One size fits all ? Female Headed Households , Access to Resources and Labor Supply to Farming in Kenya ∗

The goal of this paper is twofold as it aims (i) to investigate whether female headed households (FHH) can be distinguished by patterns of production and (ii) to look for the underlying causes of productivity differentials between FHH categories. Studies dealing with the question of productivity in FHH and male headed households (MHH) find that FHH are either less, equally, or more productive to MHH which is often explained by differences with respect to access to productive resources as land. To reconcile these findings, I propose a new way to analyze gender differences in farming that can be generalized to farm household characteristics other than the sex of the head of household. The basic idea is that both tenure insecurity and the risk of falling short of income, e.g. falling below the poverty line, affect the supply of labor, but they do so in opposite directions. That is, while tenure insecurity tends to decrease labor effort, income risks increase it as subsistence farmers want to avoid falling (deeper) into poverty. Depending on which of these risks prevails in the perception of farmers, they become more or less productive than a benchmark farmer who faces none of these constraints. The empirical results establish (i) that FHH are a very heterogeneous household category which need to be distinguished when analyzing issues of productivity and (ii) that FHH facing tenure insecurity are less productive compared to MHH. However, this result only obtains in case households are not at risk of falling into poverty. If they are, FHH increase labor supply to achieve a minimum level of income.