The University of North Texas is going to build a new 30,000 seat stadium with luxury suites (of course). It's scheduled to open in 2011.
"If you look at America's great universities, you'll see that they all have the three A's in common: great academics, great arts and great athletics," said UNT president Gretchen M. Bataille in a release. "All are key to a vibrant alumni community and continued growth. And all require great facilities. I am committed to ensuring that UNT, like many of the nation's best research universities, strives to be excellent in everything we do."
Supporters argue that athletic success can benefit colleges in many ways. It can provide positive advertising for the colleges. It can create a sense of community among students. It's a consumption product that students (as well as faculty and staff) find desirable.
One tangible offshoot of these benefits comes in the form of donations made to a college. There is some evidence that having a "successful*" football and men's basketball team leads to more donations being made to colleges. However, there is also lots of evidence that this type of athletic success does not lead to increased donations. Moreover, even if success leads to more donations, it is not clear if these are new donations or if they are donations that would have gone to academics. Lastly, those that do find a positive impact on donations usually find a modest increase.
So the research, taken as a whole, is unclear.
But never mind the evidence. Let's build a model.
Suppose that UNT will see an increase in overall donations if it has a successful football team and suppose those marginal donations end up being around $500,000 for each "successful" year. There are lots of resources that go into making a "successful" football team. There is the coaching staff and ancillary staff, there is stadium construction and maintenance, there are recruiting costs etc. What if, at the margin, it costs $1,000,000 more to have a successful team relative to an unsuccessful team. Clearly, investing in the football team for this reason is a bad idea.
Moreoever, being successful is a risk in and of itself. Winning involves uncertainty and investment in football is going to improve the marginal probability of being successful. So in the example above, the expected donations are less than the supposed actual donations, which makes the investment even more of a bad idea.
Moral of the story - use your microeconomics and think at the margin.
*In the studies I've read that have focused on Division 1 schools, football "success" is defined as being in the final AP poll, and/or making it to a bowl game, and/or a "winning tradition" (using bowl games and poll appearances over time). Basketball success is defined as being in the final poll and/or making it to the NCAA tournament and/or having some measure of tradition. There are other measures too, but these are more-or-less standard controls for success.
Via Stephen Karlson who notes that there are schools outside of the intersection of the set of successful athletic schools and successful academic schools. Like this one.