If you want to give money to an educational institution, [my emphasis] do some research and find a charter school in your metropolitan area that’s obtaining good results with a demographically unfavorable group of kids. Or find help our a regional public college of little repute that provides valuable educational services and could really use the money. Sure, if your checkbook is fat enough to finance a research endeavor that could make a major contribution to discovering an HIV vaccine or something it might make sense to invest in a world-famous university.
The correct response -- viz. that Harvard is not primarily an "educational institution" -- was given at some length by Sean Carroll last year, before the endowments went bust:
But there is one misimpression that people seem to have, that might as well be corrected before any hasty actions are taken: the purpose of Harvard is not to educate students. If anything, its primary purpose is to produce research and scholarly work. Nobody should be surprised that the gigantic endowment isn’t put to use in providing top-flight educational experiences for a much larger pool of students; it could be, for sure, but that’s not the goal. The endowment is there to help build new facilities, launch new research initiatives, and attract the best faculty. If it weren’t for the fact that it’s hard to get alumni donations when you don’t have any alumni, serious consideration would doubtless be given to cutting out students entirely. [...]
Don’t believe me? Here is the test: when was the last time Harvard made a senior tenure offer to someone because they were a world-class educator, rather than a world-class researcher? Not only is the answer “never,” the question itself is somewhat laughable.
The implications of this defense for charitable giving should be understood. A lot of institutions in the fine arts are supported as charities; Harvard etc., seen primarily as research farms, are equivalent institutions in the humanities and sciences. A great deal of research in astronomy and pure mathematics, as well as most of what goes on in humanities departments (one should note that lots of Pulitzer Prize winning writers teach at Harvard, Princeton, etc. as a day job), is "worthwhile" because a small minority of the populace values the results. Such activities used to be funded by the patronage of aristocrats, and later by that of industrialists hungry for culture. This is, if you think about it, the only sustainable form of funding: such activities aren't profitable; besides, the public isn't interested in astronomy or poetry, and it's unfair to use their tax dollars to fund such frivolities.
On the other hand, the public does care about education, feeding the poor, curing AIDS, etc.; therefore, these are sensible enterprises for the government to run, and charitable giving serves largely as a figleaf for the state's more egregious failures. In an ideal world, the government would make sure that people weren't starving and that kids were able to attend a good college; however, it would still make sense for research universities to be sustained by charitable giving. In the real world, of course, there are lots of government failures and charitable giving to feed the poor is arguably necessary, but it's still reasonable for rich people who are interested in the arts and sciences to endow universities and museums.
Besides, as Carroll says,
Students will always keep applying to those places and trying to get in, because the aura of intellectual attainment produced by precisely those scholarly accomplishments will always act as a powerful draw. Such students are by no means making a mistake; the intellectual atmosphere at such places truly is intoxicating, and if nothing else the interaction with your fellow talented students can be a life-changing experience.This is related to why overzealous attempts to recruit underprepared students to Ivy League universities are misguided. Nobel Laureates are, by and large, good at teaching classes on Chaucerian irony or quantum field theory to well-prepared students; however, remedial classes are best taught by people who are good at teaching, and don't end up at Harvard.