Unemployment Incidence of Immigrant Men in Canada

Abstract

This study compared the unemployment probabilities (UP) of immigrant and non-immigrant men in Canada during 1983-92. Data were obtained over an 11-year period from microtapes of the Survey of Consumer Finances of Statistics Canada in 1982 1983 and 1985 through 1993. The sample included all employed and unemployed men aged 24-53 years who are identified by 5 arrival cohorts. In the first model without the effects of macroeconomic conditions UP differences existed only during recessionary periods. UP of immigrants over time appeared to converge toward non-immigrant patterns. UP for both groups for the default characteristics plus college education were lower than for males with a high school education. However UP differences persisted during recessions. Countercyclical findings suggest that changes in labor demand during recessions explain immigrant unemployment differences. Findings from fixed effects specifications using pooled cross sections during 1983-92 including 2 recessionary troughs suggest misspecification but an impact from macroeconomic conditions except for cohorts arriving in 1946-55 and particularly strong effects for the 1971-75 arrival cohort. Assimilation occurred rapidly. Results from flexible form specifications indicate that the 1976-80 and the 1971-75 cohorts had significant unemployment assimilation. There was no evidence of cohort long-term unemployment effects. Unemployment assimilation rather than cohort unemployment effects better explains the convergence of immigrant UP with native UP. Policies should not encourage migration during expansionary periods.