Ethan Pollack from Jobs for the Future and Kevin Mumford, an economics professor at Purdue University, joined me to dive into Purdue’s innovative “Back a Boiler” Income Share Agreement (ISA) program. They defined what ISAs are, talked about how Purdue’s model aimed to make higher education more affordable and accessible, and discussed the findings from new research analyzing the program’s outcomes. Our conversation covered the program’s origins, regulatory challenges, its eventual pause, and what the data reveal about student outcomes, particularly regarding fairness, completion rates, and financial impacts for students from different backgrounds.
One of my takeaways? Based on the outcomes, it’s a shame that the initial momentum behind ISAs in the mid-2010s has stalled. But maybe there’s some hope now on the horizon with better guardrails in place for a resurgence behind ISAs.
Research Referenced:
* Distribution of Returns to a College Income Share Agreement: Evidence from Administrative Data
* Promising New Insights from Purdue University’s ISA Program
Michael Horn
Welcome to the Future of Education. I’m Michael Horn. And you’re joining the show where we’re dedicated to creating a world in which all individuals can build their passions, fulfill their potential and live lives of purpose. And part of that equation is thinking about how we pay for and afford what’s become a more and more costly higher education over time. So to help us learn about and think through some new research about what was a very interesting program to make higher education not just more affordable and accessible, but also focused on real value, in my view, our two individuals that we get to welcome today. First we have Ethan Pollack. He’s the senior director in the policy and advocacy practice at Jobs for the Future, or JFF as it’s commonly known, where he leads the Financing the Future initiative, which explores these new approaches to financing post secondary education. So, Ethan, great to see you.
Ethan Pollack
Great to be here.
Michael Horn
Yeah, you bet. And Kevin, we have Kevin Mumford, who’s an economics professor at Purdue University, also the director of the Research Center in Economics at Purdue. Kevin, great to see you and thanks for being here.
Kevin Mumford
Yeah, thank you. Happy to be here.
Michael Horn
Yeah. So we’re going to get into a bunch of things in a moment, but I wanted to sort of level set us start with the basics because we’re going to talk a lot about income share agreements today, ISAs, as they’re known, and we’re going to talk specifically about a program that was in effect at Purdue, the Back a Boiler ISA program, and a working paper that you published, Kevin, recently. But before we get into all that, I just want to make sure our audience is level set. We’re all on the same page. We know what we’re talking about. I know how I describe an ISA, and I see it as a pretty compelling alternative to a loan, at least on paper for a given individual. But Ethan, you’ve thought about this much more than I have. You probably have a much more concise answer.
So in brief, tell us, what is an ISA?
The Future of Education is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
Income-Based Student Loan Alternative
Ethan Pollack
Sure. So an ISA is, I think, best defined in contrast to, to what is more the standard, in particular, the standard in the private loan space, where in that latter space you really have fixed payment loans. Right. You have, you know, much more kind of like a mortgage. You’re paying a fixed dollar amount over a period of time. You have to pay back at the end of the, the, the loan you need to pay back the principal and then the interest. And ISA is different in a number of respects, but the main respect is that it is income based. You are paying on the basis of how much you earn.
So students are only making monthly payments if they earn at least a certain income threshold. They’re also making payments that are a percentage of their earnings. So if they earn above that threshold, they may pay a certain percentage. It can be anywhere from a couple percentage points. Some of the short term boot camps were charging pretty high income shares up to 17 percentage points. Purdue’s are much lower. And then the last feature of an income share agreement that’s unique is that it expires after a period of time. So it’s really de emphasizing this idea that you’ve borrowed a certain amount of money and you need to repay it with principal and interest and actually operates almost a little more like an equity investment where you’re paying on the basis of how well you do. And the person that, you know there, the, the entity that gave you that kind of, you know, financing that ISA is going to be, you know, receive a lot more revenue if you do well after graduation, and they’re going to then receive a lot less and they may receive nothing at all if you never earn above that earnings threshold.
Michael Horn
So a couple key things that stand out about that, right? It changes it from paying based on the cost or the expense to paying on the value, if you will, of it. And then there’s some more risk, obviously on the part of the institution as a part of that, it seems. Kevin, then, you know, ISA programs, they look different in different places. Ethan just alluded to that. They have different incentives and so forth. Tell us a little bit about the origins of Purdue’s spin on this through the backup boiler program and its specific mechanics as well.
Purdue’s Income Share Agreement Origins
Kevin Mumford
Sure. It, you know, it came about because of the president we had at Purdue at the time, Mitch Daniels. He had been asked in the spring of 2015 to testify before a House subcommittee. And in that testimony he had kind of a throwaway line about, you know, wouldn’t it be great if the university was doing something with income share agreements? And he got so much sort of press inquiry after that that he started thinking to himself a few weeks later, maybe, maybe we should really do this. There’s so much interest in it. So there was this funny episode where he calls in the head of the foundation and the chief legal officer for the chief lawyer for the university and says, let’s start an income share agreement. And they both were like, yeah, let’s do that. And they left the office, then turned to each other and said, what is an income share agreement? And they Googled it on their phone and the university just kind of, you know, he started the apparatus.
Twelve months later they started accepting applications and they launched the thing. You know, the idea was they’re going to allow it to be available to all majors. And in some majors people earn a lot more and in some majors they earn a lot less.
So they tried to price it according to what they thought the earnings would be in that major. They put all the majors into seven different bins and said if you’re in this kind of high earning bin, then we’re going to charge a lower income share rate. And if you’re in a lower income bin, we’re going to charge a higher rate. And the idea is to try to make it so that it would compete well with a Parent PLUS loan. They weren’t going to try to compete with the subsidized federal loan or the unsubsidized federal, the direct federal loans, but they were going to try to compete with the Parent PLUS because it has a higher interest rate. And the difference here would be, is that a Parent PLUS loan, the parents are the borrowers. Here it’s the student who is taking on this repayment obligation. And you know, there are some protections, like Ethan said, there’s a lower threshold, there’s an upper bound.
Nobody could pay more than originally it was 2.5 times the funding amount. Then it went to 2.3 times the funding amount. So that’s kind of the basic idea though, price it so that people would be paying back about what they would under a Parent PLUS.
Michael Horn
I didn’t realize. I actually did not know that. Yeah, that’s hysterical. I did not know that piece of it even. And then put it in context of like, you know, when this launched. Right. Why was it such a big deal at the time? How did it fit in? Right. With the larger movement? ISAs were gaining a lot of currency as an idea that might be this alternative to loans.
Give us a sense of why this was such an important program.
Ethan Pollack
Yeah, it was a really interesting time. So I was at the Aspen Institute at the time and I remember hearing about income share agreements and both thinking like, wow, these are like, this could be really exciting. This could totally kind of rewire how we do higher ed finance and workforce as well outside of the Title IV system. And also is like, this could also be kind of dangerous too.
Right.
We don’t have guardrails that are uniquely designed for income share agreements too. And there’s A lot of uncertainty as to which laws actually applied to ISAs. And so I was kind of in this middle camp of being really excited about the potential but also then having a bit of concern. But there was at this time, you know, when it was launched in 2016, there was this real interest in ISAs, you know, not just in higher education, but in particular, you know, this was also the rise of, kind of the time of the, of coding boot camps. And you know, with coding boot camps there, there, you know, a lot of the tech companies were feeling, look, you don’t need a college degree, we can, we can just teach you to code. And we know that you don’t believe us. So instead we’re going to take on the risk and attend my coding bootcamp. And if you don’t make above a certain, you know, amount, then you just don’t pay anything back.
Excitement and Backlash
Ethan Pollack
So I think that there was, that was happening in tandem then with the Purdue program. And but overall there was this really, especially when Purdue launched, there was this kind of real kind of boom lit and media excitement around income share agreements. But then there was also this backlash, right. And the backlash had a couple different facets which happy to get into, but I’ll gloss over it right now is just to say that, you know, there was some concerns around consumer protection and there’s some concerns around that this, you know, like any type of financing product could be used in predatory ways. I think that there, that most of the, you know, the actors in the space at that time like were trying to do the right thing. And I worked with, with many of those and in thinking about kind of where good best practices. But there were some bad actors too, right? I mean, you know, whenever you’re kind of, you know, introducing kind of a new product, there’s going to be either bad actors or people who just like, you know, are not fully understanding how this tool can be used. And so it, you know, I think that there was a real push towards using income condition financing, but it was done very quickly.
And, and I think that, and then as you know, what I lament is that the energy was really towards income share agreements for a while and then a few years later kind of was like, oh, I guess that didn’t work. And so it was, it was, it kind of like flamed out very quickly. And you know, this was kind of the origin of, of, you know, my partnership with Kevin was this feeling of like, you know, I don’t think we ever fully like studied this to see whether it actually worked or not. And everyone else just kind of moved on to the next shiny thing. And here we were saying, like, oh, we got a lot of data. We should actually, like, look at whether.
Michael Horn
Well, so I want to get to that in a moment. And you mentioned the controversy. We’ll come back to that in a moment as well, but sort of tangentially related. I think a lot of people think that the program ended because of a lot of the concerns around ISAs and so forth. Kevin and then Ethan, both of you chime in on this one, like, why did Back a Boiler—why did the program pause? What led to that? Kevin, why don’t you jump in first?
Kevin Mumford
Well, I mean, the most basic reason is that the university set this up very quickly by going with the service provider. So the service provider had been the one originating all the contracts. And in 2022, the service provider shut down and, and sold the existing contracts to another provider, but one that wasn’t, wasn’t able to or, you know, doesn’t have a business of issuing new contracts. So the university was in a situation where they said, well, we’re not sure how to issue new contracts, but we’ll continue to service the contracts we have. And I, I think that happening, like, right at the same time as when the president of the university was stepping down. The new president wasn’t opposed to this, but this was also not a champion of an income share agreement. And so the university would have had to, you know, find some other provider or, or set up the, the ability to do it on their own. And, they just haven’t put in the effort to do that.
I mean, I think part of it also was kind of because of the pandemic, and it’s because the university received a large amount of federal funding and to provide directly to students and the students that we’ve been advertising, or I guess you can’t advertise, but been offering this, this income share agreement to. They were the students who had already maxed out their federal borrowing, and that’s exactly who they decided to subsidize with the federal funds from the pandemic relief. And so there wasn’t a lot of demand for it right at that moment as well. And it kind of just contributed to saying, maybe let’s not issue new contracts, although if they were to revisit the decision today, that might be different. I also think a factor is the regulatory environment. There’s still just been an uncertain regulatory environment, which makes entering difficult.
Michael Horn
Ethan, what would you add in there? Yeah, I was gonna say, Ethan, what would you add to that last point? Because you’re much, you’re very deep into this regulatory questions around the status and the legality of ISAs and how they’re governed.
Ethan Pollack
Yeah. So ironically, the same year that Purdue paused was also the year that CFPB signed a consent order with Better Future Forward, which is a nonprofit ISA provider, mainly works with college access programs. And the consent order is they fined Better Future Forward $0 and thanked them for working with them. It took them BFF and CFPB a full year to figure out what is the disclosure for the Truth In Lending Act. It’s just focused on TILA. And so actually right around the time that Purdue paused was the same time that BFF published this, you know, this, this disclosure template that they said like anyone can use this but, but the, you know, the bureau has actually signed off on this as like a way to, to, to comply with TILA. So there actually that was a moment when there was starting to become more clarity. Since then, also California, Colorado had issued, have done rulemaking for the, on their own state basis to improve the regulatory clarity.
And then Illinois, actually just a few weeks ago, Governor Pritzker, signed legislation to kind of improve the regulatory clarity. But the main issue here is that, and what Purdue was really facing was that, you know, the rules governing these credit products were largely designed with fixed payment structures in mind. So you think of something like the Truth In Lending Act, you need to disclose the, you know, the APR, the annual percentage rate, you know, when you are, you know, presenting the loan or the credit product to the student with an income share agreement or even with something like an outcomes based loan. Because these can be structured still within a loan framework. You know, the federal IDR program is a really good example of kind of an outcomes based loan. Right. It’s a loan, but at the same time you’re going to be paying on the basis of your earnings. But if you have a product like that, what is the APR? Well, it depends on how much you earn after graduation.
So it’s really hard to know that when you’re first presenting this to a student, to a borrower. So that’s just one of the many ways in which the existing regulatory framework was really kind of poorly suited and created a lot of uncertainty. I think that in this uncertainty there was then also some criticism. So a lot of ISA providers were basically recognizing that it’s like, look, we’re not sure how to like disclose APR. We’re getting no guidance. We also feel like there’s a legal, you know, there’s. There’s some legal basis for us saying that, like, we’re just not subject to the Truth In Lending Act because we’re not credit. We’re more like equity.
You know, we don’t have principal, we don’t have interest. We’re just structured in a very different way that I think a lot of the consumer protection advocates got really freaked out by that because they, the precedent that there was a, what they saw as a credit product that was financing education that, you know, somehow you could kind of avoid complying with the, you know, all of the existing kind of consumer protection laws. Not all of them, but many of them specifically designed for, you know, student loans. They thought that as just like I saw that as a real threat, and I understand why they felt that way, but it really was just this regulatory uncertainty that was the underlying problem. They also had some concerns around things like what you might call kind of differential pricing or the underwriting. And so basically, if you are charging lower income shares in the contract to either fields of study, their majors, or graduate programs that have higher earnings, you could create some disparate impact.
There was a concern around that too. Again, this was all just theoretical. And, and then, you know, there was also, you know, then once Purdue had started, there were some examples of students that ended up paying more than they otherwise would have under Parent PLUS. Now we can guess why that was the case because they were earning a lot more than the average student. But nonetheless, there was a lot of stories and kind of newspaper and coverage around these specific students who really felt like. I felt like I overpaid. And so I think with that, that was also kind of part of, you know, as.
As Purdue was pausing, I think that that was the broader kind of, I would say, media context in which, in which kind of that decision was made. I can’t speak to, you know, exactly to what extent that was taken into account, but it certainly was something that was prevalent and, and certainly didn’t help the case for trying to really kind of continue this program.
Michael Horn
Well, so I want to dig into the research in just a moment, but quickly, Ethan, just on that because. Well, Kevin, essentially what you’re describing again is consistent with what you said up front, which is it’s really a last dollar program, if you will, of financial aid. Right. It’s not the first dollar competing against federal subsidized loans. And so that had something to do with the crowding out, if you will, from the federal funds that came in at the time. The second thing, though, Ethan, you talked about the regulatory uncertainty at the time. You talked about how it’s starting to resolve a little bit. Give us the, to the extent you’re able, give us sort of the quick snapshot of where do these vehicles, if they were to sort of make a comeback in the limelight, where would they live at the moment? Because this sort of lending classification has always felt kind of weird around this.
ISAs: Unique Loan Classification
Ethan Pollack
Yeah, yeah. So there’s a movement towards, I, I would say momentum towards resolving the question of are ISAs loans? To resolving it in the direction of they are loans, but they’re a unique type of loan. And so, and that, that kind of like solves everyone’s problem where, you know, there’s no claim that this is, you know, something that’s separate, but at the same time that you. So like the, a good example is the, you know, the Bureau’s consent agreement with Better Future Forward where, or the compliance plan where they said you’re a loan. But also they developed a very unique compliance plan for Better Future Forward, taking into account that it didn’t have any interest. And in fact there are certain boxes in the, in the temp, the TILA template that says the interest can range from,you know, in those two boxes from X to Y. And, and the Bureau ordered BFF to just put in NA into both of those boxes.
Michael Horn
Oh, interesting.
Ethan Pollack
Which, you know, if you’re a normal lender, you can’t do. But they, but it was a recognition that it’s like you’re not a normal lender. This is a unique type of product. California and Colorado when they did their rulemaking also reflected these are loans, but these are unique type of loans that deserve some, a specific type of treatment. And then the Illinois legislation that just very recently passed also treats these as loans, but unique types of loans. So I think that’s kind of the regulatory direction that we’re going on.
Michael Horn
Gotcha. Gotcha. Well, I can’t decide if NA is better than zero to zero.
Kevin Mumford
It’s using students, right?
Michael Horn
Exactly. So, okay, so now you have this research paper, Kevin, that comes out. We’ll link to it in the show notes. The working paper title is “Distribution of Returns to a College Income Share Agreement: Evidence from Administrative Data.” So this is actual evidence of the outcomes. First give us the origin story. How did the paper come about and what’s JFF’s connection to it? Kevin, why don’t you lead us off? And then, Ethan, supplement anything that should be supplemented?
Kevin Mumford
Sure. Well, I was actually away on a sabbatical when this program launched. I was focused on other research. When I’d gotten back, I read about it and thought, this seems interesting. So I contacted the group at the university working on it and just said, who’s doing your internal evaluation? I’m really interested to see. And they said, man, we set this up so quickly, we didn’t even think about that.
Nobody’s studying it at all. And so I said, well, could I see the data? And they said, sure. And they just sent it over to me. And so I thought, hey, this is great. And in 2022, I had a paper that was about selection into the program. What kind of a student is selecting in versus not. And it’s, again, it’s aimed at a small number of students.
There’s not very many that qualify because very few produce. Students actually hit the federal borrowing limits for those subsidized and unsubsidized direct federal loans. So there’s not very many students who even could consider. But I got data for all the students who were shown the terms that they saw this is what the income share rate would be for me, and then saw which ones decided to take it up on and which ones didn’t, and then looked at why some are doing it and some are not. That paper, I don’t know it, you know, 2022 is right at the time when nobody’s interested in income share agreements anymore. And so it’s. It’s still under review, but it hasn’t been one of these things that everybody seemed super interested in.
Renewed Interest in ISA Data
Kevin Mumford
And so Ethan contacted me recently and said, I think maybe there’s a moment, you know, maybe. Maybe people are interested in ISAs again. And don’t you still have all that data? And, yeah, I mean, there’s still a lot of data, particularly in that first paper there was nothing on who was paying back, what kinds of payments were we receiving? And he just said, I really think people would be interested to see how did this program do financially? I mean, did the university make a ton of money off of these students, or did they lose a lot of money on these students? And, you know, were there students paying back way more or way less than what you would anticipate from a, you know, private loan or a federal loan? And just. He just thought that people would be interested now. And so I’m always interested in doing research that people are interested in reading.
And he kind of convinced me it was worth putting in some effort for a bit on this project as opposed to all the other projects I could have been working on.
Michael Horn
Ethan, what would you add? Or, and, and perhaps also with an eye toward why the research is relevant now in your view?
Ethan Pollack
Yeah, yeah. Well, I’ll start off with that. That, you know, it’s, I mean there’s so many things happening that I think make this research relevant. Not just this research, but the Purdue ISA program. You know, if we had, if it hadn’t launched 10 years ago, I think people would be thinking now, man, we should, we should try to do something like this. Yeah. So the reason is I think a few fold. So, one, your institutions are facing demographic cliffs, right? So you have the, the number of traditional aid college students can be declining about 10% over the next two decades.
They’re going to have a huge drop off in international students too. You know, that can really put institutions in a tough spot. I know you’ve covered a lot of this in your podcast with Jeff, so listeners of both won’t be surprised by any of this. And then of course from the Big Beautiful Bill, a lot of the borrowing limits are going to have really big impact. So I published an analysis last week on GFS website. It found 1 in 10 students are going to be impacted by the caps and they’re going to be forced to either forgo education entirely or go to the private lending market. We know the private lending market has some really, you can get some cheap credit but oftentimes for many students it’s going to be more expensive. More importantly, it’s going to, you know, 95% of private loans require co-signers.
They all do credit checks, require certain credit score. And then in addition to that, it doesn’t have any type of the low wage borrower flexibility that, that the federal IDR program has or that ISA programs have, which is that, you know, if you become unemployed or you just, you, you end up having a lot lower income that you thought that you’re going to, you know, that you’ll pay less and you might pay nothing. Right. But with private loans, with some very rare exceptions, nearly all private loans you gotta pay regardless of what’s going on with you. Right. And they may do the per, you know, some, some deferments that you know, on a case by case basis, but that’s not really kind of like baked into the contract. Right. So it’s not a really firm protection.
And so, you know, we’re at this moment where it’s like, wow, we’re going to have so many students that are going into the private lending market. What types of products are they going to get? Don’t we want them to have types of products that are like kind of like ISAs that are kind of like actually the federal IDR program that have these types of low wage protections and in particular are also more accessible too because we know there’s going to be a lot of students that may just not qualify for private loans because they don’t have a cosigner, because they have bad credit. And ISAs are in particular forward looking as opposed to backwards looking. Right. They care less about the, you know, where you’ve been and more about where you’re going. What’s the program you’re enrolled in, what are the outcomes? And so the Purdue program, for example, it didn’t require a co-signer that was like really important to the model. Right. So for students that were not, that could not access Parent PLUS, then they could access something like this.
Right. So you know, I, I thought like this is such an important moment and like we also, instead of setting up a pilot and then studying it and it taking 10 years, like we just kind of, like we just kind of took a time machine, created the program that’s most relevant to today and then now we’ve got all the data, which is kind of a researcher’s dream. And so, you know, I you know, Kevin and I had talked before, but I kind of was like, we’ve got to get some data out here. Like this is, you know, this. Yeah, I think that people would be really interested in this. And it feels like the Purdue ISA program is now more relevant than ever, you know, both for policymakers, for institutions, for philanthropy, and then for researchers such as myself. And in particular it had gotten, you know, for the criticism. I was really interested to test the criticism because a lot of the criticism and a lot of the claims too, it was all based on speculation, right.
We didn’t actually have the research and I don’t like kind of making claims without a lot of your rigorous research. And so now we had enough data and I wanted to see like, was there any disparate impact? Maybe there was, like this would be really good to know, like how affordable was it compared to Parent PLUS? And so these were the types of things that Kevin and I were then able to, you know, figure out. So I think that Kevin was very much, you know, the primary researcher, and I was the one that was kind of enmeshed in the broader discourse, unknowing, kind of like this is what people would be interested in. You know, this is how it could, you know, your research kind of connect back to, kind of like, you know, to. Yeah. To actually be relevant.
Michael Horn
And so, Kevin, I want to start to get into what you found. Then in the report itself, it occurs to me it’s like an interesting sample of time because you both have some really hard economic times with a lot of job switching in those sample years. You have some boom times in there. So it’s like, it’s a very interesting set of years that I suspect you’re following these students. But talk a little bit about what you were looking for, the methodology, and then the findings itself.
Comparing Student Loan Repayment Options
Kevin Mumford
Well, the basic methodology is to look at the stream of payments coming in from those students that are participating in this program to see how much are they paying back. And then you need something to compare it to. So, you know, what I do is I take a traditional student loan and I peg the interest rate to be the same as the Parent PLUS loan. And say, what would a student have paid back had they been paying back a traditional student loan at that interest rate? And just do the comparison to see are they paying back a lot more, a lot less than they would have otherwise? And I think the time period you bring up is relevant because. Because there’s job switching, job loss, there’s uncertainty. The thing about the income share agreement is that when your income goes down, your payments go down, and when your income goes up, your payments go up. I mean, it’s adjusting to the economic conditions automatically and adjusting to your specific characteristics. Because if you lose your job and you’re unemployed for a period of time because of the pandemic or because of any reason, then the payments stop until you start working again.
And that’s just sort of built in, already. So some of the results might be colored by that. I mean, if this was sort of the best booming economic times ever, payments probably would have been higher than what we’ve actually observed. But what we observed is what happened. And there is, you know, there’s some movement up and down in people’s careers. The data comes in and, and I, I actually do it two ways because there’s a lot of ways to pause payments in this income share agreement. Again, if you’re unemployed, you’re not making payments. And, there are protections in student loans.
They’re maybe not as generous as an income share agreement, but. So one way I do it is I assume on the student loan that you just make every single payment as scheduled all the way through. And the other assumption is that anytime you pause your income share agreement, you would have also paused your student loan. And the truth is probably somewhere in between. And I compare the amount each student is paid and then the amount the university has received. And the amount the university has received is right around that same amount that they would have gotten if it was a Parent PLUS it’s a little bit less than that, which suggests that maybe they’re getting a sort of a less than a 6% internal return. On the other hand, income share agreement payments grow over time because they increase as the student’s income increases, and it naturally goes up over time.
So probably the university’s best payment years are still to come, and they’re not these initial beginning years. So it may be that the internal rate of return is a little bit higher by the time it’s all done, but it doesn’t look like the university made a lot of money. If anything, they probably would have done better just investing in the market. On the other hand, for individual students, the average student paid less in the income share agreement than they would have if they had gotten the traditional student loan at the Parent PLUS rate. But there are a handful of students at the top that paid back much more. So some students hit the cap. That’s they, they pay back 2.3 or 2.5 times the bar the amount that was originally funded. So some students are paying back a lot more, and that’s because those students have very good jobs and are earning really high incomes.
There’s a lot of students that are paying back a lot less, much, much less than they would have. And that must be because they, you know, decided to stay at home for a while. They’ve decided to do, you know, they’re stopping out of the workforce for whatever reason, and then you just don’t make payments while that’s happening. The average student, you know, is paying a little bit less than they would have under the Parent PLUS they’re doing a little bit better. And that reflects sort of the slightly lower rate of return that the university is getting on a whole.
Michael Horn
Super interesting. Ethan, before you jump in with some of your takes, one other question that I’m just curious about, Kevin, what was the impact on Black and Hispanic students and underrepresented students, because that was a big question that a lot of the consumer protection advocates were raising. And maybe you pull in some of your 2022 research on this too. Because if, if I’m recalling correctly, one of the speculations also was that some of these students, you know what Ethan just said, that it’s forward looking, not backward looking. True in theory, but maybe you would select out certain students that come from lesser advantage or less academically prepared or whatever it might be. And so perhaps they wouldn’t actually get to benefit from the protections that an ISA brings to the table. What have you found on all that?
Kevin Mumford
Well, I think we’re right to be concerned because it turns out this program has disproportionate numbers of Black and Hispanic students in it. And that’s, that’s probably because those are the students who are maxing out their federal borrowing. And, and, and so it is serving more minority students than the traditional student loan program is serving at Purdue. That said, in terms of the ISA repayment, those students tend to be getting a better deal, not a worse deal. That is just as we look at the stream of payments coming in, they seem to be paying back even, even more or less. Right? Like even less than they would have under the Parent PLUS program in comparison to the, the white student that is a participant. And so I think that, I think that we should be concerned about it because these programs are serving more minority students, but at the same time it seems to be give, offering them a better option, a better deal on average than, than it is for the, the equivalent white student. And that’s really not because there’s anything special that’s, that’s, you know, race or ethnicity focused about the program.
It just has to do with what the earnings look like afterwards. And because the earnings for the black and Hispanic students are a little bit lower than those of the traditional white student in the same program. That’s why they’re, they’re paying back less in this.
Michael Horn
Super interesting. Okay, Ethan, I’m curious your takes.
Disparate Impact in Income Agreements
Ethan Pollack
Yeah, yeah, I was just going to add to that. I mean this was always a concern, as Kevin said, I think really well, the concern around disparate impact was very valid. Whether it’s an income share agreement or even if it was just a loan that like any type of financing product, you know, this is why we have the Equal Credit Opportunity act, right. You know, is to make sure we know the history of your redlining and reverse redlining and that, you know, with an income share agreement it creates new potential opportunities to, you know, be violating, you know, fair lending laws and to be exacerbating, you know, you know, racial gaps. It is also the case that income share agreements can do the opposite and can actually, if designed well, can be more affordable. And so it was always this, this thing that, you know, when the Purdue program was being criticized for this, I felt like a lot of the critics had, you were voicing valid concerns, but they were drastically overstating their confidence in it where they were saying that the Purdue program is discriminating or the, the Purdue program or the ISAs.
I remember hearing from some consumer advocates that ISAs are inherently discriminatory. And again, it was all theory based. Right. And so, you know, this, you know, I, I kind of responded to that in 2022, published some research that we got the proprietary data from Leaf, which is a no longer, but was a servicer mainly of tech boot camps. And we did a disparate impact analysis that used the same methodology that CFDB uses when they’re figuring out disparate impact. And we found that there’s no evidence that there was any type of disparate impact here on racial or gender lines, but even still, like, it, it’s the, the frustration is kind of like, yeah, you should be concerned, but like, we don’t know until we actually crunch the numbers. And a lot of the critics just weren’t doing, you know, they didn’t.
And partly they just didn’t have the internal data right, but it should have been voiced as a concern rather than a conclusion. And so this is why kind of on the one hand, whenever we have concerns, we should voice that we should recognize the information we don’t have. Right. And recognize that, you know, we can have a concern without saying that this is bad. We need to wait for the research to come in, but then also we need to make sure that research is actually funded and, you know, it is prioritized and then do the research and then you kind of update your priors. And so, like, it’s really exciting that now we have enough data to be able to say, at least for now, it doesn’t seem like there was any type of, you know, of disparate impact at all, in fact, certainly for Hispanic students, Hispanic students, you know, we’re far more likely than white students to be in the having paid less than having paid more than Parent.
Plus, you know, this comes on the heels also of Kevin’s 2022 research which found that, you know, it looked at the impact of, of the ISA program on completion rates and found that overall completion rates were boosted by about 3 percentage points. But that was. Yeah, it was yeah, 6 percentage points for black students and 17 percentage points for Hispanic students. I think that the black student one may not have been statistically significant. Kevin, you might remember that better, but the Hispanic one was definitely, 17 percentage points is huge. So, you know, like this is a program that also is, you know, clearly benefiting students, you know, in terms of the completion is also more affordable and in particular more affordable for Eurasial minorities and then also is more acceptable accessible too. And we know in particular for black and Hispanic students, they’re going to be less likely to have either a good credit score or have available frontiers either.
Kevin Mumford
I just want to jump in because when you say that 17 percentage points, it makes it sound like it’s unbelievably large. But you got to remember that these students are the most at risk at dropping out. So these are students who, this isn’t about access to the university. This is, you know, they’re a junior or a senior who has now run out of borrowing ability. They can no longer get federal loans. They’ve come to the financial aid office saying, I’m not sure what to do here. And you think, you know, a student might come in and say, I’m thinking of a Parent PLUS, I’m thinking of this ISA. I’m trying to make the decision.
But what about the student that says, I don’t have a parent co-signer, I don’t have a parent that’ll do the Parent PLUS, you know, the private market, the private student lending market, they want parent co-signers as well. I don’t have anyone backing me. So it’s either drop out and go get a job for a semester and then try to come back, or it’s do this ISA. And it turns out the ISA is really, really effective at helping students stay in school. And so even though the university is not making money, I mean, I guess you could say they’re kind of losing money on this program because they’re not making what they would make by just putting the money on your market.
Michael Horn
Sure.
Kevin Mumford
I mean, but they’re getting a huge return in terms of students completing, students graduating. And that’s where the pressure is. I think if this is as a really cheap way to help marginal students complete, you know, universities should be thinking of this not as an innovative finance thing or getting or we’re going to make a lot of money off this. The Purdue experience suggests universities are not going to make a lot of money off ISAs. But if the university gets into it, the advantage is you’re helping a marginal student stay in school as opposed to dropping out to try to get some funding to pay their tuition.
Ethan Pollack
And even with the federal loans, you know with this kind of these new borrowing caps that are now in place, that’s going to be even a bigger problem for institutions.
Michael Horn
Well, I was going to say it probably becomes more important, right. To have these alternative vehicles out there in certain cases. Otherwise completion rates could plummet at the sort of the last mile, if you will, as you’re rounding the corner. As we start to wrap up here, I want to think about or hear your big lessons and takeaways and sort of where we go from here with this research out there. One of them that occurs to me is the importance of the design of these programs and maybe even, like, who is backing the finance. Right. Like the fact, Kevin, that you kept saying these are Purdue’s dollars, it seems like that are part of the program as opposed to an investor perhaps could be an important component as well. But I’d love your, like, wave the magic wand. What would you see done based on this research? What are the big highlights, takeaways, lessons that we should be thinking about?
University Funding via Philanthropic Investors
Kevin Mumford
I just want to say something. I’ll let Ethan take the big picture, but one thing is the university dispersed $20.8 million to these undergraduate ISA participants. And some of that money did come from some investors, but the investors tend to be sort of philanthropically motivated. These were foundations that were trying to see this thing started. They were also not expecting a big return on their money. That, you know, in terms of investors getting involved, I have a hard time seeing that there’s a huge play for this Purdue type of program because the return on investment at this price point was just not very large. What I see is universities putting their own resources into it seems to make a ton of sense because they could be putting it into sort of, you know, scholarship programs or expensive programs to try to keep kids in school. And instead you could, you could just provide this type of funding and they’re going to get their money back.
That’s the idea. I mean, I would love to see more universities try this.
Michael Horn
Well, it seems like a great way to, like an evergreen, if you will, fund that replenishes, even if it isn’t, to your point, as much as you could be earning, if you at least, you know, make back inflation or something like that, that might be worth it. Right. At the end of the day,
Kevin Mumford
yeah, that’s exactly right but, you know, in terms of big vision, maybe Ethan’s got more of a big vision.
Michael Horn
Well, so Ethan, what last word as we wrap up here? Big lessons, takeaways, magic wand you would like to wave and see done with this. Other questions we should be asking.
Impact Investment in Education Programs
Ethan Pollack
Yeah, so, I mean, first, to add to what Kevin said, I do think that, you know, not only institutions, you know, could put their own money behind it, but I do think that this can be a real good philanthropic play as well. So for, you know, for philanthropies in particular, impact investors who may be able to take less than market rate returns, I think the ROI , if you’re talking about a, you know, kind of a social impact ROI, which they’re going to be much more attenuated to, like, that can be really huge because, you know, this is something that can be an evergreen fund. And so they’re able to kind of get their money back, but at the same time they’re able to drive real impacts for students. And that’s something that I think that, you know, in particular, impact investors are always looking for those types of opportunities. And this seems like this is a perfect one for that. You know, more broadly, I think, you know, one of my lessons is that, and, and I don’t mean this as a criticism of, of Purdue at all, but I really love Kevin’s, you know, you know, Kevin talking about how he started getting involved in this is that the, you know, the, the research wasn’t baked in from the beginning. And I think that that’s something that’s not super, super unique. I think that I’ve seen so many either ISA or outcomes based loan programs that are all trying to do this type of kind of pay it forward model.
Oftentimes the research is not at the forefront, or maybe they start the program and then they’re like, okay, maybe now we’ll find a researcher. And I know as Kevin has talked about, if he had been tapped in from day one, he could have designed some more natural experiments to really get some really cool quasi experimental design. And he does do some quasi experimental design, but it was a little more happenstance. And to be able to kind of bake that into the beginning and do some randomization can lead to some really interesting results. And you know, if, if, you know, I think that oftentimes there’s a lot of people have really cool ideas that are really trying to, to build something new and they get really excited about starting to build. But I think what they forget is that if they’re really trying to drive impact, they need to recognize that the way you drive impact is by doing the thing and then having some really rigorous research that shows that it worked or that shows the ways in which it didn’t work so that the next person who wants to build something is able to learn from what you did. And otherwise, what we have is all of these, you know, people that are inspired doing kind of cool stuff, but it all kind of just like dissipates because they do it and then they know that they did it, but it wasn’t able to really scale up. And so it.
It’s really just like a plea for more research from, you know, on the front end. And then for us, you know, Kevin and I would also like to do more research on the Purdue program. There’s so many more research questions that we want to get into too, as well. So, you know, also there’s any philanthropic funders who are out there that want the third piece of artwork. We are available and happy to talk about all of the additional questions. We have this huge data set, you know, we already, they already did the work, and all we need to do is you just. We just want, you know, you know, we can spend some more time to really answer some really interesting questions there.
Michael Horn
Terrific. I love that you guys have invested in that side. It feels like it’s a lesson, and I’m going to call it a lesson in evergreen iteration, because sometimes out of research, we want the headline, did it work or did it not work? As opposed to, okay, what are the lessons learned? How do we keep improving right from there so that the impact over time is actually there? It’s not always convenient, I suppose, for the headlines, but it’s perhaps better for students. So, Kevin, Ethan, huge thanks for coming on here, talking about it. Huge thanks for doing the research, writing it up. You can also check out, Ethan, you had a blog, I believe, on the topic, headlined “Promising New Insights from Purdue University’s ISA Program.” So check that out as well. We’ll also link to it and for all you tuning in, we’ll be back next time on the Future of Education.
The Future of Education is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit michaelbhorn.substack.com/subscribe
