Original thinking in a derivatives market
How those models work
Nor have the banks completely washed their hands of the original risk, since many act as prime brokers to hedge funds in mortgage derivatives, lending money secured against holdings in instruments, which may have been created by the retail arm of the same bank.
Thus, the risk is now spread very widely in complex ways, which meant that when the sub-prime mortgages started to smell bad, no one could work out which institutions were being hurt.
Banks are quite well equipped to deal with rapid price swings in volatile markets. Some actually make good money in them, but uncertainty meant that the wisest action seemed to be not to lend money to anyone. This is how recessions or even depressions can begin, as it may become a vicious circle.
Handwashing in a risky liquid
However, the US and UK governments seem (for once) to have taken on a useful role in the markets, and pumped in liquidity. The sub-prime shakeout has caused a lot of people to question all sophisticated credit instruments, which in the short term is making it harder to borrow, and cost more, so some sort of economic slowdown is inevitable. Banks will also make less money, and I can feel your sympathy from here.
A problem with securitisation is the bank issuing the loan has some seriously bad incentives once someone else is going to take the downside. US retail bankers pretty much threw money at passers by, allowing people with tragic credit histories to borrow against property. This was made worse by borrowing the idea from credit cards where you start off paying a lot less than the standard rate, making default rates look low, before moving to the British model of floating rate mortgages, at a time when rates were going up.
Even before all this, banks had hordes of risk managers, whose job is to make market glitches survivable. They validate the models produced by quants, and try to run them into the future, to guess the probability of it all going wrong.
Of course, it's simply not possible to work out all the combinations of price movements, so they throw a vast array of randomly generated values at it to simulate the next couple of weeks. Of course the same Monte Carlo approach is also used to value them in the first place which is powerful, if more than a little slow.
The nature of these randoms means that their handling requires care. First off, the standard VBA/C++ library functions are garbage, but still get used. Then there is the question of which distribution you should choose, which is controversial.
Many use the lognormal distribution, which resembles what we see in real life. Except that it gravely under-estimates the probability of big price movements, as Nassim Taleb has been telling people for a long time. Ignoring this has led several banks to encounter issues that "should" happen only once in the life of the Earth, actually happening twice in the same month.
Although the high end of risk managers do serious maths, ultimately this scale of number crunching simply has to be done by computer. The days of lending to the right-sort-of-chap are long gone. ®
Sponsored: Hyper-scale data management