Student debt has quadrupled, to more than $1.1 trillion, over the past 14 years. Defaults are at their highest level in more than 20 years. This crisis in student debt, emerging during the past decade, has roiled students, their families, policy experts—and politicians.

But wait: This crisis is narrower and more focused than anyone had thought. New research shows that it has been fueled by a very specific segment of nontraditional students, who attend for-profit schools or two-year institutions like community colleges. Half of all student borrowers in 2011 attended a for-profit or two-year college, yet this group represented 70 percent of defaults, according to economists Adam Looney of the Treasury Department and Constantine Yannelis of Stanford University. “To the extent there is a crisis, it is concentrated among [these] borrowers,” they write in a groundbreaking paper unveiled at a Brookings Institution economic conference last week, the first research to delve into both tax data and student-loan records.

Typically, nontraditional students are older or first-generation students who don’t go directly from high school to college. Many live in poor neighborhoods and enroll in programs they are less likely to complete. And, when they do enter the job market, they don’t necessarily earn as much money, or benefit from their degrees, to the same extent as traditional students who finish private or public four-year institutions. It’s this population of students, along with the rise of for-profit universities, that account for most of the spike in student debt.

“We now know that we have this population that is not being served well,” says Beth Akers, a fellow in Brookings’s Center on Children and Families. “That is what we really want to tackle. We don’t want to perpetuate this system.” This new research offers an opening for policy experts to target their proposals at the true roots of the problem, to help nontraditional borrowers as well as the schools they attend.

Politicians, always attentive to voters’ economic anxieties, have taken up the issue of student debt on the stump. “College is supposed to help people achieve their dreams, but more and more, paying for college actually pushes those dreams further and further out of reach,” Hillary Clinton said in New Hampshire last month as she unveiled a $350 billion college-affordability plan that would (among other things) ensure that borrowers needn’t spend more than 10 percent of their incomes to repay student loans. Sen. Bernie Sanders, her rival for the Democratic presidential nomination, wants free tuition at all four-year public colleges and institutions.

Among Republican candidates, Jeb Bush has extolled the virtues of free community-college programs in states like Tennessee. Sen. Marco Rubio, who has struggled with his own student loans, has praised the idea of basing repayment plans on students’ post-graduation income.

But the problem, this latest data suggest, may not lend itself to such broad-brushed solutions. The danger of debt depends dramatically on the type of college the student attends. Students who go to most nonprofit and public four-year institutions and graduate schools default at lower rates, the Brookings research found. A whopping 27 percent of student borrowers at for-profit schools defaulted on their loans in 2011, compared with just 7 percent at selective colleges.

That’s largely because these traditional students usually finish their degrees, even in the wake of a devastating economic recession. That way, they earn more money in the labor market, reducing the burdens of repaying their loans.

Nontraditional borrowers, in contrast, borrow lots of money to attend schools with poor track records of actually granting degrees. The jobs they find pay relatively low wages and make it harder to repay their student loans. Median earnings of traditional borrowers from selective schools in 2013 came to $48,000, the Brookings paper found, more than twice the $23,200 that nontraditional borrowers from for-profit schools earned.

The Brookings research also shows that outsized loan balances, although becoming more frequent, are still relatively rare. In 2013, more than 10 percent of all borrowers owed more than $50,000, and roughly 5 percent owed more than $100,000. (In 2000-2001, a mere 7 percent of borrowers carried balances greater than $50,000.) What matters most in cases of high balances is the type of degree students obtain for the money, and the earnings that this degree might bring. Only about a third of borrowers with balances of more than $50,000 needed the loans as undergraduates, and a majority of those attended for-profit colleges. The rest of the borrowers went to graduate or professional schools, which typically offer a surer bump in potential earnings.

Another upside of this new research is that the spike in borrowing by nontraditional students for nontraditional higher education isn’t expected to continue. Why? The economy, stupid. The weak labor markets produced by the 2008-2009 recession prompted Americans to return to school in droves, in search of better credentials. But, as the economy has improved, the number of people returning to school from the workforce is already declining.

“The number of new borrowers at for-profit and two-year institutions has dropped substantially, due to the end of the recession and to increased oversight of the for-profit sector,” Looney and Yannelis write.

This may be the best news of all. If fewer people borrow heavily to attend schools that won’t help them earn big bucks, the crisis of student debt may ease on its own.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.