Le régime canadien d’assurance-emploi (AE) assure un soutien du revenu aux travailleurs admissibles qui ont perdu leur emploi. Il tente aussi d’aider les nouveaux chômeurs à garder contact avec le marché du travail pour qu’ils puissent retrouver plus facilement un emploi permanent. Ses dispositions concernant le Travail pendant une période de prestations (TPP) encouragent ainsi les prestataires à occuper un emploi occasionnel ou à temps partiel tout en touchant une partie de leurs prestations. L’objectif est ici de leur éviter de longues périodes de chômage, de les inciter à maintenir leurs compétences et leurs contacts professionnels, et de démontrer à d’éventuels employeurs leur détermination à travailler.
Colin Busby, Stéphanie Lluis et Brian McCall examinent dans cette étude si un emploi à temps partiel aide vraiment les prestataires de l’AE à trouver un travail permanent. Pour ce faire, ils analysent les données d’une série de projets pilotes lancés par Ottawa de 2005 à 2018 pour déterminer si des modifications aux règles de TPP inciteraient davantage de prestataires à accepter un emploi à temps partiel ou à travailler de plus longues heures. Ils passent également en revue les études qui ont évalué les effets de dispositions semblables en vigueur à l’étranger.
Les projets pilotes portaient sur deux modifications clés aux règles de TPP. La première, qui consistait à relever le seuil des gains admissibles sans réduction des prestations, a effectivement incité davantage de prestataires à accepter un emploi temporaire. La seconde, qui supprimait le seuil des gains et réduisait le taux de récupération des prestations sur tous les revenus, les a amenés à accepter des emplois à temps partiel aux heures plus longues et mieux rémunérés.
Mais comme le soulignent les auteurs, les données administratives de suivi des prestataires d’AE qui ont servi à évaluer ces projets portaient uniquement sur d’anciens prestataires ayant présenté une nouvelle demande d’AE. Par conséquent, l’évaluation reposait essentiellement sur les comportements de prestataires réitérants et saisonniers. On sait donc peu de choses sur ceux qui n’ont profité qu’une seule fois des règles de TPP et de l’utilité de celles-ci pour trouver un emploi permanent.
Des études menées dans d’autres pays qui appliquent des règles de TPP semblables, mais qui recueillent des données plus complètes, donnent un meilleur aperçu de leur capacité d’aider ou non les prestataires (y compris les demandeurs uniques) à se trouver un emploi permanent. En France, l’évaluation de ces dispositions a montré que la plupart des chômeurs qui avaient travaillé à temps partiel en période de prestations étaient plus susceptibles de retrouver un emploi permanent par après. Des études montrent qu’il en va de même en Belgique et en Allemagne pour les chômeurs de longue durée occupant un emploi occasionnel ou à temps partiel, notamment pour les femmes en Belgique. Des études menées ailleurs ont toutefois établi que des règles mal conçues risquent plutôt d’enfermer certains travailleurs dans un cycle d’emplois occasionnels ou à temps partiel.
Les règles de notre régime d’AE concernant le TPP peuvent aider les chômeurs à trouver un emploi permanent, concluent les auteurs, mais il faudrait les améliorer et les accompagner de nouveaux programmes pour ceux qui profiteraient peu d’un emploi occasionnel ou à temps partiel. Ils exhortent donc nos décideurs à rassembler des données de suivi sur tous les prestataires d’AE, en vue de réexaminer les règles de TPP dans le cadre d’une vaste évaluation qui pourra déterminer si elles facilitent effectivement le passage à un emploi permanent.
Les auteurs recommandent à cet effet de modifier les règles actuelles pour y réintégrer un seuil fixe de gains hebdomadaires admissibles sans réduction des prestations, tout en conservant un faible taux de récupération pour les gains dépassant ce seuil. Au Canada comme à l’étranger, les données montrent que ces règles modifiées inciteraient davantage de chômeurs à travailler en période de prestations et à accepter de plus longues heures de travail.
Ils exhortent aussi le gouvernement fédéral à prévoir des règles de TPP plus généreuses en période de ralentissement économique. Une recommandation d’autant plus pertinente à l’heure où l’économie se rétablit d’une pandémie qui a causé d’innombrables interruptions de travail. On compte généralement plus d’emplois à temps partiel qu’à temps plein en période de récession. Et pour inciter les salariés licenciés à garder contact avec le marché du travail, on devrait alors autoriser les prestataires à conserver temporairement une plus grande part de leur revenu d’emploi sans réduction de leurs prestations d’AE.
L’évaluation détaillée des dispositions concernant le Travail pendant une période de prestations d’AE permettrait d’établir clairement dans quelle mesure elles favorisent le retour au travail des chômeurs. De fait, elles pourraient profiter à certains mais être peu utiles à d’autres. Par exemple, environ la moitié des prestataires choisissent de ne pas travailler pendant leur période de prestations. C’est vraisemblablement le cas des travailleurs licenciés après de longues années de service qui ne voient guère l’intérêt d’accepter un emploi à temps partiel ou mal rémunéré. D’autres programmes comme l’assurance-salaire, qui subventionne la reprise d’un emploi à temps plein, seraient sans doute mieux adaptés à leur situation. L’instauration et l’évaluation de telles mesures seraient particulièrement bienvenues en cette période de reprise économique post-pandémie.
Globalization and technological advances cause job churn. New companies emerge and old ones become less competitive as modern economies grow and become more productive. Just over 2 million Canadian jobs (roughly 11 percent) were lost each year between 2001 and 2017, and a similar number of jobs were created. Although job churn is a common phenomenon, losing a job can be stressful for working adults. In many cases, it causes financial insecurity and a deterioration of employable skills. It can also stall career progression (Burdett, Carrillo-Tudela and Coles 2020). These concerns have taken on a whole new dimension in the context of the COVID-19 pandemic. A large proportion of Canada’s workforce has been displaced. In the first half of 2020, roughly 2.4 million Canadians were laid off or permanently lost their jobs. By January 2021, roughly 511,000 individuals had been unemployed for more than six months.
The immediate policy response for governments is to ensure that those who have suffered an unanticipated job loss can access temporary income support, typically through the Employment Insurance (EI) program. Although the program is there to provide financial support, it also strives to facilitate unemployed workers’ transition back to work. For many, this process can be long and difficult due to constraints on their mobility, their lack of required skills and limited information about available work.
In addition to training and job search support programs, there are specific provisions within EI that are meant to facilitate re-employment. Working-while-on-claim (WWC) provisions are designed to help claimants get back to permanent, full-time employment by allowing them to engage in part-time or casual work while continuing to collect part of their benefits. Holding part-time or casual work after a layoff has many benefits: workers can demonstrate an ongoing desire to work, maintain their social networks in the workforce and keep up their skills. Further, it allows individuals to continue to accumulate work experience, build self-confidence and hopefully avoid the most detrimental effects of long-term unemployment (Canada 2018). On average, just under half of EI claimants work for at least one week at some point during their claim.
Between 2005 and 2018, Employment and Social Development Canada (ESDC) conducted a series of pilot projects to test the effects of changes to the WWC provisions on the behaviour of claimants. The department wanted to see whether the changes would encourage claimants to work more while receiving benefits. Beyond these incentive effects, the use of these provisions raises important policy questions: do they facilitate transitions to permanent employment, or do they risk prolonging unemployment spells by reducing the incentive and time available for job search, pushing workers into a pattern of part-time work? This study examines the results of these pilot projects and looks at the experience of other countries that have similar measures.
In light of the Canadian and international evidence, we believe that, on average, working temporarily while collecting EI benefits improves the job prospects of displaced workers. We argue that better-designed EI provisions in Canada could encourage more claimants to take part-time jobs, facilitating their transition to permanent employment. We recommend that the provisions be made more responsive to economic downturns. This is especially important as the economy begins to recover from the COVID-19 pandemic. Policy-makers should act immediately to make the provisions more generous. This would encourage unemployed Canadians to access them at a time when the share of part-time employment in the economy is high. Policy-makers should also test new labour adjustment policies and evaluate them with improved data. For instance, wage insurance could be more effective than the EI work provisions in supporting the transition to permanent jobs for certain unemployed workers.
The Employment Insurance Act created provisions in 1996 that allowed claimants to earn up to $50 per week (or 25 percent of their weekly benefit rate, whichever was higher) in a part-time job, without facing a reduction in benefits. Earnings higher than this amount would result in a dollar-for-dollar reduction. The use of these provisions declined over time. From 1997 to 2005, the percentage of women who worked at least one week while on claim fell from 58 percent to 51 percent of all EI claims (figure 1). The decline was even sharper for men, dropping from 57 percent in 1997 to 48 percent in 2005. Also, the average number of weeks worked by all claimants decreased from over 10 weeks in 1997 to around 6 in the mid-2000s (figure 2).
In 2005, amid growing concerns that the EI work provisions were not as effective as they should be, ESDC launched a series of pilot projects that changed these rules. The pilot projects lasted until 2018 (table 1). The first project increased the allowable earnings threshold for working while on claim from $50 to $75 per week, or from 25 percent to 40 percent of weekly benefits, whichever was higher. Above this new threshold, benefits were reduced by one dollar for each dollar earned. The new threshold applied to 23 EI regions (of the 58 regions at the time) where the unemployment rate was 10 percent or higher. The government wanted to see if raising the allowable earnings threshold would encourage more part-time or casual work and lead to more permanent work after a claim.
Despite changes to the EI work provisions in these 23 regions, overall use of the provisions did not change much in 2006 and 2007. The average number of weeks worked while on claim continued to hover around the 2003-05 level of roughly seven weeks. Starting in December 2008 and lasting until July 2012, the same provisions were extended to all regions. Following this change, the average number of weeks worked while on claim jumped from 7.1 in 2008 to 12.1 in 2009 (figure 2).
However, the timing of these changes also coincided with an economic recession, which increased the number of people working on claim. Typically, economic downturns reduce the availability of full-time work. Many workers find themselves working part-time involuntarily because they cannot find a full-time job. The share of involuntary part-time work increases during recessions and decreases when the economy expands. The share of involuntary part-time work as a percentage of total part-time work grew from 19 percent in 2008 to 22.8 percent in 2009 during the last recession. It remained relatively high for many years. In August 2012, new provisions were introduced to encourage claimants to work more hours on claim. These new rules removed the allowable earnings threshold and reduced the clawback of EI benefits to 50 cents for each dollar earned, up to 90 percent of a worker’s insurable earnings for EI — earnings before taxes and deductions. Six months later, claimants were given the option to revert to the previous rules. In 2016, these provisions were extended until August 2018, at which point they were made a permanent feature of the program.
From 2013 to 2018, the average number of weeks worked on claim remained relatively stable at between 11 and 12 weeks (figure 2). In fiscal year 2018-19, the profile of EI regular claimants working while on claim remained similar to that in 2006 and 2007 (table 2). Just after the first pilot project was introduced, a slightly greater share of women (56.6 percent) than men (51.9 percent) worked while on claim. The same was true in fiscal year 2018-19, although the difference was less pronounced. Working while on claim continued to be relatively more common among claimants between the ages of 25 and 54 (table 2).
The industry breakdown for 2006 and 2007 shows that the largest shares of claimants working while on claim were typically in seasonal industries, such as fishing, logging, mining, construction and transportation, but also included manufacturing, education, and health and social services. The same pattern seems to hold in 2018-19: 56 percent of EI claimants who worked while on claim were in seasonal industries. They also remained more concentrated in coastal provinces. Less than 45 percent of claimants in Ontario used the EI work provisions, compared with more than 50 percent in Quebec, New Brunswick, Nova Scotia, Prince Edward Island and Newfoundland and Labrador.
The objective of the pilot projects was to change ei’s treatment of employment earnings to increase claimants’ take-home income (benefits plus employment earnings), thereby enhancing work incentives. Two elements came into play: (1) the allowable earnings threshold (also known as the disregard level), which is the point at which additional earnings start to reduce benefits; and (2) the rate at which benefits are reduced on each dollar earned, known as the clawback rate. In pilot projects from 2005 to 2012, benefits were not reduced for claimants who earned less than the threshold, but they were clawed back dollar for dollar on earnings above this amount. Under these rules, individuals looking to maximize their financial well-being were encouraged to work right up to the maximum allowable amount, but not beyond that. This is demonstrated by the flattening of the red line in figure 3, which shows the point at which a claimant’s total income would no longer increase were they to take additional work. Hence, we would expect a number of people to work up to this point, but no further.
Under the current rules (and those in place in the 2012-18 pilot projects), there is no earnings exemption threshold. Benefits are clawed back at a rate of 50 percent for each dollar earned. The blue line in figure 3 shows how a claimant’s total income grows with employment earnings under these provisions. It also shows that some claimants have less incentive to take a part-time job under current rules than they would have under a fixed threshold, because there are no exempt earnings. However, those who have earnings greater than that fixed threshold would be financially better off because their benefits would not be clawed back dollar for dollar. Under both the current and earlier provisions, weekly earnings can increase to a point where claimants no longer qualify for benefits and their total income comes from employment earnings. This is demonstrated most notably by the increase in the red line at around $770 of earnings, at which point total income comes exclusively from earnings. For an example of how these rules work in practice, see box 1 below.
The ESDC used econometric techniques to evaluate how changes to EI work provisions affect claimants’ behaviour. In the rest of this section, we summarize the findings of the evaluation of the initial pilot project, in effect from 2005 to 2008 (Lluis and McCall 2011), and present those from other ESDC evaluation studies of post-2008 pilot projects (Canada 2018).
The research indicates that the 2005 increase in the allowable earnings threshold for 23 high-unemployment regions was successful in increasing the number of claimants who worked while on claim and the average number of weeks they worked with partial benefits, especially among those working in lower-wage jobs. Those working while on claim tended to work up to the allowable earnings limit. Some claimants who worked so much that they received no benefits in a given week prior to the pilot project, subsequently worked less so that they would qualify for benefits under the higher threshold (Lluis and McCall 2011). Table 3 summarizes the changes in claimants’ behaviour as a result of the WWC provisions implemented under the various pilot projects.
Lluis and McCall (2011) found that the initial pilot project reduced the average number of weeks on claim by 1.2 weeks for men and 1.5 for women. Overall, the more generous threshold induced claimants to spend more of their claim period working. They also looked at the impact of an increased earnings threshold on job search efforts and concluded that it reduced the total number of hours spent looking for a job by 2.5 hours per week for men.
From 2008 to 2012, the higher allowable earnings threshold was extended to all regions that were not part of the initial pilot project. The effects on claimants working while on claim were similar to those found in the initial project: the higher threshold encouraged many to work more (while still remaining under the threshold) than they did under earlier rules (Canada 2018). Administrative data, which track repeat claimants over time, indicate that some claimants who worked so much under the prior threshold that their weekly benefits were fully clawed back, tended to work less under the higher threshold to qualify for some benefits.
Starting in 2012, new pilot projects brought major changes to EI work provisions. The threshold was eliminated and benefits were reduced on the first dollar of employment earnings at a rate of 50 cents for every dollar earned (up to 90 percent of weekly insurable earnings, at which point EI benefits were reduced by a dollar for every dollar earned). These changes were expected to encourage claimants, especially new ones, to work more hours on claim because they could continue to receive benefits at much higher levels of employment earnings, compared with earlier pilot projects. However, those who worked on claim in earlier pilot projects and earned less than the allowable earnings threshold could choose to stay with the old rules and keep all their employment earnings without losing any benefits.
Subsequent evaluations indicate that these changes encouraged new claimants (those who had not filed a claim prior to this period) to work more on claim at any level of earnings (Canada 2018). The changes did not affect job search efforts, the duration of unemployment or postclaim job characteristics (among the repeat claimants included in the data). Only 12 percent of eligible claimants chose to revert to the old rules. Claimants who were eligible to choose that option but did not were less likely to work while on claim and worked fewer weeks (between four and six weeks fewer) than they did under previous pilot projects. However, their decision to not revert to earlier rules was partly due to a lack of awareness. Less than half (47 percent) of eligible claimants knew they had this option. About one-third (33 percent) of those who did not revert would have received more benefits if they had done so (Canada 2018).
The main obstacle in evaluating whether EI work provisions facilitate a permanent return to work for EI claimants in general is that the administrative data provide only postclaim work information for those who go on to make a future claim. As a result, an outsized proportion of claims in the data are those of repeat users of EI and seasonal claimants. Despite making up less than one-third of overall claims in the past 15 years, seasonal workers make up well over half of those who use these provisions (56 percent in 2018-19).
Changes to the provisions through successive pilot projects were nonetheless successful in encouraging more claimants to work while on claim and in prompting claimants to take jobs offering more hours and higher earnings. Raising the allowable earnings threshold increased the probability of working while on claim and successfully drew claimants back into the labour force. Likewise, reducing the clawback rate on benefits encouraged many claimants to work at higher earnings (Canada 2018).
Unfortunately, we cannot evaluate the effectiveness of EI work provisions in facilitating a permanent return to work for claimants in general. Data limitations are such that we know little about the postclaim job outcomes of nonrepeat users of EI, and the evaluations of the pilot projects do not provide any information on the permanent jobs found by claimants. Somewhat encouragingly, however, the empirical evidence indicates that first-time claimants (who had only one claim over the last 10-year period) who work while on claim do not seem to develop a dependence on EI. Most do not file a subsequent claim, unlike seasonal or frequent claimants (Lluis and McCall 2011). Still, we do not know anything about the types of jobs they found, underscoring a clear need for more information on the job outcomes of nonrepeat claimants to help inform policy development on EI work provisions and other EI-related employment support programs. Collecting these data is entirely feasible.
To better understand the effects of these work provisions as pathways to permanent jobs, we turn to international studies that rely on more comprehensive data to analyze the employment outcomes (job type and job duration) of claimants after they end a claim. These research studies followed claimants’ postclaim transitions to the labour market over time, regardless of whether they later returned to unemployment insurance.
Other countries have implemented working while on claim provisions to promote part-time or casual work among claimants receiving unemployment insurance. Table 4 summarizes the basic parameters of these provisions in select OECD countries and the main research findings for each of these.
Unlike in Canada, the unemployment insurance system in the United States is administered primarily by the states, with work provisions that vary accordingly. In January 2019, 10 states had a fixed allowable earnings threshold (a disregard level), which ranged from a low of US$30 per week in Arizona to a high of US$150 per week in Hawaii. In these states, weekly earnings below the threshold do not affect benefit amount, whereas any earnings above the threshold result in an equal reduction in the benefit amount.
For 19 states, the allowable earnings threshold is set as a fraction of an individual’s weekly benefit amount, ranging from 20 percent in New Mexico, North Carolina, Ohio and Rhode Island, to 60 percent in North Dakota. For someone in North Dakota who has a weekly benefit amount of US$500, the allowable earnings threshold would be US$300, while for the same person in Rhode Island, it would be US$100. Eight states have thresholds that are set as the maximum between a flat rate and a fraction of the weekly benefit amount.
McCall (1996) examined the effects of changes to earnings thresholds within a state over time and found that raising the threshold increased the part-time re-employment rate during the first three months of joblessness. Increasing the earnings threshold shortened the duration of unemployment spells and increased the probability of claimants taking part-time work, which suggests that short-duration, part-time work could reduce the risk of long-term unemployment. However, there is evidence in Canada that working part-time while on claim can reduce search efforts for full-time employment (Lluis and McCall 2019). Fewer transitions from part-time to full-time work may be expected if the search for part-time employment takes away from efforts to find a full-time job (Ek and Holmlund 2015).
Working while on claim rules vary across Europe. Countries set the allowable earnings threshold either as a fixed amount, a fraction of the weekly benefit, a fraction of weekly earnings or as a combination of these measures. The following are a few examples:
International evaluations of these programs generally look for evidence of two things: (1) a “lock-in” effect, whereby those who accept part-time or casual employment during a claim are more likely to maintain this type of work arrangement beyond the claim, falling into a pattern of part-time casual work; or (2) a “stepping-stone” effect, whereby those who work on claim are more likely to become re-employed in a full-time job. The empirical evidence demonstrates that provisions that enable claimants to keep at least some benefits while working part-time improve their likelihood of finding a permanent job (table 4).
Much of the international evidence comes from Nordic countries. Studying work provisions in Denmark, Kyyrä, Parrotta and Rosholm (2013) find a lock-in effect as reflected by longer unemployment duration for young workers. But they also find a stepping-stone effect (along with shorter unemployment duration) for married women. In Finland, working for a short duration in full-time jobs while on claim facilitates transitions to regular employment. Also, part-time work with partial unemployment benefits may help men (but not women) in finding a regular job afterward (Kyyrä 2010). In Norway, research concludes that allowing claimants to work part-time with partial benefits reduces the transition time to permanent employment, with higher transition rates into regular employment (and high-quality jobs) during the first month of part-time employment. More than 80 percent of workers transition from part-time work to regular work with the same employer. This suggests that employers use part-time work as a screening tool (Godøy and Røed 2016).
Evidence for Germany, which has a flat earnings threshold and reduces unemployment benefits one-for-one on earnings above this amount, suggests that taking up “mini-jobs” (jobs that are exempt from social security contributions) increases the likelihood of claimants becoming re-employed in stable full-time jobs (Caliendo, Künn and Uhlendorff 2016). Cockx, Goebel and Robin (2013) look at a sample of long-term unemployed women in Belgium and find no evidence of a lock-in effect, but a significant stepping-stone effect.
Auray and Lepage-Saucier (2021) find no evidence of a lock-in effect for the French program. However, they do find robust evidence of a stepping-stone effect and they find that it is stronger among the unemployed (who start with weaker labour force attachment and therefore have the most to gain). This group includes the long-term unemployed, younger and older workers, and those who had worked fewer hours prior to becoming unemployed.
Fremigacci and Terracol (2013) also investigate France’s program and find evidence of both a lock-in effect while people are in part-time work and a stepping-stone effect later in their claim, which dominates the lock-in effect. This suggests that, when people take part-time or casual work early in their claim, they tend not to look for permanent work. However, later in their claim, perhaps thanks to having worked part-time, they are much more likely to find permanent work compared with workers with similar profiles who chose not to work on claim.
Kyyrä, Arranz and Garcia-Serrano (2017) find a similar pattern in Spain. They find that the exit rate to full-time work declines when claimants are working part-time (lock-in effect) but increases afterward (stepping-stone effect). Overall, people who take part-time work transition to full-time employment faster than individuals who do not.
One Austrian study finds no evidence of a stepping-stone effect from working a part-time job while receiving unemployment benefits. It concludes that part-time work tends to prolong the duration of spells on unemployment insurance (Eppel and Mahringer 2019). However, under the work provisions in Austria, the cost of transitioning from a part-time job to more permanent employment is particularly high because having any earnings above the threshold leads to the complete loss of unemployment benefits.
These country-specific experiences show that the design of the work provisions matters.
In Canada, the high number of seasonal claimants who use WWC provisions suggests that some may be encouraged to rely on unstable, part-time work combined with short, intermittent periods on EI. Indeed, there are indications some individuals make use of the provisions as part of an annual cycle of seasonal work and EI spells to increase their income when on claim. This is not desirable if it only delays these claimants’ transition to more stable jobs and artificially props up local economies that are dependent on seasonal work. It is critical that ESDC use a more comprehensive data set of claimants to evaluate the extent to which WWC provisions improve permanent employment outcomes for users, and adjust policy accordingly. This data-gathering effort should be coupled with new pilot projects that revise the provisions and examine the postclaim employment patterns of those who work while on claim.
Notwithstanding this information gap, there is a substantial body of research at the international level that allows us to draw some conclusions. Overall, the evidence suggests that taking part-time work during a period of unemployment can improve the ability of workers to transition to permanent work over the duration of the claim. The range of results across countries, however, shows the importance of setting appropriate work provisions to create desirable outcomes.
Canada’s experiments with WWC provisions demonstrate that individuals respond strongly to rule changes. However, it is also important to keep in mind that over half of EI claimants never work while on claim. This could be due to various factors, including a lack of information about the program, high expectations of finding a full-time job or because claimants do not perceive a part-time job as a stepping stone to more permanent work.
Domestic and international research has shown that having a fixed threshold of allowable earnings encourages a large number of claimants to take part-time or casual work, but also creates a strong disincentive to work beyond the threshold. As a result, many claimants work up to, but not beyond, the earnings limit. Further, a fixed threshold with a high clawback rate on benefits for earnings above it may have other undesirable consequences. These include claimants taking underground jobs and falsely reporting or concealing earnings (Fuller, Ravikumar and Zhang 2012, 2015). In contrast, Canada’s current provisions (no fixed threshold and a 50 percent clawback rate on every dollar earned) mean claimants with part-time work that offers many hours of work, or high pay, benefit from this design. But many will be discouraged from taking a part-time job in the first place if there is no fixed earnings exemption. This is especially true for workers displaced from low-earnings jobs, who would benefit the most from maintaining some level of labour force attachment (Yagan 2019).
We encourage policy-makers to evaluate a design that blends the two provisions, which means setting a fixed earnings exemption threshold and a modest clawback rate on earnings above the threshold. This would combine the positive work incentives of a fixed earnings threshold with the encouragement to work more hours under a modest clawback rate. Canadian and international evidence suggests that, under these proposed rules, more people would be encouraged to work while on claim and would not be discouraged from working additional hours. The red line in figure 4 illustrates the parameters we propose. Under these rules, claimants working part-time could earn up to 40 percent of weekly benefits without penalty. For every dollar in earnings above the threshold, benefits would be reduced by 65 cents (up to 90 percent of weekly insurable earnings, after which benefits would be reduced dollar-for-dollar).
A lack of awareness about WWC provisions also seems to be an issue. Many claimants in earlier pilot projects who had the option to revert to rules that would have increased their take-home income did not to do so because they did not know they had this option. The government of Canada should find new ways to increase awareness about these provisions, perhaps by incorporating automatic reminders about them during the required, regular check-ins with administrative staff over the course of an EI claim.
Moreover, setting monthly or annual allowable earnings limits instead of weekly ones might encourage work patterns that are more effective as stepping-stones to full-time permanent employment. An annual earnings threshold or one set according to the duration of the claim would give claimants more flexibility to take full-time work for a short period of time without cutting off access to benefits. This could facilitate career adjustment and provide the time needed to evaluate the suitability of a job match with a new employer before the end of a claim.
Several academic studies have found that losing a job during a major recession, especially for long-tenured workers, resulted in large and persistent earnings losses and reduced life-time earnings (Couch and Placzek 2010; Davis and Von Wachter 2011; Von Wachter 2020). Job loss can also lead to a deterioration of employable skills, as well as stalled career progression (Burdett, Carrillo-Tudela and Coles 2020). During a recession, full-time jobs decrease much faster than part-time jobs, which tend to grow as a proportion of overall employment. As a result, the number of people working on claim tends to increase during these periods and in the years soon after. Policy-makers should consider setting a higher earnings exemption threshold and a lower clawback rate during economic downturns to further encourage use of WWC provisions during these periods.
Across Canada, EI administrative rules vary by EI region. The WWC provisions could also differ by EI region. For instance, claimants in regions experiencing an economic downturn could be allowed to retain more employment earnings by temporarily increasing the allowable earnings threshold and decreasing the benefits clawback rate. As the economy picks up and more full-time jobs became available, the provisions could revert to the prior set of rules.
Figure 5 illustrates the hypothetical total income of an EI claimant working while on claim during a recessionary period under this recommendation (red line), compared with the current rules (blue line). More responsive WWC provisions would encourage unemployed workers to stay attached to the labour force by allowing them to keep more EI benefits when working during downturns. This should not only boost their take-home income but, more importantly, improve their job prospects when the economy recovers.
Despite the extensive use of EI work provisions in Canada, half of claimants choose not to work on claim. This could be because workers with a history of full-time employment may not see low-paying, short-duration, part-time or casual work as an attractive step toward permanent, full-time employment. For these workers, alternative programs could be more effective in facilitating a permanent return to the labour force. We encourage the federal government to develop a greater mix of policies to that effect.
Both in Canada and abroad, job loss imposes disproportionately high costs on long-tenured workers. Canadian evidence shows that, five years after a job loss, a little over 50 percent of displaced, long-tenured workers had not attained earnings within 10 percent of what they previously earned (Morissette and Qiu 2020). International evidence has shown that long-tenured workers on unemployment insurance were more likely to drop out of the labour force than to resume employment when their benefits expired (Card, Chetty and Weber 2007).
In addition to encouraging the unemployed to work while on claim, several countries help displaced workers find permanent work through a broad mix of worker retraining or job search programs, as well as with instruments like wage insurance. Wage insurance allows qualifying beneficiaries to accept new, full-time employment at a reduced wage, while receiving benefits that partially cover the difference between the wage they earned prior to being laid off and their current wage. Unlike work provisions that subsidize part-time work when on unemployment insurance, wage insurance subsidizes and encourages the take-up of full-time work. This option deserves more attention from policy-makers. Temporarily topping up wages in a new job could provide an incentive for claimants to increase their job search efforts and could help prevent some of these workers dropping out of the labour force when their claim ends. See box 2 for a discussion of other policy options that address the re-employment needs of long-tenured workers.
Involuntary job loss can lead to substantial and persistent earnings losses. EI provides financial support for many Canadian workers suffering job losses and work provisions encourage them to stay engaged in the job market by allowing claimants to take part-time work while keeping some of their benefits. Working part-time is expected to help claimants to maintain their skills, their confidence and their workplace contacts. Canada’s policy-makers lack a comprehensive evaluation of the extent to which EI work provisions encourage part-time work and support the transition to permanent work. The postclaim job outcomes reported in the administrative data come disproportionately from seasonal workers. Nevertheless, the available evidence indicates that the benefits of working while on claim outweigh the negatives. Indeed, the international experience with work provisions is mostly positive. However, Canadian policy-makers must collect more comprehensive data to evaluate how the provisions can be improved for all claimants.
Once there are more comprehensive data, we recommend that the federal government modify the WWC rules to encourage more claimants to take part-time work and to work more hours. This can be done by blending the elements of current and previous rule, setting a weekly allowable earnings threshold of a fixed amount with a modest clawback for each dollar of earnings above the threshold.
We also propose making the EI work provisions more generous during recessionary periods to allow claimants working part-time jobs to earn more without reducing their benefits. This is especially important now while the world is in the throes of the global COVID-19 pandemic, which is causing massive job displacement. The objective should be to ensure that unemployed workers who can only access part-time work remain active in the labour force while they seek full-time positions. Prior rules could be reinstated when the economy recovers.
Additional policies should also be introduced to support displaced workers who do not want to work part-time while on claim. This is likely the case for long-tenured workers for whom other measures such as wage insurance could be a more effective way of encouraging a return to permanent full-time employment. The government should experiment with a variety of options and evaluate how well they support transitions to permanent work. Gathering more information on postclaim outcomes is essential in this regard.
EI exists not only to provide income security in the event of job loss, but also to help workers transition to new, permanent jobs. WWC provisions help many recently unemployed Canadians keep a foot in the job market and can provide a stepping stone to permanent employment, but they are used mainly by seasonal and frequent claimants. EI work provisions should be improved, better data should be developed to fully evaluate them and additional support programs should be created to avoid the damaging effects of long-term unemployment on Canadian workers.
 Statistics Canada, “Business sector employment flow rates by firm size, provinces and the territories,” table: 33-10-0090-01 (formerly CANSIM 527-0010).
 Statistics Canada, Public Use Microdata File (PUMF) Collection, https://www.statcan.gc.ca/eng/microdata. In comparison, at its peak after the financial recession, long-term unemployment stood at around 335,000.
 Temporary income support programs had to be created in the past year to quickly help laid-off workers who were not eligible to collect EI based on existing criteria or could not access EI due to backlogs in processing applications. Although a number of these temporary programs, including the Canada Recovery Benefit, have been extended until September 2021, we expect that the EI program will continue to be the mainstay of the social safety net as the labour market situation improves.
 For administrative purposes, EI sets different qualification criteria and benefit durations according to the geographic region where one resides. The boundaries of regions are set according to labour market conditions, with the aim of keeping them comparable to other administrative regions. Revisions to EI regions are considered once every five years. Before 2014, there were 58 regions. The number rose to 62 in October 2014 and to 64 in 2016.
 The duration of pilot projects under EI was capped at three years, prompting either renewal of earlier parameters or the creation of new ones.
 A similar slow recovery was noted in the United States (Valletta, Bengali and Van der List 2020).
 The option to revert to prior allowable earnings provisions was only available to those who previously chose to use the alternate earnings rule for an EI claim. It is scheduled to be eliminated as of August 14, 2021. Also, in 2018, the allowable earnings threshold for working while on claim was extended to claimants on maternity leave, those on EI sickness benefits and eligible self-employed persons under these two categories of claimants.
 The option to revert to earlier work provisions can only be used by those who did so during earlier pilot projects. It will expire after August 14, 2021.
 The higher threshold also reduces the likelihood of deferring benefits, which is especially relevant for claimants who work on claim but earn so much that they receive no benefits. If earnings while on claim are large enough to reduce weekly benefits to zero, then the week of benefits is deferred and may be paid later. If earnings while on claim are not great enough to reduce weekly benefits to zero, then a reduced amount is paid, and the remaining initial benefit entitlement period is reduced by one week.
 The pilot project significantly increased the likelihood that claims ended when the 52-week benefit period ended (Gray and Leonard 2020). This means that those working while on claim would run out of available weeks to receive their benefit entitlements.
 Gray and de Raaf (2002) used data from Statistics Canada’s Survey on the Repeat Use of Unemployment Insurance to study WWC behaviour in Canada. They found that that those who are the most likely to report earnings while on claim tend to be those with the highest frequency of past EI claims, suggesting that those who rely on EI most frequently know best how to tailor their work patterns around the provisions.
 ESDC studies of other programs have used a combination of administrative data sets, taking advantage of the links between the administrative data from the EI claims and from the Canadian Revenue Agency (the Longitudinal Administrative Database). While the tax data are longitudinal and individuals can be followed over several years, they do not contain information on individuals’ hours of work and type of employment (full-time versus part-time or temporary employment). However, that key source of information for post-claim employment outcomes is available in the Labour Force Survey or in the census data, which have already been linked to the tax data. Creating these broad data linkages is what made it possible to evaluate the stepping-stone impacts of work provisions in other countries and can be done for Canada.
 Work provisions at the state level differ from those in Canada in that some benefits can be banked for future use when weekly earnings are above the allowable earnings threshold and benefits are reduced (but not to zero). Take for example an individual who receives unemployment insurance benefits of $300 per week and has an offer for a part-time job that would reduce weekly benefits by $100 per week. In Canada, if the person accepted that job, they would receive $200 in benefits and their remaining weeks of eligibility would be reduced by one. The $100 reduction in benefits cannot be banked for potential future use. In many states, however, the $100 can be banked for use within the individual’s benefit year (the 52-week period that follows filing a claim).
 Only New York adjusts benefit amounts based on a formula that does not use a disregard level (except that individuals cannot earn more than the maximum weekly benefit amount). In New York, the weekly benefit amount is adjusted based on the number of days worked in a week. An individual’s weekly benefit is reduced by 25, 50, 75 or 100 percent depending on if the individual works 1, 2, 3, 4 or more days in a week.
 In the remaining states, the amount of exempt earnings is either a fraction of weekly earnings or some combination of a flat amount and a fraction of weekly earnings. For example, in Michigan and Minnesota, one-half of weekly earnings is exempt. In Alaska, 25 percent of weekly earnings exceeding US$50 is exempt.
 France has another program, called the “intermittents du spectacle” that allows for part-time work while on claim for artists and technicians in the entertainment industry. The generous provisions for this group of workers provide a cautionary tale because they have encouraged a cycle of dependence for a number of the workers in this sector (Cahuc 2018). Eligible workers have little incentive to work beyond the minimum number of hours required to qualify, encouraging an indefinite pattern of low earned income supplemented with unemployment benefits.
 Benghalem, Cahuc and Villedieu (2020) report their findings on an experiment that consisted of providing detailed information on the work provisions to a random group of new and relatively younger unemployed individuals who are first-time claimants of unemployment insurance in France. Compared with the control group, the propensity to work while on claim and receive partial benefits increased by 6 percent. This shows that a lack of information is a small, but non-negligible, factor in explaining the use of work provisions and why it could be higher. The authors also find that unemployment duration increased following the use of the provisions, which they interpret as evidence of a lock-in effect.
 One anonymous reviewer of an earlier draft of this paper pointed out that this proposal and our earlier recommendation to change the existing EI work provisions are likely to increase the costs of EI. Although we agree that this is likely true in the short run, we are not convinced that these revisions would increase the total costs of EI in the long run. If our proposed provisions worked more effectively than the current rules, the average duration of claims should decrease, and claimants would be less likely to make a future claim should they find high-quality employment.
 Under the labour market development agreements, Canada’s EI program delivers the Employment Benefit and Support Measures. The benefits and measures include employment assistance services, skills development and targeted wage subsidies. A study by Handouyahia et al. (2016) evaluates the impact of the measures on the employment and earnings outcomes of eligible participants and finds positive effects among those who voluntarily decided to complete the program.
Auray, S., and N. Lepage-Saucier. 2021. “Stepping-Stone Effect of Atypical Jobs: Could the Least Employable Reap the Most Benefits?” Labour Economics 68: 101945.
Benghalem, H., P. Cahuc and P. Villedieu. 2020. The Lock-in Effect of Part-Time Unemployment Benefits. IZA Discussion Paper 14189. Bonn: IZA Institute of Labor Economics. https://www.iza.org/publications/dp/14189/the-lock-in-effects-of-part-time-unemployment-benefits
Bloom H.S., S. Swartz, S. Lui-Gurr, S-W. Lee, J. Peng and W. Bancroft. 2001. “Testing a Financial Incentive to Promote Re-Employment among Displaced Workers: The Canadian Earnings Supplement Project (ESP).” Journal of Policy Analysis and Management 20 (3): 505-23.
Burdett, K., C. Carrillo-Tudela and M. Coles. 2020. “The Cost of Job Loss.” Review of Economic Studies 87 (4): 1757–98. https://doi.org/10.1093/restud/rdaa014
Cahuc, P. 2018. “France: Social Protection for the Self-Employed.” In The Future of Social Protection: What Works for Non-Standard Workers? Paris: OECD, 99-122.
Caliendo, M., S. Künn and A. Uhlendorff. 2016. “Earnings Exemptions for Unemployed Workers: The Relationship Between Marginal Employment, Unemployment Duration and Job Quality.” Labour Economics 42: 177-93.
Canada. 2018. Employment and Social Development Canada (ESDC). Evaluation of Working While on Claim Pilot Projects. https://www.canada.ca/content/dam/esdc-edsc/documents/corporate/reports/evaluations/Evaluation_WWC-EN.pdf
——————. 2020. Canada Employment Insurance Commission (CEIC). Employment Insurance Monitoring and Assessment Report 2018-2019. https://www.canada.ca/en/employment-social-development/programs/ei/ei-list/reports/monitoring2019.html
Card, D., R. Chetty and A. Weber. 2007. “The Spike at Benefit Exhaustion: Leaving the Unemployment System or Starting a New Job?” American Economic Review 97 (2): 113-18.
Cockx, B., C. Goebel and S. Robin. 2013. “Can Income Support for Part-Time Workers Serve as a Stepping-Stone to Regular Jobs? An Application to Young Long-Term Unemployed Women.” Empirical Economics 44 (1): 189-229.
Couch, K. A., and D. W. Placzek. 2010. “Earnings Losses of Displaced Workers Revisited.” American Economic Review 100 (1): 572-89.
Davis, S. J., and T. M. Von Wachter. 2011. Recessions and the Cost of Job Loss. No. w17638. National Bureau of Economic Research.
Ek, S., and B. Holmlund. 2015. “Part-Time Unemployment and Optimal Unemployment Insurance.” International Tax and Public Finance 22 (2): 201-23.
Eppel, R., and H. Mahringer. 2019. “Getting a Lot out of a Little Work? The Effects of Marginal Employment during Unemployment.” Empirica 46 (2): 381-408.
Fremigacci, F., and A. Terracol. 2013. “Subsidized Temporary Jobs: Lock-In and Stepping Stone Effects.” Applied Economics 45 (33): 4719-32.
Fuller, D.L., B. Ravikumar and Y. Zhang. 2012. “Unemployment Insurance Fraud.” Economic Synopses 28. doi.org/10.20955/es.2012.28
——————. 2015. “Unemployment Insurance Fraud and Optimal Monitoring.” American Economic Journal: Macroeconomics 7 (2): 249-90.
Godøy, A., and K. Røed. 2016. “Unemployment Insurance and Underemployment.” Labour 30 (2): 158-79.
Gray, D., and S. de Raaf. 2002. The Impact of the Allowable Earnings Provision on EI Dependency. Social Research and Demonstration Corporation.
——————., and P. Leonard. 2020. Mind the Gap: Running Out of Employment Insurance Benefits. IRPP Study 81 (November). Montreal: Institute for Research on Public Policy.
Handouyahia, A., S. Roberge, Y. Gingras, T. Haddad and G. Awad. 2016. “Estimating the Impact of Active Labour Market Programs using Administrative Data and Matching Methods.” Paper presented at the Statistics Canada Symposium 2016, Growth in Statistical Information: Challenges and Benefits.
Kyyrä, T. 2010. “Partial Unemployment Insurance Benefits and the Transition Rate to Regular Work.” European Economic Review 54 (7): 911-30.
Kyyrä, T., J. M. Arranz, and C. Garcia-Serrano. 2017. “Does Part-Time Work Help Unemployed Workers to Find Full-Time Work? Evidence from Spain.” IZA Discussion Paper 10770. Bonn: IZA Institute of Labour Economics.
Kyyrä, T., P. Parrotta and M. Rosholm. 2013. “The Effect of Receiving Supplementary UI Benefits on Unemployment Duration.” Labour Economics 21: 122-33.
Lluis, S., and B.P. McCall. 2008. Evaluation of the Impact of the Increase in EI Allowable Earnings Pilot Project on Working While on Claim and Job Search Behaviour in Canada. Report prepared for Employment and Social Development Canada. Ottawa: Employment and Social Development Canada.
——————. 2011. “Evaluation of the Impact of the Increase in EI Allowable Earnings Pilot Project on Working While on Claim and Job Search Behaviour in Canada.” Working Papers 1106, University of Waterloo, Department of Economics, revised December 2011. https://econpapers.repec.org/paper/watwpaper/1106.htm
——————. 2019. “Employment and Job Search Implications of the Extended Weeks and Working While on Claim Pilot Initiatives.” Canadian Public Policy 45 (2): 129-72.
McCall, B.P. 1996. “Unemployment Insurance Rules, Joblessness, and Part-Time Work.” Econometrica 64 (3): 647-82.
Morissette, R., and T.H. Qiu. 2020. Turbulence or Steady Course? Permanent Layoffs in Canada, 1978-2016. IRPP Study 76. Montreal: Institute for Research on Public Policy.
Robins, P. K., and R.G. Spiegelman, eds. 2001. Reemployment Bonuses in the Unemployment Insurance System: Evidence from Three Field Experiments. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research.
Valletta, R., L. Bengali and C. van der List. 2020. “Cyclical and Market Determinants of Involuntary Part-Time Employment.” Journal of Labor Economics 38 (1): 67-93.
Von Wachter, T. 2020. “The Persistent Effects of Initial Labor Market Conditions for Young Adults and Their Sources.” Journal of Economic Perspectives 34 (4): 168-94.
Yagan, D. 2019 “Employment Hysteresis from the Great Recession.” Journal of Political Economy 127 (5): 2505-58.
This study was published as part of the Social Safety Net for Working-Age Adults research program, under the direction of Colin Busby and France St-Hilaire. The manuscript was edited by Rosanna Tamburri and copy-edited by Madelaine Drohan, proofreading was by Zofia Laubitz, editorial coordination was by Francesca Worrall, production was by Chantal Létourneau and art direction was by Anne Tremblay.
Colin Busby is a research director at the Institute for Research on Public Policy, where he leads research programs on the Social Safety Net for Working-Age Adults, and Faces of Aging. He holds a BComm from the University of Alberta and an MA in economics from the University of Ottawa.
Stéphanie Lluis is a professor of economics at the University of Waterloo. She worked for five years at the University of Minnesota Industrial Relations Center. Since 2016, she has been the director of the SouthWestern Ontario Research Data Centre. Her areas of specialization are labour economics and personnel economics. Her work is published in the Journal of Labor Economics, Industrial and Labor Relations Review and the Journal of Economic Behaviour and Organizations, and her latest publications are in the Canadian Public Policy Journal. She has a PhD from the Université de Montréal.
Brian P. McCall is a professor of education, economics and public policy at the University of Michigan Ann Arbor. He is a former co-editor of Economics of Education Review. His current research interests include the incentive effects of unemployment compensation; the effect of scholarships on college persistence, completion, and earnings; the effect of college enrolment delay on college completion and the impact of the great recession on educational attainment and earnings. He received his PhD in economics from Princeton University in 1988.
To cite this document:
Busby, Colin, Stéphanie Lluis and Brian McCall. 2021. Transitioning Back to Work: How to Improve EI Working-While-on-Claim Provisions. IRPP Study 85. Montreal: Institute for Research on Public Policy. DOI: https://doi.org/10.26070/g52y-ac79
The opinions expressed in this study are those of the authors and do not necessarily reflect the views of the
IRPP or its Board of Directors.
IRPP Study is a refereed monographic series that is published irregularly throughout the year. Each study is
subject to rigorous internal and external peer review for academic soundness and policy relevance.
ISSN 1920-9436 (Online)
ISSN 1920-9428 (Print)
Montréal — La pandémie a provoqué en un an d’innombrables licenciements et interruptions de travail. Si bien qu’en mai 2021, plus de 478 000 Canadiens étaient au chômage depuis plus de six mois. Pour les aider à retrouver un emploi à temps plein, le gouvernement fédéral devrait bonifier les règles de Travail pendant une période de prestation (TPP) de l’assurance-emploi (AE), soutiennent les auteurs d’une nouvelle étude de l’Institut de recherche en politiques publiques.
Les règles de TPP autorisent les prestataires à occuper un emploi à temps partiel ou occasionnel tout en touchant une partie de leurs prestations. Elles ont pour but d’aider les prestataires à maintenir leurs compétences et leurs contacts professionnels, à faire valoir leur détermination à travailler, et éventuellement à retrouver un emploi permanent.
« L’AE offre une sécurité de revenu en cas de perte d’emploi, mais elle a aussi pour rôle de faciliter la transition des chômeurs vers un un nouvel emploi permanent » rappelle Stéphanie Lluis, coautrice de l’étude avec Colin Busby et Brian McCall.
À l’examen d’une série de projets pilotes menés par le gouvernement fédéral de 2005 à 2018 pour évaluer l’effet de changements visant à assouplir les règles de TPP, les auteurs ont noté que les prestataires avaient favorablement réagi à ces modifications. Ils étaient plus susceptibles de travailler à temps partiel ou de travailler pendant de plus longues heures.
À l’heure actuelle, environ la moitié des prestataires tirent profit des règles de TPP. Pour inciter les autres à suivre leur exemple, les auteurs recommandent notamment de les autoriser à tirer des gains hebdomadaires modestes d’un emploi à temps partiel sans que leurs prestations d’AE soient réduites. Ils proposent aussi de prévoir des règles de TPP plus généreuses en période de ralentissement économique, lorsque les emplois à temps partiel sont souvent plus nombreux et accessibles.
Ils exhortent enfin nos décideurs à rassembler les données nécessaires pour brosser un tableau complet des prestataires qui profitent des règles de TPP et de leur situation d’emploi. Des études menées dans d’autres pays qui appliquent des règles semblables, mais qui recueillent des données plus complètes, ont démontré que le prestataires qui y ont recours ont souvent de meilleures chances de se trouver emploi permanent.
« Les règles de Travail pendant une période de prestations aident les nouveaux chômeurs à garder contact avec le marché du travail et peuvent servir de tremplin vers un emploi permanent, concluent les auteurs, mais elles doivent être améliorées pour aider davantage de Canadiens à retrouver du travail. »
On peut télécharger l’étude Transitioning Back to Work: How to Improve the EI Working-While-on-Claim Provisions, de Colin Busby, Stéphanie Lluis et Brian McCall, sur le site de l’IRPP (irpp.org/fr).
L’Institut de recherche en politiques publiques est un organisme canadien indépendant, bilingue et sans but lucratif basé à Montréal. Pour être informé de ses activités, veuillez vous abonner à sa liste d’envoi.
Renseignements : Cléa Desjardins — Tél. : 514 245-2139 — email@example.com