Examples of ''Effective Practice" in Research Support and Performance
Summary Points
__ The presentations illustrate that defining effective research performance differs at the national, corporate, and university levels. For companies, such as Amgen, Inc., research must ultimately contribute to sales and profits. For universities, building the base of fundamental knowledge and training the next generation of researchers are the key tasks. Knowledge building and training are important at the national level as well, but unless these assets are seen as contributing to broad national goals the public view of performance may be skeptical.
__ Research organizations in several countries report increased pressure to evaluate results and develop appropriate evaluation metrics and mechanisms. Companies, universities, and international research funding programs have developed a number of approaches to this task.
__ Several presentations highlighted that effective recruitment, training, and management of outstanding scientists and engineers are essential to superior research, whether in the industry, university, or government sectors.
__ Research organizations in several countries report an increasing need to maintain a global view in order to improve performance by tapping ideas and talent from around the world. Companies are already developing effective techniques for building and managing a global R&D effort.
Daniel Vapnek, Amgen, Inc., United States
General Elements
Several elements of "best practice" or "effective practice" in research support and performance appear to hold true in all situations and countries. First, it is important that the goals of research and the goals of the organization be clearly articulated, and that the goals of research are set so that meeting them clearly advances larger organizational goals. This is true in companies and universities, as well as at the national level. For example, the other presentations have highlighted how countries such as Korea and Finland have succeeded by building technological capability in focused areas where they could be competitive in global markets. A second common element is to support superior science. This calls for employing excellent, well-trained researchers. Third, it is necessary to provide adequate support to research efforts. Finally, it is necessary for research organizations to maintain a global perspective without national boundaries, in order to scan the globe
for superior ideas and people. This is becoming more important as more countries develop the critical mass of science and technology capability necessary to do outstanding research.
Factors Specific to Organizations
In addition to these general features of "effective practice," other elements of superior performance depend on the type of organization. Figure 3-1 shows the roles and interaction of universities and companies, a critical interface.
In universities, the key task is to push back the frontiers of fundamental knowledge and train the next generation of scientific and engineering talent. Publication is a critical factor in communicating and evaluating university research, particularly
the quality of publications as opposed to the quantity. Adequate support for researchers through a competitive, peer review process has been critical to the outstanding quality of health care and life sciences research at universities here in the United States.
At companies, in addition to general factors such as adequate support and clearly articulated goals, it is important to maintain a balance between focus on internal developments and opportunities to access knowledge from outside. Avoiding a "not invented here" mentality becomes more challenging as the organization grows larger. A global perspective is also important. In the biotechnology industry, we have seen a number of examples where fundamental discoveries were made in one country but were picked up and commercialized by companies in other countries.
Communication is also important in the corporate context. It is critical that top management understand the technical issues, understand the unpredictability of research, and be flexible to change as science changes. This environment and top management focus is easier to maintain in a smaller company.
We are just beginning to explore larger national and international-scale efforts in the life sciences. One example is the Human Genome Project, an effort to sequence the human genome supported by the National Institutes of Health and the Department of Energy in the United States. This program has benefited a great deal from international exchange and communication. The Wellcome Foundation and other private foundations around the world play an important role in this global effort. This has been an excellent example of international collaboration. A second example is the Human Frontier Science Program, covered next by Dr. Cuenod.
Michel Cuénod, Human Frontier Science Program (HFSP), France
HFSP is focused on promoting inter-continental cooperation in basic research in biology and neuroscience. By focusing on support for younger scientists through research grants, fellowships, and workshops, it furthers the internationalization of interdisciplinary research of very high quality. The concept was first proposed at the 1987 G-7 summit by Prime Minister Nakasone of Japan, and the program was launched in 1989 with a secretariat established in Strasbourg. In order to ensure a timely start, Japan agreed to provide the bulk of the support for the program in the initial years, with the remaining support provided by the other management supporting partners (Canada, France, Germany, Italy, Switzerland, United Kingdom, United States, and the European Union).
Basic support is provided in two fields, brain function and biological function at the molecular level. These are fields that require strong support for basic research. The problems are of high complexity. International cooperation appears to make tactical sense. Even the largest countries should profit.
A recent review by an independent panel found that the quality of HFSP awardees and proposed research were of uniformly high quality, and addressed
important questions. The quality is ensured by a peer review process modeled on the study section system of NIH, with the review panels made up of two member scientists from each participating country. From the point of view of the scientific community, the absence of political bias and the focus on high quality rather than targeted subjects are particularly attractive features of the program.
Research grants are given to groups of four members on average from at least two different countries or if possible, two different continents. The program is very competitive; only about 12 percent of the proposals are funded, while recent review committees believe that about 30 percent are worthy. Grants run for three years, and practically no renewals are given. This is a short time for a project to show results, particularly for an international team, but budget constraints dictate this. The average grant amount was $220,000 in 1997 for a team over three years. The average age of 1997 grantees was 42.6. Figure 3-2 shows the regional distribution of grants, based on results from 1990–1995. The added value includes the exposure of young scientists to different ways of thinking about science culture, and encouragement of national science agencies to think globally. This relatively small, unique program has demonstrated the efficacy and desirability of promoting international collaboration.
In 1997, 160 long-term fellowships were awarded. The success rate is about 20 percent. They run for two years at $40,000 per year. Fellows must either be from, or travel to, a member country. Most of the awardees go to the United States. This leads to most of the funding for HFSP going to the United States. The United States receives about five times more in grants and fellowships than it contributes to the program.
The total budget in 1997 is $47 million, with about 60 percent going to grants, about 30 percent going to fellowships, and a very small amount to workshops. About 7 percent goes to administration, a very low amount for an international program.
Restrictions or delays in publication are not allowed in HFSP research. Intellectual property rights have not been a problem. The center where particular research was performed is responsible for IPR.
Although the participating countries agreed with Japan's goal that other member countries should contribute about half the funding over time, this has not been achieved. Japan still funds about 80 percent of the program. Japanese decision makers are showing impatience with these trends, and achieving more balanced funding is the most challenging long-term issue for HFSP. There are several reasons for the difficulties in this area. There is budget pressure in most of the member countries, which tends to put a higher priority on national programs. HFSP supporters make the argument that promotion of international cooperation is cost effective and delivers significant leverage, but not all accept it. Also, officials in some member governments will argue that Japan's disproportional share of support for HFSP represents a small and long overdue counterbalance to the benefits the country has received over the years from basic research funded elsewhere.
Ewa Gajewska-Blaisdell, Blaisdell & Co., Poland
The end of the Cold War and break-up of the Soviet Union are resulting in new opportunities and choices for Poland and other countries that had been under Soviet domination. The science and technology research structure in Poland is mostly leftover from the Communist era. We face many challenges in building up our capabilities. Poland needs to look to other countries, both at the official level and the corporate level, in order to develop a science and technology infrastructure that can fuel economic development. Promoting Poland as a desirable location for high-technology international companies is one aspect of this task. Education of Polish political leaders about the importance of science and technology is another. Poland can benefit a great deal from increased exposure to effective practices in science and technology from around the world.
During the Cold War, research and development was dominated by large state-owned heavy industry enterprises, which have either disappeared or are facing severe pressure. The ministry responsible for R&D activities is responsible for supporting about 200 remaining government units. Recently, the ministry retained a Western consulting company to evaluate these facilities, identify best practices, the best institutes, and the ones that should be closed. Of the 200, eight were found to have promise, but the rest need to rapidly improve in order to stay open. R&D institutes have cut staffing by 30 percent in the last three years, and funding has also been cut. Government funding provides about a third of the funding for public institutes, with the rest coming from private sources for particular projects. Despite the challenges, Poland is one of the best places for developing emerging technologies. As part of activities related to joining the European Union, Poland
is setting strategic priorities for research and industrial development. The first priority is to strengthen capabilities in enabling industrial technologies, such as materials. The second is to focus on infrastructure related work. For example, Poland is in an important location in Europe, lying between Western Europe and Russia, so that advanced highway technologies and other transportation related fields would be appropriate. A final focus is on research relevant to state functioning.
Terrence Heng, Motorola, Inc., United States
The Role of R&D in Motorola
Motorola considers R&D to be its lifeblood. Annual sales total about $30 billion, and last year the company invested about $2.4 billion in R&D. This is a considerable investment, and it is expected to deliver significant returns. The R&D organization consists of corporate R&D, about ten percent of the total, and sector or business unit R&D. Corporate R&D focuses on research with a time horizon of more than five years, while business unit research focuses on what will be needed in products in the immediate future.
There are several primary reasons for making such a large R&D investment. One goal is to enhance the company's portfolio of technology platforms. Motorola lives on new product cores and new industries. Motorola created the walkie-talkie industry, the car radio industry, the mobile phone industry, and the paging industry. The company is in the process of creating the satellite telephone industry, with the Iridium joint venture.
Innovation is necessary to maintain a high rate of growth, and to avoid becoming an "also ran" company. Competition from around the world is formidable. A few years ago Fujitsu was the benchmark for a follower strategy, in that when Motorola introduced a new product Fujitsu would have something just as good within nine months. The top Korean firms are now able to do this within six months. In pursuing a business strategy based on innovation, intellectual property protection is critical. Motorola has been able to enter markets and stop entry of other companies thanks to its patent position.
Another task for the R&D organization is to eliminate technology surprises. CDMA, the emerging digital wireless technology, is a good example. Several years ago Motorola was on the leading edge with a few other companies, but now there are about 60 companies around the world working on CDMA. Motorola does constant competitor analysis. It is necessary not only to know what competitors are doing today, but to project what they are likely to be doing in five or ten years. Often new competition emerges unexpectedly from unlikely sources due to paradigm shifts.
Globalization of R&D
A final task for R&D is to play a leading role in the globalization of the company. Currently 62 percent of Motorola's sales are outside the United States. The proportion of U.S. sales is expected to fall to 25 percent by 2000. This trend
toward international sales has been apparent for some time, but R&D had not been keeping pace until recently. Five years ago almost 95 percent of Motorola's R&D personnel worked in the United States. Today there are 20 Motorola R&D organizations outside the United States, with about half of corporate R&D being performed abroad.
Our top management has mandated this globalization effort. Motorola has been reorganized into three regions—Pan American, Asia, and Europe/Middle East/Africa. While North America is not growing that quickly, sales in Latin America are growing rapidly. In Asia, growth is especially rapid in China. Five years ago Motorola had virtually no sales and no employees in China, but in 1997 the company expects $4 billion in sales and will employ 10,000 people. In addition to the direct benefits of participating in high growth markets, Motorola's global approach helps it stay attuned to the strategies of potential competitors and prevent them from maintaining domestic profit sanctuaries. This lesson was learned from Japan. By competing head-to-head in Japan, Motorola has been able to prevent Japanese companies from using a protected home base to finance market entry in the United States and elsewhere. Today, Japanese firms are not major global players in the cellular phone business or the paging business.
Governments are also encouraging Motorola and other international companies to do R&D in their countries as part of doing business. This encouragement can be implicit or explicit. Although explicit pressure may go against World Trade Organization rules, it is a fact of life in a number of countries. Governments also provide positive incentives. In Singapore, government programs support a significant fraction of Motorola's R&D costs.
Another reason for globalizing R&D is to gain access to world class talent. Motorola prides itself on hiring best-of-class people. However, competition for top scientists and engineers is tight in the United States. Motorola recruits foreign nationals studying at U.S. universities, and sends them back to their home countries. The R&D centers in India, China and elsewhere also hire the best within the country. In India, Motorola hires one of 27 who are interviewed, in China, one of 40, and in the United States, one of two. Foreign engineers can develop unique solutions to major research issues. Russian engineers, for example, often look at things in ways unique to their backgrounds.
Motorola's foreign R&D centers are organized in a number of different ways. For the most part they are wholly owned by Motorola, in order to ensure protection of R&D. Several initiatives take the form of "virtual organizations," and involve collaboration with universities and institutes. One example is a joint venture with the Chinese Academy of Sciences.
Formulating Strategy and Measuring Success
Part of how Motorola chooses its R&D programs comes from expected return on investment. Each business division has to develop a 10-year technology roadmap. Iridium, for example, was conceived 10 years ago. It is often said that technology moves quickly, but a major new platform requires 8 to 10 years of development.
R&D plans are also generated by business units. In essence, Motorola needs to create a billion dollar business every year in order to continue growing at the rate of 15 percent per year.
Research managers also generate ideas. "Minority reports," which outline ideas outside the official strategy, are an important technique. Iridium and cellular telephones both started as minority reports. Another technique utilized in starting a new business platform is to task a group with finding reasons that the new effort will not work, or factors that would essentially kill the project. If this group cannot come up with a good reason to kill the project, it is an encouraging sign.
Metrics and evaluation are also important. Motorola's most important measures at the company level are related to product sales. For example, the company's target for percentage of sales generated from products introduced in the last six years is 60 percent. This is not the case currently, but progress is being made. Cycle time from concept development to product release is also an important metric. A few years ago the company average was 36 months. Motorola's "10X Program" set the goal of becoming 10 times faster. The average is now seven months, reflecting significant progress. Metrics are also important at the individual and group levels, for example the number of patents per engineer. Motorola's current average is 1.5, with the goal being 2. Percentage of on-time completion of key milestones is another critical metric. If the company commits to sufficiently support a project to reach its key milestones, 80 percent should be reached on time.