Labour market effects of immigration : evidence from Canada
Immigration, the subject of repeated policy debates throughout the last two decades, has once again assumed a central position on the policy agenda. This debate has become more intense in recent years in Canada; the fear is over the potential job displacement and unemployment of Canadian-born workers, and the consequence to the Canadian economy. The recent immigrant incomes have been falling compared to their older counterparts helped to trigger the current policy debate. This thesis attempts to address this debate by providing an objective assessment of the displacement of Canadian-born workers due to immigration and the unemployment-immigration dynamics over the past 40 years of immigration to Canada. The thesis consists of two objectives:Objective-I: Job Displacement Effects of Immigration on Canadian-born First I address the job displacement effects on Canadian-born due to exogenous shifts in immigration flows. It is, therefore, necessary to consider the substitutability or complementarity between Canadian-born and immigrant workers. This is examined by estimating the set of wage earnings equation from the “Generalized Leontief Production Function”. The model specification abstracts from the role of capital, by assuming that labor and capital are separable in production. I then derive the iterated Zellner-efficient estimator (IZEF) (which is numerically equivalent to the maximum likelihood estimator) from the set of wage earnings equations. Then the degree of substitutability or complementarity is calculated using Hick’s (as opposed to Allen’s) elasticity of complementarity. The estimated Hicksian elasticities suggest, in the aggregate, there is no displacement of Canadian-born workers by immigration, although there is some displacement by industry.Objective-II: Unemployment and Immigration DynamicsNext, I consider immigrant not only as an additions to the existing labor force but also job creation effects through their effects for goods and services. Here immigrants are considered as endogenous and I model the dynamics of unemployment and immigration. As a first step, statistical causality is investigated between immigration and unemployment. But causality methods can suffer from omitted variable problem. So, I construct a theoretical labor market and use the cointegration analysis to determine the long run relationship among unemployment rate, immigration level, real wage, and real GDP. Then, I estimate the short-run dynamics with a specification in difference form where the parameters of the cointegrating vectors from the first-step are fixed and entered as an error correction mechanism. The causality test finds no evidence of a significant effect of Canadian unemployment on immigration. The estimation of the long-run and short-run parameter indicates that no statistically significant relationship exists between unemployment and immigration.
DegreeMaster of Arts (M.A.)
SupervisorHuq, M. Mobinul
CommitteeTran, Kien C.; Lucas, Robert F.; Li, Peter S.
Copyright DateAugust 2003