This paper reports estimates of the primitive parameters of a stochastic control model describing the labor decisions of West African small farmers. Maximum likelihood estimates are computed although optimal labor decision rules cannot be derived analytically. Vuong's non-nested model specification test shows that this method yields estimates superior to those derived by assuming that farmers solve a deterministic control problem. Low levels of labor effort are shown to be a consequence of low labor productivity in archaic rain-fed agriculture and of farmers' awareness that, in the absence of a labor market, overly ambitious production plans lead to seasonal manpower shortages.

Mazzocco and Saini (2012) propose and implement a test of efficient risk sharing that allows for preference heterogeneity. They motivate their approach as yielding different results from those of a standard efficiency test with homogeneous preferences. We show that the standard efficiency test results are misreported in their paper and that the correctly reported results do not present as compelling a case for the importance of accounting for heterogeneous preferences.

In this paper we examine whether risk sharing networks are formed so as to maximize the mutual gains from pooling income risk. The bene…t from risk pooling is largest when households have different income pro…les - e.g., different occupations - and are subjected to di¤erent sources of risk - e.g., live far apart. Gains from risk sharing therefore increase with social and geographical distance. But distance also raises the cost of interpersonal links. The net e¤ect on link formation is a priori indeterminate. We investigate this issue empirically using survey data from the rural Philippines.