Our very own redoubtable Benjamin Cole ventured onto Stephen Williamson’s blog to argue the case for an aggressive monetary policy after Williamson’s admiring post about Charles Plosser’s recent speech which prompted my own, less than admiring, post about the speech. As you might expect Benjamin was received less than cordially, but he stood his ground and gave as good as he got. Way to go, Benjamin!
I am not going to review the details of the exchange between Benjamin, Williamson, and an anonymous interlocutor (you can go there and read it for yourselves), but I was struck by what appears (to me at any rate) to be an inconsistency in Williamson’s position.
Replying to Benjamin’s call for a more aggressively expansionary monetary policy, Williamson replied as follows:
[Quoting Benjamin] “I am flabbergasted anyone thinks the Fed is doing enough. Really? Inflation is dead, unemployment is at 9 percent, and we are 10-15 percent below GDP growth trend.”
I am flabbergasted to know that you think there is any action the Fed can take now that would increase aggregate activity and/or increase the inflation rate.
Lest you think that that was a slip of the pen, Williamson makes the same point again, even more explicitly, in response to Benjamin’s reference to Milton Friedman’s advice to the Bank of Japan:
[Quoting Benjamin] “The Fed could announce it is targeting 7 percent nominal GDP growth, and follow Milton Friedman’s advice to Japan, and start printing money.”
The Fed can announce whatever it wants. If it can’t actually accomplish what it announces, what good is the announcement? “Printing money” would be, I assume, exchanging reserves for some Treasury debt. That will be essentially neutral – no effect on any prices or quantities.
Then when Benjamin invokes John Taylor’s 2006 advocacy that the Bank of Japan engage in QE, Williamson dismisses Taylor with the a wave of his hand, while dismissing the idea of any trade-off between inflation and unemployment as akin to a belief in the Phillips Curve, as if that were like believing that the earth is flat or that the sun revolves around the earth.
I would not call Taylor’s 2006 paper serious science. You seem to have a firm belief in a Phillips curve tradeoff. I’m not sure why.
Here is where things start to get tricky. Why does Williamson think that the Phillips Curve is so off the wall? Based on his earlier responses, you might think that it is because he rejects the notion that policymakers (i.e. the Fed) can control the rate of inflation. If you can’t control the rate of inflation, the Phillips Curve is useless. There is another possibility, however, which is that the rate of unemployment is unaffected by whatever rate of inflation the Fed chooses, because the Phillips Curve is vertical. Now either view is defensible, but, as far as I can tell, they are mutually exclusive. Either the Fed is powerless to affect the rate of inflation, or it isn’t.
Yet that is exactly where Williamson is headed, because in his next response to Benjamin, Williamson says the following:
[Quoting Benjamin] “No one suggests that inflation cannot be tamped down at some point–if we were so lucky as to have five years of robust growth, we could then start tightening the money supply.”[
It’s costly to reduce inflation once it gets going, right? What if all your inflation does not produce the robust growth you are expecting. Now everyone is complaining, not only about being unemployed, but about the high inflation rate, just as they were in 1980. Then you have to put them through the wringer again to get the high inflation out of the system. Do you think that will go over well?
So Williamson is no longer flabbergasted at the thought that the Fed might be able to increase the rate of inflation; it’s perfectly doable. The problem now is that the Phillips Curve is vertical, so even if you raise the rate of inflation, it won’t get you the reduction in unemployment that you thought you would get.
There are two points to make about this. First, on substance. The argument that a vertical Phillips Curve means that monetary policy is useless only works if we assume that the natural rate of unemployment is a constant of nature in the sense that the actual rate of unemployment must always equal the natural rate. If the actual rate of unemployment can exceed the natural rate, then monetary policy can hasten the return of the actual unemployment rate to the natural rate, though, to be sure, there is a risk of overshooting the natural rate. Opponents of using monetary policy to reduce unemployment like to suggest, but without saying so explicitly, that the natural rate of unemployment has risen sharply so that monetary policy can’t reduce unemployment below its current level. But to my knowledge, no one has come out and actually said in so many words that the natural rate of unemployment is now 9%.
Second, sliding effortlessly back and forth between an argument that says that the Fed is powerless and an argument that says that although the Fed can indeed raise the rate of inflation, doing so would be bad policy, because higher inflation would drive up unemployment to an even higher level in the endwhen inflation eventually had to be reduced does seem a tad, shall we say, ad hoc. And, as we all know, serious scientists never engage in ad hocery. (Query: but isn’t the Phillips Curve vertical? Answer: Well, it’s vertical when you increase the rate of inflation, but it’s negatively sloped when you reduce the rate of inflation. Go figure.)