Andrew Kelly, who will soon join UNC President Margaret Spellings’ staff, devotes a recent Forbes column to the future of Pell grants.

It’s encouraging to see that Paul Ryan’s anti-poverty agenda reserved a prominent place for the Pell Grant program. Postsecondary education is a key path to economic mobility, and Pell is the main tool the feds have to help low-income students access it. In the past, congressional Republicans have targeted Pell in spending cuts, leaving other expensive and ineffective subsidies—like higher education tax credits—untouched.

To be sure, there are valid concerns about the rapid growth in the cost of the Pell program. And reforms are necessary to ensure it encourages students to invest wisely and stay on track while providing institutions with greater incentive to focus on student outcomes. Ryan’s plan calls for some worthwhile changes, though only in broad strokes. Taking steps to put the Pell program on sound financial footing and modernize it to meet the needs of today’s students is good policy and good politics.

Why? Because Pell is a targeted, portable voucher—exactly the kind of solution that conservatives have proposed in many other areas of education and social policy. The grants help our most disadvantaged students choose a postsecondary option that best fits their needs.

In contrast, federal loans and tax credits spread subsidies across a wide array of students, and research suggests they distort the market. Federal student loans go to anybody who wishes to borrow—regardless of their income—and carry an implicit interest subsidy. And higher education tax credits primarily benefit middle and upper-income families with large tuition bills and high tax liability. Researchers have found little evidence that tax credits affect college enrollment decisions or the choice of where to attend.

To be clear, conclusive studies of Pell’s effect on educational outcomes have proven elusive, in part because the feds have never called for the type of rigorous evaluation that has become commonplace in social policy. Grant aid seems to affect enrollment, but we have only limited evidence that Pell shapes persistence and success.

Yet some recent research, from a private scholarship program in Wisconsin and a state grant in Florida, suggests that additional grant aid can increase the likelihood that poor students finish a degree.