A terrific new paper , presented today at the annual meetings of the Population Association of America, provides evidence of what some of us have long suspected: the kids who are least likely to attend college are the most likely to benefit from a college degree. Jennie Brand and Yu Xie contradict the Nobel prize winning economist James Heckman, among others, with their rigorous look at the usual contention that people who get the largest returns to a degree are most likely to go to college (because people are such rational actors, you know…). Instead, their complex yet easy-to-follow analysis finds 1) heterogeneity rather than homogeneity in economic returns to a four-year degree and 2) “negative selection”– meaning that returns decline as the probability of earning a degree increases. Their data comes from three sources (NLSY79, NLS-72, and WLS) and they employ many tests for robustness. While the paper is not yet published in a peer-reviewed journal, I’m sure it will be soon. This paper has spectacular implications for public policy. Most importantly (as Mike Hout pointed out in today’s discussion) it contradicts the common assumption that an expansion of degree-holders will result in declines in the average returns to a BA. Instead, Brand & Xie’s evidence indicates that expansion would result in an increase in the average returns, since expansion would primarily involving moving the least-likely to attend into college.
This is social science at its best: thought-provoking and informative. Keep up the good work!