The failures of central banks around the world to anticipate and control the current upsurge in inflation are now apparent to all.
What has been going on with the highly technical models which economists in these institutions build to try and explain inflation?
If we look under the bonnet, we find a debate which is almost theological in its nature. Does a concept called the Phillips curve exist? And, if it does, how do we recognise it in real life?
In the 1950s, Bill Phillips, an economics professor at the LSE, claimed he discovered a strong empirical relationship over the course of the previous century between unemployment in the UK and the increase in money wages. The higher was unemployment, the lower were wage increases, and vice versa.
This was quickly extended by economists such as the American Nobel Laureates Paul Samuelson and Robert Solow into a connection between unemployment and inflation.
Despite theoretical criticism from other economists, including Milton Friedman, the Phillips curve became deeply embedded in the central banks’ thinking about inflation.
Here’s the problem: as the years ticked by, it became harder and harder to find the relationship in the data. Across the West, we experienced a prolonged period from the mid-1990s onwards in which both inflation and unemployment were low. For a time, both of them were falling at once, in complete contradiction to the prediction of the Phillips curve.
Jerome Powell, Chair of the Federal Reserve, in 2019 went so far as to say that “the relationship between economic slack and inflation has become weaker and weaker and weaker to the point where it’s a faint heartbeat that you can hear now.”
But, by a miracle, it still continued to exist. Jim Bullard, one of Powell’s colleagues who set American monetary policy, explained that “it was the Federal Reserve who killed the Phillips curve”. The fact that the Federal Reserve had paid attention to targeting inflation successfully over the past 20 years meant that inflation had become both much lower and more stable. The link between inflation and unemployment had been broken.
Ben Broadbent, Deputy Governor for Monetary Policy at the Bank of England, made exactly the same claim in March 2020. The Phillips curve was still there, but its workings had been suspended by the success of central banks in keeping inflation low by their monetary policies.
Given the way in which inflation has taken off, this argument stretches credibility beyond its limits.
The plain fact is that the Phillips curve has been on shaky empirical foundations almost since it was first discovered. In the past couple of decades, it has been not a curve but a flat line. At other times, it has been L-shaped. It shifts around all the time.
Inflation in the West has been low in most countries for most of the past 150 years. There have been occasional surges: Germany in the 1920s, some but not all countries in the 1970s.
Over forty years ago, the Cambridge economist Bob Rowthorn argued that high inflation occurred whenever there was a lack of consensus about the distribution of income, especially between profits and wages. A wage-price-spiral would rapidly develop which could only be broken by a sharp rise in unemployment.
Rowthorn’s idea, in the current context, is worrying and suggests inflation will stick around for some time to come.