Pricing Microfinance Loans and Loan Guarantees Using Biased Loan Write-off Data

This paper discusses how micro loans could be priced using data on loan write-offs. It states that:

• The Microfinance Information Exchange (MIX) has created an online information exchange containing relevant write-off data for many microfinance institutions (MFIs);
• However, because this data is self-reported and many MFIs do not report complete information, there is a substantial selection bias in the MIX market sample.

The paper provides a simple methodology for estimating the true distribution of loan returns using only observations that are self-reported by MFIs. It states that estimating true loan return distribution not only allows pricing of loan contracts, but also loan guarantees. The paper concludes that the methodology:

• Is simple and fast to implement for pricing micro loans and loan guarantees using publicly available data on loan write-offs by MFIs;
• Takes into account the selection bias inherent in available data - that MFIs that do not report loan write-off data are likely to be poor performers;
• Can be adapted to cases in which more specific information about a particular MFI is available.

The authors also state that:

• Their quantitative analysis suggests securitizations in which the upper tranches can be made completely safe;
• The riskier tranches can be held by informed intermediaries, donors and socially responsible investors;
• The safer tranches can be marketed to tap commercial investors;
• This can expand the supply of funds available for microfinance loans multi-fold making it easier to meet the Millennium Development Goals set by the United Nations.