Essays on Household Portfolio Choice
This dissertation consists of two essays on household portfolio choice. The first essay is entitled 'Household Portfolio Choice and Retirement'. In this first essay, I empirically examine the portfolio decisions of households as they transition into retirement. I document a novel stylized fact: holding household characteristics constant, singles maintain a relatively constant share of risky assets in their financial portfolios as they transition into retirement. On the other hand, couples decrease their share of risky assets significantly. I analyze this difference in behavior, and show that it is not driven by retirement-related background risks for couples relative to singles. Instead, I show that the heterogeneity within couples can be explained by the within-couple difference in spouses' individual risk aversion levels, and that the results are consistent with a net increase in couples' effective household-level risk aversion after retirement. Further, exploiting heterogeneity in couples' relative retirement dates, I show that husbands' and wives' respective retirement events are associated with very different (opposite-signed) persistent effects on the risky share of couples' portfolios. Moreover, I show that the relative magnitude of these persistent effects is consistent with the importance of each spouse's labor income within the household before retirement. Overall, the evidence is consistent with the outcome of a household bargaining game in which wives demand a smaller share of risky assets than their husbands, with each spouse losing some bargaining power after retiring.
In the second essay, entitled 'Household Bargaining and Asset Allocation', we empirically examine the effect of intra-household bargaining on household portfolio choice over the life cycle. We find that fluctuations in the distribution of intra-household bargaining power are associated with significant asset allocation shifts between risky and comparatively safer asset classes in households' portfolios. Our results are robust to alternative risky asset definitions, including investments in stocks, real estate, and holdings in private business, as well as to alternative control specifications. We find that the implied effect of intra-household bargaining is economically large in magnitude, with changes in bargaining power driving within-household variation in risky asset shares comparable to changes in labor income and wealth over the life cycle.
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.
Rights for Collection: Duke Dissertations