Session V
Meeting the Early-stage Finance Challenge

Moderator:

Luc Soete

University of Maastricht, Netherlands and UN Univ-MERIT

THE TEXAS EMERGING TECHNOLOGY FUND

Pike Powers

Fulbright & Jaworski LLP


Mr. Powers introduced the case of the Emerging Technology Fund in Texas “as an example of what can be done to build innovation.” Innovation, he said, is more than activities of technology or business; it is also the process by which it occurs. With that in mind, he and others began in 2002 to create a Texas Technology Initiative designed to take some “conscious objective steps” toward a vision of innovation and capital formation for technology. They persuaded Texas Governor Rick Perry to include in his state-of-the-state speech a request for an “enterprise fund”; the request was successful and funded at a level of $295 million in 2003. A large portion of the early funding went to “save SEMATECH” for Texas in Austin, but subsequent funding created a $200 million Emerging Technology Fund in 2005, which also benefited SEMATECH and other entrepreneurial ventures. Both funds were reauthorized and refunded in the 2007 session of the Texas legislature.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 90
Session V ———————————————————— Meeting the Early-stage Finance Challenge Moderator: Luc Soete Uniersity of Maastricht, Netherlands and UN Uni-MERIT THE TEXAS EMERGING TECHNOLOGY FUND Pike Powers Fulbright & Jaworski LLP Mr. Powers introduced the case of the Emerging Technology Fund in Texas “as an example of what can be done to build innovation.” Innovation, he said, is more than activities of technology or business; it is also the process by which it occurs. With that in mind, he and others began in 2002 to create a Texas Tech- nology Initiative designed to take some “conscious objective steps” toward a vision of innovation and capital formation for technology. They persuaded Texas Governor Rick Perry to include in his state-of-the-state speech a request for an “enterprise fund”; the request was successful and funded at a level of $295 mil- lion in 2003. A large portion of the early funding went to “save SEMATECH” for Texas in Austin, but subsequent funding created a $200 million Emerging Technology Fund in 2005, which also benefited SEMATECH and other entrepre- neurial ventures. Both funds were reauthorized and refunded in the 2007 session of the Texas legislature. 0

OCR for page 90
1 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE Collaborative University Partnerships He also described a consortium of 12 U.S. universities (from the Big-12 athletics conference), including Texas, that have formed a virtual Center for Economic Development, Innovation, and Commercialization. The objective is to jointly create tech transfer, research, and share equipment and other resources. “It’s the paradigm of the future,” he said, “having universities collaborate.” The initial pattern for the Center was an athletic conference whose schools would “compete on the football field on Saturday, but collaborate the other six days of the week. The model is designed to break down barriers, structurally change relationships, and enhance communications. Activities of the ETF The Emerging Technology Fund with its small size and flexibility has proved to be an effective vehicle for commercialization. While the governor, lieutenant governor, and the speaker of the House approve the work of the 17-member advi- sory committee that evaluates potential candidates, there are no fund managers and only a small staff within the governor’s office to run the program. Each participating region operates a Regional Center for Innovation and Commercialization (RCIC). The ETF has three sections. $100 million goes to Commercialization Invest- ments, for which the ETF has already reviewed more than 331 applications. Of these, 22 projects at 7 RCICs received $27.5 million. The second activity is the $50 million Research Matching Grants, intended for industry consortia or single companies who already have some funding from the federal government or other sources. Of 53 requests, 14 projects were approved for $22 million in funding. Third are the Research Superiority Grants, a fund of $50 million to “discover the best researchers we can find in the country or the world and bring them to Texas.” This fund has brought researchers to Texas Tech University, the Univer- sity of Texas Health Science Center in Houston, and other laboratories. Mr. Powers concluded with the news that Governor Perry would announce the following week that $10 million of the Research Superiority Fund would go to outstanding researchers in nanotechnology. He then introduced Dr. Goodall, with whom he works on the Texas Technology Initiative. OVERVIEW OF TXAN: A NEW MODEL FOR RESEARCH COLLABORATION Randal K. Goodall SEMATECH TxAN, said Dr. Goodall, is the Texas Alliance for Nanoelectronics, which is currently being assembled as an alliance with the federal government, beginning

OCR for page 90
2 INNOVATIVE FLANDERS Collaboration Affordability —Technology Convergence —Inter-industry Collaboration 1,000,000 (Worldwide in Millions of Dollars) AP DG 100,000 R& &D 10,000 ” yR aw ’s L ore lo g o “M 1,000 hno &D R Te c 100 in g e rg 10 Em — Semiconductor Revenue — Total RD&E (Chip + Eq) 1 60 64 72 76 84 88 92 00 04 12 16 68 80 96 08 20 19 19 19 19 19 19 19 20 20 20 20 19 19 19 20 20 Year FIGURE 8 Advanced technology R&D challenge. PROC Fig 08 with the National Institute of Standards and Technology (NIST).21 He described TxAN as a statewide partnership for building innovative, networked nanotech capability “that leverages world-class researchers and R&D infrastructure and drives regional commercialization of technology.” It was the culmination of more than 4 years of work by the Texas Technology Initiative, and builds on regional university and industry strengths. He said it was both a technology-based platform for economic development and an economic development platform for federal and industry partnership. Key elements included the participation of Texas, which is now the 10th or 11th largest economy in the world, and offered considerable support in the form of the Emerging Technology Fund (ETF) and other large initiatives described by Mr. Powers. This support, he said, “is going to bring to this federal-industrial collaboration more Moore’s Law than we deserve. These industries know how to collaborate, and the Texas alliance is all about that.” Collaboration Lowers Costs of Research Goodall showed an illustration of the future of the IT research over the com- ing years, showing how collaboration among firms eventually lowers producer 21TxAN has subsequently broadened its scope to nanotechnology.

OCR for page 90
 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE R&D costs. In the United States, the Nanoelectronics Research Initiative (NRI) is located on the time scale far in the future, and the Focus Center Research Program is somewhat closer to the present. The Semiconductor Research Corpo- ration (SRC) was closer still, and SEMATECH was next to “the border between competitive and pre-competitive research.” One of the features of SEMATECH, he said, is that it is a flexible organization, so that the manufacturing side of the program could actually be split into a separate industry consortium where the 450mm wafer initiative will occur.22 A New Training Paradigm TxAN was designed to create a new innovation and commercialization paradigm. Within the competitive semiconductor world, it would include a nano- electronics workforce development initiative that would benefit all participants. This $4 million program would support 160 interns in three categories—those with 2-year associate degrees, those with a 4-year bachelor’s degree, and graduate students—to work in the fab under a standardized training routine. The objective was to demonstrate the feasibility of large-scale internships at advanced technol- ogy sites, which has not been done before. TxAN also includes the $3 million Nanoelectronics Research Initiative center, described by Mr. Scalise. This third NRI center will provide university research on innovative devices, to be funded initially by the Emerging Technology Fund. Finally, the TxAN Infrastructure Net- work will operationally link SEMATECH and the TxAN fab to Texas university labs to create a collaborative ~$500 million equivalent “State Lab.” Once these programs are in place they will assume the vital task of helping NIST and other units of the federal government fulfill the aspects of their missions that require commercialization. This is an area where the federal government needs the assis- tance of industry and the states. Advantages of a “Mezzanine” R&D Center The Nanofabrication Infrastructure Network will link most of the major university labs in Texas, many of which are world-class. The key, he said, is to link specialty research labs to the Advanced Technology Development Facility (ATDF), a higher-level industry-driven “mezzanine” R&D center—a middle floor between university labs and true industrial manufacturing fabs. The ATDF will specialize in driving the transition of university work to industry. The ~$200 million facility/capability has been developed to support leading-edge CMOS research, including full process control and a full staff around the clock. The target is to upgrade the facility and its processes for the R&D needs of the 22The International SEMATECH Manufacturing Initiative has formed and includes the 450mm initiative in its broad equipment and factory productivity mission.

OCR for page 90
 INNOVATIVE FLANDERS future using both university and private capital funds. There will be state/federal/ industry-funded “tokens” to cover processing costs for university researchers. Work that would otherwise be prohibitively expensive, he said, can be done for reasonable cost in this collaborative environment. Universities can take their own laboratory work to an additional level in the fab environment of the industrial real world. This enhanced capability and accelerated commercialization will drive startup formation and corporate-sponsored R&D. They can then tie it together across the state, using web-based documents for all resources, sharing equip- ment in novel ways, and flexibly using interchange protocols and other “insert” techniques for multiple wafer sizes and other substrates.23 The ATDF is now a separate subsidiary company that used to be part of SEMATECH. Because it is separate, each ATDF participant, including companies that are not SEMATECH members, offers support for the core, but has an invest- ment-based capacity allotment and can have private tools and an area for program operations. It can create its own flexible environment and work with suppliers on an individual contractual basis. TxAN, then, is an umbrella of universities, federal engagements, biomedical and energy consortia, and others. It also partners with SEMATECH and Albany Nanotech, where most of SEMATECH’s lithography R&D is done. The Ability to Test Products in a Real Fab The mezzanine model is the key to modern research, said Dr. Goodall, for several reasons. Manufacturing technology has become critical to the broader field of nanotechnology research, as compared with early nanoelectronics.24 Having a strong manufacturing focus and environment allows researchers to test in a “real” fab how new techniques will actually work and actually scale. The industry is making a quintillion transistors a year, and is projected to make 60 quintillion a year in 2016. New nanotech capabilities have to be scalable to the level where they intersect an industry. If a new solution is likely to require a billion or a trillion of some technology, researchers need a real facility to find out if they can actually make 100-1,000. The mezzanine concept is designed to make this possible. Dr. Goodall said that TxAN had received positive responses to its concept from several federal groups, including NIST itself and NASA, and that the TxAN virtual network provided an “overall umbrella on how to stay up with the rest of the world” through directed R&D. The central idea, he said, was to have the federal agencies link some of their mission work with this network. In the end, he said, the concept unites three forces. The first is a multi-hundred-million-dollar state fund 23The discussion on how to specifically initiate the State Lab continues with key stakeholders in Texas, and a definitive approach is anticipated by the end of 2007. 24See the earlier presentation by Kenneth Flamm for a discussion of the importance of manufacturing- based research.

OCR for page 90
 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE that drives driving commercialization. The second is the in-depth federal research of NIST, NASA, or others. These two mutually enable the third, university labora- tories networked around a core mezzanine facility—this triad attracts and supports the corporate and startup R&D that drives economic development. State Money + Federal Depth + Industrial Management The process currently under discussion, he said, is that the university system in Texas would develop management offices, tap into the research and university lab infrastructure, and tie these functions to the ETF as a mezzanine facility that is fully operating with major industrial players. The ETF and other funds can sup- port pre-commercial work, build up the infrastructure, and recruit the best talent. The university system offers management and governance. Advisory board processes, too, were under discussion, building on the long history of SEMATECH working with NIST. The concept allows a federal agency to see TxAN as an extension of itself, differing from the custom of developing new knowledge through outside contracts. A Paradigm for Other States He summarized by defining nanoelectronics in the broad sense now intended by TxAN. Microelectronics has already introduced a wide range of miniaturized specialty materials, including many applications for biotechnology. He said that TxAN was addressing the technologies that would emerge over the next 20 years throughout this new realm of the very small, from nano-electromechanical sys- tems to “bio-nano.” The participation of NIST at the outset is central, he said, because a strong focus on measurement and standardization is crucial to sustain commercialization. “Texas has all the components it takes to put this together,” he concluded, “and we’re quite confident we can make it go. Beyond this, we believe that this is a replicable paradigm that can be used by other states.” Discussion A questioner asked why TxAN was bringing the university and industry together. Dr. Goodall answered that the idea was not to eliminate university labs or teaching but to provide an industry-like middle step between university labs and actual industrial use. He gave the example of the ATDF nanofabrication facility, where several leading bio-nano device makers need silicon-based technology to build filters and other devices. They could build them in a four-inch university lab, but if they want to try for an FDA-approvable process, or build prototypes, or begin mass fabrication to test the reliability of the process with high volume, they need the capabilities offered by ATDF. At the same time, the work at ATDF is still directed by university-based researchers. Mr. Powers commented that it was

OCR for page 90
 INNOVATIVE FLANDERS an advantage to have industry driving the policy. Government officials would be most likely to support a plan in which the business community plays a dominant role, with university support. “We believe in getting everybody at the table,” said Mr. Powers, “so the elected officials can see that everybody’s in the same boat going in the same direction.” Another questioner asked what metric TxAN would use in Texas to show progress. Mr. Powers said that their current report was the initial stage, and they were asking the legislators that same question in regard to the ETF: How do you want us to show what we’ve done with the taxpayers’ money? Straight job cre- ation was not the best measure, though it is important, because new approaches always lead to disruptions and lost jobs as new jobs are being created. It was challenging to measure all those processes in a way that the non-technical citizen can understand and accept. FROM UNIVERSITY RESEARCH TO UNIVERSITY SPIN-OFF: EXPERIENCES OF VUB Bruno de Vuyst Vesalius College, Vrije Uniersiteit Brussel (VUB) and Lawfort Brussels Professor de Vuyst represented VUB, a Flemish university in Brussels that places emphasis on creating spin-offs and operates on a “project cycle basis.” That is, they start by supporting fundamental research and then examine the outputs of that research as intellectual property for potential commercialization, identifying and studying possible applications. “It is only the last 5 percent of the work that can be valorized, that may be spun off,” he said. “A spin-off may then lead to new questions for fundamental research that will feed back into the cycle.” He said that his university was the last in Flanders to start a spin-off fund, but that this could be considered an advantage, since they had been able to learn from the experience of K.U.Leuven and University of Gent. After only 30 years of existence, he said, this university of about 10,000 students had been able to develop an effective research program that generated on average seven to ten patents per year.25 “Market-oriented PhDs” In Flanders, he said, the traditional focus on fundamental research is giving way to a new paradigm. The university system has de-emphasized traditional 25The VUB is an outgrowth of an older, French-language university founded in 1834 (Université Libre de Bruxelles), both dedicated to freedom of inquiry. It became a free-standing Dutch-language institution in 1969 with a motto that “science triumphs over darkness.”

OCR for page 90
7 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE PhD programs in favor of “market-oriented” PhDs. “There is no automatic career future for traditional PhDs,” said Professor de Vuyst. “We have mobility programs for researchers. We have centers of excellence in engineering, technology, eco- nomics, and medicine. Valorization is the ‘third function’ of universities now. We have to bring our knowledge to the community. That is a big sea change for us, and it required a lot of work.” Larger reforms in the higher educational system are still needed, he said. At present, students can enter any university at age 18 for only about €500 in tuition and choose their degree freely. “We end up with far too many people who are not marketable and hence frustrated because they end up with degrees that the market does not seek. The cost of this is high to taxpayers, and is not sustainable. It is politically not correct to raise the issue, but this is something we will have to deal with at some point, as budgetary constraints will prevail.” Needed Reforms for IPR He turned to issues of IPR, saying that awareness of their importance is growing. On innovation, the public sector is more or less meeting the Lisbon goals, although the private sector is not. The related topic of patent protection faces issues that are even more complex. Patenting costs, for example, are about four times higher than in the United States. There is still no European Community patent, and there are considerable barriers to pan-European suits against patent infringement. The EU court of justice had recently ruled against that practice, which means that a patent must be defended country by country, discouraging patent development. Nor is there any grace period for publication of scientific results, meaning that university professors who think they have a patentable idea must refrain from publishing in the scientific literature, at a cost to their reputa- tion. Finally, he said, European countries needed to remove software from the TRIPS copyright framework.26 This was no longer an issue in the United States, he said, and needed to be resolved quickly in Europe. He concluded by saying that Flanders, although it was doing better than much of Europe in developing an innovation-friendly culture, still had many 26The WTO’s Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) and the Berne Convention placed software protection solely under copyright. Following the U.S. Supreme Court decision in Diamond vs. Diehr, the United States has allowed software protection through patent law; something not provided for in TRIPS or the Berne Convention. Since then, the USPTO has continuously granted patents for software. Professor de Vuyst points out that in Europe, the same evolution is hindered by the exclusion lan- guage in article 52 of the European Patent Convention (EPC). At the time of the conference reported in this volume, there was an expectation that, because of the recommendations of European Patent Office’s own expert committees, art. 52 EPC would be changed in the EPC 2000 text and the exclu- sion language would be abandoned, or at least diminished. This did not happen, as the member states did not follow the recommendations but stayed the text as is, with the result that software “as such” remains unpatentable in Europe.

OCR for page 90
 INNOVATIVE FLANDERS issues to address. There is some private equity involvement in new high-tech firms, he said, but not nearly enough entrepreneurship or venture capital risk taking. Much more attention should be given to preparing new spin-offs for growth and survival, which now takes much too long. Finally, the biggest issue for Flanders is the shortage of management that is qualified and experienced enough to lead the growing number of new firms into the future. Discussion A questioner asked whether the high cost of patenting in Europe can be reduced. Professor de Vuyst said that this was impeded by articles of EU treaties with respect to state aid, but that they would have to find a way to bring the cost down, perhaps beginning with some form of support for translation or develop- ment of claims and claim writing. COMMERCIALIzING UNIVERSITY RESEARCH: THE ROLE OF THE U.S. SBIR PROGRAM Charles W. Wessner U.S. National Research Council Dr. Wessner began by describing the “innovation imperative”—the modern realization that innovation is a prerequisite for maintaining a competitive posi- tion in the global economy. A key in responding to this imperative, he said, is the knowledge that small businesses anchor the innovation process by adopting, developing, and commercializing new ideas to a disproportionate degree. Small size and flexibility are ongoing themes in any discussion of innovation, as are the roles of the university and the faculty researchers involved in knowledge dissemination. He urged the symposium participants to think about “issues of scale,” and the power of the small but dynamic firms that bring innovations to the marketplace. The Accelerating Pace of Global Competition He addressed the issue of global competition in general, and the rapid increase in scale and effectiveness in China, where becoming the dominant global manufacturing center is a national goal. India brings its own scale advantages and an especially entrepreneurial high-value culture. France and Finland are renew- ing and funding technology programs, with France committing to a new €1 bil- lion innovation agency. The key point, he said, is that the pace of competition is accelerating. To keep up, nations need to strengthen their science and technology base, maintain an open system of trade, create incentives for R&D and knowledge transfer, and foster innovative small businesses.

OCR for page 90
 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE Small Business as Key Driver of Innovation Small businesses are a key driver of the U.S. knowledge-based economy, he said, having generated 60-80 percent of net new jobs over the past decade. In 2003, the most recent year for which full data were available, small firms created 1,990,326 new net jobs, while large firms shed 994,667 net jobs. Small firms also employ 41 percent of all high-tech workers, including scientists, engineers, and computer workers, and produce 14 times as many patents per employee as large patenting firms. The U.S. economy is large enough so that small firms have unlimited poten- tial to grow, as illustrated by the cases of Intel, Microsoft, and Google. This is seldom true for small economies; in Sweden, for example, no new large firms have developed since 1970. But even well-run small firms with promising ideas face major challenges to growth. Potential financial backers, such as venture capital firms, have the problem of understanding and forecasting what a nascent firm can do. Firms propose many good ideas, and even more bad ideas. For the private equity community, the problem is one of sorting and discovery, and suc- cess is never guaranteed. The Danger of the “Valley of Death” The greatest danger to a small firm comes at the stage of development after the end of public support and before the availability of private support, which is traditionally provided by venture capital (VC) firms. This stage, often called the “valley of death,” is critical for the small firm that must develop a prototype, develop a commercializable product, and organize a sound management team. (See Figure 9.) With the recent shrinkage of seed funding for new firms and the shift of military budgets away from basic research and toward weapons testing, however, this “valley” may have widened in recent years, making the route to commercial- ization even more daunting. Once an inventor’s personal funds and the support of friends or “angels” is exhausted, there remain significant and expensive tasks before a VC firm is likely to be interested in participating.27 “We have a powerful myth in the United States that free markets can do it all,” said Dr. Wessner. “And to many, suggesting a role for government in this process is confused with ‘picking winners’. In a sense, we are always picking winners with our public policies, and what we want to do is find the best young companies and help them perform.” 27For more discussion of this trend, see the discussion by Professor Good earlier in this volume.

OCR for page 90
100 INNOVATIVE FLANDERS Uncertainty and University Research, Distance to Sometimes DARPA funding Market Startup: Friends, Families & Fools Curiosity research Need for Supportive Seed: Angel Backers The Financial Policy Framework Strategic research “Valley of Applied research Death” SBIR and ATP The Focus of provide initial SBIR Prototype funding Product 1st Round VC development Commercialization Capital Allocation 2nd Round VC Curve Business development Expansion Total Allocated Investment Resources FIGURE 9 U.S. innovation curve. PROC Figure 09 An Environment Friendly to Entrepreneurs Dr. Wessner noted that the United States does at least maintain a policy environment that is friendly to entrepreneurs, and that the culture as a whole does support innovation. And the United States has been effective in promot- ing innovation through a variety of mechanisms, such as industry-led consortia for standards and joint research, university-based research, joint ventures with the Advanced Technology program (ATP), and the Small Business Innovation Research (SBIR) program, which grants support for proof of principle and proto- typing. Although industry provides the greatest share of funding for early-stage technology development in the United States, the federal government is estimated to provide between 20 and 25 percent. By contrast, he said, some features of European culture have restrained inno- vation, such as punitive bankruptcy laws and a culture of caution that avoids the risks inherent in small-firm formation. Crossing the Valley with the Help of SBIR The United States does have a role in helping firms across that valley of death, Dr. Wessner said, and the most assistive mechanism is the SBIR program. This program, created in 1982 and renewed in 1992 and 2001, has a budget of $2.2 billion, and participation by all federal agencies with an annual extramural R&D budget greater than $100 million is mandatory. The SBIR is not a procure-

OCR for page 90
10 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE public fund. It is funded by the sale of shares and is budget-neutral for the gov- ernment. IMEC receives some of its financing from VINNOF. A problem, said Mr. Aernoudt, is that it is very small, with only about €5 million distributed so far among 20 firms. To finance early-stage and expanding firms of larger size, the government created the Arkimedes program that is modeled on the U.S. Small Business Investment Corporation. “Flemish people have money,” he said, “but don’t like risk. So we fund this program by offering bonds with a 100 percent government guarantee.” Investors may also invest cash in the fund, receiving a 90 percent gov- ernment guarantee plus an 8.75 percent tax credit per year. If Arkimedes should lose all its money, investors are still guaranteed a positive ROI of 2 percent. The bonds have been sold and the proceeds divided into four Arkimedes funds, or ARKIVs. Mr. Aernoudt said that the Flemish mechanisms of debt financing had a theoretical basis in the Modigliani-Miller Theorem, which states: • The value of a firm is determined by its earning power and the risk of its underlying assets, and is independent of the way it chooses to finance its invest- ments or distribute dividends. • Under certain assumptions, it makes no difference whether a firm finances itself with debt or equity. As a consequence, there is a bias toward debt financing that leads to undercapitalization. He concluded that Flanders’ programs were able to make good use of abun- dant public money by using these co-investment schemes, and that they brought fiscal advantages to those who invested in innovative SMEs. His summary of the situation was that “Flanders is a paradise for innovative companies.” Discussion A questioner asked Mr. Aernoudt how his department would measure suc- cess. He responded that success would not be measured by tax revenues, because “we have perverse taxes.” He said they were using such milestones as the num- ber of innovative companies supported, and that the problem of good research results not being converted into real companies had been solved by using avail- able money. They also measured the absorption rate, by looking at how well the money was used; evaluated the effectiveness of investment behavior by analyzing the default ratio of the companies; and measured innovation capacity by examin- ing the innovative actions of companies.

OCR for page 90
10 INNOVATIVE FLANDERS YOUNG TECHNOLOGY-BASED FIRMS IN BELGIUM: THE IMPACT OF POLICY INSTRUMENTS Bart Clarysse Vlerick Leuen Gent Management School Ghent Uniersity Professor Clarysse described his efforts to quantify the impact of public policy on young technology-based firms (YTBFs) in Belgium and to answer the question, why do some firms grow faster than others? He began by saying that he and a colleague29 had identified YTBFs by looking for firms that: • Were academic spin-offs • Were in the portfolio of VC funds • Had received R&D grants • Were part of a random sample of all startups in high-tech and medium- tech sectors. They identified and studied 235 examples of small firms in operation between 1991 and 2002, doing a retrospective reconstruction of the early growth path. They were looking for factors that distinguished fast-growing firms from those growing slowly or not at all. They found that the definition of high-growth firms varied with the growth measure used, because the picture is so complex. They focused primarily on the measures of revenue growth, employment growth, and asset growth. How to Account for Rapid Growth The first task was to try to account for rapid growth. They looked at a variety of potential determinants, including founding conditions (“imprinting”); initial resources, including people and their experience; and market strategy, including focused vs. diversified and local vs. international. They were surprised to find that for firms in general, venture capital funding by itself had a negative impact on growth. Small amounts of venture capital were “significantly negatively associated with growth in revenues and employees.” However, firms able to attract more than €1 million within 18 months grew fastest. So Professor Clarysse concluded that there were two categories of firms: those that were able to produce a business plan that attracted a lot of capital, and those that had to “bootstrap” themselves with very little start-up capital. Firms winning large amounts of VC usually had business plans based on micro- electronics or similar fields, with clear exit strategies. 29This report was based on work by Professor Clarysse and Mirjam Knockaert.

OCR for page 90
107 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE Rapid growth, he said, was not equal to sustainable growth. In fact, the faster- growing firms had the highest failure rates. Growth was sustainable only for firms that created “recurrent/stable” revenue streams before burning all their cash, and several of the high-employment-growth firms failed. A combination of investor acceptance and market acceptance, with a focus on the bottom line, most often led to sustainable growth. The Experience of Firm Founders The second task was to examine the experience of firm founders, concluding that firms grew faster when their founding teams were experienced. Teams of two to three founders with a mix of commercial and technical experience were the most successful—a finding that he said has “more explanatory power than the VC variable.” Only about 16 percent of the firms had this mix. Commercial experience of the founding team was important for early growth—so important that he did an in-depth analysis of founders. The solo entre- preneur was generally not successful. The “kinship team”—brothers, mother-in- law, etc.—was often a team of convenience or comfort and usually did not endure or find capital. The most successful teams were “organic”: the team was already in place before the business opportunity arose; members might have met at school or previous jobs. The fourth type was the “matched team,” with members of similar qualities who came together specifically for the venture. These did well for the short term, but were not sustainable. Team members came into personal conflicts and split up again. The Key Factor: International Experience The third factor they examined to explain growth was the international experience of the company. They found that firms with international experience from the outset were associated with high early growth in revenues and total assets, but that there was no significant effect on employment growth in the first years. They tried to compare these observations with existing theories, including organizational learning theory, and collected data on all stakeholders and key partners. They then fit these data with control variables, such as ambition of the entrepreneur, resources at founding, and industry sector. After extensive data analysis, Professor Clarysse concluded that expe- rienced, multidisciplinary teams with international exposure were necessary for growth. While most policy measures had focused on the need for equity financing to bridge the gap between startup and commercialization, he felt it was useful to look at small-firm growth from a human resources point of view as well. He concluded by calling for a “rejuvenated industry policy” that recog- nized the importance of internationalization in determining the success of young

OCR for page 90
10 INNOVATIVE FLANDERS technology-based firms. More specifically, he said, this internationalization is most significant when present from the early stage, and when it is accompanied by founders from an international background and the participation of inter- national partners. Discussion A questioner asked why he had measured success in terms of revenue, rather than profits, and whether a business that went public early was more or less successful than one that went public later. Professor Clarysse said that they had chosen revenue as a measure because it was more likely to generate employment than profits, which could be tied to specific products rather than steady growth. As for the timing of an IPO, he said that only 6 of their sample of 235 firms went public, so that the measure would not have been helpful. CONCEPT AND EVALUATION OF THE ADVANCED TECHNOLOGY PROGRAM Marc Stanley National Institute of Standards and Technology Mr. Stanley, director of the Advanced Technology Program (ATP), began by describing its mission, which is “to accelerate the development of innovative technologies for broad national benefit through partnerships with the private sector.” He, like several other speakers, emphasized the seriousness of the U.S. seed funding gap and the “dwindling high-risk investments” in new firms, which includes decreased spending on research by industry. These conditions gave value to the question of whether the federal government should be involved in the national innovation system. The first condition he emphasized was that “the large U.S. venture capital market was not focused on early-stage firms.” Only 1.65 percent of total VC investments of $20.9 billion in 2005 went to seed or early-stage projects. 30 He illustrated the hypothetical progress of a young firm that typically faces its most severe cash flow pressures just when sources of capital were most scarce—when crossing the “valley of death.” Because this valley coincides with the stage just after technology creation, when a technology is being developed into something useful, it represents the period when a good idea may find a use in society and a commercially valuable form. 30National Venture Capital Association, 2005.

OCR for page 90
10 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE Large U.S. Venture Capital Market is Not Focused on Early-Stage Firms Startup/Seed Startup/Seed (1.65%) $346 million Early Stage (18.55%) Later Stage (34.37%) Total = $20.9 billion Expansion (45.42%) FIGURE 12 Breakdown of U.S. venture capital by stage of development. SOURCE: Based on 2005 data from National Venture Capital Association. PROC Figure 12 Technology Technology Early Commercialization Creation Development Successful Cash Cash Flow Flow Moderately Valley of Successful Death Time SBIR & ATP Unsuccessful Unsuccessful Federal Agencies, Initial Entrepreneur & Venture IPO Universities, Public Seed/Angel Investors Capitalists States Investors FIGURE 13 Public-private funding transition between innovation and invention. SOURCE: Adapted from L. M. Murphy & P. L. Edwards, Bridging the Valley of Death: Transitioning from Public to Priate Sector Financing, Golden CO: National Renewable Energy Laboratory, May 2003. PROC Figure 13

OCR for page 90
110 INNOVATIVE FLANDERS The Rationale for the ATP These conditions, he said, provided the rationale for the ATP, which is designed to help young firms through that transition. Some of the key features of the ATP are the following: • Industry leads in proposing and implementing projects. • Project selection is based on not only technical but also economic merit. • Projects must have well-defined goals and sunset provision. • Project selection is competitive, based on peer review (fewer than 15 per- cent are chosen). • The ATP is part of the National Institute of Standards and Technology (NIST), a major federal agency responsible for national measurement standards. Mr. Stanley emphasized that a good ATP candidate not only has a sound technical premise, but also a business plan, even in its early stage. Since the formation of the ATP in 1990, said Mr. Stanley, it had compiled a positive record of investing in the category of high-risk young firms. The ATP had made awards to 768 projects, which included 218 joint ventures and 550 single companies. Some 66 percent went to small businesses, and included the participation of more than 170 universities and 30 national laboratories. Of the total award amount of $4,371 million, about half was provided by industry. About 1,400 patents had been created as a result of these awards. Applicants may file for ATP awards either as single firms or as joint ventures. For single firms, there is a 3-year time limit and a maximum award of $2 mil- lion. The company pays indirect costs, and large companies must invest at least 60 percent of total project cost. Many small firms prefer to apply as part of a joint venture with a larger firm, however, which brings advantages: the time limit rises to 5 years, universities are more likely to participate, and there is no limit to the size of the awards, which average $10 million. These ventures may become partnership networks that include universities, consortia, and research labs, as well as the company. They also pay for the expenses of graduate students, and play a key role in the research goals of the federal government and the economic strategies of state governments. A Confusing Policy Background In general, he said, public-private partnering in the United States takes place against a mixed and often confusing policy background. On one hand, the Bayh-Dole Act and subsequent amendments have encouraged formation of new companies based on university research, but some universities have policies that restrict technology transfer. More generally, policymakers do not understand how limited is the amount of VC funding available to the smallest firms or the poten- tial importance of these firms to job formation and economic growth.

OCR for page 90
111 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE Short-Term Mid-Term Longer-Term Partnerships Commercialization Substantial product ECONOMIC IMPACTS (knowledge sharing) alliances revenues Increased R&D Attraction of capital Positive cash flow spending New products and Inter- and intra-industry Acceleration of R&D processes diffusion Technical Employment growth Inter-industry employment accomplishment effects (milestones, papers, Increased national output patents) Positive public ROI Total Economic Benefits Benefits to Awardees 10 or 0 1 2 3 4 5 6 7 8 9 more Years Project Start FIGURE 14 Outputs, impacts, and outcomes occur at different times. PROC Figure 14 Mr. Stanley said that the impact of the ATP, both on award recipients and on the economy in general, had been studied extensively, and was found to be considerable. The short-term and medium-term benefits for individual firms, for example, included the creation of partnerships, increased R&D spending, the attraction of new capital, and employment growth. Broader benefits included substantial product revenues, increased national output, and positive public ROI that make a difference beyond the benefits to the single firm. A History of High Returns Altogether, he said, the program had delivered a net return of almost $1 bil- lion from a government investment of about $2.1 billion. He cited the results of a recent survey of 36 projects which totaled $79.2 million in government invest- ments and delivered total revenues of more than $971.9 million. Revenues and

OCR for page 90
112 INNOVATIVE FLANDERS savings resulting from products or processes incorporating technologies devel- oped under ATP awards were said to exceed $2.7 billion. He emphasized that the ATP spent considerable effort on continual and rigor- ous self-evaluation, and gave several results of current findings: • Forty-five percent of projects had brought new technologies into commercialization. • Ninety-five percent of projects reported that ATP funding allowed for greater risk taking or longer time horizons. • Eighty-five percent of ATP projects involved collaboration. • ATP projects had produced more than 1,700 scientific publications and 1,400 patent applications. • Sixty-six percent of ATP projects were led by small companies. He concluded by calling the program a success in accelerating the develop- ment of technological advances and becoming a “strong causal factor” in moving innovations from the idea stage to the marketplace. THE CHALLENGE OF COLLECTING GOOD EVALUATION DATA Bart an Looy Flemish Policy Research Centre for R&D Statistics (SOOS) In 2000, the Flemish government decided to create a research center dedi- cated to the evaluation of its science and innovation policy. This center was charged with developing a system of indicators to quantify the results of R&D at Flemish universities, research institutes, and industry that could be used by policymakers to support appropriate science and innovation policy for Flanders. As a result, the Policy Research Centre for R&D Statistics (Steunpunt O&O Statistieken, or SOOS) was created in January 2002.31 SOOS began its activities with studies in three key areas: bibliometrics, techno- metrics, and innovation. As data sources they used the Web of Science (WoS) and ISI Proceedings databases and various sources of patent and innovation data from EC and OECD surveys. They also developed their own databases and have become “the only place in Europe where all these data sources are found together.” Creating Indicators In order to do their assessments, said Professor van Looy, SOOS had to develop an appropriate IT infrastructure and create S&T indicators for the gov- 31In 2007, the Flanders government approved a second generation of policy research centers, result- ing in, among other things, SOOS becoming SOOI (Steunpunt O&O Indicatoren).

OCR for page 90
11 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE ernment. They then launched specific studies of science and technology domains and science policy for the Flemish government, as well as original research to measure the effectiveness of policy measures and translate them into incentives. For the government they also began to publish the biannual Flemish Indicator- book on Science, Technology and Innoation and provide indicators and discus- sions of the meaning and dynamics behind the figures. They track such activities as R&D and innovation expenses by the business sector, sales share due to newly introduced products, and distribution of R&D by field. Among the long- and medium-term services for the Flemish government is an additionality analysis that asks, is public funding crowding out private investment? “We want to learn a lot about S&T policy,” he said. “Scientific Services”: Are They Good for Universities? Following a major debate in 1991 over the role of the universities, SOOS is making a significant effort to examine the role of entrepreneurial universi- ties within innovation systems and the impact of legislative frameworks. Since Flanders created a new “triple mission” for its universities, adding “scientific services” to the traditional activities of education and research, SOOS is trying to learn whether this mission is feasible, realistic, and justified. Is there added value for society? What is its overall impact—does it hamper or encourage collabora- tion between academia and industry? Part of SOOS’ objective, said Professor van Looy, is to help the universities ensure that new entrepreneurial activities do not jeopardize the traditional missions. At the same time, if the triple mission continues, he wanted to be sure that universities install procedures that result in a fair return for researchers and research groups. Some Positive Effects SOOS has already discovered a positive effect of university involvement in “science services” in the form of increased patenting. This effect was difficult to measure, he said. Many academic patents do not show up in patent statistics because the applicant is the company. Nonetheless, they demonstrated that for the period 1991-2001, when the university acted as assignee, patent activity increased, stimulating technological activity. No negative impact was observed on the collaboration between universities and companies, as measured by the transfer of ownership rights. He said that understanding innovation was complicated because it involved so many kinds of activities and talents. A contribution of SOOS was to combine many data sources, link activities in diverse fields of S&T, and find patterns based on different forms of information spanning the fields.

OCR for page 90
11 INNOVATIVE FLANDERS Long-term Commitment to Evaluation SOOS had also examined the relationship between technological performance and human capital, asking the question, Does educational strength (number of PhDs in S&T) contribute to high-technology performance? The answer was yes, there is a distinctive and considerable impact. To arrive at this answer, the researchers had to “disentangle causality,” to be sure that the human capital was driving technological performance, rather than the other way around. Professor van Looy concluded by praising the Flemish government for rec- ognizing the value of evaluating its innovation policy, and for its long-term com- mitment to the data, infrastructure, and people required. “This work could not be done,” he said, “without our proximity to scientific research, alignment with international standards, and ability to network both locally and globally.” Discussion A questioner asked what kind of jobs could be expected from the work he was describing. Professor van Looy said that “you have to have patience if you want growth—more than 2 or 3 years.” For an immediate, direct impact, he said, one could open up a flower shop. But to gain employment benefits from high- technology policies, a commitment must be made over a period of 10-15 years. He was also asked if the databases he described are publicly available. He said that SOOS had a licensing agreement that did not allow them to make the databases public. For patents, he said, they try to apply open-source logic as much as possible, but they could not infringe on license agreements they signed with researchers 5 years earlier. Professor Good turned to a more general discussion, commenting on the importance of assessment for any programs. The success rate of the ATP is high, she said, partly because the assessment of candidates is thorough at the point of application. She also mentioned the importance of entrepreneurial activity to industrial research. Too many companies are interested only in profitability, she said, with- out having a consensus on how to measure the quality and utility of their R&D. “It all comes down to people,” she said. “If you’re going to evaluate R&D, you have to assess the talent that’s put into the pool.” Dr. Spyns observed that Flemish programs are also careful to scrutinize proposals at the front end. Innovation in the Political Context Professor de Vuyst commented that an unfortunate aspect of evaluation is that skeptics can claim that researchers get the result they want. Evaluations are useful as support tools, he said, to make the case, but the significant force is the long-term political commitment. This is difficult to gain from politicians who have short-term horizons. In the Flemish case, he said, the support of the politi-

OCR for page 90
11 SESSION V: MEETING THE EARLY-STAGE FINANCE CHALLENGE cians had followed a series of historical accidents. The question was how to keep the policies in place. For the Netherlands, he said, election time was drawing near, and the innovation platform had dropped out of the debates, replaced by urgent short-term concerns such as reducing taxes. A goal was to use evaluation tools and other data to make the point that investing in knowledge is as important as investing in other basic structures. Is There a Long-term Commitment to Innovation? Professor Flamm said that the sums of money being spent on Flanders’ innovation measures were large for the size of economy. He asked if there was political support for the programs to continue. Dr. Spyns noted that the commit- ment was part of a government agreement to spend an additional €60 million each year, at least until the current government’s term ended in 2009, in addition to the investment in the innovation fund (VINNOF) of €150 million over 2 years. He also cited the publicity and awareness campaigns in the media. “The word innovation is everywhere,” he said. “We are pushing its importance into the minds of people.” Dr. Wessner asked Mrs. Moerman if she agreed. She said that Flanders has both the institutional structure and political consensus, and that she was certain the support was there. “I think that knowledge is linked with identity in Flanders,” she said, “in the same way we have a commitment to defense. What we need more of is leadership to fine-tune the policy measures across the system. We haven’t seen a lot of that.”