Blog

We’ll need more than £40m* a year to get free maps - specifically politicians willing to share

g-crescoli-364214.jpg

A commitment in the Autumn Budget to spend £20m a year for the next two years to get MasterMap into the hands of small businesses won’t help much towards getting what open data advocates have been wanting for decades - access to the treasure trove of Ordnance Survey mapping data for everyone.

Open maps! But only if you’re a small business. And just for two years.

It was with a tinge of excitement that I saw in the Twitter coverage of the Autumn Budget that it appeared there’d been a coup in getting a big open data release included in the Budget. A key geographic data set had been mentioned by name in the document and it was accompanied by a commitment to provide what appeared to be new funds to compensate Ordnance Survey (OS) for it’s release to the tune of £20m a year. On closer inspection however, the commitment included some interesting turns of phrase which suggested maybe that the open data community’s exuberance should be tempered:

“To further boost the digital economy, the government will work with the Ordnance Survey (OS) and the new Commission, by May 2018, to establish how to open up freely the OS MasterMap data to UK-based small businesses in particular, under an Open Government Licence or through an alternative mechanism, while maintaining the OS’s strategic strengths. The Budget provides £40 million a year over the next two years to support this work.”

The “Commission” referred to is a Geospatial Commission which was also announced in the Budget and would apparently provide “strategic oversight” to the various public bodies who hold this data. I’m not sure that’s traditionally what Commission’s do - they’re more usually set up to hold government to account or undertake an investigation - but this “Commission” was at least more tightly defined than other oversight bodies of the past on geographic data which might help focus the effort a bit. 

Rather than being an ongoing release of data the money referred to in the budget seems to be specifically focused on supporting some sort of analysis to work out how to get the MasterMap data into the hands of small businesses. Now £40m is a hell of a lot of money for such a piece of work and so we can reasonably assume that in reality the money will be going to pay Ordnance Survey for the access to MasterMap data itself. So despite the reference to the Open Government License in the commitment, in reality it’s probably better to describe this as support startups to use OS MasterMap data rather than a true open data release. The more cynical among us might perhaps argue that the reference to the OGL is nothing more than “open data washing” - an attempt to assuage a hungry open data community who have been very keen to get access to this data for years.

As I say, £40m is a lot of money and, in terms of recent announcements on geographic (open) data, this is a relatively big deal. Getting a politician to sign off on this sort of money isn’t easy, especially in the current climate where open data is seen as a nice to have and overshadowed by big strategic issues such as Brexit. I don’t want to detract from the team who were responsible for securing this commitment, and don’t doubt it was hard fought. Much like others I welcome this commitment, as I recognise the importance of geographic data. However, I think we’d be premature to mark this commitment up as a win for open data.  

Open-data-lite

To be fair to Ordnance Survey they do release some open data and they have been releasing it for quite a while. They have a suite of open data products available but these tend to not have as much detail in them as MasterMap and have some limitations in terms of licensing. However, they are paid to provide these products. They first released data after a consultation in 2010 and an announcement in that year’s budget. Since then there have been several changes to the package and subsequent attempts to persuade Ordnance Survey to release more data as open data. Most of these arguments are built on the idea that greater access to this infrastructural data would unleash a range of new and innovative uses, building value in startups and business in the UK. Not to mention the efficiencies in the public sector of everyone having access to the most up-to-date map for the delivery of public services.

However the reality is that Ordnance Survey hasn’t released the most detailed mapping information it has as open data for the simple reason that it makes money from selling it, and returns a profit to the Treasury every year. Clearly if all of this data was released as open data then the taxpayer would have to find some other way of keeping our friends in Southampton in business. Open data advocates have forever argued that releasing mapping data in particular would mean an increased tax take from all the innovative business which are created. This value would accrue back to the Exchequer and make up for the shortfall from selling the data. Cue many an econometric study and consultancy report giving ever more impressive estimates for the economic benefit of releasing open data to the UK economy.

Now the problem with getting Whitehall, and most importantly the Treasury, on board with this line of argument is that it relies on the idea that most of the tax will actually be able to be collected from all those new and existing businesses who are doing juicy new things with this wealth of data, and, that the increased tax revenue will cover the cost of making maps in the first place. This would be fine if it wasn’t for the fact that in reality the Treasury, and Ordnance Survey, knows that some of the main companies who would do best out of this situation may not be the biggest contributors to the UK Exchequer. In the current situation where Ordnance Survey sells data then at least some of the value is realised from these organisations as they are buying mapping products. However, if we give all the data out for free are we really sure that we’re going to be able to tax all of the additional value which has been realised and will those who are getting the most value out of this newly-free data be UK tax payers?

And so you can see how the tortuously worded commitment in the recent Budget might have come to be. This is the nearest to a commitment on open data that OS was likely to be comfortable with and which allayed their concerns about the risk of handing over all their data to the competitors. It no doubt also spoke to HMT’s concerns about not being able to recoup the benefits of the data release through tax. It seems that both OS and HMT could be persuaded to give preferential access to UK small businesses - but it seems that the precise ways by which this is to be done haven’t yet been agreed in government, hence the May 2018 deadline to work something up. An open data release of mapping data, however, it was not.

We’ll need lots more money to get free mapping data out - and to believe in sharing

So how do we get mapping data available as open data then if we’re never going to be able to convince the Treasury that they’ll truly see the results of the economic benefits? My personal opinion is that the only way that you’re going to get this type of data out is by finding a politician in power who is willing to expend some political capital with the Treasury and take a punt on funding Ordnance Survey in it’s entirety. Now this won’t be cheap and will be considerably more that £20m a year. According to the latest OS annual report it cost £93.8m to run in 16/17 and had a profit of £16m. This is a lot of money to find every year and to suggest will be recouped back in tax. It won’t be easy to make this case and so the argument would also have to be bolstered by one which was more ideological - framing the release as the right thing to do not just for UK innovation but for transparency and accountability. I also think it would have to be made in the context of a political view which was also more abundant - which not only saw the release of open data by government as the right thing to do but also an action which supported an economy in which value was created through sharing rather than rent-seeking through ownership. So if you were to follow my line of argument through, the implication is that we’re not going to get this sort of rich, infrastructural data released as open data any time soon.

Ironically, we will mostly likely need to wait for a Labour government in order to get the next chance. Arguably a Labour government would be much more likely to put in place a comprehensive regime of business tax on just the sort of foreign-owned technology companies who the Treasury at the moment worry aren’t paying their way. This might be the key issues which would convince the Treasury to back a move to release open data. It was after all a Labour government which went out and funded the first batch of releases of open data from OS.

This is not to say that there aren’t Conservative politicians who would be convinced by the potential of open data (there were a couple for instance in the Coalition Government). It’s just that they won’t win against their colleagues who will be making the case for OS and others to be charging more for their data and also those who will be arguing against a tighter tax regime for corporations. The only time a Conservative politician would be likely to be able to make a case for the full gamble of supporting the release of all OS data would be where we’re in a situation where we’d want to be seen to be supporting our own tech sector but also at the same time looking to give a non-obvious bung to American tech giants. Which isn’t a completely unlikely situation if, post-Brexit, there’s a slow down in investment in startups and the tech giants are looking elsewhere to base themselves.

Until then - Budget commitment groundhog day

And so in the absence of these enlightened politicians with deep pockets and an abundant world view we’re likely to get a succession of these types of commitments in Budgets and White Papers. They will commit to feasibility studies and access to high value data sets perhaps for particular groups who are in favour, for a limited time. But the question remains - what happens when this £40m runs out? As the open data community is keen to point out, businesses need access to reliable sources of data and two year’s worth of data isn’t reliable. Most likely what is on the horizon is yet another compromise commitment in a Budget or Autumn Statement to a feasibility study, commission or some such which will keep the dream alive. Alive until a large tech giant has mapped everything anyway and there’s no need for run a state mapping agency. At which point there will definitely be no further open data commitments in Budgets.

*Full disclosure — Updated from £20m a year when I first published. Argument still rings true though — not enough to cover open data release.

Ed ParkesComment