Introduction

From chapter 2, one is made aware of a range of cross-country experiences with decentralization. That chapter also brought to fore the many mistakes various countries had made while engaging in reforms to their administrative structures. For example, it can be seen that most countries (barring Brazil) have not put in place mechanisms for creating fiscal buoyancy. Financial autonomy of local governments is always an issue in several countries. What is interesting to note is that even if all other factors related to decentralization are weak, the importance of fiscal decentralization and, in particular, the positive effects of local governments raising their own revenues are understood and often promoted. The principle of subsidiarity has been followed to varying extents in these countries. As always Brazil, and to a certain extent, Uganda have put in place mechanisms that will enable local governments to take decisions in matters that they obviously have advantage over the central government. There are also cases of de-concentration being passed off as decentralization and state capture of local governments as in Sri Lanka and Bangladesh.

What about India? Local bodies and institutions had started to decline after the fall of the various indigenous kingdoms and empires. In particular, institutions such as the panchayats that catered to the rural population all but disappeared by the middle of the fourteenth century. Over a significant period of time (starting in the late-thirteenth century in the northern part of India) village economies were slowly being converted into revenue centres for the urban areas. Participation by citizens in governance, accountability of the local administration in particular and local autonomy slowly vanished.

This change was further accelerated by the British. Along with this came economic decline and frequent famines (even by conservative estimates 25 million died between 1790 and 1890. Another 4 million died in famines between 1900 and 1945 when food produce from India was used to feed the British war efforts (Ambirajan, 1976). These deaths were caused by the near complete absence of relief efforts provided by the administration of the day (the British were in charge of the administration either directly or through their princely agents by the late-eighteenth century when the famines started).

In this chapter, the analytics associated with the findings reported in this book have been selectively discussed. In various chapters of this book, we analyse choices of individuals and households, such as the type of schools chosen for their children, payment of bribes, voting, attendance at gram sabha meetings, coping strategies, etc. The choices made by the panchayats such as expenditures, problem solving and provision of public goods and services are also evaluated and, outcomes of the choices made by the households and the panchayats are subsequently measured. The latter include quality and quantity of public services, access to welfare programmes, transitions out of poverty, literacy rates, consumption and incomes. Outcomes are measured at the individual, household and the village levels. Similarly, the variables affecting these outcomes are measured variously at the member, household, village and on occasions at the panchayat, the taluk, the district and the state levels.

Some examples of the type of analysis done include measuring the impact of various types of grants received by the panchayats on tax revenues, impact of school choice by households on literacy rates, estimating the impact of political reservations for women on the quality and process of governance, and the impact of institutions on coping strategies undertaken by households.

In order to analyse the causal links between the choices of individuals, households or local governments and outcomes, economists often build simplified optimizing or bargaining models that are used to gain insights into the structure of the problem analysed. These models will then form the basis for inferring expected signs and the magnitudes of impacts. The models are built to be theoretically consistent. The empirical estimation equations that emerge from this structure are the ‘reduced form equations’ because the structure of the underlying problem has been reduced to an estimation equation. If the structure is well defined, then structural parameters can be recovered from the estimated equations, which will help in understanding the pathways of the impacts to be analysed.

It has often been noted that different structures can lead to similar or identical reduced form equations. Nevertheless, the reduced form equations often produce coefficient estimates that are directly policy relevant. For example, change in panchayat expenditures arising out of untied funds can be estimated to have increased infrastructure investment, suggesting that the magnitude of untied grants should be increased to accelerate infrastructure investment.

Alleviating poverty and improving the living standards especially of the rural poor has been high on the agenda of many developing countries. Despite specific programmes for economic and social development, a substantial number of poor people continue to live in appalling conditions in these countries. Experts who find nothing wrong with the design of these programmes, therefore, focus their attention on the suitability of delivery mechanisms. Some of them feel that the administrative set-up in these countries is not really conducive to promote welfare as targeting is poor and delivery is clumsy. Ensuring people's participation in the implementation of development programmes is considered a better alternative to deliver benefits as well as essential services to the poor.

The past few decades have seen an increasing trend wherein developing countries are reducing their barriers to trade and embarking on a process of domestic reforms. However, there is a concern that such a process could exacerbate the economic and social inequalities within these countries. In order to minimize the adverse consequences of trade and other economic reforms, many governments have started promoting inclusive growth as a mechanism for achieving many of their goals for economic development. From this angle also, decentralization and devolution of powers to elected local governments that could eventually lead to local self-governance is often thought of and prescribed as a tool for achieving growth.

Decentralization which, by design, ensures people's participation in planning and governance is also looked upon as an efficient instrument for reaching benefits to the people with least costs. Accountability and transparency, the cornerstones of decentralized governance, are expected to ensure minimization of misidentification of beneficiaries, better enforcement relating to asset acquisition, its retention and maintenance, capital plough-back, better provision of services etc. Besides, local wisdom and local resources being readily available under decentralized governance, local human, animal and physical resources can also be effectively utilized. This can result in creation of employment opportunities and production of goods and services relevant to the needs of society.

With these ends in view, many countries have put in place mechanisms for decentralization. In India, for example, the Constitution was amended in 1992 via the 73rd and 74th amendments to allow for increased local self-governance.

Introduction

The last few decades saw widespread fiscal decentralization across the world. Although in most cases decentralization would be partial, which means that although the local governments are given large responsibility, powers related to revenue collection were vested with the central government. Policy autonomy of local governments is greatly reduced by this type of decentralization. Studies have shown that partial decentralization have many flaws ranging from reduced accountability and incentives to pursue sound economic policies, increased rent-seeking and corruption, soft-budget constraints, to clientelism in the allocation of transfers.

Panchayats in India raise very little revenue of their own and depend mainly on transfers from above, which come in the form of block grants and a multitude of earmarked grants for specific purposes. The overall question addressed in this chapter is: how can the very low levels of own revenue-raising of panchayats be increased via changes in devolution of functions and resources, from higher levels and in the autonomy given to use the panchayats over the use of the funds?

Devolution of functions, the level of transfers, and autonomy over the use of transfers will provide positive or negative incentives to raise taxes, the incentives effect of transfers. They can either crowd out own revenue-raising, or make it more attractive in order to complement the transfers. For example, a transfer for education may crowd out revenueraising for local schools, or unrestricted block grants may provide incentives to do even more, as the public expenditure pattern can now conform more closely to the preferences of the village community.

However, the transfers can also affect the tax base and therefore revenues. This chapter uses profits per capita from farm and non-farm enterprises as the tax base. If grants are for public works or employment generation programmes, they will generate employment and, therefore, tend to increase village wages and depress profits and impact the tax revenue directly. This is called this the wage effect of transfers. In addition, the expenditures from grants can directly affect profits by the impact the public goods and services have on productivity, which is called the productivity effects of transfers.

The research contained in this book has been motivated by a seminal paper by Foster and Rosenzweig (2001) on how democracy influences the level, growth and distribution of economic resources. Central to this debate is the question of whether and to what extent democratization helps to alleviate poverty by serving the interests of the poor. On one hand, it has been argued that increased democratization promotes the welfare of the poor by improving flows of information between citizens and policymakers, and by increasing the accountability of policymakers to poor and low-status individuals. On the other hand, it has been suggested that democratization may adversely affect the welfare of the poor by increasing rent-seeking behaviour and distributing decision-making to relatively uninformed individuals, thereby shifting public resources from those with fewer private resources. Their book uses a unique panel of villages and households (the ARIS/REDS data sets) and examines some of the effects of deepening democracy in rural India. It shows that a shift towards democracy results in the implementation of outcomes that serves to reduce poverty and inequality. The outcomes are not necessarily firstbest outcomes.

The research in this book is intellectually indebted to this paper. It takes democratization at the local level as given and explores the different mechanisms and outcomes associated with local democratization (henceforth decentralization) in rural India and focuses on empowerment, service provision and the general well-being of households. By using the extremely rich NCAER-REDS panel data set, it is able to show not just correlations of the mechanisms with the impacts, but rather their causal impacts.

There are many institutions, individuals and groups whom I wish to acknowledge and thank both individually and on behalf of my co-authors.

This research was made possible by a grant from IDRC to NCAER (‘Building Policy Research Capacity for Rural Governance and Growth in India’ (grant number 105223). IDRC was the only grant making institution that recognized the importance of informed research based on economic theory for making policy. I wish to thank Stephen McGurk and Evan Due for taking an affirmative decision to fund this grant proposal and continued support.

Introduction

The primary concern of this chapter is whether panchayats can be leveraged for effective provision of services and achieves the desired development outcomes in terms of literacy, health and income. Basic services, in particular for education and health, were identified by the World Development Report (World Bank, 2000) as a key strategy for poverty reduction. A main message of that report is that improvements in services for the poor require enhanced accountability: of service providers to the citizens, of service providers to politicians and of politicians to citizens. Among other possible options these accountability improvements can be achieved by devolution of services to local governments. Choice among service providers also enhances accountability, as do payment or copayment requirements for services.

Chapter 3 raised several administrative and financial issues that form the core of service provision among panchayats in Indian villages, whereas chapter 7 has dealt extensively with qualitative evidence around provision of water and schooling in 4 villages across Maharashtra and Orissa. This chapter serves as the quantitative counterpart to the FGDs in chapter 7.

In India drinking water is the most decentralized service, with panchayats in charge of investments and maintenance. They are allowed to charge user fees, but not all of the panchayats do so. In education and health the panchayat government and the gram sabha have considerable oversight responsibility, which they may exercise with the help of statutory bodies such as VECs or pani (water) panchayats. They can also contribute financially to these services. In education and health, but not in drinking water, citizens can choose from public and private services.

The authors show that during the period of analysis involvement of citizens in the gram sabha has improved in terms of quantity and quality. This is likely to have reduced information asymmetries and therefore enabled citizens to make more informed choices about public services, as well as about their own choice of service providers and time allocation. It is also shown that, if local government panchayats are given adequate powers to manage services, then the quality of service provision, on average, improves substantially. In particular, it is shown that the involvement of panchayats in financing and management increases the availability of education, health and water, and their quality.

Introduction

The process to allow empowerment in the political sphere, reservations to increase women's political power by reserving a share of seats for women who contest in any election has become popular globally. In fact, nearly 100 countries worldwide are reported to practice gender quotas to overcome gender bias and long-standing inequalities. Still, electoral quotas remain controversial. This is due to the suspicion that the effects of political reservations may at best be temporary, or could affect only a narrow clientele. While supporters of political reservations argue that empowering members of historically disadvantaged groups can result in more inclusive processes of policymaking drawing in those previously excluded. To the extent that such measures improve access to public goods (e.g. education or roads) by those who were earlier excluded or marginalized, political reservations can be viewed as Pareto optimal strategy ensuring better development and use of a society's human potential. Critics note that such measures run a danger of bringing to office individuals who lack necessary qualifications and may then be easily manipulated by traditional elites. In addition to adverse effects, reservations, when they are in force, may also adversely affect behaviour by competitively elected leaders who, because they are prevented from standing for re-election, will have their time horizon truncated and, thus, be tempted to adopt myopic patterns of behaviour that do not maximize long-term social welfare.

This chapter explores the effects of political reservations for women in the Indian panchayats. Chapter 3 has shown that the constitution provides reserving a share of seats by rotation in the panchayats. The qualitative evidence from chapter 5 suggests that in panchayats headed by women there is a greater likelihood that both women and men participate actively in the process of governance. The Mani Shankar Aiyar Committee Report (2013) asserts that political reservations have enabled the participation by nearly a million women in panchayats as functionaries and there is not much evidence of capture of power by elite groups either. Other evidence also points to improved service provision and better functioning gram sabhas. In this chapter, empirical evidence of the development impact of political reservation for women on a range of government processes, outputs and development outcomes have been provided.

From Chapter 4, it was possible to infer some of the lacunae in the literature. One of these is the fact that much of the literature is unable to place either the household or a member of the household within a policy space. Consequently, the analysis of household or member level reactions to policy changes or the impact of such reactions on policy is quite sparse. A feature of the analysis presented in this book is the use of a unique panel data of household members, households, villages and panchayats that allows for identification of the relevant policy connections and impact among these units of analysis. For example, this allows us to evaluate the consequences of political participation and decisions of the panchayat on household-level choices and development outcomes in terms of literacy, health, consumption and income.

The basis of the data used for the analysis presented in this book is the Additional Rural incomes Surveys (ARIS) of 1968–69. Three annual rounds of these surveys were conducted ending in 1970–71. The ARIS survey was repeated in 1982, 1999 and 2006. However, starting with 1982 this was renamed as Rural Economic and Demography Survey. The details of the sample selection and some of the summary features of the data are outlined in this chapter.

Sample

ARIS was designed to obtain data on income, investment, saving and consumer expenditure in rural India, from a panel of households first selected in 1968–69. A sample of 5115 households across 17 states was selected according to a multistage stratified probability sampling. The basic unit of the sample was the household. Since it was not feasible to sample the households as the first stage, a list of all villages in a given district or a community development block (CD block) was prepared. These villages were divided into three strata. Villages, covered by the Intensive Agricultural Development Programme (IADP),1 were classified as Stratum 163 villages were selected across 15 states (except Haryana and Himachal Pradesh) for the study. All villages in the community blocks where the Intensive Agricultural Area Programme (IAAP) had been implemented were grouped to form Stratum 2.

Introduction

The preceding quotes by both members of the academic community and the political party suggest that identity defined by Jati (caste) ought to define political processes and help achieve desirable development outcomes for those groups defined by such identities. From Chapter 10, it can be noted that enabling panchayats will create avenues for households to access services. Villages have social networks formed along caste identities that may fracture the ability to arrive at a consensus. Under such circumstances, will decentralization lead to identity-based voting? In fact, from chapter 3 it is learnt that the likelihood that the possibility of such behaviour by households was the reason for many of the founding members of the constitution to oppose decentralization being provided for within its ambit.

With respect to identity-based voting, Munshi and Rosenzweig (2010) have shown that it could result in group members being able to hold elected officials to account and, therefore, allow than to elect more able representatives. This section also analyses bribes, and with respect to both behaviours it goes further by analysing how identitybased voting and payments of bribes improves the access of households to public services and their consumption in the face of imperfections of the governance system.

Introduction

Why should a study of decentralization and its attendant outcomes be taken up? Decentralization and devolution of powers to local governments has been adopted as a development strategy in many countries worldwide. However, it is not merely the sheer number of countries using decentralization as a development strategy that makes it an important reform for comprehensive analysis, rather the systemic change that is expected to be brought about in governance – making it more participatory, transparent and accountable to citizens, which makes it a reform measure worthy of detailed examination. Decentralization is viewed both as a strategy and as a system for bringing about economic development.

Decentralization, as is generally understood, implies devolution of powers to local bodies that are legally separated from the Central government. These local bodies are headed by representatives who are elected by a formally constituted electoral process. These elected representatives are, in turn, vested with powers whose jurisdiction is limited to the geographic boundary of the local body and have been vested with legally guaranteed provisions to make decisions and make use of various financial resources.

The principle of subsidiarity is central to the discussion on decentralization. This is evident in a quote from Tocqueville (1945):

Reid Buckley (2008) goes a step further and asks:

In chapter 5, it was found from the REDS data – that there have been improvements in service provision, access and quality. Such improvements have been modest in most cases. For example, average teacher absenteeism has reduced but is still quite high. The time spent by members of the households in fetching water has declined and a significant number of households in villages now have access to drinking water. However, this evidence is not a conclusive proof that the panchayats were the causal factor in engendering these changes. This chapter begins to explore whether panchayats can help in improved quality of service provision. Qualitative methods are used to examine whether panchayats can succeed in effective management of services specifically, can PRIs be used to provide services such as water and manage education? To what extent and under what conditions will these institutions succeed?

The 73rd constitutional amendment has strengthened the legal and formal structure to the PRIs. It was expected that by adequately empowering panchayats through devolution and promoting rural selfgovernance provision of services will become more efficient. Have panchayats been able to achieve this objective? To provide a few initial insights into the mechanics of service provision, Focus Group Discussions (FGDs), in which various stakeholders are brought together to evaluate and discuss the role of panchayats in service provision and management, were undertaken in two villages each of Maharashtra and Orissa. Maharashtra is industrially and economically more advanced compared to Orissa. It also is one of the states which initiated decentralized governance much before the constitutional amendment in 1992. Problems of water scarcity are common to both states, while the literacy rates in Maharashtra are higher. Both states have introduced legislations that provide improved management of water resources. pani panchayats have come into existence in these two states. The pani panchayat is a voluntary group of villagers engaged in the collective management (harvesting and equitable distribution) of water for drinking and/or irrigation. Meena Manch is an institution specific to Maharashtra to promote the education of girl children. Sarva Siksha Abhiyan has greatly energized the role of the panchayats in Maharashtra in matters related to management of schools while this is not the case in Orissa.

Introduction

The policy of inclusive growth adopted by India as part of the Eleventh Five Year Plan puts a lot of emphasis on reducing poverty and vulnerability and significantly increased government expenditures on programmes, many of which involve implementation via panchayats or in association with them. More specifically the central government has put in place two flagship programmes that both transfer income and reduce the vulnerability of households to shocks, namely MGNREGA and the Food Security Act. Under MGNREGA the panchayats select works and implement them, as well as participating in the targeting of the employment benefits. This has put control over large amounts of resources under the control of the panchayats. In the Food Security Act they participate in the targeting of beneficiaries and also have some oversight responsibilities.

Could panchayats play an even bigger role in reducing vulnerability to shocks via the use of own revenues or untied grants? Would unequal access to services, improper usage of programmes, programme capture, bribes, and inefficiencies in programme delivery, be reduced if they did so? These are the questions pursued in this chapter.

First the impact of ex ante vulnerability on ex post poverty dynamics is examined. Then the components of vulnerability are identified so that policy can be more focused. Finally, the impact of coping strategies on vulnerability and destitution and on economic welfare of households is assessed. More specifically the chapter focuses on five questions.

Does gender of the elected representatives matter, which we assess by measuring the impact of political reservations on reducing vulnerability?

• • What will be the impact of regime changes on resource reallocation and on household consumption? Are all types of regime changes (based on jati or gender) likely to have similar impacts on household vulnerability?

• • Do households fall into a poverty trap, i.e., experience chronic poverty, due to inefficiencies associated with governance?

• • Has local governance contributed to households undertaking less risky coping strategies in response to adverse shocks?

• • During periods of distress, are local governments able to provide insurance through their interventions?

• • After experiencing adverse shocks, is the insurance that the panchayats provide sufficient to eliminate the need for households to cut their food consumption to less than two meals, or to avoid selling of productive assets, and thereby reduce both vulnerability and poverty?

Introduction

In the previous chapter dealt with the impact of political reservations for women in panchayats. This impact is significant and more importantly, appears to be growing over time. There is another way to empower women. It has been shown that women who own assets are significantly empowered within their families and this leads to positive outcomes for both themselves and their family. Of particular interest is to see whether such empowerment gets reinforced by political empowerment.

But, the above quotes reflects wide-ranging prejudices against women owning property, educating the girl children and even panchayats being complicit in reinforcing such behavioural patterns. The relationship between social customs and practices germane to women's ownership of land and being cultivators, and panchayats is often not clear. In India there are often parallel amorphous institutions, panchayats, that regulate the social behaviour – mostly of women. This often includes restrictions on asset ownership, inheritance, education, etc by women.

In addition to political empowerment via reservations discussed in the previous chapter, women can also be empowered by providing them property rights or access to land. It has been shown that such measures significantly affect the life choices made by women (Deininger et al., 2013b). While political reservations lead to avenues for mobilization by women for either economic or political reasons, empowering women by providing access to property alters intra-household dynamics.

The last decade or so has been a rather tame time for analyses of democratic decentralization. Many new case studies have appeared, some of which – not least those that have focused on Africa – have been quite important, and some comparisons across continents have been valuable. But few major contributions to our understanding have emerged.

This book changes that. It is crucially important in several ways. It uses new methods to extract insights from a rich database – so that it offers us sophistication and rigour. It is a study of democratic decentralization, ‘warts and all’ – addressing key issues like corruption with admirable objectivity, and analysing things as they are and not as we might wish them to be. It does not seek to ‘sell’ or to condemn decentralization – it embraces ambiguity. It reminds us that local democracy is not a tidy process, so that we are bound to encounter great complexity. The highly credible, realistic analysis which emerges poses a formidable challenge to the pessimism in much of the literature about the impact of democratic decentralization on poverty.

That pessimism is mainly rooted in studies from Africa which stress the problem of ‘elite capture’ – whereby prosperous groups seize control of elected local councils and prevent resources from reaching poor people. One counterargument, which has by no means prevailed against the pessimists, emphasizes on an overlooked set of gains that poor people make when they engage with elected local councils. By participating, they enhance their ‘political capacity’ – i.e., their political awareness, their confidence as political actors, their political skills and their political connections to allies. Their capacity tends to increase even when they experience disappointment. Those who make this case argue that a severe shortage of political capacity is an important dimension of their ‘poverty’, alongside a severe shortage of incomes and assets.

Some of the findings in this new book resonate powerfully with that counterargument. The authors do not focus explicitly on political capacity, but they provide telling evidence of its constructive potential.

They demonstrate that despite complications and disappointments, participation – engagement with local democratic processes – can help poor people access resources and welfare programmes, reduce the probability of being vulnerable and improve the quality of local governance.