CONTACT: Stanford University News Service (650) 723-2558
University, government role in technology transfer analyzed
STANFORD -- Speeding up the transfer of technology from the university to industry became a widespread goal of governments in the 1980s, even though the processes of transfer are not yet well understood, researchers who study or participate in technology transfer said at a conference at Stanford University March 18-20.
Sponsored jointly by the Stanford Center for Economic Policy Research and the American Academy of Arts and Sciences, the conference drew about 100 participants from industrialized countries, mostly economists, sociologists and science historians who do case studies of technology transfer and model how it has worked in various industries and geographical settings. Some laboratory scientists and representatives of companies that use university research also were involved in the discussions.
Key issues of debate included whether existing laws, policies and funding programs work to maximize transfer of technology to the private sector and whether they jeopardize universities' training and open- science missions.
Current debates in Washington over whether universities should be the ones to patent technology invented with federal research grants and whether they should be permitted to work with foreign-based companies on government-funded projects also were discussed.
Surveys of industrial research-and-development managers and university personnel were presented that indicate that the research university has played an important role in development of at least a portion of new products and processes but that industry also relies most on the university for training a skilled workforce.
In addition, researchers presented surveys of university faculty and technology licensing personnel indicating that universities have concerns that their academic values do not always mesh well with those of industry.
Growth of transfer programs
In the 1980s, 43 states set up programs to transfer new technology from the university to industry, and 286 university-industry research centers were formed in the United States, more than had been established through the preceding century.
The trend was fueled by declining U.S. economic productivity, especially the decline of the Rust Belt, conference participants said. The Bush and Clinton administrations later became involved, expanding opportunities for private firms to apply for federal research funds.
Meanwhile, the government's own labs have increasingly seen their post-Cold War mission as doing publicly subsidized contract research for industry, a goal that makes them a competitor with universities for government as well as industry research dollars. Some participants questioned the wisdom of using government labs, because countries that have done so, such as France, are not known for dispersing technology widely but for concentrating science in the government sector and a few large companies.
The European Community also has developed programs to integrate university and industrial research in recent years, conference participants said, and even in Japan, where universities previously have played a minor role in scientific research, efforts are being made to integrate them into the process.
Patenting and licensing regulations in the United States were a major topic of the conference, partly because of recent proposals to limit universities' property rights to technology developed with government research grants. Some conference participants also suggested countries need to overhaul their intellectual property systems in ways that discourage secrecy concerning new knowledge.
The Bayh-Dole Act that amended U.S. patenting and trademark law in 1980 gave universities and other nonprofit institutions the right to retain the property rights to inventions derived from federally funded research, and 1984 amendments further expanded university property rights. The changes were meant to encourage private firms to develop further ideas for commercial use by making it possible for them to acquire rights from universities.
But critics say university patenting and licensing of the technology to firms creates conflicts of interest and commitment for faculty, threatens the openness of research and may shift universities to more applied research priorities. In addition, there are concerns that universities may not be able to reliably produce economic growth in the short run.
"The universities have been cast in a very utilitarian role" in recent years, said Donald Kennedy, president emeritus of Stanford and a professor of biological sciences. "It's widely believed by universities and their regents that if only they did the right thing, they could be the next Silicon Valley. I think that's heavily overstated, and there's going to be deep disappointment in universities when it doesn't happen."
Universities must be careful to guard their effectiveness as "nodes in the international transmission and reception of knowledge," said the organizers of the conference, Stanford economists Paul David and Ed Steinmueller. Through diffusion of technical information, universities promote more growth in the total stock of knowledge, they said.
Universities should examine each new initiative with industry, they suggested, with an eye to whether it "enhances or detracts from the university's ability to function effectively in the information network that characterizes open science."
Other conference participants, however, mindful of the Japanese government's recent success in spreading technical knowledge, as well as of earlier U.S. success in modernizing agriculture, urged governments to develop industrial policies that more directly involve universities in national or regional industrial problem solving.
"The U.S. university system has never been an ivory tower. It has worked for many masters," said Richard Nelson of Columbia University. "Prior to World War II, its major funding was for agriculture and local industry," similar to the state technology transfer programs that have grown up in the last decade.
David and several other conference participants objected to suggestions in Washington that universities be steered to a national economic agenda that, he said, treats the transfer of knowledge across borders as if it were a zero-sum game.
"For universities to get involved in that is particularly dubious," David said, because they are "well suited by their history to promote access to a knowledge base."
Foreign involvement in technology transfer has become an issue since the Massachusetts Institute of Technology was the subject of a congressional inquiry into its licensing of research to foreign firms, speakers said. A subsequent review by the General Accounting Office of university licensing practices found that universities proactively license technology developed with federal research grants, and that less than 10 percent has been licensed to overseas companies, reported Jon Sandelin of Stanford's Office of Technology Licensing.
Congressional criticism of foreign involvement may have been partly brought about by universities themselves, said Proctor Reid, who is directing a study of such involvements for the National Academy of Engineering. In arguing for laws that gave them the right to patent and license technology developed with federal grants, Reid said, the universities stressed their ability to contribute to national economic development.
"We have laws on the books that commit recipients of research funds to do their best to make sure the benefits go to the nation as much as possible," he said, "but these laws have not been enforced. When you ask university presidents how can the American public tell if the law is being enforced, they tell you there are no figures kept."
Others expressed concern that technology transfer will be slowed if universities lose their current rights to file for patents and collect royalties from the companies they license to use the technology.
Large companies in many industries would prefer that universities did not file patents, Sandelin said. But "patents are vital to small business and start-up companies as a means of raising funds for new product development and marketing. This is the source of job creation in the United States and may spawn new industries such as the biotech industry," he said.
One proposal in Congress would require that all federally funded inventions be put out to competitive bidding with a portion of royalties to be returned to the government, Sandelin said. The bill would severely hamper "incremental advances in science with specialized application to very small niche markets," he said.
"Industry is increasingly seen [and supported] as the source of commercially important technological innovations," said Irwin Feller of Pennsylvania State University, "with universities assuming [or being relegated to] roles as subcontractors."
University patenting has increased 15-fold since 1965, with a particular explosion in activity after 1984 at the major research universities, according to a study by Rebecca Henderson of the Massachusetts Institute of Technology, Adam Jaffe of Harvard and Manuel Trajtenberg of Stanford and Tel Aviv University. The 20 most prolific patenting institutions - including Stanford - received 73 percent of university patents in 1991, but about 150 universities are now filing for patents annually, and membership in the professional association for university technology managers has grown from 100 to 1,100.
Stanford's own licensing office has grown from six people in 1980 to 22 in 1994, Sandelin said. MIT's licensing office has initiated a formal program to foster new company formation, while in the United Kingdom, Oxford University has gone even further, setting up its own company to develop instrumentation needed to pursue some of its biotechnology research.
In the United Kingdom and in the United States, many universities have tried to follow in Stanford's footsteps by starting "research parks" on land they own. Universities also have changed their policies over time, with some allowing exclusive licenses to faculty members' companies in cases where they have felt that was necessary to successfully commercialize the technology.
(The faculty Committee on Research at Stanford recently proposed some changes in Stanford policy that would require faculty to disclose their outside consulting activities and that would vest the university with ownership of all technologies created or invented by faculty using university resources. Proposed changes will be discussed in the Faculty Senate. See story, page 1)
Ninety-eight U.S. universities received royalties of $173 million on 2,632 licenses with 1,387 new licenses signed in 1992, according to a survey published last year by the Association of University Technology Managers.
The numbers of university patents may be up but the quality may be down, Henderson said. "On average, university patents are both more important and more general than the average patent," she reported, "but this differences has been declining over time."
This decline could indicate that universities are moving toward more applied research, she said, although it is too early to tell because the more recent patents have had less time to make an impact.
Three of four university administrators involved in technology licensing reported that their university viewed patents as a "sought-after revenue-generating source," in a 1993 survey, by Dianne Rahm of the University of South Florida, of university technology managers and university researchers in the top 100 U.S. research universities. About half of researchers involved in industrial collaborations, however, maintained that their university had no clear policy on patents.
Negotiating intellectual property rights is "troublesome" to both researchers involved and university technology managers, Rahm said, "but it is not clear that managers and researchers are communicating with each other regarding the source of the strife."
More faculty researchers see the emphasis on industrial outreach as having a negative effect on the basic research mission of the university than do university technology administrators, Rahm reported.
Patenting and licensing may not always be the best way for universities to transfer technology, suggested David Hodges, dean of the School of Engineering at the University of California-Berkeley. With open disclosure of research on integrated circuit design tools, Hodges said, his school recruited a very active, generous group of industrial affiliates whose donations compared favorably to the royalty revenues they might have received by licensing the technology.
Attitudes toward patenting have changed greatly in universities since the early 1980s, said Robert Rosenzweig, former president of the Association of American Universities and a former vice president of public affairs at Stanford.
Stanford almost did not patent recombinant DNA with the University of California-San Francisco, he said, because "there was significant opposition on the part of senior members of the genetics faculty."
"It was my feeling that they had some good points," he said, such as that financial involvement would rule them out of the public policy debate about uses of the technology. Since then, he said, most of those who objected have formed their own companies and a similar debate now would be "almost inconceivable."
Some economists suggested that governments restrict the exercise of intellectual property rights in ways that encourage those involved in research and marketing to enter into more cooperative development. In biotechnology already, there is a system of "paid free access" that rewards the inventor but also maintains free circulation of genetic resources possible for research. If someone makes dependent innovations, he or she is required to get a license from the original inventor and pay royalties.
"From a social point of view, the innovation process will be accelerated if competitors are able to support, rather than ignore or block, each other," said Dominque Foray of Ecole Centrale in Paris.
New federal research programs seem, however, to be moving in the opposite direction, speakers said. The Advanced Technology Program (ATP) of the National Institute of Standards and Technology assigns ownership rights to the for-profit partner of a university, according to the ruling of a government attorney, Sandelin said.
The Technology Reinvestment Program managed by the Advanced Research Projects Agency requires universities to negotiate with their industrial partners for ownership, and company lawyers have indicated they will not permit universities to apply for patents, he said. "This removes any motivation for the university scientist to disclose inventions, and some universities, such as MIT, may not accept ATP awards."
Other findings reported
A wide variety of recent research findings were reported or discussed at the Stanford University conference on "university goals, institutional mechanisms and the industrial transferability of research." These included:
Technology transfer occurs, one biotech industry participant said, "in airplane seats and at lunch."
"The way this coupling occurs for research universities is highly dependent upon the distribution of Ph.D.s in management" of the relevant firms, said Elliott Levinthal, professor emeritus of mechanical engineering at Stanford and an early science manager at Varian Associates. "We have lots of trouble relating to the auto industry and very little with the computer and pharmaceutical industries because [the latter two ] speak our language."
Companies "have to recognize it's important to take steps to create an overlap" with universities, said Martin Packard, one of Varian's founders.
Studies show that scientific transfer "isn't just getting access to the literature but getting access to knowledge in an assimilable and usable way," said Stuart Blume of the University of Amsterdam.
"You don't know where the new drugs are going to come from, so your science staff has to do science and a lot of boundary watching. The best way is person to person [communication]," said Dru Willey, vice president of university licensing and relations at the pharmaceutical company Amgen.
In addition these executives said that about 8 percent of their firms' new products and about 6 percent of their new processes were less expensive or less time consuming to introduce because of academic research. Academic research affected new products and processes most in the drug industry and least in the oil industry, he said.
"Academic research provided new theoretical and empirical findings and new types of instrumentation that were essential for the development of new product or process, but not the specific invention itself," Mansfield said.
In a 1993 survey of 1,134 university technology managers and university researchers in the top 100 U.S. research universities, Rahm found four of 10 technology managers said that firms they had dealt with had placed restrictions on university researchers sharing information with their departments or research center colleagues. Nearly five in 10 of the researchers mentioned this restriction and said it had "created a feeling of conflict for them between loyalty to the firm and the university's value of open knowledge," Rahm said.
In addition, eight of 10 managers and four of 10 researchers said firms they had dealt with sought to prohibit or delay researchers from publishing results coming from university-firm interactions.
"My sense is that this is one of the real stresses on these centers. . . . There is some evidence that the pressure to get faculty involved in short-term projects is not working," Feller said.
This is an archived release.
This release is not available in any other form.
Images mentioned in this release are not available online.
© Stanford University. All Rights Reserved. Stanford, CA 94305. (650) 723-2300.