Pre-Separation Job Durations and Unemployment Insurance Eligibility - May 1998

Pre-Separation Job Durations and Unemployment Insurance Eligibility

View Whole Report

• Context
• Key Findings
• Evaluation Approach
• Biographical Notes

The changes to the UI system implemented in February 1994 through the Budget Omnibus Bill facilitated the reduction of the growing costs of the program (which had caused UI premiums rates to increase substantially) and began the process of structural reform of the program. A key objective of the 1994 reform was to retain existing jobs and promote job growth by containing premium rate increases and commencing the process of bringing premium rate levels down. During the recessionary period of the early 1990s, premium rates had increased by about 33%, resulting in additional job losses. The 1994 reform reduced UI premiums for 1995 to $3.00 per $100 of insurable earnings from $3.07 in 1994 and from a planned statutory rate of $3.30. In addition, changes were made to the eligibility requirements for claimants in high unemployment regions and to benefit durations and benefit levels for all claimants.

Overall, the changes to the financing and program rules were designed to promote job creation, ensure adequacy and fairness and commence the process of reducing disincentives in the system.

The key changes to the program design were:

Strengthening the link between work history and UI eligibility by changing the method used to determine how long a person could receive benefits and by increasing the minimum amount of time a person in high unemployment areas needed to work to qualify for UI from 10 weeks to 12 weeks;

Reducing the effect of regional unemployment rates on how long a person can collect UI benefits, while continuing to recognize the need for additional assistance for claimants who live in areas of high unemployment;

Establishing a dual benefit rate: a higher rate of 60% of average insurable earnings for claimants with low weekly wages and dependents; a slightly lower benefit rate of 55% (previously 57%) for all other claimants; and,

Improving the fairness of the UI program by amending and clarifying how the Voluntary Quit and Misconduct provisions are applied.

HRDC commissioned four formal evaluation studies to examine how Canadians adjusted to these reforms and one that assessed the degree to which an unemployed individual in 1995 faced financial hardship. These studies were performed by external academic subject-matter experts. Each evaluation represents a stand-alone analysis of a specific topic.

This brief summarizes a study undertaken by Drs. Peter Kuhn and Arthur Sweetman on behalf of HRDC that examines which workers were most affected by the 1994 reforms. The study attempted to identify the provinces and industries where workers were most affected by the changes and whether the changes had a larger effect on male workers or female workers. In addition, the authors attempted to measure the extent to which Canadian workers were able to adjust to the changes by altering their employment and job search behaviour.

Key Findings

The study focused only on the regular UI program and did not examine special benefit programs (e.g. maternity benefits) or the special UI program for self-employed fishers. Drs. Kuhn and Sweetman estimated both the changes in the total number of weeks of UI entitlement and the total number of weeks of UI benefits actually received. The total number of weeks of eligibility is of direct interest to a worker who becomes unemployed. The number of weeks of UI benefits they can count on will likely be an important factor in the worker's financial security and may have an important impact on the worker's job search strategy and effort. However, prior to the introduction of the 1994 changes, most unemployed workers did not use all of the weeks of benefits that they were eligible to receive. This means that in order to examine changes in expenditures by the UI system or the total benefits received by a particular province or industry, it is also necessary to examine the total weeks of benefits actually received by workers.

The authors calculated the average impact based on three different populations: (1) per UI claim, (2) per job separation and (3) per capita (for each province) or per worker (for each industry). The first measure, per UI claim, tells by how many weeks the 1994 reforms shortened the average UI claim. The second measure, per job separation, answers the question "Given that I lose my job, how many fewer weeks of benefits am I likely to be entitled to?" This measure differs from the per claim measure because not all workers who have a job separation either claim UI or are eligible for UI. Finally, the per capita measure can be used to look at how the 1994 reforms affected the receipt of UI benefit weeks in specific industries or provinces. It quantifies how the 1994 reforms caused changes in the distribution of benefits across provinces and industries, and answers the question "How many fewer weeks of UI does the average worker in industry X, or the average resident in province Y, receive as a result of the 1994 reforms?" Even though the effect of the 1994 reforms may, on average, be fairly uniform on a per UI benefit recipient or per job loss basis, the likelihood of experiencing a job separation in the first place, and therefore being eligible for UI, is higher in some industries and provinces than in others.

From their analysis, the authors found that the 1994 changes had the following results.

• The reforms reduced the number of weeks which the average unemployed worker actually received UI benefits by about 3 for unemployed men and by about 4 for unemployed women.

In addition, the authors concluded that the 1994 changes had only a modest effect on the relative distribution of UI benefit week reductions per job separation between provinces and industries. The UI benefit week reductions, however, were higher in the provinces and industries with high unemployment rates than in provinces and industries with low unemployment rates. For example, they found that Newfoundland, with a reduction of 5.1 weeks per claim, experienced the greatest reduction in UI benefit weeks received. The fishing industry experienced the largest drop in regular UI benefit weeks received per claim at 6.1 weeks for women in that industry.

• The reforms reduced the number of weeks of UI benefits that the average job leaver (separator) was entitled to receive by 7.6 for both male and female workers.

The authors concluded that the reductions in weeks of benefits per job separation were roughly evenly distributed across provinces and industries, primarily because workers in high unemployment regions were able to adjust to the increases in the minimum entrance requirements for UI benefits. Without these behavioural adjustments, very large differences in benefit entitlement would have occurred. The authors found that most workers concentrated in high-unemployment provinces and industries were able to find the extra weeks of work necessary to qualify for UI benefits. In particular, they found that most workers who would potentially have been completely disentitled as a result of the 1994 reforms had they not obtained the extra weeks of work were, in fact, able to find enough extra weeks of work to substantially mitigate the effects of the reforms. In contrast to this conclusion, the authors found no clear evidence of a similar effort to find additional weeks of work on the part of those workers who were losing weeks of entitlement (but not being totally disqualified) as a result of the 1994 reforms. This was in spite of the authors' estimate that 50 percent of workers lost between 8 and 16 weeks of benefit entitlement as a result the 1994 changes.

• The reforms reduced the average number of weeks of UI receipt per worker for Canada as a whole by half a week from 3 to 2.5 weeks for females and from 3.6 to 3.1 weeks for males.

The study found the largest effects in industries that traditionally experience high unemployment rates. The average woman employed in the fishing industry collected about 22 weeks of UI benefits per year before the implementation of the 1994 changes and 17 weeks of benefits after the changes. The average male fishery worker collected UI benefits for about seven weeks per year before the 1994 changes and for six weeks after the changes. The average forestry worker (male or female) received over 14 weeks of UI benefits per year before the 1994 changes, but this average dropped by two to three fewer weeks after the changes.

The study found that, on a per capita basis for Canada as a whole, the average number of weeks of UI receipt per person was 1.2 weeks for men and 1.7 weeks for women in the year before the 1994 changes. After the 1994 reforms, this average was reduced by 0.2 weeks for both men and women. The authors found sharp differences in the impact of the 1994 reforms across provinces on a per capita basis, with the largest effects taking place in provinces with higher unemployment rates. In Newfoundland, for example, the average man received 3.8 weeks of UI benefits and the average woman received 5.2 weeks of UI benefits in the year before the 1994 changes. After the 1994 reforms, these averages were reduced by 0.61 weeks for men and 0.80 weeks for women.

Drs. Kuhn and Sweetman concluded that the 1994 changes clearly had a disproportionate impact on the high unemployment provinces and industries that were the greatest users of the UI system prior to the change. Atlantic Canada and Quebec experienced significant reductions in the number of weeks an unemployed worker was eligible to collect UI benefits for, as well as the number of weeks that the unemployed worker actually collected UI benefits. However, these same regions continued to receive a disproportionate share of UI benefits following the changes introduced in 1994, as workers were able to meet or exceed the new, higher entrance requirement.

Evaluation Approach

This evaluation study was based on the estimated impact of the legislative changes on individuals who had job separations before and after July 1994 when Bill C-17 became effective. The authors used HRDC administrative records as their data source for the analysis. The analysis focused on a comparison of a random sample of 10 percent of the workers experiencing a job separation, as measured by the receipt of a Record of Employment (ROE), from HRDC administrative data in two different periods. The first period extended from March 1993 to February 1994, prior to the implementation of the 1994 reforms, and the second window started in August 1994 and continued until the end of July 1995.

In this study, the authors measured the impact of the 1994 changes on the weeks of benefits an unemployed worker was entitled to collect and the weeks of benefits actually collected by breaking down the entire distribution of UI entitlements and benefit receipts before and after the 1994 changes. The changes introduced in 1994 affect these variables through two channels: changes in the qualifying weeks of work and changes in the duration of unemployment once on claim. The authors first calculated the UI eligibility of each person in the pre-1994 reform sample. They then applied the new 1994 rules to this sample to determine what, in the absence of any behavioural changes, the impact of these changes would have been, had they been in place then. The resulting distribution of weeks of entitlement and benefit receipt were then compared to the results of the sample taken after the 1994 reforms were implemented. The differences shown by this comparison can then be attributed to behavioural changes on the part of workers in adapting to the new UI system under the 1994 reforms and perhaps other influences such as changes in macroeconomic conditions. Because macroeconomic conditions were relatively stable over the period of analysis, and given the nature of the 1994 rule changes, the authors believe this approach to be the appropriate one to capture the impact of the 1994 reforms on workers' behaviour.

Biographical Notes

Dr. Peter Kuhn received his Ph.D. from Harvard University in 1983. Since then he has taught at the University of Western Ontario and McMaster University, and has been a visiting professor at Princeton University, University College London, the London School of Economics, and the Australian National University.

Dr. Kuhn has published empirical and theoretical papers on several aspects of labour economics, including trade unionism, discrimination, immigration, displaced workers, unemployment, and employment contracts. He has served on the editorial board of the Canadian Journal of Economics, and is currently program director of the Canadian Employment Research Forum (an organization devoted to increasing interaction between policymakers and academic researchers) and principal investigator of the Canadian International Labour Network (an organization devoted to increasing the amount and quality of cross-national microdata research on labour markets involving Canada).

Dr. Arthur Sweetman received his Ph.D. from McMaster University in 1996, and is currently an Assistant Professor of Economics at the University of Victoria, in Victoria, British Columbia.

Dr. Sweetman studies labour markets and econometrics, and has done previous research on unemployment insurance, unions, discrimination, education, immigration and displaced workers. He is a senior researcher with the RIIM (Research on Immigration and Integration in the Metropolis) Centre of Excellence on Immigration, and is a member of the Western Research Network on Education and Training (WRNET a SSHRC sponsored initiative).