[Posted by Prof Henry G. Overman]
Following the Queen's speech my twitter feed is full of people expressing views for or against devolution to cities. Supporters push for maximum devolution to boost growth and productivity, improve efficiency in local government and build the northern powerhouse. Critics suggest the main attractions lie elsewhere - for example as a means of cutting government expenditure, a crude power grab by local politicians and a way for Londoner's to reduce taxes.
While there may be some truth in both these arguments, they won't do much to progress debate on what might constitute the right balance of powers between central and local governments. Indeed, what's a little depressing, but not very surprising, is the extent to which the vociferous public debate appears to pay almost no regard to the large academic literature that considers this issue.
Here, for example, is Wallace E. Oates one of the leading experts on fiscal federalism writing in a 1999 Journal of Economic Literature essay:
"But the proper goal of restructuring the public sector cannot simply be decentralisation. The public sector [...] consists of several different levels. The basic issue is one of aligning responsibilities and fiscal instruments with the proper level of government. [...] To realize [the advantages of federalism] we need to understand which functions and instruments are best centralized and which are best placed in the sphere of decentralized levels of government"
I've reflected on some of these issues before on this blog. For example, I'm pretty convinced that many decisions on transport and housing would better sit below national but above local authority level. After all, the evidence suggests that this is the spatial scale at which labour and housing markets operate. The governments proposals on metro mayors would help achieve this improved alignment.
In contrast, I'm still to be convinced on skills policy. Local authorities are certainly clamouring for powers in this area. But I worry that there is too much focus on higher level skills and not enough on the basics. I personally think the evidence is clear that early years and schooling should be central to any LED policy. Not only do these policies directly target one of the central problems faced by poorer families, but good local schools also help attract and retain already educated workers. But, I have no idea whether skills policy needs to be individualised, localised or centralised. I have similar reservations when it comes to active labour market policy.
I'm pretty sceptical on local industrial policy (e.g. cluster policy, complex attempts to
create innovation systems, targeting particular sectors etc) because I
don't think the evidence on its effectiveness is compelling and I worry that competition amongst local areas will lead to wasteful - and ultimately fruitless - expenditure.
Finally - I don't have a strong view on health or policing - because these are areas in which I have no particular expertise.
This isn't supposed to be comprehensive list - and my position on some of these areas isn't settled. Convincing evidence could shift my opinion either way. But we need the debate to shift from arguments over 'whether' to arguments over 'what' and 'why'. To my mind, we need answers to these latter two questions, before we can figure out how to proceed.
Friday, 22 May 2015
Posted by Paul Cheshire, LSE and SERC
We Brits are a self-satisfied lot, and seem to assume that Green Belts are a splendid British invention and a wonderful boon to the world. A signal of our praiseworthy efforts to protect the green: whether the village green or the environment: never a signal of our nostalgic hankerings for a rural England which never existed.
In fact, the truth seems remote from this English foundational myth. The way in which Green Belts developed suggests they are only belts by accident and are certainly an imported idea. As part of the work for a London First report, I have recently been looking at how our Green Belts came into existence and how their purposes have been transformed in the process. It is an extraordinary story, and I have learnt substantially from Jonathan Manns' work.
In the 19th Century, many continental European cities were getting rid of their city walls: the development of the nation-state and mechanised warfare had rendered them functionally irrelevant. This was happening in Vienna in the late 1850s and the idea of a Ringstrasse park was proposed. Although Napoleon had twice shown that Vienna’s fortifications could not withstand a modern (early 19th Century) army, the conservative Austrian military withstood their demolition until 1857. The city walls had had an area around them kept clear for military purposes, so their demolition released a substantial circular ring of space around the city. Bourgeoisie civic-mindedness was just developing. The map shows what was proposed: a Green Belt if ever there was one! But only a belt by accident. It was on the space released by the city walls and not surprisingly city walls had a roughly circular shape.
In the UK we also tend to link the idea of green belts to Ebenezer Howard’s proposals for Garden Cities. Not so. The first British person to propose a Green Belt seems to have been Lord Meath – an Anglo-Irish public figure and philanthropist. He is an interesting character and, to put it mildly, very alien to modern sensibilities. Despite being an Alderman on the original LCC he believed in the fundamentally corrupting nature of cities – so far as possible, their size should be limited and rural-urban migration halted. He also believed cities ‘corrupted’ the gene pool of the British. He realised, however, that London was going to continue to grow: so his answer was to increase the supply of green spaces and recreation grounds in London and construct a ‘Green belt or girdle’. His other solution to the burgeoning urban population was state-sponsored emigration to the British Colonies: he was a strong Imperialist - indeed the founder of Empire Day, a public holiday to celebrate the British Empire wherever the sun never set.
Meath explicitly imported the idea from the Viennese proposal but his version of a green belt or girdle had much more in common with Howard’s later ideas for green belts around his Garden Cities of Tomorrow than it did with the modern Green Belt. Apart from the Viennese Ringstrasse park, Meath was influenced by his visits to the US and seeing early ‘beltways’ – wide avenues around major cities. He envisaged such a broad avenue connecting public open spaces, some existing and some to be bought and dedicated to public use, encircling London.
An updating of his proposal as published in 1901 is illustrated below (and is taken from here). Much of the Meathian green belt was close to central London and all would have been within the GLA boundaries – mostly inside the inner boundaries of the modern Green Belt.
Source: Aalen, F. Planning Perspectives, 1989
The idea was taken up by the London Society in1919 in its proposed Development Plan for Greater London. This was still a proposal for ‘green lungs’ for a smoky, dirty and highly congested London. Official adaptation came in 1935 when the old London County Council endorsed the idea. It was implemented in 1938 with the Green Belt (London and Home Counties) Act. This empowered local authorities to buy land for public use. By the late 1940s 20,000 hectares of land had been bought around London dedicated as public open space.
In the immediate post-WWII period of reconstruction under the Atlee government, Green Belts were provided for all major cities that wanted them under the Town and Country Planning Act of 1947. But this did not propose buying land. The Act – still the basis for our planning system - expropriated ‘development rights’ from freehold land ownership. So in effect it gave power to local governments to prevent urban development on as yet undeveloped land. It did not give rights of public access or generate any change in ownership. Privately-owned land, if designated as Green Belt, would remain private. So, while still promoted in the name of green space, in practice the Green Belt provisions in the 1947 Act just stopped anything happening – rather than creating valued recreational or open spaces.
Although the 1947 Act provided for Green Belts, however, it did not implement them. That the 1947 Act was a creation of the Atlee government gives the Labour Party a sense of ownership of the idea but the implementation was under a Conservative Housing Minister, Duncan Sandys, in 1955. And the purposes by now had totally changed – both in aspiration and in reality. No more were Green Belts to be green lungs or recreational space; they were just to stop development.
As the Minister wrote: “even if…neither green nor particularly attractive scenically, the major function of the Greenbelt was…to stop further urban development”. That remains their function as confirmed in the National Planning Policy Framework of 2012. Their purpose was to be empty spaces between cities, to protect the Home Counties from the encroachment of London and force urban expansion to jump over Surrey or Hertfordshire to Northants, Cambridgeshire or Hampshire.
In the early 20th Century the Californians tried to apply zoning to confine Chinese minorities to particular neighbourhoods and later, in Louisville, Kentucky, zoning ordinances were brought in to stop Blacks from buying houses in certain areas. This was deliberate discriminatory zoning designed to protect privileges and create new ghettoes. Such discriminatory zoning was struck down by the Supreme Court in 1917, but the right to zone in a non-discriminatory way was permitted in a judgement of 1926. Local communities in the US soon found a way of keeping ‘undesirables’ out: they zoned for single family homes only and on large plots. Over time plots became ever larger and what is now euphemistically known as ‘minimum lot size’ requirements became pervasive.
It is claimed, for example, that ‘the Bay Area is built-out’. And so it is in the sense that more or less all available zoned lots are developed. But look across the Golden Gate Bridge and you see Marin County just a mile or so from downtown San Francisco. Minimum lot sizes in many of the communities in Marin County are 60 acres! There are many prosperous communities in the mid-West and West of the US which have 10, 20 and even 50 acre minimum lot size requirements. This is very effective at keeping poor people out. It is exclusionary but is does not discriminate and so is legal. It also contributes significantly to the sprawl of US urban development.
Greenbelts, as implemented, have become a very British form of exclusionary zoning. They have the effect of confining the urban poor to live at high densities in the cities and preserving the Home Counties (and other similar green places near to all our other cities with Green Belts of their own) for the aspirant gentry who had got there earlier on the tracks of the suburban railway network; or as bohemian urban refugees establishing artistic colonies such as Eric Gill’s Piggotts or the Sussex retreats of the Bloomsbury Group. It seems to be a long-standing tradition of the privileged British to want to deny for others what they enjoy themselves. Particularly if they can do so while maintaining a stance of moral superiority, presenting themselves as protectors of the real interests of those whose desires they frustrate. Prominent members of the Bloomsbury Group, happily housed in England’s pleasant fields, were solid supporters of the early Council for the Preservation of the Rural England and J. M. Keynes contributed to Clough Williams-Ellis’s seminal 1928 anti-urban tract, Englandand the Octopus. Williams-Ellis himself, of course, built a tasteful Italianate Disneyland precursor on a beautiful estuary in North Wales. But he had taste so that, of course, was alright.
Thus the idea of Green Belts, the accidental innovation – aimed at adding green space to high-density ancient Vienna in place of the city walls, then adopted in unaltered form in England as an idea to improve urban life for all – has been transmuted into a protective device for the wealthy. If you really want to plan to protect and provide better access to green space and open countryside without artificially constraining land supply and forcing up house prices, then Green Fingers (or Green Wedges) would seem to be the best solution. That is what more egalitarian Scandinavians have. Copenhagen has its Green Fingers – really brown urbanisation along the radial routes out of the city with protected countryside each side. Denmark has not just got cheaper housing: according to the Dallas Fed’s data, the real house price has increased by a factor of 1.6 in Denmark compared to 3.4 in the UK since 1975 but new houses in Denmark are a lot bigger: 80% bigger in fact.
Now green wedges seem like a quite sensible idea.
Monday, 27 April 2015
Posted by Paul Cheshire, SERC and LSE
All the Party manifestos are published now. All agree there is a housing crisis. The most recent data for housing starts in England show a fall of 19% in the last quarter of 2014 compared to the post-2008 high or the 2nd quarter of 2014; and completions and planning applications have been flat-lining since 2009. The latest English Housing Survey showed a continuing rise in age for first time buyers and that the proportion of 25 to 34 year olds who were owner occupiers had fallen from 59 to 36% over 10 years; and for the first time ever, more owner occupiers did not have mortgages than did. In other words all the evidence shows that house building is not significantly rising from what was a 100-year peacetime low in 2010, that the supply and affordability crisis continues – even gets worse – and the redistribution of housing wealth to the elderly continues.
A CentrePiece article last year explained that there had been a shortfall of building during the past 20 years of between 1.6 to 2.3 million and far too many of the houses that had been built were in the wrong places or the wrong type to satisfy demand. But – even worse – that the crisis was self-inflicted by longstanding policies.
Now we have Alice in Wonderland proposals: solutions that require us to believe six impossible things before breakfast and that words mean exactly what the manifestos say - not what they really mean. This will not build houses. Nor will promising to build 200,000 new houses a year (Labour), 230,000 starter homes (Conservative) or – any advance on 200,000? Yes, 300,000 a year (Lib Dem).
Targets and promises do not build houses. The experience of London demonstrates this. Since 2004 housing targets for London have risen from less than 20,000 to 42,000. There was a slight upturn in actual building in the boom conditions of 2004/05 to 24,000. Since then, as targets have risen, so building has fallen. In 2013/14 completions dipped to less than 18,000. To build houses one has to have land to put them on, a realistic diagnosis of the underlying causes and mechanisms for delivering them. Just raising targets or promising more houses simply means a larger shortfall between promise and delivery and an ever increasing unaffordability of housing.
Because the problem is the supply side. Urban containment boundaries were set in 1955 and the area of greenbelt has greatly expanded since then. London’s is now more than 3 times the area of the GLA (and 22% of the GLA is Green Belt); Oxford’s Green Belt is more than 7 times the area of Oxford. Oxford is now the least affordable city for housing in the UK and the City is surrounded by CPRE hotbeds telling them they cannot build here! It is very difficult to build upwards because of planning controls and every proposal to develop becomes an exhausting struggle because each decision is politicised by our mechanism of ‘development control’. Countries which do it more sensibly have plans and what the plan and building regulations permit, can be built.
Of the political manifestos perhaps the Tories’ is the most shameless. Senior Conservatives are not economically illiterate so they really know better. But what they are promising is more demand boosting: not only help to buy but extending the right to buy to houses no government owns; even major property agencies have condemned this proposal. Then the wheeze is to build 200,000 ‘Starter Homes’ to be sold to under-40 first time buyers at a 20% discount (from what?); and a London Land Commission to magically find tracts of Brownfield land that have presumably been hiding behind railway embankments for the past 15 years; and more powers to Local Authorities to resist building on the Green Belt. The Starter Homes, the small print reveals, will be either at the expense of ‘affordable’ housing built as part of planning deals or at the expense of ‘unviable’ commercial development. So not much net new construction there.
One can be entirely confident that the Conservative proposals will do nothing to increase the number of houses built net. This means their proposals will make housing even less affordable because they will subsidise demand but not increase supply. UKIP’s proposals are even less helpful. But then fair enough: one should expect less of UKIP in terms of sensible policies designed to tackle real problems. UKIP want to revise the 2012 National Planning Policy Framework (no evidence this has increased house building but it does offer some slight hope) to make it more difficult to build, increase the extent of Green Belts and extend the right to buy; but to British Nationals only.
Both the Lib Dems and Labour are, amongst other things proposing to build more houses. The trouble is neither really explains how it can be done nor sets out practical mechanisms to do it. So essentially they offer acts of faith. The Lib Dems propose 10 new Garden Cities on a revived Oxford-Cambridge rail link. But apart from some general statements about capturing land value uplift to pay for infrastructure, no mechanism is proposed for getting round the entrenched powers of opposition the current planning system provides anyone who opposes building.
Their other proposals either boost demand without adding to supply (Help to Rent) or have an old fashioned whiff of anti-landlordism and anti-speculator mindsets. They propose giving even more power to local communities to determine development. Without radical complementary changes to taxation, a new system to guarantee new houses come with additional infrastructure and losses are properly compensated this will strengthen the power of NIMBYism even more. A problem of just local decision making is that it empowers those who bear the costs of development (real – noise and dirt during construction, loss of views and more congestion – in the absence of new infrastructure – after the new houses are built) but disenfranchises those who stand to benefit – would–be residents or owners. The Lib Dem proposal to pilot land value taxation is potentially helpful but another proposal - to make planning decisions even more complex by inventing conditions requiring occupation after construction - is not. How on earth could that be imposed? Suppose the market turns down unexpectedly or the developer just gets the local market wrong? It will add yet more risk to the development process so cause otherwise viable projects to slip from viability.
Labour did have the foresight to commission the Lyons Review of Housing but their manifesto conveys the impression they almost wish they had not. They promise to increase house building to 200,000 a year by 2020 but there is no mention of how to increase land supply or funding renewed public sector house building. Instead there is coverage of what seems like an ill-thought out proposal to re-channel finance from Help to Buy ISAs; how to stop ‘land banking’ and ‘speculation’ by introducing ‘use it or lose’ planning permission and double taxation on empty homes. Then there is a proposal to tighten controls on landlords by mandating 3-year tenancies and a national register of private landlords. The most explicit proposal on the supply side is to ‘start to build a new generation of garden cities’. But while Lyons was explicit with details and delivery mechanisms, the manifesto is not. Similarly the Lyons Review had an informed analysis of the underlying problem – our self-imposed shortage of development land; the difficulty of getting across boundary agreements (see the Oxford case) and how over simplistic is the idea of relying on Brownfield land. There is no mention of these serious points in the Labour manifesto.
Overall, the manifestos confirm that all parties are unwilling to face up to the political problems they perceive would follow if they advocated solutions that might effectively address the crisis of housing supply. The illness is real but all that is on offer is snake oil; displacement activities treating some symptoms but not the underlying causes and – paradoxically - having the net effect of making the crisis worse. Perhaps that is just a little harsh on Labour but I did just hear their spokesperson offering the party’s solutions and the whole emphasis was on how the ‘market was not working so the planning system needed to be tougher’. Not so: the problem IS the planning system. It needs root and branch reform but that would take serious political courage. Fake solutions will not work: in the case of many of those on offer, including discouraging landlords, making an already mind-boggling planning system yet tougher and more complex or boosting demand when the problem is supply, the crisis will get worse. Indeed the housing crisis will get worse anyway without radical reform of the planning system, and local taxes and property taxes; and incentives.
This post originally appeared on the LSE British Politics and Policy blog.
Friday, 24 April 2015
Posted by Gabriel Ahlfeldt & Paul Cheshire, SERC and LSE
Economists, unlike cynics, do not know the price of everything – not by a long way. Certainly we do not know the price of good architecture. Apart from anything else identifying ‘good architecture’ is highly subjective. So to measure its price you have to make some brave assumptions about what it is and where to find it. And from an economic perspective the price is not the interesting issue. The interesting issue is: left to their own devices, would markets under or over provide ‘good architecture’? In the jargon, is there a problem of market failure?
For sure, public policy in this area implicitly assumes that there is market failure, and that markets alone will not provide enough good architecture. For example, policymakers intervene to control the design of new buildings, and to stop private owners from altering or replacing existing ones. But this has costs: both directly in administration and compliance, and indirectly through impacts on energy costs, on usable space in cities and on economic outcomes. A series of papers from SERC researchers explores these economic effects in detail.
Regulation for good design may lead to rent-seeking behaviour: Cheshire and Derricks find evidence that developers use ‘Trophy Architects’ to game the planning system, as a way to get more space on a given site. This study – like some others – identifies ‘good architecture’ by classifying buildings as high quality if they are designed by architects who had already won a major lifetime achievement award: ‘Trophy Architects’ or ‘Starchitects’. Such awards go back a long way, so there are 19th Century buildings in London which qualify.
Cheshire and Derricks find that for commercial buildings, a Trophy Architect (TA) adds no premium at all to the price of buildings. But it does add to both the design and construction costs: with TA design, the cost per m2 for a 25-floor building is 10 to 17.5% higher. TA design also seems to make buildings slower to let out, which suggests there could be a rental premium, even though there is no increase in a building’s selling price.
There is a puzzle here. TA design adds nothing to a building’s selling price, implying no productivity benefit from starchitecture. So why do developers pay a premium for construction? The answer appears to be that using a Trophy Architect helps game the regulatory system. Controlling for other factors, in London such buildings were an astonishing 19 floors higher than buildings on similar sites designed by ordinary architects.
An economist would say that the developers were using design to generate ‘rents’. In a city with tight planning rules and development controls, the extra rentable space in these big buildings implies a 130% increase in the value of a typical site on which there was a possibility of building tall. The problem is that these rents become a deadweight loss as the extra cost of using a trophy architect, the extra time through the planning system, and the extra risk of eventually not being successful absorb potential gains.
In London, some 25% of skyscrapers are TA designed compared to only 3% in Chicago – a city which is not just less regulated than London (so office space is cheaper) but regulation is less gameable because Chicago is a rule-based zoning system, rather than the more political development control process we use in Britain.
The deadweight loss associated with this way of gaming the planning system may be fully or partially offset because ‘Trophy Architecture’ generates welfare for residents and tourists; or even for the occupants of other commercial buildings. In both London and Berlin, for example, this kind of architecture generates a lot more visual interest, asmeasured by photos shared on sites like Flickr and Picasia. The effects are even larger for contemporary buildings like the Gherkin than for historic signature buildings. So the benefits of good modern design may go some way to offset the calculated use of design by developers to build bigger buildings.
For houses the story seems somewhat more straightforward. People do pay for better design, so good architecture seems to provide direct utility – otherwise the best known architects would not be able to charge a premium for their services. This is a private benefit and wouldn’t seem to raise issues of market failure or a case for subsidising design.
However, there is quite persuasive evidence, from a series of SERC studies, that there are external economic benefits from high quality or historic residential buildings. Property prices inside conservation areas are about 5% higher than just outside conservation areas. Given that conservation area boundaries can be quite arbitrary, and regulatory constraints inside conservation areas are costly, this effect is likely attributable to the special design character in these areas. This jump in prices at the boundary is particularly large where residents report the architectural quality of the area is large relative to nearby areas. On a four-step scale (from ‘not at all distinctive’ to ‘very distinctive’), a one-step increase is associated with an increase in prices of up to 25%.
Interestingly, this research also finds a positive price effect for conservation area properties that have modern (post-WWII) design, and even for properties outside conservation areas with a view onto buildings inside a conservation area. This is consistent with otherSERC research showing price increases in nearby areas after a conservation area has been designated.
So we can see positive house price effects from all kinds of ‘good architecture’. For example, residential buildings designed by Frank Lloyd Wright – sometimes referred to as the Greatest American Architect of all time - have a positive spillovereffect of up to 8% in Chicago’s Oak Park. The redevelopment of the Wembley stadium – designed by trophy architects Fosters and Partners – was followed by increases in propertyprices by up to 15% in nearby areas with a view.
Let’s sum up. Policymakers intervene in property markets to protect historic buildings, and to promote ‘good’ modern architecture and design quality. From an efficiency point of view, the justification for these policies is that markets may undervalue good architecture, because such features of buildings and neighbourhoods represent external benefits or a form of public good.
In the UK, regulations for historic districts or Listed Buildings can be very tight and costly: they may control the colour of paintwork and protect internal details of layout and design making it difficult to adapt such buildings for modern use or improve energy efficiency to modern standards. Even where less extreme, the cost of redevelopment behind facades is very high.
So far the evidence suggests that there are external benefits associated with high quality residential buildings and neighbourhoods: but the evidence is much weaker when it comes to commercial buildings, especially given the peculiar British practice of using expensive TAs to game the system to get more space on a given site. So where does this leave policy?
On balance the most plausible position is that there is a welfare economics case for public policy to support some good design and heritage neighbourhoods, and that this case is stronger in the context of residential than commercial buildings. But as with any policy and regulation there are costs, here in the form of restricted supply and affordability and rent-seeking behaviour. Most notably, these costs are particularly high in a context where a planning system – as in Britain – makes living and working space notoriously scarce and leaves many planning decisions the outcome of a quasi-political and bargaining approach. If there was sufficient space for development, the external cost of preserving relatively low density heritage areas would be less. And if decision making was rule-governed there would be far less incentive to employ Trophy Architects just to game the system to get more space on a given site – as opposed to improving design quality.