Sunday, August 22, 2010

Should Your Freshman College Roommate Be Your Clone?

I am in Novato, California. My friends at UCLA won't see me on campus for a few more weeks. In this post, I'd like to return to an old theme that is at the intersection of economics and sociology. The topic is the costs and benefits of living and working in a "diverse" environment. Each fall, millions of 18 year old first year college students are assigned by some unknown dean to live with a stranger (the roommate). How should your precious 18 year old be assigned this stranger? Random matching? Matching on observables (a love of guitars and tattoos)? In the past, random matching (with the exception of not pairing smokers and non-smokers) was the norm. Now, the NY Times reports on the new "marriage market" as potential roommates interview each other and shop around.

If you bother to read the article , you will see a righteous quote from Dr. Dalton Conley on his opposition to this new reality. He hints that students are choosing to avoid exposure to diversity by choosing a clone today. This benefits the suburban student in the short run but will cost him/her in the long run.

Now, as you know --- Dora Costa and I have written a serious applied econometrics paper on this very point. Using the Civil War as our laboratory, we document in this 2007 JEH paper the short run costs BUT the long run benefits of living and working in a diverse environment. Other academics have looked at samples of university students, but they have no way of following these students decades after they finish university to see how their exposure to different types of people affects their later life outcomes.