Technology Adoption, Risk, and Intrahousehold Bargaining in Subsistence Agriculture

Leveraging unique data from four villages in Southern Ghana (Goldstein and Udry, 1999), we present and test a theoretical model where agents in a two-person agricultural household make individual portfolio allocations while looking forward to an intrahousehold bargaining process. The specific context is that of subsistence agriculture where individuals within households farm separate plots but also share resources. Still, our model allows connections to other contexts approaching the question of portfolio choice in general and can be extended as such. Motivated by empirical findings that unequal access to household resources by individual members impedes investment, we show how an institution of gender bias determines allocations between spouses and feeds back onto investment strategies, ultimately affecting policy efficacy and welfare outcomes over time. In doing so, we portray the household as a simultaneous site of cooperation and conflict and offer a theoretical framework that reflects various aspects of gender inequality in this setting. Specifically, as individuals have separate spheres of influence over the household asset base they adopt heterogeneous portfolio strategies and have relatedly different means of coping with negative economic shocks. In this way, individual risk management strategies diverge from that of the a representative agent and affect prospects for growth in a risky environment. As agent expectations over the returns from bargaining alter their investment strategies, and asset choices affect one’s position at the bargaining table, household members are consequently linked over time, placing the institution of gender at the center of the household’s technology adoption problem. Individuals exercise agency in maximizing their own objectives yet this agency is necessarily bounded by the social institution determining resource shares. In the end, egalitarian households are more able to adopt risky technologies and realize gains from related policies. While gender equality is obviously an end in itself, we ultimately show that it is also a relevant means for the complementary end of economic development.