Friday, October 3, 2008

Musical Chairs

I read this earlier today, thanks to an email from a co-worker. I particularly like the part in the NYT article above where it specifically says that while the mandatory sub-prime policy looks good while the economy is performing, a downturn could prompt "a government rescue similar to that of the savings and loan industry in the 1980's". How prescient was that?

My only issue with assigning the blame here is that the sup-prime mandates from the guarantor organizations (Fannie Mae, etc), the automation of the application processes, and even consumer ignorance and predatory lending practices by brokerages really didn't cause the whole thing to fail; it would have eventually screwed a lot of individual people, and even decent swaths of the industry, but taken at face value none of those things even together could have caused this kind of catastrophic failure. What turned bad policy, individual bad choices, and denialism into a national emergency was the effect of the large firms that turned pools of purchased mortgages and loans into securities and then put them into the market, without having a fair analysis of what the loans were worth. Securitization of loans started in the 70's, and it was a bad idea then. The addition of the CRA and ARM loans later is a big contributor to the current crisis, but the mechanism at fault on a national level is the securitization process, not the loans.

Essentially, under the old system a bank held loans in a portfolio and maintained them until they were paid off. The bank's finances were tied to the centralized banking establishment, but even a large bank was never really in a position to completely hose the national economy if it failed in a vaccuum (not even if a lot of them failed at once). When loans are securitized, a company purchases pools of loans without really examining them at the same level of granularity that the bank did when the loan was signed. Now that these loans are a part of the market rather than a closed system, they are traded on speculation without giving any consideration to the risk levels the orignal lending institution would acknowledge, and the investment banks are all suddenly tied to the performance of all of their collective loans, so if one sinks they all take a hit.

The problem is that all of the involved businesses profit from the circumstances of the loan as it is initially valued, not as it ultimately performs The lender is both obligated by policy and encouraged by profit to issue these loans whenever possible, particularly since they're guaranteed anyway by the government. The broker and the lender both get paid off of the initial structure of the loan, and now with securitization of sub-prime loans they have not only a requirement by the government to pursue them, but an incentive from Wall Street to pursue them even when they are likely to become worthless, because they profit from the securitization process and do not directly suffer from the failure. The investment banks, fund managers, and other consumers of the security profit by trading on the pools, so the more there are the better the market performs, provided there is a need. That works right up until the securitization process stops because the economy dips and a significant portion of the loans fail (like all of the ARM loans that started maturing and defaulting recently), and then suddenly the securitized loans aren't worth anything because the investment banks won't trade on them without better asset liability, and now a substantial portion of the economy is invested in a bunch of worthless loan bonds. If the loans were never pooled, and never securitized, nobody but the banks who issued them and the consumers who defaulted on them would have felt the full effect; the damage would have been contained to the principle plus interest of the collective loans, rather than that plus the market price of the securities based on those loans in which the whole country was invested. Rather than twelve companies and a few hundred thousand people losing money on the deal, a few thousand companies, countless mutual funds and retirement portfolios, the Feds, the stock market as a whole, and now (with the bailout) every taxpayer is going to get bitten by this. The scary thing is that because securitization was so absurdly profitable to speculators while it was working, even as we are dropping 700+ billion on the bailout, we're still talking about extending the concept to other things, like insurance.

Even if you don't feel that the process is sort of creepy (to profit specifically off of someone else's future), it's a parasitic practice economically; essentially you are taking a set amount of money that is being borrowed and repaid, and trying to extrapolate more capital from it than it is really worth by moving it from one state of existence to another and from owner to owner. The process generates more monetary value in the market than there is actual money or asset involved (by pretty much any yardstick), and is tantamount to gambling with the national economy. Eventually, those books have to balance, and then someone gets stuck with 1000% inflation of the original, actual, tangible debt. If you think someone trading on a pool of sub-prime mortages as a bond is idiotic, wrap your head around health, automobile, renters / homeowners, and life insurance becomming a commodity soon.

The market, particularly futures trading, speculation, securities, etc, is a system of rules, so in addition to rather complicated game theory you can even reduce it down to a fair comparison with an actual game. The most telling thing about this game is that it does not conform to the normal concept of causality regarding the end of the game and the determination of winners and losers. In most games, you continue playing until someone loses, or until a certain number of players lose; winning or losing is a precondition to ending the game in completion (otherwise you just abandon the game). This functions more like musical chairs, in that there is an anomaly of causality: it is impossible to determine the outcome until the game is already over. Until the music stops, everyone appears to be winning.