In this article, we study the impact of changes of total labor costs on employment of low-wage workers in France in a period, 1990 to 1998, that saw sudden and large changes in these costs. We use longitudinal data from the French Labor Force survey (« enquête emploi ») in order to understand the consequences of real decreases and real increases of the labor cost. We examine the transition probabilities from employment to non-employment and from non-employment to employment. In particular, we compare the transition probabilities of the workers that were directly affected by the changes ("between" workers) with the transition probabilities of workers closest in the wage distribution to those directly affected ("marginal" workers). In all years with an increasing minimum cost, the "between" group (or the treated using the vocabulary of controlled experiments) comprises all workers whose costs in year t lie between the old (year t) and the new (year $1) minimum. In all years with a decreasing minimum, the "between" group comprises all workers whose costs in year t+1 lie between the present minimum cost (year t+1) and the old (year t) minimum cost. The results can be summarized as follows. Comparing years of increasing minimum cost and decreasing minimum cost, difference-in-difference estimates imply that an increase of 1% of the cost implies roughly an increase of 1.5% in the probability of transiting from employment to non-employment for the treated workers, the resulting elasticity being -1.5. Second, results for the transitions from non-employment to employment are less clear-cut. Tax subsidies have a small and insignificant impact on entry"marginal" group constitutes a good control group. In addition, there is no obvious evidence of substitution between the "between" and "marginal" groups of workers, but there is some evidence of substitution between workers within the tax subsidy zone, with wages above those of the "marginal", and workers outside the subsidy zone.