Deflation Can Happen

I’m not saying it will happen, or even that it’s likely. As a matter of fact, if you offer me less than 6:1 odds, I won’t even consider a bet on deflation, even taking into account its hedge value (since deflation is likely to be associated with bad real outcomes). But what puzzles me is that James Hamilton seems unwilling to take that bet at any odds. This despite the fact that he repeatedly acknowledges the deflation that Japan was still experiencing less than a decade ago.

He makes an argument (which I will address presently) that the Fed can stop deflation, and he seems to draw from it the conclusion that the Fed will stop deflation. It seems to me he could have made essentially the same argument 15 years ago in Japan, and it would have been wrong. I guess he thinks that, since we have learned from Japan’s mistakes, we can be confident that we will avoid making the same mistakes and therefore confident that we will avoid deflation.

I expect that Japanese policymakers thought they were avoiding the mistakes the US made in the 1930s. And to some extent, they surely were. There is far from universal consensus on what exactly those mistakes were in the first place, but perhaps the single biggest mistake the US made was in using monetary policy to defend the dollar. I expect that Japanese policymakers were congratulating themselves for long ago having had the good sense to float their currency so that mistake wouldn’t happen. (In the years leading up to the deflation, they arguably made the mistake of allowing the yen to appreciate too much on its own, but that's not something they should have learned about from the 1930s.) Another commonly alleged cited mistake the US made was in not using monetary policy to offset the effect that the contraction of the banking system had on the aggregate money supply. I just pulled up some statistics on Japanese monetary aggregates, and I see that Japan definitely did not make that mistake either.

A doctrinaire Keynesian will tell you that the biggest mistake the US made in the early 1930s was in not providing an adequate fiscal stimulus. It’s clear in retrospect that Japan did make that mistake in the 1990s (and early 2000s), but at the time I expect they thought they were avoiding it. The fiscal deficits certainly looked huge enough. And in any case, during the 1990s, the preponderance of economic opinion regarding the 1930s was tending to run against the doctrinaire Keynesian explanation. Overall, avoiding the mistakes of the past is not as easy as it might sound, nor is it guaranteed to be effective, and I have less than complete confidence that US policymakers today will succeed either in doing so or in doing so effectively. Even if the Fed can stop deflation, there’s no guarantee that it will.

But can the Fed, by itself, really stop deflation? It’s a difficult question, because it depends on what policy actions the Fed is capable of taking. There are, for one thing, institutional constraints – in particular, laws – that limit the Fed’s potential actions. There also has to be some sort of “sanity constraint.” There are things that the Fed theoretically could do, but it’s impossible to imagine that it ever actually would. There are certain actions the Fed could take that would be guaranteed to stop deflation, but before the Fed would even consider such actions, it would push Congress repeatedly for more and more fiscal stimuli and accept the fact that Congress will sometimes be slow to provide those stimuli. Meanwhile, the deflation could be well underway.

Taking Professor Hamilton’s specific examples:

First, have the Federal Reserve buy up the entire outstanding debt of the U.S. Treasury, which it can do easily enough by just creating new dollars to pay for the Treasury securities.

Let’s consider this action in two parts. First, the Fed could buy up enough of every issue to push the market yield down to approximately zero. Then it could buy up the remainder of each issue. Would either of these actions stop deflation? The second, almost certainly not, because it just replaces one zero-return asset with another. The first, maybe, but I doubt it. If people expect deflation and have very little tolerance for risk, they will just hold on to most of the money, despite it’s zero nominal return, the same way they held on to the Treasuries. Really, zero is not all that much lower than the yields on Treasury notes today. OK,

Then buy up all the commercial paper anybody cares to issue.

This is where the sanity constraint comes in. On the one hand, the Fed can buy a large but limited amount of commercial paper from reputable issuers. They’re already doing that, and it’s not working, at least it’s not stopping the economic contraction or the talk of deflation.

On the other hand, the Fed can literally “buy up all the commercial paper anybody cares to issue.” That would stop deflation, I agree, but by a very odd mechanism. Anyone could start a thinly capitalized corporation, issue commercial paper, gamble the proceeds, default (unless they win), go bankrupt, and stick the Fed with the losses. It’s roughly equivalent to dropping money from a helicopter, since the money would be out there in the economy and the Fed would no longer have an offsetting asset. In general, such policies that actually create large amounts of new financial wealth, rather than exchanging one form of wealth (i.e. money) for another (i.e. bonds), can stop deflation. But creating financial wealth by deliberately lending to borrowers that are likely to default, this is way too far afield from what the Fed is supposed to do, and it’s not something the Fed ever will do intentionally. But OK:

In fact, you might as well buy up all the equities on the Tokyo Stock Exchange.

There are really two parts to such a policy: first, the Fed would create dollars and use them to buy yen; then the Fed would use those yen to buy equities. First of all, could buying yen stop deflation? Well, yes, if we could take Japanese monetary policy as exogenous: if you devalue your currency enough, it will eventually stop deflation. But more likely Japan would respond by buying dollars, and exchange rate wouldn’t change much.

What about buying equities? In any case it would be more effective to buy US equities instead of Japanese. That would stop deflation if done on a sufficiently large scale. Unlike bonds, stocks have no inherent price limit, so if the Fed were willing to pay a sufficiently high price, it could create enough new financial wealth to stop the deflation. But is the Fed allowed to buy equities? My guess is it’s not, but I may be wrong. And if Fed does buy equities, is that still “monetary policy”? The government taking an equity interest in private sector firms sounds like fiscal policy to me.

And if the Fed can buy direct interests in real goods and services (such as equities, which represent partial ownership of capital goods), why not have them stop the deflation directly by buying consumer goods? Let the Fed put huge amounts of gift certificates on the asset side of its balance sheet. That will do the trick. I’m not sure at what point you start to breach the sanity constraint, but to me gift certificates actually seem like a more reasonable asset for the Fed to hold than junk paper.

Anyhow, I’m sure that, between the Fed, the Treasury, the president, and Congress, they can eventually come up with a way to stop deflation. Whether it’s done by expanding the Fed’s role in previously unimaginable ways or by more traditional fiscal stimuli, it will eventually be done on a scale large enough to be effective. I’m confident of that, but I’m not entirely confident that we will reach that point before experiencing a decade of deflation.