No No NAIRU

Knzn continues on his campaign to defend the honor of NAIRU in a half-hearted yet curiously dogged fashion. This is his comment in rebuttal of my previous post:

From the Hippocratic oath, “Do no harm.” That’s fine if one of your choices is to do nothing. Doing nothing, by definition, is not doing harm. So, if the humors theory is disproven, and you don’t know what to do, you just say, “Go home and hope it gets better.” But it doesn’t work that way with monetary policy, because you always have to do something. You can “do nothing” to the interest rate, but then you’ll be doing something to the money supply. Or you can “do nothing” to the money supply, but then you’ll be doing something to the interest rate. It’s always possible to set up a fixed rule and set it in motion, so that you can subsequently “do nothing” except continuing to follow the rule, but then you’re initial decision to follow the rule was “something.” Until you fix such a rule, there is always at least one choice to be made, and neither of the options can be described as “neutral” or “completely atheoretical.”

It's true -- when you have what our libertarian friends sometimes derisively call "fiat money", it is the Fed who has to control it in one fashion or another. It must make the choice to loosen, tighten or stay the course. But staying the course means maintaining the interest rate at a previous level that was also ultimately determined by the Fed. Strictly speaking, there is no neutral course. Knzn seems to be suggesting that because we don't have a natural default course, even a weak, wanting theory like NAIRU can be roped in as a better alternative to flipping the coin.

That sounds somewhat reasonable, until you realize that raising the interest rates is like slamming the breaks on the economy -- to make the conscious decision to put people out of jobs. Now does it still seem like such a good idea to go forwards on such a wobbly theory? Brock made the analogy between NAIRU and Newtonian mechanics in the comments. I find this a grave insult to Newtonian mechanics. We know that Newtonian mechanics is not the whole story, but when we want to build a bridge or design an aircraft or even just kick a ball, it's there for us. NAIRU, on the other hand...

Forecasters built the NAIRU into their models. In the 1980s, the usual estimate held that inflation would begin to accelerate if unemployment fell below 6 percent (which it rarely did). In his career as a consultant, Meyer had come to rely on such an estimate for predicting the inflation rate. Success at this modest task convinced him that the concept itself would bear the test of time and changing circumstances.

By the late 1990s, circumstances had changed. Unemployment was falling below established estimates of the NAIRU--and eventually to levels rarely observed in peacetime. Yet prices remained strikingly stable. Meyer was on the Federal Reserve Board. What to think? What to do?

Meyer chose to adhere to the NAIRU concept, while conceding ground, bit by bit, on his estimate of where the threshold of hyperinflation might be: a rationale which unfortunately always made him fear the worst. As unemployment fell, he would argue that inflation was just around the corner. Therefore (although in appointing him President Clinton had clearly hoped for better), he found himself generally favoring increases in interest rates--not because inflation was rising, but only because his model told him that it soon had to.

Knzn points out that, whatever you want to say about the NAIRU, it kept us away from hyperinflation for 30 years. At this points our disagreements move from the substantive points towards value judgements on exactly how terrible hyperinflation is. I think we'd all agree that hyperinflation is baaaaad. But is it so bad that its mere spectre is enough to spook us into start putting ourselves out of business? Isn't it soon enough to go after inflation after there have been some, you know, inflation? Knzn also mentions Japan, and how they could have spared themselves some pain by realizing that their high unemployment rates portended deflation. OK, I'm impressed by that. It's a concrete example of the link between employment levels and the strength of the currency. However, the establishment of a possible link between employment levels and the strength of the currency is very far from basing our monetary policy on this mysterious number called the Non-accelerating inflation rate of Unemployment, beneath which we cannot go without inflicting massive hyperinflation a la 1970s.