Should the Fed refrain from lowering inflation all the way to 2% for fear of imposing excessive costs on the economy? Yet letting the inflation target slip has a potential cost of its own: the anchoring of inflation expectations. The Fed is not likely to play games with its long term credibility.
The April US CPI data appear to have brought some relief after three successive months of “warm” readings, but also confirmed that US inflation remains stubborn: headline CPI is running at 3.8% annualised, core at 3.6%. 3-month and 6-month annualized core CPI are running close to 4%. We’re not out of the woods yet.
Which begs the question: would it be worth it for the Fed to inflict additional damage on the economy to push inflation to 2%?
Mohamed El-Erian raised this issue in a recent piece in the Financial Times that is great food for thought.
His argument is that in view of the many structural changes that have taken place in the economy (in a more “inflationary” direction), going the extra mile to reduce inflation to target carries excessive costs, which the Fed should refrain from inflicting on the economy.
Needless to say, how the Fed answers this question – how it will set the funds rate going forward – is probably the most important near term determinant of financial market fortunes.
The intuition about structural change and inflation is a good starting point. At times of significant structural change, it may be best for monetary policy to allow a higher rate of inflation than normally. Why?
Practically, “structural change” means that (the share of) some sector in the economy grows, while another shrinks. Workers need to move from the shrinking sector to the growing sector. For that to happen, the real wage (the nominal wage adjusted for inflation) in the expanding sector needs to rise relative to the real wage in the contracting one.
If nominal wages don’t fall – workers are reluctant to take an outright pay cut – then above-target inflation may help: the real wage in the contracting sector will be eroded by higher inflation even if pay doesn’t fall in outright nominal terms.
This way, inflation “greases the wheels” of the labour market.
Now, this is not a new idea in macro.
What’s interesting is that it has recently been revisited by some leading academic macro economists. Their paper was presented at Jackson Hole in 2021, so the Fed board and FOMC members will have heard the story.
The paper concludes that when there is downward nominal wage rigidity (workers don’t want to take a pay cut) “to get relative prices right it is easier to get inflation in the expanding sectors than to get deflation in the contracting sectors, imparting an inflationary bias to optimal policy”. That is, it’s best to run inflation above target.
Here’s another point that’s more geeky (please bear with me). Many economists think that the reluctance of workers to take outright pay cuts is a reason why the famous Phillips Curve, the relationship between unemployment and wage inflation, may be non-linear. (The idea is that unemployment has to rise a lot for workers to take pay cuts, while when unemployment is very low wages will be rising rapidly.)
Why does this matter? If the Phillips Curve is linear, then lowering inflation by a bit more will increase unemployment by a little more. If it is non-linear, then lowering inflation by a bit more may increase unemployment by a lot more – i.e. the marginal cost of inflation reduction rises.
Does all this mean the Fed should let its inflation target slip – be content with “2 point something” percent inflation? Not so fast.
First off, whether the Phillips Curve is non-linear is ultimately an empirical question, in which there has been a resurgence of interest lately. But it’s probably fair to say that the jury is still out (as with so many questions in empirical macro). Nor is it clear whether Powell and the FOMC subscribe to a linear or non-linear view of the Phillips Curve (indeed, it is wise of them not to pin their colours on either side).
But the more important thing to realise is that 2.x% inflation is not a free lunch. While there may be benefits, there are costs as well, certainly in the long run.
First, the Fed has already allowed quite a persistent inflation overshoot in the sense that they could have raised rates further and/or kept financial conditions tighter (remember December?). Arguably, they are taking their time to allow inflation to settle – precisely to avoid excessive costs. All central banks have in-built flexibility in their mandate – and they are using it. Doing so works for as long as inflation expectations remain anchored.
Which brings me to my second point.
I’d venture the guess that the reason why inflation expectations have remained anchored is because – after an initial delay – the Fed raised rates decisively. This sent a signal to markets and the private sector about its determination to return inflation to target. If the Fed now were to be content with “2 point something”, sooner or later this would affect inflation expectations.
And even if it works for now, when the next inflation shock comes around, markets and the public will say: “last time the Fed didn’t bother returning inflation to target, so this time they won’t either”. Inflation expectations would drift upwards, making disinflation more difficult – and costly. Credibility is earned hard and lost easily. In short: central banks should – and do – take a long view in their cost-benefit calculations.
Last, but by no means least, letting the inflation target slip is a communications mess for a central bank. If that’s what you’re doing, how do you communicate it? (What’s your “real” target? How long till you get there? Etc.) If you don’t communicate it, people will catch up with it sooner rather than later, and then you’ll have even more explaining to do.
The Fed won’t allow slippage in the inflation target – because they can’t afford to, in the interest of their long term credibility.