Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
PART TWO 6. Public Sector Information: Why Bother?1 Robbin te Velde Dialogic, The Netherlands I have posed the question âWhy bother?â so you are already warned about the tone of this presentation. I have been involved in studying this topic for a while, and I am a bit chastened by it, but just to put it into perspective, there have been other studies before and after, and there will be more studies in the future, so it is good to compare. One may start with the Pira study in 2000,2 which was popularized by Peter Weiss (see his âBorders in Cyberspaceâ at http://www.epsiplus.net/report/borders_in_cyberspace). A year later, in 2001, there was a study by the Dutch government. At that time the Dutch government supported the open access stream. They did a really good quantitative study on the value of geo-information, but they have lost much of the studyâs effect ever since that time. More recently, the Office of Fair Trading in the United Kingdom also published a fine study. In terms of methods, what Pira didâand what most of the other studies have doneâwas to talk to some firms and then generalize the results. That is actually the way to go, in my view. It is sensible to go directly to the users because there are no standard methods available yet. The MEPSIR3 study was a little different. That study focused more on transparency and accessibility, on a massive pan-European scale. People were asked directly about the size of the market, what their roles were in it, and so on. One of the problems was that many people just did not know. They had no clue about PSI whatsoever, so this was clearly an immature market. In this situation, the only way to get some hard data is to go directly to the firms, but it is just not feasible to do this on a large scaleâand the generalizations remain problematic. Again, Pira started it all. The reason why we are here today is because Pira said the United States has twice the investment value for PSI, but they earn 40 times more from it. Why? The answer is that in the United States you have an open access model, and in the European Union you have a cost recovery model. This was the argument brought forward by, for instance, Peter Weiss. Although I believe the argument itself is still valid, the figures used by Pira were rather doubtful. Look, for example, at the added value of PSI quoted for the United States: a staggering $750 billion (against â¬68 billion for the European Union). That is almost 8 percent of U.S. gross domestic product (GDP). These are massive numbers. I think fighting in Iraq costs the United States $100 billion a year, so you can fight another 7-8 years for the same amount of money. Regardless, the 1 Based on a presentation found at http://www.oecd.org/dataoecd/12/49/40064800.pdf 2 Commission of the European Communities, 30 October 2000. Commercial exploitation of Europeâs public sector information: Final Report for the European Commission Directorate General for the Information Society. Pira International. ftp://ftp.cordis.lu/econtent/docs/2000-1558.pdf 3 http://www.epsiplus.net/reports/mepsir_measuring_european_public_sector_resources_report 25
26 SOCIOECONIMIC EFFECTS OF PSI ON DIGITAL NETWORKS creative bookkeeping that the study did (very successfully) raised the interest of the European Commission in PSIâand especially its potential value. Nearly five years later, we did the MEPSIR study. You would expect that the value would be higher by that time, because even with a modest growth of 3 percent, the EUâs â¬68 billion would now be â¬80 billion, but we actually arrived at a lower number, a base number of â¬27 billion. To be honest, that was still pretty much an overestimate. One of the things we asked for was the market size, which is different from the value added and much larger. Because all the studies had used estimates of value added, it was important in this study to use it as well in order to make comparisons. Another issue was that it seemed the small countries tended to overestimate the value of the market significantly. Compensating for that, we divided the numbers by two. Finally, in order to fill in the missing values, we had to correct the method used. Just as Pira had done, we had used GDP as a base. But that really does not make sense because you are dealing with some big countries, such as Poland, that just do not have a vibrant information industry. So it is probably better to use the economic value of the publishing industry as a distributor of PSI, and if you do that, you arrive at lower numbers again. When we make all of these corrections, we drop from â¬27 to â¬5 billion or even â¬3 billionâtruly a big difference. Obviously, these are not precise numbers, but they give us an order of magnitudeâabout 15 to 20 times less than the Pira study estimated. Now let us look at the more recent OFT study, which covered only the United Kingdom. It arrived at an overall number of almost Â£600 million. If you take away all the distortionsâdue to trading tricks and so onâyou arrive at a value of Â£1.1 billion, or almost double the original figure. If you then use this method to calculate a value for the European Union as a whole, you arrive at a total of â¬3 to â¬5.5 billionâpretty close to the MEPSIR figures. Finally, let us return to the initial 2001 Dutch study. It was rather detailed, but it was focused only on geo-information and covered only the Netherlands. Extending this to all PSI sectors for all of the European Union is, of course, a very tricky business, but if you do so you end up with values of between â¬5 to â¬7 billion. Again, that is much lower than the â¬68 billion that was mentioned by Pira. If we take a new look at Pira, it basically said that the United States has a much stronger private information industry. Furthermore, Piraâs number included IT software, hardware, Hollywoodâyou name it, they added everything up. But that begs the questionâand this is really the key questionâof to what extent this difference is due to differences in pricing policy, that is, to the difference between an open access model and a closed access model. The assumption was that there would be a more vibrant information sector if PSI were more readily available. It is actually possible to argue the other way around. Because the United States has a much stronger private information industry, there is more of a mature demand for PSI. And there may another factor: that Americans are just better at exploiting information services, whether in the private or public sector. Or there may be no relationship whatsoever. Honestly, I do not know, and there been no research in this specific area. Gerhard Wagner from Austria is probably one of the few who has done in- depth empirical research making comparisons between countries. He has found some
PUBLIC SECTOR INFORAMTION: WHY BOTHER? 27 differences between countries in Eastern Europe, for instance. Thus, at the least we can say that there does not seem to be one single model for research. Before 2000 (i.e., before the first generation of studies), few people were aware of the value of PSI. In many countries it was locked inside the government. So the really nice added value of the whole PSI debate is that it has opened up PSI resources. Currently, in what we can refer to as the second generation of studies, the focus is on the private sector. The basic argument is that if you simply open up PSI, you will generate a lot of money. However, what is lacking in the second generation of studies, with their exclusive focus on commercial reuse, is the broader societal value of PSI. I have been arguing this from the very start to the European Commission. Unfortunately, it is a tough sell because the hard (albeit modest) figures are in the commercial reuse area and the much bigger (yet softer) numbers in the societal use area are quite difficult to measure. The current obsession with making money out of PSI is rather shortsighted and probably even damaging. What you see in practice, for instance, is that private sector reusers are now being squeezed from both sides. The public sector is doing some interesting things, such as giving away its information freely, and the private sector has its own goods that it makes available freely, such as open source software. So the market for private sector resusers of PSI may be getting smaller. This does not mean, however, that PSI is not relevant at all; we are just looking in the wrong places for value. It also means that we must change the way we measure value. Up to now we have been trying to add all the individual revenues from all these firms, and even when you add everything up, you still arrive at disappointingly low numbers. I suggest that we change the perspective and look at the cost of not giving it away to the civil society. Although this may appear strange at first sight, it is already common practice in other economic domains. A prime example here is in environmental economics. There is no directive on the reuse of water, but if you would calculate what the cost is of not having clean water or not having clean air, you would arrive at massive numbers. This is how they managed to get the Clean Air Act passed in the United States. Thus if we talk about the economic value of PSI, we should focus not only on the financial valueâwhich is the narrow economic point of viewâbut also on the broader economic value. I will mention some basic methods of how to do this, which are again derived from environmental economics. First, do not look just at the use value, but also include the non-use value. This is the value of keeping options openâfor instance, of not having a database licensed exclusively to a publisher. Second, when looking at the use value of PSI, also consider the indirect value. In the Netherlands we have the website http://www.buienradar.nl, which, translated, means âshower radar.nl.â It gives low-resolution but near-real-time images of shower clouds moving over the Netherlands. We have lots of showers in the Netherlands. We also have many cyclists. Cycling in the rain is no fun. Therefore it is nice to know where the clouds are heading, so you know that you will not get wet when you cycle home. The website is a massive success. It has millions of hits each day, and this information is free. The website has the informationâthe low-quality radar imagesâfree from the National Meteorological Office. The other commercial weather bureaus did not care about this
28 SOCIOECONIMIC EFFECTS OF PSI ON DIGITAL NETWORKS information because the images were of such low quality. Buienradar.nl, however, put the information free on the Internet and generated income from advertisements. The revenues from advertisements are a direct economic proxy, but the broader economic impacts are much bigger. You may measure these impacts, for instance, by asking people what is it worth to them that they do not get wet when riding home on their bicycles. It is not really quite that simple. You have to be rather specific in describing what the services are that may be derived from a particular piece of PSI, but there is already a lot of experience with estimating this kind of hedonistic pricing. As a first attempt, I would guess the overall benefits of making PSI freely available to society are around the original Pira figures for Europe, i.e., â¬60-â¬70 billion. So why bother? Actually, this is exactly what I said one and a half years ago, here in Paris: (1) Government is a major producer of information, and (2) there is a lot of money involved in the commercial exploitation of information. It appears that there is a huge (potential) pot of gold, which is currently the second-generation view. However, it is important to keep in mind that (1) and (2) are separate things. Public sector information is important in its own right. If you think it is important, then use taxpayersâ money to produce it, and do not mix it up with private use. If you want a dynamic private European information industry, then you will need to take various steps, such as doing something about competition policies. But this has nothing to do with PSI, per se. How then does one determine the overall total economic value of PSIâincluding its wider societal value? This depends, really, upon the view of the citizen. If citizens think it is important, then the government should spend tax money on it. I want to emphasize that we should also take this wider (and important) societal value into account; only then will we be able to arrive at some hard numbers, following the methods in studying environmental economics as an example. One should not, however, focus too much on the value of commercial reuse. That is not the huge pot of gold after all, and to focus exclusively on it may even work against getting the most economic value out of PSI.