Business

How Low Can Unemployment Really Go? Economists Have No Idea

Here are two things most economists can agree upon: They want an economy where everyone who seeks a job can get one. Yet for the economy to be dynamic, some people will always be unemployed, at least temporarily as they move between jobs.

Posted Updated
How Low Can Unemployment Really Go? Economists Have No Idea
By
NEIL IRWIN
, New York Times

Here are two things most economists can agree upon: They want an economy where everyone who seeks a job can get one. Yet for the economy to be dynamic, some people will always be unemployed, at least temporarily as they move between jobs.

There exists, in theory at least, some magic number for the unemployment rate that keeps those priorities in perfect balance, a bare minimum level of joblessness that makes room for people to move around yet ensures that nearly everyone who wants to can find work without inflation bubbling up. Economists, as they are prone to do, have created an acronym for it: NAIRU, or the non-accelerating inflation rate of unemployment.

The problem is, it is looking more and more as if we have no idea what this magic number is — an uncertainty that has huge economic consequences.

Does the 4.1 percent jobless rate in January represent something lower than this “natural rate” of unemployment and presage damaging inflation, as mainstream estimates have long suggested? Or could it fall more — maybe a lot more — putting more people to work without negative side effects?

The new Federal Reserve chairman, Jerome Powell, showed the level of uncertainty facing top policymakers in congressional testimony on Tuesday when he said that he believes the economy is at full employment when the jobless rate is in the “low 4’s” — that is, at current levels, “but what that really means is it could be 5 and it could be 3.5.”

It might not seem like a huge range, but at 3.5 percent joblessness, 2.4 million more people would be working rather than looking for a job unsuccessfully. And that’s before accounting for some of the other benefits that a very low jobless rate might bring, like higher wages and more opportunities for those who have the hardest time finding jobs, like former prisoners and those who have recovered from drug addiction.

The stakes are huge. Just five years ago, as discussion swirled about whether there had been a rise of “structural unemployment” caused by a mismatch between the skills workers had and those the job market needed, the Congressional Budget Office estimated that NAIRU was 5.5 percent, and Fed leaders’ forecasts were in the same ballpark.

At the time, February 2013, the jobless rate was 7.7 percent. If the Fed had stuck to those higher estimates of NAIRU and raised interest rates to try to head off inflation, millions of Americans who are now working would have been consigned to unemployment for no good reason.

Indeed, lately, some of the very scholars and policymakers who once put NAIRU at the center of their economic analysis have lost faith that they have a good handle on it at all. The number is shaped by such factors as the technology people use to match up with new employers and people’s willingness to relocate to find work. But like many concepts in economics, it can’t be calculated directly. Rather, it must be inferred based on what is happening to other variables, which inevitably involves guesswork, assumptions and the use of historical data in a changing world.

“It’s not a terribly useful tool right now,” said Alan Blinder, a Princeton economist and former vice chairman of the Federal Reserve. “For it to be useful you have to have at least a little confidence you know the number. You don’t need to know it to two decimal places, but within a reasonable range. If your range is 2.5 to 7, that doesn’t tell you anything.”

In the mid-1990s, when Blinder was at the Fed, he and Janet Yellen, then a Fed governor, tried to persuade the chairman Alan Greenspan that interest rate increases were needed because the unemployment rate was quickly falling below estimates of NAIRU in the 6 percent and higher range.

Greenspan won the argument, as he almost always did at the Fed in that era. And with hindsight it seems he was correct. Unemployment kept falling through the late 1990s, and reached as low as 3.8 percent in the spring of 2000, without evident flares of inflation pressure.

Indeed, the U.S. jobless rate has now fallen below 4.5 percent just three times since 1970 — the late 1990s, 2006-07, and in the last year. In none of those times did inflation flare up (and in the current expansion, it hasn’t flared up yet). Looking abroad is instructive, too. Each country has a distinct labor market that probably implies a different natural rate of unemployment. But it is striking that in Germany and Japan, joblessness levels have fallen in the last couple of years below estimates that officials in those countries have generally considered their NAIRU — without inflation taking hold.

That tells some officials here, charged with making policy in 2018, that the sensible strategy is to see just how low joblessness can go without causing problems.

“We have to have humility” about where the natural rate of unemployment really is, said Neel Kashkari, the president of the Federal Reserve Bank of Minneapolis. “Ultimately we’re trying to assess supply and demand in the market for labor, and we can start by looking at the price. The price of labor has not been growing very quickly, and we’ve been seeing job creation far in excess of what we need to keep up with labor force growth.”

In other words, the fact that the United States keeps adding jobs at a healthy clip, and that wages are rising only gradually, is itself evidence that unemployment can be allowed to drift lower than historical data might suggest. One reason, Kashkari theorizes, may be greater credibility that the Fed and other central banks won’t allow inflation to take off as it did back in the 1970s.

Meanwhile, if the unemployment rate continues drifting down and testing the lower limits of NAIRU further, 2018 may turn out to be the first time in nearly two decades to test just what good things might result from a Lycra-tight labor market. As employers find good workers to be more and more scarce, might they be willing to hire less qualified employees and train them, enhancing those workers’ long-term earning potential? Will they be more willing to take a chance on people who have been unemployed for years or have a checkered past?

Could they be more willing to invest in some of the cities that have been left behind by the expansion for the last nine years? After all, 33 metropolitan areas in the United States still have a jobless rate over 6 percent — and four, including Ocean City, New Jersey, and El Centro, California, have double-digit rates.

“No social program can substitute for a full-employment economy,” said Isabel Sawhill, a senior fellow at the Brookings Institution who studies issues around poverty and inequality. “If a job is the best anti-poverty strategy and we’re not producing enough jobs because of a misunderstanding of the macroeconomy and a failure to manage it in the best way, we are missing a huge opportunity.”

In the 1960s and 1970s, central bankers’ reluctance to tolerate any unemployment really did fuel a spiral of wages and prices that created years of high inflation and economic weakness. That was a long time ago, though, and in that time there have been huge changes in everything from how people find a job to the power of unions.

As Powell begins his term as Fed chairman, the question is how much he is willing to test the lower limits of unemployment. Is this more like the 1970s, or more like the 1990s?

Nobody said guiding a $19 trillion economy would be easy. Yet the fate of millions depends on Powell and his Fed colleagues to get the right answer.

Copyright 2024 New York Times News Service. All rights reserved.