For centuries, the United States has used imported material inputs to produce goods that it then exports. But the revolution of recent years in information technology (IT) has made it possible for U.S. firms to find overseas suppliers of services, allowing them to take advantage of cheaper foreign labor and different skills in that sector also. Between 1992 and 2000, this “Offshoring” of services by U.S. firms increased by 6.3 percent a year, on average, and it has continued to grow rapidly, spawning fears of job losses among U.S. workers. A recent IMF Working Paper by Mary Amiti and Shang-Jin Wei explores the offshoring phenomenon and finds evidence that a small number of jobs have, indeed, been lost in manufacturing industries. However, it also finds that offshoring has boosted labor productivity in the United States and that, partly because of this, there has been offsetting job creation elsewhere in the economy.
Although the offshoring of material inputs (see charts) still far exceeds that of services, progress in IT, as measured by the rise in Internet use worldwide and advances in digital telephone equipment and lines, has been rapidly boosting service offshoring. Industries that rely heavily on service inputs are more likely to respond to technological changes that reduce the cost of offshored services. Already, the offshoring of call centers and computer software development to India is common.
These developments have created concerns in the United States that U.S. jobs will be transferred to developing countries. A 2004 study by the University of Maryland found, for example, that support for free trade among white-collar workers in the United States with annual incomes of more than $100,000 slid from 57 percent in 1999 to 28 percent in 2004. Opposition to offshoring is evident in the passage of restrictions on offshoring by the U.S. Senate that barred companies from most federal contracts if they planned to carry out any of the work abroad (although this legislation was not passed in the House of Representatives).
Perception versus fact
Some elements of the media and some politicians, often using estimates provided by management consultants, have played up the connection between offshoring and job losses. A report by a leading IT analyst projects that offshoring will grow by 30–40 percent a year over the next five years and that the number of U.S. jobs that will be offshored will rise from 400,000 today to 3.3 million by 2015. Of this total, 8 percent of current IT jobs are forecast to move offshore over the next 12 years, according to the report. But, say Amiti and Wei, it is unclear how these numbers were derived.
In fact, the effects of offshoring on labor productivity and employment have not been systematically examined. Amiti and Wei’s study fills a gap in the literature by looking at these relationships between 1992 and 2000 in the United States.
Are U.S. jobs in danger?
To measure the effects of offshoring on U.S. jobs, Amiti and Wei use a standard labor demand framework and combine trade data with information contained in detailed input-output tables from the U.S. Bureau of Labor Statistics. To measure changes in service offshoring, the authors calculated the following shares of five categories of imported services in the inputs of manufacturing industry in 2000: business services (12 percent), finance (2.4 percent), telecommunications (1.3 percent), insurance (0.5 percent), and computing and information (0.4 percent).
Slow but steady
Material offshoring has been commonplace for decades and continues to grow.
Data: Authors’ calculations.
When manufacturing was finely disaggregated into 450 industries, Amiti and Wei found that service offshoring reduced jobs in manufacturing by about 0.4 percent. When the manufacturing sector was divided into 96 industries, however, they found that service offshoring had no significant effect on employment, indicating that there was sufficient growth in demand in other industries within these broadly defined classifications to offset any negative effects.
Offshoring can affect the demand for labor through three channels. First, there is a substitution effect through the input price of materials or services. That is, a fall in the price of imported services leads to a fall in the demand for labor if labor and imported services are substitutes. Second, if offshoring improves productivity, then firms can produce the same amount of output with fewer inputs. Hence, for a given level of output, offshoring may be expected to reduce the demand for labor. Third, offshoring can affect labor demand by affecting the demand for the product: an increase in offshoring can make a firm or a sector more efficient and competitive, increasing demand for its output and, hence, for labor.
Within a sector, offshoring can thus have a positive or negative effect on employment depending on whether a positive demand effect outweighs the negative substitution and productivity effects. For the U.S. economy as a whole, Amiti and Wei found offshoring had very little effect on aggregate employment.
Service offshoring, though still at low levels, is booming because of technological advances.
Data: Authors’ calculations.
Offshoring boosts productivity
The authors’ analysis indicates that service offshoring boosted labor productivity in U.S. manufacturing by between 3 percent and 4.5 percent between 1992 and 2000. With value added per worker in manufacturing increasing by an average of 35 percent over the sample period, the authors estimate that service offshoring accounted for 11–13 percent of the total growth in labor productivity in the manufacturing sector and that material offshoring accounted for 3–6 percent.
Offshoring can affect productivity through at least three channels: a static efficiency gain, restructuring, and learning spillovers. First, when a firm relocates relatively inefficient parts of its production process to another country that can produce them more cheaply, it can expand its output in the stages in which it has a comparative advantage. In this case, the remaining workers become more productive, on average, as a result of the change in the composition of the workforce. Second, the remaining workers may become more efficient if offshoring enables firms to restructure in a way that pushes out the technology frontier. This is more likely to occur, say Amiti and Wei, from offshoring service inputs, like computing and information, than from offshoring material inputs. Third, firms might perform activities more efficiently by importing services. A new software package, for example, can improve workers’ average productivity.
Offsetting job creation
Amiti and Wei find that service offshoring has led to the loss of a small number of U.S. jobs in manufacturing industries but that the loss has been offset by job creation elsewhere in the manufacturing sector in part because of increased productivity. Although the overall employment effect in U.S. manufacturing is negligible, there could be differential effects on skilled and unskilled workers that have not yet been studied. This was certainly true of materials offshoring. For example, a study by Robert Feenstra and Gordon Hanson in 1999 found that offshoring of material inputs accounted for 40 percent of the increase in the wage skill premium, as the unskilled intensive production stages were being relocated to other countries. It will be interesting to see if service offshoring actually leads to a decline in the skill premium, given that service providers are considered to have a higher average skill intensity than manufacturing production workers.
This article is based on IMF Working Paper No. 05/238, “Service Offshoring, Productivity, and Employment: Evidence from the United States,” by Mary Amiti and Shang-Jin Wei. Copies are available for $15.00 each from IMF Publication Services. Please see page 176 for ordering details. The full text is also available on the IMF’s website (www.imf.org).