OK, forget 2001 pre-September. Think back to the summer of 1998. Where were you? What were you doing? (OK, so I know the answer to that for most of you.) Think back to key events: The Russian ruble’s collapse. Long-Term Capital Management’s blow-up. Week-long canoe trips.

Here’s why the sub-prime mortgage problem cannot be contained. From today’s New York Times:

At the heart of the contagion problem is the combination of complexity and leverage. The securities that financed the rapid expansion of mortgage lending were hard to understand, and some of those who owned them had borrowed so much that even a small drop in value put pressure on them to raise cash.

“You find surprising linkages that you never would have expected,” said Richard Bookstaber, a former hedge fund manager and author of a new book, “A Demon of Our Own Design: Markets, Hedge Funds and the Perils of Financial Innovation.”

“What matters is who owns what, who is under pressure to sell, and what else do they own,” he said. People with mortgage securities found they could not sell them, and so they sold other things. “If you can’t sell what you want to sell,” he said, “you sell what you can sell.”

So — no one wants to buy my shares of Nortel, which have tumbled once again, but I need cash to make the margin call in my account, so I’ll sell some shares of Barrick Gold, even though it’s not the “problem” stock. Just like in 1998, when the markets were roiled by the Russian ruble collapse, even though few investors had any money in Russia themselves:

He recalled that the crisis that brought down the Long-Term Capital Management hedge fund in 1998 started with Russia’s default on some of its debt. Long-Term Capital had not invested in Russia’s bonds, but some of those who owned such bonds, and needed to raise cash, sold instruments that Long-Term Capital also owned, and on which it had borrowed a lot of money.

Which is why the situation today, when there are 50 million more hedge funds with black box quant strategies, is so very worrying. Another NYT article (subscription only) has a good explanation of how this works:

For example, a computer might trace the relationship and trading characteristics of two similar assets, like shares of General Motors and Ford. The fund manager then makes trades, going both short and long, based on the way these shares generally trade. If Ford typically trades cheaper than General Motors, the manager would short Ford and buy G.M., capturing what may be small profits, but on a large volume….

Seeing that such bets typically generated profits over long periods left traders believing that their stakes were conservative.

And when something changes those long-term patterns?

Fund managers experiencing losses in their fixed-income portfolios who were unable to sell their positions then tried to unwind the trades they could sell — that is, stocks.

In Canada, the effect of a credit crunch will be felt less at the retail level, initially, and more at the corporate level, as the takeover premiums built into stock prices are wiped out because highly-leveraged private equity financing dries up. The other key effect will come from demand for oil, as the Canadian economy’s growth is so tied to oil right now (Ontario’s manufacturing sector is just inches away from being an actual drag on economic growth). So headlines like this, while inevitable, are not good news for the petrodollar: Oil Trades Near Five-Week Low on Concern Demand Growth May Slow. Our economy can’t stay strong without customers, and those customers still live almost entirely in the U.S. If the U.S. consumer feels poor, the Canadian one does too — eventually.