Principale composante du filet social des adultes en âge de travailler, le régime d’assurance-emploi (AE) procure un soutien du revenu temporaire à ceux qui perdent leur emploi ou connaissent un changement de vie qui nécessite de s’absenter du travail, par exemple pour prendre soin d’un nouveau-né, d’un enfant adopté ou d’un proche gravement malade.
Mais face à la pandémie de COVID-19, qui a provoqué au Canada le plus grand choc économique des dernières décennies, l’AE n’a pas été en mesure de soutenir une proportion substantielle de travailleurs sans-emploi ni de traiter en temps voulu la hausse des demandes. Bon nombre de ses lacunes étaient toutefois manifestes bien avant le début de la crise sanitaire.
Ottawa a amorcé un examen biennal de l’AE, le premier d’une telle ampleur en plus de 25 ans. Pour contribuer à cet examen, l’Institut de recherche en politiques publiques a réuni en décembre dernier un groupe de travail formé de chercheurs et d’experts, qui ont discuté des meilleures façons de répondre aux priorités du gouvernement et proposé des pistes de réforme. Ce rapport présente plusieurs des orientations proposées lors de ces deux journées de rencontres virtuelles. Les participants ont conclu à l’inefficacité croissante d’un régime trop complexe et miné par d’évidentes faiblesses.
Les nouvelles modalités de travail et les effets économiques tenaces suscités par la pandémie ont créé une énorme pression sur un régime déjà sous tension. Et les lacunes de l’AE risquent de persister au-delà de la crise sanitaire. Car plusieurs facteurs, du vieillissement de la main-d’œuvre à la transition vers une économie à faible émission de carbone, continueront de faire pression sur l’AE. Parallèlement, le vaste bassin des travailleurs autonomes et l’essor des emplois temporaires et contractuels de courte durée pourraient perpétuer l’exclusion de nombreux travailleurs de cet important programme. Sans compter le défi particulier soulevé par le nombre croissant de personnes qui travaillent sur des plateformes numériques et sont classées parmi les travailleurs autonomes.
Les participants ont convenu de l’urgence de réformer et d’actualiser l’AE selon les besoins actuels du marché de l’emploi et l’évolution de la nature du travail. Il faudra donc faire plus qu’en rafistoler quelques éléments.
Plus précisément, ils ont dégagé un consensus sur les points suivants :
Le régime d’AE a été grandement étendu depuis sa création en 1940. Conçu comme un programme d’assurance sociale prioritairement destiné aux chômeurs temporaires, il s’est enrichi avec le temps de nouvelles indemnités (prestations de maternité, parentales, de maladie et pour soignants), d’un soutien du revenu pour les travailleurs saisonniers et de dispositifs de financement de la formation professionnelle. L’admissibilité à ces avantages, tout comme leur générosité, a varié selon la situation économique et financière du pays. Or la situation actuelle exige de moderniser le programme de l’AE pour le simplifier, le rendre plus généreux et mieux adapté aux nouvelles réalités économiques.
When the COVID-19 pandemic broke out in early 2020, causing millions of workers to temporarily lose their jobs, policy-makers could no longer ignore the long-standing flaws in the federal Employment Insurance system. It failed to cover a significant share of the workforce that was laid-off or without work, due to public health measures and restrictions, and it was not able to process the surge in claims in a timely manner. To support these workers, the federal government introduced emergency benefits geared mainly to the self-employed and other workers who didn’t qualify for regular EI benefits. These measures have since expired and have been replaced by new, temporary emergency benefits, and the eligibility requirements have been relaxed temporarily.
In a recent report, the House of Commons Standing Committee on Human Resources, Skills and Social Development concluded that the system does not reflect the realities of today’s labour market, and that it is not equipped to respond to sudden labour market disruptions like the pandemic. It said that reforms are urgently needed to make the system “nimbler and to better meet the needs of workers and employers” (Canada 2021d, 1).
This message was reaffirmed in the December 2021 mandate letter to the Minister of Employment, Workforce Development and Disability Inclusion, Carla Qualtrough. It calls for a plan to modernize EI, to build “a stronger and more inclusive system that covers all workers, including workers in seasonal employment and persons employed by digital platforms, ensuring the system is simpler and more responsive for workers and employers” (Office of the Prime Minister 2021).
As the main pillar of Canada’s social safety net for working-age adults, EI provides income support in the event of a job loss or other major life circumstances that require a temporary absence from work, such as maternity and parental leaves.
As we write this report, the federal government is in the early stages of a two-year review of the EI system (Canada 2021a). The first major review in almost 30 years, it was prompted by ongoing shifts in the labour market and the unprecedented economic context of the pandemic. Indeed, increasingly diverse work arrangements, the lingering economic effects of the pandemic and the rapid aging of the workforce have added enormous pressure on an already strained program. The time has come to modernize employment insurance.
To inform the government’s review, in December 2021 the Institute for Research on Public Policy convened a working group of researchers and experts in a two-day virtual event to discuss ways of addressing the government’s priorities and to propose reform options. The objective was not to achieve consensus among the participants, but rather to promote a free exchange of ideas and insights. This report and its recommendations are based on the policy proposals that were presented at the event. We thank the participants for their contributions and hope this report will inform the current EI review.
In 1940, following in the footsteps of other major industrialized countries, Canada introduced Employment Insurance (then known as unemployment insurance). The new program, coming on the heels of the Great Depression, was meant to stabilize the economy by buttressing household incomes during periods of high unemployment, such as the one the country had recently experienced. Its main objective was to provide temporary income support to workers who had lost their jobs. The plan was based on a social insurance model that pools unemployment risks broadly across the labour force, rather than providing universal social assistance (Canada 1962).
The program evolved quickly in the post-war years (see figure 1). Initially, it was limited to specific groups of workers (approximately 42 percent of the workforce), and it excluded those in seasonal industries such as agriculture, forestry and logging, as well as teachers and those in hospital care. From 1940 to 1955, the proportion of unemployed workers who qualified for benefits gradually grew and the duration of benefits increased. In 1955, new legislation was passed to include seasonal workers, including self-employed fishers, and sick-leave benefits were also introduced. By the end of the 1960s, close to 68 percent of the workforce was covered by the program. Another major reform in 1971 provided nearly universal access to the program and added maternity and sick-leave benefits. Eased eligibility rules meant that seasonal workers were treated in the same way as other workers, and a two-component benefit structure allowed adjustments to reflect both national and regional economic conditions. The earnings replacement rate — the percentage of a claimant’s insurable earnings replaced by the benefit — was increased from 50 to 67 percent for single persons, and it was set at 75 percent for persons with dependants and low-income earners (Yalnizyan 2021; Couturier, Hébert and Tircher 2021; Courchene and Allan 2009).
This expansionary period was followed by 20 years of contraction from 1976 to 1996, during which time EI’s generosity decreased and the eligibility criteria became increasingly restrictive. In 1976, the higher earnings replacement rate for claimants with dependants was eliminated. In the following years, the replacement rate was ratcheted down from 67 percent in 1976, to 60 percent in 1979, to 57 percent in 1993 and finally to 55 percent in 1994. In 1977, the federal government introduced variable eligibility criteria, which meant that the duration of time that one needed to work to qualify for EI and the duration of the benefits depended on the unemployment rate in the region of residence. The lower the regional unemployment rate, the longer the work period required to qualify for shorter benefit duration. In 1993, workers who voluntarily quit their jobs or refused to accept work became ineligible for benefits. For most claimants, the number of weeks worked needed to qualify for benefits increased.
The way EI is financed also saw major changes throughout this period. Although the federal government initially funded the program’s administrative costs and matched 20 percent of annual employer and employee contributions, by 1971, federal contributions were largely limited to covering the exceptionally high costs resulting from rapid increases in the number of beneficiaries during periods of high unemployment. In 1990, the program was made entirely self-financing, with workers and employers the sole contributors.
In 1996, following the last major review of EI, the program’s name was formally changed from “unemployment insurance” to “employment insurance.” At the time, the federal government was facing major pressure to consolidate its finances and reduce spending to wrestle down public debt. With persistently high levels of unemployment, there were concerns among policy-makers that dependency on some social programs, including EI, were contributing to the problem. Rebranding it as employment insurance was intended to reaffirm that its purpose was to encourage a return to work and provide temporary income support in the event of an unexpected job loss. Entry requirements were further tightened and modified, so that eligibility was based on the number of hours worked rather than the number of weeks. This proved advantageous for seasonal workers but penalized part-time workers (Bédard and Fortin 2015). In addition, several disincentive measures aimed at frequent claimants, such as seasonal workers, were introduced.
The cumulative impact of these reforms was raised by some workshop participants. David Gray, professor of economics at the University of Ottawa, expressed it well: “We get layer after layer after layer built into [the program] without any regard to coherence.” As a result of changing program objectives and ongoing structural changes in the labour market, EI coverage has shifted significantly, as have the measures used to determine the extent of coverage. The percentage of unemployed Canadians who received EI benefits fell from about 80 percent on average between 1976 and 1990 to below 50 percent between 2000 and 2010, and to roughly 40 percent between 2011 to 2019 (see figure 2). Most of the drop during the 1990s can be attributed to the changes in program eligibility criteria that restricted access to benefits to those with a valid job separation. However, 40 percent of the decline was due to an increase in the share of unemployed individuals who had not contributed to EI in the previous 12 months, because they were either long-term unemployed or self-employed, had dropped out of the labour force, or had only recently entered the labour force (Canada 1998; Gray and Busby 2016).
Since the 1990s, the way that program administrators measure EI coverage has shifted away from the percentage of unemployed workers who received regular benefits to the percentage of unemployed workers who pay into the program and have a valid job separation. Among the second group of unemployed workers, roughly 80 percent had accumulated enough hours to qualify for regular benefits in the decade before the pandemic (Canada 2021b).
At no point since its inception was EI intended to cover unemployed workers who do not pay premiums into the regime, but this group — which consists of self-employed workers and those without insurable earnings in the previous 12 months — comprised over 35 percent of unemployed workers before the pandemic, a large share of the unemployed population. Hence, labour market developments that give rise to new forms of work arrangements, and the fact that new income support programs had to be created during the pandemic, are prompting a re-evaluation of the program’s objectives.
The numerous reforms implemented over the eight decades since EI was first introduced have brought major changes to the program and serve as a reminder of both the potential and capacity to adapt it to new labour market realities.
We heard from participants that the layers of complexity added to the EI system over the years and glaring gaps in coverage have thwarted its ability to stabilize the economy in times of crisis, which was the principal reason for establishing it in 1940. Miles Corak, economics professor at the City University of New York, noted that Canada has suffered several significant economic shocks in recent decades, including the 2008-09 recession, the 2014 collapse of global commodities prices and the COVID-19 pandemic. “Each time the unemployment insurance program was too little, too late,” with the number of EI claims during these periods falling well short of the total jobs lost, he said. “This is what we have to fix.”
Keith Banting, professor emeritus at the School of Policy Studies at Queen’s University, agreed. “Recent experience has shown that employment insurance as a whole has failed manifestly and failed Canadian workers. Can EI be made crisis-proof?” he asked.
Corak said that, as the program has evolved, it has become increasingly complex, and its ability to deliver benefits in a timely manner and stabilize the economy in severe downturns has been compromised. He raised the notion of introducing an additional benefit component to the program that would take effect in times of recession, be funded through general revenues, ensure easy access and have no waiting period, much like the Canada Emergency Response Benefit in the early days of the pandemic (see box 1).
Overall, participants in IRPP’s working group agreed that changes to the EI program are long overdue. This was brought home by the federal government’s response to the economic fallout from the pandemic over the past two years, and the various contingency measures and programs that were introduced to support unemployed Canadians who didn’t qualify for EI benefits (see box 1).
The shortcomings of EI are likely to persist even after the pandemic has ended, due to changes in the labour market caused by economic, environmental and demographic forces. Worker shortages are likely to remain a challenge, as Canada’s workforce growth is expected to stagnate. In the coming decade, the number of working-age individuals in Canada for every retiree is projected to fall from 3.5 in 2020 to 2.6 in 2030.
At the same time, a significant pool of self-employed workers and the potential for a continuing shift to more short-duration, contract and temporary jobs could mean that a large swath of Canadian workers will continue to be excluded from EI, either because they do not pay premiums or because they lack sufficient hours worked to qualify. The growing number of individuals who work for digital platforms such as UBER and DoorDash and who are classified by these firms as self-employed also poses a challenge, because they do not contribute to EI and don’t have the protections that other workers enjoy (Yssadd and Ferrao 2019; Statistics Canada 2022).
Working group participants also noted that other major trends, including climate change and the transition to a low-carbon economy, as well as the potential for service-sector jobs to be globalized (as manufacturing jobs were decades ago), could make the labour market more volatile, putting additional pressure on the EI program.
A clear message that came through over the two days of discussions is that any new reforms to EI must be designed with an eye to the future of work and the workers who are likely to be affected by these changes, particularly those who rely on precarious work. There was broad consensus that the current system, which is still tailored to the labour market conditions of the 1970s, is no longer adequate, and that further tinkering at the margins will not be sufficient.
Every 25 to 30 years EI needs to be redesigned to better meet the needs of the changing workforce. The system must evolve and adapt. Some members of the working group argued that several of the changes made in the 1990s were understandable, given the economic and fiscal context of the time, but others felt that they went too far and made the system too restrictive for the realities of today’s workforce.
There was broad agreement among participants on the following points:
From the start, EI has provided temporary income support to unemployed workers in the event of job loss. In the 2019-20 fiscal year, the EI program provided $17.8 billion in benefits, of which $11.1 billion went to regular benefits (excluding the amount paid under the Canada Emergency Response Benefit). The remainder went to maternity, sickness, parental, caregiving and other benefits (Canada 2021b; see table 1).
Working group participants expressed broad support for moving away from variable entrance requirements based on regional unemployment rates toward more uniform eligibility criteria. Armine Yalnizyan, Atkinson Fellow on the Future of Workers, argued that variable entrance requirements disadvantage low-paid, part-time and on-demand workers living in larger cities, where unemployment rates are relatively low. These rules also affect workers in particular sectors such as food service, accommodation, hospitality and retail industries (Yalnizyan, Gellatly and Ritchie 2021).
Participants generally agreed that work duration requirements to qualify for benefits should be eased or the minimum-maximum range narrowed, although they differed on what the new level should be. Suggestions ranged from a uniform requirement of 360 hours to the current minimum requirement of 420 hours, which applies only in high-unemployment regions. Pierre Fortin, professor emeritus of economics at Université du Québec à Montréal, recommended returning to an eligibility requirement based on the number of weeks worked, arguing that the hours-based criterion has had a detrimental effect on full-year, part-time workers, particularly women, who account for a greater proportion of part-time employees. He recommended a universal 17-week eligibility requirement, while others called for a lower level.
As an alternative to a universal eligibility requirement, but also moving in the direction of simpler and more uniform criteria, Miles Corak recommended reducing from nine to three the number of bands of unemployment rates used to determine hours-worked requirements in EI regions, as shown in table 2.
Some participants voiced their support for reinstating the eligibility of workers who voluntarily quit their jobs, arguing that this would restore a better balance in employer-employee relations, and facilitate better job matches and longer job durations. It was suggested that, as a disincentive for quitting, these workers could face a longer wait-time penalty than other claimants before receiving benefits, rather than being disqualified outright. The potential downsides of this approach or the mechanisms for establishing a financial penalty were not discussed in detail. Of Canada’s total unemployed population in 2019, 14.2 percent were ineligible for benefits because they had quit their jobs (Canada 2021b).
Several participants expressed the view that revised eligibility criteria should address certain systemic exclusions, such as migrant workers and the self-employed. It was pointed out that under the current program many of those who work for platform-based companies, for instance, might be misclassified as self-employed or independent contractors, rather than as employees, and therefore would not pay EI premiums or qualify for benefits. The issue of employee classification and definition also raises inconsistencies between Canada Revenue Agency rules and varying provincial employment legislation in this regard.
Participants acknowledged that there is no quick fix for the needs of self-employed workers, but they agreed that the time has come to seriously consider how best to address them. In 2019, roughly 2.9 million people, or 15 percent of the labour force, identified as self-employed (Canada 2021a). However, this group is quite diverse and includes workers who are voluntarily self-employed and those who are involuntarily self-employed, contract workers, incorporated business owners and independent professionals, as well as those referred to as “gig workers” (whose needs and interests vary). Although self-employed workers can voluntarily opt in to EI special benefit programs, they currently cannot access regular benefits in case of a layoff because they do not pay premiums.
Jennifer Robson, associate professor of political management at Carleton University, said the evidence demonstrates that a voluntary program for EI regular benefits is doomed to fail, because those at low risk of losing their jobs will opt out. “If you open it up voluntarily, they will not come,” she said. The group also discussed whether potential measures for self-employed workers should be mandatory and made a part of the broader EI system. The idea of creating a separate program was not endorsed, based on the experience of other countries, where such programs failed to meet expectations.
However, a mandatory program may run into resistance from some self-employed workers. According to Kevin Milligan, professor of economics at the University of British Columbia, experience in other countries suggests that mandatory programs requiring self-employed workers to pay actuarially prescribed contribution rates are unpopular. He raised the notion of requiring app-based companies to turn over data to the federal government to determine EI premium levels for employers and employees, noting that some countries, such as Germany, have successfully used this approach to collect a social insurance surcharge.
Overall, participants generally agreed that the government should proceed with caution, and that efforts should initially focus on providing support to those most at risk of job precarity. Laurell Ritchie, a member of the Good Jobs for All Coalition and the interprovincial Employment Insurance Working Group, noted that, to qualify for regular benefits, claimants must have completely ceased working for seven consecutive days, during which time they are not permitted to earn even nominal amounts of employment income. Relaxing this rule “seems like low-hanging fruit,” she said.
We heard from several participants that the current earnings replacement rate of 55 percent is inadequate and does not allow low- to middle-income individuals to make ends meet (for a general description of how EI benefits are calculated, see box 2). They argued that a rate increase should be considered, and their recommendations on this ranged from 65 percent, the median rate currently used by industrialized countries, to 67 percent, the level in place during the 1970s. Some also recommended that the minimum benefit of $300 a week introduced during the pandemic be made permanent to better support low-income workers. Before the pandemic, there was no minimum floor for EI benefits.
Miles Corak noted that the parameters of EI are largely tailored to meet the needs of seasonal workers, especially those in coastal regions of the country, which adds a non-insurance element of income support and redistribution to the program. He recommended that more attention be paid to displaced long-tenured workers who have been in the same job and contributed to EI for many years. These workers often face longer periods of unemployment than other claimants and may need to retrain or reskill, sometimes only to find that their new jobs pay less than their previous jobs. To better support these workers, he proposed that a wage-insurance program to temporarily top up their incomes be introduced.
Some participants recommended eliminating the Family Supplement, a top-up benefit under EI that is available to low-income recipients with children, arguing that it has largely been supplanted by the Canada Child Benefit and is incompatible with the risk-insurance premise of EI.
David Gray, professor of economics at the University of Ottawa, pointed to previous research that showed that EI benefit durations have proven inadequate in times of recession. As a result, ad hoc measures had to be implemented to extend them because policy-makers were concerned that claimants would exhaust their benefits (Gray and Leonard 2020). This underscores the need to improve EI’s economic stabilization capacity.
Armine Yalnizyan, who supported adopting a uniform entrance requirement regardless of EI region, had several suggestions regarding benefit duration. She recommended allowing the duration of benefits to fluctuate with the level of unemployment in an EI region as the program currently does; modifying the grid used to calculate the duration of benefits based on Statistics Canada’s definition of full-time work, namely a 30-hour workweek rather than 35 hours; and increasing the minimum duration from 14 to 18 weeks. She also recommended that the maximum duration of EI benefits be permanently extended to 50 weeks from 45 in all EI regions.
Several participants voiced concern that the current yearly maximum insurable earnings (MIE) of $60,300 is too low and leaves a growing share of middle-income workers without full coverage. Over the past five years, close to 50 percent of EI regular-benefit claimants were at the maximum benefit level (Yalnizyan, Gellatly and Ritchie 2021). A few participants recommended raising the MIE limit by a significant amount; the proposals ranged from $83,500 to $100,000. They argued that this would bring more contributions into the program, put the plan on a better financial footing and better insure those with higher earnings against the impact of a job loss.
Some noted that raising the MIE limit and therefore payroll contributions may be unpopular among highly paid workers in secure jobs, as it would redistribute income from those less likely to make a claim to those more likely to do so. As well, they noted, it could run into political resistance. “On balance, there are likely going to be losers [among those in higher income brackets],” said Rhys Kesselman, professor emeritus at the Simon Fraser University School of Public Policy. His view was that, instead of raising the MIE, federal government contributions should be reinstated and be funded from general revenues and not payroll taxes, which is more appropriate to accomplish redistributive objectives. However, others countered that advances in digital technology, remote work and other labour market trends might make high-skilled workers more vulnerable to job loss than in the past and increase their need for improved employment insurance coverage.
In addition to regular benefits, EI also provides benefits to cover temporary absences from work due to illness, or to care for a newborn, an adopted child or a critically ill family member (see box 3).
Some working group members expressed the view that several of these benefits, referred to as special benefits, fall outside the core insurance function of the program and, as such, should be funded as social programs through general tax revenues. However, there was no consensus on this point, and some participants pointed out that removing the benefits from EI would give rise to jurisdictional disputes. Although under the Constitution employment insurance is a federal responsibility, removing special benefits from EI could result in challenges regarding whether these benefits fall under federal or provincial and territorial jurisdiction.
Previous research published by the IRPP concluded that many of the limitations of EI special benefits are a consequence of tying access to these benefits to an employment insurance system that was never meant to cover all workers (Robson 2017).
Participants generally agreed that special benefits are not broadly accessible to those who need them most, that benefit levels are too low and that the regulations governing them are too complex. Jennifer Robson noted that take-up of special benefits rises with income. For example, more than 80 percent of recent parents who earn more than $60,000 per year qualified for maternity and parental benefits in 2019, compared with 50 percent of those who earned less than $20,000 (see figure 3). Higher income earners are also more likely to receive an employer-paid top-up for maternity and parental benefits. “We have this under-coverage problem in special benefits, and [the benefits] really seem to reward those who actually have more resources,” she said.
According to Robson, there are more than 200 rules that stipulate how various EI regular and special benefits can be combined, most of which apply to combinations of special benefits. This puts some claimants at a disadvantage. For example, regular and special benefits are normally capped at 50 weeks, but eligibility is based on hours worked during the 52-week period prior to a claim. If a woman who has received maternity or paternal benefits is laid off shortly before or after returning to work, she would be ineligible for regular benefits because she would not have worked sufficient hours during the previous 52 weeks.
Robson also argued that accessing some special benefits imposes administrative hurdles on claimants at a time when they are least able to cope with them, as they are dealing with illness or caring for a critically ill family member. She recommended easing some requirements so that claimants can access benefits more promptly and provide required medical notes and other documentation later.
Since 2010, self-employed workers have been able to opt in to special-benefit programs, but they must earn a minimum of $7,555 a year (this was temporarily lowered to $5,289 during the pandemic) and pay premiums for one year before they can make a claim. Take-up for the program is low; as of February 2021, only 25,000 workers (less than 0.1 percent of eligible workers) had opted in to the program (Canada 2021d; Robson 2021). Participants were generally in favour of finding ways of extending coverage for the self-employed.
As an alternative, Miles Corak proposed establishing individual special-benefit accounts that would be funded by individual contributions and topped up from general government revenues. Individuals would be able to draw down on these accounts to cover sickness and caregiving leaves (but not maternity and parental benefits). At the end of one’s working life, the unused portion could be transferred to a Tax-Free Savings Account or a Registered Retirement Savings Plan (Corak 2021).
EI also aims to help unemployed persons maintain their attachment to the labour market, transition to permanent work and access retraining or reskilling programs for new occupations. It does this through the Working-While-on-Claim (WWC) provisions and by funding training programs and employment services. These services are delivered by the federal government and by provinces and territories through Labour Market Development Agreements and the Indigenous Skills and Employment Training program.
WWC provisions allow EI claimants to take part-time or casual work and keep a portion of their EI benefits. Their objectives are to prevent workers who have lost their jobs from being unemployed for prolonged periods, to help them maintain their skills and network of workforce contacts, and to encourage them to demonstrate their commitment to work to prospective employers (Busby, Lluis and McCall 2021). Claimants can keep 50 cents of EI benefits for every dollar they earn, up to 90 percent of their insurable earnings, before EI benefits are deducted.
Stéphanie Lluis, economics professor at the University of Waterloo, told the group that about half of EI claimants use the WWC provisions, and that many of those who do are repeat users and seasonal workers. She presented research recently published by the IRPP that calls for further investigation of the effect of WWC provisions on nonrepeat claimants, and for adjustments to the WWC provisions to encourage more claimants to take up part-time jobs or to work longer hours in those jobs. Her coauthored study also recommended temporarily making WWC provisions more generous during recessions when part-time jobs are more readily available (Busby, Lluis and McCall 2021).
EI also provides funding for employment support measures, such as job search and career counselling services, and for employer-sponsored training programs that allow claimants to receive benefits while they train.
Although these benefits and measures are not included in the federal government’s review of EI, several participants argued that they should be. Garima Talwar Kapoor, director of policy and research at the Maytree Foundation, stressed the need for a review to identify which service-delivery models work best, noting that the provinces and territories use a variety of ways to deliver training and employment services, including devolving responsibility to local and private providers, and that greater monitoring and evaluation are warranted.
Pierre Tircher, a doctoral candidate in industrial relations at Université de Montréal, noted that these types of measures were at one time funded by the federal government from general government revenues, and he argued in favour of returning to that approach. “Insurance [premiums] shouldn’t pay for anything other than social risk,” he said.
When EI was first established, the federal government was responsible for covering all administrative costs, plus an amount equal to 20 percent of employee and employer contributions. This changed in 1971 when the government reduced its funding obligations to cover only benefits paid to self-employed fishers, as well as cost increases associated with benefit extensions that kick in during periods when the national unemployment rate exceeds 4 percent. In 1990, the program became entirely self-funding — employers and employees became the sole contributors.
To this day, EI premiums are set according to the advice of the chief actuary at a level that ensures that total revenues and expenses balance over a seven-year period. EI revenues for the 2020 fiscal year were $22.7 billion and expenses were $23.7 billion, leaving a net deficit of $1 billion (Canada 2021b).
Working group participants acknowledged that many of the changes they proposed would entail higher program costs. Among participants’ suggestions for how these measures could be funded, some supported increasing the MIE to bring more contributions into the program, while others called for restoring federal government contributions. Rhys Kesselman pointed out that, by moving to a self-funding program, the tax burden for funding EI shifted away from the federal government’s general revenue base (consisting of a mix of corporate, personal and goods and service taxes) toward exclusive reliance on payroll taxes, making it more regressive. Although participants made various arguments for reintroducing a federal contribution to EI over the course of the discussions, they did not reach any conclusions. There was some agreement, however, on the need to ensure that the sources of program funding are well-suited to the policy objectives being pursued.
The idea of revisiting the idea of experience-rated EI premiums was also raised — that is, charging higher premium rates to employers or employees in those industries or sectors where EI claims were more frequent. Kesselman suggested the government’s EI review should include careful consideration of experience-rating for employer premiums. He cited his earlier in-depth analysis that challenged most of the arguments against the adoption of experience-rating (Kesselman 1983). However, other participants expressed the view that, given the current labour market context, experience-rating did not warrant being among the top priorities for EI reform.
Under the current rules, any additional program costs due to changes in eligibility, benefit duration and generosity rules must be funded through an increase in premiums. Employees (outside Quebec) currently pay $1.58 per $100 of insurable earnings, and employers pay 1.4 times that amount, or $2.21. According to the EI chief actuary, to cover increased program costs incurred during the pandemic, employee premiums will have to rise to $1.81 per $100 over the coming years, an additional 23 cents on each $100 earned ($2.53 for employers) (Office of the Chief Actuary 2021).
The research team at IRPP reproduced the EI premium rate-setting formula to estimate the potential costs of some of the proposals put forward by the working group and how these would impact the premiums that workers and employers pay.
Using EI coverage survey results reported by Employment and Social Development Canada, we made a few simplifying assumptions regarding the effects of some proposals regarding EI eligibility to determine the effect of changes to entrance requirements on the number of unemployed Canadians who receive EI.  Figure 4 shows how moving to uniform 420- and 360-hours-worked requirements, giving coverage to all self-employed workers and including those who quit their jobs would affect EI coverage.
We were then able to estimate the additional program costs involved by multiplying the number of additional EI beneficiaries per year under the proposed lower entrance requirements by the average weekly benefit. We used these program cost projections to determine the additional revenues needed to balance the EI account over a seven-year period. We then used the EI chief actuary’s estimates of future aggregate employment earnings as the basis for calculating the required change in employee EI premium rates (see table 3).
The results are perhaps most easily conveyed if we examine a hypothetical case. Consider the case of Melissa, who works full-time as a grocery store manager. Her weekly earnings at the store are $1,160 a week (rounded: $60,300 annually). Table 3 shows the effect of six proposals on the EI premiums Melissa would be required to pay under existing EI premium rate-setting rules (employers would pay 1.4 times this amount).
Our results show that easing eligibility rules under scenario 1 (uniform 420 hours worked to qualify), scenario 2 (360 hours worked to qualify) and scenario 3 (allowing job quitters to qualify) would result in higher premiums for Melissa. Under scenario 4, Melissa’s contributions would be higher, but so would her benefits should she be eligiaq ble for them. In contrast, increasing the MIE under scenarios 5 and 6 would reduce her premiums due to the larger contribution pool and have no relevant effect on her eligibility or her benefits because her salary does not exceed the current MIE ($60,300). It is worth noting that those with earnings above the current MIE would see an increase in their yearly EI contributions and the benefit amounts they would receive when eligible.
In scenarios 5 and 6, we estimate the effect of increases to the MIE on premium rates. These scenarios must take into account the following: (1) that raising it would increase average weekly EI benefit amounts, as claimants with earnings above the MIE would qualify for more generous benefits; and (2) that workers with higher earnings are, in general, less likely to make an EI claim than those with lower earnings.
Over the course of the two-day workshop, participants described the current EI system as too complex, inadequate, and suffering from glaring coverage gaps and regional inequities that have made it increasingly ineffective. Overall, they agreed that changes to the system are long overdue. This was brought home by the federal government’s response to the economic fallout from the pandemic over the past two years, as various contingency measures were introduced to accelerate claims and benefit delivery, fill the gaps in support and create new programs to cover unemployed Canadians who didn’t qualify for EI benefits.
An overarching theme of the discussion was that EI’s shortcomings would be exacerbated in coming years, even after the pandemic ends, due to the labour market impacts of technological change, population aging and the transition to a low-carbon economy. At the same time, the presence of a significant pool of self-employed workers and a continuing shift to more short-duration, contract and temporary jobs means that a large segment of the Canadian workforce could still be excluded from basic income and social supports.
Participants agreed that the time had come to re-evaluate EI’s purpose, to revise and update it to meet the needs of today’s labour market and to make it more responsive to economic shocks. They argued that it should be made more accessible, more generous and less complex. A notable concern was that the program has become overly restrictive, leaving key groups of unemployed Canadians without basic social protections. Participants generally agreed on the need to ease eligibility requirements, or at least make them more equitable for workers, regardless of where they live. They discussed the potential and rationale for increasing the earnings replacement rate and the maximum yearly insurable earnings, as well as extending the duration of benefits, although they differed on what the exact parameters should be.
Pierre Fortin, professor emeritus of economics at Université du Québec à Montréal, perhaps summarized it best when he said, “Future EI reforms should aim to strike a better balance between the permissive program that was in place until the early 1990s and the restrictive one we have today.” At the same time, all participants acknowledged that some of the reforms that they proposed would entail higher costs and design challenges and could require changes to the way the program is funded.
Keith Banting is professor emeritus in the department of political studies and the Stauffer Dunning Fellow in the School of Policy Studies at Queen’s University.
Miles Corak is professor of economics at the Graduate Center of the City University of New York. He was previously economist in residence at Employment and Social Development Canada.
Pierre Fortin is professor emeritus of economics at Université du Québec à Montréal, fellow at the Royal Society of Canada and the C.D. Howe Institute, and past president of the Canadian Economics Association.
David Gray is professor of economics at the University of Ottawa. He has worked as a consultant on program evaluation and employment policy for Employment and Social Development Canada.
Garima Talwar Kapoor is the director of policy and research at the Maytree Foundation, where she leads policy research and analysis work on income security, housing, and economic and social rights.
Rhys Kesselman is professor emeritus at the Simon Fraser University School of Public Policy. His research expertise is in taxation and social policy.
Stéphanie Lluis is professor of economics at the University of Waterloo. Her areas of specialization are labour and personnel economics.
Kevin Milligan is economics professor in the Vancouver School of Economics at the University of British Columbia. His research spans the fields of public and labour economics.
Laurell Ritchie is co-chair of the interprovincial Employment Insurance working group and a member of the Good Jobs for All Coalition.
Jennifer Robson is associate professor in the Riddell Graduate Program in Political Management at Carleton University. Previously she has worked in the voluntary sector and in the public service.
Pierre Tircher is a doctoral candidate in industrial relations at Université de Montréal and a researcher at the Institut de recherche et d’information socioéconomiques.
Armine Yalnizyan is the Atkinson Fellow on the Future of Workers. Previously she has served as senior economic policy adviser to the deputy minister of Employment and Social Development Canada.
 The level of benefits is determined by the replacement rate, which is a percentage of insurable earnings, up to a ceiling equivalent to the average annual industrial wage (which was $60,300 in 2022).
 Although currently the EI system is funded by employer and employee premiums, the federal government, on the advice of the Auditor General, includes EI in its financial statements. This affects the budget balance and the debt.
 Individuals without a valid job separation include those who voluntarily quit their jobs, are fired or refuse to accept suitable work.
 Statistics Canada, “Projected population by projection scenario, age and sex, as of July 1 (x1,000)” (scenario M3, medium growth), table 17-1-0057-01.
 In their discussions on the future of work, several participants made the point that, in comparison with the situation in previous economic restructuring periods, which affected less-educated, lower-paid workers, in the current context those with higher education and pay are also more vulnerable.
 Premiums for EI regular benefits are lower for employees in Quebec because the province opted out of the federal maternity/parental special benefits plan and instead charges separate premiums for its provincial maternity/parental insurance plan.
 Of the total number of unemployed individuals in 2019, 38.5 percent did not pay EI premiums in the previous year, 14.2 percent had quit their jobs, 8.3 percent had worked insufficient hours, and 6 percent were self-employed (Canada 2021b). For our estimates, we assumed that self-employed workers and those who quit their jobs (including those who quit to go to school) would have similar hours worked to other claimants, and would therefore have similar coverage rates. We also assumed that, of the 8.3 percent of unemployed workers in 2019 (or roughly 90,500) who did not qualify due to insufficient hours worked, 30 to 50 percent (27,150 to 45,250) would qualify under a uniform 420-hours-worked requirement and 40 to 60 percent (36,200 to 54,300) would qualify under a uniform 360-hours-worked requirement.
 Based roughly on the 2019 earnings distributions in Canada, scenario 5 assumes that contributors with annual earnings between $60,300 and $83,500 would make up 10 to 20 percent of all regular benefits paid under an MIE of $83,500, with average annual earnings of $71,900. For scenario 6, we assumed that contributors with annual earnings between $60,300 and $100,000 would make up 15 to 30 percent of all regular benefits paid under an MIE of $100,000, with average annual earnings of $80,150. To test the accuracy of our calculations, we verified our results against the sensitivity analysis outlined in the 2022 EI Premium Rate Actuarial Report (Office of the Chief Actuary 2021). We also accounted for the projected additional costs from the COVID-19 response.
Bédard, M. and P. Fortin. 2015. Onze propositions pour un meilleur régime d’assurance-emploi. Repères IRPP 7. Montreal: Institute for Research on Public Policy. https://irpp.org/research-studies/onze-propositions-pour-un-meilleur-regime-dassurance-emploi/
Busby, C., S. Lluis and B. 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. https://irpp.org/research-studies/transitioning-back-to-work-how-to-improve-ei-working-while-on-claim-provisions/
Canada. 1962. Report of the Committee of Inquiry into the Unemployment Insurance Act. Ottawa: Queen’s Printer and Controller of Stationery. https://publications.gc.ca/collections/collection_2014/bcp-pco/Z1-1961-1-1-eng.pdf
———. 1998. Human Resources Development Canada. An Analysis of Employment Insurance Benefit Coverage. Applied Research Branch. Strategic Policy. W-98-35E.
———. 2021a. Employment and Social Development Canada. “Reforming Canada’s Employment Insurance Program.” https://www.canada.ca/en/employment-social-development/programs/ei/consultation-better-program-backgrounder.html
———. 2021b. Employment and Social Development Canada. Canada Employment Insurance Commission (CEIC). 2019-2020 Employment Insurance Monitoring and Assessment Report. Ottawa: Canada Employment Insurance Commission. https://www.canada.ca/en/employment-social-development/programs/ei/ei-list/reports/monitoring2020.html
———. 2021c. House of Commons Standing Committee on Human Resources, Skills and Social Development and the Status of Persons with Disabilities. Modernizing The Employment Insurance Program. https://www.ourcommons.ca/Content/Committee/432/HUMA/Reports/RP11439397/humarp06/humarp06-e.pdf
———. 2021d. Department of Finance. “Government Announces Targeted COVID-19 Support Measures to Create Jobs and Growth.” News release, October 21. https://www.canada.ca/en/department-finance/news/2021/10/government-announces-targeted-covid-19-support-measures-to-create-jobs-and-growth.html
———. 2021e. Employment and Social Development Canada. “EI Regular Benefits: Do You Qualify?” https://www.canada.ca/en/services/benefits/ei/ei-regular-benefit/eligibility.html
Corak, M. 2021. “Employment Insurance for the Future of Work, Right Now.” On the website of Miles Corak, Economics for Public Policy. https://milescorak.com/2021/03/09/employment-insurance-for-the-future-of-work-right-now/
Courchene, T. and J. Allan. 2009. “A Short History of EI, and a Look at the Road Ahead.” Policy Options. Montreal: Institute for Research on Public Policy. https://policyoptions.irpp.org/magazines/employment-insurance/a-short-history-of-ei-and-a-look-at-the-road-ahead/
Couturier, E., G. Hébert and P. Tircher. 2021. Proposition de reform de l’assurance-emploi au Canada. Montreal: Institute de recherche et d’informations socioéconomique. https://iris-recherche.qc.ca/wp-content/uploads/2021/10/Assurance-Emploi-V6.pdf
Gray, D. and C. Busby. 2016. Unequal Access: Making Sense of EI Eligibility Rules and How to Improve Them. Commentary 450. Toronto: C.D. Howe Institute. https://www.cdhowe.org/sites/default/files/attachments/research_papers/mixed/Commentary_450.pdf
Gray, D. and P. Leonard. 2020. Mind the Gap: Running Out of Employment Insurance Benefits. IRPP Study 81. Montreal: Institute for Research on Public Policy. https://irpp.org/research-studies/mind-the-gap-running-out-of-employment-insurance-benefits/
Kesselman, J.R. 1983. Financing Canadian Unemployment Insurance. Toronto: Canadian Tax Foundation.
Office of the Chief Actuary. 2021. Employment Insurance Premium Rate. Actuarial Report 2022. Ottawa: Office of the Superintendent of Financial Institutions Canada. https://www.osfi-bsif.gc.ca/Eng/Docs/EI2022.pdf
Office of the Prime Minister. 2021. “Mandate Letter to the Minister of Employment, Workforce Development and Disability Inclusions.” December 16. https://pm.gc.ca/en/mandate-letters/2021/12/16/minister-employment-workforce-development-and-disability-inclusion
Robson, J. 2017. Parental Benefits in Canada: Which Way Forward? IRPP Study 63. Montreal: Institute for Research on Public Policy. https://irpp.org/research-studies/parental-benefits-in-canada-which-way-forward/
———. 2021. “Special benefits in EI.” PowerPoint presentation. Montreal: Institute for Research on Public Policy.
Statistics Canada. 2022. The Changing Nature of Work. Ottawa: Statistics Canada. https://www.statcan.gc.ca/o1/en/plus/249-changing-nature-work
Yalnizyan, A., 2021. “The Future of EI: Benefit Rates and Duration.” PowerPoint presentation. Montreal: Institute for Research on Public Policy.
Yalnizyan, A., M. Gellatly and L. Ritchie. 2021. Reforming EI for the 21st Century. Toronto: Atkinson Foundation. https://atkinsonfoundation.ca/wp-content/uploads/2021/11/Reforming-EI-for-the-21st-Century-Oct2021.pdf
Yssadd, L. and V. Ferrao. 2019. Self-employed Canadians: Who and Why? Labour Statistics at a Glance. Ottawa: Statistics Canada. https://www150.statcan.gc.ca/n1/pub/71-222-x/71-222-x2019002-eng.htm
With the federal government in the early stages of a two-year review of Employment Insurance, the IRPP convened a working group of 12 researchers in December 2021 to discuss ways to modernize EI and expand the range of reform options. The objectives of the working group were to: (1) share ideas on how to address some of the stated priorities of the federal government; (2) suggest complementary or other priorities that the government ought to consider; and (3) propose a wider range of reform options than might be collected in the traditional stakeholder consultation process, with clear attention to practical design and implementation. This report highlights consensus when it emerged, but consensus was not a desired outcome of the group’s meetings.
This report was written by Colin Busby, Ricardo Chejfec and Rosanna Tamburi. The manuscript was copy-edited by Francesca Worrall, proofreading was by Zofia Laubitz, editorial coordination by Francesca Worrall and Étienne Tremblay, production was by Chantal Létourneau and art direction by Anne Tremblay.
See appendix B for the list of Members of the working group.
To cite this document:
Working Group Report. 2022. How to Modernize Employment Insurance: Toward a Simpler, More Generous and Responsive Program. IRPP Report (May). Montreal: Institute for Research on Public Policy.
The authors and the IRPP would like to thank the EI working group participants for their numerous contributions to the development of this report.
If you have questions about our publications, please contact firstname.lastname@example.org.
Montréal – Le régime d’assurance-emploi (AE) doit être revu et actualisé pour en simplifier les règles, en augmenter la générosité et mieux répondre aux besoins actuels du marché du travail, conclut un nouveau rapport de l’Institut de recherche en politiques publiques.
Alors qu’Ottawa a entrepris un examen de deux ans de l’AE — le premier d’une telle ampleur en plus de 25 ans —, ce rapport présente les idées et propositions d’un groupe de 12 spécialistes de la question, qui se sont réunis en décembre dernier pour discuter des moyens de moderniser l’AE.
« Le régime d’AE est trop complexe et ses lacunes en matière de couverture des sans-emploi l’ont rendu de plus en plus inefficace, soutient Colin Busby, directeur de recherche à l’IRPP et coauteur du rapport. Ça fait longtemps qu’il est temps d’y apporter des changements. »
Une situation confirmée par les deux dernières années de pandémie, qui ont obligé Ottawa à adopter différentes mesures d’urgence pour protéger les travailleurs sans-emploi non admissibles aux prestations.
Or ces lacunes de l’AE persisteront vraisemblablement au-delà de la crise sanitaire, selon le rapport. Car plusieurs facteurs pourraient encore déstabiliser le marché du travail et accentuer la pression sur un régime déjà sous tension, des nouvelles modalités de travail au vieillissement de la main-d’œuvre en passant par la transition vers une économie à faible émission de carbone. Parallèlement, le vaste bassin des travailleurs autonomes et l’essor des emplois temporaires et contractuels de courte durée pourraient perpétuer l’exclusion de nombreux travailleurs. Sans compter le défi soulevé par le nombre croissant de personnes travaillant pour des plateformes numériques comme Uber ou DoorDash, que ces entreprises considèrent comme des travailleurs autonomes et qui ne peuvent donc profiter des mêmes protections que d’autres salariés puisqu’ils ne cotisent pas à l’AE.
Les experts du groupe ont notamment convenu de l’importance d’assouplir les règles d’admissibilité pour permettre à un plus grand nombre de travailleurs sans-emploi de toucher des prestations, d’accroître le montant de certaines prestations et d’améliorer la réactivité du régime en cas de choc économique. Leur rapport reconnaît toutefois que certaines propositions de réforme pourraient faire augmenter les coûts, susciter des problèmes de conception et nécessiter de revoir le mode de financement du régime.
« Le groupe de travail avait pour but de proposer au gouvernement un éventail de pistes de réforme », précise Colin Busby. Plusieurs participants ont plaidé pour une révision fondée sur un juste équilibre entre le régime permissif du début des années 1990 et sa version plus restrictive d’aujourd’hui.
On peut télécharger le rapport du groupe de travail sur l’AE, How to Modernize Employment Insurance: Toward a Simpler, More Generous and Responsive Program, sur le site de l’institut (irpp.org/fr).
Renseignements : Cléa Desjardins — Tél. : 514 245-2139 — email@example.com