Patrick Heller and T.M. Thomas Isaac*

In evaluating democracy in the developing world, scholars have generally focused on the nature of the regime in power, and have directed much of their attention to the role of political parties and other organized political actors in the electoral arena. Scholars have also of course directed their attention to examining how state structures shape patterns of engagement, and to detailed investigations of patterns and modes of representation. Much of this work has focused on the nation-state as the critical unit of analysis. What has been missing however is a deeper understanding of day-to-day democratic practices, and in particular the nature of the sub-national institutional field(s) on which both formal and informal political agents engage each other and the state. In the absence of such analyses it becomes very difficult to understand why so many developing world democracies have been so ineffective not only in securing social rights but even in providing for basic political and civic rights. The state has certainly been transformed, but has it, in the language that now dominates the post-transition discourses on development, become closer to the people? Why, in particular, has political participation, beyond the periodic exercise of the vote, remained so circumscribed? To what extent have transitions to electoral rule made bureaucratic institutions inherited from authoritarian or colonial pasts more open to involvement by ordinary citizens? Is the reach and robustness of public legality sufficient to guarantee the uniform application of rights of citizenship? Have developing states really changed their modes of governance, the social partners they engage with and the developmental goals they prioritize?

If making a democratic state has been difficult, making a responsive state has been even more difficult. Nowhere is this more the case than in India. The general picture of Indian democracy stands as a reminder that there is no linear progression to democracy. Much as the robustness of India’s democratic institutions have been rightfully celebrated, the effectiveness of those institutions is increasingly in doubt. Fifty-four years of almost uninterrupted democratic rule has done little to reduce the multiple exclusions of India’s subalterns. Digging below the surface, moreover, one finds that within the unitary institutional domain marked by the boundaries of the Indian nation-state, there are marked degrees of democracy, or as Guillermo O'Donnell (1993) has put it, differences in the intensity of citizenship. India’s post-transition history has produced multiple trajectories of democratization. If at one end it is possible to identify redoubts of quasi-feudal authoritarianism (the state of Bihar for example), at the other it is possible to find cases of ongoing democratic deepening. This most notably is the case of the south-western state of Kerala which stands out not only because of its storied history of popular movements led by the Communist Party of India—Marxist CPI(M) or CPM—and of an activist state that has achieved some of the most dramatic social and redistributive gains in the developing world (Heller, 1999), but also because since 1996 it has been the site of one of the boldest, most self-conscious and most extensive experiments in empowered participatory governance.

In 1996 the Left Democratic Front (LDF) coalition returned to power in Kerala and the CPM-led government immediately fulfilled one of its most important campaign pledges by launching the "People’s Campaign for Decentralized Planning". Though the People’s Campaign (Campaign hereafter) is only in its 5th year, it has already achieved a far greater degree of devolution of decision-making powers than any other Indian state. All 1,214 local governments in Kerala—municipalities and the three rural tiers of district, block and grama panchayats (the all-India term for village councils)—have been given new functions and powers of decision-making, and have been granted discretionary budgeting authority over 40% of the state’s developmental expenditures. The Campaign however represents far more than a simple decentralization of governance powers to lower-level elected bodies. In both its political and institutional design it has the socially transformative ambition—much as in the case of Porto Alegre—of compensating for the deficits of formal representative structures and bureaucratic decision-making by building and capacitating participatory institutions. On a number of counts it quite clearly stands out as a bold experiment in building participatory democracy. The first is its sheer scope and scale. The decentralisation of a wide range of developmental responsibilities to 1,214 elected local governments represents a profound reconfiguration of the state and its relationship to society and has the potential of dramatically transforming the everyday practice of democracy for Kerala’s 31 million inhabitants. Second, the institutional design of the campaign’s core institutions—Grama Sabhas (ward-level assemblies), development seminars, task forces and local governments—has self-consciously attempted to nurture and empower a model of democratic and participatory development planning. On the one hand, by having devolved planning and implementation functions to local arenas, the Campaign has for the first time in India meaningfully empowered local governments and communities to address practical problems. The entire planning cycle—which begins with the collection of local data and ends with the formulation of a comprehensive local plan that consists of hundreds of projects—is basically an extended exercise in practical problem-solving. On the other hand, both the institutional and political character of the Campaign has been centrally concerned with promoting bottom-up participation. The devolution of authority and resources to local governments has significantly reduced the transaction costs of participation, and the knowledge and capacity gap that has traditionally excluded ordinary citizens from playing an effective role in governance has been considerably narrowed by mass training programs, the active mobilization of civil society expertise and concerted efforts to empower historically marginalized groups—women, adivasis ("tribals"), and dalits ("untouchables"). Moreover, the participatory institutions of the Campaign are self-consciously deliberative—based on inclusionary and reason-based decision-making—and directly empowered because they tie project choice and formulation to actual implementation. In all these respects the Campaign marks a critical juncture in Kerala’s democratisation trajectory: it is nothing less than an explicit political project aimed at dismantling entrenched forms of bureaucratic domination and patronage politics by reinvigorating Kerala’s tradition of direct, movement-based political engagement and fundamentally democratising the institutional character of the state.

The debate on democratic transitions has understandably focused on the installation of electoral, constitutional and procedural institutions. The unit of analysis has invariably been the nation-state. While useful for typologizing regimes and differentiating democracy from authoritarianism, focusing on formal national-level institutions provides limited analytical leverage for conceptualizing democratic deepening. Because institutions and politics are relational and configurational, their attributes are never perfectly isomorphic either horizontally across different policy arenas, or vertically from one level of the state to another. As the state radiates out from its geographic and functional core its authority and its effectiveness fluctuates dramatically. Much as state-society theorists have called for disaggregating the state (Migdal et al., 1994), we need to disaggregate democracy.

As conventionally defined, formal democracy is marked by universal suffrage, regular and competitive elections, accountability of state apparatuses to elected representatives and legally codified and enforced rights of association (Huber et al, 1999: 168). An effective democracy is one in which democratic practices have spread throughout society, governing not only relations between states and citizens, but also public relations between citizens. Functionally and geographically the degree of public legality in many formal democracies remains severely constrained. In such democracies, notes O'Donnell, "the component of democratic legality and, hence, of publicness and citizenship, fade away at the frontiers of various regions and class, gender and ethnic relations" (1993: 1361). Public spaces disappear to be replaced by areas of privatized power. Local institutions and officials are colonized by bosses, chiefs, dons or caciques. Patrimonialism, clientelism and coercion eat away at democratic authority. Thus we must look beyond the macro-institutional level of parliaments, constitutions and elections and investigate the intermediate and local level institutions and consultative arenas located in the interstices of state and society where "everyday" forms of democracy either flourish or flounder. We need, in other words, a political sociology of democracy, one that specifically recognizes that a working democracy must be an effective democracy.

An effective democracy has two interrelated characteristics: a robust civil society and a capable state. A free and lively civil society makes the state and its agents more accountable by guaranteeing that consultation takes place not just through electoral representation (periodic mandates) but also through constant feedback and negotiation. Civil society is critical to democratic performance because it extends the scope and style of claim-making beyond the formal interest representation that defines political society. Social movements, associations and unions raise new issues and mobilize new actors. In doing so they not only provide a counterbalance to more bureaucratic and aggregated forms of interest representation, but they also create new solidarities, which in many instances specifically challenge existing inequalities and hence help democratize society itself. The key point here is that the health of a democracy is measured as much in the qualitative nature of its social patterns of association as in the formal character of its institutions, and that while these two variables condition each other—associational patterns are conditioned by institutional environments, and institutional responsiveness is conditioned by associational vitality—the pattern can be positively reinforcing just as it can be impairing.

The capacity of the state is also central to the effectiveness of democracy. Procedural guarantees of civic and political rights, including rights of association and free speech, do not automatically translate into the effective exercise of democratic rights. Citizenship is not a right, it is a relation. Where inequalities between social categories are so pronounced as to create extra-constitutional forms of binding authority (clientelism, patriarchy, caste subordination) the exercise of citizenship is subverted. As theorists of civil society have long argued, its associational qualities emerge only when it is doubly differentiated from the state and from primary social groupings (families, kinship groups, lineages). A precondition for the effective exercise of civic and political rights requires a state capable of securing the even, uniform and rational-legal enforcement of public authority. Individuals and groups must be protected from arbitrary state action, but also from forms of social authority that might constrain or impinge upon their civic and political liberties. And creating public spaces that are protected from non-democratic forms of authority requires far more than writing constitutions and holding officials accountable. It marks a fundamental shift in the distribution and locus of what Weber called "legitimate domination" from society to state. Given the contested and unfinished process of state-formation in much of the developing world, and most notably the institutional weakness (or vulnerability to elite capture) of local government, the writ of legally enforced public authority remains limited, producing a low-density form of citizenship. This problem is tied to both the infrastructural and authoritative limits of state power. Infrastructurally, the apparatuses of the state—the police, the judiciary, the educational system—are simply cast too thinly and too unevenly to enforce and provide for citizen’s rights. Authoritatively, the state’s legitimate realm of domination (constitutionally prescribed arenas in which its authority is binding and backed by coercion) is contested and weakened by countervailing sources of authority. And even when central level institutions are robust and fairly well insulated from particular interests (relative autonomy), the meso and local level institutions of the state—which are equally critical to securing effective rights of citizenship, including participation—can nonetheless be extremely weak.

India’s democratic institutions have withstood the test of time and the test of a fissiparous society. The basic procedural infrastructure of democracy - specifically the constitution and guarantees of the rights of association, the separation of powers, and regular and open elections at both the national and state levels - has become firmly entrenched. At a minimum, despite infamous recent episodes of communal and caste violence, democratic institutions have not only helped forge a nation from multiple nationalities but have also institutionalized acceptance of the uncertainty of rule that comes with competitive elections. The authoritarian episode of 1975-77 notwithstanding, the prospects of a democratic reversal in India are remote. India's dominant class factions, proprietary and professional, support democracy, if for no other reasons because they have benefited so handsomely from the largess of India's democratic politics of patronage (Bardhan, 1984).

The effectiveness of India's democratic institutions is an altogether different matter. Throughout vast regions of India the exercise of citizenship rights, even in the limited political sense of the term, is circumscribed by the persistence of traditional forms of social control. With more than half of India's rural households depending on landlords for access to land or labor, clientelistic ties remain key to the survival strategies of subordinate groups. With its ritualized exclusions and deeply ingrained hierarchical relations, the caste system has inscribed these material inequalities with a degree of social and cultural control that has few parallels. Low levels of literacy and discriminatory treatment by upper-caste controlled state institutions have further limited the associational autonomy of lower castes and classes. The corollary of this picture is the predominance of fragmented sovereignty. The reach and authority of the juridical and democratic state end, or more accurately are transfigured, where the writ and power of local strongmen and their caste-based followings begin. In a pattern that closely resembles both the Brazilian and African cases, extra-democratic sources of authority not only resist, but also colonize and privatize state power (Weyland, 1996; Mamdani, 1996). Local notables routinely dominate local institutions, including village governments, schools, cooperative societies and the development bureaucracy. The permeability of state authority is most dramatically exposed by the existence of private caste armies (especially in Bihar) and elite control over local police forces.

This is not to say that democratic institutions in India have been altogether lifeless. The past two decades have witnessed an erosion of traditional clientelist politics. Formal and competitive democracy in India has undermined the legitimacy of traditional social authority, spawned a whole new generation of political entrepreneurs and created spaces in which new groups have been successfully mobilized. But the political forces that have emerged are more rooted in social cleavages than ever. The basis for mobilization has shifted from patronage to identity populism. As elections have become more competitive and more groups have been brought into the political arena on their own terms, patronage has become increasingly tied to identity politics. The demand for government quotas and special privileges, whether of majority or minority communities, now dominates claim-making. This explosion of narrow demands has triggered a frantic and zero-sum scramble for preferential treatment that Bardhan has aptly described as "equal-opportunity plundering by all interest groups" (1997: 16).

In this political climate of populism and organizational fragmentation, encompassing political formations have been the exception to the rule. Labor unions have rarely extended beyond the protective confines of the organized sector (large factories and public employees) and in many instances have become little more than vehicles for the political ambitions of local bosses. Farmers’ associations have been dominated by the interests of large farmers. With the reach of public legality circumscribed by the power of local elites, lower class and lower caste efforts to organize around economic issues (outside of Kerala and West Bengal) have invariably been defeated (Brass, 1994: 334). Against this backdrop of political fragmentation, the capacity of democratic institutions to aggregate interests and in particular address pressing distributional dilemmas is more in doubt than ever. The politics of social citizenship, as Mehta (1997: 64) has remarked, are conspicuous by their absence. Thus, much as in the case of Brazil, political fragmentation has frustrated the equity-enhancing potential (and promise) of democracy (Weyland, 1996).

Despite the fact that the idea of empowered local governments has long been a staple of India’s Gandhian heritage, from Nehru's Community Development Program to national and sub-national efforts to empower India's panchayats, the history of decentralization, much like land reform, has been one of broken promises, slow political deaths, bureaucratic obfuscation and hollow legislation. Where local governments have been given some measure of power they have more often than not been captured by local elites and transformed into instruments of patronage. Regional variations notwithstanding, the balance sheet is clear: with the possible exception of West Bengal the process of shaping and implementing developmental initiatives, including the most basic of day-to-day public services, has remained a top-down affair dominated by the bureaucratic and political elites of state capitals and their intermediaries, brokers and fixers.

There are probably few examples in the world where the causal link between organized social movements and significant redistributive and social gains is as strong as in Kerala. Briefly put, a long history of social mobilization that began with caste reform movements and sporadic peasant uprisings in the 1920s and 1940s crystallized into a lower class movement under the organizational umbrella of the Communist Party which captured power in 1957. Repeated spells in power by the communists, combined with an almost continuous process of militant mass mobilization, exerted unrelenting pressure on the state to expand social programs, regulate labor markets and implement land reforms. Despite a two-decade period of virtual economic stagnation (1970-1990), social indicators have continued to climb, poverty rates have fallen dramatically and wages for unskilled workers (agricultural laborers, construction, beedi (local cigarettes) and cashew nut processing workers) remain the highest in India. The incidence of class-based mass mobilization did drop significantly beginning in the 1980s, but no government has ever rescinded or cut-back a major social program (and this despite increasing fiscal constraints) and in many sectors the state in Kerala (regardless of the party in power) continues to pioneer pro-poor reforms (Heller, 1999). There is little doubt that no other state has been more consistently pro-poor or successfully redistributive in all of India, and possibly anywhere in the democratic world.

All of which makes the launching in 1996 by a CPM-led government of the "People’s Campaign for Decentralized Planning" rather intriguing. Widely regarded as the most far-reaching and radical experiment in decentralization ever undertaken in India, the campaign’s political project has been nothing less than a frontal assault on the bureaucratic fiefdoms of the state and the patronage networks of the political system. The paradox here, as Thomas Isaac has previously noted, is that "a state government launched a movement to force its own hand to radically restructure the mode of governance. Why should any state embark upon such a mission?" (2000: 316). To answer this question requires a brief review of Kerala’s political and developmental history.

The trajectory of Kerala’s anti-caste, nationalist and radical agrarian movements reached a watershed when the Communist Party won Kerala’s first elections in 1957. The Communist Party has never, however, achieved a stable electoral majority and has consequently been in and out of power. This inability to build a sustainable majority support has been the subject of continuous debate in the Party. Its failure to extend its influence among Kerala’s large Muslim and Christian minorities in the State has been an important factor. Moreover, the success of the Communist Party itself in the area of the implementation of land reforms, the spread of education and the development of the cooperative sector, has brought about significant social and economic changes with which the Party has not yet fully come to terms. This, coupled with the fact that the Party has always harbored reservations about the scope of "parliamentary democracy"—having in fact been twice evicted from power by the central government—more than anything else explains why the Party has continually had to reinvent itself and build its mobilizational capacity. As Thomas Isaac has commented elsewhere, "[t]he Left does not have faith in the autonomous transformative power of the state government, which is only part of the overall bourgeois-landlord Indian state. Therefore, while in power or outside, they continue to mobilize the masses in support of the demands. The constant pressure from below is important in understanding the responsiveness of the state machinery" (2000: 141-42). In the 1940s the communists seized the leadership of the Independence movement. In the 1950s and 1960s the party championed the rights of poor tenants. In the 1970s, having lost the support among many who had benefited from land reform, the Party took up the cause of landless laborers. And it also moved aggressively to unionize workers in the informal sector (Heller, 1999).

A critical effect of this sustained history of mobilization was to draw the state in and to extend the reach of public legality. As the state responded to demands from below to redistribute land, to enforce labor legislation, to provide social services, it enmeshed itself in society. At the meso and local levels, this has produced a rich fabric of democratic institutions and authorities. In contrast to the modal Indian picture in which district and village-level institutions are deeply enmeshed in local power configurations and often in the hands of landed elites or dominant castes, in Kerala a wide range of institutions including district councils, panchayats, student councils, cooperative societies, and quasi-corporatist bodies have given multi-level representation to lower-class interests.

Because Kerala’s popular sectors have been successfully mobilized and effectively represented by a programmatic political party, that state has had a pronounced lower-class bias and has orchestrated significant social and redistributive reforms. But if the political character of the state has departed significantly from the Indian norm, its institutional character—and in particular its highly centralized and bureaucratic character—has not. Despite the strength of grassroots movements and institutions, state power has remained highly centralized and until the Campaign in 1996, local government in Kerala was little more than an administrative extension of the state government. Given Kerala’s sub-national status, this is hardly surprising. The Indian state—and Kerala as part of it—was born at the intersection of an imperial bureaucracy and Soviet-inspired visions of planned transformation, and deeply imbued with Nehruvian high modernism. In the hubris that marked the nationalist celebration of the state as designated agent of modernization and emancipation, top-down planning became the instrument of development of choice. As Kaviraj notes,

Without falling into the asocial and reductionist logic of public choice theory that sees a voracious, self-seeking predator in every bureaucrat and politician, the accumulation of such powers, exercised with little accountability from below, has inevitably produced interests and networks of privilege that have nothing to gain and everything to lose from a decentralization of power. The political solidity of this configuration finds its distributive logic in the rental havens that the dominant proprietary classes—including bureaucrats and politicians—have all carved out for themselves (Bardhan, 1984 and 1997).

In its demonstrated capacity to deliver social programs and its much higher degree of public accountability, the state in Kerala is a far cry from the proto-predatory states of North India. But as we have seen the difference is more in the demand-side of the equation—pressure from social movements and a vocal civil society for state action—than the supply side, as the state in Kerala has not been spared the entrenchment and ossification of rent-seeking interests. The size and power of such interests, moreover, is in no small part a product of Kerala’s redistributive project, and specifically the exponential growth of the service bureaucracy and the proliferation of (mostly unprofitable) public sector enterprises. The CPM, moreover, has always embraced "democratic centralism" and, with it, a fairly orthodox state-led and top-down vision of development.

To understand the CPM’s present emphasis on decentralization, three developments have to be singled out. First, the party has come to recognize the limits of its electoral appeal, and in a context of competitive party politics has identified democratic decentralization—with its attendant principles of non-partisanship, de-bureaucratized government and sustainable development—as the key to appealing to new constituencies. Second, the embrace of decentralization marks a tacit recognition that the redistributive capacities of the developmental state have exhausted themselves. The broad-based social movements that saw the expansion of social citizenship have been weakened and increasingly replaced by more narrow and sectoral interests. The latter vested interests have become rent-seeking and a stumbling block for democratic reforms of the state structure. If a strong, centralized and interventionist state did secure many of the benefits associated with the Kerala model (high levels of social development, extensive public infrastructure, basic institutional reforms), the second generation social development challenges Kerala faces (the quality, rather than the quantity of public services), call for a fundamentally different mode of governance. Third, despite a recent growth spurt, Kerala’s continued economic problems—in particular the lack of dynamism in commodity-producing sectors—has underscored the failures of the dirigiste state, and has prompted the recognition for the need to develop more flexible and decentralized forms of state intervention, designed to nurture, rather than regulate and control economic activity. Finally, the most palpable and devastating blow to top-down development has come from the widely perceived deterioration of public services. In comparative terms, the quality of Kerala's public health and educational services remain decades ahead of any other Indian state. But by local standards, and specifically those of a literate and increasingly middle-class society, what is probably at worst only a marginal decline in the quality of provision has produced widespread public disaffection. Though blamed in large part on a non-performing bureaucracy, the deterioration of the public sector has also been explicitly tied to the commodifying logic of globalization (and specifically the explosion of private clinics and schools in Kerala) and the cost-cutting imperatives of neoliberalism. Faced with significant challenges to the sustainability of Kerala’s social gains and the legitimacy crisis of an overly bureaucratized state beleaguered by vested interests, the CPM leadership embraced democratic decentralization in 1996 as a strategy for reinventing and reinvigorating the role of the public sector.

If the CPM’s return to power in 1996 provided the critical opening for the transformation of the state, it is the State Planning Board that has formulated, designed and driven the Campaign. In doing so, it is important to highlight that the SPB has relied closely on a stock of practical knowledge, ideas and experiences drawn from twenty-five years of local-level experiments conducted by NGOs, most notably the Kerala Sastra Sahitya Parishad (KSSP)—the People’s Science Movement. The KSSP, moreover, has played an active role at the grassroots level in implementing the Campaign. The focus of the rest of this chapter is on describing and evaluating the key institutions and processes of the Campaign and is informed primarily by the direct involvement of one of the authors—T.M. Thomas Isaac—who was a member of the SPB during the first 5 years of the Campaign and a long time activist in the KSSP, and research conducted by both authors.

As an institutional reform program the Campaign was specifically designed to nurture and facilitate greater direct participation by citizens in decision-making and was predicated on two guiding principles. The first was that local government institutions should be transformed from simple delivery instruments for national and state schemes into fully fledged governing institutions with functional, financial and administrative autonomy, and that devolution of functions and resources should be based on the principle of subsidarity (what can best be done at local level must be done there).

The second principle was that traditional representative structures need to be complemented by more direct forms of democracy. Popular participation, it was argued, would make elected representatives continuously, rather than just periodically, accountable to citizens and would introduce more transparency into the functioning of the bureaucracy. Increasing levels of direct and informed participation required both mobilizing citizens and designing institutions that enable ordinary citizens to play an active role in the selection, design and implementation of local development plans.

The campaign’s designers also realized from the outset that the instrumentalities of the state would be inadequate, both politically and practically, to the task of pushing through the necessary reforms. Given the inertia of existing institutions and the power of vested interests, legislation could never sustain such profound changes. The success of Kerala’s land reforms in the 1970s—widely recognized as having been the most far-reaching and equity-enhancing in the sub-continent—was made possible by the backing of a powerful peasant movement. A highly successful mass literacy campaign in 1991 (in which the KSSP was actively involved) also pointed to the importance of mobilizing popular initiative. Building on these lessons, and the recognition that Kerala has an impressive reservoir of mobilizational capacity, the strategic emphasis from the outset was to conduct the reforms as a campaign.

In the next section we present a detailed discussion of the campaign’s institutional design and how it has sought to reconcile the democratic objectives of extensive participation and effective deliberation with the need for technical competency and inter-level coordination in the formulation and implementation of development plans. The effectiveness of these mechanisms in achieving the objectives of democratic decentralisation are critically evaluated in the last section.

Democratic decentralization requires changes in administrative structures, in the allocation of functions and powers, and in the control of resources. All three are interrelated and to an extent need to be introduced simultaneously. In the technocratic model advocated by multilateral development agencies, decentralization is seen as an exercise in incremental institution building informed primarily by public administration and managerial sciences. Typically it is argued that certain sequenced preconditions, defined by a clear demarcation of functions among the various levels, must be met before genuine authoritative decision-making power can be successfully devolved: administrative support structures have to be created, new organizational procedures have to be put into place, government staff have to be redeployed, a new information base has to be developed and new personnel—both voluntary and official—have to be trained. Most significantly, the devolution of financial resources has to be carefully calibrated to the absorptive capacity of the nascent institutions.

What is most problematic about this linear model of decentralization is the assumption that the task of transforming the very mode in which government works can be achieved through a prescribed process of introducing a discrete set of technically and managerially rational solutions. A largely frictionless and apolitical world is more or less taken for granted. But successful and sustainable democratic decentralisation has been the exception to the rule, frustrated more often than not by bureaucratic inertia—most notably the resistance of powerful line departments—and vested political interests.

Kerala certainly has its share of entrenched bureaucratic fiefdoms and political formations with a stake in the status quo. Yet, in the short history of the Campaign, devolution has already gone far beyond the issuance of laws and executive orders.

The most dramatic step has been the devolution since 1997-98 of between 35 and 40% of the annual developmental budget to LSGIs—local self-governing institutions (grama, block and district panchayats, and municipalities). During 1997-98 the total resources devolved (the "grant-in-aid") amounted to Rs.10,250 million and in 1998-99 Rs. 11,780 million, sums that do not include funds from centrally-sponsored schemes and institutional loans to local governments. Before 1996-97, LSGIs received on average a measly Rs. 200 million in untied funds per year. There is little doubt that the administrative capacity and the management experience of the newly-elected local government representatives was hardly up to the task of accommodating such a large-scale transfer. But devolving fiscal resources and control—even while the immense task of building a new regulatory environment and administrative capacity was only getting underway—has had two critical strategic effects. First, because local governments now enjoy significant budgetary discretion, local planning exercises have a tangible and immediate character. This, as we shall see, has invited high levels of participation. Second, shifting budgetary authority to lower levels has limited the ability of patronage politicians and top-down line departments to derail the process.

The second distinctive feature of the decentralisation experiment in Kerala is the central role accorded to the planning function of the LSGIs. As a statutory precondition for receiving the grant-in-aid from the government, LSGIs must prepare a comprehensive area plan. The planning process, as prescribed by the SPB, includes holding grama sabhas (ward-level assemblies), and convening sectoral task forces in which non-official experts and volunteers directly prepare reports, formulate projects, and draft sectoral plans. The various stages of plan preparation in effect represent new participatory spaces in which citizens, elected representatives and officials deliberate and prioritize developmental goals and projects.

In order to ensure transparency and participation without compromising the technical requirements of planning, the planning process is divided into discrete phases with distinct objectives, key activities and associated training programs. Though modifications to the sequence have been made every year, the basic model that was inaugurated in 1997 (Table 1) remains the same.

A critical component of the Campaign has been an elaborate training program that has developed into one of the largest non-formal education programs ever undertaken in India. In the first year, in seven rounds of training at state, district and local level, some 15,000 elected representatives, 25,000 officials and 75,000 volunteers were given training. About 600 state level trainees—called Key Resource Persons (KRP)—received nearly 20 days of training. Some 12,000 district level trainees—District Resource Persons (DRP)—received 10 days of training and at the local level more than 100,000 persons received at least 5 days of training. All the elected representatives were expected to participate in the training program at one level or another. Each round of training focussed on specific planning activities. Separate handbooks and guides, amounting to nearly 4,000 pages of documentation, were prepared and distributed for each round.

Following the seminal analysis of Putnam (1993) it is now widely accepted that a robust civil society—defined in terms of its "norms of reciprocity and networks of civic engagement" and embodied in different types of civic institutions—is critical to securing the effectiveness of democratic institutions. Putnam’s understanding of the contribution that associational life can make to deepening democracy is however informed by an essentialist interpretation that construes civic-minded behaviour as deeply engraved in culture and history. It is, as Skocpol and Fiorina (1999) have argued, a social-psychological view that leaves little room for the role of conflict in building democratic capacities. Critics have moreover pointed out that the forms of civic life that contribute to securing developmental goods (i.e. social capital) are in fact politically constructed (Evans, 1996), and that associational life is in large part artifactual, the product of institutional environments, shifting social relations and state interventions (Cohen and Rogers, 1995).

This mutability of civil society is fully illustrated in Kerala’s contemporary history. If Kerala’s long history of social mobilization has directly contributed to the vibrancy of its civil society, it has also indirectly contributed to developments that have eroded the capacity for civic action. Class-based redistributive conflicts had two notable effects. First, they polarised Kerala’s political landscape between two highly mobilized left and right wing formations that systematically penetrated civil society organisations. Thus, schools, cooperatives, shopfloors and local institutions have all become the object of fierce political competition. With this systematic politicization of civil society it has become increasingly difficult to separate the provision of public services and goods from narrow political-organizational imperatives. Second, redistributive demands saw the expansion of the size and role of the state, and the growth of bureaucratic structures. Though large-scale interventions in education, health and social protection directly contributed to Kerala’s social development, the growth of the bureaucracy has severely circumscribed the scope for civil society initiative. Because the bureaucratic development process is top-heavy and more responsive to highly organised rent-seeking interests than popular forces, ordinary citizens retain an interest in government programs only inasmuch as narrow, individual returns are concerned. The politics of pork have increasingly replaced the politics of community improvement, and Kerala's strong traditions of popular grassroots development action have eroded over time.

The impetus behind the launching of the Campaign stems directly from a critique of the corrosive effects of these developments. On the one hand there is a recognition that a centralized, command-and-control state is no longer capable of driving Kerala’s development and that new forms of state and public action are called for. Thus, the supporters of the Campaign have been very vocal in arguing that the existing political climate of sectarian and partisan division has become an obstacle to development and that a key objective of the Campaign—much as in the case of popular budgeting in Porto Alegre—is to break the hold of clientelistic politics. On the other hand there is the recognition that civil society initiatives must be afforded more avenues and opportunities for effective engagement with public authorities.

In conceptualising planning as an instrument of social mobilization, the Campaign has sought to deepen democracy along 3 different axes. First, devolving planning and authoritative decision-making to local arenas allows for a more integrated approach to development that directly challenges the hold of hierarchical line departments and their extensive powers of control. Second, by providing visible and substantive incentives for participation, and by emphasizing deliberative processes, local development planning holds the possibility of reinvigorating civic action and loosening the grip of patronage and partisan politics. Third, by fundamentally transforming the mode and channels of decision-making, the Campaign has created new political configurations and public policy networks. Thus, elected local representatives whose functions were previously mostly ceremonial, have now been brought directly into positions of authoritative decision-making, including authority over local officials. Similarly, NGOs and CBOs have been offered new opportunities for engaging directly in development and there has been a concerted effort to create new linkages between professionals and academic institutions and communities in order to bring expertise (especially during a transitional phase in which the bureaucracy has been less than cooperative) to the grassroots. This later development in many respects parallels the dynamic blurring of state-society relations marked by the emergence of new associational networks that Chalmers et al. (1997) have identified as the defining characteristic of revitalized civil societies in Latin America.

In short, the objective of the People's Campaign for Decentralised Planning has not been simply to draw up a plan from below. The very process of planning has been conceived as a means to fundamentally transform the character and scope of participation and the nature of interest mediation. Such a transformation cannot be secured through government directives alone. It requires the creativity and the social logic of a movement (Isaac, 1999a).

4.4. Institutionalisation

One of the greatest challenges of promoting participatory democracy is to develop institutional forms that are sufficiently robust as to withstand efforts by traditional interest groups to either subvert or circumvent deliberative processes. In Kerala’s highly volatile political climate, in which the two political fronts have historically more or less alternated in power, this problem is particularly acute. Governments formed by the Congress Party have a track record of reversing decentralisation reforms, most notably by packing newly-created local institutions with political appointees.

The Campaign addressed the challenge of institutionalization by generating as much popular involvement as possible. High levels of participation wielded significant payoffs as some Congress aligned parties—and most interestingly the conservative Muslim League—expressed their support for the Campaign. The campaign’s localized planning structures have moreover created spaces in which new political alliances and commitments have been forged. By replacing the conventional systems of vertical accountability to political parties and bureaucracies with more horizontal forms of cooperation and autonomous sources of authority, the campaign’s locally integrated planning structures have provided local politicians and officials with a direct stake in the new system. Political uncertainty has also underscored the need to institutionalize the Campaign in formal terms, that is, through the passage of appropriate legislation. Thus, the LDF Government comprehensively amended the existing Kerala Panchayathi Raj Act of 1994 and the Kerala Municipality Act of 1994 with the effect of securing the autonomy of LSGIs and mandating the presentation of local plans and budgets to Grama Sabhas. New laws concerning the transparency of administration and access to information have also been passed. Moreover, hundreds of government orders creating new accounting systems, devolving authority to local officials and establishing new procedures for reporting have engraved many of the campaigns design features into the everyday workings of government.

In May of 2001, in keeping with a pattern of defeat of incumbent parties that has long been the norm in Kerala, the LDF was ousted from power by a Congress Party-led coalition. Most observers concur that the CPM’s defeat was not a judgement about the Campaign. At the time of going to press (November 2001) it is still too early to evaluate what impact the change in government will have on the Campaign. In contrast to 1991, when the UDF returned to power and immediately scuttled a much less ambitious experiment in decentralization (one that had focused on the district level) the new government has declared its commitment to the Campaign and to redressing its weaknesses. Two factors have pre-empted a frontal assault on the Campaign. One is the popularity of the Campaign at the grassroots level which cuts across the political divide. The Campaign had succeeded in building a bi-partisan coalition at the grassroots level in favour of decentralization, and any efforts to erode the autonomy and authority of LSGIs will be difficult, not only because it would require significant legislative efforts, but also because such efforts would alienate the Congress party’s own rank and file, who in coalition with other parties control roughly half the LSGIs in the state. The second is the prestige that the Campaign had gained in the national and international circles. In addition to significant media attention, the Campaign has attracted the attention of officials from other Indian states and even figured in the remarks made by the President of India in his last independence day national address.

Planning in India has historically been a highly insulated and top-down affair. The official literature on decentralised planning in India has generally been sceptical of direct mass participation in the planning process (Govt. of India, 1969, 1978 and 1984). District-level plans have generally been formulated by designated experts after what have been at best perfunctory consultations with selected stakeholders. By the early 1980s, some form of district planning machinery existed in most of the states, and the planning process was described by the Report of the Working Group on District Planning (Govt. of India, 1984) as follows:

Under the Campaign, the planning process begins at the lowest level of democratic representation, the grama panchayats and municipalities. It is only once these local governments have prepared their plans that block- and district-level panchayats come into play, and then only to ensure regional coordination. There are 990 grama panchayats, 58 municipalities, 152 blocks and 14 districts in Kerala. The councils for each of these levels of local government are directly elected on a first-past-the-post constituency system. At the block and district levels, the democratic character of planning is ensured through the involvement of elected officials and a range of citizen committees. At the municipal and grama panchayat level, the planning process is driven by direct mass participation.

Autonomous decision-making power was granted to local governments by providing untied funds ("grants-in-aid"). The heavy hand of bureaucratic traditions has been blunted by ensuring continuous, mass, non-official participation in every phase of plan preparation and implementation. In building continuous deliberative structures, the Campaign has had to tackle two micro-level design challenges. The first has been to create institutional forms that can correct for the asymmetries of power among local agents. The second has been to make local participation effective by allowing space for grassroots intervention and deliberation without compromising the technical and economic requirements of planning.

5.1. The grama sabhas

Grama sabhas, the assemblies of ward or panchayat-based residents, represent the key deliberative moment in the planning process. By law they had to be held at least two times during the initial years of the Campaign, and in later years, after amendments to the law, four times a year. The first grama sabha serves as an open forum in which residents identify local development problems, generate priorities and form sub-sector development seminars in which specific proposals first take shape. In the second grama sabha, the plans approved by the elected Panchayat council are presented to the public and departures from the original grama sabha proposals are explained. Beneficiaries for projects are also selected at the grama sabhas.

Rousseau notwithstanding, there is nothing spontaneously democratic about a general assembly, especially in a society as inflected with complex and durable inequalities as India’s. The commitment of the campaign’s architects and activists to building deliberative institutions is reflected in the time and energy that has been devoted to finding practical solutions to the problems of large meetings. An obvious innovation, but one that nonetheless required significant organizational effort, was to adopt a small group approach. In each grama sabha, after an introductory general body meeting (usually of several hundred people), participants are divided into smaller groups, each dealing with a particular development sector, to discuss issues and problems in depth. This small-group arrangement made it possible for ordinary people, particularly women, to be able to participate in the discussions. A second innovation was to provide a semi-formal discussion format and a trained facilitator for each group. Working with a basic template of questions and useful planning concepts, the role of the locally recruited facilitator is to encourage participants to list and analyse local problems based upon their real-life experiences.

5.2. Local information gathering

Asymmetries of information are a key source of domination in nominally deliberative institutions. Even in Kerala’s social climate of highly politicized and highly literate citizens, durable social and status inequalities and the hoarding of official expertise by state institutions has severely skewed access to useful information. Moreover, though available planning data is a source of significant power, it is anything but accurate or properly adapted to the requirements of local development. Taking much of its inspiration from the KSSP—which since its founding in 1962 has been dedicated to "bringing science to the people"—the Campaign has taken local information gathering as a first critical step in the planning process.

After a first round of grama sabhas, panchayats in the first year of the Campaign were required to make a formal assessment of the natural and human resources of the locality. The idea was to promote effective integration of planning and resource optimization by actually comparing expressed needs with local assets. With assistance from specially-trained Resource Persons and using techniques developed by the Campaign, a series of participatory studies were undertaken in every grama panchayat and municipality. These included the collection and organization of data available in various local-level offices, the identification and mapping of local eco-zones using a rapid appraisal technique, a review of ongoing schemes to be prepared by each local department, a social audit, and a review of local history. By and large, departments refused to cooperate, and this had serious consequences for integrating existing schemes into the new plans. The quality of the data of course varied dramatically from one locality to the other, but the exercise itself had the important effect of helping individuals develop useful skills and tapping into, and formally incorporating, local knowledge.

The outcome of the data collection exercises was a "development report" prepared according to guidelines set down by the SPB. With a five-year strategic outlook, the reports serve as the basis of the annual planning exercise. Running on average 75-100 pages, the reports provide a comprehensive overview of local development and include a chapter on local social history intended to underscore the role that social mobilization can play in meeting contemporary development challenges. The body of the reports consists of twelve chapters assessing the current status of each sector, a review of ongoing schemes and problems and a list of recommendations.

Because the recommendations of the Development Report can differ from the demands raised in the grama sabhas and because demands from different wards had to be integrated into an area-wide perspective, the reports were submitted to development seminars. The majority of delegates to the seminars were elected from the subject groups of the grama sabhas with, in principle, equal representation for men and women. Local-level government officials from the relevant departments were asked to participate, as were any experts invited by the panchayat executive committee. On average, development seminars had 231 delegates, with officials accounting for 13.8%, SC/STs (Scheduled Castes and Scheduled Tribes - the official designation for erstwhile "untouchables" and tribal groups) for 10.5% and women for only 22.1%. Extensive preparation went into the organisation of the seminars, including the distribution of the Development Report to all delegates and widespread publicity in the form of leaflets, festivals, jathas (marches) and exhibitions. The seminars were given a very high profile, with a Member of Legislative Assembly or a State Minister inaugurating 50% of the seminars. A major proportion of the seminar time was devoted to sector-wise group discussions in order to facilitate in-depth analysis of the development reports and propose amendments. The recommendations of the different groups were then presented to a plenary session and adopted.

At the conclusion of the development seminars in the first year of the Campaign, task forces of around 10 persons each were constituted to prepare the project proposals on the basis of the recommendations of the seminar (in subsequent years, task forces became the starting point of the planning process with development seminars being convened at a later stage to review the work of task forces). A key challenge in building participatory institutions is to ensure that experts, rather than simply deliberating amongst themselves, must engage in direct deliberation with citizens (Fung and Wright, 2002). The work of task forces in fact goes beyond simply levelling the playing field by in fact guaranteeing that the process of project design is informed by experts, but led by citizens. Development seminars form a total of 12 task forces, one for each development sector. The delegates selected from the development seminars are ordinary citizens, though many have undergone specialized training through the Campaign. The chairperson of the task force is an elected ward councilor. This ensures that the work of the task force will be directly linked and supported in subsequent deliberations of the panchayat or municipal council. In order to secure the relevant expertise as well as coordination with state structures, the convenor of the task force is an officer from the concerned line department.

The sustainability of a participatory institution is in large part determined by its demonstrated capacity for effective problem-solving. In order to ensure a degree of quality control and effective monitoring, task forces are required to prepare detailed project proposals in accordance with a set of criteria and standards established by the SPB. Thus, all project proposals must include a definition of objectives (as far as possible in quantitative/measurable terms), criteria for beneficiaries or areas, a time frame, an organizational overview of the role of implementing agencies, a financial analysis including identification of funding sources, a social and environmental impact review and details of the proposed monitoring mechanisms.

The fourth and final stage of the local planning process is marked by the prioritization and integration of the projects prepared by the various task into a single panchayat or plan document. The final form of the local plan is the legal prerogative of the elected council which must formally vote on the plan. There are however a number of formal and informal mechanisms that ensure that elected representatives abide by the recommendations and projects generated by the various participatory processes. Formally, the approved plan must conform to a detailed reporting format that lays out the general strategy and objectives of the plan as well as sectoral and redistributive criteria. Authorized projects must be specifically linked to the strategic statement and the full text of the proposed project must be listed in a separate appendix. This process not only guarantees accountability, but its sheer complexity ensures that the council—which has limited administrative support—has no practical alternative to building on the work of the task forces. The fact that ward councillors participate actively at every level of the participatory process, from attendance at grama sabhas and training seminars to chairing the task forces, also ensures integration between the participatory processes and the council’s final deliberations. Finally, the entire process of beneficiary selection, which is of course especially vulnerable to political abuse, is, as we shall see, the subject of an entirely separate process of regulated transparency and participation.

Since the beginning of the Campaign, plan allocations are separately indicated in the state budget, with broad guidelines regarding sectoral allocations to be made by the local body. These guidelines are both of a functional (sectoral) and redistributive character and are designed to coordinate and integrate local allocations with state-wide objectives. In order, for example, to shift public investments away from Kerala’s traditional strengths in social services and infrastructure, the SPB mandates that 40-50% of plan allocations must be directed to the productive sector. On the redistributive front, local governments are required to spend not less than 10% on projects targeted to women, and in proportion to their local population for Scheduled Castes and Scheduled Tribes.

Block and district panchayats start the preparation of their annual plans only after grama panchayats have drafted their plans. The sequential ordering is intended to ensure that the plans of the various tiers are integrated and the plans of the higher tiers complement, rather than duplicate, those of the lower tiers. A matrix-based analytical tool has been developed to assist blocks and districts in integrating the analysis and programs of the grama panchayats into their own plans. Blocks have also been tasked with integrating into their plans the different centrally-sponsored poverty alleviation schemes that have traditionally been implemented at the block level. There has been strong resistance to this move from both bureaucrats and elected representatives. In part this is due to genuine problems arising from the existence of separate guidelines for centrally-sponsored programs, but it is mostly a reaction to the prospect of losing significant decision-making powers.

In the first year of the Campaign, a sample review of the projects prepared by the local bodies revealed that a significant proportion of them had to be modified to ensure their technical soundness and viability before they could be approved for implementation.

In all, more than 100,000 projects had to be evaluated. The evaluation was not for selection or rejection of the projects, but to rectify the technical and financial weaknesses in the project proposals. This monumental task had to be undertaken within a span of three to four months. The official machinery was neither capable nor willing to cope with the task.

The SPB responded to this problem by launching the Voluntary Technical Corps (VTC). Retired technical experts and professionals were encouraged to volunteer their skills for appraising the projects and plans of the local bodies. A professional or postgraduate degree or officer-level experience in a development sector was specified as the minimum qualification for membership in the VTC. A volunteer expert committed herself/himself to spending at least one day a week giving technical assistance to the panchayats. District-level conventions were arranged for the experts who formally offered to join the VTC. More than 4,000 technical experts enrolled in the VTC. Expert committees that included a government official were then formed at the various local government levels.

The expert committees act both as advisory arms of the District Planning Committees, helping the latter to appraise the plans and projects, and as advisory committees to local planners. The committees are not empowered to modify priorities set by the local bodies. Their tasks are carefully limited to providing technical and financial advice and appraisal of projects, and suggesting modifications where necessary. The District Planning Committees approve plans on the recommendations of the expert committees. The formation of expert committees in the course of the campaign’s first year was an important organisational innovation which helped to de-bureaucratise the project appraisal and technical sanction procedures. Without this mobilization of extra-bureaucratic expertise these tasks would have bogged down in the line departments through inertia and outright resistance. Not surprisingly, these committees have been the subject of much public debate fuelled in particular by claims that the committees are a partisan attempt to create parallel structures to elected bodies.

In Kerala’s traditional system of development planning the decision-making process was the arbitrary and patronage-driven domain of elected representatives and implementation was the prerogative of the bureaucracy. A key rationale for making the decision-making process more participatory is to ensure the involvement of the beneficiaries and the public at large in the implementation phase. As Fung and Wright note, "direct participation of grassroots operators increases accountability and reduces the length of the chain of agency that accompanies political parties and their bureaucratic apparatus" (2002). Popular involvement increases problem-solving efficiency through better and more rapid feedback and increases accountability by multiplying the points of scrutiny. The Campaign has evolved a wide range of new fora and rules to maximise participation and transparency.

The Campaign’s financial procedures for regulating the flow of grant-in-aid funds to local bodies and to specific projects has been designed to maximize effective monitoring. To begin with, because the various officers transferred to grama panchayats are now directly responsible to the elected council, they can be held more directly responsible for financial flows. Financial allotments to local bodies are released in 4 instalments. All funds must be specifically tied to an approved panchayat project or state scheme, and held in special accounts that are managed by the implementing officer. Actual disbursement of funds requires co-authorization from the head of the elected body.

The creation of democratically-accountable beneficiary committees has also been an important innovation. Instead of implementing public works through contractors, local bodies are encouraged to form committees of project beneficiaries to undertake the task. The idea here is to break the ties of collusion between contractors, politicians and government engineers that have historically been the most important sources of corruption. Doing so, however, requires creating beneficiaries committees that are sufficiently autonomous and empowered to resist capture by rent-seeking interests. A key step was to shift effective authority for the technical sanction of projects from department officials to block/municipal and district level expert committees. Department officials are the convenors of the subject committees and continue to formally grant technical sanction. However, they now make decisions in their capacity as members of a committee of peers rather than officials in a departmental hierarchy. A second procedural innovation has been to shift responsibility for examining finished work and authorizing payment from official to non-official engineering experts from the Voluntary Technical Corps.

A major change introduced by the Campaign was in the procedure for selecting beneficiaries for development projects. In the past beneficiary selection has been little more than a concerted exercise in patronage that has more or less enjoyed the tacit collusion of all political parties. Campaign rules call for grama panchayats to extensively publicize the criteria for beneficiary eligibility and prioritization. Notices listing the projects and the criteria have to be prominently displayed in public places as well as printed and circulated. Applications must be printed in the Malayalam language and made freely available. The rules also provide for a system for verifying statements made in the applications. Verification can be conducted by designated officers or by a committee appointed by the panchayat. Finally, the list of applicants must be presented to the grama sabha with sector subject groups tasked with processing applications. Grama sabhas, moreover, are authorized to include sub-criteria for prioritization.

The responsibility for consolidating and finalizing the priority list of beneficiaries received from each grama sabha rests with the panchayat. The final priority list has to be created on the basis of clearly-stated norms. In no case can the relative priorities from each ward be overturned during the process of consolidation. Members of the public and the local press can attend the proceedings of this final selection. The draft list must be exhibited prominently. All public objections must be given consideration and reasons for rejection stated.

6. Critically assessing the Campaign

So far we have discussed the procedural and institutional design of the Campaign in its ideal type. But how have these new structures actually worked on the ground? Most critically, how deliberative has the planning process been, and to what degree have the activities of decentralized units been effectively coordinated with technical inputs and integrated with higher levels of planning? Given the sheer complexity and scale of the project, the inevitable teething problems and the absence of cumulative data, it is still too early to pass a definitive judgement. The institutional learning that has already taken place does however hold some important lessons for our understanding of participatory democracy, and the emergence of some fairly transparent and robust trends does allow for some tentative assessments.

As we noted in the introductory section, it was the decision in 1996 to earmark 35 to 40 percent of the plan funds for the local self-governments that kick-started the Campaign. The most important achievement of the Campaign to date has been sustaining the political will to maintain and even increase the scale of devolution in subsequent years, and this despite very severe financial constraints faced by the State Government. Local governments, in other words, have enjoyed a continuous and substantial flow of financial resources.

If the scale of resource devolution has been maintained, its redistributive character has improved significantly. In the first year financial devolution was based on a straight per capita formula that did not take levels of inter-regional poverty and development into account. What was lost in policy, was, however, gained in politics. This bland formula had the advantage of being beyond political manipulation and as such was not open to criticism by the opposition of partisanship. Moreover, the formula did effectively correct for the highly skewed patterns of patronage-driven allocation of the past (in which underdeveloped northern Kerala was inevitably short-changed) and as such did have a de facto redistributive effect. In subsequent years, the devolution formula has progressively incorporated new indices of poverty and underdevelopment.

That for the first time in India grama panchayats and municipalities throughout an entire state have prepared local area plans is a milestone in and of itself. Given the sheer enormity of the task and the lack of local experience and capacity, plan preparation in the first year ran 6 months over schedule. The dramatic returns of learning-by-doing are however reflected in the steady reduction in the time overruns that have marked each subsequent planning year.

A major objective of decentralised planning has been to match local needs and potential to actual public expenditures patterns. A rationalization of resource allocation based on more direct, informed and deliberated inputs into the decision-making process represents one of two critical efficiency gains associated with decentralized planning (the other being the increase in accountability). Because of the empirical difficulties of comparing pre- and post-Campaign expenditures patterns (there are no sub-district figures available for the pre-campaign period) a definitive assessment will have to await more intensive research efforts. Three important general trends can however already be highlighted. First, the investment priorities in the plans prepared by the local bodies differ significantly from the investment priorities in the district plans that were formulated from above before decentralization. Much greater priority is now accorded for basic needs such as housing, drinking water and sanitation by the local bodies. In the productive sectors there has been a discernible shift towards animal husbandry, garden crops and minor irrigation. Both these shifts have significant redistributive implications. Second, in contrast to past patterns the investment priorities in the special plans prepared for Scheduled Castes and Tribals differed significantly from the overall investment patterns. This points to an effort to take the weak income, asset and skill position of these marginalized communities into account. Third, in contrast to the one-size-fits-all logic of the past, there are significant interregional differences in the investment priorities of the local bodies.

The most glaring weakness of the plan preparation in the first year was the quality of the proposed projects. Many of the projects proved to be little more than modified versions of standardized department schemes. There was often little consideration of forward and backward linkages and fully integrated plans were actually rare. The reflex to mechanically allocate funds on a ward basis proved tenacious, particularly among the higher tiers (blocks and districts). Beginning with the second year, measures were adopted to improve the quality of projects and programmes. The most important measure has been to introduce subject-specific training programs for task-force members. In the second year the training program consisted of a series of locally organised stop-gap measures that produced limited results. In the third year the training program was upgraded and formalized into a state-wide program that is linked to specialized institutions such as the Kerala Agricultural University, the Institute of Management in Government, the KSSP’s Integrated Rural Technology Centre, COSTFORD (a low-cost housing NGO training institute), and a few other NGOs involved in watershed management. These specialized training programs coupled with the greater involvement of VTC members in the task forces should help improve the quality of project design.

In addition to technical deficiencies, the planning process also suffered from inter-level coordination problems. Effective decentralized planning must by definition be integrated. This is critical not only to optimizing resource allocation, reducing duplication and ensuring sustainability, but also for capturing and diffusing the innovations generated in decentralized units. The comparative advantage of "decentralized coordination" lies in increasing the "learning capacity of the system as a whole by combining decentralised empowered deliberation and centralized coordination and feedback" (Fung and Wright, 2002). This has been one of the most daunting challenges faced by the Campaign.

In the first year a number of factors contributed to weak coordination between the plans of the different tiers of local bodies and that of the state government. Though planning procedures prescribed by the SPB called for higher tiers to take the priorities and programs of lower tiers into account, in actual practice there was little coordination in the first year (in no small part because of a shortage of time). More detailed guidelines were issued in the second year, but problems persisted. In the third year the format and logic of district-level planning was significantly overhauled. More emphasis was given to the district’s role in 1) providing a macro perspective for sustainable development of the district, 2) improving integration by consolidating lower level plans and identifying gaps and duplications, and 3) providing a long-term strategic vision for future annual plans.

A major criticism of the Campaign is that all the attention to process and participation has come at the expense of actual delivery as measured by physical achievements (the process-product trade-off). The logic of this criticism is misplaced inasmuch as it fails to recognize that the quality of participation is an important objective in its own right. To focus on financial targets and expenditures, as many of the campaign’s critics have done, reflects a narrow technocratic understanding of development. But even if the building of participatory institutions can be justified on the grounds of extending citizenship alone, their long-term viability, especially under the circumstances of the liberalization of the national economy, will rely on the capacity to provide tangible developmental goods.

At this stage an accurate appraisal of physical achievements is complicated by practical problems of monitoring and aggregating existing data. Physical results, particularly in productive sectors such as industry and agriculture, will take time to materialize. And even in the case of social and infrastructural sectors, the task of actually measuring the quality of project implementation is virtually impossible given the absence of a local data gathering system.

The most readily measured physical achievements of the first two years of decentralised planning are however impressive. In the two years from 1997 to 1999, 98,494 houses have been built, 240,307 sanitary latrines constructed, 50,162 wells dug, 17,489 public taps provided and 16,563 ponds cleaned. A total of 2,800,179 individual beneficiaries received support from the plan for seedlings and fertilisers. And the 8,000 km of roads that were built far surpassed past achievements.

Because the pace of delivery has in fact surpassed expectations, the State Government has taken steps to encourage institutional financial loans to the local bodies to provide further resources. And for the first time in Kerala (or for any state in India), the government has actually set a target date (2003) for delivering shelter, sanitary latrines and drinking water (within 200 meters) to all households in the state. The universalization of pre-primary education, improvement in the quality of education and health care centres, and completion of rural electrification are also on the mid-term strategic agenda. Tangible achievements in the above sectors in the immediate future could play a critical role in sustaining and stabilising the process of democratic decentralisation.

The Campaign has created numerous opportunities for ordinary citizens to actively participate in the different phases of plan formulation and implementation. But how many citizens have made use of these opportunities? Were the discussions manipulated by locally dominant groups? Were the different forums merely a means to legitimise decisions made by the elites?

Every ordinary citizen irrespective of his/her membership in political or non-political social formations has the right and opportunity to intervene in the planning process by participating in the grama sabhas. One of the greatest achievements of the Campaign has been to demonstrate that popular assemblies can function effectively. In the year before the Campaign, grama sabhas were called after the formation of the new local bodies, but a majority failed to actually convene. In the first grama sabhas of the Campaign in August-September 1996 over 2 million people participated, with an average of 180 persons per grama sabha, representing 11.4 percent of the voting population and roughly one of every 4 households. Though participation rates have dropped slightly in subsequent years (possibly because the number of annual grama sabhas was increased from 2 to 4), these popular assemblies have become an essential feature of Kerala’s political landscape.

There are however significant limitations to the deliberative character of grama sabhas. To begin with they are obviously still too large and unwieldy for meaningful deliberation, the small group approach notwithstanding. Because of Kerala’s dispersed settlement pattern, grama sabhas participants must travel significant distances and meetings can not run more than 2-3 hours. This does not allow for serious discussion of the large number of complex issues that are normally included in the agenda of the grama sabha. Participation across socio-economic groups has been uneven. By all accounts middle class participation has been low, and most participants have been from lower classes that are the targeted beneficiaries of most development projects. In the first year the participation of Scheduled Castes and Scheduled Tribes was below their population share and women constituted a disappointing 25 per cent. In subsequent years the percentages have increased, but participation remains uneven.

The formation of Neighborhood Groups (NHGs) consisting of 40 to 50 families has been a response from below—often initiated by KSSP activists—to the limitations of the grama sabhas. Though not formally required, NHGs have been formed in around 200 panchayats. A study found that in 100 panchayats (Isaac, 1999c) NHGs function as mini-grama sabhas, discussing local issues and priorities, reviewing plan implementation and selecting beneficiaries. NHG representatives often constitute a Ward Committee which in many cases becomes the de facto executive committee of the grama sabhas. NHGs have also taken up other activities such as conflict resolution, after-school educational programs, health clinics, cultural activities, savings schemes, and project implementation. There is currently a campaign being led by the KSSP to extend NHGs to the entire state and to institutionalize what is in effect a new layer of grassroots democracy. The crowding-in effect that the Campaign appears to be having on associational life in Kerala is also evidenced in the proliferation of a variety of self-help groups, particularly women’s micro-credit schemes (Seema and Mukherjee, 2000; Manjula, 2000).

6.5. Corruption and nepotism

One of the most important criticisms of decentralization is that it often does little more than devolve corruption. Indeed, funnelling substantial funds without proper safeguards to localities will inevitably fuel rent-seeking behaviour, and possibly even community conflict. The media and opposition parties in Kerala have raised serious allegations of nepotism in beneficiary selection and corruption in the implementation of projects. Of the nearly 30,000 beneficiary committees it has been alleged that a substantial number are led by nominees of contractors (so-called benami committees). State investigating agencies have also pointed to widespread irregularities in the first year’s plan implementation (Isaac, 1999d).

In its own evaluation the State Planning Board concluded that irregularities during the first annual plan resulted more from inexperience and haste than corruption. For example, when the local bodies in the first year found it difficult to absorb and properly distribute funds, many transferred the funds to non-plan accounts or deposited the money with Government or quasi-Government agencies such as Electricity Boards or the Kerala Water Authority in order to claim full utilisation before the spending deadline. Even though regulations were bent and even broken, there was little leakage as such. Irregular expenditures that were identified by the government were disallowed and with the new rules put in place in subsequent years, such improprieties have declined sharply.

There is little doubt that many beneficiary committees have fallen prey to vested interests. But there is also little doubt that the traditional corruption nexus between contractor, engineer, and politician has been decisively broken in a large number of local bodies. For example, in the district of Kannur—a CPM stronghold—an investigation revealed that beneficiary committees have been carefully constituted and run according to the campaign’s criteria of transparency and democratic accountability. Strengthening the capacity and accountability of beneficiary committees remains one of the most important priorities of the Campaign, and a number of important reforms have already been introduced. But even if there has been and continues to be some leakage of funds due to the capture or manipulation of beneficiary committees by vested interests, most observers agree that the multiplication of checks and balances and the increased scrutiny associated with citizen participation represents a dramatic improvement of the systematic and routinized plunder that characterized the traditional system.

With respect to the process of selecting beneficiaries the returns on institutional fine-tuning and increased community experience have been visible. During the first year complaints about the selection process were registered in a majority of local bodies. The volume of registered complaints is in itself indicative of the increased transparency of the system. The traditional system was entirely based on patronage. Complaints were rare simply because the information was accessible only to the patrons and their clients. The rules for beneficiary selection have been modified in every year of the Campaign and by the third year less than a fifth of panchayats were registering complaints.

As much as the Campaign has been concerned with the efficacy of deliberative institutions, it has also, in keeping with Kerala’s long history of redistributive struggles, promoted the strategic goal of building equitable forms of participation and reducing substantive inequality. Gender justice in particular has been declared to be one of the major objectives of the Campaign. We have already noted efforts to increase participation of women in grama sabhas, and the extension of Neighborhood Groups and Self-Help organisations are clearly strengthening the associational capacities of women. Two other important strategies have been efforts to build on the constitutional provision for 1/3 reserved representation of women in LSGIs and the introduction of a special Women Component Plan amounting to 10 per cent of the plan outlay. What has been the experience so far?

The Kerala experience to date certainly bears out the importance of affirmative action ("reservations" in the Indian context) in representative structures and indeed suggests that the principle should be extended to higher levels of government. But affirmative action alone is insufficient. An in-depth study of elected representatives in Kerala revealed that while elected women representatives are better educated than their male counterparts (a social fact that is unique to Kerala in the Indian context), the women were on average younger, much less politically experienced, and inadequately equipped with a basic knowledge of rules, regulations, and administrative issues. Moreover, women representatives have had to bear a triple burden of public office, income earning activities, and domestic duties. From its outset, the Campaign has run an in-depth and continuous capacity-building program targeted to women representatives. The training program, which has evolved significantly to adapt to new challenges, has yielded impressive results. A self-assessment survey of elected women representatives shows that their administrative knowledge and management skills, as well as their ability to officiate at public functions and interact effectively with their constituencies, have improved very significantly over the last three years (Isaac et. al., 1999).

The Women Component Plan (WCP) for the first year did not meet campaign targets, both in terms of overall allocation and the relevance of projects. An obvious factor here was the insufficient representation of women among trained resource persons. This problem has been directly addressed in subsequent rounds of training. As women activists and representatives have started to play a more proactive and informed role in the Campaign, the effectiveness, content and scope of the WCP has improved. First, more than the statutory minimum requirement of 10 per cent of the plan grant-in-aid was earmarked for WCP in all districts. Second, an undue emphasis on credit and beneficiary contribution in women development projects was reduced and more realistic patterns of project financing were adopted during the second year. Third, the quality of projects improved. The tendency to include the general sector projects in WCP on the basis of notional (indirect) benefits to women has declined and the number of projects that specifically address the gender status of women has significantly increased.

The fear that the interests of Scheduled Castes and Scheduled Tribes are more readily subverted at the local level where severe caste inequality persists has often been raised by SC/ST leaders. How have SC/STs fared under decentralised planning in Kerala so far?

The Special Component Plan (SCP) and Tribal Sub Plan (TSP) in Kerala have been formulated and implemented in a decentralised manner since the mid-1980s. But this decentralisation has been purely bureaucratic and has lacked real participation by any elected representatives, let alone members of the community. Under the Campaign, 75 to 80 percent of the SCP and TSP funds were devolved to LSGIs—that is, almost entirely taken out of the hands of the state bureaucracy.

The first visible impact of decentralised planning has been a significant increase in the funds actually earmarked and spent for Scheduled Castes and Scheduled Tribes. Careful disaggregation shows that a substantial part of SCP and TSP have always been calculated on the basis of notional flows, i.e. by including general schemes that encompass, rather than target, SC and ST communities. The Campaign entirely abolished this system of calculation. As a result, the SPB estimates that real resources for the weaker sections have increased by 30 to 40 per cent as compared to the pre-Campaign period. The SPB plan appraisal also revealed that fears that local bodies would divert funds were misplaced: except in rare instances local bodies have fully accounted for grant-in-aid from SCP and TSP. And even though it was permissible to allocate up to 30 per cent of the grant-in-aid from SCP and TSP for infrastructure projects such as roads and bridges, actual expenditure under this heading was less than 20 per cent. The emphasis was on projects that could be specifically targeted for individual beneficiaries from SC and ST communities such as housing, latrines, and income-producing animals. It was also later made mandatory that special assemblies of tribal populations had to be convened to discuss and approve the tribal sub-plan. An educated local youth from every tribal hamlet was selected, trained and paid an honorarium to ensure more effective tribal participation in the planning process.

The Campaign represents a watershed in the post-independence history of Kerala in that it has made the very nature and institutional character of the state an object of contestation, with the goal of attempting to deepen and widen democracy. With every local plan that is formulated and every local project that is implemented, the new institutions and procedures of decentralized participation take root. Because this in turn strengthens civil society, and brings previously excluded or marginalized actors into the political arena, it may well be that democratic deepening becomes self-sustaining. But because the mobilizational mode that the Campaign has taken will become increasingly difficult to sustain as local planning becomes routinized, sustaining the integrity and efficacy of deliberative institutions will require institutionalizing the authority and resource base of local governments. With respect to passing necessary legislation and new regulations much has already been done. But these gains can be quickly unravelled or hollowed out if the new institutions fail to deliver. And sustainable delivery rests on first maintaining adequate levels of financial devolution, and second on successfully reforming the bureaucracy. Both factors in turn rest on permutations in the political equation.

The return to power of a Congress-led government means that the Campaign will no longer benefit from political leadership and will lose significant state support. Already, despite its public declarations of support for the Campaign, the government has weakened the institutional moorings of the Campaign by promoting parallel structures. Thus it has split the Panchayat department into two separate entities, introduced new regulatory authorities that are outside of the Campaign’s integrated structures, and has pledged to provide Members of the Legislative Assembly with funds for local development that in effect bypass panchayats. The government has also undermined the Campaign’s formal and informal support structures by demobilizing trained resource persons, providing only minimal training programs, and putting the redeployment of department officials to the local level in cold-storage.

But even if the Campaign now finds itself settling into a lower-level equilibrium, it nonetheless still represents a dramatic advance on the pre-campaign period with local government playing a far greater role in development than anywhere else in India. Moreover, five years of experimentation with decentralized planning in Kerala has created new sources of democratic authority and generated lessons that are certain to have a lasting impact. Politically, the most important lesson has been that decentralization and people’s participation can and does work. Even if only a small proportion of panchayats have even approximated the ideal of local planning, the demonstration effect of what is possible has had profound reverberations. Very concretely these hundreds of points of experimentation have brought countless innovations to project design and implementation, and these have been energetically diffused through innovative training programs in which panchayats teach each other. A once impervious and all powerful bureaucracy has, in hundreds of local communities, been displaced by the collective efforts of ordinary citizens. Ordinary citizens who have never been afforded an opportunity to effectively engage the state outside of campaign-oriented social movements now routinely deliberate and cooperate with elected representatives and local officials in deciding how to spend large sums of money. And a generalized discontent and even cynical despair about politics has in part been replaced by an open, articulate and relentless attack on patronage politics, and the beginnings, through everyday participatory practices, of a new kind of transformative politics. At a very minimum, this is reflected in the new-found respect that political parties have for civil society.

The second broad lesson is that there are no blueprints, and that any successful reform effort of this scope and depth will of necessity be one of learning-by-doing. Being confident about the normative desirability of participatory institutions thus also implies being comfortable with the notion that making such institutions work is a process of trial and error that requires continuous feedback and institutional fine-tuning. What can be asserted is that the required flexibility calls for institutional designs that strike a balance between local autonomy and initiative with enforced procedural and redistributive standards and higher-level strategic integration. What Kerala’s experience suggests, however, is that such institutions themselves are most likely to emerge from a programmatic political project that consciously reaches out to civil society and builds on the creative and even mischievous logic of social movements.

Bardhan, Pranab (1984). The Political Economy of Development in India. New York: Basil Blackwell Inc..

Bardhan, Pranab. (1997). "Sharing the Spoils: Group Equity, Development, and Democracy." (Unpublished paper). University of California, Berkeley.

Bardhan, Pranab (1999). "The State Against Society: the Great Divide in Indian Social Science Discourse," in Sugata Bose and Ayesha Jalal (eds). Nationalism, Democracy and Development. New York: Oxford University Press, 184-195.

Brass, Paul R. (1994). The Politics of India Since Independence. Cambridge: Cambridge University Press.

Chalmers, Doug, Scott Martin and Kerianne Piester (1997). "The Associative Network: Emerging Patterns of Popular Representation," in Doug Chalmers et al. (eds). The New Politics of Inequality in Latin America. New York: Oxford University Press.

Cohen, Joshua and Joel Rogers (1995). "Secondary Associations and Democratic Governance," in Erik O. Wright (ed.). Association and Democracy. London: Verso Press.

Evans, Peter (1996). "Government Action, Social Capital and Development: Reviewing the Evidence on Synergy," World Development, 24(6), 1119-1132.

Fung, Archon and Erik Olin Wright (2001). "Deepening Democracy: Innovations in Empowered Participatory Governance," Politics and Society 29(1), 4-42.

Fung, Archon and E.O. Wright (2002). "Thinking About Empowered Participatory Governance," in Archon Fung and E.O. Wright (eds.). Deepening Democracy: Institutional Innovations in Empowered Participatory Democracy. London, Verso Press.

Heller, Patrick (1999). The Labor of Development: Workers in the Transformation of Capitalism in Kerala, India. Cornell University Press.

Heller, Patrick (2000). "Degrees of Democracy: Some Comparative Lessons From India," World Politics, 52 (July), 484-519.

Heller, Patrick (2001). "Moving the State: The Politics of Decentralization in Kerala, South Africa and Porto Alegre," Politics and Society 29(1), 131-163.

Huber, Evelyne, Dietrich Rueschemeyer and J. Stephens (1999). "The Paradoxes of Contemporary Democracy: Formal, Paticipatory, and Social Dimensions," in Lisa Anderson (ed.), Transitions to Democracy. New York: Columbia University Press.

Kaviraj, Sudipta (1998). "Critique of the Passive Revolution," in Partha Chaterjee (ed.), State and Politics in India. Delhi: Oxford University Press.

Government of India (1969). Guidelines for the Formulation of District Plans, Planning Commission, New Delhi.

Government of India (1978). Report of the Working Group on Block Level Planning, Planning Commission, New Delhi.

Government of India (1984). Report of the Working Group on District Planning, Planning Commission, New Delhi.

Government of Kerala (1997). Budget 1997-98. Thiruvananthapuram: Ministry of Finance.

Mamdani, Mahmood (1996). Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton: Princeton University Press.

Manjula, B (2000). "Voices From the Spiral of Silence: A Case Study of Samatha Self Help Groups of Ulloor." Paper presented at the International Conference on Democratic Decentralisation, Thiruvananthapuram, May 23-27.

Mehta, Pratap (1997). "India: Fragmentation Amid Consensus," Journal of Democracy, 8(1), 56-69.

Migdal, Joel, Atul Kohli and Vivienne Shue (1994). State Power and Social Forces: Domination and Transformation the Third World. Cambridge: Cambridge University Press.

O'Donnell, Guillermo (1993). "On the State, Democratization and Some Conceptual Problems: A Latin American View with Glances at Some Postcommunist Countries," World Development 21(8), 1355-1359.

Putnam, Robert (1993). Making Democracy Work, Civic Traditions in Modern Italy. Princeton: Princeton University Press.

Seema, T.N. and Vanitha Mukherjee (2000). "Gender Governance and Citizenship in Decentralised Planning." Paper presented at the International Conference on Democratic Decentralisation, Thiruvananthapuram, May 23-27.

Skocpol, Theda and M. Fiorina (1999). Civic Engagement in American Democracy. Washington D.C.: the Brookings Institution.

Thomas Isaac, T.M. and R. Franke (2000). Local Democracy and Development: People’s Campaign for Decentralised Planning in Kerala. New Delhi: Leftword.

Thomas Isaac, T. M. (1999a). "The Socio-Economic and Political Context of People’s Planning Campaign." National Workshop on Decentralised Governance, organised by Kerala Institute of Local Administration and Swiss Agency for Development and Cooperation, Thrissur.

Thomas Isaac, T. M. (1999b). "People’s Planning—Towards a Handbook." Presented at the Workshop organised by the State Planning Board, Tripura, Agarthala, May 3 and 4.

Thomas Isaac, T. M. (1999c). "Janakeeyasoothranavum Ayalkoottangalum—Anubhavangalum Padangalum" [People’s Planning and Neighbourhood Groups-Lessons from Experience], in Ayalkootta Sangamam [Neighborhood Groups] Vol. I, Kerala State Planning Board, Thiruvananthapuram.

Thomas Isaac, T. M. (1999d). "Gunabhokthra Samithikalude Anubhava Padangal" [People’s Planning and Beneficiary Committees-Lessons from Experience], in Gunabhokthra Samithikalum Janakeeyasoothranavum [People’s Planning and Beneficiary Committees]. Kerala State Planning Board, Thiruvananthapuram.

Thomas Isaac, T. M. et. al (1999). "Gender and Decentralised Planning—The Experience of People’s Campaign," (unpublished Working Paper), Center for Development Studies, Thiruvananthapuram.

Thomas Isaac, T.M. and P Heller (2002). "Decentralisation, Democracy and Development: People’s Campaign for Decentralized Planning in Kerala," in A. Fung and E.O. Wright (eds.), Deepening Democracy: Institutional Innovations in Empowered Participatory Democracy. London, Verso Press.

Weyland, Kurt (1996). Democracy Without Equity: Failures of Reform in Brazil. Pittsburgh: University of Pittsburgh Press.

* Disclaimer: Thomas Isaac was erroneously identified as a researcher in this project. He was never part of the team of researchers constituted for this project. He was not a participant in any of the projects' organizing meetings. His contribution was to agree to have an edited version of a paper he had written on Decentralization in Kerala included in the volume on democratic participation. We regret any misunderstandings that his inclusion in this list may have caused.