Incomplete: The Quality Crisis at America's Private, Non-Profit Colleges

Incomplete Quality Crisis Header

Is college good enough?

There are more than a thousand four-year private, non-profit colleges in the United States. They enroll approximately 2.7 million students each year, with roughly 1.7 million full-time students taking out student loans to finance their education.1 In 2013-2014 alone, more than 1.1 million Pell Grants were awarded to recipients attending these institutions, adding up to a total of $4.5 billion federal tax dollars.2 These four-year institutions promise to educate and graduate young men and women and prepare them to succeed in the working world. We rely on these schools to be mobility engines for low and middle income people. And from one perspective, that assumption is right—today, the unemployment rate for college graduates stands at 2.6% compared to 5.4% for those with only a high school diploma.3 And it is estimated that a college graduate will earn an average of $1 million more in wages over the course of a lifetime than his or her non-college educated peers.4 

But in an analysis of full-time, loan-holding students at four-year private, non-profit colleges, we found a stunning level of institutional failure in fulfilling this mobility promise to students.5 Analyzing data from the Department of Education’s College Scorecard, we found that at the typical institution, nearly half of the students aren’t graduating, many students aren’t earning sufficient incomes even years after enrollment, and far too many are unable to repay their loans. And in a measurement we developed called the mobility metric, which ranks how well these colleges do for students of low and moderate incomes, we found levels of achievement so abysmal as to call into question the very promise of higher education at many of these schools.

This degree of institutional failure goes well beyond the current discussion about rising tuition costs to the very essence of what colleges spend billions of dollars purporting to do: provide an education worthy of the time and cost associated with it. And this analysis begs the question, what can be done about the quality crisis in our nation’s colleges?

Among our key findings:

1. Chronically Low Graduation Rates:

  • A typical four-year private, non-profit college graduates only 55% of full-time freshmen within six years of enrollment. When students don’t earn a degree within six years, they are unlikely to finish at all.
  • In only 266 schools (26% of all four-year private, non-profit colleges) did more than two-thirds of full-time freshman manage to earn a degree within six years of enrollment. The graduation rates of the remaining 761 schools are so low that if they were high schools instead of colleges they would be flagged for special attention under the recently passed Elementary and Secondary Education Act.

2. Poor Wage Outcomes for Loan-Holding Students:

  • At the average four-year non-profit, just 63% of students who received federal loans earned yearly salaries and wages that exceeded $25,000 (more than the average high school graduate) six years later.
  • At the average institution, 19% of students who had taken out loans were already behind on their repayment just three years out. And at 230 of 1,027 four-year non-profit schools (22%), at least one-quarter of loan-holding students were behind on their payments three years after leaving school.

3. No Discernible Connection between Price and Quality:

  • Using our mobility metric, the average net tuition paid by low- and moderate-income students was lowest at top-quartile schools ($15,938) and highest at bottom-quartile schools ($18,776).

4. High-Performing Schools Enroll Comparatively Few Pell Students:

  • Schools in the top decile of our mobility metric averaged 18.9% Pell enrollment. Schools in the top quartile of our mobility metric averaged 23.5% Pell enrollment. The average four-year, non-profit school had an enrollment of 38.4% Pell students.
  • Only 8.6% of top-quartile schools in our mobility metric ranking enrolled more than 38.4% Pell students (above average). In the bottom quartile, four out of five schools (80.1%) took an above average number of Pell students.

5. Poor Outcomes at Schools with Very High-Density Pell:

  • Among the 214 schools in which at least half the student body received Pell grants, only four are in the top quartile of our mobility metric ranking, and just 22 are in the top half.
  • Of the 465 four-year private, non-profit schools that have Pell enrollment greater than the mean (38%):
    • Only 27% of these schools graduated at least half of their students within six years;
    • Only two-thirds have more than half of their loan-holding students earning greater than $25,000 six years after enrollment; and
    • A mere 30% have loan repayment rates above the average (81%) three years out.

Recommendations in Brief:

The current focus on college affordability is important but too narrow. State and federal governments are the largest payers of college tuition. Recognizing their stake in the quality and outcomes of college, we recommend the following proposals to improve both:

  • Skin in the Game: Colleges should have some “skin in the game” when large numbers of students fail to graduate, gain employment, or pay back loans. Specifically, schools should be required to pay back some fraction of the federal loans their students cannot repay, a rate which can be adjusted based on the nature of the student body at particular schools. Having skin in the game will encourage colleges to care more deeply about student outcomes.
  • Better Teaching & Supports: Every school with a graduation rate of less than 67% should develop and implement a plan to increase student completion, including improving classroom teaching by professors and adjuncts and using data analytics to detect early dropout indicators. Schools need to return to the core mission of instruction.
  • Title I for Colleges: Schools that enroll high numbers of Pell students should receive special assistance from the federal government similar to Title I funding for K-12 education. We must recognize that schools that serve large low-income student populations have unique challenges that require additional resources.
  • Pell Floor: High-performing schools should be encouraged and incentivized to accept and educate far more low-income students. 134 schools have Pell enrollment of less than 19%, which is half the average enrollment for four-year, non-profits. Congress should explore whether schools with low Pell enrollment should be barred from certain funding programs.
  • Open Data: We must end the opacity of college-specific outcomes data to help students, parents, and policymakers discern whether schools are succeeding or failing. Colleges and universities should be required to make public the outcomes they are achieving with different categories of students, so that consumers are empowered to make the best choices about where to invest their time and money.

The Quality Crisis at America’s Private, Non-Profit Schools

While many have touted college as a stamped ticket to the middle class, little attention has been paid to how well these institutions keep that promise to students once they pass through their doors. Are students who go to college learning, graduating, and earning? And are colleges truly serving students for whom they’ve promised to be mobility machines?

To answer these questions we used publicly available data from the Department of Education to develop a mobility metric. This mobility metric is a scorecard of how well colleges are doing with particular cohorts of students: full-time freshmen and students who have taken out federal loans.6 This metric allows us to assess schools taking into account the number of low-income students they take, how many students they graduate, the out-of-pocket costs they pay, their ability to repay student loans, and their annual earnings six years after enrolling.

In our analysis of this data, we found that a large proportion of the 1,027 four-year, private non-profit colleges for which data is available are putting students in a deeper hole than if they had never attended college in the first place. As a result, each year taxpayers are subsidizing at best mediocre results with tens of billions of dollars in federal aid that goes to colleges without any regard to results or accountability in return.

The Appendix of this paper lists our ranking of all 1,027 private, non-profit, four-year colleges. In the future, we will do a similar ranking for 4-year public institutions. Those interested in further exploring this data can find our spreadsheet here. We encourage you to sort the data to your own choosing: by overall score, net tuition expense, Pell enrollment, future salary, location of institution, or loan repayment rate. You will also have the ability to create your own weighting system and create your own mobility metric should you choose to do so.

The purpose of this paper is to ignite a new conversation on college that is focused on quality and outcomes, not just costs. Of course tuition price is important—anyone with a child knows about the burden of rising college costs. But this conversation is too limited, often absolving colleges from taking responsibility for their outcomes. Colleges spend billions of dollars advertising their institutions to students. They take billions in state and federal money. And they make a promise to enrollees that choosing their school will make a positive difference in their lives. That difference must be a positive one, and our findings show that this is often not the case.

What follows are our five major findings. At the end of this paper, we offer recommendations to make the mobility promise of college a mobility reality.

Explaining the Mobility Metric  

Using recently released College Scorecard data for this set of institutions, we have compiled a “mobility metric” that looks at how well four-year, private, non-profit schools are doing across a variety of factors: net price (the amount paid once financial aid has been factored in) for students coming from families that make less than $48,000/year; the percentage of an institution’s students receiving Pell grants (typically students who come from families earning less than $50,000 per year); completion rates within six years for incoming full-time freshmen; the percentage of students earning more than $25,000 annually after six years; and the repayment rate for students three years after they leave school.*  

This publicly-available data is by far the most comprehensive ever available for this type of analysis, but we recognize its limitations which we lay out here. The College Scorecard reports only institution-level graduation rates for first-year, full-time students, and for other metrics, it only reports on students with federal loans. At four-year private, non-profit colleges, 65% of students are first-time, full-time students and approximately 62% of students have taken out student loans.7

Given that many institutions are loathe to share data on their outcomes, this data set is the most comprehensive public reporting of how our colleges are doing in history—and hence worth studying even with the necessary caveats. And the data limitations cannot mask what has become an astounding reality—that many of our nation’s colleges are failing to make good on their promises to students.

* For a full explanation of the data and our mobility composite score, please see Appendix A.

Finding 1. Completion rates are chronically low.

We start with completion rate because it is an important metric and a proxy for future success. Recent data has shown that a college graduate will earn an average of $1 million more in wages over the course of a lifetime than his or her non-college completing peers.8 Students who enter college but do not complete are worse off than students who never entered college at all, according to the Drexel University Center for Labor Markets and Policy.9 And with tuition prices more than doubling in real terms since the 1980’s and students taking out greater amounts of student loans than ever to earn a college degree (an average of $27,000 per student for a traditional bachelor’s degree), at the bare minimum, colleges have a responsibility to do as much as possible to see that students leave campus with a diploma in hand.10

Average Institutional Completion Rates by State

At most schools, students have just over a 50/50 shot of graduating.

  • The data from the Department of Education’s College Scorecard reveals dropout rates that rival the worst of the nation’s high schools.
  • At a typical private, non-profit school, only 55% of first-time, full-time students are graduating within six years.
  • At 415 (41%) of the colleges on this list, half or fewer students earn a degree.
  • Only 170 schools (16.6%) can boast a completion rate higher than 75%.

To put this into perspective, the odds of donning a cap and gown at one of these institutions is tantamount to the flip of a coin. And while many may be quick to blame these figures on the rising costs of college, a recent report from New America found that since 1996, out-of-pocket costs for students and families coming from the lowest two income groups at four-year private colleges has only increased by $73 and $157, respectively, per year of education.11 Writ large, private, non-profit schools are doing better than public or for-profit four-year institutions, often graduating 10-20% more of their students annually than their public and for-profit peers.12  However, with these schools boasting an average sticker price that is more than twice that of their public counterparts, a 55% completion rate is unquestionably insufficient for a system that promises to improve its students’ lives.13

Graduation Rates

There are a handful of colleges that have completion rates so low, it is unconscionable to think that they are allowed to serve students at all. There are fifty four-year private, non-profit institutions where less than one-quarter of students graduate. And at three of these schools, that number drops to the single digits. This includes Paul Quinn College in Texas, which has a 2.3% completion rate, Boston Architectural College in Massachusetts with a completion rate of 7.8%, and Bacone College in Oklahoma with a completion rate of 8.1%.

Three-quarters of schools would be considered dropout factories by K-12 standards.

At 761 of these 1,027 schools, or 74%, less than two-thirds of all full-time students earn a degree within six years of enrolling as a freshman. If these schools were a part of our K-12 system, they would be labeled as “dropout factories.” Under the latest iteration of the Elementary and Secondary Education Act (ESEA), high schools that fail to graduate more than 67% of their students are flagged for intervention and can even face possible closure if they do not improve their completion rates.14 The Alliance for Excellent Education says about 5% of high schools meet this designation.15 Yet for the 74% of private, non-profit schools that have similar outcomes, no bar exists whatsoever to trigger any form of intervention or scrutiny from the federal government—even schools that graduate students in the single digits. Instead, these schools are able to operate without consequences and have no obligation to provide consumers with notification of their poor completion rates.

Dropout factory

At these 761 dropout factories, 60% take an above average number of Pell students—often the proxy for the number of low- and moderate-income students a college takes in—and close to half charge an above average net price for tuition. In fact, 1 out of every 5 of these schools takes both an above average number of Pell students (>38%) and charges students a net price higher than most schools (>$17,620). For example, the College of New Rochelle in New York takes in more than 75% Pell students each year, charges them an average net price of more than $32,000, yet only graduates 27% of their students. As a result, there is a high number of institutions that are not only failing to graduate a large subset of their students but appear to be targeting Pell students and charging them a lot of money in the process.

Top 10 Worst Dropout Factories

*Completion rate is for students.
**Net Price refers to average net cost for students with incomes <$48,000.

There are some notable exceptions.

However, such an outcome is not always sealed in fate. A handful of schools on our list prove that it is possible for schools to have a high graduation rate while taking in a substantial number of Pell students. For example, Spelman College in Georgia boasts a graduation rate over 70%, even with more than half of their student population receiving Pell grants. In addition, when looking at the top quartile (257 schools) in our mobility metric, the following schools manage to take in more than 20% Pell students and also graduate more than 85% of their students:

 Overall, our nation’s four-year private, non-profit schools have a long way to go to ensure that more students who enter college are able to leave with the diploma they need to successfully enter the workforce and pay off any student debts. Consumers should be given more information upfront about which schools are successfully meeting these goals, and which schools are failing to set up students for success.

Finding 2. There are poor wage outcomes for many students.

For many Americans, a college degree is the gateway to financial stability and growth. A New America study found that the top two reasons students go to college is “to improve employment opportunities” and “to make more money.”16  Expecting this type of wage premium is particularly important for low-income students, who on average pay the equivalent of 84% of their earnings to attend college compared to only 15% for their wealthier peers.17 Any reading of a college brochure is replete with appeals to students on the ability for them to start a successful career.

The question is: how well are they fulfilling this promise?

Six years out, wage outcomes for loan-holding students are low.

The expected median earnings of a high school diploma-holding adult is in the vicinity of $25,000. For this analysis, we measured the number of loan-holding students who are earning more than $25,000 six years after enrollment. According to the College Scorecard:

  • At the average four-year private, non-profit schools, only 63% of students who started school with loans earned in excess of $25,000 six years later. This figure omits graduates who were in graduate school as full-time students.
  • In 597 out 1,027 (58%) of schools, fewer than two-thirds of loan-holding students earn in excess of $25,000 after six years.
  • At 17% percent of schools, less than half of students earned at least $25,000 six years later.
  • At 169 colleges (16.5%), less than half of all students earned more than $25K.

Salary Outcomes at the Average Four-Year, Non-Profit College

Loan repayment rates reflect poor wage outcomes.

Another important metric of quality is repayment rate—or the percentage of students who are able to pay at least one dollar towards their principal balance within three years of their loans becoming due. With low post-college wages, one might expect high delinquency rates for student loans. And what we see here is the data bears that out.

  • At the average private, four-year school, 19% of students found themselves behind on their student loans just three years after leaving school. This figure includes both those who have earned a degree and those who have not. 
  • At 230 of the 1,027 schools (22%), at least one-quarter of students were behind on payments within that three year span.

To put these repayment rates into perspective, during the height of the housing crisis in 2010, 90-day mortgage delinquencies peaked at around 10%.18

Loan repayment rates

In recent months there has been some bipartisan interest in making sure colleges have some “skin in the game” when it comes to schools with high delinquency rates.  Senators Jeanne Shaheen (D-NH) and Orrin Hatch (R-UT) introduced legislation last year that would create a “risk-sharing” program requiring schools with a low repayment rate to contribute to a fund to support institutions that serve a high percentage of low- and moderate-income students.19 Other proposals have also called for schools to pay a penalty if their default rate falls below a pre-determined threshold.20

There are notable exceptions.

The schools that show the best earnings outcomes are not necessarily the most recognizable names in the higher education universe. The top ten ranked schools for producing graduates who earn more than $25,000 six years after enrollment were those with a specific professional focus like pharmacy, nursing, business, and technology.

Top 10 Schools for Students Earning More than $25,000 After Six Years

Notably, three schools within the top decile of the earnings measure take in an above average number of Pell students (>38%). These schools are: 

High Pell, High Earnings

There are also a handful of schools that have exceptionally high repayment results for their students. In particular, Harvey Mudd College boasts a 100% repayment rate, while Albany College of Pharmaceutical and Health Sciences and Bates College each have a payback rate higher than 99%. Of particular note are the 168 schools with an above average repayment rate (>80%) and an above average enrollment of Pell students (>38%). The top ten in this list include:

Repayment Champs, 10 ten schools for repayment

For too many students, the expected financial stability of a college education is simply failing to materialize—and the institutions who promise but don’t deliver are in no way held responsible for this failure.

Finding 3. Tuition and outcomes are unrelated.

With most products, you assume if you pay more, you will get better quality. Of course, nothing is ever that simple with college—especially because the sticker price for tuition and the actual price for the student bear little resemblance. For many families, college is one of the most significant purchases they will make in their lives, but there is little recourse if that investment doesn’t pan out.

There is no discernible connection between price and quality.

Using both the College Scorecard’s calculation for net tuition price for students with incomes less than $48,000 and our mobility metric, we found absolutely no correlation between the price of the school and the quality of the outcome for the student.

Among students who come from families making less than $48,000 per year in pay:

  • The average net tuition at top-quartile schools from our mobility metric was $15,938 compared to $18,776 for bottom-quartile schools.
  • The quartile of schools with the worst mobility outcomes charged the most, while the quartile of schools with the best mobility outcomes charged the least.

This means that students attending schools with lower mobility outcomes (like completion rates, earnings, and repayment rates) are actually paying more for their education than their peers attending higher quality schools.

There are notable exceptions.

There are colleges offering a high quality education without making low and middle-income students pay as steep a price. When reviewing the top decile of schools for net price, we find that it is actually some of the most prestigious schools that offer some of the best prices to students coming from families that earn less than $48,000 a year. The top ten of these schools include:*

Best Bargains, Top 10 most affordable schools

*This list excludes schools from Puerto Rico, which have an average net price that is more than $10,000 less than the national average net price for private, non-profit schools.

Some of these high-performing schools enroll a large number of Pell students–Kettering, Metropolitan College, and Brigham Young, for example. However others, like Harvard, Haverford, Stanford, and Pomona do not.

For most families, college is one of the most significant purchases they will make in their lives—but unlike shopping for a car or house, there is no recourse for consumers if that investment doesn’t pan out.

Finding 4. Many high-performing schools enroll small numbers of low-income students.

For many students, college is made possible by access to federal Pell grant funding. In 2013-14 alone, 91% of Pell recipients came from families with an adjusted gross income below $50,000, as based on a formula that takes into a variety of factors, such as a students’ financial need and their anticipated price of attendance.21 Thus, not all Pell students are poor, but it is a good proxy for students who most need college to be a mobility machine.

Using Pell as a proxy for colleges’ commitment to serving as mobility engines, we found that many of the highest ranking and elite schools do the least. There are notable exceptions which we highlight below, but the broad brush for the best schools is they are lagging in their commitment to enrolling lower income students.

The schools that can do the most generally do the least.

For the four-year private, non-profit schools we reviewed in our mobility metric, the average Pell rate is 38%. Yet for schools in the top quartile of the mobility ranking—i.e. those that did the best on completion, earnings, and repayment rates—the average is significantly lower, at 24%. There is only one school in the mobility metric’s top 25—Columbia University—that has a student population comprised of at least 20% Pell students. For the top ten schools in the list, the Pell rate falls consistently within the teens, with Stanford University, Duke University, and The University of Pennsylvania rounding out the first three spots with Pell enrollment rates of 16.0%, 13.7%, and 14.2% respectively.

Very Few Pell

These schools have the capacity to take in more Pell students, but in recent years, many have chosen to increase the amount of merit pay they disburse at the expense of need-based aid—ostensibly to lure “top talent” to their schools.22 Similar findings were recently highlighted in a report by the U.S. Department of Education, which found that it is often admissions policies at highly-selective schools that favor wealthier students due to an emphasis on things like extracurricular activities and “legacies” (familial relationships with alumni).23 Schools that have a proven track record of success with Pell students have a responsibility to take in a greater share of students who could benefit from those outcomes the most. The Department of Education cites research that have found that “highly-selective schools could increase the representation of low-income students by 30% without compromising SAT or ACT standards."24

 There are notable exceptions.

But not all top-performing schools are shirking their responsibility to take in a larger proportion of Pell students. When looking at the top decile of schools from our mobility metric, 16 currently have a Pell population greater than 25%.

Top Pell performers

Overall, schools that have proven track records of success should be able to take in more students of modest means who have the most to gain from a college environment that is equipped to help its students succeed in today’s economy. We know that schools can do well with higher concentrations of Pell students, and more colleges at the top of our mobility metric should share in the responsibility to serve them in greater numbers.

Finding 5. Very high-density Pell schools often have poor student outcomes.

Since 2000, taxpayers have spent over $300 billion on Pell grants, including $31.4 billion in 2015 alone. 25 Yet most taxpayers know very little about how well this investment is paying off. Colleges and universities are not required to publicly report graduation rate data or other outcomes for students who use Pell.26 It was only in 2008 that the latest reauthorization of the Higher Education Act (HEA) mandated that institutions must disclose Pell graduation rates—but not to the public, only to prospective students who specifically ask for them. This requires organizations like the Education Trust to have to spend almost a full year tracking down Pell student outcomes, where such efforts recently uncovered a 14-point national completion gap between Pell recipients and those who do not receive Pell grants.27 Earlier this year, the U.S. Department of Education announced that it would for the first time include a cohort of Pell recipients in the IPEDS outcome measures survey for 2017-2018, making this type of institution-level information more readily available.28 This data is critically important, because we found that exceptionally high Pell density coincides with poor student outcomes.

Schools with very large Pell populations struggle.

Among the 214 schools with student enrollment comprised of 50% or more Pell grantees, only four are in the top quartile of our mobility metric and a mere 22 are in the top half. On the flipside, we see that 61.2% of these schools are in the bottom quartile of our mobility ranking. That means that many of our country’s students who are using Pell grants to attend four-year, non-profit schools are concentrated within a set of schools that are having poor outcomes and, thus, do very little to increase economic mobility.

Of the 465 four-year private non-profit schools that take in a Pell students greater than the mean (>38%):

  • Only 27% graduate more than half of their students;
  • Only two-thirds have more than half of their students earning greater than $25k six years after enrollment; and,
  • A mere 30% have a repayment rate higher than the average for private, non-profit schools (>81%).

In fact, only 73 schools on our list that take more than the average Pell (>38%) graduate more than 50% of their students and have a net price lower than the average of $17,620.

Percentage of Pell Students at Top-Ranking 4-Year Private,  Non-Profit Schools, Compared to the Average

There are notable exceptions

Luckily, some schools are bucking this trend by taking in an exceptionally high percentage of Pell students and producing relatively strong outcomes for them. In fact, when reviewing the top quartile of our mobility metric, four schools take in more than 50% Pell and have been able to achieve high success. Interestingly, three out of four of these schools are either predominately female (College of Mount Saint Vincent) or all-female (Salem College and Mount St. Mary’s College).

Mobility Machines

Though many of the premier private colleges in the nation take very few Pell students, several of them—Brigham Young University, Spelman College, and DePaul University—have more than 30% Pell enrollees and excellent mobility metrics. There are also six schools that take higher than the average number of Pell students (>38%) and see a graduation rate higher than 67% (meaning they would not be classified as “dropout factories” under K-12 standards).

High Pell, High Completion: Schools That Take Above Average Pell  and Graduate More Than 67% of Students

Low- and moderate-income students count on college the most to provide them with the tools they need to have a comfortable middle-class lifestyle. But as this data shows, too many Pell students are ending up at private, non-profit institutions that are doing little to actually help them reach this goal. Students deserve more information upfront about the type of outcomes they can expect to receive in order to select the institutions that are best equipped to help them succeed.

Addressing the Problem

The discussion on college has focused almost exclusively on access and cost. Both issues are important, but they have overshadowed what may be the most important issue of all —quality. When colleges advertise for students, accept millions of dollars in state and federal grants to subsidize it, take checks from students and their parents, and convince them to take out loans to attend, they are making an implicit bargain to provide the students who walk through their doors a better life. The data presented in our mobility metric brings into focus the startling truth that many colleges are simply not fulfilling their end of this bargain. If America’s high schools had the same outcomes as these 1,027 institutions, there would be an outcry. But for college—and in this case four-year colleges taking in high school graduates who have been accepted through an application process designed to weed out students that are a poor fit—it is met with a shrug of the shoulders. And too often, the implication is that it is the student, not the institution, who is at fault.

Congress will have an opportunity to address this during the coming reauthorization of the Higher Education Act (HEA). Specifically, new policies must focus on quality, student retention, and student outcomes. Elite and high-performing schools should be encouraged to take in more students of modest means. There must be more consumer transparency around college outcomes. Below are a series of recommendations to help colleges fulfill their mission of being mobility machines. It should begin with colleges rededicating themselves to the core mission of educating and graduating students, as well as helping them find their way into the working world.


  1. Colleges should have “skin in the game” when large numbers of students fail to graduate, gain employment, or pay back loans. Today, the risk for financing higher education falls almost entirely on students and their families. Colleges pay no price when students they admit to their schools do not succeed. Colleges shouldn’t bear all of the burden for student failure, but they should bear some of it. Colleges should pay a portion of the cost of their federal student loan delinquencies. This risk-sharing amount should be adjusted based on the socio-economic breakdown of the student body at particular schools in order to not penalize schools who take in a larger share of low- and moderate-income students. By having skin in the game, college presidents will be encouraged to think more about outcomes of students, not just acceptances. Schools with poor employment outcomes should also be required to have measurable and actionable plans to connect graduates with employment opportunities.
  2. Every school with a graduation rate of less than 67% should be required to develop and implement a plan to improve student completion. A high school with a completion rate lower than 67% is considered a drop out factory under federal law, and those schools are required to intervene to turn their failures around. Colleges that fail to graduate more than two-thirds of their students should also be required to take concrete steps to improve their outcomes. At the minimum, these schools should be required to make a plan that demonstrates a meaningful effort to help students cross the finish line, including, but not limited to: wholesale efforts to improve classroom teaching by professors and adjuncts, developing notification systems that can alert students to important enrollment and financial aid deadlines, and using data analytic systems that can closely monitor and detect early dropout indicators and intervene while there is still time.
  3. Incentivize high-performing but low-Pell schools to take more low- and middle-income students. The average four-year private, non-profit college has an enrollment that is 38% Pell. Elite schools often boast about their economic diversity, but in reality, only two schools in the top decile of our mobility metric enroll the four-year, non-profit collegiate Pell average or better. We suggest a minimum floor of Pell enrollment of 19% for all schools, particularly elite colleges. That would be an enrollment rate half the national average for non-profit, four-year schools. Yet currently, sixty of the 103 top decile schools have Pell enrollment of less than 19%. This floor would be voluntary, but schools that take fewer than 19% should lose some of the billions of dollars of federal subsidies they get from taxpayers. Given the enormous support that elite schools get from tax money, they should bear some responsibility to do more to educate those of modest means. This would also reduce the strain on schools that currently take the lion’s share of Pell recipients.
  4. Recognize, support, and reward schools that are doing well with large concentrations of Pell students. In our K-12 schools, the Title I funding formula structure recognizes that it is harder to educate concentrations of low-income students, and that the schools or districts that do so deserve additional supports. Yet colleges can access the same amount of Pell money per student regardless of whether they have 2,000 Pell students enrolled or two. Supplemental funding should be available to schools that are taking in a substantial proportion of Pell students and are able to demonstrate positive outcomes. This proposal is similar to the “Pell Bonus” idea that was championed by former Senator Tom Harkin (D-IA) in his last attempt to reauthorize HEA prior to leaving office.29
  5. End the opacity of college data to help students, parents, and policymakers discern whether schools are succeeding or failing. It is clear that there is a wide disparity in the quality of our four-year private, non-profit colleges and universities—a disparity that often has no relationship to the cost of attendance. Yet there is very little actionable information available to consumers to help them determine which institution will give them their money’s worth in education and best set them up or their children for success. While the College Scorecard website recently released by the Department of Education is a good start, it does not equip students and families with all of the information they need to understand how well a particular student will fare at any given institution—in part because institutions refuse to share that data. In exchange for the billions of dollars of taxpayer money they receive each year, colleges and universities should be required to make public the outcomes they are achieving with different categories of students, so that consumers are empowered to make the best choices about where they should invest their time and hard-earned money. More consumer power will drive college administrators to seek better outcomes for students.


If America’s high schools had outcomes like America’s colleges there would be Blue Ribbon Commissions lining the streets of Washington. As it is, there is virtually no discussion about college quality and outcomes. In this paper, we hope to ignite this conversation because the key fact remains that a college degree is the surest ticket to the middle class. The discussion on cost is important, but a discussion on quality and outcomes is also urgently needed.

APPENDIX: Methodology

Mobility Metric

The U.S. Department of Education’s College Scorecard data—compiled from 7,000 colleges and universities over the past 18 years—was used for the mobility metric featured within this report. The data was acquired from the Department of Education’s College Scorecard technical website (, which became public in September of 2015. The institutions used for analysis were four-year private, non-profit institutions that predominately awarded Bachelor’s degrees. No other filters were applied to limit the institutions used for the analysis. However, some institutions did need to be excluded due to their lack of data within the fields required by the analysis, detailed below. A total of 1,027 four-year private, non-profit colleges gave adequate data to be included within the analysis.

The following data fields were used to determine the mobility scores featured in this report:


The “mobility metric” refers to all of the data points that create the mobility composite score for each college. The composite score for the mobility metric was developed through several steps. First, each of the five variables were derived for each relevant college. For each variable, the colleges were ranked and a predetermined weight was applied to each data point. The weights allowed for net price to comprise 15% of the final score, completion rate to comprise 20% of the final score, repayment rate to comprise 25%, earnings 25%, and Pell Grant 15%. Summing the scores from the five variables created the composite score for the mobility metric. Colleges could then be ranked by composite scores to determine their impact as being mobility engines for their students. 

Colleges that were missing multiple fields of data were excluded from the analysis. Colleges that were missing only one data field and had supplementary data were salvaged if possible. For example, if the average net price for students whose families earned $0-48,000 was missing but the average net price for all students was present, the mean difference between all relevant schools that presented data for both variables was determined and applied to the average net price for all students for the schools that were missing data. By doing this, a pseudo-score was created that accounted for average differences and allowed schools to remain within the analysis rather than being excluded entirely.

Explore the data: Click to download 

Limitations of Original Datasets

There were several limitations due to the original datasets that were used to compile the College Scorecard data. The National Student Loan Data System (NSLDS), which provided average net price, repayment rate, and the percent of Pell Grant students, only contains data from students who are eligible for Pell Grants and/or federal student loans. This qualification should not impact average net price because most students whose parents earn $0-48,000 per year should qualify of Pell Grants. Similarly, the NSLDS should be the best source for determining the percent of Pell students in an institution. On the other hand, the repayment rate may not be fully representative of the institution because it includes only students within the NSLDS database. However, this should not impact the results of this report because its primary focus are the outcomes for those individuals who are not high-income and thus should be represented within NSLDS.

The Integrated Postsecondary Education Data System (IPEDS) only contains data from individuals who are first-time, full-time students and does not include students who transfer and then graduate. This qualifier is most important for colleges with high percentages of students who were part-time, previously enrolled at another school, or who transfer out of their first college. Community colleges typically exhibit the greatest percentages of part-time students and adults who are not first-time students, but community colleges are not present within this analysis. The validity of completion data for four-year private, non-profit institutions within this analysis is impacted to the degree that students are not first-time, full-time students.

Validity Concerns

Various other measures can impact the validity of the data used within this report. For example, the validity of the average net price metric decreases if there are very low number of Title IV students whose parents earned $0-48,000 because results are based on very few instances. Similarly, very low levels of Title IV students impact the validity of the repayment rate and percent Pell metrics. Very low levels of first-time, full-time students impact the validity of the completion rate metric. Additionally, to avoid variability that can take place across years, especially for colleges with low enrollment, completion rate data is pooled over two years. Repayment rate data stemming from colleges with very few Title IV students was also suppressed for low numbers. It must also be noted that the metric showing the percentage of students who obtain wages over $25,000 is the most dated metric. Not only did the students who are analyzed for the metric graduate in 2006, the data was reported in 2011, whereas data from other metrics were reported in 2013.

It is important to note that College Scorecard data is presented at an institutional level, not a programmatic level. All five metrics may vary greatly by programs within the same college. For instance, a business school within a college may result in a very high percentage of students earning over $25,000 per year, whereas students obtaining social work, philosophy, or art degrees may have different outcomes. To this end, a college may be a successful mobility engine for some students, but not for others, even if the college has a high mobility score. Ideally, future data could be reported differentiated by program to show students what specific outcomes they can best expect from particular programs, and to afford students the ability to compare specific program outcomes at various colleges. 

End Notes