For some of these things, there are market tests. Consider the performance of equally capable graduates twenty years hence. The usual metrics for prestige have no effect on their salaries or occupational status, ceteris paribus. For other dimensions, it's harder to evaluate. Consider the time spent by faculty testifying before legislative bodies or acting as expert witnesses? Does that detract in any way from their research? Or are they making more productive use of their research.
The problem in a way is that American universities and colleges don’t, can’t, think in any centralized way about what allows them to function well, what maximizes their internal productivity and generativity.
Some of it is the result of very deep-seated internal contradictions about what kinds of productivity is meant: productivity of knowledge, of engagement with the world, of numbers of “student units” churned out, of reputation?
More of the problem is that the consumers of higher education and its products (whether students, employers or the public sphere) don’t really know how to evaluate the relationship between external reputation and internal process (something that some of the discussion of Drezner’s case illustrates). They don’t know what goes on inside a university, or what it is that faculty do.
Consider, further, that there are enormous common and joint costs. A professor does some reading. That turns into a course revision and an article. To what cost center shall we charge the reading? Or does it matter? The article attracts the attention of someone at another institution. That turns into a job offer, or an opportunity to participate in a seminar, or perhaps into a public service opportunity. Again, to what extent does that reflect well on the internal organization of the university?