tip off
36

Why do cities still sprawl?

One of the perennial conceits of city policy-makers is they can contain urban sprawl. Yet throughout history, those pesky cities have persistently grown outwards.

Real house prices by distance from CBD ($09/10). Source: State of Australian Cities 2012

Australian and US cities, which largely developed in the era of mechanised transport, are notorious for sprawling. But they’re not alone.

European cities are sprawling too. The European Commission released a report in 2006, Urban sprawl in Europe: the ignored challenge, arguing that “sprawl threatens the very culture of Europe”.

The authors note that European cities were more compact in the 1950s but urban sprawl is now a common phenomenon throughout Europe. While they’re not sprawling as much as US cities, they say there’s no sign the outward spread is slowing.

London is sometimes cited as an example of a city that has been able to contain growth within its boundaries. However this argument neglects to mention that while the population within the green belt is 8 million, there are another 12 million residents living outside it.

As Wendell Cox notes, all of London’s population increase since WW2 (6 million residents) took place outside the green belt. He catalogues the metropolitan expansion of a range of European and world cities, including Milan, Barcelona, Zurich and Moscow.

Back in 2002, the Victorian Government decided in Melbourne 2030 that only 31% of all new dwellings constructed between 2000 and 2030 would be located in fringe greenfield sites. It envisaged the proportion would fall to 22% by 2030.

The centrepiece of the strategy was the Urban Growth Boundary. But faced with growing public concern about housing affordability, the same Government watered the policy down in 2008.

A new planning document, Melbourne @ 5 Million, revised the figure to 47% of all dwellings constructed over the succeeding 20 years. This simply reflected what was actually happening in the market. In fact because not all outer suburban housing construction takes place in Growth Areas, the real figure was 58%.

The Greens also appear to have effectively thrown in the towel on efforts to contain sprawl. The party’s new report on High Speed Rail (HSR) proposes that regional centres like Shepparton and Albury should become dormitory suburbs, with workers commuting hundreds of kilometres by HSR to capital cities.

Governments are strong on the rhetoric of limiting sprawl but are weak on action. That’s because the political cost of putting in place policies that actually work would be extraordinarily high.

Cities sprawl for simple reasons. The outer limit of Australian cities was originally contained by how far workers were prepared to walk, but the coming of trams and trains at the end of the nineteenth century changed all that.

The new technology gave residents the opportunity to live in the suburbs and commute to the centre. They were able to escape the (then) very low amenity of the inner city and, because peripheral land was cheap, enjoy a lot more space.

Rising incomes and car ownership after WW2 extended suburbanisation even further by filling in the gaps between rail lines. Like the country railways of an earlier era, freeways helped open up ever-new tracts of cheap land.

Suburbanites could shop locally and soon they could work in the suburbs too as more and more employers abandoned the city centre.

With the weight of population inexorably moving outwards, governments also built major facilities like hospitals and universities in the suburbs. The city centre became an occasional and discretionary destination for the great majority of suburbanites.

Today’s outer suburbs offer an attractive bundle of benefits for many people in most capital cities. They’re especially appealing to the very large numbers of residents who put a high premium on affordable space and privacy.

Yet as the boundary of the city expands, there is an increasing number of home-seekers who would prefer to live closer to the centre. Locations that’re more accessible cost more, so they accept they’ll necessarily have to live in a much smaller dwelling and may have to share their commute with others.

In theory, the supply of apartments and townhouses in long-established suburbs should increase to meet this demand, but existing residents work hard to ensure it doesn’t (e.g. see here and here).

All those new residential tower blocks indicate supply is expanding in and around the CBD. However there’s much more resistance to redevelopment in predominantly residential inner city and middle ring suburbs.

Governments could do more to contain sprawl, but they’d have to put in a bigger effort than simply drawing a line on a map.

They could, for example, override the objections of existing residents to redevelopment proposals. They could tackle the high cost of constructing buildings over four storeys.

They could increase the cost of housing on the urban fringe by, for example, levying higher infrastructure charges or simply imposing a tax on fringe growth. That would make established suburbs more cost-competitive with the fringe.

They could raise public transport fares closer to the real cost, making it more expensive to live a long way from the centre. They could discourage driving long distances by taxing cars so they pay for the congestion and environmental damage they cause.

Some of those are good ideas for other reasons, but many of them would be extraordinarily hard politically. Governments would also come under increased pressure to improve infrastructure and services within established suburbs, where it arguably costs more to provide.

Sydney has a lower proportion of starts on the periphery than other cities, partly due to high infrastructure charges but mainly because of unique geographical constraints that can’t be sheeted home to politicians. Sydney also has the highest housing costs in the country.

In Victoria, Planning Minister Matthew Guy has appointed a Ministerial Advisory Committee to oversee the preparation of a new metropolitan strategy for Melbourne. He appears to be giving them a relatively free hand and the indications are they want to achieve a very serious reduction in the level of sprawl.

Just what price Victorians should pay and are prepared to pay to reduce sprawl in Melbourne is a challenging question that the Committee will need to address. It should take a long hard look at where Melbourne 2030 went wrong.

The Committee should also think long and hard about what can be achieved – a strategy that has no chance of success is rubbish. If they can devise a politically acceptable way to improve affordability significantly while simultaneously applying the brakes seriously to sprawl they’ll be world-leaders.

36

Please login below to comment, OR simply register here :



  • 1
    Tom the first and best
    Posted December 12, 2012 at 11:51 am | Permalink

    Some of the railways in suburban Melbourne actually came earlier than the late 19th century. Not just regional lines coming through to Melbourne but suburban only lines as these maps of 1860 and 1870 show.

    http://www.vrhistory.com/VRMaps/Vic1860.pdf

    http://www.vrhistory.com/VRMaps/Vic1870.pdf

    A such Melbourne had a PT based sprawl and was one of the lowest density city in the world until the car age, when we were overtaken by most other new world cities.

  • 2
    Dylan Nicholson
    Posted December 12, 2012 at 1:47 pm | Permalink

    I’d summarize the main argument you’re making as “governments encourage sprawl because it’s politically convenient”, which sounds about right. But high land prices surely indicate that more people want to live in the denser areas than there are places to live…which would suggest there is higher demand for higher density development than governments typically recognise (or perhaps the demand comes from people with insufficient political clout!).

  • 3
    Alan Davies
    Posted December 12, 2012 at 2:10 pm | Permalink

    Dylan Nicholson #2:

    Agree with your second proposition (which I’d hoped would be clear from what I wrote). That’s where “political convenience” really comes into it.

    As for you first proposition, I’d phrase it a bit differently i.e. governments facilitate sprawl by subsidising the cost of transport. It’d be a lot less attractive if we paid the real (financial) cost of transport, particularly PT.

  • 4
    Dylan Nicholson
    Posted December 12, 2012 at 2:56 pm | Permalink

    Subsidising the cost of transport is definitely a big part of it, but I’m not so sure it’s the primary way in which government policies lead to increased sprawl – there’s a whole list of things, each of which pushes us towards low density development.

  • 5
    Robert Brown
    Posted December 12, 2012 at 4:21 pm | Permalink

    Hi Alan,

    “They could, for example, override the objections of existing residents to redevelopment proposals. They could tackle the high cost of constructing buildings over four storeys.”

    It seems to me that most inner-city suburb redevelopment proposals are ~15 storey buildings.

    I’ve been wondering why there aren’t more 4-6 storey buildings proposed. Such heights would be a great increase in density from most existing housing (i.e. 1 or 2 storey), and more agreeable (I would argue) to existing residents.

    I’m thinking or Brooklyn or even Barcelona – nice low-rise buildings that let plenty of light down to street level. Buildings that can be “walk-up” i.e. you can live on the top floor without a lift.

    I had assumed that building a 15 storey building is only marginally more costly than a 5 storey one, whilst being far more profitable (more apartments to sell), and that’s why there aren’t more low-rise buildings proposed, but your quote here suggests otherwise.

    Have you done any writing on this (i.e. increasing density, but to what?) or any linkage?

    Thanks,

  • 6
    Tom the first and best
    Posted December 12, 2012 at 4:55 pm | Permalink

    3

    It is those subsidising of the costs of cars rather than the costs of PT that is the real driver of sprawl. If the government did not do so much road building then living in outer areas would be harder and less people would do it.

  • 7
    Tom the first and best
    Posted December 12, 2012 at 4:57 pm | Permalink

    4

    Minimum parking requirements is a big cause of high costs of density. Scrap them and there will be more car park free flats in the inner-city that with thus be cheaper.

  • 8
    Tom the first and best
    Posted December 12, 2012 at 5:00 pm | Permalink

    5

    There are plenty of 3-4 story blocks of flats being built in inner-to-middle Melbourne.

  • 9
    Alan Davies
    Posted December 12, 2012 at 5:15 pm | Permalink

    Robert Brown #5:

    The argument is: taller buildings are unionised and subject to restrictive work practices. Those tall 15+ towers are around the CBD and near-CBD. Elsewhere though, it’s harder to get an adequate return on buildings of moderate height. Another theory is a building needs to be above 7 or 8 levels to justify the cost of elevators. IIRC the National Housing Supply Council’s reports on housing supply reference this issue.

  • 10
    Alan Davies
    Posted December 12, 2012 at 5:22 pm | Permalink

    Tom the first and best #6:

    Both cars and PT are subsidised but in different ways. In financial terms subsidies to PT are higher (NSW Auditor General says each rail journey in NSW is subsidised $10 on average). In terms of externalities, cars are higher. See here (although this is a topic that seems surprisingly unstudied).

  • 11
    Steve777
    Posted December 12, 2012 at 5:48 pm | Permalink

    I’ve often thought that we’d have more sensible planning if we reduced the number of local government areas. ‘Micro’ councils that administer a few suburbs in an area a few kilometres across are just too much hostage to NIMBY activists. Local papers are full of stories of people concerned about their land values and keeping newcomers and commuters off their patch while pretending to be concerned about the environment. So urban infilling, increasing density along corridors such as the North Shore line and infrastructure projects don’t proceed.

    Sydney has over 40 local councils. Eight to ten would be more than enough. If tiny local government areas were so good, why haven’t larger, outer councils that were 50 years ago semi-rural shires (e.g. Liverpool, Sutherland) broken up? Because it doesn’t make sense.

  • 12
    hk
    Posted December 13, 2012 at 9:02 am | Permalink

    How much difference in planning control outcomes is there on non-built on land when the land falls in either side of green space defined by a boundary or/and a zoned green belt? There is quite a healthy debate currently going on in the UK on the topic. The TCPA in the UK has recently (http://en.wikipedia.org/wiki/Green_belt_(United_Kingdom) put a sound case supporting a more flexible and real world approach which the Government of Victoria could apply. Unfortunately the majority of the population in Victoria do not care too much about protecting places of special significance and natural beauty. There are no comparable Acts to those in the UK to protect our beautiful country side from high profit taking developers.

  • 13
    suburbanite
    Posted December 13, 2012 at 9:18 am | Permalink

    Where do large scale developers come into the equation? Are they just responding to market demand or have they gotten involved in politics and had a hand in shaping policy? It seems strange not to included them in the discussion, after all they make large bets on future rezoning.

  • 14
    Alan Davies
    Posted December 13, 2012 at 9:17 pm | Permalink

    suburbanite #13:

    If anything, I expect developers “gaming” the market raises land prices on the fringe and thereby makes it less attractive relative to established areas. At the cost of affordability of course.

    The Land Commissions in the various states were meant to moderate prices and supply – it would be an interesting research project to see what they really achieved and what, in their current incarnations, they actually achieve today.

  • 15
    boscombe
    Posted December 13, 2012 at 10:15 pm | Permalink

    “If the government did not do so much road building then living in outer areas would be harder and less people would do it.”

    Why has no one made the obvious comment that most people want to live in low-density areas and are quite entitled to have the government serve their needs?

    I’m not for or against high density, I’m for people being able to make equally valid choices. The problem of re-developing existing suburbs into higher density ones is due to the stupidity of the way it’s being done. There are whole sections of Perth that could be redeveloped into high-rise without bothering too many people. There are leafy suburbs of quarter acre blocks that resist any subdivision, but if you, say, made all, and only, corner blocks subdivisible, I don’t think there would be too much opposition.

  • 16
    Dylan Nicholson
    Posted December 14, 2012 at 8:37 am | Permalink

    “Why has no one made the obvious comment that most people want to live in low-density areas and are quite entitled to have the government serve their needs”

    If most people wanted to live in low-density areas then land-prices in such regions would reflect that. They clearly don’t. It seems clear to me a very high percentage of people that live in low-density areas do so because there aren’t enough higher-density areas to satisfy demand, hence they’re absurdly overpriced. It’s probably not the density per se that people want of course, but the convenience and range of entertainment/shopping/job options that come with it.

  • 17
    Dylan Nicholson
    Posted December 14, 2012 at 9:02 am | Permalink

    My other suspicion is that there’s more and easier profit to be made from buying up farmland and developing greenfield sites into what’s easy to sell as some sort of apparent suburban paradise with nice big houses and lots of open space that distracts many would-be buyers from the realisation of how dependent they’ll be on long car journeys to get virtually anywhere. But I’d still say poor government policy is largely to blame for way these sites typically are developed, and for the longer term consequences.

  • 18
    Posted December 14, 2012 at 11:56 am | Permalink

    While European cities also sprawl. They generally are still small in area compared with Australian cities that take up areas the size of entire countries elsewhere.
    We were hooked on ‘;own block and free-standing house.’ Mix that in with a public transport system trying to service a population spread thinly over miles and miles and you have the typical Australian city. Endless driving to get a bottle of milk
    Of course, the cost in loneliness and mental illnesses are also a result of all this separate living. Have a look at Mumbai where a million are packed in one square kilometre and they all smile! How can that be?

  • 19
    Dylan Nicholson
    Posted December 14, 2012 at 1:10 pm | Permalink

    I’d think almost no matter what government policies you had in place Australian cities are/were always going to sprawl more than European ones – faster population growth, almost all growth occurring in an age of cheap, readily available transport, far more land available, less cultural/historical attachment to the sorts of small-scale farming that much of the land surrounding towns and cities is used for etc. etc. OTOH, increasing the density of existing European towns/cities is almost certainly more difficult now than it is (and certainly than it should be) in Australian cities. I actually don’t really think we need *any* active government plans attempting to increase density or minimize sprawl – there’s no shortage of demand for more dwellings in existing areas, and providing legislation that artificially restricts supply is kept to a minimum, and the necessary infrastructure is augmented to keep up with the increased demand, there will be far less pressure on the outskirts of cities to continue sprawling out endlessly. Further, it’s already true that many newer fringe developments now include townhouses etc. – hopefully some of them will start being built without the thoughtless zoning restrictions and lot divisions that give us vast areas of nothing but houses surrounding a single shopping complex (for which 80% of the area is given over to car parking), with no commercial or light industrial presence at all.

  • 20
    potholes
    Posted December 14, 2012 at 2:12 pm | Permalink

    Yes Gerard, a million people packed in one square kilometer Mumbai always have smiles when you’re wearing your rose coloured glasses. That aside the problem with the current process of increasing urban density is that each new development reduces amenity to open space/ gardens. Each house that gets knocked down for med/high density housing means less trees and less open space. If there was some foresight on this issue in planning, say- for every new multi-level development that a portion of land had to be reserved for open space, then it might work.

  • 21
    drsmithy
    Posted December 16, 2012 at 5:57 pm | Permalink

    The economic cost of trying to contain urban sprawl – real estate bubbles and the huge percentage of wealth dumped into them – is staggering. Australians have a couple of _trillion_ dollars tied up in their real estate bubble. How much HSR linking dispersed, low-density, self-contained, cheap population centres could that have paid for ? How much would be left over for investment in actual productive enterprise ?

  • 22
    Dylan Nicholson
    Posted December 16, 2012 at 6:28 pm | Permalink

    What makes you think real estate bubbles are a product of trying to contain urban sprawl? Indeed, as I’ve said, I’m struggling to think of a single piece of policy issued by any government in recent history that could be said to be obviously and effectually attempting to contain sprawl – but plenty of instances that clearly encourage it. Even the reserving of so called ‘green wedges’ is basically just a concession that our cities are going to sprawl anyway, so we need to at least live a bit of greenery to make the new developments seem a little less depressing.

  • 23
    Dylan Nicholson
    Posted December 16, 2012 at 6:30 pm | Permalink

    er, *leave* not live.

  • 24
    drsmithy
    Posted December 17, 2012 at 6:59 am | Permalink

    What makes you think real estate bubbles are a product of trying to contain urban sprawl?

    The Unconventional Economist (amongst others) over at Macrobusiness has done a quite comprehensive job of demonstrating it.

    Indeed, as I’ve said, I’m struggling to think of a single piece of policy issued by any government in recent history that could be said to be obviously and effectually attempting to contain sprawl – but plenty of instances that clearly encourage it.

    Urban Growth Boundaries, which allow developers to “land bank”, choking supply and driving up prices.

  • 25
    Dylan Nicholson
    Posted December 17, 2012 at 7:07 am | Permalink

    drsmithy – I’d like to see at least one of the articles in question.
    As far as urban growth boundaries grow, Alan has already pointed out here that the current boundaries allow enough room for far many more dwelling than there is expected to be demand for in the decade or so. And they exist because governments can’t exactly just allow developers to buy up farmland wherever they like and convert it to residential areas without needing to be involved in order to supply infrastructure, create individual land titles etc. etc. That anyone would think that’s the main reason for real estate bubbles (which are surely primarily caused by a combination of population growth and banks allowing borrowers to continually increase their leverage, thus forcing us all to do so in order to stay in them market) strikes me as bizarre. There’s plenty of countries (The Netherlands is one example I recall reading about) that have far more restrictions (both natural and artificially imposed) on urban development, yet haven’t experienced any sort of real estate bubble to compare with Australia’s.

  • 26
    Dylan Nicholson
    Posted December 17, 2012 at 7:11 am | Permalink

    And to clarify, I accept that of course artificially-imposed restrictions on urban development do push up prices, but by a far greater degree in inner denser areas than at the fringes, and at most cause a gradually increasing upward pressure as population increases – certainly not enough to drive a bubble.

  • 27
    drsmithy
    Posted December 17, 2012 at 2:45 pm | Permalink

    I’d like to see at least one of the articles in question.

    I don’t really have time at the moment to specifically collect them all, but here’s a few to get you started.

    http://www.macrobusiness.com.au/2011/03/jumping-the-urban-growth-boundary/
    http://www.macrobusiness.com.au/2011/09/auckland-embraces-unaffordable-housing/
    http://www.macrobusiness.com.au/2012/10/auckland-housing-crisis-an-instrument-of-poor-policy/
    http://www.macrobusiness.com.au/2011/12/why-developers-land-bank/
    http://www.macrobusiness.com.au/2011/07/rethinking-urban-planning/

    More generally:
    http://www.macrobusiness.com.au/category/global-housing/

    As far as urban growth boundaries grow, Alan has already pointed out here that the current boundaries allow enough room for far many more dwelling than there is expected to be demand for in the decade or so.

    Which is irrelevant when all that land has been snatched up by developers who only dribble it out. Land that can’t be purchased at a reasonable price is no different to land that can’t be purchased at all.

    That anyone would think that’s the main reason for real estate bubbles (which are surely primarily caused by a combination of population growth and banks allowing borrowers to continually increase their leverage, thus forcing us all to do so in order to stay in them market) strikes me as bizarre.

    Easy credit certainly adds fuel to the fire, but bubbles in places like South Korea demonstrate it’s not a requirement (and hence not a root cause). Conversely, locales in the US with few restrictions on land use did not suffer significant bubbles, despite the same access to easy credit as those that did.

    Population growth is only a problem if the land supply is not elastic enough to handle the increased demand. Ie: thanks to things like UGBs and land banking.

    There’s plenty of countries (The Netherlands is one example I recall reading about) that have far more restrictions (both natural and artificially imposed) on urban development, yet haven’t experienced any sort of real estate bubble to compare with Australia’s.

    The Dutch are also in a bubble.

    http://www.macrobusiness.com.au/2011/06/dutch-show-how-not-to-run-housing-policy/
    http://www.macrobusiness.com.au/2012/08/dutch-pay-the-price-for-poor-housing-policy/

  • 28
    Dylan Nicholson
    Posted December 17, 2012 at 5:09 pm | Permalink

    I’ll read those when I get a chance, thanks.
    But a few points of yours stand out:
    “Which is irrelevant when all that land has been snatched up by developers who only dribble it out”

    Exactly! One of the problems is that land bought by developers and not being used productively is insufficiently taxed relatively to land that is developed. The Henry Tax report definitely recognised that.
    But I can’t hardly see how getting rid of UGBs is going to change anything much.

    “Population growth is only a problem if the land supply is not elastic enough to handle the increased demand”…sure, but there’s far far greater artificially-imposed restrictions on effective land supply in already developed areas than at the fringes, so surely we should be addressing those first. I’d be fascinated to know of a single example of a case where a significant new housing development would have gone ahead weren’t it for the current growth boundaries.

  • 29
    Dylan Nicholson
    Posted December 17, 2012 at 5:12 pm | Permalink

    I’d also agree with the articles above that argue that UGBs as currently implemented probably exacerbate sprawl rather than allow density to increase naturally in existing areas. Perhaps it can fairly be listed as a policy *intended* to contain sprawl, but if legislators are not constantly vigilant in ensuring policy is being well executed and having the desired effect, it’s better to have none at all.

  • 30
    drsmithy
    Posted December 17, 2012 at 6:19 pm | Permalink

    But I can’t hardly see how getting rid of UGBs is going to change anything much.

    It’s certainly a multi-factor problem, but UGBs contribute substantially by encouraging speculative purchasing of land right on the edge. They further prevent “leapfrogging” by other developers to put housing on the cheaper land further out than that which is landbanked.

  • 31
    Dylan Nicholson
    Posted December 17, 2012 at 9:15 pm | Permalink

    Even if there were no officially published UGB, developers are going speculatively purchase land on the outer edge of existing developments, on the reasonable expectation that it where the demand is going to be, and where governments are going to agree to provide infrastructure and legal title for future residential developments. And speculative purchase of land is a problem that needs tackling regardless of UGBs.

  • 32
    Dylan Nicholson
    Posted December 17, 2012 at 9:23 pm | Permalink

    …having said that, I wouldn’t particularly see it as any great loss if UGBs were scrapped, and governments simply approved greenfield sites for urban zoning on a case-by-case basis. My main issue, as I’ve pointed out, is that the inflated land-prices are keeping huge numbers of people from enjoying the amenities of denser, more established urban areas, who then have little choice but to flock to the fringes. As it happens I also think our cities will develop into more interesting and better places to live if densities are allowed to naturally increase to accommodate demand.

  • 33
    drsmithy
    Posted December 18, 2012 at 12:24 am | Permalink

    having said that, I wouldn’t particularly see it as any great loss if UGBs were scrapped, and governments simply approved greenfield sites for urban zoning on a case-by-case basis. My main issue, as I’ve pointed out, is that the inflated land-prices are keeping huge numbers of people from enjoying the amenities of denser, more established urban areas, who then have little choice but to flock to the fringes.
    You will struggle to find anywhere in the world, with loose land restrictions and (relatively) high property prices.

  • 34
    Dylan Nicholson
    Posted December 18, 2012 at 8:51 am | Permalink

    Sure, because those areas are ones that there’s little demand for, hence no point in having restrictions!
    I’d think there must be at least a few cities in Asia (Singapore maybe?) with relatively liberal urban development policies where land is still quite expensive due to high demand and limited availability.

  • 35
    drsmithy
    Posted December 18, 2012 at 9:20 am | Permalink

    Sure, because those areas are ones that there’s little demand for, hence no point in having restrictions!
    I think you’ll find land use restrictions all across Australia are quite tight, even in places of “little demand”.

    Hence the reason a postage-stamp (~400m^2) of land way out in Springfield Lakes (near Brisbane) costs a couple of hundred grand or more.

    I’d think there must be at least a few cities in Asia (Singapore maybe?) with relatively liberal urban development policies where land is still quite expensive due to high demand and limited availability.

    I’m not discounting they might exist, but I’ve never heard of them. Certainly they’re a corner case if they do.

  • 36
    drsmithy
    Posted December 18, 2012 at 11:43 am | Permalink

    I’m not discounting they might exist, but I’ve never heard of them. Certainly they’re a corner case if they do.
    Or, I should have added, they’re going to be limited by some genuine physical boundary, which makes them completely irrelevant to the discussion.

Please login below to comment, OR simply register here :



Womens Agenda

loading...

Smart Company

loading...

StartupSmart

loading...

Property Observer

loading...