Skip to content

Women in Recessions: What Makes COVID-19 Different?

LMI Insight Report no. 39

March 2021

Home > All Publications > LMI Insight Report no. 39,…

Table of Contents

Key Findings

  • At the height of the pandemic, employment in Canada fell between February and April 2020 by 3 million (-16%), with unprecedented impacts on employment among women.
    • Employment among women fell by 1.52 million in these two months compared to less than 250,000 jobs lost in the 3 previous recessions combined.
    • Women accounted for 51% of the initial job losses compared to 17% in previous recessions.
    • More than 70% of the job losses among women during this period were in lower-earning occupations.
  • Paid hours among women also took a hit during the height of the pandemic as school closures placed pressures on caring for and educating children, especially younger ones:
    • Among households with at least one child under the age of 7, the average paid hours for women fell by 26% (from an average of 27 hours per week in February to 20 hours per week in April). This compares to a much smaller decline among men with (16% or from 38 to 32 hours).
    • During this two-month period the number of women taking a full week off work for personal or family-related reasons increased by 24% to April 2020, versus an increase of 3% among men.
  • The employment recovery has been unstable and particularly weak among women:
    • As of January, employment among women remains 485,000 (–5.3%) below pre-pandemic levels. For men, employment is down 374,000 (–3.7%).
    • Women currently account for 56% of the employment deficit, 8 percentage points higher than their share of pre-pandemic (48%).
    • More than 80% of the employment shortfall among women is in the Accommodation and food services (–227,000 or –35%) and Retail trade (–168,000 or –14%) sectors.
    • Most of this shortfall is concentrated in low-earning occupations where women’s employment is 557,700 (–14%) below its pre-crisis level (employment among women in high-earning jobs is above the pre-pandemic level). Low-income earning men’s employment is down by 339,250 (–12%).


The COVID-19 global pandemic triggered the greatest collapse in economic activity in our time, including the unprecedented loss of three million Canadian jobs (–16%) between February and April 2020. In response to the virus, governments across Canada mandated various lockdowns, causing business closures and reducing operating capacities. Further, the fear of contagion and risk of contracting COVID-19 also contributed to significantly reduced demand for many services requiring human interaction. The public health nature of the pandemic has led to unique distributional impacts across sectors for various classes of workers. Most notably, compared to previous downturns, women have borne a larger share of the labour market and social impacts and they continue to confront several challenges. By January 2021, Canadian employment has not yet fully recovered, remaining 860,000 jobs (–4.4%) below pre-pandemic levels.

During the initial impact of COVID-19, women accounted for 51% of the three million jobs lost in March and April 2020, slightly more than their pre-pandemic share of employment (48%). Since then, a rapid but incomplete employment recovery has progressed more slowly for women than men. As a result, women now account for 56% of the net employment loss between February 2020 and January 2021. This growing divergence between women and men is partly explained by the fact that initial job losses were spread widely (though not evenly) across sectors, whereas the recovery has been far slower in customer-facing service sectors (e.g., accommodation, restaurants, retail sales, etc.) in which women are overrepresented.

The distribution across sectors, however, does not entirely explain the impact on women’s employment. Across the economy, lower paying occupations have also been hit hard in terms of job losses; women make up 58% of such occupations. As of January 2021, employment for low-income earning women remains 13.8% (–557,000 jobs) below their pre-pandemic level compared to 11.5% (–339,250 jobs) for men. Further, the additional family burdens caused by the closure of daycares and schools have fallen largely onto mothers, especially those with young children.

One year into the pandemic, the labour market impacts of COVID-19 are continuing to evolve. This LMI Insight Report offers an analysis of one of the many unique dimensions of the pandemic: the unprecedented impact on women’s employment. The report provides an overview of the pandemic with a comparison to past recessions, a discussion of the sector-specific employment impacts (with a focus on women’s employment), as well as an analysis of employment loss by income group and the impact on working mothers and fathers.

Gender Impacts of Recent Recessions

COVID-19 is fundamentally different than past recessions

The economic impact of COVID-19 differs in several fundamental ways from previous recessions. First, the pandemic is driven by public health needs rather than economic fluctuations. Mandatory lockdowns ordered by governments to contain the spread of the deadly new virus across Canada closed businesses or reduced their operating capacity. The public’s fear of contagion also significantly contributed to the drop in demand for services requiring face-to-face interactions, as can be seen in jurisdictions that did not impose strict lockdowns (e.g., Sweden). Second, lockdowns were not limited to businesses but also shuttered daycares and schools, placing additional demands on parents and guardians. Even as schools responded to the pandemic by switching to virtual learning, children at home continued to require parents’ time and attention. Third, the scale of the labour market impact has no precedent. Through March and April 2020, 3 million jobs (16%) were lost — more than in the past three recessions combined and in only a fraction of the time.

The unique features of the COVID-19 pandemic led to several important distributional impacts on workers. In fact, for the first time, women account for most of the job losses. The lockdowns have had an outsized impact on service-sector jobs, particularly lower paying jobs requiring face-to-face interactions with the public in which women are overrepresented. Further, as discussed below, the burden of having children at home fell primarily on the shoulders of mothers. Whereas women sought to replace lost household income by entering the labour market during past recessions, in this case women often worked to replace the childcare and educational services normally provided by schools and daycares.

The magnitude of the labour market impact and some of its unique patterns can be seen in Figure 1, which shows that employment losses in March and April 2020 erased 15 years of gains. Overall women’s employment in Canada grew by nearly 1.47 million from 7.6 million in January 2006 to 9.1 million by February 2020, an increase of 19%. Men’s employment, overall, grew by almost the same amount, from 8.6 million in 2006 to 10.0 million in February 2020 (17%). In the subsequent two months — March and April 2020 — these 15 years of employment growth were erased as employment fell by 1.52 million (to 7.5 million) among women and by 1.46 million (to 8.6 million) among men. In other words, despite representing 48% of the work force, women bore 51% of employment losses in March and April 2020.

As government restrictions were lifted, the summer of 2020 saw a rapid but incomplete recovery in employment. By the end of 2020, initial job growth slowed and began to reverse as a second wave of COVID-19 infections spread and lockdowns were reimposed. Data as of January 2021 indicates that employment for women remains 5.3% below its pre-pandemic peak, down nearly half a million (–484,500 jobs) from February 2020 compared to 3.6% (–374,000 jobs) for men. The growing divergence in the recovery between women and men reflects the fact that the second wave lockdowns were more isolated to customer-facing service sectors (e.g., accommodation, restaurants, retail sales, etc.) in which women are overrepresented. While these sectors were hard hit in the spring of 2020, early lockdowns were also more likely to effect construction and manufacturing.

LMIC calculations,
Statistics Canada, Labour Force Survey, Table 14-100-287.
Seasonally Adjusted. Grey bars indicate recessions 

Men accounted for more than 80 per cent of job losses in earlier recessions

The 15-year employment trend shown in Figure 1 also includes the 2008–2009 Global Financial Crisis. Although the magnitudes are smaller, the job loss is evidently much greater for men than for women. Indeed, in that recession, men accounted for over 82% of all jobs lost. This outsized impact on men’s employment, as shown in Table 1, has been the norm in past recessions. Although every recession is different, men’s job losses reflect, in part, their overrepresentation in goods-producing sectors (e.g., manufacturing, construction, oil and gas) which — the COVID-19 pandemic notwithstanding — have tended to be subject to larger fluctuations in labour demand than service-producing sectors. As well, the recessions of the 1980s and 1990s occurred in a very different social and economic environment in which women represented a much smaller share of the labour force (see Box 1).

In the Global Financial Crisis, men accounted for four out of five jobs lost. From October 2008 to May 2009, employment among men declined by 3.7% (–335,000 jobs) while employment among women fell by 0.9% (–71,200 jobs). In addition, women experienced a much faster recovery, with their employment returning to pre-recession levels by January 2010. Men’s employment, on the other hand, did not recover until 18 months later, in June 2011 (see Table 1).

Similarly, in the early 1990s recession, women made up 44% of employment just before the recession began in April 1990. From this peak to the trough of employment in April 1992, women lost 38,900 jobs (–0.67%), while men lost 371,500 jobs (–5.1%). Going even further back, in the years leading up to the recession that began in June 1981, women made up less than 40% of employment. From June 1981 to October 1982, women lost 123,500 jobs (–2.7%), while men lost 464,900 jobs (–6.9%).

The COVID-19 pandemic is unique both because of the scale of initial job losses and because, for the first time, women have borne most of those losses. During March and April 2020, men lost 1.46 million jobs (–14.6%) while women lost 1.52 million jobs (–16.8%). Representing nearly 48% of employment in 2019, women suffered 51% of the employment loss. Overall, in the past 3 recessions combined, employment among women fell by less than 250,000 compared to 1.5 million during COVID-19.

Box 1. Women’s Increasing Representation in Canadian Labour Markets

Regulatory changes and evolving social norms led to a rapid increase in women’s participation in the labour force (i.e., those employed or unemployment) in the latter third of the 20th century and beyond. The share of core-aged (25 to 54 years old) women in the Canadian labour force rose from just over half (52%) in 1976 to over 80% by 2002. Since then, women’s participation rate has remained stable, hovering around 83% for the past few years. Conversely, the labour force participation rate for core-aged men has slowly declined from 95% in the mid-1970s to 91% in recent years. This means the gap in labour force participation has remained stable for nearly two decades, with men’s rate about 8 percentage points higher than that of women.

Table 1: In the past 3 recessions combined employment among women fell by less than 250,000 compared to 1.5 million during COVID-19


July 1981 to October 1982 May 1990 to April 1992 November 2008 to May 2009 March 2020 to April 2020
Duration 16 months 24 months 7 months 2 months
Employment change Women –123,500






–1.52 million


Men –464,900






–1.46 million


Job loss share Women 20.1% 9.5% 17.6% 51.1%
Employment share before recession Women 39.6% 43.9% 47.1% 47.5%


Source: LMIC calculations; LFS Table 14-100-287. Seasonally adjusted.


The distribution of job losses in recessions is in large part attributable to the variable impact on different sectors and the demographics of people who work in them. Women, who tended to be sheltered from the direct job loss impacts in previous recessions, have not fared well this time. Moreover, as the economic impact of COVID-19 continues, the relative impact on women is worsening as their share of net employment loss has increased from 51% in April 2020 to 56% as of January 2021. As we shall see, the overrepresentation of women in some of the sectors hardest hit by the pandemic is a major cause of disproportionate job losses.

Unique Sector Impact of COVID-19

Employment recovery has favoured male-dominated sectors

The upward trend in employment during the summer of 2020 slowed in the autumn and then began to reverse in December 2020 and January 2021. Recovery has stalled in a handful of sectors requiring face-to-face interactions with the public. As of January 2021, overall employment remains 858,000 jobs (–4.7%) below its February 2020 level — 97% of which is accounted for by five sectors of the economy.1

Table 2 shows the employment change for men and women in each of these sectors between February 2020 and January 2021. In the five sectors particularly hard hit, employment remains at least 10% below its February 2020 level. These sectors are:

Accommodation and food services

Retail trade

Information, culture and recreation

Other services (except public administration

Business, building and other support services

What these five sectors have in common is that they rely primarily on in-person activities to drive their business. While men have borne a great deal of job loss in these sectors, in each case the relative decline in employment has been larger for women. In fact, not only are women overrepresented in these sectors, they have also borne a larger share of the net job losses in each. More than 80 percent of the employment shortfall among women are in the first two of these sectors: Accommodation and food services and Retail trade.

Employment has recovered in several sectors initially hit hard by the pandemic including real estate and rental and leasing, educational services and notably the male-dominated construction and manufacturing sectors. For example, manufacturing employment is almost fully recovered and, as of January 2021, is only 0.3% (–6,000 jobs) below the February 2020 employment level. Men actually gained 9,000 jobs (+0.7%) in this sector, meaning that women again have borne a higher percentage of the job losses.

1 This differs markedly from the spring of 2020 when losses were more widespread, with employment declining in every sector except Finance and Insurance.

Table 2. Net Employment Change for Women and Men, February 2020 to January 2021
Women Men
Total –485,000 (–5.3%) –374,000 (–3.7%)
Accommodation and Food Services –227,000 (–35%) –144,000 (–27%)
Retail Trade –168,000 (–14%) –73,000 (–7%)
Information, culture and recreation –56,000 (–17%) –60,000 (–15%)
Other services (except public administration) –75,000 (–18%) –7,000 (–1.7%)
Business, building and other support services –46,000 (–14%) –28,000 (–7%)
All other sectors, combined +56,000 (+0.9%) –82,000 (–1.1%)

Women overrepresented in hard-hit sectors

Among all sectors, none has been hit as hard as the accommodation and food services sector. Restrictions on in-person dining and drinking have remained in place in one form or another across the country since COVID-19 first arrived in Canada. Even without rules restricting indoor dining, the public’s concern about becoming infected surely would have kept most patrons away. In addition, this sector includes many firms connected to the tourism industry, which relies on both domestic and international clientele that cannot return until borders reopen and people feel comfortable travelling again (see LMI Insight No. 30). Between February and April 2020, nearly half of all employment was wiped out in the food and accommodation sector, a contraction of 583,000 jobs (49%). Women lost 328,000 jobs during this period while men lost 255,000. The means that women accounted for 56% of job losses during the first wave of COVID-19, equal to their pre-pandemic employment share in the sector (see Figure 2).

Employment began to recover rapidly in food and accommodation during the summer of 2020 as COVID-19 cases remained low and people could dine outdoors. By September, employment had risen to 82% of its pre-pandemic level. However, as the weather turned colder and school began, COVID cases climbed rapidly through the autumn and winter. Indoor dining became risky and, in some regions, restrictions and lockdowns were re-introduced. As of January 2021, employment had fallen significantly in this sector, shedding 370,000 jobs, down 31% over February 2020. As with overall employment, the job loss burden in the food and accommodation sector fell increasingly on women, whose employment is down by 227,000 (–35%), representing 61% of the net job losses since February 2020.

Grey bars represent share of women's employment in each sector for 2019 and the red circles represent the share of peak to trough job losses for women (net change February to April 2020). 

The retail trade sector has also been severely impacted by COVID-19 with a 23% decline (–506,000 jobs) in employment between February and April 2020. Here too women’s employment was disproportionately impacted. In a sector where they made up 52% of workers, they suffered 62% of the job losses in the spring of 2020. More precisely, women lost 314,000 jobs (–27%) while men lost 193,000 jobs (–18%) between February and April 2020. The retail trade sector saw a steady recovery since April 2020. By December 2020, employment was down by only 1.4% compared to February 2020. However, new lockdown measures introduced in December 2020 and January 2021 led to the loss of some 210,000 jobs. As of January 2021, employment in the retail sector was 241,000 (–11%) below its pre-pandemic level. Most of these losses are women’s jobs; 168,000 of them (–14%) since February 2020. Women now account for 70% of the employment loss in the retail trade sector.

Other services (except public administration) include activities that require face-to-face interactions such as personal care, barbers, hairstylists, and laundry. Other services also include automotive, machinery repair and maintenance, which includes more firms deemed essential (e.g., car mechanics) that tend to be male-dominated. Between February and April 2020, employment in this sector declined by 22% overall (–179,000 jobs). Employment for women decreased by 26% (–107,000 jobs) and for men 19% (–71,000 jobs), meaning that women accounted for 60% of initial job losses. The distribution of job losses has since shifted significantly toward women. Whereas male employment has nearly fully recovered, down by only 6,700 (–1.8%) as of January 2021, female employment is down by 74,500 (–18%). Put another way, women now account for 92% of the net employment decline in this sector.

The information, culture and recreation sector includes organizations such as museums, publishing, the film and broadcasting industry, telecommunications, performing arts and recreation. Employment in this sector declined by 24% (–177,000 jobs) between February and April 2020. Women account for 90,000 of these lost jobs (–27%) while men account for 87,000, which is smaller in relative terms (–21%) since they accounted for a larger share of the workforce in this sector. As of January 2021, employment is still 16% (–116,000 jobs) below its February 2020 level.

The business, building and other support services sector consists of three sub-sectors (i) Management of companies and enterprises; (ii) Administrative support, and (iii) Waste management and remediation services. During the first wave of COVID-19, employment in this sector decreased by 12% (–86,000 jobs). Women accounted for 75% of this initial employment decline, the highest percentage out of the five hard hit sectors. Specifically, women’s employment fell by 64,000 (–19%) and men’s by 21,000 (–5%) between February and April 2020. As of January 2021, employment in this sector is still down by 46,000 jobs (–14%) among women and 28,000 (–7%) for men. Women now account for 62% of the net employment loss in this sector.

Although the distribution of men and women’s employment across sectors can partly explain why, for the first time, women have experienced a larger loss of employment than men, it does not capture all or even most of the explanation. Indeed, in each of the five hard hit sectors discussed here, women’s share of employment loss is greater than their pre-pandemic employment share. To put it simply, if two out of three workers are women in a hard-hit sector, it would be reasonable to assume that among those who lost their job in that sector, two out of three would be women. Thus, women would be bearing a “proportionate” share of the job losses. In this pandemic, however, women have taken more than their share of job losses.2 For this reason, it is essential to consider other, non-industry related reasons for this differential.

2 The difference between women’s and men’s net employment decline as of January 2021 is 1.7 percentage points (–5.7% versus –4%). Comparing this to the average employment decline differences between women and men within each sector and taking the average (weighted by pre-crisis employment level in the sector) yields a 1.4 point larger decline for women than men within sectors, which means 82% (=1.4/1.7) of the difference is due to within sector differences.

COVID-19 Has Revealed That Women Are Concentrated in Low-Paying Jobs and Reduced Hours Significantly Due to Family Responsibilities

Looking beyond the sectoral lens, our analysis reveals that women’s overrepresentation among low-income earners and their role as mothers has made them particularly susceptible to the economic impacts of COVID-19.

Female low-income earners

The impacts of recessions differ across sectors, but a common feature is that job losses in any industry tend to be concentrated among lower income earners.3 During both the Global Financial Crisis and the COVID-19 pandemic, low earners suffered more than higher earners, suggesting that those less likely to make ends meet are more likely to suffer the negative consequences of economy-wide trends. The pandemic has brought to the fore the fact that women in Canada are overrepresented among low-income occupations, accounting for nearly three out of five (58%) of these workers. The outsized job losses in these occupations have contributed to women’s greater share of overall employment losses in both the initial downturn in the spring of 2020 and going forward.

Indeed, the COVID-19 recession has impacted, and continues to impact, low-income earners drastically. During the initial downturn, employment among those in low-paying occupations fell by 26.1% (–1.8 million jobs) between February and April 2020 — nearly two-thirds of all jobs lost to that point. Further, the recovery for low-income earners has considerably lagged behind other groups. As of January 2021, overall employment remains 859,000 below its February 2020 level. However, considering that high earners — both men and women — have surpassed their February levels, low-earners account for almost the entire shortfall in jobs.4 In fact, as of January 2021, employment among low-earners remains 12.8% (–896,250 jobs) below its February 2020 level, with employment for low-earning women remaining 13.8% (–557,000 jobs) below their pre-pandemic level versus 11.5% (–339,250 jobs) for men (Figure 3). Earlier in the pandemic, low-earning women also faced a higher rate of job loss than men. During March and April, employment of low-earning women declined by 27.6% (–1.1 million jobs) versus 23.9% (–704,250 jobs) among men.

3 In defining workers by earnings categories, we follow the analysis laid out in an earlier LMIC analyses. High- and low-income earners are then defined by those working in the top 25% and bottom 25% of occupations, and middle-income earners are those working the middle 50% of occupations.

4 Employment among high-income earners fell by only 3.2% between February and April 2020. As of January 2021, they were 2.4% (+100,250 jobs) above their pre-crisis peak.

LMIC calculations; Statistics Canada, Labour Force Survey, RTRA. Not seasonally adjusted.

Paid hours fell steeply among women with young children

The COVID-19 pandemic led to large-scale closures of daycare centres and schools, meaning that children had to stay at home, be cared for and (if possible) educated. These closures have had significant effects on the ability of parents to balance work and family life. Already, prior to the onset of the pandemic, the average paid hours among women with children was substantially lower than that of men (Figure 4). With the increased responsibility of having children at home, the average hours worked among both women and men who maintained employment fell dramatically in the first few months of the pandemic. The fall in hours was particularly acute among women with young children. Among households with at least one child under the age of 7, the average paid hours for women fell by 26% (from an average of 27 hours per week in February to 20 hours per week in April). This compares to a much smaller decline among men (15.8%). By January 2021, the average hours worked has largely recovered across all groups of parents and non-parents alike.

Family care obligations during the initial wave of COVID-19 was an important driver behind women’s decline in average hours worked. Indeed, most of the hours reduced by parents are attributed to people working zero hours but who remained employed (and thus counted toward total and average hours worked). The closure of schools forced many parents to scramble to find alternative childcare arrangements, which often meant a parent taking time off work until more sustainable solutions could be found. The data suggest this burden fell primarily on women as the number of women reporting full-week absences from work for personal and family-related reasons increased by 24% between February and April 2020; for men, the increase was 3%. As schools reopened and the economy started to recover, work absences for personal and family reasons began to recede. By the autumn average weekly hours worked had largely returned to normal levels. The available data, however, do not provide insight into how ongoing family and childcare duties are being balanced by parents.


The COVID-19 pandemic has been a unique and extremely difficult time for all Canadians. For those who were — or still are — out of work, 2020 was especially challenging. The majority of those who lost their jobs have been women, which differs markedly from past recessions where men suffered the largest share of job losses. Women’s employment, as of January 2021, remains 5.3% lower than their pre-pandemic employment level, compared to 3.7% among men.

Women are overrepresented in hard-hit sectors that rely on face-to-face clientele, which accounts for part of this disproportionate impact. However, in many instances women’s employment losses are greater than their employment share. Other employment vulnerabilities were also exposed by the pandemic. Women represent 58% of employment in low-income earning occupations and have suffered a larger loss of employment than their male counterparts. Further, the impact on working mothers has been difficult even for those who retained their jobs. Their average hours worked fell precipitously, particularly for mothers with young children, during the initial phase of the pandemic in the spring of 2020.

The severe labour market impacts that we are still experiencing today are likely to continue until the pandemic is resolved. As we consider how best to ensure an inclusive and sustainable recovery, our policy and program efforts must consider those most impacted as well as the underlying vulnerabilities that have been exposed by the pandemic.


This LMI Insight Report was prepared by Zoe Rosenbaum, Liz Betsis and Behnoush Amery of LMIC. We would like to thank Tammy Schirle (Wilfried Laurier University), Sareena Hopkins (Canadian Career Development Foundation), Julie L’Allier and Marc Delisle (Employment and Social Development Canada), Henry Siu (Vancouver School of Economics), Emna Braham (Institut du Québec), Parisa Mahboubi (C.D. Howe Institute), Clemente Pignatti (International Labour Organization) and Richard Horne for their feedback and constructive comments. For more information about this report please contact Behnoush Amery, Senior Economist, at, or Tony Bonen, Director of Research, Data and Analytics, at

Contact Us

350 Sparks Street
Suite 604
Ottawa, Ontario K1R 0A4

Please enter your name.
Please enter a message.
Please check the captcha to verify you are not a robot.
Scroll To Top