Unsupported browser

For a better experience please update your browser to its latest version.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We use cookies to personalise your experience; learn more in our Privacy and Cookie Policy. You can opt out of some cookies by adjusting your browser settings; see the cookie policy for details. By using this site, you agree to our use of cookies.

We used big data to predict house prices – and this is what we found

  • 1 Comment

Thorough analysis of sales in urban areas of England revealed some predictable but also some surprising facts, writes Nicholas Boys Smith

The value of place is simultaneously the most discussed and the least understood of all things. House prices used to be the leitmotiv of a million clichéd dinner parties (no more, I think) but what we mean by the value of an actual neighbourhood and what drives that value is not only little discussed, but it has, until recently, barely even been studied.

Create Streets’ new report, Beyond Location, tries to answer this question. We have taken advantage of new techniques for analysing values as well as the ’big data’ revolution. We have conducted a uniquely wide, data-rich analysis of every 2016 property sale in six English cities (London, Birmingham, Manchester, Leeds, Liverpool and Newcastle). This has used open datasets to compute basic urban characteristics, such as street network connectivity, population density, amount of greenery and availability of different transport modes. The point of the analysis was not to investigate them separately but together. The findings are part predictable, part surprising. One thing they do show is that urban form really does matter in understanding value. Our models for urban form can predict up to 74 per cent of the official UK poverty index – the Index of Multiple Deprivation – and up to 54 per cent of sales values. Here are four key conclusions.

Price vs iden

Price vs iden

More greenery is not always a good thing. The immediate presence of attractive greenery or high-quality parks can add huge value in many situations. However, at the city-wide level, the presence of more greenery can be associated with lower as well as higher value. What it is and how it is managed really matters. For example, in London, a home closer than average to a high-quality park costs, on average, 11 per cent (or £51,000) more than one that is not, holding everything else equal. However, in Liverpool, a home located closer than average to a high-quality park is worth, on average, 7 per cent (or £7,760) less.

Land use and form also matters. We found significant relationships across the six cities between urban form and deprivation and value. Areas of high population and areas with more unbuilt land are less valuable and often associated with more deprived communities. This might be partly due to the history of post-war building but, after 30 years of Right to Buy, most people who can afford to choose, continue to avoid this type of urban pattern.

The heritage premium is more important than the new build premium. In every city studied, proximity to a listed building was associated with more additional value than the premium associated with a newly built home. A home closer than average to a listed building in London is worth 10.3 per cent (or £49,770) more than one that isn’t, holding everything else equal. The equivalent new build premium is only £8,795.

Diversity is valuable. Areas with more diversity of house types suffer from less deprivation. Areas with a more diverse offering of transport and amenities are normally worth more, other things being equal. Above average amenity diversity is associated with additional value in all cities studied.



Value is a fraught term. Extra value is not always good – certainly not for everyone. In globally successful cities, spiralling house prices are forcing out existing communities. There are ‘sorting effects’ where the better off out-compete the less well-off for the best places. The ultimate aim of this study therefore is to help developers to build and planners to permit more good places by understanding human preferences more richly. One thing is certain though. When it comes to understanding, and predicting, economic and social value, design really matters.

Nicholas Boys Smith is director of Create Streets where Alessandro Venerandi is a researcher and urban designer. Beyond Location was sponsored by U+I plc and Gascoyne Holdings and is available from www.createstreets.com

  • 1 Comment

Readers' comments (1)

  • Chris Roche

    Here is some more big and shocking Data.
    25 years ago the ratio of Part 3 students salary to first home cost was circa 1:3. Today it is circa 1:17. In 25 years time following the same relative increases in salary and property prices it will be 1: 84.
    The historic evidence is clear, proven, and tested for 3 years running with Part 3 students at Westminster University and demonstrates a doubling in Part 3 students salaries over 25 years compared to a 1000% i.e. 10 times increase in property values. The ratio of 1.17 relates to the London Borough of Wandsworth. In Kensington and Chelsea it was announced this week that the average salary is £150,000 and the ratio to average house price is already 1:34.5. Clearly future Architects will be unable to own property - this is not an irony, it is a travesty. The economic reality is we are educating too many architects, at too high a cost to the students themselves, and to society at large creating an unsustainable culture which drives down fees, salaries, and ultimately the perceived value of architects who are literally and figuratively "two a penny". It is hard to lay the blame at the door of Universities, Architecture is popular, and they are as entrenched now in Capitalism as any corporate business. The RIBA couldn't care less - it is in the interest of large employers to have an over-supply of graduates as it suppresses salaries and allows increasingly low and self-defeating fee bids. So what is the answer? Several German States have taken the brave step of reducing University intakes to demonstrably sustainable numbers.
    Only saying.

    Chris Roche ARB
    Founder 11.04 Architects / Author

    Unsuitable or offensive? Report this comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions.

Links may be included in your comments but HTML is not permitted.