AN ANALYSES DE WAEE WENDS m ELANUEACEBRING ENWSTREES IN INDSA THESE‘S FOR THE DEGREE 03" PH. 9. MEGHRGAN STATE UNWERSITY ‘11. N, KBFSfiNAN ‘i 9 6 7‘ I .ll. I‘ll ’1“ Ark ‘1 k ‘ CONTENTS CHAPTER 1 Introduction Chapter 2 Wage Structure in Manufacturing Industries I Change in Money Wages II Change in Real Wages III Change in Employment and its effect on Wages CHAPTER 3 The Inter-Industry Wage Structure I Change in the Ranking of Industries in the wage Heirarchy II The Relative Change in Money Earnings Compared to its Original Level III The Relative and Absolute Change in Money Earnings Compared to the Dispersion of the Wage Structure IV Variation in Wages and Employment CHAPTER 4 Inter-State Wage Structure I Ranking of Average Annual Money Wages By States II Government Wage Policy and Inter-State Differentials III Changes in Real Wages in the States CHAPTER 5 Issues Raised in the Formulation of a Wage Policy in India 1. Government Wage Policy and the Level of Wages 2. Wage Level and Wage Differential l. Wage Policy in Economic Development a. Surplus Labor and the Wage Level b. Skilled Labor and Wage Differential c. Variation in Wages and Employment Appendix Tables Bibliography 20 20 44 52 68 69 73 77 100 109 109 122 131 140 142 148 151 151 153 156 163 TABLE 10. 11. 12. 13. 15. 16. 17. LIST OF TABLES Average Annual Wages in Manufacturing Industries - A11 Indi‘ 1950-1960 0.00.00.00.000....OOOOOOOOOOOOCCOOOO... Industry Ranking by change in Average Annual Money Wages - All India 1950‘1960 .0........s.s.s.0.............. State Ranking by Change in Average Annual Money Wages - All India 1950-1960 CCCCOOOOOOOOOOCCOOOOOOOCOCCOOOOOOOOOOOO National Income by Industrial Origin - All India 1951-52 C0 1961-62 coco-cocoaeeeaaaaaeeeeaeaeeeeaaaaooosoao Per Capita Income and Index Numbers of Earnings by Category of Employment - All India, 1950-51 - 1960-61 ..... Range of Increase over Existing Wages and Number of Workers Benefited by Recommendations of Wage Boards ....... Index of Real Wages by Industries and States - All India, 1951-1960 0.000.000.0000..OOOICOOOOOOOOOOOO0O... Index of Emplvyment - All India, 1951-60 (1950 = 100) ..... Overall Trend in Employment (1950-60) by Industries and States O....I..........C...................... ..... O... Industry Ranking by Change in Employnent (1950-60) ........ State Ranking by Change in Employment (1950-60) ........... Effect of Fixed Versus Current Employment Weights on the Indexes of Money and Real Wages, 1950-1960 (Industries) Effect of Fixed Versus Current Employment Weights on the Indexes of Money and Real Wages, 1950-1960 (SCCCCI).soo Ranking of Ceramics, Sewing Machines and Starch by Employment Among thfl 26 IHdUStries, 1950'60 000000000000... Ranking of Assam, Bihar and Andhra by Employment Among the 12 States, 1930-1960 0....0000000000000000.00.00.000... Ranking of Industries According to Change in Average Annual Earnings (R8.) - All Indid - 1950’1960 ..00000000... Spearman Rank Correlation Coefficient for Average Annual Money Wages, 26 Industries, All India - 1950-1960 ......... PAGE 21 23 25 30 32 38 54 54 56 61 62 63 65 69 72 TABLE 18. 19. 26. 27. 30. 31. 32. 35. Ranking of Money Wages in 1950 and Percent Change in Money Wages, 1950-1960 for 26 Industries, All India ....... Inter-quartile Range and Median of Average Annual Money Wages with Absolute and Percentage Differential, 26 Industries, 1950-1960 0.00.0.0...OOOOOCOCOOOOOOOIOOOOOO. Producer Concentration in Selected Industries ............. Comparison of Fixed Capital per Worker and Money Wages for Industries 1951-1960 00....OOOCCOOOOOOOICO0.00.00.00.00 Percentage of Units Paying and Workers Receiving Dearness Allowance According to System of Payment, 1958-59 Wages and Employment Variability A1 ut AVerage Rate Of Change ’ 26 IndUSCriPS, 1950“l9h’ eeeeeeeeeeeeeeeeeeoeee State Ranking of Average Annual Money Wages for Each Year 1950'1960 eeeseeeeeeeeeeeeeeeoooeeeoeeeeeeeeeeeeeeeeee Spearman Rank Order Correlation for A erage Annual Wages by States 1950'1960 oeeeeeeeosee-eeeeeeeeeeeeeeeeeeee Wage Differential Between the Highest and the Lowest Paying State 1950’1960 e....se....ee..o.0-eee.......ee.eeee Deviation in Average Annual Wage of the High st Paying State and the Lowest Paying State from the Average oi dll States ' 1950-1960 0...O0.00000000.00.0000.0e.......0.. State Ranking of Average Annual Money Wages for 1956, With Percentage Change in Ranks for 1955 and .)60 ......... Wage Differential Between the Highest and Chi Lowest Wage Industry for Bflmbay State, 1950-1960 000......eeeoeeee Wage Differences for Similar Classes of WOFAVTS in Different States, 1958-1959 (Average Earn;ngs Per Day) ... Percentages of Basic Wages and Cash Allowance Including Dearness Allowance to Total Per Capita Axvrage Annuil Earnings in Industries by States for 1957 an“ 1959 ........ Change in the Consumer Price Index, }“ch Wages and Real Wages for the States, 1950-196“ oeeeeeeeeeeeeeeeoeeeee National Structure of Capital, Enwivmnt. Output and Value Added for Census and Samplr Sector Factories, 1959 and 1960 .................,ee.0.....00.0............0O PAGE ‘4 78 82 86 103 109 110 112 113 119 120 134 168 ""1” (‘Ir‘i'QI'ntfi . oA A. ’\~ lEE£2§B££lgfl The period 1950-1960 has been one of general economic expansion for the Indian economy in which incomes rose and industry eXpanoed. There was an increase in the wages of workers in mining and manu- facturing industries. The government's earlier pro— nouncements on wage policy showed a deep concern For upgrading the economic position of the factory .worker in the interest of social justice. Later, the claims for economic development led to a revision of this attitude in the interests of promoting saving, capital formation, and growth. A wage policy to facilitate growth had its economic and social im- glicatiohs. There was a call for wage restraint to prevent wares from exerting an inflationary pressure 9n the eegnomv. From the social point of View, wage policy had to m»ve in the direction of securing a redintitn of in0qua1ity in income distribution. This study is based on the belief that during the geriid 1950—1960, considerable changes in wage trends and differentials had taken place, and the nature 0” tnase changes ani the e‘onomic factors influencing them h I had not been clearly understood. ihe study cuiefly runnerns itself with these aspects of WSQPS 3‘ they relate to manufacturing inluSlrlPS for the above period. In the literature on economic development, a familiar theory holds that the level of wages in the industrial sector is constant during the early stages of industrialization of a developing economy, which is 1 The existence of this endowed with a surplus of labor. surplus is believed to give rise to a perfectly elastic supply curve of labor to industries which would hold industrial wage rates at a constant level. This theory raises interesting possibilities for the study of wage trends in India and for considering the issues raised in the formulation of a wage policy. These follow from the fact, that though there may be factors tending to depress the level of wages in Indian in- dustries, it is possible even at an early stage of industrialization, for wages to rise in particular industries or regions, or reveal considerable differ— entials between industries or regions, or to do both on a more disargreaated level. This makes it necessary to distinguish between the level of wages and wage differentials in order to consider the influence of 1John Fei and Gustav hanis, Development of the Labor um...- ‘0‘... §grplu§ Economy: ghegry_ang:igligy, the Economic Growth Center, Yale University, 1964, pp. 260 ff. For earlier references see, W. A. Lewis, "Economic Development with Unlimited Supplies of Labour”, Tawsnshe sEe.I:-.S.9h.0.9.1. 9.1.15.9 9.9.01“. island. .8931 al. .Styfi 1. .e__8 . V01. XXII, May 1951:, pp. 139-91; also his TUnlimited Labour: Further Notes", same Journal, Vol. XXVI, No. 1, Jan. 1958, pp.1-31. - 3 - the various economic factors, some of which would influence the level, while others would affect the shaping of differentials. From a policy standpoint, the role of wage differentials assumes importance in focussing on relative supply scarcities.for various categories of labor at an early stage of development. This chapter will be devoted to a review of the literature specifically relating to studies on wage trends and differentials in Indian industries. The various approaches adopted in previous studies will be emphasized and this will serve to show the need for undertaking the present study. I shall comment on the work of seven writers. Three of these works are doctoral dissertations in American universities and belong to a separate category of academic research. Two others are the work of researchers in government institutions. One is the outcome of research at the industry level in the private sector and one more that of an individual researcher. Of the works reviewed, two may be regarded as long term studies dealing specifically with the cotton textile industry,and the rest as focussing relatively on shorter periods. Each study is referred in a chronological order ex- cept the two long term studies which are referred to in the end, as they emphasize the historical aspects of development in a specific industry. “it; -14... l the trends in real S. A. Palekar investigated wages in India for the period 1939-1950. The main purpose of his study was to "investigate as objectively as possible such important facts about the impoverish- ment of industrial workers in India."2 He considers industry and state series of indexes on real wages, money wages, employment and cost of living for the period 1939-1950. He shows that in engineering, minerals and metals, government local fund factories, paper and printing, cement, skins and hides and miscellaneous groups of industries, real wages had declined relative to the position held by them in the base year. The period from l9fl2 to 19“? stands out rather prominently as one in which real wages were below the base year level in all the industries, whereas textiles and the. chemical and dyes group stand out prominently amongst those industries which recorded real wage gains in the post war period.3 Among the states, Bihar, Bombay, Orissa and Punjab showed a decline in real wages com- pared to their base year position. He observes that 1S. A. Palekar, Real Wages in India, International Book House (PrivatE) Ltd., Bombay, 1962. The book forms the author's dissertation for the Ph.D degree at Harvard University. 2 Ibid. p. 2. 3 . Ibid. p. 157 - 5 - the movement of real wages in the various states dis- closed generally a pattern of continuous decline between 1939 and 19h3 and of a steady rise thereafter ' till the end of the period.1 Though his study deals mainly with real wages, he also draws some conclusions on the movement of money wages. He finds that money wages rose in all the industrial groups throughout the entire period.2 In the states, the picture, according to him, was one of continuous rise with minor recessions. "Broadly speaking, money wages by 1950 reached a level of between three and four times the 1939 level in almost all the states."3 The value of Palekar's work lies in t1! prodigious amount of statistical material presented by him in this relatively neglected field of enquiry in Indian economics. In the maze of data one finds it difficult to pin point his main conclusions. One broad con- clusion he reaches is summed up when he says, "An examination of the all-India real wage series revealed that by 1950, the workers had Just managed to regain their base-year level of real wages.."u He stresses chiefly two factors for this, namely the failure of lIbid. p. 111. 2Ibid. p. 158. 3Ibid. p. 112. "Ibid. p. 175. earnings to keep pace with the steep rise in working class cost of living on the one hand and the chronic gap up to 1947 between the earnings of the workers and 1 His main the profits of the industry on the other. concern throughout has been to establish the fact of the impoverishment of the industrial worker in India. He also shows that, while money wages have gone up, real wages have not increased but have Just managed to reach their base year position. However, the study of real wage trends in India is beset by the limitations surrounding the use of the existing data on consumer price index numbers, and in the face of this, the rather empratic conclusions reached by him must be questioned. Later in this study we will show these limitations in examining the developments in real wages after 1950. Philip Kotler has studie the course of industrial wages and industrial wage policy in the post 1939 period.2 He has examined the structure as well as the level of wages. In the former, he deals with wage differentials and in the latter, has presented an in- dex of real wages of his own for the period 1950-1959. 1 Ibid. p. 107. 2 Philip Kotler, Problems of Industrial Wage Policy in India, unpublished Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, Aug. 1956. In dealing with inter-regional wage differentials, he chooses the years 1939, 19h7, 1950, and 195“ as the years to consider the changes in the ranks of nine states and the size of the differential between the high paying and the low paying states, choosing two in each case. He concludes that wide differences existed among the regions and that these differentials had shown a tendency to narrow since 1939.1 He does not, however, draw a comparison between absolute and per- centage differences. All his conclusions are based on percentage differences. He then deals with inter- industrial differences, and ranks eleven industries for the years 1939 and 19u9, according to their average annual earnings, in order to examine the rankings of the major-industries in these two years. Next he ranks several other industries based on the post 1950 in- dustrial classification for the year 1959, and has merely listed them for 1950, 1951, 1952, and 1953. He observes that between 1939 and 1909 no change at all was observed in the position of the lower paying industries. .His conclusion on wage differentials is that "the substantial wage differential between the 1 Ibid. p. 260. J! I‘lll f'llllll’ l - 8 - highest and lowest paying industries have not shown any tendency towards compression?1 He also concludes that skill differentials in India remain substantially larger when compared to some of the advanced western countries, even though they show a tendency for narrowing.2 But the examination of wage differentials is not the main part of his study and he has not undertaken any de- tailed ranking of industries or states nor has he addressed himself to both percentage and absolute changes. Kotler's approach to the study of industrial wages goes one step further than that adOpted by Palekar. While Palekar.concentrates his attention on the level of wages, Kotler has attempted to analyse wage differentials. He further suggests certain factors for the existence of differentials between the high paying and the low paying industries. In this sense he has opened the way for further research in this direction. However, neither of them have investigated the factors responsible for influencing the level of wages or its differentials by keeping these two aspects of the question separate. For instance, in explaining the narrowing of skill 1 Ibid. p. 272. 2Ibid. p. 292. differentials, Kotler cites increased acquisition of skill, growth of unions, uniform wage fixation and government intervention as factors that would unfold and develop as industrialization progresses, contri- buting to a narrowing of extreme! differences in income and the securing of equal pay for equal work. The question may be asked whether some of these factors may influence only the level of wages while others may affect wage differentials. Or again it may be asked whether we have enough empirical evidence to draw such conclusions so that the question becomes one of examining more carefully the existing data on trade union strength, governwmnt intervention in wage fixing and on increased acquisition of skills. Before we leave the review of these two attempts to study the course of industrial wages, a word must be said about the kind of data utilized in the studies. Both studies are based on earnings data reported under the Payment of Wages Act of 1936. Kotler has clearly pointed out the limitations surrounding the use of this data.2 He shows that for undertaking a study of wage 1 Ibid. pp. 273-7“. ’) “Ibid. p. 297 ft. Under the Act,returns are published only for those employees earning below Rs 200 per month. This limit on reporting earnings tends to understate the extent of increase when workers move to a higher wage bracket from one year to the next. - 10 - trends, this source tends to understate the extent 0? increase in wages.1 However, for the earlier period which Palekar has studied, data reported under this Act was the only comprehensible source available. But in 1996 the Census of Manufactures came into existence and this source has been utilized in the present study. In reporting the data the Census does not make a distinction on the basis of a worker's earnings and does not include,on this basid,categories of workers. But it provides separate data for workers and other salaried employees. Besides, the earnings reported under the Census include. the money value of various benefits or concessions received by the workers. These in many cases could be substantial amounts, and in a period of industrial eXpansion, could increase cone siderabl'. A reference to this has not been made in the data reported under the Payment of Wages Act. Owing to these differences in the Census data, it is hOped that a clearer picture of wage trends can be presented in this study than has been possible before. Though at the time of his writing Kotler could have used the Census data, he has not done so because he 4.— To the extent we have this downward bias in the use of this source, Palekar's findings on real wages must be taken with some reservations. _ 11 _ believed that an index of earnings based on either source would not be very different.1 The next reference is to a paper prepared by A. K. ChatterjeeQ for the Indian Statistical Institute, in which the author examines wage trends of persons em- ployed in twelve manufacturing industries for the period l9U6—1952. He uses the data provided by the Census of Manufactures. According to him, the cement industry has shown the highest rise in earnings by an increase of 172. 25 per cent.3 Further, he says that in six other industries the rate of increase in earnings is above 100 per cent and in five industries it is below that. The author has not made a comprehensive study of the ester industries included in the Census of Manufactures, but has made rather dubious comparison of real earnings in two periods - one based on 1939 relying on the earlier data reported under the Payment of Wages Act and the other on 19U6 based on Census figures. From this type of comparison he says that we H could assume" that in "some of the industries under l Kotler, op. cit. p. 301, footnote. 2A. K. Chatterjee, "Trends in Labour Productivity and Wages in Twelve Selected Industries, 1996-1952", working paper of the Indian Statistical Institute, Planning Division, Dec. U, 1956. 3 Ibid. p. 17. _ 12 _ our study" (obviously referring to the C.M.I. industries) per capita real earnings in 1952 is even less than the 1 On the whole, the analysis in the pre-war level. paper is brief. Palekar's second book was completed in 19592. He says "It was begun in 1955 with the somewhat limited objective of ascertaining the impact of India's First Five Year Plan, which was then in its last year of execution, on the real wages of factory workers in India."3 Once again Palekar stresses the point that the lot of the factory worker at the end of the First Five Year Plan had not improved and that they did not even receive a living wage. Since most of his calcu- lations in this book are meant to up date the data of his prevbus work which we have already considered, we will not examine them in detail. He finds that in both industries and states there has been an increase in money and real wages in a majority of cases, but in spite of this, the real wage position of the worker had 1 Ibid. p. 19. 2 . S. A. Palekar, Problems of Wage Policy for Economic Development, Asia Publishing House, Bombay, 1962. 3 Ibid. Acknowledgment. not improved. He is extremely critical of the wage policy followed by the government, and the con- cluding chapters of the book are devoted to an almost emotional plea for wage policy to be directed towards providing the worker with‘mwifiummcal means for attaining a minimum of health and comfort. Recently a member of the research staff of the Reserve Bank of India has studied the trends in money and real wages in India for the period 1951—1961.1 He has not computed figures of his own but presents the data taken from the Labour Bureau publication, viz. the Indian Labour Statistics. The study has not drawn any significant conclusions, but merely summarizes the data. Theobject of the writer seems to be to provide a factual background for a critical examination of the current wage policy.2 He finally says "though the average annual money wages per industrial worker have shown more or less a continuous rise since 1951, real wages per factory worker have fluctuated from year to year. Secondly, there was a decline in real wages per worker I H. B. Shivamaggi, "Trends in Money and Real Wages in India, 1951—1961, Reserve Bank of India — Bulletin, April 1969, pp. 421-39. ’3 Ibid. p. 1:21. -114- during the Second Plan period ........... Thirdly, as many as U3 per cent of the factory workers were earning Rs 3 or less per day in 1958-59. These point up the need for an integrated approach to wages policy."1 A study of the inter—industry wage structure has toenmade recently by C. K. Johri and N. C. Agarwal2, based on the Census of Manufacturing data for the period 1950—1961. The authors claim that they have analysed the inter—industry wage structure from the demand side. The reason they give for this (:Dth) 1 Ibid. p. A32. 2 C. K. Johri and N. C. Agarwal, "Inter—industry Wage Structure in India, 1950—1961 - An Analysis? Indian Journal of Industrial Relations, Vol. 1, No. A, April 1966, pp. 377—419. - 16 is “partly because we wanted to compare results for two highly dissimilar economics, Via. , the United States and India, and partly due to the non-availability of data on the supply side."1 In describing the main theme of the paper they point out, "The focus of the paper is on the structure of money wage earnings in different industries, and on the analysis of the factors Operating upon it from the demand side of the market. From this analysis we also draw a few pertinent conclusions for policy purpose."2 A general hypothesis is suggested by them as follows. "In an economy, such as India, which is going through the early phases of deveIOpment, the labour market is generally unorganised and fluid. This characteristic of the market permits relatively free movement of labour, in different skill categories, in response to changes in demnnd. It follows that relative changes in money wage rates are determined by the relative excess demand for labour."3 IL; k/f-efts .15 They do not explicitely testxia, but draw support for it from their finding that during the period 1950-1960, the average industrial wage rate has been steadily rising and accompanied by significant shifts in the relative positions of different industries.‘ These l'Ibid. p. 378. 21bid. p. 379. 3Ibid. p. 390. This hypothesis is taken from Bent Hansen's essay on "Full Employment and Wage Stability", in John T. Dunlap's edited essays, Theogy_of Wags Determination. The authors acknowledge this in a foot note and say that this is sufficiently general to be of use in a particular case that is analysed by then. “ibis. p. 386. -lb- two findings are based on a study of the movement of the average daily earnings in the industries. On the basis of these findings they argue that there has been a widening of the inter-industry wage structure due to the fact that the rapidly growing industries will pay more to their workers while the stagnant and declining industries will take advantage "of imperfections in the labour market" to deliberately lag behind the pace of earnings. Theoretically, if the labor market permits relatively free movement of labor as suggested under their hypothesis, then earnings differentials ought to show a tendency towards narrowing instead of widening. This would follow from the fact that labor would move from industry to industry in search of higher rewards thus minimizing the differences between them. However, the authors also refer to "imperfections in the labor market” when dealing with the wages in stagnant or declining industries. In fact, it is not clear how they CVer come to support the hypothesis they have adOpted. For instance, it is difficult to agree with their conclusions that there has been a significant shift in the relatiVe position of different industries, for they give no detailed ranking of the industries by wage changes for each year. The chief value of the work lies in the authors' attempt at an dfldthlS of the inter—industrial wage structure in which they have tried to relate differentials to specific economic variables affecting them. But the study is not solely directed towards explaining the position in India, for their desire to compare India with the United States as the starting point of their study has resulted in their neglecting to analyse the data more carefully in the case of the former. -17.. Finally we come to some long term studies on the cotton textile industry in India, a major industry employing nearly half of the factory labor force. K. Mukherhi,‘studying wage trends in this industry, has shown that the real wage(for the average fully employed operative had increased by about 2.7 tines between 1900 and 1951.1 In another study relating to the industry in western India, he observes that the internal relations of the cotton textile industry seem to have been dominated by external demand and prise relationship in the textile market. Downward revision of wage cost seem to be difficult, except in relation to small adjustments.2 In his study of the industry in Ahmedabad, he finds that the level of real wages by the fifties of this century increased 2.53 times the wage level of the first years of the century.3 These studies mainly show the increase in the level of wages for long periods extending to fifty years. They are useful as historical references and are included here for this reason. There also seems to be a difference of Opinion among the two writers cited in this review. The increase in wages to which Mukherji alludes is contrary to a different finding by M. D. Morris who says "The evidence suggests that during the critical period, the half century before 1920, mills were able to obtain new recruits without any significant upward pressure on wage rates at l which novice mill hands were employed."4 He argues that there was no 1K. Mukherji, "Trends in Textile Mill Wages 1900-1951? Arthavijana, Vol. 1. 1959, p. 95 2K. Mukherji, "Trends in Textile Mill Wages in Western India; 1900-1951," Arthavijana, Vol. 4, 1962, p. 164. r 3K. Mukherji, "Trends in Real Wages in Cotton Textile Industry in Ahmedabad," Arthavijgna, Vol. 3, No. 2, June 1961, p. 134. 4 M. D. Morris, The Emergence of an Industrial Labor Force in India, A fitudy gf EDS figmbay Cpgggg Mills. 1854~l947, University of California Press, Berkeley, 1965, p. 233 ' - 18- upward trend in wages fill-I'll'in the cotton textile mills in Bombay before 1920. The historical approach has thus not been fully explored, and it should be stressed that in the absence of more studies on this, a full appreciation of the causes that influence wage trends as well as differ- entials at the present time would be greatly hindered. But undertaking studies based on this approach is extremely difficult in India, in the absence of wage data extending backward through time. The above review has served to show the different approaches adopted in studying industrial wage trends in India. The mojor concern among the writers has been to assess the economic position of the industrial worker. Moat of the studies have concentrated on the level ofiéages. As pointed out at the begining, it is necessary to make a distinction between the level of wages and wage differentials for considering separately the influence of the various economic factors affecting them. This study has maintained this distinction and deals first with the level of wages and then considers inter-industry and inter-state differentials,in manufacturing industries for the period l950-60. In each case, the nature of the changes in wage trends are measured with Special attention to the statistical problems relating to these measurements. The attempt to consider separately the economic factors influencing the level of wages and shaping their differential has been made. In this connection, speCAial ttention has been paid in emphasizing the limitations surrounding the use of the existing data on wages in India, which has led to pinpoint areas for undertaking further research. -19- An empirical investigation is only one aspect of a wage study. The aim of constructing the indexes and subjecting them to a statistical analysis has been to examine the izsles raised in formulating an appro- priate wage policy for economic deve10pment. These issues relate to government wage policy and its influence on the level of wages and wage differentials; the role of wage differentials in focussing on relative supply scarcities for various categories of labor at an early stage of development; and finally, the need for further research to be undertaken in the specific areas opened up by this study. Chapter 2 deals with the wage structure in manufacturing industries, chapter 3 examines the inter-industry wage structure and chapter 4 the inter-state wage structure. Finally chapter 5 discusses the issues relating to the formulation of an apprOpriate wage policy for economic deveIOpment. The appendix discusses the methods involved in constructing the indexes, and the adjustments made in the data for the years 1959 and 1960, since these were drawn from a different statistical series. The tables computed in this study are included in the end. - 20 .. CHAPTER g The Wage Structure in Manufacturing Industries It was pointed out in the introduction that we are mainly concerned with analyzing the changes in the wage structure of manufacturing industries in India. This chapter undertakes such an analysis of the overall wage structure for the period 1950-1960.1 Two aggregate series of average annual wages have been compared. One of the series has been computed from the .data on average annual wages for twenty-six manufacturing in- dustries, and the other, from the average annual earnings figures of manufacturing industries in twelve states. An. analysis of money wages, real wages, and employment will be made in an attempt to (a) show the changes as revealed by the data utilized in this study and (b) provide explanations for the changes observed. I. Change in Mongy Wages For the whole period of this study there has been an increase in money wages in both absolute as well as in relative terms. The following table shows the changes. 1Throughout the study, the absolute measure relates to 1950-1960. The relative change is based on 1950. -21- Tablell Average Annual Wages in Manufacturing;;ndustries All India, 1950-1960 (1) (23 (3) (4) §1_Industries Index: 1950.100 ' By States Index: 1950-100 1950 Rs. 927 100 933 100 1951 1041 .110 1043 111 1952 1095 115 1094 116 1953 1129 119 1127 120 1954 1117 119 1134 121 1955 1102 119 1119 120 1956 1148 124 1161 126 1957 1187 . 126 1202 131 1958 1208 129 1233 134 1959 1298 135 1237 138 1960 1461 152 ‘1426 159' 1950-60 534 52% 493 59% Source: Computed from, India (Republic), Directorate of Industrial Statistics, Report on the Census of Manufactures, and the Annual Survey of Industfiesfi Note Column 1 The average annual wage for each year has been obtained by dividing the aggregate average annual wages of all the twenty-six industries by their aggregate average annual employment for that year. Column 2 The money wage index for each year is derived by taking the weighted average of money wage indexes for the twenty- six industries for that year, the current average annual employment in each industry being used as weight. AI —. 3,—1-- -_ where AI is the money wage index for all India I- is the money wage index for industry j E] is the annual employment for industry j (ct/M ,) -22- Column 3 The average annual wage for each year has been obtained by dividing the aggregate average annual wages of all the twelve states by their aggregate average annual employment for that year. The employment figure for 1959 and 1960 is the original figure reported, and not the adjusted one, used later in the study. See Appendix, pp. 168-69. Column 4 The money wage index for each year is derived by taking . the weighted average of money wage indexes for the ‘twelve states for that year, the current average annual employment in each state being used as weight. 11 2; 13‘ £5 1: . AI: ,3, ‘ ’2‘: 51 4:! where AI is the money wage index for all India I is the money wage index for state j E1 is the annual employment for state j we may now compare both series with respect to the absolute and relative changes in average annual earnings. In the industry series, there is a decline of Rs. 12 between 1953 and 1954 and of Rs. 15 between 1954 and 1955. The index of money wages is constant for the years 1953, 1954 and 1955. In the case of the state series, there is similarly a decline of Rs. 15 between 1954 and 1955 and an increase of Rs. 31 between 1957 and 1958, followed by an increase of Rs. 4 for the following year. For the industry series, there has been a steady increase in the index of money wages throughout the period, the index never once showing a decline. The state series, however, shows that there has been a sharp increase during 1950 to 1953, a slight decline in 1955, followed by a steady increase up to the end of the period. Also, one gets the impression that the index has moved in discrete jumps, e.g. sharp increase between 1950-53 followed by a ‘ 23 - leveling off, then again a jump from 1955-1956, then again a very sharp jump from 1959-1960. The index for 1959 and 1960 must be interpreted in the light of the change in the coverage of factories introduced by the-AnnualgSurvey of Industries which replaced the Census of Manufactures in 1959. The increase in the index after 1958 could be due to the wider coverage of the Annual Survey. But, even if we take the period 1950-1958 the general increase in the index is clear, as it is seen that in both series, never once has it shown a decline below the value for the base year. We may now examine the change for the whole period by seeing which industries and states have shown the greatest increase in money wages. The following table shows the ranking of industries according to the highest increase in absolute as well as in percentage terms for the period 1950-1960. Table 2 Industry Rankinggby_Change in Average Annual Money Hages- All India* 1950-1960 (1) (2) <3) (4) V (5) Industry f' g ‘1950 . Absolute Change 1 Chan e Money wage 1 Rs. (1950-1960) Rank 1951—100 (50' 37m! .1)me Sewing Machine 1589 1986 l .125 Electric Fans 846 1447 2 171 Starch 411 973 3 237 (6) Rank WNW (1) Industry Soap Iron & Steel Chemicals Paints & Varnishes Biscuit Making Sugar 'Cement Cotton Electric Lamps Tanning Jute Textiles Glass Wheat Flour Dist. & Breweries Paper Bicycles Matches Woillen Textiles Plywood and Tea Chests Vegetable Oil Rice Milling Fruits & Veg. Canning Ceramics -24- (2) 1259 Money Wage 1130 1492 927 980 706 508 906 1113 904 692 773 586 787 902 1281 987 919 586 541 266 629 760 (3) Absolute Change Rs.(1960-19§QS 841 831 806 674 597 557 510 492 427 415 355 339 333 303 300 299 268 254 232 165 141 21 -50 (49 Rank 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 (5) Z Change 1950-100 75 54 87 69 85 110 56 44 47 60 46 58 39 39 33 23 27 28 40 31 53 3 -6 (6) Rank 12 11 16 14 9 15 10 18 18 19 23 22 21 17 20 13 24 25 Source Computed from, India (Republic), Directorate of Industrial Statistics, Report on the Census of Manufactures, and the Annual Survey of Industries. Note Wheat Flour and Distilleries have the same percentage ranks. We shall also rank the states similarly as follows. . Ranking is done numerically with #1 denoting the highest rank, in C01. 3 and 5. -25- Tab e 3 State Raggigg bx Change in Average Annual Money_wages - All India 1950-1960 (1) ~' . ' 133 (3) (4) (5) (6) State 1950 Absolute Chaggg Rank Z Chagge Rank ”flutflafies Rs; $950-$269) 195Q-100 Col. 3 x 100 Col. 2 Bihar 1133 1104 1 97 2 Delhi 1134 624 2 55 7 Bombay 1160 599 3 52 8 Madras 794 536 4 67 4 Madhya Pradesh 760 464 5 61 5 Andhra 425 451 6 106 1 West Bengal 812 425 7 52 8 Punjab 714 405 8 57 6 Orissa 523 397 9 76 3 Uttar Pradesh 765 389 10 51 9 Rajasthan 670 341 ~ 11 51 9 Assam 841 ~12 12 - 1 10 Source: Computed from, India (Republic), Directorate of Industrial Statistics, Report on the Census of Manufactures and the Annual Survey_of Industries, Note: Bombay and West Bengal have the same percentage ranks. Similarly Uttar Pradesh and Rajasthan also have the same percentage ranks. Ranking is done numerically with #1 denoting the highest rank, in col. 3 & 5. It is necessary to consider both the absolute and percentage changes in analyzing the rankings. This is so because some industries and states which show l< absolute increases reveal higher rates of growth in money wages. Among the t0p five industries, Sewing Machine ranks first in the whole group of 26 industries having the largest absolute increase and Starch ranks first showing the highest percentage increase. 0n the other hand, Sewing Machine is placed third in percentage ins-tease Cow/m.) -26- increase. There is no change in the ranking of Electric Fans either in absolute or percentage terms. Thus, the first three industries show a great similarity in their rates of growth of money wages. Three industries, Chemicals, Biscuit Making and Sugar show a higher relative growth in money wages than Soap which comes within the top five. Likewise, Chemicals, Paints and Varnishes, Biscuit Making, Sugar, Cement, Tanning and Glass show a higher rate of growth in percentage terms than Iron and Steel which finds a place among the top five in absolute increase in money wages. Among the last five industries, Ceramics has shown a decline in wages. It is the only industry among the twenty-six industries showing a decline. There have been changes in the ranks of the other four in this group with respect to percentage increases. Fruit and_Vegetab1e Canning occupies the lowest rank among all industries showing an increase in wages. Rice Milling has shown a higher percentage increase than twelve other industries, even though it is in the low five group. It has shown a higher rate of growth than Cotton Textiles, Electric Lamps, Jute Textiles, Wheat Flour, and Distilleries and Breweries (same rank), Fruits and Vegetable Canning, Paper, Bicycles, Matches, Woollen Textiles, and Ceramics. Vegetable Oil shows a relatively greater percentage increase in wages than Bicyles, Fruits and Vegetable Canning, Matches, Woollen Textiles, and Ceramics. Plywood and Tea Chests -27- also exceed seven other industries as Wheat Flour, Fruits and Vege— table Canning, Distilleries and Breweries, Paper, Bicycles, Matches, and Woollen Textiles. With respect to the states, Bihar shows the largest increase in absolute, and Andhra the highest in percentage terms. Assam ranks last in both cases with actually a decline in money wages. The most interesting feature of the table is the identical ranks in the rates of growth in wages in the two states - Bombay and West Bengal - considered to be the leading industrial states. They have shown a lower percentage increase than Andhra, Bihar, Delhi, Madhya Pradesh, Punjab, and Orissa. (@vQ/L) )t I A major conclusion that emerges frim the above tables is that the ave‘are annual earnings of workers in manufacturing indus ries have increased for the xerioi 1950—1060. When the industr es and states are ranked according to the increase in their wages "\ we find that with the exception or one industry (ceramicS) and one state (Assaml average annual money wages have increased in all other cases. It is also important to bear in mind that the rather sharp in- crease fnr the terminal years 19am and 1963 should be interpreted in the light of the wider coverage of the Annual bhnuuruz)f industries. leomaxmrx be some up-- ward bias in the figures, the exact extent of which is difficult to determine. However if we focus on thw period up to 1958, both the industry series and We may now consider the factors that have in- (‘1 .iuouced the upward movement in the level of wages. General Economic ijansion and Increase in Tncomes -—-—A. . - _ -._ - ,— . ...—. -_—-—— _ --~_._.. -.-_‘- .;9 state series Show an increase in the overall index. Under the stimulus providei by development planning two :ericd 1950—1960 was marked by a considerable increase of investment.1 A study made by the United Nationa notes: "The rate of gross investment rose from an estimated 13.6 per cent of gross national product in the first plan to 18.5 per cent in the second plan. The rate of develop- mental outlays and private investment to national income came to 16 per cent in the first plan, and to 23 per cent in the second plan. At current prices, deve10p- mental expenditure increased from Rs 30 billion in the first plan to Rs 71 billion in the second plan. The outlay thus must have more than doubled even in real terms."2 If we are to rely on the above findings, we should expect to find an increase in national income with an increase in the various sectors in which it originates. lIndian statistics on investment do not provide de- tailed breakdowns of estimates of net investment by sectors. It would be highly desirable to have some estimates of net investment distribution with de- tailed breakdowns within the indifitriai sector so that we may trace the effect of increased invest- ment activity on the growth of incomes within this sector. But in the absence of such estimates, we have to rely on the breakdown of national in- come figures by industrial origin. 2 United Nations, Economic Survey_of Asia and the Far East, 1961, Bangkok, 1962, p. 83. '30 .ofigin. The following table shows the growth in national income by in- dustrial origin. The percentage of the total national income originat- ing in the various sectors is also given.‘ Table & National Income by Industrial Origin - All India 1951-52 to 1961-62 In Rs.Abja which means 100 crores, 10‘] (at current prices). a: Q) U I: U C O H E 3 ”1 g '0 ~45 a) E O a u H a ”-4 W N U (D Q) U D. W Ln 0 0 >~.-H H O a oAA >. N 00 G U Q 0 G [-1 z m “.4 a .4 C94 m .1: w c ~3 C 0-0 v- H -4 5 3 c~a u u s .4 'U .a u -H u m -4 0 w . «8 r4 s.x 3 w H s o c m m a o u u u.4 H “)6 w .cgq u-d m u Q) ‘H G O) a C O 0 gas a H G C: u 0 O H O >4 z ‘34 [II CD £2] E" U U V O m H O \5 H U U H 1 2 3 8 4 5 6 7 9 10 11 IN RUPEE§17ABJA: 100 CR0RES,,109. 1951-52 0.9 6.4 9.5 16.8 0.4 2.1 0.8 14.6 17.9 34.7 99.7 1952-53 0.9 6.4 9.7 17.0 0.4 2.0 0.7 14.7 17.8 34.8 98.2 1953-54 1.0 6.9 9.8 17.7 0.4 2.0 0.8 14.8 18.0 35.7 104.8 1954-55 0.9 7.5 9.6 18.0 0.4 2.2 0.8 14.7 18.1 36.1 96.1 1955-56 1.0 7.8 9.7 18.5 0.5 2.5 0.9 14.9 18.8 37.3 99.8 1956-57 1.2 9.0 9.8 20.0 0.5 2.8 1.1 15.2 19.6 39.6 113.1 1957-58 1.4 9.8 10.0 21.2 0.5 3.2 1.2 15.8 20.7 41.9 113.9 1958-59 1.4 10.0 10.3 21.7 0.6 3.3 1.3 16.3 21.5 43.2 126.0 1959-60 1.4 11.1 10.7 23.2 0.6 3.3 1.4 17.0 22.3 45.5 129.5 1960-61 1.6 13.2 11.2 26.0 0.6 3.6 1.6 17.6 23.4 49.4 141.4 1961-62 1.7 15.4. 11.7 28.8 0.7 3.8 1.9 18.4 24.8 53.6 148.0 .lflipfiRCENTAGE OF THE TOTAL NATIONAL INC . 1951-52 0.9 6.4 9.6 16.9 0.4 2.1 0.8 14.6 17.9 34.8 100 1952-53 0.9 6.5 9.9 17.3 0.4 2.0 0.7 15.0 18.1 35.4 100 1953-54 1.0 6.6 9.3 16.9 0.4 1.9 0.8 14.1 17.2 34.1 100 1954-55 0.9 7.8 10.0 18.7 0.4 2.3 0.8 15.3 18.8 37.5 100 1955-56 1.0 7.8 9.7 18.5 0.5 2.5 0.9 15.0 18.9 37.4 100 1956-57 1.1 7.9 8.7 17.7 0.4 2.5 1.0 13.4 17.3 35.0 100 1957-58 1.2 8.6 8.7 18.6 0.4 2.8 1.1 13.9 18.2 36.8 100 1958-59 1.1 7.9 . 8.2 17.2 0.5 2.6 1.0 13.0 17.1 34.3 100 1959-60 1.1 8.6 8.2 17.9 0.5 2.5 1.1 13.1 17.2 35.1 100 1960-61 1.1 9.4 7.0 18.4 0.4 2.6 1.1 12.5 16.6 35.0 100 1961-62 1.2 10.4 7.9 19.5 0.5 2.6 1.3 12.4 16.8 36.3 100 Source: P-T-O (Ow-or) -3l- Source Government of India, Cabinet Secretariat, Central Statistical Organization, Department of Statistics, Estimates of National Income 1948-49 to 1962-63, New Delhi, Feb. 1964, Table 2 and 2.1, pp. 2-3. We may first refer to column 10 which gives the grand total of columns 1 through 9 which seem to represent the modern wage earning sector. The percentage of the total national income originating in this sector shows a slow in- crease from 34.8 in 1951-52 to 36.3 in 1961-62. Column 9 shows a decline from 17.9 per cent in 1951-52 to 16.8 in 1961-62. But, there is an increase in the percentage of the total national income in the schxrcomprised of mining, fac- tory establishments and small enterprises, as revealed in column 8. Specifically, the percentage increase in factory establishments and mining is higher than that of small enter- prises. The increase in the national income originating from factory establishments gives us a clue as to the general economic expansion that has taken place in manufacturing industries. Under the impetus of the First and Second Five Year Plans, the net investment in manufacturing has increased, thereby leading to the expansion of existing firms and the birth of new firms in the case of existing industries and the establishment of new industries. This has had the_effect of raising the level of the average annual wages of the factory worker. A comparison of the rate of growth in per capita income with the rate of growth in the earnings of different categories of employees reveals that for the period 1950-51 to 1960-61, the percentage increase in theearnings of workers in mining and factories has exceeded that of other classes of workers as shown in the following table. T3231. 2 Per Capita Income and Index Numbers of Earnings by Category Year 1950-51 1951-52 1952-53 1953-5“ l9SN-55 19 5—56 \JW 19 6-57 \2 \JW 1957-58 1958-59 1959-60 1960-61 Source of Employment - All India,_1950:51 - 1960-61 Per Capiéa Income Faciory Miges Ra11- Cengral Rugal using 58-59 Employees ‘”“—‘ QEEE: gggfi;_ Skilled Prices Prices Worker A 52- 8 Index Numbers at current Prices(1950-51=100) 247 280 100 100 100 100 100 250 283 108 105 106 103 106 255 289 116 111 l09 lou 107 266 301 115 111 110 108 104 267 303 115 113 117 108 103 267 300 122 115 113 108 108 275 307 123 151 _116 108 109 267 297 128 168 122 113 114 280 312 130 179 125 116 117 279 310 134 192 128 120 123 293 322 143 199 137 126 ... Col. 1A, 2-6, Government of India, Planning Commission, Report of the Committee on Distribution of Income and Levels of_Living, Part I, Distribution of Income and Wealth and Concentration of Economic Power, New Delhi, 1964, pp. 59,80. Col. 18, U.N. Economic Suryey of Asia and Far East 1961, Bangkok, 1962, p. 82. C01. 5 includes railways. (Over) .33. NE. The earnings of factory employees reported above is not strictly comparable to the data computed in this study relating to workers in manufacturing, the source used in this table being different. The table shows that per capita income has in- creased by 15 per cent between 1950-51 and 1960-61. Earnings in mining and factories have shown an in- crease of 99 and 43 per cent respectively. Concentrating our attention on the manufacturing sector, it can be argued that, with the growth in income, there has been an increase in demand for goods produced inthis sector. In the absence of data it is difficult to ascertain, to what extent there has been an increase in the income elasticity of demand for manufactured goods} (gar , if we distinguish the period 1950—1960 as one of general economic eXpansion as against the period prior to 1950 as one of catching up with war time backlogs, It is reasonable to suppose that this income elasticity of demand ,JLaA increased considerably. In the face of this in- crease)-—»M ~— 9, industries have enjoyed relatively more favorable product market conditions compared to the previous period. This factor has favorably in- fluenced the profit position of industries enabling them to grant increases in wages. ' - 1Two types of elasticities have been stressed. These are, elasticity of demand for goods and services as a function of personal disposable income, and the more general elasticity with reSpect to changes in national income. See, Subbiah Kannappan, "Wage Policy in Economic Development-Some Critical Issues", Economic Weekly, Anhlan1 Klsumhnn Ira-I vv‘r M..- c: L '7 may. 1n£h -- “J -34- The Role of_Government Policy The government appointed a committee on profit sharing in May, 1948, to devise a system in which labor's share of profit could be determined. The 'committee observed that labor's share should be 50% of the surplus profits after allowance for depreciation, reserves, and fair return on capital employed.1 Thus in principle, government policy has recognized the idea of profit sharing in industry. If it can be shown that there exists a surplus of profit after'tliowing heWAU£3 for depreciation and/Jim“; labor could legitimately claim a share of this surplus. In practice, the claim has been related to tfiépayment of bonus which is now a component of the total wages received by the worker. The question whether the payment of bonus should be linked with productivity or profits has recently been re- viewed by the Bonus Commission which submitted its report in 1964. The Commission favoring a profit sharing bonus, points out: "In view of the objections to thaproposal by large sections of employers as well as by almost all the unions, and the practical difficulties inherent in any such proposal, we are unable to recommend that the concept of bonus based on profits should be re- placed by an annual bonus linked with pro- duction or productivity. It is doubtless true that properly devised incentive systems 1The reference to the committee on profit sharing has been taken from the following report. Government of India, Ministry of Labour and Employment, Report of thewfionus Commission, New Delhi, 1964, pp. 6—7. -35- in manufacturing concerns form a useful part of the wage structure and would help to increase production, but they cannot be suggested as a substitute to replace the annual profit sharing bonus." . The wage data in our study includes bonus, and hence we can establish a relation between profit and wages. The payment of bonus has become an issue before the courts and under pressure of judicial awards, many employers have been forced to adopt a permissive attitude towards wage increases. The findings made in this study suggest that the newer industries have shown a higher relative increase in wages compared to the older established industries. This can be seen by comparing industries like Sewing Machines, Electric Fans, and Chemicals, with Cotton and Jute textiles. It is likely that profits in these industries have increased owing to the growing and unfulfilled demand during the period of our study and that these industries have found it relatively 2 easy to pay higher wages- Profitability in this way 1The reference to the committee on profit sharing has been taken from the following report. Government of India, Ministry of Labour and Employment, Report of the Bonus Commission, New Delhi, 1964, p. 27 2The bicycle industry seems to be an exception in so far as it is one of the newer industries. It may be that it has eXpanded in the direction of providing more employment than in offering higher wages. -36- can definitely be related to wages. In the United States Slichter found that "wages within a considerable range, reflect managerial discretion, that where manage- ments can easily pay high wages they tend to do so, and that where managements are breaking even, they tend to keep wages down."1 In India, managements may be willing to pay higher wages out of increased profits to acquire more efficient and experienced workers, to reduce absenteeism2 and to maintain better discipline within the factory. - ( all Q/FL’ 7 1 S. H. Slichter, "Notes on the Structure of Wages," Review of Economics and Statistics, Vol. XXXII, Feb. 1950, p.88. 2Absenteeism rates are high in Indian industries. For an exhaustive and historical reference on absenteeism see, Subbiah Kannappan, Labor Force Commitment in Early Stages of Industrialization, unpublished manuscript, Department of Economics _and School of Labor and Industrial Relations, Michigan State University, East Lansing, Appendix IV, pp. 58-61. The establishment of wage tnards in different industries is another feature of givernment wage policy that must be considered as influencing the increase in the level of wages. So far, Givvph industries, namely, cotton textiles, sugar, cement, jlte, tea, coffee, and rubber planatations, iron and steel, coal mining, 1111.10 3 iron ore mining, lime stone and dolom te’have been covered by wage boards. The cotton textile and sugar industries were covered in 1957, cement in 1958; jute industry and tea planatations in 1960; coffee and rubber planatations 1n IQFI; iron and steel and coal mining in 1962; iron ore and limestone and dolomite mines in 1963.1 The following table shows the range of percentage increase over existing wages in three industries as a result of wage board recommendations. The information in this paragraph has been obtained from the following source: R. K. Malviya, "Reflections on Wage Board", Indian Labour Journal, Vol. v, No. a, :‘eptflgmf'fi‘.’ 728. Thefiafithor, when writing this was the Deputy Minister, for Labour and Employment in the central government. -xg. Tablefifi 3 Range of Increase over Existing Wages and Number of Worgers Benefited byfRecommendations of wage Boards l 2 3 4 5 Industry Date covered Ran e of Percentage Percentage of bngage Board wage increase increase over workers bene- existing_wages fited to total No. of Workers 1. Cotton Textiles 1957 Rs. 8 to Rs. 10 8.6 to 21 99.8 por mensem 22. Sugar . 1957 Rs. 21 to Rs. 47 38 to 117 96.2 per mensem 3. Cement 1958 Rs. 5. 84 to Rs. 7 to 100 100.0 45. 50 per mensem Shaurce Compiled from R. K. Malviya, "Reflections on Wage Boards", Igdian Laboug,Journa1, Vol. V, No. 9, Sept. 1964, pp. 728-30. The table shows that the range of increase over existing ]’ wages is very large in cement and sugar compared to cotton textiles. It is also seen that the percentage of workers benefited is quite high in all the three industries. We may now consider the recommendation of the wage board for the jute industry as it is another major industry which provides employment to a large number of workers. The board was appointed in 1960. At the time when it investigated into the wages of workers, the basic wage of the lowest category of workers was Rs.34. 67 and with a dearness allowance of Rs.32. 50, the total came to Rs.70. 02 per month. The board recommended a basic wage of Rs 40. 17, a dearness 1 allowance of Rs 32. 50 and an increment of Rs 8. 33. 1Government of India, Report of the Central Wage Board _fgz_lnte_1ndustry_1963, Manager of Publications, Delhi, 1964, p. 40, 68. (K! ~39. The board also gave similar recommendations for other categories of workers. Thus, the role of government policy in raising wages can be traced to these developments. These have secured significant wage premiums for the industrial worker compared to other wage earners in urban services and those in rural occupations. Trade Unionism and the Level of Wages In the west, Opinion is divided on the question of union influence.on wages and there are those who stress the importance of market forces, and those that point out that unions have significantly raised wages. The usual technique of the latter school of thought is to compare union and nonunion wages in a particular setting. If wages are higher in the union setting, it might signify that unionism was responsible for the increase in the wage of the average unionized worker.l Thus power force advocatesstress the strength of unions in raising wages, while market force supporters minimize its role, if they find, that the union-nonunion wage differential could be explained on economic grounds. In India, there has not been any attempt to study the effect of unionism by comparing union and nonunion wages in a particular setting. Dr. A. J. Fonseca, lRichard Perlman(Ed), Wage Determination-Market or Power Forces?, D. C. Heath and Co.,—Bbston, 1964, p- 77. -40- however, attempted to relate indices of the degree of unionization, (Percentage of the total industrially employed working forCe who are members of trade unions) with cost of living, money earnings, real earnings, . and employment, production, productivity.1 His work does not answer questions regarding union-nonunion wage differentials. Nevertheless, his findings have pro- voked further thought in this neglected area of en- quiry. Dr. Fonseca points out, "A l per cent increase in the degree of unioni- zation will lead to a rise of .18 per cent in the wage rate. Similarly, an increase of 1 per cent in the cost of living will bring about a rise in wages of .91 per cent. And lastly, an increase of productivity by l per cent will cause a rise in wages of 2. 09 per cent." In conclusion he says, "It is clear that there is a very close correlation between the cost of living and money wages, while that between money wages and the degree of unionization is considerably lower.3 In a following article the same writer cites the use of the well known Cobb-Douglas production function for assessing the share of labor in the domestic product and argues a case for enlarging the scope of the variables A. J. Fonseca, Wage Determinatign_gnd Organized Labor in India, Oxford University Press, 1964, pp- 194195. 21bid. p. 199. 31bid. -44l- included in the function.1 He says, ”One of tune variables would have to represent union effectiveness in promoting economic growth by raising the total output. For this purpose, a further set of variables to represent increase in union membership, its financial stability, its age, its representativeness in the industry, and similar aspects would have to be chosen."2 He stresses that the function of sound trade unionism "would be to improve the quality of labour and thereby contribute handsomely to an increase in production."3 He then ties this argument up by asserting that in the organized sector of Indian industry, where trade i unions have their largest following, great strides in production have been made.“ Dr. Fonseca places considerable faith in the power of unions to raise the level of wages. But it is diffi- cult to agree completely with his view, especially when his own calculations lead him to conclude that the correlation between money wages and the degree of unionization is considerably lower than that between money wages and the cost of living. In evaluating the role of unions in India the characteristics of Indian 1A. J. Fonsoca, "Contribution of Trade Unions to Development", Indian Journal of Industrial Re- lations, Vol. I: No. 2, Oct. 1965, p. 179180. 2 Ibid. p. 180. 31b1d. “Ibid. p. 181. He also says that this is due to the efforts of both labour and capital and it would be difficult to argue that the impact of trade unions has been entirely negative. -42,. unions are not fully grasped. Reminding that this factor is important, Dr. S. Kannappan points out, "The typical union is small, and lacks stable leadership. The labour movement as a whole is ridden by political factions. "Rational" union behavior, in the context of this acute rival unionism need not be the same as the goal of a "rational" militant leader maximizing any single or complex of objectives pertinent to the union or its members. It most certainly does not follow that a carefully calculated attitude to wage, employment and productivity relation- ships will be dominant." The above passage clearly shows that one cannot place too much emphasis on union power to raise wages. “Lao It isnpointed out that environmental conditions and government policy have discouraged the development of strong, Job—centered unionism and co-equal working relationships between worker and management.2 ‘The point regarding the size of the unions is particularly relevant because of the uncertain nature of the data on union membership. The former Deputy Chief Labor Commissioner to the government of India, 8. B. Kale, who was concerned with the development and execution of verification procedure of trade union membership has furnished verified figures for 1952-53 l S. Kannappan, "Union Adjudication and Wages", mpg Economic Weekly, Annual Number, Feb. 1962, pp- 2 - - 2 . Van Dusen Kennedy, "The Role of the Union in the Plant in India", Industrial Relations Research Assaciaiicn.222§§2%%25§. Eight Annual Meeting. New York, Dec. - , 955, Publication 16, 1956, p. 263. -45.. to 1962-63.1 For instance, membership claimed by the Indian National Trade Union Congress in 1952-53 was 1,3u7,32o where... verified figures are 919,258. In 1962-63, claimed membership was 1,828,783 where as when verified it stood at 1,268,339. Similarly, he provdes figures for the other major unions. But even he has not been able to verify the figures for all the years. This was either because, the list of members was not submitted, or venfication was not done at the request of the union. It follows that we cannot place too much reliance on the index of unionization, which is likely to overstate the degree of unionization. (gm/w“) W“. 1S. B. Kale, "Verification of Membership of Trade Unions", Indian Journal of Industrial Relations, Vol. 2, No. 1, July 1966, pp. 6HL65. -44- II. Change in Real Wages net it w Mwe will examine the change in Real Wages for all India. A comparison of Real Wage indexes computed from the Industry Series and the State Series will be made. However, it should be noted that real wage estimates for all India are at best a rough approximation as there are wide variations in the comsumption habits of workers in different regions. At present these estimates are made by deflating money wages by the consumer price index applicable to factory workers. Price relatives of fixed baskets of goods and services said to be con- sumed by workers are weighted according to consumption eXpenditure on different items to yield a set of. working class consumer price index numbers. However, the existing series of index numbers are far from satisfactory. The Labour Bureau which publishes them says, i "The index does not measure the absolute level of prices but only the average percentage change in the prices of fixed basket of goods and services at different periods of time. The index does not also measure changes in the total expenditure of the pOpuiation-group or, in other words, their actual cost of living." Government of India, Labour Bureau, A_Guide to Consumer Price Index Numbers, 1960, Simla, p. A. - 45- Besides the above drawback the index numbers available at present r~~-- > are based on family budget surveys carried out during World War II. The Labour Bureau itself admits that these surveys carried out 20 to 25 years ago are not a true reflection of the present spending habits ofworkers.l A more detailed examination of real wages would be undertaken in the chapter on inter-state wage differences. This is more likely to give a clearer indication of the extent of real wage change in each state. The most exhaustive study on real wages has been undertaken by Dr. S. A. Palekar whose main findings may be flrgt considered. — :2 He points out that for the period 1939-1950 an examination of the all-India real wage series revealed that by 1950, the workers had Just managed to regain their base year level of real wages. Real wages attained a peak in 1940 and declined -~ 1 Ibid. Introductory Note. In 1958 fresh family budget surveys were made on the basis of which new working class consumer price index numbers are now under preparation. Dr. A. J. Fonseca citing a government report on_readJusting the index numbers for industrial workers in Bombay writes, "In the latest survey of family budgets undertaken by the Labour Department of the Govern- ment of Maharashtra, in the year 1958-59, the number of items for which information was collected was 119, while in 1932-33 when the previous family budgets were collected the number of items was only “7." See A. J. Fonseca, "Contribution of Trade Unions to Development", Indian Journal of Industrial Relations, Vol. 1, No. 2, Oct. 1965, p. 183. -.46- thereafter, to register a low level in 19h3. During the war there were real wage losses in all the States except the PunJab. Real wage losses were sustained in all industries except Gins and Presses. He, g however, finds that the post war period brought with it "some additions to real wages" of workers in all the States except the Punjab which enabled the workers in most States to regain their 1939 wage position.1 Philip Kotleg presenting data on a Statewise series of real wages shows that real earnings rose by 21 per cent between 1950 and 1954. Calculating on base 1950 he says that the rise was greatest between 1950-52 whereas the period 1952-5A was marked by comparative stability.2 The index of real wages computed in this study cannot be strictly compared to the previous studies owing to basic differences in data and coverage. The studies prev busly made show that, after the war, real wage gains have been made by the workers. ‘That this trend . seems to have continued since 1950 {3 shown by the indexes computed below. 1 S. A. Palekar, Real gages in India, International Book House Ltd.-Bombay,‘l962, p.‘I75-76. 2Philip Kotler, "Problems of Industrial wage Policy in India", unpublished Ph.D, thesis, Massachusetts Institute of Technology, Cambridge, 1956. p. 336. 349. 1950 1951 1952 1953 195“ 1955 1956 1957 1958 1959 1960 .uq- gar ce Computed fru, India (Republic), Directorate of EQIEEEIZ Industrial tatiatice, rt th Mature” and the £1931 Ev g of Indultggg. 1nd“ L‘mur Journg’ Vol. II, “Ge 12. Me 1%1. pp. 12”.” Goverment of India, Central. Statistical Organi- sation, ati ct f th I Ch m. 1957-5g. p. 525 and 1&1. p. 5&1 36 F AI = 351...." ‘ '3 2‘ E 1:: 3 whu'e AI is the Real wage Index for all India I is the Real Wage Index for industry J 83 is the annual uployment for industry 3 1:21.21 nd 0 R W e nduetrie and 8t te .. Base 1950 I 100 ndu r e m 100 100 105 106 113 11“ 112 115 117 121 120 128 . 118 122 11“ 120 111 118 113 119 12!: 13" A8. The all India working class canal-er price index numbers. reduced to base 1950 were used to deflate the index of money wages for the industries. Colts) (.2 S I} E' AI .. J" 4 t: _- E :4. 3.0 whee A! is the leal Wage Index for all India I is the Real Wage Index for tate 3 8:11 is the annual .ployment for tats J For each tate, the oonmamer price index fibre relating to the centers we combined using the 1951 Population of the center as weights to arrive at a weighted index forthe date as a whole. This weighted index was used to deflate the index of money wages for that state. In the case of the industry series there is drOp by one point in 1953, by six points in 1956, by a further drop of four points in 1957 followed by a decline of three points for 1958. In the case of the states series the index declines by six points in 1956 after a steady rise up to 1955. For 1957 and 1958 there seems to be a slight decline by two points in each case. The ranking by real wages by industries will follow the same ranks as in the case of the index of money wages given previously. It may be recalled that the real wages are derived in each case by deflating the money wage by the consumer price index. But in the case of the states they would differ according as the price index differs from state to state. We may present these as follows. Bihar showed the highest increase for the whole period (1951-1960) with an in- crease of 91 points, followed by Andhra 63 points, Punjab 55 points, Orissa 5h points, Madhya Pradesh U9 points, Rajasthan ”7 points, Uttar Pradesh 39 points, West Bengal 36 points, Delhi 29 points, Madras 18 points, and Bombay 14 points. There was a decline in the case of Assam by 6 points for the whole period. No doubt there were fluctuations in each case which may be explained on the basis of the fluctuations in the working class consumer price indexes. As mentioned above, it would be more apprOpriate to consider the changes in each state in a subsequent chapter dealing with inter-state differences in wages. The main concluSion to be drawn from our real wage indexes is that there has been an increase in real wages if we take the whole period into account. The figure for-money wages includes besides the basic wage, cash allowances including dearness allowance, bonus,- and the money value of various benefits received by the workers. In recent years the government has made various attempts to link the payment of dearness allowance to the consumer price index in order to safeguard the real wage position of the worker. Three main agencies can be distinguished which lay down the principles governing the linking of the dearness allowance with the consumer price index. These are Adjudicators and Tribunals, Wage Boards and the state governments acting independently under the provisions of the Minimum Wages Act of l9u8. There are variations in the manner in which the linking is done, but the fact is, the_factory worker is partially compensated for price increases of the essential commodities he consumes. Further it can be said that such compensation has improved his real wage position. The dearness allowance forms a substantial part of total wages and it has risen in recent years. In 1957, out of an average annual per capita earning of Rs 1,216 (in the case of ,5]- factory employees earning below Rs 200 per month) in ~9 thirteen states Rs 668 formed basic wages, Rs 896 cash allowance including dearness allow- ance, Rs 21 money value of concessions, Rs 70 bonus and Rs I arrears. Thus, the other components besides basic wage formed Rs 591 which is roughly half of the total. Among the other "components"A%::: allowance including dearness allowance formed more than 82 per cent. During 1959 in the case of twelve states the figure on per capita earning turned out to be Rs 1329 of which basic wage formed Rs 727, cash allowance in- cluding dearness allowance Rs 5&0, bonus Rs A9, money value of concession Rs 6 and arrears Rs 5. Thus the "other components" besides basic wage amounted to Rs 600 Sn union cash allowances including dear- ness allowance came to 90 per cent.1 The figures for both the years were taken from The Indign Labour Year Book 1 8, p. 65 and The Tidian Labour ea; Book 1965, p. “2, published By the Labour‘Bureau. -52,. III. Chggge in Employment and its Effect on Wages The Ministry of Labour and Employment in a recent 1 points study on elployment and unemployment in India out that there are huge surpluses of man power in certain occupations while at the same time there are acute shortages in other occupations. The study also takes the stand that unemployment has increased in urban areas though it does not deal with this problem thaihlhldmgIHMNmmmelilIs , _ . in any detail. “ But.” ”much is known about the extent of unemloymentde eta-s- emen use in“ aeeeage I w “. "In the absence of unemployment surveys conducted at regular intervals, there is as yet no reliable information as to whether unemployment increased or decreased during the Second Plan period”.2 Analysing the position with respect to factories,H!nd , com- puting indexes of employment in factories for the period 1950-195 .:§§§: "The p1an(he refers to the First Five Year Plan) does not seem to have stimulated factory employment. On the other hand, employment in factories declined continuously almost during the first four years of the Plan. It was only in 1955 that the volume of factory employment was restored to its 1950 position".3 1 Government of India, Ministry of Labour and Em- ployment Service (D.G.R.E.), lo nt and Un- ggplcxaggt Study No. 8, March ¥959, p. 15. 2mm. p. 16. 3 . A- Palekar. W W, Asia ublishing cuse, Bombay, 2, p. . . -153- The above observations do not refer to a detailed breakdown of either industries or states. Though Palekar has presented indexes of employment for nine 1 he has not undertaken states and twenty-five industries any detailed examination of the trend of employment in each of them. *wo will be mainly concerned with this type of analysis. Also, we will analyse the changes in employment and its effect on wages. Changes in em- 'ployment, whether in absolute or relative terms, cannot be meaningfully ~discussed merely by a reference to its increase or decline. What is more important is an analysis of the differential rates of growth in employ- ment whether in absolute or relative terms with respect to particular industries or states. Why have certain industries shown greater increase in employment than others? We shall deal with the overall trend in employment as well as the changes in employment in in- dividual industries ‘hnd states for thezn“l “period. This will be followed by an attempt to ascertain the effect of changing employment on the overall wage structure. Before examining the data computed in our study, we may refer to the employment statistics furnished by The inle'as are based on figures reported under the Payment Wages Act of 1936 which relate to employees earning less than Rs 200 per month and also covers same establishments besides factories. -54.- Z. the Labour Bureau which is given below. tabla}. Index orggggggygent — All Ingi§_l951-6O (1950-100) Yr. 59452435511 55 i6 i] 5859 60 :Emp.! 30 32- 31 31 31 32 3" 35 36 36 39 Indx. 100 10k 103 102 103 106 111 115 116 118 122 'In Lakhs is. 100,000. Employment figures have been reported with decimals not shown here. . goggoe Indig? Labogg Journal. Vol. V, No. 8, Aug. 196.. The data relates c on y ac cry emp oyment reported under the Factory Act of l9u8. ‘The table shows that employment has declined between 1952 and 195! and theregftep shown an increase. For the.,&?,: period there is an increase of 22 points in the index. The increase is not significant up to 195“ and we may say that during the First Five Year Plan period (1951-55) there has been a slow change in employment. This probably is in accord with the finding of Palekar quoted above. we may now consider the change in employment in the case of manufacturing industries considered in this study. The overall trend in employment is shown below. - ...-.__ l.‘ ° . . .° Year '3 ustries .- ‘ Index ‘ 8 States Index finer m1 - 0 £7". LLO em .2221 - _ 1950 l,320,l05 100 l,hh7,289 , 100 :3 1951 1,318,123 101 1,~02,187 100 1952' 1,301,586 103 l,t61,568 101 1953 19% 1955 1956 1957 1958 1959 1960 -55- 1.322.250 102 1.166.185 1,376,650 106 _ l,b51,5h7 1,518,836 110 1.u96.326 1.180.139 117 1.601.767 1,u77'.226 117 1,598,111» 1,399,266 112 1,517,616 1.370.512 122 1.3119.583 1,359,863 124 1.336.503 m Computed ma, India (Republic), Directorate of Industrial Statistics, 3mg; 9:) the gag. g; ”flare! and th, mg mg 9: mafia” ' The table shows that there has been a slow growth in uployment. The index of uployment up to 1954 in both cases has hardly increased at all. There is a drop in the case of industries in 1958 and in the case of the states in 1957. an, as mentioned earlier, very little can be ascertained from the trend in the overall figures. We may consider the ranking of the individual industries for the entire period showing the change in msployment both in absolute and percentage terms, In the following table, we have ranked the industries according to the highed. ablomite increase in uployment. The percmtage change and rank according to oer cent increase is also shown for each industry. 100 101 10“ 115 11:» ‘108 108 105 ‘2‘ J 3! Industry_Rankin .56- Table _ Industry Cotton Textile Chemicals Iron and Steel Glass Bicycles Cement woollen Textiles Matches Plywood Electric Fans i,fr‘anning Sewing Machines Sugar Soap Wheat Flour Electric Lamps Paints & Varnishes Starch Distillaries Fruits & Veg. Can. Ceramics Biscuit Making Source: Cemputed from India (republic), Directorate of Industrial and the 55335; Statistics, e t on 53:13: 02 Inggetgieg. Urn. 1960 employment figure is adjusted by multiplying the reported data by 1.19.(see appendix). 2 £232 p. 602,731 26,u11 62,102 20,315 2,178 1N,388 13,851 11,800 2,998 9,929 7,5u8 1,780 107,895 5,310 0,521 1,310 3.739 1,211 0,112 1,516 16,3u5 5,610 IC‘ e in Employment Change in Eng? by Chan 1950-156%) 2 3 (1950—605 2 e Ce 8 h2,5h6 30,765 25,601 17,213 13,882 12,1u7 7,“55 5,659 u,863 0,3u0 3,696 2,698 2,689 2,u95 2,322 1,898 1,652 1,105 508 230 -USS -l,055 of Manufa N 82.21 11' \DCDNONUW 10 11 12 13 in 15 16 17 18 19 2O 21 22 5 1 Chan e T—‘ggsoréfi '2fif‘4lx’ 116 MG' 85 560 BA 50 50 162 88 “9 152 u? 51 1u2 NH 80 12 15 —;3 -1; ) 6 Rank 00' 16 5 16 7 l 8 -5.- P090? 19 0,76 —1o,050 23 -52 25 Rice aiiiihs 00,001 —13,860 20 -30 23 Vegetable Gil 06,950 ~22,l39 25 -08 20 Jute "8333... Computed frozolzgga (nepuSTZd§3SDirseto;§te of Inddstrial 21 IInastLaI_ihe.£snana_2£_lensilss:s_ Statistics, s and the Annual 33:!!! we shall also rank the states similarly. QIJE'QBQEZI-l' Table )3. gtate Rankin b Chan e in Em loyment {1955—1960: 1 - g 2 §gggg, Eggg Change in Egg? 1 Chan e Rank 4%E§§§:x 100 Uttar Pradesh 195,909 18,636 1 7 Madras 126,297 16,671 2 13 7 Hadhya Pradesh 05.068 16,012 3 36 3 Punjab 20,050 9,060 0 38 2 Rajaethan 8,713 8,115 5 93 1 Andhra 30,591 6,735 6 19 6 Orissa 9.203 2,033 7 22 5 Assam 5,285 1,688 8 32 0 Delhi 21,629 -6,027 9 -28 11 Bihar 87.758 -1c,039 10 -13 8 ‘w. Bengal 007,281 -97,100 11 -22 9 Bombay N90, 701 ~133, 339 12 ~27 10 ' source: Computed Rom India (Republic), Directorate of Industrial Statistics, . on us o and.the gngggl ‘The 1960 uploymed. figure is adjusted by multiplying the reported data by 1.19 (see appendix). .58' Paper 19,176 -lo,h59 23 -52 25 Rice silling “0,991 -13,86h 2h -3“ 23 Vegetable 011 16 ,950 -22 ,139 25 -u8 21 Statiatiea. a and the Annual 83:!!! He shall aloe rank the states similarly. W“ Jute “égégee: OOIDflfitd troa°123§n (Repu5816735hirectcgfite of Infiaetrial 21 WW Table 1} , ’State Uttar Pradesh 115,909 18,636 1 Madras 126,297 16,671 2 13 ladhwa Pradeab l5.868 16,912 3 36 Punjab 28,054 9,060 h 38 RaJaethan 8,713 8,115 5 93 Andhra 38.591 6,735 6 19 Oriana 9.203 2,033 7 22 Aaeam 5,285 1,688 8 32 Delhi 21,629 -6,027 9 -28 Bihar 87.758 —10,039 10 —11 ‘w. Bengal 917,281 -97,100 11 -22 Bombay N90, 701 -133, 339 12 ~27 ' um: Counted from India (Republic), Directorate of Induetrial Statistica, enaue c and the 533351. ' 2 1350 Chan§e in Emp! Rm 1 Chan e Rank ..21. ’ x 100 §§ate Rankin 0 Chan e in Em loggent [I§§6-1§§6)"””J' 3 “the 1%0 employ-ant figure ie adjusted by multiplying the reported data by 1.19 (eee appendix). 6 ummwmwsasz ...; \OCDH 10 N» a a. Among the tOp five industries cotton textile ranks first in absolute increase and bicycles in per- centage increase. The first five industries to have shown the highest relative rates of growth in employ— ment are Bicycles, Plywood, Sewing Machines, Electric Lamps, and Chemicals. Of these Bicycles and Chemicals are a.ao ranked within the first five in absolute increase in employment.Ian€€8n Textile and jute tertile together employed ~.~ >, 67.65 per cent of the tctal employment among tt twenty-six industries. in 1960 this went up to 70. £5 per cent. Cotton Textile has been exceeded in relative terms by 15 (ther indus‘ries, namely, the above tOp five in per- centage te $8 plus Flectric Fans, Cement, Starch, woollen ’ xtile, Wheat Flour, Matches, Tanning, Soap, Paints .nd Varnishes and Iron & Steel. This shows that .3 in the case of money wages, Cotton Textiles h'A lagged behindSZmployment growth also. Jute Tex- tile presents an even more drastic picture. Employ- ment has actually declined in this industry for the «hitire A {(PiOd. It ranks 21, with a 13 per cent decline. Thus, it is clear that the two major industries have lost ground to the other industries with respect to employs at growth. WiTh respect to the States the most striking featur of the table is the decline in employment in West T---3.n..fal and Bombay once again the two leading in- I do: :ial states in the country. Actually four out of t‘- vv~1ve states have shown a decline in employment. -59- Thus, there has clearly been a diversification of employment in that the rate of growth in employment has been greater in the case of the newer industries in relation to the older and established industries. This is seen by comparing the percentage rate of growth between Bicycles, Plywood, Sewing Machines, Electric Lamps, and Chemicals on the one hand, with Cotton and Jute textiles on the other. Even though in absolute terms Cotton and Jute textiles employ a larger number of workers.their employment has not in- creased compared to some of the smaller industries. The picture with respect to the states is even more convincing in this respect. Bombay and West Bengal, the leading industrial states, show a decline in employment. Rajasthan shows the highest rate of growth in employment followed by Punjab, Madhya Pradesh, Assam, Orissa and Andhra. We have to emphasize the importance of the gen- eral economic eXpansion and the increase in incomes that mark this period. This has further taken the form of a diversified form of growth in wages and em- ployment resulting in differential rates of growth between industries and states. For instance, Bombay and West Bengal, the leading industrial states, which ranked low in wage increases have also conceded their position with respect to employment growth. We will -60- whi le analyse these differences in greater detail A dealing with the inter-industry and the inter-state wage structure. The effect of changes in employment on the overall level of wages will now be considered. It may be re- called that our indexes of money and real wages are weighted by current year employment. If workers move from low wage to high wage industries, the effect of such a movement can be shcwn by weighting the indexes by current year employment. If movement had taken place i.e. if there had been a change in the composition of employment in favor of the high wage industries the increase in the overall index may be attributed to the increased weighting of the high wage industries.l r.) Clarence Long, using different employment weights on earnings indexes for manufacturing industries in the United States suggests that about one—fifth of the rise in average annual earnings between 1860 and 1890, was due to the shift of employment from low wage to high wage industries. The following data is given by him. Census: Annual earnings 17 industries, weighted by industrial composition of W - Ihplqnumt hm: ' 1860 1870 1880 1890 Current year. 100 131 117 1&9 1860 100 128 110 139 Effect of changing employment --— + 3 .+ 7 +10 By using daily wages he gets negative values of -11 for 1870, -12 for both 1880 and 1890 for 13 industries in the Aldrich Report and negative values of -2 and -3 for 1870 and 1880 respectively for 18 industries drawn from Weeks Report. Clarence Long, Wages and Earnings in the United States 1860-1829, National Bureau of Economic Research, Princeton University Press, 1960, pp. 77-78. -61- To find out whether the torment of workers In. low wage to high wage industries. had raised the overall level of wages, the indexes of money and real wages for the twenty-six industries were weighted by the composition of employment in 1950 and compared to the indexes based on current uploynent weight already computed in'this study. The renlts of the comparison are shown below. Table 13 11 fact of -- ' Qirrsnt Indexe 1 -n1- andRe Us i ' I...” ma .l'm- on th - .- acmm Wflfifififlfifiifi Average annual acney wages. % industries, weighted by industrial composition of ,asploynent in: Current year 100 1950 100 Effect of changing (splay-ant -- Avsrage annual real wages, a indutries, weighted by industrial cclpcaitien of .plcynsnt in 8 Wrent year 100 1950 100 Effect of changing esplcyaent -- 1'10 110 105 105 0 115 115 113 113 O 119 118 +1 112 111 *1 119 119 11.7 117 119 120 -1 12“ 125 -1 120 12“ 118 118 O 1% 127 -1 113 115 0 129 129 111 111 0 The effect of changing employment both for money and real wages 135 13“ *1 113 112 *1 is negligible as can be seen from the slight differences for the years in which they show themselves. This would ugly that the overall wage increases dawn ‘ 152 152 1% 123 *1 -62~ earlier has not been influenced by a movement of workers ‘Q from low wage to high wage industries. We may now take the state series of money and real wages and subject it to the same treatment as above. In this case the crucial difference is that we do not refer to industries on an all India basis but to the industries in each state and consider the movement of workers from low wage to high wage states. The following table shows the comparison as before. Iable |j_ Effect of Fixed Versus Current Employment Weights on the Tndexes of Money and heal Wages, 1950-7195?) (State‘) 1950 51 52 53 55 55 2:6 57 58 Average annual money wages, Vigstates, weighted by in- br.tria1 composition of employment in: Current year 100 111 116 120 121 120 126 131 13“ 1950 100 111’ 116 120 121 120 125 131 13h Effect of changing employment - 0 0 O O 0 + 1 O 0 Average annual real wages, 12 states, weighted by in- dustrial composition of employment in: 128 122 120 118 128 122 120 117 Current year 100 lO6~ 11“ 115 l 1950 100 106 115 115 1 Effect of changing employment - O - 1 O O O O O + 1 HF4 T‘JM For the years 1956, 1959, and 1960 there is a slight increase in money wages as the effect of changing employ- ment. There are no negative signs for any of the years. 59 138 137 119 117 + 2 60 134 132 f‘ {7' \A’ h») ‘ ' ‘4‘.) .T; “‘J wag, e e e a s e s o e e Bicycles biscuit Cement CEEAMICS Chemica 5 -63- For real wages there is a slight decline in the difference in 1952. The last two years shows an increase, but this is not very significant. The position with respect to the statesis not very different from that of the in- dustries. Overall wage increases has not been influenced by a movement of workers from low wage to high wage states. Earlier it was shown that Ceramics and Sewing Machines were the lowest and the highest wage industries for the period 1950-1960, when the absolute change in money wages is considered. In percentage change, Ceramics and Starch were the lowest and the highest wage industries respectively. In order to see whether there had been a change in employment in Ceramics, Sewing Machines and Starch, employment in all the 26 industries were ranked for each of the years from 1950 to 1960. The following table shows the ranking of the three industries among the other 26 industries. Table 1* ’Efianking_of Ceramics,_8ewing Machines and Starch by_ _Employmgnt Among the 2§_Industries, 1950-1960 1950 51 52 53_ 5“ 55 56 57 58 59 60 5 5 5 9 12 12 12 13 13 19 15 12 12 12 1! 9 7 7 7 7 5 5 16 17 17 15 16 15 17 17 18 19 19 17 16 16 17 17 17 16 16 16 15 1“ 2O 20 2O 2O 2O 2O 21 21 21 22 22 Cotton Textiles 26 26 26 26 26 26 26 26 26 26 26 Distilleries 8 8 ‘9 7 6 5 5 5 6 6 6 Electric Fans 10 9 7 6 5 6 U u 5 9 12 ‘lectric Lamps 2 2 2 3 3 3 3 3 3 3 3 Fruits and Vegetables 3 1 1 1 1 l 1 1 1 1 1 Glass 19 19 19 19 19 19 18 18 17 21 21 Iron and Steel 23 23 23 23 23 23 23 23 23 23 23 Jute Textiles 25 25 25 25 25 25 25 25 25 25 25 Matches 1U 1U 1U 1U 1U 16 15 1n 1“ 16 16 «a» 1950 51 52 53 5“ 55 56 57 58 59 6O . Paints and Varnishes 7 7 6 8 8 8 8 6 u 8 7 Paper' 18 18 18 18 18 18 19 19 19 u 9 Plywood 6 6 8 5 7 11 11 10 10 12 11 Rice Milling 21 21 21 22 22 22 22 22 22 2O 20 W h 3 3 u u u 6 8 9 7 u Soap 11 11 ll 12 ll 9 9 9 8 11 10 Sugar 2” 2h 2H 2U 2U 2U 2U 2U 2U 2H 2H ° 1 U U 2 2 2 2 2 2 2 2 Tanning 13 13 13 13 13 13 13 12 12 l3 13 Vegetable 0118 22 22 22 21 21 21 20 2O 2O 18 18 Wheat Flour 9 10 10 10 10 10 10 11 11 10 8 Woollen textiles 15 15 15 16 15 1“ 1a 15 15 l7 17 Source: Computed from India (Republic), Directorate 01 industrial Statistics, , epgrte on Ego Census oi‘ Magufiactuggs and S X Na! 5.le flu Oscar-tape )quc. em Are 34 {an “@710 L AM/c . Ceramics shows a fluctuation of one rank for most of the years except a change in two ranks for 1959 and three ranks for 1960. Sewing Machines shows a change of two ranks for 1956, u for 1957, and 5 for 1958. ment as the highest wage industry. In percentage change, This industry appears to have shown some change in employ- Starch was the highest wage industry for the whole period, and the ranking for this is almost constant showing minor change in employment. If there has been a movement between lowest and highest wage industries, it does not seem to be very significant. The ranking of states on the same lines was done and the lowest wage state in absolute terms was Assam and the highest, Bihar. In percentage terms these were Assam and Andhra respectively. The shows the ranking by employment of among the other 12 states. following table the three states Annual '3 ..65' Table if *Ranking;of Assam, Bihar and Andhra by 13_ployment Among the 12 States 1950-1960 1250 51 _,52 53 5“ 55 56 57 58 59 60 1. Andhra 6 6 6 6 6 6 7 7 7 6 6 2. Assam 1 l 1 1 l l 1 l 1 l 1 3. Bihar 8 8 8 8 8 8 8 8 8 8 8 h. Eoman 12 12 12 12 12 12 12 12 12 11 12 5. Delhi U u u u u 5 5 u u u 3 e. Madhja Pradesh 7 7 7 7 7 7 3 6 5 7 7 7. Madras 9 9 9 9 9 9 9 9 9 9 . 9 8. Orissa 3 3 3 3 3 2 2 2 2 2 2 9. Punjab 5 5 5 5 5 u 6 5 6 5 5 10. Rajasthan 2 2 2 2 2 3 U 3 3 3 u 11. Uttar Pradesh 10 10 10 10 10 10 10 10 10 10 10 12. West Bengal 11 11 11 11 ll 11 ll 11 11 12 11 Source: Computed from India (Republic), Directorate 01 1ndustria1 statistics, 3929;; on the go nggs o; Maggggotures and the stzies. X N01 MM 7‘“ ewe-4A ANA MaL Anal-Z flu: «(offers/- AM./<., The table shows very little change in ranks. It is striking to see that Bihar and Assam have not changed in ranks for any of the years. Thus, the overall wage level has not been influenced by the change in the composition of employment. The tables show the very slow change in employment for most of the industries and states. m m. chapter as W with as analysis of the changes is the overall figs streets" 1:.an te “lecturing industries for the period 1930-1960. N w series of average annual uses were squad. blotch-hastenaqatedtra-thedataoamrageaaaualwapa for 26 inhatriee, and the ether fra- the average annual wages of industries is 12 states. a min aoeslaaiea that merged fro- the analysis was that both may and real ages had increased for the period 1950-1960. he factors are cited for an iaareaaa is sagas. Pirat the period was worked by apasralaeee-Iaaapneiaaapdiasreaaeiaiaoaee.0aderthestia- alas previded by develop-at planing there was as increaee’ is is- rescuer W by as increase is the percentage of astieaal ma- erigiutiag is the .dara wage earning sector. Is this process, “mimetthetaeteryaadthesiaeworkaruaeededthat a! ether classes at weaken. aeeeadly, peer-eat wage policy is iahatriea. played at W role is iaflaaaciag the iaareaae is the level e! noes. Ales, m: up policy, by ans-pun u' liahthepssuaaretdonaeeallewaaeatetheooaaa-arpriae indel as “le for as iaereaaa is the dear-ass allow-ea sad other eupeasata besides basis algae is the worker's total earnings, finish ai‘a have partially asle'ardad his real wage position is the face a! prise iasreaeea a! the ea-editiaa that he boys. 7 It was eaaaeaded that'the else and characteristics of uioaa, as wall as the “rifled nature at its membership figures mat be seasidared hetero tee-d reliaaee can be placed aaaay iadss at uieaiaatiea flick .Ild ah. the influence of Indians is raisin the level of uses. . 67- M has been a divaraificatias in the rate of growth of. aployaat is as far as the saer industries have shows a greater percentage growth is amt asasrad to the older ssd established iadustrias. else Iday d bat lsngsl which ranked law is age increases Ind ceseaded their position to the other states with respect to qleyaat gawth. Finally it as shows that the overall age level had set be. influenced by the change is the capesition of “lay“ resulting ira workers avisg tro- low afi industries to high age inastries. -69- CHAPTER 3 The Inter-Industry Wage Structure In this chapter we will analyse the trend in wage differential in manufacturing industries for all of India for the period 1950-60. Th 3 has been undertaken on the basis of our present understanding of the methodsl of measuring wage differential changes. A detailed ranking of industries for each year on the basis of their average annual earnings will be considered. Then an investigation into the movement of wage dif- ferential will be taken up in an effort to ascertain whether the differential has narrowed or widened. Explanations would then be provided for the observed trend in the differential. Finally we will consider how employment and average annual money earnings have varied in each industry as shown by the coefficient of variation. 1The following works among several others too numerous to mention are particularly relevant to the methods adopted here. Donald E. Cullen, 'The Interindustry Wage Structure', American Economic Revigw, Vol. N6, Pt. 1, June 1056, pp. 35H—55. David R. Roberts, 'The Meaning of Recent Wage Changes', in R.A. Lester and J. Shister (ed) Insights into Labor Issues, MacMillan, 1948, p. E28T' G. Fottier, 'The Evolution of Wage Differentials: The Study of British Data', in John T. Dunlop (ed) The Theory of Wage Deter- mination, MacMillan, 1957, p. 2Ul. A. M. Ross and W. Goldner, 'Forces Affecting Interindustry Wage Structure', Quarterly Journal of Economics, Vol. LXIV, No. 2, May 1955, p. 263. I. Change in the~fianking of Industries in the Wage Hei rarchy ~69 The following table shows the detailed ranking of the 26 manufacturing industries on the basis of their average annual wages paid to thy a-“kers each year, for the period 1950-1960. Number 1 refers to the lowest rank and number to the highest .13 - A. rank and hence numbersrassigned in an ascending order. Ranking_of Industries According_ Table l CE: ) L. 6 to Change in Average Annual Earnings ;?E72:— All Industries a i‘tcfy'CjLeS hiscults ”‘3' I \ ‘lt31‘\inbi CS Chemicals Glass 1' ron 8‘ f‘teel Jute Textile India - 1950—I960 Rank . 59 :1 52 53, _g_ 55 56 57 58 59 60 2n 29 22 17 19 18 19 21 21 21 19 9 9 9 16 9 9 9 8 8 16 15 17 19 17 22 22 22 16 17 17 17 18 1o 10 10 11 1o 12 11 13 12 6 5 19 17 18 18 20 21 20 2o - 2o 22 22 22 23 23 23 23 23 23 23 19 19 2o 12 12 11 9 11 11 8 9 11 10 8 13 16 19 19 17 19 18 19 2 2u 2m 16 15 15 16 18 17 22 18 18 18 16 7 7 5 3 3 3 3 3 3 2 2 5 5 6 6 5 U 6 6 7 7 6 25 26 26 26 a 25 25 25 25 25 25 11 11 13 12 1B 19 19 11 10 9 10 f‘ E, db. 9 50 51 52 53 _5“ 55 5f 57 58 5 6O hMatches 21 21 21 21 13 8 10 10 16 19 19 Paints & Varnish 20 2O 2O 2O 21 20 21 22 22 2O 21 Paper 15 18 19 1M 15 15 17 15 15 8 13 Plywood 6 u a 5 u 5 u u u 5 11 Rice Milling l 1 l l 1 1 l 1 1 l l f‘ \ NI ,ewing Machine Soap 23 22 29 2M 2M 29 29 29 29 23 23 Sugar 3 5 7 14 5 7 5 ‘3 ‘> U 7 Starch 2 2 2 8 8 10 13 16 13 12 17 Tanning 8 8 8 Z 7 6 7 7 6 11 9 Veg. Oils u 3 3 2 2 2 2 2 2 3 3 wheat Flour 1“ 13 12 13 12 13 12 12 9 15 12 Woollen Tertiles 18 19 1b 15 15 16 15 1“ 1“ 13 11 Source: Computed (You, India (Republic), Directorate of Inr'iustri al Stati stics, mm ’3 \ the Census of,Manufactures and the Annuiljfiuwm§:of Industries. The most striking feature of the rankings is the minor change in rank that is shown by each industry. Rice Milling has the same rank for all the ten years, as the lowest paying industry. The tie for the highest paying industry has always been between Sewing Machine and Iron and Steel in all the years. In the case of the other industries, the ranks have not altered very much year by year. A closer scrutiny of the table will reveal the following. The maximum number of rank changes is seen to be lBland is shown by Starch. 1The rank changes are calculated as the difference between the lowest and the highest rank attained by the industry for the whole ten year period. ‘2 1‘) Qi! -7,. examine We may start from this, and V the number of rank changes for each industry. Matches changed by 13 ranks, Electric Fans and Paper by 11 ranks, Biscuit making, Cement, Ceramics, and Woollen Textiles by 8 ranks, Bicycles and Electric Lamps by 7 ranks, Wheat Flour by 6 ranks, Chemicals, Fruits and Vegetable Canning, Jute Textile, and Tanning by 5 ranks, Cotton Textiles, Distillaries and Breweriesszgzgar by‘u ranks, Glass by 3 ranks, Paints and Varnishes, Plywood and Tea Chests, Soap, and Vegetable Oils, by 2 ranks, Iron and Steel and Sewing Machines by 1 rank. Rice Milling has shown no change in its rank at all. Another striking feature of the ranking is the change in ranks of the highest and the lowest paying industrlnat the beginning of the period. Sewing Machines and Rice Milling show that they have remained high and low respectively throughout the period, with practically no change in their rankings. The latter has not changed in rank at all. The former registers a change of one rank, for the whole period. Finally, it is interesting to note that the leading industries, Cotton and Jute textiles have not shown a significant change in their ranking. In order to verify our analysis of rank changes, rank correlation coefficients were calculated for each year in the case of the 26 industries. The coefficients turned out to be as_follows: QIJ ~7a— ‘ 1 Table l! Spearman Rank Correlation Coefficient for Average Annual Money_wages,”23”lndfiégpies, All [ndia_: 1250-1260 " 1233 Rank Correlation Coefficient 1951 .98222 1952 .977u3 19)3 .93230 195“ .92273 1995 .86393 lfi55 .86119 1957 .8U888 1953 .87760 1959 .832u7 1950 .78871 Source: Commted from India (Repub‘ic), Directorate of Industrial Siatis'fi'ms,’ Revert on the Census of Manufacture; and the Annual Survey of Ind:stries Note: The ranking of the average annual money waves for tne 26 industries for each of the years 1951 to 1960 havoflbeen correlated with the ranking for 1950. The formula for Snegman RanK Correlation Coefficzent is as flilowe: is the Spearman Rank Correlation Coefficient is the difference in rant-c between pained items N is the number of oairs of observational where -73- The high coefficients confirm our previous finding concerning the stability in the rankings. Does this mean that wage differentials have neither narrowed nor widened over the years? On the basis of our rankings we may support this conclusion. If we are not careful we may even be tempted to conclude our investigation here. But a detailed survey of the existing methods of measuring wage differential changes showed that rank order coefficients do not tell the whole story. Slichter has shown that groups of industries can show a diverse pattern of change in workers' earnings over a period of years and nevertheless retain substantially their ranking throughout that period.1 Even thOUSh he does not discuss this possibility in detail, it is clear that the rank order coefficient measure only differ- entiates industries, which, at any given time, rank highest or lowest in the wage structure. It does not differentiate industries in which wages increase more or less rapidly. To find this, both absolute and rela- tive changes have to be considered in analysing the movement of differentials. ll. The*relatize change in money earnings compared to its original level To focus attention on both the absolute measure 18. H. Slichter, 'Notes on the Structure of Wages', Review of Economics and Statistics, Vol. XXXII, Feb. 1950, p. 88: and the relative change in wages, '1 '47}; V the original level (absolute terms) of wages in each industry as of 1950 has been related to the relative change in wages as shown by the index of money wages in 1960 in each of the 26 industries. Ranking of-Money_Wages in 1950 and Percent Change in Money Table 1% Wages, 1950-1960 for 26 Industries, All India l a 3 industry flank of Industry Sewing Machine Iron & Steel Bicycles Soap Jotton Textile 'V Matches Paints & Varnishes Jhemicals Woollen Textile Pement Electric Lamps Paper & Payer Products Who: at Fl our blectric Fans bistilleries Jute Textile Av. 1 2 U 10 ll 12 1M 15 l6 17 18 Annual Waves“ (Rs T'in 1950 U Rank of % Increase in 1960 (1950=100§ Starch Electric Fans Sewing Machine Sugar Chemicals Biscuit Making Soap Paints & Varnishes Tanning H m S SS Cement Iron & Steel Rice Milling Electric Lamps Jute Textile Cotton Textile Plywood Distilleries & Breweries** 1 2 .‘ I "j {3 l 2 3 Tanning 19 Wheat Flour'“ Fruits & Veg. Canning 20 Paper & Paper Products Plywood 21 Vegetable Oil Glass ‘ 22 Woollen Textile Vegetable Oil ' 23 iatches Sugar 2M Bicycles Starch 25 Fruits & Veg. Canning Rice Milling 26 Ceramics i'Numher 1 shows the highest rank, number 26 the lowest rank. ...Wheat Flour and Breweries, same rank in Column R. Source: India (Republic) Directorate of Industrial Statistics, Bepogt,on the Censgs a; hsnufactures, and Annual Survey of Industgig . 1 an we say that the percentage increase in wages show an inverse relationship to the original level? If 3 so’then it can be said that the structure has been pulled together and its dispersion around a -entral tendency such as the median has been reduced. Following the procedure adopted by Cullen3, we can choose the top five industries and the low five in- dustries for comparison. Sewing Machine, Iron and Steel, Bicycles, Soap and Cotton Textile are the high wage industries in 1950 according to our distinction. Rice Milling, Starch, Sugar, Vegetable Oil and Glass are *.._---——. _M“*‘-_-. 18 19 2O 21 22 23 2H 25 Ceramics is the only industry to show a decline in wages. (David R. Roberts, 'The Meaning of Recent Wage Changes', Op. Cit. 3Donald E. Cullen, ’The Interindustry Wage Structure', Op. Cit. l ~’76' the low wage industrfigan that year. We have to see whether the high wage industries show an inverse re— lationship with respect to the relative change in wages by the end of the period as indicated by a ranking of the index of money wage in 1960. Similarly, we have to also see whether the low wage industries have shown such a tendency. Bicycles have shown a clear inverse relationship) 03 it is ranked among the low five in 1960. Sewing Machine has retained its position among the high wage group. Iron and Steel, Soap’and Cotton Textile have slipped in ranks. Coming to the low wage industries)8tarch has clearly shown an inverse relationshp as it ranks first in 1960. Sugar also ranks among the high group in 1960. Class, Rice Milling and Vegetable Oil do not find a place among the top five in 1960. Of these, Vegetable Oil has not shown much of a relative change and remains very near the low five industries. What can be said about the relative change in earnings when its original level is taken into account? There seems to bettendency for a contraction of the wage structure in percentage terms. Cases such as Bicycles, Starch and Sugar clearly show an inverse ) relationship. But how significant is this contraction? The initial and the terminal year; alone have been con- sidered in the case of the high wage and the low wage industries. It seems desirable that we must compare a? .77- both the absolute and the relative change of the high wage and the low wage industries, along with the concommitant changes in the general level of wages for each year.1 If the dispersion in the structure was around a central tendency, then it can be said that the structure as a whole may not have shown considerable expansion or cortraction. This leads us to consider some measure of the dispersion of the wage structure, side by side with the movement of wage differential. ill. The__ relative ani absolute c ange in mt oney earnings ~— ..-“ om; _ared to the oiig+rsivn of the wage structure r ) Following Cullen's‘ method of measuring the dis- , nag persion, the interquartile range, tne median,v he absolute and percentage measures were chosen in a de— tailed ranking of the 26 industries for each year. These are presented below. .4“... - —-. 1Commenting on the emphasis laid on absolute measures by Ross and Goldner (Op. Cit.) Edwin Mansfield cautions- against drawing conclusions on this basis alone. He says there would still be the need to take account of the concommitant changes in the general level of wages. See Edwin Mansfield, 'The Measurement of Wage Differentials', Journal of Political Economy, Vol. LXII, Aug. 195b, No.wh, p. 3W7. ’) Donald E. Cullen, 'Interindustry Wage Structure', Op. Cit. p. 55. ‘ a! ‘79— Igble 19 Into -- .rt .0 d Median of Avera;e Annual Mone Wa;es . 4, w tllmmlm $9111 ergbt a i_ . ndus r ea; ' 0- 260 Year 1950 1951 1952 1953 195a 1955 1955 1957 1958 1959 1960 Note: 2 3_ n 5 6 7 831—g" 100 u6 376 816 208 as 589 53 955 859 221 05 676 58 520 909' 225 an 713 50 u72 951 222 as 731 as s51 9&6 205 09 666 u6 nan 911' 196 51 608 Rs RS6 ' 1015 . 195 51 ' 639 99 507 10u0 ' 190 53 6&5 52 565 1091 195 ' 51 719 58 636 110a 255 39 1186 53 63h 1200 201 01 1153 In columns 5, 6. and 7. the nodian or the average annual earnings in those industrio- torling the top quarter of the earnings structure for each year, and tho median of the average annual earnings in those industries forming the lowest quarter for each year, were taken to arrive at the percentage and the absolute figures. m: Co-putod bo- India (Wis). mroctonto of Industrial Sati stios. Loggrt W“ W momma- t'n. t . intorquartila rings divided by the median shows no doubt a fluctuation. But notice that the fluctuation is within a narrow range which is around 50.1 This 1Cullen concluded that the fluctuation is within a narrow range of 25 when the deviations are five paroontacsa above and six percentages below this figure. Cullen. Op. Cit. p. 361. .7q- narrow range gives a low dispersion in wages. In the light of this, what can we say about the movement of differentials? The absolute differentials have clearly widened as shown in column 7._ Percentage differentials seem to have narrowed in so far as this tendency can be noticed in five out of the ten years. For the whole period, however, there is a widening of this differentials too. We may be able to see more clearly the trend in wage differentials if we note the concomitant changes in the general level of wages. Columns 3 and U of the table have been shown for this purpose. The inter— quartile range has shown an increase up to 1953 and declined for the next two years. Thereajraz it steadily increases up to 1959 and shows a slight decline in 1960. But, the range has widened if we take the whole period into account. With regard to the median, except for the two years 1955 and 1955, there is an increase for all the other years. These two tendencies strengthen our conclusion regarding the increase in the absolute differential in wages. The question is how significant is the narrowing of the percentage differential in wages noticed in columnsS and 6. In View of the fact that the structure has been stable, its dispersion within narrow limits, we may say that the narrowing of percentage differentials for the years noticed cannot be said to be very significant. The conclusion is that .¢\ -30- there has been a widening of wage differential. The explanations for the stability in the rankings and the widening of wage differentials are as follows. 1. The Stability in Ranking Considerable interest has been shown in studying the changes in the industrial wage structure of several countries and in undertaking inter- national comparisons.1 The impression gained from these studies suggeststhat there has been a strong similarity in the changes. Rank correlation coefficients between the 'beginning- and end~period' industry earnings structures are pre- sented for Sweden, U.S.A., U.K. and Canada in a study by the Organization for Economic Co-operation and Develop- ment2 for periods varying from 8 to 51 years. It is reported that "Stability in ranking is found whether average earnings of all workers or of specific groups of workers, are studied. A separate study of Denmark, Dunlop and Rothbaum, "International Comparisons of Wage Structures", international Labour Review, April, 1955 Organization for Economic Co-operation and Development, Wages and Labour Mcbility, a Report by a Group of In— JEpendefitzfiperts on thE'Eelation between Changes in Wage Differentials and the Pattern of Employment, Paris, July 1965, esp. pp. 92-23. 2 The relationship between changes in wage differential and the pattern of employment was stulied by com- paring the changes in employment and variability of earnings. - 8;- Sweden, the United States,and the United Kingdom coverilog approximately lO—year periods in the 1950's which was made for the Group, donfirms these findings. The stability noticed in our ranking suggests that the period 1950-1960 in India was marked by the same type of changes in the wage structure found in the O.E.C.D. studies. Though a direct comparison between the findings relevant for countries like Sweden, U.S.A., U.K. and Canada,andAfindings relevant for India is not possible owing to the wide differences in industrial structure and general economic backgroud, the stability in ranking in the case of India does raise the question of examining whether a developing economy should have a different type of change in its industrial wage structure under an accelerated pace of industrial development. It is clear that whether a country is highly developed or not, changes in its industrial wages are very likely to be influenced by the same set of factors which influence them in more advanced countries. These factors would relate to product market conditions, capital intensity, profitability, trade_union pressure, managerial objectives, changes in the skill composition of the labor force and the role of government policy in issues relating to the fixation of wages. An attempt to relate the changes in the wage structure noticed in our study to these factors is undertaken below. .-?L_ 2. Product Market Conditions There is reason to ’ believe that the high paying industries in India enjoy favourable product market conditicns in that they show relatively a higher degree of producer concentration com- pared to the low paying industries. Information on producer concentration is scanty and until recently, was virtually nonexistent in India. The following table is based on the information supplied by the National Council of Applied Economic Research to the Planning Commission, in the course of the latter's investigation on the distribution of income and levels of living. Eelslgfig ‘ EL0§s.¢_e-zf..£eacsnt resign}; S e 1 c t e d I n daat r' i e 8 Industry Year - 1 share of 1 share of No. of top-most unit/tOp-few units/groups group in total units/groups under production in total pro- column (u) capacityEZ duction capa- cityél _‘ _{1) (2) (31_ (u) (5) 1. Finished steel 1958 63.70 93.36 2 groups 2. Fig iron 1958 5N.63 90.08 2 groups 3. Electric Lamps 1960 “0.00 88.70 1“ units N. Sewing machines 1960 88.00 88.00 1 unit 5. Soda ash 1958 2.25 8U.68 2 groups 6. Flectric fans 19.1 51.00 I2.00 u units 7. Paper and paper board 1958 23.50 77.90 5 groups 8. bicycles 1959 20.20 72.72 M units 9. Cement 1960 h5.00 71.90 3 groups 10. Soap 1957 30.75 69.11 u groups 11. Matches 1960 60.00 60.00 1 group with 5 units 12. Surerphosphate 1958 lb.76 53.0“ 5 groups ,_13. Hydrogenated oil 1958 19.01 “7.09 6 units ‘LB Paints and varnish 1957 ll.u0 u5.9o 6 units 15. Ceramics 1957 17.29 39.72 U groups lb. Jute textiles” 1958 12.29 37.61 U groups 17. Caustic soda 1958 1U.76 28.51 2 groups (oven) -35- (1) (E) (ii (8) (5) 18. Goal 1958 11.09 2a.70 3 grOUps I 4. Sulphuric acid 955 it.80 23.7? 2 groups zfi. Cettflh textiles 1998 H.10 11.99 3 groups 31. Sugar 1958 ------------------ 19L“) "7.36 14.56 (_3 unit-.8 ‘Tercentage of looms installed. urce Government of India, Planning 00mm ssion, Report of the F>mmittce on Listrihution and Levels of Living, Part I, New Delhi, ; .r fl-.- --.~_ ~. -------- “...-.. “me-t? ------—-- --——°°'~- -—--~-—-~-- -—- {\pr 1,477., 3. 3f}. 1e Tap most unit or grcun refers to one unit or group ani top few refers t6 more than one. It is Stunt that t1“? percerfizn“e stmnwe of tfmflnflf—fFW units/groups in total production capacity as revealed by column 5 is the highest in the case of finished steel and lowest in the case of sugar. In our previous ranking of these industries, iron and steel was placed in the high paying group, and sugar in the low payinv group, fer all the years, in the year by year ranking shown in tshle lb . There is a sliuht gain of rank for sugar but it never qualifies to re placed as a high paying industry in terms of its ahsolute earnings for its workers. Sewing Machines with 88 per cent of pro— . .iuet'ivc: carwicitgz ccwwcexALrate«i 4:1 the }1anru3 of‘rohe Ilnit. has teen the highest pay 5 ing industry For all the years, except Letween 1951 and 195?, when it was the second highest paying industry, concedinv its position to iron and steel. There is littlD doubt that this industry enjoys a virtual monOpoly and thus possesses very favourable product market advantages. n The table does not provide information on producer concentration for all the industries considered in this study. But it clearly shows that the high paying in- dustries enjoy considerable product market advantages. They are controlling the bulk of what is produced and have few or no competitors in their field. Given the condition that demand for both consumption and gp‘ital goods has gone up during this period, industries like sewing machines, iron and steel, chemicals, bicycles and electric fans and lamps, starch and soap have been in a position to raise their prices and hence pay higher wages to their workers. On the other hand, industries which face relatively a greater amount of competition like those in the food group, eg. sugar, wheat flour, vegetable oils and rice milling have not been able to enjoy similar product market advantages. Their price advantage lies in the general inflationary price in- crease which is characteristic of the period, and it is likely that the benefits of this type of price,advantage have not resulted in wage increases. Where increased demand has been associated with smallness in the number of groups or units dealing with the product, workers engaged in such groups or units have been able to get the benefit of the favcrebte product market condition. The continued enjoyment of this advantage has keptthe high paying industries at the higher ranks, compared to the lower ranks maintained by the low paying industries. This has served as one of the reasons for the wage differential to be maintained between the two. Another factor which has influenced to create favorabie product market conditions is the import control policy followed by the government. Controls were introduced as a war time measure in 19u0. After l9U7, import control policy entered a new phase when it came to be linked up with foreign exchange control. importers have been divided into three categories,nadely actual users, established shippers and new comers,and there is restriction on importse'thghsgteggods cannot be im— ported without a license. This has come in the wake of increased demand for both capital and consumption goods, creatinr more scarcity and thereby adding to the favcnuole ant: market conditions enjoyed by the industries. 3. Capital Intensity Can it be said that industries that are more capital intensive have been experiencing greater productivity and hence finding it possible to pay higher wages? In other words can an increasing trend in fixed capital per worker be associated with increasing wages? In order to ascertain this the index of fixed capital per worker for each of the twenty-six industries was constructed ani compared to tr index of money wages already calculated for the period 1950—l960. The five industries showing the highest percentage rates of growth in fixed capital per worker were compared to the five industries showing the highest rates of growth in -86- money wages. The same type of comparison was made in the case of the five industries showing the lowest rates or growth in fixed capital per worker. The following table gives the result of the comparison. um Comparison of Fixed Capital Per Worker and Monengaggs for Industries 1951-1260 _Base 1959-100 High capital intensity_ High money wage Iron and Steel (_lJ Starch LID ' Starch [23 Electric Fans L2] Sugar [3] Sewing Machines [3] Cement B] Sugar L14] Chemicals [SJ Chemicals [5] Low Capital intensity Low Monevaages Vegetable Oils [.1] Ceramics [2g Fruit Canning L2] Fruit Canning [2] Electric Lamps [3:] Bicycles [3] Rice Milling [if] Matches [14] Soap [S] Woollen Textiles [SJ N.B. Figures in bracket indicate rank. In the Indian context there has been a trend for increasing capital intensity in manufacturing industries. This was revealed by the calculation made in this study and also by others. 1 See J. C. H. Fei and Gustav Ranis, Development of the Labor Sugplus Economy: Theory and Po 0 , e Economic Growth Center, Yale University, 196! . The table shows that in the case of the high capital intensity group, starch, sugar, and chemicals provide a comparison. Whereas in the case of the low capital in- tensity group only fruits and vegetable canning pro- vides a comparison. Iron and steel in the former group sdfiging the highest rate of growth in fixed capital per worker ranked, twelfth in its rate of growth in money wages. Though the association between increased fixed capital per worker and money wages isgmronger in the high capital intensive group than the low, the rela- tionship cannot be too strongly emphasized and is by no means definite. An examination of the rest of the industries only confirmed this view. The mere increase in fixed capital does not necessarily imply that workers in such industries earn a higher wage compared to those in other industries having a lower ratio of fixed capital per worker. , -88- h. Profitability The economic rationale for linking profits with wages lies in the capacity and willingness of managements in industries having higher profits to grant wage in- creases. Reviewing studies made in the United States on the relation between profit and wages, and related variables like concentration, degree of unionization and product price, Reder feels that if a distinction between the "market power" sector and the "competitive" sector of industries can be made, an explanation can be suggested for the payment of relative increases in wages in the former sector.l Arguing that at the end of World War II both sectors were confronted with sit- uations of strong demand and high profits)he suggests, "The competitive sector acted to eliminate its excess demand faster than the market- power sector where long gestation periods of capital goods combined with a cautious outlook to hold back the investment program. This made the market-power sector's period of high profits, full capacity Operation, and "strength" in product prices last longer than the competitive sector's period did. This, 1M. W. Reder, "Wage Differentials: Theory and Measurement! Aspects of Labor Economics, A Report of the National Bureau of Economic Research, Princeton University Press, 1962, p. 293.. The distinction has been made by W. G. Bowen in his book Wage—Price_I§sug: A Theoritical Analysis, Princeton University Press, 1960. 4Thefifimarket power" sector consists of industries that are both highly concentrated and highly unionized, and the "competitive" sector of industries having the reverse character- istics. in turn, facilitated the payment of relative increases in earnings (in the market-power sector) during most of the 1950's.1 Further, he also emphasizes the crucial role of product prices, that is when "firms believe that cost increases can be passed on to buyers, they are more inclined to grant wage increases than when the reverse is the case."2 Thus, the capacity and willingness of managements to grant wage increases depends on the period of high profits, full capacity operation and "strength" of product prices. If these are prolonged the management would be inclined to pay their workers more. In India, we have already seen that the payment of bonus is based on profitability. The quantum of bonus is determined by what is know as the Full bench Formula3 by which depreciation, taxes, reserves for rehabilitation, interest on paid up capital and return on reserves em- ployed are deducted from gross profit to arrive at the available surplus for distribution as bonus. But what is not clear is the procedure for set off if the management incurred a loss. Much depends on how the management can carry over the surplus for a succeeding year if loss is incurred. Since it is necessary to relate profitability 1 Ibid. p. 29M 2 Ibid. 3 The formula was laid down by the Full Bench of the Labour Appellate Tribunal, Bombay, in Mill Owners Association verses Rashtriya Mill Mazdoor Sangh in 1950, and subsequently the parties entered into a bonus pact for the years 1952-1956. - 90- and wages for each year to study their trend, we need to know when the bonus is paid and how adjustments are made to carry over the surplus in case of loss. In the absence of data on this it is difficult’in the Indian context, to correlate profitability and wages in an attempt to find their relationship. One study, the only one known to this writer,correlates average daily wages with the gross profitability ratio for the period 1950-52, 1953-55, 1956-58 and 1959-61 in the case of manufacturing industries in India, and finds that barring the second period wage and profitability are signifi— cantly related.1 Though this study does not analyse wages and profitability with the view to explain wage differences among the industries, this finding is of value in showing that profitability does have a relation to wages in the Indian context. Given this, it can be argued that the widening of wage differentials between the high wage and the low wage industries must be due to the profit position of the former compared to the latter. 5. Government Intervention in Wage Fixation In this part of our study we are primarily interested in explaining the observed widening of the wage differ- C. K. Johri and N. C. Agarwal, "Inter-Industry Wage Structure in India 1950-61 -~Ah Analysis, Indian Journal of”Industrial‘Relations,‘Vol. l, No.‘f: April, 1956, pp. 395196. “3 .9|. ential between the high and the low wage industries. In assessing the influence of the role of government in wage fixation on wage differentials we may consider the official policy which has emphasized two things. 1. The process of‘standardization of wages should be accelerated and extended to as large a field as possible. This essentially meant the narrowing of differentials. 2. Differentials for various Jobs should be maintained at the minimum levels Justified by the degree of skill required, the strain and fatigue involved, the training and experience required, the responsibility to be under- taken, thementaland physical requirements for doing the work, the disagreeableness of the task and the attendent hazapd3,1 This essentially meant the recog- nition of wage differentials. There appears to be a contradiction between the desire to narrow differentials and the desire to recognize their incentive value. The equity and the economic points of view are both stressed and it in their combined impact on differentials is dif- ficult to determine. In May 1956, the Ministry of Labour and Employment set up an official 'Study Group on Wages' for the preparation of the wage material which would be needed by the Wage Boards and also by the government. In 1957 the group was reconstituted and representatives of state governments, employers and workers' organizations and an economist were associated with it.2 Its views constitute a 1Government of India, Planning Commission, First Five Year Plan, Ch. XXXIV, para ”7. 2Government of India, Labour Bureau, Consultative Machinery in the Labour Field. Labour Bureau Pamphlet SEPies. 1- TUHC ~11 reliable source for understanding the policy of the government on matters relating to wage differentials.1 The Study Group points out that it is necessary to bear in mind that in any attempt to redraw wage dif- ferentials, employees everywhere attach great impor- tance to prevailing differentials, be they scientific or notjand often exercise pressure to maintain those differentials.2 This view goes to show the underlying rapproach for the formulation of a policy with respect .to wage differentia13. If employees attach importance to differentials it would imply opposition to any type of suggestion for the reduction of differentials based ’ , ' 2 on scientific evaluation of Job content.“ 91 Q .0 far there has been no clear cut pronouncement of government policy with reSpect to job evaluation and its relationship to wage differentials. Recommendations Government of India, Ministry of Labour and Employ- ment, Some Papers_gn Wage Pgligy, New Delhi, 1957. See also ”Some~General Principles in the Deter— mination of Industrial Wages in India", Indian lgpour Gazette, Vol. XV, No. 6, Dec. 1957, and "Principles of Wage Fixation", Indian Labour Gazette, Vol. XV, No. 8, Feb. 1958. These two articTES have been published by the Labour Bureau in a series for giving information on the work of the Study Group on Wages. ’) ( < ‘ “W Some Papers onwflage_folicy, Op. cit. p. 3. 3 Both scientific considerations and employer- employee cooperation are stressed for imple- menting a wage differential sKBeme. Ibid. p. 3. 93+ 'cb evaluation aLd f;r evolving a standard OC- J for cupational Classifivation exists, hut these have not figured prominently in cases where the government has intervened in the fixation of waxes. Home adjudicators have merely fixed the hfnimn: LES}? wage and left the determination of wage differentials to the parties them- selves. Lthers hive made a dietih Linn between un- skilled, semi—skilled and skilled and fixad separate fur each ratequroup. Government poli.v towards fixing dearness allowance has been cited as a factor in raising the minimum wage payable and thereby compress the differentials between different classes of workers.1 This is more so if the dearness allowance is awarded as a flat rate regardless of the income group. The flat rate will tend to pull up the wages of workers in the lower category and the wage differential between the higher and lower cate- Uwries of workers will be narrowed. The following table shows the estimated percentage of workers getting dearness allowance according to the system of payment. .-.-” ---. -. _ -.-.— ..“..-‘—*- --u...-..—m--__. “W 1 Philip Kotler, Problems of Ind strial Wage Policy in India, Unpublished Ph.P. thesis, Massachusetts Institute of Technology, 1956, p. 281. Esxssniase_oCogniiimfsyiaslacassettecs Hegeivin- -44. Iatis_&3r Dearness Allowance stem_o£m£aygant,_1958-1959 . 6 t\- ..t (V "Esserdins_to c Percentage of Industry Units Workers Workers_getting D.A. according to paying Receiving Consumer Flat Income Others D.A. D.A. price rate Groups _3 _ ___i ndex l P 3 U 5 6 1. Factory industries 38.1 76.5 b0.9 27.3 30 I l l 1.? Other factory industries 33.0 U0.“ 3?.1 21.7 “1.6 U.6 1.3 Engineering “1.8 85.“ 11.0 2.1 85.8 1.1 2. Plantation industries 76.5 60.8 0.3 7.9 0.2 91.6 3. Mining industries 73.7 8Q.7 81.5 2.7 lO.U 5.“ Source: India(Republic) Ministry of Labour and Employment, Occupational EEEE_§V§Y3£4”E§DSF?l Report, 1958-1959, S mla, 1963, pp. 42-56. Factory industries refer to textile, (cotton, Jute, silk and woollen); other factory industries are, Cement, paper, sugar, chemicals, and printing presses. Engineering, refers to electrical machinery, metal extracting and refining, ship building, rail- way workshops, tramway workshops, motor vehicles, aircraft and bicycles. Plantation industries includes, tea, coffee and rubber. Mining relates to coal and manganese. The occupational wage survey was launched in July 1958 and completed by the end of August 1959. ’ Column 2 of the table shows that more than three fourths of the workers get dearness allowance in the textile About 60 group of industries, engineering and mining. per cent in plantations and U0 per cent in other industries receive it. Except in the case of plantations, workers getting dearness allowance linked to the consumer price index form a larger percentage than those receiving it at a flat rate. In view of this it is doubtful whether the granting of a flat rate allowance would have very 35'- 1 much compressed wage differentials. 6. Skill Composition_of thgwbabor Force in Manufacturing During periods of rapid economic growth one would expect short run inelasticity in the supply of skilled labor in underdeveloped countries which would lead to an increase in its wage rate. A widening of the differential between skilled and unskilled labor would result and this process would continue for some time. When training gets underway and the general level of education is raised and as economic development takes place, the difference in the wages of unskilled and Skilled labor will begin to narrow. Such narrowing can be attributed to a diminution in the number of unskilled labor and a gradual change in the skill composition of the labor force. By this is meant an increase in the proportion of skilled labor relative to unskilled labor when the total supply of labor is I- -- ... -..-...- In cotton textiles 22. 5 per cent of the workers got the flat rate while 66 per cent had their allowance linked to the price index. Only in the case of Jute textiles 95. 5 per cent of the worker had a flat rate allowance. Occupational Wage Surve , Op. Cit. Probably the compression of wage differentials would be more in the case of West Bengal, which is the center of this industry. -96- considered. Turner,for instance,points out that differentials for manual and intellectual skills 1 required by industrial development will be wide. He further points out that some of these differentials .can be attributed to wider skill differential, and that racial differentials are important in some countries. At the same time, Turner also finds some non-skill differential to exist in apparent defiance of labor market circumstances. "It is hard to see," he Sass, ”why differentials should go on widening - as it probably has in both Africa and Latin America - in face of substantial and growing unemployment."2 There has even been some doubts cast on the inevitability of a particular natural history of wages in the course of deveIOpment. For instance, a tendency toward a 'wage drift' is suggested in the light of wage rates being individually ro-negotiated.3 1H. A. Turner, w e Trends w e Policies and Collective Bargainigg: The Problems for finder- deve§opchountr1§s, University of 5ambridge, 1965, p. 0 2Ibid. p. 16-17. 3 Subbiah Kannappan, A Study of Wage Differentials in India, unpublishEd'manuscript,—Univcrsity 0?- Washington, July 13 1960, pp. 33-3". See also Walter Galenson (eds Labor in Developingkficonomics, Institute of IndustriaI”Relations, University of California, Berkeley, 1963. (Galenson says that the essays in this volume that deal with the theory of a narrowing of wage differentials as deve 0pment M 4‘ proceeds, raise considerable doubt on the; alidity,’€*" "Tf‘ p. 8.) -97- The above observations call for a detailed in- vestigation of the types of skills used in each industry. This should be followed by a more detailed assessment of the changes in the requirement of different types of skills in the different industries, over an ex- tended period of time. lThe effect of these changes on the trend in wage differentials between the different skills has to be studied. A search in this direction in the case of India revealed that at the present time, the undertaking of such a study is badly handicapped by the lack of data on skill requirement by industries. It is further hindered by a lack of data on wages by occupations (skills). Let us pursue this point at some length. The first wage Census after independence was taken in July, 1958, and completed in August, 1959.1 It is the main source of information on occupational wages. Meanwhile, in 1956 the Scientific and Technical Man- power and Perspective Division of the Planning Com- mission had made a study on the skill coefficients in different industries.2 But, neither of these reports help us in estimating the changegrthe requirement of skilled labor in different industries. lOccupational wage Survey, Op. Cit. Government of India, Planning Commission, Scientific and Technical Manpower and Perspective Division, Occupational Pattern in Manufacturing Industries,'lndia, 1956, New Delhi, #195§. 99.; There is some reason to :Igpwgt that there has been a decline in the prOportiun of unskilled labor relative to skilled labor in some of the industries. The changes in the preportion of unskilled and skilled workers, i U in manufacturing in- dustries in the priva?%tg%€¥o%9é§e anlowno lh very general terms in a study put out by the Directorate General of Employment and Training.1 Tt ShOWS that at least in nine industries (food and kindled products; textile mill products; chemicals; rubber, petroleum and coal: stone, clay and glasses; Basic Metal products; machinery and electrical equipment; transport equip- ment; and others) the overall position with respect to the change in the proportion of unskilled to skilled labor revealed that between 1956 and 1961 the pro- portion of unskilled labor declined from “7% to 2U%, wheras that of skilled labor rose from “2% to 63%. ”This study attempts to make a comparison between the earlier one referred to above by the Planning Com- mission in 1956, and its own finding in 1961 with respect to the changes. As emphasized at the beginning of this para, we can only suspect that there has been a change in the proportion of skilled workers relative to unskilled workers. But, in part, this change is —. _. ...—..— India, Ministry of Labor and Employment, Directorate General of Employment and Training, Occupational Pattern in Private Sector, India, 1961 (mimeo- ( graphed), No date, no place. clearly spurious. This is because the study itself points out "Certain occupations which have been classified as unskilled in the Planning Commission's Study in 1956 have presumably been classified as un- skilled or semi-skilled and included in the category of craftsmen and production process workers in the present study. This may be one cf the reasons for variations in the proportion of craftsmen and production process workers in manufacturinc industries in 1956 and 1961."1 Unfortunately, the study has not presented any data on wages which would have enabled us to make an assessment of the changes flmctfiwfnupose of comparing it with our own finding as to the widening of wage dif- ferentials. If skilled labor has increased, its ‘ effect would be felt on both the level of wages as well as on its differential between the skilled and the unskilled. In the light of this, the following explanation for the widening of wage differentials can be offered. In the context of a planned growth of industrialization’skilled labor would be in short supply, and wages would rise in industries requiring a higher preportion of skilled labor. In our study, iron and ilbid. p. 1U. The data on skilled workers whose change is discussed here relate to craftsmen and production process workers,as next to un- skilled labor they constituted the largest element of the labor force in the figures furnished in the report. '3 .\§ steel and sewing machine have been the high wage in- dustries compared to rice milling which has throughout been the lowest wage industry. The proportion of skilled labor is likely to be higher in iron and steel and sewing machine compared to‘that in rice milling. The differences in their wages may be attributed to the differences in the skill composition of their labor force. ...“ fit Variation in Wages a d @pigyment / The Organization for Economic Co—Operation and DeveIOpment studied the relation between changes in wage differentials and the pattern of employment. The economic rationale of their study of workers in manufacturing industries was to see whether there was an association between changes in the number of blue collar workers employed and changes in their relative earnings. In stating the rationale of the relationship, the group of independent eXperts who made the study for the O.E.C.D. point out: "The fact that a sectors variation in employ- ment is typically much greater than its varia— tion in earnings increase may be felt to carry the implication that changes in relative earnings have not been an important cause of changes in the structure of employment. But it is also consistent with the interpre— tation that the employment structure is highly sensitive to changes in relative wages (i.e. that intersectoral elasticities of substitution are significantly greater, in .mi’ aLSulute value, than unity}. 1r; veneral, the Expert Group inclines tow rxs 1hr waxe~ insensitive interpret tation of CPHV.C data, because it helieVes hat it is more strongly supported by additional eviience than the wage-sensitive hypothesis.” .49 above passage raises two questions. is change in earnings 3 cause for change in employment? It is said that a greater variation in employment than tearnirmwsrnay inujly a rmurative rnis wer tc> t. ‘liois employment highly sensitive to wage changes? hrre tne group leans towards the wage—insensitive in- terrreta tionjthough a greater variation in employment may be consistent with the wage~sensitive interpre- tation. This leaning is in part due to their finding that changes in earnings are influenced by other variables besides labor requirement(employment change) sutn as, rrofit rates and charges in profit, degree of own-,wrntration, rate of production growth, the. ratio of labor to total cost, and employment eXperience in 1 previans period. The coefficient of variation with respect to um; >lr:v:ir.-:nt is shown to be greater than earnings in ‘ Organization fir Economic ’«— reration ar:d Develop- ment, E§“:§,3nd Lab ugmflii. ity, Paris, July, 1905, p. ES. The ‘cefficient of variation is calculated fur wages and employrent for (iifierent time spans of l, 3, 5, and over 7 years for North America and Furope serarately hv 'Qeneral Industrial breakdowns'. 'Iniustry turtwikdowns with restricted cvverage' for special occupational groups for Germany and Canada and the Canadian general breakdowns for Montreal and Toronto is also given. See p. U". 8"» the O.E.C.D. study. The group points out: "Whatever the period length considered, the deviation of an industry's (or region's or occupation's) employment change from the average change in employment over the relevant group is usually not less than three times the deviation of that industry's rate of wage advance from the general rate of wage increase; and in certain countries and periods, the factor has been as much as 10 to 1." They have taken the range of variability in wages and employment. In the case of 'industry breakdowns with restricted coverage', the coefficient has ranged from O. 5 to l. O for wages, whereas for employment from 1. 5 to 2.8. They also present certain percentage changes in employment and wages. To quote them: "In the United States, between 1953 and 1960, employment in manufacturing in Arizona rose by 77 per cent and in Michigan it fell by 18 per cent, but the annual rates of increase in average earnings were respectively U. l and U. 0 per cent. In United Kingdom manufacturing, employment in radio manufacture rose by 12, per cent in 7 years (earnings increase:- 5. 7 per cent per annum); in textile machinery it fell by 20 per cent (earnings increase:- 5. per cent)." Thus, if the variation in employment is greater than that of wages, they were inclined to support the view that the wage-insensitive hypothesis is probably correct. Depending on whether one considers sectoral, in— dustrial or geographical variations in employment and wages, and the period for which the coefficient of 1 Ibid. (Ibid., p. us. 4? Wk (”N H \D -103- variation is calculated one could arrive at differenct results. Wummummm leiwoertetaenlee-emotetiottc te-eolooo chum-no an ‘IIDOHOId~IIpioyIOOQ,‘Cho-ooofifiieiont‘te-prooootod ”wawwammj figelying on the method of using the coefficient of variation as a statistic to interpret the changes in wages and employment, the coefficient was calculated for the 26 industries considered in this study. We will consider both the range of variability as well as the magnitude of the coefficient, 1e., is one coefficient much greater or only somewhat greater or less than the other coefficient of variation. Table 21. Wages and Employment Variability HO—‘fi—‘F-‘F—‘H mmszHO‘omNmmzwmr—J . O C . About Iverage Rate offlChange - 26 Industries 195011960 1 2 3 a Money Wages Coefficient of Employment Coefficient Vial-13% {on of Variatiai. . Starch 37.80 1. Bicycles “9.26 . Electric Fans 32.1” 2. Paper N1.29 . Sewing Machines 28.85 3. Sewing Machines 29.76 Biscuit Making- 19.63 H. Electnic Fans 27.“? Chemicals 19.20 5. Plywood 26.8h . Sugar 18.96 6. Rice Milling 26.59 . Soap 17.97 7. Starch 29.24 Glass 16.92 8. Vegetable Oil 23.26 . Tanning 16.12 9. Electric Lamps 20.76 Paints & Varnish 15.13 10. Ceramics 18.79 . Vegetable Oil 1h.20 11. Cement 18.66 . Rice Milling 13.93 12. Chemicals 18.h6 . Cement 12.91 13. Biscuit Making 15.70 . Plywood 12.89 19. Glass 15.h9 . Electric Lamps 12.10 15. Matches 1h.00 . Matches 11.66 16. Wheat Flour 12.h1 . Bicycles 11.29 17. Fruits and Veg. Canning 11.51 . Iron and Steel 11.02 18. Woollen Textiles 11.06 . Wheat Flour 10.8“ 19. Soap 10.03 Source: India of (01+ ' 1 2 3 ‘5 20. Distilleries 10.37 20. Tanning 21. Jute Textiles 10.02 21. Jute Textiles 22. Ceramics 9.92 22. Sugar 23. Cotton Textiles 9.0M 23. Paints & Varnish 2b. Paper 9.01 2“. Iron and Steel 25. Fruits & Veg. Canning 8.22 25. Cotton Textiles 26. Woollen Textiles 6.09 26. Distilleries All India 91.08 All India wmmmmm (Re. uhlic), Direct\rqte of Inlurtria Statistics, Report on the Census .‘lrnlf'lCI. :zres and the r‘mnuql SLAP‘WE 3“ Inizs’tries. 1 'Joeff. of Variation = G" where r- is the stade deviation and x )1. is the arithnetic mean. The data reveals a fairly wide range in the value of the coefficients. For money wages the range is from 6. O9 to 37. 80, while for employment it is still higher from 5. 52 to U9. 26. The overall coefficients for all the 26 industries shoh/ that the variation in employment is greater than the variation in wages. This means that changes in wages have not played an important part in influencing the changes in employment for the manufacturing industries as a whole. The above conclusion is subject to an important qualification in the light of two types of findings relating to high wage and low wage industries which may be considered. Earlier it was shown that Sewing Machines lFor calculating the coefficient of variation average annual eanme?5 and employment figures have been ' used. as , . The calculations are not based on rag indeXes (f earnings and employment, ; as it was felt that the true base year figure would not be reflected in the variation in this case. and Electric Fans were the high wage industries and Ceramics and Rice Milling were the low wage ones in absolute change in wages. The magnitude of the coefficient of variation in Sewing Machines for both employment and wages seem to be very close, with the former slightly higher. But, for Electric Fans the coefficient is higher for wages. For the low wage industries, we see that the magnitude of variation in Ceramics and Rice Milling is higher for employment. In percentage terms, Starch and Electric Fans. were the high wage industries and Ceramics and Fruits and Vegetable Canning were the low wage ones. The magnitude of variation in wages is higher in Starch and Electric Fans compared to that of employment. On the other hand, for Ceramics and Fruits and Vegetable Canning, the coefficient of variation is higher for employment, compared to the coefficient for wages. Thus, in considering the magnitude of variation for the low wage industries there has been a greater variation in employment, whether we consider absolute or percentage change in wages. For the high wage in- dustries, if we take the absolute change in wages, Sewing Machines shows a slightly higher coefficient for employment. But when, we consider percentage changes, th these ind at the variation in wages in?0 is 81 ea gher than the variation in employment. It appears that changes in wages have played an important part in influencing changes in employment in the high wage industries, "bargag such an influence is not apparent in the case of low wage industries. m This chsptor focusssd on tho intor-industrisl wngs structure. A distinction between high wngs snd low wsgs industries wns nods end the wngn diffsrentinl bntwoon the two wns considersd in terns of both sbsoluts snd porcentsgs chsngos. we first tried to see whether industries hsd sltsrnd their rsnhings in the wngs hoirsrchy. This showed thst very little chsngss in the routing hsd token plscs snd thst the highest snd the lowest wngs industry hsd prscticslly rstsinsd their rsnks throughout the period. He then considered the trend in wngo diffsrontisl by toting into- sccount the concsndtnnt chsngos in the gcncrsl level of wnpns. m. 1.4 to the conclusion thst tugs differentisls hnd widened both in sbsoluts snd percentsgs teens for the whole period. By wny of onplsnstion,it wss subnittod thst chsngss in the such structure in s country like Indis, undergoing an sccolsrstod tonpo of industrislisstion, would be influenced by the sons see of fsctors which influence wnges in noro sdvsncod countries. These were product nsrkst conditions, cspitsl intensity, profitsbility, governncnt intervention in the regulstion of wsgss, snd chsngss in the skill conposition of tho lsbor force. those fsctors were in turn considered so swplsnstions for the observed trend in wsgo diffsrontisls. rsvornbls product nnrkst conditions were cited for the difference in wngss between the high wngo and the low wngo industries, end for the widening of such differences. The widening of differentisl between the high wngo and low wngs industries csnnot be sttributsd to sny definite rolstionship between fixed cspitsl per worker (cspitsl intensity) snd sverngs snnusl ssrnings. The widening of diffsrontisl could be sttributod to rslstivs .to'l- profitebility betue. the inhstries. ht noro infer-stun - pundits reletive to individuel nits is required to further onbetentiete I this point. With respect to gamer—net inteluntion in the re’letien of ups in indie, it wns shown thet geverusnt policy on differentiels .hssised nsrrwing of differentiels on equity grounds. end the nintunnn of it to provide incentive for better porter-ed en thopnrtef leboe. Intnnettenptheebeennsde tosnforses policy on differsntiels. besed on job evolution end by evolving e stenhrd oesnpstie-l clessificetien. in the fees of this the influaee of W policy on diffsrentiels hes been neutrsl. It hes eentribnted neither to its nsrreving nor to its widening. It one else shown thet werhers who received e flet rete on deernese ellownnee fiich would nonpreso difforontiels eenetituted e lower porcentegs. then theeswhoroceiveditesliehedtethoconsusepriee inden. in view of this it no doubtnt whether this hed eotnelly led to oqrsssion of wngs differentiels. In considering the differentisls in wsgeo. it wns suggested thst ages my hove inereoood in inthnstries requiring e higher proportion of skilled lebor. Due to inedequete trsining fscilitdes, there nifit be e m.- u «and lshon. reletive on u. ...... an a: result in en ineroese in wnges in industries like iron end steel end sowing nscbins requiriq e higher proportion of shined lebn es epinst ties nilling end vogoteble oil where such requir-nnt in nelstively lees. Iinelly en ettqt no nshe to find out whether own in wnges plsyed en ieportent role in influencing chenges in ”lop-on. For this the verietien'in nope end qloynont wns eonsidered. It ,d' wee found thet chenges in weges have played an importent role in influencing the chenges in employment in the high wage industries compared to the low wnge ones. \-4 Andhra 1,“. s 5 am Bihar -:,’j- CHAPTER 4 Inter-State Wage Structure Let us now consider the inter-state wage structure and examine the nature of its wage differential. It was pointed out that the overall increase in wages with respect to the states for the period l950-l960 was 59 per cent. In this, Bihar showed the largest increase in absolute, and Andhra the highest increase in percentage terms. Assam showed a decline in money wages while the leading states of Bombay and West Bengal showed a lower percentage increase in wages than several other states. We will investigate these changes in greater detail and examine the nature of the wage differential by taking the aggregate of the manu- facturing industries in each state. Ranking of Average Annual_Mggpy Wages by States A year by year ranking of the average annual money wages by states is shown below. Number 1 refers to the lowest rank and number 12 to the highest rank and hence the numbers assigned in an ascending order. gable Lbr State Ranking of Average Annual Money_Wages for Each Year _—.—.--.—-o-. 1950 - 1960 50 51 52 53 5“ 52_ 56 57 58 59' 60 1 1 l 1 1 l 1 1 1 2 2 9 6 8 8 7 8 8 8 8 l 1 10 12 12 10 11 12 12 11 11 12 12 5o 51 52 53 5u___55“.5g___57 58 59 ) Bombay 12 1o 11 12 12 11 11 12 12 10 Delhi 11 11 10 11 10 1o 10 10 10 11 Madhya Pradesh 5 u u M 5 5 u 7 7 9 7 Madras 7 6 7 8 6 7 6 9 Orissa 2 2 2 2 2 2 2 2 2 3 3 Punjab u 5 5 6 6 7 6 5 5 7 5 Rajasthan 3 3 3 3 3 3 3 3 3 5 u Uttar Pradesh 6 9 7 5 U A 5 u U U 6 West Bengal 8 8 9 9 9 9 9 9 9 8 8 Source India (Republic) Directorate of Industrial Statistics, Report on the Census of Manufactures, and Annual_§grvey of Industries Once again we see that there has been very little change " in the ranks. Four out of twelve states show a change of only one rank. The position of the high paying and the low paying states undergo very little change. This is clearly seen in the ranking of Bombay and Bihar as the high paying states and Andhra and Orissa as the low paying states. The rank correlation coefficients for each year turned 01t to be as follows. Table 36' Spearman Rank Order Correlation for Average Annual Wages by_States 1950 - 1960 Year Rank Correlation Coefficients 1951 .90209 3 1952 -.9580u ' 1953 .97202 195“ .9UUOS .111. Year Rank Correlation Coefficients 1955 .92307 1956 .951ou 1957 .9510u 1958 .9510U 1959 .62237 1960 .71328 Source Computed from, India (Republic), Directorate of Industrial Statistics, Report on the Census of Manufactures, and Annual Surveyflof Industries lots The renking of the svorags ennuel nonoy wegos for the 12 stetos for eech of the yoers 1951 to 1960 beve been correlsted with ' the renking for 1950. ’fhs fornnlc used is es follows: -2- ,° l«— if! d” . NCNl—I) wha's P is the Sneenun Rank Correletion Coeffiodafi ‘ is the difference in rank between mm it-s l is the number of pairs of cheer-rations The high values of the coefficient show the slow change in the ranking by states. Up to 1958 the values are fairly close to .95. But there is a sharp dis- continuity in the value for 1959 and 1960, which may be due to the difference in statistical coverage. For the period up to 1958, a comparison of the coefficient with the actual ranking in the prkious table shows, that there has been very little change in the ranks in each state. For 1959, Assam alone shows a very sharp change in its rank while the change in the case of the other states appearsto be rather small. Nevertheless, in interpreting the change for the whole period by com- paring the base and terminal year, the discontinuity in the data after 1958 must be borne in mind. Next we will analyse the trend in wage differntials in absolute as well as percentage terms. The following 1 table presents the difference in wages between the highest and the lowest paying state for each year from 1950-1960. A n t Table - Wage Differential Between the Highest and the Lowest Paying State 1950 - 1966 Year Highest - Lowest Lowest 5 10d Bio highest 1950 734 36.68 1951 924 32.96 1952 981 31.04 1953 952 34.06 195“ 931 35.37 1955 929 35.99 1956 793 44.26 1957 830 44.41 1958 823 45.56 1959 1215 38.83 1960 1407 37.08 Sggggg Computed from India (Republic) Directorate of Industrial Statistics, Report on the Census of Manufactures, and Annual Survey of Industries. For the period 1950-1958 there has been a widening of absolute differentials. The sharp rise for 1959 and 1960 in this column must once again be due to the statis- tical discrepancy in the sources. Percentage differentials -713. appear to narrow up to 1955 and thereafter widens up to 1958, before narrowing. To some extent, the dis- continuity in the data for 1959 and 1960, may be the reason for the rather sharp change in the values. However, we are not stopping at this point to draw conclusions on the nature of the change in the differential, for we will also consider the concommittant changes in the general level of wages. For this, the overall average annual wages of all the 12 states is taken as a point of reference and the change in this overall average is re- lated to the change in the average annual wage of the highest and the lowest paying state by taking their deviations for each year. Table 27 Deviation in Average Annual Wage of the Highest Paying State and the Lowest Paying State from the Average of all States - 195611960 ’3 CHI: b0 wv C: (Ow 60 H A a) 0 $140) 5'4 c9808 C1..r\ Iv HS—a <43 (3" IV) Hid U) 430) X U) (14!!) (DE 43?; N 17.5 :38, 32mm :m .35 a... H . . “'1 333 gig 3E5 -° 5.; at; as; .98.... t: 33 9.3 922.2; 8 E73 E5; #388 6a: am an Reno 1 2 3 4 5 6 7 8 1950 1159 226 24.29 933 426 507 54.41 1951. 1379 336 32.29 1043 455 588 56.39 1952 1423 331 30.40 1092 443 649 59.52 1953 1444 317 28.19 1127 493 634 56.33 1440 306 27.07 1134 510 62» 55.05 1954 ‘9 was Bombay, and for the remaining years, Bihar. paying state from 1950 to 1958 was Andhra, and for the re- maining years, Assam. Part of the statistical discrepancy caused by the difference in sources for 1959 and 1960 can be reduced by analyzing the period up to 1958 in the above table. If we exclude the last two years, we get a narrowing, rather than a widening or maintenance of percentage differentials between the highest and the lowest paying states. But, owing to the considerable fluctuations in Columns 4, 5, and 8, there is a clear case for in- vestigating the rates of growth in money wages in each state. First, we can rank the average annual money wages in each state for the base year, 1950. Then the percentage change in wages between 1950 and 1955 can be ranked. Next the similar change between 1955 and 1960 can be ranked. Finally, the change in rank for the whole period 1950-1960 can be presented. The table .below shogfi the rankings. The lowest 1 2 3 4 5 6 7 8 1955 1451 332 29.73 1119 523 596 53.30 1956 1423 262 22.63 1161 631 530 45.72 1957 1494 228 -18.02 1266 664 602 47.58 1958 1511 274 22.23 1237 689 548 44.31 1959 1836 603 48.98 1233 622 611 49.60 1960 2236 810 56.84 1426 830 596 41.83 Sggggg Computed from, India (Republic), Directorate of Industrial Statistics, Rgport on the Census of Manufactures, and gggggl Survgy of Industries. . Note The highest paying state for 1950, 1953, 1954, 1957 and 1958 -1:5 ’- Table 2? State Ranking of Average Annual Moneinages for 1950, QUUPCG: statistics, deport tux‘V'ag 31 incis Li'led. :I‘. the 461151.113 q). UQmputed lrom indie (Hepuollc),blrectorate of industrial 9 With Percentage Change in Ranks for 1955 and 1960— 0 L0 L61: 9 L6: 8 o .8 H 9 r—1 A \o ' m ' m L ' m n ' c6 0 g 3%“ g 8 &r\ g E; t; S m m - o H «u - 0 r4 H 3 :33m :3 m x A u: <1) x O: 0) >4 Q), C m CV 330301 CV 2:05 g‘Nr-i m dCC ressed lll industries which primarily employ less skilled labor. This fact needs further investigation along; the lines of the. study undertaken here. A distinctiin can he made between industries which deal with food And hindered products like rice milling, vegetable oils and fruits ind vegetable canning, where sztly semi-skilled JHd unskilled workers are employed, and the industries like iron and steel, sewing nkudiines (uni electrit: tans, (unploying:xnzlstivelyrxnore sfllatx3d tr) time (Xflhhlh. Data for edible hydrwunun1tcd oils were run.£available for the luiSU ywuir arulliunce (duty Ln\uld lhTC lu' incliuhul. Tlflr; led tt>;1 total.<)f l. thwwvrnmcru l)f India, (kuitral Statistiruzl‘)rganizatitni, Rnport xni the mmfiiiigianufactun s, New hulhi. Vols. \'_ tfl'XIll. 3. (Q'JVUI'IIIJICUL of India, (_Iuntral Statistical Uruanization, Annual Survey .‘t ludustrigg, 1959 a 190“. [ho Annhal SUFVUE of Industries 1959 was Uni available in the United States. A microfilm oi the Survey was ob- tainvd through the International h4cumintation Centre at New Delhi. 2. GMVrrmant of India, Labour Bureau, The Indian Labour Journal, Vol II, TLi. T, LKKL. 1901, in). 1290-91. quvrumvnt of India, Lahour Buruau, Indian Labour Journal, Vol II, Jutn l‘Jhl, M). 6, p. 57.7. Uthrnmvnt of India, Central Statistical Organization, Statistical‘éh- strart oi fhc Indianfihnion, [”37-53, 9. 103 and 1961, p. 241. my... 26 industries as stated above. The choice of the twelve states was dic- tated by the fact that only in the case of these, both money wage data and the consumer price index numbers were available in the United States, for deriving the indexes of real wages. The Choice of the Base Year. There were two reasons for choosing 1950 as the base year. The period 1950 - 1960 featured an accelerated rate of industrial expansion and was particularly unique for analyzing wage trends. Specifically, in the field of industrial deveIOpment this period was char- acterized by (a) rapid expansion of firms in the existing industries, (b) growth of new firms in the existing industries and (c) growth of new in- dustries. Secondly a change in the industrial classification was intro- duced in 1950 for the reporting of data on wages and employment thereafter. Methods followed in constructing the indexes. The idea was to present an industry and a state series of indexes for the purpose of comparing the trends in the two cases. These were done as follows. Index of MongyVWaggs. The amount of annual salaries, wages and benefits reported as received by workers was divided by the average number of work- ers reported as employed to arrive at the average annual earnings. The resulting figure for each year was indexed with base 1950. Index of Real Wages. Working class consumer price index numbers relating to different centers in twelve states were available as follows: Andhra, Assam, Bihar, Bombav, Delhi, Madhya Pradesh, Madras, Orissa, Punjab, Ra- jasthan, Uttar Pradesh and West Bengal. For each state, the centerwise indexes were combined using the population of the center, as reported in the 1931 Census, as weights to arrive at a weighted index for the state as a whole. This was used to deflate the index of money wages for deriv- ing the index of real wages for each state. In the case of the industries, the above procedure could not be followed for the following reason. r/é‘o " The 0%1 and ASL in reporting the data fran year to year, bracket two or more states for any one industry. A detailed examination of them revealed that in the case of all industries except cotton textile, chemicals, and iron and steel, the practiCe of bracketing had class of 1mm e \' C on SUIHC r wages to Jurive the index EVUjlnghfni. Tlni stutrcews re;u>rt index of for each [at tory , {11(7to‘jy', days and lkir all x‘toriiws in tiu‘ statt3 or a',='i» Lt' i 3 tellt-i 8:5 tlie idindwiér (‘1. d by, r s,evtively. This figure [li”._;i_lil*“i (xrri ta]; 13): VEiEESLE' i' .1 «l rl‘;'..‘L‘t‘.“l was diviili (l hv the. H .l'i .1.” range at fixed capital per year as i llows. The on days is co di ’ uiiig l)y tht‘ ninnb(‘r (9i Wuif. T THE 1'11“ tlIC‘ ()-.'i'rall [11¢it‘-Lt's. ...—4‘- - 7-. Tull dustrius as hell as in the average ~n which manufacturing (5:31“ ml of real days. perswn was indexed Th“ x1i11ni)té1‘ WHL‘kHr . Tllt“ t he rthArtwyd Iv: lu~r .yf Vfldrht‘rj .=n,d(iyiwl. Adinstm placei of with reapect i by theSe an; fir [959 and l'flini). lr. res; Cht Annual Survey vi t'kl. to coverage of 0 Coverage of factories. differences in that 1939 industriis order. lndnst r A (vs . Vii) 'nz111 lllunl)€‘f amount of overall States, fhv dial frequently been adopted. Therefore all India working price index numbers had to be used to deflate the index wages tor the industry series. the numhrr oi persons employed iii wtnnuars (an)loyed ix/ eacii operations were carried on in the Juted by adding the number of persons employed on all these lhese averages are aggregated ilidiist rx' ;ls (lie ccise» ”nay ht}, alld tile 5 thlUVL’tl in the state or in- w'itl‘ have 193‘). annual fixed capital em- of workexs employed to arrive This iiwure was indexed with index 5 of money wages, ', emilnymvnt and iiaod vapital per Worker in the case of the in- ... . . '~\A.‘._t,.‘ have be--.i weighted by the Cinsus vi Mann12vtures was re- This resulted in ditferences and (ill {Brent es With hlfllt‘s; dral with the adjustmints posed LEdUItEI §e£ie§. The Annual Survey for 1959 contained 250 industries, and for 1969, 227 industries, compared to the 29 industries reported in the Census. The industries in the Survey included those covered by the Census, with certain changes in their classification. From the Survey, the 26 in- dustries which corresponded with those of the Census were alone taken. In the case of glass and glassware, iron and steel - smelting (including fur- nace operations), rolling and re-rolling, ceramics, paper and paper board (including straw board), and chemicals (including drugs and pharmaceuti- cals), the Survey introduced certain changes in their classification. For glass, the data related to sheet and plate glass, hollow ware, miscellane- ous ware, laboratory ware and Optical glass. For iron and steel, the clas- sifications were iron and steel, castings and forging, structurals, fer- ro alloys, and iron and steel pipes. For ceramics, it related to china ware and pottery, sanitary ware, white ware, and insulators. The classifi- cations for paper were writing paper, printing and wrapping, paper board and straw board, hard board including fibre board and chip board. In the case of chemicals there were organic and inorganic chemicals, perfumes, cosmetics and other toilet preparations, and drug and pharmaceuticals. These classifications were retained for 1959 and 1960. The data in both the Census and the Survey were reported by state breakdowns. The Survey reported data for Gujarat, Tirupura, Jammu and Kashmir and Andaman and Nicobar Islands in addition to the states report- ed by the Census. Hence for each of the 26 industries the data relating to the above states were omitted. State Series. Data relating to the 26 industries were first broken down by 12 states for 1959 and 1960. In each industry those states included up to 1958 were included. The data for each state was then arrived at by adding the data from the industries for that state. -161- Overall Employment for States. The exclusion of general engineering, alumi- num, brass and copper for 1959 and 1960 would impart a downward bias in the case of the state total since the aggregate figures for the states up to 1958 included these industries. The 1958 employment in these industries for the twelve states was included for these two years together with the employment in basic chemicals, paper for packaging and breweries for which no state breakdowns were reported in the Annual Survey. Qiiference in the Coverage of Factories. Besides the above adjustment, another type of adjustment was necessitated owing to a difference in the coverage of factories between the Census and the Survey. The Census cov- ered registered factories with power in industries employing 20 or more workers. The Survey covered registered factories with power employing 50 or more and factories withoug power employing 100 or more. Thus it left out the smaller factories employing between 20 and a9 workers and using power.1 It was decided to appraise the significance in terms of employ- ment, capital, output and value added of the factories left out by the Survey. Since there was no source in the United States that could reveal this, a letter was addressed to the Indian Statistical Institute in Cal- cutta which shared the responsibility for the tabulation and publication of the results of the Survey with the Central Statistical Organization at Delhi. It was learned2 that factories covered by the Survey cameunder the Census sector and those left out by it under the Sample sector of the A.S.I. 1. Data for non powered factories are not separately provided by the Sur- vey. It is assumed that there would not be many factories employing 100 or more workers without using some kind of power. 2. The information and the table that follows were taken from a paper ob- tained from the Institute. India, Cabinet Secretariat, Department of Statistics, Central Statistical Organisation, (Indian Statistics Wing), Estimate of Total Capital, Employment, and Output in Manufacturing In- dustry_(l959 and 196Q), Paper No. Zh-CDN (3l)/63, Calcutta. (mimeographed). ed? ./45. The former was completely enumerated while the latter was covered on the basis of probability sample. The following table shows the relative position of the two sectors. stlg 33 National Stru t s of its ozpent, Output and Value Added for Census actor Factories,» 959 and 1960 Item nsds Sector Sample Sector 1959 1960 1959 1960 (1) (2) (3) (4) No. of registered factories (Z) 22.9 21.5 77.1 78.5 Productive Capital (1) 85.9 87.8 14.1 12.2 Productive Capital per factory (Rs.'000) 2112.8 2382.9 98.9 87.2 Persons employed (2) ' 80.4 79.7 19.6 20.3 ‘ Employment per factory (No) 349.0 346.0 24.0 23.0 Gross Output (1) 79.1 80.5 20.9 19.5 Output per factory (Rs. '000) 3273.0 3754.5 248.1 240.7 Value added (X) 84.8 84.9 15.2 15.1 ~'Value added per factory (Rs. '000) 922.9 1030.1 47.0 48.1 Source: India (Republic), Cabinet Secretarit, Dept. of Statistics, Central Statistical Organisation, (Industrial Statistics Wing), Estimate of Total Capital, Employment, and Output in Manufacturingilndustry (1959 and 1969), Paper No. 24-CDN (31/63, Calcutta, p.v. The fig- ures furnished above have been assembled from 2 tables, given on the page. The table shows that the national structure of employment, output, industrial capital and value added rests very largely on the significant shares of the Census sector factories. As to the problem of adjusting for the smaller factories, it was felt necessary to do so only for the computa- tion of the employment index. In the case of the other two indexes, Viz. money wages and fixed capital per worker, the omission of the smaller fac- tories, whose share is not very significant any way, would affect both the numerator and the denominator equally. At the same time it will not dis- tort very much the representative character of the indexes. But in the case of employment, the exclusion of the smaller factories would lead to .169- an understatement of the extent of its change for 1959 and 1960. Hence an ~£5 adjustment was made to the index of employment. It may be seen from the T3 1:5 above table that in 1959, the sample sector factories employed 19.6 per cent of the workers. This related to factories employing from 10 to 49 workers. It was assumed that factories employing upwards of 10 workers with power could be taken as coming under the sample sector. On this ba- sis, factories employing upwards of 20 but below 49 workers with power would constitute 14.7 per cent of the employment.1 This percentage was arrived at by apportioning equal shares in terms of percentage covered at intervals of 10. Since the Survey covered 80.4 per cent of employment, the employment in each industry and state for 1959 was multiplied by the ‘A£:I=:fl. 18 (95.1/80.4) and this resulting figure was used to compute the index of employment. For 1960 theAWturned out to be 11.19. Andhra. A.word finally must be mentioned about the indexes relating to Andhra. The state came into existence only in 1953. Until then the Cen- sus of Manufactures reported a combined total for Madras and Andhra. So up to 1953, separate figures were computed for Andhra by dividing the total of the two states proportionately, the ratio used being the average of the ratios for the years 1953 and 1954, which turned out to be almost constant. Limitations of the Data. By way of concluding this note, it must be em- phasized strongly that the data from the Census of Manufactures and the Annual Survey of Industries suffers from the lack of a uniform coverage. The adjustments have been made by taking into account the difficulty of obtaining data otherwise and without doing violence to the representative 1. The paper obtained from the Indian Statistical Institute did not speci- fy the kind of factories covered by probability sample except to say that they were smaller in size. It is assumed therefore that these should employ at least 10 workers or more with power. «:7 character and the accuracy of (ti as p055 iblit . Our Findings must b. -170- lwdvxvs pnnsiructud, in as best a manner viewed blaring in mind this limitation. 03 r a: R . «E “3.3.... San Seaanéa 83 8a A. em a: a: Fe .9 e; made. So £85. an. a o: E. a. N a... P e; , 3.5.... 8:... Seawefi max .n A: .a. n8 n L Ea Pike! «,3; can £85.... 33 9: an RN cc; m2 AK...ne.m. 9m; $3.306.“ RS 8 2. m8 .2 m2 Rkfieefi. me...» 2363...... RE NPH Hr. nrN $4.5. ONE. “wadméflrnm‘. . NNm m2 Nm. ScrewedaWN aOH m3 8 Fe «.3 .3 $193... an..." «2.8%. me 93 NS w : c: NS Sacha; E.“ n a. Ex. 33 mg on. d w «2. sea weeKQi cka 03.2“. .2. ”no” 8.» SH 02 . 03 8a 8358 as.“ so. .96... .r 33 .95.... 9388 mm a ... .. m... oS mm a so... ...KJN 9a.? GK. 3.8.8 83 8a 2,. . a. 6 R e L. w.&..§.fi Sofia fidefifie on. a o: a m 3,; «A. a: «3. 65.5 25.. w 368.? RS d: men N3 .8 m... :56? ..n made .35 .313 Ron n2 e. n :3 «a R 93“ng «8:... max. 5.3 $3 8 re SH 8 8 95.59.» «36 Sanlcm RS «2 RN 5H .. me 3383” find «ko.awa.o~ $3 «.3 :2 n3 3 8 «£68.: an; 3:336... 93 «2 9: n3 8 9 819m; 806 3...:333 Name n2 .5 .3 R «3 «8.85; mama $3.6.» «3 8a 8a 8a 8" 8a «.3323 23.1.. Se.§.e $3 lead. mm o o a. e n a m m. H Kenn: ace Losses soc sun... nous... eaflcco... so Ea 1.5015 7380 19; Sea». a nous. as? 85:8 so .83: so See” so «via so sus: 53.35 A35... .33.» 82.... a...» 8g ... Rod 83 .. a was .85.. Pagans... are... a... E flbuhqahu Crd ..HSWC. firm 0 coma an... H Rafi Rafi ROM 3.. a anmfl RS mama “a; on: know med wad noH an noa and 30a HNH mm mma pm 93 no we” an «S 00H No # OHH no 8H 00.... can a”. nun mag and and mag and odd and um 00H noa no” OHH mm .8." .8." noa «ca cow Sfi _ m m mug-H fila COHEN RM mod 9: Naa 4% H QNH u 3&0: .30. 18%.... 1.395 9.56200 no M605” .3 Hanna uo K03 no H35 so. 5N." :NH nNH Hm... an NOH OOH SM d: ND «3 and ...: n3 «2 com con . o 0.33 d... on as and DNA 0.: 3H." N HH 0 HA QOH OCH OCH m8 and. N3“ mm a mad and“ ONH we." 50H co." m own) 50:01 3.8%.. new .08. a 5636 08686 nwmfififi n8ém6 mfi.am~.n ooc.nocan .mua.4 ohm J... 1.. 93.93... R3536.” 0.3.83 fiifik ofiéfion “Ragga 23512 Rhea}: ohm.nnm.3 083%.: ad.8o.$ «3.25.3 a 3625. G Down? £3.33 flHIWN.m .43th 2H EEQQHBE o3.“ 92} 8n.n oRJ on; . «£6 ck.“ mo...“ . «96 «.8... ma.“ 3......” 5....“ $18 no.8 $5.8 A anamm 833 93.3 Edam Ruin n «...-a 133 n2.ok.3 «2.68.3 4568.8 «3.20% .$~.~S.m~ Shanimm Ems—Sufi 25.3.}: a~u.~§.$ £5.03 .9 $053.3 03.2.3.3 8068.2: Richard o$.n~\‘.~2 0R.fin£fl «843$. nwo.mn...2 man .m. fink. §.§.R $063.8 « 92.5.3 N Handgun .85.... gflmm? 6.... 5.398. -glmg: 3.x..m9” 3 5.8.. up. n59.“ www.mfifi "\ mm." a: w 835.08.. 9838.53 08.3.32. o8.8n.n~m 08.58.58 85.85.92: 8189.83 81%.“.8fl 0863.3 0.8.8” 8a 3125. 5 33% ..N 8n.o8.o§ 5.8%“? . . 8o.kn.§ ‘ gleam? Rnahnaaflo 080’? k.“ can. 85. «on cod-gnaw? Rm.8m..:n finds; got 2: ‘ cu gamete Rad 333...: n9 .3 9363.“ «8.3 368.02 n2 3” 3.83% nRJn 8~.§~.~£ .2 RH 2.38.6. :3“ 0.5.835 8.. >3 . 98.18:: Rofia 8~.R¢.nn 9. . 2: “838..“ 5.3 36863 a: a 82....“ 53.? 568.8. on." «3 Snififl «“an 3n.fi1ncn 8n .3 Raéuuon 25.? 935.3 8 8 and; .513 §.§.R 8a 8» 5.08.8 $0.9. eqfifik an n3 $15.6». £19 fiasco.“ n3 9: «3.933 . fixw Rn.8o.nn o2 «fl «8% ‘SJN 63.8.3 .3 5 Sesame. n8.“ 8n.n~a.on a n3 $0.33.." 31h 83.58.? a and 58.813 main 89.5.9 3 ofl wagon 3“...“ 935%.... 2A .3 Sagan 83" «8.13% 8." 3.. $695 Nu.“ 3:3an «on 3 «3.58.3 8......" afink...‘ 8a 8“ 98.5.3 «35 $.23... ‘ l" a}; ‘9‘. "'f "I'O'. !‘. §§§*§aa§aaafi 0.) SA and 3.— «dd 8H 8d N“ .. .0 L i 1 x ‘4 u o I no." d..." we and 8H 00.— 8. R38... 5 fic.~.?.3 a: 8n.fiw.§ Ru 8.30;. 8. figs. .9 «8.8.... B Re.ono.m and $.83 on 5“...an 8a 5.98.: .8. 563.3. “m 3.80% 8 8o «nu 98.318. an. 252...... MM Rdfiéu a gamma .3 Reg” 8. 588.8. 8. 839.68 n . 2.} 6a... in?“ «8.3 813 am“... an... .3... 5.2 5.8 n8 .m 318... 58.2. .6...” «Rd? 93.2 can... 853 3152 anion. Ruin” .3536. 3284...“ R~.n8.fln 8n.8~...8u 58.9%.» mafiannflo 304.233 Sainofim 8.68.3 2.69.5 audunfii 33%.... 3.33.5. 80.5.8: .5583: d: 38$” .31.: 9863.8 82 uRiiR $3 8n.<.o.8a RS 353% .3 RS 8n.2n.oS 33 “EJQR RS RfiRQk $2 «3.3535 R3 canaim «“3 -o.8~..9 “3 «238.3 33 5.52m 8053.3 83 8.10362 $3 8~$~13H RS gowwmédn Run o8.818~ 33 «8.53» RS salami _ $3 Sninfio R3 nanénaé $3 $32.? do“ 2533.3 82 a u H a Ne 00H 00H 30H no R no 02 QOH mo." meg n3 30H Hod noa OOH N: n3 ma.” mg no." mo.— m3 00H 829086188 83‘83‘1 Sm mom Qua 03. EA H9 mm? m: N CA on." 00H 0mm 0% RH o3. 43 m? 00H 03 SH mg 03 on.” 2.: «EH 5.: .8." 8H 0.: no." 3.: 00H mm — i- a cmH 3% mma «Na. NNH 9: SH Nod OCH RH mm H 9.." ma." mOH 0.2“ d: m: an 00H m .. OWN. MFG 03w...“ can .5933 oao.ocu.ww: 9263.8: 08.8m63 80.80.an oHo.-o.:Ha 81mm ~62 oaa.nam.oaz CNN .NHN. ... 8.. m 085%.an 03.8053 oam.ooa.oma 08.8063 enigma? osu.mom.maa §.8m.§ ooa.nmo.~na oR.m~nJS OSJRJD oomfiR .mma 8N.m8.fia dcuconvv mmgH mmghqpm ehm.£,.m a-.~mm n :6? ncm.m~a ~i.n§ mi. 69 81c? omm.ncn mafia? men. .09 5.3: Ema? $62 32m? $~.$a .333 Sin? «EH63 3253 was? coe.ma~ «V8.9: 89:25): 83 . 00053.33." ,nma 81m3£33 $3 oodnafiamfi RS §.$~.RS RS 08.03.98 . RS oRJmnAmm $3 o:o.~nm.aam mama 93.08.?“ «a? 30.3%35 R3 0368.52 $3 .295 BM: 8N.nR.~3 83 amaanfi 33 3563.33 32 855868 R3 2N.R1.3m R3 go.~nm.~on 3‘3 8:.§.3~ $3 03.83me RS 08.58;? ~an onafimflmw Rm." “3.25.8 83 58me 52.5 N a 8 .: .HE ..ouccH 9. I‘ll Elli agar raffleémtfi no “Ilka FEES onufrvspm 2. «HVFéCH uc 17.3 . III} 33.5 -‘ 5...... a 9.2570 .NH grindpba 7:51.. .ccwuru “c, an) 70 H. a... : ,am Thacwu .Ao ... Esrcxva c «FCH ‘ .5; :58 “TEN .c par...» obcu .... .QWMM o» 30H n.7,; «Eu .7.“ ficno u we mefinu 45.75 quiruo BEL ..c mfificco 2: :c 00." m Hit? the? rm me ..8 c H 83% H 8.5 :0H OOH 5 8x19; LCC H5...ch he an. :H Mo KODCH finém ......fiflém Hiafiécu ...... (9.78....3 £25255 «:5 :9 3.. 08.6133 , «8.5:. 85.8?28 SA a? 8w.3~.omna “$.31 . 4. 80.5%83 m3 :9 02.8....oh; in“; 8063.3 93 H9 Swimmfiuca 33.8“: 25.8.ng N2 8.. 08. Exam: $5.80: “US$8.32 m2 omfl oodwmm.ifi Mafimmnnfl 08.38.33 Na Na consumfiafl “$.53 oasomxmnmm ms. 84 05.3233 “369; cooémcémfl a 3 ed 08.33.83 can .53 oodmmm .8R 9: SA 8n.fim.33 madman; 09 58.3qu 02 9: .. W 08:11.89 9&5}; 8M. .... m .anm m m ‘ a m N up: on? $5 3:32.. A93 .73. 5:9... a mmua... as? d. .1... o uc NDMEH .mowunacm 387$. C83...” .8315“? .mgbfiwdflmzw «E E. 3 m cit 8&1?me .25an $me .mgzémq .55.. 52...: .8 59.; 5...,35 m «0939M TTTTTT Winifiu‘xgnuilflfij‘ufl“MW“Wm“ 6