Although overshadowed by the financial crisis and the world recession right now, the debate over offshoring – that is, outsourcing work to foreign (often poorer) countries – seems poised to stage a comeback as a public policy concern in the not-too-distant future. Indeed, with so much protectionist talk and some protectionist action in the air, fear of offshoring may force its way back onto the policy agendas of the US and other rich countries sooner than we think.
It seems axiomatic that both the economically appropriate and the politically feasible policy responses to offshoring should differ depending on whether the share of the workforce holding offshorable jobs is, say, 2%, 25%, or 75%. In the 2% case, we should probably ignore offshoring as a detail of little consequence. In the 75% case, we should perhaps be seeking radical solutions to the manifold problems caused by massive job dislocations. But if a number nearer to 25% is more plausible, as argued here, the situation probably calls for certain marginal (and some not so marginal) policy adjustments – but certainly not panic. Thus it seems important to obtain a rough empirical handle on this number, slippery though the concept of offshorability may be.1
Several attempts have been made to estimate this fraction in recent years. Unfortunately, they have produced a wide range of estimates. Bardhan and Kroll (2003) estimated that about 11% of US jobs are offshorable. But they explicitly restricted themselves to “occupations where at least some [offshore] outsourcing has already taken place or is being planned” (p. 6) – and this was at a time when service-sector offshoring was in its infancy. Van Welsum and Vickery (2005) based their estimate of offshorability in OECD countries, about 20%, on the intensity of ICT-use by industry. McKinsey Global Institute (2005) used detailed consulting-style analysis of eight “representative sectors” in rich countries to estimate that only about 11% of worldwide private-sector service employment might potentially be offshored to developing countries within the next five years. But five years is too short a time frame. Jensen and Kletzer (2006) used geographical concentration within the US to estimate how “tradable” each occupation is. They estimated that 38% of US workers are in tradable, and therefore offshorable, occupations. Finally, I used data on job content to assess the offshorability of each of (roughly) 800 US occupations (Blinder 2009a). My “conservative,” “moderate,” and “aggressive” definitions placed 22.2%, 25.6%, and 29.0%, respectively, of all US jobs in the offshorable category. Even though these estimates are not all comparable, they present a distressingly wide range – from 11% to 38%. Can we do better?
Estimating offshorability using individual surveys
In a recent paper, Alan Krueger and I employed standard survey methods to assess the offshorability of each job in a random sample of US workers (Blinder and Krueger 2009). Moving to the individual, rather than the occupational, level is important because substantial heterogeneity exists within many occupation groups (Blinder 2009a). (For example, some accounting services are offshorable, while others are not.) In addition to improving accuracy, the other major purpose of our research was to see if we could develop a technique that could be used in standard labour force surveys, such as the Current Population Survey (CPS) in the US. We think we did.
Using a specially-designed telephone survey, which Princeton University’s Survey Research Center put in the field in June and July 2008, we experimented with three different ways to measure offshorability. In the first, professional coders used the answers to standard CPS questions to rate the offshorability of each person’s job. In the second, respondents essentially classified their own jobs by answering a single question about the need for face-to-face contact and/or physical presence on the job. (Both attributes indicate an inability to move the work offshore.) In the third, we used the answers to a series of questions on face-to-face contact, the ability to deliver one’s work from a remote location, etc. to create our own index of the offshorability of each job.
Strikingly, and surprisingly, all three measures agreed on the overall macro number – roughly speaking, 25% of US jobs are offshorable. At the micro level, the three measures agreed on the classification (offshorable or not) of a specific person’s job in 70% to 80% of all cases. In studying the detailed responses, we concluded – not surprisingly – that professional coders provided the most accurate assessments of offshorability. That is encouraging news because it implies that the Census Bureau in the US and similar agencies in other countries could easily start producing data on offshorability on a routine basis – probably without changing their survey instruments much, if at all.
In terms of major substantive results, we found that more educated workers appear to hold somewhat more offshorable jobs and that offshorability does not have many statistically significant effects on either wages or the probability of layoff. Perhaps most counter-intuitively, we found that routine work, in the sense defined by Autor et al. (2003), is no more offshorable than work that is not routine.
What might our estimate that roughly 25% of US jobs are, in principle, offshorable imply for public policy?
1. First, saying that 25% of all current US jobs are probably offshorable is not the same as predicting that all these jobs will, in fact, move offshore.
For example, even today, after decades of offshoring of manufacturing jobs, nearly 10% of American workers still work in the manufacturing sector. Virtually all of their jobs are offshorable in principle, but they have not actually gone offshore. So the requisite labour force adjustment will almost certainly be less than 25 percentage points. It will also take place gradually, over decades, as did the relative shrinkage of manufacturing employment between 1960 and today.
2. Second, the 25% estimate is roughly the same as the number of jobs – then almost exclusively in manufacturing – that were probably offshorable in the heyday of US manufacturing (around 1960).
The relative shrinkage of the manufacturing sector in the US (and elsewhere) from about 30-35% of total employment then to under 10% now was somewhat painful, especially in places where manufacturing was concentrated; it fostered some protectionist sentiment and some protectionist measures, and it induced a variety of other ill-considered policy responses. But, broadly speaking, the adjustment did not precipitate any major economic or social convulsions. This experience suggests that a similar-sized labour force adjustment can, once again, be handled by the market system – with some help from government.
3. Third, most of the policy responses that would best prepare the workforces of the rich countries for the coming wave of offshoring are conventional and not very controversial.
I refer to policies like more job retraining, bolstering the social safety net where it needs bolstering (mostly in the US rather than in Western Europe), and improving trade adjustment assistance and extending it to services. The unconventional – and therefore more controversial – policy responses may need to come in the primary and secondary educational system.
Primary and secondary schools, though well-designed to turn out factory workers for the industrial age, has not adapted very well to the information age and to the likelihood of large-scale offshoring in the service sector. I address how it might do so in Blinder (2009b). In a nutshell, I argue there that our schools will have to put more emphasis on communication skills, interpersonal contact, and creative thinking – and far less on rote memorisation.
Autor, David H., Frank Levy, and Richard J. Murnane (2003), “The Skill Content of Recent Technological Change.” Quarterly Journal of Economics, 118 (4, November): 1279–1333.
Bardhan, Ashok D., and Cynthia Kroll (2003), “The New Wave of Outsourcing.” Fisher Center for Real Estate & Urban Economics Working Paper 1103, University of California, Berkeley, Fall.
Blinder, Alan S. (2009a), “How Many US Jobs Might Be Offshorable.” World Economics, 10 (2, April-June): 41–78
Blinder, Alan S. (2009b), “Education for the Third Industrial Revolution,” forthcoming in J. Hannaway and D. Goldhaber (eds.), Creating a New Teaching Profession, Urban Institute.
Blinder, Alan S. and Alan B. Krueger (2009). “Alternative Measures of Offshorability: A Survey Approach,” NBER Working Paper 15287, August.
Jensen, J. Bradford, and Lori G. Kletzer (2006), “Tradable Services: Understanding the Scope and Impact of Services Offshoring.” In Brookings Trade Forum 2005: Offshoring White-Collar Work, ed. Lael Brainard, and Susan M. Collins, 75–134. Washington, D.C.: Brookings Institution Press.
McKinsey Global Institute (2005), The Emerging Global Labour Market, June.
van Welsum, Desirée, and Graham Vickery (2005), “Potential Offshoring of ICT-intensive Using Occupations.” DSTI Information Economy Working Paper 91, Organization for Economic Cooperation and Development (OECD), May.