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.
6 "Hybrid" Organizations One of the distinctive features of Japan's R&D system is the growth of "hybrid" organizations that enjoy support from the government but involve con- siderable initiative and operational direction from private sector companies. While observers may disagree as to whether particular organizations have achieved breakthroughs in technology, these evolving organizational experiments do repre- sent a variety of mechanisms for government-industry cooperation that have no direct parallel in the United States. MITT-SPONSORED JOINT RESEARCH LABORATORIES During the 1970s and 1980s Japan established a number of collaborative research projects with some form of government support or sponsorship. All of them bring together researchers from national and corporate laboratories in efforts to develop technology, in most cases with MITI support. These projects are established to last for finite periods of time. In recent years the direct influence of the government has diminished as organizational mechanisms have been devel- oped that allow the private corporations involved to take on larger roles.S3 The first and best known project that involved a collaborative laboratory was the Very Large Scale Integrated Circuit (VLSI) project, which ran from 1976 until 1980. The goal of the project was to develop fine pattern lithography using electron beams. It is important to recall that at the time of its establishment, there were conflicting views about the idea of establishing a collaborative laboratory. The concept has been attributed to the Industrial Electronics Division of IvIITI and leaders in ETL. 30
31 The government funded 60 percent of the budget on the condition that the companies would repay the government when they turned profits on the technolo- gies developed. Five participating companies formed an engineering research association. Most of the research was conducted in the separate laboratories of the member companies, although a joint laboratory was established. As the first experiment of its type, the VLSI project encountered a number of difficulties, most of which centered around the need to build trust among competing firms. Close observers also believe that the near-term nature of the research target, defined by a perceived need to "catch up" with IBM, made it difficult to attract the highest quality researchers from the participating firms. A second generation project, the Optoelectronics Joint Research Laboratory (OIL, 1981-1986), had a more narrow but also more futuristic target: developing multi-dimensional super-lattice technologies with atomic scale control and evalu- ation. The project involved developing optoelectronic integrated circuits (OEICs) based on existing technology, as well as basic research carried on in the OJL to lay the groundwork for more advanced OEICs. With a budget of 550 million, the OJL was organized in six research croups with researchers from the nine partici -- r~ D~^-~ A --r . . patting companies. A third generation project, the Optoelectronics Technology Research Labora- tory (OTL), was established in 1986 and continues today. Unlike the other two projects, this one was established by the Japan Key Technology Center.54 In many ways this project is a direct extension of the OJL: the same companies that participated in OJL have become stockholders (along with a few additional companies) and the R&D focus is next generation OEIC technology.55 An important feature of all three projects is that the work of the joint laboratory is in basic research, while the participating firms independently pursue more applied research, competing directly in device development. Attracting good researchers has been an issue in each case. In the case of the OJL, six research groups were established, but the fact that the researchers worked in common laboratory space encouraged spontaneous collaboration. Interestingly, research- ers in both the OJL and OTL were not at the outset specialists. As time passed, the work of research required cross fertilization of research groups. Observers believe that the neutrality of the joint lab and the focus on more basic research made possible frank discussions uncommon among researchers from competing companies. In the OTL, unlike its predecessors, researchers from different member companies are thoroughly integrated into the research groups and re- searchers themselves have more freedom in selecting subjects. Even those directly involved in the joint laboratories argue that there is no easy way to evaluate "effectiveness."56 Joint research has in these projects become more effective and deeper over time, particularly as the focus has shifted to longer-term goals. According to participants, the member companies are inter- ested and supportive of projects precisely because they are risky and futuristic- not the kind of work they are anxious to or can effectively take on independently.
32 Projects with futuristic goals can, however, be narrowly targeted to build research intensity (a comparatively large group of top quality researchers working on a well defined area). It is also important to note that as participating companies have built mutual trust, government direction has declined. At the very least, these joint projects are interesting experiments in organizational design. Consid- ering the evolution noted above, it appears that learning has occurred not only in developing technologies, but also in organizational design. Another example of a joint effort is MITI's Fifth Generation Computer Proj- ect. Planning for the project began in 1979 and in 1982 the Institute for New Generation Computer Technology (ICOT) was formally established and a ten- year development plan initiated. The purpose of the project is to develop the technologies needed for a new type of computer. There are more than 100 researchers now working at ICOT from member organizations that include na- tional labs, NTI, KDD (Kokusai Denshin Denwa), and corporations such as Fujitsu. With a $40 million budget in 198S, ICOT researchers are working on basic software development as well as parallel processing and relational data- bases. The goal of the final stage of the project, which has just begun, is to develop a prototype system. Leaders of the project, however, describe it as one involving primarily basic and "pre-pre-competitive" research.S7 It is probably too soon to draw conclusions about the success, but it should be noted that foreign researchers have joined the project and fellowships are available to support work there. "HYBRID" ORGANIZATIONS IN THE UNITED STATES While there are no direct parallels to Japan's government-sponsored joint laboratories discussed above, hybrid organizations have been developed in the United States. Whereas in Japan the locus of activity is private sector corpora- tions, in the United States universities form the core. Eighteen Engineering Research Centers (ERCs) have been established by the NSF to change the way that engineering research is done at U.S. universities. Key features of the ERCs are their cross-disciplinary research and education with a systems orientation, technology transfer, and participation by industry. The ERCs focus on a range of technology areas that were not predetermined, but arose in the process of proposal review. Seven work on manufacturing and design, for example, three in microelectronics and optoelectronics, three in mate- rials processing, two in biotechnology, and three in other areas related to energy and the environment. Due to limited funding (less than $40 million in 1989) and the large number of proposals, only a small fraction of proposals have been successful in receiving awards. The ERCs are funded on an incremental basis (in five year periods with interim third-year reviews), but a new proposal must be submitted after 11 years have passed. The ERC is a partnership where the U.S. government (NSF) provides funding
33 (half of the total), coordination, and oversight. The university provides the research and educational base, facilities, technology transfer mechanisms, and management. Industry contributes funding (about 30 percent of the total), along with advice and specialized expertise. Based on a subscription rate schedule, smaller companies pay less to participate. Participation by foreign companies in the ERCs is growing. The universities determine participation, but the general quid pro quo principle must be followed and the companies must open their doors to ERC researchers and students, as in the case of U.S. companies. Examples of ERC impacts can be cited. Research at Lehigh University has enabled a member firm to become the first U.S. competitor with Japanese and German firms in producing welded wheel trucks for railroad cars. Carnegie Mellon University has developed a design system with General Motors that is being implemented in its Fisher Body Division. The ERCs also have had aca- demic impacts in training graduate students in fields like optoelectronics engi- neering, in established cross department degrees, and new graduate curricula. Interest in collaborative research has grown in recent years in the United States and there are a number of state and regional initiatives that involve government support. Sematech, a consortium that has recently been established with U.S. government funding, will involve private sector firms working on semiconductor manufacturing R&D. A history of concern about possible collusion among competing firms working together has acted to limit collaborative activities be- tween U.S. government and industry. U.S. universities have instead been the locus of joint action here. Recent changes clarifying exemptions from antitrust law for joint R&D efforts and new thinking about the potential gains from joint approaches may well lead to new experimentation in the near future.
