Date: Thu, 17 Nov 2005 22:40:09 -0800 From: Norm Matloff To: Norm Matloff Subject: innovation to the rescue--or not To: H-1B/L-1/offshoring e-newsletter For the last couple of years some real estate researchers at the UCB business school have gotten a lot of attention by predicting dire job losses in California from offshoring. In their new paper, reported by the two articles enclosed below, they take a somewhat more optimistic point of view, latching on to the position currently fashionable among advocates/defenders of offshoring (free traders, industry spokepeople, politicians) that offshoring won't be a problem for the U.S. as long as we continue to foster innovation. I of course disagree, for reasons which I've stated in bits and pieces in the past, and will state in more organized form here, while making comments on the articles and the UCB paper. (You can get the paper itself at http://repositories.cdlib.org/iber/fcreue/reports/1005) I should begin by pointing out that I've always been skeptical about claims that offshoring will ever become big in the computer field. I showed in my CACM article (http://heather.cs.ucdavis.edu/CACM.pdf) that job losses due to offshoring have been much smaller than those due to "the other globalization," namely importation of workers under the H-1B and L-1 work visa programs. However, in the same article, I noted that it is really too early to tell just how much the offshoring phenomenon will grow, and whether the expanded levels will be sustained. Accordingly I analyzed what the impact would be if offshoring did become big, and showed that the industry's own data indicate an alarming prospect: Offshoring would have the effect of losing jobs which require a greater degree of education while gaining jobs having lesser educational requirements. This point is absolutely crucial. In our context here, it is to me the biggest flaw in the authors' paper. They are in essence saying (without realizing it) that it's OK to have the technical work for a new product done abroad as long as it produces, for instance, marketing and sales jobs for that product in the U.S. This point of view is understandable, since the authors are in a real estate research center. Their perspective is pretty much limited to the overall amount of economic activity in a given geographical region; if marketing and sales people are well paid, they can afford to buy real estate. But even those without economic background can understand that one cannot sustain prosperity on marketing and sales people alone, not in the long run. Please keep this in mind as you read on. A short and somewhat oversimplified summary of the authors' findings is that even though a number of firms are now offshoring their R&D work, the newer, smaller companies are not doing so, and since innovation and its resulting economic growth come from such firms, things may be OK. There are a number of problems with these arguments. First of all, as the Contra Costa Times reporter correctly quoted me as saying, you can't just sit people down and order them to innovate. Innovation comes serendipitously, and not necessarily from R&D people. So the focus on formal R&D is rather misplaced to begin with. Innovation comes randomly while ordinary people do their daily work in the field. The premise of the innovation argument is that Americans are more innovative. Fine, but the point is that if you decimate the number of Americans working in the field, you decimate the amount of innovation. In other words, offshoring, based on the notion that India has a comparative advantage (namely, cheap labor) over the U.S. in programming, would ironically cripple America's own putative advantage, the propensity to innovate. This is even worse in light of the fact (shown in past research) that when job opportunites in a profession are radically reduced, the students most likely to bail out of majoring in that field are the most talented students. The CCT reporter misquoted me on the issue of size, though. I was referring to the U.S. as a whole, not individual firms. What I had said was that if the U.S.' forte really is innovation, as the supporters of offshoring claim (and I agree), then by sending most of its programming work abroad it would lose all that serendipitous innovation, as the work would now be done by people in countries which have ostensibly less innovative cultures. Secondly, the authors' finding that the smaller--and they say more innovative--firms tend not to offshore R&D and thus innovation is likely to stay in the U.S. seems to be a misinterpretation. Come on, guys, isn't the first reason for the difference in behavior between the small and large firms simply that it's infeasible for small firms to establish R&D centers abroad? They just don't have the critical mass, so they couldn't send R&D abroad even if they wanted to. I also object to their claim that the smaller firms are more innovative. The history of the computer industry actually does NOT show that the smaller firms tend to produce the major advances. The small firms that have become big have done so largely by being at the right place at the right time, by having the right connections, by having business savvy, etc.--but usually NOT because they were great innovators. A point touched on by the authors is that even if innovation is done abroad, it still can create jobs here. This is true, a good example being the programming jobs generated here as a result of the invention of the Web in Switzerland. Ditto for the jobs created by the development of the Finnish (actually Swedish) Linux operating system, the British invention of relational databases, the Norwegian invention of object-oriented programming, etc. But that is missing the point. As I said earlier, the real question is what kinds of jobs IN THE U.S. which will be generated by innovation, regardless of where the innovation occurs. Say for example offshoring had already become big at the time the Web was developed. In that case, the Web programming jobs would have gone to India instead of the U.S.--while sales and marketing jobs related to the Web would grow here. Sales and marketing jobs are not enough. As I've said before, if the U.S. tries to succeed on the basis of adeptness in blarney, it will lose out to Ireland. :-) Norm http://www.contracostatimes.com/mld/cctimes/13193150.htm Thursday, Nov 17, 2005 Small high-tech firms outsource less By George Avalos CONTRA COSTA TIMES Maybe Jerry Brown had it right after all when he famously declared that small is beautiful. Two UC Berkeley researchers on Wednesday released a study that suggests smaller high-tech companies are more likely to create home-grown jobs and avoid outsourcing the work than larger technology firms. That's because tech companies in the Bay Area are often formed around an innovative idea that leads to creation of a cutting-edge product or service. The research-and-development process usually leads to the hiring of high-skilled, high-wage workers whose jobs are difficult to outsource. "By definition, the more innovative companies tend to be the smaller ones," said Ashok Bardhan, a senior research associate at UC Berkeley's Fisher Center. "Startups are often the medium through which innovations are brought to market. Most of these small companies are not going in for outsourcing." In contrast, large companies could find it easier to conduct research and development overseas, Bardhan said. The study, titled "Innovation, R&D and Offshoring," concluded that outsourcing will not doom the American economy, despite the erosion of certain kinds of jobs. The key is producing the right conditions for fledgling companies to create advanced products, according to the research by academics Bardhan and UC Berkeley professor Dwight Jaffee. "Continued innovation," according to the researchers, is "the primary way to create high-paying new jobs in this country." The study resulted in part from a survey of 48 Bay Area technology companies. Most of the companies surveyed were small- and medium-sized firms with fewer than 500 workers. About 45 percent of the firms had more than half of their current sales from products and services that were less than three years old. Among the findings culled from the companies: o Twenty-six of the 48 firms resorted to domestic outsourcing of some type of activity, primarily manufacturing. Shifting work to other parts of the United States is the most common form of outsourcing by the firms in the study. o The average size of a company that undertook offshoring was 2,931 employees. The average size of a company that did not undertake offshoring was 487 employees. o The average size of a company that outsourced at least some research-and-development operations was 4,243. The average size of a company that outsourced none of its research and development was 477. o The average size of a company that was considered innovative by the researchers was 588 employees. The average size of a company that was not considered innovative was 2,486 employees. Some academics, though, took issue with the study. They believe the UC Berkeley researchers lean too heavily on innovation as a solution and question whether the Bay Area economy can prosper if it relies too much on high-paying jobs. "If you are a low-skilled person, the argument about innovation probably does not hold a lot of water," said Nan Maxwell, chair of the Economics Department at California State University East Bay. "Maybe the innovative firms can replace the number of jobs lost, but they are different kinds of jobs than the ones that are lost by low-skilled workers. And the jobs that remain cannot be filled by low-skilled workers." One important solution, Maxwell argued, is to make sure that plenty of training is available for low-skilled workers. "The person with lower skills has access to fewer jobs, and those are the people who get hurt the most by outsourcing," Maxwell said. Norman Matloff, a professor of computer sciences at UC Davis, said innovation does not automatically arise from small companies. "It comes from groups of people working on products," Matloff said. "You need a large group of people working every day on various ideas, and that's how innovation happens." Innovation could become stunted in the Bay Area if more software and engineering jobs are shifted overseas or to other parts of the United States, Matloff warned. "You can't just hire innovators, tell them to sit down and start to innovate," Matloff said. Still, the UC Berkeley researchers believe Bay Area residents should not fear the shift of jobs overseas or other states. "Outsourcing will not be the doom of the U.S. economy," Bardhan said. _________________________________________________________________ George Avalos covers the economy, financial markets, insurance and banks. You can reach him at 925-977-8477 or gavalos@cctimes.com. http://www.bizjournals.com/sanjose/stories/2005/10/31/story6.html?t=printable Silicon Valley/San Jose Business Journal - October 31, 2005 Business News - Local News EXCLUSIVE REPORTS >From the October 28, 2005 print edition Offshoring research raises questions about valley's innovation future Timothy Roberts Cisco Systems Inc. may not be moving jobs from San Jose to India, but the $1.1 billion investment in Bangalore it announced recently may presage the day when Silicon Valley loses its edge as the center of innovation. Cisco, the giant router-maker with headquarters on the North First Street technology row, says it will invest three-quarters of that billion dollars in a research and development campus that will employ 1,400 people. Cisco and other industry voices say that technology companies have to make similar investments abroad if they want to remain strong at home, that employing engineers in Bangalore is necessary if there are to be engineers in San Jose. But a just-released study of offshoring by researchers at the University of California at Berkeley raises questions about Silicon Valley's continued pre-eminence as the engine of innovation. "The storm clouds are there for all to see," says Ashok Deo Bardhan, a senior researcher at the Fisher Center for Real Estate and Urban Economics at UC Berkeley. The new Berkeley study by Mr. Bardhan and Dwight Jaffee, is a follow up to a widely read study in 2004 that was used to raise concerns over offshoring of U.S. jobs. The new study updates the earlier one and raises a new concern -- that offshoring is now reaching beyond manufacturing and call centers into the ranks for the highest-level jobs in technology. "Offshoring has been steadily creeping up the value chain and has reached the R&D segment," the report reads. Until now, many Silicon Valley boosters have argued that the high-value jobs will remain here even as manufacturing and marketing moves to places in which it is less expensive to operate. But the same forces of competition that have lured manufacturing jobs overseas are now reaching the research lab. "The offshoring of R&D and innovation is fueled largely by the same considerations as offshoring in general -- costs, the availability of educated workers and the opening up of new markets," write Mr. Bardhan and Mr. Jaffee. "It stands to reason that India, China and other developing and transitioning countries are bound to take a larger chunk of the scientific pie." More than anything it is the opening of markets that dictates the need for overseas R&D, says Cisco spokesperson Ron Piovesan. "It's always advantageous to have engineers next to the customers to see what works and to work that into the product," he says. Cisco needs to be in India, he says, because the company has seen its sales grow by 50 percent over the past three years. "We see an enormous potential there," he says. Paul Krutko, San Jose's director of economic development, doesn't see Cisco's India investment as a loss for Silicon Valley but ultimately as a win in that it could make the company stronger. "Cisco needs to pursue markets that are growing in the world," Mr. Krutko says. "It's a smart strategy and will lead to increased employment here at the mother ship." San Jose's economic development plan is to create a business environment in which companies will form and then grow to 100 to 300 people, all of whom will be based in San Jose. But additional hires would be expected to be oversees. The focus is on keeping those high-value jobs here while recognizing that many of the lower-value jobs don't make sense in San Jose because of the costs of doing business here. "In our strategy, we are growing companies to be the next Cisco," Mr. Krutko says. The decline of manufacturing in Silicon Valley is unmistakable. The California Employment Development Department reports that in 1990 there were 248,000 people employed in manufacturing in Santa Clara County, which includes most of Silicon Valley, but in 2004 there were only 167,900 employed in that sector. Analog Devices Inc. just announced plans to eliminate 400 more manufacturing jobs by closing its plants in Sunnyvale and Santa Clara. But the company says it will keep the 300 jobs it has on North First Street in San Jose. About 200 of those jobs are in R&D. Analog is content to keep its research into power management for computer chips and into precision amplifiers where it is in San Jose, mainly because that is where the talent is. "We have to look at where people want to be," says Maria Tagliaferro, director of corporate communications. "Silicon Valley is still a good place because there are strong technical people who want to be there." The market for power management and precision amplifiers is growing, and Ms. Tagliaferro says the San Jose location may grow. Increasingly, Silicon Valley will have to compete for those types of high-level jobs, says Mr. Bardhan, but it has much of what it needs to compete. "The higher-education system still produces a lot of scientists and engineers," he says. "The cost of innovation has increased, the infrastructure of innovation is here -- the engineers and scientists but also the lawyers and entrepreneurs. And the West Coast is an integral part of the Pacific Rim." In a study released this year by the Silicon Valley Leadership Group, which includes most of the major technology companies here among its membership, Silicon Valley ranked eighth -- last -- in a ranking of high-tech centers in the United States. The criteria were cost of doing business and quality of life issues. Next year, the Leadership Group plans to compare the valley with India, China and, most likely, Ireland. "Silicon Valley may be the top tech region in the world," says Carl Guardino, president and CEO of the Leadership Group. "But we should remind ourselves that we are not the only tech region in the world." TIMOTHY ROBERTS covers public policy, corporate governance and Internet security for the Business Journal. Reach him at (408) 299-1821.