This research attempts to bring new insights on the effects of minimum wage on the employment of demographic categories obtained through regression analysis of time-series data. By focusing on employment rather than unemployment, a clearer picture of the so-called harmfull effects of minimum wage should emerged. The analysis concentrates on those less than 25 years old in the province of Quebec and covers the period 1966-1981. Four age-sex categories are retained: males 14-19, males 20-24, females 14-19 and females 20-24. Empirical studies in the U.S. have reported a differential impact between youths and other categories and the existence of a legal minimum wage differential in Quebec for those less than 18 years old may even strengthen the reported differential impact. There are four sections in the paper. The first reviews the major factors potentially associated with the observed variations in the employment of those four deemographic categories during the period 1966-1981. The empirical equations and the various measures used are discussed in the second section. In a typical equation, the dependent variable is the employment ratio and the right-hand side includes a measure of relative population, the adult male unemployment rate, a linear trend, and alternative measures of theminimum wage variable. These are expressed in real on relative terms and some attempt to explicitely account for the legal differential is made. The variables enter the equations logarithmically and both ordinary least squares and first-degree-corrected-for-positive-autoregression estimations were conducted. Various alternative specifications and measures were used to test for the robustness of the minimum wage estimates and these are generally reported in footnotes. The third section analyses the regression results and is followed by a summary and some research and policy considerations.

The major fïndings are: 1) minimum wage increases do have some negative employment effect on youths, 2) variations in the legal differential generates autonomous effect of their own although it was not possible to disentangle the two numerically, and 3) the employment effect on those 20-24 years old is more ambiguous possibly because of substitution effects between the two age groups but more generally, it is felt that the various methological limits to that kind of analysis may forbid more meaningfull results.

With respect to minimum wage policy, the author argues in favor of indexing the adult legal rate to the cost of living. Although the Quebec minimum wage has ranked relatively high in the mid-seventies with respect to other Canadian jurisdictions and North-Eastern states, this is no more the case after a four-year freeze.