The notion that politicians’ identity, including their gender, can influence policy outcomes is supported by a growing body of empirical research. This column reports a study exploring the extent to which the share of women members in national governments during the Covid-19 pandemic shaped the policy responses. Higher female representation was associated with a lower likelihood of schools being closed – not because women politicians were less concerned about the spread of the virus but due to their heightened awareness of the costs of school closures, notably for working mothers with young children. The results indicate the potential benefits of more gender-balanced governments in shaping policies that affect the wellbeing, education, and economic prospects of children and families.
Despite some progress over the past decade, women’s representation in political leadership remains limited globally. As of 1 January 2024, only 27 out of 159 countries are led by women – a modest increase from just 18 countries a decade ago (United Nations Women Headquarters Office 2024). Women represent 23% of cabinet members heading ministries worldwide in 2024; in Europe and Northern America, this share is 33%.
On the one hand, these numbers hint at the long road ahead to reach the United Nations’ Sustainable Development Goal of gender equality (SDG 5), which aims to ensure equal opportunities for leadership at all levels of decision-making in political, economic, and public life (United Nations 2015). On the other hand, they underscore the need to understand to what extent the underrepresentation of women politicians matters for policy choices, especially in times of crises.
Gender and policy decisions: Existing evidence
The notion that a politician’s identity, including gender, can influence policy outcomes is supported by a growing body of empirical research. Women politicians may matter for politics due to significant gender differences in social preferences and policy priorities (Hessami and Lopes da Fonseca 2020).
In developing countries, higher female representation increased investments in domains relevant to the needs of women and children, including health and education, enhanced institutional quality by reducing corruption and rent-extraction and lowered domestic violence against women (Chattopadhyay and Duflo 2004, Bhalotra and Clots-Figueras 2014, Brollo and Troiano 2016, Bhalotra et al. 2023, Bochenkova et al. 2023, Baskaran et al. 2024).
For high-income countries, the empirical evidence is less conclusive. While some studies of the US and Spain find no significant impact of women policymakers on public expenditure (Ferreira and Gyourko 2014, Bagues and Campa 2021), research on France and Germany reveals that higher shares of women politicians affect policy choices, prioritising women’s issues, children, childcare and health services (Lippmann 2022, Baskaran and Hessami 2023, Hessami and Baskaran 2020).
The Covid-19 pandemic has reignited academic interest in examining how women’s representation may influence policy responses. Several studies indicate that women political leaders have been more successful than men in combating the pandemic and reducing its health toll on the population as measured by infected cases, deaths and hospitalisations (Garikipati and Kambhampati 2021, Bruce et al. 2022).
First evidence on women in governments and school closures
In a recent paper (Danzer et al. 2024), we contribute new evidence to this debate by focusing not on the gender of a country’s political leader, but on the role of the share of women members in national governments in shaping policy responses during the pandemic. Specifically, we investigate whether a higher share of women in governments affects the likelihood of implementing one of the most restrictive and controversial containment measures, namely, the closure of schools.
As Figure 1 illustrates, there was substantial variation in school closures across countries and over time, with 85% of EU countries fully closing their schools in April 2020, marking the peak of these closures. To date, the effectiveness of school closures as containment measure remains unclear, with some studies suggesting reduced transmission (e.g. Lessler et al. 2021, Vlachos et al. 2021, Amodio et al. 2022), while others finding no significant impact or even increased cases (e.g. Isphording et al. 2021, Diederichs et al. 2022, von Bismarck-Osten et al. 2022).
At the same time, mounting evidence indicates that school closures have disproportionately and adversely affected children, families, and especially mothers, leading to reduced wellbeing, increased childcare responsibilities and diminished labour market participation among mothers (Albanesi and Kim 2021, Croda and Grossbard 2021, Huebener et al. 2021), significant learning losses and developmental setbacks for children (Giuntella et al. 2021, Grewenig et al. 2021), and worsening socio-economic inequalities (Adams-Prassl et al. 2020, 2022, Huebener et al. 2024, Agostinelli et al. 2022, Fuchs-Schündeln 2022).
Figure 1 Evolution of school closure in EU-28 countries from 1 March 2020 to 30 September 2021
Source: Figure 1 from Danzer et al. (2024), based on the OxCGRT database.
In our study, we examine how the proportion of women members in national governments influenced school closures in Europe from March 2020 to September 2021 – a policy disproportionately affecting children and working mothers. Since decisions on school closures and other containment measures were largely made by executive bodies, we focus on government composition.
Our analysis is based on a self-collected weekly panel dataset for 28 European countries, covering the first three Covid-19 waves, combining data from various sources. Our main outcome variable is a binary indicator for national school closures in a given week. It is constructed from harmonised daily data on school closures and other government responses from the Oxford Covid-19 Government Response Tracker (OxCGRT; Hale et al. 2021).
Our database also includes Covid-19 case numbers and information on other containment measures and overall stringency indexes. Our key explanatory variable of interest is the share of women government members in late 2019, sourced from the European Institute for Gender Equality’s Gender Statistics Database. Relying in pre-pandemic shares of women politicians reduces potential issues of potential reverse causality of changing gender compositions of governments during the pandemic.
Our findings reveal a striking negative relationship between the share of women members in national governments and the likelihood of school closures, as indicated in the descriptive scatter plot in Figure 2. Our regression results indicate that a one standard deviation increase in the proportion of women government members is associated with a 5.7 percentage point reduction in the probability of school closures.
To put these numbers in perspective: the average female share in national governments in the sample is about 31% – ranging from 9% to 58%. According to the estimates, a rise in the share of women members by about 13 percentage points is associated with a reduction in the likelihood of school closures corresponding to 24% of the average share of school closures.
This finding remains highly robust to an extensive set of sensitivity tests and when accounting for potential confounding factors, such as the severity of the pandemic, economic conditions, healthcare capacity, women’s labour supply or gender role attitudes. We also consider the potential influence of the government’s political orientation, the presence of women ministers in education or health, and the average age of government members. None of these factors alter the negative relationship between women’s political representation and school closures.
Figure 2 Share of women in national governments (in 2019) and total weeks of school closures between 1 March 2020 and 30 September 2021
Source: Figure 2 from Danzer et al. (2024)
Evidence on overall containment stringency and other measures
Supplementary analyses suggest that our main results reflect women’s preference for different policy approaches to combat the pandemic, rather than a reluctance to implement strict measures or a lower concern about the spread of Covid-19 (see Figure 3): there are no gender differences regarding the overall stringency index of implemented containment measures or the overall health response index. Similarly, we could not detect any significant gender differences regarding the likelihood of implementing more gender-neutral policies like face coverings, restrictions on gatherings, or international travel restrictions.
Overall, the evidence suggests that governments with higher female representation were equally committed to fighting the pandemic, but generally prioritised other policy tools than school closures.
Figure 3 share of women in national governments (2019) and containment stringency
Note: The figure summarises the estimated relationships between the share of women in national governments (in 2019) and (a) the likelihood of implementing selected containment measures and (b) composite indices reflecting the general strictness of all government policies (Stringency index) or health-related measures, including investment in Covid-19 vaccines and emergency funding for healthcare (Health response index) – 1 March 2020 and 30 September 2021 (coefficients rescaled for comparability).
Source: based on Tables 8 and A.9 in Danzer et al. (2024).
Micro evidence on policy priorities of women
To provide more direct evidence on individual-level gender differences in policy priorities during the pandemic, we analysed micro data from the German Bundestag during the first year of the pandemic. Lacking individual votes on school closures in European parliaments, we collected and text-analysed 7,221 written questions – a key information and oversight tool – posed by German parliamentarians to the government.
As Figure 4 shows, women parliamentarians were significantly more likely to ask questions about school closures than their male counterparts. Despite women comprising only about one-third of the parliament, they posed 65% of all school closure-related questions. Notably, this pattern cannot be explained by party affiliation, age, parental status, or marital status, thereby highlighting a greater concern among women politicians about school closures, aligning with broader evidence of differing priorities between male and female politicians.
Figure 4 Micro evidence on gender differences in policy priorities in the German Bundestag
Source: Figures based on Table 9, Danzer et al. (2024).
Conclusion and implications for future politics
The Covid-19 pandemic has demonstrated the critical role that women’s representation in political power plays in shaping policy responses. Our study reveals that higher proportions of women government members are associated with a lower likelihood of school closures.
Importantly, our results do not suggest that women politicians were less concerned about the spread of Covid-19. Instead, they indicate that women in political power preferred different policy mixes to combat the pandemic, pointing to distinct policy priorities among women politicians.
Perhaps drawing from personal experiences or information provision within gendered networks, they may have been more and earlier aware of the negative implications of school closures on children and families, particularly working mothers, and their far-reaching consequences on education, labour markets, and social inequality. By prioritising alternative containment measures, public health objectives can be better balanced with the socio-economic wellbeing of families, children and parents.
As countries prepare for future challenges, promoting gender diversity in political institutions and crisis taskforces should be therefore a strategic priority. By doing so, policymaking will benefit from a wider array of insights and social preferences to craft policies that better serve the needs of all citizens.
Source : VOXeu
