Page 23
Introduction
For the federal government, a significant task of the new century is realigning the missions of its major science and technology laboratories. As the nation's attention has turned from the tensions of the Cold War to an intensely competitive global environment, federal agencies are challenged to find productive new ways to utilize their highly skilled human resources and their extensive—sometimes unique—physical assets.
This task is changing the strategies of federal agencies in fundamental ways. Within NASA, for example, the space agency's mission now includes an increased emphasis on commercializing technologies that are developed by agency scientists and engineers. And the development of tomorrow's technologies has shifted, in the words of NASA's chief scientist, toward “highly complex, first-of-a-kind missions which cannot be accomplished or afforded using current systems.”1
To execute that shift, the agency plans to realign its own research mix by focusing on three cutting-edge areas—biotechnology, nanotechnology, and information technology. Its strategic challenge is to integrate those research areas into a new “mission triangle” for the 21st century.
A NEW STRATEGIC PLAN FOR AMES
As a major participant in this mission, the Ames Research Center, at Moffett Field, California, has undertaken a major effort to develop its own strategic plan.2
1 See the remarks of Dr. Samuel Venneri, NASA's Chief Technologist, in the Proceedings, Panel I.
2 For an overview of the Ames proposal, see the White Paper submitted by NASA in Annex A. For additional information on the park concept, see the presentations by NASA's Sam Venneri in Panel I, William Berry in Panel III, and Robert Norwood in Panel IV.
Page 24
Ames' objective is to make use of its extensive human and physical resources in ways that are both consistent with NASA's overall mission goals and effective at leveraging its own particular research capabilities and exceptional location.
The crux of Ames' strategic plan is the formation of a “research park”3 at the Ames facility that will feature a network of partnerships with private research firms and major universities. Ames brings to that plan its own substantial expertise in each of the three mission triangle fields, as well as 2,000 acres of land that includes undeveloped space and unused structures. Not the least of the park's assets is its location in the heart of Silicon Valley.
In recent years, federal agencies have tried various strategies to launch their technologies into the marketplace, with varying success. The plan to create onsite partnerships is interesting for several reasons. Most obviously, it binds NASA directly and proximately with private firms that have intimate knowledge of the high-technology markets.4 Equally important, the plan avoids generally ineffective “technology push” techniques in favor of projects that will be driven by market needs. Ames' favorable location can facilitate gaining accurate knowledge of those needs.
Doing more within existing budgets has become the hallmark of NASA over the last decade. In keeping with this imperative, Ames intends to expand its mission and promote commercialization without significantly increased budgetary support. As described in the Proceedings by Center Director Henry McDonald and Deputy Director William Berry, new research efforts would be financed by both leasing revenues from partners and from the profits of partnership enterprises. At the same time, its existing Commercial Technology Office is prepared to assist in the recruitment of partners whose work is consistent with NASA's mission and to expand its existing mission of transferring technology to the private sector.
3 “Research park” is the most common designation for an association of enterprises that focus on research and development. However, “research,” “science,” and “technology” park are used more or less interchangeably in the U.S. and Canada; the terms “science park” and “technopole” are more common in Europe and Asia. See the analysis of Michael Luger in this volume.
4 As Linda Cohen and Roger Noll point out, one of the strengths of industry-led collaborations, such as Sematech, is the market savvy of participating firms. Linda R. Cohen and Roger G. Noll, The Technology Pork Barrel, Washington, D.C.: The Brookings Institution, 1991, chapter 12. Peter Grindley, David Mowery, and Brian Silverman make a similar point, noting that the consortium's goals changed over time, reflecting the changing perceptions of its members' needs. This operational flexibility is probably essential in an industry evolving as rapidly as the semiconductor industry. See Peter Grindley, David Mowery, and Brian Silverman, “SEMATECH and Collaborative Research: Lessons in the Design of High-Technology Consortia,” Journal of Policy Analysis and Management, 13(4), 1996.
Page 25
RESEARCH PARKS
Goals for research parks vary with local conditions, the goals of their organizers, and the assets and resources available to the enterprise. As Luger and Goldstein's research has made clear, the objectives of private research park developers are often not the same as those of public sector entities.5 The concept of the research park is a modern phenomenon; most of them have been initiated in the last two decades. Research parks are defined by Luger and Goldstein as “organizational entities that sell or lease spatially contiguous land and/or buildings to businesses or other organizations whose principal activities are basic or applied research or development of new products or processes.”6 This concept evolved from the older industrial parks, whose first incarnation was the Central Manufacturing District of Chicago, established in 1905. The modern prototype of the research park appeared after World War II with the creation of Menlo Park, California, in 1948. The most successful parks to date were developed in the 1950s and 60s: Stanford Industrial Park in 1953, in northern California; Research Triangle Park in 1958, in central North Carolina; and Waltham Industrial Center in 1954, associated with other developments on Route 128 west of Boston.7
By 1980 some 20 parks had been formed—enough to attract the attention of regions across the country. During the next 10 years the number of parks increased five-fold. Since 1990, however, reductions in government and industry spending have slowed this pace of growth considerably. Between 1990 and 1995, the number of park formations dropped to 30, a more cautious pace that has continued to the present. Still, every state in the U.S. has at least one research park, and most have two or three.8
5 Michael I. Luger and Harvey A. Goldstein, Technology in the Garden: Research Parks & Regional Economic Development, Chapel Hill: University of North Carolina Press, 1991 See also the paper by Michael Luger, “Science and Technology Parks at the Millennium” in this volume. Luger and Goldstein note, for example, that Research Triangle Park was initially a state effort, whose goals were to attract high-technology industry, and thereby create jobs and improve the perception of the state as a high-tech research center. The objectives of parks created or controlled by the private sector tend to put profit and occupancy ahead of more lofty developmental—but often unprofitable—goals. It is for this reason that many research parks go “down-market” from exclusively R&D activities to manufacturing, assembly, and distribution. Reflecting this evolution, Luger observes that one quarter of the parks he reviewed failed as real estate projects and one half of the remainder changed their focus to remain viable.
6 Luger and Goldstein, Technology in the Garden, op.cit., p. 5.
7 Denise Drescher, Research Parks in the United States: A Literature Review, Department of City and Regional Planning. University of North Carolina at Chapel Hill, April 13, 1998. (www.unc.edu/drescher/litrev.htm).
8 These figures are from the Association of University-Related Research Parks, which was founded in 1986. See www.aurrp.org.
Page 26
OBJECTIVES: JOBS, GROWTH, AND SYNERGIES
Overall, most “research parks” have been initiated by real-estate developers who saw the development of research capabilities as only one of several objectives. The three most common reasons to develop research parks are to create jobs, to raise a region's status by bringing in high-growth industries, and to create synergies between different firms. The country's largest park, Research Triangle Park (comprising some 5,000 acres), was established under Governor Luther Hodges for the purpose of creating jobs for the college-trained youth of North Carolina.9
A significant number of parks are collaborations or partnerships between the public and private sectors. These are usually promoted by regional or local governments, often in association with nearby universities and with private companies that occupy the resulting space.10 Public development money is often involved, and the mission statements of many parks include active participation in technology transfer to the private sector and participation in the economic development of their cities, regions, and states.11 Some parks, like University Heights Science Park in Newark, New Jersey, have strong urban renewal missions.12 Some of these, like Kendall Square in Cambridge, Massachusetts, are tightly clustered around a single university (in this case, the Massachusetts Institute of Technology).13
AMES ADVANTAGES
NASA is following an approximate model that has become well established as a means of advancing the objectives of research and technology transfer. It is typical in being affiliated with one or more universities; it is unusual, though by no means unique, in locating on the site of a national laboratory. The Ames park is also unusual in several other respects: 1) It has the ability to construct its own campus on-site; 2) it is located adjacent to Silicon Valley, the world's largest concentration of high-tech firms and entrepreneurs;14 and 3) it has, in advance of
9 Drescher, Research Parks in the United States, op. cit., p. 2.
10 For a description of research parks from a real estate perspective, see Rachelle Levitt, ed., The University/Real Estate Connection: Research Parks and Other Ventures, Washington, D.C.: Urban Land Institute, 1987.
11 Thomas W. Durso, "Home-Grown R&D," The Scientist. 1, July 8, 1996.
12 Rachell Garbarine, "Newark's Science Park Takes Another Step Forward," The New York Times on the Web, Nov. 23, 1997.
13 Some 70 biotech firms are located within five miles of Kendall Square, a recently renovated site of urban decay. Carey Goldberg, "Across the U.S., Universities Are Fueling High-Tech Economic Booms," The New York Times on the Web, Oct. 8, 1999.
14 Ross C. DeVol et al., America's High-Tech Economy: Growth, Development, and Risks for Metropolitan Areas, Santa Monica, CA: Milken Institute, July 13, 1999 (www.milken-inst.org ). DeVol uses a measure of "high-tech spatial concentration" to describe "Tech-Poles" through the country. In his evaluation, the Tech-Pole of Silicon Valley ranks more than three times higher than its closest competitor. "As a Tech-Pole," DeVol writes, "the gravitational pull of the San Jose metro area, home to Hewlett-Packard, Applied Materials, Sun Microsystems, Intel, Cisco Systems, Oracle, and Silicon Graphics, is unparalleled." p. 6.
Page 27
construction, already secured both academic and industrial partners of considerable resources and reputation.15.
MANY GOALS—DIFFERENT METRICS
The success of research parks is difficult to measure. As suggested above, they vary a great deal in terms of objectives, types of facilities, and supporting institutions. Reflecting the many goals and different circumstances, there is a corresponding lack of agreement on metrics. There is little research or quantitative evidence that accounts for failed research parks.16
The comprehension of the research park phenomenon has also been blurred by political, ideological, and business biases, in that each participant in the development of a park has a particular definition of success.17 Parks are often seen as cure-alls by developers and local governments, who may hope for job generation, income growth, greater income equality, expanded opportunities for certain groups within the labor force, and economic restructuring of the region.18
Park developers often describe beneficial changes as results of park creation, but the park's contribution to change is difficult to gauge. For example, many of the jobs attracted to a park might have come to the region in the absence of a park. Conversely, jobs created outside a park may be a function of the park's existence. Finally, costs as well as benefits must be examined to determine “success,” including indirect expenditures on land acquisition and infrastructure development, tax expenditures from financial inducements used by government, and the opportunity cost of land used versus other types of uses.19
15 Ames has planned university partnerships with Carnegie Mellon University and the University of California at Santa Cruz, which bring considerable strengths in science and techonology, and an industrial partnership with Lockheed Martin (one division of which abuts the Ames property). It has also discussed potential partnerships with high-tech firms which have expressed interest in participating. See the presentation in this volume of M. R. C. Greenwood, Chancellor of the University of California at Santa Cruz.
16 Drescher, Research Parks in the United States, op. cit., p. 4.
17 Ibid., p. 4.
18 Luger and Goldstein (Technology in the Garden, op. cit., p. 34) write: "One of the conceptual difficulties is that there is no consensus about the definition of success... The most commonly cited goals relate to economic development. But both the literature and our data from interviews with park developers, elected officials, university administrators, business leaders, and others confirm the existence of other goals, including technology transfer, land development, and enhancement of the research opportunities and capacities of affiliated universities."
19 Ibid, p. 35.
Page 28
To address the metrics gap, Luger and Goldstein have developed multiple approaches to measure success. One is to evaluate performance against stated goals, as written into legislation and found in documents and interviews. They also use a multiple case-study approach, along with a quasi-experimental design, comparing areas with parks to similar areas without parks. Overall, they report that about half of all parks do not succeed, and of the remainder, half shift their focus from research to become office, industrial, or mixed-use parks.20
Box B. The Notion of Success
“The overall policy lesson we have drawn from this analysis is that in many regions research parks by themselves will not be a wise investment. The success rate among all announced parks is relatively low. . . Research parks will be most successful in helping to stimulate economic development in regions that already are richly endowed with the resources that attract highly educated scientists and engineers.” —Luger and Goldstein, Technology in the Garden, p. 184. |
For those parks that succeed, Luger and Goldstein and others conclude that the most notable consequence of developing a park is likely to be induced growth of R&D activity. R&D businesses are likely to cluster within the regions in order to share a specialized labor force, university facilities and expertise, business services, a certain type of social and cultural environment, and access to technical and market information. Consequently, “once a region ‘takes off with a successful research park, it should continue to experience growth in the R&D sector.”21
CLUSTERS AND GROWTH: THE ROLE OF HIGH-TECHNOLOGY RESEARCH
The suggestion that parks can induce the growth of R&D activity is mirrored by evidence that high-tech activity stimulates regional growth. The Milken Institute study referred to above compared the economic success of areas with those areas' concentration of high-tech research. The authors concluded that high-tech activity explained 65 percent of the difference in economic growth among various metropolitan regions during the 1990s, and that “research centers and institutions are undisputedly the most important factor in incubating high-tech industries.”22 There is also evidence that “. . . high-technology firms are associated
20 Ibid, p.1.
21 Ibid, p. 22.
22 DeVol, et al., America's High-Tech Economy, op. cit., p. 5.
Page 29
with innovation. . . and hence gain market share, create new product markets, and use resources more productively.” Such firms tend to perform larger amounts of R&D than more traditional industries and create positive spillover effects that benefit other commercial sectors by generating new products and processes.23
RISKS
The process of creating a new research park carries substantial risks. Many (perhaps most) parks have yet to recruit the expected number of tenants, and there is no single formula by which one can avoid these risks. An official of the AUURP commented that many parks have made the error of assuming that big companies would move into new park spaces—just as IBM became the first tenant of Research Triangle Park. More often, a park has to “grow its own” tenants, a process that requires time.24
Certain risks are inherent to the R&D universe. High-tech tenants are influenced by business cycles that traditionally experience sharp swings. Such tenants are also vulnerable to cuts in federal or industry spending, especially because long-term research is often seen by budget planners as discretionary.25
ADVANTAGES OF CO-LOCATION
One feature that favors the success of parks and attracts tenants is the co-location of participants. Many firms, industries, and regions that are successful have formed collaborative relationships with other firms, agencies and universities to leverage the benefits of cooperation.26 Such benefits already characterize the regional Silicon Valley network around Ames, which have been described as more “flexible, technologically dynamic, and tolerant of failure” than regions in which experimentation and learning are confined to individual firms.27
23 The report adds that private innovators obtain a rate of return in the 20-30 percent range with the spillover (or social return) averaging about 50 percent. Positive spillovers are often locally concentrated. National Research Council, Conflict and Cooperation in National Competition for High Technology Industry, Washington, D.C.: National Academy Press, 1996, pp. 33-35.
24 Durso, “Home-Grown R&D,” op. cit. p. 4. Also, Luger and Goldstein observe that it may take a decade or more for a park to mature or “succeed” (Technology in the Garden, op. cit., p. 44).
25 DeVol, et.al., America's High-Tech Economy, op. cit., p. 9.
26 Jane Fountain writes that social capital includes shared resources, shared staff and expertise, group problem-solving, multiple sources of learning, collaborative development, and diffusion of innovation, all of which are abundantly present in Silicon Valley. She suggests that social capital is as important as human and physical capitals. Jane Fountain, “Social Capital: A Key Enabler of Innovation,” in Lewis M. Branscomb and James H. Keller, eds., Investing in Innovation: Creating a Research and Innovation Policy That Works, Cambridge, MA: MIT Press, 1998, pp. 85-111.
27 Annalee Saxenian, Regional Advantage: Culture and Competition in Silicon Valley and Route 128, Cambridge, MA: Harvard University Press, 1994, p. 161.
Page 30
Box C. Potential Risks and Guidance for Parks
A review of recent literature on the federal role in technology development reveals a list of potential risk factors that could jeopardize the success of the Ames Research Park. Among them:
—Drescher, Research Parks in the United States, pp. 1-6. |
Another advantage of Ames' location is that the presence of business expertise and facilities allow for the incubation of young businesses. Many parks today support their own incubators, which are designed to reduce business risk for researchers-turned-entrepreneurs by providing many functions: assisting young companies to use technology for economic development; moving discoveries from the lab to the marketplace; locating support services; and obtaining consultation, funding assistance, flexible leases, and office services.28
28 Durso, “Home-Grown R&D,” op. cit., p. 2.
Page 31
Box D. Regional Advantage in a Global Economy
“Paradoxically, regions offer an important source of competitive advantage even as production and markets become increasingly global. Geographic proximity promotes the repeated interaction and mutual trust needed to sustain collaboration and to speed the continual recombination of technology and skill.” —Saxenian, Regional Advantage, p. 161 |
EDUCATION
Ames has also addressed in its strategic plan a pressing local and regional need—the education of tomorrow's science and engineering work force and the retraining of today's. Ames, with its educational and industrial partners, has planned an ambitious effort of outreach and on-site programs. As one scholar has written, “The greatest long-term threats to the Silicon Valley economy are . . . continued reductions in public funding for educational institutions—from its elementary and secondary schools to the sophisticated network of community colleges, state universities, and the University of California system—that jeopardize the rich supply of technical talent and the research base that have historically supported the regional economy.”29
The editors of a leading treatise on technology have stated the challenge from a national perspective, describing our “inadequate technical and general education and inadequate retraining at all levels” as a significant obstacle to U.S. growth. “Needed,” they write, “are education and retraining that can inspire a positive outlook toward science and technology, and an urge to maintain the American edge in technological competitiveness and entrepreneurial creativity.”30
A REQUEST FOR ANALYSIS
To stimulate a full assessment of the issues relevant to this initiative, NASA's Administrator and the leadership of the Ames Research Center asked the Board on Science, Technology, and Economic Policy (STEP) at the National Research Council to convene a symposium to review the initiative.31 Given the prominence
29 Saxenian, Regional Advantage, op. cit., p. ix.
30 Nathan Rosenberg, Ralph Landau, and David C. Mowery, Technology and the Wealth of Nations, Stanford: Stanford University Press, 1992, p. 13.
31 For additional background, see the Preface.
Page 32
of Ames as a federal presence in the northern California region and the park's relevance to the Board's current review of U.S. government-industry partnerships, the STEP Board was pleased to respond positively to NASA's request. In particular, the Board sought to address the following issues:
-
Balancing Objectives: How best to manage government-industry partnerships to accomplish NASA mission goals effectively and continue commercially relevant research, while at the same time properly managing— and balancing—access to federal facilities;
-
Developing Evaluation Metrics: The importance of developing reasonable and accurate metrics to assess successes and failures in a complex, long-term undertaking such as a research park. Such metrics, of course, hold major interest for the managements of both Ames and its parent agency, NASA;
-
Ames' Interaction with the Private Sector: The research park initiative as an element of Ames' interaction with the U.S. economy, especially small business and new start-ups, through its extensive supplier networks, existing partnerships, and expanding cooperation with industry.
THE REVIEW OBJECTIVE
These issues and the assessment challenges they entail are of great interest to policymakers and consequently to the National Academies' study of Government-Industry Partnerships. To consider the issues associated with the Ames initiatives in depth, the STEP Board organized a symposium to which it invited top members of the Ames and NASA management, leading academic experts on research parks, senior executives from the private sector, entrepreneurs in high-technology enterprises, experts from the world of private investment, and key congressional staff for an informal, but informed, dialogue.32 This exchange enabled the Ames leadership to articulate its objectives for the park and benefit from the experience of the Committee through questions, comments, and occasional caution. It was recognized by all that the Ames S&T park is a work in progress and therefore one which could benefit immensely from an informed discussion of its objectives and likely challenges.
32 A full list of participants is included as Annex C.
