The Income Elasticity for Nutrition : Evidence from Unconditional Cash Transfers in Kenya

We use a randomized controlled trial to study the effect of large income changes, through unconditional cash transfers, on the food share of expenditures and consumption of calories among poor households in rural Kenya. Our preferred estimate of the food elasticity following USD 709 transfers is 0.78 for expenditure, 0.60 for calories, and 1.29 for protein. Experimental elasticities are lower than cross-sectional estimates. These estimates are unaffected by spillovers or price changes at the village level: results are similar with vs. without an almost ideal demand system, and with a control group in treatment vs. control villages