The University of Virginia has generous financial aid but few Pell Grant recipients in its student body. (Dan Addison — University of Virginia)

As college costs have risen astronomically in recent decades (with benefits following in turn), the federal government has taken it upon itself to try to stem the tide. Even the most observant policy watchers could be forgiven for having trouble keeping track of all the programs and subsidies at work.

Perhaps the most famous is the Pell Grant program, but there’s also the American Opportunity Tax Credit, the federal direct loan program for student debt, subsidized Stafford, PLUS, and Perkins loans, and the Lifetime Learning Tax Credit.

Then there’s the tax-exempt status of scholarships and interest from state and local government bonds used to finance public higher education, as well as the deduction for charitable contributions to colleges and universities.

That's a lot of money, so much that, by friend-of-the-blog Mike Konczal's estimation, you could probably use it to make public higher education as free as public K-12. But is it making college more affordable?

Some researchers argue no. In fact, they argue that government subsidies have had the perverse effect of encouraging colleges to increase tuition so as to capture more federal dollars, an effect known in the literature as the "Bennett hypothesis". To see how this could work, imagine a school that charged $0 in tuition and fees. Suppose that the federal government then offered $5,000 vouchers to pay for college tuition, but only at schools that charge tuition. Obviously, our hypothetical free school is going to increase tuition to $5,000 a year to take advantage of that money. That's just money on the table.

The empirical evidence on whether or not that's happening is somewhat mixed, but very few studies find no upward effect on tuition from subsidies for any category of schools. Even the most skeptical find some areas of cost pressure. And the most recent work is even more dispiriting. University of California - San Diego grad student Nicholas Turner (now at the Treasury Department) found that tax-based aid, like the stimulus's American Opportunity Tax Credit, crowds out school-provided financial aid dollar-for-dollar, so students barely benefit at all. University of Maryland professor Lesley Turner found that 16 percent of the value of Pell Grants is captured by schools rather than students.

Source: New America Foundation

Now, the New America Foundation's Stephen Burd is out with a bracing report arguing that schools are actively trying to recruit high-income students with merit aid and other enticements, and making it harder and harder for low-income students to attend in the process. The share of high-income students receiving aid — be it federal, state, or institutional — went from only 13 percent in 1995-96 to 18 percent in 2007-08. Meanwhile, despite things like Pell Grants, 34 percent of public schools and 89 percent of private schools still charge a net price of over $10,000 a year to low-income students.

More troubling still, many schools with low net prices aren't enrolling many poor students. Despite the fact that Georgia Institute of Technology charges an average of $0 to low-income students, Pell recipients make up a paltry 17 percent of its student body. Same goes for elite private schools that make a big fuss about their generous financial aid packages. Harvard charges low-income students an average of $1,297 a year but Pell recipients make up only 11 percent of the student body. Even schools that have the resources to keep prices low are, perhaps not surprisingly, still overwhelmingly enrolling rich kids who can pay more.

On the other end of the spectrum, there are schools that recruit a lot of poor students and then charge them gobs and gobs of money.

It's not all bad news. The real stars of the report are liberal arts colleges like Amherst, Vassar, Grinnell, and Williams, which enroll a sizeable number of Pell Grant recipients and charge them very little. But even Amherst only has 22 percent of the student body on Pell Grants.

I'm going to be delving a bit more into these issues soon, but Burd's is probably the best treatment I've seen to date. Read the whole thing. Felix Salmon also has some useful comments.

Update: A database glitch led to Mills Colleges' net tuition being misreported in the above graphic. They actually charge $17,730 a year. New America is updating the graphic; we regret the error.