An accurate understanding of how policies and programs affect individual and household welfare requires correctly modeling household decision making. This dissertation examines standard assumptions of the collective model that household members achieve efficiency (Essays 1 and 3), have perfect information (Essays 2 and 3), communicate, and make binding commitments (Essay 3). I analyze the theoretical implications of relaxing these assumptions and empirically assess how well these assumptions reflect reality in diverse contexts. The first essay calls into question the assumption that households in rural Ethiopia operate on the Pareto efficient frontier. The second essay examines differences in spouses' responses to questions regarding women's asset ownership and participation in household decisions in Nepal, and what these differences tell us about women's well-being. The third essay analyzes how introducing cellular network access in the Philippines affects wives’ control over household resources and information asymmetries between spouses.

Do Property Rights Affect the Efficiency and Intrahousehold Labor Allocations of Rural Ethiopian Households?

If the share of land that a husband or wife claims in divorce differs from his or her share of other assets, does this induce inefficient allocations of productive resources? To address this question, the first essay examines the effects of two policies that altered the distribution of property rights upon marital dissolution in Ethiopia: (1) joint land certification, which shifted land rights from husbands to both husbands and wives and (2) changes to regional Family Codes, which shifted non-land rights from husbands to a more equal division between spouses. Using two-way fixed effects, I analyze both panel and repeated cross-sectional data from rural Ethiopia. My results suggest that, when regional Family Codes are in place, joint land certification increases real household consumption per capita and the probability of being above the poverty line relative to households governed by head-only certification. Next, I examine whether changes in time allocated to off-farm wage labor explain the implied inefficiencies in household production under certain policy combinations.

Spousal Concordance in Joint and Separate Households: Survey Evidence from Nepal

In household surveys, husbands and wives who are asked the same set of survey questions often provide different responses. Using data from Nepal, the second essay studies patterns of concordance between spouses on survey questions regarding household asset ownership and decision making. We analyze these patterns separately for couples that reside with the husband’s parents and those that do not. We find that wives are much more likely than husbands to report their own participation in asset ownership and decision making, in both joint and separate households. In joint households, wives are also more likely to report that others own assets and make decisions. Wives reporting that they own assets or make decisions is correlated with some improved measures of wives’ well-being, regardless of whether there is concordance between spouses. Concordance is not necessarily correlated with better outcomes for women, particularly when the point of agreement is that the wife does not own assets or make household decisions.

Call on Me: The Impact of Communication on Intrahousehold Information Asymmetries in the Philippines

In the final essay, I develop a conceptual framework to understand how communication affects information asymmetries and monetary transfers between spouses in the Philippines. I posit that increasing spouses’ ability to communicate increases the cost of hiding or withholding income from one’s wife and reduces the cost of monitoring one’s husband. Using data from the first randomized control trial to provide communities with mobile networks access, I find suggestive evidence that this intervention increased transfers from husbands to wives by reducing information asymmetries between spouses. I also present experimental games that could be conducted with spouses in order to test the implications of the model.