Partial employment protection and perceived job security: evidence from France

Abstract

This paper assesses the causal effect of partial employment protection on workers' subjective job security via the perceived probability of layoff. We consider the rise in the French Delalande tax, which is paid by private firms if they lay off older workers. This reform was restricted to large firms and therefore allows us to use a difference-indifference strategy. In ECHP data, we find that the change in the perceived probability of layoffs induced by the higher Delalande tax improved the subjective job security of older (protected) workers, but at the cost of a negative externality on younger (unprotected) workers. While the fall in perceived job security of younger workers is mirrored by actual increase in their layoff rate, the rise in perceived job security of older workers is more puzzling because we do not observe a reduction in their layoff rate.

A Natural Experiment on Job Insecurity and Fertility in France

“…Behaghel (2007) proposes a stochastic job-matching model to predict the impact of the changes to the Delalande tax; this shows that a rise in the Delalande tax will reduce the ECINEQ WP 2020 -532 May 2020 separation rate of those workers covered by the reform but will increase that of the younger workers in the same firms. Georgieff and Lepinteur (2018) use the same ECHP data as we do here, and confirm empirically that the 1999 rise in the Delalande tax had perverse effects on younger (aged under 50) workers in larger French firms by increasing both their perceived job insecurity and their actual risk of layoff (relative to the covered older workers).…”

“…We also looked at the effect of the announcement and implementation of the 1999 rise on the Delalande tax on the actual layoff probabilities of married and cohabiting workers using data from the French Labour Survey. As for all marital statuses inGeorgieff and Lepinteur (2018), we find positive and significant estimates from 1998 onwards (seeFigure A2).ECINEQ WP 2020 -532May 2020…”

How much does others’ protection matter? Employment protection, future labour market prospects and well-being

“…He finds no significant relationship. Similarly, Georgieff and Lepinteur (2017) show that an increase in the French Delande Tax (payable by firms whenever they dismiss elderly workers) decreases the perceived risk of job loss for the elderly, but increases it for younger workers. Kuroki (2012) showed that a deregulation of atypical work in Japan led to a decrease in permanent workers' perceived risk of job loss.…”

The asymmetric experience of gains and losses in job security on health

“…Respondents answered on a 6-point scale, 1 meaning "Not Satisfied" and 6 meaning "Fully Satisfied". This measure of perceived job security has already been used by the literature (Clark & Postel-Vinay (2009), Georgieff & Lepinteur (2018)) and is a strong predictor of individual choices such as future job quits (Clark (2001)). Boyce et al (2013) and test the loss aversion hypothesis using the following valueadded model:…”

“…This rise was restricted to firms with more than 50 employees, and as such provides a natural quasi-experiment that can be analysed via difference-in-differences estimation. Georgieff & Lepinteur (2018) demonstrated that the rise in the Delalande tax increased the perceived job security of workers aged over 50 at the cost of higher levels of perceived job insecurity of their younger colleagues in the treated firms. I take benefit of the direct impact of the reform on protected workers and its spillover effects on unprotected workers to estimate separately the causal impact of an exogenous increase and an exogenous decrease in job security on self-assessed health.…”