Opinion diverges on the impact of outsourcing, or offshoring, service activities. Some economists blithely dismiss services as nontradable, while a recent BusinessWeek cover titled “Is your job next?” reflects broad public fears. A relative lack of data on the service sector, and trade in services in particular, has hampered dialogue on this issue. Yet understanding the size, scope, and impact of services is crucial to assessing US competitiveness and interpreting developments in the US labor market.
An important way to identify which workers are vulnerable to international trade in services is to look at services that are traded domestically. These services can be classified as potentially tradable internationally, and estimates can be made of the number of workers in tradable activities in each sector. Demographic characteristics of workers in tradable and nontradable activities and employment growth in traded and nontraded services can then be compared, and the risk of job loss and other employment outcomes for workers in tradable activities can be examined. In fact, Census data reveal a rich picture of how many workers could be affected by offshoring, what kinds of workers are most likely to be affected, and what job losses can be expected.
How many workers are potentially affected by services offshoring?
A significant share of total employment is in tradable service industries. For example, more workers are in tradable industries in the services sector than in manufacturing. The share of total employment in tradable professional services is 13.7 percent, while the share of employment in tradable manufacturing industries is 12.4 percent. Some big services sectors—education, health care, personal services, and public administration—do in fact have low shares of employment in tradable industries. However, because the services sector is much larger than the manufacturing sector, the number of workers potentially exposed to international trade in services is actually larger than the number of exposed workers in manufacturing.
Some intermediate inputs into service production might be tradable even though the service industry itself is not (think computer programming or other back office operations for the banking industry). In the aggregate, the share of these sorts of workers—in tradable occupations but in nontradable industries—is not large, at about 10 percent. However, for business and professional occupations, the share of workers in tradable occupations but nontradable industries is much larger. The typical professional occupation has about 25 percent of its employment in tradable occupations but nontradable industries. To the extent that firms can disentangle intermediate service inputs from the rest of their business, workers in these tradable occupations are vulnerable to trade, even though their industry is not tradable. Thus the industry results shown in the figure above on the share of workers potentially vulnerable to trade are understated; the typical white-collar occupation involves an activity that could be traded.
What do tradable service workers look like?
Workers in tradable sectors have higher education levels and significantly higher wages. The higher incomes are not solely a result of higher skill levels—even controlling for differences in skills, workers in tradable service activities, like professional services, earn incomes almost 20 percent higher than those earned by similar workers in nontradable activities in the same sector.
What is the employment impact?
The data for employment growth differences are not definitive, yet it appears that tradable and nontradable service activities have similar net employment growth rates, except at the lowest end of the skill distribution. Low-skill tradable industries and occupations have negative average employment growth compared with positive (though low) employment growth in nontraded, low-skill services.
Some evidence suggests that displacement rates—that is, an individual worker’s loss of a job—are higher from tradable services than from nontradable. Also, there are higher displacement rates from tradable white-collar occupations than from nontradable. However, this period includes the telecom and dot-com bust, which undoubtedly contribute to, if not drives, this result. Consistent with the characteristics of employed service workers in general, workers displaced from tradable services are more educated and earn more than workers displaced from nontradable activities. Displaced tradable service workers are more likely to find new jobs than their counterparts in nontradable activities. Nonetheless, both tradable and nontradable service activities suffer earnings losses.
The data demonstrate that a significant number of jobs are potentially at risk due to offshoring. These results highlight the need to
For more information, see Working Paper 05-9.