Background

I've written previously about the campaign to unlock open address data in the UK.

One of the more recent developments was an amendment to the Data Protection and Digital Information Bill in which Liberal Democrat peer Tim Clement-Jones urged the Government to mandate regular publication of "an authoritative list of UK address data as maintained by local authorities including, but not limited to building number, name and street address, geographic coordinates, and a unique identifier."

The amendment was supported by several other peers: Natalie Bennett (Green Party), Tom Watson (Labour), Francis Maude (Conservatives), and – in debate – Dido Harding (Conservatives).

The amendment was not moved – and the DPDI Bill itself fell when Parliament was dissolved before this summer's general election – but the open addresses proposal was discussed in the House of Lords during committee debate in late March.

The Sunak Government, represented in the debate by DSIT minister Jonathan Berry (Viscount Camrose), resisted the amendment, relying on arguments that seemed to have been written by the Royal Mail. Royal Mail maintains the Postcode Address File (PAF) and is the main stakeholder in the current government-backed monopoly that controls re-use of address data in the UK.

Following a group discussion about open addresses at Open Data Camp last month, I sent an information request to the Department for Business and Trade, asking for any recent information it might hold on arguments for or against the release of national address data on open data terms. DBT leads on government policy related to Royal Mail and is the departmental sponsor for Ofcom, the UK regulator for communication services.

FOI response from DBT

DBT handled my information request under the Freedom of Information Act and returned a response this week: FOI2024.05941.zip (1.1 MB zipped).

DBT confirmed that it held information in scope of my request. An unknown quantity of that information has been withheld, mainly on the basis that it is exempt from disclosure because it relates to the formulation or development of government policy and/or is covered by legal professional privilege.

However, DBT has disclosed three redacted documents:

Draft Speaking Note for Viscount Camrose

Nearly all of the Speaking Note is redacted. However, the topline points closely prefigure the minister's response in the House of Lords debate:

1. High quality, authoritative address data for the UK is already accessible for a reasonable fee from Local Authorities and Royal Mail, with prices starting at £95 for national coverage. There is also free access to this data for developers to innovate in the market.

2. The Government also makes this data available free at the point of use to over 6,000 public sector organisations, as well as postcode, unique identifier and location data available as open data – free for anyone to use for any purpose.

3. This amendment would result in significant additional costs to taxpayers and represent low value for money given the accessibility of data currently.

4. For these reasons, the government is not able to accept this amendment.

It's unclear whether the Speaking Note reflects any advice from civil servants at DBT, or the department's own position – it seems likely that the Sunak Government was minded to resist the amendment, and only requested arguments against open addresses.

Letter requesting a review of HMG's 2016 decision on an Open Address Register

A month after the Lords debate, the peers who supported the amendment sent a letter to Viscount Camrose, care of the Department for Science, Innovation and Technology (DSIT), requesting that he consider commissioning new analysis to inform a review of the Government's 2016 decision not to support open addresses:

We believe the digital economy has changed significantly since 2016, and the economic case now strongly favours opening up. The importance of high-quality, freely available address data has only increased.

Address data is no longer just about sending letters, or even about online deliveries: it is crucial geospatial data infrastructure, used by tens of thousands of UK businesses. High-quality address data is vital for delivering commercial and public services, and for joining together other geospatial datasets; opening it up would improve its quality and support our digital economy.

In the light of this, many of our competitor economies have already made postal address data freely available. The EU now requires its member states to make address data available openly, as part of its mission to support SMEs to drive economic growth. The US, Canada, Australia and New Zealand all either have or are developing open address datasets.

We would be grateful if you would consider commissioning new analysis to inform a review of the 2016 decision, in the light of these changes.

DBT did not know whether DSIT had made any response to the Lords letter, prior to the change of Government. I have sent DSIT an information request to find that out.

Royal Mail's view on the open addresses amendment

DBT also disclosed an email from February 2024 in which Royal Mail set out its view on the open addresses amendment. Following is a slightly abridged version of the relevant text:

Royal Mail's ownership and management of the Postcode Address File (PAF) is an operational necessity, helping to underpin the universal postal service. It is intrinsically linked to the provision of our core service and obligations. PAF data, in particular postcodes, are the logistical starting point for delivering letters and parcels to people across the UK. …

We believe PAF is a good example of how a business can provide widespread access to the data it holds, while recovering its costs to maintain quality. Royal Mail recognises PAF data is of value to others. Royal Mail provides access on reasonable terms in various ways. For example, the Public Sector License (PSL) [sic] permits widespread usage and sharing of PAF data throughout the public sector, free at the point of use. Over 50k businesses are licensed to use PAF, demonstrating that usage is widespread and current fees are not prohibitive. In addition, consumers have free access to look up 50 addresses per day via Royal Mail's online Postcode Finder service, which relies on PAF.

The cost of maintaining PAF is covered by the licence fees. We keep these fees under regular review. … We make PAF available to anyone who wishes to use it on terms that are reasonable. …

Royal Mail believes that the amendment incorrectly implies that the current ecosystem for maintaining address data and making it available on free and reasonable terms does not support those groups.

Further, the minimum content of the proposed address list avoids reference to postcode even though, when ordering goods and services from organisations the most regular first step to populating the address fields online or via an agent on the phone is the postcode. Royal Mail considers that a list that does not include postcodes is not 'authoritative'. We also contend that an address based on the minimum requirements, such as a Building Number and Street Name, Geographic Co-ordinates, or UPRN, is not sufficient for most users of our address data.

One concern is that the proposed dataset would lead to software developers using multiple alternative data sources to 'approximate' a traditionally recognisable address. This would lead to fragmentation of 'standard form' addressing and confusion for businesses and consumers alike when trying order goods and services, verify customer IDs and manage their customer account data, ultimately leading to increased costs and delays with mail. For Royal Mail to meet its service obligations, it is essential that postcodes are continued to be used on mail and parcels entering our system.

For Royal Mail this amendment risks increased costs and delays to our service. The financial implications stem from both the costs of maintaining the trusted accuracy of PAF, as well as the costs associated with correcting and sorting incorrect addresses. Because postcodes are integral to post delivery, Royal Mail needs to continue to maintain PAF. Royal Mail is therefore concerned that the free alternative would lead to the dilution of the (regulated) licensing revenue for the use of PAF that currently helps offset the approximately £30mn cost of maintaining the database.

Adding to this are the financial and time implications of incorrect or incomplete addresses on letters and parcels entering Royal Mail's system. This would significantly increase sortation and distribution costs, along with increasing the time that mail and parcels are in our system which further impacts our service obligations. As such, if the amendment is adopted with the minimum content, the impact to Royal Mail would inevitably extend to businesses and people across the UK.

Comments

As I've said before, I think the debate about re-use of address data in the UK focuses too much on postal addresses.

Clearly, Royal Mail could lose out financially if Government decided to publish an open address file, and disrupt the existing monopoly, without compensation. But the extent to which Viscount Camrose simply ventriloquised Royal Mail's position in his response to the DPDI Bill amendment is extraordinary.

Royal Mail's Postcode Address File is only a downstream product of the national supply chain that produces address data, mainly from gazetteers maintained by local planning authorities. The commercial market for PAF is purely an invention of government policy. By prioritising the protection of Royal Mail's licensing revenue, successive governments have placed the interests of a multi-billion pound company over those of the thousands of smaller companies for which PAF fees are an input cost.

As Royal Mail says in its email, the Postcode Address File is an "operational necessity". Royal Mail maintains PAF principally for its own purposes. The proposal in the Lords amendment is for open address data "as maintained by local authorities" – this data is currently collated nationally by GeoPlace LLP, a public sector entity, and does not require the release of any intellectual property owned by Royal Mail. The availability of this new open source of national address data need not prevent or disincentivise Royal Mail's maintenance of PAF.

Royal Mail argues that "minimum content of the proposed address list", which does not include the postcode, would not be sufficient for most users of its address data. At the same time, Royal Mail affects to be concerned that the availability of this new dataset would encourage businesses to use non-standard addresses, and that users would start to post letters without postcodes on them.

This is all topsy-turvy. The widespread use of non-standard addresses in webforms and business applications is one of the main problems that the campaign for open address data expects to reduce. Open availability of an authoritative national address dataset would inevitably promote the use of higher quality address data throughout the UK, because it would reduce the cost of validating addresses at the point of collection and enable businesses to standardise addresses in their own customer databases without giving away any IP rights in that data.

There is no reason to think the Lords proposal would significantly reduce use of postcodes, because postcodes are already available as open data (except for Northern Ireland). If, as described, the open address file included UPRNs as identifiers, the postcodes could be appended to the file by way of UPRN lookups already available from ONS.

Royal Mail estimates a £30 million annual maintenance cost for PAF. At the moment a good chunk of that is already publicly funded – the UK Government pays Royal Mail £6.2 million each year for the Public Sector Licence (PSL), in addition to the £5.1 million it pays each year to Ordnance Survey and local government for use of AddressBase under the Public Sector Geospatial Agreement (PSGA).

Given that the lion's share of work on producing new address data is undertaken day-to-day as part of local planning functions, it seems to me that a roughly equivalent amount should be sufficient to pay for an open address file, without sacrificing maintenance or data quality – particularly if the Government required Royal Mail to start using its own resources to fund its operational need for address data.