Skip to main content

Federal Reserve Governor Lael Brainard, gave a much covered speech in Chicago on Monday.

Economists, Fed observers and others were looking for a signal whether the Federal Reserve would raise its policy interest rate next week at its September meeting of the Federal Open Market Committee (FOCM).

Hiding beneath the hoopla was something Brainard said that was more important for the longer-run: her comments on the Phillips curve. The Phillips curve shows an empirical relationship between inflation and the unemployment rate and was first introduced in an economics journal in 1958 in an article by economist William Phillips. 

The Phillips curve shows that a little more inflation is statistically associated with a lower unemployment rate. Policy makers jumped on this result, and, for the next 50-some years have built their econometric models and economic policies around this assumption.

The Phillips curve became the building block for decades of economic policy building that has led to years of governmental credit inflation in both monetary and fiscal policies to by off a little less inflation.

This period of credit inflation has resulted in several undesirable outcomes, a subject that we don't have room enough to discuss in this piece.

For some reason, Brainard stated that the Phillips curve has become almost horizontal since 2012.

What that means is that the unemployment level in the United States is now almost completely independent of inflation levels.

In other words, inflation can be at almost any level and it will have little or nothing to do with the country's employment level.

This is problematic for Federal Reserve officials, as Ms. Brainard reports, "for the past several decades, policymakers relied on the empirical relationship between unemployment and inflation embodied in the Phillips curve as a key guidepost for monetary policy."

That is, the "key" to making monetary policy doesn't exist, if it ever did exist.

TheStreet Recommends

Nobel prize-winning economist Milton Friedman debunked the Phillips curve in a 1968 article. He stated that the Phillips curve, if it existed, was relevant only for the short-term. It rested on the assumption that people did not change their expectations of inflation, given a newer, higher rate of inflation created by policy makers.

That is, people could be fooled. And, if the Phillips curve was to work in the longer-run, people would have to remain fooled. The whole idea of the Phillips curve was based on the assumption that people, even sophisticated money managers, could be tricked, over and over again.

Friedman said that in the longer-run the Phillips curve had to be vertical.

So, what are we facing now?

A flat Phillips curve implies that there is no relationship between inflation and the unemployment rate.

Another way of stating this is that the Phillips curve, being an empirical construct, was just a case of what the statistician calls "spurious correlation." Two series of data seem to have some relationship, when none exists.

Federal Reserve officials are either out of date, because their models do not reflect current relationships, or, they are out of touch because their was no real relationship there in the first place.

Either way, these results do not leave Federal Reserve officials in the best place. Their models are not working and, yet, they speak as if they had the world's revealed truth in their pocket.

Maybe Fed officials need to be a little more humble these days. They are dealing with an unprecedented situation and outdated models. They have this weird idea that they can settle markets by providing confident "forward guidance." 

Maybe they need to accept that they are working in a world of incomplete information, and that in such a world, no one has the absolute truth. They need to be more modest in their approach to understanding markets and how things work.

What they are doing now is neither helping the economy, nor their reputations.

This article is commentary by an independent contributor.