TRADE, INVESTMENT AND COMPETITIVENESS TRADE, INVESTMENT AND COMPETITIVENESS EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT Women’s Economic Participation and the Impact of Regulatory Barriers Author: Souad Adnane Under the direction of: Iva Hamel And general supervision of: Christine Qiang © 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO), http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: 2021. Women’s Economic Participation and the Impact of Regulatory Barriers. EFI Insight - Trade, Investment and Competitiveness. Washington, DC: World Bank. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third- party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Graphic Designer: Diego Catto / www.diegocatto.com >>> Acknowledgments This insight paper was prepared under the overall guidance of Christine Qiang (Practice Manager, Global Investment Climate Unit; Finance, Competitiveness and Innovation) and Sylvia Solf (Senior Private Sector Specialist, Global Investment Climate Unit; Finance, Competitiveness and Innovation). It was authored by Souad Adnane (Private Sector Development Specialist- ETC-Global Investment Climate Unit; Finance, Competitiveness and Innovation) under the direction of Iva Hamel (Senior Private Sector Specialist, Global Investment Climate Unit; Finance, Competitiveness and Innovation). The team wishes to thank peer reviewers and colleagues for their constructive comments and valuable inputs during the insight paper’s preparation, including Abhishek Saurav (Economist, ETICI), S Anukriti, Research Economist (DECHD), Paula Tavares (Senior Private Sector Development Specialist, DECWL), and Susan Caroline Schroeder (Consultant, EECF2). >>> Contents Acronyms 5 Summary 6 Introduction 8 Methodology 11 1. Childcare leave policies: Balancing work life and family responsibilities 12 2. Labor restrictions and occupational segregation 15 3. Access to finance and the role of property rights 16 4. Legal capacity within marriage: Shifting intrahousehold bargaining power 20 5. Legal protections from violence and discrimination in employment: Reducing gender power imbalances 25 Conclusions and insights about legal reform 35 Bibliography 38 Annex A: Summary of studies: Reforms and treatments that impacted women’s economic outcomes 45 >>> Acronyms ECOA United States Federal Equal Credit Opportunity Act FLFP Female Labor Force Participation GDP Gross Domestic Product ILO International Labor Organization IMF International Monetary Fund MENA Middle East and North Africa Region MLFP Male Labor Force Participation OECD Organisation for Economic Cooperation and Development UK United Kingdom US United States WBL World Bank Women, Business, and the Law Database and Report WHO World Health Organization EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 5 >>> Summary Many countries seek to improve women’s economic participation with the introduction of target- ed laws and regulations. The impact of these reforms appears significant, although the support- ing evidence is stronger in some areas than others.This insight paper considers the impact of legal discrimination and the absence of protective legislation on women’s economic outcomes, namely employment and earnings. It also explores the various pathways or channels through which laws and regulations affect those outcomes. An understanding of those mechanisms is important to the effective design and implementation of gender equal and gender sensitive laws and policies. A survey of the literature uncovered five key pathways (Figure 1) that individually and in combination can help structure and improve the understanding of how legal reform im- pacts women’s employment and earnings. The available literature offered more causal evidence in some reform areas, namely property rights, retirement, and divorce laws, than in others. Em- pirical evidence from the remaining areas (childcare leave policies, occupational segregation, legal capacity within marriage) still establishes significant and strong associations, especially in the first area, between the studied legal reforms and women’s economic outcomes. Find- ings reported in the area of legal protections from violence and discrimination remain limited and inconclusive. There is a significant gap in the literature in terms of studies covering certain reforms in developing countries which undermines the generalizability of the findings. The paper concludes that although legal reform is not enough to bring about change, it is a critical first step in initiating social change and promoting women’s economic participation and women’s employ- ment in the formal sector. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 6 > > > F I G U R E 1 . - Key findings on the effect of legal reforms on women’s economic outcomes Incentivizes women to join and/or return to the Balancing work job market life and family responsibilities Policies mandating paid maternity, paternity, and parental leaves administered by the government Increases employment opportunities for women Reducing in certain sectors occupational segregation Legislation lifting restrictions on women’s work at night and in certain industries and jobs Improved women’s Improves women’s employment business ownership Legal Reform (also their intra-household Improving access bargaining power) (lifting barriers and to property rights establishing legal & finance Legislation enabling women’s protections) access to property and finance Reduced gender Enables women to make wage gap decisions to work outside the home and have control over their earnings Shifting intra- Legislation granting women household equal legal capacity within bargaining power marriage, (to move freely, to make decisions, and to undertake legal transactions) Improves financial inclusion and prospects Reducing power imbalances in the private & professional spheres Legal protections from violence and discrimination in the private and professional spheres EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 7 >>> Introduction Women’s participation in the labor market is associated with a wide range of economic develop- ment outcomes through contributing to total household income and affecting intra-household bargaining outcomes (Heintz 2006). Women typically invest a higher proportion of their income, relative to men, in their families and communities, improving children’s nutrition, health, and school attendance (Organisation for Economic Co-operation and Development – OECD - 2010). It is no surprise that gender gaps in entrepreneurship and labor force participation are signifi- cantly associated with reduced per capita income (Cuberes and Teigner 2012). Such gaps also impede economic growth as highlighted by a growing body of literature (International Monetary Fund - IMF - 2013). According to a 2018 World Bank study that examines the economic cost of gender inequality, countries are losing $160 trillion in human capital wealth due to differences in lifetime earnings between women and men. The figure is about twice the value of gross domes- tic product (GDP) globally (World Bank 2018). A McKinsey report finds that if all countries match the progress toward gender parity of the fastest-improving country in their region, as much as $12 trillion could be added to global GDP by 2025. Considering a “full potential” scenario, if women played the same role in labor markets as men, as much as $28 trillion could be added (McKinsey 2018). An IMF paper (Ostry et al. 2019) suggests that the barriers to female labor force participation, which include legal discrimination, are equivalent to an additional tax rate of 53 percent on wom- en’s labor income in the Middle East and North Africa (MENA) region, 32 percent in South Asia, and 25 percent in Latin America and Caribbean, compared to 4 percent in Europe and Central Asia. Removing those barriers would improve welfare in the average MENA country by 16.4 percent, and fully equalizing male and female labor force participation could increase its GDP by about 64 percent. Gender equal labor force participation could increase the GDP of the average country in South Asia by 30 percent and the average country in Latin America and Caribbean by 23 percent (Ostry et al. 2019). EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 8 Although more women are present in the labor market than at labor market outcomes, as measured by the participation of any other time in history, indicating the progress that has been women in the non-agricultural labor market and the wage gap made over the past few decades, their presence remains well between men and women. While the authors do not attribute below men’s. In 2014, female labor force participation (FLFP) causality to these results, they do suggest that where women was 54 percent for the median OECD country, which is 14 are faced with less discrimination under the law, they also en- points below male labor force participation (MLFP). FLFP joy more favorable outcomes in the labor market. was 49 percent for the median middle-income country (26 points below MLFP), and 64 percent for the median low-in- This insight paper considers how laws can hinder or enable come country (13 points below MLFP) (IMF 2018). Moreover, women’s economic participation, supported by a review of the women are still concentrated in low-paid, unstable, and poor- literature. It first aims to survey the impact of legal discrimi- quality employment (Heintz 2006). nation and the absence of protective legislation, as identified by the WBL topics, on women’s economic outcomes includ- The increase in labor force participation is commonly explained ing employment and earnings. Second, it aims to uncover the by demographic drivers such as improvements in female edu- various channels through which laws and regulations affect cation and declining fertility, in addition to government expen- those outcomes. An understanding of those mechanisms is diture policies including maternity leave and child benefits as important to the effective design and implementation of gen- well as tax policies. The impact of gendered legal restrictions der equal and gender sensitive laws and policies. The paper on women’s economic outcomes has been less explored (IMF complements and builds on existing literature reviews, mainly 2015). However, there is a growing body of work addressing Roy (2019), with the ultimate objective to inform gender policy this topic especially since the World Bank Group launched its related work carried out by either international organizations, Women, Business, and the Law (WBL) database and report. governments, or local civil societies. The database provides data about discriminatory laws and regulations in 190 economies, structured around eight indica- The literature covered consists mainly of empirical studies tors or topics and 35 legal data points. The topics, with the published in peer-reviewed journals and, to a lesser extent, corresponding data points, aim to reflect the legal barriers that reports and papers published by international organizations a woman faces throughout her professional life, including mo- such as the OECD and IMF. Most empirical studies on the bility restrictions (Mobility), unsafe and discriminatory work- subject focus on measuring the impact of legal barriers and places (Workplace), occupational segregation (Pay), limited reforms on women’s outcomes without much attention to the decision-making capacity within marriage (Marriage), lack of mechanisms underpinning this interaction. Reports and pa- adapted childcare leave policies (Parenthood), limited proper- pers published by international organizations provide further ty rights and access to finance (Assets and Entrepreneurship), description of the pathways through which laws affect wom- and unequal pension rights (Pension). Hyland et al. (2020) re- en’s outcomes in addition to qualitative insights about the im- port results from ordinary least squares regressions showing pact of legal barriers. that a higher score in the WBL index is correlated with better EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 9 The survey of the literature uncovered five different pathways, 4. Shifting intra-household bargaining power in favor of or impact channels, (Figure 1), that are used to structure this women: Granting women equal legal capacity within mar- paper. While the paper considers each of the pathways identi- riage, relative to men, (to move freely, to make decisions fied separately, it is important to note that these are not mu- within the household, and to undertake legal transactions) tually exclusive; legal reform in one area could lead to eco- shifts the intra-household bargaining power in their favor. nomic outcomes for women through multiple channels. It is This shift enables them to make decisions to work outside no surprise that three out of the five pertain to gender power the home, increase their earnings prospects, and have relations, reflecting the patriarchal nature of societies in which control over their earnings. The evidence, although lim- women are more likely to face legal barriers that undermine ited in certain areas, is conclusive. Additionally, there is their economic participation. little to no evidence related to the impact of the valua- tion of non-monetary contributions during marriage and 1. Balancing work life and family responsibilities: Policies remarriage laws on women’s economic outcomes. On the mandating paid maternity, paternity, and parental leaves other hand, child marriage is identified as a barrier that administered by the government allow women to better negatively affects these outcomes. balance work and family responsibilities and incentivize them to join and/or return to the job market. Despite the 5. Reducing power imbalances in the private and profes- mixed evidence, the literature overall supports the posi- sional spheres: Legal protection from violence and dis- tive effects of childcare leave policies on women’s em- crimination in the private and professional spheres em- ployment and earnings. powers women socially and professionally by providing them with legal recourse and shifting the gender power 2. Reducing occupational segregation: Lifting restrictions balance. Equal pension rules for men and women miti- on women’s work, at night and in certain industries and gate the “deferred earnings discrimination” and increase jobs, reduces occupational segregation and improves female labor supply. While many studies explore the im- women’s access to the job market as opportunities avail- pact of antidiscrimination legislation and equal pension able to them increase. The literature supports the nega- rules on women’s economic outcomes, the impact of tive effects of these barriers on women’s earnings and anti-violence legislation on these outcomes is left unex- employment. However, it lacks empirical studies on the plored. The literature emphasizing the economic cost of impact of their removal. domestic violence and sexual harassment is abundant though, which ultimately suggests that preventing these 3. Improving access to property rights and finance: Wom- forms of violence would mitigate their economic effects. en’s access to finance is undermined by both weak Evidence around the impact of antidiscrimination legisla- property rights and discrimination in access to credit. tion is mixed, pointing to some adverse effects. The literature supporting the importance of women’s ac- cess to property rights is abundant and conclusive de- The final section of the paper addresses the impact of gender- spite some unintended consequences related to the rise related legal reforms on informality and women’s participation of marital discord and conflict. Prohibiting gender-based in the economy. The conclusion summarizes the literature, discrimination in access to credit is also associated with a highlights gaps, especially regarding the generalizability of the positive impact on women’s financial inclusion despite the findings, and raises questions for further research. It also pro- limited evidence. vides further insights into how different topics interrelate and what it takes for legal reform to bring about real change. A ta- ble summarizing all the empirical studies reviewed for the pa- per is available in Annex A. This illustrates the scale of impact of each studied reform or treatment and the methods used. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 10 >>> Methodology This paper is a literature-based study, informed mainly by empirical studies published in peer- reviewed journals, international organizations’ reports, and a few unpublished manuscripts. Ap- proximately 100 resources were consulted. Half of these reported on original empirical research, using either quasi-experimental or non-experimental research designs. Theoretical papers and secondary materials, such as literature reviews and descriptive reports, accounted for the re- mainder. All the world’s regions are represented. However, regions are not represented proportionally. The majority of studies related to childcare leave, domestic violence, and gender-based dis- crimination focus on the global North. Studies on family law reforms, including women’s access to property, focus on the global South, especially South Asia and Sub-Saharan Africa. When necessary, some historical background and the current situation are provided for context. The paper complements and builds on existing literature reviews, mainly Roy (2019) which was used as a starting point. All the resources covered by Roy (2019) were consulted and further information about the studies, such as the components of reform and scope of impact are re- ported. Resources used in those studies were also consulted to expand the literature surveyed. Further studies were identified by conducting desk research, especially to cover areas that are not sufficiently addressed by Roy (2019) including gender discrimination in pay and access to credit. Additionally, Roy (2019) provides some discussion of mechanisms linking women’s legal rights with their outcomes. However, this discussion remains brief and limited to women’s intra- household bargaining power. Channels mediating the impact of gender equal laws on women’s economic outcomes may be more clearly observed by departing from the WBL constructed cat- egories/indicators. Legal provisions belonging to different indicators make better sense if clus- tered together. The paper uncovers five impact channels suggesting a different way of clustering and examining legal reforms. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 11 1. >>> Childcare leave policies: Balancing work life and family responsibilities Policies mandating paid maternity, paternity, and parental leaves administered by the government allow women to balance work life and family responsibilities and incentivize them to join and/or return to the job market. Despite the mixed evidence, the literature overall supports the positive effects of childcare leave policies on women’s employment and earnings. Childcare leave policies have a positive effect on women’s participation in the labor force. Lack of childcare leave provisions and other work-life balance measures coupled with uneven access to childcare options negatively affect women’s employment patterns (OECD 2014). A study conducted in Egypt and Jordan shows that in both countries women working in the public sector one year before marriage are much less likely to drop out of the labor force after mar- riage than those working in the private sector. The authors explain this variation by the fact that the public sector offers better work-life balance through more generous leave policies and more generous pension and retirement benefits (Sayre & Hendy 2016). Generous maternity leave policies could also have an adverse effect on women’s out- comes, especially when the employer bears their cost. Employers who directly pay for ma- ternity leave benefits and do not pay for any form of paternity leave for men may choose not to hire women or to pay them less (OECD 2014, OECD 2017, WBL 2020). It has also been shown that as parental leave is more likely to be taken by women, it can lead employers to discriminate against them especially for positions that require costly qualification and training periods (Man- del and Semyonov 2005). Policies that provide both paternity and maternity leaves and make the former mandatory have the advantage of not encouraging discrimination and help shift the underlying gender norms about care, in addition to increasing the likelihood of mothers returning to the job market (World Bank 2012). Employment before pregnancy matters for women. Berger and Waldfogel (2004) find that pre-birth employment in the U.S. is “an extremely strong predictor” of earlier return to work. Women who were employed during pregnancy go back to work more quickly after birth than women who were not. Also, women in pre-birth jobs with leave coverage are more likely to EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 12 return to work quickly than women without leave coverage. In expansions in leave coverage also reduced post-birth their study investigating the effects of family leave coverage employment rates among women in the short run but on women’s job retention in the United States (US), United not in the long run. Those expansions consisted of ex- Kingdom (UK) and Japan, Waldfogel et al. (1999) find that tending maternity leave benefit periods from 8 weeks to such coverage increases the likelihood of women returning 6 months, and then gradually to 18 months (Ludsteck to work after childbirth in the three countries. The increase is and Schonberg 2014). The expansion in Canadian paid estimated at 16 percent in the UK, 23 percent in the US, and family leave from 25 to 50 weeks is associated with de- 73 percent in Japan. creases in returns to work in the year after birth. This change is concentrated among economically advan- Paid maternity leave positively impacts female employ- taged women (Hanratty and Trzcinski 2009). ment and the impact is even greater if the government covers the leave cost. Amin et al. (2019) highlight the empiri- Ruhm (1998) highlights the positive effect of paren- cal evidence related to the effects of maternity leave on female tal leave reforms on female employment and wages employment in developing countries. The authors focus on a from 1969 to 1993 in nine EU countries. The analysis large cross-section of countries and account for the substan- indicates that “short periods of paid entitlement around tial variation across them in who bears the cost of maternity three months lead to a 3 to 4 percent rise in female em- leave, the government or the employer. They find, using firm- ployment rates, with little effect on wages, while longer level survey data for 66 developing countries, that women are entitlements of more than nine months lead to negligible more likely to return to work if the law mandates paid maternity additional impact on employment but sizeable negative leave. They identify a positive and statistically significant rela- impacts of about 3 percent on female wages.” Olivetti tionship between the length of paid maternity leave and the and Petrongolo (2017) explain that there are caveats to proportion of female workers in the private sector. According a causal interpretation of these results due to endogene- to their most conservative estimate, an increase of one week ity: omitted variables such the implementation of other of paid maternity leave is associated with a 2.6 percentage family friendly policies (i.e., subsidized childcare) among points increase in the share of female workers in a firm. This other factors (Olivetti and Petrongolo 2017). relationship is much larger when it is the government that pays for maternity leave instead of the employer. Childcare leave available to fathers may also positively affect women’s economic outcomes; the benefits of such In Nordic and other developed countries, generous ma- leave are stronger in developing countries. Goldin et al. ternity leave policies may have negatively impacted fe- (2017) show, using census data to explore career evolution male labor force participation. While the mentioned studies 14 years after leaving school, that the gender earnings gap show a positive association between maternity leave and a expands over the lifecycle. The gap for college graduates higher probability of return to work after childbirth, others show between 26 and 39 years old expands by 34 log points, 80 that generous childcare leave policies could negatively impact percent of which occurred during the first 7 years. About 44 women’s economic outcomes. This is seen, for example, in percent of this gap is due to “disproportionate shifts by men the following studies: into higher-earning positions, industries, and firms” and about 56 percent to “differential advances by gender within firms”. Ronsen and Sundstrom (2002) show that despite women Widening is even greater for married individuals, especially entitled to paid parental leave in Nordic countries having women with children, as mothers experience less internal a much higher overall employment entry rate during the advancement (Goldin et al. 2017). Leave policies, includ- first three years following birth than non-eligible women, ing leave available to fathers, could help to address this gap very long parental leave entitlements could negatively by increasing the chances for women to return to work after impact their careers and earnings and may preserve childbirth and allowing for a more equitable share of childcare an unequal division of household labor as they prolong responsibilities between mothers and fathers. Arnarson and women’s career breaks. Mitra (2010) find that following the major changes introduced by the Icelandic Act on Maternity/Paternity and Parental Leave Lalive and Zweimüller (2009) show that an Austrian re- (2000), including longer paid leave of 3 months allocated to form increasing the duration of parental leave from one fathers, the labor market opportunities for women increased. year to two years significantly reduced mothers’ return As more men in Iceland take paternity leave and devote more to work. Levels of employment and earnings decrease in time to childcare, women work longer hours and enter high- the short run but not in the long run. In Germany, major paying occupations that demand time and fewer interruptions. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 13 The legislation also reduces possible discrimination by em- from the 2005 and 2010 rounds of Canada’s General Social ployers against women of child-bearing age (Arnarson and Survey, the author analyses the impact of the reform on both Mitra 2010). Using data for 53 developing economies, Amin et mothers’ and fathers’ participation in market and non-market al. 2016 find a positive correlation between mandated pater- activities (the sum of housework and childcare). The findings nity leave and female employment in private firms. As per their show that mothers experience “impressive gains” if exposed conservative estimate, an increase of 6.8 percentage points to QPIP. Conditional on being employed, they spend 80 min- in the share of women workers for a typical firm is associated utes more physically at the workplace (increase of 44 percent with mandating paternity leave. from baseline) and are 5.4 percentage points more likely to be full-time employed (increase of 7 percent from baseline) com- Parental leave, or leave available to both mothers and fa- pared with mothers who experienced their last birth prior to thers, may encourage a more equitable division of child- the reform. The analysis also finds that exposed fathers spend care related tasks and responsibilities among parents 37 minutes longer in nonmarket work, per day (increase of and therefore promote women’s professional success. 23 percent from baseline). The author concludes that a clear Waldfogel et al. (1999) find that mandating this type of leave pattern of reduced sex specialization has emerged as fathers’ increases women’s return to work in the UK, the US, and Ja- participation in parental leave increased following the reform pan. Blau and Kahn (2013) find that parental leave policies (Patnaik 2019). Parental leave still does not exist in most low are associated with a higher increase in female labor force and middle-income countries which explains the lack of strong participation in most OECD countries. Lack of parental leave empirical research covering these economies. in the US led to the drop in the US’s rank from sixth among the 22 OECD countries in 1990 to seventeenth in 2010 in While the above studies show that leave allocated to fa- terms of female labor participation. The expansion of family- thers positively affects women’s labor market outcomes, friendly policies in many OECD countries explains 29 percent others find no such impact. These studies (Cools et al. of the decrease in US women’s labor force participation rela- 2015, Rege and Solli 2013, and Ekberg et al. 2013) examine tive to these countries (Blau and Kahn 2013). In a study cov- the effects of the introduction of one-month paternity leave in ering the Quebec Parental Insurance Plan (QPIP), Patnaik Norway and Sweden. They conclude that the paternity leave (2019) provides a “first comprehensive causal analysis of reform had little if any effect on gender gaps in wages and the short- and medium-term consequences of a policy aim- employment. It may be the case, as suggested by Cools et al. ing to promote paternity leave”. QPIP is a system of parental (2015), that 4 weeks of paternity leave may not be considered leave benefits introduced in Quebec in 2006 that sought to enough time to really impact long run outcomes. boost fathers’ participation in parental leave. Using time diary > > > B O X 1 - Summary of findings: Balancing work life and family responsibilities The literature provides mixed evidence about the effects of childcare leave policies available to both mothers and fa- thers on women’s employment outcomes. First, there is strong evidence that maternity leave increases the likelihood of women returning to work after childbirth, especially for those who were employed before pregnancy. However, extensive maternity leave policies have also led to reducing women’s employment and earnings in some cases. These effects, though, were only observed in the short run. An important question for future research is how long paid maternity leave should be before it starts having an adverse effect on women’s economic outcomes. Second, the literature shows that in most cases leave available to fathers, either in the form of paternity or parental leave, positively affects women’s labor supply and earnings. Guaranteeing paid leave to fathers along with incentives to use it can help recognize and redistrib- ute unpaid care work and shift the traditional gender roles, allowing the fathers to be more involved in supporting their children’s early development and the mothers to allocate more time to paid labor. However, some studies show little to no effect in this regard and raise a question on the “optimal” duration that would lead to positive impact. The limited literature covering developing country experiences undermines the overall conclusions about the effects of childcare leave policies on women’s economic outcomes. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 14 2. >>> Labor restrictions and occupational segregation Lifting restrictions on women’s work at night, and work in certain industries and jobs reduces occupational segregation and improves women’s access to the job market as opportunities available to them increase. The literature supports the negative effects of these barriers on women’s earnings and employment. However, it lacks empirical studies exploring the impact of their removal. Most job restrictions prohibiting women’s access to certain sectors, or from doing jobs deemed hazardous, arduous, or morally inappropriate, or certain jobs requiring night shifts are due to legacy legislation initially intended to protect women. Many western Af- rican labor laws, for instance, still mirror the former federation of French West Africa’s 1954 ordinance restricting women’s night work and access to certain industries. Many Commonwealth economies still have restrictions that can be traced back to the UK’s Mines and Collieries Act of 1842. A remnant of the 1932 Soviet Union Law could be seen in the lengthy list of job restrictions for women in many Eastern European countries. Portuguese and Spanish legacy legislation restricting women’s work in many sectors including mining, electricity, and construction, has also shaped the labor law and regulations in many former colonies (WBL 2018). Such restrictions are supported by obsolete international norms. The International Labor Organization (ILO) had endorsed restrictions on women’s employment in multiple conventions in the past. However, given the unintended adverse outcomes of such laws on women’s employ- ment, the ILO has revised its safety standards, shifting them away from the view that women need special protection and towards the view that safety measures should apply to all workers regardless of sex (OECD 2017). Restrictions on women’s work at night are underpinned by the same safety rationale. In 2017, the ILO abrogated Convention No. 4 - Night Work (Women), 1919 and Convention No. 41 - Night Work (Women) (Revised), 1932. Prior to this, the ILO qualified the two conventions as obsolete and invited States parties to consider denunciation. In 1990, it adopted the Night Work Convention, 1990 (No. 171), calling for protection of both women and men working at night.1 Additionally, the ILO qualified its Convention No. 45 - Underground Work (Women), 1935 prohibiting women from working in certain jobs as obsolete. The organization 1 https://www.ilo.org/public/english/standards/relm/ilc/ilc89/pdf/rep-iii-1b-in.pdf EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 15 invited governments to denounce it and give favorable con- percentage points in economies with at least one industry re- sideration to the ratification of Convention No. 176 - Safety striction compared to those with none. This makes sense given and Health in Mines, 1995, which shifts the emphasis from a that many of the jobs prohibited for women are in highly paid specific category of workers to the safety and health protec- industries, such as mining and manufacturing (WBL 2016). An tion of all mineworkers regardless of gender.2 The ILO recom- OECD report confirms that mining is one of the most lucra- mends providing equal protection for both men and women tive industries in Egypt and Jordan. Prohibited from working and restricting special protection measures to pregnant and in these sectors, women are excluded from some high paid breastfeeding women only (Paul 2004). positions and are not able to gain the necessary work experi- ence to become entrepreneurs in those fields (OECD 2017). Despite recommendations that special protection mea- sures for women should relate only to pregnancy and Empirical studies also support the negative effects of breastfeeding, many countries still restrict women’s par- such legal prohibitions on women’s earnings and em- ticipation in some industries and jobs. Provisions prohibit- ployment. Ogloblin (2005) points out the negative correlation ing women working in certain industries and occupying certain between restrictions on women’s access to certain jobs and jobs still exist in 89 of the 190 economies around the world gender wage differential in Russia. She finds that women’s covered by WBL 2020. Restrictions on women’s night work monthly earnings are 62 percent of men’s and that job seg- still exist in 26 economies, most of which are found in the regation by gender accounts for about three quarters of this MENA. In Egypt for instance, women are banned from work- differential. She notes the need for making labor market legis- ing from 7 p.m. to 7 a.m., in Jordan and Tunisia from 10 p.m. lation more consistent with a market economy. Using a natural to 6 a.m., in Algeria between 9 p.m. and 5 a.m., and in Libya experiment framework, Zveglich & Rodgers (2003) find that from 8 p.m. to 7 a.m. (OECD 2017). Taiwan’s 1984 Labor Standards Law, containing provisions that restrict women’s working hours, has a negative impact on Job and industry restrictions on women can be replaced women’s actual hours worked and employment. In an empiri- with a risk-based approach to occupational health. Build- cal study covering over 59,000 firms and 94 economies, Islam ing upon a critical review of the health hazards/risks attrib- et. al (2018) find that the legal systems that allow women to uted to restricted jobs in Azerbaijan for instance, a World Bank work during night hours in the same way as men are associ- working paper (2020) finds that over 70 percent of the 674 ated with an increase of 4.1 percentage points in the probabil- restrictions in force lack evidence-based justification. The ity of having a female top manager. remainder of restrictions typically involve working with lead, mercury, and other compounds with the potential to harm Restrictions on women’s night work and access to cer- an unborn fetus. The review highlights that risks of this kind tain jobs and industries confine them to low-paid and low- would be better managed through substance-specific regula- skilled sectors. Provisions limiting or prohibiting access to tions around safe-handling practices and provides examples certain professions and occupations create occupational seg- of these regulations and practices addressing women’s re- regation confining women to certain sectors that are generally productive health within an occupational health context. The low-skilled and low paid (agriculture workers, textile workers, review also recommends repealing Ordinance #170 in its en- teachers, nurses, social workers, public sector employees, tirety, that imposes industry restrictions on women, and imple- informal jobs, etc.). Restrictions on the types of occupations menting instead a risk-based approach to occupational health, and tasks available to women and their work at night reduce ensuring that workplace hazards are accounted for adequate- opportunities available to them and may more generally dis- ly, elsewhere in Azerbaijan’s Occupational Safety and Health courage employers from hiring and promoting them (OECD (OSH) framework (World Bank 2020). 2014, OECD 2017). OECD studies (OECD 2012, OECD 2014, OECD 2017) attribute women’s concentration in low- Restrictions on women’s night work and access to cer- paid and low-skilled sectors to cultural gender perceptions. tain jobs and industries are associated with negative eco- Cultural perceptions about what types of jobs are appropri- nomic outcomes. WBL (2016) finds that restrictions on the ate for women affect their employment patterns regardless of industries where women can work, and the type of jobs they legal restrictions. The latter, however, play a significant role can occupy lower their earning potential relative to men’s. The in sustaining those perceptions. Cultures are dynamic and female to male ratio of estimated earned income drops by 12 fluid; they change and evolve over time due to different fac- 2 https://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:13100:0::NO::P13100_COMMENT_ID:2252915 EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 16 tors. Establishing and keeping legal provisions in accordance paying jobs. Two decades after Chile repealed legal restric- with social norms may, however, perpetuate these norms and tions on women working in the mining sector, in 2018 women undermine therefore social change. represented 8 percent of the labor force in the mining sector. Similarly, in South Africa, two decades after opening up the Eliminating barriers preventing women from entering cer- mining sector to women, they made up 13 percent of the min- tain sectors or occupations has positive effects on eco- ing labor force in 2016, including 15 percent of senior man- nomic outcomes. The World Development Report has found agement. In both cases, legal reforms were accompanied by that removing such barriers increased output per worker by policies to encourage, recruit, and enable women to enter and 13 to 25 percent in some countries (WDR 2012). It could also remain in the profession (USAID 2018). accelerate women’s participation in employment and in higher > > > B O X 2 - Summary of findings: Reducing occupational segregation Legal restrictions prohibiting women’s night work or access to certain sectors or jobs deemed dangerous are due to legacy legislation initially intended to protect women. They are also supported by old and now obsolete ILO norms. In most cases, those restrictions are unjustified and associated with a set of negative economic outcomes for women’s employment and earnings. Additionally, many reports published by international organizations conclude that restrictions on women’s night work and access to certain jobs and industries confine them to low-paid and low-skilled sectors and that eliminating the restrictions have positive effects on women’s economic outcomes. There is limited empirical evidence, however, on the effects of such reform. The literature lacks empirical studies evalu- ating the impact of removing these legal barriers on women’s employment and earnings. This gap in the literature could be explained by the presumed direct link between lifting restrictions on women’s work in certain sectors and jobs and female labor supply. Systematically opening up job sectors to women increases the opportunities available to them; the impact may be considered obvious. However, reforms happen in different contexts and may not always lead to the same outcomes. Empirical studies would isolate the impact of the reform and highlight the role of other variables, such as enabling policies or social norms, in promoting or undermining the success of the reform. The question merits further investment and research. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 17 3. >>> Access to finance and the role of property rights Women’s access to finance is undermined by both weak property rights and discrimi- nation in access to credit. The literature supporting the importance of women’s access to property rights is abundant and conclusive despite some unintended consequences resulting from challenging traditional gender roles. Prohibiting gender-based discrimina- tion in access to credit is also associated with a positive impact on women’s financial inclusion despite the limited evidence. Women’s access to finance is disproportionally low around the world and is affected by their access to property rights. Many studies provide evidence for the existence of a global gender gap in access to finance (Aterido, Beck, & Iacovone, 2013; Demirguc-Kunt, Klapper, & Singer, 2013; Muravyev et al., 2009). As per the 2018 Global Findex Database, women repre- sent the majority of unbanked adults and are less likely to save formally. In Sub-Saharan Africa for instance, 37 percent of women have a bank account, compared with 48 percent of men.3 Women are also more likely to be excluded from the formal financial sector in countries where laws and social norms discriminate against them in terms of access to property. They are more likely to be financially included when they have equal and secure access to land and non-land assets use, control, and ownership. A study covering 141 economies shows that the female-to- male ratio reflecting the gender inclusion gap in the use of financial services in general drops by 30.7 percentage points when women are not guaranteed the same property rights as men. It drops by 66.3 percentage points for holding a credit card and 33.3 percentage points for holding a debit card (Morsy and Youssef 2017). Access to immovable property is key to access to finance, particularly to pledge collat- eral for securing bank loans. Persha et al. (2017) find that a second wave of land certification process in Ethiopia is associated with a 10 percent increase in the likelihood of households (both male-headed and female-headed) taking out any credit for farming purposes, and a small increase in the average amount of credit obtained. In Ethiopia, land cannot be used as collateral in the formal sense. However, land certification may play that role informally by providing a signal that “the borrower is attached to a place and likely committed to improving his or her productivity on that land, and perhaps conveying capacity and ability for repayment”. The land certification is 3 https://www.imf.org/external/pubs/ft/fandd/2020/03/africa-gender-gap-access-to-finance-morsy.htm#:~:text=Worldwide%2C%20women’s%20access%20to%20fi- nance%20is%20disproportionately%20low.&text=Credit%20rationing%20through%20high%20interest,(Morsy%20and%20Youssef%202017). EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 18 also shown to have led to an “11 percent average increase in cisions for themselves to work outside the home and providing the likelihood of female-headed households (and a 12 percent them with the necessary collateral to obtain loans (Hallward- average increase in the likelihood of widows) feeling more se- Driemeier & Gajigo, 2015). cure about entering into credit-based business transactions when the transactions occur with a holder of a land certifi- Access to property is also associated with broader out- cate” (Persha et al. 2017). In a study assessing the impact of comes, interconnected with women’s access to finance India’s land-allocation and registration program in West Ben- and their economic empowerment. Other pathways may gal, that made wives sole or joint-titleholders with husbands, explain the impact of property right reforms on women’s eco- Santos et al. (2014) report increased access to credit among nomic outcomes. Mishra and Sam (2016) show that land own- beneficiaries. The average household in the program is 12 ership plays an important role in empowering women in Nepal, percent more likely to report taking out a loan from a formal by enhancing their household bargaining power in the areas bank and 88 percent more likely to use a loan for agricultural of own healthcare, major household purchases, and visiting purposes than a household not in the program. The authors family or relatives. The three variables are used as proxies to also find that compared to their peers who do not benefit from measure intra-household bargaining power. The results indi- the program, women in households within the program are: cate two major findings. First, the authors report that women “(1) 12 percent more likely to be involved in decisions to take who own land are 13.91 and 38.3 percentage points more loans from self-help groups or microfinance institutions; (2) 12 likely to have the final say in their healthcare decisions using percent more likely to be involved in decisions on whether to the probit (uncorrected endogeneity) and CMP-IV (corrected purchase productive assets; (3) 9 percent more likely to be endogeneity) models respectively. Similarly, the probability of involved in decisions related to food purchase and consump- women having the final say in decisions regarding visits to tion; and (4) more likely to be involved in decisions about the family or relatives is 9.95 percentage points higher when they family land” (Santos et al. 2014). own land, according to the more conservative model (Mishra & Sam, 2016). In Vietnam, the change in the land law, in 2000, Equal legal access to property rights is positively asso- to allow joint titles has led to a steady expansion in women’s ciated with women’s business ownership and ability to land entitlements. Granting women either sole or joint titles to pursue economic opportunities. Islam et al. (2018) find that land has increased their business activity (Kabeer 2012). Us- laws granting married men and women equal ownership rights ing a 13-year longitudinal panel from rural Tanzania, Peterman are associated with an increase of 4.4 percentage points (sig- (2011) finds that improving women’s property and inheritance nificant at close to the 1 percent level) in the probability of hav- rights are significantly associated with women’s employment ing a female top manager and an increase of 6.9 percentage outside the home, self-employment, and earnings. points (significant at the 5 percent level) for having a female firm owner. Access to property is also associated with higher Inheritance reform supports women’s asset ownership, women’s labor force participation, although effects on coun- their age of marriage, and the duration of their educa- tries vary by level of income. The effects of property rights tion. In India, the Hindu Succession Act amendment to grant related reforms are almost always stronger in middle- and daughters and sons equal inheritance rights, is positively as- upper-level-income countries as shown by Hallward-Driemei- sociated with women’s asset endowment and their socio-eco- er et al. (2013). At the median income of the study sample, nomic outcomes, namely greater access to land, higher age an increase of income of one standard deviation would raise of marriage, and higher educational attainment. While the Act women’s labor force participation 2.1 percentage points high- was amended nationally in 2005, some southern states had er for countries with equal inheritance for wives and husbands amended it earlier. The states of Maharashtra and Karnataka than countries where wives do not have the same inheritance amended the Act in 1994, allowing for the passage of enough rights. The effects of closing the gender gap in property and time for Deininger et al. (2010) to estimate the impact of this inheritance rights are strongest for girls’ primary enrollment. legislative change using data from 2006. The authors find that They are positive across all levels of country income although although the reform was not effective in fully eliminating in- they tend to be lower as income rises. At the 25th percentile equalities in access to land, it significantly increased women’s of income, for example, countries that give sons and daugh- likelihood to inherit it. Females whose fathers died after the ters equal inheritance rights have on average a 6.4 percent- 1994 reform are 22 percentage points more likely to inherit age point higher ratio of girls to boys in primary enrollment land than those whose fathers died before 1994. Females who (Hallward-Driemeier et al. 2013). Access to property rights, married after 1994, but whose fathers died after 1994 have a including equal inheritance rights, shifts the intra-household significantly higher age at marriage, by 0.54 years, than fe- power dynamic in favor of women, allowing them to make de- males who married before 1994, reducing the pre-reform av- EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 19 erage marriage age difference between men and women of fant mortality and feticide; the increased cost of daughters that 3.36 years. On educational attainment, girls who started their take away some of the economic benefits of sons, gives par- elementary education after the amendment had 0.3 years ents an incentive to decrease their investment in the health of more elementary schooling than the older cohort. In order to their daughters or reduce their own fertility (Rosenblum 2015). put those results into perspective, the authors note that pro- The reform has also led to a rise in marital conflict leading to grams such as school meals and conditional cash transfer a higher suicide rate. As explained by the study, “increased schemes in developing countries have been shown to lead to opportunities for women can accentuate tensions and marital program effects in the range of 0.1 to 0.6 years of additional discord within households, by challenging traditional roles, in- schooling. They conclude that despite the latent educational creasing the importance of negotiation, and raising the poten- effect of inheritance reform, “the large estimated effects, to- tial for conflict” (Anderson and Genicot 2015). gether with the limited cost of such reform imply that changing inheritance legislation is a potent mechanism to improve not In addition to unequal property rights, women’s financial only women’s asset ownership but also human capital accu- inclusion may also be undermined by discrimination. mulation by the next generation” (Deininger et al. 2010). Women’s lack of access to property rights, and therefore to a set of economic empowerment factors including collateral, Inheritance reform also supports greater independence is not the only factor undermining their access to finance. for women and the possibility of their accessing better Women also face greater difficulty, relative to men, in obtain- paid jobs. Heath and Tan (2018) show that the Hindu Succes- ing credit due to discrimination (Muravyev et al., 2009). In the sion Act reform positively affected female labor supply through MENA region for instance, access to loans among women- the channel of “autonomy”, as proxied by the woman’s say owned formal small and medium enterprises is less than 6 per- in household decision-making, freedom of movement, and cent (Rockefeller Foundation 2013). In many MENA countries, access to a bank account. The Act led to statistically signifi- banks ask for the husband’s co-signature, “even if he lacks cant increases of 6.6 percentage points in the probability that the financial resources or is not involved in the venture,” to a woman has say in household decisions, 8.2 percentage make sure that “the woman’s actions do not interfere with the points in the probability that a woman can go to the market wishes of the family or her husband” (Chamlou 2008). These alone, 6.9 percentage points in the probability that a woman practices occur even though banking laws do not discriminate can go to a health facility alone, 8.3 percentage points in the against women borrowers. Adopting legal provisions prohibit- probability that a woman can travel outside the village alone, ing discrimination based on sex in access to credit would re- and 3.9 percentage points in the probability a woman has ac- duce such practices by providing women with legal recourse cess to a bank account. The reform also increased women’s to complain and remedy discrimination when it happens. labor supply into high-paying jobs outside the home. The like- lihood of a woman working all year instead of occasionally Legislation prohibiting gender-based discrimination in increased 5.7 percentage points, and to work for cash instead access to credit is associated with a positive impact on of no pay or for in kind payment, 6.1 percentage points (Heath women’s access to finance. In the United States, Ladd and Tan 2018). (1982) shows that the Federal Equal Credit Opportunity Act (ECOA) may have had a favorable impact on women’s ability Women’s access to property rights may also have nega- to obtain mortgage funds, at least in the two states covered tive consequences. Despite its positive impact on women’s by the study, New York and California. ECOA was originally economic outcomes and social empowerment, India’s inheri- enacted in 1974 to prohibit lending discrimination based on tance reform was also shown to have negative unintended sex and marital status. It was amended in 1976 to further consequences. As concluded by Roy (2019), gender progres- prohibit lending discrimination based on “race, color, religion, sive legislation could have such consequences in developing national origin, age, the receipt of public assistance income, countries, especially in the presence of structural inequalities or exercising one’s rights under certain consumer protection that interact to perpetuate the status quo. The author explains laws.”4 Data on sex and marital status of loan applicants in that in order to circumvent the law, some parents disinherit New York were not available until 1977, when state law re- their daughters by “gifting” their share of land to their sons. quired state-chartered banks to collect such data. Data from However, they also generally “compensate” their daughters 1977 were then used for both states. Discriminatory practices through higher dowries or more education (Roy 2015). India’s by mortgage lenders against women in those states, which inheritance reform is also associated with higher female in- appeared in their appraisal practices increasing the size of the 4 https://www.consumerfinance.gov/about-us/blog/what-you-need-know-about-equal-credit-opportunity-act-and-how-it-can-help-you-why-it-was-passed-and-what-it/ EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 20 down-payment required, their differentially high denial prob- and debit cards in their own names in economies where gen- abilities, and their higher interest rates and loan fees, were der-based discrimination in access to credit is prohibited by all impacted positively after the adoption of ECOA. In Califor- the law (WBL 2018). Focusing on 94 emerging and develop- nia for instance, data shows that in 1977 female applicants ing economies, Islam et al. (2018) find that laws prohibiting were charged interest rates that were 0.02 to 0.04 percentage gender-based discrimination by creditors are associated with points higher than those for comparable male-female appli- an increase of 4.1 percentage points in the probability of hav- cants (joint application). This differential disappeared in 1978. ing a female firm owner and an increase of 2.4 percentage Women also faced higher loan fees than men, which was not points in the probability of having a female top manager. It is the case in 1978. Moreover, properties of women were under- important to recall that other legal rights that go beyond the praised in 1977 in all areas covered by the study; the scope of aspects measured by the WBL index and their “strength” are the practice (number of areas where it was reported) declined also relevant to access to finance. Morsy and Youssef (2017) in 1978 (Ladd 1982). focus on the institutional and regulatory level and demonstrate that the gender gap in access to loans, use of debit cards, and Formally prohibiting gender-based discrimination in ac- use of bank accounts for work purposes decline with stronger cess to credit is associated with outcomes that go be- legal rights, referring to “the degree to which collateral and yond women’s increased access to loans. WBL data show bankruptcy laws protect the right of borrowers and lenders that more women have formal accounts at financial institutions and thus facilitate lending.” > > > B O X 3 - Summary of findings: Improving access to property rights and finance Women’s access to finance is undermined by both weak property rights and discrimination in access to credit. Evidence supporting the importance of women’s access to property rights is abundant and conclusive despite some unintended consequences. The literature addresses women’s property rights both as key to accessing finance, specifically to pledge collateral for securing bank loans, and as a means for social and economic empowerment more generally. Equal legal access to property rights is positively associated with women’s ownership of business and labor force partic- ipation. It is also associated with broader outcomes, interconnected with women’s access to finance and their economic empowerment. Access to property rights, through inheritance among other means, shifts the intra-household power dynamic in favor of women allowing them to make decisions for themselves and their children. It may also have some negative consequences as it challenges traditional gender roles and increases the importance of negotiation which may accentuate tensions and raise the potential for conflict. In addition to weak property rights, women’s financial inclusion may also be undermined by other factors, namely dis- crimination. Prohibiting gender-based discrimination in access to credit is associated with a positive impact on women’s access to loans and other financial services. It is also associated with a greater likelihood of women being top managers and business owners. The literature supporting these conclusions is, however, limited. It lacks empirical studies mea- suring the impact of reforms prohibiting gender-based discrimination in access to credit. A possible area of research would be to explore such impact, there are cases of reform around the world that would provide for a natural experiment context. Lastly, the literature covers further factors that reduce the gender gap in access to loans, namely the strength of collateral and bankruptcy laws, which may be another area for further investigation. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 21 4. >>> Legal capacity within marriage: Shifting intrahousehold bargaining power Granting women equal legal capacity within marriage, relative to men, (to move freely, to make decisions within the household, and to undertake legal transactions) shifts the intra- household bargaining power in their favor. This shift enables them to make decisions to work outside the home, increase their earnings prospects, and have control over their earn- ings. The evidence, although limited in certain areas, is conclusive. Additionally, there is little to no evidence related to the impact of the valuation of non-monetary contributions during marriage and remarriage laws on women’s economic outcomes. On the other hand, child marriage is identified as a barrier that negatively affects these outcomes. Married women are more likely to be the target of legal discrimination. Although most coun- tries have adopted principles of gender equality in their constitutions, discriminatory provisions that undermine married women’s agency to benefit from economic activities persist in many national family and civil legislations. Child marriage exacerbates the effects of those provisions as it further undermines women’s agency. One explanation underpinning the legal restrictions imposed on married women could be the role of social norms and customary and religious laws, which will be addressed later in this section. Those norms and laws are reflected for instance in restrictions on married women’s freedom to move and make decisions on an equal footing to men in many MENA economies. Legacy legislation is another reason for such discrimina- tion. Many Sub-Saharan family laws, for instance, still mirror the 1958 French Civil Code that restricted women’s ability to choose where to live, be a head of household, pursue a profession, and open a bank account. The Spanish Civil Code from 1960 and the common law principle of marital power also still apply in some countries, such as Equatorial Guinea and Eswatini, pre- venting women from giving consent to enter into a contract, register a business, or open a bank account. These provisions and others, applying to married women only, undermine women’s legal capacity within marriage by limiting their freedom of movement, their decision-making ca- pacity within the household, and their capacity to undertake legal transactions, and therefore their intra-household bargaining power, as shown by the literature covered in this section. Each of these elements is key to women’s economic participation and economic empowerment more generally. Golla et al. (2011) note that, “to have power and agency to benefit from economic activities, women need to have the ability to make and act on decisions and control resources and profits.” EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 22 Child marriage negatively affects women’s labor force ing the impact of family law reform in Ethiopia. The reform participation and earnings. The United Nations International was adopted in 2000, removing the ability of a spouse to deny Children’s Emergency Fund estimates the total number of permission for the other to work outside the home. It also intro- girls married under the age of 18 at 12 million every year. Sa- duced the requirement of both spouses’ consent in the admin- vadogo and Wodon (2018) suggest that in 15 countries, child istration of marital property and raised women’s minimum age marriage leads to a loss in earnings for women in adulthood of of marriage. The reform was not initially adopted nationwide, nine percent. Child marriage has a direct impact on girls’ edu- but only in certain regions and cities. Using two nationally rep- cational attainment and fertility, which in turn affects their ex- resentative household surveys, one in 2000 just before the pected earnings and labor force participation (Savadogo and reform and one five years later, the authors find that where Wodon 2018). Another study found that ending child marriage the reform had been enacted women were more likely to work would generate an additional $7.6 billion in Nigeria and $4.7 outside the home, work in paid and full-time jobs, and in oc- billion in Bangladesh in increased earnings and productivity cupations that employ more educated workers. They highlight (Wodon et al. 2017). that “the relative increase in women’s participation in these activities was 15-24 percent higher in areas where the reform Women’s mobility is key to their economic participation was carried out, the magnitude of the impact is significant too” and empowerment. Freedom of movement may be taken (Hallward-Driemeier and Gajigo 2015). for granted in many places, at least from a legal perspective. However, legal provisions limiting women’s freedom to decide The ability to choose where to live is another impor- where to live and travel still exist in approximately 30 percent tant aspect of mobility and is positively associated with of the world economies (WBL 2020). Mobility is a necessary women’s financial inclusion. It is shown that gender gaps precondition for accessing income and resources and hence in financial inclusion are higher where married women can- for empowering women both economically and socially. An not choose where to live in the same way as men, with wom- ILO report notes that due to mobility constraints, only a small en having less access than men to bank accounts and less share of women entrepreneurs in the MENA region benefits capacity to borrow from a financial institution (WBL 2018). A from advisory or business development services (ILO 2018). similar finding was reported in a study by Demirguc-Kunt et al. Htun et al. (2019) find that provisions on women’s legal ca- (2013) covering 98 developing countries and including also pacity including their ability to move are “by far the strongest women’s capacity to save. It is important to note that mobility predictor of the share of women with bank accounts, the share restrictions may also apply to non-married women and may go of women who participate in firm ownership, and female labor beyond the aspects measured by WBL. In Jordan, guardians force participation.” (generally fathers or, in their absence, any other male relative) can decide on a woman’s travel and place of residence until Women’s ability to travel freely is positively associated the age of 30, a “right” that could be extended if a daughter with their access to economic opportunities. Islam et al. is considered “untrustworthy” by court order (OECD 2017). In (2018) show that the probability of a firm having a female Mauritania, while the law allows a man to travel outside the owner is significantly higher when the law ensures that a mar- country with his children without having to obtain the mother’s ried woman can travel outside her home in the same way as consent, it requires women to obtain consent from the father, a man. They conclude that freedom of independent travel is in his capacity of legal guardian, to be able to travel interna- positively correlated with the ownership of business by wom- tionally with her children.5 en. More generally, restrictions on mobility or interactions out- side the home may limit women’s access to finance (Interna- In addition to restricted mobility, limited decision-making tional Finance Corporation 2011). In Pakistan, women’s weak capacity within the household is another aspect of wom- bargaining position, manifested among other things in limited en’s unequal legal capacity within marriage and is asso- mobility, hamper their movements for business purposes and ciated with negative economic outcomes. Many countries access to finance (Goheer 2003). still have laws that restrict women’s decision-making capac- ity within the household and subject it to the husband’s wish Restrictions on women’s ability to work outside the home by designating him the sole head of household, granting him in the same way as men are shown to affect their bargain- alone the unilateral right to divorce, allowing him to object to ing position in the household and therefore their ability to his wife’s work, or/and requiring the wife to obey him. Such pursue economic opportunities. This finding is highlighted laws restrict women’s autonomy and undermine their right to by Hallward-Driemeier and Gajigo (2015) in a study evaluat- choose for themselves. They are often associated with nega- 5 http://www.cdhah.gov.mr/ar/images/RECUEILTEXTESDROITSDEHOMME2019AR.pdf EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 23 tive economic outcomes in terms of access to financial ser- is finalized (WBL 2020). Gray (1998), in an impact assess- vices. Demirguc-Kunt et al. (2013), using individual-level data ment study of 1970s changes in divorce laws in the US, finds from 98 developing countries, find that in countries where that the adoption of unilateral divorce laws, “allowing either married women cannot get a job in the same way as men, spouse to end a marriage simply by claiming that it has broken women are less likely to have formal accounts, savings, or down,” are associated with an increase in women’s labor par- credit. Similarly, in countries where married women cannot be ticipation. However, this is only true in states with equitable- the head of their household and are required to obey their distribution property laws (community-property jurisdictions). husbands, women are also less likely to use formal financial The author highlights the importance of marital property laws products. The use of formal credit by women is very low in in determining how unilateral divorce, compared to mutual- countries with such laws (Demirguc-Kunt et al. 2013). consent divorce, shifts intra-household bargaining power. The author explains that household power dynamic is sensitive to Greater decision-making capacity within the household is the distribution of wealth and earning power within its mem- associated with women’s improved economic outcomes. bers. What each of the spouses can potentially attain from Hallward-Driemeier et al. (2013) show in their analysis that divorce is a leverage he or she has in the overall bargaining measures of greater legal capacity, which include recogni- process. It does not necessarily mean the probability of di- tion of women as heads of household, are associated with vorce increases. Lastly, the author supports his argument by improved employment outcomes. In the absence of gender referring to data from the CPS and Census suggesting that “a restrictions on the head of household status, women are more wife’s labor supply is an increasing function of her bargaining likely to participate in the labor force and more likely to move position within marriage” (Gray 1998). Therefore, the possibil- out of agricultural employment in favor of wage employment ity for women to make a divorce decision unilaterally, on the (Hallward-Driemeier et al. 2013). Gonzales et al. (2015) find same grounds as men, increases their labor supply in states that legal capacity related provisions, including allowing wom- with community-property jurisdictions. In another study as- en to be head of household and obtain a job in the same way sessing the impacts of introducing unilateral divorce laws in as men, are related to a statistically significant decrease in the US, Stevenson and Wolfers (2006) find results indicating the gender gap in labor force participation. They show that a large decline in domestic violence for both men and women the legal right for women to be head of household coupled in states that adopted the law. with more equal inheritance rights are associated with a larger decline in gender gaps in labor force participation in develop- The possibility to account for non-monetary contributions ing and emerging economies of around 4.6 percentage points. at the dissolution of marriage may also affect the intra- household bargaining process. The bargaining power mod- Restrictions on women’s legal capacity are also associ- el underlying Gray’s (1998) argument may be used to explain ated with negative human capital outcomes, which are the role of laws providing for the valuation of spouses’ non- closely linked to macroeconomic outcomes. Hallward- monetary contributions at the dissolution of marriage. Just like Driemeier et al. (2013) show that the legal restriction on wom- equitable-distribution property laws, provisions mandating the en being head of household is positively associated with infant valuation of non-monetary contributions during marriage is a mortality and that the effect of closing such a gap is stronger leverage that either spouse may have in the overall bargaining at lower levels of country income. At the 25th percentile of process. Such laws shift the intra-household bargaining power income, granting women the same legal right as men to be in favor of women in countries where they are the ones who head of household is associated on average with a decline in assume the bulk of care responsibilities. In those countries, infant mortality of 9.3 deaths per 1000 live births; at the 10th married women are therefore more likely to participate in the percentile the decline is 12.3 (Hallward-Driemeier et al. 2013). labor force when the law allows for valuation of non-monetary contributions. It is also important to highlight that community- Lastly, spouses’ capacity to dissolve marriage is also an property regimes systematically account for non-monetary important indicator of the intra-household power dynam- contributions. Separate property regimes, on the other hand, ic and may, under certain circumstances, lead to higher can penalize women who do not earn an income during mar- women’s labor participation. Unequal access to divorce lim- riage as these regimes do not recognize property accumulat- its women’s decision-making capacity, may trap them in rela- ed during the marriage, to which the spouses may have con- tionships of dependency, and, by extension, undermines their tributed non-financially, as subject to division (OHCHR 2013). access to income and property. Unequal rights to divorce can have negative effects on women’s intra-household bargaining Restrictions on married women’s decision-making capac- power and jeopardize their financial security when a divorce ity within the household are most prevalent in the MENA EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 24 region. They also exist in some Sub-Saharan African countries same way as men. In Lesotho, until 2006, the law regarded and few economies in East Asia and the Pacific. However, all married women as minors; it limited their capacity to enter twenty MENA economies, except for Malta, have at least one into contracts, start a business, and register property, among of the four restrictions measured by WBL: unequal rights to be other things. The husband, at his own discretion, had to sign head of household, to divorce and remarry, and the require- employment contracts on his wife’s behalf. Before 2016, the ment on women to obey their husbands. More than half have Democratic Republic of Congo had one of the most restric- at least three of these restrictions. In eleven countries, the law tive family codes for women’s agency and economic participa- requires the wife to obey her husband and in nine economies, tion. Among other restrictions, married women could not sign the role of “head of household” is granted to the husband only contracts, accept employment, or open bank accounts without (WBL 2020). Throughout the region, subsidies such as hous- their husband’s permission (USAID 2018). ing allowances and parental benefits are frequently allocated to men who are seen and designated by the law as heads of Equal rights to open a bank account, register a business, household (Coleman 2015). In connection with the head of and sign a contract are associated with improved female household principle, the law may prohibit women from working employment and entrepreneurship. Drawing on a WBL outside the home without their husband’s permission or if their panel data set covering 143 economies from 1960 to 2010, work is deemed to harm the family interest (WBL, 2020). Such Gonzales et al. (2015) find that in addition to restrictions on provisions provide a clear idea of the power dynamics inside women’s rights to inheritance and property, legal impediments the household and the limited negotiation power women have to undertaking economic activities, including opening a bank within marriage due to restrictive legislation. account and signing a contract, are strongly associated with larger gender gaps in labor force participation. In many cases, These restrictions may emanate from social and/or re- the gender gaps caused by these restrictions also have a neg- ligious norms. However, it would be useful to address ative impact on GDP (Gonzales et al. 2015). The ability to reg- them as legal barriers first. The literature often points to the ister a business in the same way as men is also positively cor- role of social norms as barriers to women’s economic partici- related with women’s economic participation. It is associated pation in the MENA region (World Bank 2013, OECD 2014, with an increase of 13.3 percentage points in the probability of Coleman 2015, OECD 2017, and ILO 2018). While it is true having a female firm owner (Islam et al. 2018). Similarly, Field that restrictions may emanate from social norms, the fact that et al. (2019) find that by having access to bank accounts and they are also formalized in those countries’ laws should not be using them to deposit earned income, rural women in India overlooked. As such, one study explores whether it would be increased their labor supply in the public and private sectors. useful to address them as legal barriers first. Majbouri (2017) In this study, women who were provided bank accounts and proposes an econometric model exploring the impact of family training on how to use them and had their wages directly de- law on women’s economic participation in MENA and shows posited into those accounts (instead of the male household that Islamic family law and oil and gas rents matter jointly. Per head’s - the status quo) significantly increased labor supply capita oil and gas rents reduce female labor force participa- by 0.16 standard deviation units one year after the treatment. tion rates in countries applying Islamic family law more than Women who participated in the experiment, and their hus- others. Oil and gas rents increase the ratio of male to female bands, became more accepting of female work. Households wages and therefore men’s intra-household bargaining power. with initially lower levels of female work and stronger norms Men have more say in household decisions including who can against it saw a higher concentration of these effects. Labor work outside the home. However, this increase in men’s bar- supply significantly increased by 0.21 standard deviation units gaining power is more likely to affect women’s economic par- among “socially constrained” women, who are “less likely to ticipation when labor market institutions and laws allow men work, less empowered, and their husbands are more likely to make such decisions. When laws are equal, rents are less to subscribe to norms against female work.” These women likely to affect women’s economic participation despite the in- worked more in the long run and became more empowered. crease in male bargaining power (Majbouri 2017). According to the authors, “these patterns are consistent with models of household decision-making in which increases in Related to women’s legal capacity within marriage is also bargaining power from greater control over income interact their ability to undertake certain legal transactions in the with, and influence, gender norms” (Field et al. 2019). EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 25 > > > B O X 4 - Summary of findings: Shifting intra-household bargaining power in favor of women Women’s legal capacity within marriage is curtailed by restrictions affecting three areas: freedom of movement, deci- sion-making capacity within the household and the capacity to undertake legal transactions. Legal impediments in those areas affect women’s economic outcomes by undermining their intra-household bargaining power. Additionally, there is little to no evidence related to the impact of the valuation of non-monetary contributions during marriage and remarriage laws on women’s economic outcomes. On the other hand, child marriage is identified as a barrier that negatively affects these outcomes. Restrictions on married women’s ability to move and work outside the home in the same way as men are shown to affect their ability to pursue economic opportunities. Removing those barriers is associated with improved women’s employ- ment and entrepreneurship. Removing restrictions on women’s ability to choose where to live is positively associated with higher financial inclusion. These conclusions, however, are supported by a limited number of empirical studies. Measuring the impact of reforms improving women’s mobility is a key area for further research. Limiting decision-making capacity within the household is another aspect of women’s unequal legal capacity within marriage. This is reflected in provisions designating the husband as sole head of household, granting him alone the uni- lateral right to divorce, allowing him to object to his wife’s work, or/and requiring the wife to obey him. These restrictions are associated with negative economic outcomes in terms of women’s access to finance and labor force participation. Removing them may lead to improved women’s employment and human capital outcomes. The literature highlights that restrictions on married women’s decision-making capacity may emanate from social and/or religious norms, but that it would be useful to address them as legal barriers first. Empirical evidence related to the impact of unequal divorce and the legal requirement for women to obey their husbands remains limited. Further research is needed to investigate these two aspects. The impact related to the valuation of non-monetary contributions at the dissolution of marriage is yet to be explored. Related to women’s legal capacity within marriage is also their ability to undertake certain legal transactions in the same way as men. Existing restrictions with respect to opening a bank account, registering a business, and signing a contract apply to married women only. Granting women equal capacity to undertake these transactions is positively associated with their employment and entrepreneurship. The literature related to this area is limited, though. One explanation could be the very low number of countries with such restrictions. Women’s capacity to sign a contract for instance is currently limited in only two countries. Empirical studies to evaluate the impact of removing such barriers are still possible, espe- cially in the existence of natural experiment contexts such as the Democratic Republic of Congo and Lesotho. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 26 5. >>> Legal protections from violence and discrimination in employment: Reducing gender power imbalances Legal protection from violence and discrimination in the private and professional spheres empowers women socially and professionally by providing them with legal recourse and shifting the gender power balance. Equal pension rules between men and women miti- gate the “deferred earnings discrimination” and increase female labor supply. While many studies explore the impact of antidiscrimination legislation and equal pension rules on women’s economic outcomes, the impact of anti-violence legislation on these outcomes is left unexplored. The literature emphasizing the economic cost of domestic violence and sexual harassment is abundant though, which ultimately suggests that preventing these forms of violence would mitigate their economic effects. Evidence around the impact of antidiscrimination legislation is mixed, pointing to some adverse effects. Gender based violence and discrimination are the manifestation of unequal gender rela- tions. Feminist theory posits that societal-level power imbalances within patriarchal societies create structural factors that work directly or/and indirectly to validate a male-dominated social order and family structure that often result in men exercising power and control over women in several ways, one of which is violence (Antai 2011). Ensuring a violence-free and discrimination- free environment within family and work is important to ensuring women’s social and economic empowerment. Women around the world are disproportionally affected by violence both in the private and professional spheres. Global estimates by the World Health Organization (WHO) indicate that 1 in 3 women worldwide has experienced physical and/or sexual violence in her lifetime. Women are also much more likely than men to be victims of sexual harassment, the latter being “a manifestation of power relations”6 which are often not in favor of women. They are more likely to be affected by discriminatory practices in different areas of work, which is another manifestation of power relations, both from a gender perspective and a hierarchical one. 6 http://hrlibrary.umn.edu/svaw/harassment/explore/3causes.htm EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 27 Violence against women undermines their earnings, their from rising rates of unemployment. However, the mitigating financial inclusion, and employment outcomes. Most of the effects of such laws disappear under severe economic crisis. violence experienced by women around the world is perpe- (Cook et al. 2019). trated by intimate partners. Thirty percent of women who have been in a relationship report that they have experienced some Mandatory arrest laws reduce the chances of domestic vio- form of physical and/or sexual violence by their intimate part- lence, but they also lower the probability of reporting. Iyen- ner. Globally, as many as 38 percent of murders of women are gar (2009) finds that mandatory arrest laws actually increased committed by a male intimate partner (WHO 2020). A litera- intimate partner homicides in the US due to decreased report- ture study shows that, due to factors including gender-based ing by victims, as victims are less willing to report an incident violence, women are less likely to generate significant propor- if their abuser will be arrested. This is not contradictory with tions of household income or to control or contribute to house- the findings mentioned earlier as Dugan (2003) also reports hold financial decision-making (Hughes et al. 2015). Demir- that while households in states that mandate arrest are less guc-Kunt et al. (2013) find that in economies where there is a likely to suffer from spousal violence, police in those states high prevalence of violence against women, women are less are less likely to discover an incident due to decreased report- likely to have formal accounts, savings or credit. Domestic ing. Mandatory arrest laws therefore have a deterrent function violence is associated with high economic costs to both the which reduces the chances of domestic violence but also low- employers and employees. It results in increased employee ers the probability of reporting, which may escalate the level of turnover, high rates of absenteeism, time off work to deal with violence creating a heightened emotional environment where legal proceedings, and healthcare needed for the victims. The the perceived cost of being detained is undermined. Sherman loss in individual employee productivity leads to greater over- (1992) concludes that “the efficacy of arrest depends heavily all organizational losses in productivity (Reeves et al. 2007). on the perpetrators’ perceived cost of being detained.” It also incurs a cost to the economy in general. Intimate part- ner violence costs the US economy $12.6 billion on an an- Sexual harassment affects women’s employment and ca- nual basis - 0.1 percent of GDP - compared to 1.6 percent reer trajectory. Multiple studies highlight the effects of sexual of GDP in Nicaragua and 2.0 percent of GDP in Chile (WHO harassment on women’s economic advancement and security 2020). In Egypt, the cost of domestic violence was estimated (Newman 1995, Shaw et al. 2018). Studies highlight the is- at least $208 million in 2015 and possibly as high as $780 mil- sue of sexual harassment and violence as a safety concern lion (UNFPA 2015). Legal protection measures, although not that may hinder women’s participation in the economy, espe- sufficient on their own, may act as a deterrent by establishing cially with the lack of security infrastructure in public trans- civil or criminal sanctions against perpetrators. They also en- portation in certain countries (OECD 2014, OECD 2017). In a able women to seek help and protection (UNSD, 2015). report on the challenges on youth employment in Egypt, the authors highlight “how fears of workplace harassment discour- Domestic violence legislation plays both a preventive and age many women from joining the labor force, especially in a deterring role. Dugan (2003) shows in an empirical study smaller workplaces and in less traditional jobs” (Gatti et al. covering the 50 states of the US that domestic violence laws 2013). In North Africa, nearly a quarter of women report the reduce the chances of domestic violence and may deter po- threat of abuse, discrimination, and sexual harassment as be- tential offenders from harming family and partners as well. ing the main reason for remaining outside the labor force (ILO She also uncovers other channels through which domestic 2018). Using in-depth interviews and longitudinal survey data, violence legislation can lead more cases into the criminal jus- McLaughlin et al. (2017) examine the effect of sexual harass- tice system: increased public awareness and an enhanced ment for women in their early career in the US. They find that perception of protection. Moreover, results point out that po- sexual harassment increases financial stress, an effect that lice officers are more likely to arrest a perpetrator if domestic could be largely attributed to precipitating job change. The lat- violence is legally established as a serious offense (Dugan ter, a common response among victims, can significantly alter 2003). In a similar study covering the US, Cook et al. (2019) women’s career attainment. Legislation addressing sexual ha- examine changes in incidents of intimidation and assault be- rassment in the workplace could prevent or mitigate those ef- tween intimate partners across states with and without man- fects and improve women’s economic outcomes. WBL (2018) datory arrest laws, and under circumstances of unemploy- finds that women are more likely to have majority ownership in ment. They find that while mandatory arrest laws are not a firms in economies where such legislation exists. single solution to domestic violence, their efficacy is strongest in times of economic stress. They show that these laws are Gender-based discrimination in employment takes mul- most effective to mitigate intimate partner violence resulting tiple and overlapping forms and may affect women’s hir- EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 28 ing, promotion, retention, and remuneration. Discrimination atively affect their employment. Eberts and Stone (1985) is defined by ILO’s Convention No. 111 on discrimination in explore the effect of the Equal Employment Opportunity Act employment and occupation as including “any distinction, ex- implemented in the 70s in the US. The authors use longitudi- clusion, or preference made on the basis of race, color, sex, nal data from thousands of individual educators from Oregon religion, political opinion, national extraction, or social origin, and New York states to investigate gender differences in pro- which has the effect of nullifying or impairing equality of oppor- motions to administrative positions and the impact of the leg- tunity or treatment in employment or occupation; […]” 7 Wom- islation. They conclude that “in the early 1970s female teach- en continue to face discrimination on the basis of their gender ers were significantly less likely to be promoted than similarly in almost every aspect of employment, including hiring, remu- qualified male teachers, but by the late 1970s [after the adop- neration, benefits and promotion. In many cases, their access tion of the legislation] the difference is no longer significant. In to job opportunities is restricted on the basis of their reproduc- addition, index measures of apparent discrimination declined tive role, due to them being mothers or caregivers to other significantly by close to half.” In a study exploring the effects dependents, or of a childbearing age (ILO 2011). Relative to of US anti-discrimination legislation – the Equal Pay Act and remuneration, recent estimates indicate that globally women Equal Employment Opportunity Act - on women’s employment are paid 77 cents for every dollar earned by men (UN Wom- and earnings Neumark & Stock (2001) find mixed results. en). Pregnancy discrimination is also found to be prevalent in They find evidence that sex discrimination/equal pay laws many countries as shown by the literature review conducted increased the relative earnings of black and white females, by Salihu et al. (2012) and covering the UK, US, Canada, and but at the same time reduced their relative employment. It is the European Union. A survey of employees suggested that important to note that unlike the UK Equal Pay Act, the US “pregnant women were seen as a liability in the workforce that Equal Pay Act only accounts for equal and similar work and causes undue stress to the employer”. not “work of equal value”. Antidiscrimination legislation, covering employment and Establishing the equal remuneration for work of equal pay, is positively associated with women’s employment and value principle may help to address wage discrimination. earnings. In a study covering 53 developing countries, Amin The principle was set out in the ILO’s Equal Remuneration & Islam (2016) find a strong positive relationship between the Convention, 1951 (No. 100). 8The Principle was called for in existence of a legal provision prohibiting gender- based dis- Sustainable Development Goal (SDG) target 8.5 within the crimination in hiring practices and women’s employment rela- framework of the United Nations 2030 Agenda for Sustainable tive to men. Mandating a non-discrimination law is associated Development.9 Equal pay for equal or similar work provisions with an increase of 5 percentage points in women’s employ- do not address situations where men and women perform dif- ment. The relationship is even bigger in richer countries and ferent jobs, using different skills, and involving different work- among relatively smaller firms (Amin & Islam 2016). Similarly, ing conditions, but which are nevertheless jobs of equal value. antidiscrimination legislation has been found to have a posi- Paragraphs 672–681 of the ILO’s General Survey on the fun- tive and significant association with both relative earnings and damental Conventions (2012)10, explain the importance and relative employment of women in the UK. The Equal Pay Act, the scope of the concept of “work of equal value” for enabling requiring equal pay for the same or like work or work of equal a comparison of different jobs since, because of historical at- value by men and women, was passed in 1970. Its application titudes and stereotypes regarding women’s aspirations, pref- was delayed until the end of 1975 to allow employers time for erences, and capabilities, women and men do not occupy the adjustment. At the same time, around the end of 1975, the Sex same jobs. The Report Committee explains that “the concept Discrimination Act, requiring equal employment opportunities of ‘work of equal value’ is fundamental to tackling occupational for men and women, was also passed. Zabalza and Tzanna- sex segregation in the labor market, which exists in almost tos (1985) find that due to the anti-discrimination legislation in- every country, as it permits a broad scope of comparison, in- cluding these Acts, relative employment of women increased cluding, but going beyond equal remuneration for ‘equal’, ‘the by more than 17 percent and relative pay by about 15 percent same’ or ‘similar’ work, and also encompasses work that is between 1970 and 1980, everything else being equal. of an entirely different nature, which is nevertheless of equal value.” Only 88 out of 190 economies covered by WBL 2020 While antidiscrimination legislation is positively associat- have established this concept in their respective legislations, ed with women’s promotion and earnings, it may also neg- most of which are quite recent and have not been followed by 7 https://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:C111 8 https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/documents/publication/wcms_216695.pdf 9 https://www.unodc.org/southeastasiaandpacific/en/sustainable-development-goals.html 10 International Labor Organization, https://www.ilo.org/public/libdoc/ilo/P/09661/09661(2012-101-1B).pdf EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 29 implementation mechanisms. The UK legislation is one of the of the gap is explained by women’s fewer years of employ- oldest and most complete in this regard. It establishes differ- ment and lower salaries.” In a similar vein, Lee et al. (2019) ent implementation mechanisms for the equal pay for work find that China’s mandatory retirement law requiring women of equal value principle, including a grievance procedure and to retire 5 years earlier than men reduces welfare for women. a “comparator” procedure allowing women to make an equal Equalizing retirement ages for men and women can generate pay claim while choosing the basis for comparison.11 a welfare gain for both (Lee et al. 2019). Unequal pension rules between men and women result in Increasing women’s retirement age to equalize it with deferred earnings discrimination. Ogloblin (2002) points out, men’s, reduces the deferred gender wage gap and increas- speaking of Russia, that “no discussion of the gender pay gap es female labor supply. In another study, Ogloblin (2005) […] can be complete without taking into account the lower le- shows that given the longer life expectancy of women, com- gal retirement age for women.” This is true for any discussion pared to that of men, the difference in retirement ages has an of gender wage gap, not only in Russia but in any economy even more sizeable impact on women’s deferred earnings. In where women are made to retire earlier than men. Unequal Russia, the legal retirement age is 55 for women and 60 for retirement ages affect women not only during their profes- men while life expectancy is 73 for women and 62 for men. sional life, by undermining their chances of getting promoted An average woman is therefore expected to be on a pension and making better earnings, but also after their careers end by for 18 years and an average man for 2 years. Accounting for reducing the size of their pensions compared to men’s. Pen- this increases the ratio of female to male lifetime earnings per sions may be viewed as “deferred labor earnings” (Ogloblin month of work to 78 percent (Oglobin 2005). Moreover, in- 2002). Granting women equal opportunities to access income creases in retirement ages for women are shown to be as- after their careers end is as important as during their career. sociated with increased female labor supply in many country specific studies (Staubli and Zweimüller 2013 on Austria; Ata- Early age retirement policies undermine women’s econom- lay and Barrett 2015 on Australia; and Cribb et al. 2016 on ic outcomes. Current early age retirement policies in China the UK). The Swiss pension system reform of 1995 is also and elsewhere, may be well-intentioned and aim to “relieve” a relevant example. The reform equalized the full retirement women from work earlier and provide them with more years age and set it at age 65 for both men and women at incep- of pension. However, the adverse consequences of such poli- tion. Lalive and Staubli (2014) find that a one-year increase in cies on women should not be ignored. Women should be al- full retirement age delays market exit by 0.51- 0.66 years and lowed to make their own decisions, which may very well go increases the claiming age of retirement benefits by 0.41-0.55 against what the legislators intend for them as showcased be- years. It is important to note that the established full retirement low by the Swiss example. In addition to women losing several age was not mandatory but an eligibility condition for claiming years of contributions and therefore receiving lower pensions, the full old age pension (Lalive and Staubli, 2014). The study’s women’s earlier retirement age may also influence recruit- results are reflective of women’s preference to work longer ment decisions and limit access to top-level positions (OECD and delay their claim for pension when they can. 2017, OECD 2014). Another relevant aspect of pension systems, which also Differences in retirement ages between men and women affects the size of women’s deferred earnings, is childcare affect the period they spend in the job market and there- pension credits. If unaccounted for, childcare responsibilities fore the size of their respective pensions. In China, among can affect women’s pension rights. Women are more likely to people aged 60 and older, women receive about half the take time out of work to care for children. This can reduce amount of men’s social pensions. Zhao and Zhao (2018) find their working life, their contribution records, and the earnings that “about three-quarters of the deficit in women’s pensions used to compute pension entitlements, undermining the size is explained by their lower likelihood of receiving occupational of their pension. To offset employment interruptions, some pensions, and one-third is due to smaller benefits when they pension systems establish care credits to account for periods do receive them. Among pension recipients, nearly one-third of absence due to childcare in pension benefits (WBL 2018). 11 https://www.equalityhumanrights.com/en/advice-and-guidance/equal-pay-claims EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 30 > > > B O X 5 - Summary of findings: Reducing power imbalances in the private and professional spheres While many studies emphasize the economic cost of domestic violence and sexual harassment, there is no evidence pointing to a direct connection between legal protections against these forms of violence and women’s economic out- comes. Studies show that violence against women undermine their earnings, their financial inclusion, and employment outcomes. Anti-violence legislation plays both a preventive and a deterrent role by reducing the chances of domestic violence, although it may also lower the probability of reporting. These findings ultimately suggest that addressing do- mestic violence may reduce violence and hence improve women’s economic outcomes. Similarly, studies highlight the negative effects of sexual harassment on women’s employment and career trajectory. However, there is no empirical evidence related to the impact of protective legislation on mitigating these effects. Exploring the empirical association between legal protections addressing domestic violence and sexual harassment and women’s economic outcomes is a possible area for research. Many studies explore this relationship in reference to legislation addressing gender-based discrimination in employment and wages. While antidiscrimination legislation is positively associated with women’s employment, promotion, reten- tion, and earnings, it may also negatively affect their hiring. This could be explained by the employers’ apprehension of a higher risk of discrimination allegations if discriminatory practices are already in place. Further research is needed to confirm this assumption and determine whether possible adverse effects of anti-discrimination legislation are associated with specific contexts and variables. Further research is also needed to explore and isolate the impact of prohibiting wage discrimination through the principle of equal remuneration for work of equal value on women’s earnings. Evidence is lacking in this regard as the existing studies do not differentiate between equal pay legislations that account for work of equal value and those whose scope is limited to equal or similar work. Lastly, the literature covers “deferred earnings discrimination”, which results from unequal pension rules between men and women. Studies show that early retirement age policies reduce women’s pension and labor supply as they make them spend less time in the job market. Increasing women’s retirement age to equalize it with men’s reduces this gap. Another relevant aspect of pension systems, which also affects the size of women’s deferred earnings, is childcare pen- sion credits. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 31 > > > B O X 6 - Women & the informal economy Informal employment among women remains a significant challenge around the world. In South Asia, 95 percent of female employment is concentrated in the informal sector. It is 89 percent in Sub-Saharan Africa and 59 percent in Latin America and the Caribbean.12 In the Arab region, women constitute 61.8 percent of workers in the informal sector.13 The informal sector is not covered by labor laws, which leaves its workers, especially women, exposed to many aspects of vulnerability including low pay, unsafe working conditions, and lack of social benefits such as pension and paid sick leave (OECD 2018, UN Women 2016). Gender equality laws encourage women’s employment in the formal sector. A recent World Bank study shows, us- ing linear regressions, that more women are employed in the formal sectors where there are fewer gender inequalities in the law. A higher WBL index score is associated with more waged and salaried female workers and fewer women in vulnerable employment, as classified by the International Labor Organization (WBL 2021). Some labor laws may encourage informality through the exclusion of certain categories of workers and types of work from their protection coverage. The Egyptian and Lebanese labor laws, for instance, exclude domestic work- ers and women in agriculture from their coverage. The Tunisian labor law covers employees working in agriculture but excludes domestic workers (OECD 2017). Micro-vendors, mostly women, are often chased and harassed by the police in Cambodia. The unregulated activities of micro-businesses are sometimes confused with illegal activities as they are not covered by labor legislation (ILO 2006). Migrants represent another category of workers that are often not covered by labor legislation, as in the case of Vietnam where female migrants make up a large proportion of the informal market and lack basic social protection.14 Laws affect not only women in the formal economy but also in the informal one, which has an even greater scope. By treating women and men in the formal job market equally and covering categories of workers that are likely to be found in the informal sector, laws may prompt the shift to formality of a significant number of female employees. The coronavirus pandemic and the subsequent economic recession have confirmed once again the negative effect of informality on women; the latter must disproportionately bear the economic brunt of the crisis. By moving to the formal sector, female employees not only benefit from better protection and work conditions, they are also empowered as they are given voice and representation through trade unions and other formal channels. 12 https://www.unwomen.org/en/news/in-focus/csw61/women-in-informal-economy#notes 13 https://reliefweb.int/sites/reliefweb.int/files/resources/policy_brief_on_the_impact_of_covid-19_on_gender_relations_in_the_arab_region_en_1.pdf 14 https://www.unwomen.org/en/news/stories/2016/8/migrant-women-of-viet-nam-claim-social-protection-and-rights EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 32 Is legal reform enough to bring about change? The short answer is no. Successful legal reform is contingent on a set of factors including enforcement mechanisms and a favorable social context. In an attempt to advance the scholarship that explores the relationship between law and social and economic progress, Lee (2017) presents a general theory that explains the mechanisms by which law impacts development. These “regulatory impact mechanisms” are interrelated and influence one another; they are comprised of three categorical elements: 1. Regulatory design: “concerns how optimally a law is de- 2. Regulatory compliance: refers to “the conduct of the gen- signed to achieve its regulatory objectives”. Factors that eral public in complying with law; law would not be effec- determine the effectiveness of regulatory design include tive without compliance by the general public.” This high- accounting for a particular development objective, effec- lights the importance of a supportive public opinion and a tive institutional arrangements such as accountability and favorable social context. enforcement mechanisms, and socioeconomic condi- tions on the ground. This aspect suggests the need for 3. Quality of implementation: “assesses the degree to a thoughtful and sensible law-making process concerned which a state meets the requirements of law and under- with the achievement of expected policy outcomes. takes the mandates under the latter to fulfill its objectives”. This element is determined by two outstanding factors, state capacity and political will. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 33 Enforcement mechanisms and social context seem to be the reforms can have an immediate positive impact on the common denominators underlying the above mechanisms. lives of many women around the world. In Bulgaria, Ger- These factors jointly determine the effectiveness of regulatory gana Ivanova became the first woman to serve in the National design and compliance, and the quality of implementation. Guards’ Unit and stand in front of the Presidency in 2012, two The eradication of footbinding in China is a great example of years after the Ordinance No. 14 of 18 October 2005 of the how the two factors interplay with each other. Ministry of Defense was reformed. Before 2010, only men were allowed in such a position. The restriction was repealed Both legal reform and advocacy to shift public opinion following a gender discrimination case against the Ministry and social norms were needed to end footbinding in Chi- of Defense filed by two female students.16 In Saudi Arabia, na. Footbinding lasted for 1000 years in China but ended in a restrictions on women travelling abroad were lifted in 2019. single generation. From the mid-seventeenth century and up Saudi women over the age of 21 are now able to obtain a until the mid-nineteenth century, efforts to end the practice by passport on their own and travel abroad without a male guard- the Manchu conquerors through legal abolition and penalties ian’s permission. On the other side of the Persian Gulf, Iran’s failed. In 1874, the first anti-footbinding society was founded, “Lady Goal”, Niloufar Ardalan, missed the Asian Cup because ushering in an era of civil society activism. The society intro- her husband refused to grant her permission to travel abroad duced the effective pledge technique inspired from the Ameri- as required by Iranian law.17 can prohibitionist pledge to abstain from alcohol. Society members pledged “not to bind daughters nor let sons marry To bring about change and pave the way for the neces- bound women.” More societies were created later in different sary reforms to promote women’s economic participation cities and an increasing number of members were acquired. around the world, governments and civil societies need They “propagandized the disadvantages of footbinding in Chi- to work together. It is of utmost importance to initiate regional nese cultural terms, promoted pledge associations, and subtly and national dialogues around laws and regulations affecting conveyed international disapproval of the custom.” By 1908, women’s participation in the economy and adapt them to ex- leading Chinese public opinion shifted in favor of opposing isting social realities and the need for economic growth. The footbinding. In 1912, a decree was issued by the National- latter cannot be achieved if half of the population is left out. ist Revolution to ban the practice; the decree was effective in Women’s economic participation is indispensable to the recov- many locales (Gerry 1996). In 1915, fines were levied from ery of the global economy from the effects of the COVID-19 government inspectors on those who continued to bind their pandemic. Removing legal barriers that currently undermine feet.15 Contrary to the 1912 decree, previous legal abolition this participation and establishing enablers should be a pri- and penalties failed to reduce the scope of the practice and ority on policy makers’ agendas, especially in regions where ultimately end it, because they were not supported by public the pandemic is expected to have greater impacts on women. opinion. Legal measures alone will not eliminate certain harm- The pandemic is expected to result in the loss of 1.7 million ful practices. Advocacy to shift public opinion, informed by jobs in the Arab region for instance, including approximately an “educational and integrationist” approach, has resulted in 700,000 jobs held by women, which is proportionally high and greater and speedier successes (Wilson 2012). significant given the initially low FLFP, the lowest in the world (ESCWA 2020). There is a pressing need for comprehensive While some legal reforms, especially those associated strategies to address the situation, legal reform is a necessary with entrenched social norms, may need a longer time but not sufficient condition for effective change. and a shift in public opinion to yield a real change, other 15 https://www.npr.org/templates/story/story.php?storyId=8966942&t=1612977369638 16 https://blogs.worldbank.org/jobs/impact-legal-reforms-women-and-girls-evidence-bulgaria 17 https://www.bbc.com/news/blogs-news-from-elsewhere-34258286 EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 34 >>> Conclusions and insights about legal reform The gender legal gap is estimated by WBL at 25 percent globally, which means that wom- en have only three quarters of the rights men are entitled to. This gap is the result of dis- criminatory provisions against women and the absence of legal protections in many national leg- islations around the world, especially in developing economies. Legal barriers hinder women’s economic participation in many ways and undermine their human capital. Legal reforms may affect women’s outcomes through different channels. Laws may en- able women’s economic participation, by allowing them to balance work life and family respon- sibilities, have access to a non-segregated job market, have access to finance and property rights, improve their intrahousehold bargaining power, or by reducing gender power imbalances in the private and professional spheres. These pathways are not mutually exclusive, and reforms can impact women’s economic outcomes through multiple channels simultaneously. Addition- ally, many reforms are interrelated and so are the different topics covered by WBL. Restrictions related to women’s mobility for instance are underpinned by the limited legal capacity of women within marriage and lead to negative outcomes with respect to access to finance. Property rights are both an enabler resulting in stronger rights in terms of access to finance and an end in them- selves leading ultimately to the social and economic empowerment of women. Evidence supporting the impact of laws on women’s economic outcomes is stronger in some areas than others, namely in the areas of childcare leave policies and property rights. In both areas however, the literature points to the existence of unintended negative con- sequences. Childcare leave policies have positive effects on women’s employment and earn- ings as they allow them to balance work life and family responsibilities and incentivize them to join and/or return to the job market. Extensive periods of leave, however, may in the short run have adverse effects on these outcomes. The evidence is also strong around property rights given the abundant and conclusive literature and the relevance of property rights not only to women’s access to finance but also to human capital outcomes. However, negative unintended consequences, such as the rise of conjugal discord and violence, may result from shifting the power balance between men and women when granting the latter equal property rights. Another legal aspect related to women’s access to finance is the legal protection from discrimination by creditors in access to credit. The literature covering this aspect is limited but concludes that prohibiting gender-based discrimination in access to credit is associated with a positive impact on women’s financial inclusion. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 35 The review of the literature suggests that women’s equal and shifting the gender power balance as possible channels. legal capacity within marriage can improve their intra- Feminist literature on violence and discrimination confirms the household bargaining power, although evidence is spo- role of gender power imbalances in initiating and perpetuating radic. Women’s legal capacity within marriage affects their violence and discrimination. freedom of movement, their decision-making capacity within the household, and their ability to undertake legal transac- The available literature offered more causal evidence in tions. While it is important to explore and isolate the effect of some reform areas, namely property rights, retirement, legal restrictions in each of these areas, accounting for the and divorce laws, than in others. The reviewed empirical interconnectedness of these restrictions and addressing them papers used various methodologies that fall under two major as one block undermining the legal capacity of married wom- categories: quasi-experimental identification approaches de- en merits further research and may yield stronger findings. Ad- scribing causal effects, and non-experimental approaches ex- ditionally, there is little to no evidence related to the impact of ploring simple relations between the variables studied. Further the valuation of non-monetary contributions during marriage details about each paper are provided in Annex A. While not and remarriage laws on women’s economic outcomes. On the seeking to review the strengths and weaknesses of method- other hand, child marriage is identified as a barrier that nega- ologies used in the cited studies, it is important to highlight the tively affects these outcomes. distinction between associations and causal estimates with respect to each of the reform areas. For example, many of the studies covering childcare leave policies, namely those ex- The empirical literature on the impact of removing la- ploring the impact of major expansions in parental leave poli- bor restrictions for women’s employment and earnings cies and paternity leave quotas on women’s employment and is limited. Empirical evidence exists on the negative effects return to work, used quasi-experimental research designs, of these barriers for women’s earnings and employment but including difference-in-difference and regression discontinu- there is less on the impact of repealing them, although re- ity. The findings pointed to the existence of negative causal ports published by international organizations provide some effects of generous parental leave on women’s outcomes and clear evidence in this regard. This may be because of a pre- mixed causal effects of paternity leave quotas. The remaining sumed direct link between lifting restrictions on women’s work studies in this area, however, established positive associa- in certain sectors and jobs and female labor supply. However, tions between childcare leave policies and women’s economic further research is needed to identify whether legal reform is outcomes; with no causal relationship inferred. The use of enough to improve women’s labor participation in the indus- quasi-experimental designs was most obvious and conclusive tries in question. in papers exploring the impact of women’s property rights. Findings using difference-in-difference, inverse probability There is little to no evidence on the impact of legal protec- weighting, and instrumental variables established evidence of tions against violence and discrimination in employment causal relationships between women’s land rights and their on women’s economic outcomes. Evidence supporting the economic outcomes. Causal evidence was also reported in positive impact of equal pension rules is strong. While many most of the papers cited with respect to legal reforms increas- studies explore the impact of antidiscrimination legislation ing women’s retirement age. Oaxaca–Blinder-style decompo- and equal pension rules on women’s economic outcomes, sition, difference-in-difference, and regression discontinuity the impact of anti-violence legislation on these outcomes is were the quasi-experimental identification approaches used in left unexplored. The literature emphasizing the economic cost these papers. Lastly, although only two studies on the effects of domestic violence and sexual harassment is abundant of divorce laws were reviewed, their findings of positive causal though, which ultimately suggests that preventing these forms effects between unilateral divorce laws, enabling women to of violence would mitigate their economic effects. Further make a divorce decision on the same grounds as men, and research is needed to confirm this conclusion. There is also women’s labor supply and the decline of domestic violence, little evidence supporting the impact of ‘equal remuneration respectively, are worth reiterating, considering the identifica- for work of equal value’ legislation as opposed to legislation tion approach that underpinned them. Findings reported in accounting only for the same or similar work. Evidence around the area of legal protections from violence and discrimination the impact of antidiscrimination legislation is mixed pointing to remain limited and inconclusive. Most findings reported in the some adverse effects. The channel through which legal pro- remaining reform areas (occupational segregation and legal tections against violence and discrimination in employment capacity within marriage) did not indicate causal effects. In- impact women’s economic outcomes is not straightforward. stead, most papers reviewed used various regression meth- Some studies, however, refer to enabling women to seek help ods establishing significant and strong associations between EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 36 the studied legal reforms and women’s economic outcomes. mies especially, where discriminatory laws often still prevail. Two factors may explain the lack of quasi-experimental iden- The generalizability of the current empirical findings regarding tification approaches in the surveyed literature in these areas. certain reforms remains to be tested in these settings. Further First, this type of reform does not easily lend itself to quasi- insights about the generalizability of the empirical findings re- experimental settings due to the difficulty to establish control ported in the paper may be inferred from Annex A. groups. Second, there is a lack of data in the contexts where these barriers are most relevant, which may have limited the The literature also highlights that legal reform is not number of studies. enough to bring about change. Although legal reform is a critical first step in initiating social change and promoting There is a significant gap in the literature in terms of stud- women’s economic participation, it is not enough to uphold ies covering certain reforms in developing countries, es- genuine progress. The latter requires both enforcement and pecially in the areas of childcare leave policies and legal advocacy efforts to shift public opinion for people to desire protection against violence and discrimination, which un- and accede to change. The eradication of footbinding in China dermines the generalizability of the findings. One explana- is a great example of the interplay between legal reform and tion could be the absence of certain provisions, such as pa- evolving social norms opposing a long-lasting harmful prac- rental leave or equal pay for work of equal value, from the laws tice. Gender legal and economic inequality is not only a con- of these countries. Another driver of these uneven empirical cern from the perspective of national welfare as shown by the insights is due to data limitation (lack of gender segregated present paper, but also and more importantly a concern for data and data about certain topics) in developing countries. justice and human rights. As Nobel Prize economist Ester Du- Although these reforms do not easily lend themselves to em- flo points out, “continuous policy commitment to equality for pirical settings, further research is needed to acquire a better its own sake may be needed to bring about equality between understanding of their impact on women in developing econo- men and women.” EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 37 >>> Bibliography Amin, M., and Islam, A. 2016. “Legal Institutions and Women’s Employment,” Enterprise Surveys Enterprise Note Series. Washington, DC: World Bank Group Amin, M., and Islam, A. 2019. “Paid Maternity Leave and Female Employment: Evidence Using Firm-Level Survey Data for Developing Countries.” Policy Research Working Paper WPS 8715. Washington, DC: World Bank Group. http://documents.worldbank.org/curated/ en/606031548250184438/Paid-Maternity-Leave-and-Female-Employment-Evidence-Using- Firm-Level-Survey-Data-for-Developing-Countries Amin, M., Islam, A., and Sakhonchik, A. 2016. “Does Paternity Leave Matter for Female Employ- ment in Developing Economies? Evidence from Firm-level Data”. Applied Economic Letters 23:16: 1145-1148. http://documents.worldbank.org/curated/en/124221468196762078/Does-paternity- leave-matter-for-female-employment-in-developing-economies-evidence-from-firm-data Antai, D. 2011. “Controlling Behavior, Power Relations within Intimate Relationships and Intimate Partner Physical and Sexual Violence against Women in Nigeria.” BMC Public Health 11 (1): 511. doi:10.1186/1471-2458-11-511  Arnarson, B., and Mitra A. 2010. “The Paternity Leave Act in Iceland: Implications for Gender Equality in the Labour Market.” Applied Economics Letters 17(7): 677-680. Atalay, K., and Barrett, Garry F. 2015. “The Impact of Age Pension Eligibility Age on Retirement and Program Dependence: Evidence from an Australian Experiment.” The Review of Economics and Statistics 97.1: 71–87. Berger, L. M., and Waldfogel, J. 2004. “Maternity Leave and Employment of New Mothers in the US.”, Journal of Population Economics 17(2): 331-349 Blau, F. D., and Kahn, L. M. 2013. “Female Labour Supply: Why Is the United States Falling Behind?” American Economic Review Papers & Proceedings 103(3): 251–256 Chamlou, N. 2008. “The Environment for Women’s Entrepreneurship in the Middle East and North Africa.” Washington DC: World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/6479/448240PUB0Box310on ly109780821374955.pdf?sequence=1&isAllowed=y EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 38 Coleman, I., and Abdelgadir, A. 2015. “The MENA’s Woman Problem: Progress and challenges in women’s economic participation.” In Women and girls rising: Progress and resistance around the world edited by Chesler, E., & McGovern, T. https://ebookcentral.proquest.com Cook, J., and Taylor, T.W. 2019. “The impact of mandatory arrest laws on domestic violence in times of economic stress” Economics Letters 178: 77-8. https://doi.org/10.1016/j.econlet.2019.02.013. Cools, S., J. H. Fiva, and Kirkeboen L. J. 2015. “Causal Effects of Paternity Leave on Children and Parents.” Scandinavian Journal of Economics 117(3): 801-828. Cribb, J., Emmerson, C., and Tetlow, G. 2016. “Signals matter? Large Retirement Responses to Limited Financial Incentives.” Labour Economics 42: 203-212. Cuberes, D., and Teignier, M. 2016. “Aggregate Effects of Gender Gaps in the Labor Market: A Quantitative Estimate.” Journal of Human Capital 10 (1): 1–32. Demirguc-Kunt, A., Klapper, L., and D. Singer. 2013. “Financial Inclusion and Legal Discrimination Against Women: Evidence from Developing Countries.” Policy Research Working Paper 6416. Washington, DC: World Bank. Demirgüç-Kunt, A., Klapper, L., Singer, D., Ansar, S., and J. Hess. 2018. “The Global Findex Database 2017: Measuring Financial Inclusion and the Fintech Revolution.”  Washington, DC: World Bank. Dugan, L. 2003 “Domestic violence legislation: Exploring its impact on the likelihood of domestic violence, police involvement, and arrest.” Criminology & public policy 2.2 (2003): 283–312. Web. https://search-proquest-com.proxygw.wrlc.org/docview/200174336?accountid=11243&r fr_id=info%3Axri%2Fsid%3Aprimo Eberts, R. W., and Stone, J. A. 1985. “Male–female differences in promotions: EEO in public education.” The Journal of Human Resources 20(4): 504–521. https://www-jstor-org.proxygw.wrlc.org/stable/pdf/145681.pdf?refreqid=excelsior%3A2551e05c f6aa5c369de843acc39fa8c9 Edmeades, J., Neetu J., Kes, A., Chata, M., Murithi. L., Nayihouba, K., Onagoruwa, A., Petroni, S., Savadogo. A., Steinhaus, M., Wodon Q., and Yedan, A. 2017. “Economic Impacts of Child Marriage: Global Synthesis Report.” Washington, DC: The World Bank and International Center for Research on Women. http://documents.worldbank.org/curated/en/530891498511398503/ pdf/116829-WP-P151842-PUBLIC-EICM-Global-Conference-Edition-June-27.pdf Ekberg, J., Eriksson, R., and Friebel, G. 2013. “Parental leave – A Policy Evaluation of the Swedish “Daddy-Month” Reform.” Journal of Public Economics 97: 131–143 Field, E., Pande, R., Rigol, N., Schaner, S., and Moore, C.T. 2019. “On Her Account: Can Strengthening Women’s Financial Control Boost Female Labour Supply?” Harvard University Working Paper. https://ideas.repec.org/p/cwl/cwldpp/2201.html Gatti, R. et al. 2013. Jobs for Shared Prosperity: Time for Action in the Middle East and North Africa. Washington DC: World Bank. Goheer, N.A. 2003. Women Entrepreneurs in Pakistan: How to Improve Their Bargaining Power. Geneva: International Labour Organization. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 39 Goldin, C., Pekkala Kerr, S., Olivetti C., and Barth, E. 2017. “The Expanding Gender Earnings Gap: Evidence from the LEHD-2000 Census.” American Economic Review: Papers & Proceed- ings 107(5):110–114. https://www.aeaweb.org/articles?id=10.1257/aer.p20171065 Golla, Anne M., Malhotra, A., Nanda, P., and Mehra, R. 2011. Understanding and Measuring Women’s Economic Empowerment: Definition, Framework and Indicators. Gonzales, C., Jain-Chandra, S., Kocchar, K., and Newiak, M. 2015. “Fair Play: More Equal Laws Boost Female Labor Force Participation.” IMF Staff Discussion Note SDN/15/02. https://www.imf.org/external/pubs/ft/sdn/2015/sdn1502.pdf Dahl, G., Løken, K., Mogstad, M., and Salvanes, K. 2016. “What Is the Case for Paid Maternity Leave?” The Review of Economics and Statistics, 98(4) October: 655-670. MIT Press. Gray, J. S. 1998. “Divorce-Law Changes, Household Bargaining, and Married Women’s Labour Supply.” American Economic Review 88(3): 628-642. Hallward-Driemeier, M., Hasan, T., and Rusu, A.B. 2013. “Women’s Legal Rights over 50 Years What Is the Impact of Reform?” Policy Research Working Paper 6617. Washington DC: World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/16318/WPS6617. pdf?sequence=1&isAllowed=y Hallward-Driemeier, M., and Gajigo, O. 2015. “Strengthening Economic Rights and Women’s Occupational Choice: The Impact of Reforming Ethiopia’s Family Law.” World Development 70: 260–273. Hanratty, M., and Trzcinski, E. 2009. “Who Benefits from Paid Family Leave? Impact of Expansions in Canadian Paid Family Leave on Maternal Employment and Transfer Income.” Journal of Population Economics 22(3): 693-711. Heintz, J. 2006. Globalization, Economic Policy and Employment: Poverty and Gender Implications. Geneva: International Labor Organization. Heath, R., and Tan, X. 2018. “Intrahousehold Bargaining, Female Autonomy, and Labor Supply: Theory and Evidence from India.” Working paper. https://economics.nd.edu/assets/227115/ hsa_heathtan.pdf Htun, M., Jensenius, F. R., and Nunez, J. N. 2019. “Gender-Discriminatory Laws and Women’s Economic Agency.” Working paper. https://academic.oup.com/sp/article/26/2/193/5303946?gue stAccessKey=17612756-5b8e-40ef-bca6-18e9ef8986ff Hughes, C., Bolis, M., Fries, R., and Finigan, S. 2015. “Women’s Economic Inequality and Domestic Violence: Exploring the Links and Empowering Women.” Gender & Development 23 (2): 279–297. doi:10.1080/13552074.2015.1053216  Hyland, M., Djankov, S., and Koujianou Goldberg, P. 2020. “Gendered laws and women in the workforce.” Working Paper Series WP20-7, Peterson Institute for International Economics. International Labor Organization. 2011. “Equality at work: The continuing challenge.” International Labor Conference 100th Session 2011 Report I(B). https://www.ilo.org/wcmsp5/groups/public/@ ed_norm/@relconf/documents/meetingdocument/wcms_154779.pdf EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 40 International Labor Organization. 2018. “Constraints and good practice in women’s entrepreneurship in MENA. Case study: New evidence on gender attitudes towards women in business.” International Labour Office, Impact Report Series, Issue 10. Geneva: 2018. IMF. 2013. “Women, Work, and the Economy: Macroeconomic Gains from Gender Equity.” IMF Staff Discussion Note, SDN/13/10. IMF. 2018. “Economic Gains from Gender Inclusion: New Mechanisms, New Evidence.” IMF Staff Discussion Note, SDN/18/06. Islam, A., Muzi, S., and Amin, M. 2018. “Unequal Laws and the Disempowerment of Women in the Labor Market: Evidence from Firm-Level Data.” Journal of Development Studies. doi: http:// dx.doi.org/10.1080/00220388.2018.1487055 Iyengar, R. 2009. “Does the Certainty of Arrest Reduce Domestic Violence? Evidence from Mandatory and Recommended Arrest Laws.” Journal of Public Economics 93: 85-98. Kabeer, N. 2012. “Women’s Economic Empowerment and Inclusive Growth: Labour Markets and Enterprise Development.” London: University of London. Klerman, Jacob A., and Leibowitz, A. 1997. “Labor Supply Effects of State Maternity Leave Legislation,” in Gender and Family Issues in the Workplace edited by Francine D. Blau and Ronald G. Ehrenberg. New York: Russell Sage Foundation. Klerman, Jacob A., and Leibowitz, A. 1999. “Job Continuity among New Mothers.” Demography 36(2): 145-155. Ladd, H. F. 1982. “Equal Credit Opportunity: Women and Mortgage Credit”, American Economic Review Papers & Proceedings 72(2):166-170. http://web.a.ebscohost.com.proxygw.wrlc.org/ehost/pdfviewer/pdfviewer?vid=1&sid=4c45ebd4- 120b-4549-9234-39091fc15b2b%40sdc-v-sessmgr02 Lalive, R., and Zweimüller. J. 2009. “How Does Parental Leave Affect Fertility and Return to Work? Evidence from Two Natural Experiments.” The Quarterly Journal of Economics 124(3): 1363– 1402. Lalive, R., and Staubli, S. 2014. “How Does Raising Women’s Full Retirement Age Affect Labour Supply, Income, and Mortality? Evidence from Switzerland.” Working Paper. http://crr.bc.edu/wp-content/uploads/2014/06/Panel-7_2-Lalive-and-Staubli1.pdf Lee, H., Zhao, K., and Zhou, F. 2019. “Does the Early Retirement Policy for Women Really Benefit Women?” Working Paper, University of Connecticut. https://media.economics.uconn.edu/working/2019-12.pdf Mackie, G. 1996. “Ending Footbinding and Infibulation: A Convention Account.”  American Sociological Review 61.6: 999–1017. https://search-proquest-com.proxygw.wrlc.org/ docview/218835480/fulltext/1DEAE14FC98D4266PQ/1?accountid=11243 Majbouri, M. 2017. “Oil, Laws, and Female Labor Force Participation.” International Advances in Economic Research 23: 91–106. doi: https://doi-org.proxygw.wrlc.org/10.1007/s11294-016-9621-9 Mandel, H., and Semyonov, M. 2005. “Family Policies, Wage Structures, and Gender Gaps: Sources of Earnings Inequality in 20 Countries.” American Sociological Review 70 (6): 949– 67. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 41 McKinsey Global Institute. 2015. “The power of parity: How advancing women’s equality can add $12 trillion to global growth.” McLaughlin, H., Uggen, C., and Blackstone, A. 2017. “The Economics and Career Effects of Sexual Harassment on Career Women.” Gender and Society 31(3): 333–358. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5644356/ Mishra, K., and Sam, G. A. 2016. “Does Women’s Land Ownership Promote their Empowerment? Empirical Evidence from Nepal.” World Development 78: 360-371. Morsy, H. and Youssef, H. 2017. “Access to Finance—Mind the Gender Gap.” EBRD Working Paper 202. London: European Bank for Reconstruction and Development. Nagarajan, H., Goyal, A., and Deininger, K. 2010. “Inheritance Law Reform and Women’s Access to Capital: Evidence from India’s Hindu Succession Act,” The World Bank Development Research Group, Agriculture and Rural Development Team. Washington DC: World Bank. Newman, M. 1995. “Sexual Harassment and Productivity: It’s Not Just a U.S. Problem.” Public Productivity & Management Review 19(2): 172-186. Neumark, D., and Stock, W. A. 2001. “The effects of race and sex discrimination laws.” Working Paper 8215. National Bureau of Economic Research. https://www.nber.org/system/files/working_papers/w8215/w8215.pdf OECD. 2014. “Women in Public Life: Gender, Law, Policy in the Middle East and North Africa.” Paris: Organization for Economic Cooperation and Development. OECD. 2017. “Women’s Economic Empowerment in Selected MENA Countries: The Impact of Legal Frameworks in Algeria, Egypt, Jordan, Libya, Morocco and Tunisia, Competitiveness and Private Sector Development.” Paris. Organization for Economic Cooperation and Development. http://dx.doi.org/10.1787/9789264279322-en Ogloblin, Constantin G. 2002. “Book review of Gender, Work and Wages in the Soviet Union: A Legacy of Discrimination. By Katarina Katz.” Industrial and Labor Relations Review 56 (1): 188-90. Ogloblin, C. G. 2005. “The Gender Earnings Differential in Russia after a Decade of Economic Transition.” Applied Econometrics and International Development 5(3): 5- 26. https://www.usc.gal/economet/reviews/aeid531.pdf Olivetti, C., and Petrongolo, B. 2017. “The Economic Consequences of Family Policies: Lessons From a Century of Legislation in High-Income Countries.” Journal of Economic Perspective 31 (1 Winter): 205–230. Paul, J. 2004. “Healthy beginnings: Guidance on safe maternity at work.” Geneva: International Labour Office. https://www.ilo.org/wcmsp5/groups/public/---ed_protect/---protrav/---travail/ documents/publication/wcms_travail_pub_66.pdf Persha, L., Greif, A., and Huntington, H. 2017. “Assessing the Impact of Second-level Land Certification in Ethiopia”, Paper prepared for March 2017 World Bank Conference on Land and Poverty. Washington DC: World Bank. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 42 Patnaik, A. 2019. “Reserving Time for Daddy: The Consequences of Fathers’ Quotas.” Journal of Labor Economics 37:4: 1009-1059. Peterman, A. 2011. “Women’s Property Rights and Gendered Policies: Implications for Women’s Long-term Welfare in Rural Tanzania.” Journal of Development Studies 47(1): 1-30. Reeves, C., and O’Leary-Kelly, A. 2007. “The Effects and Costs of Intimate Partner Violence for Work Organizations.” Journal of Interpersonal Violence 22(3): 327-344. Rege, M., and Solli, I. F. 2013. “The Impact of Paternity Leave on Fathers’ Future Earnings.” Demography 50: 2255–2277. Ronsen, M., and Sundstrom, M. 2002. “Family Policy and After-Birth Employment Among New Mothers – A Comparison of Finland, Norway and Sweden.” European Journal of Population 18: 121–152. Roy, S. 2015. “Empowering women? Inheritance rights, female education and dowry payments in India.” Journal of Development Economics 114: 233-251. Ruhm, Christopher J. 1998. “The Economic Consequences of Parental Leave Mandates: Lessons from Europe.” Quarterly Journal of Economics 113(1): 285–317. Santos, F., Fletschner, D., Savath, V., and Peterman, A. 2014. “Can Government-allocated Land Contribute to Food Security? Intrahousehold Analysis of West Bengal’s Microplot Allocation Program.” World Development 64: 860–872. Sayre, E., and Hendy, R. 2016. “The Effects of Education and Marriage on Young Women’s Labor Force Participation in the Middle East and North Africa.” In Young Generation Awakening: Economics, Society, and Policy on the Eve of the Arab Spring, Sayre, E., and Yousef, T. New York: Oxford University Press. Salihu, H. M., Myers, J., and August, I. F. 2012. “Pregnancy in the Workplace.” Occupational Medicine 62: 88–97. https://academic.oup.com/occmed/article/62/2/88/1480061 Savadogo, A., and Wodon, Q. 2018. “Impact of Child Marriage on Women’s Earnings across Multiple Countries.” Education Global Practice. Washington, DC: The World Bank. Schönberg, U., and Ludsteck, J. 2014. “Expansions in Maternity Leave Coverage and Mothers’ Labor Market Outcomes after Childbirth.” Journal of Labor Economics 32(3): 469–505. Sherman, Lawrence W. 1992. “Policing Domestic Violence Experiments and Dilemmas.” In “Domestic violence legislation: Exploring its impact on the likelihood of domestic violence, police involvement, and arrest” by Dugan, Laura. 2003. Criminology & Public Policy 2.2: 283–312. Staubli, S., and Zweimuller, J. 2013. “Does Raising the Early Retirement Age Increase Employment of Older Workers?” Journal of Public Economics 108: 17-32. Stevenson, B, and Wolfers, J. 2006. “Bargaining in the Shadow of the Law: Divorce Laws and Family Distress.” Quarterly Journal of Economics 121 (1): 267–88. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 43 United Nations Population Fund. 2015. “The Egypt Economic Cost of Gender-Based Violence Survey”. Retrieved from https://egypt.unfpa.org/en/publications/economic-cost-gender-based-violence-survey-egypt-2015 UNSD (United Nations Statistics Division). 2015. “The World’s Women 2015 Trends and Statistics.” New York: UNSD. https://unstats.un.org/unsd/gender/downloads/WorldsWomen2015_report.pdf Waldfogel, J., Higuchi, Y., and Abe, M. 1999. “Family Leave Policies and Women’s Retention after Childbirth: Evidence from the United States, Britain, and Japan.” Journal of Population Economics 12: 523-545. Waters, H., Hyder, A., Rajkotia Y., Basu S., Rehwinkel J.A., and Butchart A. 2004 “The economic dimensions of interpersonal violence.” Geneva: Department of Injuries and Violence Prevention, World Health Organization. https://www.who.int/violence_injury_prevention/publications/violence/economic_dimensions/en/ Wilson, A.M. 2013. “How the Methods Used to Eliminate Foot Binding in China Can Be Employed to Eradicate Female Genital Mutilation.” Journal of Gender Studies 22.1: 17–37. https://www.tandfonline.com/doi/full/10.1080/09589236.2012.681182 Wodon, Q., and de la Brière, B. 2018. “Unrealized Potential: The High Cost of Gender Inequality in Earnings. The Cost of Gender Inequality.” Washington, DC: World Bank. Retrieved from: https://openknowledge.worldbank.org/handle/10986/29865 World Bank. 2016. “Women, Business, and the Law 2016: Getting to Equal.” Washington, DC: World Bank. doi:10.1596/978-1-4648-0677-3. World Bank. 2018. “Women, Business and the Law 2020. Women, Business and the Law.” Washington, DC: World Bank. World Bank. 2020. “Women, Business and the Law 2020. Women, Business and the Law.” Washington, DC: World Bank. doi:10.1596/978-1-4648-1532-4. World Bank. 2011. “World Development Report 2012. Gender Equality and Development.” Washington: World Bank. https://openknowledge.worldbank.org/handle/10986/4391 World Bank, FAO, and IFAD. 2009. “Gender in Agriculture.” Washington, DC: World Bank. Zabalza, A. and Tzannatos. Z. 1985. “The Effect of Britain’s Anti-Discriminatory Legislation on Relative Pay and Employment.” Economic Journal 95(379): 679-699. http://web.b.ebscohost.com.proxygw.wrlc.org/ehost/pdfviewer/pdfviewer?vid=2&sid=243b2dd7- d9b7-4dd9-83a1-098b39e66433%40sessionmgr103 Zhao, R., and Zhao, Y. 2018. “The Gender Pension Gap in China.” Feminist Economics 24(2): 218-239. Zveglich, J. E., and Rodgers, Y. M. 2003. “The Impact of Protective Measures for Female Workers.” Journal of Labor Economics 21(3): 533-555. http://web.b.ebscohost.com.proxygw.wrlc.org/ehost/pdfviewer/pdfviewer?vid=1&sid=793fade9- 59e9-4699-8f47-ddbb0dda8c24%40pdc-v-sessmgr03 EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 44 A. >>> Annex A: Summary of studies: Reforms and treatments that impacted women’s economic outcomes The following table summarizes the empirical studies reviewed for this paper, including the methods used, to better illustrate the scale of impact and key takeaways of each studied reform or treatment. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 45 Variable Reform/ Research Country Author(s) directly Scale of impact18 Takeaway Treatment methods changed U.S. Berger and Maternity Return to In weeks 13 through 52 after childbirth, moth- Cox Pre-birth employment is “an extremely Waldfogel leave work ers with leave coverage had a 68.8% greater proportional strong predictor” of earlier return to (2004) risk of return to work compared to mothers hazards work. without leave coverage. models Women in pre-birth jobs with leave Women who worked pre-birth but did not coverage were more likely to return have leave coverage were about twice as to work quickly than women without likely to remain at home 12 months after leave coverage birth as leave covered women (22.7% versus 11.0%) UK, U.S., Waldfogel et Maternity Return to The likelihood of women returning to work Probit models Maternity leave coverage increased Japan al. (1999) leave work increased by: the likelihood of women returning to 16% in the UK, 23% in the US, and 73% in work after childbirth in the studied Japan. countries. 66 Amin et al. Maternity Proportion An increase of one week of paid maternity Ordinary There is a positive and statistically developing (2019) leave of female leave was associated with a 2.6 percentage Least significant relationship between the countries workers in points increase in the share of female work- Squares length of paid maternity leave and the the private ers in a firm. regression proportion of female workers in the sector private sector. This relationship is much larger when the government paid for maternity leave rather than the employer. Nordic Ronsen and Parental Women’s Mothers who were entitled to maternity leave Cox Very long parental leave entitlements countries Sundstrom leave employment resumed employment much faster than other proportional could negatively impact women’s ca- (Finland, (2002) entry mothers. The impact was largest in Sweden hazards reers and earnings and may preserve Norway where the employment risk of entitled models an unequal division of household and Women’s mothers was more than double the risk of labor. Sweden) return to non-entitled ones. work As the number of months of paid leave increased, the relative employment entry risks of entitled mothers diminished in all countries. In Finland, the initially high employment entry risk for entitled mothers largely disappeared when the paid leave was prolonged beyond 7 to 8 months for full-time entry after second birth and part-time entry after first birth. Austria Lalive and Parental Women’s Treated mothers were significantly less likely Linear Reform increasing the duration of Zweimüller leave return to than control group mothers to have returned regressions/ parental leave from one year to two (2009) work to work three years after giving birth and the linear years significantly reduced mothers’ difference was quantitatively large: An ad- probability return to work. Levels of employment ditional 10 of 100 mothers had not returned models and earnings decreased in the short to work within three years after the 1990 birth run but not in the long run. (reform year). This difference in return to work shrinks over time but a significant three-percentage point difference still remained even after ten years. Although treated mothers worked about one month per year less during the first three years after giving birth, there were no long- run employment differences between treated and controls. During months 37–120 after birth, average employment was the same for the two groups. Treated mothers earned about three euros less (than control group mothers) from work- ing on the average day of the three first post birth years. 18 The table does not fully reflect the findings of each study. Please consult the studies for a comprehensive account of each intervention’s impact. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 46 Germany Ludsteck Parental Women’s Reform 1 increased both the job protec- Difference- Major expansions in leave coverage and leave return to tion and maternity benefit periods from 2 in-difference can induce women to delay their re- Schonberg work to 6 months after childbirth. This lowered design turn to work and can reduce post-birth (2014) the share of mothers who had returned to employment rates among women in Women’s work by the third month after childbirth by the short run but not in the long run. post-birth 30.5 percentage points. Reforms 2 and 3 employment increased the job protection period from 6 to 10 months and the maternity benefit period from 6 to 10 and from 6 to 22 months after childbirth, respectively. This led to a decline of similar magnitude in the probabil- ity that the mother had reentered the labor market by month 7 after childbirth (27.5% and 30.3%). Reform 4 extended the job protection period from 18 to 36 months and left the maternity benefit period unchanged at 18 months after childbirth. This caused a reduction in the share of mothers who had returned to work by month 19 after childbirth of 9.6 percentage points. Reform 5 raised the maternity benefit period from 18 to 24 months and left the job protection period unchanged at 36 months after childbirth. This led to a somewhat lower reduction in the share of mothers who had gone back to work by month 24 after childbirth (5.8%).  In the long run, reforms 1, 2, 4, and 5 slightly reduced the probability that the mother had gone back to work by months 52 and 76 after childbirth by about 1–2 percentage points. Reform 3, which extended the maternity benefit period beyond the job protection period, had a stronger negative impact on the mother’s return decision and lowered the probability that she had reentered the labor market by months 52 and 76 after childbirth by 4.5 and 3.8 percentage points, respec- tively. Reform 1 lowered the number of months the mother has worked through month 7 after childbirth by 1.091 months, compared to only 0.495 month through month 76 after child- birth. Reforms 2, 4, and 5 led to a somewhat larger decline in maternal employment 76 months after childbirth of 0.951 month, 1.101 months, and 1.094 months, respectively. Reform 3 caused the strongest reduction in maternal employment: the number of months the mother has been employed through the first 16 and 76 months after childbirth decreased by 1.548 and 2.839 months, respectively. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 47 Canada Hanratty Parental Women’s The expansion in paid family leave from 25 Log-linear The expansion in Canadian paid and leave return to to 50 weeks was associated with a decrease model/ family leave was associated with Trzcinski work in returns to work (for mothers) of approxi- difference- decreases in returns to work in the (2009) mately 20% both 7 and 12 months after birth, in-difference year after birth. This change was or 60% and 40%, respectively of base level design concentrated among economically returns. advantaged women. The share of women returning to work within 1 year decreased by 12 percentage points for single parents compared to 21 percent- age points for two-parent families, by 12 percentage points for women with non-wage incomes under $20,000 compared to 21 to 25 percentage points for higher income wom- en, and by 5 percentage points for women without a high school degree, compared to 19 to 25 percentage points for higher skilled women. 9 EU Ruhm Parental Female Periods of paid entitlement around three Difference- Paid parental leave of short periods countries (1998) leave employment months led to a 3 to 4% rise in female em- in- difference- increased female employment rates ployment rates, with little effect on wages. in- difference and had no effect on wages while lon- Female Longer entitlements of more than nine model ger periods had negligible additional wages months led to negligible additional impact on impacts on women’s employment but employment but sizeable negative impacts of sizable negative ones on their wages. about 3% on female wages 53 Amin et al. Leave Proportion An increase of 6.8 percentage points in the Ordinary There was a positive correlation be- developing 2016 available to of female share of women workers for a typical firm least squares  tween mandated paternity leave and countries fathers workers in (against the mean of 32%) was associated regressions female employment in private firms. the private with mandating paternity leave. sector OECD Blau and Parental Female Lack of parental leave in the US led to the Linear countries Kahn (2013) leave labor force drop in the US’s rank from sixth among the regression Parental leave policies were associ- participation 22 OECD countries in 1990 to seventeenth analyses ated with a higher increase in female in 2010 in terms of female labor participation. labor force participation in most The expansion of family-friendly policies in OECD countries. The lack of such many OECD countries explained 29% of the policies explained the US’s relative decrease in US women's labor force partici- deterioration of women’s labor force pation relative to these countries. participation. Quebec/ Patnaik Parental Women’s Conditional on being employed, mothers Sharp Mothers experienced impressive Canada (2019) leave (seek- employment physically spent 80 minutes more at the regression gains when father’s participation in ing to boost entry workplace (increase of 44% from baseline) discontinuity parental leave was boosted. A clear fathers’ par- and were 5.4 percentage points more likely design pattern of reduced sex specialization ticipation) to be employed full-time (increase of 7% emerged as fathers’ participation in from baseline) compared with mothers parental leave increased following the who experienced their last birth prior to the reform. reform. The exposed fathers spent 37 minutes longer in nonmarket work per day (increase of 23% from baseline). Norway Cools et al. Paternity Women’s No effect of the paternal quota reform on Difference- The paternity leave reform had (2015) leave quota employment fathers' work hours and yearly earnings. No in-difference little if any effect on gender gaps in (leave indication of an increase in mothers' labor model wages and employment. 4 weeks of available to Fathers’ market attachment. paternity leave may not be considered fathers) work enough time to really impact long run hours and outcomes. There is no evidence that earnings paternity leave counters the traditional allocation of parents’ labor supply. Norway Rege and Paternity Mothers’ No impact Difference-in- No evidence that the paternity-leave Solli (2013) leave quota labor supply differences quota affected mothers’ labor supply. model EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 48 Sweden Ekberg et Paternity Fathers The share of fathers taking no days of Ordinary There is no evidence for substantial al. (2013) leave quota taking parental leave decreased from 54% to 18% least squares labor market effects of the reform for parental due to the reform. The share of fathers using regression mothers or fathers. leave approximately 1 month of parental leave increased from 9% to 47%. The average Fathers amount of parental leave taken by fathers participating increased by roughly 50%. in childcare No evidence that fathers who took more Mothers’ parental leave increased their share in and fathers’ childcare. market outcomes No strong effects for mothers' or fathers’ employment. No significant wage effect for mothers or fathers. Russia Ogloblin Job Gender Job segregation by gender was responsible Heckman There was a negative correlation (2005) segregation wage for 76.3% of the gross differential in monthly model/ two- between restrictions on women’s by gender differential earnings and 73.0% of the gross differential step method/ access to certain jobs and the gender in the effective long-run wages. Oaxaca- wage differential in Russia. The effect Blinder- was more sizeable accounting for the Unequal Accounting for the difference in retirement Neumark difference in retirement ages between retirement ages between men and women increased method men and women. ages the ratio of female to male lifetime earnings per month of work to 78%. Taiwan Zveglich & Working- Women’s A 6.1% decline in actual hours worked due to Difference- Policies designed to protect female Rodgers hours employment the legal restrictions, which corresponded to in-difference workers had controversial effects (2003) restrictions a decline in an average workweek of about model on labor market outcomes. Taiwan’s 2.5 hours. 1984 Labor Standards Law, contain- ing provisions that restricted women’s working hours, had a negative impact on women's actual hours worked and employment. 94 Islam et. al. Legal gender Probability Laws allowing women to work the same Probit and The legal systems that allow women economies (2018) disparities in: of having a night hours as men were associated with ordinary to work during night hours in the female firm an increase of 4.1 percentage points in the least squares same way as men, have access to Night-hours owner probability of having a female top manager regressions property and finance, and move freely work (significant at the 1% level). are positively correlated with the Probability likelihood of women owning firms and Ownership of having a Laws granting married men and women being top managers. rights female top equal ownership rights were associated with manager an increase of 4.4 percentage points (signifi- Obtaining cant at the 1% level) in the probability of hav- finance ing a female top manager and an increase of 6.9 percentage points (significant at the 5% Registering a level) for having a female firm owner. business Laws prohibiting discrimination by creditors Mobility against women in access to finance was rights associated with an increase of 4.1 percent- age points (significant at the 5% level) in the probability of having a female firm owner compared with an increase of 2.4 percentage points (significant at the 10% level) for hav- ing a female top manager. The ability to register a business in the same way as men was associated with an increase of 13.3 percentage points (significant at the 1% level) in the probability of having a female firm owner. The ability to travel outside the home was associated with an increase of 24.3 percent- age points (significant at the 1% level) in the probability of having a female firm owner. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 49 98 Demirguc- Legal Women’s In countries where married women face no Multivariate In countries where women faced legal developing Kunt, restrictions financial legal restrictions to work outside the home, regression restrictions in their ability to work, countries Klapper, on women to inclusion women were 16.2% more likely to own a analysis/ head a household, make decisions & Singer pursue a job, (owning bank account and 25.4% more likely to have probit model/ without having to obey the husband, (2013) be head of a bank savings at a formal institution. multinomial choose where to live, and receive household, account, logit model/ inheritance, women were less likely to make having In countries where women could be head of ordinary own an account, relative to men, as decisions savings at household, women were 9.6% more likely least squares well as to save and borrow. without a formal to own a bank account, 23.8% more likely regressions having to institution, to have savings at a formal institution, and The level of violence against women obey the accessing 7.8% less likely to use informal credit. and the incidence of early marriage husband, credit) for women, contributed to explaining choose In countries where married women could the variation in the use of financial where to live, choose where to live, women were 13.6% services between men and women. and receive more likely to own a bank account and inheritance 26.9% more likely to have savings at a In economies where there was a formal institution. high prevalence of violence against Prevalence women and child marriage, women of violence In countries where married women were were less likely to have formal ac- against required by the law to obey their husbands, counts, savings, or credit. women women were 20.9% less likely to own a and child bank account and 10.4% more likely to use marriage informal credit. In countries where women (daughters and surviving spouses) had equal inheritance rights, they were 8.9% and 8.7% (respec- tively) more likely to own a bank account. In countries with high prevalence of violence against women (measured by the existence of specific legislation and the percentage of women being beaten by their partners), women were 19% less likely to own a bank account, 64.5% less likely to have savings at a formal institution, and 37.8% less likely to have access to credit. 141 Morsy and Women’s Financial Female-to-male ratio reflecting the gender Weighted Women’s access to equal property economies Youssef equal access inclusion inclusion gap in the use of financial services least squares rights relative to men was key to their (2017) to property (holding a in general dropped by 30.7 percentage model access to finance. rights debit card, points when women were not guaranteed the holding a same property rights as men. It dropped by credit card) 66.3 percentage points for holding a credit card and 33.3 percentage points for holding a debit card. Ethiopia Persha et Land Access to Land certification in Ethiopia was associ- Difference- Land certification played an important al. (2017) certification credit ated with a 10% increase in the likelihood of in-Difference role in promoting women’s access to (women’s’ households (both male-headed and female- design credit. access to headed) taking out any credit for farming pur- immovable poses, and a small increase in the average property) amount of credit obtained. The land certification was also shown to have led to an 11% average increase in the likelihood of female-headed households (and a 12% average increase in the likelihood of widows) feeling more secure about entering into credit-based business transactions when the transactions occurred with a holder of a land certificate. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 50 India (West Santos et Land- Access to The average household participating in the Propensity Including women’s names on the land Bengal) al. (2014) allocation credit program was 12% more likely to report taking score- documents improved women’s tenure and out a loan from a formal bank and 88% more weighted security and their ability to influence registration likely to use a loan for agricultural purposes regression household decisions. program than a household not in the program. models Women in households within the program were: (1) 12% more likely to be involved in decisions to take loans from self-help groups or microfinance institutions; (2) 12% more likely to be involved in decisions on whether to purchase productive assets; (3) 9% more likely to be involved in decisions related to food purchase and consumption; and (4) more likely to be involved in decisions about the family land. The share of the family land over which they were involved in decisions increased by 15% for how to use the land, 14% for what to grow on it, and 11% for whether to sell produce from it 100 Hallward- Women’s Women’s An increase of income of one standard Panel regres- Equal inheritance rights were as- countries Driemeier et inheritance labor force deviation would raise women’s labor force sion/ Lin- sociated with higher women’s labor al. (2013). rights participation participation 2.1 percentage points higher ear regres- force participation and girls’ education for countries with equal inheritance for wives sion model although effects on countries varied Women’s Girls’ and husbands than countries where wives do by level of income. ability education not have the same inheritance rights. to make Women’s ability to be head of house- decisions Infant At the 25th percentile of income, coun- hold was associated with a higher (including mortality tries that gave sons and daughters equal female labor force participation and the ability to inheritance rights had on average a 6.4 lower infant mortality. open a bank percentage point higher ratio of girls to boys account and in primary enrollment. Women’s ability to open a bank ac- be head of count without her husband’s permis- household) For countries at the 75% percentile of sion was associated with a higher income, those that did not require husband’s female labor force participation. permission for wives to open bank accounts on average had women’s labor force partici- pation that was 2.62% higher. The absence of gender restrictions on the head of household status was significantly associated with women being more likely to participate in the labor force and move out of agricultural employment in favor of wage employment; regression coefficients were not reported. At the 25th percentile of income, grant- ing women the same legal right as men to be head of household was associated on average with a decline in infant mortality of 9.3 deaths per 1000 live births; at the 10th percentile the decline was 12.3. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 51 Ethiopia Hallward- Family law Women’s A relative increase in women’s labor force Panel Where the reform had been enacted, Driemeier reform employment participation in reforming areas, 15–24% regressions women were more likely to work & Gajigo, (removing participation higher than in non-reforming areas. outside the home, work in paid and (2015) the ability of full-time jobs, and in occupations that a spouse to employed more educated workers. deny permis- sion for the Access to property rights, including other to equal inheritance rights, shifted the work outside intra-household power dynamic in the home, favor of women, allowing them to introducing make decisions for themselves to the require- work outside the home and providing ment of both them with the necessary collateral to spouses' obtain loans. consent in the admin- istration of marital property, and raising women’s minimum age of marriage) Nepal Mishra and Women’s Women’s Women who owned land were 13.91 and Probit Land ownership played an impor- Sam (2016) land decision- 38.3 percentage points more likely to have (uncorrected tant role in empowering women, by ownership making the final say in their healthcare decisions. endogeneity) enhancing their household bargaining capacity and CMP-IV power in the areas of own healthcare, The probability of women having the final (corrected major household purchases, and visit- say in decisions regarding visits to family or endogeneity) ing family or relatives. relatives was 9.95 percentage points higher models when they owned land. Tanzania Peterman A set of legal Wom- In a scenario in which there were no im- Linear Improving women’s property and (2011) reforms en's em- provements in WPIR, the predicted average probability inheritance rights were significantly improving ployment rate of employment outside the home was models / associated with women's employment women’s outside 8.83 percent, as compared to a predicted ordinary outside the home, self-employment, property and the home, average rate of 24.18 under a scenario in least squares and earnings. Property rights mat- inheritance self-employ- which all communities moved from low to regressions/ tered significantly more for poor or rights (WPIR) ment, and high WPIR. Inverse uneducated women. earnings probability The rate of self-employment under improve- weighting ments in WPIR was 2.1 times the rate under no improvements, and the gross earnings with improvements were 3.80 times the level without improvements. In communities with high WPIR, women’s predicted average individual savings were 1.35 times the predicted average amount in communities with low WPIR. India Deininger et Equal Access to Females whose fathers died after the 1994 Linear Granting daughters and sons equal (states of al. (2010) inheritance land reform were 22 percentage points more likely probability inheritance rights was positively Maharash- rights for to inherit land than those whose fathers died model associated with women’s asset tra and daughters Age of before 1994. endowment and their socio-economic Karnataka) marriage outcomes, namely greater access Females who married after 1994, but whose to land, higher age of marriage, and Education fathers died after 1994 had a significantly higher educational attainment. higher age at marriage, by 0.54 years, than females who married before 1994, reduc- ing the pre-reform average marriage age difference between men and women of 3.36 years. Girls who started their elementary education after the amendment had 0.3 years more elementary schooling than the older cohort. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 52 India Heath and Equal Decision The Act led to statistically significant increas- Ordinary The Hindu Succession Act reform Tan (2018) inheritance making es of 6.6 percentage points in the probability least squares positively affected female labor supply rights for capacity that a woman had a say in household deci- regression through the channel of “autonomy”, daughters sions, 8.2 percentage points in the probability as proxied by the woman’s say in Freedom of that a woman could go to the market alone, household decision-making, freedom movement 6.9 percentage points in the probability that of movement, and access to a bank a woman could go to a health facility alone, account. Access 8.3 percentage points in the probability that a to a bank woman could travel outside the village alone, account and 3.9 percentage points in the probability a woman had access to a bank account. Regular and paid The likelihood of a woman working all year employment instead of occasionally increased 5.7 per- centage points, and to work for cash instead of no pay or for in kind payment, 6.1 percent- age points. India Rosenblum Equal Female The reform was estimated to cause a 0.32 Difference- India’s inheritance reform was as- (2015)  inheritance infant percentage point increase in a girl’s prob- in-difference: sociated with higher female infant rights for mortality ability of death. ordinary mortality and feticide; the increased daughters and feticide least squares cost of daughters took away some of quadruple the economic benefits of sons, and differences gave parents an incentive to decrease linear their investment in the health of their regression/ daughters or reduce their own fertility. Cox proportional hazard model United Ladd (1982) Prohibiting Women’s In California, data showed that in 1977 fe- Multinomial Legislation prohibiting gender-based States discrimina- access to male applicants were charged interest rates logit models discrimination in access to credit was (California tion based credit that were 0.02 to 0.04 percentage points associated with a positive impact on and New on sex and higher than those for comparable male- women’s access to finance. York) marital status female applicants (joint application). This in access to differential disappeared in 1978. credit 15 Savadogo Child Women’s Child marriage led to a loss in earnings for Various Child marriage had a direct impact countries and Wodon marriage earnings women in adulthood of nine%. regression on girls’ educational attainment and (2018) and labor techniques fertility, which in turn affected their force The gains in earnings from ending child with expected earnings and labor force participation marriage, considering only the women who instrumental participation. married early, ranged from 1.44% to 15.60% variables of baseline yearly earning spending on the country. The gains in earnings or expected productiv- ity for women as a whole ranged from 0.49% to 4.58% of base earnings depending on the country. When including men as well, the gains in the population’s earnings ranged from 0.17% to 1.68% of the wage bill. 171 Htun et al. Restrictions Women Restrictions on women’s legal capacity led to Multivariate Restrictions on women’s legal capac- countries (2019) on women’s owning a reduction of: regression ity including their ability to move were legal a bank 0.78 percentage points in the percentage of models a strong predictor of the share of capacity account women owning a bank account. women with bank accounts, the share (limiting of women who participated in firm their ability Owning a 2.24 percentage points in the percentage of ownership, and female labor force to make firm women participating in firm ownership. participation. decisions regarding Participating 2,97 points in female labor force participa- money, in the labor tion. travel, work, force property, and children) EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 53 143 Gonzales et Women’s Female The legal right for women to be head of Panel Legal capacity related provisions, in- countries al. (2015) legal labor force household coupled with more equal inheri- regressions cluding allowing women to be head of capacity participation tance rights were associated with a decline household, obtain a job, open a bank in gender gaps in labor force participation account, sign a contract and register in developing and emerging economies of a business in the same way as men, around 4.6 percentage points. were associated with improved fe- male labor force participation. The removal of legal restrictions on married women to access property, open a bank account, sign a contract, initiate legal pro- ceedings without their husband’s permission, get a job, be head of household, individually led to a reduction of between around 2 and 3 percentage points in the gender gap in labor force participation. United Gray (1998) Unilateral Women’s The probability a wife works in the paid labor Difference- The possibility for women to make a States divorce laws labor force force increased by 2.3% with residence in in-difference, divorce decision unilaterally, on the participation a unilateral-divorce state with community probit, and same grounds as men, increased property regime. bivariate their labor supply in states with probit models community-property jurisdictions. United Stevenson Unilateral Domestic In states that introduced unilateral divorce Ordinary least The adoption of unilateral divorce States and Wolfers divorce laws violence a decline of 8 –16% in female suicide was squaresre- laws was associated with a large (2006) found, roughly a 30% decline in domestic gressions/ decline in domestic violence for both violence for both men and women, and a difference-in- men and women. 10% decline in females murdered by their difference partners. Middle Majbouri Per capita Women’s As rents rise (fall) by hundred per cent -that Simple linear When laws are gender equal, rents East and (2017) oil and gas economic is, they double, in a fully Muslim country regression are less likely to affect women’s North Africa rents/ participation with Islamic family law, FLFP decreases economic participation despite the (MENA) Family law (increases) by about 0.07 percentage points, increase in male bargaining power. and in a fully non-Muslim one, it does not change. India Field et al. Access Women’s Women who were provided bank accounts Double-lasso  Having access to bank accounts and (2019) to bank employment and training on how to use them and had regressions using them to deposit earned income, accounts their wages directly deposited into those ac- rural women in India increased their counts significantly increased labor supply by labor supply in the public and private 0.16 standard deviation units one year after sectors. the treatment. Labor supply significantly increased by 0.21 standard deviation units among socially con- strained women whose husbands has been more likely to subscribe to norms against female work. United Dugan Domestic Occurrence The odds that households in states with Logistic Domestic violence laws reduced the States (2003) violence laws of domestic mandatory arrest laws would suffer from regressions chances of domestic violence and violence spousal violence were 0.885 lower than may have deterred potential offenders households in other states. from harming family and partners as well. The statute with the strongest apparent impact on reducing violence made protection Police officers were more likely to ar- order violation a felony offense. The odds rest a perpetrator if domestic violence of family violence were 0.62 lower than in was legally established as a serious states where protection order violation was offense, but this may also have low- not a felony offense). The odds for non-mari- ered the level of reporting. tal intimate violence was 0.653 lower. The odds that officers in states with felony statutes were involved were 1.59 higher than officers in states without the statute.  The odds that officers were involved in states with mandatory arrest laws were 0.875 lower than in other states. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 54 United Cook et al. Mandatory Occurrence Starting at around 4.75% unemployment, Negative Mandatory arrest laws effective to States (2019) arrest laws of domestic mandatory arrest was seen to begin to binomial mitigate intimate partner violence re- (in case of violence suppress intimidation incidents with an in- regression sulting from rising rates of unemploy- domestic creasing effect. However, as unemployment model ment. However, the mitigating effects violence) exceeded 8.5%, the mitigating effect of the of such laws disappeared under policy disappeared. severe economic crisis. United Iyengar Mandatory Intimate Reporting declined by 4.5 percentage points Difference-in- Mandatory arrest laws increased states (2009) arrest laws partner in mandatory arrest states (about 12%) and difference intimate partner homicides in the US homicide 2.8 in recommended arrest law states (about due to decreased reporting by victims, 7%). as victims were less willing to report an incident if their abuser would be Mandatory arrest laws were responsible arrested. for an additional 0.8 murders per 100,000 people. This corresponds to a 54% increase in intimate partner homicides. 53 Amin & Antidis- Women’s Mandating a non-discrimination law is as- Ordinary least There was a strong positive relation- developing Islam (2016) crimination employment sociated with an increase of 5 percentage squares re- ship between the existence of a legal countries legislation entry points in women’s employment. gressions provision prohibiting gender-based (employ- discrimination in hiring practices and ment) women’s employment relative to men. Britain Zabalza and Antidis- Women’s Due to the anti-discrimination legislation Linear Antidiscrimination legislation had a Tzannatos crimination employment in the UK, relative employment of women regressions positive and significant association (1985) legislation and pay increased by more than 17% and relative pay with both relative earnings and rela- (Equal pay by about 15%. tive employment of women in the UK. for work of equal value and equal employment opportunities for men and women) United Eberts Equal Female In the early 1970s female teachers were Multivariate The Equal Employment Opportunity States and Stone Employment teachers’ significantly less likely to be promoted than logit model Act contributed to the decline of dis- (Oregon (1985) Opportunity promotion similarly qualified male teachers. After the crimination in promotion and apparent and New Act to admin- adoption of the legislation the difference was discrimination present in the early York) istrative no longer significant. 1970s. positions Index measures of apparent discrimination declined significantly by close to half. United Neumark Equal Pay Women’s Employment rate of black females fell by Difference-in- Sex discrimination/equal pay laws States & Stock Act and employment .032 due to sex discrimination/equal pay difference-in- increased the relative earnings of (2001) Equal and pay legislation. difference black and white women, but at the Employment same time reduced their relative Opportunity Employment rate of white females fell by employment. Act .039 due to sex discrimination legislation. Earnings of black females increased by 4.1% due to sex discrimination/equal pay legislation. Earnings of white females increased by 1.2% in non-south states due to sex discrimination law. China Zhao and Unequal Women’s About three-quarters of the deficit in Oaxaca– Differences in retirement ages be- Zhao (2018) retirement pension women’s pensions was explained by their Blinder-style tween men and women affected the ages lower likelihood of receiving occupational decomposi- period they spent in the job market pensions, and one-third was due to smaller tion and therefore the size of their respec- benefits when they did receive them. Among tive pensions. pension recipients, nearly one-third of the Ordinary least gap was explained by women’s fewer years squares and of employment and lower salaries. probit regres- sions EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 55 China Lee et al. Unequal Welfare Using the consumption equivalent varia- Linear China’s mandatory retirement law re- (2019). retirement gains for tion (CEV), on average men and women regressions quiring women to retire 5 years earlier ages women and who were born in the benchmark economy than men reduced welfare for women. men had to be compensated 2.21% and 1.93% Equalizing retirement ages for men more consumption in order to be indifferent and women could generate a welfare to those men and women who are born in gain for both. the economy with the equal retirement age policy. (CEV measures how much consumption has to be compensated to make the agents in the counter-factual model indifferent from the agents in the benchmark model.) Austria Staubli and Early Female Raising the ERA increased employment by Ordinary Increases in retirement ages for Zweimüller retirement labor supply 9.75 percentage points among affected men least squares women are associated with increased (2013) age (ERA) and by 11 percentage points among affected regressions female labor supply. increase women. Australia Atalay and Women’s Female The increase in the Australian Age Pen- Difference-in- Increases in retirement ages for wom- Barrett retirement labor supply sion increased the probability of labor force difference en were associated with increased (2015) age increase participation (LFP) by 7.7 percentage points. female labor supply. Given that an estimated 62.5% of the treated Linear cohort would receive the age pension if age probability eligible, the treatment effect conditional on model meeting the means test is a 12.3 percentage point increase in LFP. The reform increased the LFP of single women by 13.5 percentage points. With ap- proximately 70% predicted to receive the age pension if age eligible, the estimated impact conditional on satisfying the means test is a 19.4 percentage point increase in LFP. United Cribb et al. Early Female Increase in the ERA led to female employ- Multinomial Increases in retirement ages for Kingdom 2016 retirement labor supply ment at ages 60 and 61 rising by 6.3 per- probit models/ women were shown to be associated age (ERA) centage points. Ordinary least with increased female labor supply. increase squares/ Linear probability models Switzerland Lalive and Equal Women’s One-year increase in full retirement age Regression Raising women’s retirement age de- Staubli retirement labor market delayed market exit by 0.51- 0.66 years and discontinuity layed their exit from the labor market. (2014) ages exit increased the claiming age of retirement design (RDD) benefits by 0.41-0.55 years. EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 56