NEW YORK, April 12 (Reuters) - More bad news for the middle class: When the economy recovers, jobs in the middle won’t. That is the conclusion of an important new study that connects a long-term trend in the labor market with the business cycle of recession and rebound.
Nir Jaimovich, an economist at Duke University, and Henry E. Siu, an economist at the University of British Columbia, take as their starting point one of the most important continuing changes in Western developed societies. That shift is what economists, most notably David Autor of the Massachusetts Institute of Technology, have called the “polarization” of the job market. Maarten Goos and Alan Manning, extending the research to Britain, have more colorfully dubbed it the dual rise of ‘”lousy and lovely” jobs.
Their point is that, thanks to technology, more and more “routine” tasks can be done by machines. The most familiar example is the increasing automation of manufacturing. But machines can now do “routine” white-collar jobs, too - things like the work that used to be performed by travel agents and much of the legal “discovery” that was done by relatively well-paid associates with expensive law degrees.
The jobs that are left are the “lovely” ones, at the top of the income distribution - white-collar jobs that cannot be done by machines, like designing computer software or structuring complex financial transactions. A lot of “lousy” jobs are not affected by the technology revolution, either - nonroutine, manual tasks like collecting the garbage or peeling and chopping onions in a restaurant kitchen.
An extensive body of economic research has shown that job polarization is happening throughout the Western developed world. It accounts for many of the social and political strains we have experienced over the past three decades, particularly the increasing divide between the people at the top and at the bottom of the economic heap, and the disappearance of those in the middle who were once both the compass and the backbone of our societies.
What’s new about Jaimovich and Siu’s work is that they have found that job polarization isn’t a slow, evolutionary process. Instead, it happens in short, sharp bursts. The middle-class frog isn’t being gradually boiled; it is being periodically grilled at a very high heat. Those spurts of change are economic downturns: Jaimovich and Siu have found that in the United States since the mid-1980s, 92 percent of job loss in middle-skill occupations has happened within 12 months of a recession.
“We think of recessions as temporary, but they lead to these permanent changes,” Siu told me. “The big puzzle about business cycles is, why have we had these jobless recoveries over the past three recessions? These jobless recoveries are because you have these middle-skilled jobs that are being wiped off the table.”
Economists are often in the business of collecting empirical evidence of the trends many of us civilians have long experienced in our daily lives. That turned out to be the case when Siu shared his research findings with his family.
“I told my father-in-law, who used to be an executive in the oil industry,” Siu said. “He said: ‘That is exactly what happened. Every vice president had a secretary, then they fired them during the recession. But after the recession we had to pair up, and two vice presidents had to share one secretary.’”
Another example may have been hinted at in the March U.S. jobs report. Those figures showed a decline of 34,000 jobs in the retail sector despite recent improvements in store sales. Some economists attributed that apparent mismatch to the power of technology, in this case e-commerce.
“That is certainly in line with our findings,” Siu said. “Salespeople are one of the prime examples of routine jobs.”
The Jaimovich-Siu paper concludes that “jobless recoveries are evident in only the three most recent recessions, and they are due entirely to jobless recoveries in routine occupations. In this group, employment never recovers beyond its trough level, nor does it come anywhere close to its pre-recession peak.”
This is, Siu told me, “a stark finding.” David E. Altig, the research director at the U.S. Federal Reserve Bank of Atlanta, who has written a blog post about the paper, echoed that view. “One of the things you certainly note is that this is the mother of all jobless recoveries,” he told me.
Siu urged me not to be too gloomy. “In the broad sweep of history, technology is good,” he reminded me. “We’ve been wrestling with this for 200 years. Remember the Luddites.”
That is an important point. All of us, even the hollowed-out middle class, would be much worse off if the Luddites had won the day and the Industrial Revolution, whose latest wave is the past three and a half decades of technological change, had never taken hold.
But it is also true that every seismic shift, including the current one, has winners and losers. And for the losers, adapting to today’s world of lousy and lovely jobs may be even harder than it was for the artisans of the Luddite era to thrive in the Machine Age.
“What might be different today is two factors,” Siu told me. “The pace of technological change is so much faster, and we live in such a complex society, that it is harder than ever to switch to a new occupation.”
All of us are awaiting an economic recovery. We should be braced for one that offers scant comfort to the middle class.