SPECIAL COVERAGE: UNDERSTANDING OUTSOURCING

APRIL 12, 2004

Behind Outsourcing Debate:
Surprisingly Few Hard Numbers

Counting Jobs Moving Abroad
Is a Complicated Task;
It Has Some Benefits, Too

By JON E. HILSENRATH
Staff Reporter of THE WALL STREET JOURNAL

Desperate to cut costs in its struggling Internet-equipment business, executives at Infineon Technologies AG decided last year to eliminate 40 high-paying engineering jobs at its San Jose research facility and transfer the work to India.

At about the same time, the German semiconductor company started adding about 150 engineers and other white-collar workers at its operations in Cary, N.C., and Burlington, Vt., catering to a different set of customers.

Is Infineon contributing to the movement of American jobs overseas? "In my little world, it is so difficult to reconcile the numbers," says Robert LeFort, president of Infineon's North American operation.

Infineon's global job dance spotlights a fundamental question in the debate about the shift of U.S. jobs abroad. Just how do you count the number of jobs that are gained or lost as companies shuffle their operations around the world? While Infineon workers in San Jose are feeling the pinch from outsourcing, those in Cary or Burlington are overlooked beneficiaries of insourcing, or employment created by foreign investment in the U.S.

The economic picture is further complicated because there are clear -- though hard to pinpoint -- benefits to moving operations overseas, such as cheaper goods for U.S. consumers. And those benefits are easily obscured by stark examples of job loss.

The government doesn't keep count of jobs leaving the country, and the statistics available on outsourcing are sketchy. Election-year politics is only turning up the volume on the debate. Though U.S. companies have been shifting jobs overseas for decades, the latest wave has been especially scary to some because it includes well-paying white-collar jobs, not just factory workers.

The shifting of jobs is becoming a hot-button issue overseas too. Infineon's chief executive officer, Ulrich Schumacher, stepped down last month after clashing with unions and board members over plans to move jobs around the globe and other matters.

The actual number of jobs lost to outsourcing and its impact are a lot less clear than the politicians and media jumping on the issue acknowledge. Many economists estimate that roughly 100,000 white-collar jobs migrate overseas each year. That is a substantial number, though actually relatively small when measured against the size of the labor market and job losses that occur for others reasons.

For a sense of the confusion over the outsourcing numbers, consider a set of oft-cited estimates from John McCarthy, a researcher at Forrester Research Inc. in Cambridge, Mass.

In April 2002, Mr. McCarthy traveled to India in an effort to develop and sell research about offshore outsourcing. He left impressed with the country's ability to win contracts from American companies for white-collar work, such as processing insurance claims. When he returned to his office, he gathered newspaper clippings and pored through Labor Department statistics on 505 white-collar occupations. Based on his own assumptions about the vulnerability of various job categories to outsourcing, he made an educated guess about how many jobs would be shipped offshore by 2015.

His number -- 3.3 million jobs representing $136 billion in wages -- fed a growing media and political storm. BusinessWeek highlighted the number in a February 2003 cover story. The Wall Street Journal has referred to it at least five times. Lou Dobbs, a CNN business-news anchor, has made mention of the numbers on several occasions in his criticisms of businesses that move jobs overseas.

Sen. John Kerry, the likely Democratic presidential candidate, cited the Forrester research last November when he introduced legislation to regulate the call-center industry, noting in a press release, "this is 2% of the entire work force." Tom Daschle, the Senate Minority Leader, latched onto Mr. McCarthy's numbers when Democrats introduced legislation in February requiring companies to file disclosures when they send jobs overseas.

Mr. McCarthy now says his numbers were hyped and that it "makes me a little mad." He says the projected loss of jobs and income will occur over a number of years, mostly later in the decade. To date, he says, the actual number of white-collar jobs that have moved offshore is less than 300,000. That equals only about 0.2% of the total job market in any given year.

"I'm in awe that 18 months later I'm still getting five calls a day" about the report, says Mr. McCarthy. He refers to the increased attention on Indian companies as "this call center baloney."

Another widely cited estimate, produced last year by International Data Corp., a market-research company in Framingham, Mass., now appears to be flawed. Last year, an IDC analyst surveyed eight executives at technology-service companies and estimated that 23% of all white-collar tech jobs will be filled offshore by 2007, up from 5% this year. The statistics were published by the Associated Press, several midsize daily newspapers, the Los Angeles Times and the Journal, as well as trade publications.

Michael Shirer, an IDC spokesman, says the methodology in the report was "a little wobbly" and the result probably an overestimate. "We're working on a more rigorous number," he says. IDC says it will complete the new research by the middle of the year.

"There is a great deal of partial telling of the story," says Jitendra Singh, a management professor at the University of Pennsylvania's Wharton School, who has studied outsourcing. "It is understandable given the political season that we are in."

The great unknown is how much outsourcing will accelerate in the years ahead. While uncertain about absolute numbers, many economists agree that it probably will pick up. A March survey of 216 U.S. chief financial officers found that 27% planned to send more work offshore in the coming year. Among companies that already employed workers offshore, 61% said they expect to increase the offshore component, according to the survey by Duke University's Fuqua School of Business and Financial Executives International, an association of finance executives. Mark Zandi, an economist with Economy.com, a research firm in West Chester, Pa., projects that the amount of white collar and manufacturing work that is sent offshore will increase from about 300,000 jobs per year today to about 600,000 jobs per year by the end of the decade. "At this point, it is accelerating," said Mr. Zandi.

Even the highest estimates of job losses to outsourcing are small compared with the gross number of jobs lost in a given year. An average of 15 million jobs were eliminated annually in the U.S. over the past decade, said Ben Bernanke, a Federal Reserve Governor, in a recent speech. But those lost jobs typically are offset by the creation of new jobs in a labor market remarkable for its high level of churn.

Economists aren't free of the biases that accompany such debates. Most were reared on the theories of David Ricardo, a 19th century economist who laid out the principles of free trade. Mr. Ricardo believed that countries should specialize in areas in which they were relatively more advantaged than their international trading partners. He argued that when countries lowered trade barriers, everyone would benefit because they would be able to buy and produce goods more cheaply.

In political terms, it's easy to see why the outsourcing debate is dominated by critics of moving jobs overseas. While the cost to individuals who lose their jobs is obvious, the benefits of outsourcing are hard to define.

In the 1980s and '90s, two-thirds of workers who lost jobs in manufacturing industries hit by overseas competition earned less on their next job, according to a study by Lori Kletzer, an economist at the University of California Santa Cruz. A quarter of workers who lost their jobs and were re-employed saw income fall 30% or more.

The benefits of outsourcing, such as lower prices for goods and services and increased exports to fast-growing countries, are less tangible. The U.S. may be sending more jobs to India, but it also is receiving something in return. For example, U.S. educational institutions collected $1.2 billion from Indian nationals in 2002, six times the amount received from British students, according to the Commerce Department.

One method analysts have used to estimate white-collar job loss has been to look at job growth in countries grabbing U.S. business. India's National Association of Software and Service Companies estimates that between March 2000 and March 2004, employment of workers such as software developers and call-center operators, who serve clients outside India, increased by 353,000 to 505,000. About 70%, or 247,000, of those additional workers were serving clients in the U.S., estimates Sunil Mehta, a vice president at the association.

It's impossible to know if all those new Indian workers specifically replaced jobs in the U.S., suggesting that outsourcing's impact could be lower than even these numbers suggest. Some of the workers could have been hired as part of an overseas expansion or to fill positions that never existed in the U.S.

Other countries winning offshore work have seen much smaller growth in employment from offering business services to U.S. companies. In Ireland, the number of those jobs created by U.S. multinationals increased by just 2,277 between 2000 and 2002, according to the Industrial Development Agency of Ireland.

Other countries are growing fast but from a small base. In the Philippines, the number of people doing back-office work for non-Philippine companies totaled just 39,500 in 2003, compared with 25,000 a year earlier, according to SPI Technologies Inc., a Philippine company that provides business services such as call centers.

Ravi Aron, a Wharton School professor, calculated the impact of offshore outsourcing in a different way and came up with a number similar to that of other analysts. Noting the change in revenues of firms providing services mostly to U.S companies, he believes that between 2000 and 2004, about 440,000 U.S. white-collar jobs were lost to outsourcing in India and other countries.

It's easy to find examples of industries that on the surface look as though they have been hit by the latest wave of outsourcing. Between 2000 and 2003, employment in the U.S. among computer programmers, as defined by the Bureau of Labor Statistics, fell 182,000 to 563,000. The number of airline reservation agents fell 35,000 to 179,000 and tax preparation experts fell 3,000 to 91,000.

But many factors have caused these declines. They include the technology boom's sudden collapse in 2001, the subsequent recession and hesitant recovery, and, perhaps most notably, improvement in worker productivity in the U.S. that's allowing companies to produce more with less.

"There is an enormous overlap between the kinds of jobs that are being automated and the kinds of jobs that are going offshore," says Frank Levy, a Massachusetts Institute of Technology economics professor. Jobs that can be automated tend to be simple and easily describable in writing, the same criteria that often send jobs overseas.

The movement of manufacturing jobs overseas has been a larger phenomenon than white-collar jobs, many economists say. Goldman Sachs estimates that up to one million manufacturing jobs have been shifted overseas since 2001 by U.S. companies or their suppliers.

Andrew Tilton, a Goldman Sachs economist, studied relocation announcements and other sources showing employment by American companies outside the U.S. He estimates that companies have moved 300,000 to 500,000 manufacturing jobs to their own subsidiaries overseas and shifted another half a million to third parties.

In manufacturing, too, outsourcing does not seem to be the driving force behind falling employment, economists say. The jobs identified by Goldman constitute only about a third of the net decline in manufacturing employment during the past three years, as measured by the Bureau of Labor Statistics.

Joseph Carson, an economist with Alliance Capital Management LP in New York, argues that job loss in manufacturing is a global phenomenon related largely to improved productivity. He looked at employment trends in 20 large economies and found that from 1995 to 2003, 18 million jobs in the manufacturing sector were eliminated around the globe as companies found more efficient ways to work.

According to his research, China was the biggest loser, with 13 million manufacturing jobs eliminated between 1995 and 2003, mostly because of restructuring of state-run enterprises. Manufacturing employment started to pick up again in 2002. Japan lost more manufacturing jobs than the U.S. during the eight-year period. "The bigger story is improved productivity rather than outsourcing," says Mr. Carson.

Underscoring Mr. Carson's analysis are statistics kept by the Labor Department on "mass layoffs." When companies lay off 50 people or more, the Labor Department asks company officials to explain the reason. According to these figures, just 2% of workers affected by "mass layoffs" during the past five years came from companies relocating operations overseas or from import competition. That's about the same level as the mid-to-late 1990s.