Offshoring: Big Deal, or Business as Usual? by Alan S. Blinder Princeton University CEPS Working Paper No. 149 June 2007 Acknowledgements: This paper was presented at the Alvin Hansen Symposium at Harvard University, May 2, 2007. I would like to thank, without implicating, Gene Grossman and Helmut Wagner for helpful comments on an earlier draft, Lael Brainard and Alan Krueger for valuable references, and Princeton’s Center for Economic Policy Studies for financial support. I am also grateful for many useful comments from colleagues on several earlier papers on this subject. President Bush is on an eight-day tour of Asia. He’s visiting American jobs. -- David Letterman in 2006 More things are tradable than were tradable in the past, and that’s a good thing. -- Greg Mankiw in 2004 Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is long past the ocean is flat again. -- John Maynard Keynes in 1923 If there is a live intellectual debate over offshoring—which is, after all, the premise of this symposium--what is it all about? What separates those of us who worry about the effects of offshoring on the U.S. labor market from those who--like Greg Mankiw in 2004 and Jagdish Bhagwati today—see offshoring of services as just the latest expansion of international trade and, therefore, as “a good thing” for the United States-- period?1 A definition Perhaps I should start with a definition because “offshoring” is often confused with “outsourcing,” which is different. Specifically, a job is outsourced when it is contracted out of the company—presumably to another company. The country in which the job is now being done is irrelevant. So, for example, Citibank can outsource the back- office operations of its U.S. credit card business to a company in South Dakota or to one in South Korea. In the latter case, the jobs are also offshored; in the former case, they are not. Offshoring, by contrast, means moving jobs out of the country, whether or not they leave the company. Thus, Microsoft offshores (but does not outsource) jobs when it 1 There is no issue over whether or not offshoring is a positive development for the world as a whole. We all agree that it is. - 1 - moves jobs from its software laboratory in Redmond, Washington to its laboratory in Cambridge, England. But if Microsoft hires another company to provide software lab services in the United States, those jobs are outsourced but not offshored. And, of course, if Microsoft contracts with Infosys to get the work done in Bangalore, the jobs are both outsourced and offshored. The National Academy of Public Administration (2006, p. 42) suggests defining offshoring as “U.S. firms shifting service and manufacturing activities abroad to unaffiliated firms or their own affiliates.” That seems a workable definition to me. The offshoring phenomenon, which is about the location of work, does not correspond neatly any category of standard international trade data. Much U.S. service offshoring today counts as imports of services. But many U.S. service imports, e.g., tourist services consumed abroad, do not constitute offshoring because the people who do the work (in hotels, restaurants, etc.) deliver their services locally. Furthermore, some offshoring is classified as foreign direct investment (FDI), rather than as trade-- Microsoft’s building of a lab in England being a prime example. Finally, I come to the most slippery part of the concept—the one that is nearly impossible to measure. In line with the above definition, we would like to say that a U.S. company offshores jobs when it creates new jobs to serve our market, but locates them overseas. So, for example, if a U.S. manufacturer expands production by opening a factory in China, for export back to the United States, we want to say that the jobs in that factory have been offshored—even though they never existed in the United States. Measuring this particular type of offshoring requires answering counterfactual - 2 - questions—like “Would those jobs otherwise have been created in the U.S.?”--that will never be captured in official data. The debate With the definition now (hopefully!) clear, let me turn next to what the debate is not about. First, it definitely is not about the validity of the theory of comparative advantage. David Ricardo got that approximately right about two centuries ago, and I have little or nothing to add. Besides, I am not so foolish as to engage in a debate over the nuances of trade theory with the finest trade theorist of our age. Let me just state—as clearly and unequivocally as I can--that I am not claiming that the United States is about to lose comparative advantage in everything! Second, the debate is not even about the common presumption that every nation gains from trade, although that particular “theorem” does require an important footnote that I will mention shortly. Third, it is not about the comparative statics of how either social welfare or employment compares in one equilibrium state (say, after offshoring) versus another (say, before offshoring). I am willing to stipulate that, when all the dust has settled, the U.S. economy as a whole, though certainly not every American, is likely to be better off because of service offshoring. In particular, we worry-warts are not concerned that the U.S. faces a bleak future of mass secular unemployment. Thus I am happy to accept Bhagwati, Panagariya, and Srinivasan’s (2004, p. 94) assessment that offshoring “is fundamentally just a trade phenomenon; that is, subject to the usual theoretical caveats and practical responses, [it] leads to gains from trade, and its effects on jobs and wages are not qualitatively different from those of conventional trade in goods.” We will not argue about that. - 3 - What, then, is the offshoring debate about? Leaving aside the lunatic fringes (each side can name its own favorite lunatics), I believe it is about whether the offshoring of service jobs from rich countries like the United States to poor ones like India is likely to be a big deal, something I have compared to a new industrial revolution (Blinder, 2006a), or simply more business as usual—yet another routine expansion of international trade, as Bhagwati, Panagariya, and Srinivasan (2004) say. Count me as firmly in the first camp. Which makes me a worry-wart because I believe that the confluence of rapid improvements in information and communications technology (ICT) coupled with the entry of giants like China and India into the global economy is creating a situation which, while perhaps not theoretically novel, may be historically unprecedented. When I say it will be a “big deal,” I mean that offshoring will force major changes in the U.S. industrial structure, in what Americans do to earn their livings, probably in wages, almost certainly in job security and turnover, and so on. As I noted in my Foreign Affairs essay last year (Blinder, 2006a, p. 113), “Sometimes a quantitative change is so large that it brings about qualitative changes.” I suspect service offshoring will be like that. In thinking through the consequences of the confluence of ICT breakthroughs and vast new pools of labor, it is crucial to keep in mind a distinction I emphasized in Blinder (2006a) between personally-delivered services and impersonally-delivered services. Impersonal services are the ones that can be delivered electronically from afar with little or no degradation of quality—either now or sometime in the future when the technology has improved (e.g., keyboard data entry, manuscript editing). They are therefore either actually or prospectively tradable and thus potentially offshorable. Personal services, by contrast, are the ones that either cannot be delivered electronically (e.g., child care) or - 4 - that suffer severe degradation of quality when so delivered (e.g., surgery). They are therefore, for all practical purposes, non-tradable.2 We may be standing, right now, at an historical cusp. Looking backward, the crucial labor market divide has been the familiar one: between jobs that require high levels of education and jobs that do not. Roughly speaking, highly-educated workers have fared far better than poorly-educated ones for a generation. But looking forward, the more critical distinction may be the unconventional divide between personal and impersonal service jobs. And the interesting thing is that these two divisions of the workforce are almost completely unrelated. A few examples will illustrate what I mean. It seems to me unlikely that the services of either taxi drivers or brain surgeons will ever be delivered electronically by long distance. The first is a “bad job” with negligible educational requirements; the second is just the reverse. On the other hand, typing services (a low-skill job) and security analysis (a high-skill job) are already being delivered electronically from India--albeit on a small scale so far. Most physicians need not fear that their jobs will be moved offshore, but perhaps radiologists should.3 The work of policemen will not be replaced by electronic delivery, but the work of security guards who monitor sites by television might be. I could go on and on with examples like these. Briefly stated—and this is something to which I will return--the reasons why I see service offshoring as a large and potentially disruptive force for the United States (and for 2 Or so we think. Since it requires the gardener to be physically present, lawn care appears to be a quintessentially personal service. But a Chinese-American businessman recently told me about a company that is developing technology to operate a lawn mower electronically from China! 3 Levy and Yu (2006) show that offshoring of radiological services is severely restricted by regulation. So large-scale offshoring in this domain would require regulatory changes. Technology is not the limiting factor. - 5 - other rich countries) are that (a) so many Americans now earn their living providing services,4 (b) the range of services that can be delivered electronically is sure to expand as the technology improves, and (c) the number of Indian, Chinese, and other workers who are capable of providing those services will only grow over time—perhaps explosively. Does anyone disagree with any of those three propositions? That said, no one can predict the future. So why bother to debate now whether service offshoring will eventually turn out to be business as usual or a big deal? Why not just wait and see? My answer is simple: The answer matters for public policy. If this new wave of international trade constitutes no more than business as usual, then the appropriate policy response is approximately nothing. With only minor assists, laissez faire will fare just fine; the main trick is to avoid protectionism. But if offshoring will eventually amount to something approaching a new industrial revolution, then a variety of policy responses may be called for. I will return to policy responses at the end. First let me frame the intellectual debate--just to establish that we worry-warts are not all muddled thinkers.5 Some self-evident truths Since Adam Smith and Thomas Jefferson published their best work at exactly the same time, let me begin the debate by holding a few truths to be self evident. First, as just mentioned, we worriers do not question either the validity or the importance of the theory of comparative advantage. Nor do we doubt the advisability of exploiting a country’s comparative advantages rather than flailing out against those of 4 According to the BLS’s payroll survey, 83.4% of U.S. jobs in 2006 were producing services—if we count all government jobs as service jobs. 5 Bhagwati et al. entitled their 2004 paper, “The Muddles over Outsourcing.” (They meant offshoring.) - 6 - other countries. I yield to no one in my defense of free trade.6 And nothing said herein should be construed as favoring protectionism in any way. Second, I understand that trade is a two-way street. The eventual post-offshoring equilibrium cannot have the United States producing only nontradables and exporting nothing. Precisely what we will export then is a good question, for our trade patterns may have to change substantially. (More on that later.) And we need not have balanced trade in goods and services because the U.S. will surely continue to export financial assets for a long time. But America must and will remain a great exporting nation as well as a great importing nation. After all, market-driven trade patterns depend on comparative advantages, not absolute advantages. Third, comparative advantage in the modern world has relatively little to do with natural resource endowments. David Ricardo understood well why Portugal, not England, grew the grapes. These are basically the same reasons why, even today, Brazil exports bananas and Saudi Arabia exports oil. But for most of modern trade, we can mostly ignore natural endowments. Silicon Valley did not become what it is today because of a natural abundance of silicon. Nor did the U.S. develop a strong comparative advantage in aircraft because our air provides more lift. When it comes to trade in services (and much else), the skills of a country’s workforce matter much more than its climate, soil, or natural resources. It follows from this obvious insight that, in an important sense, comparative advantage is made not born. A determined and successful country can create comparative advantage for itself in industries and/or tasks where it formerly had none— as, for example, Japan did so brilliantly in automobiles and electronics. Thus, as 6 In fact, ever since my stint in government in the 1990s, many people have heard me say that my personal views on trade policy are somewhat to the right of Jagdish Bhagwati! - 7 - Bhagwati (1997) has aptly put it, modern comparative advantage (as opposed to resource- based comparative advantage) may be “kaleidoscopic,” meaning that it can move around from one country to another in response to changes in costs. Fourth, I come to the footnote mentioned earlier. Trade theorists have long understood that it is theoretically possible for a country to end up worse off when a “new entrant” country comes along and takes away its comparative advantage in one or more important industries.7 Indeed, comparative advantage does not have to be lost. As Hymans and Stafford (1995) show, the home country can become worse off if the foreign country merely gets better at producing the good that is (and remains) the home country’s comparative advantage. In the offshoring context, think about India either taking away or shrinking the United States’ former comparative advantages in a number of service occupations.8 Of course, even if lost or fading comparative advantage is the problem, protectionism is not the solution. In fact, it will probably only cause further damage— which takes us back to my first self-evident truth. However, loss of comparative advantage in major industries and occupations is a serious cause for concern in the future; and we worry-warts are worried about it. Fifth, and finally, it cannot be emphasized enough that the debate about the “threat” from offshoring is not about the nature of the eventual equilibrium position. For example, we big-dealers do not believe that the offshoring of millions of service jobs will lead to mass unemployment in the United States. However, we do foresee a massive transition as millions of workers are rudely reallocated by the market mechanism. Unfortunately, the vast majority of trade theory pertains to the analysis of full- 7 See Bhagwati (1968), Gomory and Baumol (2000), Samuelson (2004). 8 This is a prospective possibility. It recent years, the U.S. terms of trade have improved, not deteriorated. - 8 - employment equilibrium states and has little or nothing to say about either unemployment or transitions.9 Too often, economists simply label certain things as “transition costs” and then proceed to ignore them. But when it comes to a phenomenon as big as service offshoring, such a cavalier treatment strikes me as more than a trifle hypermetropic.10 In addition to job losses, it is quite likely that, by stripping away their previous immunity to foreign competition, offshoring will depress the real wages of many service workers in the U.S. who do not lose their jobs. Now, about that transition… So both my intellectual focus and my practical concerns center on the transition, not on the ultimate equilibrium state. Let us therefore pose, and attempt to answer, a series of questions, both qualitative and quantitative, about the likely nature of this transition. Here, the “truths” become less than self-evident because we are speculating about the future. I start with the hypothesis that offshoring will usher in a massive and disruptive transition—a new industrial revolution, if you will. Past industrial revolutions have changed the faces of societies, causing great dislocation before ultimately leaving those societies much better off. I expect this one to follow that same pattern. But before we reach the promised land, I suspect that we Americans will experience a nasty transition, lasting for decades, in which not just millions but tens of millions of jobs are lost to offshoring. (That’s gross, not net, losses of course.) Which brings to mind the quotation 9 See, for example, Davidson and Matusz (2000), who criticize standard trade theory for this reason. 10 Hypermetromia is the opposite of myopia. - 9 -