Essays on collective models with domestic production

In his original framework, Chiappori (1988, 1992) represents the household as a collectivity consisting of two persons, each of them being characterized by a specific utility function, that make Pareto-efficient decisions. The empirical consequences of the efficiency assumption are then analyzed in a household labor supply model, in which only total consumption and individual labor supplies are observable by the outside econometricians. Chiappori shows that, if preferences are of the egoistic type and consumption is purely private, the rule that describes how household resources are shared between spouses can be identified (up to a constant) from the sole estimation of spouses' labor supply functions. Furthermore, Chiappori et al. (2002) introduce the notion of distribution factors, defined as variables that influence the sharing of resources within the household without affecting preferences or the budget constraint, and thereby complete the preceding identification result. The initial formulation of the collective model of labor supply has been extended to consider the domestic production (Apps and Rees 1997 and Chiappori 1997) and non-participation in the labor market (Donni 2003 and Blundell et al. 2007). The PhD thesis is organized in three chapters. In the first chapter, we consider the collective model of labor supply with (marketable) domestic production. Our objectives are threefold. Firstly, we prove a new identification result for the collective model of labor supply with marketable domestic production for the case where no distribution factor is observable. We show that, even in that case, the sharing rule can be identified (up to a constant). This theoretical result, as simple as it may be, is original and important because finding (exogenous) distribution factors is always a difficult task for the econometricians. Secondly, the theoretical model is estimated using the ATUS data the American Time Use Survey (ATUS). We estimate a complete system of market and domestic labor supply equations by the SUR (Zellner, 1962) and the 3SLS (Zellner and Theil, 1962) methods, the latter method is to account for the probable endogeneity of wage rates. The selection of working couples on which estimations are performed is modeled using the Heckman method. Thirdly, we study the distortions generated by the omission of domestic production from an empirical perspective. Our results connect the discrepancies that may exist between the structural estimates of the collective models with and without household production to the properties of the domestic labor supply functions. The second chapter develops a theoretical model of labor supply with domestic production that is consistent with corner solutions. The empirical applications of collective models accounting for household production do not incorporate corner solutions. The exploited sample is restricted to the sole working couples, and the sample selection is taken into account or not. The aim of this paper is to model the decision to participate in the labor market in a collective framework with domestic production. Our results are threefold. Firstly, we describe how the decision process can be decentralized when one spouse or both spouses do not work. In the case where one spouse does not participate in the labor market, her market wage does no longer coincide with the price of leisure. For non-working spouse, the price of leisure is endogenously determined within the household and it is defined by the marginal rate of substitution between leisure and consumption computed at the corner point. Secondly, we prove that the sharing rule can be identified (up to a constant) from the observation of domestic and market labor supply functions. The information contained in domestic labor supply functions allows identifying the sharing rule even when both spouse does not work in the market. We also demonstrate that these results can be obtained without distribution factors, thereby generalizing the initial results of Apps and Rees (1997) and Chiappori(1997). Thirdly, we estimate, using the American Time Use Survey (ATUS) data, a collective model of labor supply accounting both for domestic production and for female non-participation. We estimate, by maximum likelihood, a system of three structural equations (family domestic labor supply, and husband and wife market labor supplies), considering two different regimes, that is the wife participate or not participate to the labor market. The wife's labor force participation is based on whether her desired hours of work are greater or less than zero. In the non participation case, the wife's wage, entering in the family domestic labor supply and in the husband market labor supply, is replaced by her price of leisure. The third chapter develops an equilibrium model for the household that accounts for a marketable household production, a non-marketable domestic production and non-participation in the labor market. The aim of this chapter is to provide analysts and policy makers with an instrument that can be useful to evaluate the impact of social and economic policies on individual behavior and welfare. Our objectives are threefold. Firstly, we describe the household as engaged both in a marketable and in a non-marketable domestic production. We allow also for non-participation in the labor market. The marketable produced good is entirely sold in the market at an exogenous price. The non-marketable produced good is entirely consumed by the household members and its price is endogenously determined within the household. When the household members do not participate in the labor market all their time-uses are evaluated at an implicit wage defined by the individual's marginal labor productivity. Secondly, we illustrate how the equilibrium model of the household can be used by policy makers to evaluate the impact of policies on individual behavior and welfare. In doing so, a social accounting matrix of the average Italian farm-household economy is developed using Italian ISMEA data. The household social accounting matrix describes the flows of all economics transactions that take place within the household economy. ISMEA data were also used by Menon and Perali (2009) to estimate the model presented in this chapter. The HSAM together with the production and consumption parameters estimated in Menon and Perali are used to calibrate the equilibrium model that reproduces the structural specification of the econometric model. The need for calibration is reduced to a minimum limited to the calibration of the intercepts of demand and production shares to match the levels of the HSAM. Thirdly, we use the programming model to perform simulations and policy experiments and predict the impact of changes in exogenous variables on production, consumption and labor supply decisions. Policies analyzed are the impact of an increase of the price of a crop, a decrease in the husband’s non-labor income and the increase in the husband’s market wage. These policies are shown to have an impact on the inputs needed in the agricultural production process, the outputs of the agricultural production process and individual time use and consumption of the household members.