Impact of Photonics on the National Economy
The vast diversity of applications enabled by photonics poses both economic promise and policy challenges. On the one hand, technical advances in fundamental principles of photonics may have broad impacts in many applications and economic sectors. On the other hand, this diversity means that monitoring public and private investment, employment, output, and other economic aspects of photonics is difficult. Photonics is a broad technology rather than an industry, and the economic data assembled by U.S. government agencies do not support a straightforward assessment of the “economic impact” of photonics. For example, there are no North American Industry Classification System (NAICS) codes that enable the tracking of revenue, employment, and industrial research and development (R&D) spending in photonics-related fields, and we lack data on government R&D spending in photonics. The absence of such information reduces the visibility of photonics within the industrial community and impedes the development of more coherent public policies to support the development of this constellation of technologies and applications.
This chapter takes the following form: First, a case study of lasers is used to introduce the field of photonics, and the conceptual challenges of developing estimates of the economic impact of photonics innovations are discussed. Next, company-level data are presented, and the challenges associated with using such data to provide indicators of the economic significance of the “photonics sector” within the U.S. economy are addressed. Next is a discussion of sources of R&D
investment within photonics, including government and company funding of R&D, followed by an examination of the ways in which the changing structure of the innovation process within photonics (including sources of R&D funding) reflects broader shifts in the sources of innovation within the U.S. economy. That section motivates the subsequent discussions of the role of venture-capital finance in photonics innovation, the role of university licensing, and the implications of offshore growth in the production of optics and photonics products for innovation in the field. This discussion of the changing structure of innovation finance and performance in the United States leads to the next section, which considers the implications of recent experiments in public-private and inter-firm R&D collaboration in other high-technology sectors for the photonics sector. Finally, conclusions and recommendations are presented.
THE ECONOMICS OF PHOTONICS: A CASE STUDY OF LASERS
The laser is a central technology within photonics, and a brief history of its development and expanding applications provides some insights into the economic effects of the much broader field of photonics, as well as underscoring the difficulties of measuring the economic impact of such a diverse field. First demonstrated in 1960 by Theodore Maiman of Hughes Aircraft, the laser built on fundamental research on microwave technology by Charles Townes and Arthur Schawlow at Columbia University and Bell Labs, respectively. The laser exhibits many of the characteristics of a “general-purpose technology”1 (other examples include information technology [IT], steam power, and electrical power), in that laser technology itself has been transformed by a series of important innovations, with numerous new types of lasers developed over the past 50 years. Innovations in lasers have broadened the applications of this technology, many of which have produced dramatic improvements in the performance of technologies incorporating
1 Rosenberg, N., and M. Trajtenberg. 2004. A general-purpose technology at work: The Corliss steam engine in the late-nineteenth-century United States. Journal of Economic History 64:61-99. In this paper, Rosenberg and Trajtenberg highlight four characteristics of a “general-purpose technology” (GPT): “first, it is a technology characterized by general applicability, that is, by the fact that it performs some generic function that is vital to the functioning of a large number of using products or production systems. Second, GPTs exhibit a great deal of technological dynamism: continuous innovational efforts increase over time the efficiency with which the generic function is performed, benefiting existing users, and prompting further sectors to adopt the improved GPT. Third, GPTs exhibit ‘innovational complementarities’ with the application sectors, in the sense that technical advances in the GPT make it more profitable for its users to innovate and improve their own technologies. Thus, technical advance in the GPT fosters or makes possible advances across a broad spectrum of application sectors. Improvements in those sectors increase in turn the demand for the GPT itself, which makes it worthwhile to further invest in improving it, thus closing up a positive loop that may result in faster, sustained growth for the economy as a whole” (p. 65).
lasers (e.g., fiber-optic communications). Over the course of the 50 years since its invention, the laser has been used in applications ranging from communications to welding to surgery.
The Economic Impact of the Laser
One measure of the economic impact of the laser is provided by Baer and Schlachter’s 2010 study for the Office of Science and Technology Policy (OSTP),2 which compiled data on the size of three economic sectors in which lasers have found important applications. Baer and Schlachter listed these as follows:
• Transportation (production of transport equipment, etc.), estimated to account for $1 trillion in output during 2009-2010;
• The biomedical sector ($2.5 trillion); and
• Telecommunications, e-commerce, information technology ($4 trillion).
The value of lasers deployed in each of these three sectors was respectively estimated at $1.3 billion (CO2 and fiber), $400 million (solid-state and excimer lasers), and $3.2 billion (diode and fiber lasers).
It is important to distinguish between the role of lasers as “enabling” the growth of these three sectors and the role of this technology as “indispensable” to these sectors, because the distinction is central to analyses of the economic impact of any new technology. The fundamental question that arises in this context is, What would have happened in the absence of the laser? That is, what if substitutes had been employed to realize some if not all of the benefits associated with the laser’s applications in these sectors? What would have been the cost (both in terms of higher prices and reduced functionality) associated with using non-laser substitutes? In some areas (e.g., surgery, some fields of optical communication), substitutes might well have been unavailable or would have performed so poorly as to render them useless. In other fields such as welding, however, substitutes for lasers that presented fewer cost and performance penalties might well have appeared. In some cases, substitutes for lasers might well have improved their performance and reliability over time.
In the case of the laser as with most major innovations, the data and the methodology necessary to conduct counterfactual thought experiments of this sort are lacking, which makes it difficult to develop credible estimates of economic impact. These analytic challenges are no less significant in assessing the impacts of other
2 Baer, T., and F. Schlachter. 2010. Lasers in Science and Industry: A Report to OSTP on the Contribution of Lasers to American Jobs and the American Economy. Available at http://www.laserfest.org/lasers/baer-schlachter.pdf. Accessed June 25, 2012.
photonics technologies currently in use, and they are truly forbidding where one seeks to predict the economic impact of future applications that have only begun to emerge.
Nonetheless, it seems clear that the laser has been adopted in a diverse array of applications, some of which have underpinned the growth of entirely new methods for the transmission of information.3 Equally important is the way in which continued innovation in laser technology has enabled and complemented innovation in technologies using lasers. This mutual enhancement further extends the adoption of these applications as performance improves and costs decline. Moreover, the appearance of new applications and markets for lasers has created strong incentives for further investment in innovation in lasers. All of this feedback and self-reinforcing dynamics are classic features of general-purpose technologies. Lasers are one example of such a technology within the field of photonics.
Funding of Early Laser Development
The development of laser technology shares a number of characteristics with other postwar U.S. innovations, in fields ranging from information technology to biotechnology. Like these other technologies, much of the research (especially the fundamental research) that underpinned the laser and its predecessor, the maser, relied on federal funding. Similar to the experience with IT, much of this federal R&D funding was motivated by the national security applications of lasers during a period of high geopolitical tension.4 Industry funded a considerable amount of laser-related R&D, much of which focused on development and applications, but
3 Interestingly, optical communication was the only foreseen application of the laser in 1958. See, for example, Sette, D. 1965. Laser applications to communication. Zeitschrift für angewandte Mathematik und Physik ZAMP 16(1):156-169.
4 Bromberg, J.L. 1991. The Laser in America, 1950-1970. Cambridge, Mass.: MIT Press. In this study, Bromberg emphasizes another characteristic of federally and industrially financed R&D in the field of lasers: the extent of linkage among research and researchers in U.S. industry, federal laboratories, and academia during the 1945-1980 period: “Academic scientists were linked to industrial scientists through the consultancies that universities held in large and small firms, through the industrial sponsorship of university fellowships, and through the placement of university graduates and postdoctoral fellows in industry. They were linked by joint projects, of which a major example here is the Townes-Schawlow paper of [sic] optical masers, and through sabbaticals that academics took in industry and industrial scientists took in universities. Academic scientists were linked with the Department of Defense R&D groups, and with other government agencies through tours of duty in research organizations such as the Institute for Defense Analyses, through work at DoD-funded laboratories such as the Columbia Radiation Laboratory or the MIT Research Laboratory for Electronics, and through government study groups and consultancies. They were also linked by the fact that so much of their research was supported by the Department of Defense and NASA” (p. 224). Similar linkages among industry, government, and military research characterized the early years of development of the U.S. computer and semiconductor industries, in contrast to their European and Japanese counterparts.
much of this R&D investment (particularly in the early years of the laser’s development) was motivated by the prospect of significant federal procurement contracts for military applications of lasers.
The early work in the 1950s of Townes at Columbia University on masers, for example, was financed in large part by the Joint Services Electronics Program, a multi-service military R&D program that sought to sustain after 1945 the wartime research activities of the Massachusetts Institute of Technology (MIT) and Columbia Radiation Laboratories, both established during World War II. Military funding supported early work on masers and lasers at RCA, Stanford University, and Hughes Aircraft. R&D related to lasers at the National Bureau of Standards also was closely overseen by military representatives. By 1960, according to Bromberg,5 the Department of Defense (DOD) was investing roughly $1.5 million (1960 dollars) in extramural R&D on lasers, an amount that rose rapidly after Maiman’s demonstration of the ruby laser at Hughes; by 1962, according to Bromberg’s estimates, the DOD was spending roughly $12 million on laser-related R&D, one-half of the total U.S. R&D investment in the technology. In 1963, total DOD R&D investment, including intramural projects, approached $24 million, which increased to just over $30 million in the late 1970s.6 Another tabulation of military R&D investment estimates total military laser-related R&D spending through 1978 at more than $1.6 billion (all amounts in nominal dollars).7
The military also was a major source of demand in the early laser industry, although its share of the market declined over time as civilian applications and markets grew rapidly. According to Bromberg, the DOD share of the laser market fell from 63 percent in 1969 to 55 percent in 1971.8 Although the DOD dominated the government market for lasers, other federal agencies also were important purchasers, and Seidel estimates that total government purchases of lasers amounted to nearly 56 percent of the total market in 1975, increasing to slightly more than 60 percent by 1978.9 Commercial laser sales grew from $1.985 billion in 1983 to
5 Bromberg, J.L. 1991. The Laser in America, 1950-1970. Cambridge, Mass.: MIT Press.
6 Koizumi, K. 2008. AAAS Report XXXIII: Research & Development FY 2009, Chapter 5. Available at http://www.aaas.org/spp/rd/09pch5.htm. Accessed July 30, 2012.
7 Seidel, R. 1987. From glow to flow: A history of military laser research and development. Historical Studies in the Physical and Biological Sciences 18:111-147.
8 Bromberg, J.L. 1991. The Laser in America, 1950-1970.
9 Seidel, R. 1987. From glow to flow: A history of military laser research and development. Historical Studies in the Physical and Biological Sciences 18:111-147.
$2.285 billion in 1984, according to DeMaria;10 government sales in these same 2 years amounted to $1.23 billion and $1.3 billion, respectively. The government share of laser sales almost certainly has continued to decline in more recent years.
The dominance of the early laser market by the military services had important implications for the development of the embryonic laser industry. In contrast to the military services of other NATO member nations, U.S. military procurement officials rarely excluded new firms from procurement competitions, although in many cases these firms had to arrange for a “second source” of their products to avoid supply interruptions. The prospect of military procurement contracts therefore attracted new firms to enter the laser industry and underlaid a growth in the total number of firms working with laser development. The number of new firms in the industry also grew rapidly because of the growth of new laser applications in diverse civilian markets, as well as the growth of new types of laser technologies. Clearly, the military contracts sped up the laser development. The appearance of the diode-pumped solid-state laser in 1988, however, may have triggered the exit from the industry of a large number of firms, and the number of active firms fell to 87 by 2007, during a period of rapidly increasing sales for the industry as a whole.
Although data allowing for a comparison of the structure of the laser industries of the United States and other nations are not readily available, it is likely that the number of independent producers of lasers in other nations exhibited less dramatic growth and decline. Assuming that this characterization of the laser industries of the United States and other nations is accurate, the differences reflected the prominent role of government demand for lasers in the United States, as well as the important role of U.S. venture capital in financing new-firm entrants into the laser industry.
The origin of U.S. and Japanese scientific publications appearing in Applied Physics Letters from 1960 through 2009 on the topic of semiconductor lasers was analyzed by Shimzu (2011);11 data in the study by Shimzu suggest a contrast in the sources of leading-edge laser R&D during this period. Established U.S. firms in the areas of electronics, IT, and communications dominated semiconductor-laser publications during 1960-1964, accounting for more than 80 percent of publications of U.S. origin. These firms’ share of publications dropped sharply after 1964, to 30 to
10 DeMaria, A.J. 1987. “Lasers in Modern Industries.” In Lasers: Invention to Application, J.H. Ausubel and H.D. Langford, eds. Report of the National Academy of Engineering. Washington, D.C.: National Academy Press.
11 Shimuzu, H. 2011. Scientific breakthroughs and networks in the case of semiconductor laser technology in the U.S. and Japan, 1960s-2000s. Australian Economic History Review 51:71-96.
35 percent during 1965-1974, before increasing again to 61 percent during 1975-1979 and 46 percent during 1980-1984. By 2005-2009, however, the established-firm share of U.S. scientific publications in semiconductor lasers had dropped to less than 5 percent. Start-up firms, which contributed no semiconductor-laser publications during 1960-1980, accounted for more than 10 percent of publications during 1985-1989 and 9.25 percent during 2005-2009. U.S. university-based researchers accounted for the majority of U.S. semiconductor-laser publications throughout the 1965-2009 period, as their share grew from slightly more than 57 percent in 1965-1969 to almost 85 percent during 2005-2009. (See Figure 2.1.)
The data in the Shimuzu study on publications of Japanese origin in semiconductor-laser research in Applied Physics Letters12 indicate a minimal role for start-up firms as sources of research. Although all papers published in this prestigious journal are reviewed by scientific peers, the burden of translation into English may well introduce some bias into this comparison—papers of Japanese origin effectively have to clear a higher “quality threshold” to appear in this journal. This potential source of bias should be kept in mind in comparing Japanese and U.S. publications. Established Japanese corporations, which accounted for no scientific papers during 1960-1969 (see Figure 2.1), contribute a declining share of scientific papers of Japanese origin in semiconductor lasers, although their share declined somewhat less significantly, from 75 percent during 1970-1974 to nearly 40 percent during 2005-2009. Japanese start-up firms, however, played almost no role as a source of scientific publications, appearing only after 2000 in Shimuzu’s data, with a share of 0.74 percent during 2000-2004 and 2.94 percent during 2005-2009. Japanese universities, which accounted for less than 30 percent of papers of Japanese origin in this journal and field before 1990, by 2005-2009 contributed more than 55 percent. (See Figure 2.1.)
Although covering only one area of laser technology and limited to one scientific journal, the data analyzed by Shimuzu clearly indicate that new-entrant firms in the United States accounted for a much larger share of scientific activity (as represented by publications) in semiconductor lasers than was true of Japanese start-ups, whereas established Japanese firms have maintained a more prominent role as sources of scientific publications into the 21st century than have U.S. established firms in electronics, communications, or IT. The role of university researchers as sources of published scientific research, however, appears to have grown significantly in both nations, albeit more dramatically in the United States than in Japan.
Figure 2.1 Comparison between the United States and Japan with respect to different sectors’ contributions to scientific publications that appeared in Applied Physics Letters from 1960 through 2009 on the topic of semiconductor lasers. (In the set of two bars for each time period, U.S. data are on the left, and Japanese data are on the right.) SOURCE: Based on data in Shimuzu, H. 2011. Scientific breakthroughs and networks in the case of semiconductor laser technology in the U.S. and Japan, 1960s-2000s. Australian Economic History Review 51:71-96.
Conclusions from the Laser Case Study
This brief overview of the development of laser technology and the U.S. laser industry highlights several issues that are relevant to overall photonics technology. The difficulty of measuring the “economic impact” of lasers reflects the need for any such assessment to rely on assumptions about the availability or timing of the appearance of substitute technologies. These difficulties are more serious for predictions of the economic impact of technologies currently under development. Such predictions rely on guesses about the nature of substitutes and markets, as well as predictions concerning the pace and timing of the adoption of new technologies. The laser’s development also highlights several of the features of general-purpose technologies, namely, their widespread adoption, driven in many cases by continued innovation and improvement in the focal technology, as well as the ways in which users of the technology in adopting sectors contribute to new applications that rely in part on incremental improvements in the technology. In the view of the committee, many other technologies in the field of photonics share these characteristics with lasers.
The development of laser technology and the laser industry in the United States also displays some contrasts with the experiences of other nations, particularly in the important direct and indirect role played by the federal government in the early stages of the technology’s development. Federal R&D and procurement spending, much of which was derived from military sources, influenced both the pace of development of laser technology and the structure of the laser industry, revealed most plainly in the contrasts between U.S. and Japanese scientific publications in laser technology. Moreover, the high levels of mobility of researchers, funding, and ideas among industry, government, and academia were important to the dynamism of the U.S. laser industry in its early years, with few formal policies geared toward “technology transfer” between government or university laboratories and industry such as those in place today. Although military R&D spending continues to account for roughly 50 percent of total federal R&D spending (which now accounts for roughly one-third of total national R&D investment, down significantly from its share during the period of laser-technology development),13 the share of long-term research within the military R&D budget has been under severe pressure in recent years, and congressional restorations of executive branch cuts in this spending have often taken the form of earmarks. Moreover, as the laser industry matures and nonmilitary markets exert much greater influence over the evolution of applications for this technology, the ability of military R&D to guide broad technological advances in this field has declined.
ESTIMATING THE ECONOMIC IMPACT OF PHOTONICS—INDUSTRY REVENUES, EMPLOYMENT, AND R&D INVESTMENT IN THE UNITED STATES
This section employs estimates of revenues, employment, and R&D investment for a sample of firms that are active in the field of photonics to illustrate the breadth of photonics-based industrial activity in the U.S. economy.
The data were provided by the International Society for Optics and Photonics (SPIE) and the Optical Society of America (OSA) and include 336 unique (avoiding double counting of corporations that appear on more than one membership list) corporate members in 2011; 1,009 unique companies that had exhibited at one of the two trade shows in 2011; and 1,785 unique companies listed as employers of
13 National Science Foundation. 2012. Chapter 4, “R&D: National Trends and International Comparison—Highlights.” In Science and Engineering Indicators 2012. Available at http://www.nsf.gov/statistics/seind12/c4/c4h.htm#s6. Accessed July 30, 2012.
professional societies’ individual members in 2011.14Table 2.1 lists the number of publicly traded and privately held companies in each of these three groups. Note that although the companies listed within each of the three groups described above appear only once, there is overlap among companies appearing on each of the three lists (i.e., a single firm may be listed as a member of one or the other society, as well as an exhibitor and an employer of society members). Aggregating all unique companies across the three groups produces a list of 2,442 unique U.S. companies active in some way within photonics, 285 of which are public and 2,157 of which are privately held, as shown in Figure 2.2. As a point of comparison, there were approximately 5.9 million “employer” firms (firms with payroll) in the United States in 2008, and approximately 17,000 publicly traded companies.15 Thus the present study’s count of companies across the three lists comprises approximately 0.04 percent of all U.S. employer firms and 1.7 percent of all U.S. publicly traded companies.
Data on revenues, employment, and R&D spending in 2009 and 2010 for 282 of the 285 publicly traded companies that are listed as members, employers, or exhibitors can be seen in Table 2.2.16 The total revenues associated with these 282 public companies in 2010 amounted to $3.085 trillion, they invested $166 billion on R&D (amounting to 5.4 percent of revenues), and employed 7.4 million individuals. As a comparison point, “employer” firms in the United States in 2008 created an aggregate of $29.7 trillion in revenues and employed an aggregate of 120
14 NAICS or other industry-specific public databases on economic activity in photonics do not currently exist. In an attempt to create a rough estimate of economic activity in photonics, the committee collected three types of information with help from the two largest professional societies in photonics: SPIE and the Optical Society of America (OSA). This information included (1) a list of all U.S.-headquartered member companies for each society, (2) a list of U.S.-headquartered exhibiting companies at the largest trade conference for each society, and (3) a list of all U.S.-headquartered companies associated with individual members of the professional society. The information provided by these societies covers only 2011. In the analysis of this information, the list of member companies was considered to be a rough estimate of companies with strong participation in optics and photonics in 2011, the list of exhibiting companies as a rough estimate of companies selling products involving photonics in 2011, and the list of companies associated with individual members of the professional society a rough estimate of companies with some activities in photonics in 2011. This list of firms also served as the basis for compiling estimates of economic activity during 2010 for the subset of firms for which data were available (see text). It is important to emphasize that each of these estimates is very rough, and it is plausible that some photonics-specialist firms are not captured by these estimates, while other firms for which photonics represents a small share of overall revenues, employment, or R&D investment may be included.
15 According to the U.S. Census Bureau. 2008. “Statistics about Business Size.” Available at http://www.census.gov/econ/susb/introduction.xhtml. Accessed June 25, 2012.
16 Data from Standard & Poor’s Compustat. Available at http://www.compustat.com/. Accessed June 25, 2012.
TABLE 2.1 Number of Unique Companies in 2011 That Were Corporate Members, Participated in One of the Two Largest Trade Shows, or Were Associated with Individual Members Across the Two Largest Professional Societies in Photonics
|Exhibited at trade shows||107||11||902||89||1,009|
|Employed professional society members||243||14||1,542||86||1,785|
SOURCE: Data contributed by SPIE and the Optical Society of America, compiled by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
FIGURE 2.2 Percentage in 2011 of public versus private companies across the 2,442 unique companies recorded within the SPIE and OSA databases. SOURCE: Data contributed by SPIE and OSA, and subsequently collated by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
million individuals.17 Thus, the public firms listed as active in photonics accounted for approximately 10 percent of U.S.-based employer firms’ revenues and 6 percent of U.S.-based employer firms’ aggregate employment in 2010.
Data were also used from Dun and Bradstreet to estimate revenue and R&D expenditures for the publicly traded and privately held firms listed as corporate
17 Public company listings contributed by SPIE and the Optical Society of America. Revenue, employee, and research and development (R&D) expenditure data subsequently collected from Compustat.
TABLE 2.2 Revenues, Number of Employees, and R&D Expenditures from 282 Unique Public Companies in 2009 and 2010
|Revenue ($ millions)||2,741,289||3,085,292|
|No. of employees (000s)||7,159||7,415|
|R&D expenditures ($ millions)||151,104||166,603|
|R&D: % of revenue||5.5%||5.4%|
SOURCE: Public company listings contributed by SPIE and the Optical Society of America. Revenue, employee, and research and development (R&D) expenditure data subsequently collected from Compustat. Compiled by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
TABLE 2.3 Revenue and Number of Employees for the 336 Unique Companies (in 2010) That Had Corporate Members with at Least One of the Two Professional Societies
|Company Type||No.||Revenue ($ millions)||No. of Employees (000s)||R&D Expenditures ($ millions)|
SOURCE: Company listings contributed by SPIE and the Optical Society of America. Revenue, employee, and research and development (R&D) expenditure data subsequently collected for public companies from Compustat and for private companies from Dunn and Bradstreet. Compiled by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
members of SPIE or OSA, on the assumption that photonics sales and innovation-related activities are likely to be much more significant within these firms than within those listed as exhibitors or employers of professional society members. This group of public and private corporate member companies was responsible for $503 billion in revenues in 2010 (roughly one-sixth of the aggregate revenues for the more comprehensive list of firms summarized in Table 2.1) and employed 1.5 million individuals (slightly more than one-tenth of the employment associated with the more comprehensive list of firms). Table 2.3 reports total revenues and R&D investment for the publicly traded and privately held firms within this population of “photonics specialists.”18 Clearly, the firms that can be defined as “photonics specialists” account for a much smaller share of overall U.S. employment and industrial revenues.
18 R&D expenditures were not available from Dun and Bradstreet for privately held companies.
These data suggest that a small (although non-negligible) proportion of total U.S. firms is accounted for by those U.S.-headquartered firms with sufficient activity in photonics to be active as a member, an exhibitor, or an employer of a professional-society member in the database of one of the two largest U.S. photonics-related professional societies. The aggregate revenues associated with these firms, however, represent a relatively large proportion of aggregate U.S. employer firm revenues (10 percent) and employment (6 percent). This study interprets these data as indicative of the pervasiveness of photonics innovation and technology within this economy. Data for firms identified here as specialists in photonics suggest that these firms account for a much smaller share of total U.S. industrial revenues and employment, although they are still a significant source of economic activity.
Like the data from the Baer and Schlachter study for OSTP on lasers cited above,19 these data convey some sense of the breadth of photonics-based industrial activity within this economy. While not directly measuring the economic impact of photonics within the U.S. economy in 2011, they do reflect the general-purpose nature of photonics. This field of technology influences innovation and employment across a broad swath of the economy. Once again, the estimates used in this analysis underscore gaps in existing public databases and the need for better measurement and tracking of photonics-related R&D, employment, and industrial activity to enable better understanding of the full economic impact of so pervasive a technology.
GOVERNMENT AND INDUSTRIAL SOURCES OF R&D FUNDING IN PHOTONICS AND FEDERAL FUNDING OF OPTICS20
One of the only previous attempts to estimate overall U.S. R&D investment in the field of photonics and to compare R&D investment in optoelectronics among Western Europe, Japan, and the United States is the 1992 study by Sternberg,21 which covers only 1981-1986. Sternberg in turn relies on an unpublished study for the National Institute of Standards and Technology (NIST) by Tassey,22 which
19 Baer and Schlachter. 2010. Lasers in Science and Industry.
20 This section uses rough estimates of agency-level R&D spending in areas related to optics and photonics to discuss overall trends in U.S. R&D investment in the field of photonics. To address the lack of regularly tracked data or recent published studies on U.S. R&D investment in photonics, the committee requested data on all optics- and photonics-related programs from all government agencies identified as potentially supporting R&D in these areas. The results of this data-collection effort are discussed in the text.
21 Sternberg, E. 1992. Photonic Technology and Industrial Policy. Albany, N.Y.: State University of New York Press.
22 Tassey, G. 1985. Technology and Economic Assessment of Optoelectronics. Planning report. Gaithersburg, Md.: National Bureau of Standards.
Sternberg claims includes only nondefense R&D spending for the United States. (Sternberg does not discuss whether or not Tassey’s study omits defense-related R&D spending in Japan or Western Europe.) Sternberg’s data indicate that U.S. R&D investment from industry and government sources grew from $69 million in 1981 to $339 million in 1986, while Japanese investment grew from $112 million to $344 million, and European investment grew from $30 million to $165 million. As noted above, Sternberg claims that Tassey’s data, which form the basis for these comparative estimates, omit U.S. defense-related government R&D spending, which he estimates to be as much as $230 million in 1986. If Sternberg’s revision of Tassey’s estimates is credible, U.S. R&D investment in optoelectronics as of the middle of the 1980s greatly exceeded the combined investments of Europe and Japan. In a separate calculation for fiscal year (FY) 1990, Sternberg estimates that “Science and Technology Funding” from DOD sources23 for photonics amounted to $655 million, which exceeds his estimate of U.S. photonics R&D funding from all sources for 1986.
A 1996 study on R&D spending in optoelectronics alone found that Japanese firms spent much more than U.S. firms on optoelectronics R&D during 1989-1993.24 But the study also showed that the U.S. government spent significantly more on optoelectronics R&D during this period than the Japanese government spent, investing more in R&D in 1990 alone than the Japanese government had spent in 15 years of government support. (See Tables 2.4 and 2.5.)
Since the publication of Sternberg’s monograph in 1992, studies in Canada and Europe have attempted to estimate public funding of photonics. The Canadian Photonics Consortium estimated Canadian funding of photonics from public sources in 2008 to have been approximately $136 million,25,26 roughly two-thirds of the estimated $219.7 million that the U.S. government invested in R&D in optoelectronics alone in 1993 (see Table 2.4). The European Union (EU) Framework 7 Programme invested €165 million (U.S. $210 million at average international exchange
23 It remains unclear whether or not this funding includes development work.
24 Japanese Technology Evaluation Center (JTEC). 1996. Optoelectronics in Japan and the United States. Report of the Loyola University Maryland’s International Technology Research Institute, Baltimore, Md.
25 Photonics21. 2010. “Lighting the Way Ahead, Photonics21 Strategic Research Agenda.” Dusseldorf, Germany: European Technology Platform. Available at http://www.photonics21.org/download/SRA_2010.pdf. Accessed June 25, 2012.
26 Canadian Photonics Consortium. 2008. “Photonics: Making Light Work for Canada, A Survey by the Canadian Photonics Consortium.” Available at http://www.photonics.ca/Making%20Light%20Work%20for%20Canada_2008.pdf. Accessed June 25, 2012.
TABLE 2.4 Optoelectronics R&D Spending by U.S. Firms, 1989-1993 ($ millions)
|Source of Funding||1989||1990||1991||1992||1993|
SOURCE: Japanese Technology Evaluation Center (JTEC). 1996. Optoelectronics in Japan and the United States. Report of the Loyola University Maryland’s International Technology Research Institute. Baltimore, Md.: International Technology Research Institute. Available at http://www.wtec.org/loyola/opto/toc.htm. Accessed June 25, 2012.
TABLE 2.5 U.S. Government-Funded Optoelectronics R&D, by Funding Organization, 1989-1993 ($ million)
|Source of Funding||1989||1990||1991||1992||1993|
|Armed Forces (Army, Navy, Air Force)||48.1||55.3||57.5||33.3||38.1|
NOTE: Acronyms are defined in Appendix B of this report. SOURCE: Japanese Technology Evaluation Center (JTEC). 1996. Optoelectronics in Japan and the United States. Report of the Loyola University Maryland’s International Technology Research Institute. Baltimore, Md.: International Technology Research Institute. Available at http://www.wtec.org/loyola/opto/toc.htm. Accessed June 25, 2012.
rates for the year 2010) in photonics in 2010 as part of its information and communications technology work program (EU Framework 7,27 IRS28).29
27 European Commission. 2010. “Information and Communication Technologies Work Programme 2011-12.” Seventh Framework Programme (FP7). Available at http://cordis.europa.eu/fp7/ict/. Accessed June 25, 2012.
28 Internal Revenue Service. 2010. “Internal Revenue Service Yearly Average Currency Exchange Rates.” Available at http://www.irs.gov/businesses/small/international/article/0,,id=206089,00.xhtml. Accessed June 25, 2012.
29 Photonics is also funded through other EU mechanisms and work programs, and so this is not a complete representation (e.g., medical imaging would be in the “Health” work program). In addition, individual countries, including both France and Germany, have individual programs focused on photonics. Nonetheless, the EU 2010 investment in R&D in information and communications
No recent studies have attempted to estimate the scale or agency sources of U.S. government R&D support for photonics. As part of the committee’s data—collection efforts, each agency was given a one-page description of this study, including a brief description of what the committee included in its definition of optics and photonics, along with examples of optics and photonics technologies and applications. The responding agencies were as follows: within the Department of Defense, the Air Force Office of Scientific Research (AFOSR), Army Research Office (ARO), Office of Naval Research (ONR), High Energy Laser-Joint Technology Office (HEL JTO), and Defense Advanced Research Projects Agency (DARPA); Department of Homeland Security (DHS); Department of Energy (DOE); National Institutes of Health (NIH); NIST; and the National Science Foundation (NSF). Although all of the agencies and programs contacted made a good-faith effort to respond, serious gaps nonetheless remained in the committee’s estimates of total federal R&D support for photonics during FY 2006-FY 2010. The committee obtained a total estimate of more than $53 billion for federal photonics-related R&D during this period. Nearly $45 billion of this total is based on DOD R&D investments that appear to involve photonics in some fashion but cannot be verified as limited solely or even primarily to photonics. Similarly, the estimated $4.4 billion in photonics R&D attributed to NIH includes investments in other technological fields. However, much of the NIST R&D investment in photonics is omitted from the estimates. Because of these complications, the committee recommends that an estimate of $53 billion for federal R&D investments in photonics during FY 2006-FY 2010 be interpreted more realistically as an upper bound on a “true” total that may well be anywhere from $25 billion to $55 billion.
Within individual federal programs, more reliable and precisely defined estimates of photonics-related R&D investment were obtained. Many of these agency- or program-specific investments in photonics R&D have grown during FY 2006-2010. Within the DOD, for example, DARPA R&D funding in optics and photonics has almost doubled during this period, rising to $486 million by FY 2010 (see Figure 2.3).
The second-largest funder of optics and photonics R&D, and the largest civilian funder, is NIH, which accounts for 80 percent of the reported nondefense photonics R&D in Table 2.5. Here again, the funding of optics and photonics technologies appears to have grown during the last decade. Figure 2.4 shows aggregate annual funding of optics and photonics by NIH based on a search for all proposals granted between 2000 and 2011 that included the words “optics,” “photonics,” “opto,” or “laser” in their abstract, project title, or project terms. While the data for 2011 extend
technology-related photonics still amounts to little more than the estimated U.S. government investment in R&D in optoelectronics alone in 1993 (Table 2.4).
FIGURE 2.3 DARPA funding in optics and photonics. SOURCE: Based on data collected from the Defense Advanced Research Projects Agency (DARPA) Research and Development, Test and Evaluation, and Defense-Wide Budgets Database (http://www.darpa.mil/NewsEvents/Budget.aspx), compiled by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
FIGURE 2.4 Funding by the National Institutes of Health (NIH) between 2000 and 2011 in optics and photonics based on a keyword search in the NIH RePORTer Database. SOURCE: Data collected from the National Institutes of Health RePORTer Database on October 10, 2011, compiled by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
only through October 10, NIH personnel reported that the full-year funding level for 2011 is likely to be lower than that for 2010.30
According to the data provided by NIH RePORTer search, the National Cancer Institute (NCI) has provided the most funding, at 15.8 percent of the NIH total, followed by the National Eye Institute (NEI) at 11.8 percent and the National Heart, Lung, and Blood Institute (NHLBI) at 9.9 percent.
Other data from the federal agencies also suggest that federal photonics R&D spending has grown in recent years. For example, Figure 2.5 compiles the total funding during 1977-2011 associated with grants either fully or partially funded by the NSF Electronics, Photonics, and Magnetic Devices (EPMD) Division.
Finally, data on DOE funding of solar energy and photovoltaics R&D (Figure 2.6) suggest similar upward trends in the last decade, although a longer time series suggests that federal funding in solar energy may well have been equally substantial in the late 1970s. Although federal R&D funding in this field during the 1980s and 1990s was lower than that of the Japanese and German governments, federal R&D funding has grown substantially in recent years, such that it once again exceeds these other governments’ investments (see Figure 2.7).
In summary, the data made available by federal agencies for this committee’s attempt at a complete estimate of federal photonics-related R&D investment suggest that the DOD dominates funding of optics and photonics, with NIH being the second-largest contributor.31 Federal R&D spending in optics and photonics also appears to have grown during the last decade.
CHANGES IN PHOTONICS-BASED INNOVATION IN THE UNITED STATES SINCE 1980
This section discusses several aspects of the changing structure of the public and private R&D institutions and investments that have underpinned innovation in photonics and other technologies in the United States since 1980. These structural changes have reduced the role of large industrial firms as performers of R&D and have increased the importance of smaller firms, many of which are funded through venture capital, and at least some of which rely on university-licensed intellectual
30 It is important to note that the search tool uses two different coding methods, one pre-2008 (each institute’s judgment of how a grant should be coded) and one post-2008 (an automated trans-NIH coding system called RCDC). It does not appear that this difference in coding should affect the rising trend observed from 2000 to 2006, as this is before the change in 2008.
31 It should be noted that, although every effort was made to be comprehensive in this estimate, the data provided were incomplete. The extraordinary dearth of data on the scale and sources of federal R&D funding in optics and photonics (as is the case in assessing the economic significance of photonics) leaves one “flying blind.” In the committee’s judgment, the creation of more reliable and comprehensive data on federal R&D spending in this field is essential to the formulation of a more effective and coherent public policy.
FIGURE 2.5 National Science Foundation (NSF) Electronics, Photonics, and Magnetic Devices (EPMD) Division funding in optics and photonics. NOTE: To show the trend in underlying sources, two outliers (a 4-year grant of $4.8 million in total awarded in 2002 and a 9-year grant of $6.3 million in total awarded in 1994) were removed from the analysis. In addition, three cooperative agreements (totaling $12.9 million, $13.8 million, and $17.4 million, respectively, the first awarded in 1999 and the last two both awarded in 2003) were excluded from the above plot. The first and last were agreements for the Nanoscale Science and Engineering Centers (NSECs), and the middle one was a renewal proposal for the National Nanofabrication Users Network. SOURCE: Data collected from the NSF Electronics, Photonics, and Magnetic Devices (EPMD) Division database on October 31, 2011, compiled by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
property. The role of venture capital in photonics innovation is discussed in the next section, and potential approaches to inter-firm and public-private collaboration in technology development are examined in a subsequent section. Chapter 7, “Advanced Manufacturing,” discusses another important structural change in U.S. R&D and innovation during this period—the movement of much of the production activity in photonics and other high-technology industries to foreign locations.
Since 1980, the mix of private and public R&D investment in United States has shifted from roughly 50/50 federal/private to roughly 30/70 (Figure 2.8). Moreover, within the defense-related R&D budget, the share of “research” (6.1-6.3) has declined somewhat (see Figure 2.9). To the extent that U.S. photonics technology development benefited heavily from defense-related R&D and procurement spending, the decline in the share of total R&D accounted for by defense may have reduced federal support for photonics.32 As the case study of lasers discussed above
32 As noted above, the committee lacked sufficient data to reach strong conclusions about federal agency-level R&D spending in photonics since 1980.
FIGURE 2.6 Department of Energy (DOE) funding in solar energy and photovoltaics R&D. SOURCE: Data provided by the Department of Energy on October 18, 2011, compiled by Carey Chen, Board on Science, Technology, and Economic Policy of the National Academies.
FIGURE 2.7 Federal research, development and demonstration (RD&D) budget in photovoltaics (at 2010 prices and exchange rates). SOURCE: International Energy Agency (IEA) Online Statistics © OECD/IEA 2012. Reprinted with permission.
FIGURE 2.8 Federally and non-federally funded research and development (R&D) between 1953 and 2002. SOURCE: National Science Foundation (NSF), Science and Engineering Indicators, various years.
FIGURE 2.9 Trends in defense R&D, FY 1976-2008 (House) in billions of constant FY 2007 dollars. NOTE: Contribution from Department of Homeland Security (DHS) defense R&D (shown at the top of years 2003-2008) can be difficult to discern in lower-resolution formats. SOURCE: American Association for the Advancement of Science (AAAS) analyses of R&D in AAAS Reports VIII-XXXII. FY 2008 figures are the latest AAAS estimates of FY 2008 appropriations; 2007 figures include enacted supplementals; R&D includes conduct of R&D and R&D facilities. DOD S&T figures are not strictly comparable for all years because of changing definitions. JULY ’07 REVISED © 2007 AAAS. Reprinted with permission.
suggests, defense procurement demand was an especially powerful impetus to the advance of technologies in photonics-related fields in the early years of their development. Among other things, defense procurement provided an initial market for the production of many new entrant firms. It is plausible that the U.S. market for photonics technologies now is dominated (in terms of total revenues) by nondefense demand to a much greater extent than in previous decades.33 Certainly, this general trend of a declining role in the scope and influence of defense-related procurement within the evolution of a technology-intensive sector is broadly consistent with earlier trends in information technology and semiconductors.34
During 1984-2001, the structure of U.S. industrial R&D performance also has changed significantly (see Figure 2.10). The share of overall industrial R&D performance accounted for by the largest U.S. firms has declined significantly, from 60 percent in 1984 to less than 40 percent by 2001. The share of the smallest firms (fewer than 500 employees) has more than doubled during this period, and the share of industrial R&D performance accounted for by firms with 500-9,999 employees also has grown significantly. Some of this growth in the share of smaller firms may reflect alliances with larger firms and other types of outsourcing. Nonetheless, the role of the largest U.S. firms in industrial R&D has declined significantly during this period. The committee’s studies in this chapter and in Chapter 7 suggest that trends in the structure of photonics-specific industrial R&D are likely to resemble these overall trends. These trends also may reflect growth in the role of inter-firm “markets for technology,” as smaller, specialized firms develop new technologies for sale or license to larger firms that seek to commercialize or incorporate these technologies into their products. Another factor in the growth of U.S. technology licensing activity has been the increased presence of U.S. universities in patenting and licensing faculty research discoveries, a trend that was encouraged and legitimized by the passage of the Bayh-Dole Act of 1980.35 Such “markets for technology” also have benefited from another significant development during 1984-2001 in the United States—the shift in policy to stronger patent-holder rights.
Finally, non-U.S. markets have increased their share of overall global demand for many high-technology products, and offshore sites for the production and R&D activities of many U.S. high-technology firms also have assumed greater importance, as noted in Chapter 7. In most high-technology industries, U.S. firms’ offshore R&D facilities tend to focus on product development more than on basic
33 The committee lacked data on the precise role of defense procurement in the development of the overall field of photonics or on the trends in defense procurement spending over time.
34 Mowery, D.C. 2011. “Federal Policy and the Development of Semiconductors, Computer Hardware, and Computer Software: A Policy Model for Climate Change R&D?” In Accelerating Energy Innovation, R.M. Henderson and R.G. Newell, eds. Chicago, Ill.: University of Chicago Press.
35 Mowery, D.C., R.R. Nelson, B.N. Sampat, and A.A. Ziedonis. 2004. Ivory Tower and Industrial Innovation: University-Industry Technology Transfer Before and After the Bayh-Dole Act. Stanford, Calif.: Stanford University Press.
FIGURE 2.10 Firm-size class shares of industry-performed research and development (non-federally funded) between 1984 and 2001. SOURCE: NSF, Science and Engineering Indicators, various years.
research. Their growth (and especially the growth of U.S. firms’ R&D activities in industrializing economies) reflects the increased sophistication of non-U.S. application environments (reflecting, among many other factors, greater broadband deployment in many non-U.S. economies). The United Nations Conference on Trade and Development (UNCTAD) World Investment Report 2005 noted that multi-national firms from the United States, Europe, and Japan all had increased the share of their offshore R&D activities located in the People’s Republic of China, Singapore, Malaysia, and South Korea. According to the UNCTAD report, the developing-economy share of U.S. multi-national firms’ foreign R&D investment grew from 7.6 percent in 1994 to 13.5 percent by 2002.36 The case studies in
36 Similarly, German firms increased the share of their foreign R&D investment in developing and transitional (including new Eastern European EU member) economies from 2.7 percent to 7.2 percent between 1995 and 2003. United Nations Conference on Trade and Development (UNCTAD). 2005. World Investment Report 2005: Transnational Corporations and Internationalization of R&D. New York, N.Y.: UNCTAD.
Chapter 7, as well as the experiences of the committee, suggest that both offshore production and at least some offshore R&D investment by both U.S. and non-U.S. firms in photonics have grown considerably since 1990.
The growing importance of smaller firms as R&D performers in the U.S. economy since 1980 partly reflects expanded venture-capital funding for new-firm formation and R&D in high-technology sectors. Venture capital has played a significant role in the development of the optics and photonics industry. While total venture-capital investment in the United States may be lower today than during the dot-com bubble, investments continue to be significantly higher than in the early 1990s. As shown in Figure 2.11, in 1995 the total capital invested in venture-backed companies in the United States was $3.7 billion. This investment grew dramatically, peaking at $99 billion in 2000. It then fell to a low in 2003, increasing to a minor peak around $30 billion in 2007.
The committee was able to use data on venture-capital investments in communications, equipment and services, and telecommunications, to provide insights
Figure 2.11 Total U.S. venture-capital investments, 1995-2010. SOURCE: PricewaterhouseCoopers/National Venture Capital Association MoneyTree™ Report, Data: Thomson Reuters. Reprinted with permission.
into the investments made in optics and photonics companies.37 The communications, equipment and services, and telecommunications sectors were among the leading market applications for optics and photonics technologies between 1993 and 2000. Nonetheless, venture-capital investment data for these sectors omit photonics-based applications in solar, biotechnology, and many defense applications.
Figures 2.12 and 2.13 report the estimated total venture-capital investment in network and telecommunications services during 1995-2011, as well as the share of total venture-capital investment in the United States directed toward these sectors. At the peak in 2000, optical investments in network and telecommunications services were estimated to be close to $8 billion, which represented roughly 28 percent of the total venture-capital dollars invested in network and telecommunications services. Between 1998 and 2000, the percent of dollars invested in optics and photonics companies in network and telecommunications services grew from 15 percent to 20 percent of total venture dollars invested in network and telecommunications services to a peak of 28 percent. In 2010, optical and photonics investments in communications, equipment and services, and telecommunications fell from a total of $8 billion to less than $2 billion.
Categorizing optical and photonics investments is much more difficult to do today than in 2000. Between 1993 and 2000, most of the optics investments were in the areas of communications, reflecting growth in fiber-optic networks for long-haul communications. In 1995 the primary technologies that represented optics investments included lasers, fiber optics, optical electronics, imaging instruments, and optical components. The incredible growth of the World Wide Web, the passage of the Telecommunications Act of 1996, and the new ability of individuals to invest in public stocks through online services that allow the trading of stocks at an unprecedented rate expanded demand for optical technologies that were at the center of the infrastructure needed to build the new information highway.
By 1998 an optical communications-centered view of public companies had developed within the U.S. investment community. Investment banking analysts began to follow level-three services/infrastructure systems companies (e.g., Ciena, ADC, and Nortel), components/subsystems companies (e.g., Uniphase, JDS, SDL, Inc., and LightPath), and supporting technologies companies (e.g., Newport, Aeroflex, and VerTel). By 2000, 200 to 300 optical components companies received nearly $10 billion in venture financing, 8 new firms had initial public offerings (IPOs), and 10 more filed for IPOs.38 However, the financing of so many new entrants led to intense competition for market share in a market that soon faced slower growth. In 2001, more than 100 optical companies sought financing and
37 Analysts have never collected data for photonics-specific venture-capital funding.
38 Chang, M. 2010. What is the state of investment in photonics? Laser Focus World 46(5):33.
FIGURE 2.12 Network and telecommunications percent of total annual venture-capital investments, 1995-2011. SOURCE: PricewaterhouseCoopers/National Venture Capital Association MoneyTree™ Report, Data: Thomson Reuters. Reprinted with permission.
more than 30 companies planned to file for IPOs. The optimistic growth forecasts for overall telecommunications were not realized, resulting in a crash in the market that led venture capitalists to direct their investments away from optical networking companies.
During the past decade, the investment community’s expected returns from the traditional optics and photonics sectors declined sharply. Figure 2.14 shows the change in internal rate of return (IRR) expectations from communications companies compared with IRRs in the 1990s that were greater than 100 percent. From Figure 2.14, one can see that communications companies, defined here to include significant optics and photonics, had a mean IRR of greater than 100 percent for those funded in 1993 through 1998, with the highest return occurring for
FIGURE 2.13 Trends (in U.S.$ billions) in U.S. venture series A rounds venture investments with proxy-relation to optics and photonics, 1995-2011. SOURCE: PricewaterhouseCoopers/National Venture Capital Association MoneyTree™ Report, Data: Thomson Reuters; courtesy of John Dexheimer, LightWave Advisors, Inc. Reprinted with permission.
FIGURE 2.14 U.S. venture internal rate of return (IRR) on vintage year companies, 1993-2005. (Shown is the calculated mean return on investments in terms of IRRs for companies in a given year known as the vintage year.) SOURCE: Data collected by Cambridge Associates. Reprinted with permission.
those companies funded in 1998 (a mean IRR of 272 percent). The mean IRR for communications companies with exits funded in 1999 dropped to 31 percent and was negative for those companies funded between 2000 and 2001. It can also be seen that the mean IRRs for all venture-backed companies with a successful exit performed worse than the communications sector during the late 1990s, which drove the demand for venture capital-investments during the 1999-2001 period. For companies funded in those years, the mean IRR was negative.
The expectations of the venture industry for specific sectors also are affected by trends in the overall stock market. Figure 2.15 shows the performance of the OEM [original equipment manufacturers] Capital Photonics Index relative to the S&P 500, revealing that the S&P 500 outperformed the photonics sector during the early part of 2011. If the traditional photonics industry fails to outperform the S&P 500, it is unlikely that the venture industry will shift its investment focus away from social networking and software companies, which currently account for a large share of venture-capital investments, back to photonics. Accordingly, the U.S. venture-capital industry may play a smaller role as a source of investment in photonics innovation in the telecommunications sector, for the near future at least.
Further support for this forecast of the modest near-term effects of venture-
FIGURE 2.15 Relative performance of OEM Capital Photonics Index (lower red line) versus S&P 500 (upper blue line), May 2010 through April 2011. SOURCE: Courtesy of OEM Capital, Inc., as presented by John Dexheimer to the National Research Council, August 24, 2011. Reprinted with permission.
capital funding on photonics innovation in telecommunications applications is provided by Figure 2.16, which shows the results of a venture-capital survey conducted by Deloitte and the National Venture Capital Association (NVCA) to determine the anticipated investments of venture capital over the next 5 years in the telecommunications and new media/social networking markets.
The optics and photonics companies that survived the downturn in the telecommunications equipment market have diversified into the biotechnology, energy, imaging, and defense applications. Since the venture-capital industry associations do not track the photonics industry as a separate sector, the challenge again becomes one of determining a correct metric to ascertain the role of venture financing in the contemporary optics and photonics industry. The bursting of the telecommunications bubble in March 2000 also means that investment analysts now do not track firms in these fields closely. Today only 3 investment analysts track the company JDSU (a company that produces optoelectronic components for telecommunications and data communications), compared to more than 40 in 2000. By contrast, First Solar has 30 analysts and Cree has 50, which may be interpreted as an indicator that more private investment has been directed to solar, lighting, and display technologies. Other markets in which optical technologies are used as a key enabler that have received attention from the venture-capital community include biopharmaceutical tools, environmental sensing devices, and medical devices.
With communications no longer a driving factor for many venture capitalists
FIGURE 2.16 Anticipated investment levels in terms of total capital over the next 5 years—new media/social networking (top) and telecommunications (bottom). SOURCE: 2011 Global Venture Capital Survey, June, by Deloitte; National Venture Capital Association—Next 5 Year Venture Capital Forecast of Allocations. Reprinted with permission.
and optics now spread across many applications, the optics and photonics community may benefit significantly from an initiative to regain the attention of the venture-capital community and to reduce reliance on government funding. (See Box 2.1.) In addition, for policy makers to have the ability to track and measure the impact of optics and photonics, the optics industry needs to encourage organizations like the NVCA to refine its economic and financial data collection to track venture-capital investments in optics and photonics across the diversity of markets in which optics and photonics can be enabling technologies. If returns from optics investments can be better quantified, venture capitalists may be more likely to track and invest in optics and photonics as they did for the brief period in the mid-1990s.
A Note on Government Funding of Small and Medium-Sized Businesses
As suggested in this chapter in the earlier section entitled “Government and Industrial Sources of R&D Funding in Photonics and Federal Funding of Optics,” in addition to venture capital and angel funding, government funding can contribute to the financing of small firms that are research and development performers. Major government programs funding small and medium-sized enterprises include the Small Business Innovation Research (SBIR) Program (which channels funds through multiple federal agencies, including the Department of Defense [DOD], Department of Energy [DOE], National Institutes of Health [NIH], and National Science Foundation [NSF]; the related Small Business Technology Transfer [STTR] Program; and varying programs at the National Institute of Standards and Technology [NIST], including the earlier Advanced Technology Program [ATP] and the more recent Technology Innovation Program [TIP]).
In 2010, 11 federal agencies together provided more than $2 billion in SBIR funding. Individual programs in the DOD (including the Defense Advanced Research Projects Agency [DARPA]), DOE, NIH, and NSF also can target early technology development and commercialization by small firms outside of the SBIR initiatives. An estimated 20 to 25 percent of funding for early-stage technology development comes from the federal government. Nevertheless, small technology-based firms (500 or fewer employees) received only 4.3 percent of extramural government R&D dollars in 2005 (in contrast to medium and large firms, which received 50.3 percent). Of this 4.3 percent, 2.5 percent (or 58 percent of the total) came from SBIR and STTR funds.
NOTE: The committee was unable to collect per-agency numbers on SBIR (and other small business-oriented) funding of optics and photonics. Nevertheless, it is likely that SBIR funding and other programs (including NIST’s ATP and TIP, as well as some DARPA initiatives) have played an important role in funding early-stage technology development in photonics.
SOURCES: Branscomb, L.M., and P.E. Auerwwald. 2002. Between Invention and Innovation: An Analysis of Funding for Early-Stage Technology Development. Washington, D.C.: National Institute of Standards and Technology; National Research Council. 2008. An Assessment of the SBIR Program. Washington, D.C.: The National Academies Press.
MARKETS FOR TECHNOLOGY, INTELLECTUAL PROPERTY, AND U.S. UNIVERSITY TECHNOLOGY LICENSING
As is discussed above in this chapter, structural change in the U.S. R&D system has expanded the importance of licensing transactions involving intellectual property, and the committee believes that these transactions play an important role in photonics innovation in particular. Further evidence of the increased importance of licensing transactions and patented intellectual property in general is provided by the America Invents Act (Public Law No. 112-29), which was signed into law on September 16, 2011. This act, which represented the first comprehensive overhaul of U.S. patent policy in decades, was intended in part to improve the quality of patents granted by the U.S. Patent Office and included steps to further harmonize U.S. patent policy with that of other nations.
As the discussion of lasers pointed out, U.S. universities have long been an important source of ideas and discoveries in the broad field of photonics. In many ways the role of U.S. universities in the U.S. photonics industry appears to be more significant than is true of university research in the photonics industries of other nations. The Bayh-Dole Act of 1980 (Public Law No. 96-517) was passed with broad bipartisan support in order to catalyze the commercialization by U.S. firms of U.S. universities’ research advances, and in the wake of the act’s passage, U.S. university patenting has grown. Many if not all U.S. research universities have established campus offices of technology licensing to oversee the patenting and licensing to industry of research advances.39 In the case of the photonics industry, universities have clearly over the last four decades been playing an expanding role in early-stage R&D. For example, there has been an increase in the percent of overall publications in optoelectronics with at least one academic author over the four decades between 1967 and 2007.40
The expanded licensing activities of U.S. universities have also attracted considerable criticism from at least some sectors of U.S. industry, notably firms in information technology. R. Stanley Williams of Hewlett Packard, a firm with a long history of close research collaboration with U.S. universities (and a firm active in photonics research and innovation), stated in testimony before the U.S. Senate Commerce Committee’s Subcommittee on Science, Technology and Space:
Largely as a result of the lack of federal funding for research, American Universities have become extremely aggressive in their attempts to raise funding from large corporations. . . . Large U.S. based corporations have become so disheartened and
39 Although lacking precise data, the committee believes that university-licensed intellectual property has been an important source of innovation in the U.S. photonics industry.
40 Doutriaux, T. 2009. “The Resiliency of the Innovation Ecosystem: The Impact of Offshoring on Firms versus Individual Technology Trajectories.” Work toward a Master’s Thesis. Advisor: E. Fuchs. Pittsburgh, Pa.: Carnegie Mellon University.
disgusted with the situation they are now working with foreign universities, especially the elite institutions in France, Russia and China, which are more than willing to offer extremely favorable intellectual property terms. (September 17, 2002)41
These remarks from a leading industrial research manager suggest that for at least some U.S. firms, U.S. universities’ patent and licensing policies have become an impediment to collaboration rather than a facilitator of such collaboration, which remains essential to innovation in a U.S. economy that faces challenges from foreign nations with increased technological capabilities. In some cases, frictions between university licensing professionals and U.S firms reflect an unrealistic assessment by university personnel of the financial returns associated with “driving a hard bargain” in licensing terms for a single patent. In December 2005, in response to this criticism and other industry statements of dissatisfaction, four large IT firms (Cisco, Hewlett Packard, IBM, and Intel) and six universities (Carnegie Mellon University; Rensselaer Polytechnic Institute; University of California, Berkeley; Stanford University; University of Illinois at Urbana-Champaign; and University of Texas at Austin) agreed on a “statement of principles” for collaborative research on open-source software that emphasizes the liberal dissemination of the results of collaborative work funded by industrial firms.42
In 2007, a group of technology-licensing managers that included representatives from Stanford University; the California Institute of Technology (Caltech); the University of California, Berkeley; and other leading U.S. research universities as well as the Association of American Medical Colleges, issued a list titled “In the Public Interest: Nine Points to Consider in Licensing University Technology” that emphasized the importance of ensuring access to universities’ intellectual property in the public interest.43 Finally, the National Research Council (NRC) convened a committee of experts and practitioners from industry and academia to consider best practices in university technology licensing, and issued a report titled
41 American Society of Mechanical Engineers (ASME). 2002. Statement available at http://www.memagazine.org/contents/current/webonly/webex319.xhtml. Accessed April 2, 2005.
42 The “Open Collaboration Principles” cover “just one type of formal collaboration that can be used when appropriate and will co-exist with other models, such as sponsored research, consortia and other types of university/industry collaborations, where the results are intended to be proprietary or publicly disseminated.” According to the principles, “The intellectual property created in the collaboration [between industry and academic researchers] must be made available for commercial and academic use by every member of the public free of charge for use in open source software, software related industry standards, software interoperability and other publicly available programs as may be agreed to by the collaborating parties.” Ewing Marion Kauffman Foundation. 2006. Available at http://www.kauffman.org/pdf/open_collaboration_principles_12_05.pdf. Accessed October 17, 2012.
43 Association of University Technology Managers. 2007. “In the Public Interest: Nine Points to Consider in Licensing University Technology.” White paper. Available at http://www.autm.net/Nine_Points_to_Consider.htm. Accessed July 25, 2012.
Managing University Intellectual Property in the Public Interest, in 2011.44 Notable among the conclusions of the report was that “[university] patenting and licensing practices should not be predicated on the goal of raising significant revenue for the institution. The likelihood of success is small, the probability of disappointed expectations high, and the risk of distorting and narrowing dissemination efforts is great” (p. 5).
The committee believes that the “Nine Points” document cited above, and the conclusions of the NRC’s 2011 report on U.S. university technology licensing, provide valuable guidelines for U.S. universities’ management of their photonics-related intellectual property. The committee supports the conclusions of these expert groups.
MODELS OF COLLABORATIVE R&D AND IMPLICATIONS FOR PHOTONICS INNOVATION
As is noted above, the structure of the U.S R&D system has changed since 1980. As the share of federal R&D funding has declined, so also the role of large-firm R&D laboratories has decreased in significance, the influence of defense-related procurement within maturing high-technology sectors such as lasers (and, the committee believes, other photonics technologies) has declined, and offshore R&D has grown in importance. What do these trends imply for the structure of federal R&D in photonics and the ability of such R&D investments to produce significant economic payoffs for U.S. taxpayers? In other high-technology sectors, ranging from nanotechnology to semiconductors, one policy that has proven useful is public-private collaboration in R&D.45
It is widely accepted in economics that “in the absence of policy intervention, the social rate of return to R&D expenditure exceeds the private rate, leading to a socially suboptimal rate of investment in R&D” (p. 22).46 This market failure raises important questions for how public policy should seek to foster increased investment in R&D. In addition to public funding of R&D, one useful policy tool to internalize the externalities (e.g., that the social benefits exceed the private-firm
44 National Research Council. 2011. Managing University Intellectual Property in the Public Interest. Washington, D.C.: The National Academies Press.
45 An essential first step in developing such a policy is a better accounting of the current federal investment in photonics R&D.
46 Jaffe, A. 2002. Building programme evaluation into the design of public research-support programmes. Oxford Review of Economic Policy 18(1):22-34.
benefits) of R&D is public-private partnerships or research.47,48 Past research has found a positive impact of Japanese consortia and of ATP-funded U.S. government-industry joint ventures on the research productivity of participants in the technological areas targeted by the consortia.49,50,51 Indeed, in addition to the support provided by the government’s funding, research consortia can play an important role in supporting network formation, thus increasing knowledge flows among participants,52,53,54,55 and supporting skills56,57 and the creation of new industries.58
In addition to considering research consortia, this section looks at several less widely researched models of coordinated technology development. As Bergh discusses in his paper “Manufacturing Infrastructure for Optoelectronics,”59 it considers three models for the coordination of technology development for a shorter or longer term and with more or less government funding. The first model, SEMATECH, is a not-for-profit research consortium established in 1987 to provide a research facility in which member companies could improve their semiconductor manufacturing process technology. The second, the Optoelectronics Industry Development Association (OIDA), is a not-for-profit partnership of North American suppliers and users of optoelectronic components, established in 1991 to improve the competitiveness of the North American optoelectronics industry with public
47 Spence, A.M. 1984. Cost reduction, competition, and industry performance. Econometrica 52(1):101-121.
48 Katz, M.L. 1986. An analysis of cooperative research and development. RAND Journal of Economics 17(4):527-543.
49 Branstetter, L., and M. Sakakibara. 1998. Japanese research consortia: A microeconometric analysis of industrial policy. Journal of Industrial Economics 46(2):207-233.
50 Branstetter, L., and M. Sakakibara. 2002. When do research consortia work well and why? Evidence from Japanese panel data. American Economic Review 92(1):143-159.
51 Sakakibara, M. 2003. Knowledge sharing in cooperative research and development. Managerial and Decision Economics 24:117-132.
52 Tripsas, M., S. Schrader, and M. Sobrero. 1995. Discouraging opportunistic behavior in collaborative R&D: A new role for government. Research Policy 24:367-389.
53 McEvily, B., and A. Zaheer. 1999. Bridging ties: A source of firm heterogeneity in competitive capabilities. Strategic Management Journal 20:1133-1156.
54 Whitford, J. 2005. The New Old Economy: Networks, Institutions, and the Organizational Transformation of American Manufacturing. Oxford, U.K.: Oxford University Press.
55 Fuchs, E. 2010. Rethinking the role of the state in technology development: DARPA and the case for embedded network governance. Research Policy 39:1133-1147.
56 McEvily, B., and A. Zaheer. 1999. Bridging ties: A source of firm heterogeneity in competitive capabilities. Strategic Management Journal 20:1133-1156.
57 Whitford, J. 2005. The New Old Economy.
58 Fuchs, E. 2010. Rethinking the role of the state in technology development: DARPA and the case for embedded network governance. Research Policy 39:1133-1147.
59 Bergh, A. 1996. Manufacturing infrastructure for optoelectronics. Lasers and Electro-Optics Society Annual Meeting Conference Proceedings. IEEE. Lasers and Electro-Optics Society LEOS-96. November 18-21, 1996.
funding from various agencies as well as private membership funding. The third, the National Nanotechnology Initiative (NNI), is one of the largest federal interagency R&D programs; established in 2000, today it coordinates funding from 25 federal departments and agencies for nanotechnology research and development. An effective industry coalition would take time and resources to develop and therefore would need staunch commitment by stakeholders. One goal would be to create a collective voice that is knowledgeable and credible to center activities in photonics, to provide a positive influence to the industry, and to keep government agencies and the public informed. The examples below provide insights into how these goals have been achieved through institutions in the United States historically. The model provided by the German Fraunhofer Institutes is discussed in Box 2.2.
Semiconductor Manufacturing Technology (SEMATECH)
As noted above, SEMATECH was established in 1987 to provide a research facility in which member companies could develop next-generation manufacturing
The committee heard reports about the unique and successful photonics activities of the German Fraunhofer Institutes. These institutes represent a novel approach to fostering leading-edge research by a combination of universities, companies, and government. In this approach (1) the government provides an overarching forum and core funding, (2) the industry provides a healthy percentage of the funding but is focused on key areas in which the industry has interest, and (3) the universities provide the intellectual workforce to achieve technological advances. Although it is debated whether this model would be effective in the U.S. institutional structure, this combination has been successful in providing leadership, maintaining interest, and producing impressive technical advances within the German context.
Even if the full model is not transferable, aspects of these highly respected Fraunhofer Institutes might be valuable models for the United States, given that these institutes have been playing a pivotal role in technology commercialization and enabling Germany to have a leadership position in photonics. For example, this model was instrumental in the recent formation of the €360 million High-Tech Foundation Fund. This fund includes industry and government participation to provide seed capital for start-up companies to commercialize technologies that are spun out of the Fraunhofer Institutes. This fund also enables industry leaders to steer critically needed seed capital to worthy photonics start-up companies without requiring government agencies to make the selections.
SOURCES: Fraunhofer. 2012. Available at http://www.fraunhofer.de/en/html. Accessed June 26, 2012. See also German Center for Research and Innovation. 2012. Available at http://www.germaninnovation.org/about-us. Accessed June 26, 2012.
technology.60,61 SEMATECH sought to support horizontal collaboration among U.S. semiconductor producers on the development of process technology. This initial focus, however, proved to be infeasible because of the importance of firm-specific process expertise for the competitive advantage of individual semiconductor manufacturers.62,63 As a consequence, SEMATECH altered its research agenda to a vertical collaboration model that sought to improve the technological capabilities of U.S. suppliers of semiconductor manufacturing equipment.64 This shift in its research agenda was associated with a shift in SEMATECH’s intellectual property policies—from licensing research results to member firms on an exclusive basis for 2 years, to member firms’ receiving priority in ordering and receiving new models of equipment resulting from SEMATECH-funded research.65 By the mid-1990s, SEMATECH’s interactions with equipment and material suppliers fell into four main categories: joint development projects, equipment improvement projects, provision of technology “roadmaps,” and expanded communication between suppliers and member firms.66
Although SEMATECH’s role in the improvement of the U.S. industry’s competitiveness may be difficult to prove, SEMATECH met most of its revised objectives in the development of process technology, the supply of manufacturing equipment, and collaboration between manufacturers, suppliers, and research centers.67 Some research suggests that SEMATECH reduced the duplication of member R&D spending 68,69 and that economic returns to member companies outweighed their membership costs.70 Several lessons for the design and structure of public-private consortia can be drawn from the experience of SEMATECH. First, SEMATECH focused primarily on short-term research, with 80 percent of all R&D efforts
60 Grindley, P., D. Mowery, and B. Silverman. 1994. SEMATECH and collaborative research: Lessons in the design of high-technology consortia. Journal of Policy Analysis and Management 13(4):723-758.
61 Link, A., D. Teece, and W. Finan. 1996. Estimating the benefits from collaboration: The case of SEMATECH. Review of Industrial Organization 11:737-751.
62 Grindley et al. 1994. SEMATECH and collaborative research.
63 Carayannis, E., and J. Alexander. 2004. Strategy, structure, and performance issues of precompetitive R&D consortia: Insights and lessons learned from SEMATECH. IEEE Transactions on Engineering Management 51(2):226-232.
64 Grindley et al. 1994. SEMATECH and collaborative research.
65 Grindley et al. 1994. SEMATECH and collaborative research.
66 Grindley et al. 1994. SEMATECH and collaborative research.
67 Grindley et al. 1994. SEMATECH and collaborative research.
68 Irwin, D., and P. Klenow. 1996. High-tech R&D subsidies: Estimating the effects of SEMATECH. Journal of International Economics 40:323-344.
69 Irwin, D., and P. Klenow. 1996. SEMATECH: Purpose and performance. Proceedings of the National Academy of Sciences 93:12739-12742.
70 Link, A., D. Teece, and W. Finan. 1996. Estimating the benefits from collaboration: The case of SEMATECH. Review of Industrial Organization 11:737-751.
focused on outcomes within 1 to 3 years.71 Second, SEMATECH was organized originally by industry, its operations were led and its decisions directed by industry, and it retained substantial support in the form of funding from industry.72 Third, SEMATECH operated with a fairly centralized organizational structure rather than as an umbrella consortium of independent projects (although this characterization is less true for the equipment projects), which facilitated adjustment of its research agenda and operations in response to the changing needs of the industry.73,74 Fourth, SEMATECH drew top executives from member firms into organizational decisions and management.75,76 Fifth, SEMATECH was a consortium of established companies with underlying strengths in product and process technology.77
Optoelectronics Industry Development Association (OIDA)
Similar to the situation in the semiconductor industry, the value of a photonics community coalition is apparent in providing leadership to help interface with industry and government on policy matters, as well as in informing the general public and the investment community on current matters. However, previous attempts to form photonics industry trade associations have had limited success, possibly because these organizations did not receive sufficiently broad industry participation. For example, the Laser Electro-Optics Manufacturers Association and OIDA were composed largely of photonics technology manufacturers and tended not to have support from the applications for those outcomes from industry. The case of OIDA is further discussed here.
In 1988, a National Research Council study entitled Photonics: Maintaining Competitiveness in the Information Era recommended the formation of “an industry association that could help organize consortia to conduct cooperative research and address technical problems and policy issues beyond the scope of any one
71 It is important to point out that the R&D activities of SEMATECH were complemented by two other initiatives. SEMI/SEMATECH represented the U.S. semiconductor equipment producers within SEMATECH, and operated with a modest funding base contributed by the members of the Semiconductor Equipment Manufacturing Industry Association (SEMI); and the Semiconductor Research Corporation (SRC), which enlisted the members of SEMATECH and other U.S. semiconductor firms, supported long-term R&D at U.S. universities.
72 Grindley et al. 1994. SEMATECH and collaborative research.
73 Browning, L., J. Beyer, and J. Shelter. 1995. Building cooperation in a competitive industry: SEMATECH and the semiconductor industry. Academy of Management Journal 38(1):113-151.
74 Grindley et al. 1994. SEMATECH and collaborative research.
75 Browning et al. 1995. Building cooperation in a competitive industry.
76 Grindley et al. 1994. SEMATECH and collaborative research.
77 Grindley et al. 1994. SEMATECH and collaborative research.
organization.”78 In 1991, OIDA was founded as a North American partnership of suppliers and users of optoelectronics components to improve the competitiveness of the North American optoelectronics industry.79 Early on, OIDA undertook an Optoelectronic Technology Roadmap Program, intended to identify the critical paths for the development of enabling optoelectronic technologies. This roadmap exercise concluded in 1996 that mastering volume manufacturing was essential to its members’ ability to reduce costs and improve competitiveness.80
The U.S. optoelectronics industry and OIDA have a long-standing association with NIST as well as with the Department of Defense. In October 1997, at the request of OIDA leaders, NIST organized a photonics manufacturing competition within the Advanced Technology Program (ATP) that led to the funding by ATP of 10 proposals from industry.81 In 1992, DARPA began a series of programs, which continued through 2009, to promote the development of new optoelectronics technologies, including direct funding for OIDA workshops and operating expenses. In 1994, NIST’s Optoelectronics Division was founded “to provide the optoelectronics industry and its suppliers and customers with comprehensive and technically advanced measurement capabilities, standards, and traceability to those standards.”82 As indicated by NIST,83 the division’s mission was to maintain close contact with the optoelectronics industry through major industry associations, including the Optoelectronics Industry Development Association, and to represent NIST at the major domestic and international standards organizations in optoelectronics such as the Telecommunications Industry Association and the American National Standards Institute. This division is now part of the new Quantum Electronics and Photonics Division.84
OIDA remains active in technology roadmapping and in the improvement of member-firm capabilities in high-volume manufacturing, including support for such “R&D infrastructure” as an optoelectronics foundry.85 Although the founding
78 National Research Council. 1988. Photonics: Maintaining Competitiveness in the Information Era. Washington, D.C.: National Academy Press.
79 Bergh, A. 1996. Manufacturing Infrastructure for Optoelectronics.
80 Bergh, A. 1996. Manufacturing Infrastructure for Optoelectronics.
81 Kammer, R., Director, National Institute of Standards and Technology. 1998. Prepared Remarks. Optoelectronics Industry Development Association. Washington, D.C., October 2.
82 National Research Council. 1999. An Assessment of the National Institute of Standards and Technology Measurement and Standards Laboratories: Fiscal Year 1999. Washington, D.C.: National Academy Press.
83 More information is available from the NIST Physical Measurement Laboratory at http://www.nist.gov/pml/. Accessed August 6, 2012.
84 Quantum Electronics and Photonics Division. Physical Measurement Laboratory, National Institute of Standards and Technology (NIST), website. Available at http://www.nist.gov/pml/div686/. Accessed July 25, 2012.
85 These efforts have produced few measureable outcomes.
membership of OIDA included larger companies, such as AT&T, Bellcore, Corning, IBM, 3M, Hewlett-Packard Company, and Motorola, the number of member companies has declined over the course of OIDA’s history. Further, in contrast to SEMATECH membership, many OIDA member firms are smaller, with less developed technologies, and membership fees for all are low compared to SEMATECH—in the low thousands to tens of thousands. Along with its modest industry funding, a key challenge for OIDA is that of enlisting the participation of senior industry executives in managerial positions. Perhaps most importantly, in contrast to SEMATECH, OIDA has lacked a clear R&D agenda. This consortium’s inability to develop such a focus reflects the diversity of applications characteristic of photonic semiconductors, which complicates agreement among member firms on technological goals. Inasmuch as photonics manufacturing technology is less mature than semiconductor process technologies, the optoelectronics industry might be better served by a university-government-private partnership that focuses more intensively on early-stage R&D. The Semiconductor Research Corporation (SRC) is an interesting contrast to SEMATECH in this respect. (See Box 2.3.)
Semiconductor Research Corporation
The Semiconductor Research Corporation (SRC) is another semiconductor research and development consortium whose structure contrasts with that of SEMATECH. In 1982, the Semiconductor Industry Association launched the Semiconductor Research Association as a cooperative research organization to “enhance basic research in semiconductor related disciplines” by funding “long-term, pre-competitive research in semiconductor technology at U.S. universities.” In contrast to SEMATECH—which was created in 1987 out of an SRC initiative—SRC uses horizontal collaborations between member firms and academic researchers to define and fund long-term technology developments central to the survival and success of the semiconductor industry.
In addition to SRC’s receiving funding from member firms, SRC program directors also seek matching funds from federal, state, and local governments. Federal-level collaborators have included the U.S. Army Research Office, the Defense Advanced Research Projects Agency, the National Institute of Standards and Technology, and the National Science Foundation.
In 2000, the SRC board committed to globalizing the SRC membership and research base. To date, however, little empirical research exists on the history, processes, or successes of SRC. SRC is viewed by several of its member companies as an ongoing success and warrants further study as an interesting model for research consortia focused on long-term precompetitive research.
SOURCE: Semiconductor Research Corporation. 2012. “About Semiconductor Research Corporation.” Available at http://www.src.org/about/. Accessed August 3, 2012.
National Nanotechnology Initiative
In September 1998, the Interagency Working Group on Nanotechnology (IWGN) was formed within the National Science and Technology Council of the Office of Science and Technology Policy.86 As described in the National Research Council’s 2002 report Small Wonders, Endless Frontiers: A Review of the National Nanotechnology Initiative (from which material in this paragraph is drawn substantially and, in some cases, extracted), this group formalized the operations of a set of staff members from several agencies that in November 1996 had begun to meet regularly to discuss their plans and programs in nanoscale science and technology. In August 1999, IWGN’s plan for an initiative in nanoscale science and technology was approved by the President’s Council of Advisors on Science and Technology (PCAST) and OSTP, and in its 2001 budget submission to Congress, the Clinton administration raised nanoscale science and technology to a federal initiative, referring to it as the National Nanotechnology Initiative (NNI). The National Science and Technology Council (a cabinet-level committee with membership drawn from federal agencies across the government) formed the Nanoscale Science, Engineering, and Technology (NSET) Subcommittee to focus on NNI activities. The National Nanotechnology Coordination Office, established in 2001, provides technical guidance and administrative support to the NSET Subcommittee, facilitates multiagency planning, conducts activities and workshops, and prepares information and reports. The NRC also provides feedback to NNI through its triennial review of NNI;87 such a review is currently ongoing.88
Today NNI is one of the largest federal interagency R&D programs, coordinating funding for nanotechnology research and development among 25 participating federal departments and agencies. Its federal funding grew from $225 million in FY 1999 to $464 million in 200189 and an estimated $1.639 billion in 2010.90,91 As
86 National Research Council. 2002. Small Wonders, Endless Frontiers: A Review of the National Nanotechnology Initiative. Washington, D.C.: National Academy Press.
87 National Research Council. 2006. A Matter of Size: Triennial Review of the National Nanotechnology Initiative. Washington, D.C.: The National Academies Press.
88 See National Research Council. 2012. “Interim Report for the Triennial Review of the National Nanotechnology Initiative, Phase II.” Prepublication copy. Washington, D.C.: The National Academies Press. Available at http://www.nap.edu/catalog.php?record_id=13517. Accessed October 23, 2012.
89 National Research Council. 2002. Small Wonders, Endless Frontiers.
90 National Research Council. 2002. Small Wonders, Endless Frontiers.
91 Office of Science and Technology Policy (OSTP): A Decade of Investments in Innovation Coordinated Through the National Nanotechnology Initiative. National Nanotechnology Initiative Investments by Agency from FY 2001 through FY 2010. OSTP’s “High Value Data Sets.” The White House. Available at http://www.whitehouse.gov/administration/eop/ostp/library/highvalue. Accessed December 26, 2011.
stated on the NNI webpage,92 NNI has four goals: (1) maintain a world-class research and development program aimed at realizing the full potential of nanotechnology; (2) facilitate the transfer of new technologies into products for economic growth, jobs, and other public benefit; (3) develop educational resources, a skilled workforce, and the supporting infrastructure and tools to advance nanotechnology; and (4) support responsible development of nanotechnology.
NNI has facilitated several developments to enhance dialogue and coordination among nanoscale R&D programs at federal agencies; these include working groups, an infrastructure network involving an integrated partnership of user facilities at 13 campuses across the United States, centers to support the development of tools for fabrication and analysis at the nanoscale, and NNI-industry consultative boards to facilitate networking among industry, government, and academic researchers, analyze policy impacts at the state level, and support programmatic and budget redirection within agencies. In contrast to either OIDA or SEMATECH, NNI is focused more intensively on priority setting and support for more fundamental, long-term research in this emerging technology.
Given the diversity of applications characteristic of photonics and the relative immaturity both of much of the science and much of the industry, the National Nanotechnology Initiative may provide an interesting model for the increased coordination and tracking of long-term funding of research in photonics. Since the writing of the 1998 NRC report Harnessing Light: Optical Sciences and Engineering for the 21st Century,93 there has been an explosion of new applications for photonics. Indeed, in spite of the maturity of some of the constituent elements of photonics (e.g., optics), the committee believes that photonics as a whole is likely to experience a period of growth in opportunities and applications that more nearly resembles what might be expected of a vibrantly young technology.
The preceding overview of some recent experiments in collaborative R&D makes apparent several implications for similar efforts in photonics. First, industry participation and leadership, both intellectual and financial, are essential. Second, such an industry commitment to collaborative R&D may be more difficult in a sector that spans a diverse array of applications and is populated mainly by new, relatively small firms. Finally, consortia (such as SEMATECH) in which industry plays a major role in establishing and funding the R&D agenda may not be well suited for supporting long-term research. An interesting exception may be the model
92 National Nanotechnology Initiative. Available at http://www.nano.gov/. Accessed August 6, 2012.
93 National Research Council. 1998. Harnessing Light: Optical Science and Engineering for the 21st Century. Washington, D.C.: National Academy Press.
presented by the Semiconductor Research Corporation (see Box 2.3), which has supported work by academic researchers in the United States. Nonetheless, given that SRC programs tend to focus on developing technologies to achieve specific end goals that involve stakeholders from a single industry, the long-term precompetitive research agenda supported by the SRC may be insufficient by itself to deal with the diversity of applications that are the focus of R&D within the field of photonics.
Rather than endorsing any single structure for the support of R&D that involves collaboration among industry, government, and academic researchers, the committee believes that a higher-level venue for discussion and assessment of R&D priorities is needed. Any such structure could include among its activities R&D consortia of the type represented by SRC, or SEMATECH, or still other models for collaborative R&D. A research consortium is more likely to succeed in a focused application of the optics and photonics landscape. In the absence of some coordinating initiative, it will prove difficult to develop an effective strategy for public and private R&D investment that seeks to support longer-term R&D and to translate innovation into economic opportunities for U.S. firms and employees across the diversity of emerging photonics applications. Accordingly, the committee’s judgment is that the time is overdue for a federal initiative in photonics that seeks to engage industry, academic, and government researchers and policy makers in the design and oversight of R&D and related programs that include federal as well as industry funding.
Proposed National Photonics Initiative
A national photonics initiative would coordinate agency-level investment in photonics-related R&D and could provide partial support for other technology-development initiatives, including R&D consortia funded by federal and industry sources. The committee believes that a number of experiments in coordinating across industry-government-academia in different fields of R&D and technology development are warranted. As pointed out in the next chapter, “Communications, Information Processing, and Data Storage,” one application area that may be particularly ripe for such a public-private consortium is large-scale data communications and storage, now the focus of an initiative overseen by OSTP.94 Finally, as with NNI, a national photonics initiative would assume responsibility for developing, coordinating, and measuring federal funding and national outputs (such as economic indicators) in photonics, to help inform national policy.
94 Office of Science and Technology Policy. 2012. “Big Data Press Release.” Available at http://www.whitehouse.gov/sites/default/files/microsites/ostp/big_data_press_release_final_2.pdf. Accessed July 30, 2012.
Key Finding: Photonics is a key enabling technology with broad applications in numerous sectors of the U.S. economy. The diversity of applications associated with photonics technologies makes it difficult to quantify accurately the economic impacts of photonics in the past and even more difficult to predict the future economic and employment impacts of photonics.
Key Finding: Given the diversity of its applications and the enabling character of photonics technology, data on photonics industry output, employment, and firm-financed R&D investment are not currently reported by U.S. government statistical agencies, further complicating analysis of this technology’s economic impact and prospects. Although the 1998 National Research Council study Harnessing Light: Optical Science and Engineering for the 21st Century reached a similar conclusion and recommended that members of the photonics community be involved in the next round of Standard Industrial Classification (SIC) or North American Industry Classification System (NAICS) development, no such action was taken by federal statistical agencies.
Finding: Another significant gap in the economic data on photonics is a lack of systematic collection or reporting by the federal government of its significant investment in photonics R&D. As a result, the most basic data are lacking for estimating the overall federal R&D investment in this technology field or the allocation of federal photonics R&D investments among different fields and applications.
Finding: The private organizations that monitor U.S. venture-capital investment trends also do not collect information on the full spectrum of photonics-related venture-capital investments. Changes in the structure of the U.S. R&D and innovation systems mean that the importance of venture-capital funding for the formation of new firms in photonics, as well as for these firms’ investments in R&D and technology commercialization, has grown; thus these gaps in data on venture-capital investment hamper the ability to monitor innovation in photonics.
Finding: Many of the important early U.S. innovations in photonics relied on R&D performed in large industrial laboratories and benefited as well from defense-related R&D and procurement spending. The structure of the R&D and innovation processes in photonics, similar to other U.S. high-technology industries, appears to have changed somewhat, with universities, smaller firms, and venture-capital finance playing more prominent roles. These changes in the structure of R&D funding and performance within photonics increase the potential importance of inter-firm collaboration and public-private collaboration in photonics innovation.
Key Recommendation: The committee recommends that the federal government develop an integrated initiative in photonics (similar in many respects to the National Nanotechnology Initiative) that seeks to bring together academic, industrial, and government researchers, managers, and policy makers to develop a more integrated approach to managing industrial and government photonics R&D spending and related investments.
This recommendation is based on the committee’s judgment that the photonics field is experiencing rapid technical progress and rapidly expanding applications that span a growing range of technologies, markets, and industries. Indeed, in spite of the maturity of some of the constituent elements of photonics (e.g., optics), the committee believes that the field as a whole is likely to experience a period of growth in opportunities and applications that more nearly resembles what might be expected of a vibrantly young technology. But the sheer breadth of these applications and technologies has impeded the formulation by both government and industry of coherent strategies for technology development and deployment.
A national photonics initiative would identify critical technical priorities for long-term federal R&D funding. In addition to offering a basis for coordinating federal spending across agencies, such an initiative could provide matching funds for industry-led research consortia (of users, producers, and material and equipment suppliers) focused on specific applications, such as those described in Chapter 3 of this report. In light of near-term pressures to limit the growth of or even reduce federal R&D spending, the committee believes that a coordinated initiative in photonics is especially important.
The committee assesses as deplorable the state of data collection and analysis of photonics R&D spending, photonics employment, and sales. The development of better historical and current data collection and analysis is another task for which a national photonics initiative is well suited.
Key Recommendation: The committee recommends that the proposed national photonics initiative spearhead a collaborative effort to improve the collection and reporting of R&D and economic data on the optics and photonics sector, including the development of a set of North American Industry Classification System (-NAICS) codes that cover photonics; the collection of data on employment, output, and privately funded R&D in photonics; and the reporting of federal photonics-related R&D investment for all federal agencies and programs.
It is essential that an initiative such as the proposed national photonics initiative be supported by coordinated measurement of the inputs and outputs in the sector such that national policy in the area can be informed by the technical and economic realities on the ground in the nation.