Stress and depression during the pandemic: Why gender inequalities, labour market conditions, and a reliable transfer system matter

Author: Dirk Foremny (University of Barcelona, Barcelona Institute of Economics - IEB, CesIfo), based on research with Pilar Sorribas-Navarro (University of Barcelona, IEB) & Judit Vall Castelló (University of Barcelona, IEB, CRES-UPF)

1. Introduction

Throughout this year, we have all had to learn that our individual behaviour, such as compliance with social distancing norms, can have enormous consequences on society, ultimately leading to tragic situations in terms of health and lives lost. At the onset of the pandemic, this externality made coordinated public action, such as business and school closures, curfew policies, and lockdowns, unavoidable.

Most of these polices were imposed uniformly on the entire population, but the impact of these measures on people varied between different population groups – such as the young and the old, the poor and the wealthy, but also between men and women.

First, the bright side for females. When it comes to compliance, striking differences have been documented between men and women, with women being more likely to comply with social distancing norms and other measures (Galasso et. al., 2020). It has also been argued that countries which were led by females had on average a more advantageous trajectory durring the pandemic (Garikipati and Kambhampati, 2020).

The data show that women’s mental health has been disproportionally affected by the pandemic

Women, unfortunately, do not get credit for their responsible behaviour – rather, the opposite is true. Women have been disproportionally affected by the pandemic, and data suggests that females bear a larger burden compared to males. At least in the short term, gender inequalities might increase as social distancing measures have a greater impact on sectors with high female employment rates (Alon et. al., 2020). For Spain, recent data confirm this impact on female-dominated sectors, as unemployment increased by 2.84 points for women between the end of 2019 and the third trimester of 2020, but only 2.07 points for men.

How do women deal with this situation? In a recent study (Foremny, Sorribas-Navarro and Vall Castelló, 2020), we document how different groups of society cope with the circumstances caused by the pandemic. The study first documents a substantial change in self-assessed evaluations of mental health, such as sleeping disorders and depression. We find that differences between gender, among other dimensions, are of major importance. Women are, on average, more likely to be negatively affected compared to men.

In our study, we also examine whether these differences are reflected in policy preferences. Survey participants were asked to evaluate public spending and specific public health interventions to understand whether more severely affected groups favoured different reform policies with regards to public spending.

2. The mental health gender gap in Spain

Surveys in several countries confirm that women experience greater difficulties in coping with the pandemic compared to men. However, it has been shown that even under normal circumstances women report worse mental health outcomes relative to men. The important question is if this gap has widened during the pandemic – a feature not included in many studies as reference data prior to the outbreak of Covid-19 is missing.

If some difference between men and women already existed before the pandemic (see Seedat et. al, 2009, who confirms this for many countries based on pre-pandemic data), data collected during the pandemic might overstate the gender differences if not compared to a valid pre-pandemic benchmark. Among others, Etheridge and Spantig (2020) advance this by using the UK Household Longitudinal Survey, which allows to compare data from earlier years to information collected in April 2020. The study documents an increasing gender gap in mental health conditions in the UK which is clearly pandemic-driven.

For the purpose of our study, we collected data on 2nd and 3rd of April in Spain. This was the peak of the first wave, with a death toll of 849 and 900 per day, and a country under full lockdown. We collected a second dataset between 20th and 23rd July, at the beginning of what was called the “new normal”, with daily deaths below 5 per day. In order to be able to benchmark our results to the pre-pandemic situation, we included questions which were asked in the last National Health Survey (Encuesta Nacional de Salud) carried out in 2017. We repeated questions related to mental health in our interviews and combined those with additional questions about policy preferences.

Figure 1. In the last weeks, have you felt unhappy or depressed?

Results show a severe deterioration of mental health. Figure 1 shows the responses to the question asking if interviewees had felt unhappy or depressed during the last few weeks. We documented substantially fewer people who stated that they had not felt unhappy or depressed – or, at least, no more than usual. While 85% of women and 91% of men reported this positive outcome in 2017, these figures declined to 44% amongst women and 60% amongst men in April, 2020.

On the other hand, women reported feeling much more unhappy or depressed during these difficult times, with substantial doubts about the future evolution of the pandemic and further economic and social consequences of the lockdown. One out of every ten men reported to be much more worried than usual, though this figure doubled for women: One in five women was severely unhappy or even showed signs of depression. These numbers improved slightly in July, when daily infection counts became low and citizens were allowed to move about freely. However, the persistency of the mental health impact was remarkable, as 10% of women and 6% of men continued to feel much more depressed or unhappy than usual, a large number relative to 1.6% of men and 2% of women in 2017.

In April 2020, one in five women was severely unhappy or even showed signs of depression

For a more precise measure of the pandemic's impact on women’s wellbeing, we estimated the deterioration for women relative to men during the crisis (16 percentage points in April and 11 in July) and put this value in relation to the pre-pandemic gap, when women were 6 points less likely to report that they did not feel unhappy or depressed. The increase in the mental health gender gap which can be attributed to the pandemic is estimated as high as 10 percentage points in April. The gap decreased in July but remained significant with a 5.6-percentage-point difference between genders.

Other mental health measures revealed similar results. A significantly higher proportion of people reported sleeping disorders, being constantly tense, and unable to overcome their difficulties. Again, women reported worse figures in all questions compared to men and substantially larger differences during the pandemic than in normal times.

Where does the mental health gap between men and women stem from? Two components must be distinguished here. The gender gap can arise because men and women react differently to similar circumstances. For example, it might be more difficult for one of the two genders to cope with becoming unemployed, and this in turn has a different impact among men and women. This is not necessarily bad or unfair, as men and women might just deal differently with certain situations.

Another explanation for this gap is that women during this pandemic are more likely to be affected by circumstances, such as becoming unemployed, which cause depression and unhappiness. Women and men may face situations in a similar way, but women might be more likely to be affected by bad circumstances. This can become more problematic as the gap in this case depends on unequal circumstances and inequalities between the two groups.

Figure 2. Gender gap: difference female relative to male

Probability of answering "Not at all" or "No more than usual" to the question: In the last weeks, have you felt unhappy or depressed?

Figure 2 disentangles the effects between these two factors. The left part of the chart (in blue) shows that approximately half of the gap cannot be explained by differences in observable characteristics or circumstances between men and women. Observable variables capturing important circumstances do not show that men and women react in a significantly different manner to these circumstances. This is likely explained by a worsening of pre-existing gaps in more stressful times. This increase is a concern in itself but is not directly related to underlying inequalities between the two groups [1].

More worrisome is the part explained by inequalities in observable characteristics. Approximately 53% of the gender gap in April (50% in July) can be explained by differences due to labour market conditions, income levels, and non-paid work as measured by the number of children in the household. Out of these variables, the lion’s share is driven by occupational status, which includes labour market conditions such as being unemployed or furloughed, the type of contract if the person is working, and other occupational classifications.

Approximately 53% of the gender gap in April (50% in July) can be explained by differences due to labour market conditions, income levels, and non-paid work

The biggest share of this labour market effect is driven by unemployment. The unequal increase in unemployment amongst women in 2020 explains a large part of the effect, and this might be reinforced by the higher exposure of women to temporary or part-time contracts. Our data show, for example, that it is almost 10 percentage points more likely for women to have a temporary contract than it is for men. If these circumstances cause stress during a severe crisis, women are more affected than men – not because they cope differently with such a situation, but simply by the fact that more women are in this situation.

Income inequalities explain another 12-17% of the difference. Our data show that full-time working women earn on average 15% less than men, which is in line with data from Eurostat which estimates the gender wage gap as 14% in Spain.

Our data show that full-time working women earn on average 15% less than men

The presence of children has a small impact in our analysis, but this is driven by data limitations. The number of children is similar among male and female respondents, and we do not have more detailed information about who assumes responsibility for the additional time spent on childcare and home schooling. Farre et. al. (2020) collected data on this question and found that men had increased their participation in housework and childcare slightly, while women assumed most of the burden. This effect, which we cannot capture in our data, is likely to explain the gender gap even further due to underlying inequalities.

The level of education turns out to not be very important in explaining the gap between men and women, as both genders show similar education levels in our data. Looking at Eurostat data from 2019, the share of women under the age of 64 with tertiary education was 38%, compared to 32.3% amongst men.

3. Gender differences in policy solutions

Our study analyses how individuals value health care spending compared to other public expenditures. Given that our data shows that men and women have been impacted differentially by the current crisis, we next turn to analyse whether they propose different policy solutions to combat the crisis.

Figure 3. Gender differences in policy solutions

Figure 3 shows the percentage of total budget proposed by men and women for each spending category. The dark bars are the results from our survey in April and the light bars show the figures obtained in July. Not surprisingly during a pandemic, we document that people would allocate large parts of the budget to the health sector. In July the average share was approximately 23%, an increase relative to 21% in April. It is interesting to benchmark this number against the proportion actually spent on public health which was 14.4% after combining all levels of government spending in Spain in 2018.

Women would spend 3 more euros (out of 100 euros in public funds) on public health than men

Our study reveals a substantial change relative to previous years: In 2018, slightly less than 30% believed that health spending was the most important government expenditure – a figure which almost doubled in our 2020 data. This documents a remarkable increase in people’s desired level of spending over current figures. This desired increase is also greater amongst women than men. Women would spend 3 more euros (out of 100 euros in public funds) on public health than men. This difference is statistically significant and decreased only slightly in July, when the desired spending share amongst men moved somewhat closer to that of females.

4. Policy responses

To mitigate the immediate impact of labour market conditions on mental health, appropriate and coordinated short-term responses are needed. Many countries – including Spain – have implemented furloughing schemes to soften the immediate consequences of the crisis. This, together with the intention of keeping schools and daycare centres open if the pandemic situation allows, might reduce some of the negative impact on women’s mental health.

However, our data show that the effect of unemployment in worsening mental health is not only driven by those who have lost their jobs during the pandemic, but also by women who were already unemployed when the crisis began. While furloughs provide a solution for those at the risk of becoming unemployed, those searching for jobs and receiving unemployment benefits face an additional problem. Not only does the job search become substantially more difficult; unemployment benefits also decrease over time and eligibility depends on the time workers have paid previously into the system.

To mitigate the immediate impact of labour market conditions on mental health, appropriate and coordinated short-term responses are needed

The longer the crisis lasts, the more people are exposed to these reductions and eventually the loss of this subsidy. Much of the mental health effect might be due to uncertainty about the future development of the pandemic and the resulting economic crisis. Earlier this year, the Spanish Government approved a universal basic income, which essentially is a means-tested cash transfer. This is one step in the right direction, but administrative hurdles and bureaucracy-related delays in paying families in need have left many without receiving the subsidy to date. It is important for existing subsidies to be paid in a timely fashion to avoid situations of stress and anxiety, which might increase significantly due to economic uncertainty.

Our results also suggest that citizens demand a higher share of public spending on health. While this seems an obvious result during a severe health crisis, other structural problems have also been discovered during the pandemic. Our study documents that citizens were also willing to dedicate private money to specific health care measures. We randomly asked people how much they would be willing to develop a vaccine, new medications, and improve intensive care units. Figure 4 shows that people who were asked about providing funds for ICUs proposed 58 euros on average, significantly more compared to new medications (29 euros) and developing a vaccine (35 euros) per month. These results imply that tax increases to improve the health care system and revert spending cuts from previous years would have some acceptance amongst the population.

Figure 4. Willingness to pay for health care improvements

The pandemic's psychological impact might be very long lasting. As a more specific policy, the dramatic mental health impacts show that specific programmes for the hardest hit groups are probably needed to relieve mental health pressures. Our data suggest that the increase in depression and other dimensions of distress is relatively persistent. Besides the tragic impact on physical health, additional funds should be provided to ensure proper help for those looking for support. Counselling must be available and easily accessible for the most vulnerable ones.

The dramatic mental health impacts show that specific programmes for the hardest hit groups are probably needed to relieve mental health pressures

Things will eventually return to normal, and we need to correct the inequalities which triggered the effect of this crisis sooner rather than later. Once the urgent problems are remedied and the economy recovers, we need to remember the underlying systematic problems in the labour market, as well as the tax and transfer system, which might be exacerbating mental health problems amongst the Spanish population. A broad, well-defined, and reliable system of subsidies for those not eligible to receive unemployment benefits needs to be established. At the same time, the use of temporary contracts needs to be urgently regulated.

Improvements for the most vulnerable will have costs. The tax system will be challenged after the crisis, as additional revenues will be needed to cover the ensuing costs and the tax base of several taxes might shrink. An honest debate is also needed about the degree of redistribution between high-income earners and more vulnerable groups.

A comparison with other countries suggests that there seems to be room for improvement in Spain. The tax take, defined as the share of collected taxes out of GDP, is still low. Spain collects only 35% of GDP in taxes, which, excluding Ireland, is the lowest amongst all western European countries. Germany collects 41% (which is the mean value in the EU 27) and France more than 47%. The debate about implementing a more equitable – but efficient – tax and transfer system seems unavoidable once the urgent problems are solved.

Notes

[1] This part also includes unobserved inequality, which would be explained by variables not included in our model. For this reason, this is likely an upper-bound estimate, and the part due to inequalities might be even greater.

References

• Alon, T.; Doepke, M.; Olmstead-Rumsey, J.; Tertilt, M. (2020): This time it’s different: The role of women’s employment in a pandemic recession. CEPR Discussion Paper n. 15149.
• Etheridge, B.; Spantig, L. (2020): The gender gap in mental well-being during the Covid-19 outbreak: Evidence from the UK. ISER Working Paper Series.
• Farre, L.; Fawaz, Y.; Gonzalez. L.; Graves, J. (2020): How the Covid-19 lockdown affected gender inequality in paid and unpaid work in Spain. IZA Discussion Paper n. 13434.
• Foremny, D.; Sorribas-Navarro, P.; Vall Castelló, J. (2020): Living at the peak: Health and public finance during the Covid-19 pandemic. SSRN Electronic Journal.
• Galasso, V.; Pons, v.; Profeta, P.; Becher, M., Brouard, S.; Foucault, M. (2020): Gender differences in COVID-19 related attitudes and behaviour: Evidence from a panel survey in eight OECD countries. Proceedings of the National Academy of Sciences of the United States of America, 117(44): 27285-27291.
• Garikipati, S.; Kambhampati, S. U. (2020): Leading the fight against the pandemic: Does gender ‘really’ matter?, SSRN Electronic Journal.
• Seedat S.; Scott, K. M.; Angermeyer, M.C.; Berglund, P.; Bromet, E. J.; Brugha, T. S.; Demyttenaere, K.; Girolamo, G. de; Haro, J. M.; Jin, R.; Karam, E. G.; Kovess-Masfety, V.; Levinson, D.; Medina Mora, M. E.; Ono, Y.; Ormel, J.; Pennell, B. E.; Posada-Villa, J.; Sampson, N. A.; Williams, D.; Kessler, R. C. (2009): Cross-national associations between gender and mental disorders in the World Health Organization World Mental Health Surveys. Archives of General Psychiatry, 66(7): 785-795.