Tower Hamlets Council Property Transactions Puzzle – Data

Also see
Weird Housing Status Report 1

What’s with the bells?

There are so many weird things in this story that should cause alarm bells to ring that we have decided to insert the alarm emoji 🔔 whenever there is cause for, er, alarm.

Provision of social housing is by far the most important issue in Tower Hamlets.

If a Mayor can show he or she has built lots of social housing then they are electable. If he or she cannot they have no chance of holding on to power.

Tower Hamlets is not a big place, 19.76 km², but more and more people want to live here. Its population has risen from 156,000 in 1987 to 308,000 in 2017 and is still rising to a projected 372,766 in 2030. The borough’s population density is currently 16,237 persons per square kilometre making it the second most densely populated area in the UK.

While skyscrapers shoot up in Canary Wharf they usually only have a pitiful number of homes for ordinary people. Combine this with the high cost of renting property in the private sector, huge pressures on council funding and a massive social housing waiting list of 21,000 households on the housing situation is always critical.

Working on another investigation into Tower Hamlets Council we needed to make a simple double check of an issue and submitted a Freedom of Information (FoI) request which contained far more data than expected. [LINK]

So what’s the problem?

We have numerous records of AP1 payments for council house transactions obtained by chance through the FoI request and no credible explanation of why these properties were bought using the AP1 system.

Below you can see a diagram illustrating the problem.

The main issue is that although both these data sheets are from the same organisation, Tower Hamlets Council, and they both refer to housing assets it is almost as if the data is from two different organisations relating to completely different subjects.

Tricky, innit?

You can also download the above diagram in PDF format.

The UK government has a standard way of identifying property and street information which uses the Unique Property Reference Number (UPRN) as a unique identifier for every addressable location in Great Britain. The reference for Buckingham Palace is UPRN: 10033544614.

Control your excitement, please!

The Local Government Association describes UPRNs as “key to almost everything that’s delivered or achieved by councils”. If you are really into the world of UPRNs – and you should be – you might want to read this handy guide? UPRN: A guide for councils

There is no UPRN in either of these datasets 🔔.

This is not surprising for the AP1 data. The lack of a UPRN in the standard Payments to Suppliers data is typical of the lack of detail in most Tower Hamlets Council documentation which creates an invisible barrier to 🔔 transparent government.

Remember that these two very different data sets describe transactions relating to the council’s social housing stock yet they do not look as if they have been produced by the same organisation. 🔔 🔔 🔔

Apart from Tower Hamlets Council staff and its auditors the Enquirer probably has more experience of examining council spending data than anybody.

What jumps out at us is that the AP1 payment data is so poor it seems to have been scribbled on the back of a cigarette packet 🔔 before being hurriedly entered in a spreadsheet.

One of the main giveaways here 🔔 is that of the addresses about 40% are partial or inaccurate and all have to be checked manually.

Certainly no data validation (checking data is in correct format and address exists as it is entered) has been done against a central address database 🔔 which is how IT systems usually work.

All of which is more than a little interesting.

We have submitted a couple of Freedom of Information requests which should help to reveal more details but there have been some slight problems with these.

Looking at the council documentation (such as it is) about the various housing programmes which the Enquirer believes relates this issue everything gets far more interesting and far more complex.

So complex we are still working on just representing the puzzle, let alone the answer, but will do our best to get this published as soon as we can.

As ever if you just happen to know what these odd data issues are all about please email Moley at Thanks!