Funding Options for Trading Funds and Other PSI Holders

The following is a short, policy-oriented, examination of the funding options for public sector information (PSI) holders, with application to the main data-supplying trading funds.

Those seeking a more detailed treatment of PSI issues are directed to this book chapter on the ‘Economics of Public Sector Information’.

NB: The analysis presented here is entirely my own. (And is, for example, quite distinct from the ‘Cambridge Study’ of which I was a co-author).

1. ‘Two-Sided’ Nature of PSI Holders such as Trading Funds

All information holder have two sides to their operation: the ‘write’ (‘updaters’) and the ‘read’ (‘users’) side.

For example, Companies House must collect the data for its register (write/update) and then may supply this information to third parties (read/user). Similarly the Ordnance Survey makes changes to their database in response to surveys and changes in the environment (write) and then supply this data to third parties (read). These examples can be extended to most information providers.

This fact, that is that all datasets involve both read and update operations, has important implications for policy, in particular:

2. Charges can be Made on Both Sides.

Charges to maintain and develop datasets can be levied on both the read and write side of the operation (in most cases — more on this below). That is, both those seeking to update the data (for example register a company) and those seeking to use the data can be charged.

Thus, a policy-maker seeking to fund the production and maintenance of a dataset (or datasets) has three possible options (not mutually exclusive!):

  1. Government funding: fund from general Government revenues or other external source
  2. Updater funding: fund by charging updaters, that is those who make changes to the dataset(s)
  3. User funding: fund by charging those who use the dataset(s)

Which of these should/can be used will depend on the social, technological and political circumstances. As discussed in the report and elsewhere, the evidence strongly suggests that Government or Updater funding is to be preferred over User funding — at least for bulk copies of ‘upstream’ material.

Of course, in some cases it may not be possible to implement updater funding, perhaps because it is impossible to identify/charge those making changes (one cannot charge ‘God’ for making the ‘the weather’ and hence changing the Met Office’s databases!).

However, in many cases, updater funding is feasible. This is important because politically it may be easier to alter the balance between user and updater funding than to move to direct payments from government (central or local), particularly if the immediate costs would be significant.

3. Application to particular trading funds

1. Ordnance Survey

It should be clear that OS fits within the read/update framework. In particular, many of update changes to the OS databases nowadays are likely to be anthropogenic (that is man-made or man-caused). In particular they arise from building work and the like. Furthermore, such building work is already being logged fairly precisely in the form of planning applications.

It is therefore feasible to implement updater charges — most obviously via some kind of ‘levy’ on planning applications, or, if that involves too great a change, an increase in the Local Government contribution to an External Funding Group with the suggestion that it be raised via charges on planning applications).

How large would these charges be?

Taking the total number of planning applications last year in the UK at ~ 600k/yr (figures from CLG suggest 156k granted in apr-jun 2007) a rough average levy would be: £35-55M / No. of Granted Applications = £58-91 per application (of course one would probably wanted to vary this cost with the size of the planning application).

Using the Government fee calculator for planning applications the current charge for a planning application that alters an existing dwelling is £150. For the construction of a new dwelling the basic charge is £335, for 10 £3550 and for larger applications the fees are even higher. Thus, even for the cheapest fees the levy would only be around a 45% increase, and a 10 house development it would be only 2%, and by applying the levy proportionately rather than absolutely overall it would likely equate to an increase of less than 5-10% on average (and this would be partly offset by the availability of cheaper maps — a major input into the planning application process). Moreover, these fees are likely a small part of the total cost of making a planning application and hence in overall terms the increase is going to be tiny (as a wild guesstimate something like 100th or 1000th of that: 0.02-0.002%)

2. Met Office

In the Met Office’s case updater charges appear infeasible. However the ‘funding-gap’ in the Met Office case from moving to free supply is so small (under 1M) and the existing Government payment (via the PWS) comparatively so large that one wonders whether it would not be possible for the Met Office to be encouraged to supply its main wholesale datasets at marginal (distribution) cost with the PWS increased slightly to compensate.

3. Land Registry

Land Registry is already running a two-sided charging model with the majority of their income coming from updater charges. Currently the Land Registry does not make the majority of its database available in any bulk form. Thus, here the main requirement is for them to supply a bulk product at marginal (zero) cost. Any shortfalls that result then be made up by increases in updater charges. I have not yet done any calculations but would imagine, given the scale of their operations and that update-charges already dominate read-charges, that this would require very small increases.

4. Companies House

As with the Land Registry they are already deriving the bulk of their revenue from update charges. Moving to marginal (zero) cost for their digital bulk data products will only result in a small reduction in income so this can be made up by a small increase in charges for updates.


The situation here is very similar to that of the Land Registry. DVLA’s general ‘updater’ income appears to dwarf their current ‘user’ income (at least for use of bulk products). Furthermore, the surplus of income over expenditure achieved by them in the last few years would be more than sufficient to cover any loss in revenue in moving their bulk digital products to marginal (zero) cost. That said, the question of incentives is important and so it may be wise (here as elsewhere) to link any reduction in charges in one location with an explicit mandate to make compensating increases elsewhere.


Given the special situation of the UKHO (discussed in the report) it seems best to leave the UKHO entirely to one side for the time being.

7. Summary

  • OS: full funding via updater charges is likely in which case no Govt funding needed
  • Met Office: External funding needed (but already supplied on a large scale)
    • Amounts involved are small (perhaps < 1M)
  • Land Registry: full funding via updater charges clearly feasible
    • Amounts involved probably small (not clear as no bulk product currently supplied)
  • Companies House: ditto.
    • Amounts involved are small (around 1M)
  • DVLA: ditto.
    • Amounts involved are small (around 1M)
  • UKHO: ignore for time being.