This paper analyses the process of growth of the agrarian economy of West Bengal from 1980-81 to 2002-03. There is a significantly negative trend break in 1992-93, which was the beginning of the liberalisation era in the Indian economy. The entire time period is divided into two sub-periods, namely, 1980-81 to 1991-92 and 1992-93 to 2002-03. We use the method of computing simple exponential growth rates, kinked exponential growth rates and log quadratic estimates. The former two methods suggest growth and trend breaks, but the latter shows the extent of instability. All eight variables related to the agrarian economy of West Bengal, namely, area, production, yield, consumption of fertiliser, proportion of hyv, cropping intensity, institutional credit and land reform show a decline in growth from the first to the second sub-period. All variables except area and land reform register significant deceleration. The trend break was particularly sharp for production, yield and fertiliser use.
REVIEW OF AGRICULTUREEconomic & Political Weekly december 29, 200765Agrarian Impasse in West Bengal in the Liberalisation EraMaumita Bhattacharyya, Sudipta Bhattacharyya Maumita Bhattacharyya is at the Kumud Kumari Institution, Jhargram and Sudipta Bhattacharyya is at the Economics, Department of Economics and Politics, Visva-Bharati, Santiniketan. This paper is a revised version of a part of a MPhil dissertation of the first author submitted to Burdwan University, Burdwan. The authors are grateful to Kalyanbrata Bhattacharya (who supervised the dissertation) and to Manabendu Chattopadhyay for their very useful comments. The usual disclaimer applies (Email: sudipta.bh@gmail.com).This paper analyses the process of growth of the agrarian economy of West Bengal from 1980-81 to 2002-03. There is a significantly negative trend break in 1992-93, which was the beginning of the liberalisation era in the Indian economy. The entire time period is divided into two sub-periods, namely, 1980-81 to 1991-92 and 1992-93 to 2002-03. We use the method of computing simple exponential growth rates, kinked exponential growth rates and log quadratic estimates. The former two methods suggest growth and trend breaks, but the latter shows the extent of instability. All eight variables related to the agrarian economy of West Bengal, namely, area, production, yield, consumption of fertiliser, proportion of HYV, cropping intensity, institutional credit and land reform show a decline in growth from the first to the second sub-period. All variables except area and land reform register significant deceleration. The trend break was particularly sharp for production, yield and fertiliser use. The rural economy of West Bengal underwent a process of “an agrarian impasse” during the period 1950 to 1980 [Boyce 1987]. The period manifested an agricultural growth rate of 1.74 per cent per annum, which was below the growth rates of rural (2.31 per cent) and total population (2.42 per cent) (ibid). The rural economy of West Bengal, therefore, did not find a favourable environment to diversify itself during the period preceding the 1980s. Landlords and ‘jotedars’ dominated rural West Bengal and there was an acute inequality in land and non-land assets [Das 1998, Bhattacharyya 2001]. 1 A National PerspectiveThe first change in the political environment of West Bengal took place in 1967 and 1969-70 when the non-Congress parties briefly ruled the state. During the second United Front ministry, a “radi-cal land reform from below” took place, where local peasants recovered more than 5,00,000 acres of ‘benami’ land and distrib-uted it among the landless. This was a kind of forced land re-form, undertaken outside the governmental framework, more or less spontaneous through popular participation with the support of the United Front parties [Dasgupta 1984]. This was the back-ground against which the Left Front government came to power in 1977. The major contribution of the Left Front government was to successfully implement the existing laws. The policies of the agrarian reform of the Left Front government were, first, the “operation barga” or the policy of spot recordation of share ten-ants. Second, it included the distribution of ceiling surplus land to the landless or poor peasants. Operation barga was helpful in breaking the interlinkage [Bhattacharyya 2005], and weaken the exploitative institution of sharetenancy itself. It broke the stereotype stratification of sharecropping where sharecroppers belonged to the lower class and the landlord to the upper class [Bhattacharyya 2007b]. Third, the agrarian reform made guar-antees of the implementation of the minimum wage rate for agri-cultural labourers. These reforms were carried out in the frame-work of the panchayati raj or the grass root level self-governance of the people. [GOWB 2004]. The panchayat raj institution shows a viable alternative to anti-statist model of participatory govern-ance by the World Bank [Bhattacharyya 2007a].Some studies showed a high growth performance of West Ben-gal’s agriculture since the early 1980s, which overwhelmed theo-rists and policymakers and added a new dimension for controver-sy. Saha and Swaminathan (1994), using an index number series on aggregate agricultural production, claimed that, “the expo-nential growth rate for all West Bengal for the period 1981-82 to 1990-91 was an impressive 6.4 per cent per annum. The growth performance is striking in comparison to the past. The annual growth rate of agricultural output for the period 1965-80 was
REVIEW OF AGRICULTUREdecember 29, 2007 Economic & Political Weekly66Table 1: Simple Exponential Growth Rates (1980-81 to 2002-03)Variables R2 AdjStandardConstantt-valueRegressiont-valueF-test R Sq Error of Est Co-eff ValueArea 0.6260.6090.0288.6895.933*0.00525.933*35.203* (0.012) (0.003) Production 0.8410.8340.118.904189.958*0.0360910.556*111.428* (0.01)(0.004) Yield 0.8530.8460.09167.113180.124*0.031811.048*122.064* (0.039) (0.001) Fertiliser 0.929 0.925 0.13385.641 97.850* 0.069316.560* 274.240* (0.058) (0.004) HYV area 0.957 0.955 0.063 3.174 110.913* 0.0452 21.652* 468.791* (0.029) (0.002) Cropping 0.883 878 0.03 4.911 376.114* 0.012 12.602* 158896* intensity (0.013) (0.001) Credit 0.934 0.931 204 5.655 64.401* 0.11117.269* 298.215* (0.88) (0.006) Land reform 0.887 0.882 0.02 2.604 301.156* 0.008 12.840* 164.866* (0.009) (0.001) * Significant at 1 per cent level. Figures in parentheses show standard errors.Table 2: Simple Exponential Growth Rates (1980-81 to 1991-92 and 1992-93 to 2002-03)Variables PeriodR2 Adj StandardConstantt-value Regression t-value F-test RSquaredErrorCo-efficientValueArea Pd 1 0.526 0.479 0.02868.673 493.47* 0.008 3.333* 11.11* (0.018) (0.002) Pd II0.118 0.019 0.02758.765 493.35* 0.00291.095 1.2 (0.018) (0.003) Production Pd I 0.754 0.73 0.1218 8.783 117.13* 0.0564 5.540* 30.70* (0.075) (0.01) Pd II0.751 0.724 0.047 9.402 309.31* 0.02335.215*27.20* (0.03) (0.004) Yield Pd I 0.787 0.766 0.09997 113.88* 0.05086.082*36.99* (0.061) (0.008) PdII0.8660.8510.02727.556428.91*0.01987.632*58.26* (0.018) (0.003) Fertiliser Pd I 0.955 0.951 0.089 5.418 98.86*0.109 14.57*212.36* (0.055) (0.007) Pd II0.9 0.889 0.06556.524 153.96* 0.0563 9.018*81.33* (0.042) (0.006) HYV Pd I 0.938 0.932 0.05193.115 97.60*0.053612.352* 152.57 area (0.032) (0.004) PdII0.7870.7630.05163.822114.49*0.02845.764*33.22* (0.033) (0.005) Cropping Pd I 0.823 0.806 0.03214.877 246.59* 0.0183 6.823* 46.56* intensity (0.02) (0.003) Pd II0.911 0.902 0.01415.041 553.72* 0.01299.625*92.64* (0.009) (0.001) Credit Pd I 0.944 0.939 0.1462 5.375 59.73* 0.159 13.031* 169.80* (0.09) (0.012) PdII0.790.7670.16667.05265.47*0.09245.824*33.91* (0.108) (0.016) Land PdI0.6730.640.02522.597167.32*0.00964.535*20.57* reform (0.016) (0.002) PdII0.8510.8340.01322.694315.56*0.0097.159*51.25* (0.009) (0.001) * Significant at 1 per cent level. Figures in parentheses show standard errors.2.2 per cent”. The agricultural growth rate in West Bengal is the highest among all the Indian states. Though some doubt had been cast regarding the validity of the official data [Datta Ray 1994] based on which such a conclusion was derived, no effective challenge was thrown against the findings showing high growth performance of West Bengal. On the contrary, Sen and Sengupta (1995) using an alternative source of data(comprehensive cost of production studies, ministry of agriculture, government of India) reaffirmed “a significant trend-break in the growth rate during the 1980s”. Rawal and Swaminathan (1998) also supported the trend break in the early 1980s. Some economists tried to identify the factors responsible for the high growth rate in agriculture. According to Sen and Sen-gupta (1995), West Bengal stands as a case apart because in this case, the measured input growth does not appear to be capable of explaining the magnitude of the trend break in output growth. As a matter of fact, according to Sen and Sengupta, the rate of growth of fertiliser use, percentage of high yield variety(HYV) area, and irrigated area, unlike other eastern Indian states de-clined in West Bengal in the 1980s compared to the 1970s. This, in turn, means that unmeasured technological shifters were more important here or that land and panchayat reforms (car-ried out in West Bengal but not elsewhere) caused an increase in productivity not found elsewhere. Mukherji and Mukhopadhyay (1996) attempted to determine the impact of institutional chang-es on total factor productivity in the production of rice in West Bengal. According to them, it is difficult to figure out what would have spurred this growth by simply looking at the statistics on input use. The data do not suggest an increase in factor use cor-responding to output growth. The extent of farm mechanisation has been minimal in the state and has shown no massive increase in recent times. Mukherji and Mukhopadhyay (1996) maintained that the only one change that has been taking place in favour of higher yields is the institutional change characterised by the panchayati raj and land reforms. A few scholars have opined that agricultural growth in West Bengal can be successfully explained in terms of both the technological and institutional factors, such as the belated green revolution, personal initiative and private investment, operation barga, distribution of surplus land and panchayati raj [Sengupta and Gazdar 1998, Banerjee et al 2002]. In our view, the increased land productivity in West Bengal dur-ing the 1980s is the direct result of the increased labour inten-sity in family farms belonging to marginal and small category of farmers. As the land reforms created higher purchasing power, marginal and small farmers could access green revolution tech-nology unlike in the leading green revolution states in India.The trend break in the growth rate during the 1980s in West Bengal was associated with the opening up of the horizon of diversified rural activities [Chandrasekhar 1993; Arun Kumar et al 1995, Lieten 1990, Rogaly et al 1999; Bhattacharyya 1996, 2003a, 2003b]. As a result, some positive structural changes may be observed in recent years in West Bengal. In the early 1990s, India adopted the structural adjustment programme of the International Monetary Fund (IMF) and World Bank as a loan conditionality and had to change its macroeco-nomic policies to a great extent. Indian agriculture experienced some major changes, which appeared as a sudden shock to the peasantry. First food, fertiliser and credit subsidies were curbed to a great extent. As a result, the public distribution system(PDS) was virtually wiped out. Second, import liberalisation opened up the inflow of foodgrains from abroad at cheaper prices compared to the home market. The peasantry faced two problems. (a) higher costs of production owing to the withdrawal of the fertiliser sub-sidy; and (b) selling products at viable prices. In a country where about 60 per cent of rural population is a seller of foodgrains, it leads to a loss of income for these peasant households. Due to the government withdrawal of thePDS, there was a sudden decline
REVIEW OF AGRICULTUREEconomic & Political Weekly December 29, 200767in demand for foodgrains particularly from poorer households. Government economists and policymakers explained this fact as a sign of prosperity, where the rural population voluntarily demanded less foodgrains. Hence, policymakers, under market reforms, advised peasants to diversify their production from foodgrains to export-oriented crops.According to Patnaik (2005), the Indian economy like in the co-lonial period has already witnessed an inverse relation between an increase in export-oriented production and decline in the avail-ability of foodgrains. On the other hand, the global price crash in export commodities, particularly cotton, led to suicides of thou-sands of Indian farmers. The prevailing global and local deflation-ary situation is a part of the global low interest rate regime, which is favourable for global finance capital and the international credi-tor. As a result, all the expansionary and development activities in the economy declined with the obsessed target of lowering fiscal deficit under market reforms. There was an acute increase of ru-ral employment and massive increase in rural poverty. Between 1980-81 and 1989-90, foodgrains grew at the rate 2.85 per cent per annum, and crops grew at 3.19 per cent per annum. The rate of growth of population at this period was 2.1 per cent per annum. However, from 1990-91 to 2000-01 the foodgrain growth declined at 1.66 per cent per annum, while crops declined to 1.73 per cent per annum. For the first time after independence, the population growth rate (1.9 per cent per annum) remained above of foodgrain and overall crop growth rate (ibid). One has to address the West Bengal scenario against this na-tional perspective. Though West Bengal experienced success-ful agrarian reforms and a belated green revolution during the 1980s, it had no way avoided the impact of the market reforms since the early 1990s. We have strong reasons to believe that West Bengal agriculture had a second trend break in the early 1990s in a negative direction. 2 Methodology2.1 VariablesWe have considered eight variables for our analysis. They are listed as following: (1) area under food production; (2) total pro-duction of foodgrain; (3) yield or production per area, (ie, 2 ÷ 1); (4)consumption of fertiliser per net sown area(NSA); (5) cover-age ofHYV area per gross cropped area(GCA); (6) cropping inten-sity (GCA ÷ NSA) × 100; (7) per acre institutional credit or total (direct and indirect) finance to agriculture per NSA; and (8) area under land reform per NSA (area under land reform is defined as total land distributed by the government plus total land area under barga recording). Area, production and yield are generally dependent vari-ables in a broad analytical framework. In other words, the per-formance of these variables determine the overall performance of an agrarian economy. We broadly club the consumption of fertilisers, coverage of HYV area and cropping intensity as the main technological variables. Irrigation is also a major factor but time series data on area under minor irrigation is not easily available. We hope that area under HYV and cropping intensity can broadly absorb the impact of irrigation. On the other hand, credit and area under land reform are the major institutional variables. 2.2 GrowthRatesWe consider growth rates of all the listed variables for 23 years from 1980-81 to 2002-2003. The Indian economy entered the Table 3: Kinked Exponential Growth Rates (1980-81 to 1991-92 and 1992-93 to 2002-03)Variables R2 AdjStandardConstantt-valueRCt-valueRCt-valueF-test R Sq Error Pd I Pd II ValueArea 0.6660.6330.0028.77778.01*0.00754.38*221.0419.97* (0.11) (0.002) (0.002) Production 0.885 0.873 0.095 9.47 239.72* 0.0505 8.37* 0.01732.34**76.84* (0.39)(0.006) (0.007) Yield 0.903 0.893 0.076 7.61 239.75* 0.0452 9.34* 0.01432.40**92.95* (0.032) (0.005) (0.006) Fertiliser 0.9610.9570.1016.69158.41*0.092314.31*0.04015.05*245.79* (0.042) (0.006) (0.008) HYV area 0.975 0.973 0.051 3.83 178.83* 0.0561 17.16* 0.0309 7.67* 396.45* (0.021) (0.003) (0.004) Cropping 0.894 0.884 0.029 5.08413.77* 0.01437.66* 0.00893.87* 84.62* intensity (0.012) (0.002) (0.002) Credit 0.9550.9510.1727.28101.62*0.1412.77*0.07285.40*213.10* (0.072) (0.011) (0.013) Land 0.8870.8760.0212.71317.48*0.00836.40*0.00774.843*78.71* reform (0.009) (0.001) (0.002) * Significant at 1 per cent level. ** Significant at 5 per cent level. Figures in parentheses show standard errors. RC -Regression Coefficient.Table 4: Log Quadratic Estimates of Trends (1980-81 to 1991-92 and 1992-93 to 2002-03)Variables R2 AdjStandardConstantt-valueRCt-valueRCt-valueF-test R Sq Error Pd I Pd II ValueArea 0.664 0.631 0.027 8.766 1036.86*0.00485.305* -0.00022-1.50419.792* (0.008) (0.001) (0.000) Production 0.883 0.872 0.096 9.432 317.46* 0.033410.52*-0.00136-2.683* 75.764* (0.030) (0.003) (0.001) Yield 0.907 0.897 0.075 7.585 326.12* 0.029111.73* -0.00134-3.392* 97.313* (0.023) (0.002) (0.000) Fertiliser 0.967 0.964 0.094 6.648 228.56* 0.06487 20.90* -0.00238-4.787* 291.70* (0.029) (0.003) (0.000) HYVArea 0.978 0.975 0.049 3.809 248.80* 0.043 26.29* -0.0011 -4.263* 435.15* (0.015) (0.002) (0.000) Cropping 0.902 0.892 0.028 5.079 574.74* 0.011412.10*-0.00029-1.96***92.15* Intensity (0.009) (0.001) (0.000) Credit 0.949 0.944 0.184 7.193 125.65* 0.106 17.34* -0.0023 -2.389* 185.39* (0.057) (0.006) (0.001) Land 0.8880.8760.022.711425.27*0.00811.80*-0.00004-0.32878.984* Reform (0.006) (0.001) (0.000) * Significant at 1 per cent level. *** Significant at 10 per cent level. Figures in parentheses show standard errors. RC -Regression Coefficient. 8.98.88.78.6 | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003Figure 1: Trend of Area in West Bengal(1981-2003, growth rates in %)
REVIEW OF AGRICULTUREdecember 29, 2007 Economic & Political Weekly68neo liberal regime in 1991. Since then, the economy has expe-rienced a state withdrawal, particularly in the form of a cut in fertiliser, food and credit subsidies. We use 1992-93 as the year of trend break, if any. We assume that though the new economic policy formally came into effect in 1991, its effect was realised in 1992-93. In other words, there was a small gestation period. Thus, we consider three time periods: (a) from 1980-81 to 2002-03, (b) first sub period, 1980-81 to 1991-92; and (c) second sub period, from 1992-93 to 2002-03. 2.2.1 Simple Exponential Growth RateConsider the following linear form: LnQt = a + bt Where Qt = output, t = time, b = coefficient on time, and a = con-stant. The coefficient on time, b, is the continuous rate of growth. It closely approximates to the annual compound growth rates. Therefore, the estimates of b are presented as growth rates. 2.2.2 Kinked Exponential Growth RatesThis model imposes linear restrictions in order to eliminate the discontinuity between sub-periods. As a matter of fact, it is a superior basis for comparisons of sub-period growth rates. We use the following form of the single kink case. We have normal-ised time by considering t = 0 at the break point, k. The sub-period growth rates can then be estimated with a joint intercept: lnQt = a + b1D1t + b2D2t + ut,where Qt = output, as usual, a = intercept, Dj = 1, in the jth sub-period and Dj = 0 otherwise. 2.2.3 Log-quadratic Estimates of Growth RatesIt is obvious that growth rates cannot be constant over time as it is assumed in the exponential form. The possibility of accelera-tion or deceleration of growth rates over time is incorporated in log-quadratic functional form: lnQt = a + bt + ct2 + utIf there is constant growth, c = 0. On the other hand, a signifi-cant positive value of the estimate of c indicates an accelerating growth rate and a significantly negative value indicates decelera-tion. To avoid the problem of multicollinearity we normalised time by considering t = 0 at the year of trend break 1992-93. The log-quadratic equation is the better measure of instability than the exponential form. All the data was collected from various issues of Economic Re-view, government of West Bengal. 3 AnalysisofVariablesThe results of our analysis are presented in Tables 1, 2, 3 and 4. We analyse the variables as following. 3.1 Area of ProductionThe growth rate of area of production for the period from 1980-81 to 2002-03 was very small though significant (0.52 per cent). Table 2 shows that the growth rate of area declined from 0.8 per cent in the first sub-period 1980-81 to 1991-92 to 0.29 per cent in the second sub-period 1992-93 to 2002-03. Table 3 shows simi-lar figures for two sub-periods, which are 0.75 per cent and 0.22 per cent respectively considering the kinked exponential growth rate. The log-quadratic estimation in Table 4 shows a significant deceleration between these two periods. The growth rate of area according to both estimates is not at all significant. Looking at the curve estimation it is found that during the early 1980s the area of production registered a high fluctuation. But after the second half of the 1980s, particularly from 1986-87 we found a steady increase in area till 1992-93; after that there was a steep downfall and a fluctuation started. We presume that this was due to a sudden loss in incentive for carrying out agricultural opera-tions on the part of the farmers in the beginning years of the new economic policy. From the early 1990s we observe a fluctuation but above the linear line. In 2000, there was a big pick up, which followed a big fluctuation. 3.2 ProductionAgricultural production grew at a (significant) rate of 3.61 during 1980-81 to 2002-03. The simple exponential growth rates showed a decline from 5.64 per cent to 2.33 per cent between the two sub-periods. The same estimations according to the kinked exponen-tial methods are 5.05 per cent and 1.73 per cent respectively. The log-quadratic equation shows a significant level of deceleration between these two periods. The agricultural production showed Figure 2: Trend of Production in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20039.89.69.49.29.08.88.6Figure 3: Trend of Yield in West Bengal(1981-2003, growth rates in %)8.07.87.67.47.27.06.8 | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
REVIEW OF AGRICULTUREEconomic & Political Weekly December 29, 200769a distinctly upward trend after the mid-1980s but since the early 1990s, particularly since the 1992-93, there was a downfall. The downfall became increasingly large in 2000. 3.3 YieldYield is nothing but the production per unit area. We have already observed the declining tendency of the growth rate of production in the 1980s. The overall growth rate of yield is 3.18. Considering the simple exponential growth rate, there was a decline from 5.08 to 1.98. The kinked exponential growth rate showed a decline in yield from the rate 4.52 during the first period 1980-81 to 1991-92 to 1.43 during 1992-93 to 2002-03. Log-quadratic estimates showed a significant decline in yield growth rate from the 1980s to the 1990s. Looking at the curve estimation, we observe that there was a recovery of yield growth in the early 1980s and then there was a significant rise in yield despite the fluctuation. Partic-ularly in the late 1980s there was a respectable rising tendency in yield. This might be due to the belated green revolution in West Bengal during the 1980s. 1992-93 virtually saw a crash of yield growth rate and in 2000, it fell below the linear line.3.4 FertiliserFor agricultural production, the use of fertiliser is a vital indica-tor. In our analysis, the use of fertiliser is significant (6.93 per cent) for the overall period. The growth rate in the first sub-period was 10.9 per cent and 9.23 per cent according to the simple expo-nential and kinked exponential, respectively. The same declined to almost half in the next sub-period, which are 5.63 per cent and 4 per cent respectively for the estimates mentioned above. The deceleration is highly significant according to the log-quad-ratic estimate shown in Table 4. The curve estimation of ferti-liser shows that since 1992-93, the fertiliser consumption literally collapsed from its pick. After the mid-1990s the observed curve almost coincides with the linear curve.Since 2000, the observed curve shows a downfall. The trend break in the growth of fertiliser consumption since the early 1990s definitely coincides with the withdrawal of the fertiliser subsidy and the consequent increase in fertiliser prices. The economic reform in the form of the new economic policy is likely to affect fertiliser consumption. The rise in fertiliser prices that followed the reduction in fertiliser consumption might directly affect the production and yield in a negative way. 3.5 Area under HYV The growth rate of HYV for the whole period from 1980-81 to 2002-03 was 4.52 per cent. The growth rate declined from 5.36 per cent in 1980-81 to 1991-92 to 2.84 per cent during 1992-93 to 2002-2003 following the simple exponential growth rate. Howev-er, using the kinked exponential growth rate, we observe that the same declined from 5.61 per cent to 3.09 per cent over these two sub-periods. The log-quadratic estimate also reveals the signifi-cant decline over these two sub-periods. The curve estimation of HYV shows a movement along the linear line till the early 1990s. But after the late 1990s, it shows a distinct downward tendency. The decline in the growth of HYV has indicated a process of retrograding green revolution in the 1990s. This is not merely a technological failure. Agricultural investment came down as a result of the rising cost of production with crashing prices of agricultural commodities. Following the World Trade Organisa-tion agreement, conventional seed research became increasingly difficult. At the same time, tightening fiscal reforms left meager funds for the agricultural universities and research institutes of India to carry on their work. According to some scholars, the dampening of the prospect of seed research was the main respon-sible factor behind this devastating performance of HYV. 3.6 CroppingIntensityThe growth rate of cropping intensity for the entire period was only 1.2 per cent. The slow growth of cropping intensity might be because of the slow expansion of the gross cropped area during the 1990s owing to the reduced incentive we discussed earlier. Like all other variables, the gross cropped area declined from 1.83 per cent to 1.29 per cent (using the simple exponential) or from 1.43 per cent to 0.89 per cent (using the kinked exponential) between two sub-periods. The log-quadratic exponential growth also indicates the significant decline. The curve estimation of cropping intensity shows a big downfall since the early 1990s. The decline in cropping intensity also confirms the process of reverse green revolution in West Bengal since the early 1990s. Figure 4: Trend of Fertiliser in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20037.57.06.56.05.5Figure 5: Trend of HYV in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20034.24.03.83.63.43.23.0
REVIEW OF AGRICULTUREdecember 29, 2007 Economic & Political Weekly70It might be because of the decline in public investment and public irrigation following the new economic policy. There was a disincentive on the part of farmers to go for multiple cropping as the risk of crop failure has become high with the additional trouble of crashing food prices after the harvest that Indian peasantry is facing for last few years. 3.7 CreditThe basic difference between the variables we discussed so far and institutional credit is that the latter is a policy variable in the hands of a banking authority. The expansion of rural credit took place in 1969 when the commercial banks were national-ised. The 1970s and 1980s are marked with a phenomenal ex-pansion of rural credit. One of the declared policies of the 1991 new economic policy was that like other interventionist policy the rural credit, particularly priority sector lending must be curbed[Reserve Bank of India 1991]. Like other variables, insti-tutional credit to agriculture declined. The difference is that the other variables declined as a result of market forces but here the institutional credit declined as a result of the conscious govern-ment policies. For the 23 years from 1980-81 to 2002-03, credit expanded at a (significant) rate 11.1 per cent. During the first sub-period 1980-81 to 1991-93, the credit expanded at a much faster rate 15.9 per cent (simple exponential growth) or 14 per cent (kinked exponential rate). In the second sub-period 1992-93 to 2002-03, these growth rates declined at 9.24 per cent or 7.28 per cent respectively. The log-quadratic equation also confirms this significant decline in the credit. The curve estimation of credit shows a distinct down-fall since the mid-1990s. 3.8 LandReformLand reform is also a direct policy variable, which is in the hands of the government. Land reform measures have been imple-mented in West Bengal since the late 1970s and early 1980s. But the initial tempo of land reform must decline with the satura-tion of the ceiling surplus land and the declining proportion of unrecorded bargadars. The land reform is not at all any policy agenda of the new economic policy. Therefore, it was expected that land reform should decline over time, particularly in the second sub-period. But belying our expectation, land reform did not register any significant decline at all. Land reform grew at a stagnant rate of 0.8 per cent throughout the period 1980-81 to 2002-03. The per-centage growth rates in the two sub-periods were 0.96 and 0.90 (simple exponential) or 0.83 and 0.77 (kinked exponential). Log quadratic estimates showed a relatively weak and not significant decline over these two decades. The curve estimation of land re-form shows a slight downfall in 1990-91 but a movement along the linear line thereafter. 4 ConclusionsThis paper attempted to explore the growth of the agrarian economy of West Bengal over the last 23 years from 1980-81 to 2002-03. It also tried to find a trend break in this regard. To understand this trend break, we opted for the method of simple exponential growth rate, kinked exponential growth rates and log-quadratic estimates. While the first two methods suggest growth rates and a trend break, the latter shows the extent of | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003Figure 8: Trend of Land Reform in West Bengal(1981-2003, growth rates in %)2.92.82.72.62.5Figure 7: Trend of Credit in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20038.58.07.57.06.56.05.55.25.15.04.94.8Figure 6: Trend in Cropping Intensity in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
REVIEW OF AGRICULTUREEconomic & Political Weekly december 29, 200765Agrarian Impasse in West Bengal in the Liberalisation EraMaumita Bhattacharyya, Sudipta Bhattacharyya Maumita Bhattacharyya is at the Kumud Kumari Institution, Jhargram and Sudipta Bhattacharyya is at the Economics, Department of Economics and Politics, Visva-Bharati, Santiniketan. This paper is a revised version of a part of a MPhil dissertation of the first author submitted to Burdwan University, Burdwan. The authors are grateful to Kalyanbrata Bhattacharya (who supervised the dissertation) and to Manabendu Chattopadhyay for their very useful comments. The usual disclaimer applies (Email: sudipta.bh@gmail.com).This paper analyses the process of growth of the agrarian economy of West Bengal from 1980-81 to 2002-03. There is a significantly negative trend break in 1992-93, which was the beginning of the liberalisation era in the Indian economy. The entire time period is divided into two sub-periods, namely, 1980-81 to 1991-92 and 1992-93 to 2002-03. We use the method of computing simple exponential growth rates, kinked exponential growth rates and log quadratic estimates. The former two methods suggest growth and trend breaks, but the latter shows the extent of instability. All eight variables related to the agrarian economy of West Bengal, namely, area, production, yield, consumption of fertiliser, proportion of HYV, cropping intensity, institutional credit and land reform show a decline in growth from the first to the second sub-period. All variables except area and land reform register significant deceleration. The trend break was particularly sharp for production, yield and fertiliser use. The rural economy of West Bengal underwent a process of “an agrarian impasse” during the period 1950 to 1980 [Boyce 1987]. The period manifested an agricultural growth rate of 1.74 per cent per annum, which was below the growth rates of rural (2.31 per cent) and total population (2.42 per cent) (ibid). The rural economy of West Bengal, therefore, did not find a favourable environment to diversify itself during the period preceding the 1980s. Landlords and ‘jotedars’ dominated rural West Bengal and there was an acute inequality in land and non-land assets [Das 1998, Bhattacharyya 2001]. 1 A National PerspectiveThe first change in the political environment of West Bengal took place in 1967 and 1969-70 when the non-Congress parties briefly ruled the state. During the second United Front ministry, a “radi-cal land reform from below” took place, where local peasants recovered more than 5,00,000 acres of ‘benami’ land and distrib-uted it among the landless. This was a kind of forced land re-form, undertaken outside the governmental framework, more or less spontaneous through popular participation with the support of the United Front parties [Dasgupta 1984]. This was the back-ground against which the Left Front government came to power in 1977. The major contribution of the Left Front government was to successfully implement the existing laws. The policies of the agrarian reform of the Left Front government were, first, the “operation barga” or the policy of spot recordation of share ten-ants. Second, it included the distribution of ceiling surplus land to the landless or poor peasants. Operation barga was helpful in breaking the interlinkage [Bhattacharyya 2005], and weaken the exploitative institution of sharetenancy itself. It broke the stereotype stratification of sharecropping where sharecroppers belonged to the lower class and the landlord to the upper class [Bhattacharyya 2007b]. Third, the agrarian reform made guar-antees of the implementation of the minimum wage rate for agri-cultural labourers. These reforms were carried out in the frame-work of the panchayati raj or the grass root level self-governance of the people. [GOWB 2004]. The panchayat raj institution shows a viable alternative to anti-statist model of participatory govern-ance by the World Bank [Bhattacharyya 2007a].Some studies showed a high growth performance of West Ben-gal’s agriculture since the early 1980s, which overwhelmed theo-rists and policymakers and added a new dimension for controver-sy. Saha and Swaminathan (1994), using an index number series on aggregate agricultural production, claimed that, “the expo-nential growth rate for all West Bengal for the period 1981-82 to 1990-91 was an impressive 6.4 per cent per annum. The growth performance is striking in comparison to the past. The annual growth rate of agricultural output for the period 1965-80 was 
REVIEW OF AGRICULTUREdecember 29, 2007 Economic & Political Weekly66Table 1: Simple Exponential Growth Rates (1980-81 to 2002-03)Variables R2 AdjStandardConstantt-valueRegressiont-valueF-test R Sq Error of Est Co-eff ValueArea 0.6260.6090.0288.6895.933*0.00525.933*35.203* (0.012) (0.003) Production 0.8410.8340.118.904189.958*0.0360910.556*111.428* (0.01)(0.004) Yield 0.8530.8460.09167.113180.124*0.031811.048*122.064* (0.039) (0.001) Fertiliser 0.929 0.925 0.13385.641 97.850* 0.069316.560* 274.240* (0.058) (0.004) HYV area 0.957 0.955 0.063 3.174 110.913* 0.0452 21.652* 468.791* (0.029) (0.002) Cropping 0.883 878 0.03 4.911 376.114* 0.012 12.602* 158896* intensity (0.013) (0.001) Credit 0.934 0.931 204 5.655 64.401* 0.11117.269* 298.215* (0.88) (0.006) Land reform 0.887 0.882 0.02 2.604 301.156* 0.008 12.840* 164.866* (0.009) (0.001) * Significant at 1 per cent level. Figures in parentheses show standard errors.Table 2: Simple Exponential Growth Rates (1980-81 to 1991-92 and 1992-93 to 2002-03)Variables PeriodR2 Adj StandardConstantt-value Regression t-value F-test RSquaredErrorCo-efficientValueArea Pd 1 0.526 0.479 0.02868.673 493.47* 0.008 3.333* 11.11* (0.018) (0.002) Pd II0.118 0.019 0.02758.765 493.35* 0.00291.095 1.2 (0.018) (0.003) Production Pd I 0.754 0.73 0.1218 8.783 117.13* 0.0564 5.540* 30.70* (0.075) (0.01) Pd II0.751 0.724 0.047 9.402 309.31* 0.02335.215*27.20* (0.03) (0.004) Yield Pd I 0.787 0.766 0.09997 113.88* 0.05086.082*36.99* (0.061) (0.008) PdII0.8660.8510.02727.556428.91*0.01987.632*58.26* (0.018) (0.003) Fertiliser Pd I 0.955 0.951 0.089 5.418 98.86*0.109 14.57*212.36* (0.055) (0.007) Pd II0.9 0.889 0.06556.524 153.96* 0.0563 9.018*81.33* (0.042) (0.006) HYV Pd I 0.938 0.932 0.05193.115 97.60*0.053612.352* 152.57 area (0.032) (0.004) PdII0.7870.7630.05163.822114.49*0.02845.764*33.22* (0.033) (0.005) Cropping Pd I 0.823 0.806 0.03214.877 246.59* 0.0183 6.823* 46.56* intensity (0.02) (0.003) Pd II0.911 0.902 0.01415.041 553.72* 0.01299.625*92.64* (0.009) (0.001) Credit Pd I 0.944 0.939 0.1462 5.375 59.73* 0.159 13.031* 169.80* (0.09) (0.012) PdII0.790.7670.16667.05265.47*0.09245.824*33.91* (0.108) (0.016) Land PdI0.6730.640.02522.597167.32*0.00964.535*20.57* reform (0.016) (0.002) PdII0.8510.8340.01322.694315.56*0.0097.159*51.25* (0.009) (0.001) * Significant at 1 per cent level. Figures in parentheses show standard errors.2.2 per cent”. The agricultural growth rate in West Bengal is the highest among all the Indian states. Though some doubt had been cast regarding the validity of the official data [Datta Ray 1994] based on which such a conclusion was derived, no effective challenge was thrown against the findings showing high growth performance of West Bengal. On the contrary, Sen and Sengupta (1995) using an alternative source of data(comprehensive cost of production studies, ministry of agriculture, government of India) reaffirmed “a significant trend-break in the growth rate during the 1980s”. Rawal and Swaminathan (1998) also supported the trend break in the early 1980s. Some economists tried to identify the factors responsible for the high growth rate in agriculture. According to Sen and Sen-gupta (1995), West Bengal stands as a case apart because in this case, the measured input growth does not appear to be capable of explaining the magnitude of the trend break in output growth. As a matter of fact, according to Sen and Sengupta, the rate of growth of fertiliser use, percentage of high yield variety(HYV) area, and irrigated area, unlike other eastern Indian states de-clined in West Bengal in the 1980s compared to the 1970s. This, in turn, means that unmeasured technological shifters were more important here or that land and panchayat reforms (car-ried out in West Bengal but not elsewhere) caused an increase in productivity not found elsewhere. Mukherji and Mukhopadhyay (1996) attempted to determine the impact of institutional chang-es on total factor productivity in the production of rice in West Bengal. According to them, it is difficult to figure out what would have spurred this growth by simply looking at the statistics on input use. The data do not suggest an increase in factor use cor-responding to output growth. The extent of farm mechanisation has been minimal in the state and has shown no massive increase in recent times. Mukherji and Mukhopadhyay (1996) maintained that the only one change that has been taking place in favour of higher yields is the institutional change characterised by the panchayati raj and land reforms. A few scholars have opined that agricultural growth in West Bengal can be successfully explained in terms of both the technological and institutional factors, such as the belated green revolution, personal initiative and private investment, operation barga, distribution of surplus land and panchayati raj [Sengupta and Gazdar 1998, Banerjee et al 2002]. In our view, the increased land productivity in West Bengal dur-ing the 1980s is the direct result of the increased labour inten-sity in family farms belonging to marginal and small category of farmers. As the land reforms created higher purchasing power, marginal and small farmers could access green revolution tech-nology unlike in the leading green revolution states in India.The trend break in the growth rate during the 1980s in West Bengal was associated with the opening up of the horizon of diversified rural activities [Chandrasekhar 1993; Arun Kumar et al 1995, Lieten 1990, Rogaly et al 1999; Bhattacharyya 1996, 2003a, 2003b]. As a result, some positive structural changes may be observed in recent years in West Bengal. In the early 1990s, India adopted the structural adjustment programme of the International Monetary Fund (IMF) and World Bank as a loan conditionality and had to change its macroeco-nomic policies to a great extent. Indian agriculture experienced some major changes, which appeared as a sudden shock to the peasantry. First food, fertiliser and credit subsidies were curbed to a great extent. As a result, the public distribution system(PDS) was virtually wiped out. Second, import liberalisation opened up the inflow of foodgrains from abroad at cheaper prices compared to the home market. The peasantry faced two problems. (a) higher costs of production owing to the withdrawal of the fertiliser sub-sidy; and (b) selling products at viable prices. In a country where about 60 per cent of rural population is a seller of foodgrains, it leads to a loss of income for these peasant households. Due to the government withdrawal of thePDS, there was a sudden decline 
REVIEW OF AGRICULTUREEconomic & Political Weekly December 29, 200767in demand for foodgrains particularly from poorer households. Government economists and policymakers explained this fact as a sign of prosperity, where the rural population voluntarily demanded less foodgrains. Hence, policymakers, under market reforms, advised peasants to diversify their production from foodgrains to export-oriented crops.According to Patnaik (2005), the Indian economy like in the co-lonial period has already witnessed an inverse relation between an increase in export-oriented production and decline in the avail-ability of foodgrains. On the other hand, the global price crash in export commodities, particularly cotton, led to suicides of thou-sands of Indian farmers. The prevailing global and local deflation-ary situation is a part of the global low interest rate regime, which is favourable for global finance capital and the international credi-tor. As a result, all the expansionary and development activities in the economy declined with the obsessed target of lowering fiscal deficit under market reforms. There was an acute increase of ru-ral employment and massive increase in rural poverty. Between 1980-81 and 1989-90, foodgrains grew at the rate 2.85 per cent per annum, and crops grew at 3.19 per cent per annum. The rate of growth of population at this period was 2.1 per cent per annum. However, from 1990-91 to 2000-01 the foodgrain growth declined at 1.66 per cent per annum, while crops declined to 1.73 per cent per annum. For the first time after independence, the population growth rate (1.9 per cent per annum) remained above of foodgrain and overall crop growth rate (ibid). One has to address the West Bengal scenario against this na-tional perspective. Though West Bengal experienced success-ful agrarian reforms and a belated green revolution during the 1980s, it had no way avoided the impact of the market reforms since the early 1990s. We have strong reasons to believe that West Bengal agriculture had a second trend break in the early 1990s in a negative direction. 2 Methodology2.1 VariablesWe have considered eight variables for our analysis. They are listed as following: (1) area under food production; (2) total pro-duction of foodgrain; (3) yield or production per area, (ie, 2 ÷ 1); (4)consumption of fertiliser per net sown area(NSA); (5) cover-age ofHYV area per gross cropped area(GCA); (6) cropping inten-sity (GCA ÷ NSA) × 100; (7) per acre institutional credit or total (direct and indirect) finance to agriculture per NSA; and (8) area under land reform per NSA (area under land reform is defined as total land distributed by the government plus total land area under barga recording). Area, production and yield are generally dependent vari-ables in a broad analytical framework. In other words, the per-formance of these variables determine the overall performance of an agrarian economy. We broadly club the consumption of fertilisers, coverage of HYV area and cropping intensity as the main technological variables. Irrigation is also a major factor but time series data on area under minor irrigation is not easily available. We hope that area under HYV and cropping intensity can broadly absorb the impact of irrigation. On the other hand, credit and area under land reform are the major institutional variables. 2.2 GrowthRatesWe consider growth rates of all the listed variables for 23 years from 1980-81 to 2002-2003. The Indian economy entered the Table 3: Kinked Exponential Growth Rates (1980-81 to 1991-92 and 1992-93 to 2002-03)Variables R2 AdjStandardConstantt-valueRCt-valueRCt-valueF-test R Sq Error Pd I Pd II ValueArea 0.6660.6330.0028.77778.01*0.00754.38*221.0419.97* (0.11) (0.002) (0.002) Production 0.885 0.873 0.095 9.47 239.72* 0.0505 8.37* 0.01732.34**76.84* (0.39)(0.006) (0.007) Yield 0.903 0.893 0.076 7.61 239.75* 0.0452 9.34* 0.01432.40**92.95* (0.032) (0.005) (0.006) Fertiliser 0.9610.9570.1016.69158.41*0.092314.31*0.04015.05*245.79* (0.042) (0.006) (0.008) HYV area 0.975 0.973 0.051 3.83 178.83* 0.0561 17.16* 0.0309 7.67* 396.45* (0.021) (0.003) (0.004) Cropping 0.894 0.884 0.029 5.08413.77* 0.01437.66* 0.00893.87* 84.62* intensity (0.012) (0.002) (0.002) Credit 0.9550.9510.1727.28101.62*0.1412.77*0.07285.40*213.10* (0.072) (0.011) (0.013) Land 0.8870.8760.0212.71317.48*0.00836.40*0.00774.843*78.71* reform (0.009) (0.001) (0.002) * Significant at 1 per cent level. ** Significant at 5 per cent level. Figures in parentheses show standard errors. RC -Regression Coefficient.Table 4: Log Quadratic Estimates of Trends (1980-81 to 1991-92 and 1992-93 to 2002-03)Variables R2 AdjStandardConstantt-valueRCt-valueRCt-valueF-test R Sq Error Pd I Pd II ValueArea 0.664 0.631 0.027 8.766 1036.86*0.00485.305* -0.00022-1.50419.792* (0.008) (0.001) (0.000) Production 0.883 0.872 0.096 9.432 317.46* 0.033410.52*-0.00136-2.683* 75.764* (0.030) (0.003) (0.001) Yield 0.907 0.897 0.075 7.585 326.12* 0.029111.73* -0.00134-3.392* 97.313* (0.023) (0.002) (0.000) Fertiliser 0.967 0.964 0.094 6.648 228.56* 0.06487 20.90* -0.00238-4.787* 291.70* (0.029) (0.003) (0.000) HYVArea 0.978 0.975 0.049 3.809 248.80* 0.043 26.29* -0.0011 -4.263* 435.15* (0.015) (0.002) (0.000) Cropping 0.902 0.892 0.028 5.079 574.74* 0.011412.10*-0.00029-1.96***92.15* Intensity (0.009) (0.001) (0.000) Credit 0.949 0.944 0.184 7.193 125.65* 0.106 17.34* -0.0023 -2.389* 185.39* (0.057) (0.006) (0.001) Land 0.8880.8760.022.711425.27*0.00811.80*-0.00004-0.32878.984* Reform (0.006) (0.001) (0.000) * Significant at 1 per cent level. *** Significant at 10 per cent level. Figures in parentheses show standard errors. RC -Regression Coefficient. 8.98.88.78.6 | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003Figure 1: Trend of Area in West Bengal(1981-2003, growth rates in %)
REVIEW OF AGRICULTUREdecember 29, 2007 Economic & Political Weekly68neo liberal regime in 1991. Since then, the economy has expe-rienced a state withdrawal, particularly in the form of a cut in fertiliser, food and credit subsidies. We use 1992-93 as the year of trend break, if any. We assume that though the new economic policy formally came into effect in 1991, its effect was realised in 1992-93. In other words, there was a small gestation period. Thus, we consider three time periods: (a) from 1980-81 to 2002-03, (b) first sub period, 1980-81 to 1991-92; and (c) second sub period, from 1992-93 to 2002-03. 2.2.1 Simple Exponential Growth RateConsider the following linear form: LnQt = a + bt Where Qt = output, t = time, b = coefficient on time, and a = con-stant. The coefficient on time, b, is the continuous rate of growth. It closely approximates to the annual compound growth rates. Therefore, the estimates of b are presented as growth rates. 2.2.2 Kinked Exponential Growth RatesThis model imposes linear restrictions in order to eliminate the discontinuity between sub-periods. As a matter of fact, it is a superior basis for comparisons of sub-period growth rates. We use the following form of the single kink case. We have normal-ised time by considering t = 0 at the break point, k. The sub-period growth rates can then be estimated with a joint intercept: lnQt = a + b1D1t + b2D2t + ut,where Qt = output, as usual, a = intercept, Dj = 1, in the jth sub-period and Dj = 0 otherwise. 2.2.3 Log-quadratic Estimates of Growth RatesIt is obvious that growth rates cannot be constant over time as it is assumed in the exponential form. The possibility of accelera-tion or deceleration of growth rates over time is incorporated in log-quadratic functional form: lnQt = a + bt + ct2 + utIf there is constant growth, c = 0. On the other hand, a signifi-cant positive value of the estimate of c indicates an accelerating growth rate and a significantly negative value indicates decelera-tion. To avoid the problem of multicollinearity we normalised time by considering t = 0 at the year of trend break 1992-93. The log-quadratic equation is the better measure of instability than the exponential form. All the data was collected from various issues of Economic Re-view, government of West Bengal. 3 AnalysisofVariablesThe results of our analysis are presented in Tables 1, 2, 3 and 4. We analyse the variables as following. 3.1 Area of ProductionThe growth rate of area of production for the period from 1980-81 to 2002-03 was very small though significant (0.52 per cent). Table 2 shows that the growth rate of area declined from 0.8 per cent in the first sub-period 1980-81 to 1991-92 to 0.29 per cent in the second sub-period 1992-93 to 2002-03. Table 3 shows simi-lar figures for two sub-periods, which are 0.75 per cent and 0.22 per cent respectively considering the kinked exponential growth rate. The log-quadratic estimation in Table 4 shows a significant deceleration between these two periods. The growth rate of area according to both estimates is not at all significant. Looking at the curve estimation it is found that during the early 1980s the area of production registered a high fluctuation. But after the second half of the 1980s, particularly from 1986-87 we found a steady increase in area till 1992-93; after that there was a steep downfall and a fluctuation started. We presume that this was due to a sudden loss in incentive for carrying out agricultural opera-tions on the part of the farmers in the beginning years of the new economic policy. From the early 1990s we observe a fluctuation but above the linear line. In 2000, there was a big pick up, which followed a big fluctuation. 3.2 ProductionAgricultural production grew at a (significant) rate of 3.61 during 1980-81 to 2002-03. The simple exponential growth rates showed a decline from 5.64 per cent to 2.33 per cent between the two sub-periods. The same estimations according to the kinked exponen-tial methods are 5.05 per cent and 1.73 per cent respectively. The log-quadratic equation shows a significant level of deceleration between these two periods. The agricultural production showed Figure 2: Trend of Production in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20039.89.69.49.29.08.88.6Figure 3: Trend of Yield in West Bengal(1981-2003, growth rates in %)8.07.87.67.47.27.06.8 | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
REVIEW OF AGRICULTUREEconomic & Political Weekly December 29, 200769a distinctly upward trend after the mid-1980s but since the early 1990s, particularly since the 1992-93, there was a downfall. The downfall became increasingly large in 2000. 3.3 YieldYield is nothing but the production per unit area. We have already observed the declining tendency of the growth rate of production in the 1980s. The overall growth rate of yield is 3.18. Considering the simple exponential growth rate, there was a decline from 5.08 to 1.98. The kinked exponential growth rate showed a decline in yield from the rate 4.52 during the first period 1980-81 to 1991-92 to 1.43 during 1992-93 to 2002-03. Log-quadratic estimates showed a significant decline in yield growth rate from the 1980s to the 1990s. Looking at the curve estimation, we observe that there was a recovery of yield growth in the early 1980s and then there was a significant rise in yield despite the fluctuation. Partic-ularly in the late 1980s there was a respectable rising tendency in yield. This might be due to the belated green revolution in West Bengal during the 1980s. 1992-93 virtually saw a crash of yield growth rate and in 2000, it fell below the linear line.3.4 FertiliserFor agricultural production, the use of fertiliser is a vital indica-tor. In our analysis, the use of fertiliser is significant (6.93 per cent) for the overall period. The growth rate in the first sub-period was 10.9 per cent and 9.23 per cent according to the simple expo-nential and kinked exponential, respectively. The same declined to almost half in the next sub-period, which are 5.63 per cent and 4 per cent respectively for the estimates mentioned above. The deceleration is highly significant according to the log-quad-ratic estimate shown in Table 4. The curve estimation of ferti-liser shows that since 1992-93, the fertiliser consumption literally collapsed from its pick. After the mid-1990s the observed curve almost coincides with the linear curve.Since 2000, the observed curve shows a downfall. The trend break in the growth of fertiliser consumption since the early 1990s definitely coincides with the withdrawal of the fertiliser subsidy and the consequent increase in fertiliser prices. The economic reform in the form of the new economic policy is likely to affect fertiliser consumption. The rise in fertiliser prices that followed the reduction in fertiliser consumption might directly affect the production and yield in a negative way. 3.5 Area under HYV The growth rate of HYV for the whole period from 1980-81 to 2002-03 was 4.52 per cent. The growth rate declined from 5.36 per cent in 1980-81 to 1991-92 to 2.84 per cent during 1992-93 to 2002-2003 following the simple exponential growth rate. Howev-er, using the kinked exponential growth rate, we observe that the same declined from 5.61 per cent to 3.09 per cent over these two sub-periods. The log-quadratic estimate also reveals the signifi-cant decline over these two sub-periods. The curve estimation of HYV shows a movement along the linear line till the early 1990s. But after the late 1990s, it shows a distinct downward tendency. The decline in the growth of HYV has indicated a process of retrograding green revolution in the 1990s. This is not merely a technological failure. Agricultural investment came down as a result of the rising cost of production with crashing prices of agricultural commodities. Following the World Trade Organisa-tion agreement, conventional seed research became increasingly difficult. At the same time, tightening fiscal reforms left meager funds for the agricultural universities and research institutes of India to carry on their work. According to some scholars, the dampening of the prospect of seed research was the main respon-sible factor behind this devastating performance of HYV. 3.6 CroppingIntensityThe growth rate of cropping intensity for the entire period was only 1.2 per cent. The slow growth of cropping intensity might be because of the slow expansion of the gross cropped area during the 1990s owing to the reduced incentive we discussed earlier. Like all other variables, the gross cropped area declined from 1.83 per cent to 1.29 per cent (using the simple exponential) or from 1.43 per cent to 0.89 per cent (using the kinked exponential) between two sub-periods. The log-quadratic exponential growth also indicates the significant decline. The curve estimation of cropping intensity shows a big downfall since the early 1990s. The decline in cropping intensity also confirms the process of reverse green revolution in West Bengal since the early 1990s. Figure 4: Trend of Fertiliser in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20037.57.06.56.05.5Figure 5: Trend of HYV in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20034.24.03.83.63.43.23.0
REVIEW OF AGRICULTUREdecember 29, 2007 Economic & Political Weekly70It might be because of the decline in public investment and public irrigation following the new economic policy. There was a disincentive on the part of farmers to go for multiple cropping as the risk of crop failure has become high with the additional trouble of crashing food prices after the harvest that Indian peasantry is facing for last few years. 3.7 CreditThe basic difference between the variables we discussed so far and institutional credit is that the latter is a policy variable in the hands of a banking authority. The expansion of rural credit took place in 1969 when the commercial banks were national-ised. The 1970s and 1980s are marked with a phenomenal ex-pansion of rural credit. One of the declared policies of the 1991 new economic policy was that like other interventionist policy the rural credit, particularly priority sector lending must be curbed[Reserve Bank of India 1991]. Like other variables, insti-tutional credit to agriculture declined. The difference is that the other variables declined as a result of market forces but here the institutional credit declined as a result of the conscious govern-ment policies. For the 23 years from 1980-81 to 2002-03, credit expanded at a (significant) rate 11.1 per cent. During the first sub-period 1980-81 to 1991-93, the credit expanded at a much faster rate 15.9 per cent (simple exponential growth) or 14 per cent (kinked exponential rate). In the second sub-period 1992-93 to 2002-03, these growth rates declined at 9.24 per cent or 7.28 per cent respectively. The log-quadratic equation also confirms this significant decline in the credit. The curve estimation of credit shows a distinct down-fall since the mid-1990s. 3.8 LandReformLand reform is also a direct policy variable, which is in the hands of the government. Land reform measures have been imple-mented in West Bengal since the late 1970s and early 1980s. But the initial tempo of land reform must decline with the satura-tion of the ceiling surplus land and the declining proportion of unrecorded bargadars. The land reform is not at all any policy agenda of the new economic policy. Therefore, it was expected that land reform should decline over time, particularly in the second sub-period. But belying our expectation, land reform did not register any significant decline at all. Land reform grew at a stagnant rate of 0.8 per cent throughout the period 1980-81 to 2002-03. The per-centage growth rates in the two sub-periods were 0.96 and 0.90 (simple exponential) or 0.83 and 0.77 (kinked exponential). Log quadratic estimates showed a relatively weak and not significant decline over these two decades. The curve estimation of land re-form shows a slight downfall in 1990-91 but a movement along the linear line thereafter. 4 ConclusionsThis paper attempted to explore the growth of the agrarian economy of West Bengal over the last 23 years from 1980-81 to 2002-03. It also tried to find a trend break in this regard. To understand this trend break, we opted for the method of simple exponential growth rate, kinked exponential growth rates and log-quadratic estimates. While the first two methods suggest growth rates and a trend break, the latter shows the extent of | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003Figure 8: Trend of Land Reform in West Bengal(1981-2003, growth rates in %)2.92.82.72.62.5Figure 7: Trend of Credit in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |nn1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 20038.58.07.57.06.56.05.55.25.15.04.94.8Figure 6: Trend in Cropping Intensity in West Bengal(1981-2003, growth rates in %) | | | | | | | | | | | | | | | | | | | | | | |1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
