Statistical Samplingto Measure Portfolio-at-Risk in Microfinance
Mark SchreinerSeptember 23, 2003Center for Social DevelopmentWashington University in St. LouisCampus Box 1196, One Brookings DriveSt. Louis, MO 63130-4899Tele. (314) 935-9778, Fax (314) 935-6618e-mail:
This paper describes a statistical sample design to measure portfolio-at-risk inmicrofinance. It applies the design to the microfinance portfolio of Banco do Nordeste inBrazil. Statistical audit sampling requires no special knowledge of statistics and isuseful for due-diligence inspections by possible creditors, possible owners, or inpreparation for the possible securitization of a portfolio. The sample design herestratifies by branch and by loan officer because errors in the record of arrears in themanagement-information system are likely to vary along these dimensions. Becauseerrors may also vary by loan size and are more costly for large loans than for smallloans, loans are sampled with probability proportional to size. This implicitly stratifiesthe sample by amount outstanding. Furthermore, the design samples all of the largestloans and all rescheduled loans. Given these strata, given a definition of portfolio-at-risk (for example, the outstanding balance of all loans with at least one payment atleast one day overdue), given a desired upper bound on the accuracy of the estimatedproportion of the portfolio-at-risk (for example, 2 percentage points), and given adesired precision for the confidence interval (for example, 90 percent), the paper tells(a) how many cases to draw; (b) how to estimate the proportion of the portfolio-at-risk;and (c) how to estimate the dollar amount of the portfolio-at-risk.
I thank Robert Christen, Todd Farrington, and Steve Schonberger.