Jonathan Weisman, Washington Post Staff Writer, reports:

The vast majority of the 2.7 million job losses since the 2001 recession began were the result of permanent changes in the U.S. economy and are not coming back, which means the labor market will not regain strength until new positions are created in novel and dynamic economic sectors, a Federal Reserve Bank of New York study has concluded.

The findings by Erica L. Groshen, an assistant vice president at the New York Fed, and Simon Potter, a senior economist, will be sobering news to policymakers scrambling to reverse the longest hiring downturn since the Depression. The conclusions of the study, which was published last week, were underscored yesterday by two Labor Department reports showing a surge in corporate productivity even as work hours are plunging.

The Labor Department said productivity — the amount an employee produces for each hour of work — rose at a stronger-than-expected annual rate of 6.8 percent in the April-to-June quarter.

The government will release August’s unemployment and payroll levels today, and most economists expect little change since July, even though other economic indicators have improved considerably in recent weeks.

“I hope in September and October we will see some job growth,” said Dean Baker, co-director of the Center for Economic and Policy Research, who added, however, that such a bad job market “is not what we should be seeing at this point in a recovery.”

Most past recessions have been followed by a rapid recovery of jobs, as companies that laid off workers during the downturn brought them back when business picked up. But a growing body of evidence suggests that this recession and recovery are different. Large industrial companies with such cyclical employment policies account for just 21 percent of the workforce, down from 49 percent in the early 1980s, according to the Fed study.

Now, even as the economy has slowly expanded over the past 20 months, businesses have stepped up automation, sent jobs overseas and produced more while employing fewer people.

“Instead of seeing a recession as something just to weather, managers this time seem to have seen it as an opportunity or even a mandate for permanently changing the way they operate,” Groshen said.

Researchers such as Groshen say job growth will return as some industries gain importance or new ones emerge, just as telecommunications and computers drove the employment boom of the 1990s. The problem for lawmakers is that there is no way to know when or where such developments will occur.

“The job market is vastly worse right now than it was a couple of years ago,” said Gary Burtless, a labor economist at the Brookings Institution.

By historic standards, the current jobless recovery is remarkable, the New York Fed study says. During the 1991-1992 recovery that crippled George H.W. Bush’s bid for a second term as president, overall economic output grew slowly but steadily, while job growth remained flat for more than a year.

In 2002 and 2003, the economy has grown each quarter at annualized rates between 1.3 and 5 percent, but the number of payroll jobs has fallen an average of 0.4 percent every three months. Moreover, nationally, the number of hours worked per employee has remained steady, the Fed study said, pointing to “the emergence of a new kind of recovery, one driven by productivity increases rather than payroll gains.”

The numbers are striking. Industrial machinery and equipment companies lost 160,000 jobs during the 2001 recession and 106,000 during the recovery. As consumers snapped up cars, trucks and sport-utility vehicles in record numbers, makers of transportation equipment shed 62,000 jobs during the recession and 71,000 since. Securities and commodities brokers, which saw boom times in the 1990s, eliminated 44,000 positions during the recession and 25,000 during the recovery.

And, Groshen and Potter suggest, they are not coming back.

A new study by the McKinsey Global Institute, the think tank of the consulting firm McKinsey & Co., suggests why. When a firm ships a $60-an-hour software job to a $6-an-hour code writer in India, the most obvious benefit goes to the Indian. But, the McKinsey study reports, the U.S. economy receives at least two-thirds of the benefit from offshore outsourcing, compared with the third gained by the lower-wage countries receiving the jobs.

American firms and consumers enjoy reduced costs. Larger profits can be reinvested in more innovative businesses at home. New and expanding subcontractors abroad create new markets for U.S. products. And, at least theoretically, displaced U.S. workers will find new jobs in more dynamic industries.

Forrester Research Inc., a trend-analysis firm, has predicted that 3.3 million U.S. jobs will be shipped overseas by 2015, adding that those jobs are not just assembly-line work but increasingly are white-collar positions. About 200,000 service-sector jobs will be lost each year over the next decade, Forrester predicts.

The promise of eventual economic gains are cold comfort to the roughly 9 million Americans now unemployed. The Census Bureau reported this week that more than 1.3 million more Americans were living below the federal poverty line in 2002 than 2001.

Lawmakers have focused their policy proposals on tax credits, regulatory and health care changes and trade measures designed to lower the cost of domestic production, raise the relative costs of foreign competitors and get those jobs back. But both the McKinsey study and Groshen suggest that Washington’s job preservation and recovery efforts may be fruitless.

Instead, McKinsey said, policymakers should be examining new job-loss-insurance programs funded from profits generated by moving jobs offshore.

Groshen took a similar tack. “The kind of questions that policymakers need to have answers for are ‘How can we get workers displaced from these old jobs into the next-best alternative as quickly possible, through training, through [job] placement, even through moving?’ ”

2003 The Washington Post Company