The Weakest Link: Diagnosing the Deterioration of the Youth Labour Market in Canada

Executive Summary

• Canada’s youth labour market has deteriorated markedly since early 2023, with the weakness concentrated among teenagers. By mid-2026, the unemployment rate for 15–19-year-olds exceeded 20%. However, even this startling statistic understates the scale and pervasiveness of the labour market challenges evident across a wide range of indicators.

• In addition to increased incidence, unemployment duration has lengthened. Unfortunately, for youth experiencing longer-term unemployment, spells can now last almost one year, more than twice the length in 2023.

• We construct a decomposition exercise to better understand the proximate drivers of the recent downturn. It reveals that the recent increase in unemployment was driven almost entirely (more than 89%) by weak hiring activity, rather than by increased job separations. In other words, we find that Canada’s youth are suffering from a “last-in” problem for new hires, not a “first-out” problem for layoffs, which have remained at relatively low levels.

• Indeed, we estimate that youth job finding rates are hovering near generational lows. With jobs being hard to find and discouraged workers exiting the labour force, the teenage employment and participation rates have fallen to their lowest levels in the past five decades, excluding the pandemic.

• Another concerning feature of the weak labour market is reduced labour mobility, as employed people are less likely to quit their jobs. This longer-term trend is worrying because job transitions help workers develop human capital, find better job matches, and raise real wages. In fact, we find muted wage dynamics, suggesting that labour market adjustments have occurred primarily through reduced employment opportunities rather than broader wage reductions.

• Our regression analysis confirms that youth unemployment is much more sensitive to the business cycle than core-age unemployment. As a result, weak macroeconomic conditions have been an important driver of increased unemployment. At the same time, we demonstrate that the business cycle slowdown alone cannot explain the labour market deterioration for teenagers since mid-2024, suggesting additional factors are at play.

• Factors we highlight on the labour demand side are weak hiring appetite from businesses as well as compositional changes. We first show that young workers are concentrated in entry-level jobs that require limited education and experience (in customer service occupations, such as cashiers and shelf stockers, and in industries such as retail trade and accommodation and food services). We then show that job vacancies for these positions have recently fallen more sharply than vacancies elsewhere in the economy. We attribute part of this softening labour demand to a positive short-term development: namely, the stronger recent productivity growth in several youth-employing industries. This has allowed businesses to raise output without requiring as much labour input. At the same time, we show that firms in retail trade and accommodation and food services have recently shifted labour away from teenagers to older, more highly educated workers

• The supply side provides yet another piece of the puzzle, related to the rapid increase in Canada’s non-permanent resident population following the pandemic. Policies facilitating this increase initially sought to address labour shortages, and the resulting spike in labour supply likely exacerbated challenges for Canadian youth by intensifying competition for jobs where youth employment is concentrated.

• International comparisons show that Canada’s recent spike in youth unemployment has been unusually severe relative to major advanced-economy benchmarks.

• Our research highlights a concern that the current anaemic youth labour market conditions could impose broader future economic and social costs. A large body of convincing research shows that difficulty establishing a foothold in the labour market early in a career can persistently reduce future earnings and is associated with other negative outcomes later in life, including those related to marriage, fertility, and physical and mental health.

• The immediate challenge for policymakers and businesses, then, is to take steps to improve youth employment opportunities, including by strengthening the transition from school to work, so young Canadians can gain the early employment experience needed to build strong careers and fulfill their potential.

Introduction

Canada’s youth labour market has deteriorated sharply since early 2023, with the weakness especially pronounced among teenagers. By mid-2026, the unemployment rate for 15–19-year-olds exceeded 20%. Yet even this striking statistic understates the breadth of the deterioration. Teenage employment and labour force participation rates have fallen to levels not seen outside the pandemic, job-finding rates are near two-decade lows, unemployment spells have lengthened, and employed young people have become less likely to leave their jobs for other opportunities.

These developments matter for reasons that extend well beyond the current business cycle. Early employment helps young people acquire workplace skills, build professional networks, learn about suitable occupations, and move to better job matches. A large empirical literature finds that entering the labour market during periods of weak demand can have persistent negative effects on earnings and employment. Prolonged difficulty establishing a foothold in the labour market can therefore leave lasting “scars” on young workers and impose broader economic and social costs.

Recent evidence also suggests a concerning deterioration in the well-being of young Canadians. According to the World Happiness Report (Helliwell et al., 2026), Canada ranked 71st out of 147 countries for happiness among people under age 25 in 2023–2025, compared with 24th for the rest of the population. Canada also ranked 133rd out of 136 countries for the change in happiness among those under 25 over the past two decades. While these outcomes cannot be attributed to labour market conditions alone, they reinforce the significant challenges facing young Canadians.

This report documents the recent deterioration in Canada’s youth labour market and investigates its possible drivers. The slowing macroeconomy is clearly part of the explanation: our regression analysis confirms that youth unemployment is substantially more sensitive to the business cycle than unemployment among older workers. But the magnitude and persistence of the deterioration — particularly among teenagers since mid-2024 — suggest that cyclical weakness alone cannot explain what has happened.

We examine a broad range of indicators to document the depth and nature of the recent deterioration in Canada’s youth labour market, placing current conditions in up to 50 years of historical context, depending on the indicator. Our work focuses on youth but distinguishes between teenagers aged 15 to 19 and young adults aged 20 to 24 because we find that their recent experiences have differed noticeably.

Figure 1.1 provides a useful summary of several labour market indicators, comparing the latest readings with their historical norms across age groups.¹ The figure shows that conditions are exceptionally weak across most indicators for 15–19-year-olds, while outcomes for prime-age workers are much closer to historical norms. Unemployment duration is particularly elevated, while quit rates are unusually low across all age groups. Low job finding rates, alongside low layoff rates, characterize the labour market as currently “low-hire, low-fire”.

A key contribution of our report is to distinguish between two very different ways unemployment can rise: employed workers can lose their jobs more frequently, or unemployed workers can have greater difficulty finding new jobs. Our decomposition shows that the recent rise in youth unemployment has been driven overwhelmingly by weaker job finding rather than increased job separations. The current episode is therefore much more a “last-in” problem — young people struggling to find new jobs — than a “first-out” problem of widespread youth layoffs.

To understand the sharp decline in youth hiring we examine changes in labour demand, job vacancy patterns, employment composition and industry productivity. We find that labour demand has fallen particularly sharply in entry-level occupations and in industries that traditionally employ large numbers of youth, including retail trade and accommodation and food services. Employment within these industries has also shifted toward older and more highly educated workers. Stronger productivity growth in some youth-intensive industries may be allowing firms to increase output with less employment growth. On the labour-supply side, Canada’s exceptionally rapid increase in non-permanent residents following the pandemic likely increased competition for many of the same jobs sought by young Canadians. Finally, we place Canada’s recent experience in an international context and show that the deterioration in youth unemployment has been unusually severe relative to other advanced economies.

The goal of the report is not to identify a single cause of Canada’s weak youth labour market. Instead, it is to establish the key stylized facts, quantify the proximate drivers of rising unemployment, and assess the cyclical and structural forces that may be limiting employment opportunities for young Canadians. These distinctions matter for policy. If the problem were primarily one of job layoffs, the appropriate response would differ from one in which businesses are retaining existing employees but hiring fewer new workers. The evidence presented here strongly points to the latter challenge and highlights the importance of improving pathways from school to work and ensuring that young Canadians can obtain the early employment experience needed to build productive careers.

The remainder of the report is organized as follows. Section 2 reviews the literature on Canadian youth unemployment. Section 3 documents the recent deterioration in youth labour market outcomes. Section 4 examines several potential explanations. Section 5 concludes.

Literature Review

The Effects of Youth Unemployment

The literature review consists of two sections: the first focuses on the effects of youth unemployment, while the second summarizes attempts to determine the causes of youth unemployment in Canada.

The negative effects of youth unemployment are wide-ranging. First and foremost, a large body of evidence finds that entering the labour market during an economic downturn can have negative long-term consequences on future labour market outcomes — a phenomenon known as “labour market scarring”. For example, using data on Canadian male university graduates, Oreopoulos et al. (2012) find that entering the labour market during a typical recession was associated with a 5% loss in total earnings over 10 years.² Unfortunately, labour market scarring is not restricted to recessions. In a longitudinal study using Canadian data from 1985-1999, Krahn and Chow (2016) find that cumulative unemployment between age 18 and 32 has a negative effect on total income and employment quality at age 32.

In addition to causing labour market scarring, poor labour market conditions for young workers can affect broader life outcomes. Schwandt and von Wachter (2020) find that entering the U.S. labour market during a downturn is associated with fewer marriages, fewer children, and higher rates of disability and mortality, with effects lasting until age 50. Likewise, Currie and Schwandt (2014) find that a 1 percentage point increase in the unemployment rate experienced by U.S.-born women aged 20 to 24 reduces fertility by 14.2 conceptions per 1,000 women when followed to age 40. Furthermore, using a Swedish sample, Strandh et al. (2014) find evidence that youth unemployment can cause long-term mental health scarring.

Weak youth labour markets can also cause ripple effects that negatively impact the macroeconomy. Pinto and Steinbaum (2022) show that the Great Recession significantly increased student loan indebtedness for young U.S. borrowers for at least a decade.³ Lower incomes and higher debt translate into reduced consumption, which restrains economic growth. Additionally, poor youth labour market outcomes reduce social cohesion, political participation, and trust in institutions among youth (OECD, 2020).

The Causes of Youth Unemployment in Canada

The Canadian and international literature has examined a range of explanations for elevated youth unemployment. Two recurring themes in the earlier Canadian literature are the size of youth cohorts and aggregate labour market conditions.

The influx of young baby boomers into the labour market in the 1960s and 1970s spawned several papers examining the effect of cohort size on youth employment outcomes. Foot and Li (1986) argue that “generational crowding” was the main reason for high youth unemployment in the 1970s and early 1980s. Bloom and Freeman (1986) examine the labour market outcomes of the baby boomer generation across OECD countries, including Canada, and find that cohort size had a negative effect on youth employment and expected earnings. Gunderson et al. (2000) find that if not for declines in cohort size in the 1990s, youth unemployment rates would have been much higher. Korenman and Neumark (2000) use OECD data from 1970-94 to study the effects of cohort crowding and agree that an increase in cohort size is associated with higher youth unemployment. Yet, they argue that youth unemployment responds more to general labour market conditions than to cohort size. O’Higgins (1997) arrives at a similar conclusion in a study of European countries, acknowledging the role of cohort size but emphasizing the role of aggregate demand as the chief determinant of youth unemployment.

Indeed, the consensus in the literature is that youth unemployment is more sensitive to business cycles than adult unemployment. In a 1981 study of the U.S. labour market, Clark and Summers show that youth account for a disproportionate share of cyclical employment variation, partly due to stronger cyclical fluctuations in participation. Choudhry et al. (2012) find that financial crises have a larger effect on youth unemployment than adult unemployment, with adverse effects lasting up to five years. Bell and Blanchflower (2011) corroborate this finding using microdata from the U.S. and UK, citing a shift of age-specific demand for employment in favour of older workers after the onset of the Great Recession — a shift they suspect relates to labour market regulations that make it easier or less costly to fire young workers.

Hiring slowdowns during recessions have been found to disproportionately affect the youth unemployment rate, a phenomenon known as the “last-in” effect (O’Higgins, 1997). Bernard (2013) finds that Canadian youth also experience higher unemployment inflows than adults, consistent with the “first-out” effect, as youth are twice as likely to be laid off as older workers, mainly because they have less seniority.

Taken together, the literature suggests that youth unemployment is shaped by both aggregate labour market conditions and factors specific to young workers, such as high rates of labour-market entry and exit, as well as greater job turnover. It also highlights two distinct channels through which downturns can affect youth: reduced hiring and increased job loss. The analysis that follows examines the deterioration since 2023 and which of these mechanisms best explains it.

Describing the Deteriorating Youth Labour Market in Canada

Unemployment Rate

The Canadian youth labour market — defined in this paper as the 15–24-year-old age group — has suffered a notable deterioration since the start of 2023. This section presents several stylized facts of this deterioration by analyzing time series of key labour market indicators. In each figure, the vertical dotted red line indicates the beginning of 2023, which we use as the starting point of the recent slowdown.

Youth unemployment rates generally follow the overall “prime-aged” 25–54-year-old unemployment rate trends, but exhibit larger cyclical fluctuations. For example, the Great Recession in 2008-09 caused unemployment rates to jump across all age groups, but the youngest workers saw the largest proportional increase. In the period between the Great Recession and the pandemic that began in 2020, unemployment rates for all three age groups slowly trended back towards their pre-recession levels.⁴ When the pandemic arrived in 2020, unemployment rates soared. Youth were again disproportionately affected as key sectors that employ youth — such as retail trade and accommodation and food services — were especially hard-hit. In 2020 Q2, the unemployment rate peaked at 32.7% for teenagers and 26.2% for young adults. As the economy reopened and adjusted, youth unemployment eventually fell back to pre-COVID levels. However, since the start of 2023, the youth unemployment rate has trended upwards, especially for 15–19-year-olds. In 2026 Q2, 15–19-year-old unemployment reached 20.3%, more than double the rate for 20–24-year-olds of 10.1%, which was down from its peak of 11.7% in 2025 Q3.

Although the unemployment rate for the core working-age population has also increased since the start of 2023, it has risen more slowly than youth unemployment rates. Figure 3.1.2 shows the ratios of the teenage and young adult unemployment rates respectively relative to the 25–54-year-old unemployment rate.⁵ In 2024, the teenage ratio reached an all-time high, and it remains near that peak into 2026.

By contrast, the young adult ratio did not increase as much and began trending down in late 2025. Nonetheless, it is still at levels comparable to the late 1970s, when the entry of the baby boomers into the labour market spawned a wave of concern about Canadian youth unemployment.

Participation Rate

The unemployment rate can understate labour market weakness when potential workers respond to poor job prospects by exiting the labour force. Figure 3.2.1 shows that youth participation rates began to fall at the start of 2023. The participation rate for 15–19-year-olds decreased steadily from 52.3% in 2023 Q1 to 45.9% in 2026 Q2, which is the lowest value on record outside the pandemic. At the same time, the participation rate for 20–24-year-olds dropped 2 percentage points from 78.1% in 2023 Q1 to 76.1% in 2024 Q2, before rebounding to 77.6% in 2026 Q2. The 25–54-year-old participation rate remained stable, dropping no more than 0.4 percentage points below its 2023 Q1 value at any point during that period.

Employment Rate

Youth employment rates have evolved similarly to youth participation rates. Since the start of 2023, the teenage employment rate has declined the most. Between 2023 Q1 and 2026 Q2, the 15–19-year-old employment rate fell 8.9 percentage points from 45.5% to 36.6%, which is also an all-time low excluding the pandemic. Meanwhile, the 20–24-year-old employment rate fell from 72.1% in 2023 Q1 to 67.5% in 2025 Q4, before increasing to 69.9% in 2026 Q2. It appears that while the labour market is continuing to deteriorate for teenagers, labour market outcomes for young adults have begun to improve slightly.

Not in Employment, Education or Training (NEET) Rate

Perhaps surprisingly, the decrease in youth employment has coincided with only a modest rise in the share of young people who are not in employment, education, or training (NEET), as illustrated by Figure 3.4.1. A rise in the NEET rate would be concerning because it could signal a rise in social exclusion and friction in the transition from school to work. Youth who experience NEET also suffer in the long run. After five years, they have on average lower earnings and lower odds of attending postsecondary education or being employed than their peers (Frenette and Handler, 2026). In 2025, the share of 15–29-year-olds who were NEET was 12%, up from 11% in 2023. The increase has been driven by 20–29-year-olds, since most 15–19-year-olds are students and student unemployment does not affect the NEET rate. For 20–24-year-olds, the NEET rate was 14% in 2025 compared to 12% in 2023; for 25–29-year-olds, the NEET rate was 16% in 2025 compared to 13% in 2023. Despite these recent increases, NEET rates remain near their historical averages across these age groups.   

Unemployment Duration

Concurrent with the trends above, the average duration of youth unemployment has increased since 2023 to reach record highs (Figure 3.5.1). In 2026 Q2, the average duration of unemployment for 15–19-year-olds was 14.9 weeks. That is 67% higher than the average duration of 8.9 weeks in 2023 Q1 and far larger than any value between 2006 and 2023. Average unemployment duration for young adults reached an all-time peak in Q2 2026 at 18 weeks, 58% higher than the average duration of 11.4 weeks in 2023 Q1. During the same period, the average duration of unemployment for 25–54-year-olds increased by only 33%, from 16.4 weeks to 21.8 weeks. Moreover, while average unemployment duration for youth is higher now than during the pandemic, the average unemployment duration for the core working-age population has not exceeded its pandemic peak and appears to be declining.

Figure 3.5.2 depicts the evolution of the unemployment duration distribution for 15–24-year-olds. The figure shows that long-term unemployed youth are seeing the largest increases in unemployment duration. In 2023 Q1, the 90th percentile of youth unemployment duration was 21.8 weeks; in 2026 Q2, the 90th percentile of youth unemployment duration was 49 weeks, a remarkable increase of 125%. This is troubling because long-term unemployment is associated with skills erosion, mental health issues, lower future earnings, and lower re-employment rates (Cohen et al., 2025; Feng and Gao, 2026; McCormick, 1990).

Quit Rate

Faced with a challenging labour market, employed young workers are increasingly reluctant to leave their jobs. Figure 3.6.1 shows the share of employed youth who voluntarily quit their jobs over time by age group. Since the start of 2023, there has been a clear decrease in the share of employed teenagers and young adults quitting their jobs. Furthermore, the recent decrease in quits merely continues a potentially concerning longer-term trend. The ability to shop around for new and better employment is a key feature of the youth labour market which, when functioning, allows youth to develop varied human capital and determine which jobs suit them best (O’Higgins, 1997). Also, the high turnover of young workers is an essential mechanism for wage growth. In a 1992 study of young men in the U.S., Topel and Ward found that 66% of wage growth occurred in the first 10 working years and that wage gains from changing jobs accounted for at least a third of that wage growth.

Wages

Turning to wages, shown in Figure 3.7.1, in the pre-pandemic period of 2007 Q2 to 2019 Q4, the average annual growth rate of real wages was 2.6% for teenagers, 0.9% for young adults, and 0.8% for the core working-age population. The average for teenagers was aided by the significant increase of Ontario’s hourly minimum wage from $11.60 to $14.00 in 2018. However, even prior to the increase, 15–19-year-olds had higher annual growth rates on average than 25–54-year-olds (1.8% compared to 0.6%), likely due to the smaller base. During the pandemic, wage growth artificially spiked for 20–24-year-olds and 25–54-year-olds because job losses disproportionately affected low-wage workers (Statistics Canada, 2025a). On the other hand, 15–19-year-olds were less affected because most are low-wage workers. Since 2023, real wage growth has slowed for youth while accelerating for the core working-age population. Between 2023 Q1 and 2026 Q2, the average annual growth rate of 25–54-year-old real wages was 1.60%, only 0.04 percentage points below the 15–19-year-old growth rate (1.64%) and surpassing the 20–24-year-old growth rate (0.83%).

International Comparisons

Historically, Canada’s youth unemployment rate has been below the OECD and Euro area averages (Figure 3.8.1). In the aftermath of the 2008 financial crisis, the U.S. youth unemployment rate remained above Canada’s until 2014. During the COVID-19 pandemic, Canada’s youth unemployment rate soared higher than its peers due to Canada’s more stringent health measures, but it returned to its normal ranking shortly after (Layton, 2022). Since the start of 2023, Canada’s youth unemployment rate has increased rapidly, bucking historical trends. In 2026 Q1, Canada’s youth unemployment rate was 4.1 percentage points higher than the U.S. youth unemployment rate, 2.2 percentage points higher than the OECD average, and just 1.4 percentage points below the Euro area average. These international comparisons suggest that rather than a symptom of a larger phenomenon affecting all advanced economies equally, Canada’s rising youth unemployment dynamics stand out.

Taken together, these various indicators point to an unusually weak youth labour market, with the deterioration most pronounced among teenagers. Rising unemployment has occurred alongside falling participation and employment rates, longer unemployment spells and reduced job mobility. At the same time, the muted wage dynamics suggest that the labour market adjustment has occurred primarily through reduced employment opportunities rather than broad-based reductions in real wages. The next section examines why these outcomes have deteriorated so sharply since 2023.

Explaining the Deterioration

Macroeconomic Conditions

Given that literature from Canada and abroad finds that the unemployment rate for youth is more sensitive to the business cycle than for core age workers, the first explanation for the deterioration of the youth labour market is poor macroeconomic performance. To assess this, we regressed the quarterly unemployment rate of multiple age groups on the Bank of Canada’s estimated output gap over the pre-pandemic period of 1981 Q1 to 2019 Q4. Table 4.1.1 reports the results.

Our results corroborate the literature. We find that the 15–19-year-old unemployment rate, the 20–24-year-old unemployment rate, and the collective 15–24-year-old group are all more sensitive to the business cycle than the unemployment rates of older populations.⁶ Specifically, the 15–24-year-old unemployment rate is 70% more sensitive to the output gap than the core working age unemployment rate. These differences in sensitivity between younger and older unemployment rates are statistically significant. Thus, it is highly likely that poor macroeconomic performance has contributed to the rapid rise in youth unemployment in recent years.

However, we find that the business cycle alone cannot totally explain the deterioration for the youngest age group. To see this, Figure 4.1.2 plots out-of-sample prediction errors from the regressions for these age groups in the post-pandemic period. A positive value indicates that the actual unemployment rate was higher than the model projection in a given quarter, while a negative value indicates that the actual unemployment rate was lower than the projection.

It turns out that only the teenage unemployment rate has increased more than expected since 2024 Q2, given its historical relationship to the business cycle. So, while macroeconomic conditions have played a role in the deterioration of the youth labour market, other factors are also contributing.  

Layoffs and Job Finding Rates

One natural potential culprit for increased youth unemployment could be an increase in layoffs of young workers. Literature from both Canada and abroad finds a “first out” phenomenon, by which young workers are the first to be laid off in hard economic times (Bernard, 2013; O’Higgins, 1997). In recent years, there have also been growing concerns that the adoption of artificial intelligence might have replaced workers in lower-skilled roles, who tend to be younger.

However, as Figure 4.2.1 shows, since the end of the pandemic, the share of employed young workers being laid off has been below pre-pandemic levels, and for 20–24-year-olds, the share of employed workers laid off is on a downward trend. While it remains true that larger shares of employed young workers are laid off than their older peers, there has not been a significant increase in recent layoffs that could explain the deterioration of the youth labour market.

Figure 4.2.2 reinforces the fact that youth layoffs have not dramatically increased. It shows the share of unemployed youth who are unemployed because they were laid off within the last year. Between 2006 and 2026, the share of young people who are unemployed as the result of a recent layoff has been trending down, and in the post-COVID period that has mostly continued, aside from a brief increase in 2023 for 20–24-year-olds.

In fact, the deterioration of the youth labour market appears to be the result of increased difficulty in obtaining employment, rather than an increase in layoffs. In other words, Canadian youth may be suffering from the “last in” effect, by which young workers are disproportionately affected by reduced hiring in economic slowdowns (O’Higgins, 1997). To assess this channel, Figure 4.2.3 depicts job finding rates for workers by age group.⁷ The job finding rate is an estimate of the rate at which the average unemployed worker finds employment in a given month (averaged over four quarters to smooth out fluctuations). At the start of 2023, job finding rates for workers of all ages began to decrease. However, the core working-age job finding rate did not descend below the levels of the Great Recession, whereas teenage and young adult job finding rates reached new lows. In 2025 Q2, the 20–24-year-old job finding rate bottomed out at 37%, 11 percentage points worse than its lowest value during the Great Recession. Similarly, in 2024 Q4, the 15–19-year-old job finding rate reached 45%, its lowest value since at least 2006 — and 25 percentage points worse than its lowest value during the Great Recession. While youth job finding rates seem to be recovering in recent quarters, they remain abnormally low, especially for teenagers. This evidence shows that finding a job remains historically difficult for young job seekers.

To be more precise, we quantitatively decompose the recent rise in unemployment rates into changes attributed to hiring and firing margins — i.e., weaker job finding versus increased job separations (Table 4.2.4). For the population aged 15 and over, we estimate that lower job finding accounts for the vast majority (89.3%) of the increased unemployment rate since 2023, with only a modest contribution (10.7%) attributable to job separations. Not surprisingly, job finding weakness plays the biggest role for young people (accounting for all the increased unemployment among the 20-24 age group). For core-aged workers, weak job finding remains the main driver, accounting for nearly three-quarters (72.4%) of the rise in the unemployment rate.

Job Vacancies by Occupation

The increased difficulty youth face in finding jobs would naturally coincide with weak labour demand conditions. Indeed, this section shows not only that overall labour demand has fallen, but that it has fallen fastest in occupations that have historically accounted for the largest shares of youth employment.

Table 4.3.1 lists the three occupations that accounted for the largest shares of 15–19-year-old employment between 2006 Q1 and 2026 Q1, in descending order. Combined, they accounted for 83.7% of 15–19-year-old employment over that period. The occupation that employs by far the largest share of young people is “sales and service occupations, except management”, which accounted for 50.9% of 15–24-year-old employment compared to only 18.7% of core working-age employment. This grouping includes cashiers, shelf stockers, and food counter attendants. In 2019 Q2, these three occupation groups accounted for 54.2% of total job vacancies, but by 2025 Q2 this share had fallen to only 48.3%. The weakening of labour demand for these occupations was effectively a weakening of labour demand for young workers.

Figure 4.3.2 shows the number of vacancies in the three occupations which account for the largest shares of 15–19-year-old employment, indexed to 2016 Q1.⁸ Although total vacancies in 2026 Q1 were 39.6% higher than in 2016 Q1, vacancies in “sales and service occupations, except management” were only 6.5% higher and vacancies in “occupations in art, culture, recreation and sport, except management” were only 11.7% higher. Vacancies in trades, transport, and equipment operators have fared better, but these occupations only accounted for 13.5% of 15–24-year-old employment from 2006 Q1 to 2026 Q1. The most relevant vacancies for young workers, particularly teenagers, are sales and service occupations, and they have grown much slower.

Job Vacancies by Industry

A similar phenomenon has occurred when we shift from considering occupations to industries. Table 4.4.1 reports the three industries that accounted for the largest shares of 15–19-year-old employment from 2006 Q1 to 2026 Q1, in descending order. Almost half (44.2%) of 15–24-year-old employment over that period was in two sectors: retail trade and accommodation and food services. 61.7% of employed 15–19-year-olds and 34.6% of employed 20–24-year-olds worked in those industries. By contrast, only 13.5% of employed 25–54-year-olds worked in those industries over the same period. As a result, the relatively steeper decline in job vacancies in these two industries in recent years has affected youth more than adults, and 15–19-year-olds have been especially impacted. Table 4.4.1 also contains each industry’s share of total vacancies for several quarters. In 2019 Q2, 24.4% of all vacancies in the economy were in retail trade or accommodation and food services. In 2022 Q2, at the peak of the post-COVID expansion, 26.6% of all vacancies in the economy were in those industries, but in 2025 Q2, that share dropped to 22.1%. 

Figure 4.4.2 shows the number of vacancies in the industries which account for the largest shares of 15–19-year-old employment, indexed to January 2016.⁹ While total vacancies in May 2026 were 37.1% higher than in January 2016, vacancies in accommodation and food services were only 11.1% higher, vacancies in retail trade were only 3.0% higher, vacancies in arts, entertainment and recreation were 0.6% lower, and vacancies in information and culture were 24.0% lower.

Part of the reason for the decline in vacancies in youth-heavy industries is the sluggish growth of output in the accommodation and food services industry, which was only 2.7% between 2023 Q1 and 2026 Q1, and slower than the rest of the business sector (Figure 4.4.3). Notably, the accommodation and food services industry, retail trade, and information and culture, have all recently enjoyed faster productivity growth than other parts of the economy. This has meant less growth in hours worked is needed to generate a given amount of output growth.

Improved labour composition may be part of the reason for the strong productivity growth in some of these youth-heavy industries. Figure 4.4.4 shows the stock of retail trade and accommodation and food service employment of three age groups: 15–19-year-olds, 20–24-year-olds with at least a bachelor’s degree, and 25–54-year-olds with at least a bachelor’s degree. Between 2023 Q1 and 2026 Q2, the number of teenagers working in retail trade and accommodation and food services fell 18.2%. Meanwhile, the stock of 20–24-year-olds with bachelor’s degrees working in those industries increased 54.3%, and the stock of 25–54-year-olds with bachelor’s degrees working in those industries increased 21.0%. This is evidence that the composition of labour in these two industries has improved. The average worker in retail trade or accommodation and food services is now more educated and experienced than before, and that has likely helped improve productivity in these industries.¹⁰

At the same time, the improved labour composition in these industries suggests that older, educated workers may be having difficulties finding work that makes full use of their skills. Instead of becoming unemployed, they have moved down the skills ladder to jobs that are usually filled by younger workers with less education and experience. Although this has had the positive effect of boosting productivity, it makes it hard for younger workers to find jobs. Productivity growth has reduced the demand for labour in these industries, and young workers are increasingly competing with older, more experienced workers for the same jobs. As such, the rise in youth unemployment, particularly for 15–19-year-olds, is not an isolated phenomenon. Rather, it is a symptom of a larger deterioration in the Canadian labour market, and teenagers are simply the weakest link. The effects of the deterioration are most obvious in the 15–19-year-old labour market indicators, but these underemployed older workers in accommodation and food services and retail trade are adversely affected too.

Job Vacancies by Educational Requirements

Another way to investigate the role of weaker labour demand in worsening youth unemployment is to examine the educational requirements of job vacancies. Table 4.5.1 shows the distribution of the highest level of educational attainment of each age group. In 2026 Q2, 86.0% of 15–19-year-olds and 36.2% of 20–24-year-olds held at most a high school diploma.

Because young workers have relatively little formal education and work experience, they disproportionately rely on jobs with low formal educational requirements. Figure 4.5.2 shows that these types of vacancies have decreased the most in recent years. Since most youth have few formal qualifications, they become unemployed or drop out of the labour force when they cannot find jobs that match their level of human capital. By contrast, older, more experienced workers can decide to fill vacancies which require less education than they possess, crowding out younger workers. The fact that employment of older, educated workers has increased in retail trade and accommodation and food services suggests that this is occurring. Thus, in addition to facing a decline in relevant vacancies, young workers are increasingly competing with more qualified workers for those vacancies. 

Immigration

In the wake of the pandemic re-opening, Canada significantly boosted its labour supply by adding non-permanent residents to address labour shortages (Figure 4.6.1). Many of these non-permanent residents held work permits, and many study permit holders were also able to work with some restrictions.¹¹ As a result of these changes, non-permanent residents’ share of the overall population rose dramatically from 3.6% in 2021 Q3 to 7.6% in 2024 Q4. This share has since begun to decline as policy changes have been reversed.

Data on the employment activities of non-permanent residents are unfortunately limited. Nonetheless, such a sharp increase in Canada’s labour supply likely impacted the youth labour market for several reasons. First, many non-permanent residents work in industries that account for large shares of youth employment. For example, in 2021, 17% of temporary foreign workers were employed in accommodation and food services, which represents 140,000 people (Statistics Canada, 2024). In 2023 Q2, 14.8% of businesses in accommodation and food services and 6.1% of businesses in retail trade reported hiring temporary foreign workers in the previous 12 months, which ranked them second and fourth respectively out of the 17 industries in the survey (Statistics Canada, 2023).

Second, there is evidence that new immigrants face job and skills mismatches at higher rates due to language barriers or difficulty transferring credentials (Lovei, 2026). Many temporary foreign workers and international students may have ended up working in industries such as retail trade or accommodation and food services due to the difficulty of finding work that matched their skills. Even if the work was a temporary transition to more suitable employment, this would have resulted in more non-permanent residents competing with youth for many of the same jobs. Or, if they were young and unable to find employment, they would have directly contributed to the rise in unemployed youth.

Conclusion

This report highlights the deterioration of the youth labour market since the start of 2023. Youth unemployment rates increased, employment and participation rates declined, and unemployment duration has reached unprecedented levels. The share of youth quitting their jobs has also hit new lows — so low that their behaviour is converging to that of older workers. Job finding rates bottomed out at lower levels than during the Great Recession, and although they have recovered, they remain lower than they were during the pre-pandemic period. Canada’s youth unemployment rate has been increasing faster than the OECD and Euro area averages.   

This report identifies several potential explanatory factors. First, the youth labour market has been adversely affected by weak economic growth. However, the business cycle cannot fully explain the deterioration. A reduction in job vacancies in the industries and occupations which account for the largest shares of youth employment has also increased the difficulty of finding a job. Moreover, young workers are facing more competition from older and more educated workers for these vacancies, which typically require less formal education. Some of this competition has come from temporary foreign workers, but some may also be coming from more educated workers, such as new graduates, who are unable to find work which matches their human capital. In other words, young workers, particularly 15–19-year-olds, seem to be the weakest link in a loose labour market.

More research is needed to better understand what is affecting the links further up the chain. For example, this report could not distinguish temporary foreign workers from new graduates, making it difficult to draw any conclusions about under-employment amongst 20–24-year-olds. Yet, the labour market outcomes of 20–24-year-olds are critically important as part of transitioning to full-time careers (Beard et al., 2018).

Unfortunately, the publicly available labour force data currently does not allow much deeper research. This matters, not only because it hinders our understanding of the current deterioration of the youth labour market, but also because it may hinder our understanding of the effects of artificial intelligence adoption. Early research in the U.S. and elsewhere has focused on the impacts of AI adoption on young workers on the entry margin (Brynjolfsson et al., 2025; Tucker, 2026). With the data currently available, it will be difficult to replicate such studies in Canada. Thankfully the development of improved data is underway at Statistics Canada. The costs of youth unemployment are serious and long-lasting, and the already troubled youth labour market may be in the early stages of undergoing a generational transformation.

Supplementary Material

Appendix A. Data Sources

All tables and figures in this report are constructed using publicly available data from Statistics Canada, the Bank of Canada, and the OECD. Data sources are listed beneath all tables and figures. Most data were taken from published tables; however, several series have been constructed using the publicly available monthly Labour Force Survey (LFS) Public Use Microdata Files (PUMFs) covering January 2006 to June 2026. These series were created by pooling all observations and using sample weights. Where necessary, seasonal adjustment was performed using the X-13 ARIMA SEATS procedure from the R ‘seasonal’ package. Recession shading uses business cycle estimates from Cross and Bergevin (2012). 

For most tables and figures in this report, youth are defined as 15–19-year-olds (also referred to as teenagers), 20–24-year-olds (also referred to as young adults), or 15–24-year-olds. The most common comparison group is 25–54-year-olds (also referred to as the core working-age population). 

Appendix B. Estimating the Job Finding Rate

The job finding rate was estimated using a methodology developed by Shimer (2012) and used on Labour Force Survey microdata by Bartlett and Tapp (2012).

Equation 1 describes the evolution of unemployment over time. Unemployment in the next month $U_{\{t+1\}}$ is defined as the sum of two terms. $(1-F_{\{t\}})U_{\{t\}}$ is the number of unemployed who did not find employment during the month and who remained unemployed. ${U_{\{t\}}}^s$ is the number of short-term unemployed people who entered unemployment during the month. More specifically, short-term unemployment refers to respondents who have been unemployed for 1-4 weeks.

Equation 2:

Equation 1 can be rearranged to obtain Equation 2 and isolate $F_{\{t\}}$, which is the probability of finding a job in a given month. The job finding rate can then be estimated from the probability of finding a job by using Equation 3, which assumes that job seekers find jobs according to a Poisson process.

Equation 3:

$f_{\{t\}}$ is an estimate for the rate at which the average unemployed job seeker finds a job in a given period. This estimate does not take into account movements in and out of the labour force.

Equation 4:

References

Bank of Canada. (2026). Current MPR output gap (%) [Data set]. https://www.bankofcanada.ca/?p=39818

Bartlett, R. and Tapp, S. (2012). An assessment of Canada’s labour market performance. Office of the Parliamentary Budget Officer. https://www.pbo-dpb.ca/en/publications/LIBARC-1213-155–an-assessment-of-canadas-labour-market–evaluation-de-la-situation-du-marche

Beard, R., Kim, M., and Sharpe, A. (2018). The Contribution of Full-time Students to the Canadian Unemployment Rate, 1976-2017. Centre for the Study of Living Standards. https://csls.ca/research/the-contribution-of-full-time-students-to-the-canadian-unemployment-rate-1976-2017/  

Bell, D., and Blanchflower, D. (2011). Young people and the Great Recession. Oxford Review of Economic Policy, 27(2), 241-267. https://www.jstor.org/stable/43744473   

Bernard, A. (2013). Unemployment dynamics among Canada’s youth. Economic Insights, 24. https://publications.gc.ca/site/eng/9.576621/publication.html   

Bloom, D., and Freeman, R. (1986). The “youth problem”: Age or generational crowding? (Working Paper No. 1829). NBER Working Paper Series. https://www.nber.org/system/files/working_papers/w1829/w1829.pdf   

Brynjolfsson, E., Chandar, B., and Chen, R. (2025). Canaries in the coal mine? Six facts about the recent employment effects of artificial intelligence. Stanford Digital Economy Lab. https://digitaleconomy.stanford.edu/app/uploads/2026/08/CanariesAugust2026.pdf

Choudhry, M. B., Marelli, E., and Signorelli, M. (2012). Youth unemployment rate and impact of financial crises. International Journal of Manpower, 33(1), 76-95. https://doi.org/10.1108/01437721211212538   

Clark, K. and Summers, L. (1981). Demographic differences in cyclical employment variation. Journal of Human Resources, 16(1), 61-79. https://www.jstor.org/stable/pdf/145219.pdf   

Cohen, J., Johnston, A. C., and Lindner, A. (2025). Skill depreciation during unemployment: Evidence from panel data. American Economic Journal: Applied Economics, 17(3), 208-235. https://doi.org/10.1257/app.20230195

Cross, P., and Bergevin, P. (2012). Turning points: Business cycles in Canada since 1926. C.D. Howe Institute. https://cdhowe.org/publication/turning-points-business-cycles-canada-1926/

Currie, J., and Schwandt, H. (2014). Short- and long-term effects of unemployment on fertility. Proceedings of the National Academy of Sciences, 111(41), 14734-14739. https://doi.org/10.1073/pnas.1408975111

Feng, H. and Gao, T. (2026). Impact of unemployment on skill depreciation: Evidence from China’s labor market. Acta Psychologica, 263(1). https://doi.org/10.1016/j.actpsy.2025.106165  

Foot, D. K., and Li, J. C. (1986). Youth employment in Canada: A misplaced priority? Canadian Public Policy / Analyse de Politiques, 12(3), 499-506. https://doi.org/10.2307/3550612   

Foot, D. K., and Li, J. C. (1988). Youth unemployment: A reply. Canadian Public Policy / Analyse de Politiques, 14(1), 109-111. https://doi.org/10.2307/3550457   

Frenette, M., and Handler, T. (2026). Experiencing NEETEST (not in employment, education, or training, excluding short transitions): Determinants and long-term educational and labour market implications. Statistics Canada. https://doi.org/10.25318/36280001202600800003-eng

Government of Canada (n.d.). National Occupational Classification: Hierarchy and structure. https://noc.esdc.gc.ca/Structure/Hierarchy

Gunderson, M., Sharpe, A., and Wald, S. (2000). Youth unemployment in Canada, 1976-1998. Canadian Public Policy / Analyse de Politiques, 26, 85-100. https://www.jstor.org/stable/3552504    

Helliwell, J. F., Layard, R., Sachs, J. D., De Neve, J.-E., Aknin, L. B., & Wang, S. (Eds.). (2026). World Happiness Report 2026. Wellbeing Research Centre, University of Oxford. https://www.worldhappiness.report/ed/2026/

Korenman, S., and Neumark, D. (2000). Cohort crowding and youth labor markets (a cross-national analysis). In D. G. Blanchflower and R. B. Freeman (Eds.), Youth employment and joblessness in advanced countries (pp. 57-106). University of Chicago Press. http://www.nber.org/chapters/c6802  

Krahn, H. and Chow, A. (2016). Youth Unemployment and Career Scarring: Social-Psychological Mediating Effects? The Canadian Journal of Sociology / Cahiers canadiens de sociologie, 41(2), 117-138. http://www.jstor.com/stable/canajsocicahican.41.2.117  

Layton, J. (2022). The impact of the COVID-19 pandemic on the proportion of youth not in employment, education, or training: A regional analysis and international perspective. Statistics Canada. https://www150.statcan.gc.ca/n1/pub/81-599-x/81-599-x2022001-eng.htm   

Lovei, M. (2026). Job mismatch among core working age immigrants with postsecondary education. Labour Statistics at a Glance. https://www150.statcan.gc.ca/n1/pub/71-222-x/71-222-x2026002-eng.htm  

McCormick, B. (1990). A Theory of Signalling During Job Search, Employment Efficiency, and “Stigmatised” Jobs. The Review of Economic Studies, 57(2), 299-313. https://doi.org/10.2307/2297383 

OECD (2020). Governance for Youth, Trust and Intergenerational Justice: Fit for All Generations? OECD Public Governance Reviews. https://doi.org/10.1787/c3e5cb8a-en.

OECD (2026). OECD Data Explorer: Monthly unemployment rates [Data set]. https://data-explorer.oecd.org/

O’Higgins, N. (1997). The challenge of youth unemployment. International Social Security Review, 50(4), 63-93. https://doi.org/10.1111%2Fj.1468-246X.1997.tb01084.x

Oreopoulos, P., von Wachter, T., and Heisz, A. (2012). The Short- and Long-Term Career Effects of Graduating in a Recession. American Economic Journal, 4(1), 1-29. http://dx.doi.org/10.1257/app.4.1.1     

Pinto, S., and Steinbaum, M. (2022). The Long-Run Impact of the Great Recession on Student Debt. http://dx.doi.org/10.2139/ssrn.3928927

Sax, C. and Eddelbuettel, D. (2018). Seasonal Adjustment by X-13ARIMA-SEATS in R. Journal of Statistical Software, 87(11), 1–17. https://doi.org/10.18637/jss.v087.i11

Schwandt, H., and von Wachter, T. M. (2020). Socio-Economic Decline and Death: The Life-Cycle Impacts of Recessions for Labor Market Entrants. NBER Working Paper 26638. https://doi.org/10.3386/w26638   

Schwandt, H., and von Wachter, T. M. (2019). Unlucky Cohorts: Estimating the Long-Term Effects of Entering the Labor Market in a Recession in Large Cross-Sectional Data Sets. Journal of Labour Economics, 37(S1), S161-S198. https://doi.org/10.1086/701046

Shimer, R. (2012). Reassessing the ins and outs of unemployment. Review of Economic Dynamics, 15(2), 127-148. https://doi.org/10.1016/j.red.2012.02.001

Statistics Canada. (2026a). Labour Force Survey: Public Use Microdata File [Data set]. https://doi.org/10.25318/71m0001x-eng  

Statistics Canada. (2024, November 27). Research to insights: Temporary foreign workers in Canada. https://www150.statcan.gc.ca/n1/pub/11-631-x/11-631-x2024006-eng.htm

Statistics Canada. (2025a, June 9). Research to insights: Wages in Canada, 1981 to 2024. https://www150.statcan.gc.ca/n1/pub/11-631-x/11-631-x2025003-eng.htm

Statistics Canada. (2023). Table 33-10-0678-01: Business or organization hired workers from another country through the Temporary Foreign Worker Program in the last 12 months, second quarter of 2023 [Data set]. https://doi.org/10.25318/3310067801-eng

Statistics Canada. (2026b). Table 18-10-0004-01: Consumer Price Index, monthly, not seasonally adjusted [Data set]. https://doi.org/10.25318/1810000401-eng

Statistics Canada. (2026c). Table 17-10-0121-01: Estimates of the number of non-permanent residents by type, quarterly [Data set]. https://doi.org/10.25318/1710012101-eng

Statistics Canada. (2026d). Table 36-10-0206-01: Indexes of business sector labour productivity, unit labour cost and related measures, seasonally adjusted [Data set]. https://doi.org/10.25318/3610020601-eng

Statistics Canada. (2026e). Table 36-10-0207-01: Indexes of labour productivity and related measures, by business sector industry, seasonally adjusted [Data set]. https://doi.org/10.25318/3610020701-eng

Statistics Canada. (2026f). Table 14-10-0443-01: Job vacancies, proportion of job vacancies and average offered hourly wage by occupation and selected characteristics, quarterly, unadjusted for seasonality [Data set]. https://doi.org/10.25318/1410044301-eng

Statistics Canada. (2026g). Table 14-10-0372-01: Job vacancies, payroll employees, and job vacancy rate by industry sector, monthly, unadjusted for seasonality [Data set]. https://doi.org/10.25318/1410037201-eng  

Statistics Canada. (2026h). Table 14-10-0287-01: Labour force characteristics, monthly, seasonally adjusted and trend-cycle [Data set]. https://doi.org/10.25318/1410028701-eng

Statistics Canada. (2025b). Table 37-10-0196-01: Percentage of 15-to 29-year-olds in education and not in education by labour force status, highest level of education attained, age group and gender [Data set]. https://doi.org/10.25318/3710019601-eng

Statistics Canada. (2026i). Table 17-10-0009-01: Population estimates, quarterly [Data set]. https://doi.org/10.25318/1710000901-eng  

Strandh, M., Winefield, A., Nilsson, K., and Hammarström, A. (2014). Unemployment and mental health scarring during the life course. European Journal of Public Health, 24(3), 440-445. https://doi.org/10.1093/eurpub/cku005  

Topel, R. H., and Ward, M. P. (1992). Job Mobility and the Careers of Young Men. The Quarterly Journal of Economics, 107(2), 439–479. https://doi.org/10.2307/2118478

Tucker, L. (2026). You’re (not) hired: Artificial intelligence and early career hiring in the quarterly workforce indicators. United States Census Bureau. https://www.census.gov/library/working-papers/2026/adrm/CES-WP-26-27.html