Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper proposes an approach for ex-ante welfare analysis under in-kind transfers based on demand system estimation and construction of virtual prices (i.e., the prices at which beneficiaries would be as well off as with the subsidy). As an application of this approach, I study a program which randomly transferred either a food basket, procured at a discount in wholesale markets, or an equal-cost cash transfer to poor households in rural Mexico. I find large welfare-destroying distortions from the food basket, corresponding on average to 35% of its face value. Efficiency losses are larger for more vulnerable recipients, as the transfer provided relatively large allotments of commodities consumed in lower shares by the poor. Counterfactual simulations providing better-targeted food transfers suggest that average welfare gains could be enhanced by up to 38% as compared to the cash subsidy, while also reducing the regressivity of the food transfer.