Share

Economics Discussion Paper Series EDP-1019, University of Manchester, Manchester, UK. Imai, Katsushi S., and Md. Shafiul Azam. 2010

An empirical analysis of microfinance in rural Bangladesh finds a positive impact on poverty reduction only for households who use the loans for productive purposes, suggesting that microfinance lenders should to monitor loan usage in order to meet poverty reduction goals.

The microfinance “revolution” spread a model for the poor to collectively access credit to finance micro-enterprise, allowing them to overcome a lack of collateral, the cumbersome procedures of banking, and the bias toward urban banking networks. A multitude of anecdotes and empirical evidence supports the positive impact of microfinance on poverty. However, after three decades, concern is rising over the supply of microcredit; are there too many inefficient microenterprises? Has income risen enough to provide a market for all of the new goods produced? The questions beg a reassessment of the relationship between microfinance and poverty.

The paper uses a nationally-representative panel data set and enhanced empirical techniques to study the relationship between microfinance borrowing and poverty reduction in rural Bangladesh. In addition, the authors establish a dichotomy of loans; those used for productive purposes, such as small businesses, and general loans that include non-productive purposes such as dowries. The empirical strategy measures per capita household income for 3000 survey respondents.

The data indicates, firstly, who participates in microfinance. Women are found to be most likely to obtain credit from microfinance institutions (MFIs), which is to be expected since they are a target group. Larger households and households with older heads are also more likely to have a member seek a micro-loan. Finally, proximity to the nearest market (upzilla) and employment of the head of household in the rural nonfarm economy will also increase the likelihood that a micro-loan will be sought.

Secondly, the results describe how household income levels vary in rural Bangladesh. Households headed by women tend to have much lower per capita income, representative of the economic disadvantages they experienced. Higher levels of education and employment as a professional or entrepreneur also lead to higher per capita household income. Additionally, the data indicates that income levels rose between 1997 and 2005, and the result holds after considering inflation.

The main contribution of this paper is the finding that general loans do not have an impact on household income, and thus on levels of poverty. On the other hand, loans for productive purposes are found to increase income, and thus decrease poverty. The data also indicates that the poverty reducing effect of loans has deteriorated over time. In 1998, for example, there is significant evidence that microfinance loans reduced poverty, irrespective of loan purpose. By 2008 those effects have become non-significant. Better targeting of microfinance clients would have been a way of avoiding these results.

The authors point to the supply side of microfinance to provide reasoning for the study’s findings. A larger number of competitors paired with a greater commercialization may have reduced the selectivity of MFIs when accepting clients. A higher level of monitoring and greater attention to the purpose of the loan is thus recommended in order to achieve goals of poverty reduction.