Globalization, Labor Markets and Inequality in India

India started on a program of reforms, both in its external and internal aspects, sometime in the mid-eighties and going on into the nineties. While the increased exposure to world markets ('globalization') and relaxation of domestic controls has undoubtedly given a spurt to the GDP growth rate, its impact on poverty, inequality and employment have been controversial.

This book examines in detail these aspects of post-reform India and discerns the changes and trends which these new developments have created. Providing an original analysis of unit-level data available from the quinquennial National Sample Surveys, the Annual Surveys of Industries and other basic data sources, the authors analyze and compare the results with other pieces of work in the literature. As well as describing the overall situation for India, the book highlights regional differences, and looks at the major industrial sectors such as agriculture, manufacturing and tertiary/services. The important topic of labor market institutions – both for the formal or organized and the unorganized sectors – is considered and the possible adverse effect on employment growth of the regulatory labor framework is examined carefully. Since any reform of this framework must go hand in hand with better state intervention in the informal sector to have any chance of acceptance politically, some of the major initiatives in this area are critically explored.

The book is based on the results of a collaborative research project carried out at the Institute for Human Development (IHD), New Delhi, which is an autonomous institution specializing in labor markets, employment and human development issues. The Munk Centre for International Studies (MCIS) of the University of Toronto provided administrative support for the project funded by the International Development Research Centre (IDRC), Ottowa.

Overall, this book will be of great interest to development economists, labor economists and specialists in South Asian Studies.

Dipak Mazumdar is Senior Research Associate, Munk Centre for International Sudies at the University of Toronto, Canada and Visiting Professor, Institute for Human Development, New Delhi. He is the author of numerous publications on development economics. His co-authored book, with Ata Mazaheri, The African Manufacturing Firm was also published by Routledge, in 2003. Sandip Sarkar is currently working as a Fellow with the Institute for Human Development (IHD), New Delhi, India. His main areas of research interest are industry, poverty, labor and employment on which he has experience of over two decades. His recent major research project was on the impact of globalization on the labor market in India which is sponsored by the International Development Research Centre (IDRC), Canada. He has been extensively involved in several large research projects funded by reputed national and international agencies.

Routledge studies in the growth economies of Asia

1 The Changing Capital Markets of East Asia
Edited by Ky Cao

2 Financial Reform in China
Edited by On Kit Tam

3 Women and Industrialization in Asia
Edited by Susan Horton

4 Japan's Trade Policy
Action or reaction?
Yumiko Mikanagi

5 The Japanese Election System
Three analaytical perspectives
Junichiro Wada

6 The Economics of the Latecomers
Catching-up, technology transfer and institutions in Germany, Japan and South Korea
Jang-Sup Shin

7 Industrialization in Malaysia
Import substitution and infant industry performance
Rokiah Alavi

8 Economic Development in Twentieth Century East Asia
The international context
Edited by Aiko Ikeo

9 The Politics of Economic Development in Indonesia
Contending perspectives
Edited by Ian Chalmers and Vedi R. Hadiz

10 Studies in the Economic History of the Pacific Rim
Edited by Sally M. Miller, A.J.H. Latham and Dennis O. Flynn

11 Workers and the State in New Order Indonesia
Vedi R. Hadiz

12 The Japanese Foreign Exchange Market
Beate Reszat

13 Exchange Rate Policies in Emerging Asian Countries
Edited by Stefan Collignon, Jean Pisani-Ferry and Yung Chul Park

14 Chinese Firms and Technology in the Reform Era
Yizheng Shi

15 Japanese Views on Economic Development
Diverse paths to the market
Kenichi Ohno and Izumi Ohno

16 Technological Capabilities and Export Success in Asia
Edited by Dieter Ernst, Tom Ganiatsos and Lynn Mytelka

17 Trade and Investment in China
The European experience
Edited by Roger Strange, Jim Slater and Limin Wang

18 Technology and Innovation in Japan
Policy and management for the 21st century
Edited by Martin Hemmert and Christian Oberländer

19 Trade Policy Issues in Asian Development
Prema-chandra Athukorala

20 Economic Integration in the Asia Pacific Region
Ippei Yamazawa

21 Japan's War Economy
Edited by Erich Pauer

22 Industrial Technology Development in Malaysia
Industry and firm studies
Edited by Jomo K.S., Greg Felker and Rajah Rasiah

23 Technology, Competitiveness and the State
Malaysia's industrial technology policies
Edited by Jomo K.S. and Greg Felker

24 Corporatism and Korean Capitalism
Edited by Dennis L. McNamara

25 Japanese Science
Samuel Coleman

26 Capital and Labour in Japan
The functions of two factor markets
Toshiaki Tachibanaki and Atsuhiro Taki

27 Asia Pacific Dynamism 1550–2000
Edited by A.J.H. Latham and Heita Kawakatsu

28 The Political Economy of Development and Environment in Korea
Jae-Yong Chung and Richard J Kirkby

29 Japanese Economics and Economists since 1945
Edited by Aiko Ikeo

30 China's Entry into the World Trade Organisation
Edited by Peter Drysdale and Ligang Song

31 Hong Kong as an International Financial Centre
Emergence and development 1945–1965
Catherine R. Schenk

32 Impediments to Trade in Services
Measurement and policy implication
Edited by Christoper Findlay and Tony Warren

33 The Japanese Industrial Economy
Late development and cultural causation
Ian Inkster

34 China and the Long March to Global Trade
The accession of China to the World Trade Organization
Edited by Alan S. Alexandroff, Sylvia Ostry and Rafael Gomez

35 Capitalist Development and Economism in East Asia
The rise of Hong Kong, Singapore, Taiwan, and South Korea
Kui-Wai Li

36 Women and Work in Globalizing Asia
Edited by Dong-Sook S. Gills and Nicola Piper

37 Financial Markets and Policies in East Asia
Gordon de Brouwer

38 Developmentalism and Dependency in Southeast Asia
The case of the automotive industry
Jason P. Abbott

39 Law and Labour Market Regulation in East Asia
Edited by Sean Cooney, Tim Lindsey, Richard Mitchell and Ying Zhu

40 The Economy of the Philippines
Elites, inequalities and economic restructuring
Peter Krinks

41 China's Third Economic Transformation
The rise of the private economy
Edited by Ross Garnaut and Ligang Song

42 The Vietnamese Economy
Awakening the dormant dragon
Edited by Binh Tran-Nam and Chi Do Pham

43 Restructuring Korea Inc.
Jang-Sup Shin and Ha-Joon Chang

44 Development and Structural Change in the Asia-Pacific
Globalising miracles or end of a model?
Edited by Martin Andersson and Christer Gunnarsson

45 State Collaboration and Development Strategies in China
The case of the China-Singapore
Suzhou Industrial Park (1992–2002)
Alexius Pereira

46 Capital and Knowledge in Asia
Changing power relations
Edited by Heidi Dahles and Otto van den Muijzenberg

47 Southeast Asian Paper Tigers?
From miracle to debacle and beyond
Edited by Jomo K.S.

48 Manufacturing Competitiveness in Asia
How internationally competitive national firms and industries developed in East Asia
Edited by Jomo K.S.

49 The Korean Economy at the Crossroads
Edited by MoonJoong Tcha and Chung-Sok Suh

50 Ethnic Business
Chinese capitalism in Southeast Asia
Edited by Jomo K.S. and Brian C. Folk

51 Exchange Rate Regimes in East Asia
Edited by Gordon de Brouwer and Masahiro Kawai

52 Financial Governance in East Asia
Policy dialogue, surveillance and cooperation
Edited by Gordon de Brouwer and Yunjong Wang

53 Designing Financial Systems in East Asia and Japan
Edited by Joseph P.H. Fan, Masaharu Hanazaki and Juro Teranishi

54 State Competence and Economic Growth in Japan
Yoshiro Miwa

55 Understanding Japanese Saving
Does population aging matter?
Robert Dekle

56 The Rise and Fall of the East Asian Growth System, 1951–2000
International competitiveness and rapid economic growth
Xiaoming Huang

57 Service Industries and Asia-Pacific Cities
New development trajectories
Edited by P.W. Daniels, K.C. Ho and T.A. Hutton

58 Unemployment in Asia
Edited by John Benson and Ying Zhu

59 Risk Management and Innovation in Japan, Britain and the USA
Edited by Ruth Taplin

60 Japan's Development Aid to China
The long-running foreign policy of engagement
Tsukasa Takamine

61 Chinese Capitalism and the Modernist Vision
Satyananda J. Gabriel

62 Japanese Telecommunications
Edited by Ruth Taplin and Masako Wakui

63 East Asia, Globalization and the New Economy
F. Gerard Adams

64 China as a World Factory
Edited by Kevin Honglin Zhang

65 China's State Owned Enterprise Reforms
An industrial and CEO approach
Juan Antonio Fernandez and Leila Fernandez-Stembridge

66 China and India
A tale of two economies
Dilip K. Das

67 Innovation and Business Partnering in Japan, Europe and the United States
Edited by Ruth Taplin

68 Asian Informal Workers
Global risks local protection
Santosh Mehrotra and Mario Biggeri

69 The Rise of the Corporate Economy in Southeast Asia
Rajeswary Ampalavanar Brown

70 The Singapore Economy
An econometric perspective
Tilak Abeyshinge and Keen Meng Choy

71 A Basket Currency for Asia
Edited by Takatoshi Ito

72 Private Enterprises and China's Economic Development
Edited by Shuanglin Lin and Xiaodong Zhu

73 The Korean Developmental State
From dirigisme to neo-liberalism
Iain Pirie

74 Accelerating Japan's Economic Growth
Resolving Japan's growth controversy
Edited by F. Gerard Adams, Lawrence R. Klein, Yuzo Kumasaka and Akihiko Shinozaki

75 China's Emergent Political Economy
Capitalism in the dragon's lair
Edited by Christopher A. McNally

76 The Political Economy of the SARS Epidemic
The impact on human resources in East Asia
Grace O.M. Lee and Malcolm Warner

77 India's Emerging Financial Market
A flow of funds model
Tomoe Moore

78 Outsourcing and Human Resource Management
An international survey
Edited by Ruth Taplin

79 Globalization, Labor Markets and Inequality in India
Dipak Mazumdar and Sandip Sarkar

This page intentionally left blank

Globalization, Labor
Markets and Inequality
in India

Dipak Mazumdar and Sandip Sarkar

First published 2008
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon Ox14 4RN

Simultaneously published in the USA and Canada
by Routledge
270 Madison Ave, New York, NY 10016

Routledge is an imprint of the Taylor & Francis Group, an informa business

Published in association with the International Development Research Centre
PO Box 8500, Ottawa, ON K1G 3H9, Canada
www.idrc.ca
ISBN: 978-1-55250-373-7 (ebk)

© 2008 Dipak Mazumdar and Sandip Sarkar

Typeset in Times by Wearset Ltd, Boldon, Tyne and Wear
Printed and bound in Great Britain by TJ International, Padstow, Cornwall

All rights reserved. No part of this book may be reprinted or reproduced or
utilized in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.

British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication Data
A catalog record for this book has been requested

ISBN10: 0-415-43611-7 (hbk)
ISBN13: 978-0-415-43611-3 (hbk)

Contents

Figures

Tables

Maps

This page intentionally left blank

1 Introduction
An overview of globalization, reforms and macro-economic developments in India

The process of economic reform and globalization in India

India embarked on a policy of liberalization and globalization in the latter part of the last century. There has been some discussion in the literature as to when India took steps to move away from the regime of comprehensive state control of the economy and dismantle the restrictive structure. A distinction has been made in this connection between 'reforms' and 'globalization'. Strictly speaking the former is supposed to emphasize the process of easing control of the domestic economy, while the latter refers to the attempts at liberalization on the external account. It is useful to keep the two sets of policy distinct, and will be referred to below.

In practice the former entails the latter. As Nayar (2006a, p. 10) observes:

Economic liberalization within a country creates pressures to integrate the national economy with the world economy . . . Say, for example, a country commences economic reform and removes restrictions on production by the private sector to accelerate growth. Eventually the state would have to allow imports of capital goods and intermediate goods to increase production–and that means integration into the world economy. And if it allows imports of these goods, then it must also promote exports in order to pay for them–further integration into the world economy.

In fact the process is more extensive than suggested in the above quotation. External liberalization also involved removing a good deal of restrictions on the import of consumer goods, not just capital and intermediate goods to aid production. The motivation for this was to promote competition in the domestic economy, and bring the efficiency levels in the Indian economy nearer the levels of the world economy.

It is clear that reforms of the domestic economy started earlier in India. Rodrick and Subramanain (2004) date the beginning of this process to the return of Indira Gandhi to power at the beginning of the 1980s. They ascribe the new direction to an 'attitudinal' shift in the perception of the leaders after the Congress Party had been 'chastened' by its electoral defeat in the earlier election of the late seventies. They also find a 'structural break' in several key indicators including GDP growth in this period. When the party was returned to power in January 1980 it became more inclined to support growth with the help of a more dynamic private sector. Nayar (2006b) maintains that the reform process started earlier in 1975–1976 during the regime of Indira Gandhi herself. The leadership was jolted partly by the turmoil created by the excesses of the 'dirigist' policy followed in the years 1969–1973 (including large-scale nationalization of banking and industry), and partly by external shocks (including war, droughts and oil price inflation). Besides adopting deflationary policy to stabilize the economy, the Gandhi administration undertook deregulation and export-promotion measures on top of the earlier devaluation (Joshi and Little 2000, p. 56).

While the period of 'creeping liberalization' might have been a prolonged one, it was not till the economic crisis of 1991 that there was an open endorsement of 'paradigm shift' embracing a policy of integration with the world economy and recognition of the need to follow the path of the South-East Asian growth strategy. It involved a sharp devaluation of the rupee; removal of quantitative restrictions on imports; reduction of import tariffs; and a unification of the exchange rate as the rupee was made convertible for current-account transactions. On the domestic front of the reform process the system of industrial licensing was removed and the list of items reserved for the small-scale producers was shortened considerably. The program also saw fiscal reforms though the maintenance of important subsidies, particularly on the agricultural front, continued to plague the budget (Ahluwalia 2002; Joshi and Little 2000).

The removal of quantitative restrictions on imports (QRs), an important feature of the controlled economy, came gradually in the decade of the nineties. QRs on intermediate and capital goods were removed in 1991, but they remained significant on a range of consumer goods. Over the next ten years, a series of international negotiations, starting with the 'Uruguay Rounds' of the WTO, saw a gradual whittling down of these barriers. Tariffs do, however,

remain as a deterrent to imports on a variety of goods. Table 1.1 is taken from a World Bank Report (Dahlman and Utz 2005) give the extent of the tariff barrier relative to some comparators in the early years of the century.

India's tariff rates remain high by the standards of other developing countries (Table 1.1). But a fair amount of integration with the world economy has been achieved. The following paragraphs briefly discuss the extent of the integration both in terms of the current and capital account of the balance of payments.

External trade

India's merchandise exports had a steep decline during the autarkic regime going down from 2.17 percent of world exports in 1949 to 0.44 in 1980. It hovered around 0.50 percent throughout the decade, and started going up only after 1991. Liberalization of trade has certainly had the impact of starting an upward trend and the share had reached a high of 0.8 in 2004. The share still remains quite low relative to comparator countries in Asia. China increased its share from under 2 percent in 1990 to close to 6 percent in 2003. Even a much smaller country like Korea had a share of 2.8 percent at this date (Table 1.2).

Manufactured exports have been a substantial part of the Indian export growth–reaching 74 percent in 2004 (government of India, Economic Survey–2005–2006). India seems to have performed relatively better in service exports. The gap between India and comparator countries in service exports, particularly vis-à-vis China, is not as large. India's progress in exports in computer and communications services has been much more than China's–which has done better in travel and related services (Figure 1.1).

Figure 1.1 Merchandise and service exports, India and comparators, 2002 (source: World Bank staff analysis using World Bank internal database).

Import volume has generally kept slightly ahead of export volume (Table 1.3). India has been helped in its current account by the terms of trade tilting in its favor in a majority of the years (though in very recent years there is a threat of significant deterioration of the TOT). But in any event the balance of payments position has been helped, increasingly so in recent years, by substantial inflow of foreign funds. This is due to another aspect of India's globalization–the substantial emigration of its nationals and the inflow of remittance from the overseas residents of Indian origin.

Foreign-capital inflow other than remittance has in fact not been significant. In fact India has been a significant laggard in attracting foreign direct investment. Even though the actual value of FDI in India has increased several times from its level before liberalization, it is quite small compared with global trends.

At its height in recent years it has perhaps been no more than 1 percent of GDP–compare to 4.7 percent in China in 2001 and 4.4 percent in Brazil. In value terms India received $4.26 billion in FDI in 2003, compared with $53.5 billion for China (Dahlman and Utz 2005, pp. 30–31). Dahlman, however, reports that according to the Foreign Direct Investment Confidence Index (by A.T. Kearney) India's attractiveness to foreign investors is rapidly rising–although it has been well below China's for sometime.

It is clear from this capsule account that globalization in the sense of integration with the world economy has been significant both on the trade and the capital account. It is equally clear that it has been accompanied by a spurt in the growth rate of GDP. Further, the efficiency of the economy has increased in the aggregate. The slow growth of India's constant price investment ratio (increasing from around 22 percent in the 1970s and the 1980s to no more than 24 percent at the end of the 1990s), while GDP growth was accelerating suggests that the marginal capital–output ratio must have been rising significantly. According to one author this ratio was a meager 0.12 in India's worst decade 1965–1974, but it had doubled to 0.246 in the decade of 1991–2000 (Berry 2006, p. 3).

In spite of this positive effect of globalization, doubts are widespread about the success of the economy in achieving greater equity and acceptable levels of poverty reduction. In the next section we shall review, equally briefly, the major thrust of the literature speculating on the possible links between growth and equity. While this review of the theoretical literature is not to provide exact guidelines to our empirical investigation to follow, some of the ideas explored might help to illuminate or emphasize specific results in the chapters to follow.

Labor markets, poverty and inequality in the growth process

The theoretical discussion on the changes in the incidence of poverty and inequality in the growth process of agrarian economies from low levels of income has been a major topic in development economics. The impact of growth is delineated through the labor market, and any predictions about the impact on poverty and inequality must be based on some implicit or explicit view of the structure of labor markets and their functioning.

Homogeneous labor

The classical view on economic development and its impact on inequality is found in the Lewis model and its elaboration in the early work of Kuznets and of Ranis-Fei. All these models consider the growth process to be driven by a shift of labor from the 'traditional to the developing sector', (variously identified as 'rural and urban', 'subsistence and capitalist'; 'agricultural and industrial'). Ranis and Fei in their work on Taiwan formalized the three different elements in this story which together determine the dynamics of inequality over time. First, there is the 'reallocation effect' of labor moving out of agriculture to the secondary and tertiary sectors. This shift will tend to increase inequality if the distribution is less equal in the latter. Second, we have the 'functional distribution effect' in the 'commercialized' sector (the income accruing to the agricultural sector is best treated as mixed income comprising returns to both labor and capital in family farms). An increase in the share of wages will typically increase equality. Before the 'turning point' in the labor market in the Lewis sense, the unlimited supply of labor at constant wages should induce technological change in labor-using direction and should prevent any increase in technology in a capital-deepening way that shifts the functional distribution towards capital. But after the 'commercialization point' in the labor-surplus economy the trend in the functional distribution of income depends on the nature of technical progress. Thus we have the third element in the dynamics: the 'innovation-intensity effect' that might be sufficiently biased against labor to decrease the share of labor over time and tend to increase inequality. The course of inequality through time depends on the relative strength of all these three effects. In fact the 'innovation-intensity effect' depressing the share of wages in the commercialized sector might not be delayed till after the 'turning point' as suggested by Ranis and Fei, but might already be working in a labor-saving way if entrepreneurs in the commercialized sector choose to adopt imported capital-using technology for a variety of reasons.

The Kuznets hypothesis of the U-shaped pattern of inequality dynamics follows from a theory embodying the above three elements. In the early adages of development the reallocation effect is strong and the share of labor also might, fall inducing a rising inequality. The trend is reversed when the reallocation effect weakens and the rise in wages overwhelms any effect coming from continuing bias towards capital-using technological progress.

Since the reallocation effect shifts labor from the low-income sector, the prediction is that in the early stages of the Kuznets process poverty should decline but inequality has an upward trend. We draw particular attention to this prediction because this is the trend we see in a number of developing countries in recent development history–including China and India.

Labor with different skill-levels

The basic model: skilled and unskilled labor

The economic literature in the last decade has paid a great deal of attention to the reason for rising wage inequality in a number of developed countries (including the USA). A feature of the increase in inequality has been the sharp rise of wages of 'skilled' workers relative to the 'unskilled'. The reversal of the trend towards narrowing wage differentials, which had been going on for much of the twentieth century, coincided with the increase in the share of manufactured exports going from the developing countries to USA and Japan in particular. It was natural for a number of US economists to jump to the conclusion that the critical role in this phenomenon was being played by the increase in imports of labor-intensive goods. A well-know theorem within the framework of the Heckscher–Ohlin model states that when trade is opened up, since each country tends to export the commodity using the more abundant factor, the relative price of the more abundant factor in each country will tend to increase. In developed countries, which shift to more skill-intensive products, the relative price of skilled labor would increase. The corollary of this proposition is that in developing countries the opposite will happen–the wages of unskilled labor would increase relative to those of the skilled, and wage inequality could be expected to fall.

The last corollary is less persuasive than would appear at first sight. The assumption of a homogeneous labor market, which transmits the impulses originating in the tradable manufacturing sector throughout the economy-wide labor market, even if partially acceptable in a developed economy, is wide of the mark in developing countries. The manufactured exports from developing countries originate in the formal sector. The extent to which this sub-sector is linked to the informal manufacturing sector varies from country to country, and in any case it is a small part of the labor market properly considered, which is dominated by the self-employed. We will come back to this point later on. But for the moment let us confine ourselves to some additions to the model as it is applied to the wage-labor market.

Extensions of the model: technological progress

THE DEVELOPED-COUNTRY SCENARIO

The static Heckscher–Ohlin theory is formulated in static terms with trade impacting the economy with an unchanged production function. But the recent decades have seen not only a great deal of technological progress, but of such progress being biased towards skilled labor.

Apart from considerations based on general observations, economists have noted that all explanations of rising wage inequality in the 'North' 'leave unexplained the rising skill intensity in non-traded goods as traded goods sector. In spite of having to pay more for skilled workers, employers in almost all sectors (traded as well as non-traded goods) chose to hire more skilled workers' (Gottschalk and Smeeding 1997, p. 649). 'Only technological change is consistent with rising skill intensity in the face of rising skill prices' (ibid., p. 650).

Daron Acemoglu (2002) has recently surveyed the vast literature on the technical progress and rising wage inequality in the North. The broad facts are clear. While the past 60 years have seen a vast increase in the supply of more educated and skilled workers, the returns to education in the U.S. among other countries fell during the seventies, but have begun a steep rise during the 1980s. This stylized fact is consistent with either a slowing down of the rate of increase of the supply of more skilled workers with a constant pace of technological progress, or alternatively with an accelerated pace of skill-intensive technical progress since the 1980s. It might indeed be a combination of supply and demand side factors.

Acemoglu hypothesizes that technical change is to a very large extent induced by factor market conditions. In the nineteenth century when the North had a plentiful supply of unskilled labor, technical change (e.g., during the industrial revolution) was directed to saving the use of skilled labor. With the rapid growth of education, the supply of skilled labor took a jump in its rate of growth, and the inducement for technical progress shifted towards saving the use of unskilled labor. The skill premium could be held in check as long as the pace of technical progress did not exceed that of the growth of skilled labor. But it is to be expected that with the vast growth of educated labor, its rate of growth in the North had to slow down. In fact, it is possible that the large expansion of international trade might have been a factor in increasing the pace of skill-biased technical progress. As the basic trade model noted, expansion of exports of skill-intensive products from the North tend to increase the relative price of such goods, leading to a search for technology biased towards increasing the productivity in such industries.

THE DEVELOPING-COUNTRY SCENARIO

What have these developments in the North got to do with wage-inequality trends in the South–the developing countries? The availability of a large pool of unskilled labor could be expected to promote technological progress that will economize on the use of skilled labor, as in the nineteenth century North. But there are several important factors which might suggest why this has not happened as a widespread phenomenon.

First, and foremost, it is clear from recent economic history that R&D expenditure is heavily concentrated in the North, and it seems to have the highest payoff in the advanced economies. Thus a more plausible scenario is that, rather than each country and region developing its own technology, new technology is developed in the leading economies of the North and spreads across countries.

Second, we need to emphasize that the techniques of production are not determined only by the relative supplies of different types of labor, but also by the quality of product which the market accepts. It has been noted in the literature that techniques which make use of less capital and less skilled labor often produce a final product which has attributes catering to the demand of low income consumers (Little et al. 1987, chapter 13). When a developing country enters export markets in a big way, the final consumer is located in the affluent North, and the technique of production has to be geared to producing items with superior attributes in the quality spectrum. Often these superior quality ranges would need more mechanized techniques with more skill-intensive labor. The point is reinforced by the need to achieve timeliness and homogeneous quality in the batches exported.

Third, it is useful to think in terms of different stages of production for the market. The stage of physical production might indeed be allocated to dispersed units using techniques which make use of labor of the type that is in plentiful supply (of low or traditional skills), but these have to be integrated with organizational, financial and marketing units to be able to supply the export market effectively. The garment industry, which has played such an important role in the export expansion from the South in recent decades, is a case in point. The tertiary activities needed in this export activity often use labor of high or non-traditional skills, which might be in short supply.

Fourth, the last point brings into focus an important part of the story of export expansion from the South. This is the role of outsourcing. Feenstra and Hanson (1999) among others have emphasized that the change in the degree of inequality or the relative wage of unskilled to skilled labor should be analyzed in terms of a foreign outsourcing model, which emphasizes trade in intermediate products, and not exclusively in terms of trade in final products which the H–O model stresses. The production of a final manufactured good can be broken down into several stages which can be arranged in ascending order of the skill intensity of the activity. Outsourcing from the developed country means that the some of the lower ranges of skill intensity in this chain are shifted out to developing countries. But these activities which are shifted, although they are of relatively low skill intensity in the North, are relatively in the higher rung of skill intensity in the South. The net effect of the outsourcing is then to reduce the relative demand for less skilled workers in the North, but to increase the demand for more skilled workers in the South. Thus while we can expect the skill premium in the North to increase, wage inequality in the South would also tend to increase, contrary to the predictions of the simple H–O model. This kind of outsourcing effect will, of course, be particularly important when the increase in manufactured exports from the South is being driven by direct investment by Northern businesses.

Extensions of the model: several grades of skill

Another dose of realism could be added to the basic H–O model by extending the model to accommodate more than just two types of labor–skilled and unskilled, with the former being complementary to capital in twentieth century technology. Adrian Wood (1994) makes a distinction between at least three types of labor: labor without any modern industrial skill–'raw labor' which is found in agriculture or low services, but not adapted to work in modern factories or businesses; labor with some basic skills for factory work; and labor with higher skills to perform more complicated tasks in the modern sector. Wood believes education is the basis of this classification–he calls the first category NO-ED, the second BAS-ED (those with at least primary or low-secondary education) and SKILD (with higher levels of education) the third category of labor. But the distinction need not be defined by levels by schooling alone. It is known that a significant wage gap exists in favor of labor of low skill in the 'modern' sector even in the face of plentiful supply of labor in the traditional sector (see the section below on 'segmented labor markets'). Wage inequality within the large-scale industrial sector might be squeezed but the over-all wage inequality increases because of the wage of BAS-ED labor increasing relative to that of NO-ED labor.

Even this limited prediction might be thwarted, if technological progress is skill-biased as discussed above, or alternatively, if we introduce factors of production other than labor and capital. Another strand in Wood's set of hypotheses is that factors of production in addition to labor and capital are critical in the comparative advantage of an economy–most notably land and the availability of natural resources. Countries with relatively large endowments of natural resources will tend to export more land-intensive products, while those with a shortage of such resources will tilt towards more manufactured activities. But the land-intensive primary products lead into processing industries which use less skilled labor than other industrial products. Thus expansion of industrial exports in land-abundant countries ceteris paribus would tend to dampen wage inequality, and to increase it in resource-poor economies. This is, of course, only the demand side of the story. The final outcome depends on the relative supply of educated or skilled labor over time–which is to large extent the result of autonomous state policies.

Extensions of the model: shifting boundary of the non-traded sector

In the original discussions of the H–O model there was an implicit assumption that the boundary between the traded and non-traded sectors coincided with that between the manufacturing and the tertiary sector. (For some theorists the implicit assumption was extended further to the distinction coinciding with that between the formal (modern) and informal (traditional) sectors.) Recent developments in the world economy have made this distinction quite unrealistic. For one thing, the services sector has emerged as a major exporter. Second, some products of the non-traded service sector are in close relationship to the traded sector.

Liberalization of the external sector, including devaluation which might accompany it, increases the relative price of traded goods and pushes more resources into the traded sector. But two other effects need to be considered. The first is that some non-traded goods might be complementary to the export sector. Such for instance might be infrastructure, including transport and some supporting services. An increase in the developing countries' exports, even if they are more low-skill intensive than the exports from developed countries, induces complementary expansion of infrastructure which is more skill intensive. Thus the net impact on the demand for labor of different skill levels is uncertain. Second, we should allow for substitution on the consumption side. Consider a developing country with abundant supply of unskilled labor, in which low skill services are close substitutes for the more skill-intensive traded goods (e.g., washing machines). Liberalization reduces the relative price of the latter, leading to a lower demand for low-grade services, and hence a lower demand for low-skill labor, which might offset the increase in demand for such labor induced by the expansion of labor-intensive exports.

The upshot of this discussion is that when the basic trade model is extended by successive doses of realism no definitive prediction about the movements of relative prices of skill, and hence the direction of change in the degree of wage inequality, is possible. This is not to say that empirical analysis would not yield patterns which are uniform over sets of countries or regions. Some work which has been done already has contained the intriguing suggestion that greater openness has decreased wage inequality in East Asian experience in the seventies in the expected way of the simple model, but that in several Latin American countries the opposite has been the case in the eighties. Commenting on this possible generalization, Wood (1997, p. 47) offers a hypothesis apparently based on a suggestion by Jeffrey Sachs:

It might be the case that all manufactures were import substitutes in Latin America, but only skill-intensive manufactures were import substitutes in East Asia. In that case, non-traded sectors (of a given skill intensity) might be more skill-intensive than import-competing sectors in Latin America and less skill-intensive than import competing sectors in East Asia. Hence, if greater openness (through substitution in consumption) caused non non-traded as well as export sectors to expand (and import-competing sectors to contract), the net effect might be to increase the relative demand for skilled labor in Latin America, but to decrease it in East Asia.

Education policies and the supply of skilled labor

While the evolution of the demand for skilled labor is important, countries differ enormously in the way the formal educational system develops over time. Even if skill formation is heavily influenced by on-the-job training, basic formal education is a critical variable. The impact of education on wage inequality has two different effects. The growth of educated population has a 'compositional effect' which yields an inverted U-shaped pattern à La Kuznets. Until a certain proportion of the population belongs to the more educated (and higher wage) groups, an increase in the proportion of the latter will increase inequality, but after the critical point is passed inequality falls as a larger proportion already belongs to the high wage group. The rising inequality in the earlier part of this process will be moderated if the rate of return to education does not increase, and a fall in the returns of sufficient magnitude will in fact reduce inequality. The latter possibility in fact turns on the supply of educated labor running ahead of the inversing demand as the modern sector develops. It has been noted in the literature that the decrease in wage inequality in Korea and Taiwan during their process of export-led industrialization could be traced in large measures to the prior investment in secondary education (see, for example, Gindling and Sun on Taiwan, and Fields and Yoo on Korea). The experience in these East Asian economies contrasts strongly with the development in Thailand, where post-primary education was neglected till the nineties. Thus over the period 1976–1988, Thailand had a strong upward trend in the inequality index as the export-led boom of the latter eighties put strong pressure in the market for skilled labor (World Bank 1996).

Segmented labor markets

The discussion so far has concentrated on the wage-labor market, and for the most part on labor markets differentiated by levels of measurable skills (e.g., education). But in developing countries much of the labor force is self-employed. Even within the wage-labor market discussions in the mainstream literature are generally concentrated on the formal part of the market–which typically excludes the small and micro sectors, if only because the informal sector is poorly served by regular statistical surveys on wages. If labor markets were reasonably homogeneous, trends in the formal wage-labor market would indicate trends in other parts of the labor market as well. But typically labor markets in such economies are segmented. Labor with the same measurable human capital earns significantly different incomes in different segments of the market. The trends in earnings might also diverge as between the different segments of the market.1

A schematic picture of the labor market in a developing country looks like Table 1.4 (A and B refer to the formal and the informal sectors respectively). There are large gaps in the levels of earnings between the segments of the labor market shown in the table. These gaps persist even after we have controlled for measurable human capital differences between the workers found in the different

sectors. The extent of the earnings differences would of course vary from country to country, and one of the tasks of country studies would be to quantify the more important of the wage gaps.

The impact of liberalization or other aspects of globalization on the over-all distribution of income would depend on:

First, the observed levels of employment and earnings in each segment of the stylized classification given above are the product of the intersection of demand and supply curves of labor in that segment. Thus one should be aware that factors affecting the derived demand for labor, as well as the supply conditions of labor, would be instrumental in affecting the outcomes in the segment.

Second, it should be apparent that the movements of these variables over time would be influenced, not only by the way markets for labor of different skills behave over time, but also by the working of the markets of co-operant factors, particularly land and capital. Thus the earnings of the self-employed will be more equally distributed if they are able to accumulate capital more easily over time. A more restrictive capital market would on the other hand both depress their mean earnings relative to those in the formal sector, and also perhaps lead to a more unequal distribution in this sector. For those in the farming sector, the distribution of land and of the basic inputs like fertilizer and water are of crucial significance.

Accounting for the earnings difference between the formal and the informal sectors

The formal–informal sector divide in the labor market cuts through the entire range of non-agricultural industries–both in the tertiary and secondary sectors, and in some economies even in agriculture where there is a large concentration of large Farms or plantations. In view of the importance of this phenomenon it might be useful to review the various hypotheses accounting for the gap in the levels of earnings between the two. In general these hypotheses are not mutually exclusive. They might co-exist in different degrees in any particular economy.

The institutional hypothesis

An important strand in the literature has asserted that labor in the formal sector is 'protected' in the sense that its wage level cannot be undercut by competition from outside labor. This type of 'protection' might be supported by institutions like labor laws or trade unions working independently or working hand-in-hand with the state's labor-regulatory framework. In this case, in so far as wage levels are significantly above alternative earnings outside, entry into the sector is rationed. There is an elastic supply of job seekers but only a fraction can be admitted.

The wage–efficiency hypothesis

The literature has recognized for sometime that the wage–efficiency relationship sets a floor to the wage rate in the formal sector. The most straightforward version is the nutritional one. Efficiency increases with the level of wages because better-fed workers are able to work harder. Thus no employer with a stable body of workers will offer a wage below a level at which efficiency decreases proportionately more than wages. Such a floor to wages is undermined in the informal sector because of a number of factors which include: (i) casual labor without attachment to specific employers; and (ii) self-employment working from households in which earnings of different working members are pooled together. Further, if we do not interpret the wage–efficiency relationship strictly in nutritional terms it will vary with the type of work, quality of machinery used and of goods produced, and the organization of labor. In fact it may become a hazy notion depending very much on the perception of employers. Large formal-sector employers, wary of possible labor unrest or adverse laws protecting job security, might opt for a labor system in which a small elite body of workers produces at a high rate of efficiency in exchange for stable employment at high wages. In any event a significant difference in wage per man is established between the formal and the informal sector although the difference in wage cost per efficiency unit of labor might not be that large. The extent of the differential clearly depends on the quality of labor supplied in the market as well as the institutional setting of the formal labor market.

How does this set of factors affecting the wage differential with respect to the informal sector relate to the discussion above about the skill/education related differentials? The mechanism discussed here in fact establishes a higher wage in the formal sector for all levels of education/skill. This does not, however, mean that formal-sector employers would not use education as a screening or signaling device for the selection of their workforce. If this happens we might find that the educational distribution of the workforce in the formal sector is much more skewed to the higher groups than in the informal. In addition to the wage–efficiency mechanism it would be a supplementary influence enhancing the premium enjoyed by skilled workers in the formal sector.

Constraints in the supply of co-operant factors in the informal sector

As indicated the self-employed constitute a major component of the workforce in the informal sector. Their earnings are in the nature of mixed income, consisting of returns to labor as well as to the co-operant factors used, principally capital (perhaps more of working than fixed capital). It is well known that credit constraints are more severe for the players in the informal sector. Thus the earnings profile in this sector would be critically influenced by the supply function of capital in this sector. It would affect not only the level but the shape of the earnings distribution in the informal sector. The differential with respect to formal-sector earnings is likely to be different for different parts of the distribution–and would also vary from country to country depending on the severity of the credit constraint in this sector.

Regional differences in earnings and employment

In recent work on post-reform developments spatial differences have come to the forefront of discussions. In fact the basic problem of uneven economic growth spatially is in some sense the heart of the subject of development economics. Post-reform developments in the rapidly growing economies like China have drawn renewed attention to the problem. Globalization is heavily directed in the first place to limited areas where producer links to external markets can be most advantageously established and entrepreneurs can exploit important external economies of scale. In fact the concern about unequal development exists equally in a relatively closed economy where major innovations might favor some regions more than others in a cumulative way–as might have happened in the spread of the green revolution in Indian agriculture.

It would be wrong to assume that processes of economic growth necessarily worsen inter-regional inequality. Even if the growth process is concentrated in some regions or enclaves to begin with, rising costs and links to other internal markets can and do produce incentives for producers to diversify to other areas. The net outcome in the dynamic process depends on the relative strength of these centrifugal forces and the process of cumulative causation favoring increased localization.

Plan of the work

The brief review of some of the more important theoretical ideas in the last section provides a background to the empirical investigation of the impact of the reform process and globalization on labor markets, poverty and inequality in India. The major data sources of the data utilized are the 'Thick' rounds of the National Sample Survey (NSS) which are conducted every five years. The latest round available for analysis is the 55th round for 1999–2000. At the time of the completion of this work the subsequent 'thick' round for 2004–2005 is not available for analysis with the unit-level records. But we are able to indicate some broad trends form the published reports issued by the NSS on a limited set of tabulations. This is done in the Epilogue. For the most part the bulk of the work relates to the period ending at the close of the last century.

The book is in four parts. Part I discusses the broad trends for the economy as a whole. Part II focuses on differences between major states and 'broad regions' of the country. Part III carries the analysis to the major sectors–agriculture, manufacturing and the tertiary industries. Part IV discusses issues in labor institutions–both in the formal and the informal sectors.

The analysis for All-India begins with the changes in the incidence of poverty–contrasting in particular the trends in the pre-reform years between the two thick rounds of the NSS before the 55th, and the post-reform period between 1993/1994 and 1999–2000. We discuss, within the framework of a decomposition model, the relative importance of rural–urban shift of labor, growth of the two sectors of the economy and changes in inequality within each. The varying experiences of the urban areas of different population-size groups as well as of the major states in the process are also discussed.

Chapter 3 presents the basic trends in employment and earnings over the period covered, contrasting the post-reform years with the previous periods. It goes on to document the emerging trends in inequality, both for wage earners and all wage- and non-wage-earning households together. We also document the increase in 'rural–urban dualism' in the post-reform years, which had already been suggested by the poverty analysis of Chapter 2.

Part II turns to the analysis of regional differences. This topic is of vital interest in a large country like India. We can do this regional analysis in two ways. First, we can look at differences between major states. This is important because states are political units with a good deal of autonomy in the implementation of economic policies. They are, however, not homogeneous areas in terms of agroecological areas. The latter are of great importance in Indian conditions as they have significant impact on productivity and incomes in the large agricultural sector in particular. From this point of view, working with NSS regions which are more homogeneous in character would seem to be more pertinent. We pursue the analysis at both levels. Chapter 5 is a contribution to the analysis of inter-state differences in labor-market outcomes. In the following chapter we work with NSS regions and look at differences in rural poverty in particular. Since the number of NSS regions is large, an attempt is to group them into seven 'broad regions' defined in terms of agro-economic conditions and geographical contiguity. This attempt at the analysis of rural poverty in terms of grouped NSS regions is, we believe, the first such attempt at regional analysis, and will undoubtedly be improved upon by other researchers.

Part III of the book shifts attention to individual industrial sectors of the economy. The problem of labor absorption at reasonable levels of earnings in agriculture is discussed in Chapter 7. Two important questions related to the performance of the agricultural sector are also addressed. The first is the relationship of agricultural productivity to off-farm activities; and the second is the trends in household welfare of different classes of farmers, especially in the post-reform period. The unit-level data available from the NSS are analyzed to throw light on these major issues. The last three chapters in Part III are on the performance in the non-agriculture sector. Chapter 8 is a detailed analysis of the low elasticity of employment with respect to output in the formal manufacturing sector. The fact that the (relatively) high productivity formal sector has been able to create only a low rate of employment growth, in spite of the fairly high rate of output growth, has meant that labor going into the manufacturing sector has been largely absorbed in the low-productivity informal sector. Chapter 9 in fact shows that the developments in Indian manufacturing have been somewhat more complicated than that. Apart from the labor absorption in the truly informal sector–consisting of household enterprises employing none or only a few hired workers, employment has been disproportionately concentrated at the small end of the formal sector, in units employing 6–9 workers (the so-called DME sector). The bi-modal distribution of employment in Indian manufacturing (with concentration at the smallest and the highest size-groups) has given rise to the problem of dualism in Indian manufacturing. This issue is discussed in detail in Chapter 9–where we analyze the adverse impact of the phenomenon of the 'missing middle' on healthy manufacturing growth, and also the causes leading to its origin and persistence. Chapter 10 takes on the tertiary sector which has been the major source of labor absorption from agriculture.

In the last part of the book–Part IV–labor-market institutions are studied. A critical evaluation of labor laws affecting the formal sector is followed by a review of on-going initiatives to tackle the difficult question of state intervention to improve the conditions of the large numbers of workers eking out a living in India's informal sector.

This page intentionally left blank

Part I
Trends in poverty, inequality, employment and earnings

This page intentionally left blank

2 Poverty, growth and inequality in the pre- and post-reform periods and
the patterns of urbanization in India An analysis for all-India and the major states

This chapter 1 attempts to assess the impact of the economic reforms (including liberalization) on the incidence of poverty in India. It does so by comparing the changes over the period 1993–1994 to 1999–2000 with the course of poverty decline in the previous quinquennial 1987–1988 to 1993–1994. These are the two periods covered by the six yearly ('thick') rounds of the National Sample Survey (NSS) household expenditure surveys.

A number of researchers have already worked on these particular data sets. Major contributions have been published by Sundaram and Tendulkar (2000, 2001 and 2003) and by Deaton and Dreze (2002). Our work builds on and extends these important pieces of research.

In the first section of the chapter we outline the methodology of decomposition changes in the incidence of poverty used elsewhere by Mazumdar and Son (2002). It seeks to quantify the components of the change between any two periods between those due to growth in mean income (or expenditure), those due to distribution of income, and those due to a shift of population between sectors of the labor market (e.g., the rural and the urban sectors). The first section also undertakes a reassessment of the data on measured poverty for the NSS rounds in question. As is well known from the work referred to above, there are two important issues in the empirical use of the NSS data. First, we have the problem of the correct assessment of the poverty line for different sectors, dates and regions of India. Second, we have the problem of comparability of the measured household expenditure per capita over time due to some changes in the recall period in the successive NSS surveys. In the first section we extend the Deaton and Dreze method of assessing the poverty lines, and the Sundaram–Tendulkar approach to deriving consistent estimates of the average per capita expenditure (APCE) of households at different dates. New estimates of the incidence of poverty are presented.

The second section discusses the main results for all-India. An important extension to previous work is the explicit distinction made in the analysis between metro and non-metro towns.

The third section contains a detailed analysis of the state-level data, and some progress is made in the unraveling of critical inter-state differences in the changes in the incidence of poverty over the periods studied.

The final section presents the major conclusions of the work.

The decomposition of poverty

The conventional method of understanding the dynamism of changes in poverty is done through various inequality measures including that of Lorenz curve and various Entropy measures. The decomposition exercise undertaken here does not require us to specify an inequality measure. It uses an idea of shift in that part of the Lorenz curve, which affects the poor.

A brief exposition of decomposition methodology

To get better understanding about dynamics of changes in poverty, the change in the incidence of poverty can be broken down into three elements: (i) any shift in population between the different segments with different degrees of poverty; (ii) the growth in income in each of the segments; and (iii) the change in the distribution of income, particularly at the lower end where the poor households are located. The methodology of such decomposition is set out in Appendix 1. To summarize the result, the percentage change in poverty for the whole economy can be expressed as:

where fi and Pi are the population share and poverty index of the ith group respectively;

the subscript m denotes the change in poverty due to mean income growth and the subscript I gives the measure of poverty change due to change in inequality.

The first term in equation (1) measures the effect of growth within each group on overall change in the poverty incidence, when the distribution within each group remains the same over time. This first term can be further decomposed into two terms:

The first term on the right-hand side of above equation measures the effect of growth on percentage change in poverty under the counter-factual that all groups enjoyed the same uniform growth rates and the second term in the right-hand side measures the effect of differential growth rates within groups.

So substituting (2) into (1), we get a decomposition that expresses the percentage change in the poverty incidence as the sum of four components: (1) overall growth effect when inequality in the distribution does not change; (2) effect of differential growth rates in different groups; (3) effect of change in inequality within different groups; and (4) effect of changes in population shares between groups. This is an exact decomposition and, therefore, there will not be any residual term.

The database

The database used in this analysis is the 'consumption expenditure survey' of various quinquennial rounds of National Sample Survey Organisation (NSSO).2 The purpose of this study is a comparison of the incidence in poverty between the pre- and post-liberalization periods in India. We have done our analysis on the basis of the three quinquennial rounds, i.e., 43rd (1987–1988), 50th (1993–1994) and 55th (1999–2000). We intend to capture the impact of liberalization by comparing the change in poverty for the first period (1987–1988 to 1993–1994) with that in the second period (1993–1994 to 1999–2000). The decomposition exercise for the Head Count Ratio (HCR), the poverty gap ratio (PGR) and the squared poverty gap ratio has been carried out for 16 major states and for all India (16 states combined).

Adjustment made in the database

We have used average monthly per capita consumption expenditure (APCE) as the proxy for per capital income. However, certain adjustments were made to APCE for the year 1993–1994. Expenditures of all consumer items of 1987–1988 and 1993–1994 are based on a 30-day recall period, known as uniform recall period (URP). For 1999–2000, all but five items are based on the 30-day recall period. The expenditures on the five remaining items are based on 365-day recall periods. These items are clothing, footwear, education, institutional medical expenses and consumer durables. So for 1999–2000, the reference period is known as the mixed reference period (MRP). In the year 1993–1994, for these five consumer items, expenditure data were collected for both 30-day and 365-day reference period. To make 1993–1994 data comparable with 1999–2000 data we replaced 30-day expenditures of these five items with 365-day expenditure.3 In this fashion, we could change 1993–1994 URP (uniform reference period) expenditures into MRP (mixed reference period) expenditure. A comparison of APCE on these five items by the 30-day and the 365-day reference periods for the year 1993–1994 showed that in both rural and urban areas the change of the reference period from 30-day to 365-day made substantial difference largely in clothing.

We could not convert 1987–1988 consumer expenditure data into MRP in a similar fashion because the expenditure data on the above-mentioned five items were not collected for both the 30-day and the 365-day reference periods. Hence the decomposition analysis for changes in poverty for the period 1987–1988 to 1993–1994 (pre-liberalization period) will be based on URP and for the period 1993–1994 to 1999–2000 will be based on MRP. We will surely lose continuity of poverty estimates in this fashion but avoiding this important issue would otherwise lead to an upward bias in the reduction of poverty in the post-liberalization period relative to the pre-liberalization one.

It might be objected that flow expenditure on low-frequency articles like durables might be reported differently by poor people than by rich consumers for the two alternative recall periods. In fact experiments performed in the 'thin' samples of the 51st to the 54th rounds showed that on the 365-recall period, lower-income households reported higher annual rates compared to the 30 day recall method, but richer households had exactly the opposite bias. At the same time there is an expectation that there has been a large increase on durables affecting all classes. A comparison of the change in APCE based on the 30-day recall as for the first period might not be strictly comparable to the change in the second period based on the 365-day recall. One might be missing less at the mean in the first change than in the second change.4 An examination of the detailed data on consumption by items and income groups, however, showed that the major difference for poorer groups in the reported expenditure by the two recall periods was in clothing, not in all durables. In the lowest eight income groups, ranging up to the 35–40 fractile, in the 1993–1994 (50th round) survey, the highest difference was Rs.2.5 for durables compared with Rs.17.25 for clothing (the full data are given in Mazumdar and Sarkar 2004). Thus the income-related bias in reported flow of expenditure on durables might not quantitatively of great importance in the two periods of comparison with different recall periods.

Choice of poverty line

In choosing the poverty line we deliberately did not choose the official poverty line as given by the Planning Commission of India. Historically, the rural–urban price differential as incorporated in the official poverty lines at all-India level was around 15 percent level. But the 1993 Expert Group Report recommended separate rates for each state (based on studies of interstate price differentials) and did not explicitly consider the urban to rural differentials. As a result, in 1999–2000, the urban to rural differential implicit in the official lines was around 39 percent and it is astonishingly large for some states (Deaton 2003). The effect of the adoption of the Expert Group lines was to raise measured poverty in urban relative to rural areas. The poverty-line figures, by state and sector, are calculated by using the Tornqvist price index presented by Deaton (2003).5

Following Deaton's procedure, the starting point for calculation of poverty indexes is the official rural all-India poverty line for the 43rd round, 1987–1988. The figure is Rs.115.7 per capita per month. First, rural poverty lines for states are obtained by multiplying this base poverty line by rural price indexes for each state relative to all India. Urban poverty lines, for each state as well as for all-India, are calculated from the rural poverty lines by scaling up by the respective urban relative to rural price indexes.

Similarly, for the 50th round (1993–1994), the all-India rural poverty line of 115.7 of the 43rd round is scaled up by the index for all-India rural for the 50th round relative to the 43rd round. The figure thus calculated is 196.5. Rural poverty lines for states are obtained by multiplying this base poverty line by rural price indexes for each state relative to all-India. Urban poverty lines, for each state as well as all-India, are calculated from rural poverty lines by scaling up by the respective urban relative to rural price indexes (see Mazumdar and Sarkar 2004 for the table of poverty lines by states and rural–urban location).

Results for all-India

We first discuss the pattern of poverty decline in the two periods for the whole of India–based on the new figures for the 16 states considered. The results for the decomposition analysis are presented in Table 2.1.

According to our estimates the head-count ratio (HCR) was reduced at a perceptibly higher rate in the more recent period–the decline was about 20 percent higher. This apparent acceleration is, however, largely due to the smaller base of the HCR at the beginning of the second period. The absolute decline in HCR was 6.3 percentage points in 1987–1993 and 5.3 in the 1993–2000 periods. Thus our figures support the conclusion of Deaton and Dreze that 'poverty decline has been fairly evenly spread between the two sub-periods (before and after 1993–1994) in contrast with the pattern of acceleration in the second sub-period associated with the official estimates' (p. 3734).

Growth of mean consumption accelerated in the second period, and played a larger role in the poverty reduction in this period. It can be seen that the inequality effect overall (i.e., taking the rural and urban areas together) continued to play a contributory role to poverty reduction, but the share of this factor in the reduction was much reduced.

Important changes, however, emerged in the relative importance of the rural and the urban areas in the process of poverty reduction. The share of the urban areas in the overall poverty decline increased in the later period (from 12 percent

of the total percentage decline to 21 percent). This bigger role of the urban sector in poverty decline was, however, not due to accelerated population shift to the urban sector. In fact, the 'population shift' effect, while playing a minor role in both periods, actually decreased significantly in the 1993–2000 period.

The crucial element was the higher growth rate in the urban sector. Both the sectors increased the rate of mean growth, but it can be seen from the third column of Table 2.1 that the differential effect of growth rates reduced the incidence of poverty significantly in the urban areas, but increased it in the rural areas. If the latter had grown at the same rate as the rest of the economy, poverty reduction in the 1993–2000 periods would have been 30 percent higher.

The impact of the differential growth rate was balanced to some extent by the impact of changes in inequality in the two sectors. While inequality (in the relevant range of the Lorenz curve) decreased in the rural sector, it deteriorated somewhat in the urban sector, thus canceling out some of the poverty-reduction effect of the differential growth in the sector. But the inequality effects were not as strong as the differential growth effect. This explains the larger role of the urban areas in poverty reduction in this period.

Elasticity of poverty decline and the poverty gap ratio

We have seen that both the growth rate of mean consumption and the rate of decline in the headcount ratio of poverty accelerated in the second period of our study. The elasticity of the change in HCR with respect to the growth in consumption is of interest. Table 2.2 shows the numbers for the two periods. The values of the elasticity in the rural and urban areas are very close together. In both sectors there has been a significant fall in the elasticity in the post-reform period. The results of the decomposition analysis given in Table 2.1 suggest that the reasons for this decline are different in the two sectors. In the rural economy the inequality effect increased its negative value, suggesting that ceteris paribus the effect on HCR of growth would be strengthened. But there was a significant fall in the shift of the labor force out of this sector, which weakened the impact on poverty. By contrast the urban areas–where the change in HCR benefited from the fall in the intake of labor–suffered from an adverse distributional effect.

The HCR of course only measures the number of people below the poverty line, and does not take account of the economic distance of the poor from the poverty line. This is addressed in the group of measures called the 'poverty gap ratio' (PGR) and its variants. Table 2.3 shows the values of the elasticity of

PGR with respect to mean consumption growth in the two periods. It is seen that the elasticity of the PGR with respect to growth is much higher than that of the HCR in both periods and both sectors. This partly reflects the fact that the absolute value of the initial PGR is much lower than that of the HCR. But it also draws attention to the substantive point that the growth process has affected people further below the poverty line strongly–not just those slightly below the threshold.

A second interesting point to note is that as with the HCR the elasticity of PGR fell in the second period in both sectors. However, the decline in elasticity in case of PGR seems to have been much stronger in the rural areas suggesting that the poorest of poor were worse hit.

Table 2.4 gives the results for the decomposition analysis for the poverty gap measure. The different components of the poverty reduction appear to behave in much the same way for this measure as for the HCR. As already pointed out the percentage decrease in PGR is larger than in the HCR, but qualitatively the role of all three components of poverty decline is the same in the two cases. In the post-reform period the role of population shift is much reduced. The urban areas gain in the strength of the growth effect, but inequality increases, offsetting the effect of growth to some extent. The relative slowdown in growth in the rural sector is partly countered by a favorable inequality effect.

There are differences in the relative magnitudes of the various effects. One interesting difference is that in the recent period the inequality effect on the PGR seems to be stronger than on the HCR in the rural areas, but the other way round

in the urban sector. In the rural sector the inequality effect is 38 percent of the growth effect in the HCR decomposition, but 52 percent in the PGR analysis. The offsetting effect of increased inequality is, however, weaker for the PGR in the urban areas (15 percent of the growth effect as against 25 percent for the HCR decomposition).

We are inclined to agree with the Deaton–Dreze conclusion that very little additional insight is to be gained from the detailed analysis of the poverty- gap ratio or its further refinements over and above what we learn from the simple analysis of the HCR (Table 2.5). In view of this we will make no further reference to the measures other than HCR in the subsequent discussions.

Metro and other urban areas

An interesting question pertains to the relative importance of metropolitan (population > one million) and other urban areas in poverty reduction.

The decomposition of HCR was done for metro and other areas separately. Out of the 16 major states considered for the decomposition analysis, only seven states had a metro city in the year 1987–1988, ten states in the year 1993–1994 and 11 states in the year 1999–2000. So to maintain uniformity, we analyzed seven states separately that had a metro city throughout our period of analysis. However, separate analysis was undertaken for the three states that had a metro area only since 1993–1994.

The incidence of poverty is as expected higher in the non-metro areas. We studied changes in the incidence of poverty over the two sub-periods between 1987–1988 and 1993–1994, and 1993–1994 and 1999–2000 and these have been worked out on the basis of both the URP and the MRP criteria (Mazumdar and Sarkar 2004, Table II.3).

For the seven states the decline in poverty incidence in absolute terms does not differ much between metro and non-metro areas in the 1987–1993 period. This implies that in proportionate terms the decline is much more in the metro areas. But the trend seems to have been reversed in the later sub-period.

The absolute decline is much smaller in the metros as the HCR seemed to be nearing the floor level, though the difference in percentage terms is not all that much.

Our decomposition analysis was applied to the data for the metro and the non-metro areas in the same way that we had done for the rural and the urban areas as a whole. The results are set out in Table 2.6.

It is seen that a major change in the more recent period was registered by the non-metro urban sector. Differential growth rate favored poverty reduction in a more pronounced way in the non-metros. The non-metro sector also suffered relatively much more from an increase in inequality which seemed to have affected the urban areas as a whole. Combined with a more positive contribution from population shift to poverty reduction, almost the entire decrease in urban HCR in the 1993–2000 period was accounted for by the non-metro sector.

The urban sector by size classes of towns

For some purposes it might be better to classify the urban sector by more size classes than just two. We distinguished three sub-groups: towns with a population of less than 50,000 (small); those larger than this but with less than ten lakhs or one million (medium and large); and those more than one million (metro).

For the country as whole, there is a remarkable difference between the two periods. For the 1987–1993 period the rate of growth of APCE was directly related to the size of towns, the largest towns having the highest growth rate. Accordingly the rate of decline in HCR was also varied directly with the size groups–and in fact this positive relationship was much stronger. In the post-reform

years 1993–2000 the relationship has been reversed remarkably. The reversal again is much more prominently revealed in the variations in the rates of decline of the HCR. The small towns had a rate of decline 50 percent higher than the largest group.

This interesting result poses the question: what aspect of the post-reform growth process has been responsible for this reversal of the fortunes of the small towns relative to the larger ones? It is possible to hypothesize that the result might be the consequence of either a strong 'trickle down' effect powered by a decentralization of non-agricultural activities in the urban sector. Alternatively the smaller towns might have enjoyed a stronger growth rate (and poverty reduction) in the post-reform years because of the impact of the growth and commercialization of the agricultural economy. It is probable that both influences have been present in the process observed.

It should be noted that some of the individual states show trends different from the one just mentioned for all-India. There are, in particular, five states in which the rate of growth of APCE and HCR are directly related to the size class of towns–the opposite of the result for India as a whole. These are: Gujarat; Karnataka; Maharashtra; Rajasthan; and Tamil Nadu. As it happens, these are the states which have been the largest recipients of Foreign Direct Investments. Since it is well known that FDI goes almost exclusively to metro cities, the hypothesis suggested strongly is that FDI has given an uplift to the growth rate of metro areas in those states where it has played a significant role–and that this impact has raised the growth rate of mean consumption sufficiently induce a higher rate of poverty decline than would be expected by looking at the average for all-India.

The relevant data on FDI per capita in metros by states are plotted along with the rate of growth of APCE in the individual states for the 1993–2000 period in the scatter diagram of Figure 2.1. The relationship is found to be a very strong one.

Figure 2.1 Relationship between FDI and growth rates of APCE in metro areas.

Note

See full names of different states like WB, MP, AS, etc. in first column of Table 2.8.

This result suggests that there are two different aspects to the impact of the post-reform developments including globalization on growth and poverty decline in the urban metro sector. On the one hand, there has been a distinct trend towards decentralization of economic activities to smaller towns and cities. This has led to the inverse relationship between growth and the size of towns observed in many states and in India as a whole. There are, however, a group of states in which the role of FDI is strong, and the impact is seen in a high growth rate in metro areas, so that a direct relationship between growth (and poverty reduction) and town size is observed. The only exception to this two-way classification of states is Rajasthan. The inflow of FDI per capita in metros in this state is low, yet it shares the characteristics of the high FDI states in having a relatively high growth rate in the larger towns (see Figure 2.1). It is, however, well known that even if the inflow of FDI is small Rajasthan has participated in the globalization process strongly through the promotion of international tourism in particular.

The implication of the argument of the last paragraph is that the 'trickle down' effect on smaller towns has been more important in the states with a lower level of international connection. The reform process has encouraged decentralization in these states. But what about the impulse to growth in small towns coming from the rural sector? Figure 2.2 plots the growth rate of APCE in small towns (with population of less than 50,000) against the growth of APCE in rural areas of individual states. There is indeed a positive relationship, but it is a relatively weak one.

Trends in the rural–urban dualism

An important issue in the development literature is the rural–urban gap in levels of income (and consumption) and in the incidence of poverty. Post-reform developments and globalization are sometimes viewed with concern as acting towards increasing the degree of this dualism.

Figure 2.2 Relationship between growths of APCE in small towns and rural areas.

Note

See full names of different states like WB, MP, AS, etc. in first column of Table 2.8.

We tried to throw some light on the variations in the gap between the incidences of poverty between rural and the urban areas across states. The variable to be explained is the ratio of the HCR in rural areas to that in the urban areas. Since the rural economy is large in all Indian states, a higher level of development in a state would generally imply a higher rural APCE. Also, with economic growth urbanization increases. A little reflection shows that the net impact of both these variables on the relative rural–urban gap depends on whether or not 'trickle down' is confined to the sector in question, or extends to the other sector. Take the expected sign of APCE (rural) first. If the impact is largely confined to the rural sector, then the relative gap would be reduced (the sign would be negative), but in so far as it reduces the level of the urban poor through higher demand for urban goods and services, the sign of the variable would be positive. The final outcome depends on the relative strength of the two forces. Similarly a higher rate of urbanization would mean a larger relative gap if urbanization has only a limited effect on rural incomes at the lower end of the scale, but would go the other way if the urban to rural linkage is strong. Second, as far as urban poverty is concerned, the size distribution of cities also matters since the incidence of poverty is inversely related to city size. We can use the summary measure of the share of metro population in the urban sector as a variable to capture this effect. The prediction about the sign of this variable, as with the other variables, depends on the relative strength of the linkage with the economies of sectors outside the metros. Generally, the incidence of poverty is lower in metro cities, so a greater preponderance of metro population would imply a higher relative gap in poverty incidence between the rural and the urban sector. But if 'trickle down' in states with a larger metro population is weak, the higher development of metro towns would have a limited effect on poverty incidence in non-metro urban areas, thus pushing up the HCR in the urban sector as a whole, i.e., the rural–urban ratio in HCR could be lower. The regression model with these variables is fitted to interstate variations in the rural–urban HCR ratio for different dates of the NSS rounds, and the results are reported in Table 2.7.

The results show that in the pre-reform years, between 1987–1988 and 1993–1994, the impact of the rate of urbanization is significantly negative: 'trickle down' extends to the rural economy. But the sign of the variable measuring the share of the metro in urban population is significantly positive: the growth of metro towns apparently reduces HCR in the urban sector as a whole, not so much in the rural areas. The sign of APCE (rural) is positive but not very significant.

In the post-reform years both the urbanization and metro share variables lose their significance, and the APCE is even less significant. Evidently factors other than those connected with the rural–urban 'trickle down' process discussed above now explained inter-state variations in the poverty gap. We have already seen that in the post-reform years urban growth was more important in reducing poverty across a wide range of states. This process reduced the inter-state variations in the relative poverty-gap even as it reduced the overall value of this gap.

We have also seen that within urban areas there was a distinction between the states which received a relatively large flow of FDI and those who did not. In the FDI states, it will be recalled, the pattern of poverty decline in urban areas by size of towns was different. It was inversely related to the size group of towns and was lowest in the largest cities. The relationship between HCR and town size was just the reverse in 'non-FDI' states. We tried to see if this difference was in any way related to the pattern of inter-state variability of the rural–urban poverty gap. The last equation of Table 2.7 shows that it is indeed so. FDI (urban) per capita is the only significant variable in the estimated equation and is positive, implying that those states which have a large FDI inflow have a significantly lower incidence of urban poverty relative to the rural sector. In other words, the effect of metro towns in increasing the relative poverty gap–which was true of the entire range of states in the pre-1993–1994 years–is now significant only in the FDI states. It draws attention to the point that FDI is an important player in the poverty scene in spite of the total inflow being much smaller in India than in other countries like China. FDI inflow reduces poverty significantly in the largest cities, but its 'trickle down' effect is limited outside the metro areas.

Differences by states

It is well known that Indian states differ substantially in the incidence of poverty. Also the distribution of population among the different states is uneven. The trend in the all-India measure of poverty, such as the HCR, will then be affected by the way the pattern of the difference in poverty reduction between more and less populous states. It has been hypothesized that growth rates and hence the rate of poverty reduction have not been generally stronger in the states with a larger share of the poor. The following paragraphs explore this hypothesis in more detail for the two periods we are considering.

Table 2.8 gives the shares of the individual states in the total count of those below the poverty line for the three dates (corresponding to the 38th, 50th and the 55th rounds of the NSS).

Seven states – Andhra Pradesh, Bihar, Madhya Pradesh, Maharashtra, Orissa, Uttar Pradesh and West Bengal–accounted for over 70 percent of the total poor in the rural areas in 1999–2000. Just three states–Bihar, Madhya Pradesh and Uttar Pradesh–accounted for over 40 percent of the rural poor. It is interesting to note that the same states to a large extent account for the bulk of the urban poor as well. The only difference between the two sets is Orissa which accounts for only 2.6 percent of the urban poor, reflecting the relative underdevelopment of the state. Looking back to 1987–1988 it is seen that there is not much difference in the spatial distribution of the poor–the same states account for the bulk of the rural and the urban poor. Perhaps the concentration of the poor in these states was a little higher in the earlier period.

It is interesting to see which states fell behind the all-India average in APCE growth rate in the most recent post-reform period. Because of its weight we look especially at the rural areas. The lagging states are: Assam, Andhra Pradesh,

Bihar, Madhya Pradesh, Orissa, Rajasthan, Uttar Pradesh and West Bengal (for details see Mazumdar and Sarkar 2004, pp. 24–25). These states coincide with the set accounting for the bulk of the rural poor. The only state with large HCR missing in the set is Maharashtra–which is fairly close to the growth rate rural all-India. We conclude that the spatial pattern of growth rates in the 1993–2000 period was not favorable to the cause of poverty decline in the rural sector.

Poverty decline in the two periods

Further light can be shed on the experience of inter-state differences in poverty decline in the post-reform period by looking directly at the changes in the HCR. In Figures 2.5 through 2.8 we present the scatter diagrams of the state-level changes in the HCR over the two time periods between the NSS surveys – 1987–1988 to 1993–1994 and 1993–1994 to 1999–2000. We plot the change in the HCR on the y-axis against the level of HCR in the initial year of the respective period on the x-axis. The graphs are drawn separately for the rural and the urban areas.6

We would expect that the decline in HCR would be higher in the states where the absolute value of the HCR is higher. The percentage decline in HCR is given by the ratio of the two magnitudes. Convergence between states in the incidence of poverty will occur only if the percentage decline increases with the initial value of the HCR–i.e., the relationship between the two magnitudes in the scatter diagram is non-linear. The graphs (Figures 2.3 to 2.6), however, show that there is at best a linear relationship between the decline in HCR and its initial value. There is no evidence of inter-state differences in poverty incidence to converge over time in either sector.

Second, it is seen that while the slope of the line relating the initial HCR to its absolute change is more or less the same in the urban areas, it has definitely become flatter in the rural areas in the post-reform years. For the rural sector as a whole we can no longer say that the percentage decline in poverty is directly related to the absolute value of HCR in the rural areas of the state in the 1993–1994 to 1999–2000 period. The reason for this is that several of the states suffered from a deceleration of poverty decline in their rural sector in the post-reform areas. There is naturally an overlap between the states lying significantly below the regression line of Figure 2.6 and those identified earlier in this section as being laggards in rural APCE growth. They include Andhra Pradesh, Assam, Madhya Pradesh and West Bengal.

Table 2.9 showing the classification of states into four groups and

Figure 2.3 Poverty (HCR) and declines in HCR from 1987–1988 to 1993–1994 (rural) across states.

Note

See full names of different states like WB, MP, AS, etc. in first column of Table 2.8.

Figure 2.4 Poverty (HCR) and declines in HCR from 1993–1994 to 1999–2000 (rural) across states.

Note

See full names of different states like WB, MP, AS, etc. in first column of Table 2.8.

Figure 2.5 Poverty (HCR) and declines in HCR from 1987–1988 to 1993–1994 (urban) across states.

Note

See full names of different states like WB, MP, AS, etc. in first column of Table 2.8.

their changes over the two periods might help to throw some further light on state-level changes in rural poverty decline. The groupings are from I to IV in descending order of poverty decline. Table 2.9 confirms that the flattening of the regression line in Figure 2.4 is largely due to four states slipping from Category I (high HCR decline) and category II (middling decline) in the pre-reform period to category IV in the post-reform one. This adverse effect is balanced to some extent by Maharashtra moving from IV to I, and Karnataka from IV to II.

Further details on the change in the HCR between the two periods for individual states is provided in Table 2.10 which reproduces the results of the

Figure 2.6 Poverty (HCR) and declines in HCR from 1993–1994 to 1999–2000 (urban) across states.

Note

See full names of different states like WB, MP, AS, etc. in first column of Table 2.8.

decomposition analysis applied to the data for each state. The total percentage change in poverty for the state as a whole (rural and urban combined) is given in column 10, while the components of this statistic are found in the six columns preceding it. The column of %D gives the component due to differential growth, while %I is that due to change in inequality. The differences between HCR change and the sum of %D and %I in each sector are the sum of two elements: (i) 'Uniform growth'–the hypothetical growth if the APCE in each sector had grown at the same rate as in the state as a whole; (ii) the effect of the rural–urban shift (the latter has been small in most states). The last two columns of Table 2.10 give the percentage change in HCR in each of the two sectors – rural and urban separately.

Table 2.10 Decomposition of percentage change in head-count ratio (HCR)

We can use the table to classify the states into three groups in terms of the percentage change in HCR over the two periods we are considering (taking the rural and the urban sectors together):

1 states which accelerated the decline in HCR in the post-reform period markedly (more than doubled the percentage decline): Bihar, Gujarat, Maharashtra, Karanataka, Haryana, and Punjab;

2 states which had a less spectacular, but still substantial, decline in HCR relative to the earlier period: Tamil Nadu, Rajasthan, Uttar Pradesh and Kerala;

3 states which suffered retardation in HCR decline (or actually registered an increase in HCR) Andhra Pradesh, Madhya Pradesh, Orissa, West Bengal and Assam.

It is to be noted that the list in group 3 is identical to the group already mentioned earlier as the laggards in rural poverty decline. It shows the quantitative importance of the rural sector in the overall trend in poverty reduction.

While the growth of the rural sector is naturally the dominant influence on the overall decline in HCR (the share of the rural sector in the incidence of poverty being so much more), it is important to note that the urban sector became a much more important player in several states. We have already seen in section II that this is true for all-India in the post-reform years. But some individual states stand out. The urban sector of Gujarat, Karnataka and Punjab among the group 1 states, and Tamil Nadu and Rajasthan among the group 2 states contributed to the total HCR decline of to an extent of a third or more of the contribution of the rural sector. In all these states the differential rate of growth was higher in the urban sector, contributing to the HCR decline (%D in the urban sector was negative).

Of the states in which HCR declined Bihar and Uttar Pradesh are the only ones in which the rural sector increased its contribution to the decline. Rural growth also contributed strongly to HCR decline in Kerala, but it was offset to some extent by a substantial increase in inequality in the sector, so that the contribution of the urban sector increased in the 1993–1999 period. In the case of Bihar (and to some extent Uttar Pradesh, in so far its Eastern districts are really an extension of the economy of Bihar) the strong rural growth in APCE causing poverty decline is less due to the growth rate of its domestic rural economy, as it is to the remittance sent back home by migrant labor participating in the rural economy of the North-Western states and in urban areas scattered all over India.

We have already underlined in the third section the point about the increased inequality in the urban sector in the post-reform period for India as a whole. Its relative impact in retarding poverty decline is significant, but quantitatively not substantial. The state-level data reveal that the inequality effect has been more important in some states. Running down the column (9)–%I for the urban sector–it is seen that this has been so in Tamil Nadu, Gujarat, Haryana, Kerala and Punjab. It should not, however, be concluded that inequality increased only in the urban economy of some states. There was substantial inequality increase retarding poverty decline in the rural sector of a few states as well. These include Tamil Nadu, Kerala, Punjab, and to some extent Gujarat. It is interesting to note that these states are also the ones which experienced the adverse inequality effect in the urban sector as well.

It is important to comment on the experience of the group 3 states. Andhra Pradesh suffered from a slight retardation in the rate of decline in HCR, but the incidence of poverty actually increased in the other states in this group. It is seen that all these states suffered from adverse movements in the HCR in the rural sector. The figures of column (5) reveal that the crucial problem was the differentially lower growth in the rural economy which contributed to an increase in HCR. In the case of three of three states–Andhra Pradesh, Madhya Pradesh and West Bengal–the poverty incidence would have been worse but for the stronger performance of the urban sector. Assam stands out in registering an increase in inequality both in the rural and the urban sectors. An increase in urban inequality also contributed to HCR decline in Madhya Pradesh, but its relative importance was much less than that of the slow-down in rural growth.

Conclusions

In this chapter we have contrasted the change in poverty incidence in the 1987–1993 period with that in the 1993–1999 years. The second period could be seen as one in which the impact of the reforms tied to liberalization of the economy could be expected to have had some impact. We have based this comparison from several angles on the 'consumption expenditure survey' data generated by the NSS for three quinquennial rounds: the 43rd (1987–1988), the 50th (1993–1994) and the 55th (1999–2000). We have addressed the methodological problems involved in the surveys, and produced a new set of consistent data on mean consumption of the survey households, and the incidence of poverty. The differences between our estimates and those of other researchers have been spelled out in the first section and the Appendices.

We have a used a decomposition analysis of the percentage change in poverty over the two successive periods. This method used elsewhere by Mazumdar and Son (2002) enables us to quantify the relative contribution of three elements to the overall change in poverty incidence: mean growth of consumption, population shift between defined sectors and the change in inequality. The analysis is applied to the economy as a whole and to its rural and urban sectors. It is done for all-India and for the individual states.

The more important conclusions are the following:

1 At the all-India level the absolute change in the HCR was about the same in the post-reform period as in the previous one, but the rate of change was higher because the initial base was smaller. However, the growth rate of APCE increased rather more so that the elasticity of poverty change with respect to income fell. This is true even more for the poverty gap measure.

2 The share of the urban sector in poverty reduction increased in the second period. This was not due to a larger shift of population to urban areas–in fact the rate of this shift decreased. The major reason for the change was the higher differential growth rate of the urban sector. It was offset, but only partially, by an increase in urban inequality. However, the relative impact of rising urban inequality in retarding poverty decline in the late nineties was quantitatively not substantial.

3 Turning to poverty decline by size of towns there are two different aspects to the impact of the post-reform developments including globalization. On the one hand, there has been a distinct trend towards decentralization of economic activities to smaller towns and cities. This has led to the inverse relationship between growth and the size of towns observed in many states and in India as a whole. There are, however, a group of states in which the role of FDI is strong, and the impact is seen in a high growth rate in metro areas, so that a direct relationship between growth (and poverty reduction) and town size is observed, as was the case generally in the urban sector as whole in the previous period.

4 State-level analysis showed that the states could be divided into three groups when we compare the change in HCR in the post-reform years with the period preceding it:

group 1 are those which accelerated the decline in HCR in the post-reform period markedly (more than doubled the percentage decline): Bihar, Gujarat, Maharashtra, Karanataka, Haryana, Himachal Pradesh and Punjab;

group 2 states had a less spectacular, but still substantial, decline in HCR relative to the earlier period: Tamil Nadu, Rajasthan, Uttar Pradesh and Kerala;

group 3 states suffered retardation in HCR decline (or actually registered an increase in HCR): Andhra Pradesh, Madhya Pradesh, Orissa, West Bengal and Assam.

It is seen that the unhappy performance of the group 3 states is largely due to retardation in the rate of growth of the rural economy of these states. The growth of the urban sector, however, played a significant role in the poverty decline of group 1 and group 2. We have already seen that at the all-India level, urban growth in the post-reform years was higher than rural growth. Although the urban economy is still a small part of the economy, its contribution to poverty reduction started being important in several states. The urban sector of Gujarat, Karnataka and Punjab among the group 1 states, and Tamil Nadu and Rajasthan among the group 2 states contributed to the total HCR decline to as much as a third or more of the contribution of the rural economy.

5 Increased inequality seems to have been associated with the higher growth rate in the urban sector in the post-reform period for India as a whole. Its relative impact in retarding poverty decline is significant, but quantitatively not substantial. The state-level data reveal that the inequality effect has been more important in some states. They include Tamil Nadu, Gujarat, Haryana, Kerala and Punjab. It should not, however, be concluded that inequality increased only in the urban economy of some states. There was substantial inequality increase retarding poverty decline in the rural sector of a few states as well. Theses include Tamil Nadu, Kerala, Punjab, and to some extent Gujarat. It is interesting to note that these states are also the ones which experienced the adverse inequality effect in the urban sector as well.

Appendix 1

Poverty-decomposition methodology

Let us divide the total population into k mutually exclusive socioeconomic and demographic groups. For decomposable poverty measures, then, we can write the total poverty as the weighted average of poverty within each group.

where fi and Pi are the population share and poverty index of the ith group, respectively. Further, define the change in poverty between two periods as

where and ., and P2i being the poverty incidence

in the group in years 1 and 2, respectively, and f1i and f2i are the population shares of the ith group in years 1 and 2, respectively.

Equation (1) can be written as

which shows that the change in total poverty can be written as the sum of two components. The first component measures the effect on total change in poverty due to changes in within-group poverty and the second component estimates the change in total poverty due to possible shifts in population between groups.

The percentage change in total poverty, thus, can be written as follows:

where and .

Note that the first term in Equation (2) estimates the percentage change in total poverty explained by changes in poverty within groups. The second term estimates the percentage change in total poverty due to a shift in population between groups. The shift in population is deemed pro-poor if the second term is negative because it leads to a reduction in poverty. This situation is likely to occur if migration occurs from rural to urban areas. If migration takes place from urban to rural areas, on the other hand, the second component is likely to make a positive contribution to poverty. In this case, the population shift is not pro-poor.

Kakwani (2000) has proposed a decomposition, which explains the percentage change in poverty as a sum of two components: one is the growth effect, measuring the change in poverty when mean income changes but inequality remains fixed and the other component is the inequality effect, which measures changes in poverty when inequality changes but the mean income remains constant. This methodology can now be applied within each group.

A general poverty measure is characterized as

where z is the poverty line, μ is the mean income of society, and L(p) is the Lorenz curve. The Lorenz curve measures the effect of inequality on poverty. Following from Kakwani (2000), the percentage change in poverty can be written as

where (ΔP)m is the change in poverty if mean income changes from μ1 in period 1 to μ2 in period 2 but the Lorenz curve remains fixed. Thus, (ΔP)m can be written as

where L1(p) and L2(p) are the Lorenz curves in periods 1 and 2, respectively. Note that in deriving the mean effect, we can either fix the Lorenz curve for the initial period or for the terminal period. Because we do not know a priori which period of the Lorenz curve we should fix, we have taken the average of the two periods.7

Similarly, the inequality component can be derived as

which estimates the change in poverty if inequality measured by the Lorenz curve changes from L1(p) in the initial period to L2(p) in the terminal period but mean income is fixed between the two period. The sum of the mean and inequality effects gives rise to the total changes in poverty.

We apply the decomposition in (3) within each group, which results in

where

and

ith group in year t (t = 1,2).

From (2) and (4), the percentage change in total poverty can be expressed as

The first term in equation (5) measures the effect of growth within each group on overall change in the poverty incidence, when the distribution within each group remains the same over time. This first term can be further decomposed into two terms:

where

and μ2i = μ1i(1 + g) where g being the average growth rate of the whole population is the mean income of the ith group in year 2 if the income of the ith group were growing at the same rate as the average growth rate of the whole population.

The first term in the right-hand side of (6) measures the effect of growth on percentage change in poverty under the counter-factual that all groups enjoyed the same uniform growth rates and the second term in the right-hand side measures the effect of differential growth rates within groups. Thus, substituting (6) into (5), we arrive at our poverty decomposition that expresses the percentage change in the poverty incidence as the sum of four components: (1) overall growth effect when inequality in the distribution does not change; (2) effect of differential growth rates in different groups; (3) effect of change in inequality within different groups; and (4) effect of changes in population shares between groups. This is an exact decomposition and, therefore, there will not be any residual term. This decomposition does not require us to specify an inequality measure. It uses the idea of shift in that part of the Lorenz curve, which affects the poor.

The first component will always be negative if there is a positive growth in the economy. The second component can be either negative or positive. If it is positive (negative), the disparity in growth rates of different groups has contributed to an increase (decrease) in total poverty. The third component can again be either positive or negative. If it is positive (negative), it indicates that a change in inequality within group has contributed to an increase (decrease) in the total poverty incidence. Finally, the fourth component measures the effect of migration of population between groups on the total poverty incidence.

Appendix 2

This appendix compares our calculation of HCR based on the adjustments made to the NSS figures on Average Per Capita expenditure (APCE) with others in the literature. We campared the APCE by fractile erxpenditure groups in rural and urban areas for the year 1993–1994 as given by Sundaram and Tendulkar 2003a (S&T) and our own calculation. It was observed that some differences exist only in highest fractile (95–100) but not for other lower fractiles. The effect on estimation on Level of HCR would be minimal.

In this regard, it is interesting to note that Sen and Himanshu (2004), following the S&T procedure, found all-India HCR for all areas to be 35.9 percent using URP and 30.6 percent using MRP for the year 1993–1994. S&T in their revised estimates found the corresponding figures to be 37.35 percent (URP) and 32.15 percent (MRP). This is probably because S&T use a somewhat different poverty line than the official poverty line. Sen and Himanshu (2004) further corrected the 55th round estimates of food and intoxicants for possible 'contamination' from the 7-day questionnaire. They used information from early NSS rounds to arrive at some estimates. At the lower bound, the extent of such contamination was found to be small but even then the authors calculated that the 55th round all-India poverty incidence using MRP was 27.8 percent as against the official figure of 26.1 percent. Thus they found the measured decline between 1993–1994 (MRP) and 1999–2000 (MRP revised) to be at most 2.8 percent implying an increase in the absolute number of the poor by five million. Their results are quite opposite to the official calculation of poverty decline (in HCR) by 9.8 percent implying a fall in the number of poor by 60 million and to the S&T revised estimates showing a fall of 4.83 percent denoting reduction in number of poor by 13 million in the period.

3 Trends in employment and earnings 1983–2000

In Chapter 2 the focus was on households. We looked at trends in household welfare – and the changes in the incidence of poverty at the household level in particular. But household welfare is the result of the working of factor markets, and for households near the poverty line, trends in labor markets are of paramount importance. In this chapter our attention shifts to individuals. We begin to address the question: are the observed developments in the labor market consistent with and illuminate the results obtained in the previous chapter on household-level poverty trends?

Trends in aggregate employment

The Task Force on Employment Opportunities appointed by the Planning Commission reported a sharp decline in the labor-force growth rates between the 1980s (1983 to 1993–1994) and the 1990s (1993–1994 to 1999–2000). Its estimates from the NSS showed that the growth rate fell from 2.05 percent per annum to 1.03 (GOI July 2001). Taken in conjunction with the increase in measured rates of open unemployment, this slow-down has been widely interpreted to have been the result of 'discouragement' of potential workers from entering the labor force. There are, however, two basic problems with this estimate: (i) is the question of using the correct age-structure of the population; and (ii) the question of appropriate employment category on which the estimates are based.

On (i) it has been maintained that the survey-based age-structure, on which the weighted employment rates of the NSS are based is less reliable than the age-structure reported by the nearest Census of population figures; and on (ii) the Planning Commission estimates of employment growth are based on the UPSS figures of the NSS, and does not distinguish adequately the supply side aspects of the labor force from the effects originating from changes in the demand for labor. Research on both these questions has been extensive, and we shall begin by summarizing the major conclusions from this research.

The age-structure issue

Sundaram and Tendulkar (2006) have looked carefully at the NSS-based age distribution and compared it with the data reported by the Censuses of 1981, 1991 and 2001. They reported that

using the survey-based age-distribution results in a sharp slow-down in the growth of prime-age (15–59) population in the nineties…. [But] in the context of the observed slow-down of population growth reflecting the decline in fertility–from 2.09 percent per annum (pcpa) to 1.97 pcpa–over the same period, equally problematic is the acceleration in the growth of population in the 0–9 age group–from 1.17 to 1.33 pcpa–between the two periods that shows up with the use of the survey-based age distribution.

It is generally accepted by researchers that, while surveys like the NSS are more reliable at getting at participation or employment rates for different age–sex groups, the age-structure of the population itself is better measured by the Population Censuses. Sundaram and Tendulkar thus use the age–sex-sector specific worker-population rates produced by the NSS and re-weight them by the appropriate demographic structure obtained from the Population Censuses. This adjustment has the effect of substantially moderating the slow-down in the growth rate of the labor force: from about 2.06 percent in the 1980s to 1.58 percent in the 1990s.1

The problem of measuring the labor force

The NSS distinguishes those who participate in the labor market on the basis of several criteria. A major difference is the estimate on the basis of Usual Principal Status (UPS–the activity in which the individual spent most of his time in the reference period of the last 365 days) and of the Secondary Status (SS–the activity defined in terms of some part of time spent in the reference year). Usual Principal and Secondary Status (UPSS) workers are then the principal workers as well as part-timers of various kinds. Clearly the number of SS workers could vary with changes in conditions affecting the supply of secondary workers as well as the demand for them. There is no way of judging from the numbers per se if any change observed is due to predominantly supply or predominantly demand conditions. This is particularly true of secondary women workers who form a substantial but varying part of the UPSS labor force reported by the NSS (Rustagi 2005). This important issue will be analyzed in detail in the next chapter.

Trends in employment by industry

Economic Development, in the history of both today's developed economies, and in the recent growth of developing economies, has been associated with a relative increase of employment away from the agricultural sector. This process is associated with an increase in labor productivity because the relative productivity is generally lowest in agriculture. Our first task, then, is to see how far India has been following this traditional pattern of transformation in recent decades.

Multiple occupations

We need to clarify at the outset the issue of multiple occupations in the Indian economy. How do we trace the changing structure of employment by industry or occupation when a significant number of households have members who pursue more than one occupation? There are in fact two distinct aspects of this issue. First, households would contain more than one earner of 'usual principal status' (UPS at the individual level). Second, a 'principal status' earner might have more than one activity. The first possibility creates a difference between the occupational or industrial classification of households (in terms of the activity of the 'main earner', defined as the main contributor to the household pot), and the occupations or industrial classification of individuals. The second point creates a distinction between the occupational classification of individuals based on the UPS and the UPSS status. The issue of occupational distribution by households is of importance when we are considering household income levels in different occupations. Since income (or expenditure) is available for the household as a whole we would need to define the occupation of the household by the activity of the main earner. We have done this in our work on the tertiary sector (Chapter 10). It is seen that the difference in changes in occupational classification over time by the household and the individual definitions is marginal. Here in this chapter we discuss the changes in the distribution for individuals only by the two alternatives of UPS and UPSS.

Trends in the industrial structure of individuals

It is apparent from the data in Table 3.1 that agriculture has indeed been shedding labor and it would appear that the process seems to have accelerated in the post-reform years of the 1990s. It is equally clear that the absorption of labor in manufacturing has been quite slow – even though it might have increased a bit in the nineties – and much of the increase in the labor force has been accounted for by the various types of tertiary activities, as well as construction.

We can infer from the discussion above that the decline in the share of the labor force in agriculture might have been exaggerated with the UPSS definitions because of the inclusion of secondary workers in the count. This category of workers might be disproportionately represented in agriculture which has a larger component of the self-employed. In so far as this reduction is partly due to supply-side changes affecting secondary workers or indeed due to fluctuations in the pace of technology spread in agriculture (see Chapter 7), the basic shift away from agriculture in the nineties might be overestimated. We therefore looked at the industrial distribution for labor defined on the UPS criterion. This might provide an estimate of the lower limit of the shift from agriculture.

A comparison of the two sets reveals that the UPSS definition does show a substantially larger decline in the incremental share of labor absorbed by agriculture. The UPS definition gives the decline in the incremental share in agriculture from 42.07 in the first period to 27.55 in the second. But even confining ourselves to principal workers, as the UPS definition does, the data confirm that there was a significant decline in labor absorption by agriculture in the nineties compared to the eighties. The gain of the tertiary sector in the incremental share under the alternative UPS definition was from 37.49 to 45.41.

Unemployment

What can we say about the level and trends in the rates of open unemployment in the Indian economy? The NSS data can be used to calculate unemployment rates based on either the CDS or UPS status of the labor force. The CDS estimates measure the rates of person-days which are being spent as 'not working but available for work', measured in half-day units over the reference week (see Appendix 2). These rates differ from the unemployment rates based on the UPS counts describing the 'usual status' of workers. (Note that subsidiary workers are by definition employed, still there can be estimate of UPSS rates of unemployment different from the UPS rates.) We can say that the CDS rates capture open underemployment during the week–as distinct from disguised unemployment on family farms or businesses (when some members of the household workforce are 'unproductively' employed but not declaring themselves available for work). The unemployment rate provided by the UPS measure will be necessarily less than the CDS rate since it measures the proportion of the labor force which is 'usually' unemployed during the major part of the year.

The rates of CDS unemployment in 1999–2000 was 7.2 percent in the rural areas and 7.7 percent in the urban. The UPS unemployment rates were, however, only 1.43 and 4.65 respectively (Government of India (2001), Tables 2.7 and 2.10(a)). It is apparent that underemployment, rather than round-the-week open unemployment is the real issue in rural areas. The difference, although significant, is less striking in the urban economy in the post-reform years between 1993 and 1999. It increased substantially in the rural sector–from 5.6 to 7.2. (The increase in the urban sector was hardly apparent–of the order of 0.3 percent.) A part of the increase could be attributed to the increased 'casualisation' of the labor force–the rise in the proportion of casual labor relative to the self-employed. It cannot be maintained that this is necessarily the result of deteriorating labor-market conditions. It might be partly the result of increased commercialization, as marginal farmers shifted more to wage labor. The income levels of the latter are often higher.

In any event the evidence suggests that even CDS unemployment is very much a problem of the youth, perhaps a result of waiting and job searching in the labor market. The unemployment rates for both sectors and for all rounds fall off sharply for age groups 30 and above. The increase of the unemployment rate between the 50th and the 55th rounds, after a fall between the 38th and the 50th in most age groups, perhaps does indicate a slight deterioration of labor-market conditions, but the phenomenon is one of lesser importance than other issues discussed in the book.

Trends in labor productivity by industry

The data given in the last section shows some movement outside agriculture – even though it has not been as fast as in many other Asian countries during the process of their economic transformation.

Another special feature of the changing employment structure in India has been the overwhelming importance of the tertiary sector in the absorption of labor outside agriculture. This at once raises the question: is the transformation of the employment structure – slow as it is – has really been of the type that has increased earnings of labor. A detailed examination of this point is attempted in the later part of this chapter and also in the chapters on individual major sectors in Part III. Here it is sufficient to note the relative mean productivity per worker in the major sectors and their changes over time – based on the figures given in the National Account estimates.

It is clear from the mean value of labor productivity that they are between 2.5 and 3.5 times higher in the manufacturing and tertiary sectors relative to those in agriculture – even if we take the more moderate estimates based on the UPS estimates (i.e., excluding the secondary workers who are relatively more abundant in agriculture), and even if we are looking at the less productive sub-sectors within tertiary activities (Table 3.2). Further, the productivity differential with respect to agriculture seems to have increased over time. This first cut at the data does strongly suggest that the movement of labor away from

agriculture – at a slow pace as it has been – has in fact in the direction of enhancing earnings of worker. We will see later in this chapter if this tentative conclusion is borne out by more direct evidence on wages and earnings levels.

Is India out of line with the experience of other Asian economies?

The exceptional nature of the absorption of labor moving out of agriculture in the tertiary rather than the secondary sector is seen to have been a feature of Indian development in recent decades. At the same time we have found that aggregate figures show that the relative income per worker in the tertiary sector is relatively high. Can we learn something from international experience if the Indian pattern of change is out of line with the observed pattern of development and, if so, in what way?

Papola (2005) discussed in detail the theory in the literature about sectoral shift of GDP and employment. Classical economists like Fisher and Clark explain the shift from industry to services by the changing demand patterns predicted by Engel's law. Fisher argued that services are 'luxuries' with more than a unitary elasticity of demand and so at a higher level of income increasing share of expenditure is absorbed by them and thus leads to high share of services in output and labor force. He assumed that the increase in the share of services in final demand proportionately lead to increase in the share of employment. However, Clark attributes the increase in the share of service employment

additionally to low relative productivity in services relative to manufacturing. Later economists like Bamoul and Fucho ascribed the rise in the share of service employment primarily to productivity differentials between industry and services resulting from technological, scale and geographical concentration of production in services. Further, increase in the share of service employment is also explained by the increased tendency of industry to outsource intermediate inputs used by industry to the service sector.

Popola refers to the experience of some Asian economies for comparison with India. The data for the shares of both employment and GDP and their change over the second half of the last century are given in Table 3.3.

It can be seen from the table that the share of workforce in industry increased along with is share of GDP in all countries including India, but it produced a much larger share of GDP in all other Asian developing countries other than India. It shows that the relative sectoral productivity of labor in India has been strikingly low by international comparison. By 2002 the tertiary sector in India contributed more than half the GDP in India but its contribution to employment was only 22 percent. It shows that service-sector growth has been productivity led but not employment led, contradicting views of some economists that employment grow in services because of low productivity vis-à-vis industry.2

The picture presented in Figure 3.1 of relative productivity in services vis-à-vis industry in the comparator Asian countries brings out the striking point that it is only in India – among all the countries represented – that the relative productivity in services has increased over the 40-year period. A second important point to note is that – with the exception of Thailand in 1960 when it had hardly any industry at all – the productivity in services exceeds that in industry only in India in both years, and that by a substantial percentage.

It shows that service-sector growth in India has been productivity led and not employment led, contradicting views of some economists that employment grew

Figure 3.1 Relative productivity in services and industry, various Asian countries 1960 – 2000.

in services because this sector has been a repository of low income labor 'pushed out' of agriculture. The heart of the employment problem in India would thus seem to be not an excess absorption of labor in the tertiary sector, but the relatively low productivity of the manufacturing sector, and its persistence over time. It is this low performance of manufacturing which has prevented it from being the dynamic sector playing a central role in productivity growth as well as the reallocation of labor as in other countries in the history of successful economic development.

How much of this productivity differential in favor of the tertiary sector is due to the recent developments of the information technology sector? The answer would appear to be not very much. For one thing the productivity differential in favor of the tertiary sector was substantial even in 1960 when the IT sector was non-existent. Second, in terms of the numbers employed the tertiary sub-sector dealing with IT is quite small even in recent years. Table 3.4 gives an estimate of employment in this sector based on enterprise surveys, and Table 3.5 provides the estimate from the household surveys of the NSS. The Manufacturing sub-sector includes hardware, central processing units (CPUs), communications equipment, electronic components and industrial control and supervision equipment manufacturing (not including medical equipment). The tertiary segment includes telecommunications services, computer and related services (IT and software), research and development services and also start-up companies.

The estimates show that the total employment in the IT tertiary sector is of the order of 400,000 to 600,000 (Tables 3.4 and 3.5). Considering that the total employment in the tertiary sector (in the UPS count of the NSS) was around

150 million, the percentage of tertiary employment in the IT sector was at best 0.4 per cent.

It is necessary to turn our attention to the denominator of the ratio and consider the possible reasons for the low labor productivity of the manufacturing vis-à-vis the tertiary sector in India.

Table 3.6 draws attention to the 'dualism' that exists in Indian manufacturing. The household enterprises (not employing any hired labor) contribute more than half of manufacturing employment whereas establishments with 500 and above employees contribute more than two-fifths of gross value added but employ less than one-tenth of employment. Consequently there is a tremendous difference in relative labor productivity between these two size groups and it is this which leads to very low level of labor productivity in the manufacturing sector. Such a situation does not exist in other developing countries in Asia, as will become clear from the evidence presented in Chapter 9. Unless there is substantial growth of small (10 – 100 employees) and medium (100 – 500 employees) that are relatively labor-intensive and have substantially higher labor productivity than household enterprises leading to substantial increase in the share of manufacturing in GDP with some increase in employment share in the Indian economy, we

are unlikely to follow the sectoral pattern of growth as other countries experienced in the development process.

Further discussion of this important issue will be found in Chapters 8 and 9 in Part III of this work.

Employment in the organized sector

It might be useful at this point to put the size of the formal or organized sub-sector in manufacturing in the context of total employment in the formal sector. In Chapter 10 we will examine the formal – informal distinction within the tertiary sector in detail. But for the present purposes the official estimates of different types of employment within the formal sector put out by the Ministry of Finance of GOI will suffice. These are given in Table 3.7. The stagnation of manufacturing in the formal sector is apparent from this table, as is the relatively small share of manufacturing in total formal sector employment. The total including all sectors is itself very small in 2001 – only about 7 percent of all employment. The public sector still dominates the scene in formal employment in spite of India having embarked on a process of encouragement of the private sector since the early 1980s.

Patterns of urbanization and the quality of employment

A feature of economic growth has been the increasing absorption of labor in the urban sector. The rate of urbanization has been slow in India – consistent with the slow transformation of the employment structure. There has been some concern in the literature if the reallocation of labor to higher-quality jobs in non-agriculture has been disproportionately achieved only in the urban economy. We can also refer at this point to the finding in Chapter 5 that there have been important changes in recent decades in the size structure of towns, with a redistribution of population to smaller towns. How does in the change in the industrial structure of employment differ between small and large towns – as well as between rural and urban areas? To throw light on this question we present in Table 3.8 the way the incremental flow of labor in each of the three broad sectors was distributed between the rural, and the three classes of towns. The data are presented separately for the 1980s (1983 to 1993/1994) and the nineties (1993/1994 to 1999/2000), but the classification by size of town is not available for the earlier period. Of particular interest is the relative importance of the flows of new employment in the secondary and the tertiary sectors. The importance of the secondary sector even in the most recent period is higher in the urban areas as whole, and in large towns. In 1990 – 2000 the share of manufacturing was 27 percent in large towns compared to 19 percent in small towns and only 7 percent in the rural areas. But it is apparent from the figures on incremental flows in Table 3.8 that a redistribution of employment in the secondary sector has been taking place in the recent period in favor of small towns, and also the shares of the rural and the urban sectors in the new employment has been almost the same. The importance of the small towns in the tertiary sector has, however, been increasing faster. The small towns have clearly witnessed a substantial swing away from employment in the primary sector. The redistribution of employment to small towns, which has been noticed, has been driven by non-agriculture.

Expansion of education and the quality of employment

The expansion of employment outside agriculture – and the concomitant upgrading of jobs – is closely related to the expansion of education. It has been maintained that the lopsided development of education outside the rural sector has in fact hindered the diversion of employment in the rural economy (Chadha and Sahu 2002).

Table 3.9 in particular brings out the point that it is at the education level 'graduate and above' that the urban economy plays an overwhelming role in attracting the educated. But it is also of great importance to note that the major proportionate shifts in the additional flows of educated labor are to be observed in the smaller towns in the post-reform period. These towns have been able to attract a large proportion of educated labor – with secondary as well as college qualification – in major way in the 1993 – 1999 period. This is another interesting

part of the increasing role played by the smaller towns in recent years – which had already been noticed in Chapter 2.

India has made rapid progress in upgrading the quality of its labor force. The number of workers with less than five years of education has come down steeply from 80 percent in 1983 to 65.5 in 1999 – 2000. But even then it is significantly behind most of the rapidly developing countries in Asia. The average years of schooling for the population aged 25 and over in China around 2000 were 5.7, and in East Asia 6.5 compared with 3.6 in India. The proportions with no schooling were 20.9, 22.8 and 44.5 respectively. Equally damaging is the low proportion of those with secondary schooling – known to be a critical group in the development of manufacturing and other modern-sector activities. In India it is only 17 percent at this date, much lower than India's income level would predict. It is only half that of China, and the proportion is worse for females (World Bank 2007, Table 1.3). It will be suggested in Chapter 9 that its relative neglect of primary and post-primary education in earlier years might have been a major cause of the persistence of 'dualism' and the slow growth of the dynamic manufacturing sector in India.

Trends in wages and wage inequality

Wages of casual and regular workers

The wage sector in India is substantial – even in the rural areas. Regular workers (those with a more permanent contract for varying periods of time) are more important in the urban areas, and casual wage workers (those hired on day-today contract as work is available) are in a majority in the rural economy.

Regular workers have several days of work during the week – the NSS data show that the average is between 5.6 and 5.9. Casual workers get work for fewer days of the week – generally less than four. Part of the difference in the earnings per worker between the two categories, therefore, reflects the difference in the number of days of work secured in the week of enumeration. For casual workers the seasonal element is likely to be of great importance. When making comparison between the two groups – which could be used to reflect an aspect of the formal – informal dichotomy in the labor market – it is important to be clear as to the objective if the comparison: are we interested in the levels of income of the two classes or in the returns to a unit of work?

Average wages (earnings) per day

The NSS collected data on the earnings of the workers for the preceding week (seven days) of the survey, and it also recorded in the same field the number of person-days the worker was actually at work. These data give us the earnings per day for both casual and regular workers. The casual – regular wage difference varies by rural – urban location, by gender and also by occupation (i.e., manual or non-manual). In the rural areas there were about ten million casual workers according to the 55th round of the NSS (Census adjusted 15 – 59 age group) and only two million regular. The corresponding figures for the urban sector were 1.4 and 2.5 million. Females were a third of the total in the rural casual labor market, but in all other segments, in the regular category and in urban areas, the representation of females is much smaller – of the order of 15 – 20 per cent. It is to be noted that not all regular workers were classified as non-manual. In fact both in the rural and the urban sectors almost half of the regular were manual workers. On the other hand, only 10 – 15 per cent of the casual workers were non-manual.

We examined the data from the NSS showing the difference in mean wages per day for different categories of workers in the 15 – 59 age group for the 55th round 1999 – 2000. For casual workers the manual wage rates are close to the non-manual, for both sexes, in both the rural and the urban areas. On the other hand, the difference between the two categories of workers for the regular wage earners is huge (between twice and three times as high). It shows the importance of human capital attainments in the determination of regular wages.

It is important to note, however, that even for manual workers alone, regulars earn nearly double the amount of casuals – except for females in rural areas, where the differential is more like 50 per cent higher. Since regular workers get a significantly larger number of days of work in the week (also get paid for the whole week), the difference in earnings would be even higher. While a part of this difference – even for manual workers – might reflect measured human capital attributes, a good deal of the differential really pertains to the formal-informal sector dichotomy in the labor market. This differential is partly due to institutional factors (employment in large establishments, or in the public sector) and partly due to the operation of the wage-efficiency relationship for 'established' workers with low turnover.

Distribution of wages

The data on wages from the NSS have been analyzed by Puja Vasudeva-Dutta (2004). Vasudeva noted that the dispersion in wages among casual workers is much smaller than among regular wages. This is confirmed by the graphs in Figure 3.2, which also suggests that the dispersion seems to have increased over time for regular wage workers, but not for the casual. A major reason for the difference is, of course, is that regular wage workers have a much greater variance in human capital attributes, particularly education. There is a big difference between manual and non-manual wage difference for regular workers, but not for the casual, reflecting the dispersion by skill and education for the former category.

Growth rate of wage rates

The figures 3.3a and 3.3b give a picture of the growth rates of real wage rates for different categories of labor namely rural male (RM), rural female (RF), urban male (UM) and urban female (UF).

Figure 3.2 KDF distribution for regular and casual workers for different NSS rounds.

Figure 3.3a Growth rate of wage of regular non-manual wage earners.

The wages of regular non-manual workers in the second period increased at around twice the rate of the pre-liberalization era, but there was little change in the growth rates of casual manual workers. This is an aspect of the increase in inequality in the labor market in the nineties.

Trends in wage inequality

It can be inferred from Figure 3.2 that there is strong suggestion from the KDF` graphs that wage inequality is higher for regular workers and has increased over

Figure 3.3b Growth rate of wage of casual manual wage earners.

time. Vasudeva's summary measures for wage inequality indicated that, for regular workers, GE(0) went up from 0.286 to 0.337 between 1983 and 1999, and GE(2) increased from 0.381 in 1983 to 0.430 in 1999. The level of inequality was much less for casual workers and it also declined over time. The values of GE(0) were 0.143 in 1983 and 0.117 in 1999.

It is interesting to note that while inequality among regular workers increased significantly between 1993 and 1999 – and more so at the upper end as evident from the larger increase in the GE(2) measure – the 'between group inequality' for educational groups did not change by all that much. Much the more important part of the inequality increase was accounted for by the 'within group' component. This is in line with the evidence from other countries which have experienced increase in wage inequality in the globalization era. While returns to formal education do increase, it is the differential valuation of the individual worker's non-formal attributes which seem to be more important in the increase in inequality.

Vasudeva has used the regression-based methodology of Fields to study the 'factor inequality shares' of different explanatory variables in the earnings functions estimated separately for the regular and the casual workers (Field 2000). The 'factor inequality share' gives a quantitative estimate of the total inequality in the dependant variable (in this case 'wage earnings') explained by the different explanatory variables in the earnings function.

A semi-logarithmic Mincerian (standard or augmented) wage determining function can be written as:

where a =[β1.....βj, 1] and Z = [Z1.....Zj, ε] are vectors of coefficients and explanatory variables respectively. An inequality index I can be defined on the vector of wages (w). Applying Shorrocks' theorem the relative factor inequality weights (i.e., the percentage of inequality that is accounted for by the jth factor) can be calculated as follows:

where cov[.] denotes the covariance, cor(.) the correlation coefficient and σ(.) the standard deviation.

The major results from Vasudeva's exercise can be summarized as follows:

1 As far as regular workers are concerned just over half of the variance in the log of wages are explained by the earnings function. The same variables explain much less – a third – of the variance for casual workers.

2 In terms of the explained part of the variance, human capital variables were most important for regular workers. Age accounted for about a quarter and education a third of the explained variance in 1999. The other important factor in line was industry affiliation – contributing another quarter.

3 By contrast, human-capital factors were of much less importance for casual workers – only age, and not education having any positive contribution, but at a much lower level of around 7 percent. The single most important explanatory variable was geographical difference – the state of residence contributing no less than 62 percent for casuals as against only 3.5 percent for regulars.

4 Although for regular workers the wage gap between those with graduate and primary-school qualifications increased between 1983 and 1999 (see 'Rural – urban differences' section below), the share of education in the explanation of the variance declined from 23 to 17 percent. The importance of age increased as did that of industry affiliation. Further, Vasudeva confirms that the increase in the 'contribution of selection coupled with the fall in that of education suggest a rising importance of unobservable for regular workers, possibly linked to the process of trade liberalization'.

Inequality in household welfare

Although substantial wage employment in India is still only a part of total employment, and a good deal of households are outside the wage-sector – mostly self-employed. We would want to know if the experience of non-wage households mirrors that of the wage earners. This section therefore looks at the trends of welfare of all households irrespective of the type of employment. We choose as our measure the average (mean) per capita expenditure (APCE) of the households as recorded by the NSS of successive rounds. Figure 3.4 portrays the movement of the KDF distribution of APCE over time, separately for the rural and the urban areas.

It is apparent that while the modes of the distribution have shifted outward in both sectors, but more so in the urban sector, there has been a more pronounced

Figure 3.4a KDF distribution of APCE, Rural (poverty line: Rs.196.50, at 1993 – 1994 = 100).

Figure 3.4b KDF distribution of APCE, Urban (poverty line: 227.20, at 1993 – 1994 = 100).

'flattening' of the distribution in the urban sector signifying an increased degree of inequality.

Table 3.10 gives the measures for overall inequality. Note the large values for GE(2) which is more sensitive to high incomes. While this measure has decreased substantially in the rural areas it has increased in the urban.

Rural – urban differences

The discussion in the last section has suggested that inequality has increased more strongly in the urban economy, at least in the post-reform era. Thus the disparity in household welfare between the two sectors has increased, we now look a bit more intensively at the rural – urban difference. Has the disparity increased more for some groups rather than others? Can we isolate more concretely the factors responsible for it?

Figure 3.5 brings out clearly the point that the relative difference in household welfare has increased for higher expenditure groups. We can compute the 'Blinder – Oaxaca' decomposition of mean outcome differential between the rural and the urban sectors, The difference between two groups can be decomposed into three parts: i) due to differences in endowment (E); ii) due to differences in coefficients including the intercept (C); and iii) due to interaction between

Figure 3.5 Urban – rural difference in APCE by percentile.

coefficient and endowment (CE). Depending on the model that is assumed to be 'true' model (absence of discrimination), the three-fold decomposition can be used to determine the explained (Q) and unexplained (U). By using the low group (rural APCE) as the no-discrimination base we calculated Q = E and U = C + CE.

Table 3.11 summarize the results for the 'Oaxaca decomposition' for APCE between the rural and urban areas. The calculations show that there has been a substantial increase in the 'discrimination' factor for urban households in the post-reform years between the 50th and 55th rounds. The increase in the rural – urban disparity is due not to the better endowments of the urban workers but to the higher returns to the human-capital factors secured by them in the urban economy.

Returns to education

The results discussed above suggest that educational developments have been a major player in the increase in inequality and in the growing rural – urban disparity. The increments to income from successive levels of education could be approximated by the difference in co-efficient to the education dummies in a fitted earnings functions for regular wage earners. These are reported in Table 3.12 and graphed in Figures 3.6a and 3.6b separately for the rural and the urban areas.

The difference between rural and urban economies is brought out dramatically in Figures 3.6a and 3.6b. The lift to the returns to education in the post-reform years occurs at different levels of education in the two areas. In the rural economy the sharp increase occurs at the level of secondary education, while in the urban sector the lift is observed at the college graduate level. The curves for the successive rounds, however, intersect at lower education levels in both sectors, showing that at levels less than middle, the returns to education are in fact depressed for the later years. They are nearly at the same level for middle-school leavers in the rural sector, and for secondary-school leavers in the urban. All this is consistent with

expectations about what might happen with the increase in the supply of educated labor. Demand outstrips supply in the post-reform period in the rural areas for secondary-school leavers, and for college graduates in the urban.

Analysis of the returns to education by age-groups revealed an interesting finding about the urban labor market. For the 20 – 29 group the marginal returns to secondary education actually fell in the 50th and the 55th rounds while those for college graduates showed a sharp upward movement. By contrast for the older 30 – 39 group there was a milder increase for both the secondary and college graduates in both the 50th and the 55th rounds. It is clear that the demand for the more educated has been soaring in recent years and has affected the new entrants to the urban labor market more strongly.

The literature has drawn attention to the increased demand for more educated labor in the era of globalization in a number of countries and has stressed the importance of skill-intensive technical change in manufacturing in particular,

Figure 3.6a Private return to different levels of education (urban).

Note
1, 2, 3, 4 and 5 denote the levels of education namely literate, primary, middle, secondary and graduate respectively.

Figure 3.6b Private return to different levels of education (rural).

Note
1, 2, 3, 4 and 5 denote the levels of education namely literate, primary, middle, secondary and graduate respectively.

Figure 3.7 Returns to education in urban areas by age-groups.

and/or the importance of more skill-intensive manufactured goods in international trade (see Chapter 1 above). But we have seen that the employment expansion in post-reform India has been concentrated not so much in skill-intensive manufacturing as in the tertiary sector. It is therefore useful to ask the question: is the expansion of demand for educated labor which we witness particularly in the market for college graduates in the urban areas originating mostly in the tertiary sector? Table 3.13 gives the distribution of the addition to the UPS workforce by industry for different levels of educational attainments.

It is apparent that the market for college graduates in particular has expanded relatively more in the tertiary sector. A somewhat unexpected finding is that this trend had been going on since the eighties. Equally revealing is the finding that labor with less than middle level of schooling is now almost entirely absorbed in the primary sector. The difference between NO_ED and BASE_ED education stressed in the Introduction to the book would seem to be drawn at the boundary of primary education in the Indian labor market in the late nineties. Entry into the non-primary sector would now seem to require post-primary education.

We will discuss in Chapter 10 in particular that the bias in Indian policies towards tertiary education has encouraged the growth of skill-intensive industries. It is seen from the evidence presented here that the demand for labor with college education seems to be outrunning the supply with this pattern of development – even with the historical bias in education policies. It is very likely that the return to college education has continued to increase in the years since 1999–2000.

Appendix

Employment estimates based on current daily status (CDS)

Current daily status of all individuals above the age of five is coded in the NSS for each half-day over the seven days preceding the survey. The activity of each half-day could be classified as (i) employed (ii) unemployed or (iii) out of labor force. Even if an individual is not classified as unemployed under the usual status in the week, some half-days of unemployment are possible if they are available for work for those units of time. Thus apart from the 'usual' unemployed the unemployment days would be contributed by casual workers; the self-employed who are working generally; and even by those 'usually' outside the labor force. The CDS unemployment rate is calculated by adding up these person-days of unemployment as a proportion of all days of employment plus unemployment.

There has been a school of thought in the Planning Commission and other GOI circles which has favored the use of CDS for the estimates of employment. This generally produces estimates which are much lower for 1999 – 2000 than those calculated on the UPS or UPSS basis. Thus the growth rate of employment shows a significantly higher rate of decline than the other estimates. For example, Srinivavsan (2005, Table 4) gives the growth rate of employment in the 1987 – 1999 period (between the 43rd and 55th rounds) in the rural areas at –4.59 percent on the CDS basis against –1.48 percent on the UPSS basis. But Srinivasan comments:

The total number of person-days of employment is not the same as the total number of employed persons. The reason is that a given total number of person-days of employment could be distributed among the same number of persons in many ways so as to lead to different numbers of persons employed. For example, consider a four person economy in which all four participate in the work force and together they were employed for ten person-days in the week. This yields a person-day rate of employment of 10 out of 28 or 36%. If the ten person-days are distributed in a way that one person is employed for seven days, another for three days and the remaining two are unemployed, then person-rate of employment is two out of four or 50%. On the other hand, if it is distributed in a way that three persons work for three days each and one person works for just a day, the person rate of employment is four out of four or 100%, given the priority given to the status of employment!

We know that the NSS estimates of the numbers of persons in different demographic groups are underestimated, so we have to get the population figures from the Census counts. It is inappropriate to apply the employment rates based on person-days to the count of persons obtained from the Census of Population to arrive at the total number of employed persons.

4 Accounting for the decline in labor supply in the 1990s

It has already been mentioned in the last chapter that the Task Force of the Planning Commission on Employment Opportunities (July 2001) had in its Report (popularly known as the 'Ahluwalia Report') focused attention on the sharp decline in employment growth. The Committee rightly pointed out that while a part of the slowdown in employment growth was due to an increase in the rate of unemployment, much the more important part of the decline has to be ascribed to a slow down in the growth of the labor force. This is largely because of a fall in the participation rates (PRs) as measured by the NSS.

The Committee did not take a firm position on the reasons for the fall in participation rates, but the arguments presented included non-economic factors. e.g., the expected shift in the activity status of the younger age group towards education; increase in the share of the population aged 60 and over; a reversal or 'correction' of the increase in PRs of certain age groups recorded in the 1993–1994 NSS (p. 45). At the same time its assertion that the decline in the PRs in the prime age group is 'within the margin of sampling errors', while partly applicable to males, is, as we shall see, certainly not true for females.

A follow-up Report by a 'Special Group' of the Planning Commission (May 2002) firmly put its emphasis on a slowdown in the growth of demand for labor as the culprit on the observed trends in labor force and employment growth. Noting that the fall in employment growth was accompanied by a higher rate of GDP growth, the Committee concluded: 'It means that the capacity for job creation per unit of output went down about three times compared with that in the 1980s and the early 90s' (ibid., p. 336). It suggested that the nature of economic growth had become more capital-intensive, both due to structural changes, and the 'rightsizing' of labor use.

In this chapter we shall go into some detail into the reasons for the fall in participation rates which is the main cause of the observed decline in employment rate. Our major purpose would be to see if we can confirm in any conclusive way if this is the result more of demand-side developments (as implied by the 'Special Group' Report).

The theoretical perspective

The measured labor force at any point of time (and the volume of employment) is determined by the equation of the demand and supply functions of labor. If the observed magnitude changes over time it might be because of the shift of the demand or the supply functions or by a combination of the shifts in both. Thus if the labor force or employment growth falls over time we cannot conclude that it is because of a fall in the demand for labor. It might easily be caused more by a fall in the supply of labor. We cannot measure the contribution of supply and demand factors to the observed slowdown without a fully specified and estimated model. While this might be difficult to achieve we can at least infer the relative importance of the two sets of factors qualitatively by looking at trends in wage levels over the time periods covered. This can be illustrated by Figure 4.1.

The Figure 4.1 shows the supply-and-demand framework for wage determination in a dynamic setting. The x-axis measures participation rate while the y-axis measures the wage per worker. The participation rate in effect measures the flow of labor per unit of time out of the potential stock. The demand function is then downward sloping: the lower the wage level the more is the flow of labor sought by employers from the existing stock. The upward sloping supply function is propelled by the 'substitution effect' of changes in wage levels. A higher wage will see a higher rate of allocation of time to the labor market at the expense of other activities, like leisure, household work or education.

As we have seen the post-reform period in India saw a significant fall in the rate of growth of employment compared to the previous decade. It will be shown in the next section that this decline was the counterpart of a decline in the participation rate In the competitive labor-market framework presented in Figure 4.1, this must imply that, if this decline has been caused by a shift in the demand function with a relatively unchanged supply function, the wage per worker must fall in the second period (situation 2 in Figure 4.1). We have already seen in the

Figure 4.1 Wage-determination framework.

last chapter that this has not been the case. The real wage has increased in the post-reform period. This outcome is only possible if the supply function has shifted upward. Such a shift is possible if the 'income effect' of higher welfare levels causes household members to supply less labor to the market for a given wage growth. An increasing growth rate of wages could indeed both cause and sustain such an increase in household income levels.

In the empirical work presented below we shall be investigating the possibility of such an income effect for different demographic groups and sectors.

It is of course possible that the wage determination in the Indian economy cannot be interpreted within the competitive labor-market framework in the post-reform years. The higher rate of wages might have been the result of institutional factors. In this case the rate of employment growth (and consequently participation) could fall in response to the higher wage growth, and the gap between the growth of labor supply and of labor demand would have opened up leading to a higher rate of unemployment (situation 3 in the Figure 4.1). In fact the second Report of the Planning Commission hinted at this possibility without actually analyzing it in detail. There are two objections to this alternative hypothesis. The rate of unemployment did increase in the post-reform years, but we shall see in the discussion below that the magnitude of this increase was not really large enough to account for the large fall in the growth rate of employment. Second, and more crucially, it is hard to believe that institutional determination of the higher growth rate of wages would be at all realistic in an economy in which the bulk of employment is in the 'unorganized' sector.

There is a third possibility which is suggested by the 'segmented labor market' model. Segmentation implies that labor markets in the different segments are subject to different sets of supply-and-demand factors. In this case falling demand in one segment causing reduced participation might coexist with increased demand in other segments putting an upward pressure on wages. If, then, the observed data refer to participation in the first segment and wages in the second, we might indeed have the apparently inconsistent picture referred to above with slackening demand coexisting with rising wages in the aggregate. Segmentation might be caused by many factors and depending on the focus of our analysis could refer to gender, formal-informal, caste and many others. In the context of the current issue the factor we should emphasize is the distinction between primary and secondary earners. In the methodology of the NSS the former is called Usual Principal Status (UPS) and the latter Subsidiary Status (SS). This distinction is important in the context of the issue being discussed, because as we shall see a substantial part of the fluctuation in participation has taken place in the market for SS workers, while the wage or earnings data which are reported are for the markets of UPS workers.

The SS labor market is dominated by females, the majority in self-employed status, who are generally part of agricultural households and divide their time between household activity and work on the family farm as required. The amount of time spent on the latter is not measured by the NSS which only records the number such workers during the period of the survey. During an upsurge in economic activity, as might have happened during the introduction of new technology in paddy in the early eighties, the increase in demand for labor might be partly met by an increase in workers of secondary status working for varying lengths of time. Over a period of time as the labor market settles down to the new level of activity if the technical change is of a lasting kind, a new deployment of labor would normally be worked out by farmers in which some of the secondary workers used would be replaced by UPS workers on more permanent basis. Since the time spent by the latter on economic work is normally much larger than that by secondary workers, the numbers recorded as employed by the total (including both UPS and USS workers) might actually fall although the number of UPS workers might have increased. Thus the total employment figure might give a misleading idea of the change in demand in the labor market. We have to look specifically to the numbers in UPS status to see how the employment change in relate the recorded change in wage levels.

We will follow this line of analysis in the empirical work reported below.

It might be objected that in the discussion above we might have overemphasized the importance of wage workers in the Indian labor markets. In so far there is a large presence of self-employed workers, an analysis focused on wage trends might be misleading.

How important is wage labor in the Indian labor market?

The share of wage workers in UPS (principal) employment is more than half–slightly less in rural areas. Some part of wage labor is, however, supplied by secondary workers. It is important to get an idea about what proportion of the UPSS employed (principal and subsidiary together) actually participate in the labor market as wage labor either as a first or a second job. This total gives us the proportion of the employed who actually respond to wage signals in the labor market directly. Looking at the (larger) UPSS labor force as a whole, it was seen that the proportion of wage workers goes up (relative to the UPS count) by 3.5 percent in rural areas but it marginally declines in the urban areas. Evidently in rural areas more of the total enter the labor market as wage workers in subsidiary employment. In the urban areas relatively more workers in secondary status are to be found in the non-wage employment segment.

Decline in labor-force growth and distribution of fall in LFPR among different demographic groups

The fall in the rate of growth of employment in the post-reform period can be shown to be basically due to the fall in the participation rate. It has already been pointed out in the last chapter that the rate of unemployment in the Indian labor market is low, either with or without the secondary workers. Even if the rates of unemployment might have gone up in the post-reform era a bit its overall impact on the employment growth is small.

Our discussion in this section, therefore, concentrates on the changes in the participation rates, and the possible reasons for these changes.

We first set out the change in the rates of growth of the total (UPSS) labor force between the different NSS rounds.

Table 4.1 suggests that there was substantial decline in the labor-force growth from the 1980s (38th to 50th round or 1983 to 1993–1994) to the 1990s (50th to 55th round or 1993–1994 to 1999–2000) from 1.71 percent to 1.23 percent. The decline in the labor force could be due to decline in the working-age population or due to the decline in the labor-force participation rate (LFPR). Keeping the LFPR of 1999–2000 the same as 1993–1994 we find that hypothetical labor-force growth would have been much higher at 2.16 percent–much higher that even in the 1980s. It clearly shows that it is the decline in the LFPR that is mainly responsible for slower growth of employment in the 1990s.

We examined separate graphs for the age-specific LFPRs for all the three rounds for rural male, rural female, urban male and urban female.

The overall impression from the four profiles of age-specific PRs is that the most important change in the post-reform period is the decline in female participation, particularly in the rural economy. As far as males are concerned, the profiles for both the rural and the urban areas showed marginal declines in LFPR in age groups 5–19 years and 59+ age groups in 1990s. The decline in male LFPR in 5–19 years was substantial in the 1980s but slowed down in the 1990s–in both rural and urban areas. By contrast, the female LFPR showed a decline in all age groups between 1993–1994 and 1999–2000. In the previous period between 1983 and 1993–1994 the decline in both rural and urban areas were sharper in 5–19 age groups, but there was no substantial changes in the older age groups. The distinctive change in the latest period is the marked decline in female LFPR in the working age group as well.

As a substantial proportion of females participate in the labor market in a subsidiary capacity (i.e., enter the labor market only for a part of the year) it is worthwhile to look at female LFPR separately for UPS (Principal) and SS (Subsidiary) categories. This is portrayed in Figures 4.2a to 4.2d.

Figure 4.2a Rural female UPS labor-force participation rate.

Figure 4.2b Urban female UPS labor-force participation rate.

As in the case of males, the UPS female LFPR in all areas showed marginal decline only in the 5–19 age group in the 1990s most of the decline having taken place in the earlier decade. In the urban areas a marginal decline was observed in other age groups as well. But the real dramatic changes seem to have taken place in the category of subsidiary female labor. The LFPR graphs in both rural and urban areas showed shift inwards in the 1990s–signifying a decline in LFPR in all age groups. This contrasts strongly with the movement in the previous period in the 1980s, the adult age groups of 25–49 years showing a substantial increase in LFPR in both rural and urban areas. The patterns and the nature of shift of the LFPR graphs suggest that there was an upsurge in female subsidiary labor demand in the period between 38th and 50th rounds and females belonging to

Figure 4.2c Rural female subsidiary labor-force participation rate.

Figure 4.2d Urban female subsidiary labor-force participation rate.

the age group 25–49 were in the best position to respond to it. However, between the 50th and the 55th rounds there seems to have been a substantial decline in the demand for subsidiary workers and it is reflected in the inward movement of the entire LFPR curve for female subsidiaries (It is to be remembered that for the subsidiary labor force there is no unemployment, all are employed.) This is an important point–which is valid both for the rural and the urban sectors, but quantitatively more so for the rural.

Accounting for the decline in the labor force in the nineties

Three significant points emerged from the discussion in the last section: First, the decline in the 5–19 age group of the primary labor force (UPS); Second, a

marginal decline in the 59+ UPS labor; last, a substantial decline in female subsidiary labor supply.

The question is what is the relative contribution of each of these segments to the total decline in labor supply in the 1990s? Table 4.2 throws light on this question. It presents the actual (A) and hypothetical (D) labor force, the latter on the assumption of no change in age-group specific LFPRs between the 50th and 55th rounds.1

The results show that decline in LFPR actually contributed to a fall of 23 million in the labor force. The last column shows that female subsidiary labor force (of all ages) contributed 47 percent of the total decline. The 5–19 UPS LFPR contributed 30 percent and aged LFPR (59+) contributed another 10 percent. These three factors combined accounted for as much as 88 percent of the total hypothetical labor force decline in the 1990s. Clearly the first component–the withdrawal of subsidiary females from the labor force–leads the list in terms of the diagnosis of the observed fall in employment growth. But before coming to this topic, we discuss briefly the fall in LFPR in the 5–19 and 59+ age groups.

5–19 UPS labor-force participation rate

The distribution of all persons in this age group in different principal activities is given in Table 4.3.

One can clearly see that the increase in proportion of students is the main factor responsible for decline in the work participation rate in both the rural and the urban sectors. There is a difference between the two periods: between 1983 and 1993–1994 and the subsequent period of the 1990s. The earlier period experienced substantial jump in the category of students but the shift was relatively more from the category of 'nowhere children' (doing nothing). In the latter period, it was largely a shift from UPS worker and domestic work. However, the withdrawal from work in absolute term was higher in the 1980s compared with the 1990s. We conclude that it is either demand for education and/or better educational facilities

that seem to be the prime reason for decline in labor-force participation in this age group. However, increase in the students' participation rate in this age group seems to be tapering off in the urban areas in the 1990s (Table 4.3).

We examined the relationship of the participation of this demographic group to the levels of the expenditure per capita of the households to which they belonged. The tabulation was done separately for the rural and the urban households. Participation was found to be negatively related to household expenditure levels in both sectors, but the negative relationship was stronger in the urban areas. The slope of the negative relationship clearly decreased over time, but was still strong in the 1999–2000 survey, particularly in the urban sector.

Aged 59+ labor-force participation rate (LFPR)

The post-liberalization years saw a reversal in participation trends in the rural sector for both males and females, but not in the urban areas. Participation rates fell in contrast to the positive trend in the previous decade. In the urban sector the relatively slow decline in participation already noticed in the previous decade continued at more or less the same pace. Overall the share of the post-retirement workers in the total withdrawal from the labor force in the 1993–2000 years is small.

In the 59+ age group it is generally argued that income effect predominates when withdrawal from labor force is observed over time. Only in the case of urban females do we see a consistent decline in LFPR over the three rounds and also for the top three quintiles. The trends are much more mixed for the other categories, though for both rural and urban males the decline in participation in the 1990s seems to have been strongest in the top two quintiles.

It should be emphasized that this result, although suggestive, cannot be conclusive about the income effect on the participation of seniors. This is because the earnings of the seniors staying on in market activity itself affect the expenditure level of their household.

Female subsidiary labor force

Change in the demand for subsidiary labor

As argued earlier, there was a sudden upsurge of female subsidiary labor demand in the 1980s followed by a contraction of labor demand in the 1990s. This fluctuation requires further probe. Tables 4.4a to 4.4b present the distribution of subsidiary female employment by major industries and occupations.

As we can see subsidiary female employment is concentrated in a few agricultural and allied activities, i.e., growing of cereals and animal husbandry. They contributed 80 percent of female subsidiary employment in rural areas and even a sizeable part of such employment in the urban labor market.

We, therefore, looked specifically at the growing of cereals sector in rural areas since it constituted the largest chunk of female subsidiary labor supply. This sector contributed 4.9 million out of six million of additional subsidiary female employment between the 38th and 50th rounds. At the same time it contributed

5.8 million out of 6.3 million of decline in female subsidiary employment between the 50th and 55th rounds.

However, analysis across states for this sector show that four states Karnataka, Madhya Pradesh, Maharashtra and West Bengal2 played the major role in absorption of female subsidiary labor in this sector in the 1980s and subsequent decline in the 1990s. These states include the districts which experienced a spread of the green revolution in the 1980s. Thus this piece of evidence gives credence to the argument that the spread of labor absorbing green revolution technologies in the 1980s bumped up the demand for labor in these areas, which in the short run could only be met by increasing use of female secondary labor. Subsequently in the 1990s as rising labor costs led to the introduction of labor replacing technologies, the additional demand for female labor subsided. This process might have been also helped by hurdles faced by oilseed development program in the 1990s in parts of this region.

It should be noted that most of this adjustment took place, not in the wage-labor market, but among the self-employed workers. It can be seen from Table 4.4a that agricultural wage labor accounted for a small portion of total female subsidiary employment in the rural areas, and even less so in the urban sector. Further examination of the NSS data by employment status showed that casual wage labor accounted for only 16.9 percent of all female subsidiary workers (in the age group 15 and above) in the rural areas in the 55th round–down from 21.7 percent in the 50th. The corresponding percentages in the urban sector were 12.1 and 20.6. Much the more substantial share of such employment in both sectors was accounted for by the 'self-employed' and the 'helpers' categories.

Evidence on the 'income effect'

If there is a significant income effect affecting participation then we would expect the opportunity cost of leisure to increase at all household welfare (or income) levels–but it would presumably increase more at higher levels of welfare. Consider a supply function of labor-relating participation rate (of, say, prime-age females) to the household welfare level. At any point of time, for a given distribution of household incomes, we would expect this curve to turn down quite sharply as the effect of higher household welfare begins to overshadow the substitution effect. When at a later date average income of all household increases, the supply curve relating participation of this group to the household welfare level is pushed downwards. Thus there is less participation at all welfare levels–but the point at which there is a significant fall in the slope of the curve comes earlier in the lower part of the household welfare distribution. We have seen that a very important portion of the change in participation in the post-liberalization period is accounted for by the fall in the number of females of subsidiary status. We can try to see which household welfare groups have typically contributed to the withdrawal of labor in the female subsidiary status. The index for household welfare used is the mean per capita expenditure level. We use the groupings as provided by the NSS reports.

Table 4.5 Distribution of subsidiary employment across APCE groups for ages 5+

It is seen that in the rural sector the female subsidiary workers in the 55th round area coming much more from a lower expenditure group than those in the 50th round. Notice in particular that the P2/P1 ratio, as defined in Table 4.5, has fallen from a value of 1.66 to 1.04 in the post-liberalization years. The value of this ratio for rural males has also fallen, but not by as much.3 The evidence strongly suggests that the withdrawal of subsidiary workers–which was identified as a dominant feature of the change in the rural labor markets over this period–came increasingly from higher household expenditure groups. Interestingly enough, the trend in the urban sector is the exact reverse. The P2/P1 ratio increased substantially both for males and females, suggesting that the withdrawal observed for subsidiary workers in the urban labor markets came increasingly from lower expenditure groups. We conclude that the 'income effect' seems to have been a factor in the fall in participation of subsidiary workers, particularly females, in the rural areas, but that other factors (e.g., education or social connection) might have been more important in the falling participation rate in the urban economy.

Principal-status labor-force participation

Let us now see how the supply of labor in the UPS category behaves in contrast to the supply in the UPSS status discussed in the last section. The change in the growth rates of UPS labor by gender and sector are shown in Table 4.6.

Unlike in the case of the UPSS labor force no decline in the growth rate is observed in the 1990s compared with the 1980s. However, there is an important gender difference. The male labor-force growth fell in in the 1990s, whereas female labor-force growth increased by 50 percent. The last row in Table 4.6 shows the hypothetical growth rate which would have occurred if the PRs had remained at the same levels as in the 50th round. The significant point to note is that in the rural labor market although the actual growth rate for female principal workers between the 50th and the 55th rounds was below the 'derived' growth rate for this group, it had nevertheless increased compared with the previous period between the 38th and the 50th rounds. Further scrutiny about changes in age-specific participation rates for principal females shows that this increase is really due to an increase in PRs for the prime age groups 25–59 (see Figure 4.2a).

We did a D–A analysis for rural females in the 55th round by broad age groups, showing the difference between the derived figure of the labor force (on the assumption of PRs being unchanged from the 50th round) and the actual labor force reported. A negative figure indicates that the PR for the relevant age group has increased. It was seen that the D–A statistic for the 25–59 age group was minus two million compared to the positive 0.5 million for the total rural female workers of UPS status. Thus while for the rural principal females as a whole there was a net marginal decline in PRs, the PRs for the prime age groups had increased to a significant degree, relative to the trends in all other age–sex groups in the rural labor market for UPS workers.

Considering that there had been a substantial fall in the PRs for the rural females in secondary status in this period (see Figure 4.4a), we conclude that there was some shift of employment from subsidiary to principal status during this period. It is quite consistent with rationalization of the labor force where principal workers are preferred compared with the subsidiary labor force when growth is sustained over a period of time.

It is clear, therefore, that the observed decline in the labor force–and the attendant fall in employment–has not affected the principal labor market. On the contrary the evidence suggests a tightening of labor conditions in this market. Since the reported wage rates refer to this labor market, we would expect this tightening to be reflected in an increase in real wages. This is indeed what we see across the board for different classes of principal workers.

Evidence on wage trends

We have already seen in Chapter 3 that the growth rate of the manual casual wage per day shows a slight acceleration in the rural sector as a whole, for both males and females. More detailed statistics are given in Table 4.7 for the manual

as well as the non-manual parts of the casual labor market. The wage rate accelerated substantially in the second period for both sexes in non-agriculture in rural areas and in the urban areas.

While all groups seem to have experienced an upward trend, the acceleration in the second period was stronger for non-manual workers and in the urban areas. The evidence suggests that the increase in demand for labor in the non-manual labor market supplemented the relative increase in demand for female principal workers, which seems to have been caused by the rationalization of labor deployment in the manual agricultural labor market.

Conclusion on withdrawal of labor

The observed fall in the supply of labor in the 55th round due to decrease in PRs has three major components: the 5–19 age group in UPS status (30 percent of the total decrease); the older 59+ age group in UPS (11 percent); and the females of working age group in subsidiary status (47 percent). It is the last which has been the subject of extended discussion and alternative explanations. Our analysis based on a reading of the historical record suggests that the key to an understanding of this phenomenon is the upsurge in the demand for labor in the early eighties due to the second wave of the green revolution in paddy cultivation and also in oilseed cultivation. This increase in demand was met in the short run by a lift in the participation of SS females as shown in the data for the 38th and 50th rounds. As the economy adjusted to the new level of labor demand in agriculture labor deployment was gradually changed with more use of female labor of a more regular kind. Thus we get a shift from SS females to UPS females between the 50th and the 55th rounds. Since the supply of effort by UPS workers is at a substantially higher level than for the SS workers, this led an overall decrease in female number of workers over all. While this restructuring of the female labor demand is the basic cause of the observed fall in PRs of SS females, we can also discern an element of the income effect leading to a withdrawal of female secondary workers from higher income groups. As explained the wage rates in agriculture are determined in the market for the more regular (UPS) workers. It seems that the increase in demand for such workers continued to keep ahead of the increase in supply due to natural growth of the working-age population, so that the rate of increase in wages increased in the nineties.

Appendix 1 concept of different types of labor force

Usual Principal and Subsidiary Status (UPSS) include persons in the labor force by both major and minor time criteria. In other words, it includes both principal and subsidiary status categories of persons in the labor force.

UPS labor force refers to the persons those who are included in the labor force by major time criterion.

5.0.15 Usual activity status: The usual activity status relates to the activity status of a person during the reference period of 365 days preceding the date of survey. The activity status on which a person spent relatively longer time (major time criterion) during the 365 days preceding the date of survey is considered the principal usual activity status of the person. To decide the principal usual activity of a person, he/she is first categorized as belonging to the labor force or not, during the reference period on the basis of major time criterion…. For the persons belonging to the labor force, the broad activity status of either 'working' or 'not working but seeking and/or available for work' is then ascertained again on the basis of the relatively longer time spent in the labor force during the 365 days preceding the date of survey.

(Instruction manual, 55th round, schedule 10, section 5.0.15.)

In this study, the Subsidiary Status (SS) labor force is defined as persons who are pursuing non-economic activities (out of labor force) by major time criterion (UPS) but belong to the labor force by minor time criterion. It excludes persons who are included in labor force by UPS to avoid double counting. Since unemployment status is determined by major time criterion, those belonging to labor force only on the basis of subsidiary status by default are all workers.

5.0.16 Subsidiary economic activity status: A person whose principal usual status is determined on the basis of the major time criterion may have pursued some economic activity for a relatively shorter time (minor time) during the reference period of 365 days preceding the date of survey …. It may be noted that engagement in work in subsidiary capacity may arise out of the two following situations:

i a person may be engaged for a relatively longer period during the last 365 days in economic/non-economic activity and for a relatively shorter period in another economic activity and

ii a person may be pursuing one economic activity/non-economic activity almost through-out the year in the principal usual activity status and also simultaneously pursuing another economic activity for a relatively shorter period in a subsidiary capacity.

(Instruction manual, 55th round, schedule 10, section 5.0.16.)

In our concept of subsidiary status labor force we have only included the persons who are engaged in non-economic activity for a 'relatively longer period during the last 365 days' but engaged in economic activity for a 'relatively shorter period' or 'pursued non-economic activity almost throughout the year in principal usual activity status' but pursued another economic activity for 'relatively shorter period in a subsidiary capacity'. In this fashion we managed to get UPSS labor force = UPS labor force + SS labor force which is additive.

Appendix 2: estimating the absolute number of the labor force

NSS rounds are sample surveys. They do calculate India's population but these are generally underestimated and the level of underestimation is going up over the year. We can get corrected population estimates for these three mid-year dates by interpolating population figures from three decadal population census of India–1981, 1991 and 2001.

NSS differs from decadal census in terms of age-group distribution population. To adjust for the 38th, 50th and 55th rounds of NSS we have used the five-year age-group distribution of the 1981 Census, 1991 Census and National Health and Family Welfare Survey-II (NFHS) 1998 respectively. The 13 age groups that we have considered are 0–5, 5–10, 10–15, 15–20, 20–25, 25–30, 30–35, 35–40, 40–45, 45–50,50–55, 55–60 and 60 and above. NSS also differs from decadal census in terms of rate of urbanization rate and sex ratio. To get around this problem we used the Census-adjusted NSS mid-year population and age-group distribution separately for rural male (RM), rural female (RF), urban male (UM) and urban female (UF). Thus, we calculated population for these four sections of population for 13 age groups separately.

Labor-force participation rate (LFPR) for each of these four sections of population for all three rounds have been generated for all 12 age groups (for 0–5 age group LFPR is not calculated) from unit-level data. By multiplying LFPR for each of them with the respective population cohort gives us the labor force for each of this population cohort. By adding up the labor force of all age groups and dividing it by its respective population we could derive the LFPR at more aggregate levels. Our calculated LFPR at aggregate level marginally differs from published LFPR figures of NSS.

A similar procedure has been adopted for calculating the number of workers in Chapter 3 and elsewhere.

Part II
Regional dimensions

This page intentionally left blank

5 Some implications of regional differences in labor-market outcomes in India

Ahmad Ahasan and Carmen Pages1

Introduction

This chapter attempts to examine and understand the determinants of key labor-market indicators by looking at the experience across Indian states and regions by analyzing the NSS thick round data from 1983–2000.2 It does so in two parts. First, it shows how states and regions display considerable variation in labor-market outcomes: some states and regions have been able to provide significantly more employment opportunities, and thus show higher employment and participation rates.3 Others states, not all in the same group, have lower unemployment rates.

Wage rates also vary significantly across regions leading to large differences in earnings. This leads to the second part of the chapter, which attempts to understand these patterns by trying to answer three key questions, two from the demand and the other from the supply side of labor markets. First, what are the factors that have enabled some states to create more jobs, and specifically what role did differences in economic activity and economic growth play in creating these jobs? Second, a related issue is, the role of differences in economic activity in affecting the quality of jobs as measured by earnings of workers. On the supply side the key issue that we address is the determinants of the variation in participation rates and especially in female participation rates across states and regions.

It may be useful to highlight the four main findings of this chapter. First, regional level differences in employment indicators are significant across regions and display a geographical clustering: 34 out of 78 NSS regions have statistically significantly different – better or worse – employment and unemployment outcomes than all India averages. Further, not only are these differences significant, but they have persisted over time. A related finding is that employment outcomes are clustered within certain states and regions: regions in the North Eastern states, Bihar and parts of UP, Jammu and Kashmir and Arunachal Pradesh, the coastal belts of Orissa and states such as Goa and Pondicherry have significantly worse employment outcomes. On the other hand Western and Southern States show better employment outcomes (see the third section below).

Second, wage and earning trends across regions present a complex picture with two countervailing aspects. On the one hand, significant differences in wages across regions are being balanced to some extent by evidence of convergence in wages across regions and between rural and urban areas. This finding is consistent with observations that not only are migration rates in India surprisingly low, but they have shown little signs of increasing in the 1990s. Further urbanization rates have also been much lower than predicted (Mohan and Dasgupta, 2004) and lower than in many other comparable countries. If wage rates are converging across regions and between rural and urban areas, then, other things remaining the same, the incentives to migrate to other states and to urban areas will decrease. On the other hand, dualism in wages between the formal, salaried sector and the informal, casual sector persists in that there remains a substantial premium for salaried workers (around 16 percent of the labor force) over casual labor (about 36 percent of the labor force) even after controlling for human-capital characteristics across all regions.

Third, we take up the issue of job content of growth by relating employment and earnings to GSDP levels across states as well as to changes in GSDP across four rounds from 1983 to 2000. Although GSDP growth is significant in explaining the growth of employment it cannot, per se, explain much of the variation in employment growth.4 However, we find that income differences across states and growth of income within states exert substantial positive effect on female employment levels while it also reduces unemployment rates for females in rural areas; but they are not significant in explaining employment-level differences for men. However, GSDP level changes across states do have a significant impact on raising rural earnings for males.

Fourth, regional variations in employment outcomes can be explained on the supply side, by differences in female employment and participation rates. Although female employment rates are uniformly lower than those of males, the variance in regional participation rates is also much higher for female workers. In addition, female participation rates have declined in the 1990s, a trend difficult to reconcile with the declining fertility and increasing education rates of female workers, factors that have contributed to rising female participation in other regions.5 In this chapter, we test two competing hypotheses to explain differences across regions and time in female participation: The first is that changes in participation are driven by income effects – increasing spouse earnings are driving female workers out of the labor force (see Sundaram and Tendulkar 2005b). The second hypothesis is that women are withdrawing from the labor market due to lack of opportunities (substitution effect). We find that while both forces are at work, the lack of opportunities, as indicated by unemployment rates and low expected earnings, have a greater role in explaining this trend.

The rest of this chapter is organized in the following manner: the second section briefly discusses the methodology used here and how the Chapter advances the literature. The third section documents the remarkable variance in labor-market outcomes across Indian states and regions. The following section analyzes these differences focusing on the role played by GSDP in affecting employment and earnings and the differences in participation rates. The final section concludes.

Methodology and how this chapter advances the literature

The evident disparities in economic conditions, growth and human development across Indian states have attracted considerable attention over the last few years. A sizeable literature has developed attempting to explain differences in growth and poverty-reduction performance across states. However, the cross-state and, particularly, the regional dimensions of employment remain relatively unexplored.

The literature on cross-state growth has highlighted the importance of differences in the investment climate in explaining differences in total factor productivity across states (Dollar et al. 2004); other studies have emphasized the differences in infrastructure and regulations (Lall and Mengistae 2005), the decline and variations in plan expenditures, the greater use of private capital flows and the wide variations in credit utilization (Dev 2002); the degree of urbanization (Sach et al. 2002); differences in land reforms, access to credit, education and labor-market related regulatory policies (Besley and Burgess 2004) as factors that have led to divergences in growth across Indian states. While most papers have confined their analysis to the state level, an important paper by Palmer-Jones and Sen (2004) has extended the analysis to the regional level, highlighting agro-ecological factors, irrigation and the interaction between these factors to explain divergence in agricultural growth rates. Overall the literature appears to have converged on a consensus that there is growing divergence in economic performance across states.

As noted, the literature on labor-market differences across states is relatively less developed. Some of the papers have presented at some length the differences in employment and unemployment rates across regions and have emphasized the role of divergent labor-market outcomes being a driving factor behind regional inequality (Bhattacharya and Sakkhivel 2004). A paper by Kijima and Lanjwou (2004) has estimated differences in agricultural wages across different regions. Another by Besley and Burgess (2004) has analyzed the effects of labor regulations in explaining state-wise variations in manufacturing output and employment growth. Hasan et al. (2003) estimated the effect of trade on labor-demand elasticity in industry and showed it to be positive. Extending further, they have shown how regulatory policies in states affect these demand elasticity adversely. At a broader level, the literature on labor-market differences has stopped short at two important points. First, the literature has focused more on describing differences in employment indicators across states and less on analyzing these differences, with Besley and Burgess (2004) and Hasan et al. (2004) being important exceptions. Second, the discussion has stayed focused at the state level except for Palmer-Jones and Sen (2004).

This chapter contributes to the literature on regional labor-market analysis in India in four ways. First, we construct synthetic panel data set on labor-market indicators at the regional and state levels to identify the extent of state and regional differences in Indian labor-market outcomes. The panel-based research on employment in India has so far mainly focused on using the annual survey of industries or smaller sample based data such as the NCAER surveys (e.g., Foster and Rozenweig 2004). By definition such research has excluded more than 80 percent of the Indian labor force or has been based on small nationally unrepresentative surveys. In this chapter we use the nationally representative panel to present the regional differences in employment, wages and participation, and analyze their determinants.

Second, this chapter takes the analysis beyond the state level to the (NSS-sample-based) regional level. Hitherto, analysis of regional level has been confined to a few studies on poverty and agricultural wage rate estimates. This is important since significant differences in employment indicators lie at the regional level. An examination of the data of the 55th round, as many as ten out of 32 states display a higher within state variation (as measured by the coefficient of variation) than the variation across states. Similarly, eight regions show higher within state variation in rural casual wages than the all-India variation.

Third, this chapter extends the discussion to the smaller states of the Northeast such as Arunachal Pradesh, Mizoram, Nagaland and Himachal Pradesh, where much of the variation in employment is found.

Fourth and last, this chapter uses a variety of panel data-estimation techniques using region-level data to examine the drivers of regional variations. These include estimating equations of varying degrees of complexity, fixed-effects models, and using instruments to account for endogeneity of employment, participation, wages and GSDP.

How different are labor-market conditions across states? Some stylized facts

Labor-market outcomes are significantly different across India in a number of respects. In what follows we focus on regional differences in employment, participation and earnings.

Stylized fact 1: striking regional clustering of employment

The first stylized fact is the clustering of low employment rates in the northeastern states of Arunachal Pradesh, Assam, Nagaland, Tripura and Manipur. In addition, the lagging states of UP and Bihar have particularly lower employment rates (Figure 5.1). This trend is largely mirrored in the low participation rates as shown in Figure 5.2. In general one finds a high correlation between employment rates and participation rates (0.95). This correlation is stronger for females (at 0.99 percent) than for men (0.95 percent). The relationship between these variables can be in both directions. Participation can lead to employment. Most workers seeking work in a poor developing-country labor market such as India can find work even if it is in a low-productivity job or a self-employed job. That is supply of labor creates its own demand. On the other direction, low employment rates in a region can lead to a discouragement of workers and lower participation. We examine this issue at some length later in the section 'Understanding differences across states and regions', where we try to analyze some key regional differences and trends in labor-market outcomes.

Participation rates for males and females are shown for all states in Figure 5.2. The relatively lower participation and employment rates in states such as Delhi, Kerala and West Bengal are puzzling (employment rates are presented in Figure 5.1). Given their higher income levels, the low employment rates in these states need more explanation. On the other hand, and more predictably, the prosperous states of the South and West, Tamil Nadu, Karnataka, Gujarat and Andhra Pradesh show significantly higher employment and participation rates. An interesting exception is Rajasthan where employment and participation rates are high.

Figure 5.1 Employment rates for males and female, 55th round (source: Authors estimates. Erm and Erf refer to employment rate of males and females respectively (measured in 0 to 1 scale; i.e. 0.8 refers to 80 percent employment rate). Employment rates for males and female are defined as workers in 15–59 age group as a share of the population in their age group. Derived from NSS 55th Round data).

Figure 5.2 Participation rates for males and female, 55th round (source: Authors estimates). Participation rates for men (prm) and women (prf) are defined as workers and unemployed in 15–59 age group as a share of the population in their age group (units are scaled to 0 to 1). Derived from NSS 55th Round data.

This could be one factor explaining why poverty rates in Rajasthan are low despite its relatively lower level of income.

Stylized fact 2: cross-regional differences in employment rates are much larger for women than for men

The second interesting stylized fact emerging from state and regional analysis is that the variation in female employment rates is significantly higher than for men: the coefficient of variation of female employment and participation rates is nearly four times as large as for men. However, the regional patterns in these variations are not as clear as the ones for males. In addition to low employment and participation rates in the North-eastern regions (including Tripura), and the low rates in UP and Bihar, female employment rates are also very low in West Bengal, and perhaps not so surprisingly in prosperous Punjab. But once again female employment and participation rates are much higher in the prosperous Southern and Western states of Gujarat, Maharashtra, Karnataka, AP and Tamil Nadu. One implication of this grouping is that it will be difficult to attribute low participation rates in West Bengal to schooling rates, as these are also high in the Western and Southern regions.

Stylized fact 3: regional differences in employment rates have persisted

The third stylized fact is that the divergence in employment rates across regions is persistent. We use the threshold of one standard deviation from the Indian mean to classify regions as being significantly different from all-India averages. For most variables, the number of regions significantly different from the mean have either stayed the same or increased in the 55th round (1999–2000) compared with the 50th round (1993–1994). Particularly noteworthy is the increasing divergence in rural employment rates in the 55th round compared with the 50th round. This is also confirmed when we see that employment and participation rates are highly correlated across consecutive rounds: employment and participation rates in the 55th round are closely correlated to those in the 50th round and so on. That is employment and participation rates tend to show very high persistence across regions, in sharp contrast to the case for real wages, which show a small (and negative) correlation across rounds (Table 5.1).

Another and perhaps better way to look for convergence would be to run unconditional convergence regressions (as done in the section 'Understanding differences across states and regions').

Stylized fact 4: there are some signs of convergence in wages for casual workers

As Table 5.1 has hinted, in contrast to employment indicators, there are signs of convergence in wages across regions at a time when wage growth has taken place in most regions. First, wage inequality is falling across regions for all categories of casual wages (see Table 5.2). There is a drop in the coefficient of variation and Gini coefficients in all casual wage categories – rural and urban non-agricultural between 1993 and 2000. However, there was an increase in regional inequality in salaried wages, although it was low to begin with.

Second, convergence in wages is also suggested by econometric tests of convergence that show wages in all categories to have unconditional convergence