Conclusion

International cooperation in research and development is an important activity to the U.S. government: Billions of dollars are spent on a rich and varied set of interactions every year. The nature of the activity--diverse locations, variety of topics, multiple goals--makes it difficult to track spending, assess the benefits, and ultimately defend the activity to Congress and the American public. Moreover, the disconnection between agreements signed at the executive level of government and the actual commitment of funds to collaborative research makes it difficult to know what the United States has promised to fund and what ICRD is actually being conducted. Finally, the dearth of well-developed measures for ongoing research complicates efforts to describe the benefits of these activities. This chapter discusses the positive and negative implications of this disconnection and briefly describes possible approaches to improving government's ability to track ICRD spending and assess benefits.

Scientific Opportunity Versus Political Expediency

The ability to describe the benefits of ICRD to the American public and its congressional representatives is particularly important given the opinion of some that international cooperation provides aid to foreigners at the expense of U.S. science, or that the interests of "good science" are subjugated to political interests. To test or refute this assertion, an effort must be made to describe actual, ongoing R&D activities and measure the benefits of these activities. The inventory of actual ICRD spending conducted for this study found a rich and varied amount of ICRD taking place across many agencies. Also, the case study suggests that the United States may be receiving significant benefits from this type of activity.

In the course of examining actual ICRD activities, we found little to support the idea that ICRD is primarily another form of "foreign aid." Quite the contrary, we learned that ICRD is primarily aimed at fulfilling the mission of the sponsoring agency, not to fulfill an international agreement. U.S. government agencies tend to guard their R&D funds jealously: Program managers do not spend dollars for the sake of political expediency. Given that most R&D spending decisions are made at the program level, and to the extent that ICRD is a part of that, this suggests that U.S. government funding for ICRD, on average (and perhaps better than average), is funding good science. Moreover, nothing in our research indicates that the interests of science and the interests of politics are in conflict. Certainly, federal agencies conduct ICRD to achieve multiple agendas, among them, scientific opportunity, technical efficiency, and strategic benefit. ICRD serves all three goals:

• Government grant-making activity funds scientific researchers seeking to gain access to an important global or foreign scientific resource or to collaborate with a leading foreign researcher.
• Government improves technical efficiency by cooperating with other countries on projects in which each country produces a piece of a larger system and cooperates in its final production.
• The government works with other countries to contain nuclear waste, to reduce pollution, to alleviate suffering and the spread of infectious disease, in a way that serves strategic and political needs.

Some ICRD provides only indirect or long-term benefits to the U.S. taxpayer; even in these cases, cooperation enhances the U.S. science base by increasing scientific knowledge and access to resources and data. The case study on earthquake sciences and seismology examined a number of projects that had only indirect benefit to U.S. citizens. However, the United States is leveraging foreign research dollars and greatly increasing new knowledge created about earthquakes. This benefit will eventually accrue to citizens in enhanced disaster preparedness. Another example might be efforts by AID to contain infectious disease in Africa. U.S. citizens benefit only indirectly from this activity, but U.S. researchers learn a great deal about the etiology of infectious disease in ways that will benefit the United States.

Failure to follow through on ISTAs signed between the United States and other countries paradoxically provides evidence of the relative absence of conflict between scientific opportunity and political goals. If an agency does not see scientific advantage to cooperating with another country, that agency often does not follow through on an executive-level government-to-government agreement to fund research. Most agencies report that scientific or technological opportunity determines whether a project will be funded, not whether it is required by an international agreement. Many agencies report that an umbrella agreement to cooperate with another country in a specific area of science would play only a minor role in the decision to fund research.

Indeed, government officials report that agency-level agreements, rather than "umbrella" or framework agreements, often result from joint identification of an opportunity for mutual benefit. While these projects are not always implemented--research is risky, after all--agency-based agreements may be an indicator of scientific opportunity in many cases.

Bringing ICRD Within the Fold of International Science and Technology Agreements

While the government currently maintains hundreds of agreements to conduct ICRD, not all scientific cooperation takes place under a government-to-government agreement. Some agencies keep careful track of what activities fall under which agreements, but not all agencies limit projects to those that have signed agreements. Moreover, contractors and grantees, being several steps removed from the government's diplomatic concerns, are sometimes unaware that they are working under a bilateral or multilateral S&T agreement. For cases in which intellectual property is created or trade or other disputes arise, ignorance about the regulations governing ownership, licensing, and royalties can have significant implications for where the intellectual property is commercialized. Moreover, for cases in which additional or new activities are being negotiated, it would be helpful to decisionmakers to have a map of where existing activities are occurring.

It may be useful to policymakers and agency officials to set a baseline to determine which R&D activities take place under ISTAs, and which take place outside them. Then, as agencies fund additional activities or sign new ISTAs, they could report this information on a Web page or other electronic repository. Given the advances in computer networking, this type of data collection should be relatively easy. These data would be very useful to U.S. and foreign researchers looking for opportunities to cooperate with others.

Continually refining our understanding of how best to measure ICRD will help to improve the efficiency and effectiveness of these activities. In addition, asking researchers and program managers how they would assess the effectiveness of ICRD activities would provide useful input to the process of collecting data and assessing progress.

Improving Data Collection

To better track ICRD, link it to ISTAs, and measure benefits, improved data creation and tracking are needed. Data are available to assess the benefits of ICRD. Collected efficiently, placed in proper context, and combined with qualitative testimony, these data can provide usable information to help decisionmakers track benefits and compare the returns on one activity against others. The framework suggested in this report provides a way to select measures and organize the data to streamline the assessment of and perhaps ultimately defend international R&D cooperation.

Improved data, provided in a timely way, would greatly enhance the ability of program managers and policymakers to monitor ICRD. The relevance and availability of the measures suggested in this study should be improved and additional ones should be developed. In the process, priority should be given to less intrusive measures, such as bibliometric publication and citation data and milestones, so that measurement does not unduly influence the choices researchers make when conducting ICRD projects. Given that these unobtrusive measures are scarce, additional research and development is needed to refine indicators and to suggest the best methods for collecting and aggregating measures.

Program managers and policymakers may wish to work at the agency level to develop a system of "signposts" that report on outputs and that track the relevance of those outputs to the scientific and technical community. This could include periodic reports on the excellence of ICRD taking place in a particular field. Bibliometric tools can count the number of jointly authored publications relative to U.S.-authored papers and the relative frequency of citations of jointly authored publications. Government agencies such as the DoC National Technical Information Service and government contractors such as Scientific Citations, Inc. could be tasked to refine measures of the outputs of ICRD, to craft signposts of progress, and to make these data more consistently available to government offices.

Two additional actions could improve ICRD data and, thereby, government monitoring. First, agencies should add a request for measures to international project plans. Many agencies are already developing these measures and collecting these data in response to GPRA and other government accountability practices, but the measures collected at the agency level will not aggregate to the interagency level. A request to periodically provide these measures to the NSTC could help to create the data for interagency comparisons. Second, periodic review or continual data monitoring of international science and technology agreements should be requested. This data can be collected by the agencies, or by a technical agency such as the National Technical Information Service. The data can be aggregated at the NSTC level to provide feedback on the extent to which ISTAs are reaching their stated goals.

Improving Data on Spending

Information about the benefits of ICRD accruing to the United States would provide input to agency and NSTC-level funding decisions. However, a method needs to be developed to track ICRD spending over time. This study examined only one fiscal year, 1995, of government ICRD spending, and the data collection and analysis process were labor intensive. Providing a monitoring capability of ICRD year-to-year would require one of the following:

• The easiest approach would be to collect yearly R&D data on ICRD-intensive programs. Analysts would be able to take a percentage of spending (say 4.5 percent, to track with RAND's finding of ICRD's 4.5 percent of all R&D spending) as representative of ICRD spending. The summary table in Appendix A contains a list of relevant bureaus. The outlines in Appendix B list the relevant programs. Such an approach would provide an approximate figure of ICRD activities, but it may miss significant changes and subtle shifts that take place from year to year. Moreover, it would not allow the cross-agency comparisons that provide input to the assessment framework.
• A second option would be to reproduce every few years a cross-cutting inventory like this one. This approach would provide useful snapshots of activities and a trend analysis, but it would not provide data to allow decisionmakers to track activity and make key interventions in programs that are showing significant promise or are lagging expectations.
• A third and more time-consuming, but ultimately more comprehensive, approach would be to ask agencies to "tag" ICRD activities for the purposes of budgeting and reporting. This would allow tracking over time, would show trends, and would allow analysts, program managers, and policymakers to aggregate the data for the purposes of cross-cutting analysis. A tagging system could also provide a way for agencies to identify which activities take place under international science and technology agreements, and within this category, which of these activities may produce scientific results or intellectual property of commercial value.

---

Contents
Previous section
Next section