Using over 11,000 unlicensed loans to over 1,000 borrowers in Singapore, we provide basic information about an understudied market: illegal moneylending. Borrowers and lenders interact frequently and primarily rely on relational contracts to enforce their agreements. Borrowers have high discount rates, often have gambling and/or substance abuse problems, and often repay late. While lenders sometimes resort to nonfinancial punishments, the primary cost of late repayment is the compounding of a very high interest rate. Consistent with our view that lenders cannot extract all surplus, a crackdown on illegal lending raised interest rates and lowered the size of loans.

Although both economists and psychometricians typically treat them as interval scales, test scores are reported using ordinal scales. Using the Early Childhood Longitudinal Study (ECLS-K) and the Children of the National Longitudinal Survey of Youth (CNLSY), we examine how order-preserving scale transformations affect the evolution of the black-white reading test score gap from kindergarten entry through third grade. Plausible transformations reverse the growth of the gap in the CNLSY and greatly reduce it in the ECLS-K during the early school years. All growth from entry through first grade and a nontrivial proportion from first to third grade probably reflects scaling decisions.

We examine inference in panel data when the number of groups is small, as is typically the case for difference-in-differences estimation and when some variables are fixed within groups. In this case, standard asymptotics based on the number of groups going to infinity provide a poor approximation to the finite sample distribution. We show that in some cases the t -statistic is distributed as t and propose simple two-step estimators for these cases. We apply our analysis to two well-known papers. We confirm our theoretical analysis with Monte Carlo simulations.