Essays on Household Finance and Health Economics

Abstract

This dissertation consists of three chapters. The first chapter studies why two out of three Americans claim Social Security benefits before reaching their Full Retirement Age and why even sufficiently rich people often claim early. I resolve this puzzling phenomenon by extending a standard incomplete markets life-cycle model to incorporate health dynamics and bequest motives. Relative to the existing literature, health plays a broader role, affecting medical expenses and mortality and directly the marginal utility of consumption. This role of health is disciplined using microdata on consumption, assets, income, and health from the Health and Retirement Study (HRS) and the Consumption and Activities Mail Survey (CAMS). The calibrated model successfully replicates the fraction of early claimers. Counterfactual exercises show that health-dependent preferences and bequest motives are crucial for this result. The model's success is explained by a novel channel that comes from the interaction between the negative effect of worsening health on the marginal utility of consumption, the downward health trend because of aging, and bequest motives. These two elements reduce the gains from delaying by 1) making individuals more impatient and 2) increasing the strength of bequest motives relative to future consumption. The second chapter is joint work with Christian Velasquez and Walter Ruelas-Huanca. This chapter explores the dynamics of mental health over the life cycle and assesses how sensible it is to approximate health by only considering physical health. Using data from the Panel Study of Income Dynamics (PSID) and approximating mental health with the presence of depression symptoms, we document seven facts about the evolution of mental health and contrast them with physical health. The results show two striking differences between mental health and physical health. First, while physical health consistently deteriorates with age, depression incidence follows a U-shape. Second, the likelihood of full recovery from physical health deficits decreases with age and is lower than the one for mental health, which exhibits a flat pattern. Finally, we propose and estimate a parsimonious statistical model for mental health that replicates these facts and can be easily incorporated into life cycle models. The third chapter is joint work with Tomas Rose and James Schmitz. This chapter studies the welfare implications of granting access to a standard mortgage-type credit market for financing affordable, factory-built homes. First, we briefly describe the legal regulations that have allegedly precluded low- and middle-income households in the US from accessing regular mortgage credit lines to finance the purchase of factory-built homes. We further document the current status of the credit market in the manufactured homes segment and highlight the predominance of loans featuring higher interest rates, shorter maturity, and absence of tax deductions (since some of these loans do not qualify legally as mortgages). We build a simple, dynamic, life-cycle model of housing decisions to quantify the welfare gains from changing these regulations. Using data from IPUMS (US Census Bureau), the PSID, and several other available sources, we calibrate our model to match the current home-ownership distribution at the bottom half of the US income distribution. Even at our most conservative exercise, in which we only allow for tax deductions at the factory-built homes credit segment (without modifying either the interest rate or the time to maturity), we find significant welfare gains that are equivalent to, on average, a permanent real income transfer of 6%, or to a present discounted life-time real income transfer of 94%.