Income Share Agreements are just predatory loans

Income Share Agreements (ISAs) are essentially a type of loan where the amount the borrower pays is tied to their income. ISAs are frequently touted as a solution to student debt, and all sorts of wild claims about them are made, including the claim that ISAs aren’t loans and that graduates with ISAs graduate debt free. Lambda School is a well-marketed for-profit online school that uses an ISA. Their CEO, Austen Allred, recently wrote an oped in Wired promoting ISAs in general and his school specifically. (It’s unclear to me how giving Austen a platform to publish what are essentially press releases benefits readers.) The key claim in Allred’s oped is that “If institutions only get paid when their graduates do, they’re incentivized to provide a service that actually gets students hired.” This is a common claim and is based around the belief that that schools with ISAs get more money if their students get higher incomes, and that schools are therefore incentivized to provide excellent training so students get jobs.

None of these claims are true. ISAs do not “align incentives” between students and schools. ISAs are not risk free. ISAs are not a solution to the student debt crisis, they are just more of the same.

First of all, ISAs are loans. The key feature of a loan is that borrowers are given money now and need to repay money in the future. That is exactly what’s going on with an ISA. But the terms of the ISA are misleading and opaque. Schools that provide ISAs often claim that students graduate “debt free” which is not true. This should raise red flags for any student considering a for-profit school with an ISA.

What Austen doesn’t mention in his press release/oped is that many ISAs, including the ISAs at his Lambda School, are “financed”. Here is the blog post from Lambda School announcing that they received “financing” for their ISAs. I haven’t been able to find any publicly available information about the specific details behind Lambda School’s financing. But other players in the ISA space, such as Edly, have been slightly more transparent about how the financing of ISAs would work. Charles Trafton, a cofounder of Edly, described ISAs as an asset class similar to many other forms of consumer debt: ISAs are bundled together, investors pay to own a percentage of specific bundles of ISAs, and investors then receive a percentage of the income generated by that bundle of ISAs. This means that for-profit schools with ISAs aren’t really taking risks on each student they enroll. Instead, what happens is for each student they enroll, they receive enough money from ISA investors to cover the cost of educating the student, and they probably split the profit from the ISA with ISA investors.

From the perspective of students, the financing of ISAs is risky because it makes it hard to judge whether the incentives of students and school are aligned. The terms of the ISA that the students see: the percent of income they give to lambda, the duration of the ISA, ect. are merely information about the repayment plan for the loan. Students are not given any information about the underlying terms of the loan, such as the interest rate or the cost of providing the education. Attending lambda school is like taking out a loan for a car, being told that repaying the loan will cost between $20,000 and $30,000, but not being told how much the car costs. This is very dangerous for students, because students don’t really know how much their education actually costs, and students have no idea if they’re getting a really great $20,000 education, or a really bad $200 education.

Lambda School in particular is not transparent about the costs of its education. On Lambda School’s website, Lambda School claims that students have two payment options: the ISA, or a $20,000 upfront payment. But there’s reason to suspect that the $20,000 option just exists as a red herring: to create the impression that Lambda School is expensive (i.e. provides good education), and to make the terms of the ISA look favorable by comparison. The $20,000 is a one-time lump sum payment required at the beginning of the program; Lambda School does not provide its students with the option to pay the $20,000 in monthly installments. Most of the people attending Lambda School would not be able to pay $20,000 up front, meaning that in practice, Lambda only has one payment plan: the ISA. Through conversations with prospective and current Lambda students, I’ve learned that Lambda staff respond to questions about the upfront payment plan with ignorance (i.e. the $20,000 upfront payment isn’t important enough for admissions staff to be trained to answer questions about it), and in at least one case, a Lambda staff encouraged a prospective student to choose the ISA. This was initially surprising to me because I would have expected Lambda to prefer money upfront (or money in monthly installments) to money they might not actually get (the ISA). But Lambda most likely gets money for each student they enroll through the financing of their ISA, and there are reasons to suspect that the terms of the ISA are actually quite favorable to Lambda and investors in Lambda ISAs.

ISAs are a good deal for investors (they’re described as an alternative to bonds), but the best ISAs from the perspective of investors are not necessarily the best ISAs for students. While investing in one student might be risky, due to the law of large averages investing in a collection of students is much safer, because it’s very likely that at least some of those students will get jobs. ISA investors often describe their investment decisions as choosing between different schools and different programs or majors, with the idea being that good schools that offer programs in high-earning fields (e.g. programming) would have the best ISAs to invest in. I’m not convinced that this is the full picture. Providing good education costs money, which means that ISA investors will have to pay more to buy ISAs from good schools. However, while better education leads to higher incomes, it’s not clear how strong that effect is. I wouldn’t be surprised if doubling the cost of education only leads to a 10% increase in graduate incomes. And education is not the only variable that predicts graduate income. For example, during a recession, even graduates with the best education will not make a lot of money, and in fact many people, if they can afford it, will go back to school during a recession as a way to buy time before reentering the job market. I suspect that the most profitable strategy for ISA investors would be to invest in a school that enrolls a large number of students, provides cheap education, and a low percentage of graduates find jobs immediately after graduation but a high percentage of graduates pursue additional education (and take on additional loans) after graduation. There are reasons to suspect that Lambda wants to be this type of school, even if it isn’t quite there yet.

Income is correlated with gender and race, so there is a huge incentive for schools and ISA investors to “invest” in discriminatory ways. Lambda School is developing a model to predict the “success” of each student, and the applications for this given the context of ISAs being investments are obvious. I suspect that schools like Lambda have at least thought about spending different amounts of money educating different types of students, as investing unequally could lead to bigger profits. For students that are likely to get a job after graduation (e.g. they already have degrees), it would make sense for a school like lambda to do everything to ensure that these students get the highest paying job possible. For students who aren’t likely to get a job immediately after graduation, it makes financial sense to give them a cheap education and encourage them to pursue additional credentials; the terms of Lambda’s ISA expire after five years, which is a long time for a job search but enough time for a student to get an additional degree. Schools like Lambda can execute the same strategy by segmenting students into different classes. While personalized education has merits from a pedagogical standpoint, there needs to be a lot more transparency about what education different students receive, and students should not pay the same amount for different educations. Tellingly, there hasn’t been an independent audit of Lambda’s curriculum, and there is virtually no public information about the underlying investment principles of Lambda’s ISA.

Lambda School has a documented history of being misleading about whether the terms of their ISA “align incentives” between students and school. For example, Lambda CEO Austen Allred claims that because Lambda School’s incentives are aligned with that of their students, they spend their advertising money persuading employers to hire their students. This is demonstrably false; I have screenshots proving Lambda School advertises on Google, and anecdotally I can say that Lambda advertises on twitter and facebook as well (because of the targeted nature of these ads, it’s hard to view them on demand). Austen and other employees do marketing on twitter and other social networks. While this marketing may not have the word sponsored under it, that does not mean that the marketing is free (employees are paid a salary) or that the marketing is more accurate.

The most misleading thing about Austen and Lambda School’s marketing is their website claims Lambda School students graduate with “No student loans, no debt.” In fact, the terms of the ISA do not compare favorably with the terms of many student loans in many ways. Lambda’s ISA as stated on their website offers students some genuinely great protections: they expire after five years, students don’t start paying unless they have a tech job with a yearly salary of more than $50,000, and the maximum a student will pay is $30,000 (the minimum students with jobs will pay is $16999.92). But the amount many students will pay in total is similar to the average amount of student debt recent graduates in the US have ($28,400). Unlike most students Lambda students are at a for-profit school with an unproven track record that has not been thoroughly audited, which makes it questionable whether it is worth taking on similar amounts of debt. The payment period of lambda’s ISA lasts for two years, but the size of the repayments lambda students pay each month ($708.33 monthly for students earning $50,000 a year, which is the smallest monthly payment) means that students pay much more each month than the typical student borrower. It’s unclear why Lambda students would be better positioned to make larger payments than other students with debt. In general, Lambda School’s website provides no way to compare the terms of their ISA with the terms of conventional student loans.

Since ISAs are unregulated, borrowers have virtually no protections. There are plenty of risks associated with an ISA. For example, a student might make $45,000 a year, do Lambda’s curriculum, get a job paying $50,000 a year, and end up earning less each year with the ISA. That’s not the risk. The risk is if they are unable to make the monthly ISA payments (which are much larger than those of conventional student debt), they are essentially at the mercy of Lambda School. And of course, these are just the risks that can be gleaned from the terms available on Lambda’s website. Information about the underlying finances of the ISA isn’t made available to lambda students. Austen Allred often talks about the need for regulation of ISAs, but what he advocates for in his oped is regulations of factors like the percentage of income students give up. These proposed regulations don’t address the underlying risks of ISAs, including their financing.

Given the history of for-profit universities, prospective students deserve a press that is much more skeptical when companies like Lambda School claim they have a solution to the student debt crisis. Wired has no excuse for publishing that press release of an oped, especially since their editorial guidelines state that Wired is not a place for executives to write “a piece that essentially asserts the need for your firm’s product.” The New York Times shouldn’t have been so quick to describe Lambda School as an “early success story.”. And many of the investors in Lambda School would do well to perform more rigorous due diligence before investing in a company that the state of California has issued at least one citation against. The student debt crisis is a huge problem that needs to be solved sooner rather than later, but more skepticism needs to be given to for-profit schools that claim to have easy, risk-free alternatives.