This chapter uses high-frequency data from the International Labour Organization (ILO), the LinkedIn Economic Graph team and Ipsos to offer a timely analysis of the impact of the pandemic on existing gender gaps in economic participation.
The evidence shows that while both men and women were severely affected by the pandemic, women experienced a larger impact through multiple channels. First, as women are frequently employed in sectors directly disrupted by lockdown and social distancing measures, they consequently experienced both higher unemployment rates and a more subdued re-entry into employment. Second, women’s labour force participation dropped further than that of men at the start of the pandemic. Third, women’s re-employment has been slower, with lower hiring rates and delayed hiring into leadership roles. There is also evidence that among those women who have continued to work throughout the pandemic, some have reduced their working hours more than men and some have pulled back from promotions and leadership roles.9
Underpinning these figures is a more severe “double-shift”. The overlap of work responsibilities and care (housework, childcare and eldercare) responsibilities have intensified during the pandemic, especially for households with children. While there is some indication than men have increased their contribution to unpaid work in the household since the outbreak of the pandemic, such responsibilities still disproportionally fall on women. School closures and changes to care routines are among the main factors contributing to widening labour force participation gender gaps, showing that childcare continues to be an important enabler of women’s ability to participate in the labour market on equal terms.
While in some economies labour force participation rates and employment rebounded over the course of 2020, even a temporary exclusion from the labour market may have long-term effects on the economic opportunities for both men and women, leading to a persistent scarring effect on labour markets. Job losses have the potential to not only affect livelihoods during the current economic downturn but also to affect future earnings opportunities. Mid- to long term unemployment can lead to a loss of skills or professional networks.10 Workers who lose their job during a recession can suffer a persistent earnings loss.11 Finally, women’s re-employment after a work gap can also lead to downward mobility in the status of the role performed.12
The circumstances of the COVID-19 pandemic have dramatically altered how men and women live and work as public-health measures have forced limitations to in-person work. As shown in Figure 2.1, since April and May of 2020 when the severity of the pandemic was first revealed, a large share of workplaces globally experienced widescale closures. By June 2020 workplaces in most economies had partially re-opened, but even well into 2021, may have not returned to ‘business as usual’.
Reduced business activity has led to higher unemployment rates and shorter working hours, especially in those sectors more directly affected by social-distancing measures. ILO projections and scenarios mapping the potential impact of the COVID-19 pandemic suggest that employment declined by 114 million jobs relative to 2019, and if potential employment growth in 2020 is taken into account this number could reach 144 million.13
Mass unemployment has been avoided in several economies through far-reaching fiscal support to companies as well as limitations to collective dismissals. However, hours worked (and consequently, incomes) have contracted for millions of employees and self-employed people.14 The ILO estimates that, globally in the second quarter of 2020, COVID-19 caused a loss of 18% of total hours worked compared to the second quarter of 2019, and 7.2% in quarter three. Further, in economies with more flexible labour laws, unemployment spiked in the first months of the pandemic, despite substantial stimulus packages.
Globally, employment reductions have been relatively more pronounced for women than for men. The same ILO estimates project that 5% of all employed women lost their jobs globally versus 3.9% of employed men. Although in absolute terms 64 million women and 80 million men have lost their jobs, the relative impact is higher on working women simply because there are less women in the labour market overall.
Figure 2.2 presents the real-time monthly shifts in unemployment and compares 2020 data for each month to the same month in 2019. The figure shows data for 11 of the 38 economies for which these statistics are available. On average, by September 2020, men’s unemployment had increased by just over 1.4 percentage points and women’s by just over 1.5 compared to the same time in 2019.
In Germany, specifically, women’s unemployment was 1 percentage point higher, and a similar impact on women was seen in Italy and Denmark. In the United States, an economy which has seen a significant rise in unemployment due to the pandemic, employment figures were down by 4.4 percentage points for women and 3.9 percentage points for men. That expanding gender gap has been the highest across all previous downturns in the US economy’s history since 1948.15 Data from the Bureau of Labour Statistics in the United States reveals that, in 2020, white men’s unemployment rate in the United States increased by 3.6 percentage points, but rose 4.0 percentage points for white women, 4.9 for Black or African-American women and 6.2 for Hispanic or Latina women.
Such figures, as well as further evidence,16 show that historically disadvantaged groups were more strongly impacted by COVID-19 and that the gender gap can be best understood through the prism of intersectionality with ethnicity and race.
A second and similar, yet potentially more profound scarring effect on the labour market is reduced labour force participation. Globally, ILO estimates suggest women are more likely to have exited the labour force since the beginning of the pandemic. This outlook varies by economy. Labour force participation statistics show that disadvantage by gender is pronounced in several economies, and less strong in others. As shown in figure 2.3, on average, between April and June of 2020 men and women’s reported labour force participation dropped 2.35 to 2.45 percentage points in the subset of 51 economies compared to the same time in 2019, with a marginally more pronounced decrease for male workers. Further, the statistics show that in economies such as Brazil, Canada, Colombia, Costa Rica, Greece and Iceland, women were more likely to drop out of the labour market, while in Argentina, South Africa, Mexico and the United States men were more likely to drop out of the labour market.
Seasonally adjusted, quarterly figures can provide a preview of how gender gaps in labour force participation might change in future editions of the Global Gender Gap Index. Figure 2.4 plots changes in labour force participation gender parity (women-to -men ratios) between the average of quarter 2 and 3 of 2019 (pre-pandemic) and the average of quarter 2 and 3 of 2020 plus Q3 2020 for all 33 economies covered by this particular data set.
Changes in women-to-men ratios vary significantly by economy—but have worsened in 60% of the 33 economies for which Q3 2020 data is available for this harmonized data set. Gender gaps in labour force participation are expected to widen in the future, even further than reported in this year’s Global Gender Gap Index. As a result, the projected time to close economic gender gaps discussed in chapter 1 of this report should be treated as an underestimate. Column E of Table 2.1 shows projections for possible labour force participation gender gap ratios, assuming that the same rates of change observed using quarterly data will be reflected one to-one into yearly estimates.
Underpinning such labour force dynamics is the uneven effect of COVID-19 on global industries, low-wage jobs and the most vulnerable workers. The pandemic has had a dual impact on industries— by changing consumer demand for particular products and services, or by making continued work challenging under lockdown conditions for industries which cannot move to remote work. For example, Accommodation and Food Services, Agriculture, Retail, Construction, Transportation and Warehousing are sectors that offer fewer opportunities to work from home than other sectors.17 As presented in Figure 2.5, women are frequently employed in sectors that are most affected by the current recession, contributing to the gender asymmetry in unemployment.
Further insights can be gleaned from LinkedIn Economic Graph data, which shows hiring trends by gender across industries.18 The data presented in Figure 2.6 reveals emerging gender-based asymmetries in hiring by industry and seniority. Figure 2.6a shows the year-on-year change in female hires as a proportion of all hires by industry, while Figure 2.6b shows the change in share by seniority level.
In April and May 2020, the share of women being hired into all roles dropped by 2.3 percentage points for entry-level workers, 0.9 percentage points for experienced workers and 0.5 percentage points for those in leadership positions. The most pronounced decrease was in the Non-Profits industry, followed by the Consumer industry cluster, split into Retail, Consumer Goods and Recreation and Travel. On the opposite side of the scale, the Software and IT Services industry saw the smallest decrease— and there has actually been an increase of 1 to 2 percentage points in female hires since June 2020.
A similar but less distinctive pattern can be observed in the Professional Services industry. In Manufacturing and Financial Services, the initial decrease in the hiring of women does not appear to have been counter-balanced by higher rates of hiring women between July and December. Within the Consumer cluster, the rebound of women’s hiring has been weak, with a tentative improvement in the share of women being hired into roles in Retail and Consumer Goods, but also a persistent decrease in the hiring of women into the struggling Retail and Travel industry.
During the period between June and December 2020, although the hiring of women into entry-level and experienced worker positions rebounded by 0.9 to 1 percentage points, the data shows a persistent decline in the share of women being hired into senior management positions. In fact, year-on-year comparisons show a 0.3 percentage point decrease in the share of women entering senior management positions—largely reversing progress made during 2019 and, in some instances, 2018. As presented in Table 2.2, four industries are continuing to make progress towards closing gender gaps in senior management roles by continuing to expand the share of women hired into those roles despite the disruption caused by the pandemic: Software and IT Services, Financial Services, Health and Healthcare, and Manufacturing. Other industries have seen notable declines of women hired into senior management roles; among them, Recreation and Travel, Retail, Education, and Professional Services.
In addition to workplace closures, there has been an unprecedented acceleration in the adoption of remote working and the digitalization of work processes,19 a change that has often demanded rapid, unsupported upskilling. Educational establishments have switched to remote, often digital-first learning methods20 causing additional pressures across households as carers have been forced to embrace new routines to support children’s education.
Figure 2.7 presents the share of economies that have enforced full or partial limitations on schooling due to COVID-19. Many schools closed at the outbreak of the pandemic and, although they re-opened partially in some economies, they have reopened to a lesser extent than workplaces in several geographies.
The impact of COVID-19 measures on working adults has been different for men and women following the pattern of the existing gender divide. Survey data from Ipsos presented in Figure 2.8 shows the effects of lockdown measures on working men and women. In addition to experiences of greater loneliness among all adults, people with children in the household as well as women face challenges in managing work and household responsibilities.
Those with children in the household are more likely to work unconventional hours and more likely to be concerned about their employment prospects. Men and women in such households were on average less likely to take on hybrid working arrangements, and more likely to either primarily work from home or primarily work from an office. In fact, around 1 in 2 carers reported working primarily from the office—50 % of male carers and 53% of female carers. This data contrasts with 42% for men with no children at home and 43% for women with no children at home.
Women with children in the household are more likely to report an increase in stress levels due to changes in work routines, the pressures of looking after family during the pandemic and inadequacies in their home-working environment. Most workers report reduced productivity; however, those figures are higher for women. Among the respondents, 54% of women with children in the household reported reduced productivity, compared to 46% of men with children in the household, 45% of women with no children in the household and 43% of men with no children in the household.
Female workers often perform a proverbial ‘double-shift’: a day shift in formal employment and a night or morning shift on care work in the home. OECD data on the percentage of time spent on care work between 1999 and 2010 for 20 economies illustrates this persistent gender divide. It shows that men spent 5% of their time on childcare when they had one child, while, in comparison, women spent 12% of their time on childcare.
The trend is mirrored among those with more than one child and when data is disaggregated by the age of the youngest child in the household. Data varies by economy, with economies such as Japan showing greater polarity while economies such as Norway showing greater (but not full) parity. Early evidence from the pandemic-scarred labour market shows a predictable effect of childcare establishment closures (schools and nurseries) on all carers, but in particular women. Although fathers have increased their time spent on childcare during the pandemic21 on average women continue to perform a larger double-shift, leading to reduced working hours, a reversal of gain on entry into leadership positions and a larger incidence of labour force drop-out. In the United States, women who look after young children have reduced their work hours four to five times more than men22 and they are in turn more likely to have exited the labour force.23
All three variants indicate that school closures have a negative and significant impact on labour force participation gender parity. While the provision of flexible work and existing investment in childcare infrastructure appear to be positively related to changes in labour force participation, the negative impact of school closures remains negative and strongly significant. These findings suggest that the remote provision of childcare (such as day care centres, family day care and kindergartens) that has in some instances been put in place to replace in-person care is not able to alleviate the time spent by parents, and that women have filled a larger share of that shortfall. Flexible work arrangements are also seen to be insufficient in the absence of adequate childcare. These tentative findings provide insights into not only the pandemic context but also demonstrate the overall importance of care provision for women’s labour force participation.
To see if this trend holds true across economies, the model described below explores the relationship between quarterly changes to gender parity in labour force participation and an index of school closures for the same quarter.24 The model presents three versions of the data, controlling for economy fixed effects and year-fixed effects. The results are in Table 2.3. In the first column, changes in labour force participation are only regressed on school and work closures; column two shows variation when controlling for existing childcare infrastructure;25 and column three shows a variation controlling for both existing childcare infrastructure and flexible work arrangements.
This year, we supplement the efforts of the Global Gender Gap Index to track progress across a large set of economies with timely gender-disaggregated data for a subset. While new indicators from private sector partners have allowed this report to partially track emerging trends, the current data gap should serve as a call to action for building further capacity in future data collection systems. This chapter offers an assessment of gender gaps by industry and economy. Early indicators point to potential lost ground on gains towards economic gender parity. Across economies, pre-existing gender gaps have exacerbated the asymmetric effect of the pandemic, in terms of employment and labour force participation. By industry, we have seen a widening of gender gaps in some of the sectors most heavily impacted by COVID-19, and a more pronounced emerging gender gap demonstrated by a reversal of gender parity in leadership positions.
By disrupting childcare support for families, the pandemic has had a significant impact on the lives of working parents. In the current context, this impact has been most acutely felt by working women with children who commonly continue to take on a larger share of care work in the household. The pandemic has re-emphasized the need for resilient and sustainable childcare systems. Recovery policies will need to reflect these lessons and focus efforts to invest in childcare infrastructure while reforming the care services sector to offer a more varied range of options. As a job-creating sector, investment in the care economy can prove to be a social infrastructure investment with high returns for the economy and society.
Employees have adapted to the additional challenges of working from home, and there is some early evidence to suggest that male and female workers are showing new patterns of behaviour. Additional, tentative evidence reveals those behavioural shifts might also be engendering changes to wages and progression to leadership roles, with women distinctly less likely to seek out a promotion or pay rise than men in the current context.26