Living arrangements and economic decision making in households and families

At least as much as any other institution, families can distribute resources among their members across time and space, spread risk, and foster cooperation. Accordingly, a central objective of modern applied microeconomics has been to characterize the family with sufficient sophistication. These efforts have drawn increasing attention to the critical economic distinction between households and families. The essays in this dissertation use micro-level data to empirically explore the ways in which the arrangement of the extended family into households can be understood as an economic choice.;One reason living arrangements are important is that they have important empirical implications relevant to analyzing all the other economic decisions of family members. An example of this is illustrated in the chapter 1. This chapter presents evidence that an arguably exogenous change in a public income transfer to South African blacks affected the selection of family members into households. Most importantly, the pattern of this selection was related to rarely-measured characteristics. Many previous studies which have exploited this same natural experiment have relied on the assumption that the income had no impact on such characteristics; the chapter discusses the potential implications of the finding that this assumption fails to hold.;Living arrangements are also important to economics in their own right, because they represent a mechanism by which families can improve their members' welfare. The second chapter uses new data from rural Mexico to explore how families use this mechanism to mitigate the effects of a widely discussed type of market imperfection. This imperfection relates to the fact that income risk is often shared through informal insurance arrangements. I present empirical evidence that families tailor their living arrangements to mitigate the welfare-reducing effects of incomplete information and nonbinding contracts in these arrangements.;Finally, motivated by evidence of endogenous selection of individuals into households, the third chapter reexamines a provocative recent finding in the literature. Using an alternative dataset which is not vulnerable to this problem of endogenous selection, my analysis calls into question the evidence of a causal link between Hepatitis B and population-level sex ratios at birth.