Jump to ContentJump to Main Navigation
Indian DevelopmentSelected Regional Perspectives$

Jean Drèze and Amartya Sen

Print publication date: 1997

Print ISBN-13: 9780198292043

Published to Oxford Scholarship Online: October 2011

DOI: 10.1093/acprof:oso/9780198292043.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 24 February 2017

Agrarian Politics and Rural Development in West Bengal *

Agrarian Politics and Rural Development in West Bengal *

Chapter:
(p.129) 3 Agrarian Politics and Rural Development in West Bengal*
Source:
Indian Development
Author(s):

Sunil Sengupta

Haris Gazdar

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780198292043.003.0003

Abstract and Keywords

This chapter examines recent attempts to address the challenge of rural poverty in West Bengal. One distinguishing feature of the recent history of the said state is a significant change in the balance of political power in favour of disadvantaged groups, which took a concrete expression in 1977 when the Left Front coalition came to office at the state level. This change in the balance of power has made it possible to implement a number of far-reaching social programmes that are often considered ‘politically infeasible’ in many other states, notably including land reform and the revitalization of democratic institutions at the village level. Public policies concerned with health education and related matters have also been neglected, and correspondingly, the improvement of living conditions in West Bengal in recent years has remained relatively slow.

Keywords:   rural poverty, West Bengal, political power, Left Front coalition, health education, living conditions

1. Why Look at Rural West Bengal?

West Bengal, and its neighbouring states of Bihar and Orissa, have had some of the highest proportions of rural population below the poverty line ever since these indicators have been published for Indian states. In 1987–8, the latest year for which state-wise poverty indicators were available at the time of writing, nearly three-fifths of the rural population in West Bengal lived below the poverty line.1

West Bengal is India's most densely-populated state, with a population density of 766 persons per square kilometre in 1991 compared with around 270 persons per square kilometre in India as a whole.2 Although the proportion of the population living in urban areas is a little higher in West Bengal than in India as a whole (27.4 per cent and 25.7 per cent in 1991, respectively), the pace of urbanization (p.130) has been slower in West Bengal than in the rest of the country during the entire post-independence period. Over the same period, the state's relative position in terms of industrial output has undergone prolonged stagnation and relative decline. In 1960–1, nearly 23 per cent of India's industrial output was produced in West Bengal. This fell to about 10 per cent in 1980–1, and to under 7 per cent by the end of the eighties.3 Starting out as a region of early industrial development, West Bengal lost its predominant position and failed to reverse the trend of relative industrial decline that set in from around the mid-sixties.

In the absence of industrial (and other non-agricultural) development, the agrarian economy remained the mainstay for the overwhelming majority of the rural population. Despite reasonable agricultural potential, however, output in this sector also remained stagnant from the fifties to the early eighties.4 In short, high population density, undeveloped agricultural potential, and stagnation in industry have been characteristic features of the limitations to economic opportunity faced by the people of the state. It was under these rather bleak economic conditions that West Bengal experienced a radical change in the course of its state-level politics in the late seventies.

1.1 Change On the Horizon

West Bengal has had the unparalleled distinction among Indian states, of being ruled, since 1977, by a coalition of left-wing parties, that have advocated class-based politics and are publicly committed to improving the position of the poor in rural areas as a matter of priority. There has been wide interest in the experience of the state since the Left Front coalition, led by the Communist Party of India (Marxist) or CPM, came to office over a decade and a half ago. The Left Front government came to power following a period of intense political strife in rural as well as urban areas, on the promise of vigorous agrarian and political reform. The electoral support of the Left Front in rural areas has been overwhelming. Since its initial (p.131) victory in the 1977 state elections, it has won three consecutive state elections by convincing margins, and its candidates have received a steady share of around 70 per cent of the popular vote in rural local government (panchayat) elections for four successive terms.

The antecedents of the Left Front's victory in 1977 lie in the political history of West Bengal. Not only had class-based left-wing politics found a receptive home in Calcutta, once a major centre of anti-colonial struggle, there had also been a history of active involvement in peasant movements and organizations on the part of the communists since the nineteen thirties. By the late sixties the communist parties had considerable following in rural areas, and when state politics underwent a crisis with a split in the ruling Congress party, the CPM, in alliance with other left-wing parties, formed coalition governments with the breakaway Bangla Congress in 1967 and 1969. The total period for which this coalition (known as the ‘United Front’) ruled was around thirteen months.

The late sixties were a period of unprecedented ferment in the rural areas for the redistribution of land ownership and the more effective implementation of existing agrarian-reform legislation in the state. In particular, peasant organizations encouraged their supporters to take the initiative in the identification and appropriation of landholdings above the legally permitted limits, that had been ‘concealed’ by landlords in apprehension of land-redistribution laws using various administrative and legal loopholes. This movement of the landless and the small and marginal farmers escalated after the formation of United Front governments whose constituents included parties that had led peasant organizations.5

The United Front governments were short-lived, and fell under the combined pressures of a Congress-ruled centre on the one hand, and escalating violence in the peasant revolt on the other. The Congress party returned to power in West Bengal after this period until it was defeated at the state elections in 1977 by the CPM-led (p.132) ‘Left Front’ consisting mainly of the left-wing constituents of the erstwhile United Front.

The emergence and consolidation of a political front with ideological commitment to economic equity, and with a wide base of support in rural areas, has significant implications that stretch well beyond West Bengal itself.6 If the development experience over much of India has been plagued by the political domination of propertied elites, with little or no interest in pro-poor interventions, here is the case of a state where the situation appears to be quite different. This difference deserves attention not only for what it might reveal about the potential for change within West Bengal, but also for the lessons which might be elicited from the experience of one state for the rest of India, and indeed for other developing countries.

To state that change was desperately needed, particularly in the economic conditions of the rural poor, would be no exaggeration. About the only favourable feature of the Left Front's inheritance in 1977 was its reasonably assured political base. The economic situation was characterized by extremely high rates of rural poverty, stagnation in agriculture, relative decline in the non-agricultural sectors, high and rising demographic pressure, and the near absence of economic diversification. Furthermore, by choice or by political necessity, the Left Front had to play by the rules of India's capitalist democracy. Any economic and political reforms had to be consistent with the established politico-legal framework, and had to be implemented under conditions, where, for most of this period, the constituents of the Front were on opposition benches at the centre.

Agrarian reforms and the reorganization of the system of local governance were initiated early on and they have attracted much attention in scholarly as well as political circles. The period since the late seventies has, indeed, seen many changes in West Bengal. In contrast to the situation in 1977, by 1994, West Bengal had experienced a sustained period of agricultural growth stretching over a decade. Its rural poverty indices had registered marked declines, and there was new optimism about the potential for industrial rejuvenation. But there have been positive changes in other parts of (p.133) India also. Whether, and which reforms of the Left Front played a positive (or indeed negative) role, is a matter for analysis and debate.7 We hope to contribute to this discussion by focusing on some essential aspects of the well-being of the population: poverty alleviation, health, and education.

We are interested, therefore, not only in the programme of the Left Front, and whether, or to what extent, the government was successful in implementing it, but also in how this programme relates to the conditions and needs of the poor in rural West Bengal. Section 2 discusses the political antecedents of the Left Front in West Bengal, and the context in which the agrarian reforms agenda came to occupy the centre-stage in state politics. Evidence on the implementation of agrarian-reforms is examined, and the scope that the various reform measures offer in terms of poverty alleviation and the reduction of inequality is explored. In Section 3 we evaluate the performance of agriculture in the state, and analyse trends in rural consumption, poverty, and inequality over the last two decades or so. An attempt is made to interpret these trends in output, consumption, and distribution within the context of the eastern region (i.e. the neighbouring states of Bihar and Orissa) and to assess the relative impact of land reforms on the livelihoods of the poor. Non-income aspects of well-being, such as health and education, are discussed in Section 4. West Bengal's record in the expansion of literacy and the lowering of mortality rates is compared with that of other Indian states. Section 5 offers concluding remarks and sums up the achievements as well as the failures of public policy in West Bengal in the expansion of the well-being of the rural population. We return to the political factors associated with the emergence of the Left Front, and discuss how they might have helped as well as hindered the process of social and economic development.

(p.134) 1.2 A Note about Data

A variety of data-sources are used in this paper to build a picture of West Bengal's experience over the last two decades or so. Of the primary sources of data, special mention needs to be made of the work of a research project of the World Institute for Development Economics Research (WIDER) based at the Visva-Bharati University at Santiniketan between 1987 and 1990. This project, entitled ‘Rural Poverty, Social Change and Public Policy’ collected and analysed data on a wide range of issues relating to rural poverty, incomes, ownership of land and other assets, impact of land reforms, food consumption, nutrition of children, literacy, and occupational mobility. One of us (Sengupta) was closely associated with the planning, implementation, and direction of the project from its inception.

The core of the project consisted of collecting and processing socio-economic data from six villages in the various agro-climatic zones of West Bengal, on the basis of detailed household and individual questionnaires. Much of the social and economic data, such as caste background, land ownership and operation (including the impact of agrarian-reform legislation), and household income, was based on a complete enumeration of all households. Likewise, all relevant individuals were covered to obtain information related to individuals such as educational attainment, wage earnings, and nutritional status. In all, the surveys covered 749 households, and 3,972 individuals. These surveys were carried out between 1987 and 1989, with teams of field researchers spending around six months in each of the villages. Four of the six villages surveyed are in south Bengal, two (Kuchli and Sahajapur) in Birbhum district, and one each in the districts of Medinipur (Bhagabanbasan) and Purulia (Simtuni). The other two villages are in the north Bengal districts of Jalpaiguri (Magurmari) and Koch Behar (Kalmandasguri). We draw upon results of these surveys to complement insights from a variety of published secondary data. The terms ‘WIDER villages’, ‘WIDER data’, and ‘WIDER surveys’, etc., refer to these surveys.8

The six survey villages vary greatly in their size as well as economic conditions (see Table 1). While each village may not actually (p.135) be representative of the area in which it is located, the six villages taken together capture a good deal of the diversity found in rural West Bengal. All villages save one (Magurmari in Jalpaiguri, which is close to some centres of traditional industry such as biri-making) are predominantly agricultural. Two of these (Kuchli in Birbhum and Bhagabanbasan in Medinipur) had access to shallow groundwater irrigation, and were relatively prosperous on this account. Simtuni, in Purulia, is a small, predominantly-tribal village with almost no landlessness, but with extremely poor soil and water conditions. Kalmandasguri in Koch Behar is also poorly endowed in terms of agricultural potential. Sahajapur, in Birbhum district, is the largest village and had over 200 households at the time of the survey, the majority of whom had been landless prior to the agrarian reforms. In terms of agricultural potential, this village stood somewhere between the relatively high-productivity Kuchli and Bhagabanbasan, and the poorly-endowed Simtuni and Kalmandasguri, as it had access to some canal irrigation and deep groundwater, but not to shallow groundwater.

TABLE 1. Income and Poverty in WIDER Villages

Kuchli (Birbhum)

Sahajapur (Birbhum)

Bhagabanbasan (Medinipur)

Simtuhi (Purulia)

Magurmari (Jalpaiguri)

Kalmandasguri (Koch Behar)

Number of households

142

227

134

75

99

89

Households below official poverty line (per cent)

40.4

52.3

16.5

62.5

56.6

72.7

Per cent of workers in agriculture

86

63

69

87

13

67

Mean per-capita household annual income (rupees)

1647

1544

2213

1160

1441

1212

Note. The name of the district is given in parenthesis.

Source. WIDER village surveys. 1987–9.

Secondary data sources used in this paper include socio-economic data from the decennial census, expenditure data collected by the (p.136) National Sample Survey Organisation (NSSO), agricultural output data published by the Centre for the Monitoring of the Indian Economy (CMIE), and demographic information from the Sample Registration System. Most of the arguments presented below for West Bengal draw on these widely accepted secondary data sources for empirical support. We attempt to use the primary and secondary data sources in a complementary way, with references to village-level information coming within the context of the state-level picture, by way of elaboration and substantiation.

2. Political Economy of the Reform Programme

It is not surprising that the rural programme of the Left Front has attracted interest in India and abroad. The statistics of the state's performance in the fields of agrarian and political reforms are impressive enough. By the end of 1991, for instance, West Bengal, whose share of India's cropped area was only around 3 per cent, accounted for nearly a fifth of all cultivable land redistributed under agrarian-reform legislation in India. Over 40 per cent of all land-redistribution beneficiaries in India were to be found in the state.9 While legislation for the protection of tenants from arbitrary evictions exists in many parts of the country, in West Bengal steps were taken to make this legislation effective, by official recording of tenancy leases. Over 1.4 million tenants were thus registered. In addition to these agrarian reforms, the Left Front government initiated the revitalization of the local government Panchayati Raj Institutions (PRIs) on a scale unprecedented in the country. Unlike almost anywhere else in India, regular elections to all three tiers of the PRI structure have been held on four separate occasions since 1978 in West Bengal.

Agrarian reform and democratic decentralization have been persistent themes in Indian debates on rural development since independence and even before. West Bengal has gone further than nearly all states in the implementation of the former, and arguably further than any state in the implementation of the latter. Whether it has gone far enough to substantially affect the conditions of the rural poor is an important question.

(p.137) 2.1 Agrarian Setting and Agrarian Reforms

Agrarian reforms in India have been concerned with one or several of the following issues: abolition of intermediaries in the revenue-collection system; security of tenure for sharecroppers and tenants; imposition of land-ownership ‘ceilings’ and the redistribution of private-property rights to ‘ceiling-surplus’ land from landlords to the landless and the land-poor; consolidation of fragmented landholdings. These measures either pre-suppose, or at the very least avoid conflict with, the acceptance of private-property rights in land.10 The first, that is, the abolition of intermediaries, was aimed at establishing these property rights in areas where the zamindari system of revenue intermediation had prevailed.11 This was accomplished soon after India's independence. Of the other measures, land consolidation does not have major redistributive implications.12 Among agrarian-reform measures that have been enacted in India, the reform of tenancy and implementation of ‘ceiling laws’ offer the greatest possibilities for the redistribution of economic claims from the well-off to the poor in rural areas, and it is in precisely these areas that West Bengal has a good record.13

The basis of the zamindars' economic position lay not in their outright ownership of landed property, but in their function as revenue agents of the government. While the zamindars did, indeed, hold transferable legal titles to land, these, in fact, implied unrestricted rights to land-revenue collection, but not effective control over the land itself. This remained vested with the ryots as the system (p.138) predated the revenue categories created by the colonial administration,14 With the abolition of revenue intermediation after India's independence (in West Bengal under the Estates Acquisition Act of 1953), private property rights over land became the main determinants of economic position. In the case of West Bengal, these rights reverted to the ryots.15

The fact that a very large section of the rural population in West Bengal had few rights in land was significant, both for the way in which agrarian politics developed, and also for the redistributive possibilities offered by the programme of land reforms. There are few precise estimates of the total number of landless households, but a large proportion of agricultural labourers and tenant farmers are clearly in this category. At the time of the 1981 census, men relying mainly on agricultural labour for their livelihood numbered nearly 4 million workers, comprising over a third of the rural male workforce.16 Assuming that most men reporting agricultural labour as their main occupation belonged to landless or virtually landless households, the incidence of landlessness in West Bengal must have been at least one-third, and possibly more.17 This is very high in comparison with most other Indian states. Micro-level data corroborate these observations about the high incidence of landlessness. In the WIDER villages, for instance, nearly half of all households were landless prior to the land reforms. Estimates of the number of tenant farmers (bargadars) vary between 1.5 to 2 million households, equivalent to nearly a quarter of all rural households in 1981.18

There is likely, of course, to be some overlap between the categories of agricultural labourer and tenant farmer. Some agricultural-labourer households would, undoubtedly, have leased-in land, while members of bargadar households might well have hired themselves (p.139) out as wage labourers. Despite this overlap, for an understanding of how agrarian politics developed in West Bengal, it is useful to keep an eye on the fairly distinctive class histories of these two groups in the agrarian economy. Bose (1993) traces the historical origins of the large class of landless labourers in the zamindari era to the ‘demesne-labour complex’—i.e. the substantial practice of self-cultivation by large landowners of their privately-owned (khas) holdings by means of tied labour. This, according to Bose, was more prevalent in the western parts of undivided Bengal that were to eventually constitute West Bengal. The bargadars, on the other hand, owed their origins as a class to the system of sharecropping in which the relatively more autonomous tenants were the main source of labour for the relatively heterogeneous class of landlords known as jotdars. It is this latter group of the landless and the land-poor (i.e. the bargadars), and not the tied labourers, that were the main base of the early peasants' movement in Bengal.

The post-zamindari agrarian structure of West Bengal has been alternately understood as having been dominated by big landlords and small-to-middle farmers. Bose (1986) has shown convincingly that the jotdars of Bengal were not a homogeneous group in terms of land, resources, and power.19 The absolute domination of big jotdars was confined largely to parts of north Bengal, where they enjoyed virtual monopoly over property rights in land.20 In much of West Bengal, although large landlords existed, and indeed were strengthened when their ranks were joined by former zamindars who acquired khas land in anticipation or as the result of zamindari abolition, their numbers were few, they were regionally dispersed, and they lacked state-wide collective and coherent leadership.21

Agricultural census data on landholdings are consistent with this characterization of land-ownership patterns in West Bengal. By 1971, only 2 per cent of the holdings were larger than 4 hectares, (p.140)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 1. Dutribution of Landholdings in West Bengal and India, 1971 Source, Agricultural Census 1971.

compared to 11 per cent in India as a whole. These holdings accounted for only 15 per cent of total cropped area, compared to 53 per cent in India.22 These proportions are misleading to some extent, since the average size of holdings is lower in West Bengal than in the rest of India. Fig. 1, however, shows the Lorenz curves for the distribution of landholdings in West Bengal and India as-a-whole in 1971. It is clear from the figure that landholdings were more equally distributed in West Bengal than the country as-a-whole. (p.141) The agrarian structure was thus characterized by a relatively high incidence of landlessness on the one hand, and the relatively equal distribution of landholdings (among those with land) on the other.

While bargadars had played an active part in the development of the peasant movement in Bengal,23 by the late sixties it was the demand for the redistribution of ownership rights that came to the fore of the agrarian agenda. In an economy where private-property rights in land had come to form the basis of agrarian production, the labourers and the bargadar tenants, comprising nearly half of all rural households, shared the common feature of having little or no access to such rights. The propertied classes, on the other hand, were dominated numerically as well as in terms of area of land owned by smaller farmers. It was in this context of extremely high incidence of land deprivation, and of relative political and economic isolation of the large landlords, that peasant organizations such as the Krishak Sabha built up considerable support among tenants, marginal farmers, and the landless, and were able to thrust the demand for agrarian reforms on to the agenda of state politics.24

The political power of the big landlords was effectively challenged in the late sixties when wide sections of the small and medium-sized landowners threw their lot in with the struggle of the landless and the land-poor.25 The ‘middle’ and ‘rich’ peasants played a pivotal role in turning the balance of power away from the large landlords, and very often, their involvement in left politics and their support of peasant organizations put them in positions of leadership at local levels.26 It was this broad alliance of the landless (p.142) and the land-poor, led by peasant organizations under the slogan of ‘peasant unity’, with sections of the propertied classes occupying strategic positions, which was to form the electoral base of the Left Front.

These developments which were based on mobilization for agrarian reform, and were particularly aimed at the empowerment of some of the poorest sections of the rural population (namely the landless and the land-poor), were quite revolutionary in the Indian context. However, the legal-institutional framework within which these agrarian politics were to play themselves out was the fairly conservative one that had emerged in India at the time of independence. Agrarian reforms in a West Bengal led by left-wing parties allied to peasant organizations were, as elsewhere in India, limited to the marginal redistribution of private-property rights in land, increasing the crop share for tenants, and giving them security of tenure.

2.2 Redistribution of Land Ownership

If the intended beneficiaries of the land-redistribution programme are the landless and land-poor households, our estimate of their numbers, based upon the 1981 Census, range around 2.5 million households, or just over a third of all households in rural West Bengal. The actual number of beneficiaries of land redistribution in the state by the end of 1993 was around 2.15 million. If all the beneficiaries were indeed landless to begin with, nearly 90 per cent of the landless would have received some land. This last figure may well be an overestimate (given the possibility that some non-landless have benefited from land distribution, and that the incidence of landlessness may have been higher than one-third to start with), but it does suggest that the scale of land redistribution in West Bengal has been comparable with the magnitude of the problem of landlessness.

Although land-ceiling laws had been in operation since the abolition of zamindari in 1953, much of this achievement is of relatively recent origin. Loopholes in these laws were utilized to the fullest by owners of ceiling-surplus land to take evasive action. Benami transfers, that is, the transfer of property into names of friends, (p.143) relatives, and even unwitting tenants, were frequent.27 Furthermore, implementation was the responsibility of petty land officials, most of whom were recruited from the land-revenue officials of the erstwhile zamindars, and were sympathetic to the landlords.28

The demand for the proper implementation of ceiling laws played an important role in the galvanization of support for left parties among the landless and land-poor. The parties of the United Front fought and won the state elections in the late sixties on the slogan of locating and distributing the benami lands, and the proper implementation of the existing ceiling laws. The United Front governments also drafted amendments to the land-reform legislation, and these have remained the basis of land redistribution since. Previously, the basic ceiling had been 25 acres per person. The amended ceiling laws changed the basis of ceilings from individuals to households. They also lowered these ceilings considerably. Households consisting of 2 to 5 persons are now allowed to own up to 12.35 acres of irrigated land or 17.39 acres of unirrigated land, respectively. There are higher allowances for larger households, but the absolute limit has been set at 17.29 acres of irrigated or 24.12 acres of unirrigated land.

Nearly twice as much land was vested with the state between 1967 and 1970 under the two United Front ministries, as had been vested up to then for the entire period since the abolition of zamindari in 1953.29 In terms of area of land vested and redistributed, therefore, the real breakthrough in West Bengal had occurred well before the Left Front came to office in 1977. Some 625,000 acres of land had been redistributed by then (Table 2), and West Bengal already had a commanding position in the implementation of ceiling laws in comparison with other Indian states.

After 1977, during the Left Front period, a further 287,000 acres, representing under a third of the total area redistributed, were assigned (Table 2). The number of beneficiaries, however, increased by almost a million, and more than doubled. The contrast between the pre-and post-1977 land redistribution reflects a number of important differences in the approach to land reforms between the (p.144) Left Front and its predecessors in the United Front, Before commenting on these, it is important to set the record right on some confusion that has crept into the use of official land-redistribution statistics.

TABLE 2. Implementation of Land Ceiling Laws in West Bengal

Up to 1977

Between 1977 and 1983

Between 1983 and 1991

Up to 1991

No. of beneficiaries (households)

984,032

472,443

537,141

1,993,616

Cropped area redistributed (acres)

626,284

140,417

146,688

913,389

Land distributed per beneficiary (acres)

0.64

0.30

0.27

0.46

Sources. West Bengal, Economic Review 1977–78, Statistical Abstract; Ministry of Rural Development, Annual Report 1991–92, Government of India.

A recent study of the Left Front reforms30 concludes, for instance, that there is little to distinguish between the record of the Left Front and the preceding Congress regimes in the implementation of ceiling laws. This observation is based upon the presumption that the Congress regimes were responsible for most of the land redistributed prior to 1977. This presumption is entirely unjustifiable, as it ignores the critical role of the United Front, most of whose constituents went on to form the Left Front, in the transformation in West Bengal.31 During the United Front period, benami lands (p.145) were identified and distributed by means of direct action on the part of the landless and land-poor under the guidance of peasant and political organizations.32 The legal granting of leases often lagged behind, and much of the area that appeared to have been redistributed in the early seventies had already been acquired by the beneficiaries during the ‘land-grab’ movement. It was simply a question of extending de jure recognition of de facto ownership.

In contrast to the direct approach to land redistribution of the late sixties, the Left Front redistribution programme has been less dramatic and more institutionalized. Although political and peasant organizations were actively involved, the main mechanisms of redistribution were local government bodies. Another important difference was that the average area of land distributed per beneficiary was considerably smaller. In this period, beneficiaries received under a third of an acre on average, less than half the average size of allotment prior to 1977 (see Table 2). As for the background of the beneficiaries, and whether or not they came from the target groups, published secondary data are of limited use. One indication that target groups were reached comes from the fact that over half of all beneficiaries (before, as well as after 1977) were from the Scheduled Castes and Scheduled Tribes.

Our observations from village-level data provide more detailed (though less-representative) evidence on this score. They confirm that most of the beneficiaries were people who had not owned any land at all (Table 3). The few who did, owned marginal holdings. In Kuchli, Bhagabanbasan, and Kalmandasguri, beneficiary households constituted about a third of the village population, while in Sahajapur, the proportion was around a quarter. In Magurmari there was only one beneficiary household, which was in fact landed, and which received a relatively large plot of land. In all the other villages besides Magurmari, land-reform implementation appears to be consistent with the aims of policy and the official claims of achievement. Village-level data also confirm that the total area of land available (p.146) for redistribution was a relatively small proportion of total operated area in all villages with the exception of Kalmandasguri in north Bengal. This pattern is not unrepresentative given the domination of small farmers in West Bengal's agrarian structure.

TABLE 3. Land Redistribution in WIDER Survey Villages

Kuchli (Birbhum)

Sahajapur (Birbhum)

Bhagabanbasan (Medinipur)

Simtuni (Purulia)

Magurmari (Jalpaiguri)

Kalmandasguri (Koch Behar)

Area redistributed (acres)

28.2

12.0

8.8

6.5

0.8

19.5

Per cent of total cropped area

11

4

4

4

1

18

Number of landless households

65

131

39

2

71

42

Beneficiaries of redistribution

64

56

49

13

1

30

Landless beneficiaries

43

52

34

1

0

23

Landless non-beneficiaries

22

79

5

1

71

19

Average area per beneficiary (acres)

0.44

0.21

0.18

0.50

0.75

0.65

Average household income of landless assignees from wage labour and assigned land (in rupees)

Wages

3945

4013

3643

3900

Land

1573

418

893

1527

Proportion of total household income of landless assignees from assigned land (per cent)

29

9

20

28

Note. The name of the districts are given in parenthesis.

Source. WIDER village surveys 1987–9.

(p.147) Two of the main administrative objectives of the land-redistribution programme appear to have been fulfilled in the villages surveyed in the WIDER project; there were no households with above-ceiling holdings, and beneficiaries were, by and large, from the target group. Village data, however, highlight some important limitations of land redistribution based on ceiling laws as a comprehensive redistributive system.

Of the six villages surveyed, there were four (Kuchli, Sahajapur, Bhagabanbasan, and Kalmandasguri) where the landless poor benefited from land redistribution. The two villages where the impact was minimal were Simtuni, which had no landlessness to begin with, and Magurmari, where due to proximity to a town, there was a high rate of participation in the non-agricultural economy. Sahajapur in Birbhum was the most populous village in our sample and also had a high proportion of landless households. A majority of the landless households (over 60 per cent of them) in that village received no land at all. This was in spite of the fact that the average area distributed per beneficiary was quite modest. In fact, if all landless households had received some land, the average area per beneficiary would have amounted to under a tenth of an acre. If the Sahajapur pattern of beneficiary selection was typical of the state in general, a large proportion of the 2.5 million or so estimated poor landless households would not have received any land.

The availability of ceiling-surplus land for redistribution depends, of course, on the initial distribution of land ownership, and the level at which the ceiling is fixed. For any given level of land ceiling, the more equal the initial distribution, the smaller the area of land that would be available for redistribution. Conversely, for a given distribution of land ownership, the lower the land ceiling, the larger the ceiling-surplus area (as well as potential losers from land redistribution). If the area per beneficiary is fixed, the extent to which the target group can be covered would depend on the initial distribution and the land ceiling. The relatively less unequal land-ownership structure in West Bengal (among those who own land) implies that only modest areas of land are available for redistribution unless ceilings are drastically reduced.

In actual fact none of these policy targets (land ceilings, area per beneficiary, number of potential losers, or indeed the number of intended beneficiaries) is exogenously fixed and all are subject to negotiation and political judgement including electoral calculus. (p.148) The class alignment between small-holders and the landless which emerged in the late sixties, and which formed the basis of the Left Front's electoral support could be sustained by maximizing the number of the landless beneficiaries while ensuring that most of the small-holders and indeed other landowning groups who constituted Left Front supporters were not adversely affected. The small size of the average area of land assigned appears to have been an outcome of such a c`lculus.

The impact of assigned land on the livelihoods of landless beneficiaries in the WIDER villages brings into focus three important features of the land-redistribution programme in West Bengal. Firstly, in spite of its administrative success, land redistribution under existing ceiling laws has been marginal in the sense that it ameliorated but did not resolve class contradictions, or significantly alter class relations, between the land-rich and the land-poor. For landless beneficiaries in all villages, wage labour remained, by far, the most important source of earning (Table 3).

Secondly, although they remained primarily dependent on wage labour for their livelihoods, income from assigned land made a substantial contribution to the incomes of the landless beneficiaries. It accounted for nearly three-tenths (29 and 28 per cent, respectively) of total earnings in Kuchli and Kalmandasguri, a fifth (20 per cent) in Bhagabanbasan, and under a tenth (9 per cent) in Sahajapur.

Thirdly, although implementation appeared to have been fair in all villages (with the exception of Magurmari) its impact on the livelihoods of the poor varied greatly between villages. In absolute terms, the beneficiaries in Sahajapur received marginally more land, on average, than their counterparts in Bhagabanbasan, and yet their earnings from land were under half of the latter (Table 3). The higher cropping intensity in Bhagabanbasan accounted for the difference. In Kuchli and Kalmandasguri, average areas per beneficiary were higher. The final impact of land redistribution on the livelihoods of the poor depended on the initial distribution of landholding within the village (i.e. the availability of surplus land), as well as the local agro-economic conditions (i.e. the productivity of land). As an instrument for reducing income inequality, then, the land-redistribution programme with its focus on localized redistribution was successful in reducing intra-village inequalities, but was largely ineffectual in dealing with inter-village inequalities or, indeed, inter-regional inequalities where these were determined by agro-climatic variations.

(p.149) 2.3 Tenancy and ‘operation Barga

Tenant farmers (or bargadars) bad long been regarded as a vulnerable and exploited section of the rural population. The development of the peasant movement in Bengal was closely associated with activism for the rights of tenants.33 For decades, demands of higher crop shares, security of tenure (including the right to inherit tenancy), and legal endorsement of tenancy agreements have been raised and fought for by peasant organizations.

Operation Barga, preparations for which began soon after the Left Front assumed office, has been described as a culmination of these movements.34 The Left Front government's amended tenancy legislation incorporated all the key demands over which peasant movements and organizations had fought. Tenancy was made inheritable, crop shares were fixed at between two-thirds and three-quarters depending on the sharing of inputs, and clauses in earlier legislation that provided loopholes for eviction of tenants were plugged.35 Eviction of tenants was the main obstacle to the enforcement of other demands such as higher crop shares. The threat of eviction and the availability of other tenants who might accept lower terms, made it impossible for a tenant to sustain a bargaining position vis-à-vis the landlord.36 Recording of tenancies and enhancement of their legal status were rightly identified as the necessary conditions for other aspects of the legislation to work. Operation Barga was launched in 1978 with the primary purpose of recording tenancy leases.

Government claims of having recorded 1.44 million sharecroppers have been accepted widely. Disagreements exist about whether (p.150) there are significant numbers who are not recorded,37 Since estimates of the number of tenant households range between 1.5 and 2 million, at least in terms of reaching a substantial part of the target group, Operation Barga has been a success. There is support for this assessment from other sources also. A survey of 414 kisani bargadars38 in the Nanoor area of Birbhum found that 370 or nearly 90 per cent had been registered.39 Out of these 370, however, 16 had since been evicted, most of them in exchange for outright ownership of small plots of land.40 Lieten (1992) who surveyed a kisani village in Birbhum also found that most of the tenants had registered.

Bargadars accounted for less than ten per cent of all households in the WIDER villages and the area under tenancy ranged from 3 to 17 per cent of total cropped area (Table 4). in two of the south Bengal villages (Kuchli and Bhagabanbasan) the land area under annual lease was 5 per cent or less of the total cropped area of the village. Nearly two-thirds of all leases were found to have been recorded. The only village with a relatively high share of land under tenancy (Magurmari in north Bengal) also had two-thirds of the bargadars, accounting for over 90 per cent of leased area, registered. Although a majority of the leases, and most of the area under these leases had been registered, there was still a significant proportion of the transactions that were not.

The real test of the success of the campaign is its effect on the share of the produce that actually goes to the tenant. The evidence on crop shares, mainly to be found in village-level surveys, is mixed. The Nanoor study found that only a third of the recorded bargadars interviewed received the stipulated 75 per cent share. An earlier study of villages in Birbhum and other areas found that the legally (p.151) stipulated share was rarely given.41 A survey that included both recorded and unrecorded tenants in two regions of Medinipur district found that recorded bargadars consistently received a greater share of the crop than unrecorded ones.42

TABLE 4. Land Tenancy and Registration of Leases in WIDER Survey Villages

Kuchli (Birbhum)

Sahajapur (Birbhum)

Bhagabanbasan (Medinipur)

Magurmari (Jalpaiguri)

Kalmandasguri (Koch Behar)

Area under annual leases (acres)

13.8

22.3

6.6

11.6

10.1

As per cent of total cropped area

5.0

7.8

3.0

17

9.2

Per cent of annually leased area registered

68

52

68

91

72

No. of annual leases

12

17

12

9

10

No. registered

4

10

9

6

8

Area seasonally leased (acres)

25.0

2.9

9.7

0.0

1.1

No. of seasonal leases

49

10

13

0

2

Note. The names of the districts are given in parenthesis.

Source. WIDER village surveys 1987–9.

Operation Barga has been able to extend security of tenure to tenants who previously faced the constant threat of eviction, and this has translated into higher crop shares for tenants, even though these are frequently lower than the legally stipulated shares. Time and again the propertied classes in rural India (including West Bengal) have acted in anticipation of redistributive policies by deploying legal loopholes and taking preemptive action. The expeditious recording (p.152) of land leases under Operation Barga, thus made tenancy security laws enforceable for the first time. The campaign approach that was adopted for this task showed that redistributive public policy need not stand helpless in the face of evasive and collusive strategies of propertied classes. To this extent, then, Operation Barga was successful in meeting its objectives.

There are a number of questions, however, about the economic wisdom of tenancy regulation in general, and the continued relevance of such regulation as a redistributive measure in the context of present-day West Bengal in particular. In theoretical models, sharecropping is usually understood as a way of combining different inputs under conditions of uneven endowments of land, labour, creditworthiness, and other inputs, on the one hand, and the absence of markets due to informational asymmetries, on the other. Tenancy in general, and sharecropping in particular, are not therefore, in themselves, causes of economic inequality and exploitation, but simply the means of organizing agricultural production under conditions of inequality in the ownership of assets.

Tenancy regulation and the legal enforcement of crop shares have potentially redistributive functions if tenants are universally poorer than landlords (since such reforms are primarily concerned with strengthening the tenant's position), and if crop shares are at least partly determined by the respective bargaining positions of the two parties. Even under these conditions, however, tenancy regulation is likely to be a less effective means of bringing about greater equality in incomes than the outright redistribution of property rights in land (if inequality in land ownership is, indeed, the main source of income inequality). Moreover, although the traditional pattern of rich landlords leasing land out to poor tenants may have been a valid generalization in the zamindari period, tenancy relations may no longer correspond neatly with class relations.43

The long-term effects of tenancy regulation on productivity and economic efficiency are not easy to discern by theoretical reasoning alone. On the positive side, security of tenure does provide greater incentives for longer-term investments by tenants in improving land quality. If, on the other hand, sharecropping contracts are second-best institutional responses to imperfections or failures in various (p.153) input markets, then conferring legal status on these contracts might hinder their eventual replacement by more efficient forms of input sharing. The secular decline, over the three decades preceding Operation Barga, in the proportion of agricultural holdings in the form of pure tenancy leases may well have been a response to changing economic and demographic conditions that allowed for such improvements in efficiency.44

Another argument against tenancy regulation is that statutory intervention might impose uniformity and rigidity on an otherwise complex and changing set of economic relations. Rudra (1981) has argued, for instance, that the diversity of crop shares that existed in West Bengal reflected diverse cost-sharing arrangements, and legal fixing of shares without improving the registered bargadars' access to institutional credit would leave them credit-constrained.45

Indeed, in a political climate generally favourable to recording, the failure of some tenants to register cannot be put down simply to coercive pressure by landlords. There may be advantages in preserving amicable relations with a landlord who also acts as a source of credit.46 Lieten (1992) found in his village study that most of the bargadars who had failed to register had done so because they wished to maintain cooperative relations with their landlords. Bhaumik (1993) found evidence of greater economic interaction between the landlords and unrecorded tenants than recorded tenants. He also found that the caste backgrounds of the vast majority of the unrecorded tenants was similar to that of the lessors. Both groups were predominantly from the general caste, as opposed to the domination of Scheduled Caste and Scheduled Tribe tenants among the recorded lessees. This caste-wise pattern of recording tenancies possibly reflects the different levels of cooperation and conflict between landlords and tenants with and without mutual kinship and other social ties.

(p.154) Although on the surface a tenancy contract appears to be mainly about the leasing of land, in actual fact it may embody a great diversity of economic relationships. Even the best of regulatory laws is bound to overlook some of these complexities, and may, therefore, hinder the development of mutually gainful economic transactions. There is some evidence that tenancy regulation in West Bengal might have had such an effect. This is in spite of the fact that many landlords and tenants have ignored the regulations and have chosen to continue with traditional crop and cost-sharing arrangements. In his survey of tenancy contracts, for instance, Bhaumik (1993) finds that the relatively more recent tenancy contracts are less likely to be recorded than long-standing ones. Since there is no legal restriction for a tenant filing a claim to be registered, a potential lessor is constrained in his choice of tenants to those with whom he shares a relationship of trust. Since the landless and the land-poor continue to participate voluntarily in unrecorded contracts, an unwanted side-effect of tenancy regulation from the distributional point of view might have been the restriction in the supply of new leases.

Recent years have seen the emergence of seasonal leasing contracts mainly in the summer months in regions where groundwater irrigation has been developed. This form of leasing is regarded as casual and does not come within the ambit of tenancy laws. It has become a profitable venture, as it allows households with access to groundwater to take in neighbouring fields and thus to enjoy benefits of scale economies. In Kuchli and Bhagabanbasan (the two villages in our study whhch have extensive irrigation throughout the year), the total area leased seasonally, and the number of seasonal leases, exceeded the area and number of long-term annual contracts (Table 4). Although larger farmers with access to groundwater irrigation lease in plots of land from smaller landowners in the neighbourhood of their water source, seasonal leasing is not limited to the larger farmers, by any means. In Kuchli, over 43 out of the 49 seasonal leases were taken by households with less than 2 acres, and in Bhagabanbasan, the corresponding number was 9 out of 13. In spite of the long-term decline in the incidence of sharecropping, its resilience as the basis of new rental contracts provides some support for the view that as a form of contract it fulfils important economic functions.

In sum, while tenancy regulation has, clearly, enhanced the bargaining position of tenants, the efficiency effects of such a system are (p.155) not likely to have been unambiguously positive.47 On the one hand, regulation may perpetuate contracts that might no longer be economically efficient. On the other hand, it acts as a constraint on the emergence of new contracts that are potentially profitable. Moreover, the parties to this form of economic transaction no longer correspond closely to different classes, as they might have done at the time of the Tebhaga struggle and other similar mobilizations. As a means of redistributing economic claims from the rich to the poor, therefore, tenancy regulation is less effective than it might have been in the past, and is, in any case, second-best to the outright transfer of property rights in land.

It is critical to grasp that the regulation of a particular form of economic transaction was incidental to the change in the balance of power between the dominant propertied classes and the poor. The real engine of that change was the successful mobilization of claims on economic resources on the part of large sections of the poor. The fact that this mobilization took place around the issue of tenancy security is partly a result of the history of agitation on this issue by left-wing political parties and peasant organizations. Although Operation Barga was one of the flagship reforms of the Left Front government, its effectiveness as a redistributive programme (compared to land redistribution) and its value as a productivity-raising measure need to be interpreted with some caution.

The political impact, however, of an energetic campaign for the creation of tenancy rights for nearly one-and-a-half million households immediately after the election of the new government was extremely significant indeed. It sent a strong signal to the supporters as well as the opponents of the Left Front that the change of regime was real and that the new administration was serious about implementing agrarian reforms.

2.4 Local Democracy and Political Consolidation

Besides agrarian reforms, the main plank of the Left Front rural programme was the revitalization of the local government Panchayati Raj Institutions (PRIs). West Bengal is the only state in India (p.156) where elections to panchayats have been held on a regular basis and contested along party-political lines,48 The reform of local government was initiated soon after the Left Front government was formed in 1977. The first elections were held in 1978, and since then there have been three further polls. The success of Left Front candidates in these elections has been overwhelming and consistent. Their share of the popular vote has remained around 70 per cent. Indeed, the recent central government legislation on the reform of PRIs includes key aspects of the West Bengal experience such as a multi-tier structure (with direct electoral representation at each tier), regular elections, reservation of seats for Scheduled Caste and Scheduled Tribe candidates, and a similar quota for women.

Post-reform PRIs have come to represent an alternative structure of authority to the police, civil service officers, and other official departments such as revenue and irrigation, in the rural areas.49 This is a significant departure from past practice in West Bengal, as well as from the present situation in much of India where the system of administration inherited from colonial times has retained much of its authority in spite of a plethora of formally representative bodies.

West Bengal PRIs have also been compared favourably in field-based investigations of the implementation of poverty-alleviation programmes. The Integrated Rural Development Programme (IRDP), which, in essence, is a subsidized credit scheme is one of the major anti-poverty programmes of the Indian government, whose implementation is entrusted to state governments. This provides a useful basis for inter-state comparison.50 Most beneficiaries in West Bengal were found to be from the target group, and transactions costs of obtaining the loan were relatively small. This contrasted with other parts of India where beneficiaries were often the well-off relatives of (p.157) panchayat officials, and intended target groups faced high transactions costs including the bribing of officials. This is not to say that IRDP or similar programmes have actually achieved what they set out to in West Bengal. The only aspect of implementation that we have focused on is the selection of beneficiaries, as in schemes such as IRDP where the target groups are administratively identified, the scope for leakage is quite high. In other matters, such as the choice of investment project, training of beneficiaries, backup services, etc., the record of West Bengal is not much better than most other parts of India.51 In a study focused on the implementation of the jawahar Rozgar Yojana (JRY), which is a government-run employment generation programme for the rural poor, Echeverri-Gent (1992) found that the scheme was run efficiently, served the target groups, and was relatively free of corruption.52

One explanation for the relatively better implementation of these programmes in West Bengal is that there was greater popular participation. It is now widely accepted that one of the key aspects of rural development is the participation of the rural people directly in the process of planning as well as the implementation of programmes. Local people have better information about their own needs and conditions. In the case of targeted poverty-alleviation programmes, ‘objective’ measures such as household income may be inappropriate indicators of poverty status for a wide range of reasons. Identification of the poorest using other criteria, including the perceptions of villagers, can yield quite different results from formal income criteria.53 Local participation in implementation has the advantage of bringing local information into the picture.

Popular Participation and Electoral Democracy

In West Bengal, however, participation has not, by and large, taken the form of gram sabha (village assembly) meetings where allocations (p.158) of IRDP or JRY funds have been discussed in detail. A number of studies of the working of PRIs in West Bengal indicate that decisions about expenditure priorities are rarely based on the prescribed forms of consultation (e.g. gram sabha) between panchayat leaders and the villagers they represent. This does not imply that there is no consultation or participation. In fact, there is a great deal of informal discussion of these matters. A survey of perceptions of villagers about the working of the panchayats in their villages, conducted as part of the WIDER study, also showed that in general the panchayat leaders reflected the opinions of their constituents (see Mukherjee, 1990). In particular, installation of wells and taps for potable water, and the building of roads, were considered important areas requiring attention, by villagers as well as panchayat members.

It is, in fact, mainly through the electoral process (which has been institutionalized and regularized) that popular participation has been practised in West Bengal. Village-level studies have shown that the poor, including those from socially-deprived groups such as Scheduled Castes and Scheduled Tribes, not only participate in elections as voters, but also stand as candidates.54 It can be argued, with much justification, that real empowerment involves going well beyond participation in elections, and we return to some of these concerns later.

The fact remains, however, that electoral accountability, and the responsiveness of all major political parties to the concerns of voters, have contributed to the relatively better performance of PRIs in West Bengal compared to other states. Echeverri-Gent (1992) has argued that democratic competition between political parties has given incentives to the CPM and the Left Front leadership to ‘monitor the performance of panchayat members and weed out those engaged in corruption’.55 Westergaard (1986) found similar pressures operating in her survey villages, but did come across isolated cases where corrupt individuals were retained as candidates due to their ability to deliver votes. A number of recent studies report cases of corrupt and unscrupulous panchayat leaders, including many who were Left Front supporters. It is significant, however, that in nearly (p.159) all reported cases, such leaders became electoral liabilities to their parties.56

Electoral competition alone cannot explain this situation. Instead of providing a forum for public accountability, such contest can, and does frequently, end up as an arena for factional rivalry within elite groups, with loyalties being divided along the lines of caste, faction, grouping, etc.57 This certainly appears to have been the case in West Bengal before the Left Front period.58 The CPM and the Left Front parties are themselves not immune to using caste and group loyalties to electoral advantage. This has been shown by a number of recent field-based studies of village-level politics.59 What is important, however, is that unlike many other parts of India, caste or group-based factional rivalry is not the only basis on which politics are organized in rural West Bengal. A consequence of the agrarian struggles and the mass mobilization of some of the poorest people for their economic rights has been the raising of political awareness. The poor, therefore, are no longer pliable clients of local elites, but assertive and vigilant participants in local democracy.60

PRI Revitalization and Agrarian Reforms

The distinctive features of Panchayati Raj in West Bengal relate not so much to their formal structure as to the success that has been achieved in giving these structures some vitality. The main factors that have distinguished West Bengal's PRIs from those of other states in the formal sense are that elections have been overtly party-political, that they have been held with great regularity, and that a policy of positive discrimination in favour of women and Scheduled (p.160) Caste/Scheduled Tribe candidates has been in place.61 These factors have certainly contributed to the revitalization of the PRIs in West Bengal But, at a more basic level, the real difference between West Bengal and other states lies in the fact that PRIs here are important arenas of political activity, that are taken seriously by the voters as well as the main political contenders at state and centre levels, and that they represent effective loci of power and governance. The political significance of the PRIs in the state owes much to the prominence given to them in the agrarian-reform programmes of the Left Front.

Panchayats have been involved from the start in the agrarian-reform programmes of the Left Front. As such, they have been part and parcel of the programme of change, including the redistribution of power from landed elites to an alliance of small farmers and the landless poor. Whereas implementation of agrarian-reform legislation has been largely a bureaucratic affair over much of India (as in West Bengal prior to the ascendancy of the left parties), and was mainly the result of direct action by beneficiaries and peasant organizations under the United Front periods, during the Left Front era the energies of the peasant and party organizations were channelized through the PRI structure. The presence of loci of political power in rural areas that were not subservient to old-standing propertied interests was critical in the implementation of the agrarian reforms.62 In turn, the diminished economic power of the landed elites opened the door for wider participation by the poor and socially-deprived groups in political processes. The pre-eminence of the agrarian agenda in state politics, and the central role assumed by PRIs in this regard, has been responsible for their becoming active arenas of politics.

These institutions have also served the political and electoral interests of the CPM and the Left Front government by allowing (p.161) them to consolidate their political support, and by giving rise to power structures that operate broadly under the control of the state government. This is neither surprising nor unusual, as within a pluralistic and competitive electoral system, any political party ought to be expected to look after its electoral interests. Where the Left Front can claim to have scored a remarkable success is in the institutionalization of agrarian reforms through the PRIs. The fact that all important political parties regularly contest the local government elections confers almost consensual validation to agrarian reforms which are now part of the new political reality.

Prom Activism and Agitation To Bureaucracy and Conciliation

The activities of the PRIs continue to reflect their association with the agrarian agenda of class politics pursued in the state. Not only were they involved in the early implementation of land redistribution and Operation Barga, these institutions have also acquired the role of guarantors of these newly acquired rights in property and tenancy. With the scope for further agrarian reforms seriously constrained on political as well as legal grounds, however, there have been concerns of late that the panchayats in West Bengal are ‘running out of steam’. In a report to the state government, Mukarji and Bandyopadhyay (1993) warn that with the implementation of agrarian reforms being exhausted, panchayats run the risk of losing their sense of direction. On the basis of a survey of PRIs across West Bengal, they find that the activities of the panchayats are now largely confined to carrying out public works under the JRY. According to these authors, the realization of local self-government would remain incomplete without an active programme of development.

A related issue is the change in the character of PRIs over time. Webster (1992) argues that ‘if the programme continues to change from being a political movement and increasingly becomes a bureaucratic strategy with its only political goal being the re-election of the Left Front to state government, it will fail in its original goals and disillusionment could easily result in the fall of the Left Front’.63 There is a sense, however, in which the change from ‘class mobilization’ to a more conciliatory approach is inevitable. While the rise of the Left Front was associated with the sharpening of class conflict (p.162) in rural West Bengal and the shifting of class alliances, its consolidation has necessitated a more conciliatory approach. Consequently, an important role of PRIs has been that of mediation between conflicting claims in matters such as terms of lease and agricultural wages.

The alliance between the ‘middle peasantry’ and the landless and small-holder poor, which was instrumental in the successful challenge to the position of rich landlords, blurs potentially conflicting economic claims from different groups within that alliance.64 While the landless poor and the small-holders have made political as well as economic gains, the cost has been borne, thus far, by small numbers of landlords with relatively large holdings. Other landowners (defined as middle and rich peasants) have remained relatively unscathed, and many have, in fact, acquired political power due to their association with Left Front parties.

It is easy to see that any intensification of agrarian reforms or class mobilization would pose a threat to the stability of this political alliance. Releasing more land for redistribution, for instance, would require the further lowering of land ceilings to levels that would begin to affect the smaller landowners (including ‘middle peasants’) adversely. Similarly, high wage claims by agricultural labourers would affect large landlords as well as those numerous smaller farmers who also hire labourers. The fact that panchayats are often involved in the mediation of such disputes puts them in a strategic position for the management of potential conflict between segments of Left Front support.

A Basis for Effective Governance

The decreasing emphasis on political activism on the part of PRIs (and indeed on the part of Left Front organizations) and the increasing emphasis on limited routine functions such as the implementation of central government ought not to be surprising. The scope for further agrarian reforms is restricted by legal as well as political considerations, and therefore programmes like the JRY and IRDP provide the mainstay of the panchayats' activities. It is possible to argue, of course, that the poor bear a disproportionately greater cost (p.163) of maintaining the stability of the Left Front base in terms of the opportunities of further reform that are foregone. Whether or not agrarian reforms could be taken further is a complex issue, and much turns on the political judgement of the observer.

The debate about the scope for activism on the part of the Left Front organizations and the PRIs ought not to be limited, however, to agrarian reforms alone. In post-reform West Bengal, PRIs can act as instruments of public policy much more efficiently than they are able to do over most other parts of the country. Their role in the implementation of IRDP and JRY is simply an example of the potential they represent. Other interventions, such as the policy of positive discrimination in favour of the political representation of women and people belonging to Scheduled Castes and Scheduled Tribes, and the recent literacy campaign in the state, have been implemented through the PRIs. The fact remains, however, that such actions have been piecemeal, and that after the exhaustion of the agrarian-reforms agenda, the PRIs are left without a clearly defined programme. While the agrarian-reforms agenda has its own limitations, the potential created in West Bengal due to the institution of effective governance remains largely unfulfilled. We return to some of these concerns in sections 4 and 5 below.

3. Impact on Livelihoods

3.1 Agricultural Growth

After nearly two decades of stagnation, agricultural production—particularly the output of foodgrains—took off from the mid-eighties onwards. Between 1969–70 and 1979–80, total foodgrains production increased at an average rate of 2.5 per cent a year in India as a whole, and at 1.7 per cent per year in West Bengal—lagging far behind the rate of increase of population. From 1979–80 to 1989–90, however, West Bengal's foodgrains output grew at an average rate of 3.4 per cent per year, compared to India's 2.7 per cent.65 This trend has continued into the nineties, and for the entire period from 1978 to 1991, West Bengal's trend growth rate in foodgrains output was 4.6 per cent compared with 2.8 per cent for the country as a whole.

(p.164) The break in trend appears to have occurred around 1983 when, following two years of disastrous harvests, output increased by over 50 per cent from 5.9 million tonnes in 1982 to 9.2 million tonnes in 1983. Following over a decade of high rates of growth, it is possible to refer to this break in trend as a historical point of departure.66 Saha and Swaminathan (1994) who tested for a break in trend in foodgrains output in the early eighties found it to be highly significant. They suggest a positive link between the Left Front

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 2. West Bengal's Share of India's Total Foodgrains Outpu Source. CMIE, Performance of Agriculture in Major States, July 1993.

(p.165) reforms and agricultural performance. Harriss (1993) arrived at similar conclusions about overall trends—that there was a turnaround in West Bengal's agricultural performance—from village re-surveys, but gave greater weight in the explanation to technological developments and farmers' response to relative price changes. The dichotomy between policy-led and market-induced institutional innovation that is implied in these two respective contributions is, in our opinion, too narrow a framework for understanding the recent developments in the agro-economy. Before addressing the explanations of growth, however, it is quite useful to place it in a temporal and geographic context.

Agricultural growth in recent years was preceded not just by stagnation, but by relative decline. The historically high growth rates have thus far simply restored the state's relative position in grain production in India as it stood in the mid-seventies. Fig. 2 plots West Bengal's share in India's total foodgrains output for the period 1970–1 to 1991–2, It shows that by the end of this period, despite rapid growth in output, West Bengal's share in India's total output of foodgrains (7.6 per cent) was only slightly higher than it had been in 1977–8 (7.0 per cent). Rapid growth in the eighties, then, has mainly taken the form of reversing and compensating for the relative declines of the seventies.

A longer view of the performance of West Bengal and the neighbouring states (Fig. 3) also confirms the picture of relative stagnation up to around 1982–3 and rapid growth since then.67 Fig. 3 shows that foodgrains output declined substantially in West Bengal and somewhat less sharply in Bihar. It is clear from the plot that due to the wide fluctuations in output, the direction and significance of any trend estimates would depend to a great extent on the sample period selected. While the performance of West Bengal's agriculture has been particularly remarkable throughout the late eighties, any analysis which takes a starting point in the early eighties would tend to overestimate the upward trend.68 Our own estimates of the trend growth rates for West Bengal vary from 4.3 per cent if the period chosen is 1983–91, to 6.6 per cent for the period 1981–91.

Furthermore, although West Bengal experienced the highest rate (p.166)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 3. Comparative Trends in the Output of Foodgrains.

Source. CMIE, Performance of Agriculture in Major States, July 1993.

of growth in the country during this period, the trend break in the early eighties also occurred in Bihar and Orissa, two states that share some of the agro-economic features of West Bengal.69 Compared to the all-India trend growth of 2.7 per cent, and West Bengal's trend of 6.6 per cent (between 1981 and 1991), foodgrains production grew at 3.5 per cent a year in Bihar and at 4 per cent in Orissa. Explanations for the break in trend, therefore, need not be specific to West Bengal, although the higher rates of growth in the state do require specific attention.

The proximate causes of this departure from trend have been ascribed to two simultaneous processes: improved irrigation that led (p.167) to higher cropping intensities, particularly the area under the summer boro rice crop, and the adoption of higher-yielding varieties for the main aman rice crop. While the total production of foodgrains in 1991–2 was about 4.6 million tonnes higher than 1980–1, the respective contributions of boro and aman crops to this increase were 2.0 and 2.2 million tonnes.70 The adoption of a higher-yielding variety of the aman crop is not, of course, unrelated to the availability of water. More control over water makes switching over to a higher-yielding but expensive seed more lucrative.71 Another factor behind higher yields in the aman crop, however, has been the development of HYVs that are more robust and better suited to local conditions. According to farmers in Sahajapur—a village with low groundwater exploitation—the new HYV which first came to the area in 1984 had a lower yield than the HYVs that had been around earlier, but was also less vulnerable to variability in water supply, and more resilient to pests.

The effect of the development of irrigation could be inferred from the well-known study by Boyce (1987) of agricultural stagnation in Bengal (including West Bengal as well as Bangladesh). This influential work (which was published, ironically, at a time when growth was taking off), provides a useful frame of reference. Boyce (1987) showed that there was great potential for agricultural growth in Bengal, that water was the ‘leading’ input, and that institutional innovation for the exploitation of groundwater sources was the main binding constraint. Cooperative rather than market-based innovation was considered to be the most viable mechanism for change. Subsequent experience appears to confirm the author's optimism about the agricultural potential of West Bengal, as well as the important role of water resources, although Boyce's pessimism regarding the development of private water markets appears to have been unwarranted. Private investment in irrigation has been the main source of groundwater development, and the emergence of markets for water was the main institutional innovation.

The fact that accelerated agricultural growth in West Bengal has coincided with the Left Front agrarian reforms has provided some scope for speculation about the possible effects of reforms on (p.168) productivity It has revived some early concerns in agricultural economics with establishing credentials for land reforms on efficiency grounds. Some of the early political rhetoric of left parties also carried an echo of these themes and was couched in terms such as ‘unfettering the forces of production’. The fact that accelerated growth has also taken place in eastern India as well as Bangladesh over much the same period, however, suggests that the agrarian reforms in West Bengal were not the main cause of the agricultural take-off. The relative strength of growth in West Bengal compared to its neighbouring states does, nevertheless, warrant explanation.

Specific land-reform measures, on their own, however, are unlikely candidates for this. Despite their wide coverage in terms of beneficiaries, the redistributive reforms have only affected a relatively small proportion of total cultivable area. The amount of cropped land distributed under ceiling laws, for instance, represents around 6.5 per cent of total cropped area in the state, less than a third of which was distributed after 1977. Precise estimates of the total area registered under Operation Barga are not available, but this is unlikely to have exceeded 15 per cent of the total cropped area. For land redistribution and Operation Barga to be the driving forces behind accelerated growth, these relatively small areas of land would have had to achieve extraordinarily high rates of productivity growth. This, clearly, has not been the case. Instead, there has been wide adoption of HYVs for aman, and an extensive increase in (irrigated) boro cultivation.

The dichotomization of the explanations of recent agricultural growth in West Bengal between market versus non-market innovations, or reforms versus private incentives, is problematic from both conceptual as well as empirical viewpoints.72 Private agents, after all, operate (and respond to incentives) within an overall economic context that is conditioned to a great extent by the distribution of assets as well as political power. It is true that the development of the groundwater irrigation potential has been largely down to farmers investing in tubewells for their own land and for selling water to neighbouring farmers, while the programme for public tubewells has been largely unsuccessful. At the same time, however, the improved electricity supply to rural areas, which (p.169) has enabled the installation of private tubewells, is partly a reflection of the stronger bargaining position of rural producers vis-à-vis their urban counterparts as a result of political empowerment. Similarly, while private incentives, such as the fall in the price of fertilizers relative to rice, have led to increased consumption of fertilizers, the effective operation of local-level government has probably played a positive role in the flow of information, the timeliness of supply, and the general functioning of input delivery systems.73

In short, the main factors behind the agricultural take-off in West Bengal are shared with other states of the eastern region, with similar agro-economies. West Bengal's particularly strong record may reflect the state's greater agronomic potential than its neighbours. Additionally, the combined effect of agrarian and political reforms is likely to have been positive. The mechanisms through which agrarian reforms have had a positive impact, however, are likely to be those that improved the general functioning of markets and enabled agents to participate in these, rather than efficiency effects relating specifically to land redistribution or tenancy regulation.

The tendency to seek retrospective validation for agrarian reforms on efficiency grounds is surprising, since these arguments have been largely incidental to the political mobilization around agrarian issues in West Bengal, which was based squarely in redistributive terms. For practical purposes, it is sufficient to note that the recent growth record of West Bengal dispels any possible apprehensions that redistributive reforms might have had negative effects on efficiency.74 The more interesting question, in our opinion, is whether and to what extent the poor have been able to participate in the growth that has occurred.

(p.170) 3.2 Poverty, Consumption and Distribution

The most commonly used indicator of absolute poverty is the head-count ratio, or HCR, which measures the proportion of the population with incomes (or expenditures) below the poverty line. There has been wide acceptance among analysts of Indian poverty of the basic poverty line adopted by the Planning Commission over three decades ago and periodically updated since then.75 Here, we use the series compiled by Tendulkar, Sundaram, and Jain (1993) based on the method suggested by M in has, Sundaram, and Jain (1991).76 For the distribution of income and expenditure, the main sources of information are the successive rounds of National Sample Surveys (NSS) which are designed to be statistically representative.

Fig. 4 traces the rural HCR for West Bengal, its neighbouring states, and India as-a-whole over the seventies and eighties. When the Left Front government came to office in 1977, West Bengal had the highest HCR of all Indian states. By 1983 it had overtaken Bihar, and in 1987 it left both Bihar and Orissa well behind. The decline in its HCR was particularly marked in the last period (1983–4 to 1987–8) when it experienced the highest proportionate decline of all the major Indian states. Besides a fall in the HCR, West Bengal registered consistent declines in distributionally-sensitive poverty indicators throughout the period from 1972–3 to 1987–8 (Tendulkar et al., 1993).

NSS samples are constructed to be representative of the population at large, but the surveys do not collect panel data. In other words, they select different villages and households for each round. There is some value in making comparisons over time with the same set of respondent households. This was done for West Bengal by Bhattacharya, Chattopadhyay, and Rudra (1987a, 1987b, 1987c, 1987d, 1987e, and 1988). These studies are based on re-surveys in 1985–6 of villages and households which were first surveyed in the 1972–3 and 1973–4 rounds of the NSS in the districts of Birbhum, Bardhaman, and Purulia. These re-surveys are valuable not only because they cover a period of great interest, but also because they provide a sound basis for inter-temporal comparisons, given their (p.171)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 4. Comparative Trends in the Proportion of the Rural Population below the Poverty Line

Source. Tendulkar et al. (1993) based on successive rounds of NSS.

meticulous sampling and data collection method. Further, they focus on direct measures such as the consumption of various private and public goods, rather than attempting to construct income measures (which although perhaps analytically superior, are sensitive to assumptions that must be made about relative consumption weights and prices in any inter-temporal assessment).

The ‘levels of living’ studies found improvements in the overall level of consumption of private goods in the period between 1972–3 and 1985–6. The situation varied from one commodity group to another, but the overall picture was described by the authors as having been no more than a ‘mild improvement’. The consumption of clothing, footwear, and consumer durables such as torches, wrist (p.172) watches, bicycles, furniture, and utensils remained low, but nevertheless showed improvement. Housing conditions were found to have deteriorated in terms of space per capita, and while some people improved the structure of their houses (for example, from kuchcha to semi-pukka), for others it deteriorated. The per-capita consumption of most food items increased between the first survey and the re-survey. The important exception concerned cereals, the average consumption of which stagnated.

The authors of these studies rightly highlight the marginal nature of overall improvements in consumption levels. However, they prematurely dismiss the possibility that small increases in aggregate consumption might have gone hand in hand with some significant distributional improvement. They remark ‘it is possible that even the mild improvements in non-food consumption affected only the upper strata of the rural population’ (Bhattacharya et al., 1987a, p. 1150). This, however, is inconsistent with further results subsequently published by the same authors (Bhattacharya et al., 1988). The analysis of cereal consumption by socio-economic group presented in the latter paper shows, for instance, that per-capita consumption of cereals declined from about 0.55 kgs to just over 0.50 kgs a day for owner-cultivators, and increased from 0.43 kgs to 0.48 kgs a day for agricultural labourers.

In fact, there were substantial improvements in the distribution of expenditure between 1977 and the late eighties. The Gini coefficient which was around 0.31 in 1972–3, as well as in 1977–8, fell to around 0.29 in 1983, and to 0.25 in 1989–90. The improvement in the overall distribution of expenditure is confirmed by Fig. 5, which shows Lorenz curves for the distribution of rural household expenditure using NSS data from 1977–8 and 1989–90. The Lorenz curve for 1989–90 ‘dominates’ (i.e. lies above) the 1977–8 curve, indicating unambiguous distributional improvement over this period.

The comparison with other states that experienced agricultural growth from around 1983 onwards is also enlightening. Between 1983 and 1989–90, the Gini coefficient for consumer expenditure in rural areas registered a decline for West Bengal and Bihar (Table 5). It is noteworthy however, that the decline (i.e. the improvement in the distribution of expenditure) was most pronounced in West Bengal. In Orissa, the Gini coefficient actually increased. In Bihar, the reduction was marginal, and, unlike West Bengal, there was no clear dominance of the 1989–90 distribution over the 1983 distribution. (p.173)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 5. Distribution of Expenditure among Rural Households, 1977–8 and 1988–9

Source. National Sample Survey, 1992.

TABLE 5. Distribution of Rural Per-capita Household Expenditure in West Bengal, Bihar and Orissa

West Bengal

Bihar

Orissa

Gini coefficients

1983–4

28.9

25.6

27.0

1987–8

27.8

26.5

28.5

1989–90

24.9

24.6

27.1

Stochastic dominance

1989–90 over 1983–4

Yes

No

Source. NSSO (various).

(p.174) The trend in inequality reduction observed in West Bengal between 1977 and 1983 appears to have accelerated after 1983» during the period of high growth in agricultural output.

3.3 Reform and Redistribution

Improvements in distribution imply that the consumption of the poor has risen proportionately more than that of the non-poor. This could happen if there were a transfer of income or assets to the poor (as happened with agrarian reforms), or if the economic return on their assets (i.e. wage rates) increased disproportionately, or a combination of both these processes.

Table 6 reports the modal daily wage for male labourers77

TABLE 6. Daily Wage Rates for Male Agricultural Labourers (Kgs of Rice)

1956

1983

1987

1990

Birbhum

2.60

2.00

3.60

4.05

Kuchli

2.57

3.42

4.29

Sahajapur

2.64

2.61

3.30

4.00

Medinipnr

3.00

2.10

3.40

5.43

Bhagabanbasan

3.00

3.60

Purulia

2.00

3.20

3.43

Simtuni

2.40

2.57

Koch Behar

3.00

1.90

2.30

3.54

Kalmandasguri

2.53

2.66

Jalpaiguri

4.20

2.50

3.40

3.95

Magurmari

2.85

2.80

West Bengal

2.95

2.56

3.62

5.07

Sources. Kynch (1990) for district and state average rice wages for 1956, 1983, and 1987. Agricultural Wages in India (1991) for 1990 nominal average wages for districts, converted into rice terms by applying a uniform rice price of 5.10 rupees per kg. AERC (1956) Sahajapur village survey for rice wage in Sahajapur in 1956. WIDER village surveys 1987–9 and re-surveys in 1990 for all other villages and years.

(p.175) prevailing in the WIDER villages in rice equivalent terms.78 For the purpose of comparison we also report the average wages in the respective districts at corresponding points in time. The district average wages were calculated by Kynch (1990) using official wage series for reporting centres, and deflating them by the prevalent rice prices. It is interesting to note that the district-level wage rates are quite close to the corresponding village-level wages in most cases. Both show significant increases in recent years. State-wise secondary data on agricultural wages confirms these trends (Fig. 6) and show that the recent increases in wage rates have come after a long period of stagnation.79

Wage rates can rise as a result of an increase in the demand for labour with increasing cropping intensity and the adoption of technological innovations that require more intensive and timely use of labour inputs. These factors have been operating in West Bengal since the mid-eighties, in conjunction with the accelerated growth in output. In fact, trends in wage rates have closely tracked trends in output over the entire period for which consistent data are available (see Fig. 6). Another factor that might have led to higher wages is the increased bargaining power of agricultural labourers due to unionization and collective bargaining. The threat of strike action by agricultural labourers in their bargaining for higher wages, and indeed actual instances of wage strikes, are a distinguishing feature of labour relations in rural West Bengal. The fact that village panchayats play a prominent role in intermediating labour disputes underlines the political aspect of wage-setting in the state.80 (p.176)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 6. Trends in Wage Rates and Output of Foodgrains in West Bengal.

Sources. Acharya (1988), Bipul Chattopadhyay (IEG, Delhi, 1994), and CMIE, July 1993.

The rise in wage rates in West Bengal was, in fact, within the overall context of rising wages in India as a whole, as well as in the neighbouring states of Bihar and Orissa. Fig. 7 shows the trend in real wages for West Bengal, Bihar, Orissa, and India as a whole.81 The plots show that wage rates moved in similar directions for these states through most of the seventies and the eighties.82 Although wages grew fastest in West Bengal in the eighties, so did agricultural (p.177)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 7. Comparative Trends in Agricultural Labourers' Wage Rates.

Source. Acharya (1988), Bipul Chattopadhyay (IEG, Delhi, 1994)

output.83 There is no evidence of an additional upward pressure on wages due to increased bargaining power.

In fact, the rates of growth of output and wages in West Bengal were quite similar during this period, in contrast with Bihar, Orissa, and the all-India averages, where wage rises outstripped the growth in output. Between 1980 and 1991, the trend annual growth rate of real wages in West Bengal was 5.7 per cent while output grew at 5.6 per cent per year. The corresponding rates were 4.8 and 2.7 in India as a whole, 4.8 and 2.8 in Bihar, and 5.5 and 3.3 in Orissa. Labour (p.178) organization, wage strikes, and panchayat mediation in wage-setting, therefore, do not appear to have resulted in a substantial wage premium in West Bengal. Nor are recent trends in wage rates an adequate explanation for the fact that the distribution of income improved in West Bengal, but not in the neighbouring states.

There are other aspects of the terms of employment of agricultural labourers that may have differed significantly between West Bengal and other states, and contributed to an improvement in the distribution of income.84 One such factor is the wage differential between men and women. Our time-series data apply to male labourers and it is widely recognized that wage rates for female labourers are often lower than male wage rates. Data from the WIDER villages suggest, however, that at least in these villages no such gender gap applies (Table 7). In all villages, with the exception of Kalmandasguri in north Bengal, the average daily wages of female labourers were equal to (or even marginally higher than) those of male labourers.85

TABLE 7. Wages and Employment of Agricultural Labourers by Village and Gender

Average daily wage a (rupees)

Mean number of days employed in a year

Male

Female

Male

Female

Kuchli (Birbhum)

13.47

13.43

221

61

Sahajapur (Birbhum)

12.22

12.98

189

76

Bhagabanbasan (Medinipur)

14.58

14.90

158

102

Simtuni (Purulia)

14.22

14.32

104

95

Kalmandasguri (Koch Behar)

10.97

8.31

187

142

Note. (a) Average of (total wage earnings in the year of survey/total days employed) over all agricultural labourers.

The name of the district is given in parenthesis.

Source. WIDER village surveys 1987–9.

(p.179) Women perform fewer days of agricultural labour a year than men, on average, and most of their work is concentrated in periods when seasonal demand for labour is high. The absence of female disadvantage in terms of average wage, therefore, may reflect the greater concentration of women's labour time in periods when the wages are relatively high. The similar levels of average wages for men and women, in other words, are consistent with some gender discrimination in task-specific (or period-specific) wages. But the fact that such discrimination, if it did exist, was not strong enough to result in a difference in average wages remains significant.86 Even if collective wage bargaining did not have a major impact on the general level of agricultural wages, it may have succeeded in narrowing—perhaps even eliminating—the gender gap in wage rates.

In the earlier discussion of land redistribution and tenancy reforms above (sections 2.2 and 2.3), we observed that these reforms involved a relatively small amount of land, and did not dramatically reduce dependence of the landless poor on wage labour.87 The question remains as to how far they have contributed to the reduction of rural poverty in the eighties, and to the improvement of income distribution. We showed in Table 3 that in the WIDER survey villages, the contribution of income from assigned land to the total income of beneficiary households ranged from 9 to 29 per cent. The impact of land redistribution on incomes, without being dramatic, is far from insignificant, considering that small increases in income from a low base can represent considerable improvements in well-being and economic security. Further, the income effects of land redistribution may be quite substantial in comparison with those of other recent economic changes, such as wage increases. Table 8 presents some tentative calculations on this, including estimates of the respective impact of wage increases and land redistribution on the earnings of landless beneficiaries of land redistribution in the WIDER villages. (p.180)

TABLE 8. Landless Households; Income from Wage Labour and Assigned Land

Kuchli

Sahajapur

Bhagabanbasan

Kalmandasguri

Households as a proportion of all village households

Below IRDP poverty line

0.40

0.52

0.17

0.73

Landless prior to land reform

0.46

0.61

0.30

0.43

Landless beneficiaries of land redistribution

0.30

0.24

0.26

0.23

Earnings of landless beneficiaries as proportion of IRDP poverty line a

1. Total income

0.87

0.70

0.71

0.85

2. Incremental wage earningsb

0.15

0.13

0.09

0.03

3. Earnings from assigned land in 1987–8

0.25

0.07

0.14

0.24

4. Row (2) + Row (3)c

0.40

0.20

0.23

0.27

5. Row 4/Row ld

46

29

32

32

Notes. (a) All earnings (except last row) are converted into fractions of the IRDP poverty line of 6400 rupees per year for households of 5 persons in 1987–8.

(b) This is the difference in real wage rates between 1987–8 and 1983, multiplied by days of employment in 1987–8.

(c) Row 4 is the total ‘incremental’ earnings due to wage rate increase and assigned land.

(d) Row 5 is the ‘incremental’ earnings as a proportion of total income.

Source. WIDER village surveys, 1987–9.

These calculations are based on a number of simplifying assumptions. In particular, the estimated ‘incremental earnings’ associated with wage earnings are calculated as the difference between the wages actually earned by agricultural labourers in 1987–8 and what they would have earned from the same number of employment days had wages remained at the 1983 level Similarly, the incremental income from assigned land is based on the assumption that income (p.181) earned in 1987–8 from assigned land was purely additional—e.g. there was no opportunity cost in terms of foregone wage labour. These ‘incremental earnings’ (figures presented in Table 8) are, thus, best interpreted as estimates of the ‘first-round’ effects of land redistribution and wage increases.

As might be expected, these estimates vary a great deal from village to village, but they do suggest that the income effects of land redistribution are considerably larger than those of wage increases. The only exception concerns Sahajapur, where the average area distributed was particularly small, and where cropping intensity is low. In Kuchli and Bhagabanbasan, incremental income from assigned land was almost twice as high as incremental income from wage increases, even though these villages experienced relatively rapid wage increases. These survey results lend some support to the notion that the effects of land redistribution on the overall distribution of income in rural West Bengal compare favourably with the effects of economic changes such as wage increases.

In sum, the picture that emerges from the preceding analysis of available evidence on agricultural output, wages, consumer expenditure, and related variables is fairly consistent and intuitive. The decline in the head-count ratio of poverty in rural West Bengal has occurred alongside improvements in the distribution of expenditure from the late seventies onwards. Rapid agricultural growth from around 1983 onwards was a clear break from the earlier pattern of prolonged stagnation. This break from trend, however, was not unique as it occurred also in the neighbouring states of Bihar and Orissa at around the same time. Nevertheless, West Bengal's agricultural performance from 1983 onwards has been particularly impressive, and does warrant some explanation over and beyond the explanations for accelerated growth in eastern India in general. Agrarian and political reforms of the Left Front form an important part of this explanation.

Where West Bengal's experience appears to have differed rather more significantly from those of Bihar and Orissa is in the distribution of income, and the sharing of growth. Improvement in the distribution of expenditure (or consumption) predated the rise in agricultural productivity in West Bengal, and accelerated during the period of high growth. Redistribution of private property rights in land (and possibly m land use) was one of the main mechanisms in the redistribution of incomes. This is a significant point, as it (p.182) highlights the redistributive potential of land reforms even where relatively small areas of land are involved. The rise in the collective bargaining power of agricultural labourers, as evidenced by their unionization, did not have a major impact on setting wage rates over and above what might have been expected due to growth in output. The crucial result of political mobilization, and the heightened collective bargaining power of the landless and land-poor, is the effective transfer of property rights. In contrast with the situation in West Bengal in the past, and with the current situation in most other states of India, the involvement of political organizations and elected local government made it possible to implement and defend these legal transfers.

Despite agricultural growth, and rising wages, there is little evidence of any reduction in inequality in the neighbouring states. Agrarian reforms in West Bengal, both through their directly redistributive functions, as well as through their contribution to the institution of effective local governance, have resulted in improvements in the material conditions of the poorest. Although class-based mobilization on the question of land did not culminate in radically altering class distinctions, the reforms did deliver tangible gains for some of the most deprived sections of society.

4. Change and Stagnation

The discussion so far has focused on economic indicators relating primarily to private consumption, such as household income, consumer expenditure, head-count ratios, inequality indices, agricultural wages, and direct measures of consumption of food and other commodities. Private income, however, is not an adequate basis for the assessment of well-being, given that the relationship between the former and the latter can depend a great deal on personal characteristics (e.g. whether the individuals being considered are literate or illiterate), on the social environment, on the scope and quality of public services, and related factors. International comparisons, and also comparisons of different states within India, indicate that well-being achievements do vary widely even at a similar level of private income.88

(p.183) West Bengal itself is a significant example of a state in India that has had relatively better health and educational conditions in rural areas, in spite of having some of the lowest levels of private income. In the early eighties, for instance, seven of the thirteen major states with lower rural head-count ratios than West Bengal had higher rates of infant mortality in rural areas. West Bengal's rural IMR for 1981–3 was 95 per 1,000 live births, compared to 114 and 136, respectively, for Bihar and Orissa, which had similar head-count ratios. The rural IMR was lower in West Bengal than Assam (102), Gujarat (121), Haryana (102), Madhya Pradesh (144), Rajasthan (113), Tamil Nadu (100), and Uttar Pradesh (159), all states with lower rural head-count ratios than West Bengal.89 The situation was much the same in the late eighties (see Fig. 8) when of the 15 major Indian states, West Bengal had a lower rural head-count ratio than two (Bihar and Orissa), but had a rural IMR that was higher than only three other states (Kerala, Punjab, and Maharashtra). Similarly, the rural literacy rate in West Bengal in 1981 (40 per cent) was higher than the corresponding figure not only in Bihar and Orissa but also in Andhra Pradesh, Haryana, Karnataka, Madhya Pradesh, Maharashtra, Rajasthan, and Uttar Pradesh—seven major states that had lower rural HCRs than West Bengal.90

4.1 Political Change and Public Action

Many of the factors that lie between private income and well-being are connected with the supply, demand, and utilization of public goods and services. The improvement in anthropometric indicators, for instance, depends as much on the availability of clean water, access to health services, awareness about nutritional needs, and general conditions of public hygiene, as on food intake.91 Similarly, public health measures such as mass immunization are known to have contributed to substantial lowering of mortality rates, and progress in education is often dependent on the presence and effective functioning of a public schooling system. Public intervention, (p.184)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 8. Rural Poverty (Head-Count Ratio) and Infant Mortality, 1987–8: Indian States

Source. Sample Registration Bulletin (1992), and Tendulkar et al. (1993).

therefore, can be critical for the rapid improvement of well-being even if private consumption is expanding.

Public action may also act as a lever on other factors, such as the allocation of food, health care, or leisure time within the household, that affect the relationship between private consumption and well-being but do not necessarily involve the consumption of public (p.185) goods. If gender disparities in the allocation of private (as well as public) consumption, for instance, are partly determined by the nature of female employment opportunities and property rights, then the relevant domain of public intervention might extend well beyond the supply of public goods.

On both these counts (the expansion of public services, and the promotion of social change), the political changes of the type experienced in West Bengal can be expected to have a major impact on prevailing deprivations. If the old pattern of political power being concentrated in the hands of elite groups and bureaucrats, acted as a constraint on the expansion of public services, then the changed balance of power in rural West Bengal certainly opened up new possibilities of more development-oriented governance.92 The very fact that PRIs in West Bengal do exercise political power, and have acquired authority (in contrast to much of India where they simply provide a forum for the airing of factional rivalries), implies that greater scope exists for effective supply of public goods and related interventions.

Radical political changes have often coincided with social bhanges of similar depth. Changes in political regime, associated with the redistribution of economic claims in favour of disadvantaged groups, have also resulted, in a number of countries (e.g. China), in rapid improvements in the health and education conditions of the population, and of deprived groups in particular. Some of these ingredients of change have been present, though arguably not to the same extent, in the events which led the Left Front to power in West Bengal. Even if international comparisons are not entirely relevant, due to the relatively modest scope of reforms in West Bengal, examples from closer to home also suggest a link between political change and development-oriented public action. In Kerala, for instance, political formations similar to the Left Front have been credited with playing an influential historical role in the expansion of education and health care.93 Political change, and particularly change based upon the mobilization of oppressed groups for their economic claims, can be expected to raise the consciousness amongst these groups of their social rights in other spheres. It has been suggested, (p.186) for example, that political awareness in Kerala has been an important factor in raising people's awareness of health issues and their ability to demand health services.94

For all these reasons, political change in West Bengal represented an opportunity for social change and decisive improvements in well-being. The eradication of inequalities based on class, caste, and gender in access to health care and education represented as urgent a task as the alleviation of inequalities in access to land. Indeed, the public commitment of the Left Front parties to the eradication of gender discrimination and caste-based social inequities is on record. Highly visible policies such as the reservation of scats in Panchayati Raj bodies for women and for Scheduled Caste and Scheduled Tribe candidates have been adopted. The key question, however, is whether or not this stated political commitment resulted in actual improvements. We examine some evidence below.

4.2 Progress and Persisting Backwardness

Infant Mortality

Sample Registration System surveys show that the infant mortality rate (IMR) in rural West Bengal declined from 95 to 75 per thousand live births between 1981 and 1990 (Sample Registration System, 1992). In 1990, as in 1981, West Bengal's IMR was about 19 points below the Indian average. The decline has been sustained at a similar rate since around 1970 (this is the earliest period for which a consistent SRS series is available). Of the 15 major states for which SRS data are reported, rural West Bengal's rank rose from the sixth-lowest IMR in 1981 to the fourth-lowest one in 1990. This, however, was mainly due to the poor performance of Karnataka and Andhra Pradesh, where the IMR actually stagnated over part of this period.

West Bengal's own rate of decline during the eighties was not much faster than the Indian average, and was in fact surpassed or (p.187)

                   Agrarian Politics and Rural Development in West Bengal                                               *

FIG. 9. Decline in Rural Poverty (Head-Count Ratios) and Infant Mortality Rates (IMR) between 1983–4 and 1987–8

Source. Sample Registration Bulletin (1992) and Tendulkar et al. (1993).

equalled by Bihar, Uttar Pradesh, Gujarat, Punjab, Kerala, and Tamil Nadu. Fig. 9 shows the performance of Indian states in reducing the rural IMR between 1983–4 and 1987–8 (vertical axis) and their respective changes in rural poverty (horizontal axis). West Bengal's rate of reduction of IMR over this short period was surpassed by seven states (Uttar Pradesh, Punjab, Maharashtra, Kerala, Gujarat, (p.188) Tamil Nadu, and Rajasthan), four of which (Punjab, Maharashtra, Gujarat, and Rajasthan) experienced either stagnation or increase in the rural head-count ratio during this period.

Child Nutrition

The WIDER village surveys included an anthropometric survey of all children below the ages of five years in the six survey villages. The measure chosen was weight-for-age, which is widely used in rural health programmes in India.95 Besides the classification of children between the well-nourished and the undernourished, four categories of undernourishment were used. They range from ‘slight’ and ‘moderate’ to ‘severe’ and ‘disastrous’ levels of undernourishment. If a child is found to be in the latter two categories, health centre guidelines call for clinical intervention including hospitalization if necessary.

For two of the WIDER villages (Kuchli and Sahajapur) earlier nutritional surveys—conducted in 1983—are available (Amartya Sen and Sunil Sengupta, 1983). For these villages, there were clear signs of improvement in the nutritional status of children in the intervening five years (Table 9). Particularly, the incidence of ‘disastrous’ undernourishment, which affected 8 per cent of children in both villages in 1983, was almost non-existent in 1988. The incidence of ‘severe’ undernourishment also recorded a major decline.96

These observations are based on the experience of only two villages at two points in time, and, as such, need to be interpreted with caution, but the decline of undernourishment is dramatic enough to represent a significant indication of very real improvement. ‘Disastrous’ undernutrition of a child is more likely to result from poor health and infection than from actual shortage of food. Improvements in the supply of drinking water, better awareness of health conditions, and easier access to medical services have probably been important factors in the near eradication of extreme forms of child undernutrition from these villages.97 (p.189)

TABLE 9. Percentage of Children Undernourished, by Level of Undernourishment (Children Aged 5 or Below)

Level of undernourishment

Index a

Slight

Moderate

Severe

Clinical

1983

Kuchli

20

30

26

8

48

Sahajapur

21

31

33

8

54

1988–9

Kuchli

25

37

13

0

36

Sahajapur

32

35

14

1

37

Bhagabanbasan

46

26

13

0

34

Simtuni

28

36

25

3

47

Magurmari

42

26

11

7

39

Kalmandasguri

41

37

4

4

36

Note. (a) Refer to the main text for an explanation of the index.

Sources. Sen and Sengupta (1983), and WIDER village surveys 1987–9.

The overall level of deprivation, however, has remained extremely high. In none of the villages, was the estimated proportion of undernourished children below three-quarters in 1988–9. In Simtuni, less than one in ten of the children under the age of five were adequately nourished at the time of the survey. In Magurmari and Kalmandasguri, several children were found to be ‘disastrously’ undernourished.

An index of undernourishment was calculated using the method of Sen and Sengupta (1983). This index gives a summary measure of the extent of undernourishment in any group. The higher the index, the worse the aggregate status of the group. If a child is disastrously undernourished, it is assigned a score of 4, for severe undernourishment it gets a score 3, for moderate and slight undernourishment scores of 2 and 1 respectively, and a well-nourished child gets a score 0. These scores are then summed up over the (p.190) relevant group and divided by four times the total number of children in the group (i.e. the index is 1 if all children are disastrously undernourished and 0 if all children in the group are well-nourished). Broadly, the gender and caste patterns are according to expectation—i.e. girls are more undernourished than boys, and general caste Hindu children are better nourished than children of Scheduled Caste and Scheduled Tribe families (Table 10)98

TABLE 10. Index of Child Undernutrition by Caste and GenderI a

Boys

Girls

General Caste Hindu

30

39

Scheduled Caste

33

39

Scheduled Tribe

37

50

Muslim

34

40

Note. (a) Refer to main text for an explanation of the index.

Source. WIDER village surveys 1987–9.

Literacy

Literacy rates increased between the 1981 and 1991 censuses in rural West Bengal from 48.6 per cent to 57.7 per cent for the 54+ population overall, and from 36.1 per cent to 47.2 for females (Table 11). These rates of increase were not spectacular by the standards of other Indian states. West Bengal's rank among the major Indian states remained unchanged for overall as well as female literacy rates between the two census years.

According to the 1991 census, West Bengal literacy rate for the 7+ male population was 68 per cent compared to the Indian average of 64 per cent, and the corresponding literacy rate for females was 47 per cent compared to the Indian average of 39 per cent. More recent changes in literacy patterns would appear in the literacy rates of younger age groups. According to NSS estimates for 1987–8, the (p.191)

TABLE 11. Literacy Rates in Major Indian States, by Rank

Overall 1981

Overall 1991

Kerala

81.6

Kerala

90.6

Maharashtra

55.8

Tamil Nadu

63.7

Tamil Nadu

54.4

Himachal

63.5

Gujarat

52.2

Maharashtra

63.0

Himachal

51.2

Gujarat

60.9

West Bengal

48.6

West Bengal

57.7

All India

43.6

52.1

Female 1981

Female 1991

Kerala

75.7

Kerala

86.9

Maharashtra

41.0

Himachal

52.5

Tamil Nadu

40.4

Tamil Nadu

52.3

Punjab

39.6

Maharashtra

50.5

Gujarat

38.5

Punjab

49.7

Himachal

37.7

Gujarat

48.5

West Bengal

36.1

West Bengal

47.2

All India

29.8

39.4

Sources. Population Censuses 1981 and 1991.

literacy rate for 10–14 year old males in rural areas in West Bengal (69 per cent) was lower than the Indian average (73 per cent), and was lower than all states with the exception of Bihar (60 per cent), Andhra Pradesh (66 per cent), Madhya Pradesh and Uttar Pradesh (68 per cent each). West Bengal's female rural literacy rate for the same age group (61 per cent), however, was significantly higher than the Indian average (52 per cent).99 While West Bengal's narrower gender gap in literacy rates compared to the Indian average (both according to census and NSS data) might be viewed in a positive light, it ought to be a matter of some concern that in the younger age groups this narrow gap appears to be a consequence more of an (p.192) extraordinarily low literacy rate for males rather than an extraordinarily high one for females.

Average literacy rates conceal the true nature of the problem of education in India. Analysis of literacy rates by caste and gender (Table 12) illustrates this very powerfully. The aggregate literacy rate for the WIDER villages, for example, is not very different from that for rural West Bengal as a whole. The group-specific literacy rates, however, range from, near total illiteracy for Scheduled Tribe women, to near universal literacy for general caste Hindu men. A similar picture applies with other caste groupings. In terms of public policy, then, there is not one ‘literacy problem’ but several. For some sections of the rural population, in fact, there is no literacy problem at all.

TABLE 12. Literacy Rates (Percentage of Population Aged 7 and Above, Literate) by Gender and Caste

Female

Male

Both

Rural West Bengala

38

62

50

WIDER sample

40

60

51

Caste Hindus

71

92

82

Scheduled Castes

22

43

33

Scheduled Tribes

5

24

15

Muslims

15

30

22

Note. (a) West Bengal literacy rates are from Census 1991.

Sources. Census 1991, WIDER village surveys 1987–9.

Further insights on this issue can be obtained from a consideration of changes in literacy rates over time. These can be gauged from the current literacy rates of different age cohorts. Assuming that people become literate if at all by the age of 15, and that the chances of acquiring literacy thereafter are low, literacy rates of different age cohorts can tell us something about patterns of change in literacy over time.100 Table 13 gives literacy rates by gender and caste for people who reached the age of 15 in different periods. (p.193)

TABLE 13. Literacy Rates (Percentage of Population Literate) by Caste and Age Cohort

Caste Hindus

Scheduled Castes

Scheduled Tribes

Aged 15

Males

Females

Males

Females

Males

Females

1983–7

95

88

58

30

41

0

1978–82

95

82

38

26

30

2

1973–7

95

76

40

19

19

5

1963–72

96

77

44

11

12

0

1953–62

91

53

31

7

30

0

Source. WIDER village surveys, 1987–9.

It is striking that, at the two extremes of the literacy scale, there is no discernible trend over time: for general caste Hindu males the 1953–62 cohort was already very close to full literacy, while for Scheduled Tribe females there has been no significant departure from total illiteracy during the entire reference period. The groups in between accounted for changes in the aggregate literacy rates between the fifties and the late eighties. For general caste Hindu women, there were steady improvements throughout this period, while for Scheduled Caste and Scheduled Tribe males the changes were more recent.101 The cohorts which displayed the most significant increases in literacy were the youngest ones, roughly corresponding to the period since the Left Front government has been in power.

4.3 Limited Initiatives

West Bengal's progress in improving health and educational conditions during the period of Left Front rule has not been extraordinary. The infant mortality rate has been declining, but its rate of change has been in line with past trends, and also in line with declining infant mortality rates in most other Indian states. Literacy rates have also improved but, here too, in relative terms the improvement has (p.194) not been outstanding. If anything, among the younger age groups, West Bengal's relative position might have suffered a relative decline. Traditional patterns of deprivation along caste and gender lines persist, though the recent period does appear to mark acceleration in achievement for Scheduled Caste and Scheduled Tribe males. Whether this corresponds to a one-off or sustained improvement remains to be seen. Micro-level evidence suggests that some of the most extreme forms of undernutrition have become rare, but nutritional deprivation remains widespread and acute.

The fact that, by and large, there is no break in trends of improvement in well-being, that rates of change are not very different from country-wide averages, and that traditional patterns of deprivation persist and reproduce themselves, is in itself remarkable. More could have been expected from a state that has had a distinguished record in the implementation of agrarian reforms, led the way in the establishment of local-level democracy, broken a long spell of stagnation in agricultural production, and, most importantly, experienced significant reductions in rural poverty and inequality.

It is striking, indeed, that energetic activism on the agrarian-reforms agenda went alongside a near total absence of initiative in public policy on other factors that influence well-being. The reform of local governance failed to extend in a comprehensive manner to the public health and education systems. The successful examples of Kerala as well as other states in implementing innovative programmes such as noon meals in schools and immunization drives were conspicuous by their absence in West Bengal. The emancipation of women, which is a critical factor in the expansion of public services and overall social development, made its appearance on the public agenda only with reference to the reservation of panchayat seats.102

Literacy Campaign

One significant exception to the general pattern of neglect of public services was the state's implementation of the national Total Literacy Campaign, which started in 1991. The campaign was given high prominence, and the panchayats played an active role in it. The aim (p.195) of the campaign was to mobilize large numbers of people on a non-party political basis to teach people basic literacy skills in a concentrated effort. Large numbers did indeed take part, though most of them were in fact supporters of the Left Front parties.

The precise impact of the campaign is difficult to assess at this early stage. Available results are sketchy and varied. Official claims of having completely eradicated illiteracy from some districts do seem exaggerated. The two WIDER study villages in Birbhum district (which was at one stage declared a fully literate district—the claim was later withdrawn), Kuchli and Sahajapur, were surveyed for literacy after the completion of the literacy campaign. Results indicate a significant positive impact. Nearly 60 per cent of those who were previously illiterate had achieved at least ‘semi-literacy’. This has been an important achievement, even if it is much more modest than official claims of full literacy. Another survey (Sinha, Majumder, Mondal, and Chattopadhyay, 1993), carried out in some other Birbhum villages about a year and a half after the campaign, recorded a much smaller impact. Only 11 per cent of the participants surveyed had in fact learned to read and write, as against official claims of 61 per cent.

It is difficult to form a judgement about the campaign on the basis of these two surveys. A number of basic points, however, can be made. A campaign approach, while necessary for the implementation of land reforms, and probably quite useful to raise the issue of literacy all over the state, is likely to be of limited consequence if not backed up by other support measures. Neo-literates, particularly adult neo-literates, are prone to losing their newly acquired skill if it falls into disuse. A literacy campaign, therefore, cannot be the end of the process of ‘skilling’ the population. It can only be the beginning. Without the continued acquisition and retention of literacy skills by children and young people through regular channels of schooling, a campaign risks ending up as a flash in the pan.103 Even if the literacy campaign has achieved its objectives, and the evidence on this is mixed, to say the least, it is no substitute for a comprehensive and effective system of public schooling that is able to actively address the needs of deprived groups on a sustained basis.

(p.196) 5. Beyond the Agrarian Agenda

The Left Front government has been largely successful in implementing the programme of agrarian reform legislated by governments over the decades in response to movements of small peasants and the landless poor. These reforms involved important redistribution of economic claims from the well-off to the poor. In addition, the government has been able to consolidate these gains by transforming the Panchayati Raj Institutions into effective agents of local government.

The process of agrarian reform in West Bengal, although it covered relatively small areas of land, did affect a substantial proportion of the population. Taken together, the 1.44 million beneficiary households of Operation Barga, and the 2.15 million beneficiary households of ceiling laws, constitute around half of all rural households in West Bengal. Even if we assume that some beneficiaries of these programmes overlap, a substantial section of the population, mainly the poorest people with few assets, and those from socially deprived groups, have been direct beneficiaries. While the land struggles of the sixties marked a watershed in terms of the change in class alignments, the Left Front reforms of a decade later gave that changed balance of power concrete form by instituting redistribution of claims to land affecting millions of people. This, we believe, is not an insignificant achievement. The economic benefits to some of the most deprived sections of the rural population have been tangible. West Bengal has experienced rapid declines in rural poverty, and substantial improvements in the distribution of consumption.

Two seemingly contradictory approaches to the recent agrarian and political history of West Bengal have emerged. On the one hand, Kohli (1987) focuses attention on state action alone as the engine of social and political change. While he does acknowledge the other direction of interaction—that is, the influence of civil society on state action—his analysis gives undue pre-eminence to the ideological discipline of the CPM as an explanation of pro-poor interventions. On the other hand, attempts to restore the importance of class relations and class struggle, particularly the movement of the late sixties and early seventies in the understanding of social and political change (for instance, Bose, 1993, Acharya, 1993, and Rudra, 1981), have tended to dismiss the relatively placid and bureaucratic approach of the Left Front as having (p.197) at best a marginal, and at worst a negative, impact on the situation of the poor.

There is a sense in which these two approaches describe two different aspects of the West Bengal reality. The success of the Left Front in implementing a broadly democratic system of governance has involved more than the agency of a disciplined ‘social-democratic’ party. Political mobilization and heightened political awareness on the part of the poor had contributed to the ‘presence’ of the party in the first place, and continued to play a role in imposing electoral discipline on its cadres. The reforms of the Left Front, on their part, have extended legal and institutional recognition (partly through bureaucratic channels) to the socio-political changes which were initiated in radical agrarian struggles. By so doing, the reforms have given content (be it modest) to the changed balance of power in the shape of defensible property and tenancy rights, as well as the opportunity for the exercise of civil and political rights in the shape of democratic local government.

The driving force behind the Left Front reforms has been political mobilization and the need for its consolidation. These have been the main factors in placing the agrarian question at the centre of things. The history of politicization played a major role in West Bengal, as it has done elsewhere, in setting the framework for social and political change. We have already noted some of the limitations to redistribution within this framework. The ultimate objective of the land-redistribution programme, for instance, cannot be complete equalization of land ownership, as any further lowering of the land ceiling will endanger the political constituency of the Left Front. Furthermore, the domain over which redistribution has taken place is the village, and the final impact on the livelihood of poor landless households would depend on the initial conditions (such as the distribution of land and agro-economy) in their villages.

A more serious limitation of the agrarian agenda in the West Bengal experience has been the relative continuity in some of the traditional patterns of deprivation. While patriarchy, for instance, was probably weakened as a result of collective bargaining for wages, it may have been strengthened by the creation and consolidation of private property (as well as tenancy) rights in land, mainly in favour of male household heads.

Furthermore, contrary to expectation, the political changes have not led to breaks from trend in aspects of well-being such as health (p.198) and educational conditions, for which public policy and public action are known to play a critical role. West Bengal's rankings among the major Indian states in rural infant mortality and literacy rates have not undergone substantial changes. Decreases in its mortality and illiteracy rates in recent years are in line with their historical trends. Not much of the energy and activism that attended the implementation of the agrarian-reforms agenda was transferred to these areas of public policy.

The agrarian-reforms framework in general, and in West Bengal particularly, calls for a renegotiation of class-based economic claims. Within the legal context of land-reforms legislation in India, and the political context of ‘peasant unity’ in rural West Bengal, however, this renegotiation is limited to the marginal redistribution of private property rights in land. Even under these limitations, the agrarian-reforms agenda does provide for forms of intervention that address urgent needs of some of the most deprived groups in society. The overriding concern with the agrarian-reforms agenda, however, has also been, at least partly, responsible for the almost complete neglect of claims which might arise from other sources of social and economic inequity. The failure of public policy to act on the serious gaps between the educational and health achievements of male and female children is one such case. Indeed, the near exclusive concern of the Left Front organizations with agrarian politics, and the establishment and redistribution of property rights in land, is an important factor in their neglect of individual rights to a minimal level of education and health care.

The limits to further redistribution of land, extension of tenancy rights, and arbitration of wage demands are set both by the politico-legal framework of Indian agrarian reforms, and the electoral calculus of the ruling coalition. These limits had probably been reached as early as the mid-eighties. The continued relevance of the agrarian agenda since then owes something to the growth in agricultural output and the scope created for the wider sharing of this output. In this regard, the agrarian agenda has not outlived its usefulness. It is also clear, however, that the economic, political, and social sustainability of the Left Front reforms, as well as of the gains that these reforms imply for the poor, require ventures beyond the agrarian agenda. In matters of economic policy, the West Bengal government has shown that it is quite capable of looking beyond the agrarian agenda. Once accused of having neglected the industrial (p.199) sector, the Left Front leaders are winning plaudits for their active wooing of Indian and foreign investment in the state.104 Thus far, however, there is little evidence of innovative thinking or action on urgent matters of social policy.

The activism generated by political mobilization on agrarian issues has remained confined, by and large, to the realm of relations and forces of production in agriculture. The successful implementation of agrarian reforms and the institution of local-level democracy in the state provides some of the most favourable conditions anywhere in India for overcoming traditional patterns of deprivation along the lines of class, caste, and gender. Guarantees of local political office to members of deprived groups can help in this process, but it is in the exercise of political power for the expansion of public services, and the universalization of primary health care and education that the foundations of egalitarian and participatory economic and social development can be laid. Whether or not the Left Front parties are willing and able to take the lead in this regard remains yet to be seen.

References

Bibliography references:

Acharya, Poromesh (1993), ‘Panchayats and Left Politics in West Bengal’, Economic and Political Weekly, 28 (22), 29 May.

Acharya, S. (1988), ‘Agricultural Wages in India: A Disaggregated Analysis’, Indian Journal of Agricultural Economics, 44 (2), April-June.

Agricultural Census of India 1971.

Agricultural Wages of India 1992–93.

Bandyopadhyay, D. (1986), ‘Land Reforms in India: An Analysis’, Economic and Political Weekly, 21 (25–6), Review of Agriculture, 21–8 June.

Banerjee, Abhijit V. and Maitrecsh Ghatak (1995), ‘Empowerment and Tenancy: The Economics of Tenancy Reform’, mimeo, Department of Economics, Harvard University, November.

Basu, S.K. and S.K. Bhattacharya (1963), Land Reforms in West Bengal; A Study on Implementation (New Delhi: Oxford Book Company).

(p.200) Bergmann, Theodore (1984), Agrarian Reform in India, with Special Reference to Kerala, Karnataka, Andhra Pradesh and West Bengal (New Delhi: Agricole).

Bhattacharya, N., M. Chattopadhyay, and A, Rudra (1987a), ‘Changes in Level of Living in Rural West Bengal—Private Consumption’, Economic and Political Weekly, 11 July.

———(1987b), ‘Changes in Level of Living in Rural West Bengal—Social Consumption’, Economic and Political Weekly, 15 August.

———(1987c), ‘Changes in Level of Living in Rural West Bengal—Housing Conditions’, Economic and Political Weekly, 5–12 September.

———(1987d), ‘Changes in Level of Living in Rural West Bengal: Consumer Durables, Clothing and Footwear’, Economic and Political Weekly, 31 October.

———(1987e), ‘Changes in Level of Living in Rural West Bengal: Perceptions of the People’, Economic and Political Weekly, 28 November.

———(1988), ‘Changes in Level of Living in Rural West Bengal—Variations Across Socio-Economic Groups’, Economic and Political Weekly, 28 May.

Bhattacharyya, Dwaipayan (1993), Agrarian Reforms and Politics of the Left in West Bengal, Ph.D. thesis (unpublished), University of Cambridge.

———(1995), ‘Manufacturing Consent: CPI (M)'s Politics of Rural Reforms in West Bengal (1977–90)’, paper presented at the Workshop on Agricultural Growth and Agrarian Structure in West Bengal and Bangladesh, Calcutta, 9–12 January 1995.

Bhaumik, Sankar Kumar (1993), Tenancy Relations and Agrarian Development: A Study of West Bengal (New Delhi: Sage).

Bose, Sugata (1986), Agrarian Bengal: Economy, Social Structure and Politics, 1919–1947 (Cambridge: Cambridge University Press).

———(1993), Peasant Labour and Colonial Capital: Rural Bengal Since 1770, The New Cambridge History of India III: 2 (Cambridge: Cambridge University Press).

Boyce, J. (1987), Agrarian Impasse in Bengal: Institutional Constraints to Technological Change (Oxford: Oxford University Press).

Census of India, 1981.

Census of India, 1991.

Centre for Monitoring the Indian Economy—CMIE (1993), Performance of Agriculture in Major States (1967–68 to 1991–92), July 1993.

(p.201) Chandrasekhar, C.P. (1993), ‘Agrarian Change and Occupational Diversification: Non-agricultural Employment and Rural Development in West Bengal’, Journal of Peasant Studies, 20 (2), January.

Chattopadhyay, Sujit Narayan (1992), ‘Historical Context of Political Change in Rural West Bengal: A Study of Seven Villages in Bardhman’, Economic and Political Weekly, 27 (13), 28 March.

Crook, R.C. and J. Manor (1994), ‘Enhancing Participation and Institutional Performance: Democratic Decentralisation in South Asia and West Africa’, report to the Overseas Development Administration.

Dasgupta, S. and R.S. Mukhopadhyay (1989), ‘Party Politics, Panchayat and Conflict in West Bengal’, Man in India, 69 (1).

Drèze, Jean (1990), ‘Poverty in India and the IRDP Delusion’, Economic and Political Weekly, 29 September.

Drèze, Jean and Jackie Loh, (1995), ‘Literacy in India and China’, Economic and Political Weekly, 30 (45), 11 November.

Drèze, Jean and Amartya Sen (1995), India: Economic Development and Social Opportunity (Oxford and New Delhi: Oxford University Press).

Echeverri-Gent, J. (1992), ‘Public Participation and Poverty Alleviation: The Experience of Reform Communists in India's West Bengal’, World Development, 20 (10), October.

Economic Times (1994), ‘“West Bengal: Back in Business”, An Economic Times Survey’, 12 July.

Gangopadhyay, Bodhirupa (1990), ‘Operation Barga’, mimeo, WIDER Project on Rural Poverty, Social Change, and Public Policy, Santiniketan.

Gazdar, Haris (1992), ‘Rural Poverty, Public Policy and Social Change: Some Findings from Surveys of Six Villages’, WIDER Working Paper 98, May.

Ghosh, Ratan (1981), ‘Agrarian Programme of Left Front Government’, Economic and Political Weekly, Review of Agriculture, 16 (25–6).

Government of India (1992), Ministry of Rural Development, Annual Report 1991–92.

Government of West Bengal (1993), Economic Review 1992–93.

Harriss, John (1993), ‘What is Happening in Rural West Bengal?: Agrarian Reform, Growth and Distribution’, Economic and Political Weekly, 12 June.

Islam, Sirajul (1988), Bengal Land Tenure: The Origin and Growth of Intermediate Interests in the Nineteenth Century (Calcutta: K.P. Bagchi).

(p.202) Jha, Praveen Kumar (1994), ‘Changing Conditions of Agricultural Labourers in Post-independent India: A Case Study from Bihar’, Ph.D. thesis, Centre for Economic Studies and Planning, Jawaharlal Nehru University, Delhi.

Jose, A.V. (1984), ‘Poverty and Income Distribution—the Case of West Bengal’, in A.R. Khan and Eddy Lee (eds.), Poverty in Rural Asia (Bangkok: International Labour Organisation, Asian Employment Programme, ARTEP).

Kohli, Atul (1987), The State and Poverty in India: The Politics of Reform (Cambridge: Cambridge University Press).

———(1990), ‘From Elite Activism to Democratic Consolidation: The Rise of Reform Communism in West Bengal’, in Francine Frankel and M.S.A. Rao (eds.), Dominance and State Power in Modern India. (Delhi: Oxford University Press).

Kynch, Jocelyn (1990), ‘Agricultural Wage Rates in West Bengal and Their Trends Over Time’, mimeo, Centre for Development Studies, University College, Swansea.

Lieten, O.K. (1992), Continuity and Change in Rural West Bengal (Delhi: Sage).

———(1994a), ‘What Has Really Happened in West Bengal: Land Reforms at the Centre Stage’, mimeo, Centre for Asian Studies, Amsterdam.

———(1994b), ‘For A New Debate on West Bengal’, Economic and Political Weekly, 29 (29), 16 July.

Maitra, T. (1988), ‘Rural Poverty in West Bengal’, in T.N. Srinivasan and P.K. Bardhan (eds.), Rural Poverty in South Asia (Oxford: Clarendon).

Majumder, Debanshu and Sunil Sengupta (1990), ‘Levels of Living Among the Rural Poor in West Bengal (1987–89)—A Study of Six Villages, Part 1’, WIDER Project on Rural Poverty, Social Change, and Public Policy, Santiniketan, Working Paper.

Mallick, R. (1992), ‘Agrarian Reform in West Bengal: The End of an Illusion’, World Development, 20 (5).

———(1993), Development Policy of a Communist Government: West Bengal Since 1977 (Cambridge: Cambridge University Press).

Minhas, B.S., L.R. Jain, and S.D. Tendulkar (1991), ‘Declining Incidence of Poverty in 1980s: Evidence versus Artefacts’, Economic and Political Weekly, 6–13 July.

Misra, Anshuman and Sunil Sengupta (1990), ‘Food Consumption in Selected Poor Households (1987–89)—A Study of Six Villages’, (p.203) Part 1, WIDER Project on Rural Poverty, Social Change, and Public Policy, Santiniketan, Working Paper.

Mukarji, Nirmal and Debabrata Bandyopadhyay (1993), ‘New Horizons for West Bengal's Panchayats’, A Report for the Government of West Bengal, New Delhi.

Mukherjee, Partha (1990), ‘Panchayats in West Bengal’, mimeo, WIDER Project on Rural Poverty, Social Change, and Public Policy, Santiniketan.

Mukherji, Neela (1992), ‘Villagers' Perception of Rural Poverty through the Mapping Technique of PRA’, mimeo, London School of Economics.

Nag, Moni (1989), ‘Political Awareness as a Factor in Accessibility of Health Services: A Case Study of Rural Kerala and West Bengal’, Economic and Political Weekly, 25 February.

National Sample Survey Organisation (NSSO), Sarvekshana, January-March 1989, 12 (3)

;

January-March (1991), 14 (3); and other issues.

Raghavalu, C.V. and E.A. Narayana (1991), ‘Reforms in Panchayati Raj: A Comparative Analysis of Andhra Pradesh, Karnataka and West Bengal’, Indian Journal of Public Administration, 37 (1).

Ray, S. Datta (1994), ‘Agricultural Growth in West Bengal’, Economic and Political Weekly, 16 July.

Raychaudhuri, Ajivata and Rongili Biswas (1994), ‘IRDP, Poverty Alleviation and Development: The Comparative Story of West Bengal, India’, mimeo, Department of Economics, Jadavpur University, Calcutta.

Rogaly, B. (1994), ‘Rural Labour Arrangements in West Bengal’, D.Phil. thesis (unpublished), University of Oxford.

Rudra, Ashok (1981), ‘One Step Forward, Two Steps Backward’, Economic and Political Weekly, Review of Agriculture, 16 (25–6).

Ruud, Arild Engelsen (1995), ‘Wealth, Power and Status Among CPM-Supporter Groups in Rural West Bengal’, paper presented at the Workshop on Agricultural Growth and Agrarian Structure in West Bengal and Bangladesh, Calcutta, 9–12 January 1995.

Saha, Anamitra and Madhura Swaminathan (1994), ‘Agricultural Growth in West Bengal in the 1980s’, Economic and Political Weekly, 26 March.

Sample Registration Survey (1992), Sample Registration Bulletin, June.

Sen, Abhijit and Ranja Sengupta (1995), ‘The Recent Growth in Agricultural Output in Eastern India, with Special Reference to the Case of West Bengal’, paper presented at the Workshop on (p.204) Agricultural Growth and Agrarian Structure in West Bengal and Bangladesh, Calcutta, 9–12 January 1995.

Sen, Amartya and Sunil Sengupta (1983), ‘Malnutrition of Rural Indian Children and the Sex Bias’, Economic and Political Weekly, 18, Annual Number.

Sengupta, Sunil (1981), ‘West Bengal Land Reforms and the Agrarian Scene’, Economic and Political Weekly, Review of Agriculture, 16 (25–6).

Sengupta, Sunil, et al. (1991), ‘Rural Economy and Poverty in West Bengal and the Public Policy Interventions’, Report of the WIDER project at Santiniketan, September 1991.

Singh, Ashok Kumar (1988), Peasants Revolt and Agrarian Reform (New Delhi: Commonwealth Publishers).

Sinha, A., D. Majumder, D. Mondal, and K. Chattopadhyay (1993), ‘Glimpses of Literacy Profile: An Evaluation Study of the Mass Literacy Programme in Birbhum’, Report, Agro-Economic Research Centre, Visva-Bharati, Santiniketan, West Bengal.

Swaminathan, Madhura (1990), ‘Village Level Implementation of IRDP: Comparison of West Bengal and Tamil Nadu’, Economic and Political Weekly, 31 March.

Webster, Neil (1992), Panchayati Raj and The Decentralisation of Development Planning in West Bengal: A Case Study (Calcutta: K.P. Bagchi and Company).

Westergaard, Kirsten (1986), ‘People's Participation, Local Government and Rural Development: The Case of West Bengal, India’, Centre for Development Research, CDR Research Report No. 8, Copenhagen, March.

Williams, Glyn (1995), ‘Panchayati Raj and the Changing Micro-Politics of West Bengal’, paper presented at the Workshop on Agricultural Growth and Agrarian Structure in West Bengal and Bangladesh, Calcutta, 9–12 January 1995.

Notes:

(*) We gratefully acknowledge discussion with Nripen Bandyopadhya, Boudhayan Chattopadhyay, Jean Drèze, Madan Gopal Ghosh, Barbara Harriss-White, Jackie Loh, Debanshu Majumder, Manila Murthi, V.K. Ramachandran, Ben Rogaly, and Amartya Sen. We owe a special debt to the staff of the research project ‘Rural Poverty, Social Change and Public Policy’ based at Visva-Bharati, Santmiketan, and the residents of villages they surveyed. Sri Benoy Krishna Chowdhury was a constant source of encouragement and help throughout the period of the project and after. This chapter was finalized while one of us (Gazdar) was visiting the Centre for Development Economics. We are grateful for the Centre's support.

(1) The main source of such data in India is the National Sample Survey Organization (NSSO). Reported results are from Minhas, Jam, and Tendulkar (1991). For studies on earlier trends in rural head-count ratios and the consumption of cereals etc., see Jose (1984) and Maitra (1988).

(2) See Bose (1991), pp. 49 and 58.

(3) Government of West Bengal (1993).

(4) Furthermore, there was little diversification of employment away from agriculture in the rural areas. See, for instance, Chandrasekhar (1993) and Gazdar (1992).

(5) The political formations which emerged in that period included the largely Maoist C.P.I. (M-L), or Communist Party of India (Marxist-Leninist), also known as ‘Naxalites’, who remained outside the United Front, and were, indeed, opposed to participation in government. The Naxalites initiated land seizures from big landlords in north Bengal soon after the formation of the first United Front government. This was followed within four months by a major call to their supporters by the left-wing constituents of the United Front itself for the unearthing of concealed land. The latter came to be known as the ‘land grab’ movement.

(6) This opportunity has been noted by a number of observers. Kohli (1987), for instance, notes that ‘West Bengal, ruled by a party that is communist in name and organization but “social-democratic” in ideology and practice, highlights the redistributive possibilities within India's contemporary democracy and capitalism’ (p. 9).

(7) Some of the early developments under the Left Front were reviewed, for instance, by Ghosh (1981), Rudra (1981), and Sengupta (1981) among others in a special section of the Economic and Political Weekly in 1981 (see vol. 16, nos. 5–6). Since then, there has been wide scholarly interest in public policy in the state. Besides dealing with specific aspects of economic and political development quite a number of these studies—Acharya (1993), Bhattacharyya (1993, 1994), Kohli (1987, 1990), Lieten (1992), Mallick (1992, 1993), Webster (1992), Westergaard (1986), and Williams (1995), for instance—have also attempted to formulate overall assessments of the Left Front government.

(8) More detailed information on these and other activities of the project can be found in Majumder and Sengupta (1990), Misra and Sengupta (1990), Sengupta (1991), Gangopadhyay (1990), Mukherjee (1990), and Gazdar (1992).

(9) Government of India (1992).

(10) It is worth noting that agrarian reforms in India are not institutional reforms at all, in the sense that they have been in China, where each successive reform has brought in its wake new forms of individual and collective rights over land.

(11) Under this system, which was in operation in parts of British-ruled India (having been initiated in Bengal in 1793), intermediary rights in the collection of revenue over agricultural land were made permanent and transferable. The class of intermediaries thus created was known as zamindars, who had fixed revenue obligations to the state coffers for specified areas of land. They were entitled, of course, to raise much greater amounts if they wished.

(12) Land consolidation also happens to be one area of agrarian reform where West Bengal's record compares unfavourably with other states. This is arguably a serious omission given the high degree of fragmentation of landholdings in West Bengal and the possibility of large efficiency pins as a result of consolidation.

(13) See Bergmann (1984), for instance, for a comparative study of agrarian reforms in Indian states.

(14) See Bose (1993) for a detailed analysis of the status of various claims over land during this period.

(15) It is worth noting that over time, the system of intermediary interests had evolved into a multi-layered hierarchy, with the zamindars at the top, but with many levels of ‘sub-infeudation’ in between (see Islam, 1988). The system was, for all intents and purposes, moribund, and post-independence legislation declared its demise.

(16) Census of India, 1981.

(17) Strictly speaking, this reasoning assumes that labour-force participation rate is similar among landless and landed households.

(18) On the probable number of bargadar tenants, see Lietcn (1992), p. 160.

(19) Although jotdar literally means owner of land, the term is commonly used to refer to the class of landlords who lease out land.

(20) A similar situation prevailed in parts of the 24-Pargauas district in south Bengal. Large landlords in this area were known as lotdars, and their holdings consisted mainly of allotments of recently reclaimed land.

(21) Kohli (1990) notes the relative absence of state-level caste-based mobilization of propertied classes in West Bengal in contrast with a number of other Indian regions. This might have been another factor in the relative political weakness of the large landlords.

(22) We cite figures for 1971 from Agricultural Census of India 1971, as these predate most of the land redistribution, and present the pre-reforms picture. Agricultural census data, however, are only of limited use, since they report operational holdings rather than ownership holdings. Even so, it is worth noting that West Bengal had a less unequal distribution of operational holdings than India as a whole.

(23) The famous Tebhaga (‘two-thirds share’) movement in the nineteen-fortics, for instance, was largely concerned with crop shares and tenant security.

(24) For a more detailed statement of this thesis, see Sengupta (1981).

(25) See, for instance, Sengupta (1981), Hose (1993), and Acharya (1993) among the many accounts of West Bengal's agrarian history that place the turning point of rural politics at the late sixties. An event which marked this change was the split in the state Congress party and the formation of the Bangla Congress, reflecting the political division within the propertied classes.

(26) There is considerable ambiguity in the use of terms such as ‘middle’ peasant or ‘rich’ peasant, and large landlord. They have tended, however, to dominate the political discourse, and retain some social relevance. Operationally, these categories are often thought of as representing different rates of application of family and non-family labour on owned land, with peasants, including ‘rich’ ones using family labour alongside some hired help.

(27) See Basu and Bhattacharya (1963) for a comprehensive account of the early implementation of agrarian-reforms legislation.

(28) On this point, see Bandyopadhyay (1986).

(29) On the redistribution of land under the United Front, see Ghosh (1981).

(30) Mallick (1992).

(31) Another serious misinterpretation of data contained in Mallick's (1992) study concerns the author's comparison of West Bengal's land-redistribution record with that of other states in India. In West Bengal, as in other states where the zamindari system had prevailed, most of the land that was declared surplus was under the land-ceiling clauses of the zamindari abolition legislation. Specific land-ceiling laws were enacted by states where the zamindari system (and therefore zamindari abolition laws) did not exist. Mallick (1993), however, arbitrarily compares the areas of land redistributed under various post-zamindari abolition ceilings laws in Indian states, in order to substantiate his dubious claim that West Bengal's record in redistribution was no better than that of most other states. All the data on land areas redistributed that we have presented above and in Table 2 refers to the aggregate area redistributed under all land-ceiling legislation, as reported in Government of India (1992), and other official publications.

(32) See Ghosh (1981).

(33) Communists led the peasants movement in Bengal in the mid-thirties for abolition of fixed-produce rent. This demand was met in 1938 during Provincial Autonomy rule in Bengal when the fixed-rent system was outlawed. Later in the forties the Tebhaga movement demanded a two-thirds share for the tenant, and also that threshing of the crop and its division should be carried out in public with the tenant's right to supervise.

(34) See, for example, Lieten (1992).

(35) One such loophole was the so-called ‘Resumption Clause’. This allowed landlords to evict tenants by claiming they were going to farm the land themselves.

(36) During the decisive moments of the Tebhaga struggle in the mid-forties, for instance, the threat of eviction was instrumental in weakening the resolve of the bargadars of western Bengal (Bose, 1993). There were also large-scale evictions of tenants in many areas of West Bengal in the late fifties in response to the stepping up of the sharecroppers' movements in the post-zamindari abolition period.

(37) See Lieten (1992) for a detailed discussion of contending statistics.

(38) Kisani is a form of sharecropping traditionally practised in Birbhum district and in parts of Bardhaman and Murshidabad districts adjoining Birbhum. Tenants received a one-third share of the output, in contrast with the more prevalent half-share, with the landlord providing all non-labour inputs.

(39) Sengupta (1991) and Gangopadhyay (1990).

(40) This practice of bargadars surrendering their tenancy rights in exchange for the ownership of a smaller area of land cannot be equated with the outright evictions of earlier years. The landowners in this particular area felt obliged to part with a small portion of their holding in order to guarantee unrestricted property rights over larger parts of it. This reflects the relatively better bargaining position enjoyed by bargadars.

(41) Westergaard (1986) reports, for instance, that the prevalent share in kisani areas was 40 per cent—much lower than the legally stipulated 75 per cent, but higher, than the traditional 33 per cent.

(42) See Bhaumik (1993).

(43) The rise of ‘reverse’ tenancy, that is, the leasing in of land by large landowners from smaller ones, usually to take advantage of scale economies in irrigation or other inputs, has been widely noted in West Bengal. See, for instance, Bhaumik (1993).

(44) For evidence of this decline, see the results of various NSS rounds reported in Bhaumik (1993).

(45) The law does allow for some flexibility. Tenant crop shares of 60 per cent (as opposed to 75 per cent) are set for leases where costs are shared. This still does not fully take into account the range of possible arrangements.

(46) It should be noted that Operation Barga, and agrarian reforms in general, strengthened the collective bargaining position of the landless and the land-poor. In this sense there is a ‘positive externality’ of tenancy recording that benefits unrecorded tenants also.

(47) Indeed the precise effects of tenancy regulation on efficiency need closer scrutiny of theoretical as well as empirical material than we have been able to conduct (see Banerjee and Ghatak, 1995 for one recent investigation).

(48) Notable among other states that have taken initiatives in the decentralization of power is Karnataka. Here, many policy functions were devolved to the PRIs in the early eighties. Failure to hold regular elections, however, was one of the factors that led to the reversal of these gains (see Crook and Manor, 1994). See also Raghavalu and Narayana (1991) for a comparison of West Bengal's record with PRIs with other states besides Karnataka, including Kerala and Andhra Pradesh.

(49) Bhattacharyya (1993) reports from his field study that the traditional relationship between the administrative bureaucracy and the PRIs has changed significantly to the advantage of the latter in West Bengal.

(50) See Drèze (1990) and Swaminathan (1990) for comparative assessments of the implementation of IRDP. The former is based on field studies in parts of rural West Bengal, Uttar Pradesh, and Gujarat; the latter on West Bengal and Tamil Nadu.

(51) See, for instance, Raychaudhuri and Biswas (1994).

(52) Taken together, the IRDP and the JRY constituted the bulk of central government resources committed to direct poverty-alleviation programmes.

(53) In the WIDER villages, for instance, there was significant mismatch between alternative ranking of the poorest twenty households according to the investigators' perceptions based on discussion with villagers, and those based on income measures (Gazdar, 1992). A similar exercise (with similar results) was conducted in some other West Bengal villages by Mukherji (1992).

(54) In fact, nearly all micro-level studies of panchayat elections find significant numbers of the landless, the land-poor, and people from Scheduled Castes and Scheduled Tribes among panchayat candidates and members. See, for instance, Lieten (1992, 1994b), Ruud (1995), Webster (1992), and Williams (1995).

(55) Echeverri-Gent (1992), p. 1414.

(56) See, for example, Bhattacharyya (1993), Ruud (1995), and Williams (1995).

(57) See, for example, the chapter on Uttar Pradesh in the present volume and the literature cited there.

(58) See, for instance, Chattopadhyay (1992) and Dasgupta and Mukhopadhyay (1989). For similar evidence on another state, see the chapter on Uttar Pradesh in the present volume, and the literature cited there.

(59) See, for example, Bhattacharyya (1993), Ruud (1995), and Williams (1995). AH these studies show that the CPM has been quite flexible in its approach to mobilization, and adept at using caste, religious, and factional loyalties for electoral advantage.

(60) This vigilance, ironically, is not always to the advantage of the Left Front parties, since their candidates with a record of corruption are also rejected by voters (see, for example, Bhattacharyya, 1995).

(61) These very aspects (regular party-based elections, reservation of seats for various groups) are the subject of recent central-government legislation on PRIs that the states have to follow. It is worth noting, however, that even in the case of West Bengal, these were not innovations, but that they reflected the recommendations of the Ashok Mehta Commission on PRIs in 1977, which were ignored in most other states.

(62) On the domination by large landlords of virtually all public institutions, including PRIs, prior to the Left Front reforms, see Dasgupta and Mukhopadhyay (1989) and Chattopadhyay (1992). Also see Bandyopadhyay (1986) on the resistance to change on the part of bureaucratic personnel.

(63) Webster (1992), p. 131.

(64) This alliance, incidentally, has not been an equal partnership. The ‘middle peasants’ have been the leaders, and the landless and the land-poor, for the most part, followers.

(65) Centre for Monitoring the Indian Economy (1993).

(66) The reliability and consistency of agricultural data from West Bengal have been called into question from time to time. See, for instance, Ray (1994) for a recent intervention. Different sources of data might yield different magnitudes of growth, but the treed break appears to be robust. In line with some of the more recent work in this area, we have made use of the series for West Bengal and other Indian states compiled by the Centre for Monitoring the Indian Economy (CMIE).

(67) Fig. 3 plots the trends in the output of foodgrains, with the mean for the sample period for each region set at 100. Three-year moving-averages are used for output.

(68) See, for example, Saha and Swanunathan (1994).

(69) This was also found by Abhijit Sen and Ranja Sengupta (1995). They find a greater acceleration in West Bengal compared to the two neighbouring states.

(70) Government of West Bengal (1993).

(71) There is much complementarity in the various inputs that make up the package of recent technological change. The eighties also witnessed a near trebling of fertilizer consumption in West Bengal (Government of West Bengal, 1993).

(72) Saha and Swaminathan (1994) and Lieten (1994) suggest agrarian reforms as the key explanation for growth, while Harriss (1993), though supportive of these reforms in general, has focused attention primarily on private incentives.

(73) This is broadly consistent with the finding that besides electricity, the consumption of other inputs did not accelerate in West Bengal in the eighties as it did in Bihar and Orissa (Sen and Sengupta, 1995). Part of the explanation for West Bengal's higher rate of growth in foodgrains output despite unchanged growth rates for most inputs must be in the more efficient use of inputs.

(74) One argument against redistributive reforms on efficiency grounds is that redistribution of land creates uncertainty about private property rights and discourages investment. In a situation of already heightened class tensions over the issue of land as in West Bengal, however, a once-and-for-all reform might well be a way of reducing uncertainty and insecurity.

(75) Details of the construction and updating of poverty lines in India are surveyed in EPW Research Foundation (1993).

(76) This series is, in our opinion, the most consistent one, as it takes into account changes in regional purchasing power parities.

(77) The modal wage refers to the usual wage rate for casual field labour work. In fact, many different forms of labour contracts, and corresponding wage rates, can and do exist even within a village. These other wage rates are, also, often indexed to the modal or chuto (casual or free) labour wage rate (see Rogaly, 1994).

(78) Since rice is the main staple, the choice of a rice index makes inter-temporal and inter-village/regional comparisons easier.

(79) The real wage series presented in Fig. 6 was calculated by Bipul Chattopadhyay (Institute of Economic Growth, Delhi), based on data published m Agricultural Wages of India, and using the same methodology as in Acharya (1988). Although the AWI (Agricultural Wages of India) data appear to have some methodological shortcomings, and to lead to some overestimation of wage levels, they seem to be reliable enough for the purpose of assessing broad trends in wage rates. For a good discussion of this issue, see Jha (1994).

(80) It has been noted by a number of recent micro-level studies that a strike threat has become a ritualized part of pay negotiations, m contrast with the early years of the Left Front government when strikes did raise the stakes in the bargaining process (see, for instance, Rogaly, 1994, and Williams, 1995). The fact remains, however, that wages are set collectively, with PRIs acting as arbitrators.

(81) In Figs. 6 and 7, figures on output and wages refer to three-year moving averages, expressed in terms of percentage of the period mean in the relevant region.

(82) In terms of level of real wages, West Bengal was far ahead of both Bihar and Orissa. The present discussion, however, focuses on trends rather than levels.

(83) Correspondingly, when output was falling, so were real wages, as in the first few years of Left Front rule, when, arguably labour militancy would have been the highest.

(84) We have not examined changes in employment in West Bengal and other states, for want of comparable data. This, undoubtedly, may be an important issue.

(85) Table 7 omits the village Magurmari where most of the labourers were involved in non-agricultural activities.

(86) Even in Simtuni, a tribal village where wage employment patterns are quite similar for men and women (e.g. there is little difference between them in terms of average number of days worked), the female average wage was as high as the male average wage (see Table 7).

(87) It is worth noting, nevertheless, that West Bengal is the only major state in India where the proportion of agricultural labourers in the male workforce declined between 1981 and 1991 (Census of India 1991), This confirms the trend noticed by Lieten (1992) and labelled by him as ‘depeasantization discontinued’; the author ascribes this trend to the positive effects of land reforms.

(88) These issues are discussed in greater detail in the companion volume (Drèze and Sen, 1995); see also chapters 1 and 2 in this volume.

(89) Sample Registration Bulletin, June 1992.

(90) Census of India, 1981; the figures are related to the 5+ age group.

(91) For that matter, food intake itself can also be related to public interventions in terms of the efficiency and reach of a public distribution system, or the presence of feeding programmes.

(92) Kohli (1987), for instance, characterizes the formation of the Left Front government in West Bengal as representing a change from the political economy of corruption to one of development.

(93) See, for instance, the chapter on Kerala in this volume.

(94) Nag (1989), for instance, drew attention to the contrast between Kerala and West Bengal in this regard. He observed that in Kerala health awareness grew alongside the development of political consciousness, whereas in comparison West Bengal had lagged behind on both counts. His observations, however, predate the Left Front era, when significant political changes did take place.

(95) The standards commonly used by Anganwadi health centres in West Bengal villages for the monitoring of child health are based on WHO or Harvard standards of healthy body weight at different ages.

(96) There was a corresponding rise in the proportion of less seriotsly undernourished children in both the villages (Table 9).

(97) There is some evidence, from other sources, of improvement in these factors. The re-survey of NSS observations by Bhattacharya et al. (1987b) showed that access to and consumption of public health had increased somewhat in the re-survey. The most significant improvement reported was in the access to potable water.

(98) The term ‘general-caste Hindus’ refers to Hindus other than those classified as Scheduled Castes or Scheduled Tribes. It is, therefore, a residual category used for the sake of convenience. This group does, nonetheless, comprise Hindu castes that claim ‘higher’ caste status, and in general, enjoy better economic conditions than people from the Scheduled Castes and Scheduled Tribes.

(99) National Sample Survey Organisation (NSSO), 1987–8 round.

(100) Strictly speaking, this ‘backward projection’ method also assumes that mortality rates are similar for literate and illiterate people within the relevant groups; see Drèze and Loh (1995) for further discussion.

(101) Similar patterns of change were found for primary schools enrolment in the Muhammad Bazar Block in Birbhum district over the mid-seventies onwards by Lieten (1992).

(102) While some Left Front initiatives such as collective wage bargaining are likely to have led to some improvement m the relative position of women, others, such as the redistribution of property and tenancy rights almost exclusively to male heads of poor households, may have had the converse effect.

(103) Given that a significant proportion of CPM's local-level leadership consists of village school teachers (Bhattacharyya, 1995), it is somewhat surprising that the regular schooling system has received so little attention.

(104) See, for instance, the special report on the new industrial strategy of the Left Front in Economic Times, 12 July 1994.