private sector could also do if it had access to the upstream information, then that qualifies as a downstream activity and the minimum amount charged should be full cost recovery, because anything less would be undercutting competition. It is not necessarily possible to ban public sector bodies from the downstream market altogether and, in some cases, there could be some genuine economies of scale or scope because of vertically integrated public sector bodies that have the capabilities to perform downstream activities. The main concern is to ensure a level playing field so that competition can occur, and if the private sector can make downstream products more cheaply or meet consumer demands in other ways, then the public sector body should consider pulling out of the market.
One implication of these considerations is that public sector bodies should make their information available at the earliest point that it is useful to businesses. The public entities should not complicate things by bundling their information with other data and then not allowing the private sector access to the background data. While it may be necessary to have aggregated and anonymous provision of personal data at the first stage, when it is still upstream data, the data should be made available on the same basis as to the entities’ own downstream operations—that is, for the same pricing and licensing terms.
In terms of assessing the value of PSI, there are improvements that could lead to a doubling in the value of PSI to around a billion pounds per year. How is this possible? Whereas previous studies have looked at the gross value added by PSI to the economy—a top-down approach—the tendency was for these studies to overestimate the real value of PSI to the economy because they were not looking at any possible substitutes. For example, the Pira study in 2000 used the turnover of public sector bodies, and that figure included non-PSI-related income, government grants, and so on. Other studies, such as the Ordnance Survey commissioned in 1999, also used this top-down approach. The contribution of Ordnance Survey data to Great Britain was estimated to be between ₤79 billion and ₤136 billion, representing about 12 to 20 percent of the gross value added to the entire economy during 1996. Roger Timm and Partners conducted a top-down study for the British Geological Survey (BGS), assessing its value at 8 percent of GDP, or ₤34 billion to ₤61 billion. These numbers clearly overstate the significance of these organizations, because to conclude that together they underpin over 25 percent of the economic activity in the United Kingdom is just not plausible. It would not take many more studies before the conclusion would be that PSI contributed 100 percent or more of the economic activity in the United Kingdom. Thus, the top-down approach ignores the counterfactual, because it ignores what happens in the absence of PSI.
It is better to look at the value of PSI today in terms of net economic value by estimating the willingness to pay for PSI minus the cost of producing and supplying it—that is, a bottom-up approach. To that end, OFT looked at the consumer detriment, the types of detriment that could occur, and the likelihood of them occurring for any given group of PSI holders, as well as the potential magnitude of each negative factor. The types of detriment included: unduly high pricing; restriction of downstream competition, including refusal to supply or discrimination; and failure to exploit PSI. The income of PSI holders was examined by asking how much they earned directly from their information activity, excluding government grants, and, especially, if other public sector