ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Face-to-Face Taxation in West Bengal

Set within the larger debate linking taxation with representation, the paper takes a close look at "face-to-face", i e, local government taxation in two districts of West Bengal and asks who pays and why, in situations when tax amounts are nominal. Drawing on statistical analysis, citizen survey and ethnographic research, the paper claims that while people's relative prosperity, not unexpectedly, has a positive effect on their tax contributions, even poorer social groups such as the scheduled castes and scheduled tribes may pay up, driven by a complex mix of motivations ranging from compulsion to contingent compliance. The paper concludes that as gram panchayats begin to rely more on local tax incomes, greater is the probability for a "fiscal contract" to emerge between local taxpayers and an accountable local government.

Face-to-Face Taxation in West Bengal

Compulsion, Compliance or Collusion?

Set within the larger debate linking taxation with representation, the paper takes a close look at “face-to-face”, i e, local government taxation in two districts of West Bengal and asks who pays and why, in situations when tax amounts are nominal. Drawing on statistical analysis, citizen survey and ethnographic research, the paper claims that while people’s relative prosperity, not unexpectedly, has a positive effect on their tax contributions, even poorer social groups such as the scheduled castes and scheduled tribes may pay up, driven by a complex mix of motivations ranging from compulsion to contingent compliance. The paper concludes that as gram panchayats begin to rely more on local tax incomes, greater is the probability for a “fiscal contract” to emerge between local taxpayers and an accountable local government.


Leeches suck blood, bees suck honey, and calves suck their mothers’ milk, mildly and not causing any damages; the state should collect taxes in such a non-coercive manner; also it will be easier to collect taxes if the earned revenue is properly used to satisfy taxpayers’ needs and demands, and if citizens are properly nurtured.

– The Mahabharata, as cited in Bhaduri (1998).

Just as the sun vapourises water from the earth only to return much more at the appropriate time, the king taxes his subjects to spend more on their well-being.

– Raghuvansam, Kalidas, as cited in Bhaduri (1998).

I Introduction

o these seemingly antiquated directives regarding the fiscal responsibility of the ruler vis-à-vis his subjects, and the idea of reciprocity and of a consensual “fiscal contract” between the king and his people implicit in them, have any relevance to the contemporary discourse on the politics of taxation in a third world democracy like India? Drawing upon an interesting and new line of research on the politics of taxation [Brautigam 2002; Moore 2002, 2004 and forthcoming], we aim to argue in this paper that in a developing country setting like India the fiscal issue of taxation and the political process that surrounds it are becoming a part of the democratising agenda itself. Put simply, the old debate of “taxation with(out) representation” – a hallmark of the history of representative institutions in the western world – is gaining prominence in the folklore of development and good governance too. A growing scholarly atten tion is being paid to explore the linkage between the fiscal process and the political process, that is to say, between the source of governmental revenue on the one hand and the democratic properties and potentials of public institutions on the other.

Are tax revenue-dependent states fiscally more responsible and more responsive to taxpayers’ needs and concerns than those (i e, the so-called “rentier” states in the development literature parlance) relying on “un-earned” and “un-bargained” income (such as foreign aid, or rents from natural resources like oil wealth or minerals)? How do citizens view their own tax obligations to the state and what political influence and channels of exercising governmental accountability are available to them? What is the prevalent culture of tax compliance among citizens? Does their propensity to pay taxes depend upon the level, quality and equity of services delivered by the state? Thus, some of the seemingly “technical” questions of the tax base, tax assessment, collection and compliance – issues conventionally treated as standard themes of public finance and public administration – emerge as deeply “political” issues, having a crucial bearing on such substantive democratic ideas as accountability, transparency and representativeness of pubic institutions and above all the political voice of citizens, and hence are subject matters of politics, democracy, or its “deficit”. These are the issues and questions we wish to examine in this paper, focusing on local body taxation in West Bengal.

To state the obvious, India is already a democracy, with elected local governments that cannot simply exploit taxpayers. But even in this democratic environment, taxation in poorer localities, in the absence of alternative revenue sources, may turn out to be quite burdensome for ordinary citizens, even if not totally exploitative. Alternatively, ‘gram panchayats’ (GPs) may be unable to raise enough taxes because they are too close to the people they are trying to tax and they are operating under the democratic compulsion for vote maximisation which may come in the way of revenue maximisation. These opposing forces and possibilities make it worthwhile for us to engage in a closer examination of rural taxation in India. To reiterate once more, we examine these questions not in terms of national or state level tax regimes in India, but with respect to taxation processes and relationships within the micro-jurisdiction of a GP – the lowest govern mental tier in the country. More concretely, we examine, in a situation where per capita tax amount is quite low, who pays taxes and why. And how do these local citizens see the local “tax” state – as a “punitive tax gatherer” or as an authority that responds adequately to their demands for “neighbourhood” public goods? Answers to these questions will likely to provide us some useful clues for anticipating the nature of taxation relationships between the rural public and the local government that will materialise once such local bodies are able to expand their tax revenue. More broadly, the significance of the study lies in the light it will hopefully shed, though somewhat indirectly, on the following question: in an agrarian environment of a developing country, under what circumstances could we anticipate a quasivoluntary, if not fully consensual, taxation regime to emerge such that citizens pay taxes to the state in exchange for “a voice in policymaking” and ultimately for accountable governmental institutions? Contrarily, in what circumstances is taxation likely to be coercive?1 For obvious reasons, unlike in the taxation narrative pertaining to western Europe, our focus here is not on the genesis of representative institutions but on their strength and vitality, more generally on the quality of governance and democratic practice in the contemporary south.

Even within the same national universe, however, there exist wide inter-state variations in India in terms of local-level tax performance, citizens’ propensity to pay taxes, and their confidence in local public institutions. Our focus here on the state of West Bengal is influenced by both its distinct colonial fiscal history and its contemporary political distinctiveness, namely, the near permanent incumbency of the left parties – the features on which we comment more later. Our arguments in the rest of the paper are developed as follows. In Section II, we try to embed the contemporary narrative of taxation in the developing south within the theoretical and historical debate on taxation in the deve loped north. Section III discusses the main results of statistical analyses of local government taxation in two districts of West Bengal. In Section IV we try to make sense of these findings in the light of insights about people’s views on taxation and the local “tax” state that we have gleaned from an ethnographic study of panchayat taxation in a village in one of the two selected districts. Section V offers some tentative conclusions.

II Consent or Coercion, or Even Collusion?

Theories of the link between taxation and representation draw mainly on the history of representative institutions in western Europe [Bates and Lien 1985; Levi 1999; Tilly 1992]. A brief glance at the history of European states indeed reveals that taxrelated negotiations between citizens and warring governments in need of war finances during the 16th to 18th centuries acted as a catalyst behind the emergence of representative public institutions in these countries. It is apt to cite here the pioneering work of Schumpeter (1991) that provides the conceptual building blocks of this genre of reasoning. This looks at the transition of the princely economy in pre-modern Europe, (i e, the demesne state relying on the revenue from royal estates) to the modern “tax” state which finances itself from taxes on its citizens, examines the resultant constraints on state institutions and policy-making and forcefully concludes that “public finances are one of the best starting points for an investigation of society … of its political life”. In short, his time-honoured work which remains remarkably relevant today amid greatly changed circumstances proposes to build up a “view of the state, of its nature, its forms, … as seen from the fiscal side”, that is to say, through the prism of fiscal sociology.

Inspired by this line of research, Moore, in several of his recent incisive writings (2002, 2004, forthcoming), draws upon a set of historical narratives about state formation in northern Europe – especially the Netherlands and Britain in the 16th to 18th centuries – to underline the contributions of “revenue bargaining” (a constructive social contract between the state and taxpayers) “to the emergence of representative, responsive and limited government”.

To be sure, in our analysis of local body taxation in India we cannot afford to draw simple and uncomplicated lessons from narratives of revenue bargains between princes and parliaments in early European history and transplant them in a straightforward way to a developing country setting. In particular, the nature of state-citizen interaction over revenue and the prospects for revenue bargaining are bound to get complicated by the reality of entrenched social inequalities in the depressed agrarian regions of India, and the disproportionate influence over state institutions often enjoyed by elite societal groups.

More elaborately, between the two contrasting possibilities of fiscal contractualism and coercive taxation, discussed in the taxation literature, emerges the third likelihood of collusive and co rrupt tax practices, resulting not in socially productive bargaining, but in socially sub-optimal deals between certain state actors and societal groups. The latter “receive exemptions from extra ction and other privileges simply on the basis of their clout rather than on the basis of some universal principles” [Levi 1999]. In our field research in villages, we have come across cases of local influentials, even elected representatives, evading taxes, and receiving tax concessions through unauthorised means such as under-valuation of properties. All these result in local deals and collusion, not simply coercion, and ultimately generate low tax revenues. Admittedly, such negotiations with local autho rities are often conducted piecemeal and not as a group and therefore are “individualised strategies” distinct from collective revenue bargaining, seeking tax favours through bribery or other corrupt means. Indeed the scholarly literature on taxation is cognizant of the fact that coercive taxation “ generally heavily inter-mixed with a high degree of corruption” [Moore forthcoming].

However, we are inclined to push a thought that there is something more to this kind of “devil’s deal”, to use Tendler’s (2002) pithy expression, than just piecemeal, non-institutionalised, ad hoc negotiations. Such socially suboptimal collusive arrangements, struck via both formal and informal channels of power, are widespread features of contemporary poor countries. It is in the context of such entrenched and well-established connections and transactions between formal state institutions and the socalled informal elite networks and their undeserving privileges (i e, arbitrary exemptions) that the possibility of coercive tax burden (i e, arbitrary extractions) on ordinary citizens, even in a democratic country, could be understood. Even though no overtly public mobilisation of powerful interests takes place, their activities are deeply political in their effects and significance, and hence are not simply personalised and private settlements. In a sense their influence over public policies is institutionalised though not legitimised.

The larger theoretical point that we wish to raise is that whether the tax system is perceived to be fair or not is an issue that is germane to the debate on consensual or coercive taxation and correspondingly to the demands of governmental accountability. If the government in a democracy like India taxes groups, that have secure property rights, considerable bargaining power and capacity to resist tax extraction, with a “light hand”, then in relative terms, the tax burden could be deemed weighty on “havelittles” – i e, on poorer peasants whose bargaining power is limited and political voice muted.

The current discourse on local body finances in the countries of the south takes place largely within the framework of fiscal devolution, with an emphasis on the question of the quantum of revenues generated locally to the relative neglect of such distributive concerns. For example, since rural local bodies in India until now raise only limited amount of own funds, and since the provincial governments, themselves in fiscal stress, want them to tap their respective resource bases more fully, the dominant discourse and the corresponding policy efforts are directed to augmentation of resources. There are real pressures on GPs to raise more revenues from their own citizens and through their own efforts [Marjit 1999; Ghosh and Ghosh 2004].

In this policy climate of “the more resources the better”, it is important to ask how in a democratic state like India the tax burden is distributed among citizens, across different social classes. How much can the local state extract and from whom? Alternatively, who pays and why? In other words, in tandem with the efficiency aspect of resource generation (i e, extractive efficiency), “norms of fairness concerning who pays, who is exempted and the relative equality of contributions” [Levi 1999] – the notion of “extractive equality” in short – are equally relevant focal issues in an analysis of taxation relationships and the development of democratic institutions.

To go back, albeit briefly, to the colonial fiscal history in India, taxation on poor peasants under the colonial rule seemed all the more burdensome in the light of relatively gentler fiscal burden on the intermediary, rentier classes. Indeed, the notion of taxation as coercion was not alien to the poor peasants under the colonial rule. Land tax was the main source of government revenue during the 18th and 19th centuries in the country. Indeed, tax on agricultural land, i e, “khajna”, constituted 75-80 per cent of the governmental revenue [Chatterjee 1984].

However, the history of extractive policies in the colonial era needs to be disaggregated even within India over time and space, since the colonial administrators did not follow one uniform revenue raising strategy across the entire country. On the contrary, several researchers notice and talk about a distinct north-south divide in this respect within the same national boundaries. We of course cannot cover this vast issue here. The point that is relevant for our discussion here is that it is essential to develop a spatially and temporally disaggregated account of the country’s colonial experience, taking into account both the divergent fiscal and administrative strategies that the colonial rulers pursued in different regions of the country and the variegated responses to those policies and administrative practices from the colonial subjects themselves, in the early and late colonial period. And for our immediate purpose, here we make some passing references to the distinctive features of Bengal’s fiscal past.

In east Bengal and Bihar (including Orissa), under the land tenure system introduced initially in the northern and eastern parts of the country, i e, the zamindari system, peasants had to pay khajna to landlords (i e, the intermediary classes), which was often coercive, with an increasing gap between rents extracted by them and revenue received by the colonial masters. This gap between the fixed (land) revenue payable to the government and the total rent that could be extracted from the ultimate tillers of the land led to the growth of a class of small rentiers – i e, the class of “petty landlords” – in Bengal’s agrarian economy [Chatterjee 1984]. This further increased the reliance of the colonial government on landlords and “tax farmers” who squeezed the peasantry.

Even now, when asked about GP taxes, many people in rural areas confuse it with ‘chowkidari’ (watch and ward) tax that used to be collected by the pre-independence local body, namely, the Union Board, ostensibly for the maintenance of local law and order. There was strong peasant movement and mass mobilisation in south-western Bengal – in Bardhaman, Medinipur, Hoogly – against the unfair and extractive imposition of such tax by the Union Board [Bose 1986]. This constituted a significant part of peasant understanding and cultural memory about taxation in the eastern part of the country, rendering villagers prone to perceiving taxes as coercive rents. We find the remnants of such resistance even today, as we ask people, during the course of our citizen survey, about the idea of raising revenue locally from the people. In contrast, in the southern parts of the country, under the ‘raiyotari’ system – the system of individual peasant proprietorship through which the individual cultivator was transformed into the owner of the land he tilled – peasants paid land taxes directly to the government, and this was part of the peasant culture and practice in these regions.

III Face-to-Face Taxation in an Agrarian Environment

Of the three tiers of the recently revitalised local governmental structure, direct taxing prerogatives are given only to GPs that are very small jurisdictions, with an average population of 2,892 for the country as a whole, but with wide inter-state variations in this respect.2 As the literature on the politics of taxation suggests, bargaining and negotiations between citizens and the agents of the state over taxation issues are more of a national rather than a local phenomenon – a part of the national rather than the local political process [Moore forthcoming]. One may, therefore, ask whether panchayat taxation constitutes the appropriate domain of analysis for the purpose of discerning under what propitious conditions a consensual taxation regime might prevail in a country like India. Also, in both absolute and per capita terms, the “revenue insignificance” of village-level tax income, especially in a state like West Bengal, is widely known. Indeed, taxation so far is not the primary source of GP revenues. Neither have GPs in most parts of India tried to augment their tax income in a big way. Instead, they receive most of their funding from the central government and their respective state governments, in the form of earmarked grants, block grants, equalisation grants and so on. Since the GP raises only a small proportion of its total budget locally through its own initiatives and efforts, bulk of its programmes cannot be carried out with its own funds. In other words, local level finance, with a limited resource base and a lean tax administration, is still in its infancy, as many would argue.

The house tax, i e, property tax on building and homestead land, is the main source of tax income at this level. Non-property tax revenue sources consist of various kinds of fees, rates, tolls and charges, income from public assets vested under the management of GP, revenue from sale of assets, bank interests, etc.

Indeed, the two most important taxes on private property in rural areas fall, under the present legislation, within the fiscal domain of the state government, although they come under the natural fiscal domain of panchayats [Rajaraman et al 1996]. These include land revenue on agricultural land and stamp duty on transfers of property. Under the present arrangement both are levied by the state government. This leaves only taxation on non- agricultural (mostly homestead) land and rural houses and buildings in the present fiscal domain of GPs. In short, buoyant revenue sources are kept outside the fiscal domain of GPs, resulting in their low taxable capacity and limited tax base – at times too low to generate sufficient revenues. Even so, there is actually wide variation between villages and panchayats in taxes raised, between and within states, across districts. Apart from spatial variations, temporal variations in tax collection are also expected within and between GPs. We are, therefore, tempted to argue that in years to come, with the growing fiscal activities at the GP level, we will begin to observe the effects of local taxation with a reasonable time lag.

Given the small size of panchayats, of course, taxation at this level is almost literally face-to-face taxation; that is to say, close and direct contact between tax officials and taxpayers is unavoidable. In the international context this is unusual and potentially problematic. Because it is a strongly recommended principle of contemporary tax reform that there should be a minimal direct personal contact between them to minimise the scope for extortion and corruption. As several scholarly works on China [Bernstein and Xiaobo 2000; Yep 2004] amply demonstrate, in the pursuit of resource generation local authorities in agrarian regions of the country often play a predatory role vis-à-vis ordinary peasants; extortionate local bureaucracies impose unreasonable ad hoc local taxes, fees and fines on peasant households. In recent times in China such coercive and unpredictable taxes have been bitterly resented by the peasants causing rural unrest, instability, peasant resistance movements, and even tax revolts, so much so that central leaders and agencies have been sending directives down the administrative hierarchy demanding that action be taken to lighten peasant burden. Indeed, the 2000-year old agricultural tax has been lifted very recently in the country.

In light of the experience, say in China, could one foresee that the panchayat-level revenue generation in small local arenas in India will result in similarly coercive taxation? The obvious contrasting element, of course, is democracy. Unlike the local admini strative structures in China, in India local bodies including GPs are democratically elected. Moreover, state level authorities play an important role in setting the tax rates, exemptions, assessment procedures, fee structures, etc, to minimise arbitrariness and unpredictableness in assessing tax liabilities of citizens.3 Also, there is supra-local party monitoring of these processes, especially in West Bengal, to counterbalance the influence of both local agents of the state and local elite groups. Thus, in a democratic set up with a clear mandate and guidelines from the provincial government how do taxation relationships, transactions, and bargaining go on between citizens and local politicians at the village level? Surely, the electoral and popularity calculus poses an obvious limit to the willingness of local representatives to enforce tax laws too stringently and thereby maximise revenues. On the other hand, however, in the case of GP-level taxation we are often dealing with poor agrarian environ ments, small local jurisdictions, face-to-face contact and physical proximity between tax collectors and taxpayers and also insufficient distinction between the officials who make assessments for tax liabilities and those who actually collect the money

  • the factors that are usually thought to be associated with extractive taxation. And corruption often goes hand in hand with extortion; under such situations tax collectors tend to appropriate a sizable proportion of tax revenues for themselves.4 Does the “legitimacy/democracy check” keep the tax burden on peasants within reasonable and affordable limits or does the coercive and extractive potential inherent in the tax relationship get the upper hand? This is very much an open question and worth exploring precisely for that reason. In short, since there is no inherent link between coercive taxation and local government [Moore forthcoming], it makes much sense to see how the outcomes turn out to be in Bengal’s rural landscape.
  • An official study on GP taxation in West Bengal candidly states that what is collected in the name of tax is pittance. Assessment is abnormally low, and whatever is assessed is not collected [Panchayat and Rural Development Department 2005]. Importantly, how ever, there is significant variation in per capita tax collection across GPs. But because tax payment involves only a small sum of money, ability to pay does not seem to be the only reason behind such variations; clearly there are other things at work. To examine more closely the other possible determinants of variations in tax collection/payment we focus on two districts
  • Bardhaman and Murshidabad. Bardhaman is the most prosperous agricultural region in the state, with some parts having a degree of industrialisation. Murshidabad, in contrast, is an economically depressed region, with some areas representing the subsistence Bengal. These two districts occupy different positions on the scale of socio-economic prosperity and expectedly offer wide variations in GP tax performance and hence provide a reasonably large number of parameters for a definitive comparative analysis.
  • Per capita property tax collection in the GPs of these two districts constitutes the dependent variable in our exercise.5 Per capita collection, of course, reflects a number of things: the revenue base, tax effort of local authorities and the level of tax compliance on the part of local citizens. It is, therefore, not easy to disentangle their individual effects. As a few focal explanatory variables we make use of a number of proxies – we are very much constrained by paucity of data here – which crudely measure the socio-economic features and the prosperity or otherwise of GPs. Levels of literacy are taken to be measures of both relative affluence of GPs and awareness of citizens vis-à-vis their own tax obligations. Scheduled castes and scheduled tribes (SC/ STs) are routinely deemed as socio-economically depressed population groups, and therefore a high concentration of these groups in the population may indicate a potential development lag. Hence, GPs with high proportions of SC/STs in the population may record low tax yields. But since tax amounts are small to begin with, the tax behaviour of SC/STs may indicate something other than just their ability to pay and more about their political equation with the local state called GP and the overall significance of the local government and the party to them.

    As a straightforward indicator of local people’s economic wherewithal, and by extension, of GP’s economic health, we include the proportion of “main workers” in the population – i e, those who find regular employment for six months or more in a year. It is perhaps obvious to suggest that with a greater number of people in the GP employed through most of the year, there is a greater likelihood of higher tax collection. In addition to these demographic, employment and literacy statistics, we include a variable that tries to capture the political attribute which defines the socalled exceptionalism of West Bengal, namely, the uninterrupted rule of the state by left parties (in particular, the Communist Party of India (Marxist) – the CPI(M) for nearly 30 years and its longstanding political domination [Bhattacharya 2004]. We use a crude measure of left (pre-)dominance in GPs, namely, the seat share of left parties in the total in each GP in the last panchayat election.

    The fact of left predominance in local panchayats is potentially important for our analysis in a number of ways. Since the left parties have a well-knit and organised party structure even at the village level and since many tax collectors are party appointees, the party’s organisational and disciplinary strength may work in favour of higher tax yield, at times even coercively collected, provided the party authorities themselves favour such outcomes. In contrast, if they prefer to tax villagers, especially some societal groups, with a light hand then the party’s electoral durability and the absence of much political competition may have the opposite effect.6 Also, it would be interesting to examine whether the SC/STs display a greater propensity to pay taxes because of their well known ideological proclivities to the left parties. Indeed, several psephology experts argue that one of the main reasons behind the left’s remarkable electoral success is their spectacular electoral fortunes, for years together, in constituencies reserved for SC/STs [Yadav 2006]. On the other hand, the relationship between the SC/STs and the left parties may involve an element of pressure and coercion on the part of the latter, making eligibility for party favours and for various welfare schemes contingent upon payment of taxes. Finally, we introduce a district dummy to capture the district effects, if any.7

    Table 1 furnishes the main results of regression analysis for the two districts. On an average, GPs in the district of Bardhaman collect per capita property tax of Rs 3.4 in the reference year; the corresponding figure for Murshidabad GPs is Rs 1.4.8 Admittedly, these are small amounts. But in respect of tax performance the selected GPs show wide variations. While the minimum tax collected in these two districts is less than a rupee, the maximum is Rs 18 in Bardhaman but only Rs 5 in Murshidabad in the reference years. When we look at GPs in Bardhaman as well as all GPs pooled together in the two districts, the relatively strong predictors of these variations, that spring out of our statistical analysis, include the proportion of SC/STs in the total GP population, the proportion of literates, the degree of left dominance and the proportion of main workers. All positively and significantly affect the tax outcome. Not unexpectedly, literacy, which is often taken as a measure of higher level of citizen awareness and greater sense of their own tax obligations and more generally of the prosperity of a region, produces relatively higher tax revenues. Similarly, GPs that have higher proportions of the population engaged in regular employment have understandably higher tax incomes. Simply put, relative GP prosperity contributes to higher tax takes.

    Also, GPs with stronger presence (i e, seat share) of the left parties in the panchayat board manage to perform better with respect to collection of tax dues. Politically secure and organisationally strong, the left parties, the CPI(M) in particular, are able and willing to levy a small sum that is not too taxing for ordinary villagers. As members of both opposition parties as well as of left parties other than the CPI(M) have mentioned to us in open conver sations, “It is the CPI(M) which has entrenched itself in local politics 100 per cent. They are continuously on a donationcollection drive; at times the amount is very nominal. It is actually a ploy to develop and maintain contact with the masses.” Anecdotal evidence also suggests that the CPI(M) regularly co lle cts higher amounts of donations from local elite groups, though for augmenting party revenues and not for building panchayat funds.

    In case of the district of Murshidabad, however, literacy and left prominence do not seem to be explaining much of the inter-GP variation in tax earnings. Literacy levels are much lower in this district as compared to those in Bardhaman (the average of 52 per cent as opposed to 65 per cent in Bardhaman villages). And unlike Bardhaman which is commonly considered to be the left bastion of power, Murshidabad is seen as almost the last stronghold – though at times wobbling – of the non-left opposition.9 It is perhaps not very surprising then that inter-GP variations in tax returns in Murshidabad are not decisively determined by either the literacy or the political variable. Their effects are dormant and not totally absent. Because when the data are pooled across the two districts, all four variables – social, economic, educational and political – remain important. And since the district dummy does not appear to be significant, what we capture through regressions are indeed inter-GP variations and not just the so-called “district effect”.

    The extent of long-term (if not full-time) employment among GP residents continues to be a significant predictor of tax earnings in Murshidabad too. But the result that intrigues us most is the relatively higher tax income, even in this comparatively poorer district, of those GPs that have a higher share of SC/STs in their population. Are these poorer sections of poorer localities suckers, paying under pressure and out of fear and other compulsions? Is the cost of tax evasion too high for them? Or is their tax behaviour driven by a desire to become legitimate citizens of the village republic and make a bona fide claim on the services and benefits provided by it? Or is it a mix of multiple, not necessarily

    Table 1: Regression Results for GPs in Bardhaman, GPs in Murshidabad and GPs Pooled across the Two Districts

    Dependent Variable: Per Capita Property Tax (PCTAX)

    Predictors: (Constant), Percentage of Literates in the Population (Perlit), Percentage of SC/STs in the Population (Perscst), Percentage of Main Workers in the Population (Permnwk), Percentage of Left Party Seats in the Panchayat (Perleft), Bardhaman dummy (Badum – only in case of pooled data)

    Bardhaman GPs Murshidabad GPs GPs Pooled across Two
    N = 235 N = 152 N = 387
    Adjusted R2 .159 Adjusted R2 .210 Adjusted R2 .299
    F value 12.04 F value 10.50 F value 33.96
    Model and Beta Coefficients
    (Constant) (Constant) (Constant)
    Perscst .176 Perscst .396 Perscst .252
    (2.731) (4.975) (4.089)
    Perlit .225 Perlit .107 Perlit .229
    (3.639) (1.256) (3.788)
    Perleft .212 Perleft .038 Perleft .206
    (3.222) (.508) (3.704)
    Permnwk .201 Permnwk .236 Permnwk .183
    (3.236) (2.981) (4.108)
    Badum -.042

    * t – Values are in parentheses.

    complementary, motivations? On the part of the local authorities and the ruling party in general, what is the nature of engagement with various societal groups and the mode of tax operations? Are there parallel processes of resource mobilisation through which the have – littles pay small sums to the GP and have – enoughs contribute to party funds?

    It is not possible to elicit adequate answers to all these complex questions from limited statistical exercises. Hence, to probe deeper into the subject under study we have conducted a citizen survey in two GPs in Murshidabad on the perceptions of villagers regarding GP taxation; their sense of their own tax obligations; their level of confidence in the operations of GP and expectations from the local government. This is further supple mented by an ethnographic study of tax collection processes, interaction of ordinary villagers with the tax collector and panchayat authorities and so on in one electoral constituency in one of the GPs in Murshidabad district. Our thoughts and findings gleaned from these explorations dwell on three particular issues: why poorer people, including SC/STs, pay GP taxes; what kinds of tax negotia tions and collusive arrangements go on within a microjurisdiction, and what larger discourses around the panchayat and the ruling party do animate the village republic. Taken together, these observations will hopefully help us understand the local “tax” state and its relationships with local citizens.

    IV People, the Panchayat and the Party

    “We, the poor, do not know much about taxation rules, we pay out of fear. The rich, who know the law very well, use it to evade taxes”, poignantly observes an elderly woman in one of our relatively deprived study areas. This feeling of being a sucker often gets resounded in people’s remarks. Table 2 presents the results of the citizen survey. Of the areas surveyed in Murshidabad, in the relatively backward GP a significant section (nearly 50 per cent) of the people say that they pay taxes because it is mandatory. In particular, arrears are squeezed out of them in a coercive manner. In the same area, people are especially unhappy with the quality of work taken up by the GP. The people who were most vociferous belonged mostly to poorer sections of the population. However, villagers do not have much idea about how tax money is used and therefore, though being generally dissatisfied, do not take concrete steps to give vent to their grievances. As the secretary of the District Tax Collectors’ Union remarked, “payment of panchayat taxes is still not a spontaneous affair. Spontaneity will come only when people know where the money is going and how it is being spent. Educated people understand their tax obligations and pay up. However, this is never a rule. Sometimes the poor pay up easily; it is the rich who harass us more.”

    Some amount of compulsion flows from the fact that payment of some taxes is linked to eligibility to various state services; for example the use of house tax payment receipt as proof of residence for further benefits of ration cards, birth certificate, etc. To ordinary villagers, therefore, tax is something they are bound to pay; in a sense it is a compulsory payment to the GP. Yet, there is an element here of contingent compliance too. Many taxpayers indicate that they pay taxes to secure their bona fide status. The various certificates not only provide some sort of economic assis tance to ordinary people but also act as a passport to their legitimate existence, identity and citizenship. (Such sentiments are strong especially among later settlers, mainly from Bangladesh.) As they say, “after all we have to live in the village and retain our ownership rights”. It appears that poor people, at least some of them, pay up partly because they want to wear a badge of citizenship and contribute to the public fund so that it in turn contributes to the public good. One of the fallouts of left politics in West Bengal since late 1970s is thought to be a sense of identification with party and panchayat authorities that the SC/STs and poorer people have developed over time, once the so-called “landlord” party (read the Indian National Congress Party) was removed from power. So it is a mixture of motivations, ranging from compulsion, strategic calculations, to contingent consent, which underpins common man’s tax behaviour.10

    Thus, the question as to why poorer people pay needs to be linked up with the larger question of how they view the local state, not only as a revenue-raising authority but more generally as an agency handling a lot of financial resources and making important expenditure decisions such as selection of beneficiaries of various welfare and anti-poverty schemes which may potentially make a significant difference to their lives. Most of these scheme-specific grants flow directly from the central govern ment to panchayats. But despite the stamp of “central ownership” on such poverty-alleviation programmes, it is not entirely clear that the “political pay off” from all these is enjoyed exclusively by the distant centre. On the contrary, to an ordinary villager it is the GP, the state closest to his life, which appears as the decisive authority in the distribution of “patronage funds”, and that too for some justified reasons. Contrary to the conventional wisdom, often a lot of discretion is exercised by GPs in determining spending priorities even when dealing with schemes funded by supra-local authorities. In principle transfers from the central and state governments (scheme-specific as

    Table 2: How Citizens See the ‘Local State’ and Their Own Tax Obligations: Citizen Survey in Two Selected GPs in Murshidabad

    Gram Panchayat Depressed Developed

    Pay taxes 84.4 87.7 Why pay taxes Mandatory 50.6 29.4 Voluntary payment 11.9 10.0 Contingent compliance 3.1 25.6 Others pay 8.8 8.1 Don’t know 9.4 15.6 Tax collection Go to GP office and pay 12.5 6.3 Collector comes to collect 63.8 55.6 Both 6.3 27.5 Don’t know 17.5 10.6 Tax utilisation For panchayat expenses 17.5 10.6 Local area development 26.9 43.1 Government resource crunch 6.3 10.0 Don’t know 49.4 36.3 Satisfaction and protest Dissatisfied with GP work 61.9 46.9 Quality of services not good 53.1 46.9 Dissatisfied but did nothing 53.1 41.9

    Notes 1: Figures are in percentages; figures do not always add up to 100, as there are multiple responses.

    2: We must recognise the inherent limitations of perception-based statistics. Expressed views sometime overstate the reality; there appears to be considerable over-reporting of tax payment by respondents, when compared with GP records on tax collection.

    Source: Survey data.

    well as untied funds) are under the formal control of supra-local governments. Yet there remains considerable scope for fungibility and manoeu vr ability on the part of GPs in actual expenditure decisions. Through “on-line” modifications of many prespecified guidelines and norms, GPs somewhat untie some of the tied funds in their day to day operations and thereby enjoy considerable functional autonomy in practice. It is therefore not unusual that villa gers notice the growing import of the GP and its political bosses.

    Similarly, the discretionary powers of panchayats and the influence of local party actors in the entire process of assessment and collection of taxes, especially in biasing assessment and valuation of properties, do not go unnoticed by the village folks. These quotidian revenue raising practices, the “rules of thumb” for assessing tax liabilities and the grassroots negotiations regarding these matters suggest considerable room for manoeuv rings by local level panchayat and party leaders, even when they are required to operate within the frame of general guidelines set from above. In an informal conversation, an elected member of a GP frankly admitted that undervaluation of property is a routine affair in GP taxation exercises, as these property owners are often their own friends and relatives or political comrades. It is hard to hide the fact, he continued, that panchayat members are partial in these matters to their own party supporters and sympathisers. What is more, sometimes there is hardly veiled symbiosis between local political cadre and private business, resulting in shady deals and collusive tax arrangements. The chances of political favouritism get more real since local tax collectors are often supporters or active members of the ruling party (or any of the many affiliated organisations). In the area of our ethnographic study, local people, pointing to the tax collector’s huge house on prime land and his lifestyle, often commented on the sly that he is a less than honest man.

    Finally and more generally, how people consider the tax question is influenced by how they view the whole system of party panchayats in West Bengal. Local people unmistakably see the larger presence of “the party” (read left parties in case of most GPs) in almost every aspect of panchayat activities, including its taxation operations. People seem to have little doubt that party leaders are in the driving seat as far as panchayat affairs are concerned, and that elected members as well as panchayat level administrators act as their proteges. Large resources are routed through GPs; yet there is a strong partisan control over the distribution of GP funds. Understandably, the local state assumes importance in pedestrian politics via the party. Yet, in our numerous conversations with local residents in the study areas, we have time and again noticed people’s ambivalent attitude towards the “local state”, namely, the GP and the party that controls it. Many look for and even solicit favours from GP members and yet routinely cast aspersions on them regarding their allegedly corrupt tax practices. Some seek party affiliation for economic and other strategic benefits, others consider themselves as unfortunate political minorities excluded from party favours. In short, people clearly understand the usefulness of “party-contacts”; yet they are often critical of party and panchayat leaders. In our taxation story, here leaps up the issue of whether the tax system is perceived to be fair. A budding sense of citizenship duty to pay taxes that we notice through our study and analysis appears to be juxtaposed against a feeling that GPs extend undeserving tax favours to local political and societal elites and to those close to the ruling party.

    V Conclusions

    The limitations of this study may be pointed out at this stage. We have not been quite able to say definitively whether GP taxation in West Bengal could be straightforwardly labelled as coercive or consensual, let alone what propitious conditions lead to the latter tax outcome. Like many other social science claims, ours is a tentative suggestion about the complex web of motivations that prompt people to fulfil their tax obligation, however nominal that may be. Apart from some of the obvious explanations for tax payment such as relative prosperity of people and areas, our analysis has shown that even poorer social groups such as SC/STs may pay up when payments are small and when there is an eagerness on their part to secure their legitimate status as citizens through tax compliance. But we at once hint at an additional element of compulsion influencing their tax behaviour. The domination of the left parties at the panchayat level has also appeared as a positive predictor of panchayat tax revenues. Partly this is a reflection of the organisational strength and poli tical stability of the left parties in the state and also perhaps their active engagement and grassroots contact with the masses, including SC/STs. But the flip side of this party predominance over panchayat affairs, including its taxation operations, is the clear possibility of arbitrariness, favouritsm, and even coercion in both tax assessment and collection. People know and see this very well. Thus, how people see the local tax state – as a fair and responsive revenue raiser or as a partisan and even predatory collector – depends upon which population group is doing the viewing and on which side of the ruling party it stands.

    If and when GP tax income expands – and there is genuine and visible urgency to augment panchayat revenues on part of the party leadership both at provincial and regional levels – this issue of excessive party influence on panchayat activities will appear all the more relevant to a discussion on a socially productive fiscal contract or its lack. Admittedly, party monitoring of GP activities could prove useful in ensuring GP accountability. But regular party interventions into its day to day activities and decisions can damage the constitutionally granted authority of elected members and in the end the legitimacy of the GP as a whole [Corbridge et al 2003]. Perhaps what could make a significant difference to the prospect of a consensual taxation relationship between local citizens and the GP is the latter’s “accoun table autonomy” – autonomy from excessive party influences and account ability to the people.

    These cautionary comments notwithstanding, we end our discu ssion on an optimistic note: in our field work we were struck by the fact that ordinary people have a good sense and appreciation of what is legitimate and legally permissible in the use of public funds. The corruption discourse, only faint murmurs of which we could capture through our group meetings, clearly demonstrates that local citizens have internalised some universal standards of legitimacy and rule-bound behaviour and that they are quite critical and disapproving of its violation [also see Harriss 2000]. Indeed, irregularities in panchayat taxation practices and in its various activities in general are often criticised by the local citizenry. It is quite likely that the edge of such public scrutiny will become sharper as GPs rely more on local revenues. A young, newly-elected, female GP vice-president nicely articulated this budding consciousness about GP’sgrowing dependence on public money and the consequent imperative to use it for public purposes thus: “I try to convince my constituency that since the GP does not own a money-minting tree, it has to rely on your tax contributions. At the same time, and not surprisingly, those who pay taxes expect us to deliver the goods in a timely fashion or else return their money. We certainly feel the pressure.” This is then the beginning of a story on a constructive fiscal contract between local taxpayers and a responsive local government.




    [This is part of a larger collaborative project with V K Natraj and Ganesh Prasad on Fiscal devolution in Karnataka, Tamil Nadu and West Bengal. We are most grateful to Mick Moore and V K Natraj for the intellectual inspiration and help they have generously given us in the preparation of this paper. For generous financial support to this project we are indebted to the DRC, IDS, Sussex. This paper has also benefited from the valuable comments, suggestions and criticisms of Partha Chatterjee, Pranab Das, Dilip Ghosh, Jyotsna Jalan, K Nagaraj and Ganesh Prasad. However, we regret that we have not been able to respond adequately to all of their objections and suggestions. We also acknowledge our debt of gratitude to several elected representatives, panchayat functionaries and government officials at the district and sub-district levels, NGO activists as well as the residents of the study areas who have generously given us their time, readily answered our questions and provided valuable information, data and assistance for field research. The remaining errors remain our sole responsibility.]

    1 Of course, all taxes are somewhat involuntary levies on citizens, still quasi-voluntary tax compliance is widely observed.

    2 The average population in GPs in West Bengal is 16,900. This is not a large figure by international standard, as is palpable when we take a quick look at the population size of the lowest government tier in, say, China (28217) and South Africa (149654) [World Bank 2004].

    3 Among other things, coercive taxation seems to be associated with the lack of any effective regulation of the taxation processes by higher levels of government. Interestingly, while GPs in India do not appear to be under-regulated, sometimes norms and guidelines set from above are modified by local “rules of thumb”, indicating relative autonomy enjoyed by local authorities vis-à-vis supra-local powers.

    4 In one of our study areas, there are a number of taxpayers whose names do not figure in the official tax register; their payments are probably going into “private hands”.

    5 Ideally, all GPs in the two districts should have been included in our analysis. However, there is an acute dearth of relevant demographic and revenue data at the GP level. After ensuring the availability and comparability of relevant data, we are left with a truncated set of 387 GPs. Out of the total, Bardhaman has 235 GPs and Murshidabad 152 GPs. Tax statistics collected from official sources pertain to 2002-03 for the district of Bardhaman, and 2004-05 for Murshidabad. Demo graphic and literacy figures are culled from the Census 2001. Information on GP-wise seat share of political parties is collected from the data published by the State Election Commission and relates to the panchayat election in2003.

    6 Some would argue, however, that it is uncertainty and possible electoral misfortunes in a politically competitive climate in a democracy rather than certainty of office that results in low tax yields.

    7 We have run regressions including several other potentially important variables, including a number of occupational variables, but have not obtained better and more coherent results than what we get from the set of variables used here. We admit that we do not have any direct measure of asset or property ownership in GPs. Bardhan and Mookherjee (2005) use in their analysis of panchayat developmental and poverty alleviation programmes in West Bengal a measure of land concentration in GPs and claim that with greater land concentration there is lower tax yield.

    8 A sizeable number of GP residents are outside the tax bracket. So the actual tax burden per taxpayer, the figure which is not easily available, would be much higher. In the village where we conducted our ethnographic study the average tax amount in a year per taxpayer hovers around Rs 40 to Rs 50.

    9 In Bardhaman, as per the last panchayat elections, in 91 per cent of selected GPs, the panchayat board is under the control of left parties; in Murshiadabd the corresponding figure is 46 per cent.

    10 Several studies concur that the left did deliver broad-based benefits in terms of land reforms, decentralisation and agricultural productivity improvements. However, it is well to point out that various schemebased benefits, that are distributed by local panchayats, are seen by potential beneficiaries more as personalised favours from the party in power than as entitlements [Corbridge et al 2003].


    Bardhan, P and D Mookherjee (2005): ‘Poverty Alleviation Efforts of Panchayats in West Bengal’, Economic and Political Weekly, 39(9).

    Bates, R and D D Lien (1985): ‘A Note on Taxation, Development and Representative Government’, Politics and Society, 14(1).

    Bhaduri, N (1998): Dandaniti: Prachin Bharatiya Rajshastra (in Bengali), Sahitya Sangsad, Kolkata.

    Bhattacharya, D (2004): ‘West Bengal: Permanent Incumbency and Political Stability’, Economic and Political Weekly, 39(51).

    Bernstein, T and L Xiaobo (2000): ‘Taxation without Representation: Peasants, the Central and Local States in Reform China’, China Quarterly, 163.

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

    Brautigam, D (2002): ‘Building Leviathan: Revenue, State Capacity and Governance’ in Moore and Rakner.

    Chatterjee, P (1984): Bengal 1920-1947, The Land Question, K P Bagchi & Co, Calcutta.

    Corbridge, S, G Williams, R Veron and M Srivastava (2003): ‘Making Social Science Matter-II: How the Rural Poor See the State in Bihar, Jharkhand and West Bengal’, Economic and Political Weekly, 38(25).

    Ghosh, B and S Ghosh (2004): ‘A Note on Panchayat Finances of West Bengal’, typescript.

    Harriss, J (2000): ‘Introduction’ in C J Fuller and B Veronique (eds), The Everyday State and Society in Modern India, Social Science Press, New Delhi.

    Levi, M (1999): ‘Death and Taxes: Extractive Equality and the Development of Democratic Institutions’ in I Shapiro and C Hacker-Gordon, (eds), Democracy’s Value, Cambridge University Press, Cambridge.

    Marjit, S (1999): ‘Decentralised Financing, Governance and Public-Private Cooperation’, Economic and Political Weekly, 34(20).

    Moore, M (2002): ‘Introduction: The New Politics of Taxation and Accountability’ in Moore and Rakner.

  • (2004): ‘Revenues, State Formation, and the Quality of Governance in Developing Countries’, International Political Science Review, 25(3).
  • (forthcoming): ‘Between Coercion and Contract: Narratives around Taxation and Governance’ in D Brautigam, O-H Fjeldstad and M Moore (eds), Capacity and Consent: Taxation and State Building Developing Countries.
  • Moore, M and L Rakner (eds) (2002): ‘The New Politics of Taxation and Accountability’, IDS Bulletin, 33(3).

    Panchayat and Rural Development Department (P&RD), Government of West Bengal (2005): ’Resource Mobilisation by the Panchayati Raj Institutions: A Study of Six Districts’, Typescript.

    Rajaraman, I, O P Bohra and V S Renganathan (1996): ‘Augmentation of Panchayat Resources’, Economic and Political Weekly, 31(18).

    Schumpeter, J A edited by Swedberg R (1991): ‘The Crisis of the Tax State’ in The Economics and Sociology of Capitalism, Princeton University Press, New Jersey.

    The State Election Commission, Government of West Bengal (2005). Tendler, J (2002): ‘Small Firms, the Informal Sector and the Devil’s Deal’ in Moore and Rakner. Tilly, C (1992): Coercion, Capital and European States, AD 990-1992, Blackwell, Cambridge, MA. World Bank (2004): Fiscal Decentralisation to Rural Governments in India, Oxford University Press, New Delhi. Yadav, Y (2006): ‘How West Bengal Voted: The Opportunities and Challenges’, The Hindu, May 16. Yep, R (2004): ‘Can “Tax-for-Fee” Reform Reduce Rural Tension in China?’, The China Quarterly, 177.

    Dear Reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here


    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top