Technology Transfer Systems in the United States and Germany: Lessons and Perspectives (1997)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

246 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY scription is more detailed because the related sources need important enlarge- ment or additional explanation. (See, for example, the “Universities” section, below.) THE GERMAN R&D ENTERPRISE General Structures In 1994, German institutions spent about DM78 billion (or 2.3 percent of GDP) on R&D. This was equivalent to $35.9 billion (in purchasing power par- ity), or about 21 percent of total U.S. R&D spending. The German ratio of R&D spending to GDP, though slightly lower than the American ratio (2.5 in 1994), is higher than most other large industrialized countries.1 In recent years, the public- sector/private-sector composition of German and American R&D spending has converged. Between 1989 and 1994, the share of publicly financed R&D in Germany increased from 34 percent to 37 percent. Over the same period, the share of publicly financed R&D in the United States decreased from 46 percent to 39 percent. In Germany in 1994, about 8.5 percent of the public R&D budget was spent for defense purposes; in the United States, that figure was 55 percent. Due to the primarily civilian orientation of R&D in Germany, the share of publicly financed R&D performed by industry (13 percent) is relatively low compared with the share of such research performed by industry in the United States (31 percent) (Organization for Economic Cooperation and Development, 1995). These general indicators give only a rough sense of the German R&D sys- tem. Particular institutional structures will be described here in more detail, fol- lowing Meyer-Krahmer (1990) and Schmoch et al. (1996b). In Germany, the organization of R&D activities is shaped largely by the country’s federal system of government, in which public-sector responsibilities are more evenly divided between the central government and the states (Länder) than is the case in the United States. German states are principally responsible for the educational sec- tor and consequently finance the vast majority of university budgets, including more than 75 percent of academic research. The financial flows from the state- level ministries to universities are depicted by a boldface arrow in the organiza- tional chart in Figure 3.1. Roughly 90 percent of these funds are allocated by universities for base, or general-purpose, institutional support of research. Only 10 percent of university research supported by the state is linked to specific projects. In addition to universities, other research institutions are partially supported by state-level ministries (see Figure 3.1). These include the institutes of the Max Planck Society, the Helmholtz Centers, the institutes of the Fraunhofer Society, and “other institutions” (the Blue List institutes and independent institutions es- tablished by the states, including the An-Institutes). All these institutions are described in more detail in “Technology Transfer from Universities,” below. 

TECHNOLOGY TRANSFER IN GERMANY FIGURE 3.1 Organization chart of the German R&D system. SOURCE: Schmoch et al. (1996b). 247 

248 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY Another set of research institutions, called “departmental research institutes,” are connected directly to specific state-level and federal ministries. These insti- tutes often carry out general activities in addition to R&D activities in the area of the related ministry. Federal ministries with important ties to departmental re- search institutes include the Ministry of Health, the Ministry of Agriculture, the Ministry of Transport (included in “Other Ministries” in Figure 3.1), and the Ministry of Defense (BMVg). Compared with their counterparts in other coun- tries, German departmental research institutes account for a relatively small share of total publicly funded R&D. Nevertheless, they often play an important role in the R&D landscape; some of them are leaders in special R&D sectors. In the federal government, the most important source of R&D funding is the Ministry for Education, Science, Research, and Technology (Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie [BMBF]).2 The BMBF is responsible chiefly for R&D budgets and long-term research programs on a gen- eral level. The BMBF delegates more specific decisions to program agencies (Projektträger), which manage project-related activities for nearly all fields sup- ported by the BMBF (see Figure 3.1). A further intermediate institution between the BMBF and R&D-performing institutes is the German Research Association (Deutsche Forschungsgemeinschaft [DFG]) that is responsible for supporting mainly basic research projects, especially at universities. It is noteworthy that the DFG is funded jointly by the BMBF and state governments. Other institutions that perform intermediate R&D management are the MPG, the FhG, the AiF, and the Helmholtz Association of German Research Centers (Helmholtz-Gemein- schaft Deutscher Forschungszentren, HGF). This variety of decision-making institutions would seem to indicate a high degree of flexibility in German public funding of R&D. In reality, however, the great majority of public funds are earmarked for long-term commitments; only about 10 percent of the BMBF bud- get each year is available for new tasks (Meyer-Krahmer, 1990, 1996). In the following sections, the distinct function and important role in technol- ogy transfer of German universities, the MPG, the FhG, the HGF, and the AiF will be described in more detail. What most distinguishes these sets of institu- tions from each other is the focus of their research activity along the continuum of R&D activities. The MPG is chiefly oriented toward basic and long-term applied research; the FhG, toward mid- and short-term applied research; AiF supports cooperative industrial research projects that generally have a precompetitive but application-oriented character; Helmholtz Centers conduct their activities prima- rily in areas requiring long-term investments or entailing considerable economic risks. Some Helmholtz Centers concentrate mostly on basic research, while oth- ers work in fields of strategic industrial relevance. Figure 3.2 depicts the general structure of the German R&D enterprise. Along the horizontal axis, institutions are classified according to their main sources of funding, whether public or private. Most of the private-sector institu- tions are industrial research laboratories; the number and research volume of in- 

TECHNOLOGY TRANSFER IN GERMANY 249 FIGURE 3.2 Main R&D-performing institutions in Germany, expenditures in billion 1995 DM. SOURCES: Reger and Kuhlmann (1995); Schmoch et al. (1996b). dependent private research institutes are quite low. There is an intermediate class of institutions, most notably the FhG and the institutes of the AiF, which receives funding from both government and industry. The vertical axis displays the type of R&D conducted: basic research, applied research, and (experimental) devel- opment. The shading indicates major areas of performance of the different insti- tutions. Thus, as extreme examples, the MPG concentrates on long-term basic research, whereas research in industry is mostly short term and application ori- ented, with time horizons on the order of 3 to 5 years. The sizes of the bars indicate the annual budgets of the respective institutions.3 Of course, the R&D activities of the different research organizations are not as clear-cut as Figure 3.2 makes it seem. Thus, it is not surprising that industrial laboratories perform some basic research (about 6 percent of their total internal R&D; cf. SV-Wissenschaftsstatistik, 1994) and public research institutes and uni- versities perform some applied R&D. Nevertheless, it is important to know in which major areas the different institutions are working in order to understand relevant distributions of capital and manpower. The greater orientation of institutions on the left side of Figure 3.2 toward basic research reflects their commitment to supporting the research needs of non- 

250 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY economic (in the broad sense), societal goals. The public sector tends to support earlier stages of the innovation cycle, whereas industry concentrates on later phases. Technology transfer from the left to the right side is important for the efficiency of the total system. All in all, Germany has only a small number of completely private, profit- oriented research institutes. Instead, the intermediate position between universi- ties and industry is occupied by nonprofit institutions, namely the FhG and AiF, both of which operate with some public funding; these institutions do applica- tion-oriented research. The MPG and the Helmholtz Centers largely supplement the activities of universities in the areas of basic and long-term research. Among the public or semipublic institutions, the sector consisting of departmental re- search institutes is small compared to the Helmholtz Centers, the MPG, and the FhG. The ratio of R&D expenditures in universities, other research institutions, and industry is 1:0.7:3.5. As can be seen, the institutional sector lying between universities and industry is quite large. Industrial R&D Structures ORIENTATION OF INDUSTRIAL R&D To better understand the nature and dynamics of technology transfer to in- dustry by German universities and other public and semipublic research institu- tions, it is important to appreciate the comparative R&D and technological strengths of German industry. In this context, European patent data offer a useful window on the relative technological strengths and weaknesses of German indus- try.4 A recent study by Schmoch and Kirsch (1994) compared Germany’s share of patents in 30 separate technology fields with the average share for the rest of the world in each field.5 Using an indicator of specialization, the study identified industries in which German patenting was above or below the world average. The results show a strong orientation toward fields in mechanical engineering, such as machinery, engines, handling, and transport (Figure 3.3). Indicator val- ues for consumer goods and civil engineering are also above the world average. Fields such as organic chemistry, basic material chemistry, and polymers gener- ally show average or positive values, whereas biotechnology and pharmaceutical research (which is linked to biotechnology) show values distinctly below aver- age. Finally, information technology and related fields such as audiovisual tech- nology and telecommunications show below-average values. One can conclude from these data that German industry is marked by a strong emphasis on me- chanical engineering, a conclusion that is supported by international trade statis- tics (Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie, 1997; Gehrke and Grupp, 1994; Häusler, 1989). Indeed, in Germany there are a variety of innovative SMEs conducting research related to mechanical engineer- ing that have a distinct focus on export. 

TECHNOLOGY TRANSFER IN GERMANY 251 Electrical energy Audiovisual technology Telecommunication Information technology Semiconductors Optics Control Medical engineering Organic chemistry Polymers Pharmaceuticals Biotechnology Materials Agriculture, food Basic materials chemistry Process engineering Surfaces Material processing Thermal processes Environment Machine tools Engines Mechanical elements Handling Agricultural machines Transport Nuclear engineering Weapons Consumer goods Civil engineering -100 -80 -60 -40 -20 0 20 40 60 8 0 100 Specialization index FIGURE 3.3 Specialization index of European Patent Office (EPO) patents of German origin in relation to the average distribution at the EPO for the period 1989 to 1991. SOURCE: Schmoch and Kirsch (1994). 

252 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY The moderate specialization indexes for microelectronics and information technology are confirmed by foreign trade statistics: The related specialization (RCA) index in the areas of computers and semiconductor devices is distinctly below average (Gehrke and Grupp, 1994). In part, this result reflects the fact that only a few large companies—Siemens, Temic Telefunken, Bosch, and the Ger- man subsidiaries of IBM, ITT Semiconductors, Philips, and Texas Instruments— are internationally competitive in these areas of research. This means that the German public and semipublic research institutions have only a few resident in- dustrial counterparts capable of supporting significant levels of intra- and extra- mural research. (For further details, see “Technology Transfer in Microelectron- ics,” below.) The low specialization index in biotechnology has to be interpreted in light of the very high level of U.S. activity, which largely determines other nation’s average share of biotechnology R&D. Nevertheless, the moderate indicator for Germany reflects a quite hesitant start on the part of the big chemical companies. Current activities are often based on affiliations and acquisitions in the United States, whereas research in German laboratories is still at a moderate level.6 The patent profile of the United States differs significantly from the German one. The United States has positive index values in the fields of information technology, semiconductor devices, and biotechnology and negative ones in me- chanical engineering and consumer goods (Figure 3.4). The closest correspon- dence to the German profile can be found in the fields of organic chemistry, polymers, and basic materials chemistry, which have above-average specializa- tion indexes in both countries. Also in both countries, the specialization profiles are generally stable over time. In comparing the German and American technol- ogy transfer systems, these differences in the orientation of industrial R&D have to be borne in mind. TECHNOLOGY TRANSFER TO SMALL AND MEDIUM-SIZED ENTERPRISES Growing technological and market demands have fostered considerable growth of R&D cooperation and technology transfer between large companies and noncommercial R&D institutions in Germany. Although German SMEs face many of the same challenges that have prompted large firms to seek external sources of technology and R&D, R&D cooperation between SMEs and noncom- mercial R&D institutions does not appear to be as widely established as that involving large companies. Admittedly the collaborative research activities of German SMEs have not yet been studied extensively. Some analyses, however, suggest that in recent years, SMEs, especially in the manufacturing sector, are relying increasingly on technology transfer from external research institutions. According to a joint survey by the Fraunhofer Institute for Systems and Inno- vation Research (FhG-ISI) and the German Institute for Economic Research (Deutsches Institut für Wirtschaftsforschung (Becher et al., 1989), approximately 

TECHNOLOGY TRANSFER IN GERMANY 253 Electrical energy Audiovisual technology Telecommunication Information technology Semiconductors Optics Control Medical engineering Organic chemistry Polymers Pharmaceuticals Biotechnology Materials Agriculture, food Basic materials chemistry Process engineering Surfaces Material processing Thermal processes Environment Machine tools Engines Mechanical elements Handling Agricultural machines Transport Nuclear engineering Weapons Consumer goods Civil engineering -100 -80 -60 -40 -20 0 20 40 60 8 0 100 Specialization index FIGURE 3.4 Specialization index of European Patent Office (EPO) patents of U.S. ori- gin in relation to the average distribution at the EPO for the period 1989 to 1991. SOURCE: Schmoch and Kirsch (1994). 

254 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY 25,000 SMEs of the former West Germany engage in R&D. A further study by FhG-ISI and Prognos (Wolff et al., 1994) estimated that in 1991 13,000 German firms were conducting cooperative R&D and about 3,000 to 5,000 R&D-per- forming firms offered plausible reasons why they were not engaged in collabora- tive R&D. Still, a significant number (7,000 to 9,000) of R&D-performing SMEs could potentially engage in collaborative research but have never done so (Wolff et al., 1994). The FhG-ISI and Prognos study differentiated between “hard” and “soft” technology transfer, as follows: • R&D cooperation (hard) consists of contract research by third parties (com- panies, public or industrial research facilities, universities, technical colleges, engineering offices) and joint R&D with or without a contractual basis; • Technology-related activities (soft) includes informal contacts for the pur- pose of information exchange, performance of technoeconomic studies, joint utilization of laboratories and other testing instruments and facilities, employment of university students as trainees or interns, and the prepara- tion of a graduation or doctoral thesis. Whereas 50 percent of all SMEs surveyed are involved in R&D cooperation, 30 percent declared that they practice cooperation in “less active technology-re- lated activities” (Kuhlmann and Kuntze, 1991). With respect to the importance of potential partners, there are significant differences between the two types of technology transfer. In the area of R&D cooperation, customers and consulting engineers play a vital role, whereas universities and research institutes are of medium importance, and the impact of polytechnical schools (Fachhochschulen) is negligible (Figure 3.5). With respect to cooperation in technology-related ac- tivities, polytechnical schools and suppliers are the SMEs’ most important part- ners. Universities and research institutes again occupy a middle position. Although German SMEs appear to be drawing effectively on the technology transfer abilities of customers, suppliers, and consulting engineers, some observ- ers believe that the capabilities of university polytechnical schools and research institutes are underutilized by SMEs. In general, the knowledge generated by universities, polytechnical schools, and research institutes is sought by an SME when the firm needs to understand unfamiliar techniques, wants to make use of testing equipment, or is seeking new approaches. SMEs identified three major impediments to greater collaboration with universities, polytechnical schools, and research institutes: • the low level of interest displayed by these research institutions in the specific research needs of SMEs (47 percent); • the high cost to SMEs of cooperating (mentioned by 44 percent);7 and • the perception by SMEs that these collaborations do not lead to usable results quickly enough (42 percent) (cf. Wolff et al., 1994, p. 166). 

TECHNOLOGY TRANSFER IN GERMANY 255 customers 50% suppliers: (export-)dealers raw-materials, pre-products companies in same suppliers: technology fields machinery, equip., tools 50% 50% public testing consulting laboratories engineers R&D universities/ cooperation polytechnics research institutes other forms of techno 50% related cooperation FIGURE 3.5 Partners of SMEs in R&D and technology-related activities, by percent. SOURCE: Wolff et al. (1994). Herden (1992) asked 1,349 German SMEs about the type and frequency of their contacts with universities and other research institutes during the past 5 years. The results of this study (Table 3.1) further verify the above findings. According to the Herden data, a mere quarter of all SMEs received techno- logical knowledge from research institutions or universities. Among those that did, the most frequent type of technology transfer was soft contacts without R&D cooperation. In this regard, consulting on the solution to a problem was the most important channel (cited by 69.8 percent), whereas licensing, which may be viewed as another information channel, was significantly less important (men- tioned by 9.1 percent). Training of qualified personnel at universities ranked second in importance (45.4 percent). The joint implementation of R&D projects (i.e., hard cooperation) ranked third (33.5 percent). Nonetheless, only about 8 per- cent of the 1,349 firms surveyed cooperated in joint R&D projects with academic institutions. Unfortunately, this survey did not ask firms about the temporary assignment of scientific personnel from universities to firms, therefore, the fre- quency of this kind of personnel transfer cannot be measured. The level of SME cooperation with universities and research institutions is surprisingly low compared with the potential impact of the scientific knowledge that could be gained through such partnerships (Schmoch et al., 1996b). Tech- nology transfer from universities and other research institutions to SMEs could be improved by using public subsidies to reduce the cost of collaboration. It is estimated that roughly 30 percent of the SMEs involved in R&D cooperation 

256 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY TABLE 3.1 Types of Knowledge Transfer from Academia to Industrya Channels for technology transfer (multiple choices possible) Percent Consulting on problem solution 69.8 Training of qualified personnel at universities 45.4 Joint implementation of R&D projects 33.5 Subcontracting of R&D projects 25.9 Sharing of laboratory and equipment 24.1 Information on the market potentials of new products 17.7 Directed search for R&D personnel 17.4 Directed search for recent graduates (non-R&D personnel) 14.6 Licensing 9.1 Short-term assignment of R&D personnel to universities 5.2 a The survey question was, “Have you directly obtained technical knowledge from research institu- tions and/or universities during the past five years?” Of the SMEs surveyed, 24.5 percent answered “yes,” 67.1 percent answered “no,” and 8.4 percent answered “not yet but planning to.” SOURCE: Herden (1992). could benefit from public support (Kuhlmann and Kuntze, 1991). However, much more difficult to remedy is the perception among SMEs that nonindustry research institutions are not particularly interested in the research problems of SMEs. Here, technology transfer units of universities and polytechnical schools could prove their efficiency by improving mutual understanding of the different research needs and capabilities of SMEs and research institutions (Kuhlmann and Kuntze, 1991). In general, SMEs are considered important pillars of the German innovation system (Harhoff et al., 1995). Although they contribute to new and emerging areas, their specific strength is the rapid diffusion and adaptation of existing tech- nologies. In this regard, they can draw upon the resources of a variety of R&D- performing, transfer-oriented institutions such as Fraunhofer institutes and the research institutes of industrial research associations. (For details, see “Fraun- hofer Society” and “Federation of Industrial Research Associations,” below.) Furthermore, a dense network of non-R&D-performing institutions supports technology transfer through innovation-oriented consultancy and the organiza- tion of knowledge exchange among firms. All Chambers of Industry and Com- merce (Industrie- und Handelskammern) offer consultancy services concerning not only technological innovation and potential cooperative partners but also fi- nancial problems relating to investment and public support programs. The Cham- bers of Crafts (Handwerkskammern) offers the same services for craftsmen (Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie, 1995a; Bundesministerium für Forschung und Technologie, 1993b). Both insti- tutions are legal representatives of commercial enterprises in Germany. The Chambers of Industry and Commerce are financed completely by industry; the Chambers of Crafts receive considerable public support. 

TECHNOLOGY TRANSFER IN GERMANY 257 Another important institution is the Organization for Rationalization of Ger- man Industry (Rationalisierungs-Kuratorium der Deutschen Wirtschaft), which is jointly financed by industry, trade unions, the federal government, and the states. It supports SMEs in areas of management, organization of production, and personnel training (Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie, 1995a). The diffusion of knowledge within German industry is also fostered by about 650 industrial associations (Industrie-Fachverbände) representing all industry sectors (Hoppenstedt Verlag, 1995). The principal roles of industrial associa- tions are to represent their member companies politically and to promote discus- sion of relevant commercial, political, or technical problems in their own journals and meetings. The 106 industrial research associations comprise an important subset of this larger group; they conduct cooperative research and are organized under the umbrella of the AiF.8 There are also about 400 associations for the advancement of science9 (Wissenschaftliche Fachgesellschaften), representing more than 400,000 mem- bers from across the science disciplines. About half of the members are in the engineering sciences and their associations are often called Technical-Scientific Associations (Technisch-wissenschaftliche Vereine und Gesellschaften), most of them being organized in the German Federation of Technical-Scientific Associa- tions (Deutscher Verband Technisch-Wissenschaftlicher Vereine). About 60 per- cent of the members of the engineering associations come from industry; the rest are from research institutions and universities. By contrast, in the natural sci- ences, the share of industrial members is 20 percent. The major aim of the scien- tific associations is to initiate discussion on recent research results and (especially in the natural and engineering sciences) to facilitate the transfer of knowledge between scientific institutions and industry (Schimank, 1988b; Wissenschaftsrat, 1992). Important instruments are the diffusion of knowledge through journals, conferences, or professional continuing education. Finally, the states have a key position in supporting innovation by SMEs. All states support the research and technology transfer needs of SMEs, by estab- lishing information units at universities and by providing funding for innovative projects, specific technologies, and new technology-based firms. An interesting example is the Steinbeis Foundation (Steinbeis-Stiftung) in the state of Baden- Württemberg, which established a network of 200 technology-transfer units at polytechnical schools. The foundation arranges contacts between companies looking for solutions to specific problems and the appropriate professors in the network. This system has proved very successful, and the Steinbeis Foundation plans to extend its network to other states. With respect to the federal government, the most important measures for promoting technology transfer are the establishment of transfer-oriented research institutions (see “Fraunhofer Society” and “Universities, An-Institutes and Other External Institutions,” below) and the cooperative research programs (Verbund- 

258 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY forschung) of the Ministry of Education and Research (see “Universities, Statis- tics on General Research Structures,” below). Notably, there are also 11 federal technology transfer centers, each oriented to a specific technology (e.g., biotech- nology, laser technology, production, textiles). These centers were established by the Ministry for Economic Affairs. Overall, technology diffusion in industry, particularly to SMEs, is efficiently fostered by a high level of industrial self-organization, the roots of which go back to the nineteenth century (Lundgren, 1979). These activities are backed up by and intertwined with a broad variety of public institutions and measures for sup- porting technology transfer. CONDITIONS FOR NEW TECHNOLOGY-BASED FIRMS With their adaptability and high potential for innovation, NTBFs influence the structural change of an economy. They are regarded as a stimulus for dy- namic development because they • increase the number of market competitors and therefore motivate estab- lished companies to strengthen their efforts to innovate; • increase the demand for services that support innovation; and • strengthen the regional suppliers of technical products, since their own manufacturing penetration is rather low (Kulicke and Wupperfeld, 1996). NTBFs can be found in fields such as mechanical engineering, electrical engineering, electronics, process engineering, environmental technology, biotech- nology, and medical technology. However, a significant number of economists and politicians deplore the lack of NTBFs in Germany in strategic fields of high technology. For example, more than 350 American biotechnology start-up com- panies were counted between 1971 and 1987 (Dibner 1988). The number was significantly lower in Germany. These start-ups are regarded as the basis for the outstanding position of the United States in this technological field (Kulicke, 1994). The establishment of spin-off companies by scientists presently working for other, usually much larger industry or university research units is one of the most effective channels of technology transfer. The success of U.S. NTBFs in the computer hardware and software industry prompted the German government in the early 1980s to support establishment of NTBFs by awarding competitive grants for up to 75 percent of a start-up company’s R&D costs. Although these early efforts did not meet expectations, it did demonstrate that there is a potential for NTBFs in Germany. The specific requirements of this type of start-up indi- cated a new strategy was needed, aiming at directly activating market forces by involving nonpublic investors (such as venture capital companies, private inves- tors, companies, or banks) in support of NTBFs. Instead of direct subsidies to NTBFs, incentives such as refinancing, deficiency guarantees, and co-financing were offered to venture capital investors. 

TECHNOLOGY TRANSFER IN GERMANY 259 Because of the poor employment situation and the weak industrial base in East Germany after German unification, government support for NTBFs has been revitalized and improved since 1990 (for details, see Abendroth, 1993; Kulicke, 1993). Since 1990, more than 550 NTBFs have been established in the new federal states, but with declining rates in recent years. Experts agree that despite these various public measures, the number of NTBFs in Germany is still rather low compared to the situation in the United States. An exact comparison of the annual rate of formation of NTBFs, however, is not possible because the available statistics and estimates are based on different definitions and demarcations. Neither country’s record in this area can be re- garded as optimal. The very high U.S. rate of NTBF formation is tarnished by the high failure rate of new companies, the tendency to destructive competition, and the associated costs to the U.S. economy (Florida and Smith, 1993). In Germany, the number of NTBFs is too low, but their survival rate is very high compared with other German start-ups in trade and service sectors (Kulicke, 1994). In Germany, there are a number of barriers to the successful establish- ment and operation of a sufficiently large number of NTBFs (see Kulicke and Wupperfeld, 1996). Some of the most important include: • Limited availability of capital for firm setup, the financing of develop- ments, and market entry. Failure to acquire capital is the main problem for German NTBFs. Because NTBFs do not have records of market suc- cess or because their founders cannot furnish sufficient equity guarantees, banks are reluctant to loan them money. Also, many banks lack techno- logical knowledge and for this reason hesitate to finance what they view as “risky” operations. • Limited managerial know-how on the part of the company founders. Founders of technology-based firms often come from the natural sciences or engineering and have limited management, marketing, and financing experience. As a result, they fail to develop a strategic concept, which banks need to assess the risks of investment. Links (or networks) to sales partners, cooperation partners, or suppliers are often not established, a situation that cries out for a supportive management. • Barriers to market entry. Because NTBF founders have limited knowl- edge of markets and market forces and little experience in marketing, market entry is difficult. NTBFs are further hampered by the fact that they do not possess a brand name or product image. • A shortage of qualified and experienced management personnel. Most of NTBFs cannot pay high salaries; therefore, they are less attractive to po- tential employees, including skilled managers. • Unfavorable taxation. German law does not provide preferential tax treat- ment of gains from venture-capital investment. However, analyses in the United States, Canada, and Great Britain show that high capital gains tax 

260 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY rates are negatively correlated with investment in venture capital opera- tions. In Germany, taxation is also a negative factor when investors sell their shares: Whereas the situation is more favorable for individuals, investor-owned corporations and other forms of “legal entity” have to pay income taxes on any profits, regardless of their form or value (see Pfirrmann et al., 1997). The federal and state governments are working to overcome these barriers by various support programs. In West Germany, however, there is no longer a fed- eral support program exclusively designed for NTBFs. Rather, financial support for NTBFs is now incorporated in a new program for small technology-based firms altogether. At present, most NTBFs participate in federal or state programs, and public programs are NTBFs’ primary source of external financial support. Most of the support is not given as a direct subsidy, but rather as equity stock. The promotion of NTBFs has been generally successful, and there are posi- tive examples of successful initial public offerings (IPOs). However, there are some shortcomings of public support for start-ups, including that • companies receive too little support. The aid generally covers 40 percent or less of their capital demand; for the remaining 60 percent, other private sources have to be found; • public equity stock institutions offer only financial, not managerial, support; • the programs do not offer a holistic approach; rather, they finance seg- ments of a firm’s business. Generally, such support focuses on invest- ment in capital goods and does not cover expenditures for staff; and • the programs aim primarily at the first stage of company set-up. The later stage of market entry, which demands considerable capital, is not covered. Today, most of the federal states act according to a special SME policy that promotes NTBFs and SMEs through so-called SME equity stock companies (mittelständische Beteiligungsgesellschaften [MBGs]). MBGs receive funding from Chambers of Industry and Commerce and regional banks and work in close regional cooperation with credit institutions. Their business policy is largely determined by whether public programs offer refinancing and failure guarantees, since their own funds are rather limited (Kulicke, 1990). They also have only limited capabilities to give managerial support to their portfolio firms. The German venture capital market is dominated by business investment companies (Kapitalbeteiligungsgesellschaften) of banks, savings banks, other credit institutions, and insurance companies, which are more interested in capital gains and invest little or no capital in NTBFs. Furthermore, there is a small number of independent German venture capital companies that coordinate the interests of industrial firms, banks, fund managers, and foreign venture capital firms (Wupperfeld, 1994). 

TECHNOLOGY TRANSFER IN GERMANY 261 In the United States, the use of venture capital to finance NTBF start-up is relatively commonplace. The American concept, combining equity as well as technical and management support, is not working that well in Germany, how- ever. This is in part because of legal regulations, but also because traditional ways of doing business are difficult to change. Even if habit and mind-set prob- lems are easy to identify, it is quite impossible to verify their impact and the scope of their influence on the difficulties faced by NTBFs. And while it is possible to modify habit and mind-set on an individual basis, broad structural change is taking place very slowly. Nevertheless, it is necessary to address these “soft” factors because they point to the limits of a sudden change in regulations. Legal restrictions can be addressed more easily. Of particular relevance to NTBFs are the restrictive regulations concerning bankruptcy and liability. The bankruptcy law (Konkursordnung [KO]), dating back to 1877, states that every partnership and legal entity can apply for the commencement of bankruptcy pro- ceedings and can be made personally liable. Some companies have no limits on their liability. If bankruptcy is declared, the partnership is automatically made personally liable in the event that the company’s equity capital is insufficient to pay off debts. But even with limited liability, partners can be made liable beyond the level of their investment in the firm. In general, bankruptcy is closely con- nected to the securities offered to and demanded by banks: Debtors with a weak financial background, in particular, are often made personally liable. For the GmbH (Gesellschaft mit beschränkter Haftung), banks often demand personal liability when a limited partnership has liable equity capital of at least DM 50,000. The same standard can apply to legally formed corporations (Aktiengesellschaft [AG]), where the partnership can also be made personally liable. In this case, private assets are used to repay excess business debt by means of an attachment. According to the regulations, employee claims have first priority, followed by “ordinary” business creditors, employee pension plans, and public social insur- ance and pensions (§ 61 KO). A number of other public entities follow those four, but very often private assets are not sufficient to pay back all the claims. This threat of personal liability in bankruptcy contrasts sharply with the situ- ation in the United States. If somebody goes out of business in the United States, he or she faces almost no problems starting another business. In contrast, one failure in Germany almost always ends the dream of operating one’s own busi- ness. Therefore, the risk inherent in establishing an NTBF is higher in Germany than it is in the United States. The apparent risk-averse mentality of founders of German NTBFs can be connected directly to these legal restrictions. Two-thirds of American venture capital is administered by independent funds, one-fourth is handled by corporate venture capital firms, and the remain- ing portion is held by small business investment companies (SBICs). By con- trast, the German venture capital and equity stock market is dominated by subsid- iaries of banks. Some general characteristics of banks, their goals, and attitudes may hamper their supposed supportive function. 

262 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY • Banks do not possess the necessary technological knowledge to assess accurately the risks of investing in NTBFs. Their risk-diversifying port- folio strategy limits the financing of uninsurable and risky operations. • The universal banking system in Germany and the different inherent func- tions of saving, lending, and issuing bonds means banks can lose their reputations if a business they finance fails, and this loss can have a nega- tive impact on their saving and lending functions. Therefore, banks do not want to take the risk of issuing bonds for NTBFs. Banks are also concerned about a possible loss in reputation and the profit margin, which is low for a small firm compared with a large one, when private busi- nesses go public. For these reasons, banks prefer to issue credit to NTBFs without any further commitment to the start-up. • Banks follow the strategy of constant returns. Therefore, rather than rein- vesting their capital gains, NTBFs are required to pay dividends or inter- est to the bank or the equity stock company. This policy limits the growth potential of NTBFs. One of the main barriers to success of the American venture capital model in Germany is the virtual impossibility for SMEs to go public. An equivalent to the U.S. over-the-counter market, which allows investors to sell off their shares in a start-up company, does not exist in Germany, but will be established in spring 1997. Venture capital companies therefore have faced a relatively low rate of return when investing in NTBFs. Going public in Germany is only possible for corporations (i.e., the legal form of AGs) and can only be done following strict and conservative financial requirements. An attempt has been made to remedy this problem by creating the “small corporation” (Kleine AG), which is linked to significantly fewer financial and bureaucratic requirements. In November 1996, a European stock exchange, the European Association Securities Dealers Auto- mated Quotation (ESDAQ), came into being for NTBFs and small technology- based firms. Its purpose is to promote the concept of venture capital companies, as the individual national markets of the EU are too small to match the supply and demand of the relevant actors on the market. Another barrier is that there are no tax privileges for share capital and capital share gains, which are major incentives in the United States. Resistance to venture capital investment is not encountered only on the sup- ply side. Venture capital and the underlying concept are not widely accepted by the German founders of NTBFs. Founders vehemently oppose equity stock capi- tal and venture capital because the investors are accorded executive rights (Kulicke, 1993). When faced with the financial difficulties associated with launching an NTBF, a majority of individuals change their minds about wanting to start their own companies. Most of those who do attempt to form NTBFs favor remaining independent (Kulicke and Wupperfeld, 1996). Even if NTBFs accept managerial help and agree to share the executive right of decision making, the 

TECHNOLOGY TRANSFER IN GERMANY 263 limited competence of German venture capital and equity stock companies in anything but financial matters can create problems. NTBF founders recognize the financial expertise of these firms but lament their lack of other supporting competences (Kulicke and Wupperfeld, 1996). The main current impediments to a functioning venture capital market in Germany, according to Kulicke and Wupperfeld (1996), are • a lack of attractive “exit routes” for venture capital companies to achieve high rates of return from NTBFs going public; • no favorable tax treatment for investors in venture capital companies or for venture capital companies themselves; • avoidance of risk by investors and venture capital and business invest- ment companies; and • aversion to loss of independence on the part of NTBF founders and entre- preneurs. There are a variety of steps that could be taken to increase the usefulness of the venture capital option for German NTBFs. These include: • allowing a tax reduction for investors’ contributions to special funds and reducing the applicable capital gains tax rate; • strengthening the pan-European stock exchange for NTBFs; • teaching managerial skills in natural sciences and engineering schools; and • improving the competence of venture capital and equity stock companies to assess financial and technological risks and to deepen their knowledge of technology. To sum up, in Germany, NTBF formation is discouraged by an unfavorable financial, legal, and social environment. As a result, this important instrument of technology transfer is used insufficiently. However, various steps are being taken to adapt this means of technology transfer, which has proved very successful in the United States, to the specific conditions in Germany and Europe. Impact of European Research RESEARCH PROGRAMS OF THE EUROPEAN UNION The Single European Act, ratified in 1987, formulated a European research and technological development policy. Its most important aim was to strengthen the international competitiveness of European industry in technology-intensive sectors such as information and communication technologies, the biosciences, and materials research. The policy’s main instruments are the Framework Programs of Community Activities in the Field of Research and Technological Development. Practical 

264 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY realization of the framework programs takes place in so-called specific programs, which describe in detail the scientific topics and the procedures for carrying them out. These programs last for 4 years. The Fourth Framework Program, started in 1995, is the most recently initiated, although some specific activities of the Third Framework Program (1990–1994) are still in operation. The latter established three focal areas for R&D: basic technologies, management of natural resources, and management of intellectual resources. These areas were in turn broken into six sections and a series of specific programs: • Information and communications technologies • Industrial and materials technologies • Environment • Life sciences and technologies • Energy • Human capital and mobility The Fourth Framework Program extended the specific programs within the existing sections and added another two sections, in transportation technologies and socioeconomic research; however, those two programs account for only 4 per- cent of the EU budget for research and technological development. Information and communication technologies clearly dominate with more than 36 percent of the budget. With a volume of European Currency (ECU) 5,700 million, or just under 5 percent of the total EU budget, the Third Framework Program is relatively mod- est compared with other EU operations. The significant increase in budget for the Fourth Framework Program, to ECU 9,432 million, is a further indicator of the growing importance of EU funds. In absolute terms, EU support for R&D is becoming increasingly important, especially considering the expected decreases in the flows of national R&D funds. The growing importance of the EU in sci- ence and technology becomes even more apparent if one looks at the substantial efforts that have been made since the late 1980s to strengthen the research and technology base, particularly of the less-developed regions of the EU, with so- called structural (regional, social, and agricultural) funds. EU support for research and technological development is awarded without regard to national proportional representation or quotas. The success rate of project applications is influenced mainly by the number and quality of applicants. In fact, the number of applications has risen substantially in the past few years, and application approval acceptance rates have dropped continuously. The in- crease in applications can be explained by a number of factors. The relatively high number of applications from British institutes of higher education, for ex- ample, is due to a severe cut in the national research budget for universities (Fig- ure 3.6). EU support primarily takes the form of contracted research with cost sharing. 

TECHNOLOGY TRANSFER IN GERMANY 265 60 50 40 Percent 30 20 10 0 B D GR ES F IRE I L NL P UK BIG SME RDI HEI FIGURE 3.6 Participation structure in the Second Framework Program, by country, 1987–1991. NOTE: BIG = large enterprise; SME = small and medium-sized enterprise; RDI = nonuniversity research institute; HEI = higher education institution. B = Belgium; D = Germany; GR = Greece; ES = Spain; F = France; IRE = Ireland; I = Italy; L = Luxembourg; NL = Netherlands; P = Portugal, UK = United Kingdom. SOURCE: The database CORDIS. The selection of projects is based on the following general criteria (see Kom- mission der Europäischen Gemeinschaften, 1990): • Precompetitive character of the proposed R&D activities • Transnationality of the project • Scientific and technical quality and originality of the project proposal • European dimension of the proposal (value added through European co- operation that could not be attained at a purely national level) • Technical and economic usefulness • Exploitation possibilities for the expected results However, the EU is currently redirecting its technology policy from pre- competitive research toward market-oriented projects (Klodt, 1995). Therefore, the precompetitive character of proposed projects is no longer a formal prerequi- site—and actually was not strictly applied in former programs. Compared with what is contributed by industry and the federal and state governments, the importance of EU financing is still minimal from the German perspective. From 1987 through 1991, a total of DM 1.3 billion (ECU 653 mil- lion) was received by German institutions from the EU. This represented only 0.4 percent of total German domestic expenditures on R&D. EU funding repre- sented about 1.8 percent of R&D expenditures by the federal government and about 5.9 percent of direct project support by the government (not including R&D expenditures by the Ministry of Defense). In some fields of research and technology, however, EU financing has gained 

266 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY 3 9 4 0 German 3 5 EU 3 0 2 5 2 5 2 1 Percent 1 8 1 9 2 0 1 6 1 3 1 3 1 4 1 5 9 9 1 0 4 5 0 ICT IMT Environ- L i f e Energy Human m e n t sciences capital and mobility FIGURE 3.7 R&D expenditures of Germany (1992–1993) and the EU by sections of the Third Framework Program. NOTE: ICT = information and communication technology; IMT = industrial and materials technologies. SOURCE: Reger and Kuhlmann (1995). considerable significance in Germany. The EU is relatively more active in infor- mation and communications technologies (ICT) than is Germany (Figure 3.7). In absolute terms, EU financing of German R&D activities in this field is equivalent to approximately one-fifth to one-quarter of what the German government spends on R&D in this area. As to the four focal areas of this report, the research and technological devel- opment activities of the EU are particularly relevant for information technology, microelectronics (included in information and communications technologies), and biotechnology. Production technology is supported under the heading of indus- trial technology. In 1991, German participants received 22 percent of available EU funding, a quite significant percentage. However, the allocation of those funds among dif- ferent sectors of the R&D systems is uneven. EU funding for predominantly industrially oriented programs goes chiefly to German industry. Even in the rela- tively science-oriented programs, contractors from industry predominate. How- ever, nonacademic German R&D institutes are significantly underfunded com- pared with similar institutions in other countries that receive EU support. German institutes of higher education are also underrepresented compared with the aver- age of all EU countries (see, for example, Figure 3.6 and Reger and Kuhlmann, 1995, p.25). The impact of EU-funded R&D varies significantly among recipient coun- tries (Figure 3.8). Whereas EU funds play a minor role in the German R&D system, in other, structurally weaker countries (especially southern countries), 

TECHNOLOGY TRANSFER IN GERMANY 267 3000 2500 2000 Number 1500 1000 500 0 B DK D ES F IRE I L NL P UK FIGURE 3.8 Participants in the Second Framework Program, by country, 1987–1991. NOTE: B = Belgium; DK = Denmark; D = Germany; ES = Spain; F = France; IRE = Ireland; I = Italy; L = Luxembourg; NL = Netherlands; P = Portugal; UK = United King- dom. SOURCE: The database CORDIS. EU funds support a significant portion of national R&D programs. Thus, if a country’s gross domestic product is taken into account, the dominance of Ger- many, France, and the United Kingdom is diminished. From Figure 3.8, of course, one can only imagine the relative importance of EU support for “weaker” coun- tries like Ireland, Portugal, and Spain.10 German R&D institutions have a number of concerns about EU R&D sup- port (see, for example, KoWi, 1992), including • that there is inadequate representation of some research fields among those that gain EU support; • that there is excessive amount of bureaucracy involved in the application procedure and the management of the project (see, for example, Schmoch et al., 1996b); • that the dominant role of the English language can be a hindrance in the running of EU projects and in the work of transnational project consortia; and • that low rates of approval for project proposals waste resources if the application fee is high. Despite these problems, German R&D institutions will likely become more interested in receiving EU funding. This is in part because they are becoming increasingly aware of the growing importance of international—and in this con- text, European—cooperation in R&D. In addition, the “years of affluence” in German national R&D support are over. Therefore, research institutions must search for other sources of support to compensate for the reductions in govern- ment funding. 

268 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY THE EUREKA INITIATIVE General Structures The EUREKA initiative was launched in 1985 as a reaction to the American Strategic Defense Initiative. It is not a program of the EU, but it has provided a framework for international collaboration among firms and research institutes in the fields of advanced civil technologies. Its aim is to • strengthen the productivity and competitiveness of European industry, • develop a common infrastructure, and • solve problems, especially environmental ones, affecting more than one country. EUREKA is not intended to harmonize European R&D policy, but rather to use available potentials for common goals. In contrast to the generally pre- competitive EU programs, EUREKA projects are market oriented. EUREKA projects are intended to complement existing programs of the EU. Members of EUREKA are the countries of the EU, the European Free Trade Area (EFTA) countries, Turkey, and the European Commission. Two keys to the EUREKA concept are its bottom-up approach for setting an R&D agenda and its flexible structure. This means that, in contrast to EU pro- grams, there are no predetermined technological areas. It is left to participating companies, universities, and other public- or private-sector research bodies to determine their particular areas of interest. In principle, there are no limitations to the type of projects undertaken. However, nine focal areas have been identi- fied: communication technology, information technology, lasers, transportation, energy technology, robotics, biotechnology, new materials, and environment (Fig- ure 3.9). Each EUREKA project is conceived and managed independently. There are no limitations to the size or scope of EUREKA projects. Although governments may play a role in setting standards and norms (e.g., in the environ- mental area), the particular R&D approach is left to the participants. A special 58.3 60 50 Percent 40 30 20 8 6.6 8.8 10 4.7 5.5 3.9 1.9 2.3 0 Enviro- Biotech- Robotics IT New Transport Commun- Energy Lasers nment nology materials ications FIGURE 3.9 Volume of research conducted in areas of technology, as a percentage of total EUREKA financing, status as of 1995. SOURCE: EUREKA (1995). 

TECHNOLOGY TRANSFER IN GERMANY 269 160 140 Without German participation 120 With German participation 107 Number 100 80 88 79 86 60 40 52 52 26 20 42 38 21 27 21 16 10 0 17 12 7 10 E n v i r o n -B i o t e c h R o b o t i c s I T - N e w T r a n s p o rE n e r g y Commu- Lasers t m e n t nology materials nication FIGURE 3.10 EUREKA projects, including those with German participation, according to technology, status as of 1995. SOURCE: EUREKA (1995). form of cooperation has arisen with so-called umbrellas. These serve as framing projects in which single projects are organized and carried out in a flexible but coordinated manner. Results are shared among participants as a way of promot- ing awareness (EUREKA, 1995). Important to EUREKA’s flexibility is its de- centralized structure. Each member nominates a National Project Coordinator to assist participants from that country. As of June 1995, 711 EUREKA projects were in progress, and 226 had been completed. The budget for current projects is DM 19.6 billion (including the contribution of participants). A total of 3,591 participating institutions were counted. Figure 3.10 shows the considerable German participation. Large com- panies are participating in about 43 percent of EUREKA projects, SMEs are in- volved in 24 percent (Figure 3.11). The budget figures for EUREKA and EU programs are not directly comparable, since they relate to different periods of Other organizations 5% Research institutes 28% Large companies 43% SMEs 24% FIGURE 3.11 Involvement of EUREKA participants by major organization type, status as of 1995. SOURCE: EUREKA (1995). 

270 TECHNOLOGY TRANSFER SYSTEMS IN THE UNITED STATES AND GERMANY time and involve different contributions by the participants. In any case, the budgets have the same order of magnitude. A prerequisite for EUREKA projects is sound financing, which may come from either national or EU sources. In most cases, if German participants apply for public support, the BMBF will allocate them funds out of its programs. All in all, the participating companies assess the EUREKA initiative very positively. Almost two-thirds considered that they have improved their interna- tional technological competitiveness, nearly 90 percent expect to produce new or improved products, and about 40 percent expect to achieve an increase in sales (EUREKA, 1993). The Impact of JESSI JESSI became a EUREKA project in 1989 and was scheduled to end in 1996. Its goal was to enhance the competitiveness of Europe in the areas of information technology and microelectronics. Financing sources for JESSI are shown in Fig- ure 3.12. More than 180 partners from 16 countries contributed to the JESSI program, providing approximately 3,100 person-years of effort annually. The estimated cost of this work was ECU 460 million in 1994 and will probably turn out to be the same in 1995. Approximately 50 percent of the work is carried out in France and Germany. All of the JESSI projects are funded on a cost-shared basis. The partners pay 50 percent and either national public authorities or the EU pays the remaining 50 percent. The total budget for 1989 through 1996 was ECU 2,560 million. National JESSI governments 37% partners 50% EU 13% FIGURE 3.12 Financing sources for JESSI, 1989–1996. SOURCE: JESSI (1995). 

TECHNOLOGY TRANSFER IN GERMANY 271 equipment/ basic and long- technology application materials term research competitive clean broadband advanced lithography HDTV manufacturing environment communication CMOS logic automotive chemicals silicon mobile radio technology safety electr. etching and digital audio packaging gases single projects: deposition broadcast JESSI common framework HDL component modeling technology assessment testing computer SMI support high-resolution displays FIGURE 3.13 Program structures of JESSI. SOURCE: JESSI (1995). The program was divided into four subprograms (see also Figure 3.13): • Technology: Development and testing of the basic flexible competitive manufacturing technology for advanced system applications, to be avail- able by the mid-1990s. • Application: Devising flexible, competitive system-design procedures and tools for the development of highly complex integrated circuits. • Equipment and materials: Development by the European supply industry of manufacturing equipment and materials for selected areas of micro- electronics. • Basic and long-term research: Basic and complementary applied research with the long-term perspective. JESSI was successful in establishing a pan-European platform for collabora- tive research. Europe leads the field of digital audio broadcast thanks to relevant activities in the application subprogram. The digital audio broadcast project pro- vided all the necessary components for planned field tests. Transmissions are in progress in 21 areas (10 more are planned), and 8 million people can already receive digital audio broadcasts. Significant results have been achieved in the important JESSI subprogram of technology. Technological competence has been attained in the area of micro- electronics. For example, a close cooperation between all major European inte- grated circuit companies has been established. Outcomes of this collaborative research include: • the 0.5-micron CMOS technology of Crolles (jointly developed by SGS- Thomson, CNET, and PHILIPS), which is being transferred to the new PHILIPS Waferfab. Siemens and ES2 have signed an agreement allowing ES2 to produce chips with the Siemens 0.5-micron CMOS process.