This year, the annual State of the Union hosted by the Stanford Center on Poverty and Inequality at Stanford University was dedicated to gender inequality. Researchers at the conference discussed the worrying trajectory of women's social and economic advancement over the past fifty years: fast progress until the 1990s and a slowing change thereafter [Marianne Cooper and Shelley J. Corell, on Policy]. The conference covered topics ranging from the gender earnings gap, to sexual harassment in the workplace, to early childhood education. Here, we summarize some of the main findings of the research presented.
the earnings gap
A typical measure of the gender wage gap is the ratio of female median wages to male median wages. The figure below demonstrates that the gap in median earnings shrunk significantly in the 1970-80s but has only fallen slightly since 1990, now hovering around 81 percent [Organisation for Economic Co-operation and Development. 2017. “Earnings and Wages: Gender Wage Gap”]. Emmanuel Saez presented a more comprehensive measure of the gender wage gap that shows an even larger difference between male and female earnings [Piketty, Saez and Zucman, 2018]. Unlike the median wage gap, the gap measured by Saez and his coauthors takes into account people who are not employed or self-employment, and includes earnings and fringe benefits, such as pension contributions and health benefits. Under this full accounting, women earn as little as 57% of men's earnings.
Occupational segregation is another aspect of gender inequality in the labor market and one of the causes of the gender earnings gap. Dafna Gelbsiger of Facebook explained that one facet of this is ''vertical segregation,'' the phenomenon in which women work disproportionately more in positions with lower pay, lower chances for promotion and less authority than men. As the percentage of women in an occupation increases, Gelbsiger argued, the median wages of that occupation decrease. Only 20% of women in the US work in occupations where women's median hourly wage is at least 95% of men's median hourly wage; and only 5% of women work in occupations where the women's mean hourly wage is at least 95% of men's mean hourly wage [Weeden, Newhart, Gelbgiser, State of the Union 2018, "Occupational Segregation"].
Research presented by Adina Sterling of Stanford University suggests that social ties also play into the gender earnings and employment gap. While women have larger networks than men, they on average have significantly less co-workers in their network than men. This is important, Sterling asserted, because work-related relationships increase the likelihood of finding a job and succeeding in one's career.
Women more frequently live in poverty or in deep poverty than do men, according to research presented by Luke Shaefer of University of Michigan. Shaefer presented the figure below showing that while there was a small narrowing of the gap between the number of women and men in poverty during the 1990s, there is no evidence of a change since then. As a consequence, women are also more likely to use the social safety net than men - although qualitative evidence presented by Linda Burton of Duke University suggests that part of this difference is due to gender norms that stigmatize safety net use among men.
Differences in education can potentially affect the occupation and earnings that both sexes can generate. Sean Reardon of Stanford University presented research showing that while women in the US are graduating from college at higher rates than men, they are still underrepresented in STEM majors, receiving only 35% of STEM bachelor degrees in 2016. If a degree in STEM is more highly valued in the labor market, the under-representation of women in STEM majors could explain part of the gender earnings gap.
Reardon presented evidence that male students do not consistently outperform female students in mathematics and that in reading female students consistently outperform male students throughout school, by one-half to four-fifths of a grade-level [fig. 1, page 11, Pathways : State of the Union 2018]. The figure below shows that on average, males have a negligible lead in math in fourth grade, which disappears by eighth grade. Once they reach high school, however, there is a shift and boys outperform girls in math in the US, by approximately one-third of a grade level in 2012. Reardon also cited research in psychology that finds little evidence of innate differences between men and women in math or science [“Sex Differences in Intrinsic Aptitude for Mathematics and Science?: A Critical Review.” American Psychologist 60(9), 950–958]. Surprisingly, Reardon’s work revealed that the math gap in high-school is wider in wealthier communities, though an explanation for this fact is not known.
sexual assault and harassment
Amy Blackstone of University of Maine discussed the economic relevance of sexual assault and harassment in the workplace. To begin, Blackstone cited survey evidence showing that when asked about particular behaviors, up to 85% percent of women report behaviors such as unwanted touching, leering, and offensive sexual joking at work [U.S. Equal Employment Opportunity Commission. 2016. “Select Task Force on the Study of Harassment in the Workplace”]. Counter to the popular belief that powerful men prey on less powerful women, evidence suggests that women in supervisory roles are significantly more likely than other women to be sexually harassed [Uggen, C. and Blackstone, A. 2004. “Sexual Harassment as a Gendered Expression of Power.” American Sociological Review]. Blackstone highlighted evidence that sexual harassment pushes women to change their jobs and reduce their working hours which can lead to financial stress and hampered career paths.
It is important to consider the variety of types of harassment that all people may encounter. Sexual harassment varies by ethnicity, race, age and gender expression. Individuals identifying as LGBTQ report elevated rates of sexual harassment relative to the general population. Women of color face harassment and legal barriers that are distinct in nature and require additional attention [Crenshaw, Kimberle. 1992. “Race, Gender, and Sexual Harassment.” Southern California Law Review 65, 1467–1476.].
While there is still much to do, a common theme throughout the conference was that in order to achieve equity, our society's ingrained beliefs about gender roles and identities must adjust to more flexible and fluid notions of sex and gender. Sociologists Aliya Saperstein and David S. Pedulla spoke of the non-binary gender spectrum and intersectionality with race and ethnicity as key components in understanding gender equity. These topics present an opportunity for economists to learn more about how people's sex and gender identification may impact their well-being, incentives, and opportunities in life.
In our own profession, many people are working to promote diversity and equality in the field. For example, the most recent issue of the CSWEP Newsletter focuses on "Dealing with Sexual Harassment," a crucial topic that has not been discussed widely in the profession. However, our field still has far to come and much to learn in order to achieve gender equality in our own profession and to effectively contribute to research on gender equality in the broader US.
Author: Ada Gonzalez-Torres
On January 5th, 2018, the AEA's Ad Hoc Committee to Consider a Code of Professional Conduct (hereafter, Ad Hoc Committee) released an interim report and a draft ''Code of Conduct'' for the economics profession. Clocking in at just over 200 words, the draft Code of Conduct is a concise statement on the values the economics profession strives to uphold. What is missing from the document is any such commitment or plan to uphold these values.
In order to craft an effective and appropriate Code of Conduct, the AEA must commit to a longer process that enlists and compensates a diverse group of economists to draft a robust document with a set of tangible commitments to improving conduct in our profession. We expect the group of economists crafting this document to include women, people of color, LGBTQ economists and those from a diverse set of socioeconomic, religious, national and intellectual backgrounds. These economists should be compensated financially for drafting a complete and thorough Code of Conduct that outlines concrete types of behavior that are deemed unacceptable and that institutionalizes a process through which violations can be reported and addressed.
The current draft emphasizes the importance of ''honesty and transparency in conducting and presenting research,'' as well as the AEA's support for ''participation and advancement in the economics profession by individuals from diverse backgrounds.'' The acknowledgement of these two standards is crucial to maintaining a high quality of research and respect in the profession. Given the behavior we have seen on EJMR and research showing systemic disadvantage for women and economists of color, it is clear that our profession has not been living up to to these ideals. The AEA's draft Code of Conduct takes an important step towards acknowledging the important and simple standards that all research should adhere to.
However in order to turn bold ideas into a reality, the AEA needs to establish institutions and systems that will incentivize the behavior they endorse, and address issues as they come up (and evolve over time). Professions such as sociology and law have modeled the type of robust code of conduct that a profession such as economics could adopt.
These other professions have established more specific guidelines for conduct with other professionals, clients, research subjects, audience members, and the general public. While the current draft put out by the AEA does allude to social media and spaces where comments can be made anonymously as venues that need to maintain a high standard of conduct, they do not explicitly address the specific forms of misconduct that would violate the Code. This leaves ample room for ambiguity and inaction in the case of misconduct.
Further, the Code does not offer a system of recourse for those who have witnessed or fallen victim to violations of conduct. Without a formal process of reporting and addressing violations, it is not clear that this Code will exist as anything other than wishful thinking. Again, other professions offer models for the type of committee that could receive and address reports of violations and address the ongoing need to adapt the code to an ever changing world.
Finally, the AEA has taken steps to support economists of diverse backgrounds by supporting CSWEP, CSMGEP, and the Ad Hoc LGBTQ working group. This work needs to be formalized into a concrete plan that will sustain the work of these committees without relying upon the labor and time of the underrepresented group members. It is a privilege that we have strong people leading these groups. But in order to truly make inclusion and representation a priority the AEA cannot rely solely on those who have been systematically disadvantaged.
It is in this sense that we feel that the draft Code of Conduct is incomplete. The Code states that it is the collective responsibility of economists to “develop institutional arrangements and a professional environment that promote free expression concerning economics.” However, to date, these institutions (such that they exist) have been championed primarily by those who have faced barriers to entering and contributing to the field. Without making a formal commitment to concrete and explicit support for free expression, ethical conduct, and the promotion of underrepresented groups, the AEA simply maintains the status quo - a status quo that is not acceptable by any measure.
In a supplementary document published alongside the draft Code of Conduct, the Ad Hoc Committee defended their choice to write a ''parsimonious'' Code of Conduct. In doing so, the Committee punted to the AEA journals as a venue for upholding professional conduct in the publication process. This emphasis on such a narrow venue highlights the incompleteness of the process to write and adopt a Code of Conduct that guides conduct in our profession as a whole. The supplementary document also includes a list of concrete ideas that the AEA may choose to adopt in the future to improve the profession. This list is a welcomed start to developing a full Code of Conduct (and we at WEB may respond to the specific recommendations in a future post). However, it’s relegation to the end of a supplementary document again demonstrates that the current draft of the Code of Conduct takes no explicit steps towards an improved discipline.
The Ad Hoc Committee states, in their supplementary document, that ''if the AEA decides it wants to adopt a detailed professional code of conduct ... it would need a committee with sufficient time to prepare such a document, including provisions for collecting suggestions and feedback from the profession.'' This statement echos the call that we have made here. We suggest that this work be done and that if it is not done, then the AEA will not have made progress towards a more equitable discipline.
Author: Emily Eisner
It’s recruiting and hiring season, and we know that many departments have struggled with the question of how to recruit underrepresented researchers to their departments. This is a difficult, nuanced, and controversial task. Juan Carlos Suárez stated on Twitter that
It is important to be thoughtful as to why diversity and inclusion in economics is important. And it is important to make a clear and defined plan as to how to address this need in your department. Here, we present a set of tips and tools for anyone to use while thinking about how to hire historically underrepresented researchers.
Calling your network: Professors may be particularly likely to recommend students that remind them of themselves at a young age, or their successful peers. That is problematic in a largely white, male profession. When making calls to other departments, be sure to include the question ''Who are your top female and minority candidates?'' Ask members of your network to particularly encourage qualified underrepresented candidates to apply.
Do your own search: We also recommend conducting your own broad search for potential applicants in addition to reviewing those recommended by faculty at other institutions. At Berkeley, WEB conducts an extensive search of job market websites for highly qualified candidates whose research is a good fit for the position. We pass the list of names on to the hiring departments, who often reach out and encourage these candidates to apply.
Discussion of candidates
Set evaluation criteria ahead of time: There is a great piece on this by David Wilcox, Director of the Division of Research and Statistics at the Fed, in the latest CSWEP newsletter. At the Fed,
Wilcox also stresses not including “fit” or potential for friendship in the criteria, and instead focusing on those who bring something that is missing from the group. There is a sample rubric available here from the University of Texas.
Keep common biases in mind when reading letters: Hiring committees are often looking for “geniuses,” and academics are less likely to view women or African-Americans as geniuses (Leslie et al. 2015). Take this and other forms of bias into consideration when reading letters of recommendation. The University of Arizona Commission on the Status of Women lists common differences in the way we describe men and women in letters of recommendation. Men tend to be praised with standout words such as “excellent” or “unparalleled,” whereas women tend to be described by words such as “hardworking” or “meticulous.” (Trix & Psenka, 2003). Trix & Psenka also found that letters for female applicants were shorter, had fewer mentions of research and scientific terminology, and less use of possessive phrases such as “her skills and abilities.” A letter gender bias calculator is available online if you would like to score your own letters or the letters you receive.
Set evaluation criteria ahead of time: Again, setting criteria ahead of time is helpful for combating bias. This website has some useful tips: before the first interview, create an interview rubric that will be used for all candidates. Have a single pool of questions, and choose which questions to ask which candidates before the first interview.
Confidence ≠ competence: When evaluating candidates, keep in mind that confidence and competence are not the same thing. Candidate self-perception and self-presentation can depend on a number of factors, including the potential for people do react badly to displays of confidence. One study found that, when interviewers perceived female candidates as ambitious and self-reliant, (1) the interviewers judged the female candidate to also have low social skills, and (2) interviewers shifted hiring criteria to more strongly emphasize social skills (Phelan et al. 2008).
Interview location: Be sensitive to the fact that hotel rooms can be an uncomfortable interview location, particularly with a female candidate and an all-male panel. For this reason, and many others, try to have a gender balance on the interview committee.
Use EJM to avoid EJMR: Post the status of your open position on econjobmarket.org so that candidates do not have to visit EJMR to find out information about the progress of the application cycle.
The ''two in the pool'' effect: One study found that the odds of hiring a woman for a faculty position were 79 times greater if there were at least two women, rather than only one woman, in the finalist pool. The odds of hiring a minority were 194 times greater if there were at least two people of color in the finalist pool. This may be because having only one woman or minority in the group highlights how that candidate deviates from the norm (Johnson et al. 2016). Ensure that there are at least two women and two people of color in the group chosen for flyouts.
Interpreting candidate presentation style: Again, candidate self-presentation style will vary by background and is not always a good indicator of ability. Audiences at economics job talks are notoriously aggressive, but there is evidence that white women and African-American men are penalized for responding aggressively, while white men are not (Brescoll & Uhlmann 2008; Livingston et al. 2012). Remind faculty to keep this in mind and focus on the content rather than tone of candidate responses.
Set evaluation criteria ahead of time: Have all faculty members who will be involved in the hiring conversation write down a rubric on what they are looking for before the first job talk, with as much detail as possible. Ask them to stick to this rubric.
Being alone in the department as a woman or person of color can be alienating. During the visit, demonstrate that your department has women and people of color who are thriving (if this is the case). Highlight the aspects of your school and department that may be particularly appealing to women or people of color, such as paid parental leave or grant opportunities for minority researchers. If your female or minority faculty are not currently taking a disproportionate number of administrative tasks (Guarino & Borden 2017), arrange for a meeting with a female faculty member during the visit. Or, if your department is not diverse, know that you may have to invest extra resources into hiring diverse candidates now in order to attract a broader talent pool in the future.
Continued learning and improvement
Recruiting a diverse faculty will require work over many years, and we suggest you continue learning throughout the process. In particular, commit to publishing statistics on diversity for all steps of the process, to identify where women and people of color are dropped. Solicit feedback from all candidates who turn down offers. The goal is to learn about all stages of the recruitment process: qualified candidates who don’t apply, who turn down an interview, who decline a flyout, and who reject an offer.
Author: Deirdre Sutula
This winter season, as you snuggle under a cozy blanket and reflect on the past year, consider reading some of this year’s most important new research on diversity and discrimination. We’ve put together a list of papers from 2017 that demonstrate the state of diversity in economics and reflect the challenges of eliminating gender and racial disparities in the workforce.
For each paper we’ve included a brief summary. We recommend reading with hot chocolate in hand and an eye towards how to implement more inclusive, equitable systems in our own daily lives. Happy 2018!
The Persistent Effect of Temporary Affirmative Action
American Economic Journal: Applied Economics, 2017
The unintended consequences of "ban the box": Statistical discrimination and employment outcomes when criminal histories are hidden
Jennifer L. Doleac and Benjamin Hansen
Working Paper, 2017
Gender biases in student evaluations of teaching
Journal of Public Economics, 2017
Publishing while female: Are women held to higher standards? Evidence from peer review
Working Paper, 2017
Gender Representation in Economics Across Topics and Time: Evidence from the NBER Summer Institute
Anusha Chari and Paul Goldsmith-Pinkham
Working Paper, 2017
More Reading! (i.e. some papers you may have seen already but should check out if you haven't):
Author: Emily Eisner
Emily Eisner and Megan Lang
with reflections contributed by Karl Dunkle Werner, Gabe Englander, Scott Kaplan, and Derek Wolfson
“What can I do?” We heard that question from many men in our field in the wake of Alice Wu’s paper and, more broadly, the #MeToo movement this fall. We thought about writing a “How to be an ally” post for economists. Instead, we asked four aspiring allies in our field to find out how to be an ally for themselves, then compile that knowledge for others who would like to know.
We’ve organized the content from their numerous interviews with women and minority faculty and grad students in our field at UC Berkeley here. These are resource pages that we’ll continue to build and update over time.
We also asked all four men to write reflections on their learning throughout this process. To some, these reflections may seem all too familiar. To some they may be revelatory. Some people may disagree with what they read, or feel disempowered by the thought that white men have any role to play in improving opportunities for women and minorities in the field. Some may find that these reflections resonate deeply. The one thing that is clear is that none of us has all the answers and that people’s experiences cannot be reduced into one tidy lesson.
By offering the four reflections below, we hope to
There is not one way to be an ally just as there is not one correct way to mitigate discrimination in our field. Below are four reflections from the male graduate students we asked to help us write a blog post. We hope they spark thought and conversation. Feel free to reach out to us with any reactions or stories of your own.
Before this process, I hadn’t given thought to what my gender meant in terms of my relationship with my advisor. We are both men, and we have been able to build a strong relationship both at work and outside of it. This relationship has undoubtedly improved my education and ability to do good research.
I took for granted that every graduate student may not have a similar relationship with their mentor. But a point that came up again and again in interviews and the process of writing this blog was how the power dynamics associated with advisors and advisees were inflated by gender dynamics between these two parties. I was educated by a woman, who said that “the point here is not that women want to completely shut down informal activities with advisors, but that navigating those situations one-on-one is more fraught for them.”
I came to understand that it's an additional burden for a woman to have to think about the implications of their relationships with their advisors more seriously than I do. At the same time, advisors may face the same calculation and have difficulty in offering opportunities to advisees of different races and genders. This can have serious academic consequences. We as males need to be cognizant of this calculus; we can suggest that our advisors try to plan activities where all of their advisees can participate, and when we become professors or take on other positions, we can work to create group activities that are comfortable for and inclusive of everyone.
The first year of an econ PhD program is hard. It’s an enormous amount of math, a slew of all-day problem sets, and an endless stream of opportunities to bang your head against the wall. Sitting down with my classmates who tackled first year while raising a family was eye-opening. With large chunks of time occupied by children, problem sets and research are pushed to late nights or early mornings. Choosing when I work is a privilege I had never considered. Those with family responsibilities may have schedules precluding group-work and opportunities to compare notes. Young white guys are disproportionately free of these responsibilities, one of many factors entrenching existing disparities. There have been some attempts in the field to accommodate parents, but coming up with a solution is not easy. We need to think harder about how graduate school’s demanding workload interacts with existing social structures, making the field particularly unwelcoming for women and economists of color.
At a more individual level, my conversations with classmates highlighted how important it can be, in moments of self-doubt, to hear that other people also struggle. In our conversations, people repeatedly brought up how much an affirming word can matter, and how big a deal it is to know that other people face similar challenges. I had always considered myself implicitly supportive, available when someone tells me they need a hand. That’s not enough. I need to do a better job checking in, making sure my classmates are doing well, and celebrating their successes (in grad school and beyond).
Shortly after beginning his PhD, K remembers asking the professor who helped recruit him, “Where are all the black and African people?” K, one of only a couple of black students in the department, said that it would have been helpful if someone—a staff member, faculty, or another student—had explicitly acknowledged this fact and talked about some of the things he might expect to feel over the next 5+ years.
For example, K felt an extra layer of imposter syndrome at the beginning of his PhD program (the garden variety is familiar to many graduate students). Academic setbacks were especially painful because he feared they suggested to others that he was not accepted to his program on his merits.
K also regularly wonders how his identity influences how others interact with him. For instance, when he visited Berkeley after being accepted, a professor asked him what he scored on the math section of the GRE. Does this professor ask this question to most prospective students, or did this professor ask because K is black and African?
I had not previously considered that non-white, non-male economists have less control over how they are perceived. This lack of control creates pressure to make decisions that satisfy other people’s expectations. I don’t face this pressure. Importantly, all this time spent thinking about how one is perceived takes away attention from research. I already find graduate school to be emotionally and intellectually harrowing at times; all of us should try to relieve the additional burden that women and minority students experience. So I will reflect on the stereotypes and biases that I hold, and try not to impose them on others, especially when I am in a position of power. I will also try to take K’s advice and have more conversations that explicitly acknowledge identity and how it influences individuals’ experiences in economics.
My participation in these interviews spurred an intense inner dialogue on what it means to actually be an ally. Before, I called myself an ally but now I’m beginning to understand how to put words into action. I’ve come to the conclusion that being well-intentioned is simply not good enough.
I’ve become more cognizant of the things I get for free by being a straight, white male. I don’t have to analyze the implication of walking for a coffee with my male advisor. I don’t have to act “less feminine” to gain the respect of my peers. I don’t have to worry about uncomfortable one-on-one interviews with male faculty in private hotel rooms during the job market.
These interviews also propelled me to query my own experience and reflect on if I ever consciously or unconsciously made it difficult for women and minorities in economics. I had. Did I ever call women “dudes” or refer to groups of women as “guys”? Yes. Was I more likely to interrupt women in conversations? Probably. Did I react to research differently because the author was a woman? I hoped not but I could not be 100% sure.
Truly being an ally will be an everyday learning process. My involvement with this blog post has already spurred some formative changes but I do not know everything. As a start, I now often ask myself: “Did I read this paper/treat this person/react to seminar comments/write this email/grade this paper/conduct office hours/etc. differently because this person is a woman/minority?” If I’m not sure, I will ask a friend so I may continue learning how to be an effective ally.
We’d like to extend our profound thanks to all of the people we interviewed, to our classmates in ARE, Econ, and Haas, and to WEB. We’re all going to make mistakes, but we’re keeping this conversation open so we can learn and improve. Please join us!
This month, I had the pleasure of speaking with Alice Wu, the UC Berkeley graduate who, in her thesis, exposed the gendered and sexist language used to describe women on EJMR, a website used widely by economists to disseminate information on PhD job market candidates. During our discussion, Wu and I discussed her intellectual motivation as well as the material consequences of her work. To Wu, her thesis was an exercise in producing thorough economic research and exploring how to employ modern machine learning methods to examine economically interesting labor market interactions. However, with the explosion of her work in the news and into the consciousness of the economic field, Wu recognizes the value of her work in spurring an important conversation among economists regarding how we can make our field more inclusive.
The transcript of our conversation has been lightly edited for readability.
EE: What led you to study EJMR?
AW: Honestly, it was a bit unexpected. I was picking a research topic for my senior thesis. I had worked for David Card for a semester before and he was going to be my advisor. In a conversation while I was debating research topics, he pointed out that there was a controversy going on at the time regarding EJMR and asked whether I would be interested in studying that platform. I had taken a machine learning course where I had gotten some experience with text analysis, so I decided to see if I could employ those skills while writing my thesis.
EE: Were you familiar with EJMR as an undergrad, before coming up with your research project?
AW: I first heard about EJMR during my junior year of college at UC Berkeley. I had some friends who were also undergrads and were interested in reading online reviews of their professors, and so they found the website. These were male friends and they knew I wanted to go to graduate school in economics, so I remember them showing me and saying “look, these [are] going to be your future colleagues!” I was shocked when I saw the explicit sexual content of posts and the racism and sexism on the website. When I asked my friends who were graduate students at Cal if they knew of the website they seemed really embarrassed.
EE: What has been the most challenging part of this research project for you?
AW: The most challenging part was definitely the beginning, after I had the raw text data and had done the cleaning of the dataset. I was thinking about what to look at and the first thing that I tried was sentiment analysis, which I had used in my machine learning course. However, many of the online open-source packages that perform sentiment analysis on text are built for specific styles of text or types of documents. For example, the sentiment analysis picked out the comments about women as beautiful as “positive” comments. When I saw that the algorithm I was using picked out the comments about women’s appearance as “positive” I realized the sentiments assumed by the algorithm didn’t match the purpose of the forum. It didn’t feel right or appropriate to me to discuss people’s physical appearance on EJMR, so it seemed strange to label it as positive. At that point, I had to take a step out of the pre-existing packages and think about what I wanted to get out of the data and what my actual questions were. That was the most challenging part of the research project.
EE: It’s exciting that you were able to apply machine learning and text analysis techniques in your thesis. What was your first experience using machine learning techniques and how do you see them contributing to economics?
AW: The project I did in my machine learning class at UC Berkeley was a sentiment analysis of tweets on twitter. I looked at whether tweets were positive or negative using LASSO to pick up positive or negative terms.
The machine learning literature is very different from the economics literature in terms of what type of analysis they are performing. It seems to me that overall, people who study machine learning only care about the prediction power of a model, rather than estimating and studying the relationship between two variables of interest. For example, there is a method called the “ensemble” method in machine learning where an iterative process over multiple models is used to fit data. In the end, the ensemble method aggregates results from the different models to get the best prediction. In economics, we can’t really use this method because we wouldn’t be able to make inference. If we want to use machine learning techniques in economics, we need to be able to perform some sort of causal inference. In my project, I had to think hard about what I was measuring and what economic interpretation I could infer from the data I was looking at.
EE: Turning back to the question of sexism in economics, why do you think women are underrepresented? Do you think that the online conversations on EJMR reflect people’s day-to-day experience?
AW: There are multiple components to answering this question. The stereotype literature (which is more psychology than economics) says that young girls often think they will be more successful in humanities than in quantitative fields. That belief can guide what courses women and girls take in high school and college. Since economics requires a lot of math, many women are unprepared to actually take economics courses or go on in the field. This is an entry level issue.
However, in addition to the entry level issue, there is a “leaky pipeline” in economics (higher rate of attrition for women than for men). This poses another problem and is more a workplace issue. There is an issue in the culture of economics that might turn people away, and my paper exposes a part of that culture.
I hope that the people who post these crude and objectifying comments are not representative of the profession. But overall, the whole atmosphere and culture of a profession can be shaped by a select few people. When you have a few people standing out in a certain way, that has a big impact on the culture of the field. It doesn’t really matter who these people are but just their existence and participation is disturbing.
EE: What do you see as the main contribution of your paper?
AW: I’m really happy to see that this study has triggered an open discussion about the culture of economics and how it may impact women’s representation in the field. People are acknowledging that there’s an issue and that’s the first step to addressing it.
EE: How do you see the relationship between academic research and politics/change?
AW: The honest answer is that I didn’t think about politics at all when I was writing this originally. I was working on my senior thesis and focused on the research question itself and I didn’t see it as a political tool in any way. However I’m glad that it has sparked conversation and may induce positive change.
EE: Your paper focused on the difference in language used to describe men and women on EJMR. However, there was clearly evidence of homophobia and racism in your results. What do you make of this evidence?
AW: I see very problematic language with respect to race and homophobia but the way I designed my research question I was looking at the simple difference between language used to describe men and women. The homophobia definitely shows up in my analysis with respect to how the LGBT community are described online. However, there should be more careful analysis on these topics.
Author: Emily Eisner
Hallie Jo Gist
The gender disparity in the Economics discipline at UC Berkeley is no secret. For the undergraduates, a glance around the classroom is enough evidence of an imbalance. As a part of the nation-wide Undergraduate Women in Economics (UWE) Challenge, we attempted to quantify this apparent disparity.
The UWE challenge was inspired by Professor Claudia Goldin’s research on undergraduate women in Economics at the hypothetical "Adams College," a U.S. liberal arts college. Goldin found that women are less likely than men to enter and continue in the Economics discipline. As one of 20 randomly selected universities, UC Berkeley receives funding for research on the factors that affect the gender distribution of undergraduates in Economics. The data collection described below is a first step in this research.
Our main statistic of interest is what we call the "Goldin ratio." The Goldin ratio is the male-to-female ratio of degrees earned in a given undergraduate major, relative to the male-to-female ratio in the full undergraduate population at UC Berkeley. For example, if there is a 50-50 distribution of males and females among all undergraduates, a Goldin ratio of 2 for the Economics major implies that there are twice as many males as there are females earning an Economics BA. This ratio was calculated from information on degrees earned made available through CalAnswers. For raw numbers, please see the chart at the end of this blog.
At UC Berkeley, students interested in economics choose between three undergraduate majors - Economics (Econ), Environmental Economics and Policy (EEP), and Political Economy (Poli Econ). The Economics major is the largest undergraduate major at UC Berkeley, with an average of 470 graduating students every year over the past ten years. The Economics major is housed in the Department of Economics within the College of Letters and Sciences. The EEP major is housed in the Department of Agricultural and Resource Economics within the College of Natural Resources. The College of Natural Resources describes the EEP major as the study of "economics and political institutions that affect the development and management of natural resources and the environment." The Political Economy major studies "the relationship between politics and economics in modern societies, focusing on problems of both domestic and international policy." It is one of several majors in the International & Area Studies Academic Program within the College of Letters and Sciences. Finally, UC Berkeley also offers an undergraduate Business major housed within the Haas School of Business.
While these majors diverge in the electives offered to upper year students, there is a great deal of overlap in the prerequisites and core requirements for all four degrees. The Business major has similar prerequisites to the economics majors, but a more selective admissions process. We calculated the Goldin ratio in all four majors for the last 10 years.
Figure 1 shows the Goldin ratio for each graduating class, 2007 to 2016, in the Economics (Econ), Environmental Economics and Policy (EEP), Political Economy (Poli Econ) majors, and Business (Bus.) majors. As you can see, the annual Goldin ratios tend to be highest for students in the Economics major. This indicates that the Economics major is disproportionately male relative to the university. Moreover, this disparity is greater in the Economics major than in the other three majors. The Goldin ratios are also greater than 1 for the Business major, which tells us that the Business major is also disproportionately male relative to the university. The Goldin ratios are closer to 1 for the EEP and Poli Econ majors, indicating a more equal gender balance.
Examination of the Goldin ratio for transfer students reveals that the trend is not common to students of all educational backgrounds. Figures 2 and 3 show the Goldin ratio for students in all four majors (Econ, EEP, Poli Econ, and Bus.) who were admitted as transfer students and as new freshmen. Among students who were admitted as transfer students, the Goldin ratio tends to be greater than 1, indicating a gender disparity. Furthermore, the Goldin ratio for the past ten years has shown an overall increase, indicating that this disparity is growing. On the other hand, the overall Goldin ratio for students admitted as new freshmen has decreased over the past ten years and is approaching 1, at which point the gender distribution among economics students will be equal to that of all UC Berkeley undergraduates. This tells us that while the gender disparity in the economics majors is improving for four-year students, gender inequality is growing for transfer students.
While the preceding graphs give us a good sense of the big picture, they disguise the fact that the gender disparity in Economics at UC Berkeley varies significantly by racial group.
This is especially true in the Economics department. Figure 4 reports the Goldin ratio values computed using all UC Berkeley students graduating in the periods 2007-2011 and 2012-2016. The Goldin ratios are highest for students identifying as underrepresented minority (UM) - Black, Latinx, and Native American undergraduates. Meanwhile, with averages close to 1, the gender balance among students of Asian and Pacific Islander (PI) descent, and International students is practically at parity.
Similarly, the Goldin ratios for each five-year period are particularly high for students identifying as UM in the Business major. For students identifying as Asian and PI, the Goldin ratios are about 1, indicating a gender balance reflecting that among all Asian and PI undergraduates. For International students, the average Goldin ratios are less than 1, which implies that International women are actually over-represented in the Business major relative to the undergraduate population. This trend is true for EEP students as well, though the difference between racial groups along the lines of race is less pronounced. For Political Economy students, the five-year average Goldin ratios are about 1, indicating a much more equal gender balance across racial groups.
The difference in gender balance across racial groups prompted us to calculate a new ratio, which we call the "race ratio." The race ratio is analogous to the Goldin ratio; we look at the ratio of students of a particular racial group to students of all other racial groups. For example, if 18% of UC Berkeley’s undergraduates are of underrepresented minority (UM) racial groups, then a race ratio of 2 for UM students in the Economics major implies that only 9% of Economics BA recipients are UM students. We calculated this ratio for each of the three majors within the economics and business disciplines at UC Berkeley.
Figures 8-11 show the 2007-2016 annual race ratios for Asian/PI, white, underrepresented minority (UM), and international degree recipients in the Economics, EEP, Political Economy, and Business majors. For Economics students, there is a pronounced difference in race ratios between the four racial groups. The annual race ratios for UM students are much greater than 1 and tend to be higher than those for students of other racial groups. That is, UM students are severely underrepresented in the Economics major relative to the population of undergraduates, and this racial imbalance exists disproportionately for UM students. Furthermore, with an upward sloping trend, the graph suggests that this disparity is growing. At the other end of the spectrum, we can see that international and Asian/PI students are over-represented in the Economics major, with race ratios less than 1.
We see a similar distribution of race ratios in the Business major. However, for UM students in the Business major, the race ratios trend downward. This indicates that while UM students are underrepresented in the Business major, this inequality is becoming less pronounced. The difference in race ratio across racial groups is less dramatic for the EEP major, although the disparity is still there. Again, UM students in particular are underrepresented in the EEP major relative to the university. Within the Political Economy major, the distribution of students according to racial group is much closer to that of the university. For all four racial groups, the race ratios tend to be closer to 1 than among the other majors.
The race ratio tells us nothing about the distribution of students according to gender; neither does the Goldin ratio tell us anything about the distribution of students along the lines of race. That is, just looking at the proportions of race or gender groups blurs the inequality faced by underrepresented minority women.
We struggle to address this issue. One problem is the small number of UM women graduating with degrees in the economics discipline; this renders it difficult to make meaningful statements about annual trends in degrees earned by UM women. An average of only 9 UM women have graduated with a BA in Economics each year for the past 10 years. This fact alone is alarming.
As a starting point, we created what we’ve called an "intersectional ratio." The intersectional ratio is analogous to the Goldin ratio; it is the ratio of UM women vs. all other students, divided by this ratio in the full population of Cal undergraduates. Aggregating over two 5-year periods, we created intersectional ratios—one for the period 2007 to 2011 and another for the period 2012 to 2016. These ratios for the Economics major are summarized in Figure 12.
As you can see, the intersectional ratio is extremely high for UM women in the Economics major. This suggests that UM (Black, Latinx, and Native American) women are even more underrepresented than suggested by the race ratios or Goldin ratios.
It is clear that the economics and business disciplines at UC Berkeley suffer from gender and racial disparities. In this post, we hope to shine a light on these issues. We are not sure how or why these inequalities arise - this is a topic for future research. One possibility is that women, especially UM women, are less confident in their quantitative reasoning skills. Research suggests that women in mathematics classes experience stereotype threat, which limits their performance. To be clear, we do not mean to imply that women at UC Berkeley lack quantitative proficiency. Rather, we want to suggest that the UC Berkeley could do more to support women pursuing quantitatively rigorous degrees, and that this might promote higher numbers of women in economics and business.
However, the difference in gender and race distributions among the four majors may suggest another channel, especially when we consider that the four degrees require similar lower-division coursework. In particular, it may be that some students perceive that the Economics and Business majors do not address issues of social justice, while Political Economy and EEP do. In my own experience as an Economics student, I have often felt that the undergraduate Economics program at UC Berkeley can be used to answer questions related to social justice. Perhaps our efforts to increase the numbers of UM women in these degree programs should involve making information about the content and freedom within these majors more accessible.
For those of us involved in the UWE project at UC Berkeley, these data are just a starting point. We recognize that any attempt to promote a more equal distribution along the lines of race and gender in the economics and business departments will not be successful if we do not take an intersectional approach. Neither the Goldin ratio nor the race ratio captures the particularity of the experience of underrepresented minority women within the economics discipline. We hope that the data presented above can serve as a foundation in our effort to make the economics discipline a more equitable environment for people of all genders and races, and in particular to bring attention to the severe underrepresentation of minority women in Economics.
The research described above was conducted under the direction of Professor Martha Olney in the Economics Department at UC Berkeley. The ongoing project is funded by the nation-wide UWE Challenge.
Hallie Jo Gist is a fourth-year undergraduate Economics student at UC Berkeley. She is part of a team of undergraduates involved in the UWE project at UC Berkeley. Hallie Jo is originally from the San Francisco Bay Area.
WEB Editorial Board
Most of us have probably seen coverage (New York Times, Washington Post) of an undergraduate honors thesis written by Berkeley undergraduate economics major, Alice Wu. In her paper, Wu exposes the rampant misogyny cluttering Economics Job Market Rumors (EJMR), an anonymous forum. While it comes as no great surprise that EJMR is rife with sexist (not to mention racist and homophobic) commentary, Wu's paper uses natural language processing to quantify the extent of the vitriol - and the results are disappointing. On the site, the words most commonly associated with women include "hotter," "lesbian", and "feminazi", while the words most commonly associated with men tend to be positive ("goals", "greatest") or economics-related ("adviser", "textbook", "pricing"). At WEB, we condemn in the strongest possible terms the rampant sexism on EJMR.
The commentary and attitudes on display at EJMR represent a substantial challenge to women in economics - (explicit) bias - which can discourage women from undergoing graduate training, prevent them from entering academia upon graduation, and hamper the careers of female economists. Even though EJMR clearly does not represent the views of all economists, the anonymous trolling on the site does represent real issues in the profession, and is a "canary in the coal-mine for what happens every day" (Amitabh Chandra).
While it is undoubtedly true that not all men post on it, almost *all* women (that I know) have experienced their names be publicly associated with sexist, offensive, and untrue claims.
Furthermore, it is concerning that anybody associated with academic economics use the language described in Wu’s paper to describe women. The existence of this website sends a message to the population that economists do not take misogyny (or racism, classism, etc.) seriously.
Not personally participating in this harmful behavior cannot serve as an excuse or rationale for inaction -- especially when the language used at EJMR is viewed in the broader context of women’s experiences in the economics profession. There is mounting empirical evidence that women in economics often face uphill battles. Women get less credit for coauthored work (Sarsons 2017); despite having more readable abstracts, female economists' papers take six months longer to peer review in a top journal (Hengel 2016); and women are more likely to take on department service that does not contribute to promotion (Babcock et al 2017) The share of women by PhD cohorts has stagnated for the past decade (Daniel Paserman), and women are underrepresented in economics - even relative to other STEM fields (Bayer & Rouse 2016).
A lack of diversity in the economics field - potentially exacerbated by the kind of language found on EJMR - is a serious problem (See Bayer & Rouse 2016 for more on this). Wu's paper itself serves as a reminder of the importance of including women’s voices in economics: she brought creative methods to a sensitive topic, and sparked a much-needed conversation on gender disparity.
Economists throughout the profession have spoken out over the past few days against EJMR. The discussion among economists, across twitter and personal blogs, has extended to the state of gender disparity in economics as a whole. By rejecting discriminatory and harmful actions, senior faculty and prominent economists send a clear message to young up-and-coming economists that economics does not stand for bigotry.
We applaud the ongoing commentary that condemns the unacceptable culture at EJMR. We urge the economics profession to take advantage of this conversation and take further action. Several suggestions have already been floated.
Looking for some compelling beach reading? This month we’ve compiled a list of recent prominent papers that we’ve been itching to dig into all year. The papers listed below touch on issues related to women in the workforce, and some specifically address women in economics. Taken together, these papers illustrate the breadth of recent work explaining gendered outcomes in the labor force and demonstrate the many channels that may contribute to the gender earnings gap.
For each paper we've included a brief summary, and for the CSWEP Annual Report we’ve highlighted the key takeaways. Happy reading!
Can Women Have Children and a Career? IV Evidence from IVF Treatments
Petter Lundborg, Erik Plug, and Astrid Würtz Rasmussen
American Economic Review, June 2017
Recognition for Group Work: Gender Differences in Academia
American Economic Review: Papers & Proceedings, May 2017
Gender Differences in Accepting and Receiving Requests for Tasks with Low Promotability
Linda Babcock, Maria P. Recalde, Lise Vesterlund, and Laurie Weingart
American Economic Review, March 2017
Bargaining, Sorting, and the Gender Wage Gap: Quantifying the Impact of Firms on the Relative Pay of Women
David Card, Ana Rute Cardoso, and Patrick Kline
Quarterly Journal of Economics, October 2015
Women and Power: Unpopular, Unwilling, or Held Back?
Pablo Casas-Arce and Albert Saiz
Journal of Political Economy, June 2015
Report: Committee on the Status of Women in the Economics Profession
American Economic Review: Papers & Proceedings, May 2017
Authors: Fiona Burlig and Emily Eisner
In their 2017 AER paper, “Acting Wife,” Leonardo Burstyn, Thomas Fujiwara, and Amanda Pallais show that single women in MBA programs under-invest in career-improving activities when these actions are observable to their peers. The authors argue that single women face a trade-off between investment in their careers and investment in the marriage market. In the labor market, ambition, willingness to travel, and willingness to work longer hours are rewarded. On the other hand, research shows that heterosexual men often do not want spouses with higher ambition or higher earnings. This trade-off reduces the demonstrated ambition and career-investment of single heterosexual women compared to their non-single female peers.
The authors conducted a field experiment in a first-year MBA class using a survey that they were told would be used by the career center in summer internship matching. The survey, filled out on the first day of the MBA program, asked each student about their career preferences, desired compensation, desired hours, and leadership skills. The instructions on the top of the survey stated that students’ answers would be discussed in one of their first semester courses. One group of students had instructions stating that answers would be discussed anonymously, while another group’s survey instructions simply stated that answers would be discussed. Comparing the survey answers between these groups allowed the authors to measure how single women may avoid career-enhancing activities in front of peers who they may meet on the marriage market.
The results of the experiment show that single women consistently under-report their desired compensation, propensity for leadership, hours willing to work, and travel on the job compared to their non-single peers. Importantly, this under-reporting is only seen for women who thought their responses may be observable to classmates. Because the MBA students believed that survey answers would have a direct impact on their summer internship prospects, the under-reporting suggests real under-investment in career enhancing activities and predicts real under-performance.
The results of this paper are hard to dispute - the research design leaves little room for other possible explanations of the mechanism driving women to under-perform in this context. While I was not completely surprised by the results when I saw this paper, I was saddened by the fact that this marriage-market-labor-market trade-off may leave such strong marks on women’s participation in the labor market. To learn more, I emailed Amanda Pallais herself to see if she would talk with me about the paper, its inspiration, and her next steps in researching this topic.
EE: What led you to the idea that sparked this paper?
AP: My coauthors and I had read research that finds that men prefer female romantic partners who are less ambitious and successful than they are. This led us to believe that single women might face a trade-off between investment in the labor and marriage markets. Anecdotally, we also heard single women considering this trade-off and wanted to study it formally.
EE: How do you think about this work as part of the larger literature on gender norms in the labor market?
AP: We see our work as in line with this literature - gender norms can lead women to have worse labor market outcomes. One branch of literature studies the effect of having children on men and women’s labor market outcomes and tends to find that women’s careers suffer more than men’s after having a child. Our paper is related, but shows how these norms and the marriage market can affect women’s labor market success even before they are married.
EE: How has your paper been received by the public? I would imagine that it is hard to talk about these issues with some people?
AP: It hasn’t been particularly hard to talk with people about the results of our paper. It seems many people had a feeling that single women face a trade-off between the labor market and the marriage market - and it’s been nice to be able to formalize this.
EE: Have you at all considered using variation in sexual orientation to study whether non-heterosexual women and men face any similar trade-offs? Or have you thought to use non-heterosexual women as a sort of control group for the mechanism you’re identifying?
AP: We have thought about that, and would like to be able to explore the behavior of women who are not on the heterosexual marriage market. Unfortunately we don’t have data on sexual orientation, and even if we did, I don’t think we’d have enough power to identify anything given the sample size that we’re working with.
EE: Was there anything you found while writing this paper and running this experiment that particularly surprised you?
AP: The magnitude of these results was pretty surprising to me. My co-authors and I started this project to test whether single women respond to this trade-off, so we weren’t particularly surprised that they did. But the extent to which single women made themselves look less professionally appealing due to marriage market concerns was striking. For example, single women reported being willing to travel an average of seven fewer days a month when their answers would be seen by their classmates. That is a large difference, implying potentially dramatic professional sacrifices.
EE: So, what is your next step with this work?
AP: There are two branches of things that we’re interested in. One is to extend our results to different contexts such as high schools, colleges, and jobs, looking at the long run effects on women’s labor market and marriage market outcomes. The second is what policies could help mitigate the trade-off that leads women to under-invest in their careers. For example, we might think that we want to make women’s actions private, but that’s really hard to do. Something that may be more feasible would be to change some of the defaults, like going from hand-raising in-class to cold-calling. If women participate less in class because participating signals an undesirable trait to the marriage market, cold-calling might remove this signal and lead women to participate more.
While the results in this paper highlight some of the challenges in improving labor market opportunities for women, the paper also signifies an important line of work going on in economics. In order to increase equality in the labor market, we must understand the precise obstacles that women face. I look forward to seeing what other work this paper will motivate and how the discipline will respond.
Author: Emily Eisner