Are Canadians More Likely to Lose Their Jobs in the 1990s?

Abstract

Canadians are increasingly concerned about rising job instability. Job instability can take various forms and can be measured in numerous ways. As part of a comprehensive research effort to examine job instability, this paper uses the Longitudinal Worker File (LWF) on the separations of Canadian workers from 1978 to 1993 to assess one dimension of job instability--permanent layoffs. The key question addressed in the paper is: Have permanent layoffs in Canada increased in the 1980s and early 1990s as compared to the late 1970s? We examine the time trend of permanent layoffs first by looking at the permanent layoff rate and then by logistic regressions to predict the probability of permanent layoffs. The analysis is undertaken for all workers as well as for particular sub-groups. Created by many complex processes, permanent layoffs are an on-going feature of our economy and not as cyclically sensitive as quits and other means of workforce adjustments used by firms (i.e., temporary layoffs and hirings). Every year, over a million workers are permanently displaced from their jobs, no matter whether in recessions, recovery or expansionary periods. This is as true in the 1980s and early 1990s as in the late 1970s. Permanent layoffs to 1993 have shown no overall sign of an upward trend when compared to earlier years which are comparable in the business cycle. This holds true whether using the raw data or after controlling for changes in the composition of the workforce by gender, age, province, industry and firm size. However, an increase in the probability of permanent layoffs is observed among some particular groups of workers, notably older or higher paid workers, those in the primary sector or in health, education and welfare services. We will have to wait for more recent data to evaluate trends beyond 1993. The data further show that the Canadian labor market adjusts to structural changes more through depressed hirings than increased layoffs. While the risk of permanently losing one?s job, to 1993 at least, is no higher than in earlier comparable periods, the chance of finding a new job is considerably lower, at least in the aggregate. Furthermore, most job creation in the 1990s has been self-employment, where earning may be more unstable than among paid jobs.