The Impact of Group Lending Strategies in Microfinance: Evidence from Rural Bangladesh

Presenter: Guillermo Bervejillo

Advisor: Prof. Joyce Chen (Food, Agriculture, and Enviromental Sciences)

Microfinance Institutions (MFIs) have become increasingly popular as instruments of poverty alleviation in developing countries. They can be found across the world, receiving a large amount of attention and praise for impacting the lives of poor entrepreneurs. These institutions grew even more popular with the rise of the "win-win" rhetoric in the 1990s, where, with help from a group of Ohio State economists, it was shown that MFIs need not rely on subsidies and donations to be an economically stable or even profitable industry. This paper seeks to question this paradigm, because though it has been shown that MFIs can be profitable, many are not convinced that profitability and poverty alleviation can truly coexist in the same development strategy. The cost of an economically viable MFI is the inability to offer unprofitable below market-level interest rates and the end of lending strategies such as joint liability. This paper will focus specifically on the impact of joint liability, a strategy that recently has become unpopular due to its low profitability, on the well being of participating households. The econometric evidence presented here shows that when compared to individualized lending strategies, group lending has a greater impact on well being as seen through indicators of investment in household assets and child’s educational support. These findings strengthen the claim that as microfinance has shifted its focus toward profitability it has lost some of its most effective tools in the fight against poverty, as illustrated by its move away from joint liability loans.

The data set utilized in this paper originates from a large household survey conducted by the Bangladesh Institute for Development Studies in collaboration with the World Bank during the 1990s. The Long Term Effects of Microcredit Survey is a two wave survey conducted first in 1991-92 and later in 1998-99 that interviewed close to 2,000 households in rural Bangladesh. Most of these households participated in some form of microfinance and a large portion of them listed group liability as a major form of collateral used for loans taken. Analysis of this data shows the causal relationships between household participation in group lending structures and increased investments in assets and human capital. To correct for the possibility of self-selection bias and to control for unobserved heterogeneities among households, the paper uses a fixed effects econometric model as well as a first difference approach. The proposed models find a positive relationship between group liability collateral and indicators of household assets, offering evidence of the households increased expected return on investments.