Although a quite extensive search has been made, it has not been possible to find an economic model for the conditions of the public sector information reuse. Therefore similar or potentially related academic and consultancy documents will be summarised in this section.
It is evident, but a need to be remarked, that PSI reuse economy is a fully digital economy. Digital economy differs from ‘regular’ economy in that, once created, the product reuse presents a marginal replication cost, and for most of the analysis could be approached to zero. PSI reuse is an extreme case of this economy. Difficulty in its estimation is similar to the estimation of the economic impact of investments in Information technologies.
A model to measure such impact was created by Kauffmann and Kriebel as early as in 1988, in which they apply the Data Envelopment Analysis (DEA) (Kauffmann & Kriebel, 1988). Some of these principles could be applied for a potential model to assess PSI reuse.
The PIRA analysis encompasses all firms that are in one way or another related to PSI, based on broad estimates using national accounts data (PIRA, 2000). PIRA takes the size of the information industry as an upper bound proxy for this market, particularly for estimating the US market. Based on this model, next figure shows the expected impact for Germany.
The MEPSIR analysis is based solely on the surveyed added value by all first-order re-users, focusing on how much added value can be attributed to PSI re-users. The analysis excludes second order impacts as identified in Illustration 7.
It should be noted that this economic impact is based on expectations from the sampled sources but not the ones coming from real sampled data.
The Pollock model, considers in its analysis only three main sources of PSI information: geospatial data, meteorological information and official statistics. His study is quite focused on how a single entity could be sustainable and how pricing and governance models interact. It does not take into account external effects of the information release and associated turnover for society. And in this sense this model attends only to the basics of the identified model in Illustration 7.
It is interesting to point out that incentives to subsidized models should compromise fixed costs and variable costs (those attending to the information output) in order to maximize incentives to improve performance and, at the same time, being innovative.
And most interesting of all is the statement “any of the charging policies discussed could be used successfully if a independent, transparent and coherent governance structure were in place” so it is not so dependent on the method used but in the coherence between charging policies and governance.
However the analysis only examines implicitly the impact on the overall society, impact that should be assessed in order to determine the level of subsidies to be assigned to the publisher (PSIH) entities.
This study is based on the regular price creation in this market, which is a hypothesis to be confirmed. In fact within research it is clearly identified that several industries could flourish if a marginal cost approach is used and therefore the linear hypothesis is a non-sense. Besides this, transaction costs (those associated to the payment, are also neglected as irrelevant and in favour of a more conservative hypothesis). Validation of such estimation is not provided and examples found by the author deny this approach.1
On top of that, regular price vs demand curve cannot be assumed as linear when price goes to zero, so most of the hypotheses presented in the research are not as solid as it should be required for a solid theory. Besides this valuable analysis of elasticity on the demand warns that the long-run could change as long as new applications and uses will be developed for the current information provided.
Total welfare =T(consumer surplus, producer surplus, government)
PSI holder entities, those public entities which are the owner and publishers of the information, have large fixed costs to collect, process and filter the information, whenever they have marginal costs2 on publishing and creating copies of it (Pollock, 2008).
Kalampokis refers a 5-stage model for e-government described in (EU, 2009) (see Illustration 5). In this model, open government data (which in this case is equivalent to PSI) is determined since stage 1 to be published. However this model does not pay dedicated attention to the fact that the public information released would be reusable as remarked by Alani (Alani et all, 2007). On the contrary, their analysis focuses on the complexity of the transactions that the citizens are able to interact with their administrations. Kalampokis et al. model relates added value provided to the information and organizational and technological complexity as described in Illustration 5. It also addresses one of the biggest challenges for the information reuse which is the standardisation from different public sources. Especially when several of them publish the same or quite related information. However this model defines a stage two in which Public Administrations have a connection of their data. It implies shared standards, compatible technologies and co-ordination procedures for the common publishing. This stage two is quite far from current and short term future situation. The emerging portals publishing information count on with weak implemented standards for the released information. So their applicability to current situations is limited