Cognitive Skills and Immigrant Earnings
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
by Aneta Bonikowska, David A. Green and W. Craig Riddell
Considerable research effort has been devoted to understanding earnings differences between immigrant and native-born (i.e. Canadian-born) workers (see, for example, Chiswick (1978), Borjas (1985, 1995) for the United States, and Baker and Benjamin (1994), Bloom, Grenier and Gunderson (1995), Grant (1999) and Aydemir and Skuterud (2005) for Canada). These studies establish that immigrants typically earn less than Canadian-born workers with the same amount of education and work experience. The low earnings of immigrants are often attributed to the specificity of human capital to the country where it originates. Skills generated through education or work experience in the source country cannot be directly transferred to the host country, resulting in apparently well qualified immigrants holding low paying jobs. Of course, this is not the only potential explanation for lower immigrant earnings. Another possibility is that employers in the host country discriminate against immigrants, that is, pay immigrant workers less than equally productive Canadian-born workers. Investigating these issues would be straightforward if we had access to direct measures of skill. In that case, we could compare Canadian born and immigrant workers with the same levels of measured education and experience to see whether the immigrants in fact have lower skill levels, supporting the first hypothesis. Alternatively, we could observe whether immigrants get a lower return to their observed skills, supporting the second hypothesis. In this paper, we take advantage of the rich data provided by the Canadian component of the International Adult Literacy and Skills Survey (IALSS)1 which includes both standard demographic and labour market information for the Canadian born and immigrants and results from tests of literacy, numeracy and problem-solving skills. Interpreting the test scores as direct measurements of cognitive skills, we are able to provide a closer examination of explanations for low immigrant earnings than has previously been possible. In addition, the data include more precise information on where education was obtained and age of migration than is available in most previous studies, further refining our ability to scrutinize immigrant-Canadian born earnings differentials.
The primary goal of the paper is to provide answers to four questions related to immigrants' skills. First, do the cognitive skills of immigrants differ from those of Canadian born and, if so, in what way? Second, do immigrant – Canadian born skill differences depend on where immigrant human capital was acquired? Third, do immigrants receive different returns to these skills than observationally similar Canadian-born workers? Fourth, can differences in levels and returns to these skills explain differences in earnings between immigrant and Canadian-born workers?
Recent contributions stress the need to account carefully for where education and experience was acquired in examining immigrant earnings. Using Israeli Census data, Friedberg (2000) finds that lower immigrant earnings compared to Israeli-born workers with similar education and experience can be explained almost entirely by lower returns to experience acquired outside of Israel. This is true in particular for non-European immigrants. Similarly, Green and Worswick (2002) find zero returns to foreign experience for recent immigrant cohorts but show that, in Canada's case, this is a change from the early 1980s when immigrants earned returns to foreign experience that were similar to what the Canadian born were earning for domestically acquired experience. Much of this change over time is related to changes in the source country composition of the inflow. Using data on immigrants to Ontario, Ferrer, Green and Riddell (2006) also find that low returns to foreign experience play a major role in the immigrant – Canadian-born earnings gap, especially among the university-educated. Schaafsma and Sweetman (2001) and Ferrer and Riddell (forthcoming) examine the issue of lower returns to foreign acquired education in a somewhat indirect way by using age at immigration.2 Both papers find that returns to foreign education, while lower than those to Canadian education, are still substantial. In this paper we provide additional evidence on the importance of where education and experience was acquired. This additional evidence is particularly valuable because the data used in this paper has definite advantages over the data used in previous studies. Part of the contribution of this paper is to re-examine issues about returns to foreign experience and education raised in earlier papers, using better data.
This paper also builds on work by Green and Riddell (2003) and Ferrer, Green and Riddell (2006) that uses, respectively, the International Adult Literacy Survey (IALS) and the Ontario Immigrant Literacy Survey (OILS) to examine the role of cognitive skills in earnings patterns of Canadian-born and immigrant workers. Like the IALSS, the IALS and OILS data contain both standard survey questions and literacy and numeracy tests. Green and Riddell (2003) argue that the types of literacy questions asked in the IALS are particularly conducive to using the literacy test scores as measures of cognitive skills possessed by the respondent at the time of the survey. Based on this assumption, they argue that much can be learned about how these basic skills influence earnings from an analysis of interactions of the literacy measures and other standard human capital variables. In that analysis they use a hedonic model in which observed earnings are directly determined by the basic skills an individual possesses and the implicit prices of those skills. We adopt a similar interpretative framework in this paper.
The results imply that the answer to our first main question – Do immigrant literacy skills differ from those of the Canadian born? – is yes. The Canadian-born test score distributions dominate those of immigrants, and immigrants have lower average test scores than observationally equivalent Canadian-born workers. An important part of the gap stems from a set of immigrants unable to complete the tests, and who were therefore assigned low scores. We also find strong evidence that skill differences depend on where human capital was acquired. Immigrants who completed their education prior to arrival in Canada have significantly lower skills than otherwise similar immigrants who obtained some or all of their education in Canada. Regardless of these differences in skill levels and acquisition, however, we clearly reject the hypothesis that immigrants receive lower returns to cognitive skills than the Canadian born. Indeed, an important group of immigrants benefit more than do native-born Canadians from higher skill levels. We argue that this is evidence against a discrimination explanation for differences in earnings between immigrant and Canadian-born workers.
Our earnings regression results support findings in earlier papers that returns to both foreign-acquired education and experience for immigrants are lower than returns to education and experience obtained in Canada by either immigrants or Canadian-born workers. This pattern in returns to experience does not change once we control for cognitive skills, indicating that the root of the problem does not lie in foreign experience generating lower cognitive skills. Cognitive skills themselves exert a significant effect on earnings, with a 100-point increase in the average skill score generating an earnings increase of almost 30 percent3. The combination of this return to skills and the lower skill levels of immigrants explain part of the immigrant earnings differential. We estimate that raising immigrant average skill levels to the Canadian-born level would almost eliminate the earnings disadvantage of high school educated male immigrants relative to similarly educated Canadian-born men, and would produce a substantial earnings advantage among high school educated female immigrants. Among the university educated, for whom the earnings differential is larger, raising immigrant skill levels to the Canadian-born level would reduce the male earnings gap by more than 50 percent and would more than eliminate the female earnings gap, turning the immigrant disadvantage into an advantage.
The paper is organized as follows. In the next section we present a framework for considering what we might learn from introducing cognitive skills measures into a standard earnings equation. In the third section, we discuss our data and present basic data patterns. The fourth section examines whether immigrants have different skill levels from the Canadian born. The fifth section contains the analysis of immigrant earnings, and the final section concludes.
- The IALSS study is known internationally as the Adult Literacy and Life Skills Survey (ALL).
- The Canadian Census of Population — the data used in most previous immigration studies – has no information about the origin of human capital. In addition, the age of arrival variable is coded in bracketed intervals. This enormously complicates the imputation of measures of pre- and post-migration experience.
- To put this in perspective, the mean skill score among high school graduates is approximately 280, whereas that among university graduates is approximately 320, a differential of 40 points.
- Date modified: