Fix Free College: What We Can Learn from Place-Based Promise Programs

At what should have been a sleepy school board meeting on November 10, 2005, in Kalamazoo Michigan, the school board instead unveiled a new program that would transform their town forever. A group of anonymous donors had pledged that any kindergartner that graduated from high school in the district could attend any Michigan public college for free. This Kalamazoo Promise program has been copied, in one form or another, by at least 83 school districts across 37 states. And now, Hillary Clinton has made her own promise--she wants any student from a family earning less than $125,000 a year to pay no tuition in his or her state. How these promise programs, particularly Kalamazoo Promise, have affected their communities offer some insights into how Clinton’s plan might fare.

The Kalamazoo promise was designed to boost K-12 enrollment and completion and to help revitalize a declining urban area. Indeed, many of these programs have shown enrollment and student behavior benefits for K-12 school districts. But the Clinton campaign has emphasized free college as a way to achieve affordability in higher education. And a national free college plan would likely not have the same benefits for local school districts. For that reason, our analysis focuses on the benefits that accrue to students in college, including their enrollment, completion, and earnings after college.

College enrollment went up, particularly for low-income students.

Because the Kalamazoo Promise is based in a city with a sizable low-income, minority population, it benefits a much different population than a national program might. However, the impact on postsecondary enrollment has been encouraging. An analysis of Kalamazoo Promise by the Upjohn Institute found that the scholarship significantly increased postsecondary enrollment six months after high school graduation. This was particularly true for low-income students: Before the program, students eligible for free or reduced lunch lagged their more affluent  peers in college enrollment by 25 percentage points, after the program started, that gap closed to 10 percentage points. It also encouraged more students to enroll in four-year colleges at the expense of community colleges.

Unfortunately, the effect doesn’t last. Just 12 months after high school graduation, the number of students enrolled in college is not statistically different than before the promise took effect. However, these numbers are not the result of poor retention. Instead, a lot more students enrolled more quickly after graduating from high school instead of waiting until the spring semester to enroll.

The bottom line for a national free college program:  A national program will shift students towards four-year public colleges and may speed their enrollment in college after they graduate from high school.

There was increased college completion, particularly of bachelor’s degrees.

The research also found that the number of graduates earning any credential increased by 12 percentage points, going from 36 percent before the introduction of the scholarship to 48 percent after. And most of that increase was in the attainment of bachelor’s degrees. The size of these impacts was similar for low-income students and their more affluent peers. The Kalamazoo Promise only increased credential attainment six years after high school graduation and did not affect attainment in four years.

The bottom line for a national free college program: A national free college program will likely increase college attainment, particularly of a bachelor’s degree.

There was a big return on the Promise investment.

The Kalamazoo program is funded by generous, anonymous donors, allowing them to cover the high costs of tuition at Michigan colleges for Kalamazoo graduates in perpetuity. The program spends between $10 and $11 million a year to serve around 1,400 students. But the analysis by the Upjohn Institute found that the program generated an increase in wages resulting from improved credential attainment were 4.7 times higher than total spending. They found the highest rate of returns for non-white and low-income students.  

The bottom line for a national free college program: With the increase in attainment, there should be an increase in wages for people who got degrees they would not have otherwise. However, the researchers acknowledge that Kalamazoo Promise spends a lot of money on students who would have gone to college even without the program. This would be exacerbated in a national program that wasn’t focused on a lower income, diverse school district. And adding means testing, even the high-income threshold ($125,000 a year) proposed in the Clinton plan, adds complexity to applying for the program that could discourage low-income students from participating.

The benefits of promise programs are unequally distributed.

Yet the Kalamazoo Promise scholarship does not benefit all eligible students equally. Because the scholarships cover tuition at whichever institution student’s choose, the costs to providing free tuition vary widely based on where a student enrolls. While the average student eligible for free or reduced price lunch received about $10,000 in scholarships through the Promise, the average scholarship for more advantaged students was about $24,000.

The same pattern holds for race. The average white student received about $23,000 toward tuition payments, while the average non-white student received about $12,000 in scholarship money. These differences are also affected by factors such as how long a student remains enrolled in school, and what level of degree they are pursuing.

The study also found that, nationally, students who grow up under 185 percent of the poverty line can only expect a 70 percent increase in lifetime wages for receiving a bachelor’s degree while someone who grew up above that line can expect a 135 percent return. While a 70 percent return (over a high school degree) is still significant, this data shows that students across the country who grow up low-income do not benefit equally from degree attainment. This data should caution policymakers who think that improving attainment will completely fix the lack of economic mobility in this country.

The bottom line for a national fee college program: Even though it was focused on a diverse and more low-income area, Kalamazoo Promise ended up granting the largest subsidies to those who needed it the least. We’ve written previously that this would be true of the Clinton plan in the aggregate given existing enrollment patterns.

Promise programs like Kalamazoo’s provide more evidence about the benefits and possible weaknesses of national free college. The next posts in this series will address how we can change the policy design of free college to maximize these benefits and address the weaknesses.

This is the second post in a series on ways to #FixFreeCollege. The rest of the series is available here. 

Correction: An earlier version of this post indicated that wage increases exceeded program costs by 11 percent. Indeed, wage increases exceed costs by 370 percent; 11 percent is the internal rate of return to the program. This post has also been updated to reflect the fact that while changes 12-month enrollment are not statistically significant, this is not the result of poor retention of Kalamazoo Promise recipients. Additionally, it has been updated to reflect the definition of low-income students as those from families earning less than 185 percent of the poverty line, not 125 percent. Finally, the earlier version of this post suggested that low-income students "saved" $12,000, which assumes those students would've enrolled in college and paid out of pocket in the absence of the Promise, which is not necessarily the case.