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

Under the old Unemployment Insurance (UI) system, weeks during which workers were employed for fewer than 15 hours were not covered by the program, so that these earnings were not subject to payroll tax deductions and did not contribute weeks towards UI eligibility. Furthermore, prior to Bill C-12, all weeks of employment of greater than 15 hours counted in the same way towards eligibility and benefit entitlement. Under the hours-based system, longer-hours weeks contribute more to eligibility and generate greater benefit entitlement than shorter hours weeks.

Table 1 summarizes these changes in program coverage and benefit entitlement for a number of important cases. The dark shaded rows at the bottom of Table 1 indicate cases where jobs that were previously not covered under the UI program became covered under the Employment Insurance (EI) program. Under EI, workers with these work arrangements now pay the EI payroll tax, and these hours of work contribute to eligibility and entitlement. As well as eliminating the incentive for workers and firms to distort their hours decisions, the inclusion of these weeks in the EI program was intended to broaden coverage to the growing number of workers in part-time jobs, including many who actually worked full-time hours, since broad coverage is important if the program is to continue to be effective as an automatic macroeconomic stabilizer. As the corresponding cells show, unless workers combine short weeks with longer weeks some time during the year, they will pay the payroll tax but never qualify for benefits. In cases where total annual earnings are less than $2000 per year, however, all of the payroll tax contributions of the worker and firm are reimbursed, so that the net effect of the EI system is zero.

The light shaded columns show the change in benefit entitlement under Bill C-12 for workers with a number of different work arrangements in the cases of the very lowest and highest unemployment rate categories. Scanning down the first of these shaded columns for the case of the lowest unemployment rate, we see that some workers who work regular schedules of 40 hours or more saw their benefit entitlement increase under EI. Workers employed on part-time schedules of 15 hours per week and more for a significant portion of the year saw their benefit entitlement decrease substantially. Workers employed on part-time schedules for fewer than 10 hours per week continued to be ineligible for benefits.

Scanning down the second of these shaded columns for the case of the highest unemployment rate, we see that some workers who work long work weeks gained benefit entitlement, while others lost it, depending on the number of weeks worked per year. Part-year workers who work long-hours when they do work saw an increase in benefits that is in some cases quite considerable. Again, workers on part-time schedules of 15 hours per week and more, experienced losses in benefit entitlement that are in some cases very large. Not all part-time workers lost benefit entitlement under Bill C-12, however. We see that a worker employed for 10 hours per week for 52 weeks per year who would not have been covered under the UI program is entitled to 33 weeks of EI benefit.

Obviously many workers vary their schedules over the year or hold multiple jobs. While it is not possible to quantify easily the change in entitlement in these cases, we can make some general comments. First, weeks during which workers work fewer than 15 hours and weeks in which workers work more than 15 hours in total but less than 15 hours in any one job, contribute to eligibility and benefit entitlement under EI, but would not have done so under UI. For some workers, this additional contribution to eligibility and entitlement will more than compensate for the cost of the EI payroll tax, and therefore will make them better off. For others it will not. Second, weeks during which workers work at least 15 hours but less than 35 hours contribute less to eligibility and entitlement under EI than they did under UI. Third, in most cases, weeks during which workers work more than 35 hours contribute more to eligibility and entitlement under EI than they did under UI.

A summary measure of the number of weeks of benefit entitlement for each weekly hours schedule was computed by taking the number of weeks of benefit entitlement for a person working some number of hours per week for 12, 20, 40 and 52 weeks per year, and calculating the simple arithmetic average of these values. This summary measure calculated for 1996 and 1997 is plotted in Figures 1 through 3 for low, medium and high unemployment regions respectively. Under UI, workers who worked fewer than 15 hours per week never qualified for benefits. While workers who worked at least 15 hours per week saw their benefit entitlement increase with weeks worked, they did not see their entitlement increase as they increased their hours per week. Our constructed summary UI benefit entitlement measure therefore is a simple step function, increasing from zero weeks for workers with weekly schedules of less than 15 hours per week, to some number of weeks that depends on the unemployment rate at exactly 15 hours per week. The move to an hours-based system replaced the simple step-function at 15 hours under the UI program with a more gradual increase in benefit entitlement beginning at less than 15 hours and continuing through large portions of the hours distribution. The relative magnitude of these changes in regions with different unemployment rates can be seen more clearly in Figure 4, which plots the change in the value of the summary measure between 1996 and 1997. A positive value of this change indicates an increase in this measure of benefit entitlement under EI, a negative value indicates a loss. Again, we see that the change in entitlement was substantial in some regions of the hours distribution, and varied considerably across it.