Why insist on searching for the “Holy Grail” (aka “NAIRU”)?

Nobody Knows NAIRU, and that’s a Problem for the Fed:

Economists surveyed by The Wall Street Journal have widely divergent views about how low the unemployment rate can get before job market slack disappears and inflation pressures emerge. It could be anywhere between 4% and 6%, according to the Wall Street Journal’s monthly survey of private sector economists.

On average, analysts said the “nonaccelerating inflation rate of unemployment,” also known as Nairu, was 5.1%. In all 69 analysts were surveyed, though all of them didn’t answer every question.

Nairu is a theoretical threshold at which the economy is in balance and inflation pressures are neither rising nor falling. A jobless rate below this chokepoint in theory would create inflation pressure. The unemployment rate was 5.7% in January, still a good distance above the average 5.1% estimate.

Federal Reserve officials are paying close attention to these estimates now because the jobless rate is falling rapidly, down from 6.6% a year ago. Fed officials estimate the unemployment rate’s long-run range–which is akin to a Nairu–is between 5.2% and 5.5%. Officials will update their projections in March. Some of them are thinking about revising their estimates down, because they see unemployment falling without much evidence of inflation pressure building.

Many Fed officials want to start raising short-term interest rates before the economy reaches a point of full employment. Economists surveyed on average estimated the jobless rate will reach the Nairu threshold by the fourth quarter. But here again, estimates vary widely. Some analysts said unemployment wouldn’t reach Nairu for another three years or more. Others said it is already there.

It really can be “anything”:

“Nairu is variable from cycle to cycle, much lower this time because of forces that keep inflation lower than typical,” said Allen Sinai of Decision Economics. His Nairu estimate is 4.3% and he doesn’t think it will be reached until the third quarter of 2016.

“Structural shifts in labor market since crisis push Nairu higher,” said Stephen Stanley of Amherst Pierpont Securities. His Nairu is 5.5% and he sees it being reached between April and June. The Fed, he said, is “already too late!”

That´s “Phillips Curve thinking”. And its useless thinking. For 60 years, scholars, gurus, pundits and whatnot have tried to “pin down” the “beast”, to no avail.

Unemployment does not determine or cause inflation. So to think that the state of the labor market will inform the Fed on its monetary policy decision is a waste of time.

The first set of charts covers the “Great Inflation” and “Volcker Adjustment” period (1966 – 1986).

Holy Grail_1

Note that overly expansionary monetary policy, as inferred from a rising trend for the NGDP growth rate (and not from interest rates) boosted inflation and did not stop unemployment rising due to real negative shocks (oil prices).

The “Volker Adjustment” which brought down the rate of nominal spending growth (in effect “tightening” monetary policy”) was accompanied by a reduction in both inflation and the unemployment rate.

The sequel was called the “Great Moderation”, where fluctuations were contained, unemployment mostly trended down and inflation was “down & out”!

Holy Grail_2

What we have now is the consequence of a major monetary policy error. NGDP was allowed to tank and has picked up only timidly (too low growth and below any reasonable trend level). Inflation remains “out” and unemployment, although falling is still high. It´s clearly not the unemployment rate that “informs” monetary policy. That should be signaled by NGDP. If only a target level for NGDP were set…

2 thoughts on “Why insist on searching for the “Holy Grail” (aka “NAIRU”)?

  1. This is a fascinating post and in some ways perhaps conflicts with my presentations that there is a Phillips Curve, now a “Phlat-line.” It may be that changes in structural impediments in the last 40 years have resulted in a less inflation prone economy.
    Still, the overall goal should be NGDPLT.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.