Nearly every Democrat running for statewide office this cycle will propose some type of “free” college plan. Those plans need to be well thought-out and, in particular, recognize the relationship between affordability and degree completion. Otherwise they are apt, at best, to under-deliver, and at worst do more harm than good.

Accordingly, every free college plan should be measured against five key metrics: whether the plan includes measures to improve college preparation during high school, whether the plan covers both two-year and four-year public colleges, whether the plan covers additional costs beyond tuition, whether the plan covers efforts to decrease both dropout rates and time of degree completion, and whether the plan covers students of all income levels.

Fortunately, there is a strong base of research that has helped highlight what is required to support a real college promise plan that will lead to heightened levels of timely degree completion.

1. Quality High School Academic Preparation

High school academic preparation is more predictive of bachelor degree completion than any other pre-college characteristic; it’s more predictive than race, family income, or parental education.

Academic preparation variables like curricular rigor (i.e. taking a college preparatory course track), test scores, and GPA account for 78 percent of the difference between completers and non-completers (figure below). Race, family income, gender, and teen pregnancy explain only 22 percent of the variation.

Unfortunately, a large percentage of high school graduates, including those from middle and upper middle-income families, are underprepared for college. One-in-four rising college freshmen, those entering postsecondary education in the fall term immediately following high school graduation, are placed in at least one remedial course at the “college” level where they are taught material that should have already been mastered. The average student is placed in two.

Underprepared high school graduates are 74 percent more likely to drop out of college than non-remedial placed peers. Among bachelor degree seeking students that do graduate, they take an average of 11 months longer to complete their degree than non-remedial students. And among associate degree seekers, remedial students take an average of 6 months longer to complete their degree. For bachelor degree recipients, this opportunity cost translates into roughly $150,000 in lost lifetime earnings and retirement savings compounded over 45 years.

The real shocker is that contrary to conventional belief nearly half of underprepared high school graduates come from middle and upper income, not lower income, families. Nearly half are enrolled in four-year institutions as opposed to community colleges.

A high-quality, free college promise plan needs to support all students before they enter college so they are academically prepared to succeed and go on to earn a degree in a timely fashion

2. Covering Two- And Four-Year Public Colleges

Many recently adopted free college promise plans focus on the costs of attending a two-year institution under the assumption that students seeking a bachelor’s degree can enroll at a community college and then transfer to a four-year institution after completing their two years.

While two- to four-year institution transfer pathways may work for some students and carry intuitive appeal from a cost-efficiency perspective, the research suggests beginning at a two-year college with the intention of completing a four-year degree decreases the likelihood a student will get a bachelor’s degree by approximately 30 percentage points. Peer effects, inadequate on-campus student support services, and the difficulty of transferring credits all combine to derail completion, never mind on-time completion.

A 2017 Government Accountability Office (GAO) report found that, while over a quarter of public community college students transfer to public four-year colleges, on average they also lose 22 percent of previously earned college course credits in the process. Only 14 percent of all degree-seeking students who begin at a community college go on to earn a four-year degree within six years of initial enrollment.

For those who wish only to pursue an associate’s degree or non-degree certificate, a promise of free community college combined with quality secondary school preparation is an excellent policy design. But the research does not support funneling students into community college as a cost-effective way to support bachelor degree attainment. For capable bachelor-degree seeking students, a college affordability plan that promotes initial enrollment in a four-year public college over a local community college is a superior path.

3. Covering All College Costs

Paying for college doesn’t just require paying tuition and fees. Room and board living expenses, books, and supplies all need to be financed in order to meet degree attainment goals and can easily cost far more than the combined cost of tuition and fees.

In 2017, the average published tuition and fee price at public four-year colleges for full-time students was $9,650, while the cost for room and board living expenses was $10,440. After average federal, state, and institutional grant aid—net price for all college costs—for the typical full-time public bachelor’s degree seeking student still totals over $14,000 per year. Putting that figure into perspective, median household income in the United States is just shy of $60,000. In other words, after all grant aid, the average bachelor’s degree seeking student at America’s public colleges still needs to pay out-of-pocket nearly one quarter of their household’s pre-tax income each year to cover college costs.

When remaining out-of-pocket costs, beyond tuition and fees, are unmet by financial aid packages, students typically work hours that are self-defeating for achieving their educational goals, borrow large sums, or both. To be clear, students who work modest amounts—10 to 15 hours a week—while enrolled full-time, on average do better academically than those who do not work outside of school at all. The former have been shown to take their studies more seriously and manage their time better, but too much outside-of-school work dramatically undermines academic success. Students who attend four-year colleges on a consistently part-time basis are nearly five times more likely to drop out as their peers who attend full-time.

The data reveal another important phenomenon: the undeniable disparate racial impact associated with college aid policies that fail to meet full financial need. Certain racial and ethnic groups appear to have economic and cultural perspectives that make them more debt averse. Hispanic students, for example, are significantly less likely to borrow in order to meet unmet need, resulting in part-time enrollment, full-time work, and thus, a higher likelihood of dropping out.

For those who do choose to borrow to pay for living expenses, books, and supplies, there are significant economic risks. If borrowers do not complete their degrees, they are four times more likely to default on their student loans. If they default, they can almost never discharge the debt in bankruptcy. Their likelihood of obtaining employment goes down because of credit checks that often accompany job applications. Their likelihood of marrying goes down. Their likelihood of owning a home goes down. In short, debt with no degree to show for it is one of the worst outcomes college students confront.

Any real “free college” or “debt-free college” plan must meet the financial need for students’ full cost of attendance (i.e., room, board, and other expenses included). Well-crafted college affordability plans allow students to enroll full-time and focus on their studies, thus facilitating faster degree completion.

4. Support and Accountability for Institutions of Higher Education Efforts to Boost Completion

It can’t be stated enough that similar colleges serving similar students often generate widely dissimilar completion outcomes. Consider the University of Florida that graduates students at a rate over 20 percentage points higher than University of Alabama (87.5 percent versus 65.7 percent), even though they are both flagship public universities in neighboring Southern states with similar median SAT scores and similar incoming high school student GPAs. There are scores of colleges that compare in a similar, disturbing manner like Cal State Fullerton and Rev. Jerry Falwell’s Liberty University (guess which is dramatically worse for racial minority students).

The good news is there are other examples of colleges ranging from CUNY’s Lehman College, to Georgia State University, to the University of North Carolina, that all significantly have improved completion rates or on-time completion rates utilizing a combination of increased state financial support, institutional program changes such as guaranteed course availability, and improved student support services, such as intensive academic advising and mentoring driven by use of predictive data analytics.

A quality college affordability plan ought to include direct aid to under-resourced colleges to further institutional support for students—better advising, scheduling, and counseling—so that students stay on track with coursework and make wise decisions that facilitate finishing on time. Arguably, one should go further, insisting that increased institutional aid come with mechanisms of accountability attached. While institutions could be held accountable in a number of ways, one would be to require reasonable completion goals be met within three or four years after the receipt of funds or risk exit from any statewide plan. California already includes a minimum completion requirement for institution participation in its Cal Grant program. And more than 30 states presently use some form of performance-based funding in their distribution of public higher education resources.

Targeting of Resources to Families That Need Them Most

Finally, the data is overwhelming in showing that higher education can operate as an engine of economic mobility, but also as a means of calcifying income inequality. Too often, it ends up being the latter.

Individuals with an associate’s degree are over 50 percent more likely to earn above $60,000 per year by age 35 than individuals with just a high school diploma. Bachelor’s degree recipients are over 250 percent more likely to earn over $60,000 a year by age 35 than individuals with just a high school diploma. And the wage premium associated with college degree attainment is growing markedly with each generation.

Unfortunately, students from low-income families are far less likely to complete a bachelor’s degree by age 24—the age a student traditionally matriculates straight to college after high-school graduation and then completes a bachelor’s degree within six years—than those from upper-income families. Worse, the gap has only widened. And for non-traditional students, the numbers are even worse, across the board.

In 2015, only 12 percent of individuals from the lowest income quartile held a bachelor’s degree by age 24, compared to 58 percent from the highest-income quartile. Put another way, a child born into a high-income household is nearly five times more likely than a person from a low-income household to hold a bachelor’s degree by age 24.

One can, and should, reject the notion that college affordability plans must be limited only to the poor or hard-pressed middle-class families, but we should also recognize that additional resources for college affordability are arguably limited. Therefore, within imposed constraints, time, attention, and money should be focused on those furthest behind as opposed to those most politically powerful.

It is better to provide a much larger, progressive, need-based package of college affordability and completion support than a small, regressive package available to all students. The latter is less likely to result in successful completion, and could lead to great long-term harm to a number of students, as well as taxpayers, in the form of heightened dropout levels.

Conclusion 

So-called free college plans have been passed or implemented already in New York (the tuition-free “Excelsior” plan for middle-class students attending SUNY and CUNY campuses), Tennessee (the “Tennessee Promise,” which provides a mentorship program and helps cover the cost of tuition and fees at Tennessee community colleges and colleges of applied technology), Rhode Island (free tuition for nearly all students going to the state’s community college), and Boston (the “Boston Bridge” program of free community college tuition to low-income students in the city).

Conceptually, these plans represent a major step forward. For years, families have overestimated the price of public colleges and underestimated their eligibility for and the size of financial aid. All colleges do not cost the near $70,000 a year sticker price that Harvard lists (even Harvard rarely charges that much) and every family – even Bill Gates’ family – is eligible for a federal student loan, not to mention institutional aid (some 80 percent of private college students get a tuition discount). The maximum Pell Grant is almost $6,000 a year and middle-income families are able to access tax benefits like the American Opportunity Tax Credit worth up to $10,000 over four years. The genius behind free-college efforts in New York, Tennessee, Rhode Island, and Boston is that they cut through those details and complexity underlying college financing to send a loud and clear message to all families that “Yes, You Can go to College.”

The remaining free college movement policy challenges are to improve high school student academic preparation, cover costs beyond tuition and fees at all public colleges, assure institutions of higher education are genuine partners in boosting completion, and reduce the risk of student debt for the lowest income families especially. College affordability programs that couple rigor with resources to meet these challenges offer improved chances for successful completion to the benefit of all.