Case 3.1 The rise, fall and resurgence of industrial hot spots: The experience of Silicon Valley and Boston’s Route 128 [i]
The success of high-technology industries in Silicon Valley and the Boston Route 128 region has attracted many followers, both domestically and internationally. In 1999, there were as many as 88 ‘Silicon Wannabes’: one ‘Silicorn Valley’ (Fairfield, Iowa), one ‘Silicon Sandbar’ (Cape Cod, Massachusetts), ten ‘Silicon Prairies’ (e.g., Lincoln, Nebraska; Payne County, Oklahoma) and five ‘Silicon Islands’ (e.g., Long Island, New York).
Both Silicon Valley and Route 128 prospered after WW II and faced downturns in the early 1980s, though Silicon Valley regained its vitality in the early 1990s, with Route 128 resurging only in the late 1990s. Today, Silicon Valley remains an innovation hub, while Route 128 has succeeded in other industries. The similarities and differences between the two regions help to explain the functioning of industrial clusters, especially the roles played by venture capital, local universities, local culture, industry structure and technology.
Silicon Valley
Silicon Valley did not establish a name for itself in semiconductors or any other core technology until the mid 1950s, though lucrative military contracts during WW II and the Cold War, and the engineering expertise at Stanford University, helped establish the electronics industry in the region, with the formation of firms such as Hewlett-Packard (HP), Litton Industries and Varian.
The semiconductor industry in Silicon Valley started in 1955, with the founding of Shockley Transistors by William Shockley, co-inventor of the transistor at Bell Laboratories. In 1957, the eight scientists and engineers at Shockley Transistors resigned together and set up Fairchild Semiconductor. Fairchild led the transition from the transistor to the integrated circuit, which was widely adopted in the consumer electronics and computer industries in the 1960s.
The high profits in the transistor and integrated-circuit industries attracted not only firms from other industries, but also start-ups funded by early employees who had become very wealthy. For example, by 1986, at least 124 start-ups could be traced back to Fairchild, including Intel, National Semiconductor and Advanced Micro Devices. Moreover, the founders and key employees of Fairchild’s spin-offs earned enormous capital gains, some of which were reinvested in new start-ups or venture capital funds. The availability of venture capital and the creation of successful new firms thus created a self-perpetuating process. By 1969, the San Francisco Bay Area had 150 active venture capitalists.
By the late 1970s, Silicon Valley had become synonymous with the semiconductor industry. However, in the mid 1980s, Silicon Valley’s established chipmakers faced strong competition from Japanese firms, who literally took away the semiconductor market. Throughout the 1980s, the leading chipmakers continued to struggle to compete with Japanese firms. As a result, the Silicon Valley region lost 25,000 jobs.
However, by 1991, Silicon Valley regained its prosperity, with a new wave of successful semiconductor start-ups. During the 1980s, more than 85 new semiconductor firms were started in Silicon Valley.
By 1990, electronics products exported by Silicon Valley represented almost one third of the total US exports of electronics products.
Route 128
Close industry–government interactions heavily influenced the development of Route 128 through spin-offs from government-funded research at Harvard and the Massachusetts Institute of Technology (MIT) during WW II and the Cold War. In the 1950s and 1960s, about 156 firms spun off from MIT, including Digital Equipment Corporation (DEC) in 1957. Wang Laboratories spun off from Harvard’s Computation Laboratories in 1951.
Many of these early firms became the source of later start-ups. For example, during the late 1960s and early 1970s, nearly 60 minicomputer producers were founded by engineers who used to work at DEC or other minicomputer producers. The expansion of the minicomputer industry attracted an infrastructure of support firms, including suppliers, software firms and consultants. By the late 1970s and early 1980s, Route 128 had become the centre of the minicomputer industry, representing 60 per cent of total US minicomputer production in 1982.
However, during the 1980s, the number of start-ups declined, as the customers of minicomputer companies shifted to workstations and personal computers. By 1998, Route 128 no longer had any large, dominant electronics firms.
Success factors
The two industrial hot spots shared many similarities, which distinguished them from other US regions.
Educational infrastructure
The two areas are both near excellent technical universities and business schools. Silicon Valley is near Stanford and the University of California, Berkeley. Route 128 is near Harvard University and MIT. These universities provided not only spin-off opportunities through their pioneering research, but also top scientists, engineers, technicians and managers. Between the end of WW II and the mid 1980s, four major MIT labs (the Electronic Systems Laboratory, Instrumentation Laboratory, Lincoln Laboratory and Research Laboratory for Electronics) and five academic engineering departments (Aeronautics and Astronautics, Chemical Engineering, Electrical Engineering, Materials Science and Mechanical Engineering) together produced a staggering 181 spin-offs.[ii] In the late 1980s, MIT formalized the spin-off process via the MIT Technology Licensing Office, which had recorded more than 250 spin-offs by the end of 2003.[iii] The Technology Licensing Office offers mentorship and patent assistance to entrepreneurs and inventors. In 2009, the Technology Licensing Office was issuing around “500 new invention disclosures” per year as a part of its mission “to benefit the public by moving results of MIT research into societal use via technology licensing”.[iv] The roughly 25,800 companies started by MIT alumni employed around 3.3 million people worldwide and generated US $2 trillion in yearly revenue in 2009.[v]
Venture capital
The two areas have easy access to large amounts of venture capital. For example, in 1981, Silicon Valley led all high-tech regions by receiving 32 per cent of the total venture capital in the US. Route 128 received 15 per cent. In 1985, Silicon Valley firms received US $800 million in venture capital, while Route 128 firms received US $300 million. The numbers just continued to grow, and between 1990 and 2000, the average annual amount of venture capital invested was US $1.767 billion for Silicon Valley, and US $493 million for Route 128.
The influence of the initial leading firm
Fairchild provided success stories, experienced engineers/managers and spinoffs, all of which spurred the development of the semiconductor industry in Silicon Valley. DEC and Wang Laboratories played similar roles for the microcomputer industry in the Route 128 area.
Government funding as the catalyst
The two areas received early and heavy support from US military and space programmes, which functioned as the catalyst for high-technology innovation. Between 1958 and 1974, the Pentagon spent US $1 billion on semiconductor research. Between 1990 and 2000, the average annual US federal government investment in R&D was US $883 million in Silicon Valley, and US $764 million in Route 128.
Explaining the difference
During the mid and late 1980s, both Silicon Valley and Route 128 were in crisis, although by the early 1990s, Silicon Valley rebounded while Route 128 failed to do so. The divergent performance between the two regions has been attributed to their different internal functioning, including:
Corporate culture – conventional organization versus ‘the HP way’
With an established industrial tradition and many experienced managers, firms on Route 128 were characterized by conventional, centralized hierarchical structures, with formal decision making and conservative work procedures. In most firms, including RCA, Honeywell and Raytheon, senior committee members retained the final authority on all major decisions. Such a structure facilitated vertical information flows, but largely neglected horizontal communication. Moreover, stability and company loyalty were highly valued, even more than experimentation and risk taking.
In contrast, firms in Silicon Valley were organized using a participative management style, professional autonomy and informality, known as ‘the HP way’. In Silicon Valley, firms eliminated many traditional status symbols, such as reserved parking spaces and differentiated office furniture for top managers. Informal luncheons, intramural sports teams and hallway conversations reinforced the participatory culture.
Attitude towards entrepreneurs and risk
While both regions support entrepreneurship and innovation from outside sources, subtle cultural difference may have affected their respective development. Mark Zuckerberg, the founder of Facebook attempted to secure funding in the Boston area in 2004, and was rejected by various venture capitalists. Route 128’s delay in embracing social media and networking was partly affected by the fact that investors in the area were generally older. In 2011, angel investors in Boston were on average 55 years old, while in California this average age was 32. Said one professor at MIT’s Sloan School of Management: “It became a generational issue. .. . To understand things like Facebook, you have to be 19 to 24 years old. If you’re 56, you don’t quite get it.”[vi]
Regional network – self-sufficient corporations versus relational networks
As far back as the early 1970s, Route 128 was dominated by a small number of self-sufficient and highly vertically integrated firms, such as DEC. Most of these producers designed their own computers, manufactured as many of their components and peripherals as possible, assembled their own computers and controlled all sales, after-sales service and marketing. Outsourcing was kept to a minimum, and the boundaries between firms were quite distinct. Moreover, the military contracts gained by Route 128 firms reinforced secrecy rather than open collaboration.
In contrast, Silicon Valley was not organized around a few dominant established firms. Rather, it was organized around its dense networks of social and professional relationships. These relationships strengthened repeated informal interactions. For example, after Wilf Corrigan resigned as Chairman of Fairchild Semiconductor in 1979, he started to contact his former customers and colleagues for their ideas regarding semi-customized integrated circuits. He met a former employee, Robert Walker, who had just completed some research on the custom chip business. Together, Corrigan and Walker established a new firm, LSI Logic Corporation, with venture capital financing. Later they hired a number of Walker’s former colleagues at Fairchild to develop their business. LSI went public in 1985, assisted by technical consultants, local research firms and trade associations. This exemplifies how relational networks facilitated the diffusion and exchange of intangible technical knowledge.
Engineers in Silicon Valley were more committed to the concept of advancing technology than to the firm where they worked, and they therefore moved easily from one employer to another. California law did not support ‘post-employment covenants not to compete’, in contrast to Massachusetts law. Thus, inter-firm rivalry was not critical to individual scientists, engineers or managers, as today’s competitor might become tomorrow’s customer or colleague.
Dominant design – minicomputer versus semiconductor
Silicon Valley and Route 128 had a different ‘dominant design’, instrumental to each region’s technological trajectory.[vii] For Route 128, the minicomputer represented only one segment of the computer industry. Because a minicomputer is an assembled machine – an end product, to use Prahalad and Hamel’s terminology – the minicomputer industry allowed companies to innovate new and better components, but otherwise provided few additional opportunities. In contrast, the semiconductor is a component, not an end product. As such, the semiconductor industry allowed Silicon Valley companies to innovate to create new applications in diverse areas such as consumer electronics, communications and programmable logic devices. Thus, the early entry into the two different industries determined, to a large extent, each region’s technological pathway.
The question arises: with each region locked in on the path of a very specific technological trajectory, will that trajectory lead to future success or failure?
The current situation
Silicon Valley is thriving. Of the US $23.3 billion in total US venture capital investments in 2010, Silicon Valley firms received over US $9 billion.[viii] Silicon Valley accounted for 39 per cent of US venture capital investments in 2010, a 5 per cent increase since 2003. In contrast, Route 128 only received about 3 per cent in 2010. In previous years, Route 128 would typically receive about an 11 per cent share in total US venture capital. In terms of individual firms, Silicon Valley start-ups raised almost triple in capital what equivalent Route 128 firms raised between 2006 and 2011.[ix]
Silicon Valley is still one of the most attractive places in the world to run a business, though high housing prices have become a concern. Single-family home prices in Palo Alto rose 20 per cent in 2011 from 2010 to an average of US $1.63 million.[x]
Although Route 128’s electronics business never regained its former glory, the region has successfully embarked on biotechnology as a new development path. In 2003, it ranked second after San Diego among all the biotech hotbeds in the US, according to a study by the Milkin Institute applying 44 different metrics.
The biotech industry along Route 128 has developed into a true regional network. Back in 1988, most local biotech firms were heavily dependent on formal collaboration with six public research organizations (MIT, Harvard, Tufts, Boston University, Massachusetts General Hospital and the New England Medical Center) and large pharmaceutical firms located elsewhere. At that stage, those local firms were largely isolated from each other. However, by 1998, local biotech firms had begun to work directly with one another, and public research organizations played a less dominant role in the local network.[xi]
Foreign investment in Silicon Valley and Route 128
The reputations of Silicon Valley and Route 128 as advanced electronic technology clusters and proximity to major customers have attracted many foreign firms that have engaged in traditional foreign direct investment as well as indirect investment, e.g., through venture capital funds. In Silicon Valley, foreign direct investment transactions rose from 31 in 1976 to 244 in 1987, with a total of 1,343 transactions and a total value of US $30 billion.[xii] In the Route 128 area, many high-technology firms invested without any government incentives. For example, in 1986, Toshiba opened a procurement office to buy electronic equipment in the Route 128 area.[xiii]
Although many foreign firms initially only located marketing and administrative operations in Silicon Valley and Route 128, an increasing number of foreign firms has opened technology and design centres. For example, some Japanese firms opened semiconductor design centres in Silicon Valley and Route 128 to work closely with their customers and to tailor their products to customers’ needs.[xiv]
Some foreign firms have also invested in Silicon Valley and Route 128 to absorb and develop new knowledge. For example, Samsung built an R&D lab in Silicon Valley in 1983 to develop its own DRAM technology after it failed to license the technology from major American and Japanese chip manufacturers. Samsung hired over 300 experienced Korean engineers from companies such as Intel and IBM, and these engineers led the development of 256K DRAM, trained many Korean engineers and served as information posts to identify and acquire new technologies. Samsung also built a parallel unit in Korea to facilitate the transfer of technology from California to Korea. As a result, Samsung quickly improved its technological position.[xv] Since then, Silicon Valley has continued to gain prominence as a research and development hub, especially within communications-technology research. For example, Huawei, a Chinese maker of communications equipment, set up an R&D office in Silicon Valley in 2011. That year, the firm employed about 430 engineers and technical employees, with the intent to increase staff by more than 120 in the following years.[xvi]
At the same time, telephone handset manufacturer Ericsson was also setting up shop in Silicon Valley. Rather than working from the firm’s headquarters in Sweden, chief technology officer Håkan Eriksson (CTO) chose instead to base himself in Silicon Valley. In 2011, Ericsson had more than 1,200 employees working in R&D, with the intent to continue expanding. Eriksson noted that the location has proved advantageous for facilitating collaboration with other communications giants such as Apple and Google. “The epicenter for the handset industry has shifted from Finland to Silicon Valley,” Mr. Eriksson said, “to be globally competitive, you really need to have a footprint here.”[xvii]
Foreign firms have continued to invest in the two areas, but some noticeable differences exist. In 2003, the number of foreign firms in computer hardware in California almost doubled that of Massachusetts, suggesting that Route 128 has lost some of its lustre in this technology area. However, the percentage of foreign firms in Massachusetts in photonics and pharmaceuticals rose from 10 per cent and 11 per cent in 1997 to 16 per cent and 14–15 per cent in 2003 respectively, indicating that Massachusetts has developed new strengths.[xviii]
Replicating Silicon Valley and Route 128 elsewhere
Aspiring regions have tried to replicate the ‘success formula’ of Silicon Valley and Route 128, by developing key elements such as educational infrastructure and an environment conducive to venture capital investments. These regions (or more specifically publicly funded, regional development authorities) believe that the presence of universities will spark local start-ups and thus drive regional economic development.
However, it remains unclear whether intensifying industry–university relationships, similar to those in Silicon Valley and Route 128, is likely to create new clusters. Recent research[xix] suggests that during the formative years of Silicon Valley Stanford University worked with both local startups and established firms, but Stanford’s critical contribution was its linkages with established firms headquartered elsewhere. Further, Stanford did not proactively sponsor local start-ups in those formative years. Rather, the main driver of industry–university programmes was to bring money to the university from the deep pockets of business firms. Similar to Stanford University, neither MIT nor Harvard pursued an intended strategy to promote industry concentration or local growth on Route 128.[xx]
In the case of Silicon Valley, Stanford established four major outreach programmes between 1945 and 1965, including the Stanford Research Institute (SRI), the Stanford Industrial Park, the Honors Cooperative Program and the Industry Affiliates Programs. Most participants in all four programmes were established firms. Take, for example, the Stanford Industrial Park – the Stanford programme with the greatest degree of local involvement. Of the 30 tenants in 1960, only 9 were local high-tech firms. Nine were branches of established firms (including GE), four were publishers and one was a book distributor, with the remainder including an architect, a bank, a mining company and a realtor. Moreover, the main purpose of building the park was to lease endowed land to bring revenue to the university, rather than to sponsor local start-ups.
Therefore, in the formative years of Silicon Valley, the main role of Stanford University was to connect with established firms located elsewhere, rather than to promote or incubate high-tech firms in the region. Other regions aiming to replicate Silicon Valley’s ‘success formula’ should thus be very careful when attempting to copy what they think constitutes the heart of the clustering success. These regions should not assume that, e.g., industry–university cooperation is the key driver for successful industrial clustering.
QUESTIONS
- Do Silicon Valley and Route 128 function as diamond-based clusters? Why or why not? What are their similarities and what are their differences?
- What caused the development, decline and resurgence of the two regions? What is your understanding of the different explanations for the two regions?
- How have Silicon Valley and Route 128 helped the establishment of start-ups?
- Can the ‘success formulas’ provided by Silicon Valley and Route 128 be adopted as templates for other, would-be clusters?