Three Essays in Public Policies: Unemployment Insurance, Parental Leave, And Breastfeeding Subsidy

Abstract

The three essays of this dissertation investigate the effects of various public policies, in order to inform policy design.

The first essay, ``Find a Job Now, Start Working Later: Does Unemployment Insurance Subsidize Leisure?,'' studies the extent to which unemployed individuals delay job start after finding a job. Distorting incentives is a major concern when implementing unemployment insurance (UI) programs. In this paper, I explore a new channel of moral hazard: postponing job start upon finding a job. This essay first develops a theoretical job search model that allows for a delayed job start. Then, the extent to which unemployed individuals delay job start after finding a job is assessed. I find that the delay between job offer and job start is longer when receiving UI benefits after accepting a job. It suggests that some individuals take advantage of the availability of UI benefits to postpone job start.

In the second essay, `` Wage Effects of Extended Parental Leave in Canada,'' I estimate the incidence of extended maternity leave benefits on relative wages. By increasing the effective cost of hiring women of childbearing age, the benefits given to mothers when extending maternity leave could result in a negative impact on the wages of all women of childbearing age. This essay estimates the effects of the 25-week increase in paid parental leave in Canada in 2000 on the relative average hourly wage of women of childbearing age. In line with previous research, I find no statistically significant effects in ordinary least square specifications. However, using a method that corrects for sample selection and unobserved heterogeneity, I find weak evidence that the extension of parental leave benefits decreased the relative hourly wage of women of childbearing age by 1.2% compared to older women. However, this decrease in the relative hourly wage does not translate into an effect on the relative annual earnings of women of childbearing age.

Finally, in the third essay, `` Does Breastfeeding Improve Child Outcomes? The Case of a Breastfeeding Subsidy in Quebec,'' I examine whether increasing the breastfeeding subsidy for welfare recipients in Quebec from $20 to $50 per month in 1994 increased the breastfeeding rate among new mothers. I find that the policy change increased the initiation of breastfeeding by 9.3 percentage points. I then use the increase in the breastfeeding subsidy as an instrumental variable to assess the causal effect of breastfeeding. I do not find any positive effect of breastfeeding on health and no effects on cognitive outcomes, with the exception of improved motor scores at age 1 and a decrease in the prevalence of diagnosed learning disability at age 11. However, children breastfed show some behavioral improvements in terms of decreased aggression, as well as hyperactivity and inattention score. Almost none of these effects are persistent at 15 years old.