Chart of the Day: Fixed vs Variable Subprime Default Rates

Alea today finds the most astonishing chart, which I simply have to reproduce:

The crazy thing here is the massive discrepancy, over the past two years, between default rates on variable subprime mortgages, on the one hand, and default rates on fixed-rate subprime mortgages, on the other. And I have to say that I'm at a loss for how to explain it.

I know what it's not: it's not ARM resets. Those are only just beginning to kick in now, while the gap between variable and fixed subprime default rates started gapping out as early as the end of 2005. (When the only resetting ARMs were the ones written in 2003.)

And it's not lax underwriting standards, either. Yes, underwriting standards did become too loose. But they were applied to subprime borrowers, not to subprime products. A lender wouldn't use much tougher underwriting standards when writing a fixed-rate loan compared to when writing a variable-rate loans.

OK, maybe I'm overegging the pudding a little here: I'm sure that both ARM resets and lax underwriting standards played some role in generating that huge spread between variable and fixed subprime default rates. But I suspect there's something else going on here as well.

My best guess is that subprime borrowers split into two camps. On the one hand, there are the Noble Few, who took advantage of excess liquidity to get their feet onto the bottom rung of the property ladder. These are people who wanted very much to buy a house – not as an investment, so much as to own their own home. They locked in fixed rates, comfortable that they would be able to make their mortgage payments for the next 15 or 30 years: however long it took to buy the house outright.

But the Noble Few were outnumbered by the Reckless Many, who had dollar signs dancing in front of their eyes and would read daily in the newspaper of the fortunes being made in the property market by people who followed the time-worn advice to "buy the biggest house you can afford". Maybe they were would-be flippers, who never intended to own their home for longer than the period of the initial teaser rate. Or they were people who wanted or needed cash, for a family emergency or just for a flat-screen TV, and they wanted to borrow that money at the lowest possible rate. Or they were suckers who were taken advantage of by unscrupulous mortgage brokers. Or they were simply financially naive, and determined that the cheapest mortgage they could get must be the best mortgage they could get.

In all these situations, the borrower chose the mortgage with the lowest monthly payments – and the mortgage with the lowest monthly payments was always a variable-rate mortgage.

To put it another way, what we're seeing in this chart is the entry of a whole new borrower class into the subprime market. Until recently, people taking out a subprime mortgage were much like people taking out a prime mortgage (but with less money and worse credit) – they were people whose primary motivation was to buy and own a house.

At some point, however, home-equity withdrawals and speculative purchases and dodgy mortgage originators all conspired to change the profile of the average subprime borrower – especially the sort of subprime borrower who chose a variable-rate mortgage. Which is one of the reasons why all the models which had worked in the past suddenly broke, despite no huge spike in interest rates.

The financial markets knew how people behave after taking out a mortgage. They just didn't realize that in the specific case of variable-rate subprime mortgages, they were dealing with a completely different set of people.

Okay, that's overegged as well. That might be some of it, but even that can't explain all of it. But just take another look at that heavy black line showing fixed-rate subprime default rates. They've never been lower. And that's something you'd never guess from reading the papers.