the marginal cost of distribution. The quid pro quo should be that the results of the research also should be part of the public domain.

  • The data cannot be obtained by other means. In rare cases, market failures produce a shortage of geographic data. Examples include, but are not limited to, (1) inability to obtain capital for exceptionally large or risky ventures, (2) low legal or technical barriers to copying, and (3) inability to persuade large numbers of dispersed actors to share information and resources.9 Thus, for example, it may be appropriate for a government agency to acquire broad rights when the risks of developing geographic data are large and the government must guarantee substantial spending in order to induce investment. We stress that such market failures are uncommon and should not be presumed without clear and convincing evidence of the broader public, not private, interest. The cost-benefit calculus in these circumstances will be difficult, and is complicated by the need to assess how benefits will be shared. Government should be loathe to fund investments that result in private monopolies or oligopolies.10

  • The data are required as "infrastructure" upon which other datasets or data products rely. Some geographic data supply the “infrastructure” needed to allow the integration of data among federal, state, and local agencies or to spawn new commercial products. In such cases, it may be appropriate for the government to acquire broad redistribution rights. A previous National Research Council report has suggested, for example, that environmental data should follow a “tree” model in which a government-funded

9  

Currently, there are indeed low technical barriers to copying geographic data, and the geographic data community has yet to develop the infrastructure and incentives that would allow efficient and effective sharing and exchange of geographic data among large numbers of dispersed actors (see Chapter 9, Section 9.3, for potential approaches for addressing these issues). However, it is difficult to argue that these low technical barriers and the less-than-optimal exchange infrastructure are so severe as to be causing shortages of geographic data in the United States.

10  

Oligopoly is a form of imperfect competition in which there are relatively few firms, each of which must take into account the reactions of its rivals to its own behavior (adapted from W.W. Norton and Co., 2003, Glossary, available at <http://www.wwnorton.com/college/econ/stiglitz/gloss.htm>).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement