Media: why Norwich Union had to map the UK all over again
In today’s Guardian, Why a £5m mapping project had to double up on data explained why the restrictive data practices of the Environment Agency and Ordnance Survey cost Norwich Union £5m when it decided to draw up a map to assess flood risks a few years ago.
Did the Environment Agency have the data Norwich Union needed? It did. Would it make it available? Ah, that’s a different question. Which meant that Norwich Union spent £5m (which has to come out of its customers’ pockets, in the end) to create a brand new flood map – essentially duplicating the work that already existed.
- The following posts may be related...(the database guesses):
- Environment Agency yanks flood data from OnOneMap site (30 June 2007; score: 53.86%)
- Could free data have helped the Rural Payments Agency? (19 April 2007; score: 28.05%)
- Free Our Data nominated for New Statesman award (3 April 2006; score: 15.72%)
- Trading Funds report: final totals: economy +£179m, gov't -15.4m (19 March 2008; score: 10.92%)
- In the Guardian: what happens to the Postcode Address File in a Royal Mail split ownership? (9 March 2009; score: 10.91%)
March 30th, 2006 at 8:13 am
Total sympathy with Norwich Union and their abortive efforts to get access to ‘our’ data. But are they making their more accurate flood risk maps freely available? They might even be useful to the planners proposing the infilling of the Thames Gateway?
I would certainly be interested in having access to this data to examine the long term Planning Authority and Central Government housing plans thoughout the potential flood plains of the UK.
Mike Theis, Director, Max Lock Centre, University of Westminster
March 30th, 2006 at 9:47 am
Mike’s posting asks whether the Norwich Union’s data is made freely available. I don’t know the answer to that, perhaps Norwich Union will respond?
I do know that Intermap – the company that was asked to fly the country collecting aerial data on height – do make their results commercially available. Intermap were the company awarded the contract for height data required by local government in the Mapping Services Agreement. This was one of only two elements of local government’s procurement of map based data that did not go to OS. Intermap were able to provide a superior product at less cost than OS.
Ironic, isn’t it, that the restrictions on re-use of PSI brought a new commercial player into the market, and that local government – a significant public sector player itself – was then able to benefit!
March 30th, 2006 at 11:35 am
You may find the following of interest with respect to the above comments:
Intermap Press Release, March 28, 2002
REVOLUTIONARY DIGITAL MAP COMMISSIONED TO PINPOINT FLOOD RISK
http://www.intermap.com/corporate/pressRelease.cfm?pressId=396
You may also wish to browse Jill Boulton of Norwich Union PowerPoint presentation at the ESRI Home Security GIS Summit 11-14 September 2005 in Denver, Colarado, USA
http://gis.esri.com/library/userconf/hss05/docs/boulton.pdf
March 30th, 2006 at 12:48 pm
Why is NU creating this map? To make money from people who buy insurance, by getting ahead of the other insurers.
A more balanced picture would be given by putting “profits” in: ‘NU might have been spared £5m in costs – all of which comes out of its customers’ pockets’. (The article also states that this 5m was for ‘IT’ and consultants, not just data; a very misleading statement fom this ‘campaign’).
Would all insurers give cheaper insurance if they were all given the same improved map? It’s a market which requires effort to be put in, in order to provide better services, which are then rewarded. freeourdata seems to recognise this at least.
What it fails to recognise is that the agencies such as OS are required to operate under the same conditions. They are not fully funded. They have to bring money in by commercialising their products. This in turn means they have to commercialise them well. Mastermap is an example of this: a very good, high quality dataset that is ready to be used within modern technologies. There are already web-accessible versions of it, and an XML based format (GML: see http://www.opengis.com). This would not be the case if they were forced to give the data away for free.
If data must be available for free, the taxpayer must pay more for these agencies to be fully funded, as in America. Or rather than insisting that OS, EA, etc., give data away for nothing, to the benefit of companies, we should make sure that they are required to disseminate it well, with good commercial arrangements in place available to the likes of NU. This could be enforced through penalties on their current direct funding. Going back to the bad old days of agencies with no commercial sense or aspirations would not help.
With regards to smaller companies such as start-ups, they must have a good business case that’s not just based on using free data to set up a website service cheaply and then see if advertising revenue takes off. If a business is viable it will be able to pay for data.
The example of google maps is not valid – there’s much more, and better quality, data available when the UK is looked at compared to almost any other country. Multimap uses OS 1:50k at some scales, and these are clearly better than many of the other layers used. They also offer a very cheap service to incorporate their mapping services within other websites. Some things are working, clearly.
March 30th, 2006 at 1:54 pm
If all the insurers had the same map, then they’d all identify the same properties as being at the same risk. There would be agreement there.
The £5m isn’t forgone profit; that was NU’s budget, including IT and consultants *and* data.
If data must be available for free, the taxpayer must pay more for these agencies to be fully funded,
freeourdata has always recognised this – right from the start. It’s obvious. Our argument, which is borne out by the Peter Weiss paper (find it via the links page is that the economic benefit outweighs that comparable small rise in taxes. If it stops data being resold within government at ever-escalating “prices”, that can’t be bad.
The Google Maps point is an interesting one. There’s better quality data? But who *collected* it? In many cases, local government.
Some things are working, clearly
Our point is that they would work even better without the drag that putting a made-up price on the data exerts. Adding costs to a transaction reduces the chance of it happening (see, for example, stamp tax on house sales). Arguing that because some transactions happen that therefore everything is fine misses the point.
Plus, it would do no good to a company to set up a website and reissue the free data. Anyone could do that, so there’s no commercial advantage. What companies need to do, as they always do, is add value to the data, Our argument is that focussed private organisations can do that better than government. Get the OS etc to focus on collecting the best possible data sets. Then let private organisations and individuals do interesting things with it.
March 30th, 2006 at 3:54 pm
The following information may be of assistance in understanding some of the points made and why concern is being expressed within this Guardian led campaign
14 February 2005
Ordnance Survey enhances national digital height data
Ordnance Survey is announcing a major enhancement to its detailed digital height data defining the physical landscape of Britain.
http://www.ordnancesurvey.co.uk/oswebsite/media/news/2005/feb/landformprofileplus.html
26th May 2005
Office of the Deputy Prime Minister announces the National Spatial Address Infrastructure
http://www.odpm.gov.uk/index.asp?id=1144582#TopOfPage
Ordnance Survey Press release on: ODPM announces plans for Britain’s national spatial address infrastructure
http://www.ordnancesurvey.co.uk/oswebsite/media/news/2005/may/nsai.html
The European Union (EU) has public sector procurement laws in place which all public sector organisations within each of the 25 Member States have to comply with. (Here in the UK this has not always been the case with respect to geographic information). The EU process is meant to be open and transparent to order to encourage open competition. Hence the following notices appear.
With respect to the Local Government Mapping Service Agreement
The Official Journal of the European Communities notice on the Local Government Mapping Service Agreement that announced the MSA open competition contained the following information.
Reference 74274100
Type of Contract Restricted Procedure
Period of Contract 1.04.04 to 31.03.07
Publication date of Notice 17.06.03
Closing date for obtaining document 14.07.03
Closing date for Expression of Interest 25.07.03
Dispatch of Tender document 10.10.03
Closing date for tender 21.01.04
The Official Journal of the European Communities notice on the Local Government Mapping Service Agreement that announced the award of contracts for the MSA contained the following information.
OJEC Notice Published 20 September 2005 Award of MSA contract
Ref 1: Contract Awarded to Ordnance Survey, Value – £68,700,000, Contract Reference number 102513-2003
Ref 2: Contract Awarded to Intermap Technologies, Value – £650,000, Contract Reference number 102513-2003
Ref 3: Contract Awarded to Intelligent Addressing, Value -£7,100,000, contract 102513-2003
Total value of contracts awarded: Total value £76,450,000
Type of Procedure: Restricted
Observations on the above information:
1. The type of contract was restricted but it looks as if it was a negotiated procedure as the first notice states that the Award of contract was for 3 years running from 1st April 2004 through to the 31st March 2007. Yet the contract awarded runs from 1st April 2005 through to 31st March 2009.
2. The contract did not come into being until a given percentage of Local Government bodies agreed (70%) and this did not occur until 31st July 2005. This means the contract negotiations overran by 16 months.
3. The elapsed days for the MSA procurement was 775 days, 32% was the process of inviting tenders and selecting and appointing suppliers subject to final contact negotiations, 59.5% was spent in contractual negotiations, 8.5% was spent by Local Government organisation being notified, considering and signing the MSA.
4. During the protracted negotiations the Ordnance Survey announcement on height data was announced and the Office of the Deputy Prime Minister announced on the 26th May 2005 the National Spatial Address Infrastructure.
5. A rough estimate of the cost of this procurement (based on the information available in the public domain or released under the FOI Act 2000) can be made. For local government alone it is within the range 0.75% to 7.4% of the value of the contract awarded. If the other public sector costs are added in the then the range moves up several percentage points.
March 31st, 2006 at 7:36 am
I posted a note yesterday under that part of the blog labelled OS response to the Free Our Data Campaign. It ties in to Mr. Cobin’s subsequent posting here. I will not repost again in this section, but readers may wish to reference it after reading the above.
March 31st, 2006 at 7:49 pm
AS MF points out much of NU’s investment was in IT technology and consultants not data from public or private sources, the quote selected seems to be designed to mislead the uninformed reader.
Rumour had it that NU paid Intermap £1m for the rights to exclusively use their height data in a flood model, that is a lot more than OS would have quoted for their height model (although the Intermap product is probably of higher quality).
Licensing the Environment Agency data may be a bit cumbersome but it is available at a reasonable rate as several companies who have built the data into flood models can attest.
I can’t see why NU are complaining
Incidentally at the time that NU made their £5m investment there were at least two companies offering postcode resolution flood models to the insurance industry for well under £100k per year. Presumably NU considered the enhanced accuracy of their own exclusive flood model to give them a significant enough competitive edge to justify the cost.
March 31st, 2006 at 7:56 pm
have a look at http://www.prnewswire.co.uk/cgi/news/release?id=118387 for for why NU htought £m was a good investment
April 1st, 2006 at 11:56 am
I was astonished to note someone of Jill Boulton’s eminence complaining that the Re-use of PSI Regulations (not an Act but a statutory instrument based on a European Directive) did not free access to data “in the way that was intended”. The Re-use of PSI legislation isn’t and never was about access. It puts in place some rules once quite separate decisions about allowing access have been taken. But perhaps she has been misquoted.
The insurance industry’s Holy Grail would seem to be maximised revenue in return for carrying minimal risk at minimal cost in a highly competitive world. Nothing wrong with that, you may say, but NU’s decision to go with an expensive aerial mapping solution from Intermap rather than any other was a rational commercial one taken by sensible people in the full knowledge of the options, which included waiting for OS, EA and others to move forward with accuracy and availability improvements.
Dare I say it, but Charles’ article of March 30 was just a teensy bit lacking in evenhandedness in its efforts to demonise the Ordnance Survey and the Environment Agency. His having started the debate, I recommend to him the role of impartial umpire from now on, an act which will require lots of balance and objectivity.
The Environment Agency’s web-based “What’s In Your Back Yard” service actually quite neatly illustrates an important point of principle with regard to data and information which has been gathered at public expense. The individual citizen can look up his own property’s location and be provided, free of charge, with excellent quality data on flood risk, water quality and several other strands of information, which is presented onscreen together with a map (originating from OS) and some graphics.
If you prefer the low-tech option, of course, the Public Registers are still open to the public without charge.
It seems entirely reasonable to me, though, that a private sector company wanting to acquire something like the indicative flood risk database for the whole country in order to make a commercially profitable product out of it should pay something back to the investor – the taxpayer – in the form of either royalties or a license fee. Which is what happens, isn’t it ?
While we’re on the subject of information about the environment, some other important governing principles are covered by the Environmental Information Regulations (EIR), which have been in place in the UK for a number of years and only recently updated. It would also repay students of the matter to have a look at the Aarhus Convention, to which the UK is a signatory, to see if they can bring themselves to find what’s good about the present situation as well as what’s indisputably flawed (but, I hasten to add, not totally broke).
April 1st, 2006 at 5:58 pm
Mike Clarke said…
“It seems entirely reasonable to me, though, that a private sector company wanting to acquire something like the indicative flood risk database for the whole country in order to make a commercially profitable product out of it should pay something back to the investor – the taxpayer – in the form of either royalties or a license fee. Which is what happens, isn’t it ?”
In answer to your final question, this is not what happens.
It is indeed resonable that anyone using OS or other government information for commercial gain, pay back to the treasury something in return. However, this is already done via and should only be done via, the payment of taxes.
However, at present, organisations wishing to use government information, pay for its collection or generation indirectly through taxation, albeit in some instances under the guise of contracts awarded to the likes of the OS. This income to a large degree finances the collection, storage and distributiuon of data collected by the OS etc. The OS then sells the data collected to anyone else wishing to use it, and those wishing to use it pay taxes including on any profits they may make from its use back to the treasury. In short unless you are are government or a government agency you pay twice (or at least twice) to use public information. Worse, the government contracts awarded to trading funds are effectively being used as a subsidy to finance their data collection and generation activities, something not enjoyed by private sector businessses thereby causing a distortion of the private sector information supply marketplace.
This neatly demonstrates all that is wrong with fund-holders and the current philosophy concerning how the collection of expensive information sets required for the economic and social well-being of the nation, is financed. It creates a self-limiting market, that can only give minimal financial returns to government coffers, in doing firstly, preventing the service economy expanding via entrpereneurs using this information to form the basis of new businesses, and secondly retarding the economic activity of existing businesses.
If information collection or generation is paid for by government or its agencies, this information has been paid for through taxation – public money. It is therefore, entirely reasonable and right that this information is freely available to those that paid for it (and anyone else), be they individuals or business, non-profit making organisations or charities, to use for whatever purpose they may wish.
April 2nd, 2006 at 11:27 am
Nicholas
In the case of the EA data referred to in this thread the data is freely available to any interested user on the EA website. So any taxpayer (or non taxpayer) can gain access to it at no charge.
Free access to individuals does not extend to providing the data in bulk to organisations like NU without charge. NU did not seem to be claiming at the time that they launched their flood map that they were being forced to invest £5m because of the inflexibility of the EA or OS, they were touting the significvant advantages of their greatly more accurate model (those who understand the insurance industry may question whether the NU’s pursuit of accuracy will benefit policy holders or reduce the value of several million properties by making them uninsurable in the longer term).
The idea that the tax system will redistribute the profits from companies recieving our “free data” back to the treasury more effectively than the current charging mechanism is just laughably naive. It is typically the larger companies, who have most to gain from this campaign, that are most successful at reducing their tax payments through legitimate loopholes and clever structuring. The majority of taxpayers will probably be best off with the current funding and charging model.
But this is just my opinion, let’s ask the economists rather than the geographers to provide some objective coomment on this topic.
April 2nd, 2006 at 9:17 pm
The point, steven, of the Free Our Data campaign is to make that data avaialble in bulk to organisations like NU without charge. Yes, indeed. And to anyone who wants it. The rationale is that you encourage economic activity, which generates benefits to the government through taxes.
The idea that the tax system will redistribute the profits from companies recieving our “free dataâ€? back to the treasury more effectively than the current charging mechanism is just laughably naive… let’s ask the economists rather than the geographers to provide some objective coomment on this topic.
The campaign was launched following the study of Peter Weiss’s paper (referenced on the links page. There are multiple occasions I’ve come across of people and companies who have been held back from doing projects that could have generated tax revenues by the cost of licensing. Oaks and acorns, maybe.
The majority of taxpayers will probably be best off with the current funding and charging model.
That’s a very sweeping statement which could be used to justify any amount of inertia over virtually anything. Minimum wage? The majority of taxpayers…businesses will avoid.. Child tax credit? The majority of taxpayers… businesses will avoid.. That’s not meant to be a vindication particularly of either of those policies (though they fit with The Guardian’s general policies) but to point out that “probably best off with the current..” is not the way that politics works, and there’s never been a campaign that works on that presumption.
April 3rd, 2006 at 1:59 pm
Steven,
The EA flood risk zone data you refer to is not freely available and indeed is not data in any useful sense (ie privided as a vector drawing or gridded data)
The public can only access the EA data as a dumb image served over the internet via the EA internet map server. This dumb image is not data in any useful sense, you cannot analyese it and you cannot reformat it or manipulate it in any meaningful or useful way. What you are seeing is NOT data but an image of that data! Moreover, it is licensed in a very similar way to the way OS data is licensed for viewing over the internet. You can basically look, but not touch.
As to the more general pointabout economics of providing free access to information sets paid for with public money, this model works very well in the US. There as a result there are are a lrge number of businesses providing geonformation, survey, cartography, and internet mapping services. I find very ironic in a nation that is unashamedly capitalist and to the right politically of most in Europe, has such an enlightened policy towards public access to information sets paid for from the public purse. It appears to have had no detrimental effect on the economy there, quite the opposite.
I might add that it is not only the uS that has this enlightened policy, i believe (someone correct me if I am wrong) that Denmark does too.
April 4th, 2006 at 3:45 pm
Wasn’t Ordnance Survey financing its own flood map or flooding study? I remember some information about this and then it all went quiet. What happened? Does anyone know?
April 5th, 2006 at 4:39 pm
Charles,
You have referred to it a couple of times: Peter Weiss’ standard polemic has been around for about ten years now, with occasional minor updates and adjustments for particular circumstances, the occasional new title, and a certain amount of help from his friends. But I always had a problem trying to figure out exactly what was his agenda.
It seemed to me that he could never make up his mind whether to be a) an academic, b) a lobbyist, or c) just get on with the job the US National Weather Service paid him for.
I was often puzzled by his obsession with the issues presented in Borders in Cyberspace, and I certainly worried about the apparent lack of intellectual rigour in some of his analyses and conclusions, which relied on sloppy and out-of-date third party research and less than totally accurate statements of “fact�. Some of his sweeping generalizations were quite disingenuous, and the supporting “evidence� was at best blatantly and conveniently selective.
In describing US practice and policy, for instance, Weiss failed to mark the crucial difference between federal and state/substate information. In several other ways, he also tended to gloss over some important differences between the US, Europe and other parts of the world.
In comparing European and American practice, he ignored the fact that Europe consists of a number of independent nation states which, although currently connected together by the EU and other groupings, were nearly all at war with each other within living memory. The US, on the other hand, has been a single federal state for a good deal longer (even if it did experience the agony of a civil war nearly a century and a half ago). It would indeed be astonishing if the organizational, political, constitutional, legal and cultural differences were less than they are.
In spite of what is frequently stated, various claims about the macro-economic benefits of cost-free public sector information have not been shown to be completely true. No matter which way you cut it, it still comes down to the question of Who Pays ? Somebody has to, whether it is the user or the taxpayer. This is inescapable.
The central issue in the debate is the proper extent to which the government a) needs certain information for governance purposes, b) uses it in ways other than for the purpose for which it was gathered, and c) profits from trading in and/or adding value to it.
So, is the argument we are making an economic one or a moral one ?
Peter Weiss always avoided this point, but a couple of thousand years ago another interesting chap named Cicero dismissed even this basic question with one of his own: cui bono? – who benefits?
April 7th, 2006 at 12:31 pm
This is really interesting.
There’s been repeated mention of the “we’re already paying for this data” argument. We aren’t, the government currently pays for a large proportion, with the rest coming in revenues.
It obviously comes down to a practical / political question about, ‘Who pays?’. From a political point of view it would be interesting to see if the ‘public’ are willing to pay more tax for companies (and individuals) to get free access to data (once you’ve explained the benefits to the economy, jobs for people, etc.).
I’m guessing the companies would be for it. For people (me included – I’d love to not have to buy all the OS maps for places I go walking) to make this decision they need to know the costs. Will it be possible to do this for an OS dataset, say Mastermap? If it’s so expensive to buy then it must be making OS some money (even if they spend a bit of this on ‘marketing’ or maintenance). Just a matter of summing up their potential sales (do they have to declare this in their reports?) adding on the money they make from selling maps (nobody will buy these anymore: we could all quite cheaply get a GPS with interactive 1:1250 maps if this data were free) and dividing it by the number of tax payers. I imagine there would have to be small-infrastructure improvements to serve up this data too, as it’s quite big and a lot of people could be using it. Apparently it has 40 thousand updates a day.
I expect people would rather have more money to the NHS, because we’re a sentimental nation.
The other issues have also been mentioned and are harder to quantify – removal of costs for people paying each other for this data, benefits to the economy by putting people into work, etc.
One thing hasn’t been touched upon: will the agencies disseminate data well if they don’t benefit from doing so? I suspect not. Maybe they could be punished somehow if they fail to do so. But I feel we could do this now to make sure that government agencies are obliged to have open and transparent agreements available to anyone who wants to use their data, and their effectiveness in disseminatating data is assessed, and that the costs are appropriate to the costs of collecting the data. This might resolve a lot of issues and bring some of the benefits of ‘free data’.
Otherwise: please highlight some more examples where companies are unable to do something due to costs, and where services are limited.
September 18th, 2007 at 11:22 am
The EA flood map available online does not constitute data per se. It is merely a planning tool; giving a rough (and conservative) outline of the flood zones. This is what most planning authority will use before deciding what to do with a planning application on the flood plain. At the moment, most Local Planning Authorities (but not all) have a lack of knowledge when it comes to flooding; and most planning officers, when asked about it, refer you to their colleagues from the drainage department. Perhaps, here again, the lack of freely availabe data with regards to flood risk is the problem. Yet, one cannot help but wonder why there is not a better line of communication between LPA and the EA? Especially in relation to the time it takes the EA to release data.