The Canadian earnings functions under test

Abstract

Many Canadian studies continue to use Mincer's basic post-schooling earnings function to analyse labour market earnings of individuals. Several policy recommendations are based on these analyses. The underlying assumptions in these studies are that the Mincer's specification of earnings model is correct and that the error terms of these models are normally distributed and homoscedastic. By performing several tests on four cross-sectional Canadian data sets ranging from 1981 to 1991, the present study confirms previous suggestions in literature that the standard Mincer specification of earnings model should not be adopted without testing. The hypotheses of a correctly specified functional form, and of the normality and homoscedasticity of errors, are rejected for all the data sets used in this study. These results question the validity of past policy recommendations based on Mincer type earnings functions in Canada.

^*Author to whom correspondence should be addressed

^*Author to whom correspondence should be addressed

Notes

^*Author to whom correspondence should be addressed