Sunday, August 20, 2006

Urban spawl & house prices

An idea increasingly posing a challenge to State Governments is that house prices in Australia are high not because interest rates are relatively low but because State Governments have been excessively stingy in releasing land and are charging new settlers too much in upfront infrastructure provision fees.

The Reserve Bank’s Ian Macfarlane argues that high prices are due to these inappropriate policies and PM Howard has been quick to seize on these remarks blaming the States for high house prices. Macfarlane's remarks follow recent publication of a Housing Industry Association study and a report by Alan Moran from the Institute of Public Affairs that make similar claims. The issue is important because, as the OECD has reported, Australian housing is the most 'overpriced' in the world. Something is amiss and we need to work out what to do.

I am trying to sort out my attitudes to these views for a forthcoming seminar on urban congestion to which I have been invited. I have several provisional observations and welcome comments on these.

For economic efficiency, marginal infrastructure (and indeed pollution and congestion) costs due to development on an urban boundary should be levied on those settling on the boundary. This provides incentives for people to settle where social costs are lowest and inhibits the sprawl that is so damaging in Sydney’s west, for example. This seems to me an absolute. Past generations might have got away with inefficient infrastructure subsidies but that is no reason to stick with such bad policies. New settlers should pay for the costs they impose.

Land shortages will increase the price of housing. But house prices are responding to various influences. People over–capitalize housing from a social perspective because it is a tax free way of investing their wealth. McMansions are a fact of life because of our tax system. Moreover, because the favorable tax treatment of housing has persisted for a long time, the use of housing as a successful investment has become a self-fulfilling prophecy. Consumers over-invest in housing, in a way that almost exhausts their inter-temporal budget, since this seems like (and has proven to be) a sensible way of accumulating wealth free of capital gains tax. It is not just the supply of labnd but also the inappropriately excessive investment in housing on developed land which creates high house prices.

As argued before, urban sprawl has an undeservedly bad press. Large cities which include high quality housing for a substantial majority of the population are to be lauded not condemned. Travel costs might be higher in a spawled out city but the availability of facilities in much of what is described as urban sprawl is socially and economically advantageous. To be specific – my local suburb in metropolitan Melbourne (Banyule) has a symphony orchestra, an excellent local library, fantastic restaurants, good public sporting facilities and golf-courses, a university, private and public schools, substantial biodiversity conservation zones and a major public hospital. Snobs do contemptuously describe 'suburbs' distant from the CBD as urban sprawl but often residents don’t see it that way. Often we are doing well, thank you!

Congestion is an evil and should be priced but large sprawling cities might reflect rational choices by smart people interested in enjoying the good life. Urban intensification policies need to be based on externality considerations not snobbish resentment of the fact that ordinary people can enjoy the good life that was previously reserved for the gentry. There is no a priori expectation that compact urban structures are the best way to organise our lives in cities.

Issues of conserving the environment and preserving nature are important in large cities. Indeed
conservation can be consistent with profit-seeking land development policy. But this is separate from the issue of having large 'sprawling' cities. Indeed, with significant attention to biodiversity conservation, cities might be more ‘sprawling’ than they otherwise are.

Land supply policies need to reflect marginal costs of development and the potential benefits from both more compact existing cities (if any) and the decentralisation advantages of developing new regional cities or cities in other lower cost environments. Extracting rents from land developers and from new home buyers should not be the basis for urban land-release policy.


10 comments:

FXH said...

I don't have any current figures on hand but my memory is that "most" people in the outer burbs travel less than 10 ks to work. And about teh same ina circle for other activities like school, sport, shopping etc.

The CBD and inner suburban types indulge in the fantasy that everyone commutes into the CBD.

I once worked, for a brief, but glorious time, in a relatively outer suburban council. There was a huge section of people in the area who could measure the last time they had visited the CBD in years.

hc said...

fxh, yes I think only 7% of journeys in Melbourne are radially directed towards the CBD.

The myth of primarily radial travel has lead to bad traffic planning.

Anonymous said...

There was a nice article in the Weekend Fin Review on 5 August by Robert Bruegmann on urban sprawl. He has wriyten a book on this topic (Sprawl: A Compact History).

I once lived in a McMansion because with 4 children and two dogs (and wanting to have a home study) I needed the space, and could afford to buy one.

Anonymous said...

As I understand it, the idea that bad government policy has resulted in overinvestment in housing is pretty much a consensus one amongst Australian economists.

However, given the ridiculous levels of hock people have put themselves in to buy these overvalued palaces, it would seem to be rather politically problematic to unravel the situation now by removing CGT exemptions and suchlike.

Have you put any thought into how the mess might be unravelled without the government attempting to do so (to give you some emotional stake, make it a Howard-led Liberal government) getting crucified by mortgage belt punters, for whom high house prices are music to the ears?

I'll leave the issue of the politics of sprawl to a post on my own blog, which has been rattling around in my mind for a while now.

hc said...

Sinclair, I cite Bruegmann in the link. It is very good. I also like O. Gillman, The Limitless City which presents a more traditional anti-sprawl view.

Robert, No simple answers on that one - it is tough. Gradually releasing more land will gradually depress land prices. But taxing the unearned income benefits from the family home while sensible from the viewpoint of economics is definite electoral suicide.

I'll be interested in your views on sprawl and recommend Breugmann for a informative discussion of the politics.

FXH said...

hc - do you have figures on what % of house package price in say, Berwick, Pakenham, Caroline Springs is land cost?

[net of any infrastructure contribution /levy/surcharge/tax/dead hand of gov fleece]

Anonymous said...

I'm only in third year, but I've been convinced for several years now that land scarcity is the key driver of elevated house prices in Australian cities.

One other idea I think isn't given enough thought is that if consumers actually value infrastructure such as public transport, roads, schools etc. which are only accessible to people in certain regions, then the value of that infrastructure not captured by user charges will be capitalised into house prices. this has interesting equity connotations - if the value of infrastructure isn't captured through user charges of some kind, then it is effectively a government gift to the wealthiest members of our society (land owners).

Ed Glaeser (Harvard) has written a fair bit about this stuff - you might like to look up some of his work.

hc said...

Fred thanks for that. Bob Day's piece is closely related to the Alan Moran study available online that I link to. I am reading Moran now - its a long piece.

Isn't the preference for user pays designed to reduce the government's need to finance? Do you mean government spending on big capital items such as new schools and so on. If the latter then like you I'd favour debt.

Still if schools are underutilised in established suburbs you might want to send out immediate price signals that would encourage more compact settlement.

hc said...

fxh, The Moran paper (page 60) suggests land costs in Sydney have increased from 32% of total costs of housing in 1976/77 to 62% in 2005.

For Melbourne figure in 1973 is 33% in 1973 to 49% in 2003.

Growth in land price component was lower in Melbourne than other capital cities.

Anonymous said...

Ian Macfarlane seeks to deflect blame elsewhere. The structural overhang of increasing foreign debt has been the result of approximately one decade of inflation of the money supply. Inflation of M3 has been running around 10% for years, leading to mis-investment, encouraged and abetted by banks with increasingly lax lending practices. It will end badly just like the last "banking crisis". The sheeple go along with it because they trust the banks and politicians. The air will escape from real estate - one way or another.