As the amount of student debt in the United States increases and concerns about students default proliferate, policymakers are considering many options for reducing the debt of those seriously in need of help. One such idea is “risk-sharing.” Under this proposed policy, if a student defaults on his or her loan then the university he or she graduated from is responsible for paying part of the loan.

How would colleges and universities respond to a policy like this? A new paper by Temple University economist Douglas A. Webber tries to answer that question, at least in part. One potential consequence of risk-sharing would be an increase costs for schools. If they’re on the hook for some loans then the marginal cost of admitting a student would go up.

Webber uses data from the Integrated Postsecondary Education Data System, or IPEDS, from the U.S. Department of Education. The data cover the 1987-1988 school year to the 2010-2011 school year, but not every year for every school in the data. He then uses the data to create a model that estimates the effects of risk-sharing.

The amount of increase in tuitions varies quite a bit depending upon the kind of higher-education institution. Colleges and universities that have higher default rates and students with more loans and larger amounts of loans would see the largest increases in tuition. And because those kinds of institutions are concentrated in the for-profit sector, tuition at for-profit should would jump up the most.

Webber’s model tuition at for-profit schools predicts an increase of 1 to 2 percent under the less-stringent form of risk-sharing or by 3 to 4 percent under the more-stringent risk-sharing arrangement. The increases for other types of schools would be much lower. Student-debt from the for-profit sector would decrease by $13 million per institution per year under the less-stringent arrangement and $80 million under the stronger form.

But these calculations assume that institutions would do nothing to reduce the default rate of their students. As Webber points out, this is hopefully an unreasonable assumption about the behavior of schools. History shows it to be unrealistic as universities and colleges seem to have cut default rates in response to a 1991 law change punishing institutions with high default rates.

What happens to Webber’s calculations once we assume schools will try to reduce default rates? Assuming a 10 percent drop in default rates, the increases in tuition would be smaller across all types of colleges and universities. And the decrease in student debt would be considerable larger, at about $42 million to $130 million, depending on the level of risk-sharing.

The plan does have some drawbacks. In particular, Webber finds that risk-sharing would result in a decline in the number of students attending and graduating from institutions of higher education. He points out that a reduction in college graduates might not be worrying because the higher prices will signal to some potential students that college might not be an economically sound investment for them.

But this assumes that the students turned away by higher prices would necessarily not be a good fit for college. We know that isn’t true. Or the savings from fewer defaults and loans could be plowed back into government budgets to reduce tuition prices. Unfortunately, that would only work for public colleges or universities.

So while risk-shifting seems like a policy idea with some merit, we have to be aware of how it might interact with other potential problems in the higher-education system. The unintended consequences are always in the details.