Some comments I put on Brad Delong’s web site.
Something that is also important is the time element. When things start going bad, how quickly do they start going bad. The danger is the classic bank run. Once your balance sheet starts going bad, people want their money which causes your losses to increase.
The in the case of Bear-Stearns and Lehman Brothers, this run unfolds very, very quickly. The stock in both companies plummeted within a week. The reason for this is that their business model involves borrowing short term to fund long term. If you have a bank run, the short term borrowing stops, and you are dead. In the case of Lehman and Bear-Stearns, things were happening on an hour-by-hour basis.
In the case of Citigroup, it has insured deposits and a liquidity line with the Federal Reserve. This means that when something bad happens, it falls apart much more slowly, and this gives you time to think about what to do. The thing that you have to be careful about is not to do anything that makes the problem worse. If you have a crisis of confidence, then saying or doing the wrong thing, just panics people even more. On the other hand, the other danger is that by doing nothing on the theory that doing something will panic people, you run the risk of looking as if you are out of touch and this panics people even more.
Also if you look at the systemic long-term problems at Citigroup there are two:
1) Citigroup has more employees maintaining the same amount of money as the mega-banks. The layoffs at Citi have been more than its competitors, because the pre-crisis employment numbers were higher.
2) The business model that Citigroup was advancing when it pushed for the passage of Gramm-Leach-Billey never quite gelled. The idea was that Citi would be your one stop financial services shop, but they never were able to quite to get this to work.
James Gary: The central thesis of the original post seems to be that the Citibank’s problems are due to “erroneous distributional assumptions in the risk models.” What specifically were these “erroneous assumptions?” Is there any quantitative way to describe them? Please advise.
The basic erroneous assumption involved calculating the probability that one mortgage will go bad if another goes bad. One way of thinking about it is that one way of calculate the chances that one window will break if another breaks is too look at historical window breakage rates, and think about how likely it is that one window break will cause other window to break.
The problem comes in is if you have a category five hurricane hit, then all of the windows will break at the same time. It’s worse if you are looking at a place in which a hurricane has never hit before. The two things that the risk models didn’t consider were
1) when one thing goes bad, everything will go bad, and
2) just because something has never happened before, doesn’t mean that it can’t happen. There were probably some meetings a few years ago where someone said “the only way we could get into trouble is if we have a repeat of 1929.” This is great, until you have a repeat of 1929.
This gets to a very deep philosophical problem which is causing lots of problems right now. You generally deal with the future by looking at the past, but if you are in a situation where you have something that isn’t quite like anything you’ve ever seen, what do you do? How do you predict the unpredictable or know the unknowable?
Finally there was one particularly bad assumption. In every CDO model I’ve seen, the assumption was that you’d be able to recover 40% of the value of the loan. This is partly because CDO models and CDO’s were originally designed around corporate bonds, and when a corporation goes under, there is a very well organized process for keep the corporation alive and settling losses, and 40% recovery isn’t an reasonable assumption if you are creating a security off of corporate bonds.
It turns out to be a very bad assumption for subprime mortgages. When a corporation goes bad, you still usually have a factory or some inventory that can be quickly sold, whereas when a house gets foreclosed, you end up with something that may be unsaleable. Also it’s bad because if you are creating a CDO off of a security of mortgages instead of mortgages themselves, you end up concentrating risk.
There is one more point that needs to be made, and what is why Citi started really having problems last week and not before that. After all, Citi’s balance sheets and it’s holdings of mortgages had been known for years. Why did everyone suddenly panic last week?
The reason for that is that it’s dawned on everyone that we may be in for a really bad recession. If it were the situation that all Citi had to do was to write down subprime, then there wouldn’t be a problem. What has people spooked is not the subprime (since we’ve known about that for months), but if we start getting 8-10% unemployment rates, then people will stop paying prime mortgages, credit cards, auto loans, and student loans. If all those go bad, then what had been great assets before now turn into duds.
So we really are looking at a bad cycle. If something bad happens to GM, you have unemployed workers defaulting on their auto loans and credit cards, which affects Citi, which then affects more companies, etc. etc.
What’s worse is that we were in a post-election gap in which there was no one that could get on television and say “we’re working on this and everything is going to be all right.”
The one piece of good news is that the “road to hell” happens week by week which gives you time to stop the spiral down. In the case of Lehman and Bear-Stearns, the “road to hell” was happening hour by hour.
No comments yet.
RSS feed for comments on this post. TrackBack URI
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account. ( Log Out / Change )
You are commenting using your Twitter account. ( Log Out / Change )
You are commenting using your Facebook account. ( Log Out / Change )
You are commenting using your Google+ account. ( Log Out / Change )
Connecting to %s
Notify me of new comments via email.
Blog at WordPress.com.