It’s obvious that the price of a home varies massively from place to place, but simply comparing the average cost of properties sold can hide some major differences – particularly in the size of each home.
The average cost of a home in London is £520,000 whilst in the North East it’s £160,000, but the place in London is also much more likely to be smaller. What’s much more important is the price per square metre – how much space you get for your money. This varies wildly across the country, as shown in this brand new map we’ve produced:
Map built by Freddie Poser
An MSOA – one of the divided areas shown in the map – is a statistical unit used by the ONS to split the country up into roughly equally inhabited areas, each having a population of about 8,000 people. These boundaries are from the 2021 census.
This data was collected by researchers Bin Chi, Adam Dennett, Thomas Oléron-Evans and Robin Morphet, who matched up the price people actually paid for properties (from HM Land Registry) with estimates of their floor area from Energy Performance Certificates. This process isn’t perfect – matching addresses is surprisingly difficult – but this gives us the best public data on how much space costs in different places.
So what does this mean for housing policy? Well the map shows us a few things that we know from theoretical spatial economics. The obvious lesson is clearly ‘we should build most where prices are highest’, but the picture is more complex than that.
1. Our cities are too small
Tim Leunig and Henry Overman have a fantastic paper on this, but the TL;DR is that the planning system has constrained the size of our most productive cities (think London, Cambridge, York, etc…) well below where they naturally ‘want’ to be. These places have good jobs with high wages, so people are willing to pay a significant premium to live there. Building more homes here is the single best thing we can do both to solve the housing crisis as well as boost economic growth. Letting people move to high wage areas like these means they have access to better, higher paying jobs.
This graph from the Resolution Foundation shows just how badly land is misallocated in London and the South East.
Land with permission for homes in London is worth 7x more than the same hectare used for industrial uses and 14,000x more than if it were agricultural. Not only are these values much higher in London (as we would expect), but the ratio of residential to industrial value is much higher.
In the East of England, a hectare of land worth £500,000 in industrial use becomes worth 2.4 times more with permission for homes. In London that ratio is 3 times more. The value of a hectare with homes, £36 million, is 7.2x greater than the value if the land is limited only to industrial uses.
That’s not to say there’s no demand for industrial land in London: it is the second most valuable usage type on our chart, after London residential, narrowly beating out residential use in the wider South East. But these figures show us clearly: that there is too little land for homes so people are willing to pay more for land for homes than for other uses.
2. Some areas are expensive, but building there is tricky
Leunig and Overman also make another key point that is often missing from the housing debate: some areas are expensive, but building more homes there may damage the amenity value that makes people want to live there. Building more homes in London will generally be value-creating: it creates more economic value through providing access to good jobs. Building more homes actually grows the economy, because of the increased agglomeration effects.
But consider a small town, nestled in a National Park or beautiful landscape: here prices per square metre may be very high indeed. The Lake District is immediately obvious on the map for its expensive prices, but if we built thousands and thousands of unattractive homes around Windermere it’s unlikely we’d see high tech industry spring up. Instead we might destroy economic value by tarnishing the very natural beauty of the area which creates that value.
3. Beware high capital stock
The Resolution Foundation’s Economy 2030 Report makes a key point that is easy to overlook. In some areas of the country, like Sunderland, the high productivity observed derives not from easily expandable agglomeration effects of knowledge workers, but through high existing capital stocks in specialised factories. If we build homes in places like this, we cannot assume that people who move there are easily able to get those same, higher than average productivity, jobs. This is because those jobs need significant capital spend to support them. Simply adding homes here won’t overcome the other serious hurdles to fixing growth in these sectors.
Conclusion
Location matters. To solve the housing crisis we need to pay attention to economic geography and build in the highest productivity places. This is not just because that’s where people want to live (though they demonstrably do), but because building homes in places like Cambridge, Oxford, York and London is the closest thing we have to a surefire magic growth serum.