The Offshoring of Engineering: Facts, Unknowns, and Potential Implications (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

An Academic Perspective on the Globalization of Engineering Charles M. Vest When I was asked to speak at the beginning of this ses- a relatively constant production in natural science and engi- sion, I pointed out that I do not know a great deal about the neering, and China’s production is rising rapidly. If you look topic of offshoring and that everyone else in the room prob- behind those facts and separate science from engineering, ably knows more. So, my purpose today is to provide some you see that the United States continues to lead in science context along with my personal views before you begin your degrees, but not in engineering degrees (Figures 1 and 2). deep exploration of the topic. These figures have generated a great deal of debate over My main message this morning is that I wish you well in the last few years. I learned many years ago that if you want sorting out, as the workshop subtitle says, the “facts” from to write a paper that gains a high rank in the science cita- the “myths,” and in coming to a deeper understanding of tion indexes, you should make a very obvious error so that the nature of globalization, particularly for engineers and everybody will write papers correcting it, thereby driving up engineering work. This understanding is very badly needed. the ranking. The first draft of Rising Above the ­ Gathering Above all, we need guidance on how we as a nation can stop Storm (COSEPUP, 2007) quoted inaccurate statistics on thinking about globalization as a set of awful problems and Chinese and Indian degrees, contributing not only to debate begin thinking of it as a set of opportunities for America and, over the report, but to a feeding frenzy of pundits focusing on indeed, for the world. this error. Vivek Wadhwa, who is speaking later today, more substantively pointed out the problems with those statistics (Wadhwa et al., 2005). THE CURRENT SITUATION Indeed, I do not disagree very much with what Vivek has Let’s start with the basics. Where is the expertise going to say, which can be summarized in four major points. First, to be in the future? We know that natural shifts and changes all degrees are not created equal. That is absolutely true. are occurring in where engineers and scientists are being edu- There is a significant disparity in quality among degrees cated, although there is some debate about the accuracy and from various institutions in the United States, and far greater meaning of the statistics. Asia now accounts for a growing disparity among degrees from institutions in places where the share of first science and engineering degrees (NSB, 2006). higher education system is developing very rapidly, such as However, if we look at doctoral degrees, the picture is quite China and India. different, with Europe, as a collection of nations, ahead of Second, in proportion to population, there is no obvious both Asia and North America. imbalance in the numbers of engineers being produced by the If we look at first degrees in science and engineering—the United States, China, and India. After all, the United States bachelor’s level—country by country, the United States has has less than 5 percent of the world’s population, a percent- age that is expected to drop going forward, so why should   Charles M. Vest is president of the National Academy of Engineering and we expect our absolute number of engineering graduates to President Emeritus of the Massachusetts Institute of Technology. be as large as those in much larger countries? 184 

AN ACADEMIC PERSPECTIVE ON THE GLOBALIZATION OF ENGINEERING 185 180 160 140 US 120 China 100 Japan S. Korea US 80 China UK Germany 60 UK 40 20 0 00 02 01 0 96 4 3 5 6 8 9 94 95 98 99 92 93 7 97 91 8 8 8 8 8 8 8 9 20 20 20 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 FIGURE 1  First natural science degrees. Source: NSB, 2006. vest_1.eps Third, salary trends and other labor market information than those of China and India. I agree with that. I don’t know indicate that there is no shortage of engineers in the United if that situation is fleeting or will last forever. But I believe States. This may be true today, but here I must raise a caveat. we should aim at making it last forever. In fact we currently Everyone I know who has looked at current labor market have major advantages over the rest of the world in the way conditions and predicted what they mean for the future, most of our institutions educate most of our engineers. especially in engineering, has fallen on their sword. I claim No matter how you look at it, there are mixed messages no particular wisdom about the “right” number of engineers out there. Earlier this month, on October 12, 2006, The New we should be graduating. But I do think we have to be very York Times carried a story with the headline “Profit Rises careful about basing decisions on today’s marketplace condi- 53% at Infosys, a Top Indian Outsourcing Company” (Rai, tions. We really should focus on the future. 2006). A mere five days later, on October 17, there was an- Finally, the fourth point is that our universities are better other headline in the Times, “Skills Gap Hurts Technology 300 250 200 China China Japan S. Korea 150 US UK Germany 100 Japan US 50 0 00 02 01 0 96 4 3 5 6 8 94 98 99 9 95 92 93 7 97 91 8 8 8 8 8 8 8 9 20 20 20 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 FIGURE 2  First engineering degrees. Source: NSB, 2006. vest_2.eps 

186 THE OFFSHORING OF ENGINEERING Boom in India” (Sengupta, 2006). One important point in the only do universities often spawn new enterprises, but they second story was that many software and service companies also tend to play a very important role in bringing people in India report that the quality of engineering and computer together, in effect forming a centroid for the boiling and science education is sufficiently bad in their country that they perking that leads to the development of small, technology- consider only about one of every four engineering graduates based companies. employable. Another factor that is not mentioned so much is that ven- Notwithstanding these mixed signals, I believe the broad ture capitalists often prefer working in a small region where trend in graduation numbers matters and should give us they know everybody and can stay close to the companies pause, for two very simple reasons. First, we must compete they invest in as they build their networks. in the global economy, while simultaneously maintaining our American standard of living. This is a daunting challenge. Manufacturing Migration Second, I believe that prospering in the Knowledge Age requires people with knowledge, much of it relatively deep My own view is strictly middle of the road, namely, that in the areas of science and engineering. So we must closely both camps are correct in that there are some aspects of glo- monitor trends and not base decisions just on our current balization that make location less important and some that position or where we have been. The important question is make it more important. I will start with one obvious trend, what happens to the next generation. What should we do now what I will call “manufacturing migration,” or the idea that to prepare young people for their personal and professional many industries, particularly industries that manufacture lives in the future? products, may first develop in the United States, but then migrate to, say, Taiwan and then, perhaps, to Korea, to China, to Vietnam—and who knows where next? One of the THE IMPORTANCE OF LOCATION questions before you is whether this migration is inevitable. Much of the debate in the popular press and in politics, What are its pluses and minuses? and certainly in our own profession, has to do with location. Whether or not migration is inevitable—and I suspect For example, if we think very broadly about industrial R&D that it is—it is serious business. Just a few factoids here and innovation and the importance of location, there appear (COSEPUP, 2007; Palmisano, 2006): to be two camps out there, as one might expect. The first says, fundamentally, that location does not matter any more • Between 2000 and 2003, foreign firms are estimated and is going to matter even less in the future. This view has to have built 60,000 manufacturing plants in China. been popularized and communicated extremely effectively • In 2004, chemical companies closed 70 facilities in the in The World Is Flat by Thomas Friedman (2005). His basic United States and tagged 40 more for shutdown. view is that the Berlin Wall came down in 1989, but Micro- • Of the 120 major chemical plants currently under con- soft Windows went up; and one day we woke up and found struction, at least as of about two years ago, one was that $1.5 trillion worth of optical fiber had connected all of in the United States, and 50 were in China. us around the world. The interesting story behind that, of course, is that most of the businesses that laid the fiber failed, So good, bad, or indifferent, manufacturing migration is and some of their managers are sitting in jail. Nonetheless, happening. we ended up connected in a way that we could never have What does this mean for the quality and quantity of jobs imagined. Friedman goes on to say that globalization has ac- in the United States? What are we really losing and gaining? cidentally made Beijing, Bangalore, and Bethesda next-door In the brief period from the beginning of 2000 to the end of neighbors, and many jobs are now just a mouse click away 2002, it is estimated that about 400,000 jobs in IT manufac- from anywhere. turing were lost in the United States (PCAST, 2004). While There is another camp that tends to take the position that overall employment in U.S. manufacturing declined by 6 location does matter. While I am not sure Michael Porter percent between 1997 and 2001, employment in computer would appreciate me viewing him as a representative of manufacturing declined by 20 percent. one end of this discussion, because he is a very broad and thoughtful person, he has built a very powerful case over the Changes in Innovation and R&D years about the importance of regional innovation clusters in the United States and elsewhere. These clusters are group- Consider the evolution of U.S. corporate innovation and ings of industries related to one another, and their proximity R&D over the past several decades. We might think of the and interaction leads to an accumulation of human capital, 1970s as the golden age of corporate research laboratories, expertise, synergy, communication, and so forth. some of which still exist, generally in rather different forms. The importance of proximity to universities for small The key point here is that corporate R&D labs of that era not companies and corporate laboratories is well established. Not only generated new ideas for their own companies, but also 

AN ACADEMIC PERSPECTIVE ON THE GLOBALIZATION OF ENGINEERING 187 contributed enormously to the science and engineering com- Internet and the World Wide Web, and tax and trade policies. mons by virtue of publications, participation in meetings, and However, the fundamental reason is that innovators and the collaborations with universities and each other. innovation system are just reacting to the increased speed In the 1980s, due to the near-death experiences of many and complexity of business, technology, and markets. segments of U.S. industry, the R&D function was dramati- Figure 4 makes this point. If you go back to the introduc- cally transformed and was largely absorbed into product de- tion of the automobile around the beginning of the twentieth velopment. This was necessary, because it enabled a number century, it took essentially a lifetime from the first marketing of our companies not only to survive, but also to prosper, of this product to the point at which it had reached 25 percent at least for a period of time. But this trend did represent a of the U.S. population. For the telephone and then the radio, change in the U.S. innovation landscape, as corporate labs it took something on the order of a professional career for became less active as sources of non-proprietary ideas. a similar diffusion. In the case of the World Wide Web, it In the 1990s, of course, having largely turned away from took, astoundingly, only about eight years to reach a quarter longer term R&D, although there are obvious exceptions of the U.S. population. So things certainly are speeding up. here and there, many of our large companies began acquir- Stuart Feldman at IBM put together a chart confirming ing their innovation rather than carrying it onboard, by, for something we all probably know (Figure 5). In 1800, virtu- example, purchasing high-tech start-ups. ally everybody in the United States worked in agriculture, a One indicator of this trend is provided by Robert Lucky, sector that now accounts for an almost immeasurably small who plotted the affiliations of authors of papers published share of our employment. Manufacturing rose but has now in the IEEE Transactions on Communications by percentage declined, being replaced very rapidly by services, particu- (Figure 3). In 1970, the vast majority of papers were actually larly services based on information technology. written by computer scientists and engineers working for Population and development are also shifting regularly. U.S. companies, with only a small percentage authored by A paper from Goldman Sachs a year or two ago estimated, academics. This has shifted and changed in two directions. in just one decade, about 80 percent of the world’s middle- The percentage of authors from both U.S. and non-U.S. income consumers will be living in nations that we currently industry, at least in this field, has declined to almost nothing consider to be outside the industrialized world (Wilson and these days. This has been accompanied by a rise in author- Purushothaman, 2003). ship by academics, with academics outside the United States In addition, consider two facts. First, people everywhere now slightly ahead of U.S. academics. in the world are smart and capable, and when given the op- So, with migration between countries and shifts in the portunity, they will do amazing things. Second, the Internet roles of companies and universities, the innovation landscape and the World Wide Web are major democratizing forces is changing. The question is why. There are some very basic that have opened up opportunities and possibilities for people reasons—economics and wage rates, the availability of the who may not have had them before. 70 60 50 40 30 20 10 0 1970 1975 1980 1985 1990 1995 2000 2005 US Industry US University Non-US Industry Non-US University FIGURE 3  Percent authorship of papers in IEEE Transactions on Communications. Source: Lucky, 2006. Reprinted with permission. vest_3.eps 

188 THE OFFSHORING OF ENGINEERING World Wide Web Cell Phone Personal Computer Radio Telephone Automobile 0 10 20 30 40 50 60 Years to reach 25% market penetration FIGURE 4  Why everyone is in a hurry. Source: Charles M. Vest (compiled from NSB, 2006). New Business Models vest_4.eps the “globally integrated enterprise.” That is increasingly the way his company and many others are being run. Globally If we add all these trends together, we can see why people integrated enterprises are driven by globally shared technolo- are thinking about new models of conducting business. One gies and standards and linked by information technology, example is the concept of open innovation that has been and their focus is shifting from products to production. New popularized by Henry Chesbrough of the Harvard Busi- borderless strategies, management, and operations for inte- ness School. He points out that companies increasingly find grated production and value delivery are being developed. they have to reach beyond their own boundaries—perhaps So life, and innovation, today are not simple. Take the beyond their own countries, perhaps even into competing recent example of Sony and Toshiba in Japan, which excel organizations—to find the people who do particular things at conceiving, designing, and building computer games for best, where the best ideas originate. Companies have to reach young people. IBM, based in the United States, excels at out, grab those people, and somehow bring them together. designing and manufacturing sophisticated chips. Those This has stimulated debate in the business world because of companies got together and, in Austin, Texas, developed issues about licensing, partnering, joint-venturing, and so new processors designed to drive the next generation of com- forth. But, clearly, some form of openness is developing in puter games. Then, a few weeks ago, Los Alamos National our innovation system. Laboratory ordered what will probably be the world’s largest More recently—and more radically—Sam Palmisano, the supercomputer based on these chips, which were designed CEO of IBM, traces the history of corporations over the last for the gaming industry. two centuries and asserts that we are now shifting away from Those of you who have a few gray hairs will remember the model of the multinational corporation to what he calls the furor a decade or two ago in this country when someone 100 90 80 70 Services (Info) 60 Services (Other) 50 Industry (Goods) 40 Agriculture 30 20 10 0 1800 1850 1900 1950 2000 2050 FIGURE 5  Percentage of U.S. employment by sector, history and projection. Source: Stuart Feldman, 2005. Reprinted with permission. vest_5.eps 

AN ACADEMIC PERSPECTIVE ON THE GLOBALIZATION OF ENGINEERING 189 attempted to buy a Japanese Fujitsu supercomputer to use for new meta-university, a platform on which institutions all government-sponsored research. We have come a long way around the world can share teaching materials, information, when the Japanese game-chip industry ends up driving Los methodologies, and so forth. Educators can pick and choose Alamos’s most advanced computer. and shape the best material from everywhere and integrate it in ways that fit the local context. In addition, there is a growing number of experiments out there in telepresence, the THE GLOBALIZATION OF HIGHER EDUCATION ability to operate laboratories from a distance, particularly What does this all this mean for education? Let’s look at from poorer parts of the world, running expensive educa- how the research university has evolved over time and how tional laboratory equipment in wealthier states. it has globalized. We begin in the nineteenth century with The next development I will call Research University Humboldt University in Germany, which developed the Globalization, Part III. A lot of groups are beginning to model of the research university as we know it. That model work together and think about the best way to educate and was transplanted in the United States with Humboldt rather prepare our engineers for the coming century. One example directly inspiring Johns Hopkins University. In the second is a study sponsored by Continental AG that has been going half of the nineteenth century and first half of the twentieth on for about a year now called Global Engineering Excel- century, Berkeley, Stanford, Michigan, Illinois, and others lence: Educating Engineers for the 21st Century. The study began to adopt, and adapt, the research university model. involves faculty members from ETH Zurich, Georgia Tech, Institutions such as MIT, RPI, Caltech, and so forth took the MIT, Shanghai Jiao Tong University, Technical University model in a somewhat different direction. of Darmstadt, Tsinghua University, E.P.U. Sao Paulo, and Then, in the 1960s and 1970s, this model was literally the University of Tokyo. Thus excellent minds representing transplanted into India through the founding of the IITs (In- several continents, several approaches, and some of the best dian Institutes of Technology). In my view, the development engineering schools in the world can think together about of IITs over the past 50 years is one of the most amazing suc- the nature of the curriculum and the experience we owe our cess stories in the world. In 2006, the European Union started students. to establish EIT, the European Institute of Technology. We To summarize, a number of things are going on in the do not know how the EIT will develop, but it is interesting globalization of higher education. I do not think it is a matter to note that Germany and, indeed, Europe are in the process of which approach wins, but we will see which approaches of re-importing the very research university model they first succeed—propagation/emulation, overseas campuses and sent to us. facilities, multinational alliances, distance education, the So that was Research University Globalization, Part I. meta-university (or, as those in industry prefer to call it, Part II encompasses three trends. First, some institutions are “digital convergence”), or plain old-fashioned redefining of establishing a physical presence in other countries. Many our curricula and goals for globalization. U.S. universities are either opening or have opened campuses abroad, primarily to give their students a different perspec- LOOMING PROBLEMS tive and different experiences. In addition, laboratories, research facilities, medical schools, and other operations are I leave you with the thought of some real policy clouds being opened in Singapore, the Middle East, and elsewhere. looming over globalization. First, there are serious unre- Some of these are already being dissolved! solved issues about the control of “deemed exports.” A A second trend is that strategic alliances are being built deemed export, of course, means that an export license is among universities around the world. This is an old tradition required when sensitive information is shared with a non- in basic sciences, such as physics, but somewhat newer in U.S. citizen in a research context. We have to resolve this a lot of other areas, including engineering. The Cambridge- issue. Second, although great improvements have been made MIT Alliance is a good example. in visa policies, there are still problems with the issuance of The third, and perhaps most interesting and exciting visas by the U.S. government, particularly for short-term trend, is virtual presence, which tends to take two different visitors, scholars, participants in joint research and technical forms. There is a big argument going on about which is best, meetings, and so forth. although there is probably room for both. One is distance A third looming issue that must be approached carefully education, both synchronous education—for example, the is overdependence on foreign graduate students. I think one MIT-Singapore link using Internet2 to conduct classes that of the greatest absolute strengths of this country is that won- we conceive of as occurring in a big room, half of which derful, bright young men and women come to us from all is in Cambridge and half in Singapore—and asynchronous over the world, and I am fully behind as much openness as education through various Web-based tools. we can have. At the same time, we must educate more U.S. The other form of virtual presence is the open-content citizens in science and engineering and encourage them to movement, which I believe represents the emergence of a contribute at advanced levels. 

190 THE OFFSHORING OF ENGINEERING CONCLUSION Friedman, T. 2005. The World Is Flat: A Brief History of the Twenty-First Century. New York: Farrar, Straus and Giroux. I believe we are the most innovative nation on the planet, Lucky, R.C. 2006. Personal Communication. and we still have the best research universities in the world. NSB (National Science Board). 2006. Science and Engineering Indicators. Arlington, Va.: National Science Foundation. We are still the king of the hill in R&D in most fields. We Palmisano, S.J. 2006. The globally integrated enterprise. Foreign Affairs have these comparative advantages—a strong science and 85(3): 127–136. technology base and a free-market economy built on a sub- PCAST (President’s Council of Advisors on Science and Technology). 2004. strate of democracy and freedom. Sustaining the Nation’s Innovation Ecosystems: Report on Information But I leave you with this paranoid thought. The enemy I Technology Manufacturing and Competitiveness. Available online at http://www.ostp.gov/PCAST/FINALPCASTITManuf%20Report fear most is complacency. We have work to do in this coun- Package.pdf. try. I very much look forward to hearing your thoughts and Rai, S. 2006. Profit Rises 53% at Infosys, a Top Indian Outsourcing Com- what you can learn and teach us all about the real, evolving pany. New York Times, October 12. nature of globalization. Sengupta, S. 2006. Skills Gap Hurts Technology Boom in India. New York Times, October 17. Wadhwa, V., G. Gereffi, B. Rissing, K. Kalakuntla, S, Cheong, Q. Weng, REFERENCES and N. Lingamneni. 2005. Framing the Engineering Outsourcing De- bate: Placing the United States on a Level Playing Field with China and COSEPUP (Committee on Science, Engineering, and Public Policy). 2007. India. Pratt School of Engineering, Duke University. Available online at Rising Above the Gathering Storm: Energizing and Employing America http://memp.pratt.duke.edu/downloads/duke_outsourcing_2005.pdf. for a Brighter Economic Future. Washington, D.C.: The National Acad- Wilson, D., and R. Purushothaman. 2003. Dreaming with BRICS: The emies Press. Path to 2050. Global Economics Paper 99. New York: Goldman Sachs. Feldman, S. 2005. Presentation at Carnegie-Mellon University, June 29, Available online at http://www2.goldmansachs.com/insight/research/ 2005. IBM Research. reports/99.pdf.