The research program on rural finance has six main thrusts:
- Long-run impact of micro-credit programs
- Informal rural financial markets and their interaction with formal financial intermediaries, including micro-credit providers
- Impact of new product/design innovations in micro-credit programs using randomized placement and a fully experimental evaluation design
- Consequences of increasing competition among micro-finance providers on borrowers, the market for credit, and the sustainability of micro-credit
1. Long-run impacts of group-based micro-credit programs
Researcher: Shahidur Khandker
This study, conducted jointly by the World Bank and Bangladesh Institute of Development Studies (BIDS), examines the role of micro-credit programs on rural household welfare. It follows up on an earlier 1991/92 survey and study of the same households.
The 1991/92 study reported the short-term impacts of three major micro-credit programs on several household and individual welfare outcomes. Using a quasi-experimental survey design technique this cross-sectional study handled the two significant biases involved in the earlier impact analysis: non-random program placement bias and program participation heterogeneity bias.
The study investigates whether the benefits of micro-credit programs are sustainable or diminishing. A comprehensive analysis of the long-term impacts of the micro-credit programs looks at a number of household and individual outcomes (per capita consumption, non-land asset, employment, schooling, and fertility behavior). The analysis also investigates whether micro-credit program participation causes market saturation, whether the timing of credit matters, and whether credit and non-credit effects of micro-credit programs can be distinguished.
The analysis confirms the findings of the cross-sectional data analysis of the 1991/92 study. Micro-credit programs have positive effects on household welfare. However, the impacts have diminished over time, implying diminishing rates of return to borrowing.
The results suggest that micro-finance helps reduce poverty and has spillover effects that improve the welfare of the non-participants. Micro-credit also increases women's empowerment. Women with access to credit take a greater role in household decision-making, and have greater access to financial and economic resources, greater social networks, greater bargaining power vis-à-vis their husbands, and greater freedom of mobility.
2. Seasonality, hard-core poor and micro-finance in Bangladesh
Researcher: Shahid Khandker
This study, carried out in collaboration with the Bank’s South Asia Region SASPR and in partnership with the Palli Karma Shahayak Foundation’s (PKSF's) Institute of Micro-finance (InM), attempts to quantify the role of micro-finance interventions in mitigating seasonal deprivation and extreme poverty in Bangladesh.
Seasonal deprivation, locally known as monga, is common in the North West (NW) region of Bangladesh and is caused in part by seasonal lack of employment. The lack of consumption smoothing loans constrains many households from mitigating seasonality in consumption. PKSF, the country's wholesale MFI, partly funded by the World Bank, has launched a major monga mitigating program to eradicate seasonal deprivation.
The proposed research study plans to collect pre- and post-intervention household data in both program and non-program areas, and use propensity score matching and double-difference to assess the welfare impacts of alternative instruments introduced by PKSF through its partner organizations (POs) on household welfare. These instruments include cash for work, consumption smoothing loans, and production loans to improve households' coping ability to mitigate monga on a sustainable basis.
3. Informal rural financial markets and their interaction with formal financial intermediaries, including micro-credit providers
Researchers: Ghazala Mansuri (WB), Sanjay Jain (University of Virginia, Charlottesville), Xavier Gine (WB)
A traditional motivation for intervening in financial markets in rural areas has been to provide an effective alternative to informal lenders who have been conceived of either as extracting monopoly rents in certain market niches, or as being burdened by very high transactions costs resulting in high reported interest rates. While it has been clear for a while that the development of formal financial institutions has done little to displace the informal lender, no attempt has yet been made to systematically examine the effect of financial intervention via micro-finance programs on the size or structure of informal rural financial markets. Although such schemes are often conceived as creating competition with informal lenders, it cannot be taken for granted that increased MFI competition will necessarily reduce the role of informal credit, especially for the poorest borrowers. Interestingly, evidence from a number of recent studies suggests that households involved in micro-finance programs in Bangladesh continue to have substantial dealings with money lenders as well as other informal sources of credit.
Past work on this topic includes two applied theory papers . The first ("A Little at a Time") focuses on the interaction between informal credit providers and micro-finance institutions. The analysis suggests that it is quite possible for the informal credit market to expand, and for informal credit to become costlier, in environments where micro-finance has entered the market. The second ("Credit Layering in Rural Financial Markets") looks at the layering of credit in informal financial markets and its implication for the cost of credit to farm households.
New work, using a field experiment, builds on these studies. Specifically, the objective is to assess the impact of microfinance entry on the size and structure of the informal market. The study location is rural Pakistan where the informal credit market remains large and there are several distinct types of informal lenders. The study area includes 5 districts all over rural Pakistan where a microfinance institution (the National Rural Support Program (NRSP)) is active. In these districts, 150 villages were randomly selected from among the set of villages where NRSP was not active. Roughly 2/3rds of these villages were randomly allocated to “treatment” (i.e., NRSP starts work) and “control” (NRSP will not enter for 2 years). The baseline survey, for this study includes a census of all informal lenders in both control and treatment villages. The baseline is planned for December 2007. We expect to do the first follow-up in June 2009.
4. Impact of New Products/Design Innovations in Micro-Credit Program
Researchers: Ghazala Mansuri (WB), Xavier Gine (WB), Shahid Khandker (WB)
Two studies underway examine the impact of a specific intervention introduced by a micro credit granting institution:
Identifying Constraints to Asset Accumulation and Entrepreneurship Among the Poor
The Welfare Impact of the Thailand Village Fund
Identifying Constraints to Asset Accumulation and Entrepreneurship Among the Poor
Researchers: Ghazala Mansuri (DECRG) and Xavier Gine (DECRG)
Micro-credit is generally perceived as an important tool to reduce poverty. It remains unclear, however, whether credit alone is enough to lift people out of poverty. There are several reasons why credit alone may not suffice. First, micro loans may simply be too small to enable those with entrepreneurial ability to establish viable projects. Second, the poor may also lack sufficient information on viable business opportunities and the organizational skills to manage a profitable enterprise-thus lifting the liquidity constraint, even with larger loan sizes may not be effective without a significant enhancement of information on business opportunities and management. Finally, market risks for small producers may simply be too large to enable successful entrepreneurial activity for most.
The study design described below allows an assessment of each of these issues using a field experiment. This study is being done in rural Pakistan with the community organizations (COs) of a CDD program, the National Rural Support Program (NRSP), which also provides micro credit to its members.
Four program areas distributed across the country were first selected. Village branches in these four areas were then randomly allocated to either receive business management training or not. From each village branch, clusters of COs were then randomly selected for inclusion in the study. Therefore, all COs falling in one village branch were either assigned or receive training or not depending on whether they were located in training village branches or non-training village branches.
In addition, all eligible CO members in both training and non–training COs have the opportunity to participate in a loan lottery. The lottery is designed to induce a 50 percent chance of winning.
This study was initiated in the fall of 2006. The baseline was completed in early 2007. Business training was completed in September 2007. Orientations for the loan lottery are underway. We expect to do a follow up in April/May 2008.
The Welfare Impact of the Thailand Village Fund
Researcher: Shahid Khandker
The research study proposes to use panel household survey data from Thailand to assess the welfare impact of a country-wide rural credit scheme, known as the Thailand Village Fund. The goal of the program is to reduce poverty by mitigating the credit constraints of poor households and increasing rural non-farm employment.
Under this program a revolving fund of 1 million Baht is given to each village that meets a set of pre-specified conditions, including registration with the National Cooperative System. The study uses data collected in 2002 and 2004 by Thailand's national statistics organization (NSO). The sample for these surveys includes roughly 40,000 households. A panel of households was surveyed in both years in order to enable an evaluation of the impact of the Village Fund. In addition, the larger cross-sectional sample can also be used to study the some short-run effects. The main research question is: Does the Thailand Village Fund really matter in rural Thailand, given the presence of Bank for Agriculture and Agriculture Cooperatives (BAAC), another major rural credit scheme?
5. Consequences of increasing competition among microfinance institutions on borrower welfare, the market for credit, and the sustainability of credit institutions
Researcher: Ghazala Mansuri (WB)
Increased competition among MFIs raises two sets of important questions: one, what form does the competition take? Two, what are the implications of this competition, both for the financial soundness of the institutions themselves, and for borrowers? To take the first set of questions: when different MFIs operate simultaneously in the same geographic region, or ‘market’, to what extent is their relationship likely to be competitive, or co-operative? How much information should MFIs share? What kinds of information should they share? Do non-formal, trust-based mechanisms (reputation, repeated interaction, loyalty etc.) constrain strategic behavior by borrowers? Which aspects of program design are most crucial in this respect?
A theoretical paper examines the mechanisms MFIs can use to enforce contracts in an environment characterized by many lenders, when information on borrower histories is costly, the borrower pool is collateral poor, and no legal or financial infrastructure exists for contract enforcement. The analysis differs from earlier studies in that lenders in these models are profit-maximizers, whereas the objective of the MFI in our study is to achieve sustainability though maximizing market outreach. This affects the analysis considerably, and yields a sharply counter intuitive result: increasing competition can actually enhance sustainability by constraining the MFIs ability to engage in market outreach. The analysis also explores how the incentives for information-sharing are affected by the fact that non-profit MFIs’ objective functions are different from those of traders or profit-seeking banks. The model provides an interesting comparison with Padilla and Pagano’s model, under various information sharing regimes. Our analysis also identifies the conditions under which competition expands access to credit, or reduces the cost of credit, for the poorest households.
An empirical study, examines the dynamics of competition in local markets using data from 120 villages in Bangladesh where a complete census of microfinance providers was done. The objective is to answer the following types of questions: Is there a spatial pattern of MFI concentration? Does this lead to market saturation and excessive competition in some areas, and thin to non-existent services in others? What form is the strategic interaction among proximate MFIs likely to take? Are concerns about opportunistic default by participants in high concentration areas warranted? The survey, completed in 2005-06 was undertaken jointly by researchers at the Bank (Ghazala Mansuri (DECRG)), IRIS, University of Maryland (Thierry Van Bastelear) and the University of Virginia, Charlottesville (Sanjay Jain).
6. Group versus individual incentives in micro-finance
Researcher: Shahid Khandker (DECRG & WBI)
The role of groups in micro-finance has been debated for a long time. Grameen Bank has been the pioneer in introducing group mechanisms as a joint-liability scheme to reduce loan defaults. While anecdotal evidence suggests that groups matter in loan repayment behavior, no rigorous study has yet done to determine whether group incentives influence the loan repayment behavior of individual borrowers.
The study examines the role of group incentives in individual and group default behavior in Grameen Bank. Borrower-level panel data on loan repayment was used loan repayment behavior. The study attempts to show the endogenous group effects on individual loan repayment behavior after controlling for exogenous and correlated group effects.
The study proposes that both individual and group incentives are present in the loan repayment behavior of group members at Grameen Bank. But extent of causal impacts of endogenous group behavior in individual loan repayment behavior is unclear
The study demonstrates that the probability of loan repayment of a group member in Grameen Bank depends on the behavior of the group, even after controlling for the exogenous characteristics of the group or the assortative matching of individuals into groups.
|Policy Research Working Papers|
|For more policy research working papers from the World Bank's institutional archives, search here using author’s last name, title or working paper number (“wpsxxxx”).|
|WPS6821||Dynamic effects of microcredit in Bangladesh||Khandker, Shahidur R.; Samad, Hussain A.||2014/03|
|WPS6333||Does access to finance matter in microenterprise growth ? evidence from Bangladesh||Khandker, Shahidur R.; Samad, Hussain A.; Ali, Rubaba||2013/01|
|WPS6273||Replicating replication : due diligence in Roodman and Morduch's replication of Pitt and Khandker (1998)||Pitt, Mark M.; Khandker, Shahidur R.||2012/11|
|WPS6255||The surprising effects of the great recession : losers and winners in Thailand in 2008-2009||Haughton, Jonathan; Khandker, Shahidur R.||2012/11|
|WPS6204||Grameen bank lending : does group liability matter ?||Khandker, Shahidur R.||2012/09|
|WPS5331||Seasonal and extreme poverty in Bangladesh : evaluating an ultra-poor microfinance project||Khandker, Shahidur R.; Khalily, M. A. Baqui; Samad, Hussain A.||2010/06|
|WPS3826||Incomplete contracts and investment : a study of land tenancy in Pakistan||Jacoby, Hanan G.; Mansuri, Ghazala||2006/02|
|WPS2998||Does micro-credit empower women : evidence from Bangladesh||Pitt, Mark M.; Khandker, Shahidur R.; Cartwright, Jennifer||2003/03|
|WPS2945||Microfinance and poverty - evidence using panel data from Bangladesh||Khandker, Shahidur R.||2003/01|