Speeches and Papers

The Globalization of Research and Development and Innovation

by Martin Neil Baily, Peterson Institute for International Economics

Testimony before the Committee on Science and Technology
United States House of Representatives
June 12, 2007

 


Globalization has provided many benefits to the US economy. My Peterson Institute colleagues, Gary Hufbauer et al., have estimated that the United States is a trillion dollars richer today than it would have been if there had been no reduction in trade barriers after the end of World War II.1 Many studies of productivity carried out at the McKinsey Global Institute have shown that productivity in an industry is enhanced when it is exposed to global competition, particularly competition against the world’s leaders.2 You have to compete against the best if you want to be the best. The Organization for Economic Cooperation and Development (OECD) found that openness to international trade had provided an important stimulus to growth among the member countries of that organization.3

The United States benefits from globalization because it results in lower prices for US consumers, provides greater access to new technologies and business practices from around the world, allows US companies to take advantage of economies of scale, and forces other companies to improve their own performance. One sign of the benefits of the open and competitive market in the United States is the fact that productivity growth has been strong for the past 10 years. From 1995 to 2006 output per hour in the nonfarm business sector of the US economy has been nearly 2.9 percent a year, much faster than the pace achieved for 20 years prior to 1995 and faster than most other advanced economies.

At the same time, there are legitimate concerns about the impact of globalization on Americans. There is concern over the impact of globalization on the skilled workforce and on the science and technology base of the US economy—the topic of this hearing. Strength in science and technology has been a key part of the success of the United States over its history. In addition there is concern over the huge trade and current account deficits and the slow growth of wages and incomes for lower skilled workers.


Scientific Research Has Always Been a Global Endeavor

The history of science tells us that major contributions have been made to scientific knowledge from countries and regions around the world. The United States came to the fore in scientific research during the 20 th century, relying on its great universities and taking advantage of outstanding scientists and engineers that came to the United States from the rest of the world. Today, the United States remains unquestionably the global leader in science, judged by the size and quality of its research community and on the metric of Nobel prizes.

US leadership is not unchallenged, however. Other countries are determined to build up their own scientific research and are funding research projects. What are the lessons for US policy?


Offshoring Services and Science and Technology

Historically, the United States has been a preferred location for employment in science and technology and has a robust comparative advantage in services. In 2006 the United States ran a $72 billion surplus in services trade, despite the fact that goods trade was in a huge international deficit. As part of the $72 billion services surplus, the United States ran a surplus of $35 billion in royalties and licenses, much of that coming from technology, as well as movies and other media. These figures in fact greatly understate the global revenues generated by technology activity in the United States. US- and foreign-based multinational companies draw on the technological base they have developed through research and development (R&D) and business development here in the United States and use it in operations throughout the world. The returns come back as net income to US companies.

The United States also runs a trade surplus in education reflecting the foreign students that are educated in US institutions. The only major service categories in which the United States ran a deficit were insurance and transportation.

The very large trade deficits in manufactured goods experienced by the United States have been the result largely of a value of the dollar that has made US production too expensive relative to other countries, and the dollar has also hurt US services trade. The values of the euro, the British pound, the Canadian dollar, and other currencies have adjusted upwards and this has made the United States a more competitive economy for locating production facilities and also R&D and other technology facilities. This should help to boost US employment in technology fields going forward. Some Asian currencies, notably the Chinese renminbi and the Japanese yen, remain undervalued, according to several of my Peterson Institute colleagues, and if these currencies adjust upwards in the future this will add to the desirability of the United States as a location for high technology research, as well as tradable services more broadly.

On balance, the US service sector as a whole has sustained its position as a net exporter through a challenging overall environment for trade. Many countries around the world have offshored their R&D and technology employment to the United States—pharmaceutical R&D by US and European companies in New Jersey, for example.

This is not to downplay the competitive challenges now facing the US service sector and the pressure being felt by some mid-level occupational categories in the United States. Table 1, prepared by the Peterson Institute’s Jacob Kirkegaard, shows employment in a number of computer and technology related occupations, as well as employment in lower-skilled service occupations that are subject to relocation off shore. The upper half of the table reveals that call center type occupations and low-wage technology workers have experienced a substantial decline in employment, about 800,000 between 1999 and May 2006. This decline is in part the result of offshoring, moving these jobs to lower-cost locations. Not all the employment decline is trade-related, however. Some of the largest declines are for data entry keyers, word processors, and typists. These occupations have been heavily affected by changes in the technology itself, making it easier to read and transfer data electronically and allowing many white collar workers to enter their own documents or spreadsheets directly into the computer, bypassing the need for secretarial assistance.

Table 1 Detailed US IT-related occupations 1999–2006 (May)


Occupations

1999

May 2006

Total change

Percentage change

Annual wage May 2006


Call center occupations

 

Telemarketers

485,650

385,700

–99,950

–20.6

24,190

 

Telephone operators

50,820

26,350

–24,470

–48.2

32,710

           

Low-wage technology workers

 

Switchboard operators, including answering service

248,570

172,060

–76,510

–30.8

23,640

 

Computer operators

198,500

123,750

–74,750

–37.7

35,010

 

Data entry keyers

520,220

295,650

–224,570

–43.2

25,640

 

Word processors and typists

271,310

153,530

–117,780

–43.4

30,540

 

Desktop publishers

37,040

30,440

–6,600

–17.8

36,120

 

Electrical and electronic equipment assemblers

387,430

211,460

–175,970

–45.4

27,510

 

Semiconductor processors

42,110

41,520

–590

–1.4

34,730

           

Total call center and low-wage tech. workers

2,241,650

1,440,460

– 801,190

– 35.7

$ 27,227

             

Mid-level IT workers

 

Computer support specialists

462,840

514,460

51,620

11.2

44,350

           

High-wage technology workers

 

Computer and information scientists, research

26,280

27,650

1,370

5.2

96,440

 

Computer programmers

528,600

396,020

–132,580

–25.1

69,500

 

Computer software engineers, applications

287,600

472,520

184,920

64.3

82,000

 

Computer software engineers, systems software

209,030

329,060

120,030

57.4

87,250

 

Computer systems analysts

428,210

446,460

18,250

4.3

72,230

 

Database administrators

101,460

109,840

8,380

8.3

67,460

 

Network and computer systems administrators

204,680

289,520

84,840

41.5

65,260

 

Network systems and data communications analysts

98,330

203,710

105,380

107.2

67,460

 

Computer hardware engineers

60,420

74,480

14,060

23.3

91,280

 

Electrical engineers

149,210

147,670

–1,540

–1.0

78,900

 

Electronics engineers, except computer

106,830

131,880

25,050

23.4

82,820

           

Total High-wage tech. workers

2,200,650

2,628,810

428,160

19.5

$ 75,819


Sources: Bureau of Labor Statistics CES Data ( 1999, 2000, 2001, 2002, May 2003, November 2003, May 2004, November 2004, May 2005, and May 2006); National Occupational Employment and Wage Estimates; Kirkegaard, Jacob F.

This is an important point. The book by Frank Levy and Richard Murnane points out that the characteristics that make it possible to offshore a particular job also make it possible to automate that job.4 This means that offshoring and automation are often alternatives. It is misleading to look at jobs that have “moved” to India and assume these jobs would have remained in the United States. In many cases, the jobs would have been automated if there had not been the opportunity to buy the service overseas.

The lower part of the panel shows employment for mid-level workers and high-wage technology workers. The mid-level employment has risen nearly 52,000, and the high-wage workers have increased by about 428,000 between 1999 and 2006. Despite the impact of the technology crash in 2000–2001, and despite the impact of service sector offshoring, employment in these job categories on average has increased substantially—by nearly 20 percent. Within the high-wage categories, however, there is one that stands out: computer programmers have seen a decline in employment of about 133,000. The decline in employment in this area comes because of the end of the tech boom, but also because many programming jobs have been relocated offshore. The person who heard that programming was the way to ensure a good job and took some courses to learn the basics has found that the jobs are not there. Those that upgraded their programming and computer systems skills have been in demand.

The Economics of Service Sector Off-Shoring.5 One of the things that scare Americans is the idea that almost any job today could be offshored. That is not true. A careful estimate has found that about 11 percent of all jobs could theoretically be carried out in a remote location. There are higher estimates around, but these do not take into account adequately some of the difficulties of performing tasks remotely, including the difficulty of complex, one-on-one interactions that are required in many operations.6

Even though 11 percent of employment is a lot smaller than some of the scare numbers out there, it is still a very large number of jobs. Civilian employment in the United States was about 146 million in 2006, so 11 percent would be over 16 million. But in fact the likely number of jobs that will be offshored over the next few years is much smaller than this. The main determinant of the number of jobs offshored is the extent to which US businesses judge that it is economic to do so. For some sectors the cost advantage from moving off shore is very small and not worth the risks involved. This is becoming increasingly true for offshoring to India, where wages are rising very rapidly for skilled workers. For many sectors it is not possible to disaggregate their value chains and move parts of them overseas because the business processes are just not suitable. Many small businesses do not have the scale to make offshoring worthwhile. For some sectors there are issues of regulation or intellectual property protection that preclude offshoring. On balance, it can be expected that no more than four million jobs will be offshored over the next five years, or about 2.7 percent of civilian employment in the United States. Figure 1 illustrates the different factors that influence the offshoring decisions companies make.

Figure 1 Factors that Influence Degree of Adoption

Source: McKinsey Global Institute, The Emerging Global Labor Market.


Overall, the growth of offshoring is demand driven because there is an adequate supply of workers located in other countries that are qualified to perform the tasks that US companies will look for. There are a couple of important qualifications on the supply side, however. One of the arguments often used to argue that US jobs and wages are threatened is to claim that there are billions of new workers in the global labor market competing directly with American workers. This is not the case. After careful interviews with a number of companies, the McKinsey study found that the number of suitable workers available is much, much smaller. Based on educational qualifications alone there were about 33 million workers available in 2003, but after assessing their language skills and suitability and availability to work for multinational companies, the number dropped to about 4 million. The number of suitable workers is growing over time, of course, and so the overall supply will be more than adequate to meet the US demand of around 4 million over the next five years, but talking about billions of competing workers is just misleading.

The second qualification is that the number of suitable engineers, particularly software engineers, in the global economy may not be adequate to meet demand, leaving unmet engineering needs and/or rising relative wages for this group. Countries such as India and China are growing at an amazing pace and increasing their own demand for skilled workers. High tech in the United States is a rapidly growing sector again. If demand growth exceeds current estimates there will be a shortage of trained workers globally.

Globalization and Technology: Evolving Models. The nature of service sector offshoring is changing. Initially, companies took part of their value chain and sent it overseas—call centers or basic programming. What is happening now is that US companies are forming partnerships with companies in India and elsewhere. The new models have the following characteristics:

As is to be expected, the opening up of service activities to globalization has triggered a new round of interactions. The overseas suppliers of services are developing skills that allow them to work with US multinationals to increase productivity, the range of activities that can offshored and the different geographies that supply services. As offshoring matures as an activity, it takes on new roles which focus on improving productivity and efficiency in US operations, not just moving jobs. Note also that leading Indian offshoring companies are rapidly increasing their operations in the United States and Europe. Many of the outsourced services being provided to US companies are being supplied by employees of outsourcing companies that are based in here in the United States, creating American jobs.

The Shifting Mix of Jobs. The US economy has sustained low rates of unemployment for the last 20 years and currently has an unemployment rate of 4.5 percent, so our economy can create jobs. Indeed many companies report they have trouble recruiting workers. The challenge for the US labor market is that the distribution of wages has become much wider over time. How serious this problem is and the extent to which it is the result of trade or technology is a matter of controversy that I will not address here, but there is no question that the offshoring process has resulted in a shift in the composition of employment. As we saw in table 1, in computer and other occupations that have been subject to offshore competition, there has been a decline in basic jobs and an increase in higher skill jobs, on balance. Although offshoring is not large enough to be a main driver of the distribution of income in the United States, it will contribute to some extent.


Policy Implications of Offshoring


Education, Globalization, and the Science and Technology Workforce

We know that the American education system is not providing adequate skills to many Americans, skills that would allow them to get better jobs and that would increase the number of people that can work in R&D and technology jobs here in the United States. This is a hard problem to fix, and part of the difficulty is that many students are unwilling to study technical subjects. We could help, ho wever, by increasing opportunities and incentives.

Higher education has become more expensive for low-income families because the value of government scholarships and awards has not kept pace with rising education costs. Congress could help solve this problem by providing additional grant money for students that lack the resources to attend.

Americans do respond to incentives. Many people, including myself, believe that it is in the interest of the economy as a whole to have an increase in the number of people educated in science and technology and hence a case for public support of science and technology education. Having a strong science and technology workforce based in the United States helps generate good jobs and preserve our current strength in this area. Congress could add to the size of this workforce by providing more graduate scholarships in science and technology subjects that are available to US citizens and permanent residents. It is contradictory to talk about the need to protect our technology infrastructure if we are unwilling to pay the modest amounts needed to strengthen it directly.


Conclusions

Globalization is being blamed for problems that have been created by failures in other areas. The United States does not save enough; job transitions are too costly because they can cause a loss of health insurance; workers that lose or leave jobs are not given adequate income or retraining support to help them find new jobs that are better than the ones they may have lost. Denmark has developed a system of “flexicurity” that gives them a flexible labor market but provides substantial but tough-minded support for workers. Most of the rest of Europe has income support but not enough flexibility. The United States has flexibility but not enough support. The Danish model is not one that could be translated directly to the United States, but there are lessons for the United States here. Denmark has more people employed than does the United States, relative to population, and sustains a lot of good jobs.

For a number of years the value of the US dollar against many currencies was out of line with the level that would allow US workers to compete effectively and exploit the underlying strength and productivity of the US economy—it is still out of line against some currencies. The most important way to make sure the US economy retains its strength as a center of technology jobs is to increase national saving and reduce our dependence on capital inflows from overseas, inflows that are the counterpart and enabler of our trade deficit. The federal government has run very large cumulative budget deficits for many years. We need a fiscal policy in which there are budget surpluses during periods of full employment.

Trying to strengthen the R&D and technology jobs base of the United States by subtle or overt protectionism is a mistake. The United States is already an attractive location for these activities, and it will become more attractive if we can take advantage of the global economy to reduce costs. In particular, Americans will be much better off if we can use the global economy to reduce the crushing costs of health care.


Notes

1. Scott Bradford, Gary Clyde Hufbauer, and Paul Grieco. 2006. The Payoff to America from Global Integration. In The United States and the World Economy: Foreign Economic Policy for the Next Decade, ed. C. Fred Bergsten and the Institute for International Economics. Washington: Institute for International Economics.

2. For a list of productivity studies, see www.mckinsey.com/mgi

3. OECD. 2007. Economic Policy Reforms: Going For Growth 2007 . Paris: Organization for Economic Cooperation and Development.

4. Frank Levy and Richard J. Murnane. 2005. The New Division of Labor: How Computers are Creating the Next Job Market. Princeton , NJ: Princeton University Press.

5. This section draws on The Emerging Global Labor Market (2006), a study of the McKinsey Global Institute on which I was an advisor. See www.mckinsey.com/mgi.

6. Alan Blinder (Off-Shoring: The Next Industrial Revolution. 2006. Foreign Affairs [March–April].) makes a rough estimate that 28 to 42 million jobs are susceptible to offshoring. Blinder does not mention the possibility of service jobs that come to the United States as a result of trade. J. Bradford Jensen and Lori Kletzer (Tradable Services: Understanding the Scope and Impact of Services Outsourcing, 2005, Peterson Institute Working Paper 05-9, Washington: Peterson Institute for International Economics) use an original empirical approach and indicate a pretty large number of jobs that could theoretically be offshored, although the authors believe only a fraction of this total are actually vulnerable.



© 2014 Peter G. Peterson Institute for International Economics. 1750 Massachusetts Avenue, NW.
Washington, DC 20036. Tel: 202-328-9000 Fax: 202-659-3225 / 202-328-5432
Site development and hosting by Digital Division