Naturally

One of the best ways to lose a lot of money or mess up the economy, depending on your day job, over the past several years has been to obsess about the natural rate of interest.

As a theoretical concept, the usefulness of the natural rate is pretty dubious.  But the real mischief comes from the term itself, which encourages the view that it is possible and perhaps desirable for the price of money (or really short-term debt) to be set without intervention by humans.

That view is incorrect. What is worse is that the people who hold it are not even aware they are making a judgment call. So not only are they wrong. They are confident too.

Before getting to the unfortunate behavioral psychology of the natural rate, let me comment on the technical concept, even though it is secondary, to me.  The natural or Wicksellian rate of interest is sometimes  (e.g. by Yellen) defined as the rate that would hold the economy at full employment and the target inflation rate in the absence of shocks.

There are three things to point out about this construct:

  • In reality, the economy is always experiencing shocks and is away from macro equilibrium. So the natural rate does not have much influence of the funds rate the Fed actually sets, as we have clearly seen.  The natural rate is more for making speeches about.
  • Just because we can refer to something does not mean that the thing exists. In this case, we can define what the natural rate is in terms of the unlikely-ever-to-exist conditions under which we might observe it.  But it does not necessarily follow from this that a manifestation can be found out there in the world.  People have a bad habit of falling for essences, which I think would be a good subject for a book, not of philosophy but of psychology.
  • Econometric estimates of this thing that might or might not exist have a standard error of about +/- two percentage points. The confidence interval, then, is far too wide for the natural rate to be used as a guide to monetary policy, as Yellen emphasized in her December speech to the Economic Club of Washington.  Speculating on the level of the natural rate would seem to be dominated by just looking out the window and observing the state of the economy and then speculating on how that view might change if the funds rate were moved by 25, 50 or 100 basis points.

The problem with the natural rate of interest, though, is much more with the term itself than with how economists use it.  The five or six people reading this blog – including that guy apparently in Brazil – might not be representative. But when most people hear the term “natural” rate, their instinct is not to go to Wicksell or to Yellen’s take. Rather, their reaction is much more visceral.  What the vast majority of people hear is that there is such a thing as a short-term interest rate that would prevail absent “intervention” or “distortion” by human intention.

How that interest rate might be determined is never made quite clear, because the whole idea is muddled.  But we can identify several vague intuitions. For some people, “natural” is what might obtain if the Fed followed a money supply growth rate rule.  For the truly benighted, it is what a gold standard might deliver.

For too many, the natural rate is the average that prevailed while their father was working or whatever their own personal bias on this or that political question requires or whatever is needed to explain away having missed the tripling of the stock market while the Fed was aiming that way.  This whole rally is just propped up by “fake money” and interest rates set at “unnatural levels!”

What these guys have not internalized is that money (the supply of which has a dominant influence on the short-term interest rate) does not occur in “nature.” It is a social convention or human artifice, which means that things associated with it are unavoidably artificial.  The search then for a “natural”, in this behavioral sense, rate of interest is not just impractical and hugely unprofitable. It is also utterly impossible as a matter of logic and semantics.