Last spring while I was finishing my fellowship at Columbia Business School, much of the student body was busy trying to overturn the school’s grade disclosure policy. Back then, Columbia was one of the few top MBA programs that did not practice grade non-disclosure, meaning recruiters were allowed to ask Columbia students about their grades. By the end of the year, the issue had passed a student referendum, and this semester Columbia became the latest business school to have a grade non-disclosure policy, which encourages students not to disclose their grades to employers until they’ve been hired.
Grade non-disclosure policies are a quirk of MBA programs. You won’t find them in medical or law school. In fact, the only place you do find them is among top business schools. Of the 15 most selective MBA programs, 9 of them have some form of a grade non-disclosure policy. But of the schools ranked from 20 to 50, none do.
A new paper from a pair of Wharton economists examines why this is. Wharton, it should be noted, has grade non-disclosure, and while the faculty (according to the authors) is fairly vocal in its opposition to the policy, it’s typically approved by the students each year by a wide majority. The paper begins with a discussion of what economists call signalling, where trusted information is transferred through signals we send. The job market is one of the leading areas where we do this, as explored by economist Michael Spence in his 1973 signalling model. The idea is that without knowing whether you’re good at a job like valuing bonds or putting together marketing proposals, employers rely on the signal we send by way of an MBA from a top business school.
The authors argue that students at these schools “reduce the accuracy of their signal by passing grade non-disclosure policies.” Why would they do that? Students typically argue that non-disclosure policies allow them to take greater risks and harder courses without having to worry about an embarrassing transcript. But in the face of evidence that self reported levels of learning have mostly fallen since the introduction of grade non-disclosure policies, this seems a little dubious. According to a Wharton Journal article, surveys showed a 22% decline in the time students spent on academics during the first four years after the school passes a grade non-disclosure policy.
The authors say the real reason behind grade non-disclosure goes back to signalling, and the risk-management properties of pooling:
We show that students at elite schools are the most likely to adopt a non-disclosure policy, subsequently reducing their effort. Intuitively, a non-disclosure policy allows the median voter to study less and then pool to receive the expected (mean) wage, which might be more valuable to her than receiving the median wage with effort. For plausible wage distributions, the desire to pool becomes more valuable at more selective schools.
A vote for non-disclosure, however, allows the median voter to essentially “free ride” off of the expected pooled wage, which will be advantageous when there are enough students who are more productive than the median voter.
Since the signal they’re sending to employers is already strong, students at top b-schools are more likely to vote for non-disclosure if it reduces their effort without substantially reducing their perceived value to employers, which is already pretty high as they were able to get into a top program. Students at less-selective schools, though, don’t feel they have that luxury.
In response to grade non-disclosure, some schools have implemented minimum grade requirements to encourage effort, or at least punish an egregious lack of it. For example, Wharton students are dismissed if they score in the bottom decile in at least five credit unit courses during their first year, or eight credit unit courses over two years. Strangely though, the authors find that minimum grade requirements actually increase student support for grade non-disclosure by enhancing the incentive for collusion, and decreasing the value for students to compete against each other.
As long as employers keep valuing a degree from Harvard Business School over MBA’s from mid-tier programs, grade non-disclosure policies aren’t likely to go away. Are we surprised then that the smartest business students in the country have figured out a way to game the system in their favor?