Will free college lead to more degrees?

In August, presidential candidate Hillary Clinton announced her plan to reduce the cost of higher education. Following on the heels of President Obama’s January 2015 proposal to make community college free, a key component of Clinton’s plan is to improve higher education access and affordability by reducing the amount of tuition paid by students who attend public colleges and universities. Some states, including Tennessee and Oregon, have already implemented free community college policies. Similar to other need and merit-based financial aid programs, students will receive the tuition subsidies if they meet certain academic benchmarks (such as maintaining a 2.0 GPA, in the case of President Obama’s plan). Tuition-free policies will no doubt increase enrollment. Studies have found that reducing the cost of college through both need and merit-based aid programs increases the probability of enrollment from 3.6 to 4 percentage points per additional $1000 worth of aid.

However, if the goal of these programs is to increase degree-completion, which is the outcomes that most benefits students and society, then increasing enrollments by reducing the cost of college is only the first step. Research examining students who entered college in 2003 reveals that 54 percent of those in the lowest income quartile did not earn a degree or certificate within six years. The power of need-based financial aid programs to improve persistence and degree completion is less certain than the power of such programs to increase enrollment. Dynarski’s 2003 study on the loss of the Social Security Student Benefit Program does not find statistically significant evidence that the reduction in financial aid decreases degree completion. On the other hand, other research finds positive effects of programs such as the Pell grant and other state-based grants on students’ persistence and degree attainment.  

Though some research suggests that need-based financial aid programs do affect the probability that students will persist through college and complete degrees, some of the additional federal investments in higher education which are currently being proposed should be more explicitly tied to behaviors and outcomes that we know improve student success. One option is to put some of the funding towards performance-based scholarships.

In some sense, financial aid programs such as the Pell grant and the proposed tuition subsidies are performance-based, because they require that students maintain particular GPAs and course loads. However, there is a time-lag between the end of the semester, when a student accomplishes a particular GPA, and the next semester, when they lose their financial aid if they have not made satisfactory academic progress. This time-lag may limit the power of this kind of aid to directly affect student behavior if students do not associate the loss of aid in one semester with their academic performance in the previous semester. Another option is to use financial aid dollars to reward students for earning a particular GPA, completing a certain number of credits, or demonstrating other behaviors that may increase student persistence and completion, such as meeting regularly with an advisor.

In recent years, the MDRC has run a number of experiments, testing the effectiveness of exactly this type of performance-based scholarship.  Experiments were run in six states, at eight institutions, and included over 12,000 students. These experiments compared the effectiveness of multiple different scholarship designs on different populations, and the program often targeted at-risk students, such as those with children. Participants in the treatment group were paid (cash in hand, on top of need-based financial aid) if they attained certain benchmarks, including maintaining a “C” average or better, and earned a certain number of credits.  Finally, scholarships were paid to student based on performance in the current term, regardless of performance in past terms.

At most of the sites, students in the treatment groups earned more credits by the end of the term or semester than students in the comparison groups. In Ohio, students earned 1.7 more credits by the end of the year (the maximum scholarship amount in Ohio was $1800). Students in the treatment group were also more likely to earn their degrees. The number of students in the treatment group who earned degrees two years after random assignment was 3.6 percentage points higher than in the comparison group. The evidence from these experiments suggests that need-based financial aid that is immediately tied to academic outcomes can improve persistence and completion.   

This is not to say that all need-based financial aid should be packaged like these performance-based scholarships. Students also need financial aid dollars up front, at the beginning of the semester, in order to pay tuition and living expenses. On the other hand, the higher education affordability plans that President Obama and candidates Clinton and Sanders are proposing call for additional federal investment in higher education, beyond that already devoted to Pell grants. Some of this additional funding could be put towards performance-based scholarships that incentivize persistence and completion, rather than simply reducing the enrollment cost of college.

There is no doubt that these higher education affordability plans are going to increase enrollment, and make college more affordable for some students.  Recent enrollment numbers suggest that the Tennessee Promise has increased fall enrollment by almost 25 percent. However, a more efficient use of additional state or federal dollars invested in higher education would be to tie some funding explicitly to outcomes that we know lead to increased rates of college completion such as credit accumulation.