ISAs aren't a lending innovation, they're an alignment tool

Thanks to @LambdaSchool's success, it seems like ISAs are having a moment. With that, it's being treated as a lending innovation, from my standpoint it isn't. The value of ISAs is not as a financial instrument, but as a tool for incentive alignment and marketing. 1/x

I'm seeing a lot of "ISAs are great for borrowers because there's less risk". While that's true, the risk hasn't gone away, it's just been transferred to the lender. That means for it to make financial sense, the lender has to charge more. /2

As a kicker, ISAs also create a perverse incentive (to earn less) which increases the risk further. That combined with that natural adverse selection problem means that for a typical borrower ISAs must inherently be more expensive than a traditional personal loan. /3

There's one exception to that, cases where the lender is better positioned to mitigate the risk than the borrower. For situations where the lender is able to invest directly in the borrower (e.g. by making them more employable) ISAs could become cheaper. /4

That's the incentive alignment use case. ISAs more tightly couple the success of borrower and lender giving lenders a reason to invest further in borrowers. But, for this to work that impact has to meaningful and you need strong underwriting /5

I think the second major use case is marketing. ISAs can act as a strong signaling device for brands. A good analogy here is Patagonia's lifetime warranty. It creates some naturally perverse incentives, but it's a strong signal by Patagonia about their commitment to quality. 6/

Lambda combined both of these masterfully. The ISA acted as a huge marketing tool that then forced them to have a strong commitment to delivering results for their students. This to me is the killer use case - situations where you can combine signaling with investment. 7/

For me, examples like @placement and Merittian are clear examples of companies carrying that out (though I've got some questions about the social dynamics created by the latter). 8/

Even for these use cases, I still think if you can afford it you're probably better off paying up front. If I'm confident in the outcome, I'd much rather pay $18k for Lambda instead hitting the $30k ISA cap. Not everyone has that luxury and for them the ISAs are great. /9

All of this makes me skeptical of use cases like @AvenifyHQ though I could easily be proven wrong, especially if they have an underwriting edge. But, from a pure lending standpoint, ISAs aren't new. @Upstart tried doing that 7 years ago and failed. /10

I'm sure there are other examples in the graveyard that I just don't know about because the lender dynamics are tricky. I'm excited to see the boom, I think ISAs are really interesting and powerful mechanism, but I hope we're getting the right takeaway from Lambda's success. /11