Estimating Intra-Household Bargaining Power When Outside Options Are Endogenous w/ Jean-Paul Chavas
Economists’ ability to empirically study household power dynamics is hampered by the need to make strong assumptions about partners’ next best alternative to the collective outcome. These assumptions lead to misspecification of the sharing rule. In this paper, we obviate the need for such assumptions, and specify a measure of the sharing rule without the corresponding misspecification error. We accomplish this by applying Harsanyi’s (1986) generic games framework to the limited-commitment collective model setting. We demonstrate uniqueness and existence of the solution by applying Rubinstein, Safra, and Tomson’s (1992) appeals-immunity solution concept. We demonstrate that the expected value of bargaining power is semi-parametrically identified in an ordinal framework. That is, the research need not assume specific utility or social welfare functional forms. In addition to empirical tractability and analytical precision, this update to the collective model provides a more nuanced view of intra-household bargaining power: partners with a greater
capacity to specify more damaging threats will have more control over the decision-making process, regardless of whether they act on those threats. This update brings to light a drastically different view of power than the standard perception in the family economics literature: power in the family is contested, and frequently updating in response to small environmental changes.