Female Labour Supply with Time Constraints

Abstract

The Italian labour market seems unable to allocate a significant fraction of the working age population efficiently. The gap between the employment rate in Italy and in the other developed economies is foremost attributable to the low employment rates of youth, seniors and women. The low employment rates of these three groups are due to several factors limiting both labour demand and labour supply. For women in particular, constraints on the allocation of time play a crucial role in determining labour supply behaviour. In this thesis we try to understand how non-standard time constraints may affect the behaviour of women, and their labour supply in particular. In the first chapter we study how the constraints on work-schedules affect the time allocation of workers in Italy. For a large fraction of employed individuals the work schedule is very rigid, as a consequence of outdated industrial relations. In order to understand whether constraints on the work-schedule produce significant effects on the allocation of time of wage/salary workers in Italy, we exploit the intrinsic differences between them and self employed workers. In fact, one of the main features of self-employment is the greater control over the days worked and daily hours of work. We use the last wave of the Italian time use survey (2008-2009) to provide evidence that the distribution of hours of work of self-employed workers is much more dispersed than that of wage/salary workers and that average standard deviation of their daily minutes of work within a week is significantly larger. Then we show that self-employed workers respond more to shocks affecting the value of leisure. We show that on sunny days the increase of leisure and the reduction of work are significantly larger for self-employed workers. We address whether unobservable characteristics, such as preferences for leisure and for outdoor activities in particular, determine this differential response and find no evidence for this. We interpret the differential response to weather shocks as a consequence of the time constraints on work-schedules. This evidence is relevant for female labour force participation since in Italy a large fraction of women choose not to work because they would otherwise not be able to reconcile family and work responsibilities. In the second chapter we study the Added Worker Effect (AWE). The retrospective questions provided by the new labour force survey allow identification of transitions between labour market states in a 12 month time-window. Since we are able to identify the reason for the husband’s job loss, we distinguish between transitions associated with low or high income losses. We find that both the wife’s probability of joining the labour force and that of finding a job increase when the husband is dismissed or he is forced to quit his job for health reasons, two cases of usually high income losses. Moreover, we estimate the wife’s full transition matrix between labour market states and we find that the loss of a job by a husband increases the probability that his wife will enter the iv labour force. Finally, we provide some descriptive evidence that time constraints can also impact the magnitude of the AWE. Focusing on mothers with young children, we show that the estimated AWE is positively correlated with the regional provision of child care services. The third chapter is based on the time use files of the Canadian General Social Survey. We study how Sunday shopping deregulation changed the time allocation of women, with a particular focus on those with children. The empirical analysis relies on the provincial variation in the time of the policy change. Our results suggest that women with children, who usually face stringent time constraints, respond to the policy change by substituting weekday shopping with Sunday shopping. The amount of time these women save from doing shopping on weekdays allows them to increase their minutes of work. On Sunday, shopping increases at the expense of leisure. The main result of this chapter is that the labour supply of mothers may change even when non-obvious constraints on the allocation of time change.