FILE COPY Some Aspects of Wheat and Rice Price Policy in India SWP381 World Bank Staff Working Paper No. 381 April 1980 Prepared by: Raj Krishna and G.S. Raychaudhuri, Consultants South Asia Programs Department Copyright e 1980 The World Bank 1818 H Street, N.W. +1ington, D.C. 20433, U.S.A. _____ iiews and interpretations in this document are those of the autho ! hould not be attributed to the World Bank, to its affiliated FI L iizations, or to any individual acting in their behalf. The views and interpretations in this document are those of the authors and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting in their behalf. WORLD BANK Staff Working Paper No. 381 April 1980 SOME ASPECTS OF WHEAT AND RICE PRICE POLICY IN INDIA This paper analyzes India's pricing and procurement policy for two major foodgrains: wheat and rice. Its main purpose is to provide a vigorous statistical analysis of the factors influencing the fixing of procurement prices, and the influence of the procurement price and other important factors on the volume of procurement. The paper also examines the effects of price movements on the output of wheat and rice, and the impact of zoning on interstate price dispersion. . O ') Prepared by: Rai Krishna and G.S. Raychaudhuri, Consultants South Asia Programs.Department Copyright t 1980 The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. SOME ASPECTS OF WHEAT AND RICE POLICY IN INDIA Table of Contents Page No. 1. INTRODUCTION ...................................... 1 2. PROCUREMENT PRICE, COST OF PRODUCTION, HARVEST PRICE AND WHOLESALE PRICE ..... ................. 2 3. THE EMPIRICAL BASIS OF PROCUREMENT PRICE DETERMINATION ................................... 16 4. PROCUREMENT FUNCTIONS ............................. 20 5. INTERSTATE PRICE DISPERSION AND ZONING .... ........ 28 6. SUPPLY FUNCTION ................................... 34 7. CONCLUSIONS AND POLICY IMPLICATIONS .... ........... 44 APPENDICES ........... .................................. 47 BIBLIOGRAPHY ......... .................................. 61 SOME ASPECTS OF WHEAT AND RICE PRICE POLICY IN INDIA 1/ Raj Krishna G.S. Raychaudhuri 2/ I. INTRODUCTION 1. The main purpose of this paper is a rigorous statistical analysis of: (a) the factors influencing the determination of procurement (gov- ernment purchase) prices; (b) the influence of the procurement price and other important factors, on the volume of procurement; (c) the effects of price movements and other shifter variables on the output of wheat and rice; and (d) the effect of zoning on interstate price dispersion. 2. The paper is organized as follows. In Section 2 the relationship between the procurement price, the cost of production, the harvest price, and the market wholesale price is analyzed in detail. The evidence suggests a strong link between the procurement price and the lagged average wholesale price. This link is formalized in estimated procurement price functions in Section 3. Trends in the procurement-output ratio for wheat and rice are reviewed in Section 4. This review indicates hypotheses about factors deter- mining procurement. The corresponding procurement (volume) functions are presented in Section 4. Section 5 is devoted to zoning. The hypothesis emphasized some years ago (Krishna, 1965) that zoning increases interstate price dispersion is tested with an equation in which a measure of price dis- persion is regressed on a scale variable and dujmmy variables representing the zoning policy every year. 1/ Thanks are due to Meera Garg, Chandra Mazumdar and Tara Nair for competent statistical assistance, and Sartaj Alam Kidwai for computer programming. 2/ Professor of Economics and Reader in Economics of Delhi School of Econo- mics respectively. - 2 - 3. Supply functions are presented in Section 6. They are characterized by two new features. First, the supply response elasticity is estimated directly instead of being approximated by the acreage response elasticity as in earlier studies. Second, the relative price used in the specification is the wholesale price of wheat or rice deflated by an input price index rather than the wholesale price deflated by an index of the prices of competing crops. With the increasing role of purchased inputs, the allocation of land and other inputs is likely to be strongly influenced by variations in input prices as well as output prices. 4. The policy implications of estimated relationships are indicated in each section and summarized in the concluding section. 2. PROCUREMENT PRICE, COST OF PRODUCTION, HARVEST PRICE AND WHOLESALE PRICE 5. The most important quantitative indicator of official policy for a crop is the procurement price 1/ at which the government stands ready to pur- chase grain, subject to the annual procurement targets. This price is fixed for each State on the basis of the recommendations of the Agricultural Prices Commission, but the prices fixed by the States have not always been the same as the prices recommended by the Commission. 6. In Tables IA and IB, we have computed for wheat and rice the mean excess of the prices actually fixed in each State over the prices recommended by the Commission during the period 1964/65 to 1976/77. It is clear that actual State-level prices have usually exceeded the recommended prices. 1/ The "support" price which is lower than the procurement price is not discussed in this paper for it. is a notional price which is suppased to become operational only when the market price sinks much below the pro- curement price. This has happened only rarely. - 3 - The average of the excess in different years has varied between 1 and 9 per- cent for wheat. In 1967/68, the mean excess was as high as 34 percent. Except Gujarat and Madhya Pradesh all the important wheat-producing States have been fixing procurement prices in excess of recommended prices by 6 to 9 percent. For rice, the excess has ranged from 1 to 7 percent in most years. In 1966-67 and 1967-68 it was as high as 16 and 21 percent respectively. In 1966/67, Bihar fixed a procurement price which was 79 percent above the price recommended by the Commission and in 1967/68 Assam fixed a price as high as 46 percent above the price recommended. These two States and Kerala and West Bengal have been prominent in fixing rice procurement prices higher than the recommended prices in some years. A small part of the difference between actual and recommended prices is due to the divergence of paddy-rice conversion rates across States. But the observed differences remain significant even after an allowance is made for conversion differentials. TABLE IA MEAN EXCESS OF THE WHEAT PROCUREMENT PRICE FIXED BY STATE GOVERNMENTS OVER THE PRICE RECOMMENDED BY THE AGRICULTURAL PRICES COMMISSION /a Average of Eleven Years Average of Seven States State Percent Year Percent Bihar 7 1966-67 1 Gujarat 3 1967-68 34 Haryana 9 1968-69 7 Madhya Pradesh 4 1969-70 9 Punjab 6 1970-71 6 Rajasthan 8 1971-72 3 Uttar Pradesh 8 1972-73 6 1973-74 3 1974-75 0 1975-76 1 1976-77 6 /a Calculated from primary data in reports of the Agricultural Prices Commission and issues of the Bulletin on Food Statistics. - 4 - TABLE lB MEAN EXCESS OF THE RICE PROCUREMENT PRICE FIXED BY STATE GOVERNMENTS OVER THE PRICE RECOMMENDED BY THE AGRICULTURAL PRICES COMMISSION /a Average of Twelve Years Average of Twelve States State Percent Year Percent Andhra 3 1964-65 4 Assam 8 1965-66 7 Bihar 12 1966-67 16 Haryana 4 1967-68 21 Kerala 7 1968-69 4 Madhya Pradesh 5 1969-70 1 Maharashtra, 3 1970-71 0 Orissa 4 1971-72 2 Punjab 5 1972-73 2 Tamil Nadu 3 1973-74 12 Uttar Pradesh 4 1974-75 -2 West Bengal 8 1975-76 0 /a Calculated from primary data in reports of the Agricultural Prices Commission and issues of the Bulletin on Food Statistics. 7. The phenomenon of States fixing prices higher than the recommended prices is explained by the influence exercised by farmer interests on policy at the State level. But it is also due to the realistic economic reasoning that the procurement targets recommended by the Center could not be realized with unattractive purchase prices, especially in bad crop years. 8. In its numerous reports the Agricultural Prices Commission (APC) has repeatedly mentioned the considerations which have influenced the prices recommended: (a) the cost of cultivation; (b) the trend in the open market wholesale price reflecting overall shortage; (c) the need for securing a balanced growth of the output of related crops; (d) the need to reduce interstate price dispersion; and - 5 - (e) the need to limit the rate of inflation in the interest or the consumers of grain, industrial development and overall stability. Typical extracts from the Commission's reports enunciating these criteria are given in Appendix A. The record shows, however, that the various criteria listed above were applied and emphasized in an uncoordinated way. Some of them were stressed and used for some decisions and others on other occasions. They were never integrated into an objective model to compute the price to be recommended. 9. Therefore, the effective influence which different criteria did have on the recommendations of the Commission can only be analyzed by reviewing actual empirical relationships. The effect of the criterion of inter-crop balance cannot even be analyzed empirically because prices for related crops were not worked out systematically and simultaneously. Also, the Commission's concern with inflation only set a limit to the recommended increases in pur- chase prices when increases were warranted on other grounds. But the use of the other three criteria - cost, market price trends, and inter-regional dispersion - can be examined in some detail. 10. It has been argued that the procurement price, which acts as the effective support price, should cover the full cost of production, 1/ includ- ing the imputed value of family resources, in order that it may provide suffi- cient incentive to farmers to maintain and increase production. A price fixed on this basis would insure commercial farmers against losses in good harvest years, and provide to subsistence and semi-commercial farmers a surplus above paid-out (cash) costs which they can use to finance innovations (see Krishna, 1/ This full cost is denoted "cost C" in the Indian Farm Management data. (See Appendix B). - 6 - 1967). Thus, it is necessary, first of all, to compare the procurement prices fixed with the available estimates of the cost of production. 11. Second, it is necessary to compare the procurement price with the open market wholesale price because the distribution of the marketed surplus between government procurement and sale in the open market should be expected to depend on the ratio of the procurement price to the open market price. If the procurement price falls shorts of the open market price by a substan- tial margin, procurement is likely to remain satisfactory, even if administra- tive pressure is applied in the form of compulsory levies and other penalties. 12. The results of a comparison between the procurement price and the full cost of production of wheat, for the years and States for which cost observa- tions are available, 1/ are presented in Table 2. The Table shows that in five States (Uttar Pradesh, Punjab, Haryana, Bihar and Rajasthan) accounting for more than three-fourths 2/ of national wheat output, the procurement price was clearly below the full cost of production by about 14 to 38 percent in all three-year periods in the 1950s and 1960s for which cost data are available. But in the case of irrigated wheat in the Punjab, the deficiency of procurement price in the 1950s was negligible (only 1.5 percent) because the yield per hectare was high and cost per quintal was low due to irrigation. In two dis- tricts of Western Uttar Pradesh in the 1950s, the procurement price even exceeded the cost of production by a small margin in the case of irrigated wheat for the same reason. 1/ Cost data relate only to samples in districts, selected as representative of States/regions for Farm Management Studies. 2/ See Table A.1 in the Appendix, - 7 - TABLE 2 PROCUREMENT PRICE AND COST OF PRODUCTION OF WHEAT Average Excess of Procurement Number of Price Over Cost State (District) Period Observations of Production (Percent) Haryana /a (Karnal, Rohtak and Jind Tehsil of Sangrur) 1961-62 to 1963-64 3 -23.3 Haryana /b (Karnal, Rohtak and Jind Tehsil of Sangrur) 1961-62 to 1963-64 3 -28.6 Bihar (South Monghyr) 1957-58 to 1959-60 3 -27.0 Bihar (Shahabad) 1960-61 to 1962-63 3 -37.9 Rajasthan (Pali) 1962-63 to 1964-65 3 -29.2 Punjab /a (Amritsar and Ferozepur) 1954-55 to 1956-57 3 -14.0 Punjab /b (Amritsar and Ferozepur) 1954-55 to 1956-57 3 -1.5 Uttar Pradesh /a (Meerut and Muzaffarnagar) 1955-56 to 1956-57 2 -21.0 Uttar Pradesh /b (Meerut and Muzaffarnagar) 1955-56 to 1956-57 2 +8.0 Punjab /c (Ferozepur) 1967-68 to 1969-70 3 +11.0 Punjab /d (Ferozepur) 1967-68 to 1969-70 3 +30.0 Eastern Uttar Pradesh /a (Deoria) 1966-67 to 1968-69 3 +39.6 Eastern Uttar Pradesh /b (Deoria) 1966-67 to 1968-69 3 +21.0 Uttar Pradesh /a (Muzaffarnagar) 1966-67 to 1968-69 3 +41.0 Uttar Pradesh /b (Muzaffarnagar) 1966-67 to 1968-69 3 +66.0 /a Unirrigated Wheat. /b Irrigated Wheat. T7 Desi Wheat. /d Mexican Wheat. Source: Various issues of Bulletin on Food Statistics and Studies in the Economics of Farm Management. -8- 13. The relationship between procurement price and cost of production appears to have changed significantly in the first four years of the green revolution period beginning 1966/67. All the available observations for this period (Table 2, lower panel) show that the procurement price exceeded cost by a comfortable margin varying between 11 percent in the case of Desi wheat in the Punjab and 66 percent in the case of irrigated wheat in Western Uttar Pradesh. 14. This remarkable change was due to the fact that the cost of produc- tion increased less than the procurement price as new technology was adopted. In Table 3, for example, we can see that between the 1950s and 1960s, the procurement price increased in Western Uttar Pradesh by Rs 33 per quintal and the cost of production by only Rs 19 per quintal for irrigated wheat, and Rs 11 for unirrigated wheat. In Punjab also the procurement price increased by Rs 42 per quintal and the cost by only Rs 22 to Rs 38 per quintal (for irrigated, unirrigated and Mexican wheat). 15. We can conclude from this evidence that the wheat procurement price did not cover the full cost of production, or exceeded it only by a small margin in some irrigated tracts in the 1950s, but it provided a comfortable margin above full cost in the early years of the green revolution. 16. In contrast with the procurement price, the farm harvest price, namely, the price prevailing during six to eight weeks immediately after the harvest in each State, has been above the cost of production in all impor- tant wheat-producing States in the 1950s as well as the 1960s. The few exceptions to this relationship pertain to unirrigated or inadequately irri- gated areas of Uttar Pradesh, Punjab, Bihar and Rajasthan (Table 4). - 9 TABLE 3 CHANGES IN PROCUREMENT PRICE AND COST OF PRODUCTION, PUNJAB AND UTTAR PRADESH Increase in Increase in Procurement Price Cost Average Average 1967-68/1969-70 1967-68/1969-70 Minus Minus 1954-55/1956-57 /a 1954-55/1956-57 (Rupees Per Quintal) (Rupees Per Quintal) Punjab 42.33 Ferozepur Local Irrigated 37.98 Ferozepur Local Unirrigated 32.67 Ferozepur Mexican Irrigated 27.12 Ferozepur Mexican Unirrigated 21.81 Average Average 1966-67/1968-69 1966-67/1968-69 Minus Minus 1955-56/1956-57 /a 1955-56/1956-57 Uttar Pradesh 32.64 Muzaffarnagar Irrigated 19.23. Muzaffarnagar Unirrigated 11.38 Sources of basic data: Farm Management Reports and issues of the Bulletin on Food Statistics. /a The procurement prices relate to 1951/52/1952-53 but are assumed to be the same even for 1954-55/1956-57 for the purpose of this comparison, because no revisions were announced and there was no procurement in the latter years (Table 8). 17. Tables 5 and 6 document the additional fact that in all important wheat-producing States, without exception, the procurement price has been less than the harvest price and the annual average wholesale price. - 10 - 18. The two main wheat price hierarchies prevailing in the 1950s, 1960s and 1970s revealed by the evidence in Tables 2, 4, 5 and 6 can be summarized as: 1950s Procurement Price < Cost of Production < Harvest Price < Annual Average Wholesale Price. 1950s and 1970s Cost of Production < Procurement Price < Harvest Price < Annual Average Wholesale Price. The latter ranking implies that even when the procurement price has covered the cost of production, it has been consistently below the two market prices. - 11 - TABLE 4 FARM HARVEST PRICE AND COST OF PRODUCTION OF WHEAT Average Excess of Farm Harvest Number of Price Over Cost State (District) Period Observations of Production (Percent) Bihar (South Monghyr) 1957-58 to 1959-60 3 -6.0 Rajasthan (Pali) 1962-63 to 1964-65 3 -16.0 Uttar Pradesh /a (Meerut and Muzaffarnagar) 1956-57 to 1956-57 2 -16.0 Punjab /a (Amritsar and Ferozepur) 1954-55 to 1956-57 3 -2.0 Haryana /a (Karnal, Rohtak and Jind Tehsil of Sangrur) 1961-62 to 1963-64 3 +30.0 Eastern Uttar Pradesh /a (Deoria) 1966-67 to 1968-69 3 +38.0 Bihar (Shahabad) 1960-61 to 1962-63 3 +78.0 Punjab /b (Amritsar and Ferozepur) 1954-55 to 1956-57 3 +12.0 Uttar Pradesh /a (Muzaffarnagar) 1966-67 to 1968-69 3 +69.0 Haryana /b (Karnal, Rohtak and Jind Tehsil of Sangrur) 1961-62 to 1963-64 3 +21.0 Punjab /c (Ferozepur) 1967-68 to 1969-70 3 +9.0 Eastern Uttar Pradesh /a (Deoria) 1966-67 to 1968-69 3 +54.0 Uttar Pradesh /b (Meerut and Muzaffarnagar) 1955-56 to 1956-57 2 +15.0 Uttar Pradesh /b (Muzaffarnagar) 1966-67 to 1968-69 3 +92.0 Punjab /d (Ferozepur) 1967-68 to 1979-70 3 +28.0 /a Unirrigated Wheat. 7@ Irrigated Wheat. 7T Desi Wheat. 7W Mexican Wheat. Sources: Various issues of Farm (Harvest) Prices of Principal Crops, Agri- cultural Situation in India and Studies in the Economics of Farm Management. - 12 - TABLE 5 PROCUREMENT PRICE AND FARM HARVEST PRICE OF WHEAT Average Excess of Procurement Number of Price Over Farm State Period Observations Harvest Price (Percent) Bihar 1951-52 to 1953-54 3 -22.0 1966-67 to 1975-76 10 -20.9 Gujarat 1966-67 1969-70 to 1972-73 5 -23.0 Haryana 1966-67 to 1972-73 7 -12.0 Madhya 1952-53 to 1953-54 2 -16.0 Pradesh 1965-66 to 1975-76 11 -14.9 Punjab 1951-52 to 1952-53 2 -12.0 1966-67 to 1975-76 10 -18.9 Rajasthan 1951-52 to 1953-54 3 -12.0 1967-68 to 1975-76 9 -5.7 *Uttar 1951-52 1 -6.0 Pradesh 1966-67 to 1975-76 10 -14.1 Source: Various issues of Bulletin on Food Statistics, Farm (Harvest) Prices of Principal Crops, and Agricultural Situation in India. TABLE 6 PROCUREMENT PRICE AND ANNUAL AVERAGE WHOLESALE PRICE OF WHEAT Average Excess of Procurement Number of Price Over Whole- State Period Observations sale Price (Percent) Bihar 1951-52 to 1953-54 3 -28.7 1965-66 to 1975-76 10 -29.3 Gujarat 1951-52 to 1952-53 2 -32.4 1965-66 to 1974-75 10 -27.3 Haryana 1965-66 to 1975-76 11 -11.6 Madhya 1952-53 to 1953-54 2 -9.1 Pradesh 1965-66 to 1975-76 11 -17.2 Punjab 1951-52 to 1952-53 2 -5.3 1966-67 to 1975-76 10 -11.4 Rajasthan 1951-52 to 1953-54 3 -16.1 1967-68 to 1975-76 9 -19.9 Uttar 1951-52 to 1952-53 2 -16.1 Pradesh 1966-67 to 1975-76 10 -18.5 Source: Various issues of Bulletin on Food Statistics, Agricultural Prices in India, Statistical Abstract of Punjab, and Statistical Abstract of Haryana. - 13 - Cost/Price Comparisons: Rice 19. Table 7 shows that in three Southern rice-producing States (Andhra Pradesh, Tamil Nadu and Kerala) and West Bengal, accounting for about 40 percent of the total national output, 1/ the procurement price has been below the full cost of production by about 25 to 50 percent in all three-year periods in the 1950s and 1960s for which cost data are available. 20. For the Coimbatore district of Tamil Nadu some cost data are avail- able for the 1950s as well as the early 1970s. The average deficiency of the procurement price seems to have declined slightly from 31 percent in the middle 1950s to about 21 percent in the early 1970s. In the Cuddapah district of Andhra Pradesh the difference between cost and the procurement price narrowed from about 30 percent to 6 percent between 1967/68 and 1971/72. Thus, it seems that in Andhra Pradesh and Tamil Nadu this gap diminished in the early 1970s. 21. But the fact remains that in the two earlier decades the procure- ment price never covered the full cost of production in the major Southern rice-producing States and West Bengal. 22. In Assam, Bihar and Orissa (accounting for about 25 percent of total rice production in the country as a whole), the procurement price has been significantly higher than the cost of production. Madhya Pradesh, Punjab and Western Uttar Pradesh have also had procurement prices in excess of the cost of production (Table 7). 1/ Table A.2 in Appendix. - 14 - TABLE 7 RICE PROCUREMENT PRICE AND COST OF PRODUCTION (ACTUAL) /a Average Excess of Procurement Number of Price Over Cost State (District) Period Observations of Production (Percent) Andhra (Cuddapah) 1967-68 to 1969-70 3 -29.7 1971-72 1 -5.5 Kerala (Alleppey, Quilon) /c 1964-65 1 -29.0 Tamil NaduTSalem, Coimbatore) /b 1954-55 to 1956-57 3 -31.0 Tamil Nadu (Thanjavur) 1967-68 to 1969-70 3 -10.0 Tamil Nadu (Coimbatore) 1970-71 to 1972-73 3 -21.0 Eastern U.P. (Deoria) 1966-67 to 1968-69 3 -40.7 West Bengal (24 Parganas) /b 1954-55 to 1956-57 3 -24.6 West Bengal (Hooghly) /b 1954-55 to 1956-57 3 -49.7 Assam (Nowgong) 1968-69 to 1970-71 3 +13.7 Bihar (Shahabad) /b 1960-61 to 1962-63 3 +34.8 Bihar (South Monghyr) /b 1958-59 to 1959-60 2 +5.0 Madhya Pradesh (Raipur) /c 1964-65 1 +22.2 Orissa (Cuttack) 1967-68 to 1969-70 3 +38.7 1971-72 to 1972-73 2 +32.8 Western U.P. (Muzaffar- nagar) 1966-67 to 1968-69 3 +19.6 Punjab (Ferozepur) 1968-69 to 1969-70 2 +19.0 /a Based on the Farm Management Studies (FMS) cost of production data. lb In the absence of a common period for procurement period price (P) and cost of production (C) we have taken. (a) F (1951-52/1953-54) x C (1954-55/1956-57) for Tamil Nadu (Salem) and West Bengal, (b) F (1951-52/1953-54) x CH (1958-59/1959-60) for Bihar (South Monghyr), and P FH (c) FH (1964-65/1973-74) x C (1960-61/1962-63) for Bihar (Shahabad). where FH is farm harvest price. /c Cost data are available for the period 1962-63/1964-65, but the procure- ment prices are available only from 1964-65 onwards. Hence there is only one observation. Sources: Various issues of Bulletin on Food Statistics; Studies in the Eco- nomics of Farm Management (Various States and periods), APC Report September 1974, Indian Agriculture in Brief, 14th edition. - 15 - 23. This inter-regional difference in the cost/procurement price rela- tionship is explained by the fact that costs are much lower in the northern States than in the south. These cost differences are in turn due mainly to differences in labor cost. Information available in the Farm Management Studies clearly reveals that labor cost, both in absolute terms and as a pro- portion of total cost, tends to be lower in the eastern and northern States than in the southern States (Appendix Table A.3). 24. As in the case of wheat, the farm harvest price of rice has been above the cost of production in most of the States. But in Andhra Pradesh, parts of Bihar, West Bengal and Uttar Pradesh, even the farm harvest price has been less than the full cost of production. 1/ 25. In all States, the rice procurement price has been less than the harvest price and the annual average wholesale price. 2/ The only apparent exception in Tamil Nadu where in recent years the procurement price has been slightly above the wholesale price but even this exception is due to the fact that the wholesale price reported for that State is the controlled price. 26. Thus, we are led to conclude that the rice procurement price has been below the cost of production in West Bengal and South India, though in recent years the gap has tended to shrink. In the Northern States the procure- ment price (P) has usually covered the full cost of production (C). But it (P) has always been below the two market prices - the harvest price (H) and the market wholesale price (W) - in almost all States. 1/ Appendix Table A.4. 2/ Appendix Tables A.5 and A.6. - 16 - 27. The rice price hierarchy prevailing in most States, revealed by the evidence in Table 7 and Appendix Tables A.4, A.5 and A.6 can be depicted as follows: P < C < H < W 3. THE EMPIRICAL BASIS OF PROCUREMENT PRICE DETERMINATION 28. Hypotheses to explain the behavior of procurement will be tested in the next Section. But we investigate here the empirical basis of the pro- curement prices actually fixed. The notion suggested by the evidence is that the procurement price is strongly influenced by the trend of the wholesale prices in the recent past. We have defined this past period as two preceding years and estimated the elasticities of the procurement price (PP) with respect to the lagged average wholesale price in two preceding years (PWL). These elasticities for wheat are tabulated in Table 8. They range from 0.61 in Madhya Pradesh to 0.86 in Haryana. The all-India elasticity 1/ turns out to be 0.68. The squared correlation coefficient (r ) is satisfactory in the case of five of the six States and in the all-India regression, and all the elasticity coefficients in the estimated (log-linear) regressions have signi- ficantly high t-values. The synchronous movements of the all-India PP and PWL can be seen in Figure 1. 29. For rice the movements of PP and PWL are depicted in Figure 2. And statewise elasticities are tabulated in Table 9. They range between 0.6 and 0.9 in nine States and between 0.3 and 0.5 in MP, Maharashtra and West Bengal. The all-India elasticity is 0.65. Therefore, we can confidently 1/ The all-India procurement price is the weighted average of States' procurement prices, the weights being the States' shares in all-India procurement in 1972-73. - 17 - infer that in the case of rice as well as wheat, the procurement price has simply followed the trend of the market wholesale prices with a time-lag. In other words, the market price has been the leader price and the procurement price a reluctant follower. Regardless of the principles of price fixation, the prices actually fixed by the States seem to have been based on the simple rule of thumb that if the market price rises for one or two years, the pro- curement price should be raised by about half to two-thirds of the percentage increase in the market wholesale price. 30. In the few years when the lagged average wholesale price of rice or wheat declined in particular States, the procurement price was kept constant or increased. It could never, of course, be reduced because a reduction would arouse very strong farmer resentment. On the other hand, in periods of abnormally short harvests and steep increases in market prices (as in the late 60s and early 70s) the adjustment of the procurement price could only be a small fraction of the rise in the (lagged) wholesale price and the gap between the two became usually wide. (See Figures 1 and 2.). TABLE 8 ESTIMATED ELASTICITY OF THE WHEAT PROCUREMENT PRICE WITH RESPECT TO THE AVERAGE WHOLESALE PRICE IN TWO PRECEDING YEARS IN DIFFERENT STATES (1965-66/1975-76) 2 Elasticity Values R Bihar 0.83 (5.17) 0.75 Haryana 0.86 (6.22) 0.81 Madhya Pradesh 0.61 (5.10) 0.74 Punjab 0.78 (3.35) 0.56 Rajasthan 0.67 (5.92) 0.80 Uttar Pradesh 0.62 (2.83) 0.47 All India 0.68 (5.11) 0.74 INDIA PROCUREMENT PRICE AND LAGGED WHOLESALE PRICE OF WHEAT Price Index (1965-66/1975-76) 150 PWL - Average of all- India wholesale price index of wheat tagged one year and wholesale price index of wheat lagged two years (1972-73 = 100) PP -Weighted procurement price index of wheat producing states in India (1972-73 = 100) - - _____ ________PW L _ _ _ _ _ _ _ 100 tiP o lll|||tl l s s l 50 1965-66 67 69 71 73 1975-76 Marketing year beginning April !World Bank -21305 INDIA PROCUREMENT PRICE AND LAGGED WHOLESALE PRICE OF RICE PWL (1964-65/1975-76) 240[ 220- PWL-Average of wholesale price index (1965-66 = 100) lagged one year and lagged two years PP -Weighted procurement price index of rice producing states in India (1965-66 = 100) 200 180 160 140 195 120 00 16 6 6' 69 71 7 4 75 f165 Marketing year beginning November 4p `1Wrl 135 40~~~~~~~~~~~~~~~~~~~ 4~~~~~~~4 80- 1964-65 66 67 68 69 70 71 72 73 74 75 76 Marketing Year beginning November World Bank - 21304 - 20 - TABLE 9 ESTIMATED ELASTICITY OF THE RICE PROCUREMENT PRICE WITH RESPECT TO THE AVERAGE WHOLESALE PRICE IN TWO PRECEDING YEARS IN DIFFERENT STATES (1964-65/1975-76) It' 2 State Elasticity Values R Andhra 0.64 (9.10) 0.88 Assam 0.93 (7.17) 0.82 Kerala 0.71 (5.03) 0.69 West Bengal 0.34 (4.97) 0.68 Orissa 0.70 (8.26) 0.86 Madhya Pradesh 0.49 (6.75) 0.80 Tamil Nadu 0.83 (5.82) 0.75 Bihar 0.64 (5.12) 0.70 Uttar Pradesh 0.59 (8.50) 0.87 Punjab 0.67 (8.13) 0.87 Haryana 0.73 (9.52) 0.90 Maharashtra /a (1964-65/1965-66) (1968-69/1975-76) 0.29 (2.45) 0.38 All India (12 States, including Tamil Nadu) 0.65 (10.84) 0.92 /a As the wholesale price for 1965-66 was not reported, we have lost 1966-67 and 1967-68 in the lagged equation. Sources: Various issues of Bulletin on Food Statistics, Agri- cultural Prices in India, and Statistical Abstract (Different States). 4. PROCUREMENT FUNCTIONS 31. Before estimating procurement functions it is useful to review trends in the procurement of wheat and rice as a proportion of total production. 32. The time-series of the all-India wheat procurement/output ratio (POR) and the ratio in seven major States for the 26-year period (1951-52/1976-77) are tabulated in Table 10. In the early fifties (1951-52 and 1952-53) about 10 to 12 percent of output was procured. For the next fifteen years (1953-54/ 1967-68) the all-India procurement/output ratio remained negligible, evidently TABLE 10 RATIO OF WIEAT PRQCURPEMNT TO PROUCTION INDIAN STATES (1951-52/1976-77) Madhya Uttar Year Bihar Gujarat Haryana Pradesh Punjab Rajasthan Pradesh All-India 1951-52 - - - 2w81 25.08 14,63 8.39 12.55 1952-53 - - - 2,73 16.50 6.75 .11.49 10.02 1953-54 - - - 1.46 2.80 2.83 - 3.43 1954-55 - - - - - - - - 1955-56 - - - - - - - - 1956-57 - - - - - - - - 1957-58 - - - - - - - - 1958-59 - - - - -- 1959-60 - - - 2.58 7.61 1.24 1.78 2.98 1960-61 - - - 5.06 9.21 - 1.31 3.53 1961-62 - - - - - - - 1.82 1962-60 - - - - - 1963-64 - - - - 3.05 - - 0.84 1964-65 - - - _ 3.07 - - 0.95 1965-66 - - - 3.46 10.65 0,46 0.78 5.78 1966-67 - - 3.68 6.78 3.97 - - 1.90 1967-68 - - 5.69 8.54 23.36 2,18 3.69 7.84 1968-69 - - 14.59 5.47 40.51 5.31 0.83 13.89 1969-70 0.79 - 17.81 1.05 41.46 0.42 3.27 12.80 1970-71 - - 22.74 0,36 49.33 0.08 5.10 15.89 1971-72 1.03 5.64 30.27 1.93 57.10 7.89 14.89 21.40 1972-73 0.08 0.11 34,05 2.04 55.69 2.44 11.17 18.96 1973-74 1.59 - 26.27 8,.40 50.43 8.33 10.85 18.32 1974-75 0.12 2.32 13.75 4.88 20.90 7.09 5.43 8.98 1975-76 2.25 2.31 22.16 5.87 44.55 10.17 11.01 16.80 1976-77 4.40 11.81 37.23 9.97 51.57 20.71 21.04 23.30 Sources: Various annual issues of the Bulletin on Food Statistics. - 22 - because the availability of concessional imports led the government to relax domestic procurement effort. In six of these fifteen years no procurement operation was undertaken at all. The average procurement/output ratio for the whole fifteen-year period turns out to be only 1.8 percent. (If the years 1965-66 and 1967-68 are excluded the average sinks to only 1.2 percent.) The procurement ratio recovered to a significant level only in 1968-69, after the big increase in procurement price in the previous marketing season and a record harvest in the rabi season of 1968. This was the first normal season after the introduction of the new HYV technology. In the following three years (1969-70/1971-72) the procurement/output ratio (POR) rose from 14 to 21 percent. Punjab and Haryana farmers sold as much as one-half and one-third of their wheat crop to the government. In 1972-73 and 1973-74 the government introduced the "socialization" of the wheat trade involving monopoly purchase of market arrivals by State agencies. The POR remained high in these two years (18 to 19 percent) but lower than the peak reached in the pre-socialization year (21 percent). In the season 1974-75 the procurement price was raised by 38 percent to a record level of Rs 105 per quintal (from Rs 76 in the preceding year), but POR suddenly sank from 18 percent to a mere 9 percent. Two reasons seem to explain this decrease: the decline in output by about 3 million tonnes and the withdrawal of the socialization of the wheat trade in April 1974 after two years of experimentation. In the next two years, 1975-76 and 1976-77, the POR recovered to 17 percent and then rose to the record level of 23 percent. 33. Apart from price policy the main factor which accounts for the satisfactory trend in wheat procurement since 1968-69 is that all-India wheat procurement is dominated by Punjab, Haryana and (Western) Uttar Pradesh, where the irrigation ratio is very high, the new wheat technology has been rapidly - 23 - adopted, and the wheat crop is raised on medium and large-scale farms as a commercial crop. 34. In the case of rice, Table 11 shows that the all-India procurement/ output ratio (POR) has recorded different levels in four distinct periods. In the first three years (1951/52-1953/54) it averaged 7.6 percent. In the next three years (1954/55 to 1956/57), there was hardly any procurement and the POR averaged only 0.57 percent. Then over a seven-year period (1957/58 to 1963-64), the ratio stagnated at a little more than 2 percent. In 1964/65, when the procurement prices began to be fixed regularly in all the States and procurement activity was intensified the ratio suddenly recorded a threefold increase from 2.75 percent in 1963/64 to 7.4 percent in 1964/65 and 10.3 percent in 1965/66. Over the decade ending 1973/74, the ratio has averaged 8.1 percent which is comparable to its level in the early 1950s. However, it should be noted that after reaching the peak level of 10.3 percent in 1965/66, the ratio declined slowly to about 7 percent in 1972/73 and has recovered to about 9-12 percent only in the recent three years (1973-74/1975-76). 35. In view of the record, the hypothesis set up to explain the volume of procurement from year to year (PR) is simply that it depends on (a) the level of output (Q) and (b) the ratio of procurement price and the open market wholesale price (PPW):(1) PR = f(Q, PPW). In this equation, the price ratio is expected to catch the substitution effect:the division of the farmers' marketed surplus between sales to the government and sales in the open market. The output variable captures the scale effect. 36. Equation (1) for wheat 1/ has been estimated with all-India data for the period 1965/66-1975-76. 1/ Both the PR and Q figures relate to the period April-March every year. TABEI 11 RATIO OF RICE PRCUR8M NT TO PRODUCTION (INDIAN STATES) Year Andhra Ass8n Bihar Haryana Kerala Madhya Maharashtra Orissa Punjab Tasil Uttar West All India Aver- (November Pradesh Nadu Pradesh Bengal (Pereent) age to October) (Percent) 1951-52 - 4.43 0.85 - - 10.17 - 7.60 55.21 35.65 5.54 9.69 9.86 1952-55 - 7.94 0.76 - - 11.15 - 12.94 41.75 9.97 2.55 5.79 7,8 7.60 1953-54 6.90 6.16 0.52 - - 6.45 - 11.77 25.44 2.76 0.46 5.22 5.1 1954-55 - - - - - - - - 21.17 - - - O.98 1955-56 - - - - - - _ _ - - - - 0.12 0.57 1956-57 - - - - - - - - - - - - 0.67 1957-58 6.52 0.06 - - 1.75 0.09 - 2.95 51.58 - - 1.69 1.81 1958-59 2.44 - 0.45 - 5.24 11.92 - 7.24 25.65 5.52 5.40 1.79 5.54 1.72 1959-60 0.56 - - - - 14.41 - 2.04 50.48 - 5.56 - 2.64 1960-61 - - - - - 4.69 - 0.55 55.08 - 5.68 - 1.54 2.15 1961-62 - - - - - 5.41 - 0.38 36.62 - 2.65 - 1.17 1962-63 5.56 - - - - 2.94 - 1.08 535.51 1.14 4.21 - 1.82 1963-64 4.26 3.76 - - - - 4.05 - 1.67 34.98 1.96 4.85 2.14 2.75l 1964-65 15.44 5.54 0.75 46.15 5.05 12.58 9.14 5.54 80.92 9.44 4.06 5.82 7.58 1965-66 16.08 3.03 1.64 47.83 1.90 6.25 8.62 6.51 80.41 25.15 2.99 11.96 10.26 1966-67 12.47 2.51 0.21 40.56 7.01 5.05 JD.99 5.42 38.17 19.48 5.48 2.55 8.11 8.58 1967-68 14.55 5.10 0.85 58.19 6.58 7.89 12.58 4.26 57.55 18.71 4.54 5.09 8.60 1968-69 8.69 8.55 1.29 61.51 6.86 10.58 14.10 6.11 58.70 17.44 6.84 7.04 8 .65 1969-70 5.60 4.42 0.82 64.15 6.67- 10.84 15.20 6.95 65.64 7.79 6.05 6.47 7.28) 8.54 1970-71 9.74 4.19 0.72 54.78 5.47 15.44 14.19 5.92 76.31 2.22 7.00 4.55 7.55s 1971-72 5.66 4.56 0.55 58.96 4.51 11.56 15.50 4.67 85.81 5.04 7.84 5.90 7.24 1972-73 7.66 4.09 1.25 59.96 4.14 7.88 8.85 5.25 78.95 2.55 6.94 5.08 6.90 8.71 1973-74 15.02 5.42 1.58 77.20 5.25 5.88 9.81 4.82 81.09 4.86 11.17 2.74 8.89 1974-75 15.58 6.50 1.34 70.92 5.15 6.27 6.16 1.74 82.21 10.54 8.11 5.44 9.45 1975-76 20.37 10.13 1.22 76.28 2.14 7.02 2.52 4.70 85.04 15.12 17.17 4.25 12.25) Source: Various issues of Bulletin on Food Statistics and Economic Survey (1976-77). - 25 - (1) PR = -11.50 + 0.332 Q + 7.324 PPW (-7.168) (17.429) (5.468) -2 R = 0.968 (t-values are given in parentheses) The coefficients of output and the price ratio are highly significant with expected signs. The elasticity of procurement with respect to output turns out to be as high as 2.32; and the elasticity of procurement with respect to the procurement price/wholesale price ratio is even higher (3.72). 37. In the case of rice a modification of equation (1) was necessitated by the fact that in some major rice States the output of a year is so highly correlated with the wholesale price of that year that it also gets highly correlated with the price ratio (PPW) via the denominator of the ratio. Accordingly, the absolute procurement price (PP), instead of the price ratio (PPW) has been included in the equation, leaving the wholesale price effect to be reflected in the output variable itself. The amended equation is: (2) PR = f(Q, PP) 38. The all-India (11 States excluding Tamil Nadu 1/) regression for rice corresponding to equation (2) reveals the elasticity of procurement with respect to procurement price to be 0.42 (Table 12). The elasticity with respect to output is almost unity (1.01). Specification (2) explains 85 percent of the variation in procurement, with the regression coefficients of procurement price as well as output being individually significant. 39. The regression for six rice surplus States 2/ (Andhra, Haryana, Madhya Pradesh, Orissa, Punjab and Uttar Pradesh combined) again explains 1/ Tamil Nadu is excluded because the available data revealed a high negative elasticity (-4.35) with respect to output. The noneconomic reasons which caused this odd relationship need further investigation. 2/ See Appendix B for the identification of surplus and deficit States. - 26 - 89 percent of the variation of procurement (Table 12). The elasticity of procurement with respect to output (1.16) in the surplus States is comparable with the elasticity (1.01) for the whole group of 11 States. With respect to prices the elasticity is 0.71 in the six surplus States, as compared with 0.42 in the group of 11 States. 40. The results for the group of five rice deficit States 1/ (Assam, Bihar, Kerala, Maharashtra and West Bengal combined) are different and interest- ing (Table 12).. In the deficit States, unlike the surplus States, procurement is significantly responsive to the ratio of the procurement price to the market wholesale price. In fact, in the deficit States the price elasticity with respect to the relative procurement price is as high as 1.2; in the surplus States it is only 0.8 (with respect to procurement price alone.) On the other hand, in the deficit States, the elasticity with respect to output is less (0.75) than in the surplus States (1.16). Also, in the case of deficit States the coefficient of output is only marginally significant at the 10 percent level. 41. This result is important because it implies that, given the output, the volume of procurement in deficit States is much more sensitive to the ratio between the procurement price and the wholesale price. Substantial pro- curement can be made even in the deficit States provided that this ratio is attractive for farmers. 42. All the estimated procurement functions bring out the importance of the procurement price as a stimulant of the quantity of procurement. As it refutes the view of some policy makers that the volume of procurement can or should be raised substantially by mere administrative pressure, without an 1/ See Appendix B for the identification of surplus and deficit States. TABLE 12 RICE PROCUREMEN REGRESSIONS: INDIA (1964-65/1974-75) Serial Dependent Regression2 Numberial untry Variablent ConstanCoefficients R Elasticity Number Country Variable Constant -pp pP/PW Pp PP/PW Q 12.1 India (11 States Excluding PR -1329.49 8.52 0.08 0.85 0.42 1.01 Tamil Nadu) (-2.13) (2.70) (4.00) 12.2. India (6 Surplus PR -1640.28 10.37 0.14 0.89 0.71 1.16 States)* (-3.26) (4.16) (3.69) 12.3 India (5 Ddficit PR 541.44 882.35 0,03 0.64 1.27 0.69 States)* (-1.51) (3.05) (1.09) PR t Quantity procured PP t Procurement price index (1965-66 = 100) PW s Wholesale price index (1965-66 = 100) Q t Production *See Appendix B. - 28 - improvement in the price offered to the farmers. Our equations show that the supply of any crop to the government is responsive to prices just like total output or the total marketed surplus. Therefore, the importance of an appro- priate procurement policy for raising the level of procurement cannot be over- emphasized. 43. Our results also underscore the obvious truth that the propitious seasons for a stock build-up are not the bad ones in which stocks are urgently needed but those in which output levels are comfortable. 44. It remains to review the recent evidence on overall supply response, but a brief study of interstate price dispersion is in order here. 5. INTERSTATE PRICE DISPERSION AND ZONING 45. The zoning policy under which interstate or interzone movement of grain under private auspices is restricted has been a controversial subject in India. The supporters of zoning have argued that it facilitates procure- ment at relatively low prices by restricting the export of grain from surplus States. The opponents have pointed out that it hurts producers in the sur- plus States and consumers in the deficit States by pushing prices down for surplus State producers below the free trade level and pushing them up for deficit State consumers above this level. The total imports of deficit States and the total exports of surplus States are likely to be less with zoning than without it. (See Bhagwati and Chakravarty 1969 and Krishna 1965). 46. It is difficult to estimate what procurement and interstate grain transfers would be under hypothetical movement policies other than the ones actually followed. But the hypothesis that the actual zoning policy increases - 29 - interstate price dispersion can be tested with the specification: (3) CVPt = f(Zt, St, Ct) where CVP is the coefficient of interstate variation of wholesale wheat prices in year t; Z is a dummy variable having the value zero in years of free trade and value I in years of large (multi-State) or single-State zones; S is total wheat supply defined as all-India net wheat output plus net imports plus the initial government stock; and C is an index of the concentration of wheat output. 1/ 47. The variable Z reflects zoning policy and is expected to have a positive sign, for when the interstate movement of wheat is restricted, the coefficient of interstate price variation is likely to be higher. The supply variable is included for the obvious reason that, given the zoning policy, the larger the overall availability of wheat, the greater will be the volume of interstate movement under official auspices, and, therefore, the smaller the interstate price dispersion. However, data show that this negative effect of availability on interstate price variation is effective only after avail- ability exceeds a critical level. In other words, the relationship between the CVP and availability ceteris paribus is parabolic. 48. The inclusion of the concentration index became necessary because the geographical distribution of wheat production in India is highly skewed. The share of Punjab and Haryana in total all-India output has increased from about 20 percent to 30 percent over the 25 years (1950/51-1974/75). The share of Uttar Pradesh has come down from about 42 percent to 30 percent. And the 1/ The index is | r. where ri is the percentage share of State i in all-India output. - 30 - share of other States has changed marginally. But the overall index of concentration has declined. It is obvious that, apart from other factors, a higher concentration of production should be associated with greater inter- state price variation. Therefore, the concentration index is expected to hav a positive sign. 49. The second-degree wheat equation estimated with 34 observations (1951-53 to 1970-75) turned out to be: 2 (4) CVP = -112.63 + 5.39Z + 3.70S - 0.08S + 1.93C (-2.67) (2.44) (3.24) (2.31) (2.38) 2 R = 0.563 (t-values are shown in the parentheses). 50. In the estimated equation, the coefficients of all the parameters are significant with the expected signs of the concentration (>0) and zonal restriction (>0) variables. The coefficients of S (>0) and S (<0) signify that the inverse relationship between supply and price dispersion becomes effective only at a high supply level of 23.42 million tonnes. 51. The equation supports the proposition that interstate price disper- sion increased systematically with the intensity of zonal restrictions. It is interesting to note that the coefficient of variation averaged only 11.7 percent in the seven years of unrestricted movement, 14.93 percent in the five large-zone years and 19.42 percent in the twelve single-State zone years (Table 13). This relationship is faithfully captured by the estimated equation. - 31 - TABLE 13 WHEAT ZONING AND PRICE VARIATION: INDIA (1951-52/1974/75) Supply Coefficient of (Million Concentration of Year Variation (CVP) Average Zoning (Z) tonnes)(S) Production (C) (1) (2) (3) (4) (5) (6) 1951/52 19.26 1.00 L 8.71 51.54 1952/53 13.68 1.00 L 7.78 52.07 1953/54 8.30 1.00 L 8.57 48.68 1954/55 13.10 13.59 1.00 L 8.10 49.38 1955/56 10.46 0.00 F 8.30 48.00 1956/57 9.14 9.80 0.00 F 8.95 47.08 1957/58 14.46 1.00 S 10.59 47.39 1958/59 15.55 1.00 S 10.05 48.34 1959/60 19.99 1.00 S 12.12 46.07 1960/61 12.15 15.54 1.00 S 12.39 47.36 1961/62 13.14 0.00 F 12.78 48.48 1962/63 10.09 0.00 F 13.90 47.42 1963/64 7.63 10.29 0.00 F 13.36 45.87 1964/65 30.39 1.00 S 15.58 46.41 1965/66 32.43 1.00 S 16.49 48.11 1966/67 23.27 28.70 1.00 S 17.18 45.47 1967/68 20.31 20.31 1.00 L 16.66 47.06 1968/69 21.12 1.00 S 18.07 45.50 1969/70 10.72 15.92 1.00 S 19.13 43.61 1970/71 16.94 0.00 F 20.13 44.35 1971/72 15.03 15.99 0.00 F 20.84 43.71 1972/73 14.97 1.00 S 25.99 41.77 1973/74 22.33 1.00 S 24.88 43.19 1974/75 15.64 17.65 1.00 S 23.18 41.62 Sources: Various issues of Bulletin on Food Statistics, Agricultural Prices in India and Statistical Abstracts (Different States). L = Multi-State Zones. S = Single-State Zones. F = Free Movement. - 32 - 52. For rice, the following equation, estimated with 25 observations (1950-51/1974/75), also turned out to be highly satisfactory: (5) CVPt = 36.46 + 7.18 Zl+ 27.88 Z2 - 1.02 Qt (4.87) (2.38) (7.24) (4.85) 2 R = 0.782 (Absolute t-,values are given in parentheses). Here the dummy Z1 is zero in years of free movement and unity in large-zone years. Z is zero in free trade and large-zone years and unity in years of single-State zones. The variables are all significant with the expected signs. Again, the coefficient of variation averaged 8.66 percent in the three years of free movement, 10.74 percent in 7 large-zone years and 24.79 percent in 11 single-State zone years (Table 14). Thus the introduc- tion of single-State zoning doubled the average interstate price dispersion. 53. The two price dispersion equations for wheat and rice provide strong support for the view that with zone restrictions surplus producers earn less and deficit area consumers pay more than they otherwise would; and both lose more in bad crop years than in normal years. This proposition would hold even if it is true that the government can procure more grain with zoning; for price dispersion clearly increases in spite of the transfer of this "extra" grain from the surplus to the deficit areas. The transfer must be less than it would be without zoning; otherwise dispersion could not increase. Whatever zoning may be achieving it is in all probability not increasing the sum of producer and consumer welfare as it is measured in welfare economics. - 33 - TABLE 14 RICE ZONING AND PRICE VARIATION: INDIA (1951-52/1974/75) Coefficient of Year Variation (CVP) Average Zoning (Z 1) (Z 2) (1) (2) (3) (4) 1951-52 24.7333 1 0 1952-53 16.3264 18.28 1 0 1953-54 13.7902 1 0 1954-55 9.2854 0 0 1955-56 7.6632 8.66 0 0 1956-57 9.0269 0 0 1957-58 12.7894 1 0 1958-59 13.7317 1 0 1959-60 11.2268 1 0 1960-61 10.1580 10.74 1 0 1961-62 8.6056 1 0 1962-63 10.0990 1 0 1963-64 8.5910 1 0 1964-65 21.2049 0 1 1965-66 31.7946 0 1 1966-67 43.0606 0 1 1967-68 30.5808 0 1 1968-69 15.0192 0 1 1969-70 18.0620 24.80 0 1 1970-71 16.6435 0 1 1971-72 18.2834 0 1 1972-73 27.1014 0 1 1973-74 29.0047 0 1 1974-75 22.0070 0 1 Sources: Various issues of Bulletin on Food Statistics, Agricultural Prices in India, and Statistical Abstracts (Different States). - 34 - 6. SUPPLY FUNCtIONS 54. The fact that the acreages of individual crops in India vary system- atically in response to intercrop price movements is now widely accepted on the basis of numerous acreage response studies. Therefore, we did not consider it necessary to run another set of routine acreage response regressions. On acreage response it is sufficient to recall the evidence from a few recent studies. The early papers (Krishna 1963 and Dharam Narain 1965) demonstrated the presence of positive acreage response with pre-Independence data. But in recent work, post-Independence data have been used. In some cases, the period covered by the studies includes the early years of the Green Revolution. Table 15 summarizes some estimated wheat acreage elasticities. 55. The most numerous elasticity estimates based on an identical model have been generated recently by Cummings (1975). The model includes lagged price, lagged acreage, rainfall and trend. Cummings reports short-run and long-run elasticities of wheat acreage for eight States estimated with post- Independence data. For Punjab the short-run wheat estimate (0.1) is close to the estimate made with pre-Independence data. Of the other estimates in Table 15, very few are statistically significant. (The long-run elasticity estimates are particularly unreliable.) The significant short-run estimates from well-specified equations lie mostly in the interval .08 to .24. Only the Cummings equation for Gujarat yields a significant higher estimate of the order of 0.9. This estimate is out of line with most other significant estimates. The reasons for this unusual estimate need to be investigated. TABLE 15 WHEAT ACREAGE ELASTICITY ESTIMATES Region Period Elasticity Signifi- Source SR LR cance 1. Punjab Pre-Independence (.22) S Krishna, 1963 2. Punjab Pre-Independence .08 .14 S Krishna, 1963 3. Punjab 1950-67 .10 .13 NS Cummings, 1975 4. Punjab NA .002 .004 NA NCAER, 1969 5. Punjab Post-Independence (.898) S Sud & Kahlon, 1969 6. Punjab NA .08 .86 NA Ramesh, 1965 7. Haryana 1961-73 (.01 to .05) NS Singh & Kumar, 1976 8. U.P. 1939-58 (.06) NA NCAER, 1962 9. U.P. NA .06 .06 NA NCAER, 1969 10. M.P. NA .09 .20 NA NCAER, 1969 11. Rajasthan 1951-68 .02 .03 NS NCAER, 1969 12. Rajasthan NA .04 .12 NA NCAER, 1969 13. Bombay 1939-58 (.64) NA NCAER, 1962 14. Gujarat 1954-67 .93 1.00 S Cummings, 1975 15. Maharashtra 1955-67 .24 .24 S Cummings, 1975 16. Mysore 1954-67 .23 .33 S Cummings, 1975 17. Delhi 1948-67 .17 .25 S Cummings, 1975 18. H.P. 1949-66 .02 .01 NS Cummings, 1975 19. West Bengal 1946-67 .23 .20 NS Cummings, 1975 20. India 1939-58 (.16) NA NCAER, 1962 21. India 1949-66 (.67) S Maji et.al. 1971 22. India NA .11 .72 NA Ramesh, 1965 NS : Not significant ) : Short-run and long-run not distinguished NA : Not available S : Significant SR : Short-run LR : Long-run - 36 - 56. Since wheat is a major subsistence crop with a large acreage in the growing States, it is reasonable that the short-run acreage elasticity should be low (0.08 to 0.24). But even a 2 percent shift in a large acreage, induced by a 10 percent change in the realized relative price of wheat, can cause a substantial variation in production. 57. Cummings also reports short-run and long-run elasticities of rice acreage for as many as 13 States/Territories, estimated with time series for the period 1949-67. The price elasticities of acreage are, however, statistically significant only in the case of five States, Andhra Pradesh (0.48), Kerala (-0.14), Maharashtra (-0.12), Manipur (0.20), and West Bengal (autumn crop) (0.37). Only short-run elasticities are shown in the parenthesis). Where the elasticity is positive and significant it ranges between 0.2 and 0.5 as in Manipur, West Bengal and Andhra Pradesh. In Kerala and Maharashtra the elasticity has turned out to be negative with the Cummings specification. The positive and significant elasticities cited above for the recent period are comparable with the elasticities estimated for the pre- independence period averaging 0.35 (Krishna 1963 and Parikh 1972). Another estimate of the short-run elasticity for Andhra (Rao and Pandey, 1976) is also of the same order of magnitude (0.35). 58. While acreage response has been the subject of considerable study, relatively little work has been done on output response. The elasticity of output with respect to the appropriate relative price is the sum of the elasticity of acreage and the elasticity of yield per acre; therefore, the elasticity of acreage can be a good approximation to the elasticity of output only if yield is stagnant or insensitive to price changes. In the pre- Independence period, the yield per acre of wheat was not ony stagnant but - 37 - declining at a low rate. During the 30-year period 1914-15/1943-44, for example, wheat acreage increased at the rate of 0.64 percent per annum, but yield per acre decreased at the rate of -0.14 percent. In the post- independence period (1950-51/1973-74), the acreage growth rate escalated to 2.73 percent and the yield growth rate escalated, even more, to 3.08 percent. For this period, the elasticity of output with respect to relative price should be much higher than the elasticity of acreage with respect to relative price. In response to price changes (and other exogenous variables such as the overall irrigation capacity created by the government, the availability of new technology, etc.) farmers would have adjusted not only the acreage in wheat but also the quantities of other inputs applied from year to year. In other words, a yield response on their part would have been added to acreage response to generate the recorded response to output. 59. In the case of rice, too, yield was stagnant before Independence but after Independence it has been growing significantly and at a higher rate than acreage. During 1957-58/1970-71, 1/ rice yield was growing at 2.29 percent per year whereas rice acreage grew only 1.4 percent. For such periods of substantial yield growth the price elasticity of output should be expected to be significantly higber than the price-elasticity of acreage. 60. These considerations suggest that for the post-independence period supply functions rather than acreage functions need to be estimated for wheat as well as rice. The specification of the price variable in the supply function for the post-independence period also needs modification. When acreage growth or acreage substitution is the main phenomenon the relevant 1/ Excluding the exceptionally bad years 1965/66 and 1966/67. - 38 - price variable to explain the variation in acreage is the intercrop price ratio. But when yield growth is the major factor causing output growth, and acreage is relatively stable, the appropriate price variable would be the price of the crop deflated by the input cost index. For the growth of yield is associated with an increase in the ratio of purchased inputs to unpaid family inputs, and the intensity of input application should depend upon the output-input price ratio. 61. Accordingly, we have specified the following supply equation for rice: (6) Qt = f (PW t-/ IPt, RNt, Qt-1 where: i) Qt is all-India rice output for the crop year July-June; (ii) PWt 1 is all-India wholesale price index for rice for the period January-June, i.e., the last six months of the previous crop year; 1/ (iii) IPt is the input price index (5 States) for the crop year t; and (iv) RNt is the all-India rainfall index 2/ for rice, for the crop year t. The last term of the equation (Q 1) comes from the reduced form of a Nerlovian adjustment lag model (Nerlove, 1958 and Krishna 1963). 62. The estimated rice regression corresponding to equation (6) is given in Table 16. Every explanatory variable is significant and the speci- fication explains 87 percent of the variance of output. The short-run price elasticity turns out to be 0.45. The implicit long-run elasticity is 0.85. 1/ This lag is chosen after a trial with alternative lag periods (3 months, 9 months and 12 months of the previous crop year). 2/ We are grateful to S.K. Ray (1977) for the rainfall index. TABLE 16 RICE OUTPUT REGRESSION INDIA 1957-58/1970-71 Dependent Regression Coefficients 2 Elasticity Variable PWt- / t RN(t) Q(t-1) R PWt-/ 'Pt RN(t) Q(t-1) SR LR Q (t) 15930.30 222.99 0.47 0.87 0.45 0.85 0.62 0.45 (3.72) (4.08) (3.81) Q(t) Current output. Q(t-1) Output lagged one year. RN(t) Rainfall in the current year. PW :-I Index number of wholesale prices for six months (January-June), lagged one period Pwt-l (1965-55 = 100) IPt Index number of input-prices (1965-66 - 100) for the current year. SR Short run. LR Long run. t-values are shown in the parentheses. - 40 - 63. As expected, the estimated post-independence (short-run) output elasticity is greater than the pre-indenpendence and post-independence acreage elasticities. This should be so when yield is growing significantly. The acreage elasticities average about 0.35 as noted above. But the all-India output elasticity is about 0.45. This is, of course, no more than a very rough comparison because the acreage response elasticity with respect to the inter-crop price ratio is strictly not comparable with the output response elasticity with respect to the output-input ratio. 64. These results have two implications. First, an all-India rice output elasticity with respect to the output-input price ratio of the order 0.45 implies that in an average year a 10 percent increase in the price ratio would produce ceteris paribus a 4.5 percent output growth (i.e., an increase of about 1.6 million tons above the current base output). Second, it is obviously important to maintain input prices as well as procurement prices at incentive levels. For it is the ratio between the market price and the input price index which influences production, given the rainfall conditions. 65. For wheat, the specification of the supply function is: (7) Qt = f(PW t-/IPtV t- where Vt 1 is the ratio of the irrigated area in wheat to the total lagged one year in wheat. 66. From the estimated wheat equation (Table 17, Equation 3) the elasticity of output with respect to the output-input price ratio turns out to be 0.59 and the elasticity with respect to the irrigation ratio 0.81. 67. In order to verify additivity, acreage and yield elasticities have been estimated from equations 17.1 and 17.2 in Table 17. These equations TAB LE 17 WHEAT SUPPLY RESPONSE REGRESSIONS (1957-58/1969-70) Equation Dependent Regression Coefficients -2 Vlasticlty Number Variable Constant Ptl Vtl R Pt-i Vt_ INDIA 1731 It 7.764 3.285 0.078 0.673 0.221 0.211 (5.821) (2.248) (1.482) 17.2 Y 0.779 3.281 0.139 0.939 0.338 0.575 t (0.696) (2.675) (3.146) 17.3 Q -5.023 7.996 0.273 0.820 0.594 0.812 (-1.927) (2.796) (2.640) PUNJAB 17*4 At 0.735 0.498 0.715 0.840 0.276 0.269 (3.255) (2.914) (1.266) 17.5 Yt -0.305 0.690 1.840 0.891 0.432 0.781 (-0.967) (2.389) (2.239) 17.6 Q -1.865 2.207 3.770 0.874 0.824 0.954 (-2.023) (3.164) (1.635) At : Current acreage. At yt t Yield in quintal/hectare. at s Current output. Pt 1 : Index of wholesale price (base 1965-66 = 100) in pre-sowing period (October-September) t-1~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~- lagged one year, deflated by input cost index for current year. Vt 1 : Ratio of irrigated wheat acreage to total wheat acreage, lagged one year. The equations are linear in form. t-values are shown in parentheses. - 42 - have the same explanatory variables as the output equation. The price elasticity of acreage given by equation 17.1 is 0.22 and the price elasticity of yield given by equation 17.2 is 0.34. These elasticities add up to 0.56 which is very close to the price elasticity of output (0.59) estimated from equation 17.3. 68. It is interesting to note that the price elasticity of acreage (0.22) for the period covering the late 1950s and the 1960s is close to the upper limit of the range of elasticities (0.08 to 0.24) estimated in earlier studies. 1/ But as we hypothesized, the output elasticity is much higher than (almost three times) the acreage elasticity. The difference between the output elasticity and the acreage elasticity is accounted for by the elasticity of yield. 69. The elasticity of output with respect to the irrigation ratio (0.81) is also roughly equal to the sum of the acreage elasticity (0.21) and the yield elasticity (0.58) with respect to the irrigation ratio. The irrigation coefficient is clearly signficant in the yield and output equations but not in the acreage equation. This result suggests that yield now responds to an increase in the irrigation ratio much more than (almost three times as much as) acreage. Thus, the effect of the growth of irrigation capacity on output accrues mainly via the resulting yeild increases. 1/ Tie one-period elasticities remain comparable even through the pre- Independence equations had the lagged acreage variables derived from a Nerlovian adjustment model and the new equation (17.1) has the lagged irrigation ratio. The reason is that these variables (Q 1 and Vt ) are highly correlated. t- - 43 - 70. Since Punjab is the most important wheat-growing State, equations similar to the all-India equations have been estimated for the State (Table 17). The price elasticities of acreage, yield and output are, respectively, 0.28, 0.43 and 0.82, as compared with 0.22, 0.34 and 0.59 for the country as a whole. The additivity condition is again fully verified. But every elasticity is somewhat higher in the Punjab than in the country as a whole, because of the higher technological level and the greater commercialization of Punjab agriculture. 71. The hypothesis that input-price variations have a significant effect on acreage, yield and output is also verified, for the coefficients of the output-input price ratio are clearly significant in all equations. 72. It will be noted that in the case of wheat we have used irrigation ratio as the non-price variable, rather than rainfall and the lagged dependent variable derived from a Nerlovian adjustment mechanism. The reason is that irrigation ratio is the most important proxy for all technological changes which shift the supply function. It is the critical pre-condition for the use of all other modern inputs such as hybrid seeds, chemical fertilizers, pesticides, equipment, etc. A real causal variable which obviously has a strong influence on output from year to year is preferable t*o the lagged dependent variable which reflects a hypothetical adjustment or expectation process. 73. The estimated wheat equations bring out an important fact relevant for policy, namely, that the responsivenes of output to the irrigation ratio is usually higher than the responsiveness to relative price changes. The irrigation elasticity of all-India output (0.8) is about 33 percent higher than the price elasticity (0.6). This finding shows the pointlessness of - 44 - the superficial controversy which often turns up about the relative price or technological variables alone being important determinants of output. It is clear that both are important but the responsiveness of output to technological change is, at the margin, higher than the responsiveness of output to relative price movements - though one can legitimately argue that the rate at which farmers adopt technological innovations, other than the expansion of public sector irrigation capacity which is independent of farmer decisions and price movements, is itself a function of the prevailing output- input price ratio regime. The policy implication is that for sustaining a satisfactory rate of output, policy-makers must maintain both a satisfactory price regime and a rapid growth of irrigation capacity, input supplies and technical knowledge. 7. CONCLUSIONS AND POLICY IMPLICATIONS 74. The main findings of the preceding analysis can be summarized as follows. First, it is clear that the procurement price actually fixed by the government has been influenced by the open market wholesale price in the recent past, except in abnormal years, and not by the cost of production. The elasticity of the procurement price with respect to the average wholesale price in two preceding years has been 0.68 in the case of wheat and 0.65 in the case of rice. Across States, it varied within the narrow range of 0.5 to 0.9 in the case of wheat as well as rice. 75. In the 1950s, the wheat procurement price did not cover the cost of production in important wheat-producing States but in the late 1960s, it allowed a margin of profit above full cost. - 45 - 76. In the case of rice, too, the procurement price remained below the cost of production of rice in the major rice-producing States of the South and in West Bengal in the 1950s and 1960s. But the difference narrowed in the 1970s. An upward adjustment of southern and West Bengal prices has clearly been overdue. In the northern and eastern States, on the other hand, the rice procurement price has been significantly higher than the cost of production. 77. The procurement price almost always remained below the market harvest price and the market wholesale price. And whereas the market prices did cover the cost of production in most State and periods, the procurement price did so only in some regions. 78. Second, the volume of procurement has been systematically responsive (a) to the volume of output and (b) to procurement price. With respect to output the elasticity of rice procurement has been about unity, and wheat procurement as high as 2.3. With respect to procurement price or the procure- ment price/market ratio the elasticity has been 0.4 in the case of rice and 2.7 in the case of wheat. In the case of rice deficit States the responsiveness of procurement to the procurement/market price ratio has been found to be even higher than in the surplus States. The policy implication of these findings is that, given the level of output a satisfactory procurement price is critical for government obtaining the desired quantity. And if an attractive procure- ment price/market price ratio is maintained, procurement levels can be raised even in deficit States. 79. Third, zoning restrictions have been shown to increase interstate price dispersion nearly 2 to 2-1/2 times and thus to hurt the producers of surplus States and the consumers of deficit States. But this perverse effect of zoning is moderated in good crop years when greater open market availability tends to reduce price dispersion. - 46 - 80. Fourth, it is shown that the short-run sensitivity of wheat output to relative price movements has increased in the post-independence period to 0.6 from about 0.2 in the pre-independence period. This is due to the rfact that in the early period acreage responded to price movements but yield remained stagnant; but in recent years yield response has been added to acreage response. Production is now sensitive not only to intercrop price ratios but also, and much more, to the output-input price ratio. And, as in the past, the effect of the technological supply shifters, of which irrigation capacity is the most basic, on output, remains much stronger than the effect of price movements. Thus for maintaining satisfactory output growth, steady growth of irrigation capacity and inputs supplies is as essential as a favor- able output/input price regime. 81. The production of rice too has been more sensitive to the relevant relative price ratio in the recent period than in the pre-independence period. The post-independence short-run elasticity of output with respect to the price ratio comes out to be 0.45 as compared with 0.35 in the pre-independence period. Rice output is also significantly sensitive to rainfall - the elasticity being 0.65. 82. On the whole it seems that for maintaining incentives for production and procuring sufficient grain for public distribution and buffer stocks the best policy-mix would be a reasonably high procurement price, covering the full cost of production (C), and moving up with the market wholesale price, in a regime where free movement of grain is slowed throughout the country. - 47 - APPENDICES Appendix A Appendix B Table A.1 Table A.2 Table A.3 Table A.4 Table A.5 Table A.6 Bibliography - 48 - APPENDIX A ABSTRACTS FROM THE AGRICULTURAL PRICES COMMISSION'S (APC) REPORTS REGARDING FACTORS INFLUENCING THE PROCUREMENT PRICES RECOMMENDED Cost of Cultivation "...the minimum price should be related to the cost of cultivation, properly defined and measured." (APC Report May-July 1965, p. 2). "The Commission has taken cognizance of the fact that, mainly because of external factors, the expenses of cultivation have been affected since last year. It feels that minimum support prices should reflect these altered conditions." (APC Report December 1966-March 1967, p. 3). "The formulation of price policy can be severely handicapped by the absence of adequate data on either cost of production or level of output individual years and for different crops and regions of the country." "...Faced with this difficulty, the Commission has often had to rely on indirect measures for.reaching a judgment on the level of costs." (APC Annual Report 1967/68, p. 13). "...The price policy for foodgrains should embody a structure of minimum support prices which is sustainable over time. No doubt, such support prices should not only cover the cost of production but, in addition, provide for adequate incentives which may help maintain a socially desirable cropping pattern and be also conducive to the introduction of improved agricultural practices." (APC Report April 1968, p. 1). - 49 - "...higher foodgrain prices, in so far as they find reflection in increased agricultural wages, have the effect of enhancing the cost of cultivation, thus feeding the cost-price spiral in agriculture." (APC Report September 1968, p. 14). "...The Commission is beset with the difficulty that the available estimates of cost of production of agricultural commodities suffer from various deficiencies to which the Commission has had occasion to refer in the past; more particularly, those for coarse grains, besides being fragmentary, display a high degree of erraticity ... with the result that representative and dependable estimates of unit cost production of Kharif cereals are not available at this stage." (APC Report May 1963, p. 3). "In the immediate context of the decision with regard to the appro- priate level of the procurement prices for the Kharif cereals to be marketed in the 1974/75 season, the recent increases in the prices of fertilizer and other inputs need to be taken into account." (APC Report September 1974, p. 6). Market Wholesale Prices "Despite the general buoyancy of market prices which, for the last two seasons, have stayed away above the announced minimum price for Kharif cereals, possibility, however, exists that certain pockets may energe where, for some crops, prices may be unduly depressed in particular seasons "By and large it nonetheless remains true that, for the coming year, minimum support prices can play only a marginal role in stimulating agricultural growth. The dominating influences will be (a) the open market - 50 - prices, (b) the purchase prices fixed for government procurement, and (c) the relative price movements between Kharif cereals and competing crops in different States." "The downward adjustment in market prices discernible during 1968 needed to be reflected in the procurement prices." (APC Annual Report 1968/69, p. 2). "...it has necessarily to take note of the trend in market prices... downward shift - even if mild - has occurred in the market prices of foodgrains during 1968." (APC Report September 1963 - April-September 1969, p. 11). "Since the immediate effort ought to be to contain the incipient pressures which the prices of some grains have started experiencing in recent months, it would not be advisable to effect an increase in the general level of procurement prices lest it should accentuate these pressures." (APC Report September 1971, p. 13). "In determining the level of the procurement prices, the Commission has considered the current trend of the market prices, the need to contain their rise ... in all cases of increase in paddy prices, procurement prices for rice may be upgraded correspondingly. As for the Kharif coarse cereals, the existing procurement prices may be continued for the 1972/73 season; over and above these, an ad hoc premium of Rs 2 per quintal may be given." (APC Report September 1972, pp. 5-6). "The current year has witnessed an unprecedented increase in food- grain prices, the pace of rise having been almost double of what it was last year. The index number of foodgrain prices in July 1974 has recorded an increase of 35.6 percent over its level in July 1973 as against an increase of 18.8 percent in July 1973 over that in the corresponding month of 1972." - 51 - "...an important part of the overall increase was the result of the sharp increases affected in the administered prices of foodgrains and their impact on the open market prices..." (APC Report September 1974, pp. 2-3). Interstate Price Dispersal "There is little doubt that the zonal arrangements have introduced an element of disparity in the prices of even the same foodgrain as between neighboring States. "For several States, the prices recommended by the Commission were adjusted upwards by the Government while announcing the procurement prices during the last two years. Moreover, in a number of States, and on a number of occasions, particularly during the last year, procurement prices as fixed by the Government were raised sharply in the middle of the season... In several States, "bonus" payments over and above the procurement prices were also announced. "Such mid-season raising of procurement prices has put into disarray the structure of interstate parity of prices which the Commission has been endeavoring to bring about." (APC Report September 1967, pp. 8, 10-11). "In the fixation of procurement prices, another guiding principle is the desirability of narrowing the interstate disparities... Unfortunately, in the course of the last two years, the prices actually fixed have resulted in widening such disparities. For example, the coefficient of variation of procurement prices fixed by the State Governments for common wheat increased from 4.0 in 1966/67 to as much as 10.9 in 1967/68." (APC Report March 1968, p. 8-9). - 52 - "As a result of the factors underlying the fixation of procurement prices during the last two years, a wide hiatus has emerged in the prices of the same grain as between the neighboring States. In making its recommenda- tions on the procurement prices for Kharif cereals for 1967/68, the Commission has urged a narrowing of the range of prices. For example, in the case of paddy, it had recommended procurement prices ranging from Rs. 42.50 to Rs 46.00 per quintal for the different States. The price as finally fixed by the State Governments, however, ranged from Rs 45.00 to Rs 56.25 ... inclusive of the bonus, the range was from Rs 45.00 to 65.00." (APC Report September 1968 - April-September 1969, p. 11-12). "An important element contributing to interstate disparities in the procurement prices of the same grain has been the practice on the part of some State Governments of paying a bonus over and above the procurement price. This bonus, which is tantamount to an unilateral increase in the procurement price by a State Government beyond what has been concurred to by the Union Governent, also raised the price to the consumer." (APC Report September 1968 - April-September 1969, p. 16-17). "... the Commission had pointed out that the grant of bonus by the State Governments ... accentuated the interstate disparities in the procurement prices of the same grain." (APC Report August 1970, p. 43). Intercrop Disparity "Shifts in intercrop ratios perform a useful function of allocating land and other production factors in response to the relative changes, during and between the seasons, in the supply and demand positions of different - 53 - agricultural commodities. Variability is thus the essence of intercrop ratios of market prices, but as mentioned earlier, the minimum prices have to be more stable." (APC Report May-July 1965, p. 2). "In the world market, the export price of wheat is generally quoted below the export price of rice. In India to, the market prices of wheat have been traditionally lower than those of rice. A similar relationship should, therefore, be normally maintained between the procurement prices of the two cereals. This indeed was the consideration which had weighed with the Commission in recommending in March 1967 the procurement prices for wheat for the major producing States at levels lower than the procurement prices of rice fixed by the Government earlier. But the actual prices for procurement announced by the Government in April 1967 were at levels much higher than the corresponding procurement prices for rice." (APC Report September 1967, p. 11). "The Commission would be in favor of the procurement prices for wheat for the 1968/69 season begin set at levels which would reflect in general the trends in open market prices. While deciding on these levels, the normal relationship between the prices of wheat and rice has also to be borne in mind. Until the beginning of last year, the procurement prices for wheat were gen- erally fixed at levels lower than those for rice; this was consistent with the relative prices in the open market for the two cereals. As a result of the unusual developments last year, this relationship was distorted and the procure- ment price for wheat in several States stood higher than the corresponding prices for rice." (APC Report March 1968, p. 8). - 54 - APPENDIX B Full Cost "C" The cost concept 'C' includes: All cash and kind expenses actually incurred, less rent. It comprises wages of hired human labor, value of bullock labor (owned or hired or both), seeds, manures, fertilizers, insecticides and pesticides, interest paid on capital, depreciation and other repair charges. Rental paid for leased in land. Rental value of owned land plus interest on fixed capital plus land revenue on owned land. Imputed value of family labor. Surplus and Deficit States Surplus States have been identified as States where per capita cereal production is higher than the all-India average and per capita net supply of cereals from the Central Government Pool is less than the all- India average. The deficit States are identified as those were per capita cereal production is lower and per capita net supply from the Central Pool is higher than the all-India averages. The reference years are 1965/66, 1970/71 and 1974/75. Surplus States include Andhra, Haryana, Madhya Pradesh, Orissa, Punjab and Uttar Pradesh. Deficit States include Assam, Bihar, Kerala, Maharashtra, and West Bengal. - 55 - TABLE A.1 PROPORTION OF ALL-INDIA WHEAT OUTPUT AND PROCUREMENT IN DIFFERENT STATES (AVERAGE 1972/73 TO 1974/75) Proportion Proportion of of Output Procurement State (Percent) (Percent) Uttar Pradesh 28.9 17.06 Punjab 22.7 59.55 Madhya Pradesh 10.3 3.96 Bihar 9.6 0.41 Haryana 8.5 14.00 Rajasthan 7.6 3.55 Maharashtra 3.4 0.40 Gujarat 3.1 0.36 Andhra Pradesh 0.1 0.00 Others 5.8 0.71 Total 100.0 100.00 Source: Bulletin on Food Statistics, Ministry of Agriculture and Irrigation, Government of India, 1976. -56 - TABLE A.2 PROPORTION OF ALL INDIA PADDY OUTPUT AND PROCUREMENT IN DIFFERENT STATES (AVERAGE 1972-73/1974-75) Proportion of Proportion of States Output Procurement (Per Cent) (Per Cent) * Surplus States: 1. Andhra 12.6 18.4 2. Orissa 9.4 4.6 3. Uttar Pradesh 8.6 9.0 4. Madhya Pradesh 7.4 5.8 5. Punjab 2.7 25.7 6. Haryana 1.1 9.4 Sub Total 41.8 72.9 Deficit States: 1. West Bengal 14.6 5.4 2. Bihar 10.7 1.7 3. Assam 5.0 3.2 4. Maharashtra 3.0 3.0 5. Kerala 1.4 Sub Total 36.5 14.7 Tamil Nadu 12.4 8.1 Total: 90.7 95.7 * For definition, see Appendix B. Source: Bulletin on Food Statistics, Ministry of Agriculture, Government of India, 1975, 1976. - 57 - TABLE A.3 RICE: AVERAGE LABOR COST (MANDAYS) OF CULTIVATION Proportion of Human Labor Mandays Cost to Total Per Cost 'C' State (District) Period Hectare (Percent) (1) (2) (4) A. Procurement Price < Cost of Production Andhra (Cuddapah) 1967/68 to 1969/70 143.71 (19.38) Kerala (Alleppey, Quilon) 1962/63 to 1964/75 147.87 (33.53) Tamil Nadu (Thanjavur) 1967/68 to 1969/70 129.60 (30.25) West Bengal (24 Parganas) 1954/55 to 1956/57 111.35 (48.25) West Bengal (Hooghly) 1954/55 to 1956/57 130.39 (46.70) West Bengal (Hooghly), 24 Parganas combined) 1954/55 to 1956/57 120.09 - B. Procurement Price > Cost of Production Assam (Nowgong) 1968/69 to 1969/70 74.50 (29.49) Orissa (Cuttack) 1967/68 to 1969/70 112.60 (34.60) Orissa (Sambalpur) 1957/58 to 1959/60 101.10 (36.94) Madhya Pradesh (Raipur) 1962/63 to 1964/65 99.50 (33.85) Western U.P. (Muzaffarnagar) 1966/67 to 1968/69 69.34 (16.89) Bihar (South Monghyr) 1957/58 to 1959/60 279.87 /a (23.22) /a In view of the level and the proportion of human labor cost to total cost 'C' and the current wage rate, this figure appears to be unreliable. Source: Studies in the Economics of Farm Management (Different States and periods). - 58 - TABLE A.4 RICE: FARM HARVEST PRICE AND COST OF PRODUCTION Average Excess Number of Harvest Price of Over Cost of State (District) Period Observations Production (1) (2) (3) (4) A. Procurement Price < Cost of Production Andhra (Cuddapah) 1967/68 to 1969/70 3 -20.0 Andhra (West Godavari) 1957/58 to 1959/60 3 - 0.3 Bihar (South Monghyr)/b 1958/59 to 1959/60 2 -39.5 West Bengal (24 Parganas) 1954/55 to 1956/57 3 - 3.7 West Bengal (Hooghly) 1954/55 to 1956/57 3 -35.6 U.P. (Deoria) 1966/67 to 1968/69 3 -28.3 Andhra (Cuddapah) 1971/72 1 +29.9 Tamil Nadu (Salem, Coimbatore) 1954/55 to 1956/57 3 +11.6 Tamil Nadu (Thanjavur)/c 1969/70 1 + 5.0 Assam (Nowgong) 1968/69 to 1970/71 3 +20.0 Tamil Nadu (Coimbatore) 1970/71 to 1972/73 3 + 4.0 Orissa (Sambalpur)/a 1957/58 to 1959/60 3 +17.3 Orissa (Cuttack)/a 1967/68 to 1969/70 3 +55.7 1971/72 to 1972/73 2 +52.7 Bihar (Shahabad) 1960/61 to 1962/63 3 +28.7 Kerala (Alleppey, Quilon) 1962/63 to 1964/65 3 +10.7 Madhya Pradesh (Raipur) 1962/63 to 1964/65 3 +27.0 U.P. (Meerut, Muzaffarnagar)/d 1956/57 to 1968/69 1 +37.0 U.P. (Muzaffarnagar) 1966/67 to 1968/69 3 +47.3 Punjab (Ferozepur) 1968/69 to 1969/70 2 +25.0 /a For Orissa wholesale prices during the harvest period (January and February) have been used as the farm harvest price data was not available. /b The year 1957/58 has been dropped as it was an abnormally bad crop year. /c Cost data are available for the period 1967/68-1969/70, but the farm harvest prices are not available for the period 1966/67-1968/69. The wholesale price during the harvest period being controlled we have presented only one observation here. /d Cost data are available for the period 1954/55-1956/57, but farm harvest prices are not available for the years 1954/55-1955/56. Hence we have only one observation. Sources: Various issues of Farm (Harvest) Prices for Principal Crops, Agriculture Prices in India, Agricultural Situation in India, August 1975, Studies in the Economics of Farm Managment (Various States and periods). - 59 - TABLE A.5 RICE: PROCUREMENT PRICE AND FARM HARVEST PRICE Average Excess of Procurement Price Over Farm Number of Harvest Price State Period Observations (Percent) Andhra /a 1964/65 to 1974/75 11 -17.9 Assam 1951/52 to 1953/64 3 -13.0 1964/65 to 1974/75 11 - 9.7 Bihar 1951/52 to 1953/64 3 -24.7 1964/65 to 1974/75 11 -18.7 Orissa /b 1951/52 to 1953/54 3 - 4.3 1964/65 to 1974/75 11 -13.0 Kerala /a 1964/65 to 1973/74 10 -50.0 Madhya Pradesh /a 1952/53 to 1953/54 2 -15.0 1964/65 to 1974/75 11 - 8.1 Tamil Nadu /c 1951/52 to 1953/54 3 -38.3 1964/65 to 1965/66 7 -11.7 1969/70 to 1973/74 Uttar Pradesh /a 1964/65 to 1974/75 11 -11.6 Maharashtra /a 1964/65 to 1974/75 11 -43.5 Haryana /d 1967/68 to 1972/73 6 -20.0 Punjab /e 1964/65 to 1972/73 9 -12.0 West Bengal /f 1951/52 to 1953/54 3 -21.7 1964/65 to 1973/74 10 -26.8 /a Procurement price data for the period 1951/52-1953/54 or one of these years are not available. /b In the absence of farm harvest price data, the wholesale harvest prices (average of January and February) for the period 1964/65-1974/75 have been used. /c Farm harvest price data for the period 1966/67-1968/69 and 1974/75 are not available. /d Haryana State came into existent from November 1966. For the years 1973/74 and 1974/75, farm harvest price data are not available. /e Farm harvest price data for the periods 1951/52-1953/54 and 1972/73- 1974/75 are not available. /f Farm harvest price data for the year 1974/75 is not available. Sources: Various issues of Bulletin on Food Statistics, Farm (Harvest) Prices of Principal Crops, Agricultural Prices in India, and Agricultural Situtation in India, August 1975. - 60 - TABLE A.6 RICE: PROCUREMENT PRICE AND ANNUAL AVERAGE WHOLESALE PRICE Average Excess of Procurement Number of Price Over State Period Observations Wholesale Price Andhra /a 1964/65 to 1974/75 11 -21.5 Assam /b 1951/52 to 1953/54 3 -11.7 1964/65 to 1966/67 1968/69 to 1974/75 10 -12.7 Bihar 1951/52 to 1953/54 4 -20.7 1964/65 to 1974/75 11 -37.3 Madhya Pradesh 1951/52 to 1953/54 3 -23.7 1964/65 to 1974/75 11 -17.5 Orissa /b 1952/53 to 1953/54 2 -14.0 1964/65 to 1974/75 11 -21.0 Uttar Pradesh /a 1952/53 to 1953/54 2 -25.5 1964/65 to 1974/75 11 -34.6 West Bengal /b 1952/53 to 1953/54 2 -16.5 1964/65 to 1974/75 11 -39.0 Kerala /a 1966/67 to 1974/75 9 - 9.2 Punjab 1951/52 to 1953/54 3 - 6.0 1964/65 to 1974/75 11 -14.0 Haryana /d 1967/68 to 1974/75 8 -25.0 Maharashtra /a 1964/65,1966/67 to 1974/75 10 -15.4 Tamil Nadu /c 1952/53 to 1953/54 2 -52.0 Tamil Nadu /d 1964/65 to 1973/74 10 + 5.1 /a Procurement price data for the period 1951/52-1953/54 are not avail- able for Andhra, Kerala and Maharashtra. /b Wholesale price data for the year 1951/52 is not available for Orissa, U.P., West Bengal and Tamil Nadu. For Assam the year 1967/68 is ommited as the wholesale price was controlled by the State. /c In Maharashtra for 1965/66 no wholesale prices were recorded. /d Wholesale prices in Tamil Nadu are under State Control since 1964/65. For the year 1974/75 the price figure is available only for five months; hence it has not been included. Sources: Various issues of Bulletin on-Food Statistics, Agricultural Prices in India, and Statistical Abstract (Different States). - 61 - Bibliography 1. Bhagwati, J. and Chakravarty, S. (1969): "Contributions to Indian Economic Analysis: A Survey," American Economic Review, Vol. 59, September 1969 (A Supplement). 2. Cummings, J.T. (1975): "The Supply Responsiveness of Indian Farmers in the Post-Independence Period: Major Cereal and Cash Crops," Indian Journal of Agricultural Economics, vol. 30. No. 1, January-March 1975, pp. 25-40. 3. Directorate (Bureaus) of Economics and Statistics of State Governments: Statistical Abstracts. (Various Annual Issues.) 4. Government of India, Ministry of Agriculture and Irrigation, (1978): Agricultural Prices in India, Delhi. 5. Government of India, Ministry of Agriculture and Irrigation: Reports of the Agricultural Prices Commission. (Various Issues). 6. Government of India, Ministry of Agriculture and Irrigation, Directorate of Economics and Statistics: Agricultural Situation in India (Monthly). 7. Government of India, Ministry of Agriculture and Irrigation, Directorate of Economics and Statistics: Bulletin on Food Statistics, Annual Issues. 8. Government of India, Ministry of Agriculture and Irrigation, Directorate of Economics and Statistics: Farm (Harvest) Prices of Principal Crops. 9. Government of India, Ministry of Agriculture and Irrigation, Directorate of Economics and Statistics: Indian Agriculture in Brief, 14th Edition, New Delhi. 10. Government of India, Ministry of Agriculture and Irrigation, Directorate of Economics and Statistics: Studies in the Economics of Farm Management. 11. Government of India, Ministry of Finance: Economic Survey 1976-77, New Delhi, 1977. 12. Krishna, Raj (1963): "Farm Supply Response in India - Pakistan: A Case Study of the Punjab Region," The Economic Journal, Vol. 73, No. 291, September 1963, pp. 477-487. 13. Krishna, Raj (1965): "Rice Zone Policy: A Note of Dissent," in Ministry of Food and Agriculture, Government of India, Report of the Agricultural Prices Commission on Price Policy for Kharif Cereals for 1965/66 Season, May-July 1965. - 62 - 14. Krishna, Raj (1967): "Agricultural Price Policy and Economic Development" in Herman M. Southworth and Bruce F. Johnston, (eds.), Agricultural Development and Economic Growth, Cornell University Press, Ithaca, New York, 1967. 15. Maji, C.C. et al. (1971): "Dynamic Supply and Demand Models for Better Estimations and Projections: An Econometric Study for Major Foodgrains in the Punjab Region," Indian Journal of Agricultural Economics, Vol. XXBI, No. 1, January-March 1971. 16. Narian, Dharam (1965): Impact of Price Movements on Areas Under Selected Crops in India: 1900-1939, Cambridge, U.K. 1965. 17. National Council of Applied Economic Research (1962): Long Term Projections of Demand and Supply of Selected Agricultural Commodities, 1960-61 to 1975/76, New Delhi, 1962. 18. National Council of Applied Economic Research (1969): Projections of Demand and Supply of Agricultural Commodities, New Delhi, 1969. 19. Netlove, Marc (1958): The Dynamics of Supply-Estimates of Farmers' Response to Price, John Hopkins, Baltimore, 1958. 20. Parikh, Ashok (1972): "Market Responsiveness of Peasant Cultivators: Some Evidence from Prewar India," Journal of Development Studies, Vol. 8, No. 2, January 1972, pp. 291-306. 21. Ramesh, D. (1965): "Long and Shortrun Price Elasticities of Acreage Under Crops", Agricultural Situation in India, Government of India, Ministry of Food and Agriculture, August 1965, pp. 319-324. 22. Rao, K.P.C. and Pandey, V.K. (1976): "Supply Response of Paddy in Andhra Pradesh," Indian Journal of Agricultural Economics, Vol. 31, No. 2, April-June, 1976, pp. 46-53. 23. Ray, S.K. (1977): Variations in Crop Output: An Attempt to Analyse and Estimate, the Effects of Weather and Prices on the Output of Major Food and Commercial Crops of India: (Mimeographed), Institute of Economic Growth, Delhi. 24. Singh, I.J. and Kumar P.: "Impact of Price and Price Variability on Acreage Allocation in Haryana," Indian Journal of Agricultural Economics, Vol. 31, April-June 1976, pp. 31-37. 25. Sud, L. and Kahlon, A.S.: "Estimation of Acreage Response to Price of Selected Crops in Punjab State," Indian Journal of Agricultural Economics, Vol. 24, July-September 1969, pp. 46-50. RECENT PAPERS IN THIS SERIES No. TITLE OF PAPER AUTHOR 350 Energy Options and Policy Issues in D. Fallen-Bailey Developing Countries T. Byer 351 Gtowth and Equity in Semi-Industrialized J. Bergsman Countries 352 Capital Flows and Developing Country Debt J. Katz 353 Trade Policy for Developing Countries D. Keesing 354 Development Problems of Mineral-Exporting G. Nankani Countries 355 The Global Framework R. Cheetham, S. Gupta A. Schwartz 356 The Distribution of Income in Brazil G. Pfefferman, R. Webb 357 Estimating Shadow Prices for Colombia in an W. Schohl Input-Output Table Framework 358 Inter-Country Comparison of "Real" (PPP) P. Isenman Incomes: Revised Estimates and Unresolved Questions 359 Price Distortions in Agriculture and Their E. Lutz Effects: An International Comparison M. Bale 360 Costs and Benefits of Agricultural Research: G.E. Schuh The State of the Arts H. Tollini (consultants) 361 Investment in International Agricultural G. Scobie Research: Some Economic Dimensions (consultant) 362 Identification and Appraisal of Rural Roads H. Beenhakker Projects A. Chammari. 363 Small Enterprises in African Development: J. Page (consultant) A Survey 364 Income, Consumption and Poverty in Thailand, 0. Meesook 1962/63 to 1975/76 365 A Survey and Critique of World Bank Supported R. Marris Research on International Comparisons of Real (consultant) Product -2- No. TITLE OF PAPER AUTHOR 366 Paradigms in the Study of Urban Labor Markets D. Mazumdar in LDCs: A Reassessment in the Light of an Empirical Survey in Bombay City 367 Incentives for Resource Allocation: S. Acharya A Case Study of Sudan 368 Why the Emperor's New Clothes Are Not Made D. Morawetz in Colombia (consultant) 369 Economic and Social Analysis of Projects K. Cleaver and of Price Policy: The Morocco Fourth Agricultural Credit Project 370 The Tokyo Round and the Developing Countries B. Balassa 371 Bus Ownership and Efficiency in Urban Areas A. Walters C. Feibel 372 The Tokyo Round: Results and Implications R. Kemper for Developing Countries (consultant) 373 Malnutrition: Some Measurement and Policy Issues T.N. Srinivasan 374 The Prevalence of Calorie Deficient Diets S. Reutlinger in Developing Countries H. Alderman (consultant) 375 The Design of Organizations for Rural F. Lethem, B. Thoolen Development Projects - A Progress Report W. Smith (consultant) 376 Fostering the Capital-Goods Sector in LDCs: H. Pack (consultant) A Survey of Evidence and Requirements 377 Macroeconomic Implications of Factor H. Pack (consultant) Substitution in Industrial Processes 378 Household Income or Household Income per Capita G. Datta in Welfare Comparisons J. Meerman 379 Managing Information for Rural Development: G. Deboeck Lessons from Eastern Africa B. Kinsey 380 Women in the Urban Labor Markets of Africa: N. Shields The Case of Tanzania