The mansion tax map: where the money comes from

We’ve modelled the impact of the “mansion tax” by analysing land registry data on every property transaction since 1995. This lets us estimate how much each postcode and Parliamentary constituency will pay. It’s an approximate and lower-bound estimate – see methodology details below.

This interactive map shows the results. It also shows English and Welsh data¹ on the current council tax bands, median house prices, and the changes in house prices since 1995 (which demonstrate quite how out of date council tax is):²

You can view the map fullscreen here. It’s important to stress that this is only showing postcodes – the markers on the map are at the centre³ of the postcode and do not represent individual properties.

This chart shows our estimate of the revenue from the mansion tax for each constituency. You’ll be unsurprised to see that most affected properties are in London (you can move the mouse over individual constituencies to see full details).

My view is that the tax is good policy. There will be inefficiencies and unfairnesses, as with all taxes, but the basic concept is right: ending the anomaly that someone in a £10m home pays the same council tax as someone in a £1m home – and only twice the council tax of someone in a £400k home. I wrote more about my views here.

Methodology and limitations

Our methodology was straightforward: use the Land Registry price paid database to find every property transaction since 1995 that looks residential⁴, and uprate prices by the change in median house prices in each constituency (whilst also displaying council tax data).

The code for the analysis and webapp is available on our GitHub.

It would be technically straightforward for our analysis and map to work at the level of individual properties, but we were uncomfortable with the privacy implications (although there are many property price websites that let you see “price paid” data for specific properties). We therefore limit the map to postcodes (which has the side benefit of making it load and respond much faster).

Our very simple approach has obvious limitations: inflation is higher in some areas within a constituency than others, it doesn’t take account of improvements etc to properties, it ignores properties that haven’t been sold since 1995, and in some cases (particularly high value property) the price is hidden. This means our approach is generating what is very much a lower-bound estimate of the actual static revenue from the tax. The total estimated revenue is around £420m.

This is much less than the OBR’s static revenue estimate of £600m – that will be because they used more sophisticated approaches, for example more granular house price inflation corrections, better detection of residential property, inclusion of properties that aren’t in the transaction data. The OBR then adjusts the static estimate to reflect behavioural effects (clustering below thresholds) and losses to other taxes – this brings their total estimated revenue to £400m. The fact their post-behavioural adjustment estimate is so close to our static estimate is a coincidence.

So our figures, and our map, is missing properties and undervaluing properties, which together amounts to an error of 50% – that’s pretty significant. However we expect that the overall picture presented is an accurate reflection of the areas from where the mansion tax revenues will come.

---

Contains HM Land Registry data © Crown copyright and database right 2025. This data is licensed under the Open Government Licence v3.0.