The past several decades have seen college grow less affordable at every level, spurring public consensus that families need better options.1

Congress has done little to remedy this crisis. States, for their part, have enacted a growing number of “free college” programs, including the Tennessee Promise, signed into law in 2014, and others modeled after it. In just the five years from 2015 through 2019, fifteen states created scholarships that guarantee all qualifying families pay no community college tuition after aid has been applied.2 As a form of cost containment, they are all limited to certain subsets of students, some more limited than others.

Building on this momentum, many have argued for a nationwide “promise” to eliminate community college tuition, which would bring the United States into the fold of many other developed nations that make workforce training cost-free.3

In his 2015 State of the Union address, President Obama announced a proposal to eliminate community college tuition nationwide, a plan called America’s College Promise (ACP).4 A version of this proposal has been introduced every session of Congress since, and it was included as a banner provision of President Biden’s Build Back Better agenda.5 President Biden proposed the plan again this year as part of his Fiscal Year 2024 budget proposal.6

While there is across-the-aisle agreement that tuition-free community college is good for American families and students, ACP has not yet passed Congress. As the momentum for ACP has gone through cycles of starting and then stalling, some have argued that free college is best advanced by states.7 Others have suggested that the states are increasingly nearing that goal themselves.8

But lost in the debate is a crucial detail: America’s College Promise is not a fifty-state version of the state promise model. In fact, it goes significantly farther, setting community college tuition at $0 before other aid is factored in, contrasting with the “last-dollar” nature of most state promise scholarships. This makes the total aid received by students greater than what they would receive from a state promise program, especially for working-class families. This “first-dollar” quality of ACP’s tuition benefit is a feature, not a bug, and it gives ACP a fundamentally different structure than most state promise programs. In addition, ACP is for all students, not only recent high school graduates or those who are enrolled in certain programs of study.9

In this report, I leverage survey data from the U.S. Department of Education to show how key design differences between state promise programs and America’s College Promise yield strikingly different benefits for students. When compared with a sample aggregated from six states with active last-dollar promise programs, ACP would deliver 56 percent more total combined federal and state aid for promise-eligible students, potentially more depending on whether states turn existing promise dollars into non-tuition stipends. Recipients would see their grant aid available for non-tuition expenses by $2,800 on average, more than doubling the typical aid for housing, food, child care, and health care that the typical community college student currently receives in a promise program state.

Even if all fifty states enacted a last-dollar promise program, ACP would still deliver five times as much funding per year to reduce students’ costs of attendance.10 But the greatest benefits to students would come if ACP is passed and states rework their promise programs as non-tuition stipends, helping students cover their living costs, transportation, child care, and more.

Even if all fifty states enacted a last-dollar promise program, ACP would still deliver five times as much funding per year to reduce students’ costs of attendance.

In the eight years since President Obama first proposed ACP, a growing number of states have enacted their own statewide promise programs. However, a federal plan remains badly needed, and overstating states’ progress risks jeopardizing support for Congressional action that would combine the financial might of the federal government with states’ existing efforts to eliminate tuition and put a workforce-ready credential within every American’s reach.

The next time ACP is under consideration by Congress, the bill’s authors should strengthen the incentives that already exist in the bill to ensure states do not scrap their promise programs—or worse, view membership in the federal–state partnership as a binary choice between state-based free college and federal free college.

Where the Status Quo Has Gotten Us

In the United States’ federalist system of governance, higher education has been largely left to states. States charter public universities, many states set parameters for tuition pricing through legislation, and college presidents are hired by boards filled through state-level appointments or elections. This extends to the community college level, where counties and cities also play key roles.

The federal government steps in only within a limited capacity, mainly through the provision of aid directly to students: the Pell Grant, federal student loans, federal work-study, and targeted programs, such as grants for veterans.11 Although the federal government can cut off an institution from the flow of federal student aid, there is no existing lever in statute or regulation that allows the federal government to limit or curb the cost of college placed on families.12 In brief: the federal government plays virtually no role in the setting of prices families must pay.

This is the federal–state governing balance that has brought us where we are, with large shares of Americans expressing dissatisfaction with the cost of college, and many choosing not to enroll in the first place. At public institutions, net tuition per full-time student (i.e. tuition after financial aid) has grown 62 percent over the past twenty years, from $4,452 to $7,244 in constant 2022 dollars.13 Over the same period, education appropriations per full-time student have fallen and slowly recovered, but new growth in college costs have been entirely shouldered by the student and their family.14 This may help explain the persistent gaps in educational attainment by race, with just 34 percent of Black adults, 28 percent of Hispanic adults, and 25 percent of Native American adults holding a college degree.15 Racial gaps in degree attainment are no better among young people. Absent change, the status quo is here to stay, and it entrenches pre-existing inequality.

A bipartisan array of state leaders have now taken action, passing last-dollar tuition scholarships, or “promise” scholarships. State promise programs offer students a good deal: if a student enrolls and stays in-state, the student will not pay out-of-pocket for tuition, because the state government will pick up the remaining tab after all aid is factored in.

The announcement of a promise program is usually a political win, and rightly so. But there are also important points of nuance that we must note.

First, some state promise programs are narrowly tailored to small subsets of students. Eligibility criteria based on time out of high school, GPA, and full-time enrollment status reduce the number of students who qualify. So do bottlenecks such as income and transcript verification, and in one state, applicants must even pass a drug test, paid for by the student.16

These factors make programs easier to fit within constrained state budgets, but they also push participation rates downward. Consider these examples:

  • The Work Ready Kentucky Scholarship only applies to five approved areas of study.17 In the 2020–21 award year, 3,033 students received the grant, out of 61,000 degree/certificate-seeking community college students in Kentucky.18
  • Only 2,025 Maryland community college students out of 120,000 degree/certificate-seeking community college students received the Community College Promise Scholarship in the program’s second year; families reported confusion about paperwork requirements and the state cut the program’s funding.19
  • Arkansas Future Grant is only available on a first-come first-served basis. In the 2020–21 award year, 604 students received the grant, out of roughly 38,000 Arkansas community college students seeking a certificate or degree that year.20

By some counts, nearly two-thirds of state governments offer promise programs.21 But the share of community college students who receive grants from a promise program, or would be if they met income limits, is far lower.

Moreover, some promise programs shift to a “first-come, first-served” allocation system when demand exceeds funding. For example, the statewide promise programs in New York, Oregon, Nevada, and Maryland all include procedures for triaging applicants under certain budgetary circumstances, resulting in waitlists and temporary income limits.22 These can also limit the number of participating students, in ways that the students and families may not be able to predict.

Figure 1 shows state promise programs and their reported counts of promise program recipients,23 compared to total community college students seeking a certificate or degree in the 2020–21 award year.24 Aggregated together, 31 percent of students enrolled in one of these states’ community colleges that year received a state promise grant, a percentage that is driven upward by high shares in three large states (Tennessee, California,25 and Washington).26 In the median states, West Virginia and New Jersey, 9.0 and 8.1 percent of community college students received a promise program scholarship, respectively.

Figure 1

Of course, it is commendable that states offer these programs, even as limited as some of them are. Every financial aid program a state legislature enacts is an important contribution to the great project of making upward mobility in the United States attainable for all.27

But how we talk about these existing state programs matters. We cannot assume that the general public knows the nuances of these programs’ designs. Voters may misinterpret well-intentioned headlines, such as “Is Community College Free? Yes, in These 31 States,” taking the message to mean that most people can already enroll in community college tuition-free.28 This is not only a messaging problem in terms of adequately informing prospective students and their families: it also jeopardizes the success of proposals that would supplement and improve on existing promise efforts. Unless the nuance of promise programs is understood, the public may believe federal action to eliminate tuition would be overkill.

This is not only a messaging problem in terms of adequately informing prospective students and their families: it also jeopardizes the success of proposals that would supplement and improve on existing promise efforts.

The reality is just the opposite: a federal free college program such as ACP is badly needed for low-income students to afford college. This holds true even for students who already receive a state promise grant, as I show in the next section.

Comparing the Federal First-Dollar Plan to Last-Dollar State Grants

If enacted by Congress, the America’s College Promise Act would establish a federal–state partnership through which the federal government sends block grants to states on the condition that they charge community college students $0 in tuition and fees. For qualifying students–roughly three out of every four students who apply for federal aid29–tuition simply ceases to exist.30

ACP’s “first-dollar” benefit covers tuition before other aid is factored in. Most state promise programs contrast with ACP in that they cover tuition after other aid, making them “last-dollar” programs.31 Under a last-dollar promise program, a low-income student whose Pell Grant exceeds tuition receives no promise dollars. At the same time, someone whose family does not receive any federal assistance would receive a promise grant equal to tuition, minus other state aid if applicable.

The “first dollar” versus “last dollar” design choice has a strong influence on who receives money from the program. Scholars have observed that middle- and higher-income students benefit the most from last-dollar awards.32 Due to its last-dollar design, New York’s Excelsior Scholarship has been found to predominantly serve students with family incomes above the median for New York City, and who would be in the top 20 percent among applicants to the City University of New York (CUNY).33 Other analysis has found that last-dollar programs deliver grants to more white students, and fewer to Black or Latino/a students, than what would be expected based on enrollment.34

Others respond that the simple guarantee of enrolling without having to pay out-of-pocket for tuition breaks through to low-income students who would otherwise be on the fence about enrolling, and who would otherwise only learn about their net tuition cost after completing substantial paperwork.35 In other words, the indirect benefits of last-dollar programs are greatest for low-income students. However, it remains true that the direct beneficiaries of last-dollar programs are often higher-income.

It is also important to note that roughly half of state promise programs have income caps, and that students from high-income families tend not to enroll in community college.36 Last-dollar programs seldom “waste” money on well-resourced students. At the same time, low-income students are often left needing more assistance.

Figure 2, below, shows combined Pell Grant and last-dollar promise aid for community college students enrolled full-time who applied for federal aid, grouped by their family’s adjusted gross income (AGI).37 As a student’s income decreases, combined grant aid remains roughly the same.

Figure 2

America’s College Promise maintains all the upside of last-dollar programs while creating a progressive system of college costs, where no student pays tuition and the Pell Grant and state financial aid offset low- and middle income students’ non-tuition expenses. Freeing the Pell Grant for non-tuition costs is no trivial matter: the maximum Pell Grant is $7,395 for the 2023–24 award year, and non-tuition costs contribute 60 percent to 80 percent of the total cost of attendance at community colleges, depending on the student’s living arrangement.38

Figure 3, below, shows how a first-dollar program, such as ACP, delivers more than $8,000 in combined grant aid for the average community college student from the lowest income bracket.39 By comparison, combined grant aid for that same student under a last-dollar program is under $5,000 (Figure 2 above).

A first-dollar program, such as ACP, delivers more than $8,000 in combined grant aid for the average community college student from the lowest income bracket.

Figure 3

Students Everywhere Stand To Gain from ACP, Including Those with State Promise Grants

In 2021, researchers at the Urban Institute examined how ACP would affect community college students who already have no net tuition from Pell Grants and/or state grants. They found that America’s College Promise would reduce the average amount students would pay out-of-pocket for living expenses by 33 percent, from $10,661 to $7,193.40 Combining the ACP benefit, the Pell Grant, and other aid, students who currently pay no net tuition would see their total grants rise from $7,258 to $10,725, a 47 percent increase in total grants.41 The subject of the Urban Institute’s analysis is not exactly the same as the population that receives a promise program: a maximum Pell Grant will exceed community college tuition in most states, whether the student is in a promise program or not. But the analysis demonstrated how ACP would improve affordability by cutting down students’ unaided living costs.

Since then, new survey data allow us to look within specific states’ community college sectors.42 In this analysis, I examine how ACP would change how promise-eligible students pay for college in a sample of six states with active last-dollar promise programs: Michigan, Nevada, New Jersey, New York, Oregon, Tennessee. A full description of the methods of my analysis can be found in Appendix B.

Total Grant Aid Received

Figures 4 and 5 show how students’ total grant aid would increase through ACP: specifically, for those who are already eligible for their state promise program. Orange bars indicate the estimated total grant aid under the status-quo of each state promise program, gray under ACP, and yellow under ACP if state promise programs are converted into non-tuition stipends.

Among promise-eligible students, increases to total grant aid would range from 39 percent (Nevada) to 74 percent (Oregon). (See Figure 4 below.)

Figure 4

When we aggregate by enrollment, the typical promise-eligible student in one of the sample states would see their total grants increase by 56 percent, from $5,714 to $8,926.43

Importantly, ACP incentivizes states to maintain their existing need-based financial aid by converting it to non-tuition stipends. The yellow bars show the estimated total aid students would receive if states maintained their promise programs as non-tuition aid, assuming no change in which students receive state promise program dollars.44

Comparing the yellow bars (ACP plus state promise program stipends) to orange bars (state promise alone), the average student across these six states would see their total grant aid increase 85 percent, from $5,714 to $10,581.

Figure 5 below shows the same estimates for the Promise-eligible Pell recipient population in each state. The bars are higher than in Figure 4 by about $1,000, but it remains true that the bars reflecting ACP (gray) and ACP plus state promise (yellow) are the highest.

Figure 5

Among promise-eligible students who receive Pell, increases to total grant aid range from 39 percent (Nevada) to 67 percent (New Jersey) with the enactment of ACP. Aggregated by enrollment, the typical promise-eligible Pell recipient in one of the sample states would see their total grants increase by 51 percent, from $6,970 to $10,574.

The typical promise-eligible Pell recipient in one of the sample states would see their total grants increase by 51 percent, from $6,970 to $10,574.

If states convert their promise programs to non-tuition stipends, then the typical Pell recipient would see their total grants increase 72 percent, from $6,970 to $11,973.

This confirms what we would expect: ACP makes a sizable difference in the financial means of students, even those who already qualify for free college through a last-dollar promise program.

Grants for Non-Tuition Expenses

For three states, we can calculate the expected change in the grant aid available for students’ non-tuition costs in each scenario: state promise alone (orange), ACP alone (gray), and ACP with state promise as stipends (yellow).45

Among promise-eligible students, increases to non-tuition grant aid would range from 51 percent (Nevada) to 283 percent (Michigan). In the aggregate, a typical promise-eligible student would see their non-tuition grant aid more than double, a 226 percent increase from $1,239 to $4,035, a gain of about $2,800. (See Figure 6 below.)

Figure 6

In other words, the move to ACP would significantly increase community college students’ grant support for costs other than tuition: housing, food, transportation, supplies, and more.

Figure 7 below shows the same analysis for promise-eligible Pell recipients. Again, Pell grant recipients would have a large amount of non-tuition grant aid, as much as $7,700 in Nevada. The percentage increase from the state promise to ACP ranges from 63 percent (Nevada) to 270 percent (Michigan).

Figure 7

Aggregated together, the typical Pell Grant recipient in one of these states would see their non-tuition grant aid rise 227 percent under ACP, from $1,675 to $5,491.

For many students in the United States today, the concept of receiving substantial government assistance with living costs is unthought of, though it is a reality in many other countries with fewer resources than ours.46 When it comes to students’ and families’ pocketbooks, the greatest shift ACP ushers in may be how it reduces under-resourced students’ the struggle of living and learning at once.

“Stacking” Promise Programs to Offset Living Costs: The Politics and the Policy

The prior sections of this report have argued that lawmakers and the public must recognize the need for a federal plan to achieve nationwide free community college on a first-dollar basis, such as America’s College Promise. Even the states that have promise programs face limits of breadth and depth. Cost constraints force many states to limit eligibility to certain subsets of students by age, enrollment intensity, or occupation (a breadth problem), and the last-dollar design limits how much these programs can help students cover their living costs (a depth problem). ACP would facilitate major strides forward along both of those axes.

But the existence of a large number of statewide promise programs also enables a golden opportunity: if ACP is enacted, states can stack their existing promise aid47 on top of the first-dollar ACP benefit to offset students’ living costs. The combination of ACP, the Pell Grant, and state grants can, beyond simply eliminating tuition, deliver students critical financial assistance for books and supplies, housing, food, transportation, and more: the exact costs that so often impede students’ progress from matriculation to degree completion.

The combination of ACP, the Pell Grant, and state grants can, beyond simply eliminating tuition, deliver students critical financial assistance for books and supplies, housing, food, transportation, and more.

As a federal–state partnership, ACP does not mandate states to convert state financial aid for community college students into stipends, but it incentivizes them to do so. Specifically, the text of the America’s College Promise Act states the following:

  1. States must provide a sum of money called a “state share” to join the federal–state partnership.48 The state share must be used to eliminate tuition at the state’s community colleges, except that leftover funds after eliminating tuition can be allocated towards other educational purposes.49
  2. States must also adhere to maintenance of effort provisions which prohibit states from cutting other spending on higher education.50 Specifically, states must maintain their levels of fiscal support for higher education per full-time equivalent student, support for the operating expenses of public four-year institutions, and support for need-based financial aid. These three “maintenance of effort” requirements (MOEs) are all calculated separately, and each year they are indexed to the prior three years’ average.51
  3. A state can count in its state share the state financial aid it provides to ACP participants.52 In addition, state financial aid does not need to be included in the MOE requirement for fiscal support for higher education, even while it can still be counted in the MOE requirement for need-based financial aid (if it is in fact based on need).53 I will call this the “Section 786 Exemption,” after the bill section where the rule is located.

In other words, the text of the America’s College Promise Act incentivizes states with promise programs to keep providing these grants—and to keep providing them to community college students.

But there’s more to the story. Even if states maintain their promise programs and allocate them as stipends, will they do it equitably? How likely is it that states can claim the Section 786 Exemption? And how can governors who are reluctant to join ACP be further incentivized to join, given that they can tell voters, “We have free community college already”?

The following sections unpack these questions that, far-off as they may seem, could one day become battlegrounds and may determine whether states can capitalize on the opportunity to harness the power of both federal- and state-based free college.

Stacking Equitably

The Section 786 Exemption incentivizes states to continue giving the same level of financial aid to community college students, but it does not require states to distribute that financial aid in any particular way. (The only requirement is that the aid not be awarded predominantly based on merit.)54

Every state that joins ACP will have to rewrite its formulas for which community college students get financial aid and how much, be it a promise program or not. This marks a fork in the road for states. On the one hand, a state can allocate the most funding to students with the greatest financial needs, such as by making their award amount a function of the student’s Student Aid Index.55 (See Figure 8 below.)

Figure 8

On the other hand, the state can designate a certain amount of non-tuition grant aid it wants every student to receive and rewrite the promise program’s formula so that the converted promise program, combined with Pell, will guarantee that amount.56 The same students who receive last-dollar promise aid would receive the converted promise: namely, upper-income students. (See Figure 9 below.)

Figure 9

This presents a déjà vu moment in the longstanding “promise” debate. Is it better to target limited resources at those with the greatest need, or to empower a simple, powerful message about college affordability that applies to all students before they’ve even filed the paperwork? Arguably, there could be upside to the simple message of, “Enroll in community college, pay no tuition, and get a $3,000 stipend.” There is also an understandable political appeal of delivering more money to the middle class.

But, as a technical matter, such a distribution would be regressive. Furthermore, students on the high-income side of the distribution would be paid to go to college: it is not simply that they will not pay out-of-pocket, but also that money would go into their pocket.

To garner political support in Congress and statehouses around the country, we should ensure that states invest in the students most in need—and with the most potential to create better economic opportunities for their families.

To garner political support in Congress and statehouses around the country, we should ensure that states invest in the students most in need—and with the most potential to create better economic opportunities for their families. The text of ACP should require states to submit the allocation formulas of the state financial aid for which they claim the Sec. 786 Exemption to the Department of Education.57 Moreover, the secretary of education should be permitted to withhold the exemption from financial aid programs that skew too heavily towards upper-income students.

To some, such a decision may seem like federal overreach. But the passage of America’s College Promise would mark the single greatest advancement since the 1960s towards ensuring a college degree is attainable for every American across the socioeconomic spectrum. The lowest-income students, especially those in high-cost-of-living areas, should not still be working thirty hours a week or taking out debt to earn an associate’s degree under ACP. While federal lawmakers have the tools to promote affordability for working-class families, they should use them to the fullest.

Enabling Stacking in High-Tuition States

Some states will find it difficult to claim the Section 786 Exemption, limiting their ability to stack aid and posing a question about how the federal government should step in to help high-tuition states maintain their financial aid programs.

States receive federal funding under ACP under one condition above all others: that they charge participating students $0 in tuition.58 That tuition must be replaced, or else community colleges will become revenue-starved.59 The replacement of that revenue comes primarily from the federal share and the state share.60 For every state alike, the federal share is four or more times larger than the state share; during the economic nadir of severe recessions, the federal government will step in to cover the entire state share.61

The mandated amounts of the federal and state shares are calculated using the state’s full-time equivalent community college enrollment and the national median in-state community college’s tuition charge. That last factor, the median in-state community college tuition charge nationwide, is the same for every state’s calculation.62

Roughly half of states’ community colleges fall below the median, and they will have an easier time eliminating tuition. Roughly half of states’ community colleges fall above the median, and they will have to add in additional money beyond their state share to participate.

The Section 786 Exemption makes sense for a state only if it does not need to spend its entire state share on covering lost tuition: in other words, a low-tuition state.

However, many promise programs are in states on the high-tuition side of the scale, meaning they can’t claim the Section 786 Exemption. When they join ACP, they will have to cover the cost of tuition, and because of the MOE requirements, they will also need to continue their promise programs (plus all other financial aid), without assistance from the federal government for those costs.63

TCF’s projections in Table 1 below shows that only a handful of states with promise programs would be able to claim the Section 786 Exemption to count their promise program spending towards their required state share while delivering those dollars to students as stipends.

Table 1
State ACP Federal + State Shares Real cost of eliminating tuition  Balance after eliminating tuition  Cost of promise program Can state claim Sec. 786 Exemption for promise program?  
AR $85.4 M $88.8 M – $3.4 M $2.1 M No
CA $3.04 B $980 M + $2.06 B $654 M Yes, for some costs64
HI $47.3 M $50.3 M – $3.1 M $2.9 M No
KY $122 M $153 M – $31.1 M $9.3 M No
MD $229 M $293 M – $64.1 M $6.3 M No
MI $359 M $471 M – $111 M $14.2 M No
MO $181 M $193 M – $11.8 M $48.8 M No
NJ $337 M $483 M – $146 M $26.4 M No
NM $101 M $56.2 M + $45.1 M $3.9 M Yes, in full65
NV $102 M $83.4 M + $18.3 M $2.8 M Yes, in full66
OK $130 M $126 M + $4.3 M $10.3 M Yes, for some costs67
OR $168 M $245 M – $76.6 M $19.9 M No
RI $35.3 M $46.4 M – $11.1 M $7.0 M No
WA $436 M $418 M + $18.3 M $122 M Yes, for some costs68 
WV $37.6 M $42.8 M – $5.2 M $3.9 M No
Source: Author’s projections of America’s College Promise costs using 2020-21 data from the Integrated Postsecondary Education Data System, as well as states’ total reported promise program awards for the 2020-21 award year. Not all states that had promise programs in 2020-21 provide sufficient reporting to populate their data in this table: this includes CT, DE, IN, NY, and TN.

Based on projections using 2020–21 data, only five of the fifteen states with promise programs (and sufficient data for calculations) could at least partially cover the costs of their promise program through the Section 786 Exemption.

The design of ACP clearly signals that its authors want state financial aid programs to persist as non-tuition assistance for community college students. However, states may need more federal help following through.

A revision to ACP could establish a “State Financial Aid Continuation Fund,” whereby the federal government would refund states $1 for every $2 they spend on need-based state financial aid for ACP recipients.69 Restrictions could apply, such as limiting the eligible student population to Pell Grant recipients or another indicator of financial need. By adding this fund to ACP, the federal government can make sure that the decision to join the federal–state partnership will not be framed as a choice between the free college program state lawmakers created and the free college program Congress is establishing.70

Easing the Political Choice to Join ACP

State promise programs are political achievements. Governors, including liberal governors, may be reluctant to join ACP if their promise program will no longer keep credit for covering tuition. Lawmakers will not want one of their signature legislative accomplishments to become an add-on to someone else’s bigger, better win. Even if they like Joe Biden, they don’t want Joe Biden to eat their lunch.

As a workaround, a revision to the text of ACP could specify that states can label the elimination of tuition on financial aid award letters and promotional materials with the name of their promise program.71 For example, a student in Tennessee would not see “America’s College Promise” on their award letters, but instead, “Tennessee Promise.” To a state’s residents, it would be as if the federal–state partnership is supercharging their state promise program, not replacing it, similarly to how the exchanges created by the Affordable Care Act increased their uptake on the exchanges by using names like “Kynect” in Kentucky or “CoverME” in Maine.72

Allowing states to rename ACP within their borders may also make the program more intuitive for students and their families. A student in Rhode Island cannot take their federal ACP benefit into Massachusetts or Connecticut, and it may help a Rhode Island family understand the plan if it is called “Rhode Island Promise,” the name of the state’s current promise program.


Both the federal government and the states have duties to ensure their residents can afford the education that is needed for participation in the economy. In the absence of Congressional action, states have blazed a path forward through promise programs. However, a rising recognition of students’ challenges covering non-tuition expenses underscores how the wave of last-dollar programs linked to tuition can only carry the nation so far.73

For sixteen states with promise programs and sufficient public reporting, Figure 10 compares current annual promise program expenditures for community college students versus projected funding for America’s College Promise. These programs deliver $1.1 billion per year to students; America’s College Promise would fund first-dollar free college in these same states at a level of $5.6 billion per year.74

Figure 10

It follows that, even if every U.S. state adopted a last-dollar promise program in the vein of existing programs, community college students would receive only 19 cents for every dollar that America’s College Promise would make available.

If states had the deep pockets of the federal government, they would not have to make the tradeoffs that can be found in state promise programs’ designs. Every promise program is an honest attempt made with limited resources. The aim of this research is not to point out the limitations of state promise programs, but rather, to quantify what would be newly possible if Congress activated the power and weight of the federal government to do what few states can do alone and eliminate community college tuition for good and for all.

It follows that, even if every U.S. state adopted a last-dollar promise program in the vein of existing programs, community college students would receive only 19 cents for every dollar that America’s College Promise would make available.

States would still decide their own destinies if Congress passes the America’s College Promise Act. Under ACP’s federal–state partnership, states must apply to join, committing to certain rules affecting state funding for public four-year colleges, transfer pathways, alignment with K-12 education, and more.75 States can simply decide not to join, as was true with the expansion of Medicaid under the Affordable Care Act. But if that federal–state partnership is any indication, many states that initially hold out could eventually join, with the number of nonparticipating states approaching zero.

This report has proposed ways to improve states’ ability to join ACP while maintaining the political and affordability upsides of their homegrown promise programs. It need not be an either-or choice, and the combination of the two can create a stronger pathway to a college degree than either can create alone.

Appendix A, containing the data and methods for Figures 1 and 10, can be found here.

Appendix B, containing the data and methods for Figures 4 through 7, can be found here.

The code used for the original analysis in this report can be found at this GitHub repository.

This research was generously supported by the Kresge Foundation. Thanks to Michelle Miller-Adams and Tom Harnisch for feedback on the research in this report, and thanks to Mary Rauner for feedback on an early draft of the report.


  1. Hannah Hartig, “Democrats overwhelmingly favor free college tuition, while Republicans are divided by age, education,” Washington, DC: Pew Research Center, August 2021. In addition, recent polling by the New America Foundation finds that 82 percent of respondents agreed with the statement, “States should spend more tax dollars on education opportunities after high school to make them more affordable,” with a similar share saying the same of the federal government. See Figure 9, “Varying Degrees 2023,” Washington, DC: The New America Foundation, August 2023.
  2. Jen Mishory and Peter Granville, “Policy Design Matters for Rising ‘Free College’ Aid,” Washington, DC: The Century Foundation, June 2019.
  3. Peter Granville, “President Biden’s Free Community College Plan Matters Now More Than Ever,” Washington, DC: The Century Foundation, September 2021.
  4. “FACT SHEET – White House Unveils America’s College Promise Proposal: Tuition-Free Community College for Responsible Students,” Washington, DC: The White House Office of the Press Secretary, January 2015.
  5. “FACT SHEET: How the Build Back Better Plan Will Create a Better Future for Young Americans,” Washington, DC: The White House Office of the Press Secretary, July 2021.
  6. “Budget of the U.S. Government: Fiscal Year 2024,” Washington, DC: The White House Office of Management and Budget, March 2023.
  7. Beth Akers, “Free College Movement Will Likely Head Back to the States — Where it Belongs,” Real Clear Politics, October 2021.
  8. Eric Kelderman, “Free College Is Dead in Congress, but It’s Alive and Well in the States,” Chronicle of Higher Education, November 2021.
  9. A student must enroll at least half-time to qualify for America’s College Promise. This is also a feature of state promise programs, and many state programs require full-time enrollment.
  10. See Figure 10 for details and methods.
  11. Exceptions include emergency stimulus, such as HEERF funds passed in 2020 and 2021, and targeted funds for certain categories of colleges, known as Title III.
  12. Carolyn Fast, Bob Shireman, and Alex Edwards, “College Tuition Is Out of Control. Voters Want Government to Do Something,” Washington, DC: The Century Foundation, July 2023.
  13. See Figure 2.1: “2022 State Higher Education Finance (SHEF) Report,” Boulder, CO: The State Higher Education Executive Officers Association, June 2023.
  14. Ibid.
  15. “Stronger Nation: Progress toward Racial Equity,” Indianapolis, IN: The Lumina Foundation, n.d., accessed June 2023.
  16. “Drug Screening,” Charleston, WV: West Virginia Higher Education Policy Commission, n.d.
  17. These are Advanced Manufacturing, Business/Information Technology, Construction, Healthcare, and Transportation/Logistics.
  18. See Appendix A.
  19. See Appendix A. See Danielle Douglas-Gabriel, “Why 3,000 people are still waiting for Maryland’s community college scholarship,” Washington Post, October 2020, and Danielle Douglas-Gabriel, “Maryland’s free community college initiative gets off to a slow start,” Washington Post, September 2019, and
  20. See Appendix A. See Ashley A. Smith, “A Promise With Limits,” Inside Higher Ed, March 2017.
  21. Map of statewide promise programs, The Campaign for Free College Tuition, n.d., accessed August 2023.
  22. Danielle Douglas-Gabriel, “Why 3,000 people are still waiting for Maryland’s community college scholarship,” Washington Post, October 2020. Sami Edge, “Lawmakers poised to keep Oregon Promise for now, but could limit it to low-income students,” Oregon Live, April 2023.
  23. See Appendix A for data sources.
  24. Students not seeking a degree or certificate were excluded from enrollment totals to avoid counting dual-enrolled high school students.
  25. I include California as a promise program state, but its tuition-free model differs from other states’. California’s public universities and community colleges charged no tuition until the 1970s and 1980s, respectively. When the state’s community colleges started charging tuition in 1985, a waiver was introduced for families below a specified poverty-level threshold. This waiver has been continued since and operates as an entitlement for qualifying families, but it was renamed the California College Promise Grant (CCPG) in 2017. In 2017 and 2019, state legislation allocated funding to create the California College Promise, a separate program that is state-funded but administered by individual community colleges. Community colleges are permitted to spend their College Promise funds in different ways. Some reduce tuition and fees for students who don’t receive the CCPG; others offset students’ non-tuition costs, such as books as supplies. Eligibility criteria may also vary. In this report, I treat the CCPG and the College Promise as a combined program for interstate comparisons, though they are separate in practice. For further details on the state’s tuition-free programs, see Mary Rauner and Sara Lundquist, “College Promise in California: Recommendations for Advancing Implementation, Impact, and Equity,” San Francisco, CA: WestEd, November 2019.
  26. If California were removed from the calculations, only 15 percent of eligible students participate.
  27. That said, increases in state financial aid sometimes accompany cuts to state fiscal support for public colleges and universities, which contributes to a “high-tuition high-aid” financing model that carries negative implications for upward mobility.
  28. In fact, these lists often include state programs that are not free college programs at all. For example, the MassGrant Plus program in Massachusetts covers the balance of tuition after all other grants and the family’s Expected Family Contribution (EFC). For all families except those with an EFC of $0, out-of-pocket tuition payments must be made. The MassGrant Plus program is often included in these lists.
  29. I estimate that 56 percent of community college students would qualify for ACP based on current enrollment factors, after meeting qualifications for enrollment intensity, in-state residency, and FAFSA completion. However, the free college incentive could lead to more students enrolling at least-half time, enrolling in-state, or completing the FAFSA. If FAFSA completion rates were not a factor, then this figure would rise to 76 percent. Source: author’s analysis of data from the National Postsecondary Student Aid Study: 2015-16, via NCES Datalab.
  30. See Section 791(5) for the definition of an eligible student under ACP. To receive the benefit, students must be an undergraduate enrolled at least half-time, qualify for in-state or in-district tuition (or would if not for immigration status), not be a dual enrollment or early college high school student, and if they are a U.S. citizen, complete the Free Application for Federal Student Aid (FAFSA).
  31. A third formula, “middle-dollar,” largely resembles the last-dollar formula but sets minimum award amounts that ensure qualifying students receive a nonzero amount of financial assistance from the program. The Oregon Promise Scholarship is regarded as a middle-dollar scholarship. For the purposes of this report, I group middle-dollar scholarships with last-dollar scholarships.
  32. Meredith Billings, “Understanding the design of college promise programs, and where to go from here,” Washington, DC: The Brookings Institution, September 2018.
  33. Judith Scott-Clayton, CJ Libassi, and Daniel Sparks, “The Fine Print on Free College: Who Benefits from New York’s Excelsior Scholarship?”, Washington, DC: The Urban Institute, May 2022.
  34. Anthony P. Carnevale, Jenna R. Sablan, Artem Gulish, Michael C. Quinn, and Gayle Cinquegrani, “The Dollars and Sense of Free College,” Washington, DC: The Georgetown Center on Education and the Workforce, October 2020.
  35. Sara Goldrick-Rab and Michelle Miller-Adams, “Don’t Dismiss the Value of Free-College Programs. They Do Help Low-Income Students,” The Chronicle of Higher Education, September 2018.
  36. Jen Mishory and Peter Granville, “Policy Design Matters for Rising ‘Free College’ Aid,” Washington, DC: The Century Foundation, June 2019. In 2019-20, only 5.7 percent of dependent community college students came from families making above $150,000, compared to 9.4 percent of public four-year students and 10.9 percent of students in all other sectors. Source: author’s analysis of data from the National Postsecondary Student Aid Study, 2019-20, accessed via NCES Datalab using retrieval code uqtwcw.
  37. The distribution of students by AGI category is as follows: 18.5 percent have AGIs of $0 to $15,000, 24.7 percent have AGIs of $15,000 to $30,000, 16.5 percent have AGIs of $30,000 to $45,000, 9.9 percent have AGIs of $45,000 to $60,000, 12.4 percent have AGIs of $60,000 to $90,000, 8.8 percent have AGIs of $90,000 to $120,000, and 9.3 percent have AGIs over $120,000. See NCES Datalab, retrieval code arydlu.
  38. Table 330.40, Digest of Education Statistics, Washington, DC: National Center on Education Statistics, n.d., accessed June 2023.
  39. The distribution of students by AGI category is as follows: 18.5 percent have AGIs of $0 to $15,000, 24.7 percent have AGIs of $15,000 to $30,000, 16.5 percent have AGIs of $30,000 to $45,000, 9.9 percent have AGIs of $45,000 to $60,000, 12.4 percent have AGIs of $60,000 to $90,000, 8.8 percent have AGIs of $90,000 to $120,000, and 9.3 percent have AGIs over $120,000. See NCES Datalab, retrieval code arydlu.
  40. Jason Delise and Jason Cohn, “What Free Community College Means for Students Who Already Have It,” Washington, DC: The Urban Institute, September 2021.
  41. Ibid.
  42. Specifically, the National Postsecondary Student Aid Study: 2017-18 Administrative Collection.
  43. Weighting by enrollment means this estimate is driven disproportionately by New York (39 percent of students) and New Jersey (25 percent of students), the two largest states by degree/certificate-seeking community college enrollment in the sample of six states.
  44. However, a state could choose to deliver those stipends in a way that favors low-income students. This is unpacked further in the next section of this report.
  45. This is not feasible for the three states that had active promise programs in 2017-18, the award year reflected in our data source, NPSAS:18-AC. The problem this poses is that we cannot distinguish between state promise dollars and non-promise state financial aid.
  46. Lara Takenaga, “4 Years of College, $0 in Debt: How Some Countries Make Higher Education Affordable,” The New York Times, May 2019.
  47. And other state-funded scholarships.
  48. Section 786. The amount of the state share is a function of the number of the state’s full-time-equivalent community college students, the nation’s median in-state community college full-time tuition charge, and the percentage of total partnership costs that will be covered by the federal government. See Section 786(b).
  49. Section 789.
  50. Section 788(c).
  51. Ibid.
  52. Section 786(b)(2)(A).
  53. Sections 786(b)(2)(A) and 791(10)(b)(iii).
  54. Section 786(b)(2)(A).
  55. The Student Aid Index replaces the Expected Family Contribution measure, a change made through the FAFSA Simplification Act starting in the 2024-25 award year.
  56. If no new money is allocated to the program, the state could peg this amount to the pre-ACP full-time tuition charge.
  57. That is, the allocation formulas that would be implemented once ACP takes effect in the state: not the allocation formulas of programs as they are currently written.
  58. Section 788(a).
  59. That is, any more than they already are revenue-starved. See “Recommendations for Providing Community Colleges with the Resources They Need,” Washington, DC: The Century Foundation, April 2019.
  60. I say “primarily” because high-tuition states will need to supply additional resources beyond the federal and state shares to replace the totality of tuition revenue.
  61. Section 790.
  62. This value falls between $4,000 and $5,000, depending on what you count as a community college and what year you use as the reference point.
  63. It is also worth noting that states that cannot claim the Section 786 Exemption are not incentivized to commit their MOE-mandated financial aid to community college students.
  64. Assuming an 80 percent federal share of costs, California’s state share is $607 million, all of which can be put towards the Section 786 Exemption. However, the total cost of its promise program is $654 million, which means only about 93 percent of the promise program’s cost can be accounted for by the exemption.
  65. Using 2020-21 data, New Mexico’s remaining balance after tuition would be sufficient to cover its entire promise program using the Section 786 Exemption. However, New Mexico has greatly expanded its promise program since then, meaning these projections are likely off by a meaningful margin.
  66. Nevada can fully claim the Section 786 Exemption for its promise program, using its leftover $18.3 million. Nevada’s state share is $20.3 million, greater than the $2.8 million cost of the promise program in 2020-21.
  67. Oklahoma has a leftover balance of $4.3 million. If it devotes all of that leftover balance to its promise program through the Section 786 Exemption, it will cover about 42 percent of the promise program costs.
  68. Washington State has a leftover balance of $18.3 million. If it devotes all of that leftover balance to its promise program through the Section 786 Exemption, it will cover about 15 percent of the promise program costs.
  69. This is a lower federal share of costs than the federal government’s contributions toward eliminating community college tuition, which is 80 percent in the final year of the program under the Build Back Better Act’s text. Supporting students’ non-tuition costs is a lower priority goal of America’s College Promise than the elimination of tuition, hence the lower share (50 percent) through the State Financial Aid Continuation Fund.
  70.  It would also have the benefit of helping states satisfy the maintenance of effort requirement for need-based financial aid, especially those that cannot claim the Section 786 Exemption.
  71. Other federal-state partnerships demonstrate this. The Supplemental Nutrition Assistance Program (SNAP) and Medicaid are often named something other than SNAP and Medicaid within a state: for example, “CalFresh” and “MediCal” in California and “FoodShare” and “BadgerCare” in Wisconsin. Of course, it may be warranted for the text to also require the state to receive the Secretary’s approval before changing the program’s name.
  72. Jon R. Gabel, Heidi Whitmore, Sam Stromberg, and Matthew Green, “Why Are the Health Insurance Marketplaces Thriving in Some States but Struggling in Others?”, New York, NY: The Commonwealth Fund, November 2018.
  73. For a review of this issue, see Ann Coles, Laura Keane, Brendan Williams, “Beyond the College Bill: The Hidden Hurdles of Indirect Expenses,” Boston, MA: uAspire, June 2020, and Sandy Baum and Jason Cohn, “Nontuition Expenses: Implications for College Affordability and Financial Aid Policies,” Washington, DC: The Urban Institute, June 2022.
  74. See Appendix A for details.
  75. For more about those rules, see Section 788(b) of the 2021 bill text.