Any Economists still read this?
by Tom Temple
30 September 2008
Back in February, I wrote this article describing an ominous credit crisis as well as I could understand from my brother-in-law. Roughly speaking, it was a prediction of what has happened in the last few weeks. (Actually, he predicted this almost exactly as it happened, I just didn’t understand him.)
We had some excellent commentary that largely assuaged my fears. In particular, you guys convinced me that there would be sufficient liquidity to prevent any wide-spread problems. I am hoping we could continue that discussion given the current situation. Do you guys think that the government should do nothing? Should it buy the securities at less than the “hold to maturity” price? Do you think the bailout plan is good/necessary?
Sep 30, 09:12 PM
THC lives!
I didn’t participate in your last post Tom, because I didn’t feel like I knew jack shit about that stuff. I knew there was a housing bubble. And I knew that the downside of the housing bubble would likely be bad (for some people, at least). And finally, I knew that since housing markets are inherently local, different communities would feel the sting to varying degrees.
Boy have I learned a lot over the last 2-3 weeks. Enough to know that I really don’t know shit about complex finance. Or maybe just enough to make myself look dumb.
Anyway, I have two points:
(0) Would anyone like to comment on the fact that this whole mess could plausibly be laid at the door of some shitty credit scoring models (i.e. machine learning models!) from the early 90’s? Specifically, the fact that these models were “trained” during a period of continual increases in housing values leading them to discount the risks of subprime lending. Check out Arnold Kling’s discussion recently on bloggingheads.tv.
(1) We should carefully decide whether our goal is to fix the problem (i.e. restart the credit markets) or to punish the people who fucked up. I feel like there’s going to be a tradeoff there.
Anyway, it’s pretty astounding to read about what exactly mortgage backed securities and credit default swaps actually are (if anyone actually knows!). Mortgage backed securities I get: it seems the lessons here are pretty simple, do a better job of assessing risk for individual loans and raise the capital requirements for these sorts of financial institutions.
Credit default swaps I still don’t really get. Is it basically like buying insurance for debt? Like if someone owes me money, and then I buy an insurance policy on that debt from a third party?
I dunno. The more it all tumbles around in my head, the more I feel like the whole thing was just a house of cards built on a whole crap load of shitty risk evaluations on mortgages. Like if we actually knew now what the default rates were going to be, these assets would have prices and the banks would suddenly know which among them were about to disappear and which were just going to be really fucked for a while.
Oct 1, 10:27 AM
I was thinking about that post during all this action the last couple weeks. The prediction was remarkably good for timing and the timing of the market slump, though I don’t think it got the mechanism quite right. The prediction involved baby boomers panicking and irrationally pulling money out of equities.
In practice the equity price drop reflects unraveling of the credit market (which affects the true underlying value of stocks). And the big change in the recent weeks has been a decrease in the supply of loanable funds, not in the demand side.
Financial organizations have lost a lot of money, and they are overvaluing assets. That make it dangerous to loan to these firms. So those who make loans demand a higher interest rate. This in turn means that only those who need loans badly will take loans, but those are the institutions in the worst shape. Thus firms are even less likely to make loans. This “cycle” freezes the credit market.
Stocks have lost value because they are truly less valuable when credit markets don’t work.
The intent, as I understand it, of the bailout is to give these institutions something of known value (cash) so that financial institutions can simply look at each other’s assets and know who is credit-worthy. It comes with big costs (the oft-talked about moral hazard of bailing out these firms), but the credit market as it stands promises a serious crisis. I won’t pretend to know what to do.
My hunch of an explanation doesn’t involve bad credit scoring models. The guys making investment decisions for these firms could make millions in bonuses if they took big risks and it worked out. When it doesn’t work out, they don’t personally lose millions. They might get fired, but the see more of the upside than the downside. So their contract structure encouraged them to be very aggressive.
As a result, financial institutions took more risk then they could handle. Those risks went poorly, and they’ve lost a lot of money.
Some of them, in an effort to appear credit-worthy, are hiding how much they’ve really lost. And that brings us back to lenders not knowing who they can trust.
Oct 1, 02:28 PM
Dan said: My hunch of an explanation doesn’t involve bad credit scoring models. The guys making investment decisions for these firms could make millions in bonuses if they took big risks and it worked out. When it doesn’t work out, they don’t personally lose millions. They might get fired, but the see more of the upside than the downside. So their contract structure encouraged them to be very aggressive.
As a result, financial institutions took more risk then they could handle. Those risks went poorly, and they’ve lost a lot of money.
The difference being that they all knew the default rates on these mortgages would be much higher than advertised? I thought the point of all this restructuring of mortgage debt was to convince buyers (of the securities) that all that risk had been “washed out”. So doesn’t that mean that at some point in the chain here, someone was either (a) dishonest or (b) ignorant of the true risks of the investment? Maybe some of both?
Oct 1, 08:28 PM
Joran:
I’m not clear who you are saying was being dishonest or ignorant. The CDS buyer or seller?
AIG employees raked in huge bonuses selling CDS’s. When they lost their shareholder’s money, they walked away unscathed.
The banks making bad loans and buying CDS’s made a lot of money in lending fees. My understanding is that the subprime lending era was lucrative, and they were giving big bonuses.
So the guys making decisions all did fairly well for a while.
I also think there was a potential upside. Credit derivatives allowed AIG and others to take huge gambles on the default rate, and there could conceivably have been fewer defaults than they predicted. It also let the banks gamble on AIG’s solvency.
As luck would have it, both gambles were a loss… but I’m not sure that was inevitable.
Oct 1, 10:35 PM
0) To be fair to Brian, I mostly filled in the baby boomer stuff from my own head.
1) I agree about the incentivizing risk at the brokers. In fact, back in February when I asked him about why other people weren’t acting like him, he actually said something like that. He said that there were other people who knew what he knew and were still making those trades for the bonuses. They knew that at worst they would lose their jobs at the end of the line.
2) I think it’s safe to say that the days of buying a house with >=10% down are over.
Oct 2, 11:38 AM
My main question is still unaddressed. Why does the government have to pay the “hold to maturity” price on the securities? Your earlier argument supposed that there would be a less artificial price which would find buyers (i.e. not the government). Are there simply not buyers besides the government (because of illiquidity)? Or is that “natural” price somehow unacceptably low?
Oct 2, 01:22 PM
I don’t see why the hold to maturity price is artificial. Though no one knows what the hold to maturity price is, because we don’t know what the default rate will be. Normally the market does a good job of predicting these prices, but the market isn’t functioning right now (illiquidity as you suggest).
My understanding was that Bernanke proposed buying securities using a reverse auction. I’ve heard that the reverse auction is expected to result in hold-to-maturity prices. I don’t know if that’s true or not.
But reverse auction doesn’t seem unreasonable to me.
You’re probably right that government could use another mechanism to get a lower price. I’d guess pricing these things is going to be difficult under any mechanism.
I also don’t know whether buying securities at a cheap price and forcing sellers into big markdowns is counterproductive or not.
Oct 9, 01:41 PM
More thoughts on my point about the downsides in the shift to a purely analytical, quantitative risk assessment scheme, via Tyler Cowen:
http://www.voxeu.org/index.php?q=node/2101
Dan: I’m probably being unclear, because I don’t really have a great grasp of CDS’s at all. I certainly take your point about the relevant incentives for middling employees; that definitely seems relevant.
My point is that it seems to me that there’s been a massive shift in the last 15-20 years in assessing financial risk in ways that are purely analytical and oftentimes staggeringly complex. And that part of our current problems might stem from a certain amount of overconfidence in these models. Basically the same problem that happens with statistics all the time: the people who create these complex risk models aren’t the ones making decisions based on the output of the models. It’s like the old Feynman quote about the Challenger disaster, about how there was a disconnect between the engineers actually studying temp effects on the shuttle and the “suits” making the decisions.
Complex mathematical models are hard to use well. The more complex and subtle they get (even when they really do have useful information to impart) the harder it is to communicate the output of these models to someone who hasn’t been hip deep in the technical details for some time.
Oct 9, 11:37 PM
Hi Tom!
Brian’s understanding and prediction of this mess has been pretty unbelievable. If you want more info. you should ask him. I’m sure he could lay it out for all of you. I’m also pretty sure that there is a lot more to his explanation than you initially described in February. Either way we took everything out of the stock market months ago. Turns out “chicken little” was right.