Wages and Family Time Allocation - New version 29 August 2016
This paper examines how married people's allocation of time responds to wages and the gender wage gap. In the US, the gender wage gap has narrowed down by as much as 25% over the last three and a half decades. At the same time, women's labor supply has increased and, while couples spend less time on household work, men's share of household work has risen. I develop a life-cycle collective model for individuals in a household (spouses) who differ in preferences and bargaining power but share a common budget constraint. The spouses allocate their time across market work, home production and leisure, and they also decide about public consumption and savings. The model features lack of commitment to lifetime marriage and the gender wage gap can affect the spouses' bargaining power in the household. I estimate gender-specific preferences and how intra-family bargaining power responds to the gender gap using data on married and divorced individuals from the PSID. The results suggest that the narrowing gender wage gap improved women's bargaining power in the family resulting in a shift of household work from women to their husbands. The model is used to assess, counterfactually, the implications of gender wage equality for family time allocations. If the gender gap was eliminated altogether, the proportion of women in full-time market work would increase by up to 32% whereas couples' time into home production would decrease by up to 21% as women would reduce their household work by up to 7 hours per week.
Consumption Dynamics and Allocation in the Family - New version 7 September 2016
This paper studies how individual and total consumption in the family respond to idiosyncratic wage changes using a collective life-cycle model for a family of two decision-making spouses. The model incorporates endogenous family labor supply, public and private consumption, asset accumulation, correlated wage shocks, and general nonseparable, spouse-specific preferences. Wages enter the household budget constraint, but also the spouses' intra-family bargaining powers implying lack of spousal commitment to future allocations. I derive analytical expressions for the dynamics of earnings and consumption; I show how those can be used to identify the household structure (spouse-specific preferences, allocation of consumption between spouses, a rich set of bargaining effects) with panel data on hours, earnings, assets, and household-level consumption only. The identifying assumption is that spouses have the same preferences with their single counterparts. Preliminary evidence from the PSID suggests strong labor and consumption response to wage shocks and that hours and consumption are substitute goods at the intensive margin of labor supply. Wages have an economically significant effect on intra-family bargaining power, but not statistically so.
Consumption Inequality across Heterogeneous Families - New version 2 June 2017
This paper studies the transmission of wage shocks into consumption in two-earner unitary families that exhibit unobserved preference heterogeneity.
Heterogeneity is nonparametric and nonseparable from household preferences. I derive analytical expressions for any moment of the cross-sectional joint distribution of consumption and earnings relying on approximations as in Blundell, Pistaferri, and Saporta-Eksten (2016). These expressions enable the
decomposition of consumption inequality into distinct components pertaining to wage inequality, preference heterogeneity, and heterogeneity in initial
conditions. I establish identification of any moment of the cross-sectional distribution of policy-relevant parameters, such as consumption and labor supply elasticities with respect to wages, from panel data on consumption, hours, and earnings. Identification does not rely on any specific parametrization of household preferences or their distribution. To illustrate these points empirically, I fit second and third moments of consumption, earnings, and wages in the PSID. After controlling for a rich set of observables, I find: (1) substantial heterogeneity in consumption preferences but not in preferences over labor supply; (2) the distributions of permanent and transitory wage shocks are left-skewed implying that, on average, a negative wage shock is more unsettling than a positive one; (3) transitory wage shocks are, on average, fully smoothed but there is substantial heterogeneity in the ability families have to insulate consumption from such shocks. The findings are consistent with an environment where households are differentially affected by liquidity constraints.