Evaluation of Initiatives to Extend EI Regular Benefits

Evaluation approach

This evaluation report is primarily based on the technical study “Extended Duration of EI Regular Benefits: Evaluation Technical Study” (ESDC, 2015), which is a descriptive and econometric study of a sample of completed EI claims where only regular benefits were paid.

For both initiatives, the evaluation examined the take up by EI claimants of the additional weeks of benefits provided and the impact on the length of EI claims as well as on the likelihood of exhausting entitlements and benefits during the pilot periods.

An analysis of the total benefits paid associated to the additional weeks of entitlement was conducted for both initiatives. Although the assessment of the macro-economic impact of the EDB initiative was not in scope of this evaluation, the total benefits paid analysis provide an estimate of the direct spending stimulus of the EBD initiative.

In the case of Pilot 15, the evaluation explored the extent to which the pilot helped address the income gap issues related to seasonal workers in the pilot regions and the effectiveness of the termination trigger provision.

Evaluation findings

The evaluation provides key findings on the EDB initiative and Pilot Project 15.

The evaluation found EDB to be successful in that the initiative provided an estimated spending stimulus of 2.5 billion dollars between March 9, 2008 and September 11, 2010 for which 57% ($1.41 billion) is attributed to the additional EDB weeks paid and 43% ($1.09 billion) is attributed to increased use of regular entitlement among all EI claimants.

The EDB initiative resulted in 34% of regular claimants using additional benefits, of which 76% used the entire additional 5 weeks. In addition, the EDB initiative allowed all claimants, not only claimants who used the additional weeks, to remain on EI for 1.6 weeks longer on average and to reduce entitlement benefit exhaustion by 5.3 percentage points.

Results for Pilot Project No. 15 are mixed. Pilot Project No. 15 met its goal of reducing the number of seasonal workers facing an income gap with a 3.3 percentage point probability decrease and also led to a decrease of the income gap length of seasonal income gappers of 2.2 weeks on average. However, only 3.2% of the total 558 million of additional benefit paid during Pilot 15 period was used by seasonal income gappers. This indicates that although Pilot Project No. 15 improved the situation of seasonal gappers, the measure was too broad for the stated goal.

Furthermore, based on the facts that claimants' exhaustion of benefits and number of weeks used were similar before and after the trigger, the regional termination trigger included with Pilot 15 was found to be successful both in its timing and design.

Background

The testing of increasing the number of weeks of benefits available to some claimants began with EI Pilot Project No. 6. Pilot 6 was introduced for a two-year period in 2004 in 24 regions. The pilot was re-introduced as a new pilot project, Pilot Project No. 10, for a period of 18 months in 21 regions in 2006 and was extended in 2007 for a further period of 18 months until May 31, 2009. Both pilot projects tested the costs and impact of extending the number of weeks of benefits in selected economic regions. More specifically, they tested whether providing more weeks of benefits would reduce the number of seasonal workers facing an income gap with neither EI benefits nor employment income (referred to as "gappers"), and whether there would be associated behavioural effects. Pilot 10 ceased to have effect in February 2009 during the economic recession, when the EI Act was amended via the Budget Implementation Act, 2009 to introduce the Extended Duration of Employment Insurance Regular Benefits (EDB).

Extended Duration of Employment Insurance Regular Benefits (EDB)

The EDB was intended as a temporary stimulus measure as part of a broader response to the recession of 2009 and to reduce EI exhaustion rates during the recession. EDB temporarily provided additional support to unemployed workers facing transitions in tough economic times by providing claimants with five additional weeks of benefits. In addition, the maximum duration of benefits available under the EI Program in regions of high unemployment has been increased from 45 weeks to 50. The temporary measure came into effect on March 1, 2009, with extra weeks available to individuals who had an open claim on March 1, 2009, and were still unemployed. EDB concluded on September 11, 2010 and was followed seamlessly by Pilot 15.

EI Pilot Project No. 15

Pilot Project No. 15 was based on the same parameters and included the same economic regions as Pilot Project No. 10, and allowed further testing through a period of economic recovery. In addition, it tested the impact of a regional unemployment rate-based trigger, which removed some participating regions from the pilot when their regional unemployment rate fell below 8% for 12 consecutive months. More specifically, the pilot ended on September 24, 2011 in St. John's (Newfoundland and Labrador); on March 24, 2012 in Chicoutimi-Jonquière (Quebec) and on June 23, 2012 in Sudbury (Ontario). The pilot ran until September 15, 2012 and was allowed to sunset at that time.

Considerations

The EDB and subsequent Pilot 15 were introduced in different economic periods with different policy objective: the first one in a time of recession and the second one in a time of recovery. This could explain the broad assistance beyond the seasonal gappers that the evaluation rightfully points out for Pilot 15, but may also put the cost of the initiatives into perspective, as Pilot 15 was seen as a stimulus measure.

In addition, a number of EI pilots were in force during the EDB and Pilot 15 which were intended to change claimant's behavior both on and off claim. These include the Best 14 Weeks, Working While on Claim, and NERE pilots. While this report controls for all of these pilots, the changing parameters within each pilot complicates the analysis over time.

It is also important to remember other key differences between the EDB and Pilot 15. Specifically, the EDB was available nationally and increased the maximum insurable weeks from 45 to 50 weeks. The report does reflect these differences and this is only pointed out here for further emphasis.

Finally, the inclusion of an unemployment based trigger in Pilot 15's design was a first and the Skills and Employment Branch is pleased that the evaluation finds that termination trigger was successful in terms of its timing and design.

Key conclusions

The Skills and Employment Branch welcomes the findings in the report and since EDB and Pilot 15 both were temporary measures with different policy objectives, no overall recommendations are made.

The evaluation points out that both initiatives were successful in terms of reducing the entitlement exhaustion rate. However, this came at a cost as the increased use of benefits was observed among all claimants, not just among those who used additional weeks of benefits. As discussed in the considerations section, the wide reach of the assistance was part of stimulus package in response to a recession.

The Branch also appreciates the lessons learned from the evaluation with regard to the testing and evaluation of the unemployment rate based trigger.

1.1 Overview

This report presents the findings of the evaluation of the Extended Duration of Employment Insurance Regular Benefits and Pilot Project Relating to Extended Benefits (Pilot 15). These are two successive Employment Insurance (EI) initiatives geared towards increasing EI entitlement for Canadians filing regular benefit claims, each with different policy objectives.

Budget 2009 introduced several stimulus measures designed to soften the effect of the economic downturn on Canadians. A number of measures were geared specifically towards workers who experienced job separations during that period and were expected to experience job market impacts due to the decrease in labour demand and tightening labour market conditions. One of those measures was the EDB. Following the downturn, Pilot Project No. 15 was introduced to test the impact of extending EI benefits during a period of economic recovery for seasonal claimants who regularly face income gaps after exhausting their EI claim.

2.1 Extended duration of employment insurance regular benefits

The EDB was introduced and implemented nationwide as a response to the 2008/2009 economic downturn. Between May, 2008 and August, 2009, the overall Canadian unemployment rate increased from 6.3 to 8.7 percent. It was expected that the average duration of unemployment would likewise increase.

The initiative had two policy objectives: 1) to stimulate spending; and 2) to reduce EI exhaustion rate.

More specifically, the rationale for the initiative as stated in the 2009 Budget plan was to respond to a need for additional support in uncertain times, and to:

"…provide all regular EI clients with the reassurance that they will have financial support for a longer period while they find new employment, should it be required. It will also provide additional income support to unemployed Canadians who would otherwise have exhausted their benefits."^{Footnote 1}

EDB received Royal Assent on March 12, 2009 and came into effect on March 1, 2009. It extended EI entitlement by five weeks and increased the maximum allowable EI entitlement from 45 to 50 weeks.

The extended benefits applied retroactively to EI claims with benefit periods that did not end prior to March 1, 2009 (i.e. for claims established as early as March 9, 2008) to claims established up to September 11, 2010. Depending on their insurable hours of employment and the unemployment rate in their regions of residence; some claimants received the additional benefit entitlement after exhausting their original entitlement and prior to finding employment.

2.2 Pilot Project no. 15

Pilot Project No.15 was a two-year re-introduction of Pilot Project No 10. Pilot No.10 was established for the purpose of testing the costs and impact of extending the number of weeks of benefits in 21 economic regions of relatively high unemployment. More specifically, it tested whether providing five additional weeks of benefits would reduce the number of seasonal workers facing a gap between the exhaustion of their EI benefits and the resumption of their seasonal employment income and whether there would be associated behavioral effects.

Pilot Project No.15 was introduced from September 12, 2010 to September 15, 2012 based on the same parameters and included in the same 21 EI economic regions (see Appendix A – Pilot Regions), as the previous Pilot Project No.10 to allow for further testing during a period of economic recovery as well as to assess the impact of a regional unemployment rate-based automatic termination trigger. The automatic termination trigger provision excluded regions from the pilot if their unemployment rate was below 8% for 12 consecutive months.^{Footnote 2}

The three excluded EI regions and the date of exclusion are as follows.

• St. John's, excluded from Pilot 15 on September 24, 2011
• Chicoutimi-Jonquiere, excluded from Pilot 15 on March 24, 2012
• Sudbury, excluded from Pilot 15 on June 23, 2012

Pilot 15 terminated for the remaining 18 regions on September 15, 2012.

2.3 Evaluation approach

This evaluation report is primarily based on the technical study titled: Extended Duration of EI Regular Benefits: Evaluation Technical Study (ESDC, 2015). A comprehensive discussion on the methodologies as well as detailed econometric results can be found in the technical study. Also, for an overview of the methodology, see Appendix F – Data, Definitions and Methodology.

The econometric analysis used for this report takes into account and controls for a number of temporary EI measures in effect during the EDB and the Pilot 15 initiatives. These temporary EI measures included:

• The Extension of EI Regular Benefits for Long-Tenured Workers (EEILTW) which provided, in addition to the EDB, further entitlement in the number of weeks of regular benefits to claimants who had not drawn on benefits extensively in the previous five years, a group of claimants referred as long-tenured workers (LTW). The EEILTW provided up to 20 weeks of additional regular benefits for claims established between January 4, 2009, and September 11, 2010.
• Calculating Benefit Rate Based on Claimant's 14 Highest Weeks of Insurable Earnings Pilot Projects (Pilot Projects Nos. 7, 11 and 16) which were implemented between October 30, 2005 and June 23, 2012. Under the Best 14 Weeks pilots, the average weekly earnings were determined by taking the average of the highest or best 14 weeks of insurable earnings that fell during the qualifying period. If a claimant accepted additional weeks of work beyond the 14 weeks at a lower salary, the level of EI benefits would remain the same.
• Pilot Projects Providing Increased Access to Employment and Unemployment Benefits for New Entrants and Re-entrants (NERE) (Pilot 9 and 13) which were implemented between December 11, 2005 and December 4, 2010. NERE pilots tested the labour market impacts of decreasing the number of hours of insurable employment required for new entrants and re-entrants to the labour force to qualify for benefits. The NERE pilot reduced the EI entrance requirements for NEREs from 910 hours to 840 hours and encouraged the use of EI training benefits by marginal NEREs.
• Four different working while on claim (WWC) pilot projects (Pilot Projects Nos. 8, 12, 17 and 18) were implemented between July 5, 2005 and March 9, 2013. Pilot 8, 12 and 17 increased the maximum allowable earnings while on claim without a reduction to benefits from the greatest of $50 or 25% of the weekly benefit amount to the larger of $75 or 40% of the weekly benefit. Pilot 8 was implemented in 23 economic regions from December 11, 2005 to December 6, 2008, where Pilot 12 and 17 were implemented nationwide between December 7, 2008 and August 4, 2012. Under Pilot 18, claimants can keep 50 cents of their EI benefits for every dollar earned, up to 90% of the weekly insurable earnings. Pilot 18 was introduced on August 5, 2012 and concluded on August 1, 2015.

3.1 Extended duration of employment insurance regular benefits findings

3.2 Pilot project relating to extended benefits (pilot project no. 15) findings

Table E.1 Evaluation matrix (in text for web version)

F.1 Lines of evidence

A literature review was conducted to address questions related to the rationale of the EDB and Extension of EI Regular Benefits for Long-Tenured Workers (LTW) initiatives. The full list of questions regarding EDB and Pilot 15 evaluation can be found in Appendix E – Evaluation Matrix.

This evaluation report is primarily based on the technical study "Extended Duration of EI Regular Benefits: Evaluation Study 2014", which is a descriptive and econometric study of a sample of completed EI claims to which only regular benefits were paid.

Three previous technical updates to ESDC, 2015 were also submitted to support the annual EI Monitoring and Assessment Reports (MARs), which only examined the impact of EDB using EI administrative data. But due to reporting and initiative, timelines were limited to claims completed at the time.

F.2 Data source and restrictions

The administrative data used are from the Status Vector (SV) file, a derivative of the department's Benefits and Overpayments file which is used to administer EI claims, and the Record of Employment (ROE).

The SV includes information on claim applications, bi-weekly claimant reports, and EI Commission claim and benefits decisions. The data used for this analysis is based on a 10 percent sample of completed EI claims starting from June 5, 2005 to March 9, 2013. Claims receiving regular benefits alone were retained in the sample. Claims deemed to be LTW are excluded from the data sample when they received a larger entitlement than they would have under the EDB benefit schedule. In addition, due to the retroactive nature of the EDB initiative, claimants who initiated a claim between March 9, 2008 and November 16, 2008 and who could have experienced a gap in their benefit periods were excluded from the sample (30% of claims during EDB time period).

The ROE is completed by an employer when an employee stops working for that employer and contains information on the employment history of an individual. Only ROEs that were completed within 52 weeks after the end of the current claim have been considered.

F.3 Definitions

The study of extended week initiatives requires the use of terms and concepts that are not immediately obvious. The following are some of the key terminologies that are used throughout the report.

Profile

Pure regular claims: A claim where only regular benefits were paid. Other special benefits available through EI at the time include maternity, parental, sickness, and compassionate care leave benefits. Any claims with special benefits received are excluded from the sample.

First-time claims: No regular claims in the 5 year prior to the current claim.

Occasional claims: One or two regular claims in the 5 year prior to the current claim.

Frequent claims: Three or more regular claims in the 5 year prior to the current claim.

Seasonal claims: A subset of frequent claims i.e. those with 3 or more regular claims in the 5 years prior to current claim, all starting within the same eight week calendar period as the current claim.

Entitlement exhaustees: Claims for which all weeks of regular entitlements are paid.

Benefit period exhaustees: Claims that reach the final week of the benefit period before all regular benefits are paid.

Gappers: Entitlement exhaustees who experienced an income gap longer than 1 week between the exhaustion of their regular claim and their first day of work following 52 weeks of the current claim. An individual exhausting his or her claim on a Friday and starting a new job on the following Monday is considered to have a one week gap; this is due to the calculation of the week code variable in the administrative database. In order to avoid erroneously assuming these individuals experienced an income gap, only those claimants with an income gap longer than one week were considered to be gappers. Note that this study considers only seasonal gappers for impact analysis; accordingly gappers were extracted only to identify seasonal gappers.

Seasonal gappers: Gappers who had 3 or more regular claims in the 5 years prior to current claim, all starting at the same eight week calendar period as the current claim. The definitions of gapper and seasonal gapper are explained in Figure F2.

Image description

Figure 2 is a chart used to provide a definition of seasonal gappers (at the bottom of the chart). EI claims (at the top of the chart) are subdivided into pure regular claims and other EI claims. Pure regular claims are selected and subdivided into claims for which EI benefits were not exhausted and claims for which the benefits were exhausted. Only claims with exhausted benefits are relevant to identify gappers. These claims are classified into claims with an income gap of 1 week, claims with an income gap longer than 1 week within 52 weeks following the current claim and claims with an income gap surpasses 52 weeks following the current claim. Claims with an income gap longer than 1 week within 52 weeks following the current claim are gappers. Seasonnal gappers are gappers with 3 or more regular claims in the 5 years prior to current claim, all starting at the same eight week calendar period as the current claim.

F.4 Methodology

Groups of claims were selected prior to, during and after the EDB period from regions included in the extended weeks pilots (pilot regions) and those that were not (non-pilot regions). To assess the impact of the initiatives, the sequence of events is divided into four different time periods in order to control for the large changes in the economic environment. Table F.1 presents the sequence of events for these two groups.

Descriptive results on EI benefit weeks and the proportion of the extra entitlements used are obtained by claim and claimant characteristics. A linear model is adopted to analyze benefit duration changes and a logit model is used for probability of exhaustions (ESDC, 2015). The following table lists the covariates in the regression model.

Difference-in-difference regression methods (Wooldridge, 2010) are applied to measure the impact of the extended week initiatives. Difference-in-difference regression methods are applied over the four periods (Pilot 6 and 10, EDB, Pilot 15 and No-extension) to evaluate the impact of the extended week initiatives. The method compares the average change in EI outcomes in pilot regions with non-pilot regions before and after the policy change. In addition, a regression discontinuity design (Imbens and Lemieux, 2008) has been applied to evaluate the impact of the termination trigger in Pilot 15. A comprehensive discussion on these methodologies as well as detailed econometric results can be found in the study titled Extended Duration of EI Regular Benefits: Evaluation Technical Study (ESDC, 2015).