In a speech on Thursday at the University of Buffalo, President Obama put forward proposals to help rein in the growing cost of college tuition and mounting load of student debt to help cover those costs. One of those proposals will be “paying for performance”: His plan would set up a new ratings system that would measure colleges’ performance on a variety of metrics and then eventually tie federal student aid to how well they perform. This would, in theory, incentivize institutions to improve on metrics such as tuition costs, graduation rates, and how much student debt their graduates carry.
But will it work?
It turns out that many states have already experimented with tying aid to performance, and there is some evidence of how these plans play out. According to a report from the Center for American Progress, 26 states tried this model out between 1979 and 2007, but many dropped it thanks to design flaws like rigid and arbitrary requirements or failing to allocate enough funding to really create an incentive to make changes. The idea has regained popularity in recent years, however, with proponents rewarding progress over completion, recognizing differences between community colleges and universities, and dedicating more funding. Now Ohio, Pennsylvania, Indiana, Tennessee, Washington, and Louisiana have implemented these models.
While these policies have not been in effect for very long, making it difficult to measure the outcomes, there do seem to be improvements. The CAP report points to Pennsylvania in particular as a success story. It set aside 8 percent of its budget, or about $36 million, to reward schools based on performance. It measured success on eight indicators such as how many students complete degrees, the retention of students, and faculty productivity, all of which had to be met to get a share of the funds. Those that exceeded the requirements got more money. Since it went into effect in 2000, the state’s public colleges have seen a 10 percent increase in graduation rates and a 15 percent increase in the retention of Hispanic students. College officials also say there is an institutional culture more focused on efficiency.
After some hiccups with its first version, Washington has also seen success with a second iteration. Its model gives colleges money for each metric of achievement that they meet through supplemental funds, which leaves the base funding untouched. Points are awarded based on the number of students who improve their scores on skill tests, complete a math course, earn a certain number of credits, complete a degree, and others. Between 2007 and 2011, while this program was in effect, colleges increased their point totals by 31 percent on average. Yet only about half of students contribute to getting points.
Another study looked at the research on these programs as a whole to see what the impact has been. Evidence suggests that tying funding to performance has the immediate effect of raising awareness in institutions of state goals for higher education and of their own performance as well as increasing competition among them. This has meant colleges are using more data and have changed academic programs, staffing, and courses as well as student services.
But the evidence of whether all of this leads to better outcomes — such as higher graduation rates and fewer students dropping out — is mixed and there is little data to sort through. Florida, Washington, and Ohio saw big jumps in graduation rates, but it’s unclear how much of this was due to performance pay. Tennessee didn’t improve retention from 1995 to 2003, while Ohio administrators claim they did but there is no objective data.
The Center for American Progress collects some best practices that could help a federal plan void pitfalls and experience success like Pennsylvania and Washington. It suggests that these models have to set aside enough funding to actually create an incentive to change. It also argues that stronger incentives will come if the money is taken out of base funding, rather than tacked onto it. They should track progress, rather than base everything on completing goals. Phasing the system in and gaining support from institutions can boost success. But given the scant data on these programs, Obama may have to just try it out to see.