In 2011, we were opening a brand new graduate school of education. Ours was the typical start-up problem. We had no track record. Why should customers—recent grads from selective colleges who wanted to become schoolteachers—choose us?

The top graduate schools, like Harvard, had established reputations and gorgeous campuses. The upstart graduate schools, often operating online, spent huge amounts to acquire a customer—as much as $14,000 in marketing expenses to acquire a single masters student.

We had none of that: no prestige, no campus, nor that kind of advertising budget.

What was our competitive advantage? We were willing to live or die on job placement. Many education school graduates get their degrees in May without a job offer in hand. New teacher hiring often happens in June, July, and even August. If students borrow with conventional student loans, they must repay them even if they don’t get hired. For those who earn a master’s degree in teaching, this happens quite a bit.

We believed we could generate teachers that principals would really want to hire, and that our graduates would get multiple job offers in springtime. So why not finance things differently?

We drafted an offer to prospective students: “We believe principals will covet teachers who get our degrees. Therefore we’ll absorb all of the risk that you won’t get hired. Assuming you get hired into a new school of your choosing, and you’re earning $40,000 to $60,000 as a starting salary, you’ll pay us back the tuition over 2 years, interest free.”

That offer is an example of what is called an income share agreement. There are different ways to shape such offers, which are becoming increasingly common in higher education notwithstanding a lack of federal regulatory clarity about them. University of Utah, for example, offers a deal by which students can pay 2.85% of their income for 3 to 10 years, up to two times the amount of what the student borrows.

We settled on an income share agreement that was more “fixed.” Contingent on being hired at a school they like, students pay our graduate school $4,000 a year for 3 years (those are the 2018-19 numbers; the amounts have varied over the years, and there is no interest charged as the school bears the risk of inflation). The simplicity was more palatable to our prospective students. They appreciated the certainty around tuition, expressed as a fixed number instead of a percentage of future income.

Many students still perceive the income-based repayment as “weird,” even though there are income-based student-loan repayment programs at the federal level for public service.

It may make sense for coding bootcamps such as Juno College of Technology to use a percentage in their income share agreements, so that the more their graduate earns in her first 2 years, the more Juno gets. That logic is flipped, though, when the employers are schools and nonprofits. Nobody particularly wants to steer a newly minted teacher towards a job at School A over School B because of a salary difference if there are other factors that make School B a better fit for the graduate. That’s why our income-share agreement, without the percentage, avoids it.

Conventional loans create a particularly big problem when it comes to masters degrees in teaching. Often students borrow upwards of $40,000 to attend private colleges. That’s not easy to pay back on a teacher’s salary. It’s not just that it’s a barrier to entry for many potentially great teachers. That debt quickly becomes a barrier to exit for teachers who feel “trapped” by their debt. Teachers who wish to quit because they don’t like the job feel they can’t walk away from the sunk cost of an expensive degree. So they stay in the profession unhappily, to the detriment of their students.

The income-share agreement has one lovely additional benefit. Some education schools have acknowledged they don’t have data on precisely where their graduates end up teaching. Indeed, some defied the Obama administration’s efforts to capture this data. Because of the way our students finance their degrees, though, we’re bound to our graduates for their first three years. We have every incentive to support and coach them. That incentive is magnified by our other revenue source — the teacher placement fees schools pay for the right to hire one of our graduates. All of these interactions with school principals, in turn, keep improving our actual graduate school curriculum: we’re constantly taking from our alumni’s real life K-12 experiences and incorporating those lessons back into our graduate classrooms, to help our newer cohorts of masters students.

We’ve had 259 students graduate having taken one of these ISAs. 100% got teaching jobs. That perhaps sounds implausibly high, but our graduate school is unusual in that it combines academic exams with performance assessments. These performance assessments require students to demonstrate mastery of fundamental teaching skills in simulated classrooms. Folks who were perhaps better suited for other professions than teaching exit the program after the performance assessment (without any debt whatsoever, and with help from us to find what they “really” want to do). Roughly 95% of our graduates are on track to fully pay back, or have already done so.

The tuition is low because we have an unusual second revenue stream: the schools pay us a finder’s free for the right to hire one of our graduates. That is, they can either hire Candidate A from Any School of Education in the USA for free, or hire Candidate B from our graduate achool for $8,000…and principals choose “B” quite often. We acknowledge that sort of fee structure is not plausible for most education schools.

Income-share agreements are attracting increasing attention not only from students and from colleges and universities but also from policymakers and politicians. We hope to see this way of financing education grow over time, including a variety, like ours, with “fixed” rather than percentage-based terms. Such agreements reduce risk for students, limit the marketing spending needed to attract applicants, and increase transparency of outcomes.

Mike Goldstein is the founder and Scott McCue is the dean of the Charles Sposato Graduate School of Education, based in Boston.