Household Decision Making with Violence: Implications for Transfer Programs

In this paper I study how intimate partner violence responds to transfers to women, and whether this response depends on the transfer being in-kind or in-cash. To this end, I develop and estimate a model of household decision... [ view full abstract ]

In this paper I study how intimate partner violence responds to transfers to women, and whether this response depends on the transfer being in-kind or in-cash. To this end, I develop and estimate a model of household decision making in the presence of in-kind and cash transfers, in which the weights of the husband and the wife in the household utility are endogenously determined through violence. Only men can inflict violence to increase their relative weight, but violence destroys female labor productivity. Under this framework, the utility gains the husband can appropriate through violence are lower when the transfers are in-kind. As a result, in-kind and cash transfers have different effects.  I estimate the model using data from a randomized controlled trial which provides  transfers to poor families in Ecuador, either in-kind or cash. The results indicate that, on average, violence destroys 4% of female productivity with a market value of 10 dollars a month. Violence also reduces the female relative weight in the household utility by 12%. A cash transfer equivalent to 10% of the average household  income would reduce the prevalence of violence from 17% to 10%.  If the the same transfer were in-kind, violence would decline by 3 additional percentage points. This differential effect amplifies with the size of the transfer.