New AIR Report on Income Share Agreements is Deeply Flawed

Income Share Agreements (ISAs) have gained some attention lately as a promising alternative to traditional student loans, particularly private student loans and federal Parent PLUS loans. With an ISA a funder gives money to an individual in exchange for the individual agreeing to pay a percentage of her income for a set number of years—with no requirement that the original amount be paid back. Proponents support ISAs for a number of reasons, including their potential to offer low-income, debt-averse students a financing option that mitigates the risk of traditional loans.

While this idea may be promising, however, the ISA market in the U.S. is currently very small, consisting of only a few small providers and nonprofits just starting to get off the ground. (Many, including the authors, have argued that regulatory uncertainty has played a significant role in inhibiting this market’s growth, the basis for recent legislation on this topic.) In this context, the American Institutes for Research (AIR) issued a new report last week attempting to project whether an ISA industry—which barely exists right now in the U.S.—would serve low-income students if the industry were to grow.

Predicting a future ISA market is a speculative exercise of questionable value

Given the limited number of ISA providers, predicting how this market will develop and which types of students might benefit is already a speculative exercise of questionable value. ISAs are fundamentally different from loans, so we can’t look at existing loan underwriting standards. We have limited data on individual earnings, so projecting on that basis is difficult. And predicting consumer preferences with respect to new products is highly uncertain.

But the AIR report takes this difficult task of trying to predict the future and compounds its challenges by using a deeply flawed methodology in two respects. First, the study’s sample consists largely of companies not focused on education ISAs, and, in some cases, that don’t offer ISAs at all. Specifically, AIR’s researchers surveyed the underwriting criteria of existing ISA providers (neglecting one, Education Equity) and some private lenders. They then tossed out two candidates with a track record of serving low-income students on the grounds that they are only in a pilot phase (13^th Avenue Funding) or aren’t in the U.S. market (Lumni). That left six companies that served as the basis for the study’s conclusions: SoFi, CommonBond, Pave, Upstart, Cumulus Funding, and App Academy.

Four of the six companies surveyed aren't current ISA providers

Unfortunately, four of these companies are not even current ISA providers. In fact, SoFi and CommonBond have never been ISA providers—they just issue and refinance loans, with CommonBond specifically focused on MBA programs. To compare a loan company to an ISA provider is to completely miss the point: ISAs are fundamentally different financial vehicles, and thus will differ in the types of investments compared to private loans. So comparing underwriting standards between the two is nonsensical. And while Pave and Upstart were once ISA providers—though, importantly, ones clearly marketed to would-be entrepreneurs rather than aspiring undergraduates—the fact that neither company issues ISAs anymore raises questions as to whether they are the best examples from which to gauge future education ISA provider successes. That leaves Cumulus and App Academy. While an ISA from Cumulus can be used for education, it is marketed as an alternative to a personal loan (i.e. paying medical expenses or consolidating debt), and therefore have underwriting more akin to personal loans. That makes App Academy the only company currently offering something like an education ISA, though even they characterize it as a first-year “placement fee.” In short, it’s hard to see how a sample of companies largely not focused on education ISAs—or ISAs at all—can be predictive of how an education ISA market might develop.

Second, the report’s authors attempt to assess the scope of a potential ISA market by looking at a sample of students six years after enrollment—one to two years into repayment, in most cases—and counting the fraction whose loan payments consumed a manageable share of their income. That's an odd standard to use in determining which students are well-suited for an ISA. A student’s after-school earnings are typically lower initially—meaning many students who ultimately earn a positive return on their education might have burdensome loan payments the first two years out of school. Additionally, ISA providers will likely pool risk, taking losses on some students and doing well on others. Said differently, the report’s methodology is like saying the scope of a car insurance market is only those people guaranteed to never have an accident.

At the end of the day, no one knows if a robust ISA market would fund many low-income students, and aside from promising examples, no one can know. This type of financing option has few analogs, and the market barely exists currently. Looking at loan providers and ISA experiments in other arenas doesn't get us any closer. Ultimately, we would be better off simply putting in place an effective regulatory regime for ISAs, including strong consumer protections, and seeing what types of new options emerge."