Here's Paul Krugman's rehash of Keynesian business cycle theory:
So, how does this all end?
I’ve been saying for a long time that this isn’t your father’s recession — it’s your grandfather’s recession. (I actually used the phrase about the last recession, too.) That is, it isn’t something like the 1981-82 recession, which was brought on by the Fed to control inflation, and ended when the Fed decided that we had suffered enough. Instead, it’s like the 1929-33 recession — or the recession of 1873-1879 — a slump brought on by the collapse of an investment and credit bubble...
Maybe it’s time to dust off Keynesian business cycle theory.
Keynes himself actually didn’t have much to do with this theory. In fact, one of the key moves in his development of the General Theory was the decision to focus on how economies stay stuck in depression for extended periods, rather than on the more complex question of explaining the economy’s ups and downs. But he did devote a brief chapter at the end to the subject, and Hicks elaborated on this quite a lot.
What’s notable about this theory is that it made no use of the self-correcting mechanism expounded in every principles textbook, mine included — the mechanism in which falling prices lead to a rising real money supply, which
shifts the aggregate demand curve outmoves the economy down the aggregate demand curve. Why? Well, as we’ve now learned the hard way, a sufficiently severe bubble-bursting pushes you into the liquidity trap, and makes the aggregate demand curve more or less vertical.
Instead, recovery comes because low investment eventually produces a backlog of desired capital stock, through use, delay, and obsolescence. And eventually this leads to an investment recovery, which is self-reinforcing.
And what do we mean by use, delay, etc.? Calculated Risk had a nice piece on auto sales, which I find helps me to think about this concretely. As CR pointed out, at current rates of sale it would take 23.9 years to replace the existing vehicle stock. Obviously, that won’t happen. Even if the desired number of vehicles doesn’t rise, people will start replacing vehicles that wear out (use), rust away (decay), or just are so much worse than newer models that they’re worth replacing to get the spiffy new features (obsolescence).
As autos go, so goes the capital stock. In the long run, we will have a spontaneous economic recovery, even if all current policy initiatives fail. On the other hand, in the long run …
Note the casual tone here. That's not just Krugman's style. It's essential to Keynesianism. As became clear during the monetarist and new classical (rational expectations) counter-attacks in macroeconomics during the 1960s and 1970s, you can't make Keynesianism rigorous. You can't base it on rational optimizing agents in competitive markets. Keynes' original argument relied on various ad hoc behavioral and/or institutional assumptions, and was not always rigorous even when those assumptions are accepted.
Think about the argument here. A "a backlog of desired capital stock?" What does that mean? Surely desire for capital (meaning capital goods, not money) is never in short supply. I would be glad to have a house and a few cars and some functional machinery to make stuff with! What limits people's willingness-to-pay (not their desire) for capital is that they expect the returns to be low. "Obviously, that won't happen," writes Krugman about replacing the nation's stock of cars in 23.9 years. No, that's not obvious. If the economy deteriorates enough, we'll use our cars longer, and when they give out we might not replace them, just ride the bus or something. Yes, the utility of cars rises as they become fewer, so there might be some pickup in demand as more of them fail, but that might just cause another drop in demand for something else (houses, say, or computers) as people cut other expenses to replace their cars.
Krugman diagnoses "an investment and credit bubble." Something looks like a bubble if it pops. That is, if variable x drops like a stone this year, x's level last year looks like a bubble. But it might be, not that x was at an unsustainable level before, but that there was some shock, some disaster that caused it to fall. Housing was a bubble, yes. Credit -- maybe. But investment, no. First, investment, except residential construction, hasn't been so high for the past few years. Second, the concept of an investment bubble is problematic. It's clear what it means to say there's a bubble in, say, tulip bulbs: tulip bulbs are nonsensically expensive relative to other goods. But investment is just a general term for all the stuff we make today that makes us richer tomorrow. How can you have too much of that?
There's a much more parsimonious story to be told here. A story that can be told in non-casual ways and make sense. In the past few months, there's been an explosion of political risk, resulting in particular from the victory in the 2008 elections of Democrats who want to tax and spend more. Investors have to think: If I invest, and succeed, how much of my returns will get confiscated by the state? The answer to that question has just gotten a lot worse, ergo investment has fallen off a cliff.
UPDATE: By the way, for the record, I think Keynes might have been right that in 1936 what the economy needed was massive government spending. The policy insanity of the 1930s had so devastated investor confidence that moderate measures and efforts to commit to non-expropriation couldn't work, at any rate not while FDR was president. (If Wilkie had won in 1940, maybe.) So having the lumbering government force the economy to move may have been the best we could do then. And that's sort of what happened. WWII was helpful because it forced a return to some kind of means-ends rationality: we had to make stuff and do things, even if it was just for the sake of killing rather than, say, the good life, and so we had to give doers and makers room and respect. The 1930s cult of envy and escapism and paranoia withered and we became grown-ups again, and the habits of making and doing, combined with the Red Scare, brought a return of capitalism and prosperity (which WWII was not, despite high GDP numbers). Keynes had one foot in theory and one foot in policymaking, and while I think the theoretical pretensions of the General Theory are hugely inflated and the biggest misstep in 20th-century economics was to buy into those pretensions, Keynes' advice may well have been largely right in an immediate sense. Maybe Keynesianism can be compared to feudalism, which emerged during the 9th-century Viking invasions of Europe: it was a crude institution that may have been appropriate for a particular time, but whose continuance was unfortunate later. Keynes was bad for economic theory but maybe good for economic policy, in the short run. Today's Keynesian-revivalism, by contrast, is not theoretical improvisation to justify a sound policy hunch, but a bungling application of outworn economic doctrines that were invented for a very different economy, long ago.
UPDATE II: By the way, I was a little unfair above: the explosion of political risk began, not with Obama's and the Democrats' victory, but to some extent with the Bear Stearns bailout earlier this year, and especially with Hank Paulson's ultimatum last October. Obviously these events were forced by markets to some extent, and there's no way for the authorities to "do nothing" in such cases (although they could have "done less"); since the state is the ultimate guarantor of the contracts of a bewildering web of which the big banks are at the center, it can't not be involved. Paulson's ultimatum, in particular, was a breaking-point of sorts: I think prior to that moment no one had quite conceived that such a thing-- such a demand for authority, such volumes of money-- could occur. Everyone's political model of what the government is and does vis-a-vis the market was thrown into chaos. Obviously that's not Obama's "fault" even if he supported it.
But the handover of power in the wake of the Paulson bombshell was peculiarly unfortunate, because a revolution in the economic constitution was already underway and then was handed off into the extremely inexpert hands of Obama, and the hungry paleoliberal Democrats with their backlog of ancient agendas. That Obama has capable lieutenants may be of little use. What is needed is not advice but decisions, or rather, commitments. Something to fix, and coordinate, expectations, to eliminate the sense of flux, the toxic counter-point of high-stakes grabbing and dreamy improvisation.