In the mid noughties I lived for two stretches in St Albans, a commuter city nestled in the Hertfordshire green belt. The first time I lived and worked in the town centre, after leaving school. Some years later I lived there again, commuting into a job in London. It’s just 18 minutes from central London by train.
Each year there are 7.5 million entries and exits through the station gates. It’s a fantastic place for commuters to live, if you can afford the exorbitant prices.
It’s also a very easy place to live without a car if you are happy to walk and cycle about town.
Yet nine in ten residents owns a car, half drive to work, and almost 7 in 10 trips are taken by car overall. Train journeys account for 7.5% of overall trips, just over a tenth the number made by car.
The city is growing, and more housing is being built to accommodate that. Partly as a consequence, traffic is forecast to increase by 12% from 2014 to 2031.
My mind casts back to my time there whenever I read another think tank, politician or pundit promote the idea that we should build homes in the green belt near train stations.
The latest is a report by Paul Cheshire and Boyana Buyuklieva for the thinktank Centre for Cities, suggesting that millions of homes should be built within 800m of train stations that have a commute of 45 minutes or less into a big town or city.
Their assumption – oft stated but never evidenced – is that by building close to these train stations all the new residents will commute into work by train, so you get a sustainable settlement. Better that, they argue, than people commuting much longer distances as they’re priced out of the cities.
The problem is that this assumption is demonstrably false. The Royal Town Planning Institute looked at the evidence in 2015, using the 2011 Census for towns near London, and found that:
building one million homes around railway stations in the [London] Metropolitan green belt could result in between 3.96 and 7.45 million additional car journeys per week on roads which are already struggling with congestion and delays.
They studied the commuting patterns of five towns with fast commutes into central London – Hemel Hempstead, Watford, High Wycombe, Maidenhead and Bracknell. The findings were almost identical to St Albans:
only 7.4% of commuters actually travel to London by train on a regular basis, despite living within easy walking or cycling distance of a station. The majority of commuters (72%) instead travel by private vehicle, mostly driving to jobs within their hometown and to other places not in London.
There are two key factors that explain these findings, and the travel patterns in St Albans.
The first is that most people don’t work in central London and commute by train. It’s an unexamined premise fixed in the minds of professional middle class researchers, for whom this is their daily experience. Most people in commuter towns work in their local town in shops, the health service, local offices and factories, or the same in a neighbouring town, or from home.
The second factor is that commuting is only one of many reasons that people travel. Most journeys are taken to go shopping, visit friends, take children to school and so on. Not many of us will take a train into central London to do those things.
The Centre for Cities report compounds the error with other details. For example, the authors propose building homes at a density of 40 homes per hectare. This is too low to sustain public transport and local shops, which means people will have to walk or cycle (or much more likely, drive) into the town centre or train station. If we do build homes in commuter towns and villages we should be thinking about intelligently improving the housing density and rethinking local transport strategies to help people out of their cars and onto buses, bikes and foot.
The report suggests a way of capturing land value uplift for ‘social purposes’ but talks of investing that money in commuter rail (which only accounts for some 7% of journeys for work, never mind the other journey purposes) and in ‘roads and other transport’. There is no consideration given to whether investment could be focused on making a dent in the 70% of journeys taken by car, let alone an objective to stabilise or event reduce traffic levels in the area (thinking back to the predicted growth in St Albans).
Another fleetingly addressed question is whether there is capacity on the national rail network for all these new commuters. Transport for London undertook an assessment of the investment needs for London, accounting for commuters from elsewhere in the UK, in a report called 2050. This research did not assume millions of homes would be built near train stations for commuters into cities like London. TfL identified £1.3 trillion of investment over the next forty years, including upgrading many of the main lines and building three new main lines into London, just to cope with the growing demand. Read the excellent London Reconnections analysis for some idea of the challenges. What would happen if you added hundreds of thousands of extra rail passengers coming from green belt homes?
The Centre for Cities report, like most in this genre, covers other topics like ecology and local services in a similarly peremptory and feeble way.
There is a reason that the National Planning Policy Framework runs to 76 pages. The planning system has to balance a wide range of factors – housing, the local economy, transport, agriculture, ecology, health, communications, good design, flooding, schools and other local services, sustaining town centres, to mention just a few! – and to draw these together in a way that promotes sustainable development. Applying this framework, local planning authorities rarely propose large scale homebuilding in their green belt near train stations.
By contrast, calls to ‘build homes near train stations in the green belt’ rarely reach beyond housing and some half-formed thoughts about local infrastructure.
The very idea is essentially a sop to environmentalists – if you want to build in the green belt you suggest doing it near train stations to ‘tick the box’ on the environment.
None of this is to necessarily rule out the idea of building homes near train stations. There may well be places where it’s a good idea, and can be well planned. New homes in St Albans, Watford and Bracknell could bring investment to make roads safer for walking and cycling and to improve bus services; homes could be built in locations, and at densities, and with street patterns, that discourage car usage; and so on.
But the ‘near train stations’ idea is just the wrong place to start. The question to explore is: are there locations in the green belt where you could build lots of homes and where you could plan them in a way that promotes sustainable development? That would consider the scope to reduce local traffic and congestion, enhance the local ecology, improve the viability of local town centres, and so on. It may be that ‘near this train station’ turns out to be a good location, but it equally may be that current train station locations are a minor consideration.
Every time someone starts with the ‘near train stations’ premise, they are – to my mind – signalling that they aren’t too interested in sustainable development, and just want to tick a box so they can focus on their main interest – huge housebuilding programmes in the green belt.
At a time when most acknowledge the climate emergency, isn’t it about time think tanks stopped making the simplistic call to build homes near train stations in the green belt?
Nothing important is simple is it? Does this analysis also apply to outer London boroughs like Enfield?
Interesting question. I can imagine it’s similar where the local bus network isn’t great and it’s not nice/convenient to walk and cycle. But it should be much easier to change that in Enfield than somewhere like Bayford or Broxbourne, which happen to have train stations.
The RTPI study makes a classic error – it looks at the consequence one thing without evaluating alternatives. The people who might live in the green belt are currently living somewhere and are probably commuting – mosyly by car. So what matters is the difference. Now if they move from Islington to Broxbourne there will almost certainly be more road travel even if they keep the same jobs in inner London. Might not be true if they moved from Maldon.
You are right to say that a train station isn’t sufficient. But I also fail to see how it isn’t necessary.
Part of this is the broken state of public transport in this country. What you need is frequent service on a strong network. What you get is a bunch of bus companies trying to backstab each other and their customers. The train network remains better than that but has also degraded compared to what it should be (e.g. in other European countries), and it seems that even you have the idea that trains are only for American-style commutes ‘car park to city centre, 9-5’. It doesn’t have to be that bad, that’s a conscious choice made by policy makers over the 20th century and beyond.
Insofar as a sustainable development needs to have a train station (or more than one), it also needs to have all-day frequent service that fits into an all-day frequent network of public transport. That will make it useful for many more trips than commutes. Furthermore, the station needs to integrate into the new development’s centre, with intense land uses immediately around it, connecting buses, easy permeability by foot and cycle, and cycle parking. This is anathema to typical British developers and planners, who think too much like Americans, and want to stick massive car parks all around stations that destroy all their potential. As you note, many jobs and daily destinations are not in city centres and are scattered around the countryside. However, this also comes down to poor planning. Had those destinations been near train stations / public transport nodes, then the frequent, all-day network of public transport would serve them just as well as any city centre location.
With regard to local transport patterns you are absolutely right that it requires sufficient housing density, shopping amenities and street patterns that promote walking, cycling and local bus service while discouraging the use of cars for such trips. Again, developers and planners have no idea how to do that for the most part, and little interest in learning. Without explicit guidance and policy requirements, they’d rather just repeat the same old broken patterns that have resulted in the mess we currently know.
On the other hand, the people who already oppose development on the green belt are going to doubly oppose building at those density levels you indicated, because too many people don’t understand that low density development leads to sprawl and even more green belt destruction.
So what’s the way forward? Most larger development in this country seems to proceed at the convenience and whim of local landowners, who may or may not have parcels that are well-located, and who don’t care either way. They have political backing already by virtue of wealth and connections.
The ‘near train stations’ concept is overly simplistic, yet it is also easily understood. If you want to build a political coalition of people in favour of sustainable development as replacement for unsustainable development, then it can’t all be difficult concepts.
Ideally a new development would put all housing within easy cycling distance of a rail station and a shopping centre. Plus bus services to take people to at least one bigger shopping centre and to places of employment. The key issue here is distance so we might call the key concept the Compact Community or the Liveable Community. I see this as an extension of the Low Traffic Neighbourhood that many of us support.
Hi Tom
In High Wycombe, commuters are moving out of London to buy our relatively cheaper housing (forcing young people of my children’s age to move as far out as Bicester because they cant afford the housing). If you design for commuting to London, you are likely to force local people out, often causing greater commuting distances to be driven by others.
To add to the points in your article for High Wycombe. It is very hilly and buses often run in the valleys. That is why some of us are forced to use our cars or why there are so many taxi journeys. Buses dont go where we want them to go, or they are too slow, too expensive for a family and too unreliable.
There is no sensible rail connection to Reading or Maidenhead from High Wycombe, where lots of people work, but we do now have one to Oxford.
Children very often do not attend their local school. There is not necessarily any transport to get to schools, so they are driven there. This is a politcal decision on catchment areas.
Partly because of the local topograhy, retail parks have been built away from the town centre, as in many places. These are designed to be accessed by car. Same for one of the cinemas and the sports centre. We do have a cinema and 10 pin bowling alley in the Town Centre. That was factored in when the last shopping centre was built.
Amenities should be sited in town centres and other areas of dense housing, with public transport running between. Perhaps that is the redesign that will keep town centres alive, replacing empty retail outlets with cinemas and sports centres etc.?
We also need far more public transport and at a very heavily subsidised rate. This would have to include express services to make bus commuting a realistic proposition. If you own a car, there is little incentive not to use it at the moment. If road tax was scrapped and transferred to the cost of petrol, this might help. Driving a car is still too cheap to encourage people to travel another way. There are not enough “other ways”.
The move to online shopping should help reduce car journeys and could be eco friendly if delivery vehices were “green” and journeys shared (e.g. deliveries to certain post codes were made on set days).