the repository could build on this model. Many of the participants thought that it could be valuable to share more information and approaches to measuring impacts of PSI policy.
Another topic of discussion was data collection. One obstacle to data collection is that countries have different national accounting practices, which makes analysis of PSI market development very difficult. Some kind of international cooperation therefore might be needed in this area. While at a national level, PSI sectors and PSI products may be isolated, international industry and product classifications usually do not allow such separation and generally lump several different content or service industries together with PSI-based ones.
The difficulties of obtaining data on the use and reuse of PSI (or PSI market development more generally) were discussed as well. The academics and sometimes even the private-sector bodies that conduct surveys on PSI use generally do not have the authority to demand the submission of data. For example, the OFT indicated that it is sometimes in a better position to elicit responses from government bodies. Therefore, it may be useful to involve government competition bodies in this process, because in certain cases they may have better access to data.
The need for a theoretical model of expected benefits was highlighted as well. The e-government economic programs were mentioned as one potentially useful example. There may be some parallels between them and PSI use in terms of the theoretical model of expected benefits; PSI research may thus benefit from examining these models.
Another area of discussion was how best to learn about the issues that users and reusers of PSI are facing. Participants raised a number of ideas, including suggestions for data collection methods, such as how to get in touch with users and reusers and how to group respondents. One suggested approach is to announce new PSI research that is relevant for reusers so that reusing businesses could be identified. Publicly funded libraries can be used for disseminating PSI to citizens and for research on the users and reusers of such information. Publicly funded libraries are often used as a cost-effective way for disseminating government data, especially in North America and Scandinavia.
The mapping of the value chain of PSI came up a number of times. The suggestion was made that the value chain could be modeled in terms of activities or business processes, attaching costs to the value chain. The underlying idea is to make studies more comparable by linking them to value chains.
Participants also suggested a number of ideas for further studies on PSI, including the transition costs of switching PSI policy regimes (e.g., the cost-benefit analysis of moving away from the U.K. trading fund model that was introduced in the 1980s), the substitution of PSI by private-sector-generated data, examining licensing costs in different areas of PSI, and the extent of network externalities of online PSI.
Finally, participants thought it might be interesting to look at other subject areas in order to learn from comparable situations. For instance, one might examine value generation from free online access in terms of the parallels it has with the economic impacts of the liberalization of telecoms.