Time to Discard the NAIRU Jacket?

When the University of Chicago’s Milton Friedman unveiled the concept of the "natural" rate of unemployment, in a 1968 presidential address to the American Economic Association, he let loose a rabbit that economists have been chasing ever since. In a symposium in the Journal of Economic Perspectives (Winter 1997), a number of them slow down long enough to consider whether the pursuit is still worthwhile. NAIRU, as the "natural rate" rabbit has come to be known, is an ugly acronym for "nonaccelerating inflation rate of unemployment"—which means, more simply, the rate of joblessness that is consistent with an unchanging rate of inflation. The assumption is that inflation is largely determined by the labor market and its upward pressure on wages. The implications for monetary policy, not to mention workers, are great. If the Federal Reserve Board wants to maintain a stable rate of inflation, then it should try to keep unemployment at the NAIRU level; if it wants to reduce inflation, then it should maintain unemployment above the NAIRU level.

When Friedman first hurled his thunderbolt from what passes among economists as Mount Olympus, it seemed, says Joseph Stiglitz, chairman of the President’s Council of Economic Advisers (CEA), as if the natural rate had been established by "royal edict...as another one of the universe’s invariant constants." For many years, the NAIRU was assumed to be about six percent. Today, however, it is apparent "that if a NAIRU exists, it must be changing over time," Stiglitz observes. Between August 1994 and August 1996, for instance, the unemployment rate was below six percent, so inflation should have risen; instead, as measured by the consumer price index, it fell, dropping from 2.9 to 2.6 percent. NAIRU proponents draw the conclusion that the "natural rate" has declined. Indeed, research at the CEA suggests that it has fallen by about 1.5 percentage points since its peak in the early 1980s. But uncertainties abound, Stiglitz notes.

What brought the NAIRU down? For one thing, Stiglitz says, demographic change, particularly the aging of the baby boomers. Older people are less likely to be unemployed, and so their natural rate of unemployment is lower. Also, after the post-1973 slowdown in productivity growth, workers eventually moderated their demands for increased real wages. Competition in the product and labor markets also held wages down.

The link between the NAIRU and inflation is obviously not a simple one, Stiglitz notes. But that does not mean the concept of NAIRU is worthless, he believes. By the CEA’s analysis, unemployment alone accounts for at least 20 percent of the variation in the inflation rate. Policymakers need NAIRU as a guide. "If there is no clear, systematic relation between inflation and unemployment," Stiglitz asks, "why wouldn’t policymakers simply keep trying to push unemployment lower and lower?"

That is just what they should do, argues James K. Galbraith, of the University of Texas at Austin. The NAIRU is dubious as theory, the collective attempts to estimate it have become "a professional embarrassment," and there is little empirical support for the proposition that cutting unemployment below the NAIRU promptly sparks inflation. The United States "has not experienced wage-led inflation since the 1950s, except briefly in 1973," he says. "Since 1973, average real wages have by most measures been stable or falling. All accelerations of inflation have been led by commodities, especially oil, or by import prices via devaluation. Why not therefore conclude that the economy has almost always been above the NAIRU during this time?" But Stiglitz contends that the uncertainty about the precise level of the NAIRU does not invalidate its usefulness as a guide. If the Fed action on interest rates turns out to be based on a mistaken estimate of the NAIRU, the consequences are likely to be modest, and the course can be reversed. So, after nearly 30 years, should economists stop running after the NAIRU rabbit? Not surprisingly, perhaps, the authors of the six articles in the symposium are far from consensus. But the fact that two articles, and, to an extent a third one, are, in Stiglitz’s words, "openly hostile" to the concept of NAIRU, suggests, at the very least, a growing impatience with the elusiveness of the quarry.