In a week loaded with important economic news, no piece of data will garner more justifiable attention than Friday's April employment report. The importance of that report was driven home by Chairman Bernanke's comments at last week's press conference (emphasis added):

"REPORTER: Mr. Chairman, given what you know about the pace of the economy now, what is your best guess for how soon the [Committee] needs to begin to withdraw its extraordinary stimulus for the economy.


"CHAIRMAN BERNANKE: Well, currently as the statement suggests, we are in a moderate recovery. We will be looking very carefully first to see if that recovery is indeed sustainable, as we believe it is. We will also be looking very closely at the labor market. We have seen improvement in the labor market in the first quarter relative to last year. We would like to see continued improvement, more job creation going forward. At the same time we are also looking very carefully at inflation. The other part of our mandate."


In February and March payrolls expanded by an average of 205,000 jobs each month, a pace that is probably sufficient to make progress toward reducing the still-elevated unemployment rate. But at that pace it will take about three years before we see the same number of jobs that existed as of December 2007.

The significant lag between gross domestic product recovery and employment recovery has been particularly extreme in the wake of the most recent recession, but this pattern was a characteristic of the previous two recessions as well. You know the facts: In the post-WWII recessions up to 1989, the average time it took to regain recession-generated job losses was 10 months. The recovery time expanded to 23 months and 38 months in the recoveries following the 1990–91 and 2001 recessions. And we are on track to shoot past those records this time around.

Explanations abound, but one popular belief is that the answer hides somewhere within the somewhat ambiguous phenomenon labeled "globalization." A few weeks back, David Wessel of the Wall Street Journal provided some pretty compelling facts:

"U.S. multinational corporations, the big brand-name companies that employ a fifth of all American workers, have been hiring abroad while cutting back at home, sharpening the debate over globalization's effect on the U.S. economy.


"The companies cut their work forces in the U.S. by 2.9 million during the 2000s while increasing employment overseas by 2.4 million, new data from the U.S. Commerce Department show. That's a big switch from the 1990s, when they added jobs everywhere: 4.4 million in the U.S. and 2.7 million abroad."


Two obvious questions: What jobs are we talking about, and what is the meaning of the differential in job growth? Is this a story of "offshoring"—the shifting, if you will, of jobs to foreign locales for production that still fundamentally satisfies U.S. demand? Or is it more a reflection of the different pace of growth in foreign markets relative to U.S. markets?

It's difficult to come to definitive conclusions on these questions, but we do have some information about the types of jobs that underlie the aggregate job-growth picture drawn in Wessel's statistics. Here's what we know based on data from the U.S. Bureau of Economic Analysis's International Economic Accounts from 1999 to 2008:

Combined-by-industry-us-foreign-employees

A couple of things jump out. First, among U.S. multinational employers, some industries added U.S. employees, and some shed them. On net, these corporations lost 1.903 million U.S. jobs from 1999–2008. During this same period, manufacturing multinationals in the United States lost 1.938 million jobs (see the table). Also, foreign employment in manufacturing represented less than 13 percent of U.S. employment losses and only 10 percent of the total foreign employment gains generated by those multinationals.

By industry, the largest U.S. job losses after the manufacturing industries were created by finance and insurance firms. But, as the table shows, foreign employment in these types of firms also fell. In fact, there were only two types of industries listed above—manufacturing and information—in which foreign employment by U.S. multinationals grew while U.S. employment fell.

Finally, the largest category of foreign job gains was "other industries," which breaks down as follows:

Breakdown-of-other-industries-by-foreign-employment

Sixty-nine percent of the foreign employment growth by U.S. multinationals from 1999 to 2008 was in the "other industries" category, and 87 percent of that growth was in three types of industries: retail trade; administration, support, and waste management; and accommodation of food services. Some fraction of these jobs, no doubt, reflect "offshoring" in the usual sense. But it is also true that these are types of industries that are more likely than many others to represent production for local (or domestic) demand as opposed to production for export to the United States.

We certainly don't present this information as a definitive answer to the question about the role of offshoring in the slow U.S. jobs recovery. But if you forced us to choose between global or domestic factors as the place to look for solutions as we struggle with persistent underperformance in U.S. labor markets, we'd choose the latter.

Update: At the Street Light blog, Kash Mansori looks at the issue focusing on the global distribution of sales. Different take, similar conclusion.


Photo of Dave Altig By Dave Altig
senior vice president and research director at the Atlanta Fed