Understanding the Psychology of Mass Default: A Case Study of Eastern MaharashtraPrincipal Investigators: Lisa C. Nestor, Deeptha Umapathy
Research Team: Lisa C. Nestor, Sachin Srivastava
LEAD Centre: Centre for Microfinance
Focus Area: Credit
Project Geography: Maharashtra, India
The microfinance industry has witnessed rapid growth over the last 30 years. India in particular has demonstrated an intensifying demand for microfinance services with the number of domestic microfinance institutions (MFIs) rising from 10 in the year 2001 to 90 by 2010, while the number of active borrowers increased from approximately 1 million in 2001 to an estimated 30.7 million today (mixmarket.org).
Generally, the industry has proved incredibly resilient, pushing past the challenges of high transaction costs, poor infrastructure (particularly in rural locations) and low financial literacy. However, microfinance, both globally and in India, continues to face challenges to its stability and long-term sustainability. In particular, the industry has witnessed a series of mass community defaults, such as the Andhra Pradesh ‘Krishna Crisis’ of 2006, the Karnataka ‘Kolar Crisis’ of 2009 and more recently the Maharashtra ‘Marathwada Crisis’ of 2010/2011.
In each case, community-driven resistance has encouraged historically responsible borrowers to partake in a mass default on their current loans. Although these movements are normally small in size, they have consistently proved difficult to curtail and contain, often spreading to surrounding communities and resulting in a permanent state of non-repayment. These community defaults have both short-term and long-term consequences for the MFIs and local communities, impacting local access to finance, trust in the market and the operating capacity and liquidity of the regional MFIs.
In an effort to better understand the cause and motivation behind these community- based loan defaults the Centre for Micro Finance, in association with RBI-CAB and MFIN, has initiated an independent study of the recent community defaults in the Marathwada districts of Eastern Maharashtra. Using quantitative data gathered from local MFI branch offices and various qualitative stakeholder interviews, the study will investigate the nature (i.e. pace of growth, spread, etc) of the defaults as well as identify both primary and secondary causes of individual and group loan defaults. In particular, the study aims to model the incentive structure and decision-making model behind individual and group loan defaults.
Methodology and Research Design
The Maharashtra Default study will test the hypothesis that defaults tend to occur at the Centre level) and examine the effect of current loan cycle (week) at the time of the community shock on default rates(Breza, 2010). This will be examined through a combination of individual client interviews, facilitated group discussions, credit officer interviews and branch level data gathered from partner MFIs.
The study follows a non-randomized research design, investigating client behaviour through a series of ‘post-event’ interviews. However, the sample will be drawn using targeted random selection. This is achieved by randomly eliciting group and client participation in targeted communities which have been pre-identified as having high or low rates of loan default. Sample selection will also follow a 3:1 ratio for defaulting vs. non-defaulting clients.
The total research sample will comprise approximately 120 client interviews (90 default: 30 non-default) and approximately 50 facilitated group discussions. The individual client interview participants will be selected from the group discussion participants, with two to three candidates selected from each group discussion. Researchers will also conduct 8-12 credit officer interviews and 6-7 MFI branch manager interviews. These qualitative interviews will then be combined with branch level portfolio data from 3 MFIs. This macro-level data will compliment the stakeholder interviews by providing a broad perspective on the timing, pace and breadth of the community defaults.
From piloting exercises, researchers have found that there is strong local demand for microfinance loans within both defaulting and non-defaulting communities.
It also appears that there is no singular cause for default, even within the closely-linked local communities. Researchers think it is more likely that a several minor causes for default converged once the momentum of the community default began.
Lastly, defaulting clients do not seem to target specific MFIs. Once clients decided that they would default, they generally defaulted on all current loans, not loans administered by a specific MFI.
Research and preparation for the Maharashtra default study began in late August, 2011. Field work began in early November 2011 and will be complete by the first week of December 2011. Initial data analysis is scheduled to be completed by the end of 2011.
Researchers hope to present their findings at industry seminars and conferences throughout early 2012.