Monday, August 12, 2013

New Evidence on Need-Based Grant Aid

Ben Castleman and Bridget Long of Harvard University just issued a terrific new paper on the impacts of a Florida need-based grant distributed to students across the state.  Using a rigorous regression-discontinuity design, the authors make several contributions to the study of the impacts of financial aid by tacking a couple of of tough questions:
  • Does need-based aid promote college completion?
  • Who benefits most from need-based aid? Is it the highest-achievers to whom merit aid is often targeted?
They find that YES, need-based aid (without any performance criteria) produces strong and statistically significant impacts on credits earned and degree completion.  Specifically the authors find that $1,000 more of grant eligibility increased the probability of staying continuously enrolled through the spring semester of students' freshman year by 3.3 percentage points, increased the cumulative number of credits students completed after four years by 2.3 credits, and increased the probability of earning a bachelor’s degree within five, six, and seven years by 2.5, 3.5, and 4.0 percentage points, respectively. This is very similar to what my team is learning from studying a Wisconsin grant program.

They also find that the impacts of need-based aid are strongest for students who did well in high school but are below the cut-off for the Florida Bright Futures grant, again mirroring findings from Wisconsin. 

Critically, the authors also note that most students got this need-based grant just once-- as in Wisconsin, many lost it between the first and second years of college. How much more effective might financial aid be if we made it easier for students to retain it?  

In short, the answers do not surprise me at all and lend important empirical evidence to a debate that has been tilted towards merit & performance aid mainly because of a lack of tests on the need-based aid.


Policymakers:  Please read this carefully before jumping to the conclusion that you must change the structure of need-based aid to promote college completion -- it is already doing so.  Changes could be positive or could undermine effectiveness.  In fact, the authors conclude: "Overall, our results suggest that not only does need-based aid have a positive effect on persistence and degree completion, but also that increasing the award amounts of current aid programs could have beneficial effects." 

States and Colleges and Universities:  Investments in well-prepared students who aren't making your merit cutoffs are good bets for cost-effective investments in need-based aid.  Mark Schneider and I have been saying this for a long time.   Consider making your money really pay off.