For over a year now, we've had the financial and general media referring to the collapse of the real estate bubble as "the subprime crisis". It's been said so often that it's now the accepted truth.
The storyline has been that the entire problem with real estate has been "subprime borrowers with poor credit" who are "losing their homes." So everybody with good credit is A-OK, right?
Wrong.
Welcome to Housing Crash 2.0... "the Alt-A mortgage crisis."
The real issue with housing is not now, and never has been, subprime lending. The problem was, and still is, prices. Prices that increased to levels far beyond what any buyer's income could support.
If we look up the income and credit score chain a bit, we go from subprime lending to Alt-A lending. An Alt-A loan is a loan made to a borrower with good credit, but which does not conform to the underwriting standards for a prime loan. No-doc loans, "NINJA" (No Income, No Job, no Assets) loans, interest-only loans, payment-option loans, 100% financing, pretty much anything beyond a fully-amortizing, 80% loan-to-value, fixed or adjustable-rate, ordinary, boring, traditional mortgage.
In the areas of high appreciation, like Southern California, Arizona, Florida, Las Vegas... you know, just those places where the people are, a large proportion (70-80% in San Diego) of the mortgages to borrowers with good credit during the 2004-2006 boom were Alt-A. Why? Because the buyers could not qualify to borrow the amounts needed to pay the prices under traditional underwriting. They didn't make enough money.
Let's take an upper-middle class, college educated couple. Give them a combined income of $120,000 a year, and a FICO in the mid 700s. Under traditional (IE, pre-2003) lending practices, such a couple could probably borrow around $300,000 - $325,000. With ~20% down, they'd have somewhere between $350,000 and $400,000 to spend.
Were these people buying $350,000 houses? Of course not. They were buying $800,000 houses, even million dollar houses. I know. I worked for a company that installed alarms and home entertainment systems for all of the major housebuilders. When you're a service tech, you talk to your customers. Very few will tell you what they make, but everyone will tell you what they do.
So how did people with $120,000 incomes buy $1,000,000 houses? Enter Alt-A lending.
The buyer doesn't make enough money to qualify? Easy, go no-doc... just write the income in. In San Diego, 95% of no-doc loans overstated the borrowers' income... 80% of those overstated by 50% or more.
Payments are too high? Easy, use a pay-option ARM with a low initial rate. The minimum payment is low enough that they can make it, and who cares about negative amortization when prices are gong up 30% every year? Industry estimates are that 70% of payment option borrowers are only making the minimum (negative amortizing) payment.
So how are these loans doing?
Not so good.
From
housingwire.com, referencing a report from Clayton Fixed Income Services, Inc.
The picture being painted for Alt-A is increasingly beginning to look a whole lot like subprime, as a result, even if peaking resets in the loan class aren’t expected until the middle of next year. In particular, loss severity continues to ratchet upward — a trend that portends some likely further reassessment of rating models at each of the major credit rating agencies, as they catch up with the data.
Loss severity — the average amount lost relative to unpaid principal balance — reached 41.4 percent for all Alt-A first liens in REO during the most recent rolling six month period through May, Clayton said; that was up from a 37.6 percent rolling average one month earlier, and compares to a similar 49 percent loss severity average for subprime first liens liquidated in REO through May.
So Alt-A losses are trending up to where subprime losses are. And the loan amounts tend to be bigger, so we're talking about even more money than is being lost in subprime.
But hey, these are people with great credit scores! So good, in fact, that that's all the lenders asked them for when they qualified them. Surely they are a lot less likely to default, right?
Maybe not...
Add in soaring borrower defaults, and the picture doesn’t get much better. Clayton reported that the 2006 vintage saw 60+ day borrower delinquencies among Alt-A first liens reach 21.22 percent in May, up 10.5 percent in a single month; 2007 fared even worse, with 60+ day delinquencies ratcheting up 22 percent to 18.55 percent.
Even the 2004 Alt-A vintage, what’s left of it, is defaulting at a fast pace: 60+ day delinquencies in the vintage shot up by nearly 24 percent in just one month.
We've also been told that it's ARM resets that are causing the defaults and foreclosures. Again, not so fast:
"Amid all the attention being paid to rate adjustments, however, it’s important to note that out of all the active delinquent ARM loans in Clayton’s portfolio, approximately 70 percent were already delinquent prior to the first rate change date," analysts at the firm noted in their report.
(emphasis added by me)
Last, but not least, this graph is truly terrifying:
Alt-A Resets By Month
Alt-A loans are blowing up at a far higher rate than the lenders predicted, and the peak in resets is still a year and a half away.
And, since I like graphs, especially really, really scary ones, I'll close with this one, from Professor Piggington's Econo-Almanac for the Landed Poor. It plots San Diego house prices as a multiple of per capita income.
Graph of the San Diego Real Estate Boom
--Shannon