Job creation without higher government spending (including education spending), inflation, or trade barriers: http://jobcreationplan.blogspot.com/

Right. My example (incorrectly) assumed that just as many students receive aid as don’t. There is a distribution, of course, but for illustrative purposes I ignored it. You could use the “correct” one, but the underlying point would be the same: there are winners, there are losers, and evaluating current levels of tuition assistance relative to what would be ideal requires carefully measuring what winners are winning, what losers are losing, what changing the level of tuition assistance would do to those numbers moving forward, and then you’d have to make some value judgements (e.g., “would a 4% tuition hike for 100 non-recipients be worth an 8% reduction for 50 recipients?” and the like)

It’s just not enough to ask “Does Federal College Aid Really Help or Is It Just Capitalized into Higher Tuitions?” (the title of Jared’s post). It’s obviously not all capitalized into higher tuitions, but that doesn’t mean it “really helps.” The policy effects and value judgements are more nuanced.

What I mean to say, I guess, is that there’s more in it than, “If we decrease one student’s tuition, then the other students’ tuitions are all going to go up by the amount of that aid absorbed into the tuition.” That is, if you’ve got one student getting aid and an arbitrary number of students not getting aid, then the amount of the aid recapitalized into the tuition is going to be spread out among all of the other students. If most of the students receive aid, then the ones who don’t are going to bear a disproportionate burden (and perhaps get nudged up into an area where they do need aid, intensifying the feedback loop if indeed the feedback loop is there).

Another lesson that you can’t fight economics. Not until we get decent financial incentives to teach (at any level) will our schools and universities improve.

I think this short-changes the Center for College Affordability and Productivity’s “Bennett Hypothesis 2.0.” Andrew Gillen doesn’t baldly claim that all federal aid gets absorbed, just that certain types of federal aid gets absorbed in certain circumstances, and he carefully defines those circumstances. For instance, Pell Grants work better for poor college students than Grad PLUS loans do for law students. Distinguishing the effects of different types of federal aid is not something researchers have been doing.

But is that all the cost shifting that matters? Is our goal really to say that not *all* of our federal aid is going to administrators and (some) professors (etc.)?

For example, say we subsidize the tuition of a low-income student to the tune of one dollar, 80 cents of which will be absorbed into the now-higher tuition at the private, for-profit school he’s chosen to attend, because that’s the best school he got into, and federal aid is going to cover it anyway. I think this is actually a reasonable thing to assume, given that both demand for and supply of higher education is actually rather inelastic! (if it’s not, just adjust that 80 cents number downwards to whatever you like, and consider my argument based on that number).

OK, so now his tuition is effectively 20 cents cheaper. Great… it cost $1 to make his tuition 20 cents more affordable. We have now subsidized the education of a low-income student, in an extraordinarily inefficient way. This seems like a raw deal for taxpayers (does this kid’s parents pay taxes..?). Plus, we have now subsidized a private, for-profit, and I should add *tax exempt* corporation, using that 80 cents. Gross!

And what about all those other students who are *not* low income, or who *don’t* get that $1 in federal aid? You know, like, middle-income students. Or, *most* students.

Well, their tuition is still 80 cents more expensive, isn’t it. They didn’t get any aid, but the school is still 80 cents (or whatever number you like, the elasticities aren’t 1 you know) more expensive. They (or their parents) will now have to either dissave more, or borrow more.

I think you have to make some tough value judgements here: how do we want to spend taxpayers money, who do we want to subsidize, and are we comfortable with any negative effects that has on people we’re *not* subsidizing? I’m not saying the federal government shouldn’t provide tuition assistance, I’m just saying it’s more complicated than you’ve laid it out here. And perhaps more importantly, what other policies besides Spending Money could we explore to improve education? I mean really, an 8-year-old could come up with tuition assistance as a worthwhile policy — economists can do better than that.

Full disclosure: I am (or was) a middle-class grad student at a private, for-profit school, $65k+ in debt.