The new Government research uses machine learning, or artificial intelligence, to look at the characteristics of different towns across the UK and how they correlate with house prices. They found that some characteristics were far more likely to predict high prices than others, and they are not all as obvious as you might think.
What makes a town’s properties more pricey
The research, which can be found here, looked at 23 different statistics surrounding Britain’s towns and cities, as various as whether the town was on the coast, if it had a university, and how quickly the population had grown in recent years.
The most important characteristics when it came to predicting higher house prices were as follows:
-
Distance to London
By far and away the biggest predictor of high prices is proximity to the UK capital. The government research found that, for every 50 kilometres further away a town was from London, the average price of property dropped by £50,000.
It is a startling figure but only applies to towns within a 200km radius of London – after that, the link begins to weaken and further decreases in the distance have little effect on house prices. For context, both Cardiff and Lincoln are 200km from London, so this effect covers a wider area than you might think.
-
The types of job that people do
The second-biggest correlation between a town’s characteristics and the house prices is the types of job that the inhabitants typically do. For each 10 percentage point increase in inhabitants doing skilled ‘Level 4’ jobs (which typically require a degree) house prices increased by £20,000. For each 10 percentage point increase in inhabitants doing ‘Level 2’ jobs which are typically clerical, retailing or machine operation, house prices decreased by a similar amount.
-
Income deprivation
A measure known as income deprivation is the third-biggest reason why one town’s properties are worth less than another. This measure shows the percentage of people in a town experiencing deprivation from low income.
The average price of a property reduces by around £10,000 for every 5 percentage point increase in deprivation level.
And the factors that don’t really matter
While these three factors make a huge difference to property prices, there are other factors that make very little.
These include:
-
Whether a town is coastal
Despite our fondness for a sea view, the study found that being by the seaside has only a tiny, but positive, effect on house prices.
-
Whether a town has a university
The presence or absence of a seat of higher education had no impact at all on house prices in a town or city
-
The distance to the nearest city
While distance to London had a huge impact on the price of homes, distance to any other city appeared to have very little. This was only the twelfth (out of 23) most important factor affecting house price.
Remote working is the elephant in the house price game
These findings look conclusive, but one major event could turn them all on their head. The towns and cities analysis was only published this week, but used 2019 data. Since then, the pandemic has prompted an exodus towards the countryside due to increased home working, and it is still too early to say whether this will have a longterm effect on the desirability of different towns and cities.
With London house prices rising at their fastest rate since 2016 , there is some evidence that the pandemic effect is already moderating, but it is still possible that a similar study carried out in two years’ time will show a very different picture.
In the meantime, if you’re looking for somewhere new to live and wonder how prices will perform, looking at these town characteristics might give you an idea of why some towns are more expensive than others.
The ONS report also includes a useful tool that allows you to see the average house price in every town, and which of these characteristics makes the biggest difference to prices. Together with your own research on areas to live in, or to invest in residential property, it provides an entirely new way to pick a property ‘hotspot’.