The subprime mortgage crisis (useful primer here) has provoked the usual partisan reactions, with the left griping about "corporate governance" and "transparency", and the right telling everyone to just "suck it up". I fear that both sides are underestimating the severe and fundamental nature of the problem.

First, some history: in the 1970s, Western economies all seemed to be going haywire simultaneously. Inflation was running rampant, governments were running huge deficits, unemployment was skyrocketing, and the world's major economies were lurching from crisis to crisis. In retrospect (most modern economists will tell you), the problem was that the old Keynsian approach to government intervention in the economy had reached its limits. Further government spending only seemed to exacerbate inflation and deficits, without providing much of a stimulus to the economy. The problem was fixed (again, most modern economists will tell you) by governments shifting to monetarism as the new, better way to guide the economy to prosperity.

What this story doesn't explain, however, is why Keynsianism, which had fueled two and a half decades of postwar prosperity, suddenly stopped working. Some economists will say that it was always too imprecise a tool, and its wielders were bound to lose control eventually. Others will say that it was "abused" by governments eager to keep the good times rolling. Like an antibiotic, the latter would explain, Keynsian pump-priming loses its effectiveness when overused unnecessarily during prosperous times, rendering it incapable of mitigating the inevitable downturn.

But economies aren't bacteria. How do they become immune to economic stimuli? Why wouldn't "overuse" of Keynsian stimulus, whatever its side effects, at least succeed in its basic purpose of warding off recessions?

The answer can be summed up in one word: expectations. Once enough people start assuming that the government's response to any economic slowdown will be more government borrowing and spending, they can place financial bets on that assumption. For example, they can bet that the government will ensure that lots of money remains available in the economy, and that they can therefore raise prices accordingly, or demand higher wages. As these bets pile up, they dampen the effect of the intervention they anticipate, forcing the government into an even more extreme intervention to achieve the same result--further heightening expectations for future interventions. Eventually, expectations match the government's maximum feasible effort, and all interventions fail. Only a completely new, unanticipated form of intervention can hope to work.

Let us return now to the subprime crisis. In 1987, Alan Greenspan massively expanded government credit and cut federal interest rates in response to a stock market crash. He was to do this multiple times over the course of his career--in 1998, and again in 2001--in response to similar economic crises. This maneuver--which came to be known as the "Greenspan put"--was remarkably effective at mitigating the effects of economic shocks. However, it has also been blamed for the stock bubble of the late 1999s and the real estate bubble of the early 2000s.

The connection between these bubbles, the policies that preceded them, and the lethal effect of expectations is nicely illustrated by the investor behavior that led to the subprime crisis. Why did so many sterling financial firms pour money into highly questionable investments based on mortgages of dubious quality? Were they fools or maniacs, caught up in some kind of frenzy? More likely, they were sensible people making a very reasonable bet: that when the bubble burst, the Federal Reserve would come to the rescue with a massive interest rate cut, and most if not all of their bubble profits would be preserved. And indeed, plenty of subprime mortgage investors, Bear Stearns notwithstanding, have ended up netting a hefty profit from this bet.

Unfortunately, the bet only increased the magnitude of the crisis, while dampening the government's capacity to resolve it. The subprime-driven real estate bubble was not only far bigger than previous bubbles, thus requiring a much bigger liquidity infusion than previous ones--it also juiced the economy enough to nudge inflation upward, limiting the government's leeway to cut interest rates without triggering an inflation spiral. As in the 1970s, the reigning paradigm for government management of the economic cycle has become too predictable, and has thus lost its curative power.

I don't know what the next paradigm will be, but it had better be pretty darn complicated. Financial firms have enormous analytical resources at their disposal these days, and are much better at predicting government policies than they used to be. It'll take an even more inscrutable policymaker than the famously Buddha-like Greenspan to keep them from catching on to the government's methods very quickly--and promptly undermining them.