Tag Archives: signaling

What is College All About?

What is a university for? This comes up quite a bit when my wife and I discuss the future of online education. I am a bit more bullish than her, but being the common sense person she is, I know I have to pay attention to what she says. She is not against online classes per se, but wonders how they can replace the university because she thinks the university is as much about learning as it is about socializing and learning how to communicate with people. As generally is the case, she is right.

When David Brooks attempted to answer the question in a New York Times piece, he mentioned technical knowledge and practical knowledge. Both are important for sure, but the picture is somewhat incomplete. A more complete answer comes probably from Sal Khan (Alex Tabarrok of the Marginal Revolution fame has argued a quite similar point here.) In One World Schoolhouse, which I reviewed here, Khan argues that the university primarily performs the following three functions:

  • Teaching/learning;
  • Socialization;
  • Credentialing.

Notably, research is not on this list. Khan does acknowledge that research is an important function for the society, but he implies, correctly, that it doesn’t have a first-order impact on the student. This especially rings true these days because most star researchers don’t even teach classes anymore. This is simply a result of misaligned incentives. Outside maybe MBA settings, you don’t get much by being a rockstar teacher and your time is better spent on producing yet another paper. The teaching is mostly left to adjunct professors, lecturers, or even PhD students. Let’s look at this from the perspective of the student. Does the student really care, or does it benefit him in any way, that the faculty is home to five Nobel laureates? Not really, unless the student takes a class with them, which is generally not the case.

Now, having those star researchers on the faculty is certainly good for reputation. But this already manifests itself in the credentialing function, and its natural extension: signaling. Once you have a degree from that university, you indirectly benefit from the cache of that university. Research, then, is arguably a second-order benefit for the student, even though it is still a first-order benefit for the society as a whole.

Now, Khan’s vision is that teaching/learning and credentialing should be separated. In the world Khan envisions, credentialing is not necessarily a function performed by the university; instead, it is accomplished through rigorous assessments, which Khan calls microcredentials. Khan is quick to point out that this does not eliminate the need for the university, instead it makes it optional like an MBA. Khan also notes an unemployed adult can use these assessments to signal that he has the necessary skills.

I think he is spot on with this vision and this may be a path to solving some of our broader problems, such as the long-term unemployment. There is some recent evidence that a person who has relevant experience but has been out of the long force for more than six months may be skipped over for someone who has been employed even without relevant experience (If you are so inclined, read the original research here). This may be rational for the employer, but bad for the society as it creates sustained long-term unemployment.

To be clear, the problem does not have an easy solution but Khan’s vision would be a good first step in that direction. If a prospective employee can signal his competency through these assessments, that could increase the chances of him getting hired. At the college level, it provides a sustainable path for the student who doesn’t have the money, or connections, to get into a good school. More than anything, what the society needs is reassurance that the American dream is still alive, and meaningful long-term change can come through successful unbundling of the functions of the university.