Local and global networks of immigrant professionals in Silicon Valley *.

• Author: Saxenian, Anna Lee
• Position: Report

INTRODUCTION

Entrepreneurship and globalization concern scholars and policymakers interested in economic transformation. However, researchers typically treat these phenomena in isolation. Most studies of entrepreneurship focus either on the attributes of individual entrepreneurs or on their connections to the local or regional environment. (1) Studies of globalization focus on multinational corporations and nation-states. (2) As a result, entrepreneurship and globalization are rarely linked.

Recent research suggests, however, that globalization and entrepreneurship are related: Foreign-born entrepreneurs are becoming agents of globalization by investing in their native countries, and their growing mobility is in turn fueling the emergence of entrepreneurial networks in distant locations. In Silicon Valley, for example, Taiwan-born entrepreneurs have built social and professional networks to support U.S. ventures, which they use to accelerate new firm in Taiwan. (3) There is evidence of a similar process among Indian immigrant entrepreneurs, (4) and scholars have begun to document emergence of strikingly similar transnational activities among Latin-American immigrants in the United States. (5)

We know little about the extent and contours of this phenomenon. In what ways are globalization and entrepreneurship linked? Do foreign-born counterparts? What role do ethnic networks play in the process of new firm formation? To what extent are first-generation immigrants creating transnational networks that link their native countries and the Unite States? What is the nature of these connections? Is the "brain-drain"-the migration of the best and brightest from poor to rich nations-accelerating, being reversed or being replaced by "brain circulation". That is, are there more complex two-way flows of skilled workers between developed and less-developed economies? (6)

Policymakers, face challenges resulting from the increasingly open flows of skill, technology, and capital across national boundaries. These processes have transformed debates about trade, immigration policy, and intellectual property rights, forcing creation of new institutions and mechanisms for adjudicating conflicts. This study will help to identify significant, and often unanticipated, areas of policy concern.

This study contributes to our understanding of entrepreneurship, globalization, and their interrelations by documenting findings of the first large-scale survey of foreign-born professionals in Silicon Valley. The survey explores the scope and organization of the local and transnational networks constructed by the region's immigrant engineers and scientists. It focuses on first-generation Indian and Chinese immigrants, the two largest groups of skilled immigrants in the region, and compares their participation in local and global networks to one another and to that of their U.S. born counterparts.

METHODS

Surveying foreign-born professionals is unusually difficult. Most daunting is developing a sampling frame because the target population (foreign-born engineers and other professionals) is difficult to identify and, once identified, difficult to reach.

We have only rough estimates of the population of immigrant professionals in a region, making difficult to determine representativeness survey. Nevertheless, we have attempted to maximize the study's validity.

We estimate from Current Population Survey (CPS) data that there were about 320,000 professional workers in the high-technology sectors of the San Francisco Bay Area economy in 2000, including approximately 20,700 born in Greater China (Mainland, Hong Kong, and Taiwan) and 18,400 born in India. [The estimates of the representation of foreign-born workers in the Silicon Valley workforce are based on data on place of birth and employment from the Current Population Survey 1994-2000 sample for the five-country Bay Area (San Francisco, Oakland, and San Jose). The totals are calculated using employment totals from the Bureau of Labor Local Area Statistics for San Jose, Oakland, and San Jose metropolitan statistical areas (MSAs). More recent date (1998-2000) suggest a substantial increase in Indian as well as Chinese high-technology, high-skill workers in the region; however, the sample size for the two-year period is too small for reliable estimates. Thanks to Peter Hall for his help with this analysis.]

The survey was deployed on the web because of the nature of this research--particularly its focus on computer-literature professionals. The relative lack of scholarly experience with web-based surveys led us to rely on the expertise of CustomerSat.com, an independent survey vendor in the San Francisco Bay Area, for this project.

We designed the survey questionnaire using insights gained from several years of interview-based research on immigrant entrepreneurship in Silicon Valley. In addition, a pretest was conducted with approximately 25 foreign-born professionals to identify any confusing or potentially misleading questions.

The sample was drawn from the membership of 17 leading immigrant professional associations in Silicon Valley. Foreign-born engineers, the great majority from China and India, mobilized these associations during the 1980s and 1990s, often in response to the experience of invisible barriers to professional advancement, or "glass ceilings," in the region. However, these associations quickly became important forums for the mobilization of ethnic resources to support information exchange, career advancement, and entrepreneurship within the region's immigrant communities. The associations that agreed to participate in the survey are among the largest and most active professional and technical associations in Silicon Valley, with memberships ranging from 500 to 5,000 (Saxenian, 1999).

The initial goal was for the associations to provide email addresses for all their members. Because many associations were concerned about preserving the confidentiality of their members, two methods of deployment were used. Six associations provided email membership lists directly to CustomerSat.com for deployment. Another 11 associations took the responsibility for sending the invitation to participate in the survey directly to their membership. In both cases, reminders were then sent out approximately two weeks after the initial invitation had been issued.

This sampling approach was our only available option, but it created two types of selection bias. Lists of association members used for the survey do not include all foreign-born professionals in the region, or even all the Indian and Chinese professionals. At most, these associations represent one-third of the Indian and Chinese immigrant population in Silicon Valley. [We estimate that out of about 20,700 Chinese professionals in the area, approximately 7,500 Chinese immigrants are members of local professional associations. Their total association membership is 15,000, but most Chinese professionals belong to more than one association, so their numbers have been deflated accordingly. Likewise, 6,461 Indians are members of local professional associations out of approximately 18,000 Indian professionals in the region. This means that close to one-third of both Chinese and Indian immigrant populations in the region belong to professional associations. Some of these association members are U.S.-born Indians or Chinese.] Furthermore, the association lists do not represent random samples but rather the most active members of respective communities.

The survey was on-line for two months, between May 15 and July 12, 2001. CustomerSat.com sent out 10,837 invitations to participate in the survey and received 2,273 responses, a 21 percent responses rate. Although this rate is consistent with those of other business surveys in California, a higher response rate would provide greater confidence in the findings. Moreover, the response rate varied depending upon the method of deployment. The response rates for associations that sent survey invitations directly to their members ranged from 1 percent to 19 percent.

The representation of foreign-born Chinese and Indian workers in the survey is difficult to calculate because of data limitations. The survey sample includes 788 respondents from Greater China, or 3.8 percent of the region's total professional population from that region. Likewise, the sample includes 769 respondents from India, or a 4.2 percent of the Bay Area's estimated foreign-born Indian professional population. The representation of the other foreign-born (189) and U.S.-born (260) populations is substantially lower, with 0.3 percent of the former and 0.1 percent of the latter.

One consequence of our sampling strategy is that the results are biased toward immigrants who are members of professional associations. There is potential for bias as well from the self-selection of the respondents. These limitations are not as severe as they might be because the focus of this research is on immigrants who play active leadership roles in their respective communities, particularly in starting companies and building both local and long-distance networks. Previous research has demonstrated that these foreign-born entrepreneurs are responsible for substantial wealth and job generation in the region (Saxenian, 1999).

The largest groups of foreign-born respondents that are not treated separately here are those from Hong Kong (4 percent) and South Korea (2 percent). We lack a sufficient number of responses from either group to reliably treat them separately, and except for the analyses of Greater China (which includes Hong Kong), they are included in the other foreign-born category. There is also a sizable cohort of Asian-American respondents, particularly those who identify themselves as Indian-American (26 percent of U.S.-born respondents to the survey) and Chinese-American (17 percent of respondents). Unfortunately, these samples are not large...