Orin - Exactly!. It would take decades for us to see what the consequences of such an experiment would be. And there would be the potential for so many confounds in there (other social, technology, climate changes that would also take place in the interim) that attribution to the scholarship/no scholarship variable might be hard. And what of US News? All of this is interesting as a thought experiment. And I think you put your finger right on the problem. But we don't have enough information. So with what we know, what school (especially in the middle ranks, neither the top or bottom) could responsibly recommend "first we get rid of all the scholars"? Few moves would prove as likely to result in an immediate plummeting of reputation, while having almost no impact on cost (unless we replace all the faculty with adjuncts). And that would surely hurt job prospects for our students. My point in the above is that calculating the value of scholarship, and thus by extension comparing its value to other costs, is not as easy as some suggest because it accumulates slowly and sometimes without easy means to count or quantify. That does not mean it does not exist.