Adjusting tuition through grants and scholarships is a form of price discrimination, one long researched in economics, and practiced for finding students of merit. Grant McCracken observes a note on the practice in yesterday's WSJ and writes:

So why are they discounting the price of the �product.� It can�t be to drum up more business. They�re oversubscribed as it is. I think it�s to create more generous alums. Down the road, better students should enjoy more accomplished careers, earn bigger incomes, and give bigger gifts to alum mater. It�s a long term bet, but universities are well positioned for long term bets.

That certainly would be a hypothesis based on rational self-interest. It could also be that schools want high application rates so that they can be more selective (and look better in rankings by US News and the like), and handing out a few more merit scholarships early on leads to more applications down the road. But McCracken continues,

Would this be the time to think about quality control? There is no point bringing better students to campus, if we are going to inflict a substandard education on them. They will graduate unhappy. Or they will just leave and end up giving their alum dollars someplace else.

So it�s time to do something about those academic dead beats. You know the ones I mean. (If you don�t have them as colleagues, you had them as teachers.) Almost every department in almost every university has academics who just gave up years ago. Usually, they don�t teach very well. Usually, they hardly think at all. Now in mid-career, they appear to struggling to qualify as late entrants in that rather large club identified by George Bernard Shaw. �Most people would rather die than think. Most do.�

The answer to this is that it may be cheaper to keep handing out merit scholarships than to clean up the academic deadbeats. The problem is large. And if there are enough good faculty around, the deadbeats are circumvented. The good faculty attract good students. I'm amazed how many of the students I have in economics have had the same faculty in other departments, the ones I think are good teachers and researchers. But it makes sense: Universities have sorting processes that shunts poorer students towards poorer faculty.

McCracken tries to assign a cost to incompetent faculty and there's no doubt they cost, but unless you account for students' strategic behavior in sorting towards good faculty when the students are also good, you've overstated the problem. Still, I think McCracken has a useful exercise here, for the reason he concludes:

If we can get this number right, we can calculate not only the costs inflicted on alumni support created by Dr. Lunatic, but the amount that must be laid at the door of university administrators who refuse to move against him. Now we have a metric that can be used to assess performance in the high offices of the university. Dr. Lunatic has no shame. If he had, he would have restored himself to usefulness years ago. But Dean Robertson? Actually, she has pretty active sense of pride. If we say she is costing us hundreds of thousands of alumni dollars by suffering Dr. Lunatic, there�s a pretty good chance she will do something about it.

This can apply by extension to showing the numbers to the boards of trustees of campuses with university administrators who will not act against their incompetent faculty. And it should.