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 formula for calculating the level of weekly benefits paid to EI recipients was changed as part of the move to an hours-based EI system in January 1997. Under the new system the weekly benefit rate is based on average weekly insured earnings in the rate calculation period (RCP), which is the twenty-six 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 that lower their average weekly earnings and therefore their subsequent benefit levels.

In 1997, HRDC introduced temporary benefit rate adjustment formulas that are intended to undo this disincentive. Under the Small Weeks Projects, claimants were allowed to exclude or bundle “small” weeks, during which earnings were less than $150, from the calculation of average weekly earnings. Over 11 percent of all claimants in regions included in the Projects worked at least one small week in the Rate Calculation Period, and therefore benefited from the program. Women were far more likely than men to benefit: 7.6 percent of male claimants participated compared to 14.6 percent of female claimants. The exact number of small weeks worked could not be measured directly because, although data on weekly earnings are available in unedited form, their preliminary nature made them unsuitable for this purpose. Instead, these important statistics are inferred from the difference between a claimant’s actual benefit rate and status quo rate. The results indicate that the average female participant worked between 3.2 and 5.3 small weeks, and the average male participant worked between 2.9 and 4.1 small weeks.

Participation in the program does not necessarily mean that individuals altered their behaviour in response to the program. Some small weeks might have been worked anyway. We refer to an increase in benefits that arises simply because a different formula has been applied to an unchanged work pattern as a “passive” increase in benefits; however, some behavioural response might be expected. The small weeks benefit formulas altered work incentives both by eliminating the disincentive to work small weeks, and creating a new incentive to convert below-average weeks when earnings exceed $150 to small weeks. Claimants may have responded to the first incentive by working additional weeks, and/or to the second incentive by converting below-average big weeks to small weeks.

Econometric evidence provided in this paper indicates that the average female participant would have worked no small weeks had the program not been in place, and did not convert any big weeks to small weeks. All of the between 3.2 and 5.3 small weeks worked by the average female participant appear to have arisen as an active increase in weeks worked in response to the program. All but 1.5 of the small weeks worked by the average male participant, that is, between 1.8 and 2.6 weeks, similarly appear to have arisen as an active increase in weeks worked. The remaining 1.5 weeks could have arisen passively, that is, they might have been worked anyway, or could reflect the conversion of big weeks to small. Given the very high effective marginal tax rates on small weeks in the pre-program period, and the almost equally high effective marginal tax rates on below-average big weeks during the program period, the latter interpretation is much more plausible. The program increased the income of the average female participant by an estimated $450 to $750, and of the average male participant by an estimated $495 to $615.

These findings indicate that the Small Weeks Projects succeeded in encouraging a significant proportion of EI claimants to work additional weeks. By encouraging additional work and by eliminating the penalty for small weeks that would have been worked anyway, the program generated increased incomes for these primarily low-income, female claimants in high unemployment regions.

While the Small Weeks Projects succeeded in attaining their stated objectives, the evidence strongly suggests that there has been an unintended side effect in addition. The average male participant appears to have reduced working hours in some below-average big weeks in order to increase both leisure and income. Given the very high effective marginal tax rates on earnings in excess of $150 in below-average weeks, this response is unsurprising. This unintended effect of the Small Weeks Projects should be monitored carefully in future.