Past experience suggests that the new towns policy can accelerate development in certain areas. Government should combine it with wider national reform to make the planning system more efficient.
The Government has announced that it is planning a new generation of new towns. It says that they could deliver ‘hundreds of thousands’ of homes through developments of at least 10,000 new homes each, 40 per cent of which will be ‘affordable’.
While the Government hasn’t put a timeline by which a certain number of houses should be delivered by new towns, the implication is that it sees these projects as an important part of delivering the 1.5 million homes promised in their manifesto.
But what will their contribution be? This blog, using newly available data on housebuilding since the Second World War, looks back at the last time the UK built new towns and considers what the findings mean for the Government’s ambitions today.
In England, eleven new towns were initiated by the Labour Government between 1946 and 1950 following the 1946 New Towns Act. Five more were started under Conservative government between 1961 and 1964. And the final six were led by the Labour government between 1967 and 1970. Each new town was managed by an individual development corporation, answerable to central government, that was responsible for masterplanning, land assembly and infrastructure development, as well as attracting private investment.
There was also a programme of 36 ‘Expanded Towns’, started in 1952 by the then Conservative Government that preferred urban expansions managed by large urban authorities to the centrally organised new towns projects.
Between 1947 and 1993, New Town Development Corporations (NTDCs) in England coordinated the building of 307,000 houses. This includes 231,000 built by public authorities and 76,000 by private housebuilders operating within NTDC designated land. By comparison, 880,000 houses were built in Greater London during the same period.
This represents 3.3 per cent of all housebuilding in the 40 years after the New Towns Act. As Figure 1 shows, this peaked in the 1970s, with nearly 5 per cent of houses in that decade delivered by NTDCs.
That these are relatively small percentages shouldn’t really be surprising. Total English housing stock increased from 11.6 million in 1950 to 20 million in 1993. Splitting all of this between 20 new settlements would have built Bristol 20 times over.
While the overall number of homes built was modest compared to total housebuilding, new towns built at a faster speed than other parts of the country. Figure 2 shows that the average housebuilding rate (houses built in a given year divided by housing stock in the previous year) for districts ever containing new towns outstripped the housebuilding rate of other districts both during and after the period that the NTDC was coordinating building there.
Interestingly, the same is true for places with expanded towns, despite their having substantial existing stock. Through the 1960s, these places matched the new towns’ housebuilding rate. While data on expanded town housebuilding separated from total building in their respective local authorities doesn’t exist, the implication of the above data is that they, as much as the new towns, significantly changed the conditions for housebuilding in their towns.
Of all the houses built by NTDCs, 60 per cent were located in the Greater South East (GSE). The average New Town in the GSE built 5,000 more homes in the first 20 years after its designation, compared to a New Town in the rest of England (1,500 more if you exclude Milton Keynes).
As Figure 4 shows, local authority districts with new and expanded towns in the Greater South East also had a higher housebuilding rate than their comparators in the rest of England throughout the post-war period.
When people talk about new towns, it’s often about Milton Keynes. As the graph below shows, Milton Keynes was a relative outlier – building more, and more quickly, than any other NTDC. The average NTDC built less than half the total Milton Keynes did, despite Milton Keynes building its first house in 1970. Milton Keynes shows the art of the possible but history suggests that it was also somewhat of an outlier.
Here are some key takeaways.
The Government should follow the example of their post-war forbears and concentrate the new towns and urban extensions where need is highest. These places are most likely to grow quickly and continue growing, even after the main work of the development corporation is done.
The Government has already indicated that they expect most of the ‘new towns’ to actually be urban extensions. The evidence on the success of the expanded towns programme suggests that this is the right path.
There are international examples that show it can be done from a relative standing start – for example, the Dutch 10-year VINEX programme which ran from 1996-2005 delivered 450,000 homes across 90 schemes – but the Government should not underestimate the challenges in achieving this. 300 extra planners certainly won’t be enough to build the required capacity and expertise in masterplanning, land assembly and project management in new development corporations and local authorities.
Ultimately, new towns will only be one of the tools the Government will need to use to increase housebuilding and improve planning outcomes. Increased housing targets that make it more challenging for local authorities to prevent housebuilding are an important start, but more will be needed.
Increasing the use of Local Development Orders, and Supplementary Planning Documents, would help push numbers up further by removing uncertainties from the planning process. Eventually, the Government should aim to move from using these piecemeal de facto zoning tools to a more comprehensive UK-wide flexible zoning system.
Centre for Cities has digitised UK Government statistical reports on housebuilding from 1945 to 2001. This data reports the number of houses built by local government, private sector, and housing associations in each local authority district in England. There were roughly 1,200 local authority districts before 1973, and roughly 300 after that. The pre-1973 data has been reconciled to 1971 local authority boundaries, and the post-1973 data to 2023 local authority boundaries. We are able to construct a continuous series from 1945 to 2022 using 1981 county boundaries. Where we report data at district level, there is a break in 1974 due to country-wide local government reorganisation.
Contained in this data is information on housebuilding by the New Town Development Corporations (NTDCs), including private housebuilding within their designated areas. This gives us unique information on the numbers coordinated directly by the NTDCs. Our data omits the Central Lancashire New Town, started 1970. Our data also counts the Shropshire new towns, Dawley (1963) and Telford (1968), as a single NTDC for geography consistency reasons.
List of new towns: Aycliffe; Peterlee; Washington; Runcorn; Skelmersdale; Warrington; Corby; Northampton; Telford & Dawley (see note above); Redditch; Peterborough; Basildon; Harlow; Hatfield; Hemel Hempstead; Stevenage; Welwyn Garden City; Crawley; Bracknell; Milton Keynes.
List of expanded towns: Cramlington; Killingworth; Widnes; Gainsborough; Macclesfield; Winsford; Grantham; Rugeley; Cannock; Brownhills; Tamworth; Daventry; King’s Lynn; Thetford; Mildenhall; Bury St. Edmunds; Melford; Haverhill; Huntingdon; St. Neots; Wellingborough; Sandy; Sudbury; Letchworth; Braintree; Witham; Aylesbury; Banbury; Swindon; Andover; Basingstoke; Ashford; Hastings; Droitwich; Weston-Super-Mare; Bodmin; Plymouth.
Leave a comment
Be the first to add a comment.