The gender furlough gap: Why did women stop working at higher rates than men?

Profile picture of Bozena Wielgoszewska

Bozena Wielgoszewska, Research Fellow at the Social Research Institute, University College London and former UK Data Service Data Impact Fellow, discusses the gender furlough gap.

The COVID-19 pandemic affected women’s employment rates more adversely than men’s. Why?

In March 2021, a year into the pandemic, it was already widely documented that women were furloughed at higher rates than men, leading to the pandemic being labelled as a ‘shesession’. However, the mechanisms behind this phenomenon were less clear.

In our research, we used data collected at that time from the surveys conducted among participants of the longstanding British Cohort Studies. Our study shows raw gender furlough gaps of three percentage points. We also show that women were around five percentage points less likely than men to remain in active paid work, and four percentage points less likely to remain in the same job. The differences in furlough rates are particularly striking, especially for women who lived with their partner and children.

Occupational segregation

One potential reason for this is that women were ‘in the wrong place at the wrong time’, working in jobs that were more affected by social distancing measures. Women are generally more concentrated in jobs requiring face-to-face interaction (such as travel and services), but they are also more likely to work in jobs considered critical at the time (such as health, social care and education). Our research shows that, once job characteristics (namely occupation, being a key worker and working part-time) are taken into consideration, the initial gaps are reduced, but do not disappear. Even in critical occupations, women were more likely to be furloughed.

Efficient household allocation

Decisions taken within the family about how to trade off domestic and paid work were another potential reason we investigated. According to the efficient household allocation theory, women’s paid work is typically given lower priority, as they tend to have lower earnings than their male partners. It would therefore minimise the potential loss of household income if it were women who took a step back from employment. If this were true, we would expect the remaining gaps in employment rates to be reduced once accounting for the partner’s job characteristics, but we do not find evidence for this.

Childcare responsibilities

Childcare responsibilities, which typically fall on women, could have been another potential reason. Previous research shows that gender pay gaps open gradually following the birth of the first child and continue to widen for many years after that point. This is because such employment gaps impact on how much employment experience women accumulate and because women tend to switch to part-time work, which is often less well paid per hour.

While school closures during the pandemic added to the domestic responsibilities, we hypothesised that childcare responsibilities could be to blame. However, accounting for the number and age of children within the household, on top of the job characteristics, does not close the remaining gaps, so this also does not appear to be a key factor.

Social norms

Social norms are another important factor, although these are particularly hard to measure and therefore to test empirically. The residual gap that remains after controlling for job characteristics of both partners, as well as the number of children and age of youngest child, may be an indication that social norms are at play. Both men and women face social pressures related to looking after children on a ‘breadwinner status’, which may have been translated into higher furlough rates for women.


Alternatively, the remaining gaps may reflect preferences. Women may have chosen to be furloughed because, facing the choice between paid work or the conditions offered under the furlough scheme (not working while still receiving 80 percent of their pay), they may have simply preferred the latter. This notion is particularly difficult to test empirically. Even if we had data on expressed preferences in our survey, these may still reflect internalised social norms, concerns about countering social expectations or the reputational consequences of deviating from prescribed behaviours.


An alternative explanation for the residual gap is employer discrimination, based on inferred statistical characteristics of mothers, stereotypes, or simply favouritism of men. Although discrimination based on gender is illegal in the UK, the pandemic presented an unprecedented setting, and few established procedures were in place. This novel context may have reinforced existing prejudices and fixed ideas about traditional gender roles.


The differences in employment and furlough rates are concerning as they may jeopardise women’s positions in the labour market, in a similar way to the loss of work experience following childbirth, but on a larger scale. This could potentially lead to the reinforcement of gender inequalities or even reversal of the progress towards gender equality.

Other radical changes to the labour market currently taking place, such as the increase in remote working, have the possibility of both narrowing and exacerbating gender inequality. Our study sheds some light on the reasons behind these inequalities, but more research needs to be done to fully understand and prevent such inequalities from being further entrenched.

This blog post was originally published on Transforming Society and you can read the research that is based on here.

The project is funded by the ESRC grant number ES/S012583/1.

About the author

Bozena is a former UK Data Service Data Impact Fellow and a Research Fellow at the Social Research Institute, University College London, where she works across two centres: the Centre for Longitudinal Studies (CLS) and Quantitative Social Science (QSS). Within the CLS, she leads two research themes: ‘Inequality, Poverty, and Social Mobility’ and ‘Labour Market and Skills.’ Bozena has previously contributed to projects such as the COVID-19 Longitudinal Health and Wellbeing National Core Study, where she was involved in generating new, data-driven insights into the Covid-19 furlough scheme. She also played a key role in the harmonisation of income and earnings data across cohort studies and was involved in a project examining the language used in essays written by participants of the National Child Development Study at age 11, shedding light on their future social mobility. In her role at the QSS, Bozena works on a project investigating the gender wage gap over the life course and across cohorts.

Leave a Reply

Your email address will not be published. Required fields are marked *