Over the weekend I received a comment on my post about Clark Johnson’s new paper from someone who disputed Johnson’s assertion that it is a myth that the Fed has been following an expansionary monetary policy since the 2008 financial crisis. Here is the relevant part of the comment:
The Fed cannot fix the economy by changing “expectations” because they have no tools to follow through. Just saying that they should “change expectations” is not enough. I cannot change expectations of the whole economy, because I have no tools to follow through, the same applies to the Fed.
Even though I disagree with the commenter that the Fed cannot affect expectations, I understand and sympathize with the commenter’s skepticism that Fed could actually do so. As I have written here before, I don’t think that our current theory of fiat money explains very well how the value of a fiat money (i.e., the inverse of a comprehensively defined price level) is determined. Without such a theory, it is hard to specify the exact mechanism or channel by which a central bank can control the price level. Nevertheless, it seems clear that an essential element in that mechanism is control, though perhaps limited and imperfect, over the price-level expectations of economic agents.
WASHINGTON — The Federal Reserve made a rare promise on Tuesday to hold short-term interest rates near zero through at least the middle of 2013, in a sign that it has all but written off the chances of an expansion strong enough to drive up wages and prices. . . .
By its action, the Fed is declaring that it, too, sees little prospect of rapid growth and little risk of inflation. Its hope is that the showman’s gesture will spur investment and risk-taking by convincing markets that the cost of borrowing will not rise for at least two years.
The Fed’s statement, with its mix of grim tidings and welcome aid, contributed to wild market oscillations as investors struggled to make sense of the economy and the path ahead.
The juxtaposition of my commenter’s skepticism about the ability of the Fed to affect expectations with the news item quoted by Scott suggests to me (or, to be more precise, reinforces for me) the following observations about expectations:
- Expectations are partly autonomous, partly induced by policy rules or by policy announcements made by policy makers;
- Expectations sometimes affect outcomes;
- Expectations can therefore be self-fulfilling (referred to by Karl Popper as the Oedipus effect);
- Expectations are often contagious;
- Expectations can be cyclical (even exhibiting bubble-like characteristics);
- Expectations sometimes are, and sometimes are not, consistent with equilibrium
- There may be multiple equilibria corresponding to various sets of expectations;
- Keynes’s famous characterization (General Theory, chapter 12) of the stock market as a beauty contest has an important kernel of truth to it and does not presume that traders act irrationally;
There is no clear distinction between expectations and fundamentals, because expectations are fundamentals.
If these observations about expectations are right, then conventional rational expectations (DSGE) models, which assume a unique equilibrium determined by fundamentals, are flawed at the most basic level, because they exclude a priori the existence of multiple potential equilibria. If there are many possible equilibria, each corresponding to a particular set of expectations, economic policy can affect outcomes by altering expectations, leading to the realization of a different equilibrium from the one that would have been realized under the old set of expectations. What real business cycle theorists identify as productivity shocks could just as easily be regarded as expectational shocks, possibly induced by policy choices. Put another way, by affecting expectations, monetary policy can affect not only the expected rate of inflation, it can affect the real rate of interest, so that the standard interpretation of the Fisher equation, in which expected inflation is added to an unvarying real rate of interest, is valid only on the assumption that there is a unique real equilibrium independent of the expected rate of inflation. But if there are multiple possible equilibria, whose realizations depend on what rate of inflation is expected, the observed nominal rate is not simply the sum of expected inflation and a uniform real rate, because the real rate is not uniform with respect to the expected rate of inflation.
This is what Keynes meant when (General Theory, p. 219) he rejected the concept of a unique natural rate (which in his terminology corresponded to the Fisherian real rate), because there is a different real rate corresponding to each level of employment. In fact, it is even more complicated than that because in Keynesian terminology, the real marginal efficiency of capital may shift as the expected rate of inflation changes. But we can save that complication for another time.