Industrial location policy : the Indian experience

SWP620 Industrial Location Policy The Indian Experience A. Uday Sekhar WORLD BANK STAFF WORKING PAPERS Number 620  WORLD BANK STAFF WORKING PAPERS Number 620 Industrial Location Policy The Indian Experience A. Uday Sekhar INTERNATIO??EL MO'?-EARY FuND JOINT MEH99Y [A/AR 2'. 1984 INTEIRNATIONAL D:piiK FOR rL'.E:O5i'JTrCTION i!JD DIVELCPIENT WARlSl2NOTCN,'L .C. 2Ve91 The World Bank Washington, D.C., U.S.A.  Copyright 0 1983 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. First printing October 1983 All rights reserved Manufactured in the United States of America This is a working document published informally by the World Bank. To present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The views and interpretations in this document are those of the author(s) and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting on their behalf. Any maps used have been prepared solely for the convenience of the readers; the denominations used and the boundaries shown do not imply, on the part of the World Bank and its affiliates, any judgment on the legal status of any territory or any endorsement or acceptance of such boundaries. The full range of World Bank publications is described in the Catalog of World Bank Publications; the continuing research program of the Bank is outlined in World Bank Research Program: Abstracts of Current Studies. Both booklets are updated annually; the most recent edition of each is available without charge from the Publications Sales Unit of the Bank in Washington or from the European Office of the Bank, 66, avenue d'Iena, 75116 Paris, France. A. Uday Sekhar is deputy manager of the Industrial Development Bank of India and a consultant to the World Bank. Library of Congress Cataloging in Publication Data Sekhar, Uday. Industrial location policy. (World Bank staff working papers ; no. 620) Bibliography: p. 1. India--Industries--Location. I. Title. II. Series. HC44o.D5S44 1983 338.6'042'0954 83-21774 ISBN 0-8213-0273-6  ABSTRACT The Government of India has been utilizing a wide-ranging set of measures to influence the location of industry. They are generally intended to disperse industries from large metropolitan areas and to promote the development of less developed regions. This paper probes into the working and impact of individual location policies, and demonstrates certain serious deficiencies and lacunae in the justification, design and implementation of many of them. These short- comings call for a thorough overhauling in the approach of the Government towards industrial location policy. In particular, this paper recommends that whenever there is evidence that the existing concentration is inoptimal or inequitable, policy makers should operate with incentives rather than controls, as this form of industrial location policy instrument appears to provide the least damage to the economy while achieving the locational objectives. ACKNOWLEDGMENTS The views expressed in this paper are personal and do not reflect those of the institution in which I am employed. I wish to record my deep indebtedness tco Rakesh Mohan, Professor Edwin Mills, Douglas Keare, Peter Townroe, Koichi Mera and other members of the former Urban and Regional Economics Division of the World Bank for their invaluable assistance in the preparation of this paper.   TABLE OF CONTENTS SUMMARY ................................................................... viii Chapter I -- INTRODUCTION ................................................ 1 Chapter II -- INDUSTRIAL LOCATION POLICIES IN INDIA ....................... 1. Policies Influencing Inter-State Distribution of Industry .. 15 i. IndustriaL Licensing ........................ 15 ii. Location of Public Sector Plants ............................ 22 iii. Distribution and Pricing Policies for Basic Industrial Inputs ...................................... 31 iv. State Government Incentives ................................. 35 2. Policies Influencing the Intra-Regional Distribution of Industry ............................................ 36 i. Policies to Encourage the Small Scale Sector ................ 37 ii. IndustriaL Estate Program ................................... 43 iii. Rural Industries Projects ................................... 49 iv. Metropolitan Planning ....................................... 56 v. Incentives to Develop Backward Districts .................... 59 Chapter III -- TRENDS 'IN INDIAN INDUSTRIAL LOCATION ...................... 69 1. Trends in the Inter-State Distribution of Industry ............... 69 2. Trends in the :Intra-Regional Distribution of Industry ............ 77 Chapter IV -- AN OVERALL ASSESSMENT OF INDUSTRIAL LOCATION POLICY ......... 91 1. Rationale of Industrial Location Policy . . 91 i. Resource ALlocation and Growth ............................... 92 ii. Equity .................... ..... ...... ............... 96 2. Limitations of Industrial Location Policy . . 97 POSTCRIPT..102 POSCRPT.................................................................10 1. National Committee on the Development of Backward Areas .......... 102 i. Policy Recommendations for Medium and Large Industries ............................................. 103 ii. Policy Recommendations for Small Industries, Anci:Llary Industries and Agro-Industries ............... 104 iii. Recommendations for Maximizing the Local Impact of Industrial Development ................................. 105 2. A Critique of lthe NCDBA Report ................................... 106 BIBLIOGRAPHY .............................................................. 109 APPENDIX I: Maximum and Minimum Values of Hirschman-Herfindahl and Thei:L Indices .......................................... ll APPENDIX II: A Shift-Shasre Analysis of Industrial Growth in the lMajor Industrial States of India ................... 113  LIST OF TABLES Table Page 1. Share of Manufacturing in GDP (7) in some Developing Economies, 1977 2 2. Structure of the Indian Manufacturing Sector, 1974 4 3. Trend Growth Rates (7 per annum) in the Manufacturing 6 Sector 4. Value Added in Factory Manufacturing Sector as a Proportion of State NDP (7) 7 5. VAM/NDP For Each State Divided by VAM/NDP for All-India 8 6. Share of States in Total Value Added in Factory Manufacturing Sector (7) 10 7. Share of States in Total Factory Employment in Manufacturing (7) 11 8. Distribution of Urban Employment and Population in Non-Household Industry by the Size of Cities, Towns, and Town Groups (TG)/Urban Agglomerations (UA) 12 India, 1961 & 1971 9. Approved Licenses as a Percentage of License Applications, 1959-66 18 10. Correlation Coefficients between State-Wise Proportions of Licenses Approved and Indices of State Development 20 11. Share of States in Cumulative Investment (Gross Block) in Central Public Undertakings (7) 24 12. Share of States in Annual Investment In Central Public Undertakings (7) 25 13. Correlation Coefficients Between State-Wise Per Capita Investment in Central Public Undertakings and Indicators of State Development 26 14. Distribution of Cumulative Public Sector Investment (7) 28 15. Share of Employment in Central Public Undertakings in Total Employment in Manufacturing and Mining (7) 29  16. Orders placed by Public Sector Undertakings to Small Scale Units as a Percentage of Output of Modern Small Scale Sector 30 17. Region-Wise Installed Capacity of Cement (%) 31 18. Share of Unregistered Sector in Value Added in Manufacturing (%) 38 19. Employment and Output in Modern Small Scale Sector, 1972 39 20. Percentage of Sheds Completed but not Functioning 46 21. Share of Industrial Estates in Industrial Employment and Output (%) 50 22. Public Expenditure on Rural Industries Projects 53 23. Proportion of Saimple Units Availing of R.I.P. Assistance (%) 54 24. Estimated Public Sector Expenditure on Small Scale Industries 55 25. Share of Backward Districts in Area and Population, 1971 60 26. Reimbursement of Central Investment Subsidy by Government of India (Rs.m) 62 27. Share of States and Union Territories in the Central Investment Subsidly Disbursed during 1972/73 to 1978/79 63 28. Terms of Concess ional Finance for Units in Backward District/Areas, 1975 66 29. Trends in the Share (%) of Backward Districts in Assistance Sanctioned and Disbursed by the Industrial Development Bank of India, 1964/65 to 1980/81 67 30. Value Added per Employee (Rs. '000) in Industry 72 31. State Share in Value Added/State Share in Employment in Factory Manufacturing Sector 73 32. Measures of State-Wise Concentration of Industry (Factory Sector) 74 33. Distribution of Industrial Employment in Cities, Towns, and Town Groups (TG)/Urban Agglomerations (UA), 1961 & 1971 79  LIST OF FIGURES Figure Page 1. Share of Backward Districts in Factory Employment (7), Maharashtra 86 2. Share of Backward Districts in Factory Employment (7), Haryana 87 3. Share of Backward Districts in Factory Employment (7), Punjab 88 4. Share of Backward Districts in Factory Employment (7), Andhra Pradesh 89 5. Share of Backward Districts in Factory Employment (.), Kerala 90 Map 1. Map of India 118  SUMMARY As in most other developing countries, India had been experiencing since her independence a high and perhaps increasing concentration of population and economic activities at selected locations such as Bombay and Calcutta and severe disparity in welfare among different regions within the country. In response, the Government of India adopted a series of measures to promote a more decentralized and spatially "balanced" development. Most of the measures are directed at influencing the location of manufacturing establishments as they are viewed as the prime determinant of spatial development. This policy has been followed by state and local governments as well. However, these measures cannot be without unintended side effects. The purposes of this paper are first to identify the intended objectives of industrial location policies which have been instituted in India in order to assess to the extent possible their effectiveness in achieving the objectives, and then to evaluate their possible unintended as well as intended effects in order to arrive at a more balanced view on this subject. In the first chapter, the manufacturing sector of India is reviewed with emphasis on identifying its spatial pattern of distribution. On the state basis, Maharashtra, West Bengal, Gujarat and Tamil Nadu are most industrialized, although the share of West Bengal has been eroding consistently since the early 1960's. Chapter 2 presents a review of industrial location policies currently in effect. Those influencing interstate distribution of industry comprise (1) industrial licensing, (2) location of public sector plants, (3) distribution and pricing policies for intermediate industrial inputs and (4)  state government incentives. Since 1951, the Central Government has been authorized to license a new industrial establishment or a substantial expansion. This instrument has been used to achieve a "balanced" regional development. This policy orientation has been strengthened since 1965, and since 1977 an explicit decision has been made to prohibit the setting up of licensable industries within certain geographical limits of large metropolitan cities with a population exceeding one million and urban agglomerations with a population greater than half a million. Location decisions of public sector plants are also an important policy instrument, because many basic and capital goods are produced by the public sector in India. The less developed states of Madhya Pradesh and Bihar received large shares of investment, followed by Orissa and West Bengal. Since location decisions were made on the basis of technical and economic considerations as well as regional balance, it is impossible to isolate the degree of regional consideration which went into location decisions, but an analysis shows that the per capita cumulative investment in this sector has been consistently negatively related to state per capita income and undoubtedly it contributed to balancing the distribution of industry among states. With regard to the distribution of such basic commodities as cement, steel and coal, "freight-pooling" is exercised by the Central Government so that firms in every location may equally benefit from the provision of these goods. This policy, as well as the preceding ones, disturts firms' locations from the economically rational locations.  State governments provide a host of incentives aimed at attracting industries to their own states. They include financial assistance for investment, the provisioin of infrastructure and industrial estates, concessions in power and water tariffs and concessions in taxes. The policies Lnfluencing intra-regional distribution of industries include (1) those for encouraging village and cottage industries as well as modern small-scale enterprises, (2) industrial estate programs, (3) the Rural Industries Project Program, (4) metropolitan planning and (5) incentives to promote industrial development in backward districts. Generally they are designed to promote the development of rural and backward areas by encouraging industries in those areas or by encouraging traditional or small enterprises which are more equally distributed over space. One exception is metropolitan planning which is frequently oriented to zoning out industries to peripheral areas. One example is the Bombay Metropolitan Regional Plan of 1973 which attempted to divert most industries to the "counter magnet" across the bay. Then, Chapter 3 analyzes changes in the spatial pattern of industrial location during recent decades. As far as interstate distribution of industry is concerned, various indices indicate that the concentration of industry subsided from the 1960's to the 1970's. For example, converging trends among states were observable from 1961 to 1975 in the share of the manufacturing sector in the state product and in the value added per employee in the manufacturing sector. Also, the industrial value added and employment became more equally distributed among states from 1960 to 1975 as measured by the Theil's Inequality and the Hirschman-Herfindhal Indices. Since the trends in industrial location are results of both market forces and government  policy, it is not possible to conclude that the industrial location policies have successfully narrowed disparities among states. However, there are good reasons for speculating that some of the policies were instrumental in narrowing the disparities. The intra-regional distribution of industry was examined next. By comparing the degree of concentration of industrial employment in 1961 and 1971 by grouping cities by size, the following conclusions were obtained: for India as a whole, the employment in the household industry lost the degree of concentration substantially from 1961 to 1971, due to a reduction of concentration (relative to population) in the smaller classes of towns. However, the non-household industry maintained its level of concentration during the period. This means that the industrial policy apparently had not resulted in any substantial shift of industry away from large cities to smaller towns. Chapter 4 presents an overall assessment of industrial location policy. From the viewpoint of efficiency in resource allocation, one should not expect a regionally balanced industrial development because natural resources are not evenly distributed and there are economies of scale and agglomeration in production processes. Industrial policies by the public sector are needed for the provision of infrastructure because in principle they cannot be provided adequately by the private sector and for correcting externalities which arise in the form of congestion and pollution. However, the former can best be provided by identifying gaps in infrastructure based on the demand for infrastructure and the costs of providing it. Externalities can best be dealt with by adjusting prices to reflect externalities.  Therefore, from the viewpoint of efficiency, the best set of measures to correct inefficient locational distribution of industries operates via readjustments in infrastructural investment and the prices charged for their utilization. Direct industrial location policies such as licensing and input rationing are poor substitutes. Even if the best solutions are not workable, it would be advisable to operate through incentives and disincentives rather than through controls, because the latter were crude devices and may lead to extremely large efficiency loss in the national economy without being detected by any observer. A stronger argument may be made for industrial location policy for the sake of spatial equity. However, as population and labor are mobile within a country and particularly within a state, the need of achieving spatial equity itself can be questioned. In addition, there is serious doubt about the effectiveness of industrial location policy in achieving spatial equity. One serious limitation of locational control is that, contrary to expectations, controls cannot induce industrialists to invest in non-viable areas. Therefore, if licensing is utilized actively for locational purposes, the result will be a decrease in investment in the restricted areas without a compensating increase in the other areas. This involves not only a loss in efficiency, but also a defeat of the objective of industrializing non-restricted areas. This and other examples of serious inefficiencies generated by industrial location policies imply that not enough thought has been given to the rationale of these policies. Our exploration of their ramifications reveals that a thorough overhauling is required in the approach toward locational concentration of industry.   -1- Chapter I INTRODUCTION Economic development, by its very nature, is a dynamic process characterized by various types of "imbalances." Nowhere is this more striking than in the spatial dimensions of growth, growth which tends to get concentrated in certain regions within the economy and, in the case of non-agricultural activities, at a few nodal metropolitan centers. Whether this spatial "imbalance" is a "natural" and dynamically efficient (in the Pareto sense) concomitant of the growth process or whether it is partly a result of imperfectly functioning markets and externalities is a debatable issue. The answer varies, of course, from country to country and must be settled by empirical evidence. Even if spatial concentration is efficient, one is still faced with the equity issue, which is extremely important in a geographically large federal polity such as India. Motivated by these welfare and efficiency considerations, the Indian government has been adopting a series of measures to promote a more decentralized and "balanced" form of development. In this study, we are only concerned with those measures affecting the manufacturing sector. Our objective is to examine the various explicit policies that have been used by the state to influence industrial location and to assess their effects. But before embarking on a study of these policies, it would be useful to set the stage by briefly surveying the pattern of industrial growth in India. The origins of modern industry in India can be traced as far back as the nineteenth century. Even the diversification of the manufacturing sector from its predominant emphasis on consumer goods started early in the twentieth  -2- century when Jamshedji Tata succeeded in setting up a steel plant in 1911 against tremendous odds. Because of these early origins, the Indian manufacturing sector at the time of independence in 1947 was already quite large in comparison with other developing countries. After 1947, the era of planned development began, and this gave further impetus to industrialization. The State also took an active role in setting up industries producing basic and capital goods, especially during the Second Five-year Plan (1956-61). However, despite these early traditions of entrepreneurship, agriculture continues to dominate the Indian economy. In 1977, the manufacturing sector accounted for only 157. of GDP, which is rather low compared with other developing nations with relatively well developed industrial sectors like Brazil, Mexico, Argentina and Korea (Table 1). But, given the size of the Indian economy, the manufacturing sector is extremely large in absolute terms. In 1973, India had the sixteenth largest manufacturing sector among all market economy countries. Table 1 Share of Manufacturing in GDP (C) in some Developing Economies, 1977 Argentina 33 Korea 27 Mexico 24 Brazil 23 India 15 Pakistan 14 Kenya 11 Nigeria 8 SOURCE: United Nations, Yearbook of National Accounts Statistics, 1978.  -3- As Table 2 shows, the Indian manufacturing sector is quite diversified, with capital goods constituting as much as 22% of value added in manufacturing in 1974. The corresponding shares for consumer and intermediate goods were 32% and 46% respectively. The institutional milieu of Indian manufacturing displays an interesting diversity. The most remarkable aspect is the importance of production by traditional and small scale enterprise. The so-called "unregistered sector," consisting predominantly of cottage and village industries utilizing traditional techniques and employing household labor to a significant extent, contributed more than a third of total value added in manufacturing in 1975. Small modern enterprises (which, in 1980, were defined as factories with fixed capital value of less than Rs. two million) accounted for 16%. In addition, there are imanufacturing units operated by the State, whose output constituted only 12% of total manufacturing output in 1973. However, this understates their real importance, since public sector units are mainly engaged in tlhe production of certain critical items like steel and various types of machinery and transport equipment. In comparison with certain other deveLoping countries, foreign capital plays a relatively minor role in the Indian industrial sector. Only 8.5% of total paid-up capital in Indimn joint stock companies in 1969-70 originated from foreign sources. Value added in foreign subsidiaries constituted only about one tenth of total value added in manufacturing in the same year. The remaining part of the manufacturing sector is, of course, the private sector steered by domestic entrepreneurship.  -4- Table 2 Structure of the Indian Manufacturing Sector, 1974. Share in Total Manufacturing Value added (IL) Food Products 8.6 Beverages 0.6 Tobacco 2.1 Textiles 21.2 Wearing Apparel and Footwear 0.5 Leather and Products 0.3 Wood Products 0.6 Paper and Products 3.5 Printing and Publishing 1.9 Industrial Chemicals 10.2 Other Chemical Products 7.1 Drugs and Modicine 2.8 Rubber Products 1.8 Plastic Products 0.5 Non-metallic Mineral Products 3.3 Metals and Metal Products 15.6 Non-electrical Machinery 7.8 Electrical Machinery 6.9 Transport Equipment 7.1 MANUFACTURING 100.0 NOTE: The data pertains to establishments with ten or more workers using power or 20 or more workers without power. SOURCE: United Nations, Yearbook of Industrial Statistics, New York 1978.  -5- Table 3 shows the growth rates of value added in the registered and unregistered manufaLcturing sectors and in the index of manufacturing production during the periods 1951-64 and 1965-78. The major feature of industrial development during Lhe post 1965 period that has attracted attention is the slowdown in growth of the organized sector, which declined sharply from 7.37 per annum to about 3.47 per annum in the two periods. The deceleration of the unregistered manufacturing sector has been more moderate, though significant. The net result has been that the growth rate of real value added in the manufacturing sector as a whole fell from 6.37 per annum in 1950-64 to 3.67 per annum during 1965-75. One of the striking characteristics of Indian industrial development is its uneven geographical spread. In order to identify the relative degree of industrialization of different states, we computed the share of value added in the factory manufacturing sector in NDP for each state (Tables 4 & 5). Comparing the ratios for 1976, one finds that Maharashtra, West Bengal, Gujarat and Tamil Nadu were the industrial leaders, in the sense that they had an above average manufacturing value added/NDP (VAM/NDP) ratio. In 1961, also, these states (along with Delhi) were relatively more industrialized (with the exception of Tamil Nadu). However, during 1961-76, the manufacturing sectors of the other states have been growing at a faster pace relative to NDP. As a result, a trend towards equalization of state VAMP/NDP ratios is apparent. In 1976, Maharashtra, West Bengal, Gujarat, and Tamil Nadu together accounted for about 557 of value added and 527 of employment in the manufacturing sector (Tables 6 & 7). The contribution of Maharashtra alone was about 257 of value added and 197 of employment. However, the shares of these states hiave been falling since 1961, when they contributed as much as two-thirds of value added and 587 of employment.  -6- Table 3 Trend Growth Rates (. Rer annum) in the Manufacturinz Sector 1950-64 1965-75 Value added in the registered manufacturing sector (in constant prices) 7.3 3.4 Value added in the unregistered manufacturing sector (in constant prices) 5.0 3.9 Value added in manufacturing (in constant prices) 6.3 3.6 Index of manufacturing production 6.7 3.9 (1951-64) (1965-78) NOTE: The growth rates have been estimated from log-linear trend lines. The contant prices are 1960-61 prices for 1950-70 and 1970-71 prices for 1970-75. The two have been merged and the trend growth rate computed with the help of a dummy variable. Changes in base years in the index of manufacturing production have also been accomodated with the help of a dummy. SOURCE: Sekhar (1981), p.6  -7- Table 4 Value Added in Factory Manufacturing Sector as a Proportion of State NDP (%) Karna- Maha- A.P. BihaLr Gujarat taka Kerala M.P. Rashtra 1961 3.4 7.1 14.8 5.9 6.7 3.9 19.2 1962 3.4 8.2 13.5 5.9 7.0 2.8 19.2 1963 3.5 7.4 13.6 6.7 6.7 3.9 18.8 1964 3.6 7.5 12.4 6.7 5.9 4.4 18.7 1965 4.2 7.31 14.1 7.2 6.2 4.9 19.8 1966 4.0 8.7 13.2 7.1 7.8 5.4 18.7 1967 3.4 6.4 12.5 7.1 8.5 4.2 17.6 1968 3.8 6.8 13.9 7.3 8.5 6.0 17.9 1969 4.9 7.1 13.6 8.5 9.9 4.8 20.2 1970 5.0 7.7 13.0 9.2 7.3 5.6 21.1 1971 5.0 8.3 12.7 10.1 8.7 5.0 21.9 1972 1973 6.2 6.7 14.3 7.8 7.3 6.9 20.8 1974 5.6 9.2 20.3 8.1 8.3 7.6 21.4 1975 7.3 12.2 16.0 10.1 7.6 7.7 20.7 1976 8.3 9.C0 16.9 10.9 8.0 9.6 21.1 Punjab & Haryana Orissa Rajasthan T.N. U.P. W.B. Delhi India 1961 3.3 2.0 8.7 3.7 15.9 11.6 8.3 1962 4.0 2.3 9.7 3.6 18.0 10.4 8.6 1963 4.8 2.6 9.6 3.7 18.1 12.0 8.5 1964 5.1 2.4 10.5 3.5 18.2 11.8 8.3 1965 7.2 7.5 2.9 11.4 4.2 18.1 11.5 9.0 1966 6.3 4.4 2.7 10.7 3.5 16.1 10.7 8.5 1967 5.0 2.8 2.9 10.9 2.6 14.0 9.6 7.5 1968 6.7 4.6 4.1 10.9 4.3 13.4 9.8 8.0 1969 6.1 4.7 3.9 12.2 4.6 13.7 11.0 8.7 1970 6.6 5.5 4.2 12.8 4.9 13.4 10.0 9.1 1971 7.1 4.8 4.9 11.6 4.8 13.9 10.0 9.3 1972 1973 6.4 5.6 3.7 12.2 4.8 14.1 9.2 9.1 1974 7.8 6.3 4.9 15.2 5.6 15.6 8.6 10.2 1975 8.7 5.4 6.3 13.9 6.0 16.4 8.1 10.5 1976 8.9 6.5 16.3 6.8 16.9 10.8 11.3  -8- Table 5: VAM/NDP* For Each State Divided by VAM/NDP For All-India Karna MIaha A.P. Bihar Gujarat taka Kerala M.P. Rashtra 1961 0.4 0.9 1.8 0.7 0.8 0.5 2.3 1962 0.4 1.0 1.6 0.7 0.8 0.3 2.2 1963 0.4 0.9 1.6 0.8 0.8 0.5 2.2 1964 0.4 0.9 1.5 0.8 0.7 0.5 2.3 1965 0.5 0.8 1.6 0.8 0.7 0.5 2.2 1966 0.5 1.0 1.6 0.8 0.9 0.6 2.2 1967 0.5 0.9 1.7 0.9 1.1 0.6 2.4 1968 0.5 0.9 1.8 0.9 1.1 0.8 2.2 1969 0.6 0.8 1.6 1.0 1.1 0.6 2.3 1970 0.5 0.9 1.4 1.0 0.8 0.6 2.3 1971 0.5 0.9 1.4 1.1 0.9 0.5 2.4 1972 1973 0.7 0.7 1.6 0.9 0.8 0.8 2.3 1974 0.6 0.9 2.0 0.8 0.8 0.7 2.1 1975 0.7 1.2 1.5 1.0 0.7 0.7 2.0 1976 0.7 0.8 1.5 1.0 0.7 0.8 1.9 Punjab & Raja- Haryana Orissa sthan T.N. U.P. W.B. Delhi 1961 0.4 0.2 1.0 0.4 1.9 1.4 1962 0.5 0.3 1.1 0.4 2.1 1.2 1963 0.6 0.3 1.1 0.4 2.1 1.4 1964 0.6 0.3 1.3 0.4 2.2 1.4 1965 0.8 0.8 0.3 1.3 0.5 2.0 1.3 1966 0.7 0.5 0.3 1.3 0.4 1.9 1.3 1967 0.7 0.4 0.4 1.5 0.4 1.9 1.3 1968 0.8 0.6 0.5 1.4 0.5 1.7 1.2 1969 0.7 0.5 0.4 1.4 0.5 1.6 1.3 1970 0.1 0.6 0.5 1.4 0.5 1.5 1.1 1971 0.8 0.5 0.5 1.3 0.5 1.5 1.1 1972 1973 0.7 0.6 0.4 1.3 0.5 1.6 1.0 1974 0.8 0.6 0.5 1.5 0.6 1.5 0.8 1975 0.8 0.5 0.6 1.3 0.6 1.6 0.8 1976 0.8 0.6 1.4 0.6 1.5 1.0 * Value Added in Factory Manufacturing Sector as a Proportion of NDP.  -9- The distribution of industrial employment (in household and non-household industry) among different classes of cities and towns in 1961 and 1971 is presented in Table 8. Factory industry is heavily concentrated in cities with population exceeding 100,000 - cities which in 1971 employed over 70'S of factory workers. We shall ana:lyze trends in industrial location in greater depth later in this paper. Here, we have merely tried to present the marked inter- and intrastate differences in industrialization which are the motivating forces behind the design and implementation of locational policies, to which we now turn our attention.  -10- Table 6 Share of States in Total Value Added in Factory Manufacturing Sector (7.) Karna- Maha- A.P. Bihar Gujarat taka Kerala M.P. rashtra Orissa 1961 3.1 6.3 10.5 3.8 2.7 3.0 27.0 1.1 1962 2.9 7, 8.8 3.7 2.7 1.9 27.1 1.6 1963 3.1 6.5 8.5 4.3 2.4 2.8 26.3 1.9 1964 3.2 6.5 8.3 4.4 2.2 3.3 25.4 2.0 1965 3.4 6.5 8.6 4.5 2.3 2.9 24.8 2.6 1966 3.5 6.8 8.7 4.9 2.9 3.3 25.3 1.7 1967 3.3 6.2 9.4 5.1 3.7 3.5 26.1 1.2 1968 3.3 5.9 9.1 5.2 3.7 4.4 25.8 1.9 1969 3.9 5.6 8.7 5.4 4.0 3.3 26.6 1.8 1970 4.0 5.5 9.1 5.7 2.9 3.6 26.b 1.9 1971 4.0 5.7 8.3 6.0 3.1 3.3 27.4 1.5 1972 1973 5.1 4.9 9.5 5.1 2.8 4.8 26.8 2.0 1974 4.3 b.2 10.0 4.5 2.8 4.7 26.2 1.7 1975 5.0 7.8 8.9 5.1 2.5 4.3 24.6 1.6 1976 5.1 5.4 9.4 4.9 2.5 4.8 24.4 2.2 Punjab & Raja- harjana sthan T.N. U.P. W.B. Delhi Total 1961 3.0 1.1 8.5 6.1 18.9 1.9 100.0 1962 3.1 1.1 8.8 5.6 21.0 1.7 100.0 1963 3.6 1.1 8.2 5.7 21.6 1.9 100.0 1964 4.0 1.2 8.8 b.0 21.0 1.9 100.0 1965 4.1 1.3 9.1 6.6 20.0 1.8 100.0 1966 4.5 1.4 9.1 6.0 18.4 1.7 100.0 1967 4.1 1.7 9.6 5.0 17.8 1.7 100.0 1968 5.1 1.8 9.1 7.2 16.0 1.7 100.0 1969 4.5 1.7 9.5 6.9 14.8 1.8 100.0 1970 4.7 2.1 9.8 6.6 13.6 1.7 100.0 1971 5.0 2.1 9.4 6.2 14.4 1.7 100.0 1972 1973 4.5 1.8 9.3 6.4 13.3 1.5 100.0 1974 4.6 1.9 9.5 6.5 12.9 1.4 100.0 1975 5.2 2.4 8.5 6.5 13.3 1.4 100.0 1976 5.4 2.3 9.4 7.4 12.1 1.7 100.0  -11- Table 7 Share of States in Total Factory Employment in Manufacturing (X) Karna- Maha- A.P Bihar Gujarat taka Kerala M.P. rashtra Orissa 1960 5.8 5.0 9.2 4.6 4.5 4.2 20.8 0.8 1961 5.8 4.9 9.2 4.5 4.4 4.3 21.1 1.0 1962 5.7 4.9 9.2 4.7 4.3 4.2 20.7 1.1 1963 5.6 4.9 9.0 5.1 4.0 4.3 20.7 1.2 1964 5.6 5.0 8.8 5.1 4.1 4.3 20.5 1.4 1965 5.5 5.1 8.8 4.8 4.2 4.5 20.2 1.4 1966 5.4 5.1 8.8 5.1 4.3 4.5 19.9 1.4 1967 5.6 5.3 8.7 5.2 4.3 4.4 19.8 1.4 1968 5.9 5.4 8.5 5.3 4.3 4.7 19.7 1.6 1969 5.4 5.5 8.6 5.4 4.3 4.4 20.4 1.5 1970 5.2 5.6 8.8 5.6 4.1 4.6 20.1 1.5 1971 5.2 5.6 8.8 5.5 4.1 4.4 20.7 1.4 1972 6.0 5.4 8.9 5.9 4.3 4.4 19.8 1.3 1973 6.3 5.5 9.1 5.8 4.5 4.3 19.8 1.3 1974 6.5 5.6 9.2 5.9 4.6 4.3 19.2 1.3 1975 6.9 5.9 9.2 6.0 4.6 4.3 18.7 1.3 Punjab & Raja- Haryana sthan T.N. U.P. W.B. Delhi Total 1960 3.4 1.5 8.6 8.3 19.2 1.8 100.0 1961 3.4 1.5 8.4 b.6 18.9 1.8 1U0.O 1962 3.5 1.b 8.3 8.5 19.3 1.8 100.0 1963 3.5 1.5 8.3 8.5 19.4 1.8 100.0 1964 3.6 1.6 8.2 8.7 19.2 1.8 100.0 1965 3.7 1.6 8.2 8.7 19.3 1.8 100.0 1966 3.7 1.6 8.6 8.8 18.6 1.9 100.0 1967 3.8 1.7 8.8 8.5 18.2 1.8 100.0 1968 3.8 1.7 8.6 8.1 17.9 1.8 100.0 1969 4.0 1.7 8.8 8.3 17.1 1.9 100.0 1970 4.2 1.7 9.0 8.4 16.8 1.9 100.0 1971 4.4 1.7 9.0 8.3 16.5 1.9 100.0 1972 4.4 1.8 9.0 8.3 16.1 1.9 100.0 1973 4.5 1.8 8.8 8.2 15.9 1.9 100.0 1974 4.4 1.8 9.1 8.2 15.6 1.9 100.0 1975 4.5 1.9 8.7 8.3 15.2 2.0 100.0  -12- Table 8 Distribution of Urban Employment and Population In Non-Household Industry by the Size of Cities, Towns, and Town Groups (TG)/Urban Agglomerations (UA) India, 1961 & 1971 Share (.) of Different Classes of Cities/Towns in City/Town/TG-UA Employment b/ Population Class a/ 1961 1971 1961 1971 Class I 62.6 71.8 48.4 55.8 Class II 11.5 8.8 11.9 11.3 Class III 15.1 10.9 18.5 16.3 Class IV 7.2 5.9 13.0 11.3 Class V 3.2 2.4 7.2 4.7 Class VI 0.3 0.2 0.9 0.5 100.0 100.0 100.0 100.0 NuTE: a/ Cities, towns and town groups/urban agglomerations are categorized into the following classes: Class I: Population of 100,000 or more Class II: 50,0UU to 99,999 Class III: 2U,000 to 49,999 Class IV: 10,000 to 19,999 Class V: 5,000 to 9,999 Class VI: Below 5000 For other concepts and definitions, see Table 33 b/ Employment in non-household industry Source: Government of India, Census of India 1961 and 1971.  -13- Chapter II INDUSTRIAL LOCATION POLICIES IN INDIA Before examining the content of the industrial location policies, we shall aLtempt to throw some light on the factors that influenced the policy-makers in adopting these measures. Unfortunately, these factors are not clearly formulated in any of the policy documents, for example, the five-year plans. The justifications for state intervention in industrial location are couched in efficiency and equity terms, but there is no precise idea about the trade-offs involved (wherever these exist). This probably reflects, to a significant extent, a major weakness of the theoretical development literature and its failure to develop an empirical methodology to assess the most "desirable" spatial distribution of industry. However, a clear difference in motivation can be perceived between policies influencing inter and intra regional distribution of industry. In India, It is individual states that form the basis of regionalization by the decision makers, though sometimes the North-South-East-West divisions are also utilized. The original demarcation of the states was on a linguistic basis, with a corresponding cultural and ethnic homogeneity. Thus, strong sociological bonds tie the inhabitants of each state together, and create effective demands for receiving "just" shares of the national pie. The Indian Republic attained independence only about three decades back, so that these demands can be backed by arguments about unequal treatment during the colonial period. The setting-up of infrastructure and transportation networks during British rule was determined largely by political exigencies and the colony-metropolis tracle relationship. This obviously meant that, at the  -14- time of independence, the regions started with different infrastructural endowments - a problem with which the new government had to come to terms. Apart from these historical reasons, physical endowments and various socio-economic factors are responsible for the diverging performances of different regions. However, even in instances where these factors dominate, it is politically infeasible to let these differences in growth exceed certain limits in keeping with the character of the federal polity. All these political and equity considerations have led the Indian government to attempt to implement policies to promote a more balanced inter-state distribution of industry. Policies targeted at affecting intra-regional industrial location have been motivated both by efficiency and welfare arguments. Efficiency justifications have focused on the inoptimal distribution of infrastructure and imperfectly functioning markets that have led to the concentration of industry in certain metropolitan regions to the neglect of smaller towns and backward areas. It is argued that the latter possess industrial potential but have not developed because of insufficient infrastructural investment, high risk aversion on the part of entrepreneurs in the face of uncertainty, and imperfectly functioning capital markets. Proponents of these policies also point out that agglomeration economies and economies of scale have been exhausted in certain cities, and their continued expansion implies inefficient resource allocation. However, apart from these efficiency issues, equity factors are also seen as extremely important justifications of policies such as those encouraging the village and handicrafts industries that have traditionally formed an important component of rural economic life. These industries utilize highly labor-intensive techniques which may be able to  -15- absorb some part of the under-employment that characterizes Indian agriculture. It is felt that encouragement of modern small-scale enterprises, which tencl to be concentrated in the metropolitan areas, will promote a moro equitable distribution of income by increasing the returns to small entrepreneurs. All these efficiency and equity issues and their implications for industrial location and policy will be explored at length in Chapter 4 where we draw lessons from the Indian experience with industrial location policies. First, however, we look at those designed to promote a more balanced inter-state distribution of industry. 1. Policies Influencing Inter-State Distribution of Industry Chronologically arranged in order of their implementation, these policies have been of the following kinds: (i) Industrial licensing; (ii) Location oDf Public Sector Plants; (iii) Distribution and pricing policies for intermediate industrial inputs; and (iv) State Government incentives. (i) Industrial Licensing The basic attitude of the Government of India towards the direction and character of industrial development is still epitomized by the "Industrial Policy Rtesolution" (IPR) that was adopted in 1956. 1/ One of the important objectives of industrial policy listed in the IPR is 1/ This replaced an earlier resolution formulated soon after independence.  -16- balanced regional development.l/ The mechanism of implementing industrial policy was the Industries (Development and Regulation) Act of 1951. This Act provided the legislative framework for the licensing and regulation of industrial investment. A licence from the Central Government was required for the establishment of a 'new' industrial undertaking or for 'subsLantial expansion' of existing units. 2/ There is heuristic evidence that the state governments have tried to influence the licensing authorities in order to attract industry. The testimony of officials to the Estimates Committee (1967-68) illustrates this: "It has been sLated by the official representative of the Ministry that 'Quite a number of states keep on representing that their areas have not received adequate share of licenses' we have no compulsive means of attracting people to particular areas. We consider the applications as we receive them and provided the location in a particular area is not too uneconomic, other things being reasonably equal, we do give preference to the more underdeveloped areas." 3/ 1/ Other, perhaps more important, objectives are: rapid industrialization in conformity with plan priorities and targets, development of basic and capital goods industries, expansion of the public and co-operative sector, protection and encouragement of the small scale sector and the curbing of monopoly in industry. 2/ The precise rules relating to the licensing system have varied over time. It must also be noted that additional licenses were required for imports of equipment. Furthermore, there are a number of other Government controls, e.g. price and distribution controls. Fresh investment involving the large business houses or foreign collaboration were subjected to further scrutiny by special statutory committees. 3/ However, the correlation coefficient with per capita income is only 0.07.  -17- The Industrial Licensing Policy Inquiry Committee Report (1967) also reported that "there have been cases where, because of persistent pressure by the state authorities concerned, licenses have been granted for location within those states. Our case studies show that there were certain states which followed up applications for location in their territories very systematically and persistently and these states were often able to ensure that applicants for licenses within their territory succeeded." (p. 113) However, the actual weightage assigned to regional location by the licensing authorities is a moot point. The Hazari report, which examined industrial planning and licensing policy, contains information relating to state-wise licensing applications and approvals for the-period 1959-66. This is presented in Table 9. In order to assess the importance of the "balanced regional development" objective, we correlated these approval ratios with an index of state industrial development, i.e., the share of industry in GDI?. The correlation coefficient turned out to be 0.40. 1/ Now, this result cannot be interpreted from a purely regional perspective. This is because the approval ratio Is a function of many other variables, like the techno-economic character of license applications, and the other industrial policy objectives including the relationship with plan piriorities, encouragement of the public and small-scale sector, and the curbing of monopoly. Despite these qualifications, the positive correlation does Indicate that the "balanced regional development" objective has not prevented the more advanced industrial states from faring better in the licensing decision process. 1/ However, the correlation coefficient with per capita Income is only 0.07  -18- Table 9 Approved Licences as a Percentage of Licence Applications, 1959-66 Approved Licenses Share of MIanufacturing Per capita income Licence Applications (X) in GDP (%) (Rs. at current prices) 1960 1965 1959-6 1961 1965 1960 1965 A.P. 48 47 48 3.4 4.2 275 387 Bihar 63 48 48 7.1 7.3 215 332 Delhi 54 32 49 11.6 11.5 6W8 887 Gujarat 50 40 49 14.8 14.1 362 498 Kerala 50 64 53 6.7 6.2 259 380 M.P. 51 40 44 3.9 4.9 260 305 Tamil Nadu 47 57 57 8.7 11.4 334 403 Miaharashtra 56 48 53 19.2 19.8 409 534 Karnataka 56 37 55 5.9 7.2 296 448 Orissa 52 63 55 3.3 7.5 21b 329 Punjab, Harayana & H.P. 39 49 44 6.0 7.2 347 506 Rajasthan 36 37 41 2.0 2.9 284 373 U.P. 46 50 44 3.7 4.2 252 373 West Bengal 55 43 50 15.9 1i.1 390 532 Source: L.K. liazari - Industrial Planning and Licensing Policy, Final Report (Government of India, Planning Commission, 1967)  -19- An important contributory factor - apart from techno-economic superiority of the applications from the advanced states - could be that the big and more established producers in these states had the resources to successfully lobby for allocation of licenses. For example, it was financially less burdensome for them to maintain liaison offices at New Delhi for this purpose. This general conclusion pertains to the period 1959-66 as a whole. The period is too small to permit trend analysis. Nevertheless, it is interesting to compute the relationship between license approvals and state development separately for 1960 and 1965 and compare the two years. The results are presented in Table 10. The picture for 1960 is essentially the same as that for the period 1959-66 as a whole, the license approval ratio having a significant positive correlation with the level of state industrial development while being almost unrelated to state per capita incomes. However, for 1965, there is a dramatic switch - the approval ratio displays a significant negative correlation with state industrial development, while continuing to show little relationship with per capita income. Thisi may be indicative of an increased importance assigned to the objective of allocating licenses on an equitable basis to the states.  -20- Table 10 Correlation Coefficients between State-Wise Proportions of Licenses Approved and Indices of State Development Correlation of state-wise "approved licenses as a proportion of license applications" with 1960 1965 1959-66 3/ Value added in manufacturing GDP for States 0.44 -0.09 0.40 State per capita income 2/ -0.10 -0.50 0.07 Notes 1/ Certain small states and union territories have been left out for lack of data. 2/ In current prices. 3/ The indices of state development used in computing the coefficients are the averages of 1960 and 1965.  -21- However, it would be premature to reach a definitive conclusion until it is possible to extend our licensing data beyond 1966. In our discusslon of industrial licensing, we have focused only on inter-state allocation of licenses. In 1971, another regional dimension was added to the licensing criteria by the decision to prohibit the setting up of licensable industries within certain geographical limits of large metropolitan cities with a population exceeding one million, and urban agglomerations with a population greater than 0.5 million as per the 1971 census. 1/ In 1980, it was decided that undertakings in a large number of industries could increase their capacity by 257 over a five year period without having tco apply for a licence. Moreover, undertakings in certain industries "of importance to the national economy" and "engaged in the production of articles of mass consumption" which had installed capacities in excess of licensed capacity were permitted to regularize these. Both these types of liberalizations were available even to undertakings located in the metropolitan areas mentioned earlier which had been barred from expanding under the 1977 policy. Moreover, this locational policy was also relaxed in certain other cases including those units exporting 1007 of their output. 1/ Earlier, in 1973, it had been decided that units normally exempt from the licensing requir,ement (for example, the small scale sector) would have to apply for a license if they were located in these areas.  -22- (ii) Location of Public Sector Plants Soon after independence, in 1954, the ruling Congress party adopted the resolution of "constructing a socialistic pattern of society" in India. Since then, the industrial policy statements of the Indian government have given considerable priority to the development of the public sector. In 1972, public sector units contributed to 127. of total value added in the manufacturing sector. The Indian goverment has explicitly sought to restrict the production of many types of basic and capital goods to the public sector. Therefore, the public sector is the major supplier of items like iron and steel, heavy electricals, petroleum products, shipping and railway equipment. V/ Direct public investment in manufacturing commenced on a large scale during the Second Fivo-year plan (1951-61). Regional considerations have always been quite important in the location of these units. For example, the Third Five-year plan (1961-66) stated that "in the location of public sector projects, the claims of relatively backward areas have been kept in view wherever this could be done without giving up essential technical and economic criteria. The location of several important projects like the steel plants have been determined on the basis of expert study and on economic considerations. But as they are situated in areas which were hitherto industrially backward, the latter will benefit." (3rd Five-year plan, p. 145) 1/ For example, in the early '70s, the share of the public sector was over 90% of output of saleable pig iron, heavy electrical equipment like synchronous condensers, steam and hydro turbines, generators and alternators, ships and locomotive boilers.  -23- The shares of the states in cumulative and annual investment in central public undertakings during 1963-78 are exhibited in Tables 11 and 12. In 1963, about 80* of this investment was concentrated in the four states of Bihar, Nadhya Pradesh, Orissa and West Bengal. Since 1963, many important public undertakings have come up in the other states, so that the distribution across states has become somewhat more balanced. By 1978, the share of thet four states (just mentioned) in cumulative investment had come down to 57*. We attempted an empirical test of the importance of regional considerations in the location of public sector units. Just as in evaluating industrial licensing, this was done by correlating Investment in Central Government public sector undertakings per capita for each state with indices of state backwardness like per capita income and the share of manufacturing in state income. The results are presented in Table 13. It appears that before 1970, states with low per capita incomes have been favored in their share of public sector projects. However, as the just-cited quote from the Third Plan suggests, this result may be explained by the plentiful supply of appropriate raw materials in the backward states rather than explicit locational considerations. To illustrate this, consider the composition of the public sector in the major low-income states favored before 1970 - Bihar, Madhya Pradesh and Orissa. One finds that in 1969, steel plants accounted for one-third of cumulative investment in Bihar, two-thirds in Madhya Pradesh and 90% in Orissa.  -24- Table 11 Share of States in Cumulative Investment (Gross Block) in Central Public Undertakings (%) A.P. Assam Bihar Gujarat Karnataka Kerala Harayana Punjab 1963 0.8 1.5 15.9 0 3.6 0.2 - - 1964 1.0 1.8 16.9 0.1 3.5 1.5 - - 1965 1.5 2.1 17.7 0.1 3.0 1.8 - - 1966 2.4 1.5 17.0 1.4 2.9 2.0 - - 1967 3.0 1.4 17.2 1.4 2.9 2.2 - - 1968 2.9 2.0 17.8 2.9 2.7 2.6 0.3 1.2 1969 2.9 2.1 20.8 2.9 2.7 3.4 0.2 1.1 1970 2.9 2.2 23.0 2.9 2.7 3.4 0.2 1.0 1971 3.0 2.1 24.9 4.2 2.7 3.4 0.2 0.9 1972 3.1 3.1 25.9 4.3 2.8 3.3 0.2 0.9 1973 3.5 3.2 26.9 4.7 2.9 3.2 0.2 0.8 1974 3.8 3.0 27.8 4.7 3.0 3.3 0.2 0.8 1975 4.3 3.2 26.8 4.8 3.0 3.2 0.3 1.2 1976 4.2 3.6 25.1 5.8 2.8 3.3 0.7 2.2 1977 4.2 3.4 27.2 5.7 2.9 3.0 1.5 2.1 1978 4.4 3.3 25.2 5.6 3.7 2.9 1.3 2.0 Punjab & Tamil M.P. Maharashtra Orissa Harayana Rajasthan Nadu U.P. W.B. Total 1963 23.7 2.0 21.4 2.8 0.2 8.3 0.2 19.4 100.0 1964 23.5 2.6 19.2 2.6 0.2 9.1 0.7 17.4 100.0 1965 23.9 2.7 17.4 2.4 0.2 9.3 1.8 16.2 100.0 1966 23.3 3.1 15.2 2.0 0.2 9.1 2.6 17.2 100.0 1967 21.8 3.2 14.3 1.8 0.3 9.3 3.5 17.7 100.0 1968 19.6 3.5 14.8 1.5 0.7 8.9 4.8 15.3 100.0 1969 18.2 3.4 14.2 1.3 0.9 8.8 4.6 13.8 100.0 1970 16.5 3.6 13.4 1.2 1.0 9.2 4.6 13.5 100.0 1971 15.5 3.5 12.6 1.1 1.1 8.8 4.3 12.7 100.0 1972 14.8 3.6 11.7 1.1 1.4 8.4 4.2 12.6 100.() 1973 14.0 3.9 11.0 1.0 1.8 7.5 4.0 12.5 100.0 1974 13.8 3.9 10.1 1.0 2.2 6.9 4.1 12.3 100.0 1975 13.4 4.9 9.3 1.6 2.6 6.2 4.1 12.6 100.0 1976 18.2 5.0 8.3 2.9 2.5 6.7 4.1 7.6 100.0 1977 16.2 6.8 7.0 3.7 2.5 5.1 4.1 8.3 100.0 1978 16.1 8.1 5.9 3.3 2.5 5.1 4.4 9.5 100.0 Source: Bureau of Public Enterprises, Annual Report on the Workinz of Industrial and Commercial Undertakings of the Central Government- various issues.  -25- Table 12 Share of States in Annual Investment in Central Public Undertakings (%) A.P. Assam Bihar Gujarat Karnataka Kerala Harayana Punjab 1964 2.1 3.1 21.6 0.3 2.8 7.9 1965 3.7 3.5 22.2 0 0.8 3.1 1966 7.5 -1.9 12.9 8.7 2.2 3.2 1967 8.1 0.9 18.6 1.6 2.9 4.3 1968 2.5 4.7 20.6 9.4 1.7 4.2 1969 2.9 3.0 44.1 2.6 2.6 9.5 0 0 1970 2.8 2.7 39.6 2.6 2.6 3.7 0 0.3 1971 4.4 1.1 42.8 16.3 3.0 2.8 0.2 0.2 1972 3.5 12.6 35.4 5.3 3.3 2.9 0.1 0.2 1973 6.1 4.0 34.3 7.7 3.6 2.3 0.2 0.3 1974 6.9 1.3 35.2 5.1 4.5 3.9 0.3 0.9 1975 6.9 4.3 21.3 5.4 2.8 3.0 0.9 3.7 1976 3.3 5.8 16.8 10.5 2.0 3.6 2.6 7.0 1977 4.6 2.4 36.0 5.2 3.2 1.6 5.2 1.9 1978 5.1 3.0 15.9 5.4 7.6 2.7 0.1 1.4 Punjab & Tamil M.P. Maharashtra Orissa Harayana Rajasthan Nadu U.P. W.B. Total 1964 22.4 5.3 8.8 1.7 0.1 13.0 3.3 7.5 100.0 1965 26.0 3.1 8.0 1.1 0 5 10.4 7.2 10.1 100.0 1966 19.9 5.7 3.1 0.1 0.2 8.1 7.1 23.1 100.0 1967 9.4 3.2 6.6 0.1 1.1 10.4 11.1 21.7 100.0 1968 10.4 5.0 17.0 0.1 2.1 7.5 10.0 4.8 100.0 1969 7.3 2.4 9.3 0 2.9 7.7 3.3 2.4 100.0 1970 3.4 4.9 7.6 0.3 1.6 12.7 4.3 12.0 100.0 1971 6.5 3.0 5.0 0.4 2.1 5.0 2.2 5.4 100.0 1972 7.8 4.1 2.8 0.3 4.0 3.7 2.8 11.6 100.0 1973 8.3 6.1 6.4 0.5 4.5 1.5 2.3 12.2 100.0 1974 12.3 3.8 2.0 1.2 5.6 2.2 5.4 10.7 100.0 1975 11.3 10.7 4.9 4.6 4.7 2.0 4.1 14.1 100.0 1976 42.1 5.2 3.4 9.6 2.2 9.1 3.9 -17.5 100.0 1977 7.3 14.9 1.5 7.1 2.3 -1.8 4.1 11.6 100.0 1978 15.6 14.5 0.4 1.5 2.6 5.0 5.7 15.1 100.0 Source: Bureau of Public Enterprises, op. cit.  -26- TABLE 13 Correlation Coefficients Between State-Wise Per CapitalInvestment In Central Public Undertakinms and Indicators of State Development Per Capita Per Capita Aggregate Cumulative Investment Investment durinR 1965 1970 1975 1950-65 1965-70 1970-77 State per capita incomes -0.44 -0.52 -0.50 -0.44 -0.56 -0.16 Share of Manufacturing in state GDP -0.01 * -0.1 0.01 -0.01 -0.01 -0.04 a/ Per Capita income, share of manufacturing in GDP and population figures used to compute the correlations are the averages of the end-points of the time period.  -27- Clearly the availability of iron ore and coal was the major reason for the location of the steel plants in these states. In this context, it is interesting to find that during 1970-77, the correlation of per capita investment in public undertakings with state per capita income turns out to be extremely weak (though negative). Thus, it could be hypothesized that once the steel plants had been constructed and the public sector started diversifying into other types of products (Table 14) where the techno-economic advantages of different states were more balanced, the link between investment and regional backwardness becomes tenuous. This argument seems to be further buttressed by the low correlations of public investment with the share of manufacturing in state GDP. However, a more important explanation for this result is that the operation of the public sector plants in the backward states has directly contributed to their industrial development. This is quite clear from Table 15. One finds that in states like Assam, Bihar, Madhya Pradesh and Orissa, employment in Central Public Undertakings constitutes more than 307 of total employment in the factory sectors. Apart from the direct contribution of public enterprises to the state economy through their production and employment, they also promote industrial development through their backward (input demand) and forward (supply of output) linkages with the other sectors. 1/ These are difficult to measure with the exception of one type of linkage. Public sector undertakings have made special attempts to develop small-scale ancillary units to suplply various kind of components and sub-assembled 1/ Moreover, public sector units have invested significantly in townships for their employeeis as a means of fulfilling broader social goals. The spatial implications of these townships have not been analyzed since they fall somewhat outside the scope of our paper.  -28- Table 14 Distribution of Cumulative Public Sector Investment (7) 1965 1970 1977 Steel 44 33 24 Engineering Products 18 24 11 Chemicals 10 11 20 Mining and Minerals 8 9 19 Petroleum 12 9 6 Others 8 14 20 Total 100 100 100 Source: Bureau of Public Enterprises, op. cit. products, and, in many cases, the undertakings have set up industrial estates for these units with infrastructural and technical facilities. Information relating to the orders placed by the public sector plants with these small entrepreneurs is available and is presented in Table 16. One finds that the contribution made by the public sector in promoting these small scale units is still rather small, with the most significant impact in Orissa, Karnataka, Bihar and Madhya Pradesh.  -29- Table 15 Share of Employment in Central Public Undertakings in Total Employment in Manufacturing & Mining (%) A.P. Assam Bihar Gujarat Haryana Karnataka Kerala M.P. 1973 9.9 30.8 31.7 3.5 0 22.9 4.6 14.0 1974 11.0 29.7 38.2 3.5 0 24.3 5.9 35.8 1975 12.5 30.1 39.8 4.2 0.1 26.8 7.1 37.0 Tamil Maharashtra Orissa Punjab Rajasthan Nadu U.P. W.B. J & K 1973 1.9 34.1 6.1 4.3 7.5 5.1 6.3 0 1974 9.4 34.4 6.6 6.8 7.5 17.9 7.6 6.6 1975 12.7 36.2 8.4 9.1 9.5 20.4 8.5 8.2 SOURCE: Bureau of Public Enterprises, op.cit.  -30- Table 16 Orders placed by Public Sector Undertakings to Small Scale Units as a Percentage of Output of Mbdern Small Scale Sector 1974-75 1975-76 A.P. 1.8 1.2 Bihar 1.0 2.2 Gujarat 0.7 0 Harayana n.a. 0 Karnataka 1.9 4.1 Kerala n.a. 0.1 M.P. 1.7 4.4 Maharashtra 0.1 0.2 Orissa 6.6 0.5 Rajasthan 0.2 0.2 Tamil Nadu 0 4.1 U.P. 0 0.3 West Bengal 0.9 0.5 SOURCE: Data on orders by public sector undertakings in Bureau of Public Enterprises, op.cit. Data on small sector output from Ministry of Industries, All-India Report on the Census of Small Scale Industries, 1976. The coverage of the census was restricted to units registered with the Directorates of Industries. These comprise essentially the 'modern' small- scale sector (and exclude the large traditional sector operating in rural areas and small towns). Small scale units are defined as those with capital of less than Rs. 750,000 in original value of plant and machinery. The Census relates to 1972/73. However, the Report estimates total output for 1974/75 and 1975/76. We have applied this growth rate to the output for individual states during 1972/73 to derive the outputs for 1974/75 and 1975/76.  -31- (iii) Distribution and Pricinz Policies for Basic Industrial Inouts There are certain products that are crucial for the establishment and operation of industrial projects; chief among these in the Indian context are cement, steel and coal. As with any other comodity, regional production of these inputs is determined by factors like the availability of raw materials, transport costs and proximity to markets. The Indian Government has followed a policy of control on the prices and distribution of these cozmodities with the objective of regional equity. Given the crucial nature of these products in industrialization, the official justification is that states blessed with easy availability of these inputs should not gain in relation to others. The mechanism Of operation of these controls and their economic effects will be illustrated with the case of cement. The regional distribution of cement capacity is presented in Table 17. Table 17 Region-wise Installed Capacity of Cement jfl Year North East West South All - India 1971 17.6 16.9 28.6 36.9 100 1975 17.5 17.6 29.3 35.6 100 Source: Govt. of India, Office of the Cement Controller: Cement Production and Despatches, Various issues. Note : States covered in different regions. North - Haryana, Himachal Pradesh, Jaiu and Kashmir, Punjab, U.P., Chandigarh, Delhi and Rajasthan. East - Bihar, Orissa, Meghalaya, West Bengal, Manipur, Nagaland, Sikki, Tripura, Arunachal Pradesh Mizoram and Assam. West - Gujarat, Kaharashtra, M.P., Goa, Daman, Diu, Dadra and Nagar Haveli. South - Tamil Nadu, A.P. Karnataka, Kerala, Pondicherry and Andaman, Nicobar and Lacadive Islands.  -32- Production of cement is concentrated in the Western and Southern regions which accounted for almost two-thirds of cement capacity in 1971 and 1976. Furthermore, within these regions, there is considerable variation in production by individual states. The price control exercised by the Government to balance these regional variations takes the following shape. In the first instance, the Government fixes the 'retention price' - the price payable to the manufacturer. Second, through suitable subsidies, railway freight charges for the transportation of cement to all destinations are kept the same. Thus, the F.O.R. (free on rail) price of cement is uniform in all parts of the country. However, the actual retail prices vary according to the incidence of state taxes. These prices and distribution controls have resulted in various kinds of distortions of the market mechanism and led to certain irrationalities in the industrial location process. The most important effects are explored below. . First, the "freight-pooling" arrangement has obvious implications for the location of cement plants. Since transportation charges to markets are the same irrespective of distance, entrepreneurs will locate their plants in accordance with raw material availability - the major ones being limestone, gypsum and coal. As a corollary, these controls lead to the disappearance of natural protection in the form of transportation charges that would have led to some development of cement capacity in states less well-endowed with these raw materials. Second, freight-pooling was introduced in 1966. Thus, cement units set up earlier in proximity to markets have found themselves at a disadvantage vis-a-vis new units.  -33- Third, there is less of an incentive for the utilization of substitutes for cement in "deficit" states. For instance, in the North Eastern Region, states like Kashmir and other areas with plentiful timber resources, the utilization of these is sub-optimal because of the artificial lowering of cement prices. Fourth, there is8 an excessive use of railway transport especially to distant consuming areas. This is particularly important in the context of frequent bottlonecks in transport that have affected Indian industrial growth. Fifth, a fiscal burden is imposed on the Government to the extent that a subsidy has to be provided to support this freight equalization mechanism. Finally, the transportation subsidy reduces the scope for "split location" of clinker and grinding plants, i.e. setting up of clinker plants near raw materials and grinding plants at consuming centers. A major advantage of this division is saving in transportation costs (because of reductions in tonnage that needs to be transported) and convenience in transportstion (clinker can be dispatched in bulk in open wagons). Because of the subsidy on transport, these advantages are not enough to outweigh the additional expenses involved in setting up grinding facilities in consuming areas. There has been it persistent shortage of cement in India despite the ready availability of' essential raw materials. There is little doubt that price and distribution controls are important reasons for this phenomenon. Cement produc:ers have resented the "unremunerative" prices  -34- fixed by the Government. 1/ Besides this, irrationalities in the location of production and utilization of cement resulting from these controls have also played their part. It is not easy to assess the precise locational impact of the freight equalization schemes without a detailed study of the regional distribution of raw materials, markets and other factors influencing the industrial location process and the policy changes and their linkage in a general equilibrium context. It is obvious that, in the case of cement, the concentration of capacity in the Southern and Western Regions implies that freight equalization has benefited the Northern and Eastern Regions. 2/ On the other hand, for steel and coal, the Eastern Region which dominates in production has lost vis-a-vis the rest of India. A careful analysis of these policies is urgently called for to eliminate some of the resulting irrationalities and to gauge the extent of efficiency loss in relation to the gains in regional equity. 1/ Recently, in February 1982, the Government introduced a policy of partial decontrol of the cement industry whereby units could sell cement output in excess of two-thirds of their capacity at free market prices. In the short while that this policy has been in force, available information indicates that there has been a dramatic improvement in the supply situation of cement. Moreover, a number of companies have started diversifying into the production of cement in response to the increased profitability of this industry. 2/ For a detailed analysis, see National Council of Applied Economic Research. Cement Industry in India, 1979, New Delhi.  -35- (iv) State Government Incentives A variety of ilncentives are offered by individual states to attract entrepreneurs to establish industries within their borders. There are also state-level financial institutions and industrial development agencies whose major taisk is the provision of industrial finance and promotion of industry. The major incentives fall into three broad categories: 1/ a. Reduction in capital cost and infrastructural expenditure Provision of financial assistance (both in the form of loans and equity); infrastructural facilities and availability of plots and sheds in industrial estates; assistance in acquisition of land; and assistance in the preparation of feasibility reports. b. Incentives related to the prices and availability of inputs Reductions in power tariffs and electricity duties; financial assistance to units for setting up power generating sets; reduced water rates; and assistance in the procurement of controlled raw materials. c. Tax incentives Concessions (on loans) for payment of sales taxes, octroi stamp duties and property taxes. These three types of incentives exist in almost every state. However, the precise content of each incentive varies from state to state. It would be interesting to determine the extent to which these differences have influenced the decisions of entrepreneurs In plant location. Unfortunately, there has been no research study of this nature essentially because of the paucity of detailed information required for a rigorous analysis. For example, one needs state-level data relating to 1/ The magnitude of the incentives are often increased further for small scale units or units set up in "backward" districts.  -36- land rents, location of estates, power tariffs, water rates, tax levels etc. The task is also complicated by the fact that these incentives keep changing over time for every state. A priori, it can be argued that if the Governments of individual states were serious enough about industrialization within their boundaries, they would not allow the overall attractiveness of their incentives to fall much below those offered by other states. However, "rationality" cannot always be taken for granted in Governmental decision making. Moreover, there are other objective factors which influence the offering of incentives - the budgetary position, for instance. Finally, it is not merely the announcement of incentives but the actual effectiveness with which they are implemented that will influence industrial location. Here again, there is little doubt that the efficiency of the implementing agences varies across states. Finally, there is the similarity between state incentives and the phenomenon of different developing countries competing with each other to attract foreign investment. The advantages that are derived from inter-country co-operation in this analogy may be of some relevance. Otherwise, there is the danger that unhealthy competition between the state Governments will unnecessarily drain state budgetary resources (in the form of transfers to the private sector) with no net gain from a national viewpoint. 2. Policies Influencing the Intra-Regional Distribution of Industry Having looked at the locational policies aimed at a more equitable inter-state distribution of industry, we now turn our attention to those affecting intra-regional industrial growth. Chronologically arranged in the order of their implementation, these consist of: (i) policies for encouraging (a) village and cottage industries and (b) modern small scale enterprises;  -37- (ii) the indusltrial estates program; (iii) The Rural Industries Project Program; (iv) metropolitan planning in the major cities; (v) incentives to promote industrial development in backward districts: (a) Central Government incentives, (b) Concessional finance from all-India financial institutions and (c) State Government incentives. 1/ (i) Policies to Encourage the Small Scale Sector A major aspect of Indian industrial policy has been its emphasis on encouraging the small scale sector. The reasons for this are summed up in the following quote from the Industrial Policy Resolution of 1956: 'They (village and small scale enterprises) provide immediate large scale employment, they offer a method of ensuring a more equitable distribution of national income and they facilitate an effective mobilization of resources of capital and skill which might otherwise remain unutilized. Some of the problems that unplanned urbanization tends to create will be avoided by the establishmaent of small centers of industrial production all over the country.' Thus, the desire to promote a decentralized form of development is quite central in the rationale for supporting the small scale sector. The small scale sector can conceptually be broken up into two categories - the traditional village and cottage industries operating in the rural and semi-urban areas and modern small scale enterprises. It is extremely difficult to draw a precise dividing line between the small scale sector and the rest of the manufacturing sector. The Government uses capital investment as the criterion for defining the modern small sector. The precise 1/ We have described State Government incentives in an earlier section  -38- cut-ott has obviously been increasing with time. In 1963, when factories with fixed capital less than five lakhs V/ comprised the small sector, the latter contributed to 20.77 of factory value added. In 1973-74, the limit was raised to Re. 7.5 lakhs. By this criterion, in 1973 and 1974, the sector's share in value added was 17.37 and 14.67 respectively. In 1975-76, the figure was again increased to Rs. 10 lakhs. The sector's contribution in this year was 16.17. Obviously because of these arbitrary changes in definitions, these figures cannot be used as indicative of trends in the development of the modern small scale sector. A crude index of trends in the small scale sector as a whole (both modern and traditional) is provided by the share of the "unregistered" sector (i.e. units employing less than ten workers with power or 20 workers without power) in national income. This is displayed in Table 18; in 1975, the share of the unregistered sector was as high as 37t. Table 18 Share of Unregistered Sector in Value Added in Manufacturing Year Share (%) 1965 39.5 1966 39.4 1967 41.4 1968 40.7 1969 38.0 1970 37.3 1971 38.1 1972 38.2 1973 37.4 1974 36.7 1975 37.4 Note: The unregistered sector consists of units employing less than ten workers with power or 20 workers without power. Source National accounts statistics presented in Reserve Bank of India, Report on Currency and Finance. I/ A lakh is a unit used in India and is equal to 100,000.  TABLE 19 Employment and Output in Modern Small Scale Sector, 1972 EMPLOYMENT GROSS OUTPUT Ratio of Ratio of Number Percentage small scale sector Value Percentage small scale sector to (000s) distribution to factory sector(%) (Rs.crs)-/distribution factory sector(%) Food Products 131.2 7.9 1.9 152.3 5.9 4.5 Beverages 4.6 0.3 2.3 7.4 0.3 1.4 Hosiery & Readymade garments 75.3 4.6 393.8 155.4 6.0 322.0 Wood Products 94.7 5.7 125.5 102.6 3.9 81.9 Paper Products, Printing, etc. 89.1 5.4 35.5 126.4 4.9 20.9 Leather Products 31.8 1.9 71.3 88.6 3.4 49.2 Rubber & Plastic Prods. 81.7 4.9 87.7 151.2 5.8 62.5 Chemicals 159.0 9.6 49.5 346.7 13.3 15.1 Glass & Ceramics 202.3 12.2 250.4 125.5 4.8 106.6 Basic Metal Indus. 109.6 6.6 24.0 294.1 11.3 14.4 Metal Products 300.1 18.2 170.8 469.1 18.0 87.9 Machinery & parts 145.3 8.8 44.8 211.0 8.1 20.3 Electrical & Electronic Prods. 65.9 4.0 26.1 151.7 5.8 13.6 Transport equip. 83.5 5.1 21.3 134.8 5.2 12.0 Miscellaneous 40.0 2.4 - 62.7 2.4 - Repairing, servicing & job work 39.0 2.4 26.3 23.3 0.9 9.8 TOTAL 1653.2 100.0 31.3 2602.7 100.0 14.1 2/ Crore is a unit used in India and is equal to ten million. SOURCE: Computed from Development Commissioner, Small Scale Industries (Ministry of Industry & Civil Supplies), All-India Report on the Census of Small Scale Industries (1976).  -40- In sharp contrast to the concentration of traditional enterprises in simple consumer goods production, modern small scale units are engaged in manufacturing a wide variety of commodities (Table 19). In fact, basic metals, metal products, machinery and transport equipment are the most important activities, together accounting for 437 of employment and 487 of gross output in this sector. Despite the problems of definition, there is no doubt about the critical importance of trends in the small sector in an analysis of Indian industrial development. Within the small scale sector, it is the traditional industries that dominate. Because of the labor-intensive character of production in traditional enterprises, their share in employment is striking. The 1971 census estimated that as much as 77.4* of the labor force in the manufacturing sector was employed in household units. The most important activities in traditional enterprises are textiles and textile products; in 1971, almost one-third of household workers were engaged in this activity. Another indication of the significance of these industries is the estimate that, in 1976, about 50* of total cloth output in the country was produced by the decentralized sector, i.e., the non-mill sector where the utilization of handlooms is quite common. Apart from textiles, other major household industries are agro-processing (like the processing of cereals and pulses, &ur 1/ and khandsari 1/), carpentry and blacksmiths, coir and coir products, sericulture and handicrafts. The Indian Constitution assigns major responsibility for development of small scale industries to the State Governments. However, the Government of India through the Small Scale Industries Development Organization (SSIDO), renders advice and coordinates the state programs. The SSIDO also performs an 1/ Crude varieties of sugar.  -41- industrial extension service to small scale enterprises by providing access to improved technology and products and imparting training to them in management, finance and marketing. Two other important agencies set up by the Government of India to assist the Small Scale sector are the National Small Industries Corporation (NSICO)and the Small Industry Extension Training Institute (SIETI). The NSICO performs the following duties: 1. it supplies machinery on a hire-purchase basis; 2. it acts as a contractor for Government orders; 3. it operates EPrototype Production and Training Centers; 4. it distributes scarce items on a bulk basis; and 5. it markets thle export of products produced by the small scale sector. The SIETI conducts research on small scale enterprises and provides training to officers concerned with extension services. These three organizations are primarily concerned with modern small scale industries. Responsibility for household and village industries 1/ is assigned at the Central Government to the Khadi and Village Industries Comnission. At the State Government level, the Director of Industries is in overall charge of development of all industries (both large and small). As we have stated earlier, the most important activities in traditional enterprises are textiles and textile products most of which are produced on handlooms. The All-India Handloom Board assists the handloom textile industry through a variety of policies: 1/ Excluding textiles other than "khadi" (crude handspun cloth).  -42- 1. reservation of certain types of cotton textiles for production by the handloom industry; 2. encouragement of co-operative societies; 3. loans to weavers to subscribe to the share capital of the co-operatives; 4. supply of yarn to the co-operatives; 5. schemes for improvement of handlooms; 6. aid for marketing and rebates on sale of handloom cloth; 7. training facilities for weavers. Especially after the nationalization of the major commercial banks in 1967, the financial needs of the small scale sector are being taken care of on a priority basis, with concessionary interest rates and collateral. In the states, the State Financial Corporations make loan capital available to small industry. Finally, since 1978, a number of "District Industries Centers" have been set up as district-level governmental organizations to oversee the development of the small scale sector. Up to this point, we have examined policies operating mainly through Government expenditure on the small scale sector. However, many of the policies are in the form of controls. For example, the licensing system bans the production of certain items in large factories, reserving them for the small sector. Moreover, small scale enterprises do not have to go through the licensing system to set up new units or to expand capacity. These sorts of restrictions are also important to protect the handloom sector. As we pointed out, certain items of clothing are reserved for production by handlooms. In addition, the Government also directs spinning mills to allocate part of their yarn production to satisfy the raw material requirements of the handloom sector.  -43- Having examined the general state policies used to encourage the small scale sector, Wv now turn our attention to policies specifically targeted at the location of small scale units - namely, the industrial estates program and tlhe rural industries projects program. (ii) Industrial Estate Proaram Industrial esstates have been defined as "a group of factories constructed on an economic scale in suitable sites with facilities of water, transport, electricity, steam, bank, post office, canteen, watch and ward and first aid and provided with special arrangements for technical guidance and common service facilities." 1/ The Indian Government has taken the initiative of setting up industrial estates with two objectives in mind. One is the encouragement of the small scale sector, through the provision of various infrastructural facilities and economic incentives which are essential characteristics of the industrial estate. The second igi the dispersal of industry away from central metropolitan regions through suitable location of the industrial estates and their catalytic influence on economic activity in their area of operation. The industrial estates program was initiated in 1955, toward the end of the First Five-Year Plan (1951-56). The first industrial estate to commence operation was the one at Rajkot (in Gujarat). The program gained signlficant momentum during the Second Five-Year Plan (1956-61). The Plan made a provision of RE. lllm. for the construction of 110 industrial estates, of which Rs. 104 m. was actually spent. By the end of the Plan, 67 estates were constructed of which 53 were actually functioning. In l/ Alexander, P.C. Industrial Estates in India, Bombay: Asia Publishing House, 1963. In practice, an industrial estate does not have all these characteristics  -44- terms of factory sheds, 2077 had been completed though only 1049 were operating The geographical distribution of the 119 estates sanctioned during the First and Second Plans is as follows: 1/ Places with population of less than 20,000 25 Places with population of 20,000 - 50,000 25 Places with population of 50,000 - 100,000 21 Places with population of over 100,000 48 Total 119 The momentum of the industrial estates program was maintained during the Third Plan (1961-66), which allocated Rs. 282 m. for this purpose. The actual expenditure incurred during the plan amounted to Rs. 227 m., which was more than double the expenditure of the Second Plan period. 2/ 216 estates were constructed during 1961-66, of which 145 were actually functioning. The Third Plan placed a much greater emphasis on utilization of industrial estates as an instrument for dispersal of industry in towns and rural areas. For this purpose, the plan fostered the idea of "rural Industrial estates," i.e., those in villages with less than 5,000 inhabitants and located at a "sufficient" distance from large cities and towns. The Plan also subdivided the other individual estates into "urban" estates (in cities and towns with a population exceeding 50,000) and "semi-urban" estates (in small towns with population between 5,000 and 50,000). At the end of the Third Plan, out of the 198 functioning Industrial estates, 187 were rural, 307 were semi-urban and V/ From Bandyopadhaya, 1969 2/ This comparison is unadjusted for prices.  -45- 52* were urban. During 1966-69 (which was the period of the three Annual Plans ), Rs. 76 m was spent on industrial estates. In 1969, there were 285 functioning industrial estates of which 227 were rural, 327 semi-urban and 467 urban. During the mid-sixties, a number of evaluation studies were undertaken to examine the working of the industrial estates during the previous decade. 1/ 'rhese studies cast grave doubts on the extent to which the industrial eastates program was achieving its twin goals of promoting industrial development and industrial dispersal. This was reflected in non-occupation of the constructed sheds and only partial functioning of the occupled sheds. Table 20 shows that during the sixties, more than 401L of the completed sheds were not operating. The rural industrial estates fared the worst; in 1969, 50* of the completed sheds were functioning. The estates in semi-urban areas did not fare much better with only 57* of sheds operating. The urban industrial estates performed much more satisfactorily; 767 of the completed sheds were working in 1969. The reasons for this disquieting performance, particularly of the rural and semi-urban industrial estates, have been made clear by the evaluation studies. The most important factor was the wrong choice of location for the industrial estate. In many cases, careful techno-economic surveys were not conducted to ascertain the suitability of the area in terms of proximity to markets, availability of raw materials, labor, transportation facilities, etc. For example, in their enthusiasm to promote the program, 1/ These included studies by the State Governments, the Central Small Industries Organization and the Estimates Committee (Lok Sabha).  -46- TABLE 20 Percentage of Sheds Completed but not functioning December 1961 48 December 1962 43 March 1964 38 March 1965 40 March 1966 41 March 1967 42 March 1968 41 March 1969 41 March 1972 35 March 1974 29 SOURCE: D. Nagaiya (1971), pp.41 Bharti (1978) Sanghvi (1979)  -47- certain states decided to locate one or two estates in every district. "In many of the states, it was considered a matter of political prestige to locate an estate in each district as the district was not considered to have come of age otherwise." 1/ Another case olf economic irrationality was the response to the general advisability of locating the estates outside municipal limits in order to encourage dispersal. While this made sense for cities and large towns, some states decided to locate the estates two to four miles away from the center even in the case of small towns. This inevitably resulted in serious bottlenecks in transport facilities and the supply of other essential inputs like water and electricity. There were many other cases of poor pre-investment planning. After a critical examination of the operation of the industrial estates program, the Estimates Committee noted that "a little delay in the selection of a location in the beginning is a small price compared to the waste of public funds and private resources resulting from the wrong location of an industrial estate." 2/ The problems encountered in the choice of location reflect the limitations of using industrial estates as tools in the industrialization of backward areas. Merely setting up an estate in a backward region is quite insufficient; there has to be coordinated provision of infrastructural investments in transport, communications, power, water supply, etc. Otherwise, it is quite unprofitable for entrepreneurs to set up production units in rural and semi-urban areas. In fact, this has been a major problem encountered in the implementation of the industrial 1/ Nagaiya, 1971, p. 66 2/ Government of India, Estimates Committee of the Third Lok Sabha, 106th Report, New Delhi, 1'966.  -48- estates program. In some areas, the "District Industries Officers had to practically plead with industrialists to set up industries in the newly constructed estates." 1/ Another disconcerting feature of the performance of the program has been the inordinate delay that has characterized the various stages of implementation, like administrative approval of the scheme, land acquisition, construction of sheds and other facilities, allocation and occupation of sheds and the commencement of production. These have arisen because of administrative inefficiency and poor coordination between the various agencies involved. The poor performance of the industrial estates program aroused a great deal of concern, and policies followed since the mid sixties have emphasized consolidation rather than expansion in the number of estates. In contrast to the actual expenditure of Rs. 226 m. on the program during the Third Plan, the estimated expenditure in the Fourth Plan (1969-74) was only Rs. 157 m. (in current prices). While expenditure on industrial estates amounted to 12, 19 and 24* of total expenditure on small scale industries 2/ in the First, Second and Third Plans respectively, the share during the Fourth Plan was only 16S. The outlay for the Fifth Plan (1974-79) is Rs. 210 m., which is only 9.57 of total expenditure on small scale industries. By 1975, there were 469 functioning estates of which 227. were rural, 297. semi-urbin and 49S urban. However, the performance of the rural and semi-urban industrial estates has still been rather unsatisfactory. In 1974, only 457 of the V/ Nagaiya, 1971, p. 71 2/ i.e., expenditure on modern small scale industries, industrial estates and rural industries projects.  -49- completed sheds in rural estates were operating (the corresponding ratios for urban and semi-urban estates were 817 and 667 respectively). The impact of the industrial estates program can be assessed by relating production and employment on the estates to overall trends in the industrial sector. In comparison with the entire registered sector (i.e., factories employing more than ten workers with power or 20 workers without power), output and employment on the estates are rather small (Table 21). However, these shares have been gradually increasing over the sixties and, in 1969, amounted to 1.0 and 1.6 percent respectively. By 1974, the share in total output and employment had gone up further to 1.7% and 3.07.. The Annual Surveys of Industries conducts yearly sample surveys for enterprises employing between 10-49 workers (with power) or 50-99 workers (without power). We can treat this "sample sector" as the modern small-scale sector (in contrast with the traditional village and household units that constitute the "unregistered" sector). The share of industrial estates in sample sector output amounted to six percent in 1969. The share in employment was more significant - nine percent in 1969 and 167 in 1974. (iii) Rural Industries Projects The Rural Industries Projects program was initiated in 1962-63 to promote rural industrialization in certain parts of the country. This objective is to be realized by "identification and moLivation of prospective entrepreneurs, guiding them in selecting suitable products and appropriate technology, lproviding them technical extension services and helping them in obtaininlg their inputs including credit facilities, raw materials and skilled labor." 1/ The program was to be coordinated with 1/ R.V. Rao, 1978, p. 31  -50- TABLE 21 Share of Industrial Estates in Industrial Employment and Output (X) Share (Z) of Industrial Share (%) of Industrial Estates in Employment in Estates in Output of Sample/a Factory /b Sample Factory Sector Sector Sector Sector 1961 2.2 0.4 1961 1.6 0.3 1962 3.3 0.6 1962 2.2 0.4 1963 n.a. n.a. 1963 n.a. n.a. 1964 4.0 0.6 1964 3.1 0.5 1965 6.6 1.0 1965 3.7 0.5 1966 7.0 1.1 1966 4.5 0.8 1967 9.4 1.6 1967 6.5 1.1 1968 8.3 1.4 1968 6.5 1.1 1969 9.3 1.6 1969 6.0 1.0 1974 16.0 3.0 1975 n.a. 1.7 NOTES: /a Sample sector consists of units employing between 10-49 workers Twith power) or 50-99 workers (without power). /b Factor sector consists of units employing more than 10 workers Twith power) or more than 20 workers (without power). SOURCE: Computed from data in Nagaiya (1971) and Annual Surveys of Industries, various issues.  -51- other area development plans so as to achieve integrated regional development. However, the rural industries projects (R.I.P.) program was conceived essentially as a pilot scheme to gain experience and to identify the crucial factors for promloting rural industrialization. Initially, the program covered 49 areas each with a population of about 350,000. These areas were chosen on the basis of the following criteria: "(i) areas with favorable agricultural conditions and considerable organized agricultural effort and also a heavy pressure of population. (ii) areas with agriculture undertaken mainly under unirrigated conditions and consiiderable need for additional employment; (iii) areas with considerable under-employment because of unfavorable natural conditions and lack of development potential; (iv) tribal and otber backward areas; (v) areas with large industrial projects established or to be established; or (vi) areas in the neighborhood of rural universities and institutes." 1/ During the Thircl Plan (1961-66), an expenditure of Rs. 48 m. was incurred on the R.I.P. program, which constituted 2.27 of total public expenditure on village and small scale industries (Table 22). In the Fifth Plan (1974-79), planned eixpenditure had risen to Rs. 211 m., and the share to 3.97. In 1969, the value of production in rural industries projects formed only 1.1 and 0.27 of output in the sample and factory sectors respectively. In 1973, these proportions were almost exactly the same (1.2 and 0.27). Because of the labor-intensive nature of the products manufactured in the rural industries projects and the technological processes utilized by them, 1/ O.P. Jain, 1975, p. 49  -52- their contribution is much more significant in terms of employment - 13.1 and 2.3* in relation to sample and factory sector employment. By 1973, these shares had risen to 15 and 2.87. In the Fifth Plan (1974-79), the decision was taken to extend the R.I.P. areas to entire districts, excluding towns with population exceeding 15,000 (1961 Census). 1/ By 1975-76, the number of R.I.P. projects had more than doubled from the original 49 to 111. In 1974, the Programme Evaluation Organisation (PEO) of the Planning Commission conducted an evaluation or the working of the rural industries projects. 2/ This study brought out a number of deficiencies in the working of the program. A key component of the program was an industrial potential survey of the area to guide the priorities of assistance to dirferent types of industrial units. However, the evaluation study found that the quality of these surveys left much to be desired, and the actual pattern of units that sprung up bore little relation to the "potential" as brought out in the surveys. There were various critical administrative inefficiencies. The State Governments did not attach much priority to the implementation of the program and the quality of staff assigned to administer the scheme was poor. The P.E.O. tried to assess the impact of the R.I.P. program on industrial development in the areas covered, by detailed investigations of a large sample of industrial units in existence in these areas. The percentage of units which availed themselves of the various components of assistance 1/ O.P. Jain, 1975, p. 49. 2/ Programme Evaluation Organisation, Planning Commission - Evaluation Study of Rural Industries Projects, Govt. of India, 1978.  -53- Table 22 Public Expenditure on Rural Industries Projects Share in Total Public expenditure on village and Actual expenditure. (Rs.m) small-scale industries (7). Third Plan (1961-66) 48 2.2 Annual Plan Period (1966-69) 66 4.9 Fourth Plan (1969-74) 101 3.4 Fifth Plan (1974-79) 211 a/ 3.9 a/ Planned outlay SOURCE : Planning Commission, Various Five Year Plan documents. provided by the program is given in Table 23. One finds that with the exception of loan assistance, the other components have only had a marginal impact on the industrial units. The principal objective of the R.I.P. Program is to promote industrial development in the rural areas. However, a clear perspective is lacking of both the character of rural industrialization to be promoted and the economic viability of' such a strategy. In particular, there is  -54- Table 23 Proportion of SaSUle Units Availing of R.I.P. Assistance (1) Type of Assistance i) Technical assistance provided 5.9 ii) Supply of raw materials etc. 12.0 iii) Supply of machinery and tools 4.3 iv) Marketing facilities 1.0 v) Loans for a) term finance 5.8 b) working capital 8.8 c) term finance as well as working capital 4.0 vi) Miscellaneous and other assistance 2.9 Source: Evaluation study of Rural Industries Projects: Programme Evaluation Organisation, Planning Commission, Government of India, March, 1978. ambivalence regarding the priorities to be attached to the promotion of villase industries and artisanship versus the development of small industries utilizing relatively modern techniques of production. The former are scattered in numerous villages and small towns while the latter  -55- are concentrated in towns of medium to large sizes. Thus, the strategy regarding which type of industry to aid would have an impact on the locational pattern of rural industrialization. The current policy is to exclude towns with population exceeding 15,000 (by the 1961 census) or 25,000 (by the 1971 ce!nsus) from the orbit of the program. However, even with this cut-off point, locational issues like the choice between concentration of assistance in certain towns or villages (to act as growth centers) versus the spireading of assistance among all locations of industrial units have yet to be resolved at a policy level. This section, has described the gamut of policies designed to influence the growth and location of the small scale sector. To give some idea about the relative importance of these various programs, Table 24 displays estimated public expenditure on the various components during 1966-69 and 1969-74. Table 24 Estimated Public Sector ExDenditure on Small Scale Industries (Rs. Crores) 1/ 1966-69 1969-74 Small Scale Industries 52.46 70.33 Industrial Estates 7.35 15.73 Handloom Industries 13.37 28.57 Powerlooms 0.46 3.90 [hadi and Village Industries 54.03 102.66 Sericulture 3.75 8.39 Coir Industry 1.21 4.49 Handicrafts 4.80 6.24 Rural rndustries Projects 6.70 10.13 Total 144.13 251.01 1/ A Crore is a unit used in India and is equal to ten million. Source: Government of India (Planning Commission), Fifth Five-Year Plan.  -56- (iv) Metropolitan Planning Attempts at regional planning in India have been made essentially in the context of metropolitan areas. While metropolitan plans exist for most big towns and cities in India, the most well-established and effective ones are those for Delhi, Bombay and Calcutta. Unfortunately, we do not have the detailed data that would permit a study of their operation and impact. Therefore, we shall restrict ourselves to a cursory and descriptive look at these plans. Delhi has the distinction of having been the object of the first attempt at city planning. The Delhi Development Authority was established in 1955 and prepared the "Delhi Master Plan" for the period 1962-81. V/ The plan indicated the directions for optimum socio-economic development of the ci,ty. As far as the industrial development of Delhi was concerned, the plan was in favor of restricting and controlling the process. The growth of industrial workers during 1962-81 was to be limited to 257. Industry was to be restricted in urban Delhi and deflected to the "ring towns" of Ghaziabad, Faridabad, Ballabgarh, Gurgaon, Bahadurgarh and Loni. Only small and medium-sized industries satisfying local consumption needs and services would be considered for location in urban Delhi. The Delhi Master Plan also recognized another level of spatial planning - the "National Capital Region," consisting of the area (within a radius of 80 km) around Delhi and encompassing tehsils in the neighboring states of Uttar Pradesh, Haryana and Rajasthan. The development of the "ring towns" and the National Capital Region in synchronization with plans for the Delhi urban area requires a high degree of coordination between the 1/ Delhi Development Authority, Delhi Master Plan, Vols. I & II, New Delhi, 1961.  -57- metropolitan organizat;ions of Delhi and those of neighboring states. Unfortunately, this Was lacking for a number of years. However, recently, serious attempts are being made to establish an effective political and administrative apparatus for this purpose. 1/ The CalcuLta Metropolitan Planning Organisation was established in 1961 and prepared the "Calcutta Metropolitan Regional Plan" in 1966. 2/ The plan was more in the nture of an "indicative" rather than a master plan detailing the physical development of the Calcutta region. Another distinctive feature was that "the Calcutta plan looks far beyond the immediate environs of Calcutta and emphasizes the need for linking the hierarchy of towns throughout the city's hinterland, which includes not merely the whole of West Bengal but also the neighboring states and territories, to establish a meaningful reciprocity between agriculture and industry and to integrate the development of material and human resources." 3/ The first major step towards metropolitan planning in Bombay was taken in 1964 with the preparation of the Development Plan for Greater Bombay. As far as industrial location was concerned, the thrust of the Plan was to strongly discourage the coming up of new industrial units or the expansion of existing ones in the Greater Bombay area. The state Government accepted this as a general policy, but was quite lax in permitting expansions of textile firms. However, in 1968, the Government 1/ See Town and Country Planning Organisation, National Capital Regional Plan 1972. 2/ Calcutta Metropolitan Planning Organisation, Basic Development Plan for Calcutta Metropolitan District, Calcutta, 1966. 3/ Nisra et al., p. 136  -58- reviewed this industrial location policy. A distinction was introduced between 'conforming" and "non-conforming" zones. Units in the former would be freely permitted to expand and a liberal attitude was adopted towards the setting up of small-scale units. The other major development was the setting up of the Gadgil Comhittee in 1965 for recommending policies for development of the Metropolitan Regions of Bombay-Panvel and Pune. On the recommendations of this Committee, the Bombay Metropolitan Regional Plan was drafted 1/ and formally adopted by the State Government in 1973. As far as industrial location was concerned, dispersal and decentralization of industries were accepted as the major policy goals. Industries were not to be encouraged in certain areas (like the Kalyan-Ulhasnagar-Ambarnath and Thane-Kavesar complexes) and would be diverted to new zones (like Nhava Sheva and Bhiwandi and Bassein tehsils). Only consumer-oriented industries (satisfying local consumption and service needs) would be allowed in Greater Bombay. These industrial policies outlined in the plan were formalized by the State Government of Maharashtra with the announcement of an "Industrial Location Policy" in 1974. The Metropolitan region was divided into four zones with restrictive policies for the central areas and more liberal ones for the periphery. A system was introduced whereby new industrial units or existing units wishing to expand operations or shift location in the Bombay Metropolitan Region required "No Objection Certificatesm from the Commissioner of Industries of the Government of Maharashtra. I/ Bombay Metropolitan Regional Planning Board, Report of the Draft Regional Plan of Bombay Metropolitan Region, Vols. I & II, 1970.  -59- (v) Incentives to Develo2 Backward Districts After 1979, a number of steps were initiated to promote the industrial development of backward areas. The Planning Comission recomended certain criteria for the identification of these backward regions consisting of per capita agricultural production and industrial output, agricultural workers, factory employees and other workers as a proportion of the population, per capita consumption of electricity, and the length of surfaced roads and railway mileage in relation to population. However, the states have introduced their own -variations to these criteria. The definition of backward areas varies accordinlg to the type of incentive (elaborated below). The shares of those backward districts eligible for concessional finance in area and population in indlvidual states and union territories in 1971 are listed in Table 25. 1/ For India as a whole, these backward districts account for 607. of the population and 707. of area. The policies adopted for encouraging industrial growth in the backward districts operate through the following incentives: (a) Capital InvesBtment Subsidy; (b) Transport subsidy; (c) Income tax concessions; Cd) Concessional finance from all-India financial institutions; and Ce) State Government incentives. 2/ 1/ The correlation coefficient of these population shares of backward districts in individual states with state per capita incomes in 1971 was 0.54. Thus backward districts form a relatively larger segment of population in backward states. 2/ These have been described earlier.  -60- TABLE 25 SHARE OF BACKWARD DISTRICTS IN AREA AND POPULATION, 1971 Backward Districts as % of total for the States States Area Population Andhra Pradesh 72.4 59.3 Assam 77.4 75.6 Bihar 47.2 55.0 Gujarat 67.2 49.1 Haryana 50.0 39.0 Himachal Pradesh 88.0 73.0 Jammu & Kashmir 100.0 100.0 Karnataka 66.7 62.5 Kerala 42.3 50.0 Madhya Pradesh 87.0 83.0 Maharashtra 54.7 41.7 Manipur 100.0 100.0 Meghalaya 100.0 100.0 Nagaland 100.0 100.0 Orissa 61.0 48.0 Punjab 38.9 35.0 Rajasthan 61.9 54.2 Tamil Nadu 64.4 60.1 Tripura 100.0 100.0 Uttar Pradesh 68.4 61.9 West Bengal 82.5 68.4 Union Territories* 100.0 100.0 All India 71.3 59.0 "Backward Districts" in this Table refer to those selected for concessional finance from Financial Institutions. *Excluding Delhi and Chandigarh SOURCE: K.S.V. Menon, 1979, pp.54  -61- The Central Government capital investment subsidy scheme was introduced in 1971. Under this scheme, new or expanding units in selected backward districLs were entitled to a subsidy of ten percent of their total or additional fixed capital investment. The ceiling on investment, eligible for subsidy, was fixed at Rs. 5 m.; however, units with investments exceeding this ceiling would also be considered at the discretion of the Government though the maximum amount of subsidy would still be Rs. 500,000. In 1973, major modifications were introduced. The rate of subsidy was raised to 157 and the investment ceiling to Rs. 10 m. The discretionary clause for units with investment exceeding this limit would still hold, subject to a subsidy limit of Rs. 1.5 m. One indicator of the progress of the scheme is provided by the reimbursement of Central subsidy by the Government (Table 26). In the initial years, the scheme did not evoke much response. However, from 1974-75 onwards, there has been a significant and almost continuous increase in the reimbursement of central subsidy. The state-wise picture of disbursement during 1973-79 is depicted in Table 27. Four states consisting of Tamil Nadu, Maharashtra, Gujarat and Karnataka account for over 427 of total disbursement during 1973-79. Simultaneously with the introduction of the capital investment subsidy in 1971, a transport subsidy was also announced to aid industrial development in hilly backward areas. Under this scheme, new industrial units in the States or Union territories of Jammu and Kashmir, Assam, Manipu, Meghalaya, Nagaland, Tripura, Arunachal Pradesh, Mizoram, Andamans and Nicobar Islands, Lakshadweep, Himachal Pradesh and the hilly districts of U.P. were eligible to a subsidy amounting to 501 of the transportation costs of both raw  -62- TABLE 26 Reimbursement of Central Investment Subsidy by Government of India (Re. million) 1972-73 1.2 1973-74 5.9 1974-75 38.6 1975-76 60.0 1976-77 111.7 1977-78 199.6 1978-79 154.1 SOURCE: National Committee on the Development of Backward Areas, 1980, pp. 21  -63- Table 27 Share of States and Union Territories in the Central Investment Subsidy Disbursed durin& 1972/73 to 1978/79 State/Union Share (%) in Subsidy Disbursed Territory uDtil 1978/79 1. Andhra Pradesh 12.1 2. Assam 2.3 3. Bihar 2.1 4. Gujarat 7.5 5. Haryana 2.8 6. Himachal Pradesh 3.4 7. Jammu and Kashmir 2.6 8. Karnataka 7.0 9. Kerala 4.8 10. Madhya Pradesh 5.0 11. Naharashtra 11.7 12. Manipur 0.1 13. Meghalaya 0.3 14. Nagaland 0.7 15. Orissa 1.1 16. Punjab 3.3 17. Rajasthan 8.3 18. Sikkim 0.1 19. Tamil Nadu 15.5 20. Tripura 0.3 21. Uttar Pradesh 2.6 22. West Bengal 1.9 23. Andaman & Nicobar 0.1 24. Dadra & Nagar Haveli 0.3 25. Arunachal Pradesh 0.2 26. Goa, Daman & Diu 3.3 27. Lakshadweep - 28. Mizoram 0.1 29. Pondicherry 0.5 Total 100.0 Source: National Committee on the Development of Backward Areas, 1980, 9. 21.  -64- materials and finished goods. Expanding units are also eligible for this subsidy for their expansion programs, provided that the increase in production exceeds 257 of average annual output during the last three years. During the first five years of its operation, the transport subsidy scheme failed to make any impact on industrial development in these regions; only two small claims (from Tripura) were received during these years. The decision relating to the desirability of continuing the scheme is currently under consideration by the Ministries of Industry and Finance. One of the criticisms voiced against the transport subsidy scheme is that it may lead manufacturers to substitute raw materials imported from outside for those available locally, for example, to utilize cement instead of timber in construction. This would adversely affect the objective of promoting industries based on local resources. This criticism has remained largely academic because of the poor response to the scheme. Income tax concessions for new industrial units (including hotels) in backward districts were announced in 1974. These units were allowed a deduction of 207. of profits while computing taxable income. All projects consencing operation after 1970 were eligible for this concession for a period of ten years thereafter. Unfortunately, data pertaining to the number availing themselves of these income tax concessions are not readily available. In 1970, schemes of concessional finance for investment in backward areas were introduced by the all-India financial institutions - the Industrial Development Bank of India (IDBI), the Industrial Finance Corporation of India (IFCI), the Industrial Credit and Investment Corporation of India (ICICI), the Industrial Reconstruction Corporation of India and the National Small Industries Corporation. These included direct assistance to industries as well as refinance facilities for the state-level institutions. The precise  -65- terms of the concessions (especially the rate of interest) have varied over the years. Those prevailing in 1975 are summarized in Table 28. Table 29 presents trends in assistance provided to units in backward areas by the Industrial Development Bank of India (IDBI) which is the apex development bank for industry in India. One finds that there has been an impressive increase in the share of backward areas in sanctioned assistance since 1970-71, with a corresponding lagged effect on disbursements. This trend has been particularly noticeable after 1974-75. In 1980-81, almost half of sanctioned assistance from IDBI was going to units in backward regions. Even though these trends pertain only to assistance from IDBI, they would be broadly reflective of trends in overall corporate sector investment. This is because most medium and large projects in the corporate sector seek assistance from one or more of the term-lending financial institutions operating in the industrial sector; 1/ and IDBI is the largest of these institutions. 2/ Thus, the data appear to indicate that the capital investment subsidy and the schemes of conce!ssional finance have had a significant impact on the industrial development of backward regions. However, an important qualification needs to be made before such a conclusion can be drawn. This relates to the distribution of assistance among the 247 districts that qualify for concessions. Only a small proportion of the districts have benefited from these incentives. For instance, 15 districts account for over 567 of the 1/ During 1971-79, on average about half of corporate investment in the private sector was financed by these institutions. See IDBI, Report on Development Banking in India, 1980-81, p. 2 2/ In 1980-81, for example, assistance sanctioned by IDBI (excluding export finance and subscription to shares and bonds of financial institutions) constituted 53* of total assistance sanctioned by all the term-lending financial institutions operating in the industrial sector. See IDBI OD cit. pp. 1, 19.  TABLE 28 TERMS OF CONCESSIONAL FINANCE FOR UNITS IN BACKWARD DISTRICTS/AREAS, 1975 Rate of Interest Grace Period Amortisation Participation Underwriting Com- (Per Cent) (Years) Period in risk mission (per cent) Name of Institution (Years) capital Industrial Development 8.5 5 15 to 20 Relatively 1.25 for shares Bank of India (10.25) (3) (10 to 12) heavily on (2.5) merits 0.75 for debentures (1.5) Indus. Finance On rupee loans-- 9.5 5 15.to 18 Relatively 50 per cent of Corp. of India (11.25) (3) (10 to 12) to a greater the normal On foreign currency 10.5 extent on loans-- (11.5) merits Indus. Credit and On rupee loans-- 8.5 5 20 -- 50 per cent of Investment Corp. (10.25) (2 to 3) (12) the normal of India On foreign Currency loans-- 9.5 (10.5) Indus. Reconstruction 7 -- -- -- -- Corp. of India (8.5) National Small Indus. 11 to 13 -- -- -- -- CoERoration* (13 to 15) COMMITMENT CHARGE REFINANCING FACILITIES PROMOTERS' OTHER CONCESSIONS CONTRIBUTION Industrial Development Reduced by 0.5 per To State Finance Cor- Bank of India cent or waived al- porations and Banks together in excep- up to Rs. 30 lakhs at tional cases 5.5 per cent Industrial Finance Cor- 50 percent of the -- Lower than usual 50% reduction in other charges poration of India normal Industrial Credit and 50 percent of the -- Lower than usual Invest. Corp. of India normal Industrial Reconstruc- tion Corporation of -- -- 7% (8.5%) interest on loans for India reconstructing/modernizing Industries National Small Indus- -- -- 10%(15%) interest on earnest tries Corporation money for supplying machinery and NOTE: Figures in brackets indicate normal terms for advanced areas. equipment under the hire-purchase SOURCE: M.D. Godbole, 1978, pp.72-73. scheme.  -67- TABLE 29 Trends in the Share (Z) of Backward Districts in Assistance Sanctioned & Disbursed by the Industrial Development Bank of India 1964-65 to 1980-81 Year -/ Share (X) of Backward Districts in Total Sanctions Total Disbursements 1964-65 10.7 18.5 1965-66 8.7 11.6 1966-67 12.9 6.6 1967-68 14.2 15.8 1968-69 32.2 22.7 1969-70 23.7 18.2 1970-71 24.2 16.6 1971-72 33.2 20.4 1972-73 33.2 24.1 1973-74 36.8 34.5 1974-75 35.2 36.8 1975-76 42.9 39.5 1976-77 52.7 41.4 1977-78 47.3 51.7 1978-79 39.4 50.9 1979-80 47.9 40.6 1980-81 47.2 41.8 1/ Defined as July of any year to June of the following year. Source: Industrial Development Bank of India, Operational Statistics, various issues.  -68- Central Investment subsidy disbursed up until 1978-79. 1/ Similarly, 22 districts received 497 of Lhe total disbursals of concessional finance by thefinancial institutions. 2/ Data relating to IDBI disbursement of concessional finance reveals that the Lop 50 districts accountted for about 70* of this assistance as of December 1979. 3/ This concentration of investment in certain districts is natural. Only a small number of the districts that qualify for incentives possess thc potential for icii(ustrial growth in terms of factors such as the availability of raw materials, infrastructure and proximity to markets. Entrepreneurs will only locate their uniits in such areas. Certain other limitations of these incentives for backward area development will he elaborated on in the concluding chapLer 4, where we shall provide a critique of the Indian experience with industrial location policy. 1/ NCDRA, 1980, p. 14 2/ NCDBA, Op cit,_p. 15 3/ N.R. Shenoy, and S.K. Guptya, Regional Pattern of IDBT's Assistance, published in IDBI, 1980, p. 278.  -69- Chapter III TRENDS IN INDIAN INDUSTRIAL LOCATION In the last chapter, we described the various policies used by the Indian Government to influence industrial location and their method of operation. We have also attempted to throw some light on the effect of each of these policies, to the extent permitted by data. On the basis of our analysis, we find that the effectiveness of each of the locational instruments has varied considerably. What, then has been the overall impact of these policies on the inter- and intrastate distribution of industry? An investigation of this issue involves an analysis of the actual trends in the location of industry. These trends are the result of both market forces and Governmental policy. The separation of these two determinants is an analytically complex task, one which is ruled out because of data constraints. In lieu of this, we shall merely present the findings relating to the inter and intra-state trends in the distribution of industry and draw some simple conclusions, which would at least be useful as a background for more intensive studies to be conducted in a future research project. 1. Trends in the Inter-State Distribution of Industry We first examine trends in the inter-state distribution of industry. The data that we have used comes mainly from the following sources: (a) the Annual Survey of Industries (ASI); (b) statistics on factory labor collected under the Factories Act, 1948 by the Labor Bureau;  -70- (c) state income data from Central Statistical Organisation, Monthly Abstract of Statistics, June 1979; and (d) state population estimates from Institute of Applied Manpower Research, Facts Book on Manpower, 1977. The concepts, tabulation and time-frame utilized in these sources constrain us in a number of ways. First, the industrial data only relates to the factory sector - i.e., units employing more than ten workers if using power or 20 workers without power. The ASI is conducted in two parts. A Census is conducted for units employing more than 50 workers with power or 100 workers without power. The rest of the factory sector is covered on a sample basis. Thus, the entire so-called "unregistered" sector, consisting of producing units employing less than ten workers with power and 20 workers without power, falls outside the ambit of these industrial and labor surveys. Secondly, the ASI was launched only in 1960. Before 1960, a "Census of Manufacturing Industries" was conducted annually only for about half of industry groups into which all establishments were classified. A "Sample Survey of Manufacturing Industries" covered the entire manufacturing sector on a yearly basis, but, unfortunately, no state-wise data were presented. Because of the absence of comprehensive regional data, we have only analyzed trends since 1960. Certain problems crop up because of creation of new states and re-definition of state boundaries. Most of these alterations in political sub-division occurred before 1960. However, in the north-eastern region, there have been some changes after 1960. Since the statistical documents are unclear about the handling of these alterations,  -71- we have chosen to neglect the union territories and states in north-eastern India. In any case, this region accounts for a very small portion of industrial output and employment. The same kind of reasoning justifies the exclusion of certain other minor states and union territories. In the introductory chapter of this paper, we looked at trends in the levels of industrialization in the states and their shares in value added and employment in the Indian manufacturing sector during 1961-76. We tried to explore the differences between state shares in value added and employment by examining the value added/employment ratios for states and relating these to the all-India average (Tables 30 and 31). One finds that, in 1975, value added per worker was higher than average for Maharashtra, Bihar, Rajasthan, Orissa, Punjab and Haryana and significantly lower than average for Kerala, Andhra Pradesh and Delhi. If we look at the period 1961-75 (and, in particular, the decade 1961-71), there seems to be a trend towards equalization of value added per employee across states. Variations in value added per employee across states and over time are the result of many factors, such as capital intensity of production, inherent labor productivity (owing to, say, skill levels) or variations in product rnix of the manufacturing sector. Unless one examines each of these in depth, it will not be possible to identify the precise contributory factors to the trends noted above. Finally, two iindices were computed in order to come to an overall assessment about trends in state-wise concentration of industry (Table 32). These are Theil'El inequality index and the Hirschman-Herfindahl index.  -72- TABLE 30 Value added per employee (Rs '000) in Industry KARNA- MAHA- A.P. BIHAR GUJARAT TAKA KERALA M.P. RASHTRA 1961 1.6 3.8 3.4 2.5 1.8 2.0 3.8 1962 1.6 4.5 3.0 2.5 1.9 1.4 4.1 1963 1.9 4.4 3.1 2.8 2.0 2.2 4.2 1964 2.1 4.8 3.4 3.1 1.9 2.7 4.5 1965 2.5 5.1 3.9 3.7 2.2 2.5 4.9 1966 2.8 5.9 4.3 4.2 3.0 3.2 5.5 1967 2.6 5.2 4.8 4.4 3.8 3.6 5.9 1968 2.7 5.3 5.2 4.7 4.1 4.5 6.3 1969 4.1 5.9 5.8 5.8 5.4 4.3 7.6 1970 4.8 6.2 6.6 6.5 4.3 4.9 8.4 1971 5.1 6.9 6.4 7.4 5.1 5.1 9.0 1972 1973 6.8 7.5 8.8 7.4 5.4 9.5 11.4 1974 7.1 11.9 11.6 8.2 6.5 11.7 14.6 1975 8.1 14.8 10.8 9.4 6.1 11.3 14.7 PUNJAB & RAJA- HARYANA ORISSA STHAN T.N. U.P. W.B. DELHI INDIA 1961 2.7 3.5 2.3 3.0 2.1 3.0 3.1 3.0 1962 2.8 4.4 2.2 3.3 2.1 3.4 3.0 3.1 1963 3.4 5.3 2.5 3.3 2.2 3.7 3.5 3.3 1964 4.0 5.3 2.8 3.9 2.5 4.0 3.7 3.6 1965 4.4 7.2 3.2 4.4 3.0 4.1 3.8 4.0 1966 5.2 5.1 3.6 4.6 2.9 4.3 4.0 4.3 1967 4.7 3.7 4.4 4.8 2.6 4.3 4.0 4.4 1968 6.5 5.9 5.0 5.1 4.3 4.3 4.6 4.8 1969 6.5 6.9 5.6 6.2 4.8 5.0 5.5 5.8 1970 6.9 7.9 7.8 6.9 5.0 5.1 5.5 6.3 1971 7.7 7.3 8.2 7.0 5.0 5.9 6.0 6.8 1972 1973 8.6 12.8 8.8 8.9 6.6 7.1 6.7 8.4 1974 11.1 14.0 11.7 11.2 8.5 8.9 7.9 10.7 1975 12.9 13.6 14.2 10.9 8.8 9.8 7.6 11.2  -73- Table 31: STATE SHARE IN VALUE ADDED/STATE SHARE IN EMPLOYMENT IN FACTORY MANUFACTURING SECTOR KARNA- MAHA- A.P. BIHAR GUJARAT TAKA KERELA M.P. RASHTRA 1961 0.5 1.3 1.1 0.8 0.6 0.7 1.3 1962 0.5 1.4 1.0 0.8 0.6 0.5 1.3 1963 0.6 1.3 0.9 0.8 0.6 0.7 1.3 1964 0.6 1.3 0.9 0.9 0.5 0.8 1.2 1965 0.6 1.3 1.0 0.9 0.6 0.6 1.2 1966 0.6 1.4 1.0 1.0 0.7 0.7 1.3 1967 0.6 1.2 1.1 1.0 0.9 0.8 1.3 1968 0.6 1.1 1.1 1.0 0.8 0.9 1.3 1969 0.7 1.0 1.0 1.0 0.9 0.7 1.3 1970 0.8 1.0 1.0 1.0 0.7 0.8 1.3 1971 0.8 1.0 1.0 1.1 0.8 0.8 1.3 1972 1973 0.8 (.9 1.0 0.9 0.6 1.1 1.4 1974 0.7 L1.1 1.1 0.8 0.6 1.1 1.4 1975 0.7 1.3 1.0 0.8 0.5 1.0 1.3 COEFFICIENT PUNJAB & RAJA- OF HARYANA ORISSA STHAN T.N. U.P. W.B. DELHI VARIATION 1961 0.9 1.2 0.8 1.0 0.7 1.0 1.0 0.273 1962 0.9 1.3 0.7 1.1 0.7 1.1 1.0 0.334 1963 1.0 1.6 0.7 1.0 0.7 1.1 1.0 0.314 1964 1.1 1.'5 0.8 1.1 0.7 1.1 1.0 0.286 1965 1.1 1.8 0.8 1.1 0.8 1.0 1.0 0.328 1966 1.2 1.2 0.8 1.1 0.7 1.0 0.9 0.257 1967 1.1 0.83 1.0 1.1 0.6 1.0 0.9 0.216 1968 1.3 1.:2 1.0 1.1 0.9 0.9 1.0 0.189 1969 1.1 1.:2 1.0 1.1 0.8 0.9 0.9 0.178 1970 1.1 1.2 1.2 1.1 0.8 0.8 0.9 0.189 1971 1.1 1.1 1.2 1.0 0.7 0.9 0.9 0.176 1972 1973 1.0 1.5 1.0 1.1 0.8 0.8 0.8 0.248 1974 1.0 1.3 1.1 1.0 0.8 0.8 0.7 0.246 1975 1.2 1.:2 1.3 1.0 0.8 0.9 0.7 0.263  -74- TABLE 32 MEASURES OF STATE-WISE CONCENTRATION OF INDUSTRY (FACTORY SECTOR) VALUE ADDED EMPLOYMENT Theil's Hirschman- Theil's Hirschman- Inequality Herfindahl Inequality Herfindahl Index Index Index Index 1960 11.70 1961 15.65 14.09 11.04 11.70 1962 16.43 14.72 10.87 11.66 1963 15.75 14.38 10.75 11.65 1964 14.72 13.82 10.38 11.51 1965 13.79 13.33 10.17 11.43 1966 13.29 13.03 9.70 11.16 1967 14.33 13.30 9.37 10.97 1968 12.86 12.79 8.99 10.80 1969 12.48 12.79 9.10 10.84 1970 11.97 12.60 8.92 10.72 1971 12.55 12.93 9.06 10.81 1972 1973 11.59 12.52 8.71 10.46 1974 10.84 12.25 8.28 10.25 1975 9.40 11.54 7.83 9.99 1976 11.27 NOTE: The Hirschman-Herfindahl Index is defined as Ep 2/100, where pi is the percentage share of each state i in value added or employment. The value of the index rises with the degree of concentration. We have defined the Theil's Inequality Index with respect to state share of population. The Index is defined as Epi log (p /qi), where qi is the percentage share of each statei in population. Intuitively, the index provides us with a measure of the divergence of state shares of industry from population shares. The value of the index rises with the degree of con- centration.  -75- We have defined Tlheil's inequality index with respect to state share of population. The index is defined as Ep log (pi/qi), where Pi is the percentage share of each state (i) in value added or employment while qi is the percentage share of each state in population. Intuitively, the index provides us witlh a measure of the divergence of state shares of industry from populatlon shares. The value of the index rises with the degree of concentration. Further elaboration of the Hirschman-Herfindhl and Theil indices is prov:Lded by Appendix I where, in particular, the maximum and minimim values of the indices are discussed. The Hirschmani-Herfindhl index is defined as E pi 2/100. The value of this index also rises with the degree of concentration. The measures (particularly Theil's index) indicate a significant decline in state-wise concentration - especially in manufactured value added. For example, comparing the two years 1961 and 1975, Theil's index exhibits a decline of 40% for value added and 30% for employment. The decrease in the case of the Hirschman-Herfindhl index is 18% and 15% respectively. We have seen earlier that certain types of locational policy were explicitly targetted at influencing the inter-state distribution of industry. However, our evidence does not permit an assessment of the precise contribution of these policies to the reduction of inter-state imbalances in industrial developm,ent that has been observed. But, certain tentative hypotheses could be advanced., For instance, we have examined the policy cf locating public sector undertaking in backward states. Our analysis indicates that this policy has been implemented in practice - even though pratmatic considerations (like availability of raw materials) may have been dominant in the locational decisions. Moreover, we have found  -76- that these undertakings have had a significant impact on the industrial development of the backward states. Our examination of the industrial licensing system is somewhat inconclusive but suggests that, at least after 1965, increased importance may have been assigned to the objective of allocating licenses on an equitable basis among the states. Evidence pertaining to the other major policies directly affecting inter-state distribution of industry (like state government incentives, distribution and price controls) does not warrant any conclusions regarding the direction or magnitude of impact. There are other types of Governmental policy which have influenced inter-state distribution of industry even though this has not been their immediate objective. For instance, considerations of regional equity are extremely important when deciding on the inte;-state allocation of tax revenue collected by the Central Government. Similarly, grants are provided by the Central Government to the states, and the quantum of these going to different states is also decided on, to a significant extent, by equity considerations. These revenue transfers and grants are used for both developmental (infrastructure, agriculture, transport, etc.) and non-developmental (mainly drought and famine relief) purposes by the states. To the extent that this expenditure influences economic development in the states, there would also be a differential impact on industrial growth through both supply and demand factors. We have not examined these types of policies in this study since we are concerned only with direct policies affecting industrial location. However, in the last analysis, as we have emphasized earlier, trends in industrial location are the result of both market forces and Governmental policy. For instance, trends in inter-state variations in  -77- factor prices, natural endowments and incomes are important explanatory variables. Unless one studies these complex interdependencies, it would not be possible to judge the magnitude of contribution of locational policy to the reduction in inter-state imbalances. 2. Trends in Intra-Regional Distribution of Industry We now turn Ito trends in the intra-regional distribution of industry. The data used comes essentially from two sources: (a) Population Census; and (b) Labor Bureau Statistics The Labor Bureau statiistics contains data only for the factory sector while the population census also includes information for household industry. The population censuses of 1961 and 1971 contain information relating to the population engaged in industry in different classes of towns, cities and town groups/urban agglomerations categorized in the following manner. 1/ Class I: Population of 100,000 or more Class II: Population of 50,000 to 99.999 Class III: Population of 20,000 to 49,999 Class IV: Population of 10,000 to 19,999 Class V: Population of 5,000 to 9,999 Class VI: Population of below 5,000 The distinction between the concepts of "town group" and "urban agglomeration" used in the 1961 and 1971 censuses is explained in the notes to Table 33. This distinction obviously poses problems in comparing 1/ Unfortunately, the 1951 census contains this type of classification only for total population but not for industrial workers.  -78- the two censuses. However, a much more serious problem of comparison is the change in definition of "worker". Put briefly, the change consists of the dichotomy of persons into "workers" and "non-workers" according to labor time disposition in 1971, rather than gainful occupation irrespective of time spent on it in 1961. The measurement of the work-force engaged in household industry is especially sensitive to this change in definition because a significant proportion of workers allocate part of their time to agricultural activities. Moreover, this allocation of time among different activities is a more important characteristic of the smaller towns in comparison with the large ones. For these reasons, one cannot draw strict conclusions regarding changes in the distribution of industry between different sizes of urban groupings based on a comparison of the 1961 and 1971 censuses. Furthermore, comparability is also affected by the fact, that between the two censuses, towns move between size classes because of population growth. For example, a number of class III towns in 1961 became class II towns in 1971. Despite these serious problems, we shall persist with comparisons of the two censuses because of the absence of alternative data. Table 33 contains data relating to the distribution of population and industrial employment (in household and non-household industry) among different classes of cities and towns in 1961 and 1971. The data is presented for India as a whole and for the eight most important industrial states. We have also computed the employment to population shares and the Theil inequality indices (defined earlier). The first thing to note is that, as one would expect, relative to population, employment in household industry tends to be more concentrated in the smaller classes of towns where the demand of the population for  TABLE 33 Distribution of Industrial Employment in Cities, Towns, and Town Groups (TG)/Urban Agglomerations(UA), 1961 & 1971 SHARE(%) OF DIFFERENT CLASSES OF CITIES/TOWNS IN: INDIA POPULATION EMPLOYMENT- IN: EMPLOYMENT SHARE/POPULATION SHARE (1) (2) (3) (4) (5) C/ ~~~~C, Household-z Non-Household-/ Household Non-Household City/Town/TG-UA Industry Industry Industry Industry Class a/ 1961 1971 1961 1971 1961 1971 1961 1971 1961 1971 Class I 48.4 55.8 31.1 42.8 62.6 71.8 0.6 0.8 1.3 1.3 Class II 11.9 11.3 12.9 12.9 11.5 8.8 1.1 1.1 1.0 0.8 1 Class III 18.5 16.3 22.6 20.5 15.1 10.9 1.2 1.3 0.8 0.7 % Class IV 13.0 11.3 20.3 16.2 7.2 5.9 1.6 1.4 0.6 0.5 1 Class V 7.2 4.7 11.9 7.1 3.2 2.4 1.7 1.5 0.4 0.5 Class VI 0.9 0.5 1.2 0.5 0.3 0.2 1.3 1.0 0.4 0.5 All Classes 100.0 100.0 100.0 100.0 100.0 100.0 Theil Inequality:- 3.05 1.60 2.38 2.52 Index (TII) of Industrial Employ- ment in Relation to Population MAHARASHTRA Class I 65.0 70.8 51.6 56.0 82.6 86.7 0.8 0.8 1.3 1.2 Class II 6.9 6.9 9.8 10.3 4.7 5.0 1.4 1.5 0.7 0.7 Class III 12.3 11.1 14.6 15.1 7.4 4.7 1.2 1.4 0.6 0.4 Class IV 10.6 8.1 15.9 14.4 4.0 2.7 1.5 1.8 0.4 0.3 Class V 4.9 2.8 7.7 4.0 1.2 0.9 1.6 1.4 0.3 0.3 Class VI 0.4 0.3 0.5 0.3 0.1 0.1 1.1 0.9 0.2 0.3 100.0 100.0 100.O 100.0 100.0 100.0 TII: 1.76 2.27 3.73 3.40  Contd. TABLE 33 WEST BENGAL Class I 56.5 71.0 26.1 49.7 60.9 85.7 0.5 0.7 1.1 1.2 Class II 17.8 11.8 29.3 22.0 17.4 4.5 1.6 1.9 1.0 0.4 Class III 17.0 9.6 22.8 14.3 17.9 5.8 1.3 1.5 1.1 0.6 Class IV 5.8 5.0 16.5 10.4 2.7 2.5 2.8 2.1 0.5 0.5 Class V 2.4 2.5 4.5 3.2 1.0 1.5 1.9 1.3 0.4 0.6 Class VI 0.3 0.1 0.7 0.4 0.1 0 2.1 3.9 0.3 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 TII: 9.38 4.61 0.85 2.75 GUJARAT Class I 43.5 49.0 34.2 39.2 65.0 68.3 0.8 0.8 1.5 1.4 Class II 11.7 15.4 10.6 15.3 8.5 11.1 0.9 1.0 0.7 0.7 Class III 22.5 15.7 24.4 17.9 14.6 9.2 1.1 1.1 0.6 0.6 Class IV 13.2 12.8 19.0 17.0 7.5 7.3 1.4 1.3 0.6 0.6 Class V 8.5 6.8 11.2 10.5 4.1 3.9 1.3 1.6 0.5 0.6 Class VI 0.7 0.3 0.7 0.1 0.2 0.3 1.1 0.5 0.4 1.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 TII: 1j17 1.20 4.20 3.41 TAMIL NADU Class I 41.3 57.7 29.5 46.3 51.4 68.5 0.7 0.8 1.2 1.2 Class II 16.2 14.2 19.6 18.2 16.6 13.5 1.2 1.3 1.0 1.0 Class III 20.5 15.1 22.2 17.4 17.8 11.2 1.1 1.2 0.9 0.7 Class IV 14.5 9.9 13.9 11.1 9.9 4.9 1.0 1.1 0.7 0.5 Class V 6.8 2.8 14.0 6.5 4.0 1.6 2.0 2.3 0.6 0.6 Class VI 0.7 0.3 0.7 0.5 0.4 0.2 1.1 1.5 0.5 0.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 TII: 2.24 1.67 1.27 1.45  Contd. TABLE 33 UTTAR PRADESH Class I 54.4 57.1 41.2 50.4 67.5 67.2 0.8 0.9 1.2 1.2 Class II 11.8 10.8 9.6 14.1 12.0 8.3 0.8 1.3 1.0 0.8 Class III 16.7 16.7 20.9 16.3 11.6 13.8 1.3 1.0 0.7 0.8 Class IV 11.0 10.4 18.6 12.7 6.2 7.9 1.7 1.2 0.6 0.8 Class V 5.9 4.7 9.6 6.3 2.6 2.8 1.6 1.3 0.4 0.6 Class VI 0.2 0.2 0.1 0.1 0.1 0.1 0.6 0.5 0.4 0.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 TII: 2.43 0.59 2.07 1.01 ANDHRA PRADESH Class I 42.7 48.4 19.8 25.8 52.8 56.3 0.5 0.5 1.2 1.2 Class II 8.5 13.3 9.5 12.5 8.0 11.5 1.1 0.9 0.9 0.9 Class III 24.2 20.9 29.9 36.4 24.2 19.7 1.2 1.7 1.0 0.9 Class IV 15.8 13.4 28.5 20.2 11.3 9.2 1.8 1.5 0.7 0.7 Class V 8.7 3.8 12.2 4.8 3.6 3.2 1.4 1.3 0.4 0.8 Class VI 0.1 0.2 0.1 0.3 0.1 0 1.5 1.9 1.6 0.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 TII: 5.65 5.53 1.64 0.71 KARNATAKA Class I 41.3 51.1 20.0 31.5 56.8 67.3 0.5 0.6 1.4 1.3 Class II 12.6 8.2 10.3 6.9 13.4 6.2 0.8 0.8 1.1 0.8 Class III 16.0 15.4 27.2 30.7 15.2 13.7 1.7 2.0 1.0 0.9 Class IV 19.8 19.3 30.2 24.6 10.4 9.5 1.5 1.3 0.5 0.5 Class V 8.1 4.7 8.8 5.5 3.4 2.6 1.1 1.2 0.4 0.6 Class VI 2.4 1.2 3.4 0.7 0.8 0.7 1.4 0.6 0.4 0.5 Total 100.0 1.0 100.0 100.0 100.0 100.0 100.0 TII: 5.57 4.87 3.35 2.83  Contd. BIHAR Class I 43.1 45.4 33.9 43.8 56.4 64.3 0.8 1.0 1.3 1.4 Class II 12.9 11.1 12.7 8.6 8.6 8.0 1.0 0.8 0.7 0.7 Class III 21.6 23.9 21.3 23.0 20.2 17.0 1.0 1.0 0.9 0.7 Class IV 14.9 14.4 22.1 19.4 9.5 6.9 1.5 1.3 0.6 0.5 Class V 7.0 4.8 9.4 5.0 5.0 3.3 1.3 1.0 0.7 0.7 Class VI 0.5 0.5 0.5 0.2 0.3 0.4 1.1 0.4 0.7 0.7 Total 100.0 100.0 100.0 100.0 100.0 100.0 TII: 1.28 0.55 1.82 3.36 NOTES: a/ In the 1961 Census, the concept of "urban" group was introduced to identify well-formed urban clusters based on facility of road and railway transport and the interchange of population on account of business and work. In the 1971 Census, the concept of "town group"was replaced with that of "urban agglomeration" with a stricter and more uniform definition. According to the Census document, "an urban agglomeration was formed by taking into account a contiguous urban spread constituting a town and its adjoining out- growths or two or more physically contiguous towns together with contiguous well-recognized urban outgrowths, if any, of such towns." These cities, towns and town groups/urban agglomerations are categorized into O the following classes: Class I: Population of 100,000 or more Class II: 50,000 to 99,999 Class III: 20,000 to 49,999 Class IV: 10,000 to 19,999 Class V: 5,000 to 9,999 Class VI: Below 5,000 b/ The differences in definitions of 'workers' in the 1961 and 1971 population censuses pose problems of comparison. In brief, the conceptual difference revolves around "the dichotomy of persons into 'workers' and 'non-workers' according to labor time disposition in 1971 rather than gainful occupation irrespective of time spent on it in 1961." c/ Industry is defined as 'manufacturing, processing, servicing and repairs'. A household industry is defined as an "industry conducted by the Head of the household himself/herself and/or mainly by the members of the household at home or within the village in rural areas and only within the premises of the house where the household lives in urban areas". d/ The Theil Inequality Index is defined as E pi log (pi/q ), where pi & qi are the percentage shares of the different classes of towns/cities in employment ana population respectively. Intuitively, the index provides us with a measure of the divergence of share in employment from population shares of the different classes. The value of the index rises with concentration.  -83- small quantities of simlple manufactured products are met by household production units. 1/ The reverse is true of non-household industry where economies of scale and t:echnological and infrastructural requirements dictate location in the larger classes of cities and towns. Comparing 1961 and 1971, we find markedly different trends in the distribution of employment in household and non-household industry. For household industry, the Theil inequality index for India as a whole declines by almost 50% from 3.1 in 1961 to 1.6 in 1971. This reflects a reduction of concentration (relative to population) of household industry in the smaller classes of towns. Part of the reason for this may be spurious, and may be connected with changes in the definitions of the work force. As we have stated earlier, in smaller towns there is a greater intermingling of industrial and other activities; in the 1971 Census, their inclusion in the work force depends on their allocation of time, whereas in the 1961 Census, gainful occupation (irrespective of time disposition) is used to define "workers." Thus, there may be a downward bias in the estimate of household industry workers in the smaller classes of towns in the 1971 Census compared to the 1961. Apart from this definitional reason, the phenomenon of household industry moving away the smaller towns is a natural concomitant of the development process. As the level of technology in household units improves, economies of scale become more important and there is a shift to larger towns. 1/ In 1971, about 50% of household industry workers were operating in towns falling in classes II, III and IV. These towns accounted for about 40% of the population which was relatively evenly distributed among the three classes.  -84- The Theil inequality index has also fallen significantly between 1961 and 1971 in West Bengal, Tamil Nadu, U.P., Karnataka and Bihar. It has gone up for Maharashtra and remained roughly the same for Gujarat and A.P. For non-household industry, there is hardly any change in the Theil inequality index; the index was 2.4 in 1961 and 2.5 in 1971. Thus, between 1961 and 1971, industrial policy has not resulted in any substantial shift of industry away from large cities and towns to the smaller ones. However, it must be pointed out that this conclusion only holds in the aggregate because the pattern varies for different states. Between 1961 and 1971, the Theil inequality index rose significantly for West Bengal and Bihar, fell considerably in Maharashtra, Gujarat, Uttar Pradesh, Andhra Pradesh and Karnataka and remained approximately the same in Tamil Nadu. The analysis that we have been conducting thus far of intra-regional industrial distribution ends at 1971, which is the year when the last population census was conducted. However, as we have seen earlier, certain important policies were introduced after 1970-71 to promote the industrial development of backward areas. Fortunately, district-wise data relating to factory employment exists for some states. This data can be utilized to investigate whether these policies have had any impact on industrial employment in the backward districts. The different schemes described earlier adopted different definitions for identifying the backward districts. In computing trends in the share of backward districts in total factory employment, we have chosen those qualifying for concessional finance, since these have been defined in the broadest sense and encompass the set of districts eligible for the other  -85- incentives. These shares have been plotted in figures for the states for which data is available - namely, Maharashtra, A.P., Punjab, Haryana and Kerala. Since the incentives were introduced around 1971 and data is only available till around 1975, it may be somewhat premature to judge the success of these incentives on the basis of these graphs, especially because of the lag between policy formulation and implementation and because the oil shock of 1973 might have had a much greater impact. Nevertheless, the figures reveal a change in the trend after 1971-72 for most of these states. Ihe share of the backward districts in factory employment was falling up to 1971-72 and since then has either risen or stabilized. However, the magnitude of these changes has been relatively minor (a few percentage points at the most) in most states. Thus, it would be premature to make definite inferences about the efficacy of post- 1970 locational policies based on these trends.  Share of Backward Disticts In Factory Employment (%) Maharashtra 8.5 85 8.0 8.0 7.5 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~7.5 1g / ' 1 1X LD 7.0 ' 7.0 6.5 6.5 6.0 I 6.0 1961 '62 '63 '64 '65 '66 '67 '68 '69 '70 '71 '72 '73 '74 1975 Time Woild Bcx*-25347  Share of Backward Districts In Factory Employment (%) Haryana 17.0 17.0 16.0 16.0 15.0 15.0 14.0 14.0 13.0 I I l l l 13.0 1965 '6 '67 68 '69 '70 '71 '72 '73 '74 '75 1976 Tir8e '1dd SOM-25M4  Share of Backward Districts In Factory Employment (%) Punjab 15.0 15.0 14.0 - 14.0 13.0 13.0 12.0 12.0 11.0 11.0 10.0 109. 9.0 9.0 8.0 I 8.0 1964 '65 '66 '67 ' '69 '70 '71 '72 '73 '74 '75 1976 Wold Bank-25349  Share of Backward Districts in Factory Employment (%) Andhra Pradesh 30.0 130.0 28.0 28.0 26.0 26.0 24.0 24.0 22.0 22.0 1960 '61 '62 '63 '64 '65 '66 '67 '68 '69 '70 '71 '72 '73 1974 Time Worid Bar*-25350  Share of Backward Districts In Factory Employment (%) Kerola 30.0 30.0 29.0 29.0 28.0 28.0 > 27.0 27.0 26.0 - 26.0 25.0 I I l 25.0 1963 '64 '65 '66 '67 '68 '69 70 '71 '72 '73 1974 Tine fkdd Bon*-253S1  -91- Chapter IV AN OVERALL ASSESSMENT OF INDUSTRIAL LOCATION POLICY In Chapter II, we examined the operation of the various types of policies that have been pursued in India to directly influence industrial location. Thereafter, Chapter III was devoted to an exploration of statistical trends in the spatial distribution of industry that have emerged as a result of market forces and the working of these policies. In this concluding Chapter, we make an overall evaluation of industrial location policy, which involves probing into the justifications and rationale of these policies. It appears that these fundamental issues have not been adequately sorted out by the policy makers. As a result, critical deficiencies have emerged in the context of locational policies that have been pursued so far, many of which have already been pointed out in Chapter II. 1. Rationale of Industrial Location Policy As we saw in Chapters I and III, there have been marked regional disparities in industrial development in India. These disparities have taken the form of sharp inter-state differences in industrial development as well as the concentration of industry in certain metropolitan areas. Now why should the Government take any remedial action relating to these disparities? The answers can be grouped into two broad categories: i. the disparities are a result of inefficient resource allocation and, in a dynamic context, will hinder further industrial and economic growth; and  -92- ii. these disparities are inequitable and therefore, in the interest of social justice, should be curbed. Let us examine each of these in detail. (i) Resource Allocation and Growth The location of industries is determined by a variety of factors. On the supply side, the important ones are the availability of raw materials, infrastructure (especially power and transport) and labor (both skilled and unskilled). On the demand side, the crucial factor is the size of the market for the commodity and its proximity to the production site. We will not elucidate on the various theories of location which have been developed because these are well known. However, the crucial point to recognize is that it would be naive to expect regionally balanced inudstrial growth in an economy. In the first instance, natural resources are unevenly distributed and therefore, industries utilizing these will tend to develop in an uneven fashion. The tendency for industries to concentrate in large towns and cities can also be easily explained in terms of economies of scale and agglomeration economies. Thus it is natural to expect concentration of industry in certain regions and metropolitan areas. But, having said this, it must be recognized that there are certain factors that could lead to this concentration exceeding "optimal" levels. This may result in an inefficient resource allocation with harmful implications of growth. Now, what are these factors? A basic requirement for industrial growth is the availability of infrastructure (power, water, transport, communications, etc.). The natural economies of scale that characterize infrastructure sectors have meant that these are operated by the public sector in most countries. The  -93- Governments must ensure that the provision of infrastructure in every location is in keeping with the current economic levels of different regions and their growth prospects. This requires, on the one hand, timely and flexible responses to regional requirements and careful long-term planning on the other. There are good reasons to believe that the provision of infrastructure in India has not fulfilled these requirements. First, the decision to undertake infrastructure investment has been more frequently based on political judgment than sound economic analysis. Second, resources for infrastructure investment are frequently supplied from other than the beneficiaries; in the form of Central or State government grants. The absence of a cost recovery system allows deviation of supply from demand. Apart from faulty allocation of infrastructural investment, there are certain other factors that may lead to distortions in the location of industries. First, there are certain externalities which private entrepreneurs do not incorporate in their investment decisions. For example, pollution and congestion that arise from setting up industrial units in cities are social costs which entrepreneurs neglect. Thus, the resulting concentration of industry in cities is inoptimal from a social viewpoint because of tlhe divergence between private and social costs. However, the introduction of measures to eliminate such externalities do not necessarily lead to a reduction in concentration. It may, as Tolley (1974) demonstrated, lead a big city to further expansion. Second, it is argued that certain regions and metropolitan areas possess industrial potential but entrepreneurs are not willing to invest there because of high risk aversion on their part.  -94- The extent to which these factors are important in the Indian economy should be carefully studied before corrective policies are undertaken. Unfortunately, there is a gross deficiency of research in this important area and policies have been initiated without a sound empirical base. In particular, it is surprising that there are no satisfactory published studies which compare the industrial potential of different sizes of cities and towns, and assess whether the provision of physical and social infrastructure is in keeping with these potentialities. 1/ To be sure, there are numerous industrial potential surveys of towns, cities and regions 2/; but these tend to be mere identification of possible industries that could be established. The surveys are non-analytical in character and cannot be used as a basis for making decisions regarding the precise allocation of scarce budgetary resources for development of infrastructure in different cities and towns. Granting that the existing locational pattern of industry is inoptimal because of the various factors outlined earlier, the question arises as to the best policies for rectifying this. The best policies are clearly those which tackle the root of the problem. As far as the distribution of infrastructure is concerned, certain regions and metropolitan areas should be identified where there is significant industrial potential but which suffer from deficiencies in 1/ A few isolated studies do exist, for example, Kulkarni & Kulkarni (1968) and Ministry of Works& Housing (1977). However, these are marred by serious lacunae. To illustrate, both the reports just referred to concentrate on infrastructure provision and do not relate it to industrial potential. 2/ See, for example, I.D.B.I. et al. (1972), F.I.C.C.I (1977).  -95- infrastructure. Adequate physical and social infrastructure should then be provided to these areas. The user-charges for public services in cities should be raisecl to reflect their true costs. This will not necessarily reduce or increase in-migration to large cities. User-charges are desirable not only as a way of rationing demand for services but also for checking whether investments are justified based on revenues and costs of increments in infrastructure. Further, the prices that industries in large cities have to pay for the utilization of infrastructure should be raised to incorporate externalities like pollution costs. From the viewpoint of efficiency in the allocation of resources, the best set of measures to correct inefficient locational distribution of industries operates via readjustments in infrastructural investment and the prices charged for their utilization. Direct industrial location policies such as licencing and input rationing are poor substitutes for these measures. However, if some of the latter are not workable (for example, large increases in public transportation fares are often politically infeasible), then, as a second-best solution, one may have to resort to industrial location policies. However, even here, the choice of instruments must be devised carefully. In particular, it would be advisable to operate through incentives and disincentives rather than through controls. This is because it is possible to fine tune the former according to the extent of excess concentration of industry. For example, taxation could be used as a substitute for increases in the prices charged for the utilization of infrastructure. Controls are, by nature, crude devices and also result in large discretionary powers being granted to inflexible bureaucracies.  -96- Industrial location policies operating via incentives would probably also have to be utilized to induce entrepreneurs to invest in new areas that have just been provided with the necessary infrastructure. There is always a natural reluctance on the part of industrialists to venture into new regions even though these may be profitable. Besides, it takes some time for these enterprises to overcome certain initial problems that arise from the newness of the region. Thus, analogous to the "infant" industry argument in the theory of international trade, these industries may have to be subsidized to some extent to get over these initial problems. However, this should be done only for a limited period and gradually phased out, otherwise the infants will never grow into adults. (ii) Equity The other argument cited for industrial dispersal is that regional "imbalances" in economic growth are inequitable. By influencing the location of industries, the Government seeks to correct these "imbalances" and thus promote a more equitable pattern of economic growth. Two basic issues can be raised here. The first relates to the basis for characterizing regional "imbalances" as inequitable. The second is whether industrial location policies are the right instruments for balanced regional growth. Let us take these issues in turn. In the world economy today, there are marked differences between the developed and less developed countries. The reasons for the evolution of this international economic order are complex. Factor endowments, technology, trade, entrepreneurial motivation, the role of the state and good fortune have all played their roles. The same factors can also be expected to operate within an individual economy. Thus, in all economies,  -97- areas displaying rapid growth co-exist with those undergoing stagnation and retardation. In the international economy, there are all sorts of barriers to international migration so that one could decry the sharp "north-south" division as inequitable to the extent that low income populations are deprived of the opportunity of bettering themselves by migrating. However, in an individual economy with free mobility, such an argument is untenable. But, in the Indian context, strong sociological bonds - resulting from cultural and ethnic homogeneity - tie the inhabitants of individual states together. This inhibits the process of inter-state migration as; an automatic mechanism for equalizing state incomes. Thus, there is; something to be said for the equity argument at the inter-state level. However, at the intra-state level, the relative absence of obstacles to population migration from depressed to comparatively prosperous regions weakens substantially the case for direct state intervention to promote development in depressed geographical areas based on equity considerations. Now assuming that some corrective measures have to be adopted to reduce these inter-state disparities in economic levels in the interest of equity, is the location of industry the best measure? The answer is negative. The experience with the policy in the past speaks eloquently for this. We shall turn to this in the next section. 2. Limitations of Industrial Location Policy The preceding analysis casts serious doubts on the fundamental justifications for initiaLting direct measures to influence industrial location. Without giving; adequate thought to these basic issues, a wide range of industrial location policies have been pursued in India. Some of these have had little effect on the geographical distribution of industries. In general, most of the policies have resulted in a wastage  -98- of resources and led to distortions in allocative efficiency and growth. Let us expand on these observations and provide illustrations from Chapter II. First, we look at some policies operating through controls. Consider the use of the licensing mechanism to promote industries in backward regions. Now, the basic fact is that licensing can be used to prevent industries from being set up in certain regions but cannot induce industrialists to invest in non-viable areas. Therefore, if licensing is utilized actively for locational purposes, the result will be a fall in investment in the restricted regions without a compensating increase in the others, thus defeating the basic objective. Moreover, by favoring applications from certain states or backward regions over other more viable propositions, a loss in efficiency results. Policies to maintain a uniform price in all regions of basic inputs like cement, iron and steel (through freight pooling arrangements) have resulted in various distortions. For instance, producers of these inputs located close to raw materials find themselves at an advantage vis-a-vis those located in proximity to markets. Thus, ironically, these policies encourage locational concentration in production of these inputs which offset to some extent their broader impact on development of the other industries. Even regarding the latter, the uniform-price policy discourages the use of substitutes (for example, wood for cement) in areas located far from the producers of these inputs. The other types of distortions have been described in Chapter II. There has been great disappointment regarding the lack of spread effects of large public sector units on the local economy. However, this  -99- is quite natural in an area characterized by low agricultural productivity and poor infrastructural support (the state of Bihar, for example). Furthermore, there has been concern about the setting up of inefficient public sector units because of the importance given to location in backward regions. But, in practice, as we have seen, techno-economic considerations have been predominant in public sector location and, to some extent, their location in low-income states is fortuitous. Let us now look at the case of policies operating through the market mechanism. The various types of incentives for setting up and operating industrial units in backward districts have already been described. The backward districts that are eligible for these concessions have been defined in a very broad manner, for example, those that qualify for concessional finance constitute more than two-thirds of the area of India as a whole. It wouLd be naive to expect that, merely by virtue of this broad definition, industrial development would be generated in all these areas. Only a small handful of these districts possess the environment (especially in terms of infrastructure) for industrial growth; shrewd entrepreneurs would only venture into these few areas. They will set up units in the other regions only under exceptional circumstances - for example, the availability of a specific kind of raw material. This sort of locational decision would be made even in the absence of incentives. Thus, the taking up of concessions by these units represents an inefficient use of Government expenditure. To the extent that industrial development is only feasible in towns and metropolitan areas of certain size, the question may be legitimately raised whether the areas eligible for incentives should be  -100- defined in terms of towns rather than districts. In practice, of course, industry will typically congregate in the most viable towns of districts defined as backward. However, defining backwardness in terms of districts implies that certain viable but industrially under-developed towns in "non-backward" districts are not benefited. Finally, another serious criticism of concessional finance and the investment subsidy is that they operate through cheapening the cost of capital. In a labor surplus economy like India, this will add to distortions in factor markets. The poor response to the transport subsidy scheme to encourage industrial development in hilly backward regions is again a natural result of the absence of basic preconditions for industrial growth in these areas. Moreover, just as in the case of price equalization policies, the transport subsidy could result in misallocation of resources by inducing manufacturers to substitute raw materials from outside for those available locally. Finally, yet another example of the futility of industrial location policies where preconditions for viability (especially proximity to markets) and heavy infrastructural support (like transportation and communications) are lacking is provided by the rural industrial estates program. As we have seen, efforts were made to promote rural industrialization by setting up rural industrial estates in remote areas. The outcome, naturally, was that these estates performed very poorly as reflected by low occupation rates of sheds. It proved to be extremely difficult to attract small entrepreneurs to these estates. Thus, again, there was enormous wastage of resources from the expenses incurred in constructing these estates.  -101- These are some examples of serious inefficiencies generated by industrial location policies. They have arisen largely on account of not enough thought being given to the rationale of these policies. Our exploration of their ramifications reveals that a thorough overhauling is required in the approach towards locational concentration of industry. In particular, we have argued that wherever it is thought that this concentration is inoptimal or inequitable, policy makers should operate with incentives rather than controls. Of two typical forms of incentives, financial incentives and infrastructure provisior, the latter has several desirable properties. Iit provides long-lasting support for industrial development and its benefits are received by all activities within the service areas. Although infrastructure provision may be more costly than the alternative, it would be more consistent with the development objectives for which industrial location policies are maintained. Incentives should be used sparingly to overcome lethargy or risk aversion on the part of entrepreneurs toward locational shifts. If the current battery of industrial location policies is allowed to continue operating in an unmodified manner, there is the real danger that, instead of correcting the inoptimality of industrial location, these policies will add further to the distortions.  -102- POSTSCRIPT National Committee on the Development of Backward Areas In November 1978, the Planning Commission set up a National Committee on the Development of Backward Areas (NCDBA) to " formulate appropriate strategies for effectively tackling the problems of backward areas." 1/ To assist the NCDBA, a Working Group on Industrial Development in Backward Areas was formed with instruction, among other things, "to recommend programs and policy mesures for influencing and controlling the locational pattern of industrial activity." 2/ The views of the NCDBA on industrial location policy were submitted in October 1980 in their "Report on Industrial Dispersal". The recommendations of the report are currently being considered by the Government. The NCDBA report is an important policy document which may result in significant modifications in industrial location policy. Therefore, it is necessary for us to assess its recommendations in the light of our study of the working of industrial location policies in India The NCDBA makes the distinction between two sets of industries: (i) medium and large industries; and (ii) small, ancillary and agro - industries. It recommends different types of policies for influencing the location of each of these sets. 1/ NCDBA (1980), Page 197. 2/ NCDBA (1980), Page 198.  -103- The major recommendations are summarized below. (i) Policy Recommendatios for Medium and Large Industries The NCDBA advocates the "growth center" concept for developing these industries in backward areas. A hundred new industrial centers should be identified and provided with sufficient infrastructure for attracting industry. Three basic criteria have been used for selecting these centers: "1) They should have a population of 50,000 or more as per the 1971 census; 2) They should have less than 10,000 workers in non-household manufacturing as per the 1971 census; 3) They should not be near existing centers. Existing centers may be defined as all centers with a level of employment in non-household manufacturing exceeding 10,000. Nearness may be defined in terms of the following cut-off distance from each category of existing centers: Level of employment in non- Cut-off distance beyond household manufacturing in which new centers should existing centers be chosen Over 150 thousand 150 kms. 50-150 thousand 100 kms. 25-5 thousand 75 kms. 10-25 thousand 50 kms. " 1/ 1/ NCDBA (1980), pp. vi and vii.  -104- Furthermore, to make allowances for disparities in industrial development between states, the NCDBA recommends that these hundred centers be allocated between industrially "developed" 1/ and "backward" states 1/ in the ratio 30:70. Further preference may be shown to the hilly states of Jammu & Kashmir and Himachal Pradesh and the states (excluding Assam) and Union territories of the North-eastern Region by allocating them ten of the 70 centers reserved for backward states. The state-wise distribution of the centers should be proportional to area and population (with equal weights for both). Once the hundred industrial centers have been identified, the NCDBA recommends tht an Industrial Development Authority (IDA) should be established in each of them. The IDA should coordinate the efforts of the various implementing agencies required for developing the center, concentrating on the provision of infrastructure in the forms of power supply, access to regional and national transport networks, telecommunications, industrial estates, housing and urban infrastructure. (ii) Policy Recommendations for Small Industries, Ancillary Industries and Agro - Industries The NCDBA is of the opinion that existing industrial estates in the selected industrial centers should be rehabilitated and new estates should be established in those centres which do not possess one already. Moreover, to further decentralize industry, an additional 100 industrial estates should be constructed in areas away from the industrial centers identified above. l/ These categories have been defined according to whether value added per capita in manufacturing in a state exceeds or falls below the national average.  -105- As we noted in Chapter II, a number of "District Industries Centers" (DICs) have been established since 1978 as district-level governmental organizat:ions to oversee the development of the small scale sector. The NCDBA reiterates its faith in the DICs as coordinating bodies for assisting the sma]Ll scale sector. It also advocates more effective implementation of existing Government policies like priority purchase of small sector output, adequate provision and allotment of raw materials, and technical and financial assistance. The NCDBA feels that the rehabilitation of existing industrial estates and the development of new ones will be made possible, among other ways, through the encouragement of ancillary industries. It recommends that the latter should be developed by exerting active pressure on public and private sector medium and large units to promote ancillaries. These should be provided with raw materials and the "mother" units should guarantee the purchase of their output. Agricultural development will generate the need for certain types of industries like repairs, services and agro-processing industries. The NCDBA recommends the encouragement of such industries through appropriate policies. (iii) Recommendations for Maximizing the Local Impact of Industrial Development Finally, the NCDBA advocates certain measures for maximizing the impact of industrial development in backward areas on the local economy. As far as unskilled labor is concerned, it recommends that the current obligation of public sector units to recruit through the local employment exchange should be continued and the possibility of extending this requirement to private sector units should be considered whenever the latter receive concessions for location in backward areas.  -106- For skilled labor, the NCDBA feels that the establishment or improved functioning of local training institutes would be necessary. Local entrepreneurship should also be actively encouraged through entrepreneurship development and training programs. A Critique of the NCDBA Report It is in regard to the location of medium and large industry that the NCDBA has made innovative policy suggestions. In general, their recommendations for small and ancillary industries seek to operate through programs already being implemented. Therefore, we shall consider first the suggestions for developing industrial centers to attract medium and large industry. There is a healthy recognition of agglomeration economies by adoption of the growth center approach. As the report states: "Left to itself, industry goes to urban areas because of the availability of infrastructure and ready demand for products. .... Public cannot ignore these advantages of agglomeration and any attempt to distribute large and medium industry, at any rate, throughout the country in small lots is bound to fail. Hence, the aim of policy must be to develop viable industrial growth centers in backward regions." 1/ Furthermore, the NCDBA is correct in isolating insufficient infrastructure in small towns and urban centers as a major inhibiting factor in their industrialization. However, there are certain problems with the criteria that the NCDBA has used to identify the most viable industrial centers. The mere utilization of the indices of population size and number of workers in 1/ NCDBA (1980), Page 80.  -107- manufacturing will nolt give an accurate picture of the industrial potential of particular towns. A proper assessment would involve thorough techno-economic surveys to ascertain the raw material base, market size, linkages with larger iindustrial centers, characteristics of the labor force and existing availability of infrastructure. It is only on this basis that non-industrialized towns can be arranged in some sort of hierarchy of potential for industrial development. Therefore, the cut-off point can be defined atccording to the availability of funds for infrastructure developiment. In terms of these prerequisites for proper identification of growrth centers, the NCDBA's criteria are far too simplistic. Moreover, the NCDBA introduces the additional dimension of inter-state equity by giving additional weightage to "backward states." From a purely efficiency viewpoint, a national perspective is preferable since intra-state disparities are just as marked as inter-state differences in industrial development. However, as we pointed out in Chapter 2, political realities make it imperative that economic disparities across states be taken into account in all national policies. In recommending the development of small and ancillary industries, the NCDBA does not examine the critical issue of the efficiency of the smaLl scale sector in relation to the medium and large sector. Furthermore, iit is not at all obvious that modern small scale and ancillary industries are more easily decentralized than medium and large industries. In fact, the NCDBA does recognize the convenience of locating ancillaries close to large "mother" units. Despite this, it recommends that ancillary units should be located in industrial estates away from industrial centers. It is clear that the lessons from the unsuccessful experiment of developing semi-urban and rural industrial estates have not been learned. Moreover, compelling public sector units to purchase from  -108- small ancillary units where economic considerations dictate otherwise, would mean that there is an implicit subsidy for developing these units. This should be taken into account when implementing policies for promoting ancillarization. The NCDBA advocates the development of agro-based industries. However, it fails to consider household and handicraft industries (which are the most dispersed industries in the small sector) and to re-evaluate the plethora of programs and policies that exist to assist these. The process of economic development implies gradual integration and homogenization of labor markets. The recommendation of the NCDBA that units should be compelled to recruit from local areas militates against this concomitant of economic growth. It implies discouragement of migration and further segmentation of labor markets which is quite retrogressive from the viewpoint of promoting growth. The NCDBA's recommendations to promote training institutes and entrepreneurship development and training programs are in the right direction. However, here again, limited budgetary resources imply that decisions regarding the location of new training centers should be carefully taken, keeping a national perspective in mind.  -109- BIBLIOGRAPHY Alexander, P.C., Industrial Estates in India, Asia Publishing House, Bombay (1963). Armstrong, H.W. and Taylor, J., Regional Economic Policy and its Analyses (1980). Bandyopadhyaya, K., Industrialisation through Industrial Estates, Bookland Pvt. Ltd., Calcutta (1969). Bhagwati, J.N. and Desai, P., India : Planning for Industrialisation, Oxford University Press (1970). Bharti, R.K., Industrial Estates in Developing Economies, National Public House, Delhi (1978). Bombay Metropolitan Regional Planning Board, Regional Plan for Bombay Metropolitan Region, 1970-91 (1974). Bureau of Public Enterprises (Government of India), Annual Report on the Working of Industrial and Commercial Undertakings of the Central Government. Calcutta Metropolitan Planning Organisation, Basic Development Plan for Calcutta Metropolitan District (1966). Central Statistical Organisation (Government of India), Statistical Abstract. Delhi Development Authority, Delhi Master Plan, Vols. I & II (1961). Federation of Indian Chambers of Commerce and Industry, Industrial Development Potential Survey of Selected Towns of India, Delhi (1977). Godbole, M.D., Industrial Dispersal Policies, Himalaya Publishing House, Bombay (1978). Hanson, A.H., The Process of Planning, Oxford University Press (1966). Hazari, R.K., Industrial Planning and Licensing Policy, Final Report, Government of India, Planning commission (1967). The Industrial Credit and Investment Corporation of India Ltd., Attracting Industries to Developing Areas (1975). Industrial Development Bank of India, et. al., Industrial Potential Surveys of Various States and Union Territories (1972). Industrial Development Bank of India, Industrial Development of Backward Regions (1974;).  Industrial Development Bank of India/National Committee on Development of Backward Areas, Seminar on Industrial Development of Backward Areas, Bombay (1980). Jain, O.P., Rural Industrialisation, Commercial Publications Bureau, Delhi (1974). Kulkarni, G.S. and Kulkarni, A.P., Community Cost of Industrial Location at Bombay and Nasik, The State Industrial and Investment Corporation of Maharashtra LTD., Bombay (1968). Menon, K.S.V., Development of Backward Areas through Incentives, Vidhya Vahini, Bombay (1979). Ministry of Works & Housing, Government of India, Report of the Task Force on Planning and Development of Small and Medium Towns and Cities, New Delhi (1977). Mishra, R.P., Sundaram, K.V. and Prakasa Rao, V.L.S., Regional Development Planning in India, Vikas Publishing House, Delhi (1974). Nagaiya, D., Industrial Estate Programme, Small Industry Extension Training Institute, Hyderabad (1977). National Committee on the Development of Backward Areas, Report on Industrial Dispersal, Planning Commission, Government of India, New Delhi (1980). Office of the Registrar General, Census of India (General Economic Tables), 1961 & 1971. Planning Commission, Government of India, Report of the Working Group on Identification of Backward Areas (1969). Rao, R.V., Rural Industrialisation in India, Concept Publishing Company, Delhi (1978). Sanghvi, R.L, Role of Industrial Estates in a Developing Economy, Multi-tech publishing company, Bombay (1979). Sekhar, A.U., Factors in India's Industrial Development 1965-75, Unpublished Ph.D. dissertation, Princeton University (1981). Sundaram, K.V., Urban and Regional Planning in India, Vikas Publishing House, Delhi (1977) Theil, H., Economics and Information Theory, North-Holland Publishing Company, Amsterdam (1967). Tolley, G.S, "The Welfare Economics of City Bigness," Journal of Urban Economics, 1., 321-345 (1974).  - 111- APPENDIX I Maximum and Minimum Values of Hirschman-Herfindahl and Theil Indices. Definitions Pi percentage share of a region i in total value added in manufacturing qi : percentage share of a region i in total population N number of regions Hirschman-Herfindahl (HH) Index E p2/100 Theil Inequality Index : E p log(p /qi) The maximum value of the HH index is 100 when the entire manufacturing sector is located in one region. The minimum value is obtained when industrial value added is distributed equally in all regions. Then, the value of the index becomes 100/N. In Table 32, where we have considered 13 states while calculating the index, the minimum value is 7.7. The waxi.T-im value of the Theil inequality index is infinitely large when p.>q= 0 for some i. Intuitively, there is an enormous divergence between the share of a region in industry and its share in population. On the other hand, when pi = qi for every i, i.e. the shares of regions in both industry and population are exactly equal for every region, the index attains a minimum value of 0. Since the two indices are constructed on entirely different lines, it is obvious that any value of one index cannot be compared with that of the other. It is only valid to compare values of the same index and these must also be seen in relation to the maximum and minimum values of that index.  -112- The Theil inequality index has certain advantages over the HH index. The chief advantage relates to its "aggregation" properties. It is beyond the scope of this appendix to elaborate on this; a detailed explanation is provided in Theil (1967).  -113- APPENDIX II A Shift-Share Analysis of Industrial Growth in the lMjor Industrial States of India In explaining divergences in the industrial performance of different states, one of the important factors to be kept in mind is the product mix of industry in the states. A technique of isolating this factor is "shift-share analysis." The method is best explained by Armstrong and Taylor (1980): "Shift-shaLre analysis is a method of calculating the extent to which the difference between a region's growth and the nation's growth can be explained by the region's industry-muix. The method is easily explained. We start with three definitions. 1. Regional growth rate (g r Er t- Zr0 g m Zr0 1ii where: r. = regional employment in industry i Erl t sum of employment across all industries in the region t - finaL year of study period o = initial year of study period  -114- 2. National growth rate (g ) t 0 n Ent - En0 i i