1. Built environment
March 25, 2019

Why did we stop building at density? On planning, and why the CPRE is still wrong about brownfield

By John Myers

Today saw another CPRE report trumpeting about space for a million homes on brownfield sites. Sadly that’s nowhere near enough to end the housing crisis – and most of those sites are not in places like London with the worst shortage.

But there is plenty of room to build more homes by making existing suburbs more walkable and attractive. We’re just really bad at doing it. In his book Shut Out, Kevin Erdmann fears we have built nearly all the attractive, dense places there will ever be.

“Density has become a sort of natural resource, like gold. We have a fairly fixed amount of ‘reserves’ of high-density urban centres, which can only be expanded with great effort.”

Why did building friendly, walkable, dense, mid-rise – like Barcelona, Paris, Vienna, Bloomsbury or South Kensington – become so much harder in the 20th century?

The current English planning system has never been good at graceful densification. There is no credible evidence it can do so at sufficient scale.

It was never designed to: when it was created in 1947, London’s population was shrinking. New Towns were all the rage. Planners aimed to decant people out to new garden cities in the countryside.

Sadly, moving jobs around was much harder than they thought. Network and ‘agglomeration’ effects matter. New Towns like Milton Keynes, well-linked to vibrant existing cities, did much better than those like Skelmersdale.

Content from our partners
The key role of heat network integration in creating one of London’s most sustainable buildings
The role of green bonds in financing the urban energy transition
The need to grow London's EV infrastructure at speed and scale

But as the number of homeowners increased, it became ever harder to build new towns anywhere within reach of the best job opportunities. We never built as many New Towns as intended. And then places like London started growing in population again, and we have never kept up with demand.

To create more homes within existing towns and cities, you need lots of flats, not houses, on an unprecedented scale.

England, for example, has never built more than 100,000 flats a year. We would need over three times that to meet demand with flats alone.

Source: Holmans; MHCLG Live Tables 209 and 254.

And even that flatters our record, because the 1960s and ‘70s involved demolishing swathes of existing homes for council estates. As you can see, since WW2 we have never grown the net housing stock at the percentage rate of the 1830s, let alone the far higher rate of the 1930s:

Source: Neal Hudson of Residential Analysts.

Over three-quarters of London is just low-density sprawl: endless two-floor semi-detached houses.

Source: Emu Analytics.

It’s not just an English problem. Issi Romem of Trulia Research has shown how most US cities have given up on densifying their suburbs. These days they are mainly a sea of dark sprawling suburban blue, building little or no housing, like the San Francisco Bay Area.

Source: Issi Romem; BuildZoom.

There are a few exceptions: places like Miami or Dallas which, like Tokyo, have a majority of renters, who tend to be less NIMBYish; or places with no zoning, like Houston. See how much more orange Houston has today, where people are still building apartments in the suburbs?


Source: Issi Romem; BuildZoom.

Erdmann shows that. in the 1970s, as a share of GDP, the US invested three times more in building apartments than it does today. The US now spends four times more on home improvements than on building apartments, and even more on building houses:

Source: Kevin Erdmann, Shut Out, Fig. 7-9; BEA Table 5.4.5.

And those houses are mainly built in car-dependent sprawl, far from the best job opportunities.

So the problem is not just that we don’t build enough new flats in places awash with jobs, like London, Oxford, or San Francisco. We don’t build enough flats anywhere.

Most places in the US and England can only manage lots of flats with high-rise towers – the spots of bright red on Romem’s maps. That works for some, but less well for those on low incomes, and it inevitably causes a backlash: so much so that places like San Francisco, New York and London have failed to build enough housing for decades.

If we want to stop the housing crisis from worsening forever, we have two options: improve the planning system to build more flats, or reform the green belt to build more houses. No wonder brilliant academics like Paul Cheshire and Christian Hilber have been calling for green belt reform for so long.

You’d think CPRE would be fighting to upgrade the planning system so that lots more beautiful mansion blocks are built near stations in existing cities, reducing the pressure on the countryside. Sadly not. We build fewer than half the flats today that we did in the 1960s. A cynic might note that having a little green belt under threat is quite good for CPRE recruitment – but I like to think they’ve just been indecisive.

We can build attractive, dense places like Covent Garden again. The only question is how to solve the politics and change the system to do that today. The huge rise in the percentage of homeowners over the 20th century made it much harder.

Realistically, to build hundreds of thousands of beautiful new flats a year, making existing places better and more walkable by adding more homes, we need to be more creative.

That’s why our YIMBY campaigns are pushing for residents of a single street to be allowed to vote to choose a design code to allow more housing on their street. We think that’s one of the only ways to sustainably make existing places better and, ultimately, end the housing crisis.

John Myers is co-founder of YIMBY Alliance and London YIMBY, which campaigns to end the housing crisis with the support of local people.

This article is from the CityMetric archive: some formatting and images may not be present.
Websites in our network