The training divide: a Canada-US comparison of employee training

In the paper, we use data from the [International Adult Literacy Survey] IALS (1994), a unique data set that is administered similarly across Canada and the United States, in particular, that contains answers to questions on literacy and numeracy skills, as well as information on the demographic background of the respondents. This allows us to explore many dimensions of training previously overlooked. Two broad themes emerge from our analysis. The first one is that while there is some evidence of cross-country differences in the provision and in the intensity of training in terms of magnitude, we think that the within-country factors related to ethnicity, language, or gender differences play a major role. The most solid statistical evidence in this paper concerns lower levels of training incidence and intensity for French Canadians than for English Canadians. We also find similar evidence, although not quite as strong, for a difference between African Americans and White Americans. Secondly, where we find strong evidence of major differences between Canada and the United States, it is not so much in terms of the amount of training provided to its workforce, but rather in terms of the very different effects that the literacy measures seem to have on training. In particular, the surprisingly negative association between quantitative literacy and training found in the United States is not part of the training structure in Canada. The benefit of having standardized questions asked on both training and literacy to each individual in both countries makes for a meaningful analysis of the determinants of training by country. Together these two sets of findings are strongly suggestive that two fruitful avenues for future research would be first to further explore the link between language/ethnicity type factors and post-schooling human capital accumulation. The second possible research avenue would be to further investigate the reasons why, for example, more quantitatively literate workers in the United States are less likely to receive training whereas the opposite is true in Canada, which is the traditionally expected outcome.

Published abstract reprinted by permission of the copyright owner. Show less