The final straw was the letter a lawyer for a private nonprofit college sent just before 5 p.m. last Wednesday. It said any statement about the college facing an imminent risk of closure would be false.
“The publication of such a statement would be grossly irresponsible and would cause great harm to the college, and demand is hereby made that you refrain from such publication,” the lawyer wrote.
The letter followed a flurry of similar calls and emails from colleges and their representatives about a list a college advising company planned to release the next day. The list featured projections of how many years 946 private colleges have until they could run out of money and close.
This communiqué, however, was addressed in part to Edmit, the start-up advising company. The letter convinced Edmit’s two co-founders to scuttle the release of their financial modeling tool.
Inside Higher Ed had planned to write about the projections, but the company's decision changed that plan, too. The goal was to foster an open discussion about what has become an elephant in the room for higher education: Which financially stressed colleges are likely to shut down, and what can be done to protect students and taxpayers from abrupt closures?
“At this point, there’s zero consumer protection like this. And I don’t think that’s the optimal amount,” said Doug Webber, an associate professor of economics at Temple University. “If a school goes under, it’s really bad for the students.”
Webber and several other experts on higher education finance had examined the source data, methodology and projections from Edmit, which were created with assistance from graduate students and two economists at Brandeis University’s International Business School.
Webber also is one of six experts on Edmit's "data integrity council," an advisory group that helps the company vet its practices. He is unpaid in the role and was not involved in the creation of the company's forecasting tool on private college finances. (Note: This paragraph has been added to a previous version of the article to include new information from Edmit about its affiliation with Webber.)
The projections used qualitative and quantitative data, from federal sources, to estimate how long before the net expenses for the 946 private colleges exceeded their net assets. After that, the model assumed the colleges would fail, because no enterprise can continue to operate without taking in enough to pay its bills. The model provided that information in a single number so it would be accessible. That number was the estimated time until closing for each college.
The model analyzed “what the school’s finances are like if they stay on their current path,” said Stephen Cecchetti, one of the two Brandeis economists who helped oversee the project. Put simply, it’s an estimate of financial resources divided by the college’s “burn rate,” he said.
Edmit based the model on four primary variables: investment return on endowment funds, tuition prices, tuition discounting and faculty and staff member salaries.
Most of the underlying federal data was from 2017, creating a lag time that could skew projections. The potential for errors with this sort of statistical modeling is substantial, experts said. And the model’s four variables might fail to fully or adequately capture a college’s projected financial sustainability.
Yet several experts said the Edmit model wasn’t missing crucial data sets.
“Speaking generally about this kind of data, it should be out there,” said Webber. “It could be really useful.”
‘The Wild West’
Nick Ducoff and Sabrina Manville rolled out Edmit in early 2018. The Boston-based company caters to prospective students by providing college advising services and tools so families can “treat college like an investment.”
Edmit offers admissions and pricing data for more than 2,000 colleges, features a college cost comparison tool and conducts personalized financial aid estimates for students.
A few months after the company went live, Mount Ida College shut down abruptly. The closure of the private college in Massachusetts set off a cascade of problems for its 1,500 students and added to worries about regulators’ lack of preparation for private college failures.
In the aftermath of Mount Ida's closure, some of Edmit’s customers began asking about the financial health of colleges as part of their enrollment decisions, said Ducoff, the company's CEO and co-founder.
“We didn’t really have good answers,” he said. “The consumer certainly doesn’t have a playbook around this.”
Ducoff and Manville, Edmit’s other co-founder, decided to try to create a modeling tool that would gauge private colleges' ability to survive financially. They focused solely on private institutions, in part because it would be unworkable to make projections on the fiscal health of public colleges without being able to factor in future shifts in state funding.
The sample included 946 nonprofit colleges that provided sufficient data to make projections based on the U.S. Department of Education’s Integrated Postsecondary Education Data System (IPEDS) and other federal reporting sources, such as tax filings. The period examined was from 2002 to 2017.
Edmit's founders wanted to start a conversation about the stability and viability of college business models, aiming primarily at trustees, faculty members and others who work in and around higher education, Manville said. The company has not used the projections with its customers.
“Colleges would be doing this kind of analysis primarily to support their own goals: to inform them about their competition, to put themselves in the context of what else is happening in their region or in the sector,” said Manville. “We wanted to create something that was more transparently available to families and also simple enough to understand for a broader audience so that it can actually inform people’s decision making and understanding of what’s happening.”
Students, families, policy makers and even higher education leaders increasingly are questioning the business model of some private colleges amid a steady drip-drip-drip of campus closings. At the same time, accreditors and state and federal agencies -- higher education’s regulatory triad -- are considering methods for how to better monitor the financial sustainability of colleges, in part to warn students and parents about institutions that appear to be on the brink.
Part of the problem is that these agencies, which are tasked with protecting students and taxpayers if a college shuts down, tend to lack clear standards for what’s considered acceptable for teaching-out a campus and otherwise ensuring that students can finish earning their credentials or have their loans forgiven, according to Ben Miller, vice president for postsecondary education at the Center for American Progress.
“Right now, it’s the Wild West,” said Miller.
But government agencies and accreditors could do more, Miller said. If regulators are worried about a college's finances, they need to take action.
“They have a moral responsibility to warn students,” he said.
The stakes are high for students. An abrupt college closure can thrust them into chaotic situations. Even students at colleges that close in an orderly fashion can suffer negative effects. Their credits might not transfer fully, or they might have to travel farther from home to get to class. Some never re-enroll after their college closes.
Alumni can feel the pain, too, as their alma mater’s brand takes a hit after a closure. College faculty members and other employees also can have their lives upended. And donors might wonder why they gave money to a failing enterprise.
Edmit shared the projections with Inside Higher Ed this summer, after requiring that they be kept private until the company was ready to post them publicly.
The company's co-founders planned to publish the projections on Github, a platform for open-source projects, under its own logo and the Inside Higher Ed banner. The source code was available, and a lengthy explanation was planned, saying the list was an early measure, that its developers were seeking feedback and potential improvements, and that students and parents shouldn't base college decisions on it.
Inside Higher Ed's planned coverage included a news article that described the list, interviewed the presidents of some colleges that it described as being at imminent risk of failure and discussed the limitations of such a data set. A second news article reviewed 27 years of higher ed's uneasy history with public financial metrics before delving deeply into the limitations of any data or predictive model. An opinion piece, from Ducoff and Manville, was to explain why it was important to share the information in question.
About a week before the target publication date, a reporter reached out to several colleges on the list. They were picked to represent a range of institutions facing different degrees of projected money woes.
For example, Utica College was selected because of previous reporting on strategies the college had taken to strengthen its finances. Utica seemed like a possible case where the projections were upended by canny moves by college leaders.
Utica threatened to sue if Inside Higher Ed published an article on Edmit's projections.
Another institution, Herzing University, was selected because of the projected vulnerability of its branch campuses. Confusion about federal data on branch campuses also could skew the projections.
While the projections might not fully capture the financial health of some colleges, Edmit had evidence that the forecasts could be accurate. That's because several colleges included in the modeling tool have shut down during the last several years. Almost all of those colleges had precarious finances, according to the projections.
Here are the model's estimates for how long it would be before those college would have been at risk of closing: Southern Vermont College (four years), Green Mountain College (six years), Marylhurst University (six years), Concordia College of Alabama (six years), Marygrove College (seven years), Newbury College (seven years) and Grace University (seven years).
Likewise, the model was generally consistent with projections about the annual number of private college closures made by Moody's, the financial ratings firm.
Fragile College Turnarounds
A letter from a law firm representing Herzing called the projections false and misleading.
“It would be reckless for a respected higher education publisher such as Inside Higher Ed to make such predictions based on old, incomplete, and inaccurate data and an admittedly flawed model,” the lawyer wrote.
The university insisted that the model was incorrect in referring to federal data sets for the finances of its branch campuses. It said any projections instead should cover the full institution.
This issue isn’t new and has been called the “longtime nemesis” of higher education researchers. The use of federal Office of Postsecondary Education identification codes by some college systems has proven confusing when regulators have sought to oversee their finances and outcomes. The codes can split institutions into different entries for their branches, and the finances for individual campuses might look weaker than for the university as a whole. This issue cropped up with the former Corinthian Colleges, for example.
After several exchanges, Herzing was asked how to find federally collected financial data that accurately represented its full system. (Note: This paragraph and the following one have been changed from previous versions in response to a statement Herzing sent after the article was published.)
"The Herzing IPEDS information is not aggregated," a university spokesman said. "But even if it was, it includes less than 20 percent of our population because it counts only first-time, full-time students."
Several colleges and the National Association of Independent Colleges and Universities were shown a Github site featuring the data and projections, as well as a raw paper from Brandeis students on the methodology.
Shortly thereafter, the projections and methodology began circulating between private colleges and their lobbying associations.
“I get Edmit’s desire to be partnered with you with but I gotta say, I’m surprised IHE has hitched its wagon to this,” Pete Boyle, a spokesman for NAICU, said via email. “How much sense does it make that four short-term data points can define a college’s long term future, and that colleges do not change and adapt to challenges over time?”
The report on the methodology lacked specificity, explanation and breadth, Boyle and others said. The supporting data regarding school closures were questionable, the background research on the choice of explanatory variables and method was lacking, and supporting arguments for choosing the variables were absent. The assumption that the explanatory trends will continue was the most troubling aspect of all, they said.
“To look 10 years down the road in higher education is dangerous (what will happen with HEA, for example?),” Boyle said, referring to the long-delayed reauthorization of the federal Higher Education Act. “And to look 50 or 100 years down the road is worse.”
However, Boyle said private colleges recognize the real concern about institutions in dire straits that stay open too long, then close abruptly and leave students out in the cold.
“It’s a tough conversation,” he said.
Economists tend to make lots of projections about the future. Done responsibly and presented with caveats, such projections can have value, the consensus holds.
One college president emailed with the subject line “IHE Article Puts Students and Colleges at a Greater Risk?” before going on to explain that his college was fixing its problems.
“Given the fragile nature of a successful higher education turnaround,” he wrote, “even small external threats can undermine or complicate such efforts, even putting present students at greater risk.”
It wasn't a surprise that many college leaders and their associations fought the publication of this data.
Many said no number can capture all of the factors that go into keeping the doors open at complex, multimillion-dollar operations like colleges. Others may have felt their colleges were on the brink of collapse and had to fight against unflattering media coverage with every available resource or risk that collapse accelerating.
Several private college representatives said they generally trusted Inside Higher Ed to start this conversation with the right care and tone, but they worried about less nuanced chatter that might follow. This could create a “run on the bank” effect of worsening a college’s fiscal situation by publishing details about its survivability.
Ducoff and Manville shared this concern. That’s why they tried to avoid false positives, such as by not requiring a cash cushion for colleges in the forecasts. That means the model was too conservative in some cases. For example, Mount Ida was projected to last indefinitely.
“The point of this is not to wag fingers at financially precarious colleges,” said Ducoff.
When contacted, not all colleges questioned the newsworthiness of the projections.
One college president said recent college closures had caught the attention of his peers. “Right now, everyone’s gotten the cold water and realizes they have to do something,” he said. “The question is what to do.”
The president said trustees need to provide reality checks for colleges that are facing financial challenges, in part by asking if the institution realistically can fulfill its mission.
In the end, Inside Higher Ed could not publish its news article about Edmit’s scuttled projections.
Information about the model and its forecasts have made the rounds between private nonprofit colleges. But anything like those forecasts remains out of the hands of students and families.
These are indeed tough conversations. Yet for now the industry would prefer to keep those discussions private.