In the past few years, dozens of states, localities, and schools have proposed or enacted “free college” policies, also known as College Promise programs. While a handful of states have run Promise-like programs for decades, the 2014 launch of the Tennessee Promise and the Obama administration’s focus on the concept catalyzed several states to pursue their own versions of a statewide free or debt-free college proposal:

  • A total of sixteen states now have at least one statewide Promise program, with two states running two different versions of a Promise program.1
  • Of those sixteen states, ten have enacted and funded a Promise program since 2014, with eight states enacting a Promise program in 2017 alone.2

These numbers make it clear that, after decades of decline in the percentage of state budgets going to higher education,3 Promise programs are becoming an increasingly common pathway for states to pursue urgently needed new—though frequently narrow—investments in higher education.

Often spurred by a broader desire to grow state economies and provide greater access to economic mobility and financial security for their residents, policymakers pursue two policy goals when designing statewide Promise programs: (1) to address a growing concern around rising college costs and student debt burdens felt by a wide swath of students, and (2) to capture the positive effect that a clear affordability message can have on spurring college attendance amongst students who might not otherwise enroll, or who might qualify for aid but not realize it. Because Promise programs are easy to explain, and might reach more people and a wider political constituency than typical financial aid programs, they also have the potential to build levels of public support similar to those held by universal public benefits like K–12 education.4 And finally, it is worth noting that some policymakers have begun to include requirements in Promise programs in order to pursue objectives less directly related to the core goals of financial aid or college affordability measures.

The specific policy design choices made by policymakers, combined with the level of funding allocated, will ultimately impact how well they meet those core goals. Promise programs have spurred states to make welcome investments in higher education, though these initial first steps have often been small. At least in their initial stages, few states have recharged their higher education investments enough to make significant progress toward a more universal benefit—and each state faces their own unique hurdles to getting there, some more challenging than others. Without that investment, as states launch programs with rationing policies to contain program costs, the choices they make will have very different impacts on who benefits, how well it measures up against the goals of spurring enrollment and lowering debt, and how their program impacts the progress their state makes in closing gaps in enrollment and attainment rates by race and income.

This report first identifies and analyzes the range of design choices made by states and the challenges and opportunities created by those choices, finding that states that have enacted Promise programs since 2014 have imposed significant eligibility requirements. It then provides guiding principles to help states considering future Promise programs to avoid inequitable cost containment mechanisms, maintain a clear message to students and families, and ultimately make bolder investments in higher education.

Defining and Tracking State Promises

What constitutes a statewide Promise program? There is no single definition, but for the purposes of this document, Promise programs are distinct from existing state financial aid in that they provide at least free or debt-free tuition to a significant subset of students who are not chosen based primarily on merit considerations.5

The structure of Promise programs differ from most state financial aid programs. Non-Promise financial aid programs generally structure eligibility criteria and award levels to meet one of three objectives: to provide students that demonstrate financial need with a discount on tuition and fees or on the total cost of attendance, to provide high-achieving students with free or discounted tuition (merit aid programs), or to direct discounts on tuition or cost of attendance to certain populations, such as foster youth. The amount of the award, or discount, is rarely defined as covering all or a percentage of the cost of tuition or total cost of attendance, and instead is typically determined by the revenue made available through legislative appropriations in any given year.

Finally, dozens of localities and campuses have enacted non-statewide programs in the past decade.6 Research on those programs can at times be applied to state-level program design, but this report only tracks programs funded through state-controlled dollars that are available to students across the state.

Using this definition, it can be said that sixteen states have active Promise programs. While the most recent Promise programs are too new for a robust evaluation of their impact on reducing debt and spurring enrollment, we can, to some extent, learn from decades of research on need-based and merit-based financial aid, local Promise programs, early commitment and older Promise programs, and the financial challenges facing today’s students to analyze the structure of existing programs and guide future program design.

Promise Programs Features Vary

While states often describe their programs as universal, in reality they include extensive eligibility requirements intended to either ration the benefit in order to to bring down costs or direct the benefit to certain populations of students (or both). These limitations still allow states to offer a “guaranteed” benefit, but one that is not universal to all students enrolled in postsecondary education in the state. In fact, the eligibility requirements imposed by some states limit the programs to just a small percentage of college students.


Most Promise programs offer free tuition at community colleges but not four-year institutions: only one program enacted since 2014, the New York Excelsior Scholarship, includes a guarantee of free tuition at four-year institutions.7 On the one hand, limiting Promise programs to community colleges targets aid awards to a population (community college students) that tends to be lower-income and need the support the most. The lower number of eligible students, combined with the fact that tuition and fees tend to be far lower at community college than at four-year institutions, also brings down the cost of the programs more substantially than if it covered both two- and four-year institutions.8 But one study also raises the question as to whether it may also encourage “undermatching,” meaning that some low-income students who may be able to enroll in more selective four-year institutions—where data show that they may have better outcomes on average—instead choose two-year schools.9 Another study suggests this effect may be overstated.10

Limiting Promise programs to community colleges targets aid awards to a population (community college students) that tends to be lower-income and need the support the most.

Enrollment Intensity and Timing

More than half of statewide Promise programs require students to attend full-time in order to be eligible, and eleven states have programs that restrict eligibility to recent high school graduates. Part-time students are more likely to be on their own financially and shouldering a larger share of the cost of attending college, due to lack of family support and lack of state aid support. Furthermore, almost half work forty or more hours a week, and almost 40 percent have dependents.11 The combined effect of these two limitations mean that many, if not most, non-traditional students are ineligible for their state’s Promise programs unless their full aid package makes it realistic for them to attend full-time. Research shows that full-time students are more likely to graduate,12 but today’s limited Promise programs do not provide enough funding to allow working students to cover existing financial obligations, drop their work hours, and enroll full-time.13

Today’s limited Promise programs do not provide enough funding to allow working students to cover existing financial obligations, drop their work hours, and enroll full-time.

First-, Middle-, and Last-Dollar Funding

All but one of the eleven newer statewide programs are last-dollar, meaning that they require students to first use Pell dollars and other grant aid toward the cost of tuition, and then cover the remaining gap. Oregon instead uses a “middle-dollar” approach: it provides last-dollar coverage but guarantees at least $1,000 of support to all students, regardless of whether they have other grant aid that covers their tuition. Those students can instead use that $1,000 to help cover the cost of living beyond tuition.14 In addition, the older programs in Louisiana, Oklahoma,15 and Mississippi16 all provide a first-dollar scholarships. Washington state uses a combination approach: its income-capped scholarship fills the gap between other state aid and tuition and fees, but does not take into account Pell awards.17

This design element has important equity implications. A last-dollar program will send fewer state resources to low-income students by virtue of the fact that Pell grants or other programs may already cover part of their tuition. In Oregon, even with the “middle-dollar” feature, before the state added an income cap, about 60 percent of the funding went to students from families in the top two expected financial contribution (EFC) quintiles of participants.18

In contrast, a first-dollar program would cover tuition and fees regardless of other aid, allowing low-income students eligible for Pell or other grant aid to use those dollars to cover books, transportation, housing, and other costs that students must finance while studying. Paying for living costs is likely to pose a bigger barrier to college access and reducing debt for a low-income student than paying for tuition costs might pose for wealthy students. The most positive findings on the enrollment and persistence impact of a Promise program comes from an evaluation of the local Kalamazoo Promise, which is a first-dollar program.19


Of the sixteen statewide programs considered in this report, eight contain a minimum GPA and ACT/SAT requirement.20 While the programs included here do not use the rigorous merit requirements in traditional merit aid programs like the Georgia HOPE scholarship, it is important to note that research shows that merit aid programs can have inequitable racial and socioeconomic impacts.21


Several programs created in the decades prior to the recent surge of interest in Promises limit their commitments to students from low-income families, often defined as families who qualify for free or reduced lunch.22 Only two of the more recent Promise programs limit eligibility by income. In New York, the Excelsior program cuts off eligibility for families earning over $125,000, and Oregon also added an income cap after they decided not to fully fund the original uncapped program in the second year.23

Post-graduation Residency

Three programs require residency post-graduation (Arkansas, New York, and Rhode Island). New York and Arkansas tied those requirements to the number of years that the student receives the benefit; if the student leaves the state before then, then the grant converts to a loan, a requirement with significant potential for confusion for students and families.

Student Supports

Several programs provide significant resources to support students in completing their degrees, including mentorship and community service. Evaluations of those attempts are still in their early stages.

Targeting High-Demand Fields

There is a growing trend toward offering free tuition only for college programs providing degrees or certificates in growing fields, particularly in states in the south and midwest (AR, IN, KY, MN). These requirements narrow their respective Promise programs’ reach: some states include any STEM or high-demand field, while others designate only specific programs for eligibility.

Tuition +

Only two programs cover costs beyond tuition and fees: Hawaii covers books, transportation costs, and supplies if students have unmet need, while Oregon provides a middle-dollar scholarship that gives a minimum of $1,000 to cover non-tuition costs for low-income students who already have tuition covered. Because Hawaii only covers costs for students who have unmet need, it is the only Promise program structured as a “debt-free” program, in that it requires students who can afford to pay for tuition, books, transportation, and supplies to do so, but eliminates the need for any student to take on debt in order to cover those costs.24

Table A

Download this Chart as a PDF

Key design features of Promise programs created in the past thirty years25

Year Passed


/Last Dollar

Inst26 Tuition  and/or other costs Income


Occup limits Age 


Residency post-grad Student Supports in College Part-time/Full-time HS GPA/

Curriculum Requirements



L CC Tuition/


No Yes No Yes (3) Mentoring + CS*** PT/FT No


L CC Tuition No No HS No No FT 2.5


L CC T/F/B/ TR/S** Unmet need No No No No 6 cr/




L Cert Tuition/


No Yes Indep No No PT/FT No


1st 2/4 Tuition/


$46K28 No 8th grade No No FT 2.5/




L CC Tuition/


No Yes No No No PT/FT 2.0
LA ’98 1st 2/4 Tuition No No HS No No FT 2.5/ACT/


MO ’93 L CC29 Tuition/


No No HS+4 yrs No No FT 2.5/ACT


MN ’15 L CC Tuition/


$90K Yes HS No Mentoring FT No
MS ’97 1st 2/4 Tuition/


$39.5K No HS+1 No No FT (30 credits) 2.5/ACT/curric
NV ’17 L CC Tuition/


No No <20 No No FT Men/CS
NY ’17 L 2/4 Tuition30 $100K-$125K No No Yes (=yrs) No FT (30 credits)31 No
OK ’92 1st 2/4 Tuition $55K32 No 10th grade No No PT/FT 2.5/


OR ’15 M CC Tuition


EFC33 No HS No 1st yr exp PT/FT 2.5
RI ’17 L CC Tuition/


No No HS Yes (= yrs) No FT (30 credits) No
TN ’14 L CC Tuition/


No No HS No Mentoring/


TN ’17 L CC Tuition/


No No >24/Indep No College success PT/FT No
WA ’07 F/L34 2/4 T, F, B $46K35 No 8th grade No No PT/FT 2.0
 *Additional features not included in this table: number of semesters of availability, limits around existing AA/BA/certificate holders, budgetary limitations (some run out of money), amount available at non-public institutions, SAP or GPA requirements once in college, coverage of development courses, small “co-pays,” state residency requirements before enrolling, eligibility of undocumented students, proactive notification by the state, **T = tuition, F = fees, B = books, S = supplies, TR = transportation. ***Community Service

Older Trends

Recent energy to pass Promise programs builds on two trends in state financial aid, both stemming from the 1990s and early 2000s.

Early Promises

Several states—Mississippi, Washington, Oklahoma, Louisiana, Missouri, and Indiana—passed early versions of a Promise-like program in the past two decades. Other than Louisiana’s, these programs focus on students from families earning below a certain income threshold—often tied to eligibility for free and reduced lunch programs. All but Missouri provide free tuition at both two- and four-year institutions.

Most of these programs were created as “early commitment” programs, designed to provide low-income students in middle and high school with a message that they could afford college, and to give them a pathway to applying and enrolling. Several require students to take a specific curriculum and meet minimum GPA requirements. Evaluations have shown these programs can have a positive effect on college enrollment.36 And even with income limitations, many of these programs reach the same or more students as do modern Promise programs.37

Merit Programs

The recent uptick in interest around Promise Programs also comes on the heels of a trend in past decades to offer a guarantee of free or highly discounted tuition for in-state residents meeting significant merit standards. For example, the Georgia HOPE program, Florida Bright Futures, and the West Virginia Promise all started as tuition-free guarantees, though several of these programs have reduced the percentage of tuition and fees that they cover. Research shows that these programs can be regressive and are biased toward sending aid dollars to white students.38 For example, in Florida, the addition of stricter ACT/SAT standards for the Bright Futures scholarship in 2012 disproportionately harmed black and Hispanic students.39High GPA requirements for maintaining scholarships have also been shown to cause students to avoid math, science, and other majors or courses that tend to grade more stringently.40

The Promise programs included in this analysis have lower merit requirements and are more inclusive than are traditional state merit aid programs. However, Louisiana’s program, with a 2.5 GPA requirement and ACT minimum, also disproportionately benefits white students.41 And it is possible that over time, a state Promise program might further raise GPA requirements in order to limit enrollment, moving their programs closer to the Georgia HOPE or Florida Bright Futures programs.

The Promise of the Future

More than a dozen states are likely to consider Promise program proposals in 2018. State-level renewed focus on affordability, coming after years of per-student cuts to higher education budgets, is a welcome change. But as state legislatures debate the contours of a program for their respective states, they should follow a set of guideposts to ensure their proposals do not exclude, or underinvest, in students who need the most help. Particularly when launching early stage programs, states should:

1. Avoid inequitable cost containment measures.42

  • Target limited investments. A first-dollar program will do far more to reach the students who most need aid. Short of that, a policymaker weighing a first- or middle-dollar approach that excludes the wealthiest students against a last-dollar program with no income cut-offs should pursue the former approach.43 Similarly, a state will do more to improve college affordability by covering a wider range of costs, such as tuition, fees, books, supplies, and transportation, for those who have unmet need, than just covering tuition for everyone. If policymakers do not make budgetary investments necessary for a universal or first-dollar program and must limit who qualifies, income caps that exclude the wealthiest students, a debt-free model, or at least a middle-dollar component are better choices than limiting access by, for example, GPA or age.44
  • Skip the merit requirements. Programs with high GPA or ACT/SAT cut-offs begin to look like merit aid programs, which have resulted in inequities in state aid programs. Policymakers should leave merit considerations to schools’ admissions offices.
  • Do not exclude nontraditional students. Programs open only to recent high school graduates or full-time students leave behind students who work, who have caregiving responsibilities, and who are more likely to be low-income.45 Programs should cover a prorated cost of attendance or tuition, depending on the program, and ensure that any bonuses provided to students for enrolling in more credits makes converting to full-time enrollment more feasible for a significant portion of working students.
  • Include undocumented students. Promise programs should also be open to undocumented students, who have no access to federal financial aid and limited access to in-state tuition or existing state financial aid programs.46
  • Avoid undermatching. Provide sufficient counseling resources for Promise applicants to understand the full range of options available, particularly in states with two-year free Promise programs and in states with high-quality private institutions.
  • Fund and publish transparent assessments of impact. Particularly when launching a narrow Promise program, publish and analyze the impact of the program on low-income students and students of color so policymakers can quickly address inequitable design elements.

2. Maintain a clear message, excluding useless eligibility requirements, and making use of the communications opportunity.

  • Do not include post-college residency requirements. At best, residency requirements make accessing a program confusing and overcomplicated; at worst, it changes a Promise program from a free-college program to a loan program where you might qualify for forgiveness, and further discourages loan-averse students from taking advantage of it.47
  • Keep it simple. While research on the effectiveness of Promise programs is still in its early stages, researchers have identified at least one emerging theme: the simpler the better.48 Eligibility requirements, such as confusing GPA standards, detract from one of the most promising characteristics of a free or debt-free structure. And additional hurdles such as drug testing requirements proposed in West Virginia would both complicate the program and limits an opportunity for those suffering from addiction.49
  • Leverage the spotlight. Legislators should allocate funding for counselors, schools, and coordinating agencies to use the attention created through passing legislation to galvanize a large-scale public outreach campaign about the program. Stakeholders can also incorporate resources about student support programs and benefits, such as emergency aid or SNAP, for low-income students who still have unmet need.

The recent momentum around Promise programs has shone a welcome light on the increasing cost burdens borne by low- and middle-income students. As debates around Promise programs continue, state legislators serious about spurring enrollment, lowering debt, and addressing inequities in our higher education system should ensure that proposed Promise programs provide both a clear message and a clear benefit to those who need it most. Doing so may also require a more serious conversation about revenue: while Promise programs may be free for students, building a more equitable, evidence-driven program will require a bolder investment from states—and would benefit significantly from an enhanced partnership with the federal government as well.



  1. These states include: Arkansas, Delaware, Hawaii, Indiana, Kentucky, Louisiana, Mississippi, Missouri, Minnesota, Oklahoma, Oregon, Nevada, New York, Rhode Island, Tennessee, and Washington. Two programs, Minnesota and Nevada, have been launched as pilots, so their status may change this year.
  2. Arkansas, Hawaii, Indiana, Kentucky, Nevada, New York, and Rhode Island all enacted new legislation creating Promise programs in 2017; Tennessee expanded its existing Promise program to include working adults.
  3. Michael Mitchell, Michael Leachman, Kathleen Masterson, “A Lost Decade in Higher Education Funding,” issue report, The Center on Budget and Policy Priorities, Washington DC, 2017,; Matthew M. Chingos, Sandy Baum, “The Federal-State Higher Education Partnership,” issue report, The Urban Institute, Washington DC, 2017,
  4. Many proponents also view long-term fiscal investment in aid as a goal, seeing the broader political constituency behind a more universal benefit as a way to sustain public financing over time. Elise Swanson, Angela Watson, Gary Ritter, and Malachi Nichols, “Promise Fulfilled? A Systematic Review of the Impacts of Promise Programs,” Working Paper Series, no. 16 (October 2016): 57,
  5. As discussed later, this is a murky distinction. For the purposes of this report, programs with 2.5 GPA requirements are not considered primarily merit-based. In addition, several states offer programs to students over the age of sixty or sixty-five: the subset of students participating in these programs is so narrow that they are not included in the list of sixteen statewide programs considered here. See e.g., “South Carolina – Senior Citizen Education, Elder Learning,”
  6. College Promise Campaign, “Bring the Promise of Affordable College to Your Community,” Washington DC,
  7. “New York’s Promise to Students: Ever Upward,” Excelsior Ever Upward, New York State, One other program, South Carolina, offers free tuition at two- and four-year institutions, but its other limitations—specifically the limitation to those over sixty—make it a far smaller program.
  8. Sandy Baum and Jennifer Ma, “Trends in Community Colleges,” College Board Research,
  9. In Massachusetts, researchers found that the Adams Scholarship encouraged high-achieving, low-income students to enroll in less selective institutions than they could have enrolled in otherwise. Sarah Cohodes and Joshua Goodman, “Merit Aid, College Quality, and College Completion,” American Economic Journal: Applied Economics 6 (4): 251-285, See also Laura W. Perna,”Understanding College Promise Programs,” report, University of Pennsylvania, Pennsylvania, 2016,; However, researchers have also found that this effect may be less relevant in a state where the public institutions available are on par with other options, and that a Promise program can instead push students up the quality ladder. Timothy J. Bartik, Brad J. Hershbein, Marta Lachowska, “The Effects of the Kalamazoo Promise Scholarship on College Enrollment, Persistence, and Completion,” Upjohn Institute for Employment Research Working Paper 15-229, 2017, In Tennessee, there was a 10-percent increase in freshman enrollment in the first year of the program. However, enrollment at four-year institutions decreased significantly. The decrease at four-year institutions followed a trend in decreasing enrollment that existed before the Promise program, so it is too early to estimate how much the Promise program has encouraged students to move from four-year institutions to two-year institutions. “Tennessee Promise Annual Report,” Annual Report, 2017,Tennessee Higher Education Commission, Tennessee Student Assistance Corporation,
  10. Jennie Brand, Fabian Pfeffer, and Sara Goldrick-Rab “The Community College Effect Revisited: The Importance of Attending to Heterogeneity and Complex Counterfactuals,”
  11. Marcella Bombardieri, “Hidden in Plain Sight: Understanding Part-Time College Students in America,” issue brief, Center for American Progress, Washington DC, 2017,
  12. Clive Belfield, Davis Jenkens, Hana Lahr, “Momentum: The Academic and Economic Value of a 15- Credit First Semester Course Load for College Students In Tennessee,” publication, Community College Research Center, Tennessee, 2016,
  13. About 40 percent of community college students work twenty or more hours per week. Unless Promise scholarships can make up for those lost work hours, it is unlikely that full-time enrollment will be feasible for many students. Lumina Foundation: Think you know the Facts? Today’s College Students Are…” (Print, 2015), For example, even with the Kalamazoo Promise, which provided a full-tuition, first-dollar program, the increase in average units accumulated was about two classes over eight semesters. This represents a large gain, but did not on average boost enrollment intensity such that most recipients were going full-time. Timothy J. Bartik, Brad J. Hershbein, Marta Lachowska, “The Effects of the Kalamazoo Promise Scholarship on College Enrollment, Persistence, and Completion,” Upjohn Institute for Employment Research Working Paper 15-229, 2017, The decision to drop work hours is often complicated, and small aid amounts may not make up for the risk of dropping hours or changing jobs to one that requires fewer hours. Marcella Bombardieri, “Hidden in Plain Site Understanding Part-Time College Students in America,” issue brief, Center for American Progress, Washington DC, 2017, And researchers found that the impact was for four-year schools—for two-year students, they were more likely to take fewer credits because those were the students who would not have enrolled at all (32-34).
  14. Higher Education Coordinating Commission, Office of Student Access and Completion, “Oregon Promise,”
  15. “Office of Administrative Rules,” David Lopez Ohio Secretary of State,
  16. Higher Education Legislative Plan for Needy Students (HELP), Miss. Code Ann. § 37-106-75, Mississippi provides first-dollar awards, but does not allow the total aid received to exceed total cost of attendance.
  17. Revised Code of Washington (RCW), Title 28B, Chapter 28B.118, Section 28B.118.010, Total aid cannot exceed cost of attendance.
  18. “Senate Bill 81 Legislative Report: The First Term of Oregon Promise,” 2016, Higher Education Coordinating Commission, analysis does not account for the recent changes made to the program, which will be discussed later on in the report.
  19. Timothy J. Bartik, Brad J. Hershbein, Marta Lachowska, “The Effects of the Kalamazoo Promise Scholarship on College Enrollment, Persistence, and Completion,” Upjohn Institute for Employment Research Working Paper 15-229, 2017,
  20. For purposes of this document, we draw a line between Promise program and a purely merit-based scholarship program by a 2.5 GPA requirement. The average high school GPA is 3.0, so a 2.5 at least allows a significant cohort of students earning below the average to qualify. For this reason, the Louisiana TOPS scholarship, often described as a merit program, is included as a Promise program.“America’s High School Graduates,” The Nation’s Report Card, NCES 2011-462, 2011, National Center for Education Statistics, United States Department of Education,
  21. Susan Dynarski, “The Consequences of Merit Aid,” National Bureau of Economics Working Paper 9400, 2002,
    The HOPE scholarship’s 3.0 requirement was shown to grow the racial gap in enrollment. See also
    Donald E. Heller, Patricia Marin, and Gary Orfield, “State Merit Scholarship Programs and Racial Inequality,” report, The Civil Rights Project Harvard University, Massachusetts, 2004, (Dynarski, cited in Bangs, Davis, Ness, Elliot & Henry, 2002 Bridget Terry Long, “What Is Known About the Impact of Financial Aid Implications for Policy,” working paper, National Center for Postsecondary Research, 2008,
    Kathleen Mcgrory, “Feds: Florida Scholarship Program does not Violate Anti-Discrimination Laws” article, Miami Herald, 2014,
    “Financial Aid Policy: Questions and Concerns”,Undergraduate Financial Aid in the United States, Section 4, American Academy of Arts, The Louisiana TOPS program requires a 2.5 GPA but also includes an ACT requirement. This program disproportionately benefits white students. Louisiana Board of Regents, “TOPS Report: Analysis of the TOPS Program from 2005 -2014,” (December 2015),
  22. See e.g. 21st Century Scholars: Indiana’s 21st Century Scholar Enrollment & ScholarTrack (2017), The program is targeted at families who qualify for free or reduced lunch. Katie Lobosco, “Oregon Promised Free Tuition. Now it’s Cutting Back,” news report, CNN, Washington DC, 2017,
  23. The New York program currently cuts off eligibility at $110,000, but that cut-off will soon increase to $125,000. High Education Services Corporation, “Excelsior Scholarship FAQs,”, New York State, “Oregon Promise,” Office of Student Access and Completion, High Education Coordinating Commission, The current cap stands at $20,000 in Expected Financial Contribution. Researchers in Pennsylvania also recently proposed a program with an income cap of $110,000. Mark Price, Stephen Herzenberg, Diana Polson, “The Pennsylvania Promise Making College Affordable and Securing Pennsylvania’s Economic Future,” Figures A1-A3, report, Keystone Research, Pennsylvania, 2018,
  24. A program that covers all or more of the cost of attendance, and not just tuition, would make a last-dollar structure much more useful to low-income students: if the state Promise covered costs not covered by other grants, it would presumably leave low-income students with no unmet need.
  25. There are also two programs (in California and Montana) approved by states but not yet funded; those were not included in this table. Additionally, California’s existing aid programs could conceivably be considered a Promise structure. The Cal Grant program covers tuition and fees for students just out of high school whose families earn below an income ceiling of almost $100,000 (for a family of four) and who earn at least a 3.0 in high school. Students earning a 2.5 are eligible for Cal Grant B, which has a lower income threshold of just $52,000 for a family of four and only covers a small living allowance in the first year. In the second year it covers tuition and fees as well as the living allowance. Both programs also have asset ceilings. Furthermore, California waives tuition for almost half of all students at community colleges. The determination is made at the campus-level and made based on a set of metrics to indicate financial need.
  26. Some programs cover associate’s or certificates that are typically awarded at community colleges but that a student pursues at a four-year institution.
  27. Requirements that a student enroll after high school vary from the semester following graduation to several years after high school.
  28. Maximum for a family of four; the ceiling increases for larger families.
  29. Schools must be listed as participating—most community colleges participate. “Eligible A+ Community Colleges and Vocation Technical Colleges,”
  30. The Scholarship covers up to $5,500 in tuition. If tuition is higher than that and no other aid makes up for it, SUNY or CUNY must award a “tuition credit.”
  31. If does not maintain 30 credit hours, will owe money for the second semester.
  32. $55K when they enroll, must be below $100K when they start receiving benefits.
  33. Did not originally have income limits, now has a maximum EFC limit of $20,000 set by their Commission.
  34. The College Bound Scholarship is last-dollar when it comes to state aid, but does not take into account Pell when it comes to covering the gap in tuition.
  35. $46K when they enroll, $55K when they graduate from high school.
  36. Thomas L. Harnisch, “State Early Commitment Programs: A Contract for College Success,” policy brief, American Association of State Colleges and Universities, 2009,
  37. For example, New York serves about 1.2 million two- and four-year students, but Excelsior reached 22,000 students in its first year (just under 2 percent of undergraduates). Rick Selter, “New York Free College Goes to 22,000 and Counting,” Inside Higher Ed, October 2, 2017,; National Center for Education Statistics, “Digest of Education Statistics,” 2015 Tables and Figures, Table 304.10, About 246,000 students attend college in Oregon; the Promise reached 6,700 students last year. Higher Education Coordinating Commission, “Senate Bill 81 Legislative Report: The First Term of the Oregon Promise,” National Center for Education Statistics, “Digest of Education Statistics,” 2015 Tables and Figures, Table 304.10, Even assuming the number of Promise recipients at any given time triples, it would reach about 17 percent of all students enrolled at community colleges in the state and 8 percent of students overall. For comparison, the Oklahoma Promise reached 17,749, approximately 9 percent of students enrolled in college (assuming the 2014 enrollment numbers held constant). Ibid; Oklahoma Higher Learning Access Program, “Oklahoma’s Promise – 2016-17 Year-end Report,”
  38. “State Merit Scholarship Programs and Inequality,” The Civil Rights Project, Harvard University, Ed. by Donald Heller and Patricia Marin,
  39. Caitlin R. Mcglade, Scott Travis, “Minorities, Poor Hit Hardest by Stricter Bright Futures Requirements,” article, Sun Sentinel, Florida, 2016,
  40. Maureen Downey, “Georgia College Students Will Now Earn HOPE GPA Boost in Many Science and Math Classes,” article, Georgia, 2017,
  41. In Louisiana, 75 percent of Taylor Opportunity Program for Students (TOPS) recipients were white; in comparison, 55 percent of enrollees at state colleges and universities were white (almost all TOPS recipients—94 percent—went to public institutions). Louisiana Board of Regents, “TOPS Report: Analysis of the TOPS Program from 2005 -2014,” (December 2015),; Louisiana Board of Regents, “Higher Education Factbook,” (2017),
  42. There may be additional expenditures needed to ensure that even robust programs do not have the unintended consequences of reducing the number of seats available for traditionally underserved students. For example, any program that creates more universal access but does not increase the capacity of the system to serve students runs the risk of creating unintended barriers to accessing more limited seats—with the result that upper-income students become more likely than their lower-income counterparts to be able to figure out how to navigate the system and enroll.
  43. The reach by income quartile will depend on the socioeconomic makeup of the student body, the level of aid already available, and tuition levels in the state. See e.g. Mark Price, Stephen Herzenberg, Diana Polson, “The Pennsylvania Promise Making College Affordable and Securing Pennsylvania’s Economic Future,” Figures A1-A3, report, Keystone Research, Pennsylvania, 2018, If a state implements income thresholds, they should ensure that those thresholds demarcate a phaseout of an award rather than creating a cliff effect that would provide a full benefit just below and no benefit just above the limit.
  44. Targeting aid to low-income students who would not otherwise apply for financial aid has an added benefit of bringing in more federal Pell dollars to the state.
  45. CLASP, College Students Aren’t Who You Think They Are,
  46. Access to in-state tuition or state financial aid programs varies significantly by state. Delaware’s program, for example, gives undocumented students specific background on how to qualify.
    Delaware Technical Community College, “Scholarship Requirements,” Delaware,
  47. For example, research has shown that Hispanic students are more likely to be loan-averse than are white students. Angela Boatman, Brent J. Evans, and Adela Soliz, “Understanding Loan Aversion in Education: Evidence from High School Seniors, Community College Students, and Adults,” AER 3 No. 1 (March 2017),
  48. Timothy J. Bartik, Brad J. Hershbein, Marta Lachowska, “The Effects of the Kalamazoo Promise Scholarship on College Enrollment, Persistence, and Completion,” Upjohn Institute for Employment Research Working Paper 15-229, 2017,;
    Celeste K. Carruthers and William Fox, “College coaching, financial aid, and post-secondary persistence in Tennessee.” Economics of Education Review 51: 97–112, 2016; “Financial Aid Policy: Questions and Concerns”, Undergraduate Financial Aid in the United States, Section 4, American Academy of Arts,; Susan Dynarski, Judith Scott-Clayton, “Financial Aid Policy: Lessons from Research” NBER Working Paper Series, no. 18710 (January 2013):47, Researchers also highlight the role that univerality can play in building trust in the system. Sara Goldrick-Rab and T. Kolbe, “A Matter of Trust: Applying Insights From Social Psychology to Make College Affordable.” Policy Insights from the Behavioral and Brain Sciences, 2016; 3 (2): 237 DOI: 10.1177/2372732216656457.
  49. Mark Huelsman, “On Free College, West Virginia is a Case Study on What To Do and Not To,” Demos, January 17, 2018,