June 20, 2014 - June 20, 2014
By Susan Dynarski and Judith Scott-Clayton
Any college student who wants a federal loan or Pell grant has to file a Free Application for Federal Student Aid. With well over 100 questions about income, assets and expenses, the Fafsa approaches the IRS Form 1040 in length, and is longer and more complicated than the 1040A and 1040EZ, the tax forms filed by a majority of taxpayers.
Senators Lamar Alexander and Michael Bennet, in a New York Times Op-Ed on Thursday, said they were introducing legislation that would significantly simplify the process of applying for aid for college. The legislation would slice the Fafsa to just two questions: family size and income.
As economists, we have worked on this topic for years. The senators cite our research in their Op-Ed, and the proposed legislation closely hews to aproposal we wrote in 2007. Recent research provides strong evidence that the complicated aid application process discourages people from applying for aid and going to college.
Four economists conducted a randomized trial in which potential students (or their parents) were provided with assistance in completing the Fafsa. A control group received information about college costs and financial aid, but no such assistance in filing the form. The result: The college enrollment rate of low-income high school seniors rose 8 percentage points, while among older people with no college experience enrollment rose about 2 percentage points. These are big boosts, both given the low enrollment rates of low-income people and given the low cost of the intervention: less than $100.
You can pick up this stick from either end. You can try to help students navigate through a complicated system (as did the Fafsa experiment described above), or you can eliminate the complexities so students don’t need the help. We’re in favor of the latter, as is the Alexander-Bennet proposal. It’s a waste of social resources to have government build a complicated bureaucracy that then requires an army of high school counselors, college financial aid staff, community-based workers, private consultants and volunteers just to help people navigate through it.
Advocates have been sounding the bell about complexity in the aid system for some time. But many were hesitant about moving forward, since it seemed impossible to target aid to the neediest students without the detailed data about finances that the Fafsa collected. Financial aid officers often told us that the complexity of aid was a necessary evil, without which we could not target aid to students with the greatest need. The Fafsa needs to be long, they argued, so that we can precisely measure who needs aid the most.
We decided to take this argument at face value and measure empirically how much complexity in aid applications contributes to the targeting of funds. We examined detailed data from thousands of aid applications and aid packages, calculating how the distribution of federal aid would shift if we were to drastically scale back the Fafsa.
How much does complexity help with targeting? The answer surprised even us. The vast majority of the questions on the Fafsa contribute nothing to the targeting of federal aid. It turns out that income, plus family size, turns out to be enough information to let you closely mimic the current distribution of federal aid.
We ran a lot of regressions to get to this conclusion, but the results were straightforward. If you know a potential student is very poor — say she grew up in a family with income of $15,000 — then you know she is eligible for the maximum Pell grant. You don’t need to ask how much her parents have in their (probably nonexistent) investment accounts. Similarly, if you know a potential student comes from a family with income of $100,000, you know that she is not eligible for a Pell Grant. You don’t need to ask about her summer job.
When we compare aid eligibility based on the current lengthy Fafsa to what it would look like using just a few questions, the differences are minor. We found 74 percent of students would see no change at all in their Pell, while 91 percent would see a change (increase or decrease) of less than $500.
We don’t argue that the current distribution of aid is perfect. Some might want more money to go to those with low incomes, for example.
Our point is that you can get needy students more aid with a lot less paperwork, and a lot more transparency. Evidence indicates that this will make our aid programs much more effective in doing the job they were designed for: getting more people into college.