Don’t count on brownfield: it won’t solve the housing crisis alone
12 June 2014
SUGGESTED
Regulation
Uncategorized
Trade, Development, and Immigration
4 November 2024
In today’s Mansion House speech, George Osborne is expected to outline new plans to encourage housing development on brownfield sites. The speech follows in the wake of an IMF report, which urged the UK government to relax planning rules that constrain both brownfield and greenfield development in order to sort out the grave imbalances in the housing market. Unfortunately, whatever brownfield incentives Osborne outlines today, they are unlikely to be enough.
Britain’s housing market imbalances have been building for a long time. By international standards, British housing construction levels have been extremely low for over three decades, with only between 30 and 40 new housing units built for every 10,000 inhabitants per year. But after 2007, construction fell to a new low, even for the UK. And while aggregate figures are returning to their inadequate pre-recession levels, this masks an ongoing near-standstill in areas where demand is highest.
The general problem in housing is that there is almost no overlap between the economically sensible and the politically palatable. Virtually every study on the determinants of house prices that has ever been conducted concludes that inflexible planning restrictions, especially growth boundaries like green belts, are key price drivers; but those blanket development bans are also protected by well-coordinated vested interests, to which the government does not dare stand up. This is where brownfield sites come in. Unlike greenfield land, brownfield land has few organised defenders. Could we simply bypass the Nimby lobby by unlocking the potential of these sites?
At its best, brownfield redevelopment can be a win-win situation. Derelict industrial sites can have negative knock-on effects on their surrounding areas, in which case redevelopment would not just increase housing supply but also remove a source of blight. They are often within or close to existing settlements, and thus connected to an existing infrastructure network. And yet brownfield redevelopment is being held back. Decontamination can be a legal minefield, involving unpredictable long-term liabilities for developers. Clean-up costs are partially tax-deductible, but the tax situation is so complex that developers cannot readily assess to what extent they will be able to make use of those options. If Osborne’s plans can address some of those issues, they will make a positive contribution to the housing situation.
But we should not exaggerate the potential for brownfield site redevelopment. It can be a sensible complement, but it cannot be a substitute for a wider liberalisation of the planning system. We need a lot more housing on both brownfield and greenfield land. Let’s have a look at the numbers.
The Department for Communities and Local Government estimates that, if every square inch of redevelopable brownfield land was used for housing, it would be enough to fit 1.5m new homes. The figures are not perfectly reliable and are subject to frequent revisions, with previous estimates indicating much smaller numbers. But let’s give the highest estimate the benefit of the doubt. How should that figure be interpreted?
First, not every development that is physically possible is also economically viable. For some of those sites, decontamination costs will, for the time being, be prohibitively high. Greater legal certainty could make those costs more predictable and insurable, but they still have to be incurred, of course. Secondly, not all of these sites are in places where people want to live. Less than a third of the brownfield land area is in London and the South East, and even this is still a very high level of aggregation. Thirdly, the 1.5m figure is based on the size of recently developed dwellings, which are the smallest that can be found anywhere in Europe. If we allowed for just a little bit more space and comfort, the numbers would dwindle. Fourth, development is already heavily biased in favour of brownfield sites, with a brownfield-to-greenfield ratio of 2:1.
So, yes, of course we should make the best possible use of previously developed land. But we should not pretend there is a limitless supply of such land in the right places. If brownfield redevelopment was really such an obvious no-brainer, developers would have snatched up that profit opportunity long ago.
What the whole discussion really shows us is the inadequacy of rigid, bureaucratic forms of land categorisation. An ex-industrial site that is used as a car park is not really ‘derelict’, and green-belt land that is used for high-intensity agriculture is hardly pristine countryside. We should judge land by its actual environmental and amenity value, not by a box-ticking template from the 1940s.
This article was originally published by City AM.
Britain’s housing market imbalances have been building for a long time. By international standards, British housing construction levels have been extremely low for over three decades, with only between 30 and 40 new housing units built for every 10,000 inhabitants per year. But after 2007, construction fell to a new low, even for the UK. And while aggregate figures are returning to their inadequate pre-recession levels, this masks an ongoing near-standstill in areas where demand is highest.
The general problem in housing is that there is almost no overlap between the economically sensible and the politically palatable. Virtually every study on the determinants of house prices that has ever been conducted concludes that inflexible planning restrictions, especially growth boundaries like green belts, are key price drivers; but those blanket development bans are also protected by well-coordinated vested interests, to which the government does not dare stand up. This is where brownfield sites come in. Unlike greenfield land, brownfield land has few organised defenders. Could we simply bypass the Nimby lobby by unlocking the potential of these sites?
At its best, brownfield redevelopment can be a win-win situation. Derelict industrial sites can have negative knock-on effects on their surrounding areas, in which case redevelopment would not just increase housing supply but also remove a source of blight. They are often within or close to existing settlements, and thus connected to an existing infrastructure network. And yet brownfield redevelopment is being held back. Decontamination can be a legal minefield, involving unpredictable long-term liabilities for developers. Clean-up costs are partially tax-deductible, but the tax situation is so complex that developers cannot readily assess to what extent they will be able to make use of those options. If Osborne’s plans can address some of those issues, they will make a positive contribution to the housing situation.
But we should not exaggerate the potential for brownfield site redevelopment. It can be a sensible complement, but it cannot be a substitute for a wider liberalisation of the planning system. We need a lot more housing on both brownfield and greenfield land. Let’s have a look at the numbers.
The Department for Communities and Local Government estimates that, if every square inch of redevelopable brownfield land was used for housing, it would be enough to fit 1.5m new homes. The figures are not perfectly reliable and are subject to frequent revisions, with previous estimates indicating much smaller numbers. But let’s give the highest estimate the benefit of the doubt. How should that figure be interpreted?
First, not every development that is physically possible is also economically viable. For some of those sites, decontamination costs will, for the time being, be prohibitively high. Greater legal certainty could make those costs more predictable and insurable, but they still have to be incurred, of course. Secondly, not all of these sites are in places where people want to live. Less than a third of the brownfield land area is in London and the South East, and even this is still a very high level of aggregation. Thirdly, the 1.5m figure is based on the size of recently developed dwellings, which are the smallest that can be found anywhere in Europe. If we allowed for just a little bit more space and comfort, the numbers would dwindle. Fourth, development is already heavily biased in favour of brownfield sites, with a brownfield-to-greenfield ratio of 2:1.
So, yes, of course we should make the best possible use of previously developed land. But we should not pretend there is a limitless supply of such land in the right places. If brownfield redevelopment was really such an obvious no-brainer, developers would have snatched up that profit opportunity long ago.
What the whole discussion really shows us is the inadequacy of rigid, bureaucratic forms of land categorisation. An ex-industrial site that is used as a car park is not really ‘derelict’, and green-belt land that is used for high-intensity agriculture is hardly pristine countryside. We should judge land by its actual environmental and amenity value, not by a box-ticking template from the 1940s.
This article was originally published by City AM.
8 thoughts on “Don’t count on brownfield: it won’t solve the housing crisis alone”
Comments are closed.
SHARE
Housing or houses, Kristian?
We have unusually low population densities in our towns and cities by European standards because we enforce low rise developments (and your numbers assume more low rise developments). In most European countries there is much more ‘medium rise’ housing – it benefits us by using existing infrastructure, reducing car use and accessibility of local amenities, etc.. Not only should we be building on brownfield sites we should encourage redevelopment of existing low density, low quality, housing, into higher quality higher density housing, perhaps by such incentive measures as a land value tax forming a part of local taxation.
There are several problems with increasing urban population densities by developing brownfield sites. In many instances decontamination costs are so high that very large subsidies are required (e.g. the Olympic site, North Greenwich, Ebbsfleet etc.).
Also, there are substantial and ongoing ‘diseconomies of agglomeration’ such as congestion, greater exposure to crime/anti-social behaviour, the requirement for collective services (e.g. in blocks of flats compared with standalone houses) and large subsidies for public transport.
Rather than imposing locational decisions from above, why not leave locational decisions to landowners, developers and homebuyers, so that the outcomes reflect individual preferences?
Richard Wellings’s assertion that higher population densities in urban areas cause greater congestion, greater crime and anti-social behaviour and greater need for public transport can’t be justified by any reference to evidence. We also know that collective services are cheaper per person to provide where the population density is higher. The opposite of what he says is true.
Nobody is talking about ‘imposing decisions from above’. That’s what we have at present and it has produced low density sprawl and poor housing. Other countries have fewer regulations, and, guess what, higher densities and better housing.
@HJ In an unhampered market economy urban size and density reflect trade-offs between the economies and diseconomies of agglomeration. Government ‘compact city’ policies tend to ignore the latter, leading to inefficient outcomes.
@Richard Wellings. You are missing the point. You made a series of negative assertions about higher density housing that cannot be supported by any reference to evidence. We have perhaps the most restrictive planning laws in Europe – and they explicitly encourage low rise, low density, housing – not ‘compact cities. Other countries that don’t have such restrictive regulations have higher densities.
As for decontamination costs of brownfield sites – this is a quite separate issue, unless you are arguing that historically contaminated land should never be decontaminated.
@HJ – Draconian spatial planning policies have also been imposed in other European countries, imposing high-density housing on the population. And in the UK, a key rationale for the 1947 Town and Country Planning Act was to prevent 1930s-style ‘urban sprawl’ (i.e. families choosing to live in bigger houses with bigger gardens, well away from the slums). Since the early 1990s ‘compact cities’ have indeed been government policy, with even detached houses packed in very tightly, but with a major emphasis on small flats located along public transport corridors, often with restricted car parking.
Moreover, there is a wealth of evidence on the diseconomies of agglomeration. See, for example, our 2004 monograph on road pricing which showed that c. 30% of UK traffic congestion costs are in London. Also look at the pattern of violent street crime, heavily concentrated in a small number of high-density London boroughs.
@Richard Wellings – Your historical assertions are not very convincing in view of the deliberate low density housing in new towns such as Milton Keynes. Planning authorities stop developers building upwards and try to limit the number of dwellings in new developments and developers try to increase the number buy making them smaller but still low rise. London has low density housing compared to other large European cities – if your argument was correct, it would enjoy less congestion and lower crime as a result. It dies not. Low density means more car use and more congestion because people need to drive to get to local facilities because they are more spread out – most car journeys are less than 5km.
@Richard Wellings.
If the economies of agglomeration didn’t outweigh the disadvantages, land wouldn’t have any value.
As it is, land values are highest where population is the highest and most dense.
The reason being, every time a city doubles in size, its GDP per capita increases by 15%. As does the level of amenities enjoyed.
The problems start when this collective value is privatised by landowners. No work, effort or enterprise required on their part. See Ricardo’s Law of Rent.
As explained here, in the BBC documentary on maths, The Code. Skip to 50.15
https://www.youtube.com/watch?v=O9n5cWpiyD0
Which is why, if you build homes where demand is highest, prices rise, not fall