(updated in 2016)

Like Google and Yahoo, our Anderson finance area is regularly get skewered for why we don't have more women finance faculty. I thought after some more recent bursts of adverse press attention, I would bring some real numbers to bear on the issue. I cannot speak for Google and Yahoo, though, but only for my own area: academic finance at Anderson. And it's time to clarify with some more transparency what is going on here. It is not a secret cabal.

PS: What I write here applies to finance. I do not know enough about economics overall. Finance is also different, because unlike economics, it is not regularly offered in undergraduate programs.

Over the last two decades, we managed to hire only 3 women (out of a total of about 25 hire, staffing an area of about 15 faculty): Monica Piazzesi, Liu Yang, and Andrea Eisfeldt. All three are great. Monica left for Chicago and then Stanford. Liu went back to D.C. (Maryland), because of a joint career problem. And Andrea Eisfeldt is still with us, but visiting a hedge fund in CT (where she is paid many times more than academics are paid). We hope to get her back. [2015: We did! Hurray!] All three are the right people to ask whether the gender climate at Anderson was negative in any way. I would hope that they would say that Anderson was great for them—and as encouraging (or discouraging) as it is to all professors.

In 2013-14, we could hire one "rookie." So, we invited for job interviews at the AFA meetings about 50% women (which is not easy, as you will see below). We made four (simultaneous) offers, two of them to women. We only got one acceptance: a man.

What is going on here?

PhD Application Pool

Candidates for PhDs at top international universities in finance tend to apply to all top PhD programs at the same time. Thus, the Anderson PhD pool is likely representative of the global pool of applicants to the top 50 finance PhD programs. In 2012, 242 candidates applied to the Anderson finance program. Our PhD program accepts about 10-15 students each year, and about 3-5 enroll. (The others usually turn us down for competitive programs, like Stanford, MIT, etc.) Unlike in some academic disciplines (such as psychology or sociology), there is really very little "diversity perspective" in finance as a discipline. There is really no minority or female perspective on net present value. (There may be some niches, but they are few and far in between.)

The finance PhD program is highly quantitative, so test scores are believed to predict admissions and success. Scores are also objective, which is an advantage when evaluating 242 applicants from vastly different backgrounds on the basis of resume-padded documents. (Verbal test scores are difficult to compare, because foreign students tend to score lower on verbal tests.) Therefore, I simply calculated the highest percentile from the overall GMAT, the quantitative score, the verbal score, the analytical writing score, and the analytical score. This procedure should still favor domestic candidates, who thus have an extra shot at being ranked highly. Only 15% of applicants that applied with complete provided test scores were women. In the top end of the applicant distribution, there were 64 students with at least one score above 95%. 8 of these 64 applicants (12.5%) were women. Expanding this classification yields a simple table:

Percentile 95 96 97 98 99
Total Applicants 64 58 40 34 15
Female 8 7 4 2 0
Pct 12.5% 12.1% 10.0% 5.9% 0.0%

In my mind, all of these test scores are appropriate for admission, but higher test scores are more appropriate. How should we choose?

In 2012, the Anderson School’s finance area accepted 11 PhD applicants. One of our admits was a woman (who ended up going to Stanford). The program enrolled 3 students in 2013, all male. Only one of our current PhD students (in any year) is female (Mindy Zhang).

Anderson finance may have an indirect admission bias in favor of US-based applicants, because we need to be able to judge candidate performance and the meaning of recommendations.. It is easier for us to understand the application attributes of U.S. applicants. For example, recommendation letters are easier to interpret if we know who the person is who wrote the letter of recommendation. This bias does not extend to foreign students already in the US, much less to U.S. students of certain ethnicities. (The majority of our applications are foreign, so this favors (does not harm) our effort to have a more representative and diverse pool.)

I mention this because “foreign diversity” issue can make “gender diversity” issues even more difficult. In 2012, all 8 women out of the 64 top candidates with at least one score above 95% were applicants from China and Korea. Frankly, our finance PhD program does not receive (m)any applications from top-scoring women from the U.S. (If you are one, please apply next year!)

Naturally, we would expect that, further down the pipeline, the percentage of women with doctorates applying for faculty positions should be similar to the number of women entering PhD programs. The pipeline intake suggests we should see women faculty applicants at a rate of about 10-15%. (As a profession, we have a more serious problem: why are there so few women applying to our PhD program [especially US women] in the first place? But this is not an issue that we can solve at the faculty hiring stage.)

Recent Graduates in the Academic Market for Finance PhDs

The second piece of evidence comes from my effort to collect faculty availability data by gender directly. I conducted an informal survey, requesting basic statistics about women graduates from PhD programs. I contacted mostly individuals who I knew, with a strong preference towards surveying women and minority faculty. I have no reason to suspect that they would not tell me the truth. Here is what I found: The number of women students in current PhD programs is similar to the number of women that apply to PhD finance programs, i.e., about 10-15%. The same 10-15% ratio applies to successfully graduating and placed students from these finance programs. (Finance PhD programs suffer attrition on the order of 20-40% from the beginning of the program to completion.)

As one of the individuals who I contacted for the survey pointed out, economics PhD programs offer another pipeline into finance. There may well have been more women economics PhD students that applied into the finance market. There is also some foreign inflow and outflow into the assistant professor faculty market. Thus, this evidence should be viewed as suggestive, not definitive.

The third piece of evidence in my study is the Directory of Finance professors at Ohio State. Unfortunately, it is not gender coded, and it is often difficult to classify correctly foreign first names. My best attempt suggested about 25 out of 260 listed faculty from graduating years 2006 to 2011 were women. This number is in line with the 10-15% estimates that seems to be best.

Professional Summary

We have a universal country-wide pipeline problem in finance and academic finance. The shortage has come about well before we in academic finance areas even get to look at applications into our PhD program (and thus 4-6 years later at applications for faculty positions). Only 10-20% of candidates are female. Almost none are minority. I wish we could fix this, but I don't think we can. We simply already sit too late in the pipeline. We simply need more STEM (Science-Technology-Engineering-Math) high-school and college students that are female and minority, with an interest in economics and finance.

Interestingly, longer-term, we may be facing a reverse issue. There are now more women graduating from universities than men. (It is an open secret that many university admission departments are conducting affirmative action in favor of men.) The underrepresentation-of-women problem could possibly fix itself, especially if we can direct more college undergraduates towards STEM. Alas, there is one thing that will always be hard to fix: it is difficult to bring up a family and have a 12-hour/day career at the same time; and from what I have casually observed, conditional on having children, women have been more likely to (want to) drop out from the workforce than men. This has reduced the number of tenured senior women faculty further.

Personal Summary and Dilemma

I am personally offended by the implication that I am a member of a cabal (of professional financial economists) that does not make enough effort to admit women and minorities. Most of the time, it is implicit. Sometimes it is explicit.

I am not a grudging consenter to diversity. I love diversity. I want us to be more diverse, perhaps even outright Bohemian.

I admit I may be tone-deaf. It escapes me how anyone with a brain could want to discriminate against women or minorities. Then again, our home business was run by a woman, my mother. My wife has a Ph.D. My boss, our business-school dean, is a woman. The best university president I have been with was Ruth Simmons at Brown. Many of my RAs have been women. (Unfortunately, in my MBA finance classes, the male-female ratio is already 2-to-1. I hope it is not me.) My kids are racial mixes. I myself am an ethnic (but not racial) mix. My friends are of all types and creeds. I prefer rainbow-like worlds—although I still wish I had more minority friends. To me, diversity is an interesting aspect, just like an interesting history or background.

Alas, for anyone who has a sense of justice: recognize that there is a dilemma.

We want more female and minority professors, but we also need to be fair and admit students and faculty based on merit and promise. I have no problem considering applications more favorably when the candidates had to persevere through more difficult circumstances, which can have bearing on both merit and promise. (For minorities coming from poorer socioeconomic backgrounds, differential admission choices for them [and them only] are easy to justify.)

Now, Let's say that we have ranked all the candidates. Is it just and right and proper to discriminate against X in favor of Y, because X happens to be white? Or black? Or asian? Or african? Or woman? Or transgender? Or muslim, jew, hindy, bhuddist, christian, mormon, shaman, or atheist? Or illegal? Or alien? Or ... ?

How many X should we skip over in order to get a Y? 5? 10? 100?

Obviously, it is more easily defensible when Y is as good as X (Y=X); but, when X>Y, then how can we not be fair to X?

By the way, when both applicants are equally good (X=Y), is it clear that we should always take X over Y? Or but the other way around? This is just because other individuals, also of person X's background (that X had nothing to do with and does not benefit from, either), are more or less numerous? Lotteries can be fair ex-ante, but they are not ex-post. (Someone equally deserving loses.) The just-described choice is not even a reasonable ex-ante lottery, though: we just exclude one in favor of the other on the basis of race, gender, etc.

I don't have the answers. I hope our candidates will not be judged by the color of their skin, but by the content of their character, and the merit, and promise of their application—and what they had to overcome and achieve in order to make it to this spot.