A Colombian welfare program made the formula used to assign benefits public to local officials in 1997. Compare the distribution of households just below (left) and above (right) of the eligibility threshhold in 1994 and in 2003. Seems an awful lot of households just barely qualify in 2003…
Kudos to Adriana Camacho and Emily Conover for the nice detective work in this paper: Manipulation of Social Program Eligibility: Detection, Explanations and Consequences for Empirical Research.
David Evans' personal blog 