In this post, there's a conversation about research into Why Some Loan Pools Stink So Much. It's one thing if it's a bad credit score and 100% financing on a house that isn't worth what people thought it was worth (I realize that's a bit tortured and apologize, but I think that's the correct formulation). It's wholly another thing if the pool is characterized by low LTV and good credit scores. AMBAC is apparently digging all the way down through the servicer paperwork to see What Really Happened, and while they aren't committing themselves yet, the implication seems to be massive, endemic fraud. That can probably be identified and will therefore be prosecuted. Stay tuned, apparently.
Really disturbing is the idea that in some pools of second mortgages, the loss rate is 81% -- as in, the borrower walked and the lender got more or less nothing out of the collaterol. I don't care _how_ good a deal someone offers on that kind of distressed debt. You'd be hard pressed to break even paying 10 cents on the dollar on shit that stinky.
The guy being interviewed wraps it up by pointing out a graph that shows a real drop in the rate of rate of loss (again, sorry about the tortured language), and notes that's because otherwise, you'd have more than 100% loss of collateral, which doesn't make sense either.
Terminal velocity, apparently. Or maybe paste on concrete?
Bubbliciousness, but not fraud, I was looking elsewhere on Calculated Risk and saw a post about CRE. What the heck, I go, is CRE and google is not instantly helpful. Context, however, was: commercial real estate. I had been commenting, a few weeks ago, about Las Vegas real estate to some of my husband's relatives and my brother-in-law popped up with some comment about how Las Vegas' real estate was still going hot and heavy. This made me go, wha' hunh? Having just read, well, never mind. Anyway, he was talking about commercial real estate and I had it in the back of my head that CRE (that's one useful TLA, it turns out) lags (just like some regions of the country tend to lag) residential real estate. Calculated Risk gives a time frame (4-7 quarters) and then proceeds to speculate about what we can expect based on previous bubbles (S&L crisis, etc.).
They then discount the S&L one as non-representative and try to figure out how overbuilt CRE is this time around based on vacancy rates, suggesting things aren't quite as bad this time around. I don't know. Driving around the other day looking at houses, I was really struck by all the vacant retail and light industrial we also happened to spot. Also interesting to see that CRE drops tend to happen faster (more compressed in time) than residential, which makes sense to me.