The Economist has a live debate going on since last week. The “motion”:
This house believes that a 2% inflation target is too low.
Brad DeLong defends the motion while Bennett McCallum argues against. Ryan Avent, of The Economist, moderates. In this second round, rebuttals, Andy Harless is a featured guest and ends up siding with DeLong. According to Andy (emphasis mine):
As a practical matter, stability is the main advantage cited for a low-inflation regime. We are comfortable with 2% inflation, but higher rates could be a slippery slope, as in the 1960s and 1970s. And even if we do not slide down the slope, its potential existence will make people nervous, be a drag on the economy and produce excessive volatility. Yet that does not seem to have been a major problem from 1983 to 1991, when the Fed successfully maintained a stable inflation rate in the 4-5% range. If economic stability is our goal, the real resilience provided by a higher inflation rate should trump the illusory comfort provided by a lower one.
The figure below shows that between 1992 and 2007 (just before the economy capsized) there was some gain in real output (RGDP) stability relative to 1983-91. Maybe that´s not the most relevant point Andy is trying to make. He´s saying that the lower inflation (1.5% to 2.5%) provides an “illusory comfort”, so it´s possible that there would be a welfare gain from a higher (4% to 5%) inflation.
The next figure shows the hatching down of inflation over the periods that Andy mentions. Given all the excitement about proposals of higher “inflation targets”, maybe inflation is considered a “reverse addiction” type of “good”. Addictive goods are those that once you get more of it you won´t settle for less. With inflation it might be that once you get less of it you just won´t settle for more, regardless…
Ryan comes to the “rescue” with some pointed remarks on the rebuttals. Notably (my emphasis):
It seems to me that the discussion is focusing on two key questions. The first is to just what extent the zero bound binds. Both debaters acknowledge that the Fed is not entirely helpless when interest rates fall to zero. Ben Bernanke has gone a bit further, suggesting that the Fed’s quantitative easing policy is not qualitatively different from interest-rate reductions. The less of a constraint zero is, the weaker the case is for a higher target. But we should also be careful not to ignore either the psychology or the political economy of the nominally independent central bank. If the zero bound does not actually bind, then some other factors have constrained monetary responses in the wake of the Great Recession and, earlier, in Japan: culture, politics, or something else. If central banks are systematically reluctant to take extraordinary steps even when additional expansion is technically possible, then the zero bound may be a real constraint, whatever the models say.
I believe these remarks show how useless the debate on inflation really is. In addition to the difficulties in defining the proper measure of inflation, it indicates that we should not discuss “symptoms” like inflation. In medicine, for example, sometimes a low fever can be a very bad sign, while a high fever may not be too worrisome; it all depends on the underlying causes of those fevers. That´s what should be discussed.
In Ryan´s remarks, identifying the “something else” is what´s important to “free” monetary policy. In a post several weeks ago, I defined the “something else” – the factor that is constraining monetary responses – as the “obsession bound”; the fact that “come hell or high water” an inflation (even if not a “target”) higher than 2% is “taboo”.
Maybe the “motion” should have been: “Instead of inflation, what should monetary policy target”?
I´m in the camp that favors targeting nominal spending, level targeting. Some at the Fed, including Bernanke in the case of Japan, have proposed a Price Level Target (PLT). In the present case it would be infinitely better than IT, but could be misguiding in the case of productivity (or real) shocks.