Taxation, Marriage and Labor Supply: Evidence from a Natural Experiment in France

Abstract

This paper uses the French family quotient ("quotient familal") reform of 1995 to analyse the impact of the individual income tax on marriage behavior and labor supply decisions. An important feature of this reform was the cancellation of fiscal subsidies aimed at cohabitant couples with children. Before 1995, the system of the family quotient granted one extra half unit to each single parent with children as defined for tax purposes. The 1995 family quotient reform cancels the benefit for cohabitants with children by introducing the notion of isolated parents with children in the tax declaration. This measure thus compensates the marriage penalty for couples with children but does not change anything for couples without children. As a result, the tax for a one-earner cohabitant couple with one child that earns 35, 000 euros a year has increased by about 1, 200 euros after the reform. To assess the impact of the reform, we use the difference-in-differences estimation approach. Using the panel structure of the French employment survey (1990-2000) we find that the probability of marriage has increased for stable couples by about 4 points because of the reform. In a second stage, the response of married women with children is analyzed, using as a control group women who did not have children before they married. We therefore identify the tax effect as the differences between the change in the labor supply of women with children before they married and the change in the labor supply of women without children before they married. We find strong evidence that the labor supply of married women has decreased due to the 1995 family quotient reform. Nevertheless, although our results support the working hours response, they are inconclusive as to participation response.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 21 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.