Is Industry-Academia-Government Collaboration a Panacea?*1

The Japanese economy remains gloomy. In February this year, The Economist featured an article headlined "JAPAiN," profiling Japan's continuing economic slump.

What needs to be done to pull the Japanese economy out of the doldrums and put it back on the path to growth? Innovation is essential for economic growth. Robert M. Solow, the Nobel Prize laureate in economics, says that economic growth is mostly attributable to "technical change," his term for what is today popularly called "process innovation," that is, increasing productivity by improving the production process. To stimulate new demand and realize domestic consumption-led growth, product innovation, which is for developing completely new or significantly improved products with unprecedented quality, also plays a very important role.

There are several ways to encourage such innovation. One way is to promote industry-academia-government collaboration in which private-sector companies conduct research jointly with universities and public research institutes. According to a study conducted by Edwin Mansfield, approximately 10% of industrial innovations would have never happened or, even if they did, would have been significantly delayed were it not for academic research. Conversely, this means that research findings by universities and/or public research institutes have been vital to at least 10% of these innovations.

Innovation is now understood as important for overcoming Japan's economic stalemate, and industry-academia-government collaboration is available as a means to promote it. But will such collaboration provide a quick cure for industrial challenges regardless of their field? Indeed, with increasingly diversified and sophisticated technologies incorporated into single products, it is becoming difficult for companies to on their own develop all the technology and scientific knowledge required for innovations. On the other hand, however, collaboration among multiple organizations involves coordination costs. So, in what fields would the benefits of industry-academia-government collaboration exceed the costs? To answer these questions, I examined patents resulting from industry-academia-government collaboration.

For this research, the Institute of Intellectual Property patent database (IIP patent DB), developed by Akira Goto (The University of Tokyo) et al. was used. About 8 million patent applications published during 1972-2002 were classified into three groups based on their type of right-holding "applicant" - patents applied for by a private-sector company, by a university, and by a public research institute. Patents jointly applied for by a private-sector company and a university or public research institute are defined as industry-academia-government collaborative patents. In reality, there are also cases where university researchers, who have filed for or obtained patents for their inventions, subsequently transfer their patent rights to a private-sector company for commercialization of the technology. However, for the purpose of this research, such cases have been excluded. Thus, the number of industry-academia-government collaborative patens used in this research may be somewhat smaller than the actual number of such patents if they fall under their customary definition.

The number of patent applications filed with the Japan Patent Office (JPO) has been increasing. The total yearly number of applications rose from some 150,000 in the 1970s to about 350,000 in the late-1980s. The pace of increase slowed in the 1990s with the yearly number of applications remaining in the 350,000-400,000 range.

Industry-academia-government collaborative patent applications published during 1972-2002 totaled 6,988, accounting for only 0.8% of the roughly 8.76 million published during the period. Inventions made by industry-academia-government collaboration thus have a very minor presence.

However, the yearly number of industry-academia-government collaborative patents has been increasing at a very rapid pace. The number, which stood at less than 50 per year in the late-1970s, exceeded 200 in the late-1980s, reached some 300 in 1995, and jumped to around 800 in 2000. The enactment of the Act on the Promotion of Technology Transfer from Universities to Private Industry (Act No. 52 of 1998) can be cited as one reason behind this trend.

Viewing industry-academia-government collaborative patents from a university standpoint, as of around 1972 most university patents were filed for independently by the universities themselves. In contrast, nearly half such patents are today filed for jointly with other entities. This indicates that universities are increasingly collaborating with industry and other organizations.

In what areas of technology is collaboration among industry, academia, and government particularly active? The table below shows collaborative inventions broken out by "subclass" level (subclass 798) based on the International Patent Classification (IPC) and arranged in descending order by the number of patent applications. The "Rank among all" column shows the pro forma rank of each technology area among all patent applications, not only those applied for under industry-academia-government collaboration, classified in accordance with the IPC. By comparing the two rankings, we can see that the trends of inventions under industry-academia-government collaboration differ substantially from those of inventions in general. For instance, "C12N: genetic engineering...," "B01J: chemical or physical processes...," "C02F: treatment of water...," and "E02D: foundations..." - none of which are within the top 50 in terms of their rank among all - are placed among these 10 in terms of rank among industry-academia-government collaborative patents. On the other hand, among the top five technology subclasses in terms of rank among all, "G06F: electric digital data processing," "G11B: information storage based on relative movement between record carrier and transducer," "H04N: pictorial communication," and "G03G: electrography; electrophotography; magnetography" are not within the top 10 in terms of rank among collaborative patents. To examine the degree of influence of industry and academic institutions (including public research institutes) in their collaboration, industry-academia-government collaborative inventions were further classified, based on their characteristics, into 1) private-sector company inventions; 2) university inventions; and 3) public research institute inventions, with the numbers of patent applications by area of technology clarified for each type of invention, and the product-moment correlation coefficient (PMCC) with the corresponding numbers for all industry-academia-government collaborative inventions was calculated. The coefficient value was found to be higher for 2) and 3) than for 1), that is, the characteristics of industry-academia-government inventions in terms of their technological attributes are more in line with, or affected by, those of academia than those of industry. Within academic institutions, the influence of public research institutions was also shown to be greater than that of universities.

This study showed that the number of inventions resulting from industry-academia-government collaboration increased notably from 1996. Areas of technology subject to this collaboration were also found to be biased compared to the areas where the concentration of overall patent activities is observed; collaboration activities are disproportionately concentrated in areas where public research institutes and universities have strengths. This implies that private-sector companies should seek strategic collaboration with academic institutions by clearly defining the areas where universities and public research institutes have strengths, rather than haphazardly seeking collaboration in areas of their own weakness.

This study provided some suggestions about industry-academia-government collaboration by using patent applications jointly filed by private-sector companies and academic institutions. However, there are various other means and forms of industry-academia-government collaboration. For instance, universities can serve as a vehicle for human resource development, release their findings in the form of research papers, accept researchers from private-sector companies, and transfer their rights to research findings at an early stage when they are merely rights to obtain patents. Research activities can benefit in the future by capitalizing on diverse means and forms of industry-academia-government collaboration.

>> Original text in Japanese