 Joshi, Vijay, 10, 15, 21n. 17, 22, 23, 24, 53n. 10 Kim, Linsu, 145n. 20, 156n. 30 Kingdon, G. G., 308, 323, 324 Kletzer, Kenneth, 53n. 9, 56n. 13 Kochar, Anjini, 185, 329 Krueger, A. P., 307, 308 Krueger, Anne O., 182, 185, 355n. 1 Lal, Deepak, 100, 124 Lall, Sanjaya, 143t, 254 Lau, Lawrence, 330 Lavy, V., 307 Lazear, E. P., 307 Lewis, Arthur, 11n. 3 Little, I. M. D., 10, 15, 21n. 17, 22, 23, 24, 53n. 10 Lowell, B. Lindsay, 179 Mahalanobis, P. C., 12 Mahalingam, Sudha, 146n. 22 Mieszkowski, P., 324 Ministry of Finance, 32n. 31, 53n. 10, 60, 169, 170 Mohan, Rakesh, 100, 254, 256

Morawetz, David, 10, 11t, 32n. 32 Mukherji, Joydeep, 185, 186 Murthy, N. R. Narayana, 182 Nagraj, R., 96n. 5 Naidu, 180–81 Narasimha Rao, P. V., 21, 25 NASSCOM-McKinsey, 33, 169, 170t, 174– 76, 179, 188 Natarajan, I., 100 Nath, H. K. A., 48n. 2, 50n. 4 Nayar, Baldev Raj, 143n. 17 NCAER (National Council of Applied Economic Research), 132–33, 218 NCERT (National Council of Educational Research and Training), 303, 325t Nehru, Jawarhalal, 11, 143 Nelson, Richard, 142n. 14; 330 Ninan, T. N., 141 NSF (National Science Foundation), 143t Oates, W. E., 324 Olekalns, N., 51, 53n. 10, 77 Organisation for Economic Cooperation and Development (OECD), 11n. 2, 170 Parthasarathy, Balaji, 172, 173, 175n. 11, 176t, 177–78 Patel, U., 49, 53n. 10, 77 Phelps, Edmund, 330 Planning Commission, 32n. 31 Prabhakar, Mohana, 181n. 21 Premji, Azim, 131 PROBE Team, 308n. 6 Psacharopoulos, George, 329n. 2 Radhakrishna, R., 62 Ramaswami, B., 63 Ramaswamy, K. V., 243 Rao, M. G., 48n. 2, 50n. 4, 55n. 12 Reserve Bank of India, 25, 26, 50n. 5, 51 Robinson, Joan, 130 Rodrik, D., 311 Rosenzweig, Mark R., 330, 333n. 4, 343n. 7 Sachs, J., 55n. 12, 96n. 5 Saez, Lawrence, 184 Sahay, R., 96n. 5 Sahota, G., 48n. 2, 62n. 16 Sandesara, J. C., 244 Sauvageau, Y., 324 Saxenian, AnnaLee, 182, 190 Schmitz, Hubert, 263

Author Index Schultz, Theodore W., 330 Schwab, R. M., 324 Scindia, Madhavrao, 134 Sen, Amartya, 17, 95n. 4, 303, 329n. 1 Sen, Pronab, 172, 176 Seshagiri, N., 172, 181, 203 Singh, Manmohan, 21, 49, 64 Singh, N., 55n. 12 Sipahimalani, V., 307 Smith, Sean Eric, 157n. 32 Spiegel, Mark, 330 Sridharan, Eswaran, 172 Srinivasan, T. N., 10, 14n. 7, 19, 36, 37n. 39, 42–43 Stern, N., 53n. 10 Stiglitz, Joseph E., 52 Subbarso, K., 62 Subramanian, C. R., 64, 171

371

Tendulkar, Suresh, 18n. 12 Tilak, J. B. G., 323 Trivedi, Kamakshya, 123 UNCTAD, 273–79t, 280–86t, 287–90t Vajpayee, A. B., 169, 183 Varondakis, A., 96n. 5 Veganzones, M. A., 96n. 5 Weingast, B., 59, 79 Welch, Finis, 330 Wood, Adrian, 229 World Bank, 11t, 17n. 10, 18, 47, 48, 49, 50, 51, 55, 55n. 12, 57, 60, 63, 66 World Economic Forum, 32 Zeira, J., 311

Subject Index

Agricultural sector: crop productivity (1970–71, 1980–82, 1999), 18, 339–43; electric power subsidies for, 115; infrastructure needs provided at state level, 113; land prices reflecting future productivity growth, 331–32, 345–46, 350– 53; needed serious reform, 80; needed tax reform for, 61; post-independence performance of, 11 Balance of payments: capital account liberalization with deficit financing, 77–78; crisis (1966–67), 14, 47, 76; crisis (1991), 1–2; current account deficit (1980s–90s), 16, 21–22 Bangalore: contrast with Silicon Valley, 198–203, 205; entrepreneurship in, 199 Banking system: credit allocation and rationing system, 20, 217; current borrowing practices, 77; government-owned commercial banks, 265; nationalization (1960s), 14; post-1991 reforms related to, 24–26; pre-1991 debts, 49 Bodyshopping: defined, 172; development and managerial tasks related to, 178; related to software exports, 174, 176, 204 Bureaucracy: need to reduce state-level size of, 116; as obstacle to IT development, 186; state-level spending for, 54, 81, 116 Capital markets: government-subsidized borrowing in, 77; post-1991 liberaliza-

tion in, 132; regulatory framework of, 201; SSI problems in, 264–66 Competition: introduced into banking system, 26; post-1991 growth in industrial sector, 134; under post-1991 reforms, 147; protection of SSIs from large companies’, 215, 259; with reform in telecommunication services, 36–37; SSIs need to compete, 260; for statelevel private investment, 58; technical capability as core of, 142–43 Computer policy (1984), 181 Currency: devaluation (1966), 14; devaluation (1991), 21–22 Data sources: about SSIs in India, 267–72; to analyze effect of reservation policy, 241–43; capex database of investment spending, 103–6; to estimate number and composition of Indian SSIs, 227– 29, 267–72; for state- and central-levels of plan spending, 106–8; state-level GDP, 92; to test determinants of school building and schooling investment, 330, 332–33 Debt, external (1990–91), 48–49, 116 Disinvestment: fiscal impact of, 65–6. See also Privatization East Asian countries: export composition, 246–57; labor markets in, 40; manufacturing employment in, 224, 227, 230

373

374

Subject Index

E-commerce: constraints on components of, 201–2; WTO provisions for development of, 183 Economic growth: as determinant of demand for schooling, 331; effect of fiscal deficits on, 51–52; effect of good governance on, 117–18; export- and importrelated (1950s–60s), 13–14; fiscal deficit as factor in, 20–21; in India (1980s– 2000), viii–ix, 2, 47; infrastructure quality as determinant of, 110–12; 1992– 2000 period, 27–29; policy issues to enhance state-level, 112–21; rate of investment as factor influencing, 103–6; variations in state-level, 101–12 Economic performance: effect of infrastructure inadequacy on, 32–33; of India compared to other developing countries, viii; of IT industry, 169–70; preand postreform state-level, 91–101; of small scale manufacturing sector, 223– 57 Economic policy: effect of deficit financing on, 77–78; Five Year Plans, 1, 11, 12; liberalization (1990s), viii; related to improvement of living standards, 5; response to financial crisis (1991), 2. See also Fiscal policy; Monetary policy Economic reforms: after 1991 crisis, 21–27; agenda for, 138–42; beginning of stabilization period (1991), 47; as cause for regional inequality, 101–3; priorities for reform (2000), 29–43; related to World Bank and IMF arrangements, 79–80; for software and computer industries, 171–75 Education: factors influences enrollment levels, 307–10; government spending on elementary (1980–96), 305–6; home tutoring, 304; Indian Institutes of Education, 209–10; as investment in human capital, 329; national council for teacher education, 306; need to improve system of, 6; quality of elementary schooling, 303–4; related to computer science training, 176; related to quality of labor force, 39; returns to schooling, 329–31; role of information technology in, 202; school enrollment rates (1960s–90s), 17; to support IT industry, 186–87; technical training in IT, 194. See also Schools

Electrical power industry: effect of shortages, 33–35; needed reforms for, 114 Employment: in formal and informal labor markets, 40; in household and nonhousehold industries in India, 230–37; in IT industry (1990s), 194; manufacturing sector, 223–37, 257–58; post1991 growth, 132–3; in small-scale manufacturing, 223–37; in state-owned enterprises (1970s–80s), 19 Enrollment. See School enrollment Entrepreneurship: in Bangalore and Silicon Valley, 199; effect of reservation policy on, 298–99; incentives for, 190 Federalism. See Market preserving federalism Financial crisis (1991): economic reforms emerging from, 21–27; factors influencing, 16–17, 21; fiscal policy after, 22; response to, 2; trade policy after, 23–24 Financial sector: effect of reforms (1990s), 217; factors leading to crisis (1991), 16– 17, 21; post-1991 liberalization, 24, 30– 31; post-war public ownership in, 20; as priority for reform, 30–31; recommendations for venture capital in, 183–84; regulation of non-banking sector of, 26. See also Venture capital industry Firms, domestic: emergence of new, 135–36; saving of “sick,” 40–42, 67; subsidies to loss-making, 40–41; technology collaboration with foreign firms, 150, 156 Firms, foreign: investment in post-1991 period, 133, 136–38; IT operations and overseas development centers in India, 185, 197; in post-1991 markets, 147 Fiscal deficit: effect of monetization of, 51– 52, 77; effect of overspending (1980s), 15–16; growth influenced by, 20–21; increases (1990–2000), 72–73, 115; reduction (1990–91, 1996–97), 21, 49–51, 72; rising from expansionary fiscal policy, 48–49; of state- and central governments (1999–2000), 118–19; state-level (1990–99), 53 Fiscal policy: after 1991 financial crisis, 22; expansionary (1980s–90s), 1–2; need for correction of, 67–70, 77; proposed reforms, 60–62 Five Year Plans, 1, 11, 12, 14 Foreign direct investment (FDI): in Indian

Subject Index IT sector, 194; post-1991, 134; in software industry, 204–5 Government, central: infrastructure provisions by, 113; under market preserving federalism, 59; plan spending to explain investment, 106–8; provision of public goods by, 55; role of, 118–21; SSI support programs of, 217–19, 259; subsidies as proportion of spending of, 20, 62–65; transfer payments to states, 54–56 Government, state-level: borrowing to finance negative balance from current revenues, 114–15; central government assistance to, 119–20; financing of private schools by, 304; fiscal deficits of, 53–54; impact of reforms on growth in, 101–3; IT policy innovations, 179–84; lending institutions of, 217; major responsibilities of, 55; plan spending to explain investment, 106–8; restrictions on borrowing, 57; schooling equality related to spending by, 311, 313–20; SSI support programs of, 217–19, 259; subsidies provided by, 62–65. See also Schools Home tutoring, 304, 314, 319–21 Human capital: as benchmark for a country’s performance, 329; education as investment in, 329; of Indian labor force, 39. See also Education; Literacy rates; Schools Income per capita: growth (1990s), 17; in India, 1, 10–11 Industrialization: as focus of Second Five Year Plan, 12, 14; pre-1984 importsubstitution, 171 Industrial policy: effect of SSI policies and potential for reform, 298–302; licensing and regulation as instruments of, 258– 59; pre-1991, 143–44 Industrial sector: effect of abolition of licensing, 102; firm restructuring, 149– 50; growth (1985–88), 49; growth (1960s–90s), 18–19; infrastructure provided at state level, 114; post-1991 changes, 132–36; post-1991 technology development, 148–58; state industrial development corporations, 218; successes and failures in, 158–65. See also Manufacturing sector

375

Industry, large-scale (LSI): growth of, 241– 42, 257 Industry, small-scale (SSI): clusters in India, 260, 291–97; continued protection from competition, 134; credit support for, 217, 259–60; eligibility under reservation policy, 215–16; idea of clustering of, 263–64; list of clusters in India, 291– 97; needed tax reform for, 61; need financial support for startup, 264–66; potential effects of reform in policy for, 298–302; reservation policy effect on growth, 38; reservation policy to protect, 213–15; resistance to policy reform of, 299–300; sources of information about, 267–72; value added, 237–41 Inequality: impact of reform on regional, 101–3; in Indian society, 206, 310; in interstate regional development, 96–101; in school attainment, 304; of schooling, 304, 310–12, 316, 320 Information: in concept of SSI clustering, 264; to promote SSI credit history and availability, 265; on SSIs, 267–72 Information technology: factors needed to diffuse, 207; role in future infrastructure development, 202; WTO Information Technology Agreement (1996), 82–83 Information technology (IT) industry: advantages and disadvantages for Indian, 194–200; dominance in industrial sector, 135–36; exports of, 169–70; factors influencing future growth, 184–88; hardware development and manufacture, 170–71, 198; innovative applications, 188–89; revenues and market share (1990s), 194; strategies for future, 188–92 Information technology: Information Technology Action Plan (IT Action Plan), 180–84 Infrastructure: central government role in development of, 120–21; development to support IT industries, 186–87; effect on IT industry of limited, 202; IT Action Plan concerns related to, 181–82; post-independence Indian, 15, 20; potential role of information technology for, 202; as priority for reform (2000), 29–38; quality as determinant of growth, 110–12; spending for schoolrelated, 305–6

376

Subject Index

International Monetary Fund (IMF): longterm borrowing from (1991), 21; medium-term arrangement with, 79– 80; short-term borrowing from, 2 Investment: of foreign firms, 136–38; growth expectations influence school building, 331–49; in human capital, 329–31; infrastructure required to attract, 114; in IT (1990s), 187; post-1991 boom, 132– 33; reallocation in postreform period, 102–3; in science and R&D before 1991, 143–46; in SSIs, 215–16; statelevel public and private activity (1995– 96), 104–6 IT. See Information technology (IT) industry Labor markets: effect of regulation of, 39– 41, 214; formal and informal, 40; needed reform of, 38; needed reform of labor laws, 118. See also Employment; Workforce Legal system: laws related to IT industry, 197; needed cyber law development, 183; needed reform of labor laws, 118 Literacy rates: increases (1950–2000), 303; as proxy for quality of human resources, 108–10; rise in (1950s–90s), 17 Macroeconomic policy, pre-1980s, 48 Manufacturing sector: employment in, 257– 58; employment in small-scale firms, 223–37; labor-intensive part of, 298; relation of employment levels to trade, 244–46; reservation policy for smallscale firms, 219–22; value added for SSIs in, 237–41 Market preserving federalism, 59, 79 Monetary policy (1980s–90s), 1–2 Multi-Fibre Arrangement (MFA), 256, 262 Nationalization of banking system (1960s), 14 Population growth (1950s–90s), 17–18 Ports: central government responsibility for, 113; required investment to modernize, 37–38 Poverty: central government programs related to, 119; post-war levels of, viii, 18, 49; trends in state- and centralgovernment levels, 98–101, 123–24

Power. See Electrical power industry Private sector: controls over, 19–20; labor force in, 40 Privatization: plans for SEBs, 115; recommended for public-sector enterprises, 116, 121; of state-owned enterprises, 42–43, 133–34 Public opinion: about reservation policy, 299–300; economic growth as topic for, 141; need for reform lobby, 142 Public sector: growth of (1970s–90s), 19; labor force in, 40; post-1991 reduction in, 133. See also Government, central; Government, state-level Railroads: central government responsibility for, 113; priority for improvement of, 37 Research and development (R&D): post1991 spending for, 148, 156–58; pre1991 indigenization policy, 144–46; Telecommunications & Computer Networks Group, 190–91 Reservation policy: arguments for abolition of, 261–63; effects of, 38, 241–44; impact on exports, 244–57; potential effects of reform of, 298–302; to protect SSIs, 215; rationale for and features of, 219–22, 243–44; for small-scale manufacturing, 20, 44. See also Industry, small-scale (SSI) School enrollment: increases in (1951–92), 305–7; in private schools, 304, 314–18, 324, 326; relation to school quality, 307–10 Schooling: returns to, 320, 329–30; rise of home tutoring, 304, 314–16, 320 School quality: demand for schooling related to, 307–11; differential effect of, 311, 314; of elementary schools, 303–4; in government and private schools, 324–25; relation of schooling outcomes to, 307–8 Schools: effect of growth expectations on building of, 331–49; effect of quality on schooling attainment, 307–11; factors influencing quality of, 311–14; growth in number of and infrastructure for, 305–6; regulation of private, 323–24 Schools, private: aided and unaided, 323–24; enrollment in, 304, 314–18, 324, 326; as factor in school inequality, 304; factors

Subject Index influencing growth of, 324–25; growth of, 314–16; recognized and unrecognized, 324; student-teacher ratios in, 324 Schools, public: quality of governmentfinanced, 310–11 SEBs. See State Electricity Boards (SEBs) Securities and Exchange Board of India (SEBI), 26 Services sector: needed tax reform for, 61 Silicon Valley: contrasted with Bangalore, 198–203, 205; entrepreneurship in, 199 Social sector: infrastructure provided at state level, 114; social policy issues, 80, 198; state-level spending and subsidies for, 58, 78, 115 Software industry: in Bangalore, 199–200; evolution (1984–98), 171–79; growth rates and exports, 158, 170; National Information Technology and Software Development Task Force, 170, 180–81; revenues (1998–99), 169–70; skill levels of workforce, 209 Software Policy (1986), 181 SSIs. See Industry, small-scale (SSI) State Electricity Boards (SEBs): arrears of, 57; deregulation in Uttar Pradesh, 131; losses of, 115; provide subsidies, 34; as state monopoly, 114 State-owned enterprises: employment in (1980s), 19; under Second Five Year Plan, 12, 14; soft-budget constraint for, 41–42 Subsidies: of central government, 20; need to reduce, 75; provided by SEBs, 34; for for SSIs, 218; state-level cross subsidization, 119; of state-level governments, 54, 58, 115 Taxation: advances in reform of, 74; integrated VAT taxation, 74–75; of intermediation, 77; needed reforms, 60–62; proposed VSAT system, 61; state-level, 55; transfer system from central to state, 55–56 Tax policy: needed state-level modernization of, 116; reforms, 81–89

377

Technology: Indian Institutes of Technology, 191, 194; post-1991 changes in, 146–58; pre-1991 industrial policy, 143– 46; provided by foreign firms to India, 137. See also Information technology (IT) industry Telecommunications: central government infrastructure provision for, 113; effect of over-regulation, 36; New Telecom Policy (1999), 36; recommendations for liberalization of, 182; reforms during 1990s, 35–37 Trade policy: after 1991 financial crisis, 23– 24; countries liberalizing post-war, viii; developing countries’ commodity group ranking, 287–90; developing countries’ export structure, 273–86; export- and import-related, 19–20, 23–24, 29–30, 134; IT Action Plan recommendations for, 182–83; post-war importsubstitution and licensing, 12–13, 19– 20; protection of IT industry, 171; tariff reduction as part of reform, 60–61. See also Reservation policy Transportation. See Ports; Railroads Venture capital industry: in Bangalore, 200; current status and regulation of, 183– 84; to encourage innovation, 208; funding of, 184; in India, 265; IT Action Plan recommendations for, 183; in United States and United Kingdom, 208, 265–66 Workforce: bodyshopping, 172, 174, 176, 178, 204; migration among states, 99– 100; post-1991 growth in productivity, 132–33; quality and quantity in Indian software industry, 176–79, 194; skill levels in software industry, 209; in SSIs, 38. See also Education; Employment; Human capital; Labor markets World Bank, 79–80 World Trade Organization (WTO): Information Technology Agreement (1996), 82–83