Malinvestment and Recessions

Frequent commenter Pithlord asks a good question in the comments to my last post:

[M]arkets can be rational-but-wrong. But if that’s the case, and if the markets are revising their beliefs about future real productive capacity (as opposed to nominal demand), then there must have been some new piece of information in September 2008 that meant it was rational to revise expectations of future productive capacity downwards. A lot. I have trouble seeing what that piece of information would be.

That’s an extremely good question, and I’ll try to answer it as I see it. But first, I want to stress that I’m not an economist, and that I come at this with, basically, the instincts from my Wall Street career, as well as my strong inclination to believe that the people who run the Fed are not idiots. They may certainly be wrong – we’re very much in uncharted territory – but if the solution to our economic problems were obvious, and in their hands, then you need an explanation for why they are not pursuing that solution. They might be more willing to tolerate unemployment than someone else – they may have other priorities. But any dovish argument needs to grapple with those priorities. Simply asserting that there is no other side to the question – that it is obvious what is to be done, and it’s simply mysterious or evidence of malicious intent if the powers that be don’t do it – well, it’s just a poor mode of argumentation, it seems to me.

So, here’s how I look at it. I don’t believe the wealth effect is terribly significant. In general, people don’t spend (much) of rises in asset values, and they don’t save (much) more in response to asset price declines. That’s because most people don’t own much in the way of assets. Most people consume what they earn – or more. And the people who own assets don’t reallocate between savings and consumption that much in response to market fluctuations, even big ones. Markets fluctuate for all sorts of reasons; that’s why they say the stock market has predicted twelve of the last four recessions.

The kinds of market declines that presage recessions are ones related to the creation of large amounts of bad debt. When large amounts of debt go bad, it’s a problem in and of itself – money gets “stuck” cleaning up debt rather than being deployed more productively. But it’s also an indication that prior investment decisions – on a broad scale – were based on exceedingly faulty information about possible returns.

In the 2000s, we had reasonably good growth in consumption, and therefore reasonably good economic growth. But a huge percentage of this consumption was underwritten by rising housing values. The market behaved, for some time, as if either this would continue forever, or it would be eventually replaced by some other driver of consumption growth. When the housing market turned negative, the market initially treated this as a cyclical downturn – there would be less construction in the near future, so money would flow into other goods or services. But when the sheer scale of the bad debt became clear, that revealed two things. First, it revealed massive weaknesses in our financial system – which is what caused the financial panic. But it also caused a significant revision in expectations about how much consumption would grow in the future. The debt overhang was directly going to cause people to consume less as they saved to pay off their bad debts. But indirectly, it changed people’s expectations because the scale of the bad debt made people realize how much of the consumption growth of the 2000s was tied up in a rising housing market. Tied up, in other words, in an illusion, one that was not going to be repeated or easily replaced.

The mutual fund disclaimer says “past performance is not indicative of future results” but that’s all we have to go on in the real world. If past performance was good, we think future performance will be as well. If the past performance is revealed to have been not as good as previously thought, then future performance expectations will be revised sharply downward.

I actually saw this happen once before, on a much smaller scale, in the 2000 recession. I remember when the market’s perception changed from the notion that this was just a correction to the expectation that something was going to change in the real economy. We could have vaporized trillions on paper without any material consequence in the real world – as I say, I’m skeptical that the wealth effect is very large. But two high-profile bankruptcies changed the way the correction was perceived: Enron and Worldcom. Both were instances of fraudulent accounting. Even though both were large companies, their bankruptcies alone wouldn’t have been a big deal. But because their frauds were so longstanding, what the market understood was that these companies had not really been profitable for some time. Moreover, the market understood that other companies – their competitors – had invested heavily, with borrowed money, to copy these apparently moneymaking strategies that were now revealed to be fraudulent. All these other telecommunications and energy companies were now perceived to be at risk – projections for profit went out the window, and projections for bankruptcy replaced them. So, once again, what changed was expectations for how much return the economy was likely to produce. What had apparently been produced profitably in the past was revealed to be fraudulent – value-destroying rather than value-creating. That changed expectations for what might be produced in the future. On a much smaller scale, but a similar process.

The line between what I argued in my last post and what the inflation doves are arguing isn’t actually that sharp. I’m not denying that a rise in inflation would boost demand, all else being equal. I’m arguing that all else won’t be equal – that we need to worry about a rise in inflation resulting in a drop in real growth, via a variety of mechanisms. That would mean the boost to NGDP from an inflationary strategy would be lower than expected, which would lead to demand for more inflation. And the tradeoff gets worse the further down that road you go.

That’s why I’m biased toward a hard-money monetary policy – a policy that focuses on price stability (and yes, I think there’s a case for some degree of “catch up” given the deflation of 2009) rather than the dual mandate that the Fed formally operates under (though they don’t seem to be following it in practice, and haven’t been for some time) of achieving price stability and full employment. But the more important focuses should be twofold: on alleviating human misery by tackling unemployment directly, and on taking action on the supply side that would improve expectations regarding future growth. And I cannot stress enough that I believe there are lots of left-wing opportunities that properly should be classified as supply-side ones. For example: liquifying the housing market by some kind of debt cram-down.

We’re now actually overdue for a revival in construction. But it isn’t happening because the housing market has seized up due to bad mortgage debt. The amount of inflation it would take to solve that problem is terrifying – we don’t want to go there. But we could tackle the debt directly. I’d call that a supply-side action, government clearing away the dead wood so the economy can grow again – just like patents are a government-granted monopoly, the process of foreclosure is a creation of the legal system. The goal should be to get transactions happening again, and that requires cleaning up the mountain of bad debt, even if it means banks take another hit. Again: the focus should be on creating conditions that encourage wealth creation, not on preserving the position of incumbents. (Some people argue Japan couldn’t get out of its liquidity trap for years because it wouldn’t inflate the economy enough – I’m inclined to agree with those who argue that Japan couldn’t get out of its liquidity trap because it wasn’t willing to take on the incumbent banks, who would not write off their enormous portfolio of bad debt.)