Guest Blogs

This is an underpinning and frequently avoided question throughout all the commercial sectors that rely upon the re-use of Public Sector Information (PSI) to create value-added products for end users. But it should not be so¹.

Public Sector Information, is just that. It is information (of whatever kind) generated by enterprises that are government owned, run or commissioned at public cost (i.e. paid for out of revenues raised through taxation). One might therefore suppose that, fundamentally, such data “belongs” to the “public”; that is it belongs to those who have ultimately funded its generation, its management, handling and storage. But is this how ownership of these data are seen by those who, on behalf of the “public”, undertake the generation, management handling and storage of them?

Sadly, no. In fact, those government organisations responsible for these tasks on behalf of the public all too often regard the data as “theirs”, that is they see it as “owned” by them and to be provided to others only as a concession in return for some consideration such as a license fee or re-use fee, and not at all if, in their judgement, to do so might somehow make life too much harder for them. This is particularly (but not exclusively) so when those “others” lie within the commercial sector of the economy and intend to re-use the data to provide value-added products and services to the overall benefit of the economy in terms of employment and the generation of relative wealth.

A clear example of this thinking is to be found in the field of meteorology. Here the World Meteorological Organisation (WMO), the international Agency of the United Nations that co-ordinates meteorology world wide, has passed a Resolution of its Executive Council (WMO RESOLUTION 40 CG XII (1995)) that effectively allows all National Meteorological and Hydrological Services (NMHS) to withhold the free provision of meteorological data that they have generated or hold (other than a strictly limited sub-set that in practice is too difficult to protect) to any organisation other than another NMHS. This deeply protectionist measure is rooted in the concept that the data “belongs” to the NMHS and not to the “public” who paid for it. In publicly available documents that can be found (with some searching) on the WMO web site, one can find statements such as:

“However, in some cases, the competition with private service providers has led to conflicts. Among the key issues that cause concern are access to NMHSs data and information (my italics) and the possible reduction in services provided by NMHSs.”

This mindset, is surely not one that a UN Agency charged with representing the interests of the entire meteorological sector world wide and not just those of the NMHSs, should support or encourage? The data are not “theirs” they are “ours”. They belong to the public and, subject to a reasonable charge for the additional work of retrieval and re-transmission; they should be freely available to us all.

[1]A detailed exploration of the nature and status of meteorological data can be found in:
Ton W. Donker; Access to and re-use of public-sector environmental data
And information. Policy developments with a focus on the european hydro-meteorological scene. Polish Academy of Sciences, Geographia Polonica, Vol. 80 No.2 (Autumn 2007).

Scientists at Imperial College, London have reported in the journal Science[1] that shorter, milder winters experienced since 1985 on St Kilda Island in the Hebrides, have resulted in smaller Soay sheep. This seems to be one of many impacts that are arising from changes in climate. Others, such as short term flooding, increasing storminess, longer and more frequent hot and cold spells, will give rise to significant social and economic consequences and the meteorological industry could potentially make a significant contribution to their management. But the charges levied by the National Meteorological Services (NMS), the custodians of climate data, for access to those data, are preventing this in many cases by making the resulting value-added contributions uneconomic.

“Climate” is actually the historical, integrated effect of “weather” and establishing what the past climate was and the current climate is requires the analysis of what are, on a day to day basis, mainly normal weather observations. Some observations are made specifically to allow the climate to be more fully defined than would be possible using only routine weather observations but often these “climate” observations are available to and used by the NMS for routine weather work as well. Which makes it all the more strange that some of these same NMS do not make such observations available to the Private Sector in real time. Moreover, when they are eventually made available as “climate data”, the NMS often charge huge sums of money for access to them.

It is true that data on which climate records are eventually based require additional quality control and computation beyond that which is normal for purely “weather” observations. A “daily” average temperature, for example, may be compiled from several hourly temperature measurements but in the age of automatic observing, processing and communication, this hardly justifies either preventing real time access to the original observational data or, when access to the climate data is available, the massive costs that are often charged for them. To obtain the full set of daily average temperatures (that is one number per day!) from a single station in Latvia for ten years (not at all an unusual requirement for the simplest climatological task) costs about 27,010 Euro! And this is for PSI for which the industry and the citizens, through their taxes, have already paid!

Is it any wonder that the information economy in weather and climate is growing at less than 2% per annum in Europe and is less than half the size of that of the USA?

[1] Ozgul A, et. Al. The dynamics of phenotypic change and the shrinking sheep of St Kilda. Science DOI: 10.1126/science.1173668

At this time of shrinking national economies and falling Treasury revenues it is hard to imagine why European nations should simply forego a minimum of almost 300 million Euro of taxation annually. But they do.

Using 2006 figures, the value-added meteorological market in the US and Europe should be rather similar in total potential size at around $2 x 1011 p.a. In the US, penetration of this market is around 0.7%, at $1.4 billion per annum, whereas in Europe, it is around 0.3% at $650 million per annum Moreover, net growth in this market in the US has been around 17% per annum over the period since 2000 while in Europe this growth has been around 1.2% per annum. It is a simple calculation to show that if the European market performance matched that of the US, then European Treasuries would annually gain a net 290 € million in taxation revenue from this small sector alone.

The major difference between the US and Europe in the exploitation of the inherent market value of PSI in this sector lies in the re-use model. In the US the data are available to all at the marginal cost of re-distribution and re-use is governed in general by simple, overarching, non restrictive license terms. The US government, as originator of the PSI, makes no attempt to compete with the private sector in exploiting the PSI to generate revenue from the market. By contrast, in Europe, most (though not all) governments, operating a model developed in the 1970s, charge significant sums for the PSI, apply restrictive, often complex and expensive licenses and have downstream trading arms that compete with the private sector for the value added business. The effect is that they inhibit market growth in the private sector and to fail to grow the market themselves, particularly at the low price end, because their trading arms are fighting high fixed costs and, often, entrenched bureaucracies.

At a time when every government needs to stimulate its economy, increase employment opportunities and increase tax revenues it is crazy to continue to operate in this 1970s mindset. The late Peter Weiss (see Borders in Cyberspace: Conflicting Public Sector Information Policies and their Economic Impacts U. S. Department of Commerce National Oceanic and Atmospheric Administration National Weather Service 2002) has argued that reason why the US market thrives and the European market does not is primarily because of the differences between the two models for the supply of wholesale data (PSI) outlined above. Although this argument has been challenged anecdotally, there seems to be little, if any, convincing published argument against it. Moreover, it is very difficult to find any clear, credible alternative explanation of why this very large difference in the development and growth rates of the respective markets should exist.