An Evaluation of the Impact of the 1997-98 Small Weeks Projects on EI Program and Labour Market Outcomes - October 2000

An Evaluation of the Impact of the 1997-98 Small Weeks Projects on EI Program and Labour Market Outcomes

View Whole Report

• Purpose
• Background
• Methodology and Data Sources
• Key Findings
• Biographical note

In 1997, HRDC put the Small Weeks Adjustment Projects in place to enable workers in 29 high-unemployment regions (above 10 percent) to take all available work without affecting their EI benefit rate. Phase I of the projects applied to certain claims filed after May 4, 1997 in selected high-unemployment regions, and phase II applied to certain claims filed after August 31, 1997 in the remaining high-unemployment regions. Both phases ended by November 15, 1998. The objectives of this study were to develop a theoretically justifiable model to determine (1) the nature and magnitude of any changes in work patterns in response to the Small Weeks Adjustment Projects, and (2) how the projects affected participants' incomes.

The formula for calculating the weekly benefits paid to EI recipients was changed in the move to an hours-based EI system. Under the new system, the weekly benefit rate is based on average weekly insured earnings in the rate calculation period (RCP), which is the 26 weeks preceding the last day of employment. Average weekly earnings are calculated by dividing total earnings during the RCP by the greater of the number of weeks worked or the minimum divisor. This formula creates a disincentive for individuals to accept weeks of work during the RCP, as it lowers their average weekly earnings and therefore their subsequent benefit levels.

The Small Weeks Adjustment Projects were designed to undo this disincentive. The projects' definition of a small week was any week with earnings less than $150. Both phases of the projects used two ways of calculating EI benefits to increase a claimant's incentive to accept small weeks of work. The bundling-small-weeks method allowed a worker to use all regular weeks and the best small weeks to bring the number of insurable weeks up to the minimum divisor, which was defined as equivalent to the number of 35- hour work weeks required in the region to qualify for EI benefits, plus 2. The remaining small weeks were bundled for the purposes of calculating the benefit rate. The excluding-small-weeks method was similar: The regular weeks and the best small weeks were used to meet the minimum divisor, but the remaining small weeks were excluded from the calculation of the benefit rate.

The study used administrative data supplied by HRDC. The evaluation primarily follows the comparison group methodology of nonexperimental program design. It compares individuals' labour-market activities when the projects were in effect with those when the projects were not in effect. When necessary, econometric techniques are used to remove the effects of all external factors.

Almost 1.5 million EI claims were filed in regions covered by the projects between their inception and November 14, 1998. Of these, 52 percent were filed in the period beginning May 4, 1997 in phase 1 regions, and 48 percent were filed in the period beginning August 31, 1997 in phase 2 regions. In total, 7.6 percent of male claimants participated in the Small Weeks Adjustment Projects, compared with over 14.6 percent of female claimants.

Table 1
Selected data means: all claims filed in regions and months when the projects were in effect.

Table 1 presents mean values of some important variables for participants and non-participants in the program period. Participants lived in regions with somewhat higher unemployment rates and worked fewer hours in the qualifying period than did non-participants. The variable "Previous" is defined as the number of weeks since June 1996 during which the individual received benefits. On average, project participants had spent a greater number of weeks collecting EI benefits in the recent past than had claimants in general. The profile of a typical beneficiary of the Small Weeks Projects therefore is a female claimant from a particularly high unemployment region, who has a history of previous benefit receipt and has worked approximately 25 percent fewer hours in the qualifying period than the average EI claimant.

Men and women's participation rates were seasonal, rising in the fall and early winter. Women's participation rate was greater in every month in the sample and had a more pronounced seasonal pattern than men's. With the exception of August, the participation rate was higher in 1998 than in comparable months in 1997.

The study concludes that the average female participant worked between 3 and 5.3 small weeks and that the average male participant worked 2.9 and 4.1 small weeks during the program periods. These may not be taken as the incremental impacts of the projects, without further investigation, for two reasons. First, in the absence of the projects, some individuals would still have accepted some small-weeks work. Second, a worker with some work weeks in which he or she earned more than $150 per week but less than his or her average weekly earnings might convert some of the below-average big weeks into small weeks during the program period. The contribution of this study is to sort out what part of the increase in small weeks worked could be attributed to the Small Weeks Adjustment Projects. The findings are as follows:

• Eleven percent of all EI claimants in the relevant period, benefited from the Small Weeks Projects.
• The participation rate of female claimants, 14.6 percent, was much higher than that of male claimants, which was 7.6 percent.
• The profile of a typical beneficiary of the Small Weeks Projects was a female claimant from a particularly high unemployment region, who had a history of receiving UI/EI benefits, and who worked approximately 25 percent fewer hours in the qualifying period than the average EI claimant.
• Participation rates were higher in 1998 than in 1997. Econometric analysis of the behaviour of the participation rate over time reveals a pattern that was consistent with gradual learning about, and adjustment to, the program.
• The evidence points to a significant increase in the number of small weeks worked in response to the projects. The econometric estimation shows that the average female participant worked between 3 and 5.3 additional small weeks, while the average male participant worked between 1.8 and 2.6 additional small weeks and converted an estimated 1.5 big weeks into small weeks.
• Taken together, the change in the rate calculation formula and the behavioural response to the program increased the income of the average male participant from $370 to $490 and that of the average female participant from $450 to $750.

The author concludes that the Small Weeks Adjustment Projects succeeded in encouraging a significant number of EI claimants to work additional weeks, during which earnings were low. By encouraging additional work and by eliminating the penalty for small weeks, the program generated increased incomes for many low-income, female claimants in high-unemployment regions.

There is some evidence that the average male participant reduced his working hours in certain below-average big weeks in order to increase his leisure and income. Given the very high effective marginal tax rates on earnings over $150 in below-average weeks, this is not surprising. It is a by-product of the $150 small-week cut-off point, and is an issue for future legislative fine-tuning.

Jane Friesen is Associate Professor of Economics at Simon Fraser University. She is a specialist in labour economics, with a particular interest in labour-market programs and labour legislation. Her previous research has examined the effect on the labour market of advance notice and severance laws, minimum wage legislation, overtime pay regulation, and several aspects of the Canadian unemployment insurance program.