How much are companies shaped by the business and social cultures around them? A lot, judging by the divergent fates of the nation's two great high-tech centers

During the 1970s northern California's Silicon Valley and Boston's Route 128 attracted international acclaim as the world's leading centers of electronics innovation. Both regions were widely celebrated for their technological vitality, their entrepreneurship, and their extraordinary economic growth.

The enchantment waned in the early 1980s, when the leading producers in both regions experienced crises. Silicon Valley chip makers relinquished the semiconductor market to Japan, while Route 128 minicomputer companies watched their customers shift to workstations and personal computers.

The performance of these two regional economies diverged, however, later in the decade. In Silicon Valley a new generation of semiconductor and computer companies, such as Sun Microsystems, Conner Peripherals, and Cypress Semiconductor, as well as the region's established companies, such as Intel and Hewlett-Packard, experienced dynamic growth. The Route 128 region, by contrast, showed few signs of reversing its decline. The "Massachusetts Miracle" ended abruptly, and start-ups failed to compensate for continuing layoffs at the region's established minicomputer companies.

Why has Silicon Valley adapted successfully to changing patterns of international competition, while Route 128 is losing its competitive edge? Because, despite similar origins and technologies, the two regions have evolved distinct industrial systems since World War II. Their responses to the crises of the '80s revealed variations in local economic structure and organizational philosophy whose significance was unrecognized during the rapid growth of earlier decades. Far from superficial, those variations illustrate that local factors play an important role in determining how well a company will adapt to changes in an industry. And it's possible to pinpoint the factors that enable one region to capture and nurture the entrepreneurial spirit -- and allow another to let it slip.

Silicon Valley has a regional-network-based industrial system -- that is, it promotes collective learning and flexible adjustment among companies that make specialty products within a broad range of related technologies. The region's dense social networks and open labor market encourage entrepreneurship and experimentation. Companies compete intensely while learning from one another about changing markets and technologies through informal communication and collaboration. In a network-based system, the organizational boundaries within companies are porous, as are the boundaries between companies themselves and between companies and local institutions such as trade associations and universities.

The Route 128 region is dominated by a small number of relatively vertically integrated corporations. Its industrial system is based on independent companies that keep largely to themselves. Secrecy and corporate loyalty govern relations between companies and their customers, suppliers, and competitors, reinforcing a regional culture that encourages stability and self-reliance. Corporate hierarchies ensure that authority remains centralized, and information tends to flow vertically. The boundaries between and within companies, and between companies and local institutions, thus remain distinct in the independent-company-based system.

The performance of Silicon Valley and Route 128 in the past few decades provides insights into regional sources of competitiveness. Far from being isolated from what's outside them, companies are embedded in a social and institutional setting -- an industrial system -- that shapes, and is shaped by, their strategies and structures.

Understanding regional economies as industrial systems rather than as clusters of producers, and thinking of Silicon Valley and Route 128 as examples of the two models of industrial systems -- the regional-network-based system and the independent-company-based system -- illuminate the different fates of the two economies.

Consider two pairs of comparable companies, one pair located in Silicon Valley, the other on Route 128. The comparison of Apollo Computer and Sun Microsystems -- start-ups in the same market, the former on Route 128 and the latter in Silicon Valley -- shows how small companies benefit from external sources of information, technology, and know-how in a decentralized network-based industrial system. And the case of Route 128's Digital Equipment Corp. (DEC) and Silicon Valley's Hewlett-Packard -- the leading computer-systems producers in the two regions -- shows how regional networks facilitate the reorganization of large companies.

The experiences of Apollo and Sun show how the isolating structures and practices of Route 128's independent-company-based system put start-ups at a disadvantage in a fast-paced industry. Apollo pioneered the engineering workstation in 1980 and was enormously successful. By most accounts, the company had a product that was superior to that of Sun (which was started two years after Apollo, in 1982). The two companies competed neck and neck during the mid-'80s, but in 1987 Apollo fell behind the faster-moving, more responsive Sun and never regained its lead. By the time it was purchased by Hewlett-Packard, in 1989, Apollo had fallen to fourth place in the industry, while Sun was number one.

Apollo's initial strategy and structure reflected the model of corporate self-sufficiency that had been followed by its region's large minicomputer companies. In spite of its pioneering workstation design, for example, the company adopted proprietary standards that made its products incompatible with other machines, and it chose to design and fabricate its own central processor and specialized integrated circuits.

Sun, in contrast, pioneered open systems. The company's founders, then all in their twenties, adopted the UNIX operating system because they felt that the market would never accept a workstation custom-designed by four graduate students. By making the specifications for its systems widely available to suppliers and competitors, Sun challenged the proprietary and highly profitable approach of industry leaders IBM, DEC, and Hewlett-Packard, each of which locked customers in to a single vendor of hardware and software.

That strategy allowed Sun to focus on designing the hardware and software for workstations and to limit manufacturing, choosing instead to purchase virtually all its components off the shelf from external vendors. As Sun grew into a multibillion-dollar company, that focus enabled it to rapidly introduce complex new products and continually alter its product mix.

As a result, the Sun workstations, while vulnerable to imitation by competitors, were significantly cheaper to produce and priced lower than the Apollo systems. Apollo, like the Route 128 minicomputer producers, was slow to abandon its proprietary systems and as late as 1985 still refused to acknowledge the growing demand for open standards.

Sun's strategy succeeded because it drew upon Silicon Valley's sophisticated and diversified technical infrastructure. Apollo not only failed to respond quickly enough to industry changes but also suffered from a more limited regional infrastructure. Its commitment to formality, hierarchy, and long-term stability -- typical of most Route 128 companies -- could not have offered a greater contrast to the "controlled chaos" that characterized Sun.

The successes of the '80s generation of start-ups were the most visible sign that Silicon Valley was adapting successfully, but changes within the region's large companies were equally important. Established producers such as Hewlett-Packard decentralized their operations, creating intercompany production networks that formalized the region's social and technical interdependencies and strengthened its industrial system.

Adaptation in the Route 128 economy was constrained by the isolating organizational structures and practices of its leading producers. The region's large minicomputer companies adjusted very slowly to the new market conditions, and by the end of the decade they were struggling to survive in an industry they had once dominated.

By 1990 both DEC and Hewlett-Packard were $13-billion companies, and they're now among the largest civilian employers in their regions. Both faced comparable challenges, but each responded quite differently: Hewlett-Packard gradually opened itself by building a network of local alliances and subcontracting relationships, while strengthening its global reach. DEC, in spite of its formal commitment to decentralization, retained a substantially more self-sufficient organizational structure and corporate mind-set.

The lessons from Sun and Apollo, DEC and Hewlett-Packard are clear: local economies with industrial systems built on regional networks are more flexible and technologically dynamic than those in which learning is confined to individual companies. Sun and Hewlett-Packard are not unique in Silicon Valley -- the region is home to hundreds of specialty high-tech producers that adjust to one another's needs through shifting patterns of competition and collaboration.

Since 1980 Route 128 has continued to generate new companies and technologies, but its companies have failed to commercialize their technologies rapidly or consistently enough to sustain regional prosperity. The regional economy continues to flounder today as cuts in defense spending compound the difficulties caused by ongoing layoffs at DEC and other minicomputer companies.

What can be done to promote local health? Our comparison suggests that networks flourish in supportive regional contexts. To survive, networks need a region's institutions and culture to ensure the repeated interaction that builds mutual trust while also intensifying rivalries. When industrial networks are embedded in such a supportive local environment, they promote a decentralized process of collective learning and foster the continual innovation that is essential in the current competitive environment.

Yet the clustering of companies in a given area does not by itself create such mutually beneficial interdependencies. Companies in an industrial system may be geographically clustered and yet have limited capacity for adaptation if the area's leading producers are independent-minded. As in the case of Route 128 -- and many of the older industrial regions of the United States and Europe -- the legacies of a history of economic self-sufficiency that are passed on to the institutions and infrastructure of a regional economy mean that the prospects for regeneration are neither easy nor fast. Adopting an industrial system that breaks down the institutional and social boundaries that divide companies represents a major challenge for Route 128; it's a challenge that will be even more daunting for regions with less sophisticated industrial infrastructures and skill bases.

AnnaLee Saxenian is the author of Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Harvard University Press, 1994).