This week I’m attending the inaugural conference of the Working Families Success Network, where one of the workshops is about income smoothing. This is an accounting practice designed to help families maintain a steady flow of resources and avoid fluctuations that can be disruptive. In recent years, behavioral economists like those at Ideas42 have expressed concern that federal financial aid is disbursed to students in lump sums, suggesting that students may face difficulty budgeting those funds so that they have enough to get through the term. Efforts such as MDRC’s Aid Like a Paycheck pilot program are intended to address this problem.
I agree that the way that federal financial aid is distributed to students is problematic and that we can and must do better. Over the last five years following a cohort of Wisconsin Pell Grant recipients, I have witnessed numerous fixable problems. But in this case, I think researchers are focusing on a very small issue relative to the big ones affecting student retention rates. Specifically, I strongly suspect that the number of students dropping out of colleges because of fiscal mismanagement of federal aid dollars is very small, relative to the large number of students suffering due to (a) ever-changing availability of financial grants and work-study, (b) annual increases in the cost of college attendance, and (c) the need-analysis used to allocate federal aid. These factors drive the significant instability in income that financially-vulnerable undergraduates face, and we need broad efforts to address them now.
Consider the student who decides to attend college in 2013 and is provided with a financial aid package that includes the federal Pell Grant, a state grant, an institutional grant, and federal loans. Let’s say the student manages to make it on that aid plus a part-time job, and persists to enroll for a second year.
The costs that student faces in the second year vary depending on all of the following factors:
This list is long and yet probably incomplete. Yet it makes the point: if any of these factors changed, or if the changes together do not result in no net change from one year to the next, it is very likely that the cost of college is going to rise as a student moves through college. Moreover, the student who is progressing through coursework and pursuing programs or majors is facing more difficult academic circumstances while trying to survive on fewer resources.
When volatility occurs, the consequences are significant. Consider what just happened in Georgia, when the state raised academic standards for the HOPE scholarship: half of them did not return for another year of education. What a complete and inefficient waste of state resources. The costs of college attrition for colleges and society are high enough, and the loss of those 11,000 undergraduates just added to them. I have similar data from Wisconsin, including from my randomized trial, that I am not yet at liberty to share or publish, but which shows the same impacts. Undergraduates from low-income families do need their income smoothed to encourage degree completion, no question about it.
Unfortunately, the current approach to addressing smoothing seems aimed at “personal problems” rather than policy solutions. Financial coaching and budgeting are very individualistic solutions. Structural change is needed. While “Aid Like a Paycheck” is a proposed structural shift, it also creates administrative burden at the institutional level that most colleges cannot tolerate. Instead, the structural changes needed should occur at the federal and state levels. For example, we might recommend:
Yes, these are big lifts, but they will also generate big impacts. We should undertake some demonstration projects to verify this, to be sure. But overall, I think we have an obligation to partner to tackle the real challenges, rather than spend time tinkering around the edges. The risk of the tinkering is that it leads policymakers to believe there are ways to skate by with cheap and simple solutions, which this nation knows full well simply isn’t true.