Tag: Gender

Assessing a Model for the Implementation of an Equal Pay Review and Reporting (EPRR) Methodology in Georgia

20211102 Assessing a Model of Equal Pay

Georgia’s gender pay gap has started to attract the attention of the population and policymakers alike. The gap persists despite working women generally reporting better labor-market skills and personal characteristics. It has been argued that this could be the result of systematic gender-based workplace wage discrimination, resulting in unequal pay for equal work. The discussion that ensued highlights how the fight to guarantee equal pay for equal work could benefit from establishing an Equal Pay Review and Reporting Mechanism. In response, the ISET-PI team – after reviewing the best international practices – devised and tested an excel based tool that could help companies and governmental agencies identify, monitor, and fight gender discrimination in Georgia. The main quantitative result of the exercise identified that, should reporting be made mandatory, extending the obligation to companies that employ up to 50 people would make the administrative costs for companies and public administration up to twenty times higher; thus, the usefulness of the tool was found to be substantially limited when applied to smaller companies. Finally, the exercise emphasized the reluctance of companies to provide the data required, leading to the conclusion that the successful implementation of such an initiative would require the enforcing agency to have the legal authority to sanction failures to provide the necessary data.


One of the key gender inequality indicators is the gender pay gap – or gender wage gap – calculated as the average difference between the remuneration for men and women in the labor market. Its evolution is monitored worldwide, and closing this gap is considered a key step towards more inclusive and prosperous economies and societies. According to the World Economic Forum, as of 2020, no country (including the top-ranked ones) had yet achieved gender parity in wages.

In Georgia, the unadjusted hourly gender pay gap amounts to 17.7 percent of the average male hourly wage (UN Women, 2020). Moreover, when controlling for personal characteristics of men and women, the adjusted hourly gender pay gap in Georgia is estimated to be 24.8 percent (UN Women, 2020). This implies that women, on average, have better observable labor-market characteristics but are still paid less than men.

These findings prompted a core discussion within the Georgian society on the presence of unequal pay for equal work in Georgia as one of the possible reasons for the gap and how to tackle the problem. The idea of equal pay for equal work entails that individuals in the same workplace are given equal pay if they perform the same type of work. Consequently, this potential source of the pay gap can only be verified at the individual employer level. This is accomplished by calculating the unexplained gender pay gap at the organizational/employer level and validating whether, and why, these differences exist.

Given the attention the topic holds in the national discourse, ISET Policy Institute created and tested an excel tool, built in line with the international best practices and adapted to the Georgian context, to help employers and government offices identify and measure the differences in wages between men and women performing equal work. During this process, the team learned several noteworthy lessons, as summarized in this policy brief.

International Experience

There is growing consensus that transparency is critical when dealing with pay inequality and, therefore, gender pay reporting should become the norm. Since 2010, several (mostly developed) countries have introduced reporting schemes to monitor gender pay gaps, promote awareness about gender equality issues throughout society (particularly among employees), and increase organizations’ accountability to address gender inequalities (Equileap, 2021).

However, the gender pay gap is a key issue for which the disclosure of information remains particularly low. Equileap’s 2021 report revealed that 85 percent of organizations worldwide did not publish information on remuneration differences between female and male workers in 2020.

Three countries, according to Equileap, lead the way in gender pay gap reporting: Spain, the UK, and Italy (Figure 1). In each of these top three countries, reporting is mandatory.

Figure 1. Percentage of organizations publishing gender pay information, per country

Source: Equileap, 2021. The figure only includes countries for which more than 49 surveyed organizations were included in the Equileap dataset.

However, even in these countries, and, more generally, in all countries scrutinized by Equileap but Iceland, firms with 50 or fewer employees are not required to report on gender pay gaps.

The Case of Georgia

Georgian legislation clearly establishes the principle of equal pay for equal work for all employees. The requirement applies to both public and private organizations. Nevertheless, enforcement of the law remains a significant challenge.

At present, Georgia has no reporting requirements regarding employee salaries for private organizations. It has not yet designed a reporting scheme for equal pay for equal work, nor has it assigned the task of collecting this information to any governmental body.

Moreover, Labour Inspectorate representatives state that few wage discrimination cases are currently being filed in the country. The main reason behind this is that norms regarding equal pay for equal work have never been properly specified. In addition, there are no explicit criteria defining the concept of ‘equal work’. Thus, employers and employees alike do not seem to fully understand the phrase – equal pay for equal work.

The Excel Tool

After a careful review of the three tools presently utilized to calculate gender pay inequality (the Swiss Logib, the German Logib-D, and the Diagnosis of Equal Remuneration (DER) tool developed by UN Women), ISET-PI built a Georgian model as a modified version of the DER tool that is adapted to the Georgian context and includes some variables from the Swiss tool.

The tool itself is an excel file with several worksheets. The two main facets are the inputted data sheet and the results sheet. Companies may input information on their employees in the data sheet, and the findings will then be demonstrated in the results sheet. The tool first identifies people performing the same work, and classifies jobs based on their official titles, alongside managerial responsibilities and skill requirements. After individuals are grouped by job, the tool calculates the average salary within each group separately for men and women. Thereafter, the pay gap is calculated based on the average salary for the two gender groups.

With the support of the Employers’ Association, several companies of all sizes were approached to test the tool. Unfortunately, only a few agreed to participate, and just two completed the trial: one small-sized enterprise (with 50 or fewer employees) and a large-sized enterprise (with 250 or more employees).

While low participation rates have significantly limited our analysis, we still obtained several important insights which are discussed in the next subsection.


Firstly, it is important to note that companies’ willingness to share anonymized salary data was very low, even among the companies that completed the test.

Secondly, the usefulness of the tool for obtaining a comprehensive view of equal pay for equal work in small companies (with 50 or fewer employees) appeared fairly limited as few people within the same firm perform the same job.

Thirdly, we performed a simple cost assessment exercise to evaluate the compliance costs – to both companies and the government – of collecting and reporting the gender pay gap. We found that extending the data collection requirement to small companies would increase the compliance costs by up to 20 times (high-cost scenario) compared to an example where small companies are exempt. This is because there are many more small companies in Georgia (146,802), than those classified as medium or large ones (2,752 and 609, respectively).

In addition, during the implementation of the exercise, we became aware of the following:

  • Under the existing legal provisions, it would be extremely difficult to introduce the EPRR in a mandatory format – no governmental agency could sanction companies for failing to comply.
  • Opting for the mandatory option and sanctioning the emergence of unequal pay in certain job categories could incentivize companies to manipulate the data input. In this case, therefore, it would be ill-advised to provide the full tool to companies, as they could more easily adjust data inputting to obtain more favorable indicators through successive iterations.


Setting up an EPRR system is one way to contribute to the implementation of the equal pay for equal work principle.

Designing the Georgian Model for the Implementation of an Equal Pay Review and Reporting Methodology generated several useful insights that might prove valuable for policymakers in Georgia and other developing countries:

1) The EPRR instrument can be utilized for the analysis of gender pay gaps within companies with more than 50 employees. Within smaller companies, evaluating the gender pay gap significantly increases the costs to society, while providing rather limited additional information.

2) The decisions about whether to provide the analytical part of the tool to companies, and whether reporting should be voluntary or mandatory should be taken jointly. If the goal is to provide an instrument to the agency enforcing the equal pay for equal work principle and to facilitate appeals from workers, the tool should be made mandatory. However, in this case, companies should only provide the input data, without having access to the part of the tool that assesses pay gaps at the job level. On the other hand, if the goal of the reform is to support willing companies in their efforts to eliminate unequal pay for equal work conditions, a non-mandatory form may be preferable. In this instance, companies should have access to the full version of the tool. This would allow them to better understand the dynamics that lead to unequal pay and thus put in place internal remedial actions.

3) If the goal is to provide a tool to the agency enforcing the equal pay for equal work principle, it is crucial that any gaps in the associated legislation are closed. As such, the enforcing agency should be capable of sanctioning failures to provide the required data, and prosecuting violations of the equal pay for equal work principle.

Finally, it is important to note that testing the application of the equal pay for equal work principle at the company level through an EPRR system, while useful for identifying potential causes of the gender pay gap and the existence of gender disparities within companies, is just a first step in a longer and more complex process. Once disparities are identified, both companies and enforcing agencies should follow up with additional research and analysis to determine whether these disparities are linked to discriminatory practices, and what type of remedial options could be adopted.


Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Female Representativeness and Covid-19 Policy Responses: Political Representation and Social Representativeness

20210928 Female Representativeness and Covid-19 Policy Responses Image 01

There is anecdotal evidence that countries with female leadership in policymaking are more efficient in combating the Covid-19 pandemic. This paper studies whether countries with high female representativeness in political and social layers respond differently to the Covid-19 outbreak. We explore patterns at a cross-country level, which enables us to consider the variation of gender implicated institutions. Our findings indicate that it is women’s social representation, rather than female political leadership, that has the potential to capture cross-country variation in Covid-19 policy responses. Our study confirms that well-functioning and effective institutions are not established from the top-down but rather from the bottom-up.


In light of the Covid-19 outbreak and the resulting actions developed and implemented by countries worldwide, questions have been raised about government policy responses and what can trigger them. The pandemic brought forward the need for measures that help mitigate the spread of the virus such as hand washing, reduced face touching, face mask policies, and physical distancing. In many countries, the implementation of lockdowns and social distancing measures had a large impact on employment, including reductions in working hours, furloughs, and work from home arrangements (Brodeur et al., 2020; Coibion et al., 2020; Gupta et al., 2020). There are notable concerns about the potential damage non-pharmaceutical interventions can inflict on economies and labor markets (Andersen et al., 2020; Kong and Prinz, 2020). Further, the implementation of these measures requires certain institutional and individual behavioral changes. While some countries were successful in developing and implementing policy responses that addressed the challenges of the pandemic, others have experienced considerable difficulties.

There is anecdotal evidence suggesting that countries with female leadership in governmental policies are more efficient in combating the Covid-19 pandemic. Several articles from prominent media outlets, such as CNN, The Conversation and Forbes, hypothesize that female leaders are systematically better at managing the pandemic and that this divergence can be attributed to gender differences in management style and risk-taking behavior.

This policy paper explores whether countries distinguished by higher female representation in government policies, both in development and implementation, responded differently to the Covid-19 outbreak, and if so, how the response differed from other countries. For this purpose, we identify two layers of female representation: political representation and social representativeness. The layer of political representation considers the role of women’s representation in public policy design and implementation at the top level of executive and legislative institutions. Social representativeness captures women’s representativeness in different layers of society and spheres of life. It reflects social norms, legal inequality between men and women in different spheres of private, economic, and business life, as well as realized gender inequality, e.g., in labor market participation, education, or local leadership.

With respect to political representation, we address the question of whether countries distinguished by a higher female representation at top executive and legislative levels differ in terms of policy responses to Covid-19. With respect to social representativeness, we aim to capture the variation in these responses that may originate from differences in the expected reaction of the public, which in turn is driven by women’s representativeness in different layers of society. We derive evidence-based conclusions capturing the role of female leadership at the country’s executive and legislative level, as well as the role of gender representativeness in other layers and institutions of society.

The motivation for this research stems from the extensive literature on differences in values and social attitudes between men and women. For example, women have been shown to be more trustworthy, public-spirited, and likely to exhibit ‘helping’ behavior (Eagly and Crowley, 1986), vote based on social issues (Goertzel, 1983), score better on ‘integrity tests’ (Ones and Viswesvaran, 1998), take stronger stances on ethical behavior (Glover et al., 1997; Reiss and Mitra, 1998) and behave more generously when faced with economic decisions (Eckel and Grossman, 1998). Thereby, one may ask to which extent these differences transmit to public policies in societies where women are better represented, either politically or socially. While our study primarily concerns Covid-19 policy responses, we discuss other related literature on the relationship between women’s representativeness and public policy in the next section.

Our analysis shows that it is the women’s social representativeness layer, which can explain government reactions to the Covid-19 pandemic. This goes in line with the institutionalist literature, suggesting that more a gender-balanced character of institutions translates into policy measures and related outcomes. With this finding, our study suggests further evidence on the central role of institutions. Consistent with the existing evidence, we claim that well-functioning and effective institutions are not established from the top-down, but rather from the bottom-up (Easterly, 2008; Dixit, 2011; Greif, 2006). In such institutions, women’s participation in labor markets, businesses, and other spheres is essential as these are factors that distinguish countries in their response to the pandemic. While the evidence provided is suggestive, it opens further avenues for studies to assess causal relationships.

Covid-19 Policy Measurements

To conduct our analysis, we collect data from a number of different sources. For data on the Covid-19 situation and government policy responses, we use the Our World in Data portal. This online platform compiles a number of data sources, most of them updated on a daily basis. Statistics on female participation and leadership is retrieved from the World Bank and UNDP. Summary statistics of the variables are reported in Table A1 of the Appendix.

The policy response variables are based on a number of different measures implemented by national governments. These are aggregated into three composite indices: Stringency, Containment & health, and Economic support. (The index methodology can be found here.) We present the components of the three indices in Table 1 and a detailed description of the policy measures and their scoring in Appendix C.

As seen in Table 1, the Stringency and Containment & health indices have some common dimensions; containment & closure policies (C1 – C8) and public information campaign (H1). Both are rescaled to a value from 0 to 100 (100 = strictest). The Economic support index records measures such as income support and debt/contract relief and does not share any common dimensions with the other two policy response indices. The scale of the index also ranges from 0 to 100 (100 = full support). The extent of heterogeneity in government policy responses across countries is illustrated in Figures 1 – 3. While containment and closure policies are stricter in many Asian and Latin American countries, economic support is more extensive in many European countries, Canada, New Zeeland, and few other countries.

 Table 1. The structure of the Covid-19 policy measurements.

Note: Categories and assigned values of policy measurements are in Appendix C.

Figure 1. Stringency Index

Note: A choropleth map shows countries/territories by their Stringency index score, based on data collected from the portal of https://ourworldindata.org/policy-responses-covid. Countries are grouped into five groups (quantiles), from the lowest to the highest values of the index.

Figure 2. Economic support index.

Note: A choropleth map shows countries/territories by their economic support index score, based on data collected from the portal of https://ourworldindata.org/policy-responses-covid. Countries are grouped into five groups (quantiles), from the lowest to the highest values of the index.

Figure 3. Containment & health index.

Note: A choropleth map shows countries/territories by their Containment and health index scores, based on data collected from the portal of https://ourworldindata.org/policy-responses-covid. Countries are grouped into five groups (quantiles), from the lowest to the highest values of the index.

Female Representativeness: Layers and Indicators

Multiple studies in economics and political science suggest that the gender of public officials shapes policy outcomes (Chattopadhyay and Duflo, 2004; Iyer et al., 2012; Svaleryd, 2009). Evidence suggests that increasing the number of women in higher ranks of public administration (legislative bodies and ministries) has a substantial impact on the political office and policymaking (Borrelli, 2002; Davis, 1997; Reynolds, 1999). On the other hand, a number of studies demonstrate that gender has no association with policy outcomes (Besley et al., 2007; Besley and Case, 2003; Bagues and Campa, 2021). The role of the institutional setting and environment can, thus, be decisive in this regard. Women are also found to be more concerned about social policy issues and prefer higher social spending than men (Lott and Kenny, 1999; Abrams and Settle, 1999; Aidt and Dallal, 2008). Further, women are more likely to use a collective or consensual approach to problem and conflict resolution rather than an approach founded on unilateral imposition (Rosenthal, 2000; Gidengil, 1995).

In our study, the political representation layer is measured as female leadership at a country’s executive level (representation in government cabinets) and participation at the legislative institution (parliament) level. To assess this, we consider the following indicators: 1) the presence of a female president or prime minister and proportion of women in ministerial positions, and 2) women’s representativeness in legislative bodies measured as the proportion of seats held by women in national parliaments. The variation of these indicators across countries is illustrated in Figures B4 – B6 in the Appendix.

Our approach to social representativeness is in line with social role theory. This framework provides a theoretical explanation of a structural approach to gender differences (Eagly, 1987; Eagly and Karau, 2002; Wood and Eagly, 2009). It claims that men and women behave according to stereotypes associated with the social roles they occupy, and these differences can, in turn, influence the role of women in local governance and leadership. In line with other research on gender, the social role theory proposes a rigorous framework for analyzing the gendered aspect of government organizations. For instance, evidence shows that women tend to be more collaborative and democratic, hence demonstrating a more caring and community-oriented behavior (Eagly and Johannesen-Schmidt, 2001).

The gender aspect of local governance indicates that the personal preferences and opinions of leaders predominate and shape policymaking (Besley and Coate, 1997). Female leaders (including municipality heads) are more inclined to favor the inclusion of citizens in the decision-making process (Fox and Schuhmann, 1999; Rodriguez-Garcia, 2015), implying that the society is a more informed and engaged stakeholder in the public policymaking (Ball, 2009).  Given that municipalities are taking on a greater and more interactive role in citizens’ well-being, they become a key channel in reinforcing trust in government. Furthermore, the literature finds an interrelationship between female voters and government outcomes, whereby women’s enfranchisement affects government size and spending (Lott and Kenny, 1999; Miller, 2008, Aidt and Dallal, 2008). As such, this can lead to improvements in government outcomes and policy effectiveness. The evidence from Bloomberg’s Covid-19 Resilience Ranking suggests that success in containing Covid-19 while minimizing disruption appears to rely more on governments fostering a high degree of trust and societal compliance.

Furthermore, the patterns of gender relations in societies reflect formal and informal institutional rules and policies. Gender equality enhances good governance and helps to further improve relationships between government and citizens (OECD 2014). Similarly, Elson (1999) argues that labor markets are structured by practices, norms, and networks that are “bearers of gender”. Societies with better legal frameworks for women have more balanced gender participation in labor markets, governance, and leadership, along with more equal gender roles and less gender-biased stereotypes. We anticipate that better representation of women in policymaking in such societies is also reflected in the choice and effectiveness of Covid-19 policy measures.

Building on the above theories explaining the relevance of women’s representativeness in diverse societal layers for policy development and implementation, we identify three indices that have the potential to capture the effect of social representativeness – Women, Business and the Law index (WBLI), Gender Development Index (GDI) and Gender Inequality Index (GII). The WBLI is composed of eight indicators, covering different areas of the law related to the decisions women make at various stages of their career and life. These indicators include mobility, workplace, salary, marriage, parenthood, entrepreneurship, assets, and pension. Hyland et al. (2020) show that, globally, the largest gender inequalities are observed in the areas of pay and parenthood. That is, women are most disadvantaged by the legal system when it comes to compensation and how they are treated once they have children. The index scales from 0 to 100 (100 = equal opportunities). The diagram in Figure 4 illustrates how the components of the WBLI index measure key activities of economic agents throughout their life.

Figure 4. The linkages of 8 indicators in Women, Business and the Law index (WBLI)

Source. Women, Business and Law, 2020. World Bank Group.

The second index, the GDI, measures gender inequality in the achievements in three basic dimensions of human development: Health, measured by life expectancy at birth; Education, measured by expected years of schooling for children and mean years of schooling for adults aged above 25; and Command over economic resources, measured by estimated earned income.  The same dimensions are included in the Human Development Index (HDI), and the GDI is defined as the female-to-male HDI ratio (i.e. perfect gender equality corresponds to a GDI equal to one).

Turning to the third index measuring social representativeness, the GII reflects gender-based disadvantages in the following dimensions—reproductive health, empowerment, and the labor market. The index measures the loss in potential human development due to gender inequality in achievements across these dimensions. It ranges from zero, where women and men fare equally, to one, where one gender fares as poorly as possible in all measured dimensions. One of the dimensions of the GII, women’s empowerment, has a sub-dimension – “Female and male shares of parliamentary seats”, one of our indicators measuring political representation. Generally, we do not consider the two layers being as mutually exclusive, but intersections are expected to be minimal.

Central to our study, the three indices capturing social representativeness in a country encompass the institutional quality of its society from a gender development perspective. The distribution of each index across countries is shown in Figures B1 – B3 (See Appendix B).

Women’s Representativeness and Covid-19 Policy Responses: Partial Correlation Analysis

In this section, we explore the relationship between Covid-19 policy responses and the measures of political representation and social representativeness. For this purpose, we explore (i) correlations between the indicators and indices of the political and social representation layers and (ii) partial correlations between these measures and policy response indices.

We start with a correlation analysis of the different indicators in the layers. It shows that the WBLI is in high correlation with other representativeness variables. This index captures the legal equality between women and men which has been shown to be “associated with a range of better outcomes for women, such as more entrepreneurship, better access to finance, more abundant female labor supply, and reductions in the gender wage gap”. (WB, 2021). One can think of the GDI and GII indices, as well as the political representativeness indicators, as reflections of a broad policy framework in diverse areas of social, business, and legal activities. A legal environment that promotes gender equality, even if not sufficient by itself, is likely to lead to progress in these areas. Indeed, Hyland et al. (2020) show that greater legal equality between men and women is associated with a lower gender gap in opportunities and outcomes, fewer female workers in vulnerable positions, and greater political representation of women. This way, the WBLI may capture key predispositions for women’s representativeness in society. Further, Hyland et al. (2021) show that the WBLI index is in high (partial) correlation with country GDP per capita, polity score, legal origin, religion and geographic characteristics. This evidence suggests that the WBLI may have the capacity to reflect important country characteristics which ultimately shape cross-country institutional variation.

Table 2. Scatterplot table for GDI, GII and Women, Business and the Law Index, Proportion of seats in parliament held by women and Proportion of ministerial seats held by women.

Note: Scatterplots are constructed for 149 countries. Fitted lines are based on a quadratic function, shaded areas indicate 95 percent confidence intervals and country population is used for weights applied for fitted lines and bubbles. For each scatterplot, correlation coefficients and their significance are reported. *** p<0.01, ** p<0.05, * p<0.1.

Next, we explore partial correlations of these indicators with Covid-19 policy responses (Table 3). In this analysis, we control for a number of factors that potentially confound the relationship between a particular policy response and representation layer. Specifically, we control for (i) the number of infected cases per million inhabitants, (ii) the number of deaths per million, (iii) GDP per capita, and (iv) life expectancy. The number of infected cases and deaths enter the model in order to control for country differences in the spread and consequences of the virus. GDP per capita captures the stage of country development, accounting for cross-country differences in resource capacities and constraints. Both of these control variables are claimed to have an important role in Covid-19 related research (Coscieme et al., 2020; Aldrich and Lotito, 2020; Elgar, Stefaniak and Wohl, 2020; Gibson, 2020; Conyon and Thomsen, 2020). Life expectancy is an important proxy for country inhabitants’ resilience against the virus, conditioned by health and health infrastructures.

Significant correlations are observed between the WBLI and the three policy response indices. The correlation between the WBLI and Stringency (and Containment & health) index is negative, implying that lighter restrictions have been imposed in countries with better business and legal conditions for women. A positive correlation is observed between the WBLI and the economic support index, suggesting that countries with better conditions for women in diverse business and societal areas have provided more extensive economic support in the pandemic. This finding is in line with existing evidence showing that women are more concerned about social policy issues and prefer higher social spending than men (Lott and Kenny, 1999; Abrams and Settle, 1999; Aidt and Dallal, 2008). Also, lighter restrictions and more generous economic support do not presume any trade-off in terms of the allocation of financial resources constrained by a state budget.

Interestingly, we do not observe significant correlations between policy responses and other indicators of women’s representativeness. The only exception is a correlation between GDI and the Containment & health index, which is significant at the 10% level and hinges heavily on two outliers (if we drop the two outliers, the P-value of the correlation increases from 0.0931 to 0.2735).

Table 3. Scatterplots of policy responses and social representativeness and political representation variables.

Note: Scatterplots are constructed for 133 countries. Fitted lines are based on a quadratic function, shaded areas indicate 95 percent confidence intervals and country population is used for weights applied for fitted lines and bubbles. Correlation coefficients are reported with significance levels: *** p<0.01, ** p<0.05, * p<0.1.

In our partial correlation analysis, we do not control for the direct effects of the gender dimension of social norms and practices. Social norms, practices, as well as informal and formal rules can, however, explain a substantial part of the gender gap (Hawkesworth, 2003; Mackay, 2009; Franceschet, 2011; Elson, 1999; Froehlich et al., 2020) relevant for making decisions. Our measures of women’s political and social representativeness do not fully cover gender differences in norms and practices. As Hyland et al. (2020) point out, de-jure female empowerment does not necessarily translate into de-facto empowerment, especially in countries with social norms and informal rules that result in low representation of women in diverse societal spheres. The authors indicate that laws are actionable in a short period, while more time is needed to bring changes in social norms.  In our paper (Grigoryan and Khachatryan, 2021), we attempt to address this issue by incorporating the Social Institutions and Gender Index (SIGI) into the model and evaluating the confounding effect on the covariates of the model. We show that the WBLI captures the effect of the gender gap owing to social norms and practices on Covid-19 policy responses as measured by SIGI. This result suggests that the endogeneity arising from the omission of a measure of such a gender gap is likely to be minimal.

Discussion and Conclusions

Our correlation analysis suggests that it is the layer of women’s social representativeness that can explain the policy reactions of governments in times of the Covid-19 pandemic. This result is in line with the institutionalist literature on gender inequality and social role theory, which suggests that a more gender-balanced character of institutions translates into policy measures and related outcomes. Among the three indices constituting the social representativeness layer, the WBLI is, by construction, more inclusive in terms of capturing women’s role in diversified societal areas. From Table 2, we observe that the WBLI is the only index that is in strong correlation with all other indicators. We also identify strong dominance of the WBLI in correlations with policy responses: it is the only indicator that is significantly correlated with all three policy response measurements (Table 3).

To conclude, our results establish an association between female social representativeness, as measured by the (legal) equality of opportunities between men and women, and Covid-19 related policies. One potential interpretation of these findings concerns the central role of the gender balance in different institutions and layers of society in understanding policy responses to the Covid-19 pandemic. While it was parliaments and governments that implemented policies, we find that the measures undertaken correlate more strongly with factors related to the social representativeness of women rather than those related to their political representation. This suggests a dominant role of gender-balanced institutions at the ‘grass root’ level in terms of the scale and scope of the crisis response. Naturally, these institutions may result (or be correlated) with more gender-balanced political representation, but the latter alone is not helpful in explaining the variation in the reaction to the pandemic.  These results underline the importance of balanced gender representation in the labor market, business, and other spheres of social life.  Further investment and development of ‘grass root’ institutions that improve women’s socioeconomic opportunities, could provide a fundamental foundation for policy development in a crisis situation.

There could also be alternative interpretations of our findings. There is rich evidence that the gender dimension is deeply implicated in institutions (Acker, 1992; Chappell and Waylen, 2013; Lovenduski, 2005). Gender norms and gender practices have been shown to have an influence on the operation and interaction between formal and informal institutions (see, for instance, Chappell, 2010; Krook and Mackay, 2011; Chappell and Waylen, 2013) and the gender dimension of political institutions is reflected in their practices and values, hence affecting their outcomes (such as laws and policies), formation, and implementation (for instance, Acker, 1992). In turn, governmental policies and rules shape societal norms and expectations. These considerations imply that our results could be driven by the overall values, culture, and institutions of respective societies. These factors would both result in a more gender-neutral legal environment and ‘grass-root’ institutions, and ultimately, distinguish countries in their response to the Covid-19 pandemic. In this way, our results open an avenue for future studies in this important domain to better understand the causality of observed relationships.


  • Abrams, B. A., & Settle, R. F. (1999). Women’s suffrage and the growth of the welfare state. Public Choice100(3), 289-300.
  • Acker, J. (1992). From sex roles to gendered institutions. Contemporary sociology21(5), 565-569.
  • Aidt, T. S., & Dallal, B. (2008). Female voting power: the contribution of women’s suffrage to the growth of social spending in Western Europe (1869–1960). Public Choice134(3), 391-417.
  • Andersen, A. L., Hansen, E. T., Johannesen, N., & Sheridan, A. (2020). Pandemic, shutdown and consumer spending: Lessons from Scandinavian policy responses to Covid-19. arXiv preprint arXiv:2005.04630.
  • Bagues, M., & Campa, P. (2021). Can gender quotas in candidate lists empower women? Evidence from a regression discontinuity design. Journal of Public Economics, 194, 104315.
  • Ball, C. (2009). What Is Transparency? Public Integrity 11 (4): 293–308. doi:10.2753/PIN1099-9922110400.
  • Besley, T., & Coate, S. (1997). An economic model of representative democracy. The Quarterly Journal of Economics112(1), 85-114.
  • Besley, T., & Case, A. (2003). Political institutions and policy choices: evidence from the United States. Journal of Economic Literature41(1), 7-73.
  • Besley, T., Folke, O., Persson, T., & Rickne, J. (2017). Gender quotas and the crisis of the mediocre man: Theory and evidence from Sweden. American economic review107(8), 2204-42.
  • Borrelli, M. A. (2002). The first lady as formal advisor to the president. Women & Politics 24 (1), 25–45.
  • Brodeur, A., Gray, D., Islam, A., & Bhuiyan, S. (2020). A literature review of the economics of Covid-19. Journal of Economic Surveys.
  • Chappell, L., & Waylen, G. (2013). Gender and the hidden life of institutions. Public Administration91(3), 599-615.
  • Chappell, L. (2010). Comparative gender and institutions: Directions for research. Perspectives on Politics, 183-189.
  • Chattopadhyay, R., & Duflo, E. (2004). Women as policy makers: Evidence from a randomized policy experiment in India. Econometrica, 72(5), 1409-1443.
  • Coibion, O., Gorodnichenko, Y., & Weber, M. (2020). The cost of the Covid-19 crisis: Lockdowns, macroeconomic expectations, and consumer spending (No. w27141). National Bureau of Economic Research.
  • Davis, R. H. (1997). Women and power in parliamentary democracies: Cabinet appointments in Western Europe, 1968-1992, Volume  2.  U of Nebraska  Press.
  • Dixit, A. K. (2011). Lawlessness and economics. Princeton University Press.
  • Eagly, A. and M. Crowley (1986, November). Gender and helping behavior. a meta-analytic review of the social psychological literature.
  • Eagly, A. H., & Karau, S. J. (2002). Role congruity theory of prejudice toward female leaders. Psychological review109(3), 573.
  • Eagly, A. H., & Johannesen‐Schmidt, M. C. (2001). The leadership styles of women and men. Journal of social issues57(4), 781-797.
  • Eagly, A. H. (1987). Reporting sex differences. Eckel, C. and P. Grossman (1998). Are women less selfish than men? evidence from dictator experiments. Economic Journal 108 (448), 726–35.
  • Easterly, W. (2008). Institutions: Top down or bottom up?. American Economic Review98(2), 95-99.
  • Elson, D. (1999). Labor markets as gendered institutions: equality, efficiency and empowerment issues. World development27(3), 611-627.
  • Fox, R. L., & Schuhmann, R. A. (1999). Gender and local government: A comparison of women and men city managers. Public Administration Review, 231-242.
  • Franceschet, S. (2011). Gender policy and state architecture in Latin America. Politics & Gender7(2), 273.
  • Froehlich, L., Olsson, M. I., Dorrough, A. R., & Martiny, S. E. (2020). Gender at work across nations: Men and women working in maledominated and female-dominated occupations are differentially associated with agency and communion. Journal of Social Issues, 76(3), 484-511.
  • Gidengil, E. (1995). Economic man—social woman? The case of the gender gap in support for the Canada-United States free trade agreement. Comparative Political Studies28(3), 384-408.
  • Glover, S. H., M. A. Bumpus, J. E. Logan, and J. R. Ciesla (1997). Re-examining the influence of individual values on ethical decision making. In: From the universities to the marketplace: the business ethics journey, 109–119.
  • Goertzel, T. (1983). The gender gap: Sex, family income and political opinions in the early 1980’s. JPMS: Journal of Political and Military Sociology 11 (2), 209.
  • Greif, A. (2006). Institutions and the path to the modern economy: Lessons from medieval trade. Cambridge University Press.
  • Grigoryan, A. & K. Khachatryan (2021). Covid-19 policy responses through the lenses of female political and social representation. Mimeo.
  • Gupta, R., Ghosh, A., Singh, A. K., & Misra, A. (2020). Clinical considerations for patients with diabetes in times of Covid-19 epidemic. Diabetes & metabolic syndrome14(3), 211.
  • Hartl, T., K. W. and E. Weber (2020, 4). Measuring the impact of the German public shutdown on the spread of Covid-19. Covid Economics 1 (11), 25-32.
  • Hawkesworth, M. (2003). Congressional enactments of race-gender: Toward a theory of raced-gendered institutions. American Political Science Review, 529-550.
  • Hyland, M., S. Djankov, and P. Goldberg. 2020. Gendered Laws and Women in the Workforce. American Economic Review: Insights, 2 (4): 475-90.
  • Hyland, M., S. Djankov, & P. Goldberg. 2021. Do Gendered Laws Matter for Women’s Economic Empowerment? Peterson Institute for International Economics Working Paper No. 21-5.
  • Iyer, L., Mani, A., Mishra, P., & Topalova, P. (2012). The power of political voice: women’s political representation and crime in India. American Economic Journal: Applied Economics4(4), 165-93.
  • Kong, E., & Prinz, D. (2020). Disentangling policy effects using proxy data: Which shutdown policies affected unemployment during the Covid-19 pandemic?. Journal of Public Economics189, 104257.
  • Krook, M. L., & Mackay, F. (2011). Introduction: Gender, politics, and institutions. In Gender, politics and institutions (pp. 1-20). Palgrave Macmillan, London.
  • Lott, Jr, J. R., & Kenny, L. W. (1999). Did women’s suffrage change the size and scope of government?. Journal of political Economy107(6), 1163-1198.
  • Lovenduski, J. (2005). Feminizing politics. Polity.
  • Mackay, F., Kenny, M., & Chappell, L. (2010). New institutionalism through a gender lens: Towards a feminist institutionalism?. International Political Science Review31(5), 573-588.
  • Mackay, F., & Waylen, G. (2009). Feminist institutionalism. Politics & Gender5(2), 237-237.
  • Miller, G. (2008). Women’s suffrage, political responsiveness, and child survival in American history. The Quarterly Journal of Economics123(3), 1287-1327.
  • OECD. (2014). Women, Government and Policy Making in OECD Countries: Fostering Diversity for Inclusive Growth. OECD. doi:10.1787/9789264210745-en.
  • Ones, D. S. and C. Viswesvaran (1998). The effects of social desirability and faking on personality and integrity assessment for personnel selection. Human Performance 11 (2-3), 245–269.
  • Reiss, M. C. and K. Mitra (1998). The effects of individual difference factors on the acceptability of ethical and unethical workplace behaviors. Journal of Business Ethics 17 (14), 1581–1593.
  • Reynolds, A. (1999). Women in the legislatures and executives of the world: Knocking at the highest glass ceiling. World Politics 51 (4), 547–572.
  • Rodriguez-Garcia, M. J. (2015). Local Women’s Coalitions: Critical Actors and Substantive Representation in Spanish Municipalities. European Journal of Women’s Studies 22 (2): 223–240. doi:10.1177/1350506814549424.
  • Rosenthal, C. S. (2000). Gender Styles in State Legislative Committees: Raising Their Voices in Resolving Conflict. Women & Politics 21 (2): 21–45. doi:10.1300/J014v21n02_02.
  • Svaleryd, H. (2009). Women’s representation and public spending. European Journal of Political Economy, 25(2), 186-198.
  • Wood, W., & Eagly, A. H. (2009). Gender identity. Handbook of individual differences in social behavior, 109-125.


(The Appendix can be found in the PDF version of the brief)

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Five Years in Operation: the Polish Universal Child Benefit

Family in the golden hour representing Child Benefits

Over the last five years, Polish families with children have been entitled to a relatively generous benefit of approximately €110 per month and child. Initially granted for every second and subsequent child in the family regardless of income and for the first child for low-income families, the benefit was made fully universal in 2019. With the total costs amounting to as much as 1.7% of Poland’s GDP, the benefit reaches the parents of 6.7 million children and significantly affects these families’ position in the income distribution. Its introduction has led to a substantial reduction in the number of children living in poverty. However, since families with children are more likely to be among households in the upper half of the income distribution, out of the total cost of the benefit, a proportionally greater share ends up in the wallets of high-income families. While the implementation of the benefit has significantly changed the scope of public support to families in Poland, there are many lessons to be learnt and some important revisions to be undertaken to achieve an effective and comprehensive support system.


One of the principal commitments in the 2015 Polish parliamentary elections of the then-main opposition party – Law and Justice (Prawo i Sprawiedliwość, PiS), was introducing a generous child benefit. The purpose of this benefit was to support families and encourage higher fertility, which had been one of the lowest in the European Union for a long time. Following PiS’s electoral victory, the new government introduced a semi-universal child benefit of approximately €110 per month (exactly 500 PLN per month, thus the Polish nickname of “the 500+ benefit”) in April 2016. Initially, the benefit was granted for every second and subsequent child in the family regardless of income and for the first child in low-income families. Since July 2019 (nota bene three months before the next parliamentary elections), it was made universal – all parents with children under the age of 18 are entitled to 500PLN per month for every child.  The benefit is relatively generous (for comparison, it accounts for 17.9% of the minimum wage in Poland in 2021), and universal coverage implies substantial costs for the government budget, totalling about 41bn PLN per year (1.7% of the Polish GDP).

Over the last five years, a number of analyses of the consequences of the benefit’s introduction have been conducted. These have encompassed a variety of socio-economic outcomes for Polish families with children – from a comprehensive assessment of these consequences (Magda et al. 2019) to analyses focused on specific effects of the benefit, such as the impact on women’s economic activity (Magda et al. 2018, Myck 2016, Myck and Trzciński 2019) or poverty (Brzeziński and Najsztub 2017, Szarfenberg 2017). The fifth anniversary of the benefit’s implementation seems to be a good opportunity for a summary and update of previous evaluations of the distributional consequences and financial gains for households resulting from this policy (an overview of all the previous CenEA analyses of the child benefit can be found in CenEA 2021). The results presented in this brief are based on analyses conducted using the Polish microsimulation model SIMPL on data from the 2019 CSO Household Budget Survey (more details in Myck et al. 2021). It should be noted that the analyses do not account for the impact of the Covid-19 pandemic on the material situation of households, as the data was collected before the outbreak. As previous studies suggest, the consequences for households of the pandemic and the series of resulting lockdowns varied greatly depending on various factors, such as the sources of income, sector, and form of employment, thus making it impossible to estimate precisely (Myck et al. 2020a).

The Child Benefit on Household Incomes

Due to its universal character, the distributional consequences of the child benefit payments are directly related to the position of households with children aged 0-17 in the income distribution relative to those without. As households with children are more likely to be in the upper half of the distribution (taking into account the demographic structure of households through income equivalisation), out of the total budget expenditure on the benefit, a proportionally greater share goes to high-income families (Table 1). Families with children in the two highest income decile groups (i.e., belonging to the 20% of households with the highest income) currently receive almost 25% of the total annual expenditure on the child benefit. On the other hand, among the 20% of households with the lowest incomes, families with children receive only 11.7% of the total annual cost of the benefit.

Table 1. Household gains resulting from the child benefit by income decile groups

Source: Myck et al. 2021. Notes: Income decile groups – ten groups each covering 10% of the population, from households (HH) with the lowest disposable income to the most affluent households, calculated on the basis of equivalised incomes.

Compared to the poorest 10% of households, families with children in the highest income decile receive 2.5 times more of the total funds allocated to the benefit.

It is also worth noting that the proportion of benefit in the disposable income is relatively evenly distributed if one considers all households in a given decile (with and without children). The proportional benefits in the first nine income deciles are in the range of 3.4% and 5.3% and only fall to 1.9% in the highest income group. A significant differentiation of the benefit in proportional terms can only be seen when accounting solely for households with children within each income decile. The benefit amounts to as much as 26.9% of the disposable income of households with children in the first decile, and the effect falls in subsequent groups – from 18.9% and 16.4% in the second and third deciles, to only 4.1% in the top decile.

The Child Benefit and the Position of Families With Children in the Income Distribution

Taking into account the magnitude of the policy, the position of families with children in the income distribution relative to other households may, to some extent, be the result of receiving the benefit itself. It is, therefore, reasonable to ask what role the benefit plays in shaping this relative position in the income distribution. Figure 1 presents the number of children under 18 in households by income decile groups when the benefit is included in total household income (left panel) and in a hypothetical scenario when the child benefit payment is withdrawn (right panel). As we can see, the withdrawal of the benefit would cause a substantial change in the relative position of families with children in the income distribution, significantly increasing the number of children in the lowest income groups. While in the current system, the poorest 10% of households include 342 thousand children aged 0-17, this number would be 553 thousand in a system without the benefit. However, the benefit also changes the relative position of high-income households with children. In the current system, the richest 10% of households include 762 thousand children. Subtracting the benefit from their household income would reduce this number to 687 thousand.

Figure 1. The child benefit and its impact on the position of families with children in the income distribution

Source: Myck et al. 2021.

Thus, even when taking into account the income distribution without the benefit, the number of children among the richest 10% of households is almost 25% higher than the number of children in the poorest 10% of households. Looking at the income distribution after including the benefit, there are more than twice as many children in the richest 10% of households than among the poorest 10%. This, in turn, inevitably means that, out of the total cost of the benefit, over twice as much money is transferred to households belonging to the richest deciles as compared to the funds transferred to families belonging to the poorest 10% of households.


With the total costs amounting to 1.7% of Poland’s GDP, the child benefit introduced in April 2016 substantially raised the level of direct financial support for families with children. As shown in this brief, the benefit reaches the parents of 6.7 million children aged 0-17 and significantly affects the position of these families in the income distribution. While, on the one hand, a large proportion of families with children have incomes high enough to be in the highest income groups even without this support , the lowest decile group would include over 200 thousand more children in the absence of the benefit. This confirms that the child benefit alone contributes to a significant improvement in the material conditions of families with children and to a significant reduction in poverty (cf. Brzezinski and Najsztub, 2017; Szarfenberg, 2017). However, the scale of this reduction is modest given the size of the resources involved. This is not surprising given that the bulk of the total costs of the benefit comes from the 2019 program extension to cover all children regardless of family incomes, which largely ended up in the wallets of higher-income families (Myck et al. 2020b). One of the key goals of the benefit upon introduction was to increase the number of births in Poland by easing the material conditions of families with children. Yet despite a radical increase in the level of support, the number of births in Poland over the period 2017-2020 has essentially remained the same as that forecasted by the Central Statistical Office in its long-term population projection of 2014 (Myck et al. 2021). It is thus difficult to consider the benefit a success in terms of this major objective. Moreover, the withdrawal of the income threshold has largely eliminated the negative disincentive effects of the benefit with regard to employment (Myck and Trzcinski 2019). However, it is unclear whether the post-pandemic economic situation will allow for an increase in female labour force participation, which declined following the introduction of the benefit in 2016 (Magda et al., 2018).

The effects of every socio-economic programme should be assessed by comparing cost-equivalent alternatives. Despite all gains the “500+” child benefit has brought to millions of families in Poland over the last five years, the flagship programme of the ruling Law and Justice party does not fare well in this perspective. The need for change seems much broader than the reform of the benefit alone. The benefit was introduced on top of two other financial support mechanisms focused on families with children, namely family allowances and child tax credits, and the three elements have been operating in parallel since 2016. A number of suggestions on creating a streamlined, comprehensive system have been made a long time ago (e.g., Myck et al. 2016). However, a major restructuring of the entire support system with clearly defined socio-economic policy goals in mind seems all the more justified now, when many families may require additional assistance due to the difficult financial situation related to the Covid-19 pandemic.


This Policy Brief draws on the CenEA Commentary published on 31.03.2021 (Myck et al. 2021). It has been prepared under the FROGEE project, with financial support from the Swedish International Development Cooperation Agency (Sida). The views presented in the Policy Brief reflect the opinions of the Authors and do not necessarily overlap with the position of the FREE Network or Sida.


  • Brzeziński, M., Najsztub, M. 2017. The impact of „Family 500+” programme on household incomes, poverty and inequality”, Polityka Społeczna44(1): 16-25.
  • CenEA 2021. Childcare benefit 500+ in CenEA analyses. https://cenea.org.pl/2021/04/06/childcare-benefit-500-in-cenea-analyses/
  • Magda, I., Brzeziński, M., Chłoń-Domińczak, A., Kotowska, I.E., Myck, M., Najsztub, M., Tyrowicz, J. 2019. „Rodzina 500+– ocena programu i propozycje zmian. (“Child benefit 500+: the evaluation of the programme and suggestions for changes”), IBS report.
  • Magda, I., Kiełczewska, A., Brandt, N. 2018. “The Effects of Large Universal Child Benefits on Female Labour Supply”, IZA Discussion Paper No. 11652.
  • Myck, M. 2016. “Estimating Labour Supply Response to the Introduction of the Family 500+ Programme”. CenEA Working Paper 1/2016.
  • Myck, M., Król, A., Oczkowska, M., Trzciński, K. 2021. “Świadczenie wychowawcze po pięciu latach: 500 plus ile?”(„The child benefit after 5 years – 500 plus what?”), CenEA Commentary 31/03/2021.
  • Myck, M., Kundera, M., Najsztub, M., Oczkowska, M. 2016. „25 miliardów złotych dla rodzin z dziećmi: projekt Rodzina 500+ i możliwości modyfikacji systemu wsparcia” („25 billion PLN to families with children: Family 500+ programme and possible modifications of the family support system”),  CenEA Commentary 18/01/2016.
  • Myck, M., Oczkowska, M., Trzciński, K. 2020a. “Household exposure to financial risks: the first wave of impact from COVID-19 on the economy”, CenEA Commentary 23/03/2020.
  • Myck, M., Oczkowska, M., Trzciński, K. 2020b. „Kwota wolna od podatku i świadczenie wychowawcze 500+ po pięciu latach od prezydenckich deklaracji” („Tax credit and child benefit 500+ after five years since electoral declarations”, in PL), CenEA Commentary 22/06/2020.
  • Myck, M., Trzciński, K. 2019. “From Partial to Full Universality: The Family 500+ Programme in Poland and its Labor Supply Implications”, ifo DICE Report 17(03), 36-44.
  • Szarfenberg, R. 2017. “Effect of Child Care Benefit (500+) on Poverty Based on Microsimulation”, Polityka Społeczna 44(1): 25-30.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Economic Perspectives on Domestic Violence | Insights from the FROGEE Webinar | Part 1

Broken mirror with a man's hand representing domestic violence during COVID-19 pandemic frogee

The COVID-19 pandemic and the resulting lockdown restrictions have amplified the academic and policy interest in the causes and consequences of domestic violence. With this in mind, the FREE Network invited academic researchers to participate in an online workshop entitled “Economic perspectives on domestic violence“. This policy brief is the first in a series of two briefs summarizing the papers presented at the workshop. The current brief addresses the presentations that had a more general focus on domestic violence. The second brief will discuss the papers devoted to the domestic violence implications of the pandemic.


Domestic violence (DV), as well as one of its main forms – intimate partner violence (IPV) – are societal issues of massive proportion. The World Health Organization estimates that 1 in 3 women across 80 countries worldwide are victims of IPV during their lifetime (WHO, 2013). IPV imposes huge costs on society: its victims, for instance, are estimated to be twice as susceptible to depression and alcohol abuse, and 16% more likely to give birth to a low birth-weight child (WHO, 2013).

IPV separates itself from other types of violent offenses in several aspects. To start with, the intimate victim-perpetrator relationship causes IPV to be vastly underreported. The victim may have feelings of shame, guilt, and self-blame, which could deter her from seeking support.  Further, IPV and more generally DV cases also have high rates of attrition within the justice system. These distinct characteristics highlight the level of difficulty in developing policies aimed at helping victims of intimate partner abuse. The fact that the prevalence of IPV is widespread and at the same time vastly under-reported, casts doubt on the policy measures and legislation in place today.

This policy brief is the first in a series of two that summarizes the recent economic research on IPV presented in the workshop entitled “Economic Perspectives on Domestic Violence”. The workshop was organized as a part of the Forum for Research on Gender Economics (FROGEE) supported by the Swedish International Development Cooperation Agency (SIDA).

Economic Determinants of Domestic Violence

A number of presentations in the workshop were devoted to the economic determinants of domestic violence.

Andreas Kotsadam presented a paper on the relationship between women’s employment and IPV in Ethiopia. The link between the two is twofold: employment could increase women’s empowerment and, thereby, decrease IPV; however, the boost in empowerment could threaten the man’s status in (male-female) relationships, and lead to violent retaliation. Violence could also be used to extract economic resources from working women. To study which of these mechanisms prevail, the authors conducted an extensive field experiment collaborating with shoe and garment factories in Ethiopia. From a list of qualified job-candidates provided by employers, they randomly assigned 1500 equally qualified women living with partners to either getting a job (treatment group) or not (control group). Prior to treatment, women from both groups were interviewed and asked to answer various questions regarding intimate partner abuse. They were also called to a follow-up survey 6 months later. The statistical analysis of these answers fails to establish a causal link between employment status and the incidence of IPV.

Taking a more theoretical approach, Paul Seabright‘s preliminary work on the determinants of IPV offered a dynamic framework modeling how (unpredictable) economic circumstances and (predictable) individuals’ traits influence domestic violence, as well as formation and dissolution of partnerships. The model distinguishes individuals in their ability to control resources within relationships without the use of violence (“skills”), and in their costs of engaging in violence (“temperament”). The model assumes that individuals with more violent temperaments are on average endowed with lower skills. It predicts women’s income and their risk of IPV should be negatively correlated cross-sectionally, but that positive shocks in income should increase IPV for married women while decreasing it for women with easier exit options. The authors test the model on survey data from Brazil and data on randomized expansions of a food-program in Ecuador. The results support the cross-sectional prediction and confirm that the effect of income shocks depends on exit options, though does not support the prediction of an increase for married women.

Sonia Bhalotra’s presentation addressed the DV consequences of another type of economic shock, namely female and male unemployment, and also considered the role of unemployment benefits as a mitigating factor. By exploiting an extensive dataset covering every court case in Brazil between 2009 and 2017, and information on mass layoffs at the local level, the study finds that the probability of a male being prosecuted for a DV crime increases by 32% when he loses his job and persists at similar levels 4 years after. For female job-loss, the corresponding effect is significantly larger and amounts to 52%. Bhalotra and her co-authors argue that the fact that unemployment of either the man or the woman leads to an increase in domestic violence is consistent with unemployment constituting a negative shock to income and a positive shock to time spent at home. They further argue that the larger impact of female relative to male unemployment is potentially consistent with the “household bargaining model”, which encapsulates the idea that it becomes more difficult for a woman to leave a violent relationship when she is more economically dependent on her partner. Additional analysis shows that eligibility for unemployment insurance increases DV once benefits expire and that this is in turn a result of unemployment benefits increasing peoples’ time in unemployment.

The Role of Police

Part of the workshop was dedicated to the role of the criminal justice system. A fact that stresses the importance of studying police behavior is that domestic abuse cases generally suffer from high legal attrition and most of them are dropped before reaching the court. Variation in the characteristics of law enforcement could likely play a role in explaining differences in DV across contexts.

In this vein, Sofia Amaral introduced a study on the relationship between gender diversity of the police force and domestic violence in the UK. The gender-distribution within law enforcement is believed to directly influence DV in two ways: First, gender-based differences in attitudes and norms may influence police-handling in DV cases. Second, if the gender of the victim aligns with that of the officer, the victim may be more willing to cooperate and disclose evidence. The data shows that the total share of women in the police force is almost equal to that of men, but the tasks performed differ systematically across genders. Women are found to be overrepresented among call-handlers and underrepresented among first-response teams. For each position, Amaral and her co-authors investigate whether changes in gender-distribution influence the rate of legal attrition, rate of repeat victimization, and the amount of time spent at a scene (response duration). By analyzing police force and crime data the study shows that there are substantial efficiency gains from increasing gender diversity, particularly in first-response teams. An increase in the share of females in first-response teams increases response duration, reduces legal attrition, and decreases repeat victimization. There is an even larger effect when a female is the most experienced officer in the team. The gender of the call-handler has no significant effect on the outcomes of interest.

Along somewhat similar lines, Victoria Endl-Geyer presented research on the link between the quality of police response and DV in the UK. More specifically, the research explores how increased police response times, caused by police station closures in 2012, affected the rate of repeat victimization in DV cases. Faster police response times are believed to improve the victim’s cooperation: If the police are quick to arrive at the scene, the victim gets less time to revise the initial assessment that she needed support. The results show that faster police responses are associated with a higher conviction rate. However, they also increase the likelihood of repeat victimization. A potential explanation could be the so-called “reprisal effect” – the perpetrator retaliates with more violence as a response to being reported by his partner.


Many studies on IPV, including some that were presented at the workshop, highlight that an inherently good policy such as improving police response, sometimes leads to unintended negative consequences to victims. In the keynote speech, Leigh Goodmark addressed this topic by critically discussing the history, consequences, and alternatives to criminalization of IPV in the US. As suggested by her recent book, domestic violence has fallen in the US since the introduction of criminalization and mandatory arrest of IPV crimes. However, historical trends show that the overall crime rate has fallen to a greater extent. Goodmark provided several reasons why criminalization has likely been unsuccessful in deterring IPV.  Some studies emphasize that it is the accountability and monitoring of perpetrators (even after incarceration) that has been effective in deterring IPV crimes and not the punishment itself. In fact, there are vast costs of DV criminalization occurring to victims of domestic abuse, such as financial instability caused by unemployment of (in many cases) the primary breadwinner in a household. Also, criminalization has been shown to exacerbate other correlates of IPV such as aggressive and hostile tendencies of the perpetrator. Goodmark proposed alternatives to DV criminalization that avoid such costs and thereby, are potentially more effective in reducing domestic abuse. First, there are solutions rooted in economics such as cash-transfer programs, employment training, and micro-financing. These types of measures can help to reduce the economic penalties of seeking support and strengthen the victim’s financial independence. Also, more social solutions were suggested such as community organizing, restorative justice, and community accountability. Moreover, Goodmark underlined the fact that individuals with adverse childhood experiences, often involving violence, are significantly more likely to commit violent crimes such as IPV. Identifying and intervening at an early age to educate these individuals about intimate relationships has been shown to be effective in dealing with the problem.  In a nutshell, Goodmark stressed the importance of constructing a balanced policy approach that targets the origins of DV and argued that the time has come to reconsider punishing violence with more violence.


Problems related to IPV misreporting were a recurring subject of discussion at the workshop. A lot of the previous research on IPV relies on direct surveys asking women whether they were a victim of different instances of IPV. The main problem associated with such surveys relates to accuracy: social factors such as stigma, shame, and/or self-blame, as well as privacy concerns, are likely to influence respondents’ answers. A practice that has proven successful for sensitive questions is the use of an indirect method called list experiments, where the structure of the survey mitigates much of the above concerns on the respondent’s side (see, e.g., https://blogs.worldbank.org/ impactevaluation/list-experiments-sensitive-questions-methods-bleg).

Veronica Frisancho presented a study on the gap in reporting originating from direct questionnaires vs. list experiments based on experimental evidence from Peru. The experiment considers two groups of 500 women each. Women in the first group participate in a survey that uses direct questionnaires, whereas those in the second group answer a survey using indirect questionnaires. Based on the answers, the authors obtain an IPV prevalence rate for each group and define under-reporting as the difference in prevalence between them, under the assumption that the rate of under-reporting in the presence of indirect questionnaires is minor. Unexpectedly, yet encouraging, they find no evidence of misreporting in the direct-questions method. However, when looking closer at different education levels, they find that under-reporting is significantly more prevalent for highly educated women. In other words, less educated women are more truthful when answering questions about IPV. Frisancho emphasized that these types of patterns make it more difficult to identify the most vulnerable groups, implying that direct methods could increase the risk of mistargeted policies.

More generally, there are several reasons why respondents may be less truthful when answering questions related to IPV. On the one hand, individuals may be aware that they are victims of abuse, but perhaps are unwilling to confess due to stigma. On the other hand, it could be that individuals fail to identify themselves as victims of abuse at all, and do not consider their relationship unhealthy. Against this background, Nishith Prakash presented preliminary results of an ongoing study on behavioral barriers to the demand for DV-support services. The baseline results of the survey indicate belief gaps among women who scored high on levels of abuse: a significant majority of abuse victims rated their relationship as healthy. While 46.43% of respondents report some form of physical, emotional, or sexual violence, the portion of those with the prior belief that they are in an abusive relationship is only 1%. The study also finds that stress about Covid-19 correlates with higher levels of self-blame, abuse, and lower levels of understanding of what abusive behaviors are.

The covid-19 pandemic and its massive repercussions on determinants of DV such as mobility, economic insecurity, and social isolation have offered new possibilities for researchers to study the underlying causes of DV, while also making DV research ever more important. The next policy brief in this series will summarize the presentations which were specifically devoted to the consequences of the pandemic on DV. On behalf of FROGEE and SITE, we would like to thank the speakers for their contributions to the understanding of this topic, which will be indispensable both to the academic community and to policymakers in their efforts to design more effective policies for the future. We would also like to thank SIDA for generous financial support.


  • WHO, Department of Reproductive Health and Research, London School of Hygiene and Tropical Medicine, South African Medical Research Council. “Global and regional estimates of violence against women”. Reference No. 978 92 4 156462 5. 2013.

List of Participants

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Domestic Violence in the Time of Covid-19

20201012 Domestic Violence in the Time of Covid-19 Policy Brief image 01

 Since the outbreak of Covid-19 in the spring of 2020, media outlets around the world have reported increases in domestic violence. United Nations secretary-general António Guterres has even referred to it as a “shadow pandemic”. Besides news outlets, academic researchers have also taken an interest in the issue, which is crucial if we are to draw the right conclusions from the patterns we see in the statistics. Preliminary evidence shows that the incidence of intimate partner violence has also increased in Sweden, notwithstanding the absence of a strict lockdown. This is likely related to the socio-economic changes brought about by the pandemic.

A Shadow Pandemic?

In response to the Covid-19 pandemic, governments around the world introduced a variety of measures aimed to stave off the contagion, and billions of worried people adapted their behavior and lifestyle. But did the pandemic, and the changes brought by it, also lead to an increase in domestic violence?

Were we to simply look at the number of domestic violence offenses reported over time, we would not be able to answer this question. Historical trends and seasonal patterns in domestic violence would confound this observation, while the crisis might affect the reporting of crimes independently of their occurrence. More rigorous statistical analysis is needed for understanding not only the true situation with domestic violence under the pandemic, but also the reasons behind it. Investigating the driving factors is crucial for informing policy reactions already in the short run — is it a loss of income that generates violence, or could it simply be increased exposure? Do we need more unemployment benefits or shelters for victims?  ­­Moreover, the rather special conditions created by the pandemic can contribute to our general understanding of how domestic violence occurs in relation to other societal dynamics, unveil some of the causal mechanisms that are still open questions in the literature and help to fight this issue further, even after the pandemic is over.

Socio-economic Theories of Violence

Within social science research, studies that focus on the relationship between domestic violence and factors at a societal level can be divided into several different branches. A large corpus of theories interprets violence as a result of power imbalance within households. This perspective is associated with explanations such as bargaining power, exit options, and status, theoretical concepts that are often embodied and approximated by observable factors such as (relative) education, income or employment status. For example, Aizer (2010) provides results in line with the bargaining power hypothesis showing that a decrease in the gender wage gap in the US is associated with a decrease in domestic violence against women. Along the same lines, Anderberg et al. (2016) use UK data to show that an increase in unemployment among men reduces the incidence of intimate partner violence (IPV) while an increase in unemployment among women increases it. In contrast, a study from Spain documents the opposite relationship in provinces characterized by stronger traditional gender roles (Tur-Prats, 2019). It finds that a decrease in female relative to male unemployment causes an increase in violence, which is more in line with the “backlash explanation” — when a woman improves her economic position and independence, the man in the household feels that his identity as breadwinner is threatened and retaliates with violence as a result. Studies such as Iyer et al. (2012) and Miller and Segal (2018) highlight the importance of improving the position of women in society, which can be achieved, for example, through role models and female representation in critical positions. They associate the proportion of women among elected politicians and among the police, in India and the United States respectively, with a significant increase in reports of crimes against women and at the same time a significant decrease in the incidence of such crimes.

An alternative interpretation of domestic violence puts more emphasis on its emotional and irrational nature. In this case, particular events or negative emotional shocks, such as an unexpected negative result of an important football match (Card and Dahl, 2011), are believed to trigger violent reactions in the heat of the moment. The likelihood of such incidents is exacerbated by stress and emotional climate within a household, which in turn are influenced by economic conditions or financial uncertainty. For example, several studies from developing countries associate improvements in general economic conditions with a reduction in domestic violence (Hidrobo et al., 2016; Kim et al., 2007; Haushofer et al., 2019).

Finally, there is a common perception that domestic violence increases during holidays and weekends as families spend more time together and potential victims are more isolated from their social networks, in line with the so-called exposure model in criminology. So far, research on this hypothesis is limited and incomplete. However, it is precisely one of the areas where studies from the recent months may fill the knowledge gap: the fact that lockdowns and work from home  forced many families to spend more time together at home while retaining full wages, gives a unique opportunity to examine exposure in isolation from other economic factors.

The opposite of exposure is known as (self-) incapacitation theory: no aggression will occur while a (potentially violent) partner is occupied with something else, whether imposed or self-chosen. Several studies focusing on this hypothesis have documented that the incidence of violent crimes declines, on the street or in the home environment, when potential perpetrators are in school (Jacob and Lefgren, 2003), in prison (Levitt, 1996), at the cinema (Dahl and DellaVigna, 2009) and when they have access to a legal prostitution market (Cunningham and Shah, 2018; Ciacci and Sviatschi, 2018; Berlin et al., 2019). During a lockdown, the availability of such activities is restricted, both to violent people as well as potential victims.

Research on Domestic Violence During Covid-19

The list of studies analyzing data from the past few months is growing by the day. Although full consensus is yet to be reached, the results that have emerged point towards a few patterns: spikes in domestic violence can be credibly connected to strict limitations of movement, at least in some contexts (India, Ravindran and Shah, 2020; Peru, Agüero, 2020; 15 large US cities, Leslie and Wilson, 2020);  unemployment could be an important mechanism (Bhalotra et al., 2020; in Canada, Beland et al., 2020 find no impact of unemployment or work arrangements per se, but do associate spikes in violence to financial difficulties); alcohol does not seem to amplify domestic violence during the pandemic, at least in some context (Silverio-Murillo and Balmori de la Miyar do not find any effect of the prohibition to sell alcohol in parts of Mexico City); and by and large barriers to reporting might be a serious issue (Spencer et al, 2020).

A selection of studies on domestic violence during the Covid-19 crisis, many of which are as yet unpublished, were presented at the recent FROGEE Workshop “Economic Perspectives on Domestic Violence”. Two FREE Policy Briefs summarizing the event are forthcoming.

Domestic Violence in Sweden During Covid-19

Studying Sweden against this background can be particularly interesting for at least two reasons. Sweden regularly occupies the top positions in international rankings of gender equality in many dimensions and is seen as having advanced progressive norms and attitudes in this area. As pointed out by the literature on the economic determinants of domestic violence, underlying norms and attitudes can play a significant role in shaping the impact of other factors, such as unemployment (Tur-Prats, 2019). Therefore, the Swedish case can offer a valuable comparison to studies focusing on countries that have different attitudes and norms.

According to estimates by the National Council for Crime Prevention (BRÅ), at least 7% of the Swedish population is exposed yearly to domestic violence, both men and women in roughly equal parts. However, women are much more likely to report recurring violence and to end up hospitalized.

When it comes to the particular situation of the Covid-19 crisis, Sweden is also close to unique in its contagion-management strategy. Swedish policy relied much more than elsewhere on voluntary participation and individual responsibility rather than coercion. Certainly, working from home when possible was encouraged, the use of public transport discouraged, and indoor events with more than 500, and thereafter 50 participants were forbidden, which included many sports and cultural events. In fact, the Google mobility index, based on location data from Google Account users, shows patterns of clear deviation from the baseline since week 11 of 2020, when the authorities declared a very high risk of community spread.

Figure 1. Mobility patterns in Sweden during Covid-19

Source: Author’s aggregation of Google mobility index. The lines show the deviation from baseline, in percentual terms, of total user presence in different urban areas by category.

The plots in Figure 1 show that the presence of Google Account users was about 10% higher in residential areas (the pink line) and much lower in workplaces, despite some variation over the period: the initial decline was roughly half as large as the impact of summer vacation, as shown by the blue line. Also, visits to retail centers and grocery stores, recreation places (such as restaurants, cinemas, and theaters), and transit stations decreased, especially during the beginning of the period. Mobility in parks and green areas, shown separately, follow to a larger extent a seasonal pattern.

Nevertheless, the general population was never forbidden or even discouraged from leaving their homes, which clearly makes a stark difference for many of the mechanisms that, based on the literature, we think could play a role in explaining domestic violence.

According to BRÅ, during the first half of 2020, there was a 1% increase in total reported crime compared to the same period of the previous year. However, there is wide variation among the crime categories: 9% more violent assaults against women were reported, and 4% more against men, but 6% fewer rapes of women and 9% fewer rapes of men. As discussed above, it is not straightforward to draw conclusions from simple comparisons over time. Preliminary analysis utilizing the variation in mobility patterns over weeks and municipalities reveals that a 10% increase in residential mobility is associated with a (lower bound) increase in reported non-battery crimes against women committed by an intimate partner by 0.015 crimes per 10,000 individuals (a sixth of the mean). The corresponding figure for a 10% reduction in mobility in retail and recreation areas and transit mobility is around 0.0025 additional crimes (3% of the mean) (see Figure 2). Crime categories include attempted or planned homicides; sexual molestations, sexual assaults, and rapes; violations of integrity and privacy (including limitation of freedom, coercion, threats, persecutions; battery crimes are not included for the time being because of a coding mistake in the police system pertaining this particular category).

Figure 2. Mobility patterns and IPV in Sweden during Covid-19 – non-battery crimes

Source: Author’s analysis. Crime data provided by the police, mobility index provided by Google.

We consider this a lower bound because of the voluntary nature of the Swedish ”lockdown” – if people have the freedom to choose, then it is reasonable to expect that individuals more exposed to the risk of domestic violence would decide to be less at home, which would reduce the strength of the relationship observed. In the opposite direction, we might be worried that when more people are at home, more crimes are reported by a third party, such as neighbors, and thus not implying that more crimes are being committed. However, we differentially see more reported crimes with a female victim than with a male victim, which is not necessarily easy for a third party to distinguish by the sounds. Therefore, it seems likely that, based on the changes in mobility patterns, IPV against women has increased in Sweden during the Covid-19 crisis. Other consequences of the crisis that might also play an important role in shaping IPV and domestic violence, including the huge increase in unemployment and changes in alcohol sales, remain to be investigated.


In conclusion, research from the past months finds some limited support for hypotheses originating from previous literature on the relationship between different socio-economic factors and domestic violence. When these factors were affected by the pandemic and the associated economic crisis, domestic violence responded as well, to a varying extent depending on the context. This can be seen as an indirect and hidden cost of the pandemic.

Preliminary evidence indicates a similar case for Sweden, notwithstanding the absence of a strict lockdown. This implies that a significant part of the changes in behavior, which in turn can be expected to affect domestic violence, have occurred as a response to the pandemic itself and not necessarily as a result of policy measures.

While the shock of the pandemic will help us to better understand some of the underlying mechanisms behind the phenomenon of domestic violence, many questions are still open, and it is important to look beyond the pandemic. Domestic violence existed before Covid-19 and will, unfortunately, remain part of our societies when the pandemic is over. Investigating and understanding its determinants is important in order to formulate proper policies to combat it during and after the crisis.


Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Removing Obstacles to Gender Equality and Women’s Economic Empowerment – What Can Policy Makers Learn from Global Research on Gender Economics?

20200224 Removing Obstacles to Gender Equality FREE Network Policy Brief Image 01

On November 15-16, 2019, the FREE Network and the ISET Policy Institute organized and conducted an international gender economics conference in Tbilisi, Georgia. The conference was organized as part of the FROGEE initiative – the Forum for Research on Gender Economics – supported by the Swedish International Development Agency (SIDA) and coordinated by the Stockholm Institute of Transition Economics (SITE). The conference brought together researchers, policymakers, and the broader development community to discuss obstacles to gender equality and women’s economic empowerment, as well as policies to remove existing constraints, with a particular focus on Eastern Europe and Emerging Economies. This policy brief provides an overview of the main takeaways from the presentations, with a special focus on policy-relevant lessons.


In November 2019, Tbilisi welcomed its first international academic conference on gender economics, “Removing Obstacles to Gender Equality and Women’s Economic Empowerment”. The conference focused on the state of economic policy and gender issues around the world and more specifically in the ECA (Europe and Central Asia) region. The opening remarks were offered by two prominent keynote speakers – Dr. Caren Grown, Senior Director for Gender at the World Bank Group, Washington D.C, and Dr. Shahra Razavi, Chief of Research and Data at UN Women HQ in New York. The key addresses offered a global perspective on the current state of gender equality and progress made during the last 20 years. The global overview was followed by a policy panel discussion featuring prominent members of the policy-making community in Georgia. The panel participants reflected on how various policies have impacted gender (in)equality in the South Caucasus and in Georgia in particular. Later in the day, plenary presentations offered a preview of the South Caucasus Gender Equality Index, which is being developed by the ISET Policy Institute, and new research in gender economics done by academics in Georgia, Armenia, Belarus and Sweden.

The second day of the conference showcased research conducted by academics from over 15 countries covering 4 continents. It presented a range of diverse topics in gender economics, including, most prominently the links between childcare policies and labor supply decisions of women, female labor force participation (LFP) and happiness, evolving family structure and gender-selection preferences, the impact of economic, financial and public policies on women’s empowerment, the male-female earnings gap and gender aspects of international trade.

Below, we summarize the results and policy lessons that emerge from the body of work presented at the conference.

Gender Equality Progress in the ECA Region and Worldwide: Key Takeaways

First, as recent global data shows, the progress in women’s access to resources, in particular their access to the labor market, has on average stalled worldwide in the last 20 years. The labor market participation rate of women in 2018 stood at 63% globally, which is largely the same as in 1998, with some notable progress observed only in Latin America and the Caribbean (increase from 57% to 67% between 1998 and 2018), Australia and New Zealand (70 to 79%), as well as Northern Africa and West Asia (29 to 33%). The labor force participation gap between men and women is most pronounced for women who are married or in unions (44% gap, as opposed to 20% for single/never married or 17.9% for divorced/separated women).

Second, the ratio of time spent on unpaid care work by females was about 3-4 times that of males in most countries in the world, with some notable outliers: 11 times in Pakistan, 10 times in Cambodia and 9 times in Egypt. Only in Australia and New Zealand, the ratio of female to male time spent on unpaid work was slightly below 2. Thus, around the world, family responsibilities and unpaid work at home have clearly disproportionately burdened women, potentially preventing them from having an independent source of labor income, and generally weakening their financial position and bargaining power within the family unit. The recent UN Women report on Families in the Changing World (2019) argues for implementing a comprehensive package of family and women-friendly policy measures, which would include, among others, universal childhood education and care, universal healthcare coverage, long-term care for the elderly, etc. Such a comprehensive package would cost between 2-4% of GDP for most countries covered by the study. At the same time, the report argues that it would generate jobs, new investments and be a sizeable source of new tax revenue to the economies. Hence, the costs of such a program would be partially offset by the economic and tax benefits of formalizing the informal care economy. The study also details the ways in which countries could mobilize resources to pay for such packages, including improving tax collection, eliminating illicit financial flows, and leveraging aid and transfers.

For the South Caucasus in particular, the state of gender equality has not systematically been tracked until now. While there exists a number of thematic studies, surveys and narratives, as well as a more general Gender Inequality Index (GII) compiled by UNDP for all countries, a deeper systematic approach has recently been pioneered by the ISET Policy Institute, which started the ambitious project of developing a Gender Equality Index for the South Caucasus and, going forward, for the broader region of transition economies. The methodology behind the index is similar to the one adopted by the European Institute for Gender Equality, which tracks the Gender Equality Index for 28 European countries across a number of dimensions. Obviously, issues of data availability make it more challenging to build such an index in the context of transition economies. Thus, ISET-PI is working to construct some of the measures for the transition economies, using country-level data and household-level databases.

Childcare Policies and Labor Supply

One of the key messages emerging from the academic research in the area of childcare policies and labor supply was that gender-focused social policies need to be crafted carefully, with a focus on the binding constraints of the specific country context. A paper by Vardan Baghdasaryan and Gayane Barseghyan looked at how child-care service availability (affordability) affected the female labor force participation on the intensive and extensive margins in Armenia. The stage for a natural experiment in economic policy was set at the time when the Municipality of Yerevan unexpectedly decided to abolish childcare services fees (roughly 15% of average wage). The researchers hypothesized that such an intervention would have resulted in increased female LFP, as was the case in other (mostly developed) regions and countries around the world (e.g. Quebec in Canada). In the context of Armenia, however, the authors observe that there was no significant effect on female LFP rate on the extensive margin, meaning there was no evidence of inactive women entering the labor force. One possible explanation is that in the context of a developing country such as Armenia, the limiting factor to female participation in the labor force is the lack of market demand for the skills profile of non-active mothers. In such an environment, as the authors conclude, the monetary incentives do not suffice to lift the binding constraint on female LFP.

Yolanda Pena-Boquete presented a study on the case of Australia which analyzed how the labor hours and LFP of both women and men in the family are affected when either the mother’s or the father’s wages increase or when the price of childcare changes. The study finds that the mothers’ working hours respond positively and much stronger to a change in hourly wage than the fathers’. The policy implication is that an increase in mothers’ hourly wage would potentially result in a significant increase in their working hours and labor force participation. The wage effect on women’s working hours and LFP is much more pronounced even compared to the scenario when childcare prices decline.

Overall, the studies in this area demonstrated the need for a careful, multi-faceted approach in designing effective and cost-efficient labor market policies aimed at increasing labor force participation by married women with children.

Labor Force Participation and Happiness: Evidence from the South Caucasus

The paper by Norberto Pignatti and Karine Torosyan looked at the differences in the reported happiness levels between women of different labor market status in the three South Caucasus countries. The intriguing finding of the study is that while in Georgia, there is no difference in the reported happiness level between working women and housewives, in Armenia and Azerbaijan, working women with similar characteristics are much less likely to report being “very happy” than housewives. The interesting finding is that the overall results for Georgia also apply to the Armenian and Azerbaijani minority women in the country, implying that “cultural factors” may play a minor role in the reported differences between countries.

Family Structure and Gender-Selection Preferences

Gender-biased sex selection (GBSS) has been on the forefront of gender policy issues in the South Caucasus, as Armenia, Azerbaijan and, until recently, Georgia struggled with skewed sex ratios at birth (SRB). Understanding the driving forces behind GBSS, and in particular son-preference as a socio-economic phenomenon, is especially important. One of the recent studies on the issue was presented by Davit Keshelava of the ISET Policy Institute. The study “Social Economic Policy Analysis with Regard to Son Preference and Gender-biased Sex Selection” looked at the factors underlying GBSS rise and fall in Georgia over the last 15 years. The study also gleaned facts about the changing attitudes towards GBSS and son-preferences in different regions of Georgia. One of the study’s main findings is that the fall in the sex ratio at birth has been statistically significantly correlated with real income growth in the regions, reduction in poverty, and female employment. Among other factors significantly affecting the reduction in sex ratio at birth, was, surprisingly, the level of male education, while female education was statistically insignificant. The study documented a persisting son preference in Georgia, but also high awareness and strong negative attitudes towards gender biased sex selection in those regions that showed the sharpest improvement in sex ratio at birth over time.

Looking at the issue of gender preferences in the context of transition economies in Europe, Izabela Wowczko presented joint work with Michał Myck and Monika Oczkowska which investigated how preferences for the gender composition of children in the family might have changed in Central and Eastern European (CEE) countries after the fall of communism. The results showed that gender-neutrality was observed in almost all CEE countries before the transition. After the transition of the 1990s, many of the same forces which operated in the South Caucasus have affected the countries of Central and Eastern Europe – namely, decline in incomes, decimated traditional social safety nets and better access to ultrasound and family planning technologies. However, in the post-transition CEE countries, the authors observe a clear preference for a mix (boy/girl) or possibly boys at parity three (i.e. having two boys or a boy and a girl in the family reduced the likelihood of having a third child significantly, as opposed to having two girls). It was also observed that in most CEE countries (except Romania), there was an increased likelihood of having a second child if the first child is a boy – thus demonstrating a girl preference at parity two.

Policy Impact on Women’s Empowerment

A study from India by Mridula Goel and Nidhi Ravishankar looked at the impact of policy interventions on the long-term indicators of women empowerment. It shows that public policies were responsible for improving the so-called “power enablers”, such as literacy rates, financial access, property rights, political voice, etc. However, there is some evidence that not all traditional power enablers, e.g. having a bank account or working for money, are correlated with higher indicators of empowerment, measured by a woman’s autonomy in decision-making within the family. For example, working for money (receiving cash compensation) or having a bank account was found to be negatively correlated with a woman’s ability to decide how her own money is spent – possibly pointing to the existence of prejudice or negative attitudes within the household in such cases.

Another interesting study on this topic by Maria Perrotta Berlin, Evelina Bonnier and Anders Olofsgård looked at whether foreign aid projects foster female empowerment in the surrounding community using data from Malawi. It finds support for a small positive impact of aid on men’s and women’s attitudes related to domestic violence and sexual rights. There is, however, little systematic difference in the impact of gender-targeted aid versus general aid – with exceptions being the impacts on women’s experience of violence and women’s participation in decision-making.

Male-Female Earnings Gap and Gender Aspects of International Trade

The male-female earnings gap is a recurring topic in gender economics. Whether the gap is driven by differences in education and skills of men and women, labor market discrimination, choices of working hours, the “glass ceiling” or “sticky floor” phenomena, the gap is evident and persistent in both developed and developing countries. One of the papers presented by Dagmara Nikulin looked at the impact of trade liberalization on the gender wage gap in Europe. Generally, the economic literature does not provide conclusive evidence in this regard, and the link remains ambiguous. The paper, examining evidence from Europe, finds in particular that participation in global value chains (GVC), which the authors measure by foreign value added in exports, is correlated with reduced wages overall, but the negative effect on wage is lower for men than for women.

Echoing the results of the previous study, the paper by Marie-France Paquet and Georgina Wainwright-Kemdirim, “Since the effects of trade liberalization are not gender neutral, how can we improve its gender outcome? – Crafting Canada’s Gender Responsive Trade Policy” focuses on the problem of identifying and addressing potentially negative impacts of trade on female jobs. The study details a diagnostic modelling approach, which is to use CGE modeling combined with sectoral employment data (a labour module within CGE). The proposed model uses an overlapping generation framework and includes an occupational matrix to allow movements between occupations. This approach allows for specific potential impacts of generic FTAs by gender, age group and occupation.


To sum up, the first international academic conference on gender economics issues in Tbilisi highlighted the diversity and complexity of gender issues around the world and in the South Caucasus region in particular. It also became a powerful catalyst for new research and collaboration ideas among participating institutions and individual researchers. Finally, it demonstrated how policy-oriented research can help inform the policy-making community about the areas where intervention is most needed, design the most effective policies, and calculate the associated costs and benefits of interventions.

References to Selected Presentations

  1. Shahra Razavi “Policies for Gender Equality in an Unequal World: Challenges and Opportunities”, keynote presentation.
  2. Vardan Baghdasaryan and Gayane Barseghyan “Child Care Policy, Maternal Labor Supply and Household Welfare: Evidence From a Natural Experiment”.
  3. Michal Myck and Kajetan Trzcinski “From Partial to Full Universality: the Family 500+ Programme in Poland and its Labour Supply Implications”.
  4. Karen Mumford, Antonia Parera-Nicolau, Yolanda Pena-Boquete “Labour Supply and Childcare: Allowing Both Parents to Choose”.
  5. Norberto Pignatti, Karine Torosyan “Employment vs. Homestay and Happiness of Women in the South Caucasus”.
  6. Davit Keshelava et al. ISET Policy Institute Report “Social Economic Policy Analysis with Regard to Son Preference and Gender-biased Sex Selection”.
  7. Izabela Wowczko, Michał Myck and Monika Oczkowska “Gender Preferences in Central and Eastern Europe as Reflected in Family Structure”.
  8. Mridula Goel, Nidhi Ravishankar “Has Public Policy Succeeded in Enhancing Women Autonomy and Empowerment in India Over the Last Decade?”.
  9. Maria Perrotta Berlin, Evelina Bonnier and Anders Olofsgård “The Donor Footprint and Female Empowerment”.
  10. Dagmara Nikulin & Joanna Wolszczak-Derlacz “Gender Wage Gap and the International Trade Involvement. Evidence for European workers”.
  11. Marie-France Paquet, Georgina Wainwright-Kemdirim, “Since the Effects of Trade Liberalization are not Gender Neutral, How can we Improve its Gender Outcome? – Crafting Canada’s Gender Responsive Trade Policy”.

From Partial to Full Universality: The Family 500+ Programme in Poland and Its Labour Supply Implications

20191216 From Partial to Full Universality FREE Network Policy Brief Image 01

The implementation of the ‘Family 500+’ programme in April 2016 represented a significant shift in public support for families with children in Poland. The programme guaranteed 500 PLN/month (approx. 120 euros) for each second and subsequent child in the family and the same amount for the first child in families with incomes below a specified threshold. As of July 2019, the benefit has been made fully universal for all children aged 0-17, an extension which nearly doubled its total cost and benefited primarily middle and higher income households. We examine the labour market implications of both the initial design and its recent fully universal version. Using the discrete choice labour supply model, we show that the initial Family 500+ benefits generated strong labour supply disincentives and were expected to result in the withdrawal of between 160-200 thousand women from the labour market. The recent removal of the means test is likely to nullify this negative effect, leading to an approximately neutral impact on labour supply. We argue that when spending over 4% of GDP on families with children, it should be possible to design a more comprehensive system of support, which would be more effective in reaching the joint objectives of low child poverty and high female employment combined with higher fertility rates.


Following the 2015 parliamentary elections in Poland the ruling Law and Justice Party was quick to fulfill their campaign promise of implementing a generous quasi-universal family support programme. In April 2016, all families began receiving PLN 500 (approx. 120 euros) per month for each second and subsequent child, while households that passed an income means test were granted the same amount for their first or only child.  At a cost of nearly PLN 22 billion (5.2 billion euros, approx. 1.1% of GDP) per year, the Family 500+ benefit became the flagship reform of the Law and Justice government’s first term.

With new elections approaching in October this year, the government announced a significant expansion of the programme in May, which made it fully universal. The extended programme is nearly twice as expensive with an additional cost of PLN 18.3 billion (4.3 billion euros) per year, valuing the whole package at over 2% of GDP. This takes the total value of financial support for families with children, including family benefits and child-related tax breaks, to 4% of GDP and it means that as far as family support is concerned, the ruling party has brought Poland from one of the lowest-spending countries in the EU to one of the highest over the course of 4 years.

The initial design of the benefit had a significant impact on childhood poverty in Poland, with an absolute and relative decrease from 9.0 to 4.7 percent and 20.6 to 15.3 percent respectively between 2015 and 2017 (GUS, 2017). While a more targeted design could have made a far greater impact, these changes still reflect a significant improvement in the material situation of families with children. The policy may have also had a modest upward effect on fertility rates in the first years following its implementation, although this is difficult to assess given the parallel roll out of several other fertility-oriented policies and other changes which could have played a role in family decisions. Simultaneously, as argued in the ex-ante analysis by Myck (2016) and ex-post analysis by Magda et al. (2018), these positive outcomes came at the cost of reduced female labour market participation. This reduction primarily affected women with both lower levels of education and living outside of large urban areas (Myck and Trzciński, 2019).

The Family 500+ Reform: Design and Distributional Implications

The initial Family 500+ programme directed funds to 2.7 million families in addition to any already existing financial support and has been excluded from other means-tested support instruments.  Since families that had a net income of less than PLN 800 per month per person could receive the benefit for the first or only child, the policy had a distinct redistributive element and meant that  the bottom half of the income distribution received nearly 60% of the funds. However, the design was characterised by clear labour market disincentive effects, which were particularly strong for second earners and single parents.

In a one-child household (53.3 percent of families with children, GUS, 2016) with the first earner bringing in an income equivalent to 125% of the national minimum wage, the second earner needed only to earn PLN 940 per month in order for the family to cross the means test threshold and stop receiving the Family 500+ benefits. The benefit design is presented in Figure 1 in the form of budget constraints for the first earner (Case A) and the second earner (Case B) in a couple with one child. In the latter case the first earner is assumed to receive earnings equivalent to 125% of the minimum wage. The disincentive effects of the means test are clear in both cases and we can see that for the second earner, the benefit withdrawal comes at a very low income level – far below the national minimum wage of PLN 2100 per month. The “point withdrawal” of the benefit implied that it was enough for the family to marginally exceed the means test threshold for it to completely lose eligibility for the Family 500+ support for the first child.

The expansion of the Family 500+ programme, which came into effect in July 2019, eliminated the means-tested threshold thus making the policy fully universal. It came, however, at the cost of the redistributive character of the programme. Over 32% of the additional expenditure resulting from the universal character of the policy has been passed on to the top quintile of the income distribution and in its new version, the bottom half of households only receive 45 percent of all spending. The expansion of the programme is thus unlikely to further reduce child poverty significantly and – since its beneficiaries are mainly families with middle and high incomes – it is not expected to bring noticeable changes in fertility levels.

Figure 1: Family budget constraints for the first and second earner

Source: Authors’ calculations using the SIMPL microsimulation model.

Partial and Full Universality of the Family 500+ Programme and the Implications on Female Labour Supply

With the use of modelling tools to simulate the labour market response to changes in financial incentives to work, we have updated the initial simulations of Myck (2016) using the latest pre-reform data and examined the simulated labour supply decisions to the expanded fully universal programme, as if it were implemented instead of the initial version of the benefit. The analysis was conducted with data from the 2015 Polish Household Budget Survey, a detailed incomes and expenditure survey conducted annually by the Polish Central Statistical Office.

Table 1: Effects of the initial and the expanded Family 500+ programme on female labour supply

Results of the simulations are presented in Table 1. Simulations were conducted separately for single women, and under two scenarios for women in couples assuming that both partners adjust their behaviour (Model A) and that the labour market position of the male partner is unchanged (Model B).  The simulated labour supply response to the initial reform confirms the magnitude of earlier results and suggests an equilibrium effect of 160-200 thousand women leaving the labour force. This is also consistent with results presented by Magda et al. (2018), who found that female labour market participation decreased by approx. 100 thousand women after the policy had been in place for one year.

However, as we can see in the right-hand part of Table 1, the response to a fully universal design – modelled as if it was introduced in 2016 instead of the means-tested version – is essentially neutral. For single mothers the reduction is only about 3000, while for women in couples, the model suggests a small positive reaction under the Model A specification and a small negative one under Model B. In total, the universal design of Family 500+ benefits can be described as labour supply neutral. Since the reaction has been modelled on pre-reform data, and because some women have already withdrawn from the labour market after the introduction of the initial benefit design in 2016, the remaining uncertainty is whether the new set of incentives will motivate these mothers sufficiently to return to work.


The introduction and subsequent expansion, of the Family 500+ programme has substantially increased financial resources of families with children in Poland. The policy rollout of the initial, partially universal programme has seen substantial changes in the level of child poverty in Poland and may have contributed to a modest increase in fertility in the initial years following the introduction of the reform. The means-tested design of the benefit, however, incentivised a significant number of women to leave the labour market. One year after the introduction of the policy approximately 100,000 women were estimated to have left the labour market (Magda et al. 2018), while the equilibrium effect of the policy suggested long-run implications of over 200,000 (Myck, 2016). The updated simulation results using the latest available data suggest slightly lower, though still substantial equilibrium implications of the initial partially universal design of the Family 500+ programme in the range of between 160,000-200,000. However, as we show in our latest analysis, these labour market consequences could be reversed after the expansion of the programme to a fully universal set-up. The simulated effects of the universal design of the programme, which has been in place in Poland since July 2019, modelled as if it was implemented instead of the initial means-tested version, are broadly neutral for female labour supply. The only question is how likely the mothers who left employment in response to the initial policy will return to work given the new set of financial incentives. Considering these positive implications of the fully universal programme, one has to bear in mind that the extended programme, which will cost over PLN 40 bn per year (approx. 2% of GDP), is unlikely to contribute to the other key objectives set by the government, namely reducing child poverty and increasing fertility. Including the Family 500+ programme, the Polish government currently spends about 4% of GDP on direct financial support for families with children. Given the design of the policies which make up this family package, it seems that the joint objectives of higher fertility, reduced poverty and higher female employment could be achieved more effectively under a reformed structure of support that would be better targeted at poorer households, include specific employment incentives, and incorporate support for childcare, early education and long-term care.


This brief summarizes the results presented in Myck and Trzciński (2019). The authors gratefully acknowledge the support of the Swedish International Development Cooperation Agency, Sida, through the FROGEE project. For the full list of acknowledgements see Myck and Trzciński (2019).


  • Goraus, K. and G. Inchauste (2016), “The Distributional Impact of Taxes and Transfers in Poland”, Policy Research Working Paper 7787, World Bank.
  • GUS (2016), “Działania Prorodzinne w Latach 2010-2015, Główny Urząd Statystyczny Polish Central Statistical Office, Warsaw.
  • GUS (2017), “Zasięg ubóstwa ekonomicznego w Polsce w 2017r.”, Główny Urząd Statystyczny – Polish Central Statistical Office, Warsaw.
  • Magda, I., A. Kiełczewska, and N. Brandt (2018), “The Effects of Large Universal Child Benefits on Female Labour Supply”, IZA Discussion Paper No. 11652, IZA-Bonn.
  • Myck, M. (2016), “Estimating Labour Supply Response to the Introduction of the Family 500+ Programme”, Working Paper 1/2016, CenEA. Jacobson, L., LaLonde, R. and Sullivan, D. (1993). “Earnings losses of displaced workers”, American Economic Review, 83, pp. 685–709.
  • Myck, M. and Trzciński, K. (2019) “From Partial to Full Universality: The Family 500+ Programme in Poland and its Labor Supply Implications”, Ifo DICE report 3 / 2019.

Gender Gaps in Wages and Wealth: Evidence from Estonia

20191118 Gender Gaps in Wages and Wealth FREE Network Policy Brief Image 01

This policy brief introduces two related papers examining two types of gender gaps in Estonia. First, it presents the work of Vahter and Masso (2019), who study the wage gender gaps in foreign-owned firms and compare this gap with the situation in domestic ones. Then it summarizes a paper of Meriküll, Kukk, and Rõõm (2019), who focus on the wealth gender gaps and highlight the role of entrepreneurship in this gap.

Gender inequality is not only a moral issue. An extensive literature has highlighted the cost of gender inequality in terms of economic (in)efficiency. Most of the academic work has, however, focused on either the US and Western Europe or developing countries. Research focusing on systematic gender disparities in Eastern Europe is rather scarce. Yet, there is much to be learned from this region. The purpose of the FROGEE (Forum for Research on Gender in Eastern Europe) project is to study several issues related to gender inequality in former socialist countries.

This policy brief summarizes two papers presented at the 2nd Baltic Economic Conference at the Stockholm School of Economics in Riga, on June 10-11, where a special session on gender economics was held with the support of the FROGEE project. The event, organized by the Baltic Economic Association (see balticecon.org), gathered more than 85 researchers from the Baltics and all over the world. These two papers focus on Estonia, one of the most successful economies among the transition countries, where however the gender wage gap is among the largest in the European Union.

Firm ownership and gender wage gap

An important source of wage inequality originates in firm-specific pay schemes (see for instance Card et al. 2016). Understanding the characteristics of firms associated with a gender pay gap is thus a necessary step to design relevant policy responses. In a paper entitled “The contribution of multinationals to wage inequality: foreign ownership and the gender pay gap”, Jaan Masso and Priit Vahter, both at the University of Tartu, compare the situation in foreign-owned firms with domestic ones. The fact that foreign-owned firms provide on average higher wages to their employees is well documented. However, the question of whether this premium differs between men and women remains largely overlooked.

A potential channel linking firm ownership and gender wage gap is the transfer of management practices from the home country of the investor to the affiliate. The great majority of FDI in Estonia originates from Finland and Sweden, two countries that regularly top international rankings on gender equality and that have set the fight against gender inequality as a top priority. Observing a lower level of gender wage gap in firms owned by Swedish and Finnish capital would suggest the existence of such a mechanism, even if there is evidence that Scandinavian countries do not stand out in a positive way when it comes to women in the top of the distribution (see for instance Boschini et al., 2018, and Bobilev et al., 2019).

On the other hand, Goldin (2014) has shown that a large part of the gender wage gap in the US can be explained by differences in job “commitment”: firms disproportionately reward workers willing to be available 24/7, more flexible regarding business trips, spending longer hours in the office, etc. Such workers happen to be more often men than women. Multinational firms may require such commitment and flexibility to a larger extent than domestic firms, due for instance to their higher exposure to international competition. This would imply a larger gender pay gap in foreign-owned firms compared to local firms.

To investigate this issue, Masso and Vahter (2019) rely on Estonian administrative data, providing information on the whole universe of workers and of firms in the country between 2006 and 2014. This matched employer-employee dataset allows to track the wage of individuals over the years, but also to compare wages both across and within firms. It thus becomes possible to estimate the gender wage gap at the firm level (controlling for relevant individual-level factors affecting wages, such as age and experience), and then to check whether this measure systematically differs between domestic and foreign-owned firms.

However, simply comparing the gender pay gap between these two types of firms could lead to spurious conclusions. Foreign-owned firms have on average different characteristics than domestic ones: they do not operate in the same sectors, they do not have the same size nor the same productivity. To overcome this issue, the authors rely on a matching method: for each foreign-owned firms, they match a domestic firm with similar (observable) characteristics.

They find that in domestic firms, women are on average paid 19% less than men, even after accounting for many other factors associated with wage. In foreign-owned companies, both men and women are better paid. However, both genders do not benefit from the same premium: men are paid roughly 15% more in foreign-owned firms, whereas the premium for women is only 5.4%. This difference implies an even larger gender wage gap in multinational firms. To illustrate the economic significance of these results, for a man and a woman earning a monthly wage of 1146 euros (the average gross wage in Estonia in 2016), the premium for switching from a domestic to a foreign-owned firm is respectively 171 and 62 euros. Further, they provide some evidence that lower “commitment” is associated with a stronger wage penalty in foreign-owned firms. All in all, these results suggest that there is not necessarily a relationship between a multinational wage policy (especially in its gender wage-gap dimension) and the gender norms prevailing in its country of incorporation.

Gender and wealth gap

The vast majority of academic papers studying gender inequality focuses on the wage gap. But gender inequality can affect other types of economic outcomes, such as labor force participation, unemployment duration, or wealth. The latter is of particular interest since wealth can greatly contribute to empowerment. Merike Kukk, Jaanika Meriküll and Tairi Rõõm, all at the Bank of Estonia, extend the literature with a paper entitled “What explains the Gender Gap in Wealth? Evidence from Administrative Data”. This paper is one of the first to study the gender wealth gap in a post-transition country. The literature on the gender wealth gap is rather scarce because of a lack of suitable data: wealth measures are often computed at the household level, while individual-level data is necessary for such a study.

The main aim of this paper is to depict a precise portrait of this phenomenon in Estonia. In particular, the authors do not simply estimate the overall wealth gap but investigate the magnitude of the gap across the wealth distribution. In other words, is there a difference between the poorest men and the poorest women? Or on the other side of the distribution, are the richest men more wealthy than the richest women?

For this purpose, Kukk, Meriküll and Rõõm combine administrative individual-level data on wealth with survey results. The administrative data are generally considered of much better quality than the other, but they do not provide a lot of additional information on individuals. On the other hand, survey data provide a wealth of information about individual characteristics. Merging allows getting the best of both worlds. Regarding the methodology, the authors use unconditional quantile regression to track gender differences at different deciles of the wealth distribution. They further decompose this “raw” gender gap into two components: the “explained” part, i.e., the part of the gap resulting in differences in characteristics between men and women (demographics, education, etc.), and the “unexplained” part.

This study estimates the raw, unconditional gender wealth gap in Estonia to be 45%, which is of similar magnitude as in Germany. Interestingly, this difference is essentially driven by differences in the top of the distribution: there is a large gap between the richest men and the richest women. This “raw” difference is however explained by a single variable: self-employment, as men are much more likely to have business assets than women. Once controlling for the entrepreneurship status, the wealth difference between the richest Estonians becomes insignificant. This suggests the need to support policies encouraging female entrepreneurship and to remove barriers particularly affecting women. For instance, the literature has previously pointed out that women have less access to external sources of capital than men (e.g., Aidis et al., 2007). Such distortions can ultimately result in a wealth gap at the top of the distribution, as documented by this paper.

In addition, the literature has proposed several mechanisms that could result in gender-specific patterns of wealth accumulation. The simplest channel is through the wage gap, as it can be seen as the accumulation of the wage gap over time (e.g. Blau and Kahn, 2000). The authors thus compare the gender gaps in wealth and income. They uncover a strong wage gap, with men earning significantly more than women starting at the 6th decile: the higher we go in the income distribution, the larger the wage gap. How to reconcile this finding with the absence of a wealth gap conditional on entrepreneurship status? A possible explanation suggested by the authors is that women simply accumulate wealth better than men do.


These two papers illustrate two different mechanisms explaining gender-specific economic outcomes. The larger wage gap observed in multinational companies can be explained by a stronger commitment penalty for women, mostly because of childcare. This asks for two potential policy interventions. First, the development of childcare could facilitate the reduction in the “commitment gap” that disrupts women’s careers. Second, institutions could support a more flexible repartition of childcare responsibilities. Note however that Estonia already has the longest duration of leave at full pay (85 weeks), and that this leave can be freely split between parents. As for the wealth gap at the top of the wealth distribution, it can to a large extent be explained by the entrepreneurship status. This difference could partly be explained by differences in preferences and risk-aversion, which would require long-run policies to be mitigated. But in the short run, there is room for specific policies supporting female entrepreneurship and removing barriers particularly affecting women, such as a tighter credit constraint.


  • Aidis, R., Welter, F., Smallbone, D., & Isakova, N. (2007). Female entrepreneurship in transition economies: the case of Lithuania and Ukraine. Feminist Economics13(2), 157-183.
  • Blau, F. D., & Kahn, L. M. (2000). Gender differences in pay.  Journal of Economic perspectives14(4), 75-99.
  • Bobilev, R., Boschini, A., & Roine, J. (2019). Women in the Top of the Income Distribution: What Can We Learn From LIS-Data?. Italian Economic Journal, 1-45.
  • Boschini, A., Gunnarsson, K., & Roine, J. (2018). Women in Top Incomes: Evidence from Sweden 1974-2013. IZA Discussion Paper No. 10979 .
  • Card, D., Cardoso, A. R., & Kline, P. (2015). Bargaining, sorting, and the gender wage gap: Quantifying the impact of firms on the relative pay of women. The Quarterly Journal of Economics131(2), 633-686.
  • Goldin, C. (2014). A grand gender convergence: Its last chapter. American Economic Review104(4), 1091-1119.
  • Meriküll, J., Kukk, M., & Rõõm, T. (2019). What explains the gender gap in wealth? Evidence from administrative data. Bank of Estonia WP No. 2019-04.
  • Vahter, P., & Masso, J. (2019). The contribution of multinationals to wage inequality: foreign ownership and the gender pay gap. Review of World Economics155(1), 105-148.

The Gender Wage Gap in Belarus: State vs. Private Sector

20190830 The Gender Wage Gap in Belarus FREE Network Policy Brief Image 02

This brief is based on research that studies gender difference in wages in Belarus using survey data from 2017. According to the results, the unconditional gender wage differential equals 22.6%. The size of the wage gap is higher in the state sector than in the private sector. Additionally, it increases in the state sector throughout the wage distribution and accelerates at the top percentiles, indicating the presence of a strong glass ceiling effect.


The causes and consequences of the gender wage gap in the labor market, that is the difference between the wages earned by women and men, continue to attract increasing attention in empirical studies worldwide.

Belarus’ labor market is not an exception and faces the problem of wage inequality like other neighboring and transition countries. According to the National Statistical Committee of the Republic of Belarus (Belstat), the average gender wage gap in terms of monthly wages was 19% in 2000, it increased up to 23.8% in 2015, and reached 25.4% in 2017.

In this regard, this brief updates the estimates of the gender wage gap in Belarus. And it summarizes the results of the study on what the role of the state and private sectors are in the distribution of gender wage differences in Belarus (Akulava and Mazol, 2018).

Data and methodology

The data used in the research is from the Generations and Gender Survey (GGS) conducted in Belarus in 2017. This survey is a nationally representative dataset that is based on interviews of about 10,000 permanent residents of Belarus, aged 18–79, covering the whole country disaggregated by regions. The GGS contains information on a range of individual (age, gender, marital status, educational attainment, employment status, hours worked, wages earned etc.) and household-level characteristics (household size and composition, land holding, location, asset ownership etc.).

The analysis is based on the typical Mincer model of earnings that estimates individual wage income as a function of various influencing factors using the OLS approach (Mincer, 1974). Specifically, the Mincerian wage equation is defined where the log of the hourly wage rate is regressed on a set of male and female workers’ personal and job characteristics (educational level, working experience, occupational type, organization type, family characteristics, and region).

Next, we use the Oaxaca-Blinder (OB) methodology (Oaxaca, 1973; Blinder, 1973) to identify and quantify the contribution of personal characteristics and the unexplained component (which is referred to as differences in returns) to the wage difference between males and females.

Finally, we apply the Machado-Mata (MM) technique (Machado and Mata, 2005) to look into the nature of the wage gap at various points of the income distribution and also to test the difference for individuals employed in the state or private sectors. For the Machado-Mata procedure, we estimate our specifications at the 10th, 25th, median, 75th and 90th percentiles of the wage distribution.


The analysis shows that women’s wages are lower than men’s wages all over the wage distribution. The average raw gender wage gap equals 22.6% and it increased substantially compared with 9.0% in 1996 and 17.8% in 2006, the numbers obtained in the study conducted by Pastore and Verashchagina (2011).

Figure 1. Gender differential by quantile of the wage distribution

Source: Authors’ estimates based on GGS.

The level of female earnings is lower than the male regardless of the occupational type, educational background, work experience and organizational type. Moreover, the underpayment of women is lower for low earning workers, but increases up to the end of the wage distribution (see Figure 1).

The OB decomposition shows that female educational attainment and job-related experience help to decrease the level of the wage gap slightly (see Table 1).

Table 1. Oaxaca-Blinder decomposition results

Source: Authors’ estimates based on GGS.

However, the occupational choice is leading to an expansion of the difference in earnings. However, its effect is also small, indicating that occupational segregation plays a minor role in explaining the gender wage gap. The major share of the gender wage gap is formed by the unexplained part, which is likely to be attributed to discrimination.

Next, the level of remuneration is higher among private companies. However, contrary to other countries in transition, the average gender wage gap in Belarus in the private sector is lower than in the public sector.

Moreover, the MM decomposition estimates presented in Table 2 demonstrate that the gender wage gap in the state sector shows evidence of the glass ceiling effect (the size of the total wage gap expands at the top of the wage distribution), while no evidence of either glass ceiling or sticky floor (the size of the total wage gap increases at the bottom of the wage distribution) in the private sector.

The negative coefficient near the characteristics part in the private sector shows that female endowments outweighs their male counterparts. Thus, controlling for personal characteristics, if the labor market rewards males and females equally, the wages of females in the private sector should be substantially higher (see Table 2).

Table 2. Machado-Mata decomposition of the observed gender wage gap by organization type

Source: Authors’ estimates based on GGS.

Finally, the results also suggest that female workers are better off being in the private sector at the lowest and the highest percentiles (i.e. the size of the gender wage gap is lower there compared to the 25th and 50th percentile).

A possible explanation for all the above is that institutional differences seem to play a crucial role here. First, Belarusian private firms work under stronger regulation than in other transition economies which makes it harder for them to set low wages. Second, they also operate under stronger competition (compared to state companies), which force them to identify individual productivity more correctly, narrowing the gender difference in pay. In contrast, the paternalistic attitude to women left as a legacy from the Soviet Union further increases the gender wage gap in the public sector.


In this brief, we present new evidence on the existence of a gender wage gap in the Belarusian labor market and analyze the differences in its distribution between the state and private sectors.

Our results show that the unconditional gender wage gap in terms of hourly wages equals 22.6%. Thus, jointly with a previous study (see Pastore and Verashchagina, 2011) and recent official indicators, all these indicate that the pace towards gender equality in Belarus seems to be sluggish. For the moment, all institutional changes accomplished by the Belarusian government to reduce gender discrimination are not enough and require additional efforts to cope with that problem.

However, the gender wage gap is shown to be much wider in the public sector than in the private sector. At the same time the private sector appears to be more attractive than the public sector in the country in terms of the level of remuneration. Therefore, additional structural shifts of the economy accompanied by the growth of competition are needed to induce a further reduction of the gender wage gap.


Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Gender and the Agency Problem

20190520 Gender and the Agency Problem Image 02

Is it good for a firm to have a female CEO? Are countries with more female politicians less corrupt? An increasing attention to female representation in key roles in society has called for research exploring the outcomes and implications of such representation. A useful approach to investigate the impact of gender in such contexts is the so-called principal-agent framework which studies situations in which one party acts on behalf of another party. The idea is that the gender of participating parties is likely to affect motives, behavior and outcomes, predicted by the principal-agent framework. This brief reviews the use of the principal-agent framework for analyzing the effect of gender in two important areas of research: corporate finance and corruption. It outlines postulated theoretical channels for gender to matter, summarizes empirical findings and points to some of the policy challenges.

Increasingly, arguments in favor of more women in key positions are being put forth in society. Many European countries have by now introduced gender quotas for corporate board participation, with Norway being the first one to mandate a quota of 40% female board membership in late 2003. The United States joined the trend in 2018, with California being the first state to require women on corporate boards. The 2019 share of female CEOs in Fortune 500 companies is 5 %; while this number sounds very low, it is twice as high as a decade ago. Women’s presence in politics and bureaucracy is also increasing in many countries worldwide.

This tendency is clearly positive news in the fight for more gender equality, and it is likely to improve the position of women in the society. However, its implications for other economic and societal outcomes are not immediately clear. For example, is a more gender-balanced board or a female CEO good news for company performance? How would female politicians affect policy and societal outcomes?

One useful approach for answering such questions is based on the so-called principal-agent framework (developed to study what is known as “agency problems”). This framework, widely used in economics, political science and other related disciplines in the last half century, addresses the problem of incentivizing one person (referred to as an “agent”) to act on behalf of another person or entity (referred to as a “principal”). Many situations in real life are well described by this basic framework and it has been used in a wide range of different contexts, from relationships within a firm, or between a lawyer and her client, to insurance, real estate, policy choices by elected officials or appointed bureaucrats, and even situations involving corruption.

The relevant question is then whether, and if so, how, the gender of the agents can affect motives, behavior and outcomes, predicted by the principal-agent framework. This brief will focus on two main areas of studies within gender economics that use agency theory to motivate their findings: the role of gender in corporate governance, and in corruption. The brief will outline the theoretical channels through which the gender of the actors may act in these contexts, summarize the empirical findings of this literature, and shortly comment on policy implications. While the focus on two areas only may seem to be relatively narrow, it will allow identifying a number of common gender effects across the contexts, which may suggest implications for the other potential applications of the approach.

The basic principal-agent framework

Effectively any situation in which one party acts on behalf of another party for monetary or non-monetary compensation can be analyzed within an agency framework. A typical feature of such situations is that the parties have different objectives: for example, the board of the firm (the principal in this case) would be interested in maximizing the firm value, while the CEO (the agent) would probably be more concerned about her personal compensation. This difference is not necessarily problematic per se as long as the principal can get the agent to act as the principal wants. However, if parties do not have the same information – which is typically the case in the reality – the misalignment of their objectives becomes an issue.

Two main problems may arise in such situations. The first one is referred to as the problem of hidden action (moral hazard) – that the agent is likely to act in line with her own objectives, rather than in the principal’s ones. This is likely to occur as long as her effort cannot be perfectly monitored by the principal. For example, shareholders typically cannot directly attribute the evolution of the firm’s value to the actions of the CEO, which may result in the CEO making decisions that are, for instance, too risky from the firm’s value maximization perspective. The second one is the problem of hidden information – when the agent is better informed about the issues at stake than the principal, which again may result in the agent not acting in the best interest of the principal. For example, shareholders may have a poorer knowledge of the market than CEO, which may result in the CEO making decisions maximizing her own compensation rather than the firm’s value.

To lessen the extent of these problems, one needs to think of the spectrum of tools/decisions under the agent’s control, as well as of the design of her compensation schemes so as to align her private objectives with those of the principal. For example, to motivate a CEO to behave in the interests of shareholders, his/her compensation package typically includes company stock options. In some cases, the way to provide better incentives for the agent is to delegate more decisions, allow her more discretion and link her compensation closely to the outcome of her actions. One possible example of such a mechanism is franchising: on average franchisees retain about 94% of franchise profits, which would make them very motivated to achieve good franchise performance. However, the cost of high incentivization is the potential misuse of decision power, especially if the set of the decisions for an agent to have control over is not chosen wisely and if sufficient alignment (or intrinsic motivation) is not achieved. Another obstacle when implementing the principal’s preferred outcome is the trade-off between agent’s incentivization and risk aversion. The agent is typically seen as more risk-averse than the principal (for example, firms’ shareholders would typically diversify their risks by investing in a number of companies, while the CEO’s main source of income would be associated with the company she manages). As a result, the agent may avoid undertaking the principal’s value-maximizing actions because of the risks associated with them.

The bottom line of this discussion is that the task of incentivizing the agent may be difficult, and the principal’s best-preferred outcome may not be achievable.

Gender and the agency problem

There are many twists and modifications of the basic framework described above aimed at better modelling the specific problem at hand. One particular feature of the principal-agent relationship that has received increasing attention in the literature is the gender of the participating parties. The main strands of this literature have studied the relevance of gender for corporate governance and corruption.

Gender and corporate governance

The corporate governance part of the literature focuses on the impact of the gender composition of the board of directors or of the gender of the CEO on firms’ (or banks’) performance, risk-taking, capital allocation decisions, firm reputation etc. One standard approach to this set of questions is to consider the principal-agent relationship between the agent – the CEO – and the principal(s) – the board of directors (and sometimes other firm stakeholders) – and ask how, and why, the gender of either party may affect the relationship between them and the outcomes of this relationship.

There are several channels suggested by the literature. First, women and men may have different personal characteristics – such as risk aversion, level of confidence or ethical values (though there is not necessarily agreement on the direction of the difference: while most studies argue that, on average, men are typically more overconfident than women (e.g., Barber and Odean, 2001; Lundeberg et al., 1994), there is no consensus about risk attitudes – e.g., Jianakoplos and Bernasek (1998) or Croson and Gneezy (2009) show that women are more risk-averse than men, while Adams and Funk (2012) document the opposite). These differences in personal traits may affect the decision-making of a board/CEO in an incomplete-information environment and ultimately the firm’s performance.

Second, women and men may face different employment opportunities in case they lose their job, which, again, is likely to affect their decision-making and risk-taking (e.g., Faccio, Marchica and Mura, 2016).

Third, more gender-diverse boards may better reflect the preferences of (gender-mixed) firm stakeholders; in terms of the agency theory this would imply more aligned interests between the principal and the agent. It may matter because mixed-gender groups (and, by implication, boards) may exhibit different decision-making processes than same-gender groups, which, again, may introduce frictions into the agency relationship (e.g., Amini et al., 2017 or Van Knippenberg and Schippers, 2007).

Finally, the gender composition of the board may matter because female board members may improve monitoring over the actions of the CEO, since they are more independent not being part of the same “old boys’” social networks as the male members of the board and the (male) CEOs (Adams and Ferreira, 2009).

Empirically, this literature is largely inconclusive: while the majority of studies does find that the gender of the firm’s decision-maker(s) matters, the sign of the effect differs between studies, datasets and specifications. For example, based on a US sample of firms, Bernile, Bhagwat and Yonker (2018) find that more gender-diverse boards lead to lower firm risk, and better performance. In turn, Adams and Ferreira (2009) document negative effects of more diverse boards on performance. Sila et al. (2016) find no relation between board gender diversity and risk. Similarly ambiguous are the findings on the effect of CEO’s gender on firms’ performance, as measured by risk exposure, capital allocation, propensity to acquire, business strategies etc.

One possible reason for this variability of findings is the endogeneity of the presence of female CEOs/board members and firms’ outcomes, which is difficult to account for empirically (Hermalin and Weisbach, 1998; Adams et al., 2010). For example, female CEOs may self-select into firms with lower risks due to their own risk-aversion. Alternatively, corporate culture may affect the relationship between the gender of the CEO/board members and firm performance, etc. (see Adams, 2016 for an overview of this problem). There has been a number of attempts to address the causality/endogeneity issues in this context. For example, Bernile, Bhagwat and Yonker (2018) and Alam et al. (2018) exploit variation in the gender composition of boards created by the diversity of potential directors residing a non-stop flight away from the firm headquarters. Their motivation is that the personal travel costs of directors decrease with the availability of non-stop flights. Faccio et al. (2016) attempt to resolve the endogeneity issue by proxying the likelihood of hiring a female CEO by a measure of how many other firms that share board members with the firm in question have female CEOs. The idea there is that working with female CEOs in other firms may make board members more familiar with working with female executives, and more willing to hire a female CEO in the firm in question. A subset of the literature exploits reforms introducing gender quotas in corporate boards. These studies argue that the reforms are introducing an exogenous variation in the proportion of mandated changes in board gender composition – firms with more women in the board prior to the reform would need less adjustments to comply with the reform (see, e.g., Bertrand et al., 2018 for a state-of-the-art example of such an approach). Still, the endogeneity concern remains very valid for this literature. A recent literature overview by Kirsch (2018) or somewhat more dated, but still be relevant one by Terjesen et al. (2009) can be a good starting point for more detailed information on this field.

Gender and corruption

Similarly, there is a sizeable literature of gender aspects of corruption. This literature addresses a variety of topics, including the impact of corruption on women and gender inequality, gender-associated forms of corruption, and most importantly for us in the current context, gender attitudes and behavior towards corruption. One of the predominant theoretical mechanisms in this literature, again, uses agency theory. The main difference to the version of agency theory applied in the corporate governance case above is, perhaps, that in the case of corruption there is not always a clear pattern of subordination between the principal and the agent. More specifically, the principal for a (potentially corrupt) agent official may be either a higher-level official, or the direct recipient of her services or the electorate in general (of the agent official is elected). However, just as in the corporate governance literature, the gender vs. corruption literature asks the question how the outcome of an interaction between the principal and the agent would be altered by the gender of either party. It argues that women may behave differently from men in a corrupt environment through a number of channels, most of which resemble the ones in the corporate governance literature outlined above.

For example, gender differences in behavior and attitudes to corruption may be due to of personal traits, such as risk aversion or gender-specific conformity with social norms (e.g., Esarey and Chirillo, 2013 suggest that women are more likely to conform to the local social norms, so they are less likely to engage in corruption in an institutional environment where corruption is condemned, than in the societies when it is more accepted).

These differences may be due to differences in outside options of the corrupt official in case corruption gets detected (such as alternative employment opportunities). They may also be due to women not being part of business/political network(s), or having less experience in how things are done in decision-making positions. This could make them better monitors when they are in a principal role, or less able (or willing) to engage in corruption when in the role of agent. Thereby, it may result in a negative link between women in government and corruption, but only a short-term one (e.g., Pande and Ford, 2011). However, Afridi et al. (2017) argues for an opposite view, that a newly appointed female bureaucrat’s lack of experience may increase corruption due to inability to handle matters efficiently. Their empirical results indeed support it: in India newly appointed female council heads are less efficient than male ones due to lack of experience; this efficiency gap also includes higher corruption levels in female-led villages. With time, as the female council heads gain experience, the difference disappears.

As can be expected, empirically this field is again not entirely conclusive. The early empirical research suggested a negative link between gender and corruption, or, more specifically, found that a higher presence of women in government is associated with lower levels of corruption (e.g., Dollar, Fisman, and Gatti, 2001 or Swamy et al., 2001). However, there has since been a wide discussion about the causal mechanisms of this relationship. One of the arguments has been that this correlation is due to institutional mechanisms: greater representation of women in power is observed in a more developed institutional environment, which is also providing more effective checks on corruption (e.g., Sung, 2003). Still, the discussion is ongoing, as other scholars argue that the relationship is still in place even after controlling for the institutional factors, though not in all power positions (e.g., Jha and Sarangi (2018) show that female presence in parliament decreases corruption while other measures of female participation in economic activities have no effect). There is certain evidence of female bureaucrats being less aggressive in extracting bribes (Dabalen and Wane, 2008) or female business owners paying less bribes (Breen et al., 2017), but the determinants and the causal relationship of these findings are again, unclear.

There has been a number of attempts to resolve the causality issue of the gender-corruption link. Similarly to the corporate governance literature, researchers have used an instrumental variable approach (e.g., Jha and Sarangi (2018) use number of genders in a country’s language to instrument for female labor force participation, as it has been shown that gender discrimination is higher in countries where the dominant language has two genders as opposed to countries where it has no gender or three or more genders. The same authors use the year of universal suffrage to instrument the female participation in parliament). Unlike in corporate governance literature, a large part of this literature uses experimental approach, relying both on lab experiments to study gender attitudes to corruption (e.g., Rivas, 2013), and natural experiments (Afridi et al., 2017 study the reform in India that randomly allocated a third of council headship positions to women) and quasi-experiments (Brollo and Troiano (2016) look into close elections in Brazil and use a regression discontinuity design to show that female mayors are less likely to be corrupt). A useful overview of the literature is offered in Rheinbay and Chêne (2016).

Summing up and policy implications

There is an active public and academic debate about the greater involvement of women in key positions in society, its implications and outcomes, and potential policies to achieve it. A natural way of analyzing the implications of having more women in strategic positions utilizes the principal-agent modelling approach, with the presumption that the gender of the parties is likely to affect the model’s predictions and outcomes. A substantial attention in this literature has been devoted to the impact of gender in corporate governance and corruption. Importantly, these two strands of literature outline several common channels through which gender is likely to have an impact, such as risk aversion, outside opportunities in case of losing employment, etc. This similarity suggests that the same channels are likely to play a role in other gender-relevant agency contexts.

Another similarity between these two areas of research is the ambiguity of the results in terms of both theoretical predictions and empirical findings. One possible source of this ambiguity is, likely, suboptimality of the empirical methods used, which might not allow to adequately establish the causal relationship between the characteristics and outcomes of the agency relation and gender of its participants. Differences of the contexts of the empirical studies are another probable contributor to the variation in predictions and results.

However, this ambiguity obviously does not mean that policies to empower women should not be undertaken at all. First, even if the results of a particular narrowly-targeted policy are so far found to be ambiguous, it may still be highly useful in changing social norms, with all the benefits attached to it. For example, there is no sufficient evidence that establishing gender quotes in corporate boards would improve firms’ performance. For example, Ahern and Dittmar (2012) find that introduction of quota in Norway had a negative effect on Tobin’s Q. However, a quota reform in Norway resulted in the appointment of better qualified female board members and raised the career expectations of younger women post-reform (Bertrand et al., 2018). Second, this ambiguity stresses that there is no universal “silver bullet” policy applicable to all countries and contexts: the design of policies that address gender inequalities, as any other policy, needs to carefully account for the local institutional and cultural context. Further, recent contributions to this literature has become much more informative for the policy makers. An active development of this field and its methods suggests that we are about to learn much about the role of gender and other compounding factors in the above contexts. In other words, modern informed gender policy is just around the corner.


  • Adams, R. B., (2016). Women on boards: The superheroes of tomorrow? Leadership Quarterly, 27 (3). pp. 371-386.
  • Adams, R. B., Hermalin, B. E., & Weisbach, M. S. (2010). The role of boards of directors in corporate governance: A conceptual framework and survey. Journal of economic literature, 48(1), 58-107.
  • Adams, R. B., & Ferreira, D. (2009). Women in the boardroom and their impact on governance and performance. Journal of financial economics, 94(2), 291-309.
  • Adams, R. B., & Funk, P. (2012). Beyond the glass ceiling: Does gender matter?. Management science, 58(2), 219-235.
  • Afridi, F., Iversen, V. & Sharan, M.R. (2017), Women political leaders, corruption, and learning: evidence from a large public program in India. Econ. Dev. Cult. Change, 66 (1) pp. 1-30.
  • Ahern, K. R., & Dittmar, A. K. (2012). The changing of the boards: The impact on firm valuation of mandated female board representation. The Quarterly Journal of Economics, 127(1), 137-197.
  • Alam, Z. S., Chen, M. A., Ciccotello, C. S. & Ryan, H. E., (2018). Gender and Geography in the Boardroom: What Really Matters for Board Decisions? Mimeo. Available at SSRN: https://ssrn.com/abstract=3336445
  • Amini, M., Ekström, M., Ellingsen, T., Johannesson, M., & Strömsten, F. (2016). Does gender diversity promote nonconformity?. Management Science, 63(4), 1085-1096.
  • Barber, B. M., and Odean T.  (2001). “Boys Will Be Boys: Gender, Overconfidence, and Common Stock Investment.” The Quarterly Journal of Economics 116, no. 1: 261-92.
  • Bernile, G., Bhagwat, V., & Yonker, S. (2018). Board diversity, firm risk, and corporate policies. Journal of Financial Economics, 127(3), 588-612.
  • Breen, M., Gillanders, R., McNulty, G., & Suzuki, A. (2017). Gender and corruption in business. The Journal of Development Studies, 53(9), 1486-1501.
  • Brollo, F., & Troiano, U. (2016). What happens when a woman wins an election? Evidence from close races in Brazil. Journal of Development Economics, 122, 28-45.
  • Croson, R., & Gneezy, U. (2009). Gender differences in preferences. Journal of Economic literature, 47(2), 448-74.
  • Dabalen, A., & Wane, W. (2008). Informal payments and moonlighting in Tajikistan’s health sector. The World Bank Policy Research working paper 4555, https://elibrary.worldbank.org/doi/pdf/10.1596/1813-9450-4555
  • Dollar, D., Fisman, R., & Gatti, R. (2001). Are women really the “fairer” sex? Corruption and women in government. Journal of Economic Behavior & Organization, 46(4), 423-429.
  • Esarey, J., & Chirillo, G. (2013). “Fairer sex” or purity myth? Corruption, gender, and institutional context. Politics & Gender, 9(4), 361-389.
  • Faccio, M., Marchica, M. T., & Mura, R. (2016). CEO gender, corporate risk-taking, and the efficiency of capital allocation. Journal of Corporate Finance, 39, 193-209.
  • Hermalin, B. E., & Weisbach, M. S. (1998). Endogenously chosen boards of directors and their monitoring of the CEO. American Economic Review, 96-118.
  • Jha, C. K., & Sarangi, S. (2018). Women and corruption: What positions must they hold to make a difference?. Journal of Economic Behavior & Organization, 151, 219-233.
  • Jianakoplos, N. A., & Bernasek, A. (1998). Are women more risk averse?. Economic inquiry, 36(4), 620-630.
  • Kirsch, A. (2018). The gender composition of corporate boards: A review and research agenda. The Leadership Quarterly, 29(2), 346-364.
  • Lundeberg, M. A., Fox, P. W., and Punccohar, J. (1994). Highly confident but wrong: Gender differences and similarities in confidence judgments. Journal of Educational Psychology, 86( 1), 114
  • Pande, R., & Ford, D. (2011). Gender Quotas and Female Leadership. Background Paper for World Development Report, World Bank.
  • Rheinbay J. & Chêne, M. (2016). Gender and corruption topic guide, Transparency International, https://www.transparency.org/files/content/corruptionqas/Topic_guide_gender_corruption_Final_2016.pdf
  • Rivas, M. F. (2013). An experiment on corruption and gender. Bulletin of Economic Research, 65(1), 10-42.
  • Sila, V., Gonzalez, A., & Hagendorff, J. (2016). Women on board: Does boardroom gender diversity affect firm risk?. Journal of Corporate Finance, 36, 26-53.
  • Sung, H. E. (2003). Fairer sex or fairer system? Gender and corruption revisited. Social Forces, 82(2), 703-723.
  • Swamy, A., Knack, S., Lee, Y., & Azfar, O. (2001). Gender and corruption. Journal of development economics, 64(1), 25-55.
  • Terjesen, S., Sealy, R. & Singh, V. (2009). Women Directors on Corporate Boards: A Review and Research Agenda. Corporate Governance: An International Review, 17(3), pp.320–337.
  • Van Knippenberg, D., & Schippers, M. C. (2007). Work group diversity. Annu. Rev. Psychol., 58, 515-541.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.