The private sector meets randomized evaluations

For nearly a decade now, the gold standard in international development in the public sector has been the randomized evalution. There are numerous examples, perhaps the most famous being that of Mexico's PROGRESA program. Until recently financial and private sector development were more or less ignored or assumed unfit for this type of evaluation - I suspect the (mistaken) assumption was that survival in the marketplace was enough of an evaluation.

However, a new newsletter called FPD Impact promises to rectify this oversight. And it looks like new evaluation work on private sector development is bringing to light some very provocative findings. The first edition looks at Which Microenterprises have High Returns to Capital?. David McKenzie, the author, examines experiments where microenterprises in Sri Lanka and Mexico were randomly selected to receive grants of between US$100 to $200:

• Real returns per month averaged 5.7 percent in Sri Lanka and between 20 to 33 percent in Mexico - higher than market interest rates. 
• In Sri Lanka, "the average male-owned enterprise has very high returns to capital, on average female-owned enterprises saw no gain in profit from the grants." This will surely raise eyebrows, as much of the microfinance world has focused on women, in part because they were assumed to be more credit constrained. McKenzie suggests that more focus should be given to poor urban males.
• Perhaps most interesting - and hopeful - of all, microenterprises are not poverty traps: "Our results suggest high returns from relatively small amounts. Firms can therefore start small and grow." This seems to run at least partly contrary to the views of Abhijit Banerjee, one of the heavyweights in the world of randomized evaluation.

>> Download a copy of FPD Impact.