How Pay vs Place works
Every figure is calculated from official open data. This page says where it comes from, how we work it out, and where it can mislead.
Areas covered
316 local authority areas in England and Wales are covered. Scotland and Northern Ireland are not, because their data comes from different sources.
Data sources
| Figure | Source | Period |
|---|---|---|
| House prices (median, by property type) | HM Land Registry Price Paid Data | Sales registered 28 August 2025 to 28 August 2026 |
| House price history (average price) | HM Land Registry UK House Price Index | 1995 to July 2026 |
| Rents | ONS Price Index of Private Rents | August 2026; history from 2015 |
| Pay | ONS Annual Survey of Hours and Earnings (Table 8.7, place of residence) | 2025 (provisional); history for 2005, 2010, 2015, 2020 |
| Crime | Home Office police recorded crime, published by ONS | year ending March 2026; change against year ending March 2025 |
All sources are published under the Open Government Licence v3.0. Pay and rent figures are in cash terms and are not adjusted for inflation.
How the figures are worked out
Price ÷ pay
The typical home price in an area divided by the median gross annual pay of full-time employees living there. A result of 8× means a typical home costs eight years of median pay.
Income needed to buy
Illustrative only. It assumes a 10% deposit and a lender willing to lend 4.5 times income, so the income needed is the price after the deposit, divided by 4.5. Real lending depends on your circumstances, and many people will be offered more or less.
Rent as a share of pay
A year of average rent divided by median gross pay. This is before tax, so the share of take-home pay is higher.
Where your salary goes further
The calculator uses the same rules: you can afford to buy where the price is within what a lender would typically offer against your income and your deposit, and to rent where rent is no more than 35% of your gross income (adjustable on the page).
Crime
Police recorded crimes per 1,000 residents, using ONS population estimates. Residential burglary is per 1,000 households. Police publish figures for community safety partnerships, which sometimes cover several council areas; where that happens we say so on the area page.
The ratings
The labels (Affordable, Fairly affordable, Expensive, Very expensive) are our own judgement, not an official standard. We chose the cut-offs to give a useful spread of areas:
| Rating | Buying: price ÷ pay | Renting: a year of rent ÷ pay |
|---|---|---|
| Affordable | under 5× | under 25% |
| Fairly affordable | 5× to 8× | 25% to 33% |
| Expensive | 8× to 11× | 33% to 40% |
| Very expensive | over 11× | over 40% |
Crime is rated against the England and Wales average: Low (under 75% of the average), About average (up to 115%), Above average (up to 160%), High (above that).
Limits you should know about
- Averages hide variation. Every area has expensive and cheap streets. Use the figures to shortlist, not to decide.
- Pay is where people live, not where they work. Commuters to London raise pay in some surrounding areas.
- Pay is for full-time employees only, and some area estimates are less reliable because they come from small samples. Where ONS withholds a figure as unreliable, we leave it out.
- Two house price measures. The tables use the median price of sales in the last 12 months. The history charts use the Land Registry's average price, which is calculated differently, so the two will not match exactly.
- City-centre crime is overstated per resident. Rates use the resident population, so areas with lots of visitors, students or nightlife look worse than they feel to residents.
- Recorded crime is not all crime. It depends on what is reported and how police record it. One year's change can reflect recording changes, not just a real trend. Crime for some Kent areas is missing because of problems with the force's systems.
- Council changes. Where councils have merged, history before the merger may be missing or left out.
- Written summaries are generated from the numbers on each page, so they are consistent but formulaic.
Spotted a mistake?
Please tell us, and say which page and figure.