The Central Bank as a Deus ex Machina

When I saw the link to Buttonwood’s latest Economist article, subtitled Why low interest rates could also encourage saving, I was hoping that maybe I could just let this post I’ve been working on sink into the depths of my drafts folder.  Unfortunately Buttonwood’s excellent article covered related, but distinct, points, so I’ve gone ahead and reworked my own thoughts on the matter.

A few weeks ago Andy Harless asked whether umbrellas cause rain (quoting Narayana Kocherlakota in the first sentence) :

“To sum up, over the long run, a low fed funds rate must lead to consistent—but low—levels of deflation.”
OK, that is (almost) pure nonsense. It’s true that a low fed funds rate can exist, in long run equilibrium, only if people expect deflation (or if money is worthless). But the causation goes in the opposite direction. People lend at a low interest rate because they expect deflation. People carry umbrellas because they expect rain. An equilibrium with umbrellas must include a significant possibility of rain, but we don’t say that carrying umbrellas must “lead to” rain. If we take away people’s umbrellas, it will not prevent rain, and if we require people to carry umbrellas, it will not cause rain.

The causality argument that Andy Harless makes is far less clear than he implies.  Remember that the Federal Funds Rate is a policy rate and that the Federal Reserve is effectively controlling the nominal return that savers can get on short-term Treasury bills — with clear effects on the nominal returns of other short-term assets.  Thus the statement “People lend at a low interest rate because they expect deflation” reads to me as having the implication that the Fed is trying to engineer an environment where people expect deflation, which clearly is nonsense.

The Fed is forcing individuals (who are unable to borrow at these extremely low rates) to invest their savings either in highly risky assets or at extremely low nominal returns (see Bruce Krasting).  The question at hand is what is the effect of this policy on economic activity — and through this channel on price expectations.  The hope underlying the arguments proposed by Harless et al. is that these negligible returns will have the effect of discouraging saving and increasing consumption — i.e. that monetary policy can be used to address the Paradox of Thrift.

It is not, however, clear to me — and to many others — that when you engineer an environment where saving sucks, you actually convince people that it’s a good idea to consume more.  It is possible that precisely because standard safe investment vehicles are almost worthless, people become even more uncertain about the future and seek even greater precautionary savings though they may become creative about how they store those savings.  In other words, it is possible that maintaining policy rates at absurdly low levels is not only an ineffective means of addressing the Paradox of Thrift, but furthermore that this policy has the opposite effects — it may reduce economic activity and, thus, generate deflationary expectations.

Harless either disagrees with or has not thought about this possibility.  He writes:

What does it mean to say that “the real economy will improve sufficiently that the real return to safe short-term investments will normalize at its more typical positive level”? It means there will be opportunities for real investment that have more attractive expected returns. Even if the Fed’s actions lead people to increase slightly their expectations of falling prices, people will also notice these real investment opportunities and will start investing in those rather than in safe short-term investments. Or they’ll take money and spend it on consumer purchases in anticipation of continued employment. Either way, there will be more purchases made, which will tend to drive up prices, and the deflation prophecy will not fulfill itself.

Ultimately, as people notice the economy improving, they will come to expect rising rather than falling prices, no matter what the Fed does. Ultimately, the effect of having the Fed keep interest rates too low for too long will be inflation, not deflation. Of course, the Fed will notice this and then raise interest rates to slow down the economy and stop the inflation rate from rising further, so it shouldn’t be a big problem.

He (along with Kocherlakota) assumes that eventually “the real economy will improve sufficiently” for everything to come back to normal.  The question that needs to be asked is whether the Fed’s ZIRP policy will slow down or perhaps even forestall this expectation of recovery.  In my view assuming that economic circumstances have to improve is naive and similar to assuming a can-opener — policy should be designed to deal with the possibility that economic conditions get worse — afterall we didn’t have a depression and we’re nowhere near an economic nadir.

The economic circumstances from which the only way out is up, are horribly worse than what we face now.  Nobody wants to go there.  But one path to the nadir is to assume that things are so bad already that improvement is just around the corner, so there’s no need to face up to our current losses.  Thus, even though I understand that normalizing interest rates (by raising them to say 1% and possibly further once the economy has adjusted to the consequences of the first hike) may cause distress, I also think that there’s a measure of distress that can only be delayed, not avoided — and that the policy of delay will only make things worse.  1% interest rates are halfway to normal, we need to start heading in that direction or we may never get there.

In short the zero lower bound is not a good place to be, in large part because we have very little idea how the economy behaves around the zero bound.  Simply assuming that lower rates are always stimulative seems extremely naive to many of us.  Our economy is facing serious problems, but its far from clear that the central bank has the means to solve them.