Social Capital and enforcment of Informal Contracts

General Presentation

Over the past decade, many scholars suggest that social capital, defined as “features of social organization, such as trust, norms, and networks that can improve the efficiency of society by facilitating coordinated actions” (Putnam 1993: 167), contributes significantly to the alleviation of poverty. Thus, Robert Putnam (1993) showed that dense associational networks within civil society correlate positively with indicators of political democracy and economic growth. As the argument goes, agents interacting in networks and associations develop a framework of common values and beliefs. The trust that emerges from common understanding in turn generates norms of reciprocity that can help solving problems of ‘collective action’ (Olson, 1965). For instance, these norms can ensure that social sanctions are applied to punish non-collective behavior. In this sense, social capital can play a major role to correct state and market failures, and tends to be viewed by major actors in development, such as the World Bank, as a key element of anti-poverty policies. This idea underlies the recent insistence on development programs based on ‘micro-finance’ or ‘community driven development’.

            However, in spite of its intuitive appeal and its obvious relevance to the analysis of development processes in poor, developing economies, the concept of social capital has received relatively little theoretical attention. This is partly due to its conceptual vagueness, which lends itself to many different, and sometimes conflicting, interpretations. Moreover, social capital plays many different roles in economic exchanges: it serves in the creation of networks, in exchanging critical information, in shaping beliefs and norms supporting particular equilibria or in providing powerful sanctioning mechanisms. (For a recent survey, see Durlauf and Fafchamps, 2004).

            In the absence of formal contracts and sanctioning agencies, many economic exchanges are based on informal arrangements that cannot be enforced through courts or monitored by external parties. In this project, we study the role of social capital in generating social sanctions that agents can use to enforce informal arrangements. Typically, in the literature, social sanctions are posited parametrically, and are supposed to be used unilaterally whenever there is a breach in the ‘contract’ (see e.g. Besley and Coate, 1995; Anderson, Baland and Moene, 2007). They are thus conceived essentially as an instrument to sustain existing agreements. This argument however relies on unduly restrictive assumptions on the nature and the use of social sanctions. These assumptions are essential to many results in this literature but have never really been questioned or discussed. First, sanctions are assumed to be costless to the sanctioning agent. Indeed, if they are costly, it is hard to see how they can credibly be used to prevent defection and enforce existing arrangements. Second, the use of sanctions is restricted to the ‘cooperating’ members in order to punish the ‘defecting’ ones when they do not follow the norm. However, it should also be possible for the ‘defectors’ to use similar sanctions against those punishing them, making the threat against them less credible. Indeed, social sanctions are often bilateral: if some agents are excluded from economic interactions with others, they can also reciprocate by excluding others from interacting with them. We intend to go beyond the literature by investigating the sustainability of informal agreement when these two assumptions are relaxed.

To understand this better, we believe that it is crucial to properly model what constitutes a social sanction, so as to provide stronger micro-foundations to this idea. We therefore intend to first investigate the mechanisms through which social capital, in the sense of ‘dense’ interdependence between agents, ‘generates’ social sanctions. In a second step, we will explore the impact of social sanctions in sustaining existing agreements. In the process, we will call into question the prevailing view that social sanctions necessarily help agreements. Indeed, they can also be used to force agents to renege on their obligations if such defection is beneficial to their particular group. To give an example, in the context of micro-credit, group members can use social sanctions to enforce repayment of the loan to the bank but, with appropriate norms and beliefs, they can also use the same sanctions to enforce collective default against the lending agency (for more details, see below).

We shall investigate this issue both at a theoretical and at an empirical level. At the theoretical level, we explore the role of social sanctions in micro-credit groups, in collective action problems and in informal insurance arrangements. We will develop an approach in which:

(1) social sanctions are explicitly derived from social interactions (‘social capital’) between members of a given community (e.g., under the form of a collaboration to a collective activity);

(2) social sanctions may be costly and cannot therefore always be credibly used to sustain existing agreements;

(3) social sanctions are reciprocal, so that nothing prohibits the defecting member to retaliate against other members of the group;

(4) social sanctions can also be used by the group to sustain other behaviors than the one specified in the original arrangement, and this depends on the beliefs and the norms prevailing in the group.

In this perspective, the theoretical effort we intend to pursue in this project defines a completely new approach to the role of social sanctions in informal agreement.[1] In the context of microcredit, contracts which hold a group of borrowers jointly liable for repayment could improve welfare by allowing banks to offer greater loan sizes or lower interest rates than would be profitable with individual contracts. As a first step, we will consider a problem of ex-post moral hazard (repayment incentives). Our main interest in this work is to understand:

 (1) how groups interactions generate the types of sanctions used to enforce group lending contracts. To this end, we will explicitly introduce social capital under the forms of voluntary participation to a collective activity, separate from micro-credit. Social sanctions can then take the form of excluding an agent from participating in that collective activity.

(2) What are the circumstances under which collective default, instead of collective repayment, emerges as an equilibrium strategy? In particular, we shall focus on the cases where the same processes of social sanctions and exclusion that made possible group repayment are at work. The choice between different equilibria then follows from the prevailing norms, beliefs and attitudes at work in the group.

(3) More generally, while social networks and non-market institutions have often been shown to facilitate the functioning of markets, the effects of markets on these types of institutions have been little studied. While banks might find the existence of these social sanctions useful because they provide borrowers with improved incentives for repayment, groups may also use the behavior of group members towards the bank to infer their interest or productivity in the group collective activities. Both sanctions and group contracts are therefore part of a ‘social equilibrium’ and the relationship between these market and non-market structures is of primary interest: while social networks do bring the poor within ambit of formal credit institutions, group lending contracts can also be used by communities to sort households and exclude some of them from participating in collective activities. We shall therefore explore how temporarily greater access to credit for some households may in fact result in their exclusion from social networks and community level public goods.

In a separate line of work, we intend to investigate the potential of social sanctions to sustain behaviors that are more conducive to collectively beneficial output. In the past, we have extensively studied the relation between inequality and collective action, where differently endowed agents contribute in a decentralized way to a collective good (see e.g., Baland and Platteau, 2003). We wish to pursue this issue further by exploring how, depending on the norms prevailing in the group, beliefs about an agent’s general value to the group can be formed depending on his contribution. To this end, we shall start by developing a game with a sequence of two decisions, so that information gathered in the first game about the agents can be used for punishment by collective sanctioning in the second game.

            In the context of risk-sharing arrangements, we intend to focus on two dimensions. First, we will again investigate the role of social sanctions in sustaining risk-sharing programs. Agents participate to risk-sharing agreements, but also in other collective activities, so that sanctions may involve their exclusion from this other set of activities, and provide an additional tool to enforce risk-sharing agreements. As we discussed in the context of micro-credit, an agent’s behavior in one sphere produces information that is used in the other sphere: for instance, an agent’s refusal to provide help may be interpreted as a signal of low interest not only in the risk-sharing agreement, but also in the other collective activities.

Also, in heterogeneous communities, the possibility of separate collective activities leading to distinct risk-sharing networks will be explored. The roles of norms will be particularly important there, as a generalized risk sharing network may be viable if the norm is that one should help each other, irrespective of which groups he belongs to (a view closely related to the notion of ‘generalized morality’ developed by Granovetter), while it may not be under the alternative norm that one should help members of his own community but not others (referring to the norm of ‘restricted morality’). We wish to investigate the emergence of such norms and beliefs, and their implications for the sustainability of risk-sharing arrangements.

We shall also carry out two empirical projects based on original data sets. The first one is on microfinance groups in India, where we have collected a completely unique data set which tracks micro-credit groups and their participants over ten years, and intend to understand better the dynamics of social sanctions, enforcement and exclusion within groups, as well as the determinants of the group survival and success. This empirical effort is unique in that (1) there are very few studies (and none in the domain of ‘self-help’ groups or microfinance) that followed a large number of groups over time, collecting information on all major events of the group, and followed all members, including those who have left the group, with information on their economic evolution; (2) the data set we collect is very detailed in social background information (including the detailed subcast –‘jati’– of each member) as well as family relationships between members (for each group, we have recorded a family network matrix: for each member, we know all the members in the group with whom he has a family connection and the nature of that connection). This data set will be very useful to understand the role of social capital and the dynamics of social exclusion in micro-credit.

The other empirical project aims to investigate solidarity arrangements in Cameroon, the pressures for interpersonal redistribution of wealth, and the strategies used to avoid these pressures or escape the potential social sanctions. In spite of its prevalence in Africa and its obvious implications for poverty, inequality and growth, this phenomenon and its implications have never been studied in a systematic way. This project originates from a small survey we carried out last year in Cameroon. In rural areas of Cameroon, members of the credit cooperatives we investigated engage in a puzzling borrowing behavior: more than 20% of loans are fully collateralized by liquid saving available on the borrowers’ saving account.

From a number of interviews we carried out there, it appears that members of those credit cooperatives who took loans fully collateralized by their savings, claim that they use loan repayments as an excuse to oppose demands for transfers made by friends or family. One member summarizes it as follows: “When I take a loan from my savings, my children and my wife think I have no money. I do it on purpose. If I simply withdraw my money, it will end, so I tell them that I borrowed. Then, when one complains he has a problem, I say I have to pay back my loan. It protects me from my children’s demands.” (for more details, see Baland et al, 2007).

Unfortunately, we have not been able to collect a systematic data set, and our explanation relies on a small number interviews. As a result, we intend to conduct an empirical analysis of the saving and borrowing behaviour of members of credit cooperative in Cameroon. The specific objective of our empirical analysis is threefold. First, we want to identify the nature of individuals’ impediments to save and the strategies used to protect savings. We will pay special attention to the pressures individuals may face to donate money to their relatives and the social sanctions they may face if they do not abide by their obligations. We will also explore potential problems of self-control limiting individuals’ ability to save. Second we will verify whether individual behaviour is consistent with the use of credit as a signal of poverty. In particular, we wish to investigate the degree to which savings are kept secret while debts are advertised to relatives, and whether the information about loans effectively discourages requests for cash transfers. Finally, we want to propose a more formal test of the hypothesis that credit is used as a signal for poverty to oppose demands made by relatives. On the one hand we will evaluate whether, ceteris paribus, an individual who faces more pressure for redistribution or more possible sanctions is more likely to take a loan rather than withdraw his money from his saving account. On the other hand, we will verify whether the same individual is more likely to engage in that behaviour in times where he suffers from greater social pressure. To this end, we shall carry out a large household survey in Cameroon.

Current Outcomes and Work in Progress: October 2011

Project 1: Social Capital and Micro-credit

1. Jean-Marie Baland, Siwan Anderson and Karl-Ove Moene, 'Sustainability and organizational design in informal groups, with some evidence from Kenyan Roscas', Journal of Development Economics, 2009, 90(1), 14-23; BREAD WP.
2. Jean-Marie Baland, Rohini Somanathan and Zaki Wahhaj: ‘Repayment incentives and the distribution of gains from group lending’, CEPR WP6520 and BREAD WP, 2011, submitted to the Journal of Development Economics.
3. Jean-Marie Baland, Rohini Somanathan and Zaki Wahhaj: ‘Social sanctions and repayment incentives in group lending’, working paper, 2011.
4. Jean-Marie Baland, Lata Gangadharan, Pushkar Maitra and Rohini Somanathan: ‘Social Exclusion in a Laboratory Microfinance Experiment’, working paper, 2011.
5. Timothee Demont: theory paper, working paper, 2011.

Project 2: Social Capital and mutual insurance

1. Jean-Marie Baland, Karl Ove Moene: ‘Informal and Formal Social Security in the presence of moral hazard’, work in progress, 2011.
2. Wouter Gelade: ‘Social Sanctions and Attitude to risks in mutual insurance networks’, work in progress, 2011.

Project 3: Social Capital and Collective Action

1. Jean-Marie Baland, Sanghamitra Das and Dilip Mookherjee, ‘Poverty, Collective Action and Forest Degradation in Northern India and Nepal’, Environment and Development Economics: Essays in Honour of Sir Partha Dasgupta, Oxford University Press, forthcoming.
2. Jean-Marie Baland, P. Bardhan, S. Das, and D. Mookherjee: ‘Forests to the People: Decentralization and Forest Degradation in the Indian Himalayas’, World Development, 2010, vol. 38(11), 1642-56.
3. Jean-Marie Baland, P. Bardhan, S. Das, D. Mookherjee and R. Sarkar: 'The environmental Impact of Poverty: Evidence from firewood collection in Rural Nepal', Economic Development and Cultural Change, 2010, vol. 59(1), pages 23-61.
4. Jean-Marie Baland, Francois Libois and Dilip Mookherjee: ‘Evolving Patterns of Firewood Collections in Nepal: A Household Panel Analysis 1995-2003’, working paper, 2011.
5. Jean-Marie Baland and Francois Libois: ‘The impact of civil conflicts in Nepal’, work in progress, 2011.
6. Jean-Marie Baland, Marc Bellemare and Subrendhu Pattanayak: ‘Leaders and Followers in the adoption of private latrines in Indonesia’, work in progress, 2011.
7. Isabelle Bonjean, Jean-Philippe Platteau and Vincenzo Verardi: Innovation Adoption…
8. Petros Sekeris: 3 projects: deluca vargas; gift exchange; Congo RDC.

Project 4: Microfinance in India

1. Jean-Marie Baland, Rohini Somanathan and Lore Vandewalle, ‘Socially disadvantaged groups and microfinance in India’, working paper, 2011.
2. Lore Vandewalle, ‘The Role of Accountants in Indian Microfinance Groups: A Trade-off between Financial Benefits and Mutual Assistance’, Working paper, 2011
3. Paolo Casini and Lore Vandewalle, ‘Public Good Provision in Indian Rural Areas: the Returns to Collective Action by Microfinance Groups’, Working paper, 2011.
4. Jean-Marie Baland, Timothée Demont, Rohini Somanathan and Michel Tenikué, ‘School choices and credit constraints for microfinance participants in India’, work in progress, 2011.
5. Timothee Demont: ‘Poverty, Access to credit and Absorption of Income Shocks’, work in progress, 2011.

Project 5: Solidarity Arrangements in Cameroon

1. Jean-Marie Baland, Catherine Guirkinger and Charlotte Mali: ‘Pretending to be poor: borrowing to escape forced solidarity in credit cooperatives in Cameroon’, Economic Development and Cultural Change, 2011, vol. 60(1), 1-16.
2. Jean-Marie Baland, Isabelle Bonjean, Catherine Guirkinger and Roberta Ziparo: ‘Family pressures and investment incentives in Cameroon’, work in progress, 2011.
3. Jean-Marie Baland, Isabelle Bonjean, Catherine Guirkinger and Roberta Ziparo: ‘Intra-Household transfers and solidarity arrangements in Cameroon’ , work in progress, 2011.

Other related work

1. Jean-Marie Baland, K. Moene and J. Robinson, ‘Governance and Development’, in D. Rodrik and M. Rosenzweig (eds), The Handbook of Development Economics Volume 5, Elsevier, North-Holland, Chapter 69, pp. 4597-4656, 2011.
2. Jean-Marie Baland and J.-P. Platteau: "Institutions, Property Rights and Development", in A.K. Dutt and J. Ros (Eds), The International Handbook of Development Economics, Vol. 2, Edward Elgar, Cheltenham, UK and Northampton, USA, pp. 394-406, 2009.
3. Jean-Marie Baland, Cedric Duprez: ‘Are labels effective against child labor?’, Journal of Public Economics, 2009, 93, 1125-30; CEPR WP4542 and BREAD WP.
4. Jean-Marie Baland and Jim Robinson: The political value of land: political reforms and land prices in Chile, CEPR working paper, 2010, Revised and Resubmitted to the American Journal of Political Science.
5. Jean-Marie Baland and Kjetil Bjorvatn: ‘The distributive impact of privatization of common property in a dynamic perspective’, submitted to the Journal of Public Economics.

1. Jean-Marie Baland, Cedric Duprez: ‘Social labeling, labor standards and fair trade’, working paper, 2010.
2. Jean-Marie Baland and Francois Libois: ‘The Poor and the Commons: the case of the Indian Himalayas’, work in progress, 2011.