Directorates
·	Director General's Office
·	Data Development
·	Strategic Evaluations
·	Partnership and Continuous Evaluations
Publications
·	Evaluation Reports
·	Evaluation Briefs
·	Labour Market Development Agreement Series
·	Lessons Learned Series
·	Evaluation Tool Kit Series
·	Evaluation Forum Newsletter
·	From Theory To Practice

The goal of this study is to look for any changes in the distribution of the hours of jobs in the economy that may have resulted from the move from Unemployment Insurance (UI) to Employment Insurance (EI). It also quantifies the changes in benefit eligibility and entitlement that followed from the same policy change. The key aspects of the legislation that are the focus of the study are the movement to a system that, first, uses hours as the unit of account, and second, increases coverage to include low hours jobs. Implicit in the coverage of low hours jobs is the possibility that jobs can be combined to be eligible for EI benefits in ways that are not possible under UI.

This move meant that starting January 5th, 1997, all paid workers were covered by EI. In contrast, under the previous UI legislation only jobs of more than 15 hours per week, or with earnings over a specified amount (the amount varied over time, but was about $150–163 in the period around the change) were covered. UI created a distortion in the labour market arising both from obtaining the value of coverage, and from increasing the price of labour because of premium payments as weekly job hours crossed this coverage threshold point. EI eliminated this distortion.

This study uses the Canadian Labour Force Survey (LFS), and the 1996 Canadian Out of Employment Panel (COEP) survey combined with Human Resources Development Canada (HRDC) administrative files, to explore labour market changes resulting from the legislation. The two surveys are quite different in their sampling frame and measure very different aspects of the labour market. The LFS is a cross-sectional survey of the population and as such is a “stock” survey that permits an analysis of the labour force’s behaviour and composition, but it does not measure the “flow” of job separations very well. The COEP, however, does exactly this, and is an excellent source of information on movements across jobs and labour force states.

A few caveats must be borne in mind when interpreting the observations contained in this report. First, only a short amount of time has passed since the legislation was enacted, and some aspects of the EI Act were phased in over an extended period. This study’s findings can, therefore, only be seen as measuring the short-run impact of the policy change. Second, a single aspect of the legislation may induce opposing reactions from different “worker-firm” pairs with some increasing their hours, and others decreasing them, but all that we can observe is the net macroeconomic effect. In as much as this net effect is the result of underlying movements in different directions, we may be missing the impact on individuals. Third, there was a marked improvement in the economy across the two policy periods that may confound some of the results slightly. Finally, the Bill itself is a very complex piece of legislation that reforms many aspects of the UI/EI system. A particular element of the reform might induce a reaction in one direction, while another element could cause the opposite effect. Once again, given the nature of the data available, only the net effect will be observable.

Summary of Findings from the LFS Data

Multiple Job Holding

There is no observed change in multiple job holding for men. There is no observable change for women overall, but women in industries with high rates of part-time employment may have experienced a small increase in multiple job holding.

The Hours Distribution of All Jobs

For men outside of the Atlantic region there appears to have been a small decrease in jobs under 15 hours per week that is coincident with the UI-EI transition. There is also a concomitant increase in the percentage of jobs over 30 hours. No change is observed for women.

The Hours Distribution of New Jobs

New jobs (less than one month old) might be a margin on which firms and workers can react to the legislation quickly. For both sexes there is a drop in the percentage of new jobs that are less than 15 hours per week across the UI-EI transition period, and an increase in the percentage that are over 30 hours per week. There is, therefore, some evidence, although it is limited given how recently the policy change occurred, that over time the weekly hours of work distribution will shift to fewer low hours jobs in reaction to the policy change.

Summary of Findings from the COEP Data

Eligibility

1. About 5.1 percent of all job separators who are ineligible for benefits under UI become eligible under EI, while 2.8 percent of all job separators become ineligible.
2. Women, and youth (<25), are about 2-3 percent more likely to be in both the groups that experience an increase and a decrease in eligibility than males, or other age groups.
3. There are also differences across provinces with workers in the Atlantic provinces being 2-8 percent more likely to become eligible among those who are eligible, and 1–2 percent less likely to become disqualified among those who are disqualified. Those in the Territories, on the other hand, are about 4 percent less likely to become eligible among those who become eligible as a result of the policy change.

Entitlement

1. Entitlement changes are summarized for various job types. Some workers experience large reductions in entitlement, while others experience large increases as a result of the policy change. On average, however, there was only a slight reduction in entitlement, about 0.3 weeks.
2. The entitlement changes resulted from two distinct aspects of the EI reform — the move to hours and the reduction in the maximum weeks of entitlement from 50 to 45 weeks. A counterfactual is, therefore, performed that isolates the effect of the hours change. In the absence of the maximum benefit reduction, average entitlement increases, rather than decreases, by an average of about 0.4 weeks as a result of the move to an hours based system with first hour coverage.
3. Women experience a reduction of about 2.2 more weeks of entitlement than men under the EI policy relative to the UI one. Further, workers with post-secondary education experience, on average, an entitlement decrease of about a week and a half more than those whose highest level of education is a high school degree, and about 2 weeks more than those with less than a high school degree. Older workers (>55) also have, on average, about 1.7 weeks greater reduction in entitlement than others. The impact across provinces appears to be quite even except for a small increase in average relative entitlement in PEI, and a small reduction in the Territories. These small average entitlement changes across identifiable demographic groups suggest that the very large changes in entitlement observed for a small percentage of job separators are spread relatively even across the same groups.

Overall, it appears that there have been many small changes in the distribution of hours, and in the nature of new (low tenure) jobs that are coincident with the onset of EI. While the UI-EI transition is likely associated with many of these effects, interpreting them as resulting exclusively from the UI-EI transition may not be correct given the lack of controls for changing business cycles across the regimes.

The eligibility and entitlement results, mentioned above, are clearer. The UI-EI transition has had very diverse impacts, imposing costs (both large and small) on some individuals and bringing benefits to others. Overall, those who become eligible outnumber those who become ineligible, and overall, the hours portion of the UI-EI transition is associated with a small increase in entitlement.

Finally, the measured responses to, or effects of, the policy changes need to be put into perspective. Relative to the entire labour force every one of them is small. But, for the subgroups involved, especially workers whose weekly hours are close to the former UI coverage “cutoff” of 15, the changes can be quite large.