Association for Public Policy Analysis & Management

As  the microfinance industry has grown both in level of financial assets and number of beneficiaries over the past three decades, more technical and administrative concerns have been arising. Many researchers have studied different features of microfinance products in both demand and supply side. Among those studies, focusing on behavioral economics of beneficiaries, who are the end-users of microfinance institutions, had a critical importance. Seeing the extent to which behavior clusters are correlated with beneficiaries’ characteristics opened numerous research questions. 

There are several surveys, gathered by international organizations, which include daily or weekly transaction data of families or individuals. Making a merged database of all these resources over different years and regions, presents a great opportunity to look at financial patterns of low-income people. Although different occupations, cultures, and environments restrict households’ financial management in different ways, there are still similarities among people’s economic behaviours. Indeed, most of the targeted groups are suffering from a lack of savings and a high risk of losing income. Thus, by applying clustering methods on detailed transaction data, we can achieve a better understanding of demand side of the microfinance market.

Among numerous criteria, which are defined in design of microfinance tools, this research is focusing on how to measure the financial risk of micro credits, based on income and expenditures of targeted groups. As microfinance institutions are facing potential defaults in their loan repayments, measuring those risks is a fundamental tool to improve their portfolio. Quantifying potential risks associated with specific individual characteristics, such as gender, age, occupation, marital status, education, and geographical location, provides the flexibility in design of micro credits. More specifically, seasonal effects on transaction data are in the spotlight in this study. By considering the mentioned approach, the research will provide recommendations for microfinance administrators to adjust their portfolio faced with predictable fluctuations in their assets.

Keywords: Micro credits, Risk management, Seasonality, Transaction data