BY LISA TREI

How did Silicon Valley get to be so successful? And can that success be duplicated in other parts of the world?

Those are key questions that an international team of economists, business entrepreneurs and management specialists has been asking for the last two years. The group has compared Silicon Valley's experience to those of new centers of technological innovation developing in countries such as Ireland, Israel, India and Taiwan, and in northern Virginia and San Diego.

The group's findings will be presented in a forthcoming book, Silicon Valley and Its Imitators, that is scheduled to be published this year. On March 20, project members, sponsored by the Stanford Institute for Economic Policy Research (SIEPR), met to discuss some of their research with leading players in the valley.

"Silicon Valley has been awfully hard to clone," said Gordon Moore, the co-founder of Intel who is widely regarded as one of the valley's founding fathers. "I'm not sure if it can be planned; a lot of luck is involved. I don't think you could go back and repeat it here today."

However, Moore said, given the right combination of talent, opportunity and timing, new clusters can develop. Important elements for success include:

c The existence of a technology-anchored tenant in the area with many exploitable niches;

• Lots of well-trained engineers;
• Entrepreneurs with no fear of failure;
• Available capital to back ventures;
• Accommodating government policies.

Moore added that Silicon Valley's very success might also lead to its eventual demise. "What keeps the area vital is the inflow of young technical talent," he said. But with the high cost of living and the difficulty of working in a highly congested region, growth is becoming an increasing problem. "Young people are not being brought in," he said. "I have a concern that we're all growing old together."

Alfonso Gambardella, professor of economics and management at the Sant'Anna School of Advanced Studies in Italy, agreed that Silicon Valley's model can be duplicated but said certain constants are required. These include a ready supply of talent, such as the large numbers of unemployed skilled engineers found in Ireland, Israel and India. International demand based on a technological breakthrough, such as the Internet- and network-security markets exploited in Ireland and Israel, is another prerequisite.

Ashish Ahora, associate professor of economics and public policy at Carnegie Mellon University, said that some countries, such as India, are essentially "also rans" to Silicon Valley. "Being ahead is a huge advantage," he said. "In India, if you have a good idea you're encouraged to come here." However, this has started to change as the Indian government becomes less repressive and funding becomes more available. The biggest challenge for companies in India, Ahora said, is to learn how to grow and develop. "They can figure out what to do, but doing it is going to be a lot harder," he said.

In addition to Moore, Gambardella and Ahora, project members included SIEPR-affiliated economists Professor Timothy Bresnahan, Acting Assistant Professor Scott Wallsten and Michael Horvath, associate professor of business at Dartmouth College.

According to Bresnahan, who led the project with Gambardella, present-day centers or clusters are more similar to the Silicon Valley of the 1960s than of today's valley. "New clusters offer substantially less support for entrepreneurship in the start-up or pioneering phase than does a mature cluster like Silicon Valley today," Bresnahan described in a project overview. "Network effects -- benefits to particular technology firms that arise from the presence of other firms or from support structures such as venture capital -- play only a small role in our regional studies. Those benefits appear to come later."

Furthermore, Bresnahan said, new centers must take advantage of a technological and market opportunity that is not already being exploited, be it the integrated circuit industry in 1960s-era Silicon Valley or Finland's present-day cell phone industry. He said successful new clusters turn away from established sources of income, seeking new ones, "and want to make their relationship with existing technologies and clusters cooperative rather than competitive in the start-up phase."

Another similarity is the amount of investment, effort and development required before a cluster takes off. "Development of the integrated circuit business [in Silicon Valley] was not the magic exploitation of some 'new economy' rules," he said. "Instead, it took years of firm-building and market-building efforts." Long-term investment in the education of a skilled labor force has been critical in a number of regions, notably Ireland, India, Israel and Taiwan. "The specific supply of highly skilled workers can come from any of a number of sources," he said, "but it needs to be plentiful: Stanford-trained scientists and engineers to Silicon Valley, Russian émigrés to Israel and an educated population earning far less than world-standard wages in Ireland and Israel."

Although it is critical to invest in the assets that permit the emergence of a successful new center of technology, Bresnahan stressed that no "magic recipe" for success exists. "Sometimes these long-[term] investments in national or regional capabilities can grow for a long time in what seems like a low-returns mode before they take off," he said.

Based on the project's research, the group concludes that:

c Growing pressure on the limited amounts of skilled labor in this country may favor the emergence of new clusters elsewhere. A reverse "brain drain" of foreign-born professionals hit by the recent economic downturn in Silicon Valley may help spur growth in their native countries.

c New clusters have to deal with the reality of U.S. dominance of the most important existing areas of technology. Emerging companies have to learn how to stay linked to existing clusters while starting new ones. The best solution appears to be participation in a worldwide innovation network that is an extension of Silicon Valley itself.

c Efforts to jump-start clusters or make top-down efforts to organize them do not work. However, governments playing a supportive, but not leading, role have made entrepreneurship easier in many regions, notably in Ireland, Taiwan, Israel and Virginia.