Let’s start in your early years. You finished your BA at California State University. How did you come to study economics?
That’s a good question. I originally planned to major in history and I made an appointment with the chairmen of the Department of History to talk about it, but he happened to be out of town. And I never rescheduled. In the meantime I had a friend who said, “I took a really interesting course called Principles of Economics. You might want to take it.’ And so I took it the next semester and I loved it. It was a discipline that had mathematics and business. It had everything. After that I was captivated.
What motivated you to focus on the gender gap?
I took graduate labor economics and I had an interest in racial discrimination. And I knew a lot of people had done work in that field. But I thought that there had not been a lot of work done on gender discrimination—at least not by economists—and I was right.
So I started thinking about how to apply a model by Gary Becker, that he originally intended to be used to look at racial differences in labor market outcomes, and I decided that that could be applied to research on gender.
So after you graduated, it took you two years to publish your most famous papers, and you got famous for this ‘Oaxaca-Blinder decomposition.’ The decomposition has become so famous that there is now even a Stata command. How did you feel that someone had made this ?
I found out from one of my graduate students. He rushed to my office and said ‘There is a Stata command with your name on it!’. So he downloaded it on my computer, and I was amazed. In fact, it really made doing the work easier!
Beyond the gender gap, what are your other research interests? I see you are going towards crime economics?
There is still a story about decompositions in the paper I presented at CERGE-EI, but instead of looking at labor market outcomes between men and women, this paper looks at differences in prison sentences between men and women. It tries to understand how much of the gap we observe can be explained by circumstances such as women committing less serious crimes, and how much of it is unexplained, and thus may be the contribution of judges’ preferences.
So what are the conclusions from this research?
We show that, unlike in the labor market, in prison sentences women are in fact favored. There is an unexplained gap. If we look at the majority group, white males and white females, we find that male prison sentences are on average 20 months longer. When you control for the nature of crimes committed, then we can explain 14 months of that gap. But there are still six months that you cannot explain by anything other than judges having a preference in favor of women.
It is very interesting, and one can conjecture as to why that is the case. Some have said, ‘what if judges believe that women learn their lesson faster than men, so then it would be socially inefficient to put them in prison as long as men?’ There are two problems with that idea: one is that we already control for their past criminal history. If they learn their lesson quicker, they wouldn’t have a criminal history as severe as men, so we already controlled for that. The other problem is that even if we really believe that’s true, there is no way you can implement that as part of the legal system. Because where does that end?
Is there any other research you are interested in?
I’ve worked on discrimination based on monopsony, and it’s a new model by Alan Manning. The idea there is that if the labor supply elasticity of men and women are different to a firm, and if the labor supply elasticity is less for women, you can show that if a firm wanted to maximize profits it would make sense for them to pay women less than men who are equally qualified.
So you actually gave a reason for why a company would pay women less.
Well yes, but it’s pretty interesting. So we have this case of less elastic labor supply for women at the firm level. But for the labor market as a whole, it’s the opposite. The labor market elasticity for men to a labor market is less elastic than for women; the difference is that there is a labor force decision that has to be made, and that’s what makes the labor elasticity supply for men less elastic, even while it’s more elastic to an individual firm.
Where do you see a research gap in economics?
I see a gap in our attempts to understand labor market discrimination. We have refined measurement techniques. But even if we come up with econometric techniques which we feel are truly measuring discrimination, most of the time the technique itself doesn’t tell you the source. It only identifies the gap.
How should a research question be chosen?
I think that scientifically, we feel that one ought to sit in the corner of their office and think deeply, come up with a model, and then go out and fine the data. But this is not a practical way to come up with a topic. Quite often what we actually observe is an academic saying, “here is a very nice data set which is available for research” and then starting to think about what kind of models can be developed that will take advantage of that dataset. I know scientifically it should be the other way around, but as a practical matter it’s often not the best way to proceed. But there are two views on this.
Most CERGE-EI students are from transition economies. What is their competitive advantage in terms of pursuing research questions?
CERGE-EI was set-up as a way to try to understand economies in transition. Now, many years later, we might argue that none of the central European economies are in transition. So one might ask, ‘where do you go next after coming out of that transition research?’ I think one possible example is developmental economics. It is right now on the frontier again. When I was a student, it became a very hot field, and then it got quiet. But now it is back, and new work is being done. Field work, experimental work, a lot of work in Africa. I know a lot of people at CERGE-EI have an interest in that. And I think from the background of transition economies at CERGE-EI, this is a natural path that someone might take at the institute.
What is your impression of CERGE-EI ?
Almost from the very beginning I knew about CERGE-EI and its mission. And it has become very prestigious, even from the beginning, so I’ve always had a very high opinion of the Institute.
My impressions today are exactly the ones I had ex-ante. I’m very impressed with the students, the faculty, and the program. Nothing has changed in that regard.
Ronald L. Oaxaca is the McClelland Professor of Economics at the University of Arizona, Tucson. His areas of research include labor economics, applied econometrics and applied microeconomics
Interviewer: Jasena Kukavcic
Monday 8 October 2012