SI 2020 Gender in the Economy

An NBER conference on Summer Institute 2020 Gender in the Economy took place online July 24-25. Research Associates Jessica Goldberg of University of Maryland, Claudia Goldin of Harvard University, Seema Jayachandran of Northwestern University, Claudia Olivetti of Dartmouth College, and Tom Vogl of University of California, San Diego and NBER organized the meeting, sponsored by Bill & Melinda Gates Foundation. These researchers' papers were presented and discussed:

Papers on Women and Household Finance Issues

Francesco D'Acunto, Boston College; Ulrike Malmendier, University of California, Berkeley and NBER; and Michael Weber, University of Chicago and NBER

"Gender Roles and the Gender Expectations Gap" (NBER Working Paper 26837)

Expectations about macro-finance variables, such as inflation, vary significantly across genders, even within the same household. D’Acunto, Malmendier, and Weber conjecture that traditional gender roles expose women and men to different economic signals in their daily lives, which in turn produce systematic variation in expectations. Using unique data on the contributions of men and women to household grocery chores, their resulting exposure to price signals, and their inflation expectations, the researchers show that the gender expectations gap is tightly linked to participation in grocery shopping. They also document a gender gap in other economic expectations and discuss how it might affect economic choices.

Simone G. Schaner, University of Southern California and NBER; Erica M. Field, Duke University and NBER; Rohini Pande, Yale University and NBER; Natalia Rigol, Harvard University and NBER; and Charity M. Troyer Moore, Yale University

"On Her Own Account: How Strengthening Women's Financial Control Impacts Labor Supply and Gender Norms" (NBER Working Paper 26294)

Do consumers show a strong bias toward low deductible insurance plans, as many field studies imply? Schaner, Field, Pande, Rigol, and Troyer Moore report on a controlled experiment intended to see whether subjects have a predisposition toward such plans and whether that preference is consistent when their default plan and premiums are changed. Subjects were presented with a scenario where they had to make a decision on whether to purchase a plan with a low deductible (LD) or high deductible (HD) when faced with an illness having a specified probability and cost. Participants had to choose between these plans in two rounds with the identical risk of an illness and specified premiums. If their default option was an LD plan in Round 1, then it was an HD plan in Round 2. The experiment did not show a strong bias toward low deductible health plans. Only slightly more than half of the respondents chose an LD plan even when it was optimal for them to do so. When faced with a default option that was switched in Round 2, 58% of the respondents chose the same plan as they did in Round 1, implying that some but not all subjects resisted the default option in their decision process. Subject choices were correlated with their responses to questions about risk aversion and a desire for peace of mind.

Mention of Three Additional Papers on Women and Household Finance Issues

Emma Riley, University of Oxford

"Resisting Social Pressure in the Household Using Mobile Money: Experimental Evidence on Microenterprise Investment in Uganda"

Riley examines whether changing the way microfinance loans are disbursed to utilize widespread mobile money services impacts the businesses of female microfinance borrowers. Using a field experiment of 3,000 borrowers of BRAC Uganda, Riley compares disbursement of a loan as cash to disbursement of a loan onto a mobile money account. After 8 months, women who received their microfinance loan on the mobile money account had 15% higher business profits and 11% higher levels of business capital. Impacts were greatest for women who experienced pressure to share money with others in the household at baseline, suggesting that providing the loan in a private account gives women more control over how the loan is used.

Luigi Guiso and Luana Zaccaria, Einaudi Institute for Economics and Finance

"From Patriarchy to Partnership: Gender Equality and Household Finance" (slides)

Using Italian survey data, Guiso and Zaccaria document a marked shift in household financial decision-making power from men to women. The share of wife-headed households increased from almost zero in the early 1990s to over 35% in recent years. This reflects a slow but steady social norm transformation that changed family governance from a patriarchal system to an egalitarian one. The researchers use the variation of social norms across cohorts and regions to identify the effects of gender equality on households' financial decisions. They find that less male-biased norms have a positive effect on households' participation in financial markets, equity holdings, and asset diversification -- these effects are stronger when the benefits from information and cost-sharing between spouses are larger. Importantly, equality increases returns from financial investments. Taken all together, this evidence suggests that gender roles in household financial management can have large economic costs. Consistent with this, the researchers show that the patriarchal system began to be abandoned when a pension reform in the early 1990s made it too costly for the younger cohorts.

Raimundo Undurraga and Ana Maria Montoya, Universidad de Chile; Eric Parrado, New York University; and Alex Solis, Uppsala University

"Bad Taste: Gender Discrimination in the Consumer Credit Market" (slides)

When evaluating observably similar loan applications from men and women, do loan officers favor men? Undurraga, Parrado, and Montoya randomly assigned loan requests of variable amounts to a balanced sample of male and female prospective borrowers who then submitted the assigned loan requests to randomly assigned loan officers in Chile. It is found that loan requests submitted by women are 18.3 percent less likely to be approved. The researchers explore the presence of taste-based mechanisms by eliciting gender preferences among loan officers and show that discrimination against female applicants is 32% larger among pro-male officers, with most of the effect driven by male pro-male officers. Official statistics show that women in Chile have higher repayment rates than men. Hence, an alternative hypothesis is that gender differences in approval rates are due to inaccurate statistical discrimination, i.e., loan officers discriminate female applicants not because they have animus against them but because have biased beliefs about their repayment capacity. The researchers experimentally test for this by implementing an information treatment aimed at "correcting" loan officers' biased beliefs about women's repayment rate. The treatment is found to be ineffective, reinforcing the hypothesis of taste-based discrimination.

Victimization, Gender, and Covid-19

Heidi Stöckl, London School of Hygiene and Tropical Medicine, and Gerry Mshana, National Institute for Medical Research, Tanzania

"The Effect of COVID-19 on Women, Livelihood, and Violence in Mwanza, Tanzania"

Across the world, reports of increases in intimate partner violence due to COVID-19 related lockdowns or social restriction measures are rising, with the fear that the predicted recession and loss in economic livelihood will lead to further increases. To curb the spread of COVID-19, the Tanzanian government took several measures including closing schools and colleges, banning large gatherings, restricting travel from affected countries and recommending hand washing and social distancing throughout the country. To the researchers’ knowledge, there is no published evidence on how COVID-19 and the control measures impact women’s physical, psychological and economic well-being, their relationships including intimate partner violence, and their children in low- and middle-income countries. Tanzania already has a comparatively high rate of intimate partner violence, with one in four women reporting physical and/or sexual violence in the last year (Kapiga et al. 2018), and the majority of people are employed in the informal sector. Tanzania is therefore an important setting in which to investigate and monitor the effect of COVID-19 control measures on women’s physical, psychological and economic well-being, and their relationships including their experiences of intimate partner violence to advise future policy directions and interventions. Stöckl and Mshana’s study aims to examine this by conducting a two-wave quantitative phone survey of 421 women and two in-depth phone interviews with 15 women currently involved in a longitudinal study on predictors and consequences of intimate partner violence in Mwanza, Tanzania.

Amalia R. Miller, University of Virginia, and Carmit Segal, University of Zurich

"Effects of the COVID-19 Pandemic on Domestic Violence in US Cities"

Around the world, news outlets have reported increases in domestic violence (DV) as a result of the COVID-19 pandemic and the attendant restrictions on individual mobility and commercial activity. Although there are several reasons to expect that DV would increase for married and cohabiting couples, there are also forces that could reduce DV, such as COVID-19 sickness in a household or less time spent together by couples that live apart. It is also likely that victims who wish to report their abusers or seek assistance will face further barriers during the pandemic, which would depress reporting rates. The effects of the pandemic on both true and reported incidences of DV are therefore ambiguous. Miller and Segal empirically examine the effects of the COVID-19 pandemic on DV and reported DV. They initially focus on reported DV using daily, city-level data from individual police departments in major US cities. Starting with real-time police data allows them to track the effects of the pandemic on cases reported to police within a relatively short timeframe. This can help inform public policy responses to the ongoing pandemic. The high frequency of the data will also be useful for tracking the dynamics of reporting over the course of the pandemic. For example, one may expect a decline in reporting during stay-at-home orders that is followed by an increase when restrictions are lifted. Next, the researchers integrate additional data sources to the analysis that may shed light on DV reporting as they become available.

Sonia R. Bhalotra, University of Essex; Emilia Brito Rebolledo, Brown University; Damian Clarke, University of Chile; Pilar Larroulet, University of Maryland; and Francisco Pino, University of Chile

"COVID-19 and Domestic Violence – Evidence from Rolling Quarantines in Chile"

COVID-19 has led to a surge in domestic violence (DV). It is unclear whether this results from income shortfalls and income uncertainty, or from families being locked down together and distanced from their social networks. Identifying causes of domestic violence has always been hard, as systematic data and relevant natural experiments are scarce. Policy responses to COVID-19 provide an opportunity to isolate the underlying mechanisms. Chile has implemented fine-grained rolling quarantines, such that different neighborhoods within a city have been under quarantine at different times. Quarantines have covered at least half the population, and in some cases lasted more than 100 days. Bhalotra, Rebolledo, Clarke, Larroulet, and Pino gather rich administrative data covering calls to state-sponsored hotlines and to new support channels such as ‘silent’ WhatsApp lines as well as police records. The data show that calls to the national DV hotline have trebled, while formal crime reporting has tended to fall. This suggests a worrying situation in which just as abuse has spiked, avenues for formal redress have become less accessible. The researchers propose primarily to use the rolling nature of quarantines and geographic dispersion in intensity of COVID-19 contagion to examine the causal effect of COVID-19 related lockdown on different measures of DV exposure and reporting, characterizing their dynamic path during the pandemic and lockdown period conditional on seasonal/secular trends. The researchers will then attempt to illuminate the relative contributions of economic stress and confinement. They will also be able to comment on the effectiveness of newly designed services to encourage reporting during the crisis.

Keith Finlay, U.S. Census Bureau; Michael G. Mueller-Smith, University of Michigan; and Brittany Street, University of Missouri

"The Determinants and Aftermath of Victimization in US Households and the Implications of COVID-19"

Estimates suggest nearly one-third of all women in the United States experience domestic violence in their lifetime, with higher rates among low-income and minority women (Smith et al. 2017; Currie et al. 2018). Due to a variety of data constraints, it is unclear how domestic violence events fit into the broader context of the lives of women in the United States. These issues are particularly relevant amid the COVID-19 pandemic, as many women and children may be forced to shelter-in-place in undesirable, vulnerable arrangements and face heightened barriers to economic self-sufficiency. Finlay, Mueller-Smith, and Street measure domestic violence events through the National Crime Victimization Survey (NCVS) and the newly created Criminal Justice Administrative Records System (CJARS). These data, along with longitudinal residential and family relation crosswalks built as part of the CJARS project, allow us to identify victims and offenders from survey and administrative records through dwelling-level cohabitation histories. These data are linked at the person level with extensive survey and administrative microdata held by the US Census Bureau to understand how earnings, benefit receipt, family structure, and cohabitation evolve leading up to domestic victimization and in its aftermath. Additionally, the combination of victimization measures based on surveys and administrative criminal justice records, e.g. arrests and charges, allow for exploration of known reporting hurdles specific to domestic victimization that may be exasperated during this pandemic. To understand the potential impacts of shelter-in-place orders, the researchers also estimate the number of individuals at increased risk of domestic victimization using geographic locations, cohabitations, and past victimization histories.

Papers on Victimization, Vulnerability, and Violence against Women

Eleonora Guarnieri, Ifo Institute Munich, and Ana Tur-Prats, University of California, Merced

"Cultural Distance and Conflict-Related Sexual Violence"

Guarnieri and Tur-Prats examine the role of ethnic-based gender norms in explaining the occurrence and intensity of sexual violence in conflict. They generate a novel dyadic dataset that contains information on the ethnic identity of the actors involved in 33 ethnic civil conflicts in Africa between 1989 and 2009 and their use of sexual violence. After exploiting ancestral economic, family, and societal arrangements, the researchers construct and validate an ethnic-based gender inequality index. They control for a large set of fixed effects and find empirical support for two interrelated hypotheses. First, gender-unequal armed actors are more likely to be perpetrators of sexual violence. Second, the researchers consider the perpetrator's gender norms relative to the victim's. Applying a gravity approach, the researchers find that sexual violence is driven by a specific clash of conceptions on the appropriate role of men and women in society: sexual violence increases when the perpetrator is more gender-unequal than the victim. Guarnieri and Tur-Prats show that (i) these patterns are specific to sexual violence and do not explain general violence within a conflict, (ii) differences in other dimensions of culture unrelated to gender do not explain conflict-related sexual violence.

Girija Borker, The World Bank

"Safety First: Perceived Risk of Street Harassment and Educational Choices of Women"

Borker examines the impact of perceived risk of street harassment on women's human capital attainment. Borker assembles a unique dataset that combines information on 4,000 students at the University of Delhi from a survey they designed and conducted, a mapping of potential travel routes to all colleges in the students' choice set using an algorithm developed in Google Maps, and crowd-sourced mobile application safety data. Using a random utility framework, it is estimated that women are willing to choose a college in the bottom half of the quality distribution over a college in the top quintile in order to travel by a route that is perceived to be one standard deviation (SD) safer. Furthermore, women are willing to spend INR 18,800 (USD 290) per year more than men for a route that is one SD safer -- an amount equal to double the average annual college tuition. These findings have implications for other economic decisions made by women. For example, it could help explain the low female labor force participation in India.

Roee Levy and Martin Mattsson, Yale University

"The Effects of Social Movements: Evidence from #MeToo"

Social movements are associated with large societal changes, but evidence on their causal effects is limited. Levy and Mattsson study the effect of the MeToo movement on a high-stakes decision -- reporting a sexual crime to the police. The researchers construct a new dataset of sexual and non-sexual crimes in 30 OECD countries, covering 88 percent of the OECD population. They analyze the effect of the MeToo movement by employing a triple-difference strategy over time, across countries, and between crime types. The movement increased reporting of sexual crimes by 13 percent during its first six months. The effect is persistent and lasts at least 15 months. Because a strong effect on reporting before any major changes to laws or policy took place is found, the effect can be attributed to a change in social norms. Using more detailed US data, the researchers show that the movement also increased arrests for sexual crimes in the long run. In contrast to a common criticism of the movement, they do not find evidence for large differences in the effect across racial and socioeconomic groups. The results suggest that social movements can rapidly change high-stakes personal decisions.

Mention of Three Additional Papers on Victimization, Vulnerability, and Violence

Ing-Haw Cheng, Dartmouth College, and Alice Hsiaw, Brandeis University

"Reporting Sexual Misconduct in the #MeToo Era"

What deters individuals from reporting sexual misconduct, and what are the effects of #MeToo for reporting? Cheng and Hsiaw show that individuals under-report sexual misconduct if and only if a manager's misconduct is widespread. The reason is that individuals face strategic uncertainty over whether others will also report misconduct and corroborate a pattern of behavior. The model is applied to study a manager's decision to mentor subordinates, the coordinating effect of raising public awareness of misconduct, and the policy effects of confidential holding tanks for reports and rewards for whistleblowers. Overall, the study highlights several unintended and intended consequences of #MeToo.

Andreas Kotsadam, University of Oslo, and Espen Villanger, CMI

"Jobs and Intimate Partner Violence – Evidence from a Field Experiment in Ethiopia"

Kotsadam and Villanger identify the effects of employment on Intimate Partner Violence (IPV) by collaborating with 27 large companies in Ethiopia to randomly assign jobs to equally qualified female applicants. The job offers increase formal employment, earnings, and earnings shares within couples in the short and medium run but relatively small effects in any direction on the main outcome, physical IPV, can be rejected. In the short run, job offers reduce emotional abuse and there are indications of heterogeneous effects whereby women with low bargaining power at baseline experience increased risks of abuse if offered a job.

Anke Becker, Harvard Business School

"On the Economic Origins of Restrictions on Women's Sexuality"

Becker studies the origins and function of customs aimed at restricting women's sexuality, such as a particularly invasive form of female genital cutting, restrictions on women's freedom of mobility, and norms about their sexual behavior. The analysis tests the anthropological theory that a particular form of pre-industrial subsistence - pastoralism - favored the adoption of such customs. Pastoralism was characterized by heightened paternity uncertainty due to frequent and often extended periods of male absence from the settlement, implying larger payoffs to imposing restrictions on women's sexuality. Using within-country variation across 500,000 women in 34 countries, Becker shows that women from historically more pastoral societies (i) are significantly more likely to have undergone infibulation, the most invasive form of female genital cutting, (ii) adhere to more restrictive norms about women's promiscuity, (iii) are more restricted in their freedom of mobility. Instrumental variable estimations that make use of the ecological determinants of pastoralism support a causal interpretation of the results. Becker further shows that the mechanism behind these patterns is indeed male absence, rather than male dominance, per se, or historical economic development.

Victimization, Gender, and Covid-19

Erica M. Field, Duke University and NBER, and Ursula T. Aldana, Institute for Peruvian Studies

"The Impact of COVID-19 on Intimate Partner Violence in Urban Peru"

There are several reasons health experts worldwide fear an increase in intimate partner violence (IPV) during the COVID-19 pandemic, including economic stress, greater physical proximity of family members, and limited access to support networks. Indeed, anecdotal evidence from several settings suggests increases in police and hotline reports of IPV. In Peru during the first week of the lockdown, the government’s domestic violence helpline registered a 30 percent increase in calls. However, higher reporting of IPV episodes may be driven by changes in reporting behavior rather than incidence of violence. For instance, because victims are isolated from friends and family, they may be more likely to seek help from hotlines. Hence, the degree to which the worldwide Covid-19 pandemic has increased rates of IPV remains an important empirical question. To shed light on this, Field and Aldana are working with the Peruvian Ministry of Women to conduct two phone-based surveys of 1800 women located in municipalities across the country and believed to be at heightened risk of IPV. The surveys measure IPV incidence at two points in time post-pandemic, along with changes in household resources and conflict. The analysis evaluates the causal effect of pandemic-related restrictions in economic and social activity on reported IPV by exploiting spatial variation in the degree of sheltering enforced by local governments. The study seeks to: (i) understand the causal impact of restrictions implemented to contain the spread of COVID-19 on IPV; and (ii) characterize couples at heightened risk as a result of these measures.

Rebecca Thornton, University of Illinois at Urbana-Champaign and NBER; Scott Cunningham, Baylor University; and Gregory DeAngelo and Anuar Assamidanov, and Yunie Le, Claremont Graduate University

"COVID-19, Shelter-in-Place, and Domestic Violence"

There are widespread claims and concerns of global increases in domestic violence and child abuse due to the COVID-19 pandemic. Multilateral organizations, countries, and local governments are bracing for increased need to support domestic violence victims. In March, the United States budgeted $47 million for domestic violence in its coronavirus relief package and several state governments set aside their own emergency funds. Thornton, Cunningham, Le, and Assamidanov use a rich set of data and robust empirical causal methods to measure the effects of COVID-19 and subsequent public health policies on domestic violence against women, men and children in the United States. Using 911 and 211 calls for service, police records, and arrest/bookings, the researchers exploit the differential timing of shelter-in-place orders, school and childcare closures, prisoner releases, and foot traffic/mobility to study the impact of the pandemic on violence within families. The researchers also explore heterogeneity across census tract characteristics and where possible, across individual-level demographics and poverty status. Results from the study will help inform policy makers with understanding the ways domestic violence survivors can be supported.

Sarah J. Baird, George Washington University, and Manisha Shah, University of California, Los Angeles and NBER

"The Shadow Pandemic: COVID-19 and Violence Against Adolescent Girls in LMICs"

COVID-19 has rapidly disrupted the lives of individuals across the globe. Of particular concern is the ‘Shadow Pandemic’ of increased violence against women and girls. For adolescent girls in low- and middle-income countries (LMICs), this increased risk of violence is particularly acute due to already existing poverty and gender-based inequalities, as well as decreased access to existing social safety nets including school-based resources. These short-term increases in violence are likely to persist with disruptions in education increasing pressure to marry early and unintended pregnancies, resulting in long-term consequences for the well-being of the adolescent. Through the use of a rapid phone survey combined with ongoing panel data collection as part of the Gender and Adolescence: Global Evidence research program, Baird and Shah provide evidence on the near-term impacts of COVID-19 on the experience of violence by over 2,600 adolescent girls and its consequences across three diverse countries—Bangladesh, Jordan and Ethiopia—as well as across distinct contexts within each country (urban vs. rural; refugee vs. non-refugee), and other important characteristics of the adolescent (e.g. marital status). This analysis will add to the relative dearth of research on the impacts of health crises on adolescents’ exposure to violence in LMICs. Given the alarmingly high prevalence of gender-based violence faced by adolescents and young women, and its expected increase due to COVID-19, it is crucial we begin to understand the causes and consequences of violence in order to develop and advance interventions and policies to reduce this experience of violence.

Bilge Erten and Silvia Prina, Northeastern University, and Pinar Keskin, Wellesley College

"COVID-19 Movement Restrictions and Domestic Violence: Evidence from the US"

Recent anecdotal evidence suggests that there has been a considerable increase in cases of domestic violence in several countries after the introduction of movement restrictions to slow the spread of COVID-19. Erten, Keskin, and Prina take advantage of variation across space and time in shelter-in-place policy changes implemented at the state level in the United States, as well as compliance with those restrictions, to estimate the causal impacts of social isolation on the risk of experiencing domestic violence. Their estimates show that online searches for domestic violence hotlines and calls to police departments for domestic violence incidents begin to increase prior to the implementation of these mandated restrictions and continue afterwards. The findings are consistent with the decline in mobility, which started prior to such policies. Results appear to be driven by high-income households who have a greater ability to reduce their movement outside home.

Papers on Women's Well Being and Children's Health
(Joint with the Children Group)

Daniel Halim, Hillary C. Johnson, and Elizaveta Perova, The World Bank

"Preschool Availability and Female Labor Force Participation: Evidence from Indonesia"

At 50.9 percent, female labor force participation in Indonesia is far below the regional average of 60.8 percent. Is it being hindered by a lack of affordable childcare services in the country? Halim, Johnson, and Perova exploit the joint variations in preschool-age eligibility and access to preschool across regions and over years in a difference-in-differences framework. Using a constructed panel of mothers to account for unobserved individual career and family preferences, they estimate the elasticity of maternal employment to preschool access. The researchers find that mothers of age-eligible children are more likely to be employed as a result of improved access to public preschools. Mothers are absorbed in informal sector occupations that do not require full-time commitments.

Sarah Miller, University of Michigan and NBER; Laura R. Wherry, University of California, Los Angeles and NBER; and Diana G. Foster, University of California, San Francisco

"The Economic Consequences of Being Denied an Abortion" (NBER Working Paper 26662)

Restrictions on abortion are pervasive, yet relatively little is known about the financial and economic impact of being denied an abortion on pregnant women who seek one. Miller, Wherry, and Foster evaluate the economic consequences of being denied an abortion on the basis of the gestational age of the pregnancy. Their analysis relies on new linkages to administrative credit report data for participants in the Turnaway Study, the first study to collect high-quality, longitudinal data on women receiving or being denied a wanted abortion in the United States. Some women had pregnancies close to the facility's gestational age limit, but below it, and received a wanted abortion (Near Limit Group). A second group of women had pregnancies just over the facility's gestational age limit and were turned away without receiving an abortion (Turnaway Group). The researchers link study participants to ten years of credit report data including several years prior to their recruitment into the study. Using these data, they compare differences in credit report outcomes for the two groups of women over time using an event-study design. They find that the trajectories for these outcomes are similar for the two groups of women prior to the abortion encounter. However, following their visit to the abortion provider, the researchers find evidence of a large and persistent increase in financial distress for the women who were denied an abortion that is sustained for several years. Being denied an abortion increases the amount of debt 30 days or more past due by 78 percent and increases negative public records, such as bankruptcies and evictions, by 81 percent. The researchers conduct additional analyses that use a regression discontinuity design to compare outcomes for women just above and just below the gestation limit at each clinic and find results that are consistent with the event study analyses. Miller, Wherry, and Foster explore the mechanisms behind these findings by taking advantage of existing survey data collected for the study participants and compare the effects sizes w document to those experienced by similar women following a "typical" birth. The results highlight important financial and economic consequences of restrictions on abortion access.

Wolfgang Keller, University of Colorado, Boulder and NBER, and Hâle Utar, Grinnell College

"Globalization, Gender, and the Family" (NBER Working Paper 25247)

Keller and Utar show that in the presence of labor market shocks child-bearing and child-rearing has far-reaching implications for gender inequality, household specialization and family structure. Employing population register data on all births, marriages, and divorces together with employer-employee linked data for Denmark, the researchers show that lower labor market opportunities due to Chinese import competition lead to a shift towards family, with higher rates of fertility, parental leave, and marriage and lower divorce rates. This shift is driven by women, not men. Correspondingly, the negative earnings implications of the trade shock are concentrated on women, and gender earnings inequality increases. Central to this difference is that a woman’s ability to give birth during a specific period of her life, referred to as the biological clock argument. Women have a higher reservation value for staying in the labor market when young, and a given negative labor shock induces women to substitute more strongly to family activities than men. There is no gender difference (1) for workers past their fertile age, (2) in the size of the negative labor shock, and (3) due to occupational composition since the researchers exploit only within-worker variation. Despite the lower labor earnings, fertility increases in Denmark also due to insurance payments and government transfers that sustain personal income so that the positive substitution effect towards fertility dominates the negative income effect that lowers the demand for children.

Mention of Three Additional Papers on Women's Well Being and Children's Health

Scott Kim, University of Pennsylvania, and Petra Moser, New York University and NBER

"Women in Science. Lessons from the Baby Boom"

How does parenting influence gender inequality in science? Kim and Moser investigate this question by examining data on children, productivity, and promotions for nearly 83,000 American scientists in 1956, the height of the baby boom (1946-64). Using patents to measure productivity, they find that parenting reduced the productivity of mothers but not fathers. Mothers were less productive in their 20s and early 30s but became more productive after age 35, reaching peak productivity several years after other scientists. Event study estimates show that the productivity of mothers declined after they married but recovered 15 years later. In contrast, fathers and other women were most productive in the early years after marriage. These differences in the timing of productivity have important implications for promotions. Specifically, the researchers find that mothers were 21 percent less likely to be promoted to tenure compared with fathers and 19 percent less compared with other women. In contrast, fathers were slightly more likely to get tenure compared with other men. To interpret these findings, the researchers investigate selection into marriage, parenting, and “survival” in science. Mothers were no less productive than other women, but female scientists married late and had fewer children than male scientists. Linking the data with faculty records, the researchers show that female scientists, and especially mothers, were less likely to survive in science. Employment data reveal a dramatic decline in entry by women who were in their 20s at the baby boom, suggesting that the disparate burden of parenting created a lost generation of female scientists.

Hanno Foerster, Boston College

"Untying the Knot: How Child Support and Alimony Affect Couples' Dynamic Decisions and Welfare" (slides)

In many countries divorce law mandates post-marital maintenance payments (child support and alimony) to insure the lower earner in married couples against financial losses upon divorce. This paper studies how maintenance payments affect couples' intertemporal decisions and welfare. Foerster develops a dynamic model of family labor supply, housework, savings and divorce and estimate it using Danish register and survey data. The model captures the policy trade off between providing insurance to the lower earner and enabling couples to specialize efficiently, on the one hand, and maintaining labor supply incentives for divorcees, on the other hand. He uses the estimated model to study various counterfactual policy scenarios. The researcher finds that alimony payments come with strong labor supply disincentives and as a consequence fail to provide consumption insurance. The welfare maximizing policy involves increasing the lump sum component of child support, increasing the dependence of child support on the payer's income and reducing alimony payments relative to the Danish status quo. Switching to the welfare maximizing policy makes women better and men worse off, but comparisons to first best allocations show that Pareto improvements are feasible, highlighting a limitation of child support and alimony policies.

Sandra Aguilar, Columbia University; Eva O. Arceo-Gomez, CIDE; and Elia De la Cruz Toledo He, Loyola Marymount University

"Inside the Black Box of Child Penalties" (slides)

As developed countries have been unable to completely close the gender wage and participation gap, recent literature has revisited the old findings regarding the existence of child penalties in the labor market. Developing countries, however, present different challenges to the ability of women to work for pay. Aguilar, Arceo-Gomez, and De la Cruz Toledo He produce the first formal estimation of the child penalties in the Mexican labor market, and the second in a Latin American context. Using an event study approach and an instrumental variable strategy as a robustness check, they estimate the short-run impact of children on labor force participation, wages, shifts from hours at work to hours in unpaid labor at home and on transitions between the informal and formal sector. The researchers find significant gaps between men and women in the short-run impacts of children on both paid and unpaid work. Results show that the arrival of a child affects members of the extended family unevenly, reinforcing gender roles. The researchers also find that childbirth has a greater effect on lower-income women’s wages, and some heterogeneity by income deciles in the magnitude and trends of the effect of children on hours worked, paid and unpaid.

Papers on Women and Education across the World

Claudia Senik, University Paris IV Sorbonne, and Naomi Friedman-Sokuler, Bar-Ilan University

"From Pink-Collar to STEM fields. Cultural Persistence and Diffusion of Socialist Gender Norms"

The fall of the Iron Curtain in 1989 led to a mass wave of migration from the Former Soviet Union (FSU) to Israel. Senik and Friedman-Sokuler document the persistence and transmission of the Soviet unconventional gender norms, both vertically across generations of immigrants, and horizontally through neighborhood and school peer effects. The researchers follow an entire cohort of Israeli girls throughout their educational and occupational choices. Among young Israeli women who immigrated as infants from the FSU, the researchers identify the persistence of two important features of the Soviet culture: the prioritization of science and technology, and the strong female attachment to paid-work. In high school, these women are significantly more likely than natives and other immigrants to major in STEM. In tertiary education, they remain over-represented in STEM, but also differ significantly from other women by their specific avoidance of study fields leading to “pink collar” jobs, such as education and social work. They also display a specific choice of work-life balance reflecting a greater commitment to paid-work. Finally, the choice patterns of native women shift towards STEM and away from traditional female study fields as the share of FSU immigrants in their lower-secondary school increases.

Itzik Fadlon, University of California, San Diego and NBER, and Frederik P. Lyngse and Torben Heien Nielsen, University of Copenhagen

"Early Career, Life-Cycle Choices, and Gender"

Josefa Aguirre, Pontificia Universidad Catolica; Juan Matta, New York University; and Ana Maria Montoya, Universidad de Chile

"Joining the Men's Club: The Returns to Pursuing High-earnings Male-dominated Fields for Women"

The low participation of women in high-earnings fields such as technology and engineering (TE) is believed to contribute to the gender wage gap. This paper investigates the labor market returns to pursuing majors in TE for men and women using data from Chile. Aguirre, Matta, and Montoya link administrative records on postsecondary application and enrollment to labor earnings and fertility data and exploit discontinuities in admission generated by Chile's centralized system of admission to higher education. They find that enrollment in TE as opposed to humanities, arts or social science (HASS) increases men's earnings and employment by 74% and 29%, but does not increase earnings or employment for women. The absence of returns for women seems to be the consequence of them failing to fully integrate into the men's club: enrollment in TE increases the probability of employment at high-paying and male-dominated industries for men, but not for women. Finally, the researchers show that enrollment into TE does not affect women's fertility or their partners' test scores and earnings.

Mention of Three Additional Papers on Women and Education Across the World

Carolyn Sloane, University of California, Riverside; Erik Hurst, University of Chicago and NBER; and Dan Black, University of Chicago, "A Cross-Cohort Analysis of Human Capital Specialization and the College Gender Wage Gap" (slides)

Eric V. Edmonds, Dartmouth College and NBER; Benjamin Feigenberg, University of Illinois at Chicago; and Jessica Leight, International Food Policy Research Institute

"Advancing the Agency of Adolescent Girls" (NBER Working Paper 27513) (slides)

Can life skills be taught in early adolescence? Using a clustered randomized control trial, Edmonds, Feigenberg, and Leight analyze the impact of a school-based life skills intervention in grades six and seven within a sample of 2,459 girls in Rajasthan, India. Their evidence suggests that the intervention is successful in developing stronger life skills including increased agency, more equitable gender norms, and stronger socio-emotional support. Girls also drop out of school at a lower rate: the researchers observe an approximately 25 percent decline in dropout that persists from seventh grade through the transition to high school.

Yana Gallen, University of Chicago, and Melanie Wasserman, University of California, Los Angeles

"Informed Choices: Gender Gaps in Career Advice"

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us