The high level is ISAs are very, very hard to make work. We designed them to be *incredibly* student-friendly.

In their current form a school has to be executing nearly perfectly to make money on them. https://twitter.com/mattsuthe/status/1380355585833504769

To use an ISA a school takes equity-like risk but gets debt-like returns.

If a student doesn’t do well in the school they probably pay zero. If they don’t get hired they pay zero. If they get hired below the income threshold they may pay but potentially not for years.

Meanwhile, in every student’s mind, they have/are paying the full ($30k) cap.

Most students won’t pay the full cap ever, and there’s no interest, so the school pays that. The net present value of them having paid x dollars 2+ years from now is far, far less than x.

Basically ISAs take everything that sucks about loans (risk, interest) and the school eats all of that.

It’s difficult to explain just *how much better* ISAs are than traditional loans, but you’ll see evidence for that in schools starting to push students to loans more and more.

It’s entirely possible that in the next 12 months nearly every school stops offering ISAs. In fact, I think that’s the most likely outcome.

Lambda School is in it for the long haul, though. It’s interest-aligned tuition or nothing at all.

I think the next iteration of ISAs will make it clearer than the ISA cap is not the *price* it is the *maximum* - it's a failsafe.

We made the cap so low it's actually confusing.

I also think they won't be called ISAs because that's a pretty stupid name in my opinion.

The best alignment is probably a lower percentage of income, longer duration, higher cap.

If the ISA payment is x%/yr the goal should be to help one increase his or her income by 2x% or 3x%.

Then your education is effectively free.