This one weird trick could fix the British economy
Free the data, free the people
Update (2023): Given how this post has become an important part of the O’Malleyist canon, if you’re reading it for the first time you might like to skip straight to the second section (“Liberate The Postcode Address File”), where the meat is. The opening is mostly scene-setting.
This is going to sound crazy but Liz Truss was right.
No, not that we should have blown up the economy for spurious ideological reasons. Or that it is a good idea to cut taxes when facing down inflation. She wasn’t even right when she claimed the Queen was good.
But she did accurately identify a major problem in her one conference speech as Prime Minister, and she was right: Britain’s economy needs to grow.
The grim reality is visible on a number of economic metrics: We’ve got a terrible productivity problem, and wage growth is stagnant, while business investment mysteriously plateaued in 20161 and has still not recovered even to that point post-COVID. And to put it in perspective, if current trends continue, the average British family could be poorer than the average Polish family in around a decade.
These woes are widely acknowledged on both sides of the political spectrum. But what makes them particularly annoying for the government, as Sam Freedman identifies, is that the solutions are almost universally toxic to the Conservative Party’s electoral coalition. Building more houses, getting closer to Europe and allowing more immigration would all make the economy stronger and make us collectively better off. Similarly, embracing O’Malleyism and carpet-bombing the countryside with wind-farms and nuclear reactors would turn Britain into a glistening utopia.
But as much as I hate to admit it, none of these ideas stand much of a chance of getting past the ruddy-faced, cosplay Churchills on the Conservative backbenches.
Unfortunately I’m not smart enough to figure out what other big macro-interventions could be made to fix the economy. But what I do know is that there is one small policy intervention that would make the country just a little bit better – and very gently help nudge our sluggish economy in the right direction.
That’s right folks, I’m finally making the case for opening up the Postcode Address File.
If you follow me on Twitter, you’ll have been waiting for this one for a while. If you don’t, you’re in for a treat. In any case, why not sign up and get more blazing hot politics, policy and media takes directly in your inbox?
Liberate the Postcode Address File!
The chances are that last time you bought something online, most of the address fields were filled in automatically for you. You typed in your postcode, pressed a button and the website or app was smart enough to figure out your full address. It’s simple, and makes buying stuff slightly less tedious.
But what you might not realise is that the reason that websites can do this is because they have licensed access to the Postcode Address File (PAF). This is a massive database of every postal address in the UK, which is legally owned and controlled by the Royal Mail2. It contains over 30 million records in total.
And here’s the mad thing: Even though address data is pretty basic information, it’s actually really annoying to get hold of.
Imagine you’re a small business that needs to use address data, or if you’re a bedroom coder and you’ve got a clever idea for a new app that uses address data in a creative way.
If you want to access the data, you need to pay the Royal Mail a hefty fee for the privilege.
For example, a one-off copy of the database costs £360, but this isn’t enormously useful because the dataset is always changing, as businesses shut down, new buildings are built, and so on. So instead, you can pay for quarterly updates - for a mere £900 a year, or get monthly updates for £1,0803.
But this is just a “data supply” fee. On top you need to pay £85 for every person who is going to use the data. Oh, and there’s VAT charged too. And this is only the fees where the imagined scenario is basically workers sat at desks using Microsoft Excel or updating a company CRM system.
If you have the sheer audacity to imagine that you might want to use address data on a website, then the fees hike even more, to at least £6,150. Or if you want to use the data across an organisation, you can look forward to paying £18,400. Every. Single. Year.
The costs spiral from here. Royal Mail charges “solutions providers” and larger corporations even more money. And it isn’t enormously clear, but if you want to use some of the other associated data the Royal Mail holds, such as “alias” data, which is things like the names of houses, rather than just the numbers, then… there’s fees for that too. You get the idea.
To be absolutely clear and fair to the Royal Mail, there are special “microbusiness” and charity dispensations. You can apply and theoretically access the address data for free. But for the former, this is only for developing your cool new thing. Once it launches and you want people to use it, it’s time to pay up, before you’ve even made a penny.
And in my view this is absolutely insane.
Unleash the nerds
This is obviously about more than just address auto-complete at checkout. It is about innovation. If we can invent new ways of working more effectively, our productivity increases and our economy will grow. It is innovation that will mitigate our terrible economic malaise.
Opening up public data is an important part of this. Open data makes it possible for anyone to build tools, services and businesses that can improve the way we work and live, without us having to wait for the government to do it for us.
Transit data is a particularly striking example of this in action. Because Transport for London and National Rail share their schedule and real time data openly, there has been an explosion of tools and apps built on top of it.
It’s why Google Maps can tell you when your next train is due, and it’s how CityMapper can mash together Tubes, buses and other forms of transport to help you figure out how to get to your destination. And it’s why instead of battling with a terrible National Rail website, the real nerds know that RealTimeTrains is a much better place to figure out if your train is on time (and which bits of infrastructure it is passing through en route).
And just as importantly, it also means we can make awesome home decorations that light up based on real time transport data.
It is sad that as things currently stand, address data is currently locked away by the Royal Mail, making it harder for this sort of innovation to occur. If you’ve got a great idea for a business, or even just a browser plug-in that uses address data, then forget it – unless you have upwards of six grand in your wallet.
So this is an enormous missed opportunity for innovation.
And if you're still not persuaded, let me put this in the starkest possible terms: The fees for the PAF is probably why no enterprising nerd has yet done what we all need, and created an app that will send a push-alert to remind you which coloured bins to take out each week4.
What makes breaking open the PAF urgent too is that in the future, address data is likely going to become even more important. If we’re going to live in a world where home deliveries are not just common, but are often carried out by robots or even drones, they will require high quality, accurate geographic data to function.
And even if you’re sceptical that robot pizza delivery will ever come to pass, I still think there is a strong case for liberating the PAF.
To borrow an analogy I first heard made about, er, crypto, by the technology analyst Benedict Evans: When 3G and 4G were invented, few people would have predicted Uber and Instagram. But what was easily predictable was that faster internet on mobile phones would turn out to be a useful and good thing.
I think the same is true for the PAF. Even if we don’t know exactly how liberated PAF data might be used in the future, it feels like something that will be useful – because our address is a foundational part of our lives, and is an obvious digital building block for services.
This is perhaps why in 2013 the Open Data Institute called the PAF the “critical missing dataset” in UK open data. And why it pointed to a Deloitte analysis that apparently showed that when Denmark’s equivalent address datafile was opened up in 2002, it led to a 30-to-1 financial benefit-to-cost ratio.
In other words, it seems clear to me that there is a strong economic case for liberating the Postcode Address File from the clutches of the Royal Mail.
And hopefully it is a cause that politicians of all stripes can actually get behind, as it’s pretty much the perfect technocratic policy: It’s a bit left-wing, as it involves effectively renationalising a Royal Mail asset that was sold off during privatisation, but it’s also a bit right-wing, as it’s in service of unleashing a torrent of business innovation5.
Banging the drum
I’m far from the first person to identify this problem. The tragedy of the PAF has long been on the radar of data nerds. Charles Arthur was banging the drum for it eight years ago, as part of the Guardian’s #freeourdata campaign, and open data policy guru Peter Wells has over the years put in the actual hard work of talking to policymakers and actually knowing what he is talking about6. And as far back as 2009 campaigners were pushing for it - and even future Labour Deputy Leader Tom Watson got involved.
Sadly, all of this great work did not lead to a change in policy. That’s why I’m writing about it, all these years later.
But it does seem like now could be the time for another push. Not only is there compelling economic case, but the Geospatial Commission is holding a consultation.
This is the government body that looks after geographic data7, and according to the consultation’s page on GOV.UK, a key focus is going to be looking at how innovation can be unlocked across the geospatial value chain.
It doesn’t mention the PAF or address data specifically, but this is clearly relevant. Plus it seems crazy to me that the consultation is asking questions about using AI to analyse satellite photos, and completely speculative technologies like quantum computing, when the Postcode Address File is right there, as the lowest of low hanging fruit.
So if I’ve succeeded in suitably firing you up, maybe go and tell them what you think.
A better world is possible. And maybe one day we’ll get there, thanks in part to an open Postcode Address File dataset helping us figure out how to actually find it.
Congratulations! You made it to the end! If you enjoyed this content and would like more of This Sort Of Thing, make sure to follow me on Twitter (@Psythor) and subscribe to my Substack for more takes on a wide range of different subjects. And please do share or forward this on to anyone you think might be interested – I know a lot of you nerds are well connected, so let’s see if we can get anyone in government to pay attention and finally free the PAF!
I guess this is just one of those completely unknowable things that we’ll never be able to explain.
To be clear, this doesn’t include your name or personally identifiable information - it’s a database of buildings and premises, and how to access them. The separate ‘Alias’ dataset also includes business names. But as far as I’m aware, individual householder names are not included.
Just to show you how antiquated this system is, you can only either download the data via FTP, or (of course) pay even more money to get it on a DVD.
Even if they could build a scraper for each local authority, the prohibitively expensive bit would be licensing the postcode lookup data.
If you want to read about a similar thing, check out this article I wrote for E&T at about the National Underground Asset Register. In that case, the government is building a big map of underground pipes, cables and other infrastructure so that utilities companies can work better together. Like opening up the PAF, it’s exactly the sort of thing the government should be doing.
There are surely many other people I’m unaware of too, given that I’m late to the issue.
And amusingly, until he resigned a couple of weeks ago it was technically overseen by Gavin Williamson, who I’m sure paid close attention to goings-on at the Commission.
I've just moved from the UK to Australia, and of all the things I thought I'd miss about the UK, postcodes wasn't high on my list of guesses. Here, a postcode is a 4 digit number. If you're lucky, it has enough granularity to identify your suburb.
So instead, Australia has a highly regularised database for identifying streets and numbers, even to the point of regulating st/ln/av/rd etc.
Of course, that ain't free either....
Coming to this later through your parking platform post… it seems the Consultation is still underway, open until 16th Dec 2023.
And I consider the OS open data reform @charlesarthur ‘s greatest legacy!