2% inflation changes everything
By Scott Sumner
One can think of the past 100 years of US history as featuring three macro regimes:
1. Commodity money, with alternating periods of rising and falling prices.
2. Unanchored fiat money, with rising and falling inflation rates.
3. 2% inflation targeting.
Both transitions occurred gradually, so it’s not possible to assign precise dates, but 1940 and the early 1990s are plausible estimates. Here is the CPI inflation rate since 1913:
The Fed has a flexible inflation target, where it does not try to achieve precisely 2% inflation each year. In addition, they target the PCE price index, not the CPI. Thus if oil prices rise sharply, they generally allow the CPI to rise modestly above 2%, for a time. Here’s the core PCE inflation rate, but only since 1930:
Over the last 25 years, core PCE inflation has fluctuated in a tight range, from a high of 2.28% at the peak of the housing boom (2006) to a low of 1.16% in the depths of the Great Recession (2009.) Even that overstates the actual variation in non-food/energy prices, as core inflation is still somewhat affected by oil prices (think airfares, freight charges, etc.)
Many people focus on how the Fed often misses its 2% target, or treats it as a ceiling, and those are important issues. But doing so risks overlooking more important trends. Thus if the Fed were to treat 2% as an inflation ceiling, one might think of them as actually targeting inflation at something like 1.6% or 1.8%. I’m not sure that’s true, but even a 1.6% inflation target would have important implications.
The transition away from commodity money that begins to show up in the data after 1940 was essentially complete by 1968, at which time macroeconomics had to be reinvented. With fluctuating inflation expectations, we needed new ideas such as the Natural Rate Hypothesis and the Taylor Principle, which were not required under a commodity money regime with roughly zero percent expected inflation.
The current regime of 2% inflation targeting calls for another reinvention of macro. I’m not sure people have fully grasped how much everything changes under inflation targeting. Consider this recent comment by Tyler Cowen:
Finally, if the carbon tax is revenue-neutral, just sending money to everyone (in what proportions?) doesn’t give them anything in return as measured by real resources. Maybe it would give Jay Powell a slight headache, however, since he and others at the Fed would have to decide whether and how to do an offset, or not.
I agree with Tyler that it’s unwise to rebate carbon tax revenue to the public (I’d use it for deficit reduction.) I also agree that the Fed might need to consider how to offset this policy. (Although a combination of tax and rebate would probably not have much impact on demand, requiring only a small offset.) Where I disagree is the claim that “the Fed would have to decide whether” to offset. Perhaps it’s a throwaway line, but I see it as the type of thinking that is now outdated.
Suppose a big rebate program is announced, and suppose it would materially boost AD if not offset. How will the Fed react? There is almost no chance that the next FOMC meeting would feature Fed officials saying “because of the rebate, we should shift our target from 2% inflation to 2.2% inflation”. Is there a chance that they would nonetheless do exactly that, but keep it secret? I suppose anything is possible, but I don’t see any evidence that they’d even wish to do this. That’s certainly not how they act; they are frequently nudging rates up and down to try to keep inflation close to 2%.
So if I am right, why is this not better accepted? I believe that many people put too much weight on the fact that inflation targeting isn’t perfect, and that the Fed rarely achieves exactly 2% inflation. Maybe they are even slightly biased toward “lowflation”. But that fact has little bearing on the question of whether the Fed would try to offset any particular demand shock, rather it suggests their current tool kit is far from perfect. The Fed has moved well beyond the “whether” question; now it’s all about the “how” question.
In this new world of inflation targeting it is still possible that a fiscal shock (or some other sort of demand shock) will impact aggregate spending. Fed offset is far from perfect. Rather, what’s new about the post-1990 world is that fiscal shocks are no longer expected to have any specific impact. It’s equally likely that the Fed will do too much to offset, as too little. The new baseline assumption is no effect on average, not in every single case.
This new 2% inflation world has also proved challenging for me. My toolkit is mostly monetarist, ideally suit for the high and volatile inflation rates seen all over the world between 1950-1990. Ideas like the Fisher effect and the Natural Rate Hypothesis are (or were) my primary tools for thinking about macro issues. And now we live in a world where the Fisher effect has virtually gone away, and the Natural Rate Hypothesis seems almost useless. Inflation expectations are stuck at roughly 2%. It’s increasingly hard for me to find useful things to say our 21st century economy.
But that’s not because the world has changed in any “structural” sense, rather that the policy regime has changed. When I was young, idealistic reformers thought the Phillips Curve was structural, and could be exploited to create millions of jobs for unemployed workers. Later we discovered that this would not work. Robert Lucas showed that structural parameters that look stable (like the Phillips Curve) are not reliable when the policy regime changes.
Recently I see more and more idealistic young people (who don’t recall the Great Inflation) recommending policy initiatives that are based on the notion that “inflation is dead”, not worth worrying about anymore. Yes, inflation is currently dead (or at least low and stable), but that’s only because we are not trying to take advantage of low inflation expectations by using monetary and fiscal policy to finance “populist” policies. It is not because demographics, or global competition, or technological change are creating a structurally low inflation rate.
PS. For the purposes of this post, I assume that fiscal shocks have little impact on aggregate supply. In most cases, I believe that’s a reasonable assumption. The recent corporate tax cut, however, probably slightly boosted AS.