Ben Southwood is Head of Housing, Transport and Urban Space at Policy Exchange.
The Government means to reform planning in order to allow more houses to be built. The Local Government Association tells us the Government is wrong.
Changes to the planning system are unnecessary, they argue, for it is developers who hold back housing delivery, not planners.
They tell us that around 90 per cent of planning applications are being approved – far from being in thrall to NIMBYs, planners seem to say yes to virtually every application they receive. The real problem, the LGA says, is that once developers have got permission to build, they frequently fail to do so: permission for hundreds of thousands of houses has been given without those houses ever having been built.
But while these arguments are superficially persuasive, they are fundamentally mistaken. Planning as it currently operates is clearly an obstacle to development, and land banking is a consequence of the existing system.
In the Planning White Paper released today, the Government announces its intention to bring in a zoning system which increases the amount of development that can be done without a planning application, similar to the system proposed by Jack Airey (then PX, now No10) in Rethinking the Planning System for the 21st Century.
Under such a system, households automatically have the right to develop land they own, assuming that it is within various basic rules (e.g. about height, overshadowing, safety, and aesthetics). This would be the biggest change to the system in seventy years.
As things stand, there are very significant restrictions on development. We can see this in two main ways: firstly, land with planning permission is worth dramatically more than land without planning permission. Secondly, the cost of building houses in places like London is far less than the price those houses fetch on the market – typically around a quarter – suggesting that something beyond build costs is preventing developers from building more homes.
This argument would be uncontroversial among most economists. However one popular response, most recently made by the Local Government Association, is that the shortfall in new homes is mostly not due to the planning system in general, and specifically not due to planning departments at local planning authorities, but due to the behaviour of property developers. They give two arguments on this point: the percentage of applications approved, and the problem of ‘land banking’.
With nine in ten planning applications approved by councils, and more than a million homes given planning permission in the last decade not yet built, planning is not the problem.
While planning officials do the job they are given as well as possible, and the LGA is correct that they have faced tough budgetary constraints, these facts do not exonerate the planning system as a whole in leading to the undersupply of housing we experience.
It is true that somewhere between eight and nine in every ten planning applications that get to the final stage are approved. But this statistic is misleading, because the cost and restrictiveness of the system means that applications are only made if the applicant expects them to be approved. This cannot tell us about the overall restrictiveness of the system, since it leaves out all the applications that people might want to make but do not because they expect them to be rejected.
If 95 per cent of criminal cases brought to trial got a conviction, that does not tell us much about whether a high percentage of people who actually committed crimes were being convicted – since it doesn’t include those cases that never see trial due to a lack of sufficient evidence.
Most potential planning applications never make it to that final stage, or indeed any stage. Were there no planning system at all, housing starts would be dramatically higher. A more relevant statistic is the percent of potential housing starts that are allowed by the planning system. One estimate, based on Ian Mulheirn’s estimates for the elasticity of prices to supply, would suggest that in London alone we would build 1.5 million more houses to meet pent-up demand, on top of the 600,000-700,000 we are likely to build in the next ten years under the current system.
In other words, the system is building around 30 per cent of the housing that would be demanded in a system where only supply and demand determined what was built.
The true figure demanded would be more like five million, if London was to grow to home a similar share of the national population as cities like Copenhagen and Dublin do. This suggests that planning is allowing just ten per cent of the housing that would be demanded in a freer system.
The most recent review found that around six per cent of annually granted permissions had not been started within 12 months of being granted. Over time, these six per cents build up. But why is land banking a feature of planning systems like ours?
The reason is that ‘banking’ makes sense for assets whose supply is scarce and uncertain. Property developers have a number of different factors of production that they combine to build houses: things like building equipment, workers, and land to build on. All are needed to build houses, but while the supply of things like equipment and workers is relatively guaranteed for the near future (because they own or lease the equipment, and have employment contracts with the workers or know they can hire them if they need them), the supply of land you can build on is less guaranteed – unless they “bank” it.
The alternative to “banking” developable land would be to risk being in a situation where they have paid for equipment and workers in advance but have no land on which to use them.
As the risk, cost, and time invested in getting hold of land and planning permissions rises, the amount of land that it’s rational to bank rises too.
This difficulty, complexity, and cost also leads to a situation where there is no alternative to big housebuilder development. Britain has one of the lowest self-build rates in the developed world: under ten per cent, compared to over 50 per cent in Germany, over 40 per cent in Japan, and nearly 30 per cent in France. SME builders and self builders have little incentive or need to land bank, but cannot negotiate a system where it is essential.
Thus while the statistic around permissions is misleading, the LGA’s concerns about land banking are understandable. But land banking is an entirely rational, and in itself reasonable, response to a system that makes land banking necessary. If the LGA is worried about land banking, they might be pleased with what these reforms deliver.