Housing bubble? Not here.

It’s nice to see that Jeremy Grantham has rolled into town this week and stirred up the property bubble argument once more.  Mister Grantham, the co founder and chief investment strategist of US based investment management firm Grantham, Mayo, Van Otterloo & Co (or GMO), has been widely quoted during his tour across Australia saying that Australian property prices are in a bubble and need to fall 42 per cent to revert to trend (42 per cent, he’s even more bearish than one noted local bear who bet on 40 per cent falls).  Whilst in Australia Mister Grantham has also commented on Australia’s proposed Resources Super Profits Tax, the UK and Chinese Property markets.

Whilst one can’t argue that capital city prices are expensive and unaffordable for many, capital cities do not make up the entire Australian property market.  There is such a focus on the capital city property markets due to the fact that this is where the majority of Australian’s live.  Over the three months to April 2010 the median house price in Australian capital cities was $492,000 and the median unit price was $410,000.  When you combine all areas of the country including those outside of capital cities the median price falls to $425,000 for houses and $382,500 for units.  The median price for all dwellings (houses and units combined) is $414,000.  Clearly national median prices are heavily influenced by the capital city markets however prices are generally much more affordable once you move out of the capital cities.

The big difference between Australia and most other parts of the world is that we are a highly centralised society with only six states and two territories.  60 per cent of all home sales take place across 0.5 per cent of the land mass.  Based on ABS figures to June 2009, the capital cities account for an estimated 14,039,373 persons of a total of 21,955,256 persons nationally.  This means that about 64 per cent of the Australian population lives within the capital cities with an estimated 55 per cent of the population living within the four largest cities (Sydney, Melbourne, Brisbane and Perth).

When you compare this to other countries you can see there are fundamental differences not just in the size of Australia’s population but the fact that it is so heavily centralised.  As some examples:

In 2007 the USA had an estimated population of 302.2 million persons, the 20 largest cities in the country, in terms of population, accounted for just 10.7 per cent of the population.  The 50 largest US cities account for only 15.5 per cent of the population.  Sydney makes up an estimated 20.5 per cent of Australia’s population.

There are similar results in some European nations.  Although the data is a little older, 2001 Census data shows the United Kingdom had a total population just over 49 million persons.  Whilst London accounted for 14.6 per cent of the total population, the 20 largest cities accounted for 27.7 per cent of the population.  Sydney and Melbourne combined account for an estimated 38.7 per cent of Australia’s population.

In Germany, 2002 statistics show the estimated population was 82.4 million persons.  The capital Berlin accounted for 4.0 per cent of the country’s total population whilst the 20 largest cities accounted for 17.8 per cent of the population.

Finally, 2002 statistics for France show that the 20 largest cities in the country account for 11.7 per cent of the total population.

Getting the idea?  As a population we spread ourselves very thinly across the country.  Obviously we can’t all move to the desert in the middle of Australia but there are plenty of habitable regions of the country which are under populated.  64 per cent of the country’s population on 0.5 per cent of the land mass is completely nonsensical.   The capitals are also experiencing the highest increases in population, which creates a major strain on infrastructure in these regions and leads to higher property prices (particularly when new housing construction is in such short supply).  As an outsider looking in you can understand why international experts believe we are in a housing bubble but when you look beneath the surface the fundamentals of the Australian property market are very different here. 

We are also yet to mention the fact that Australia’s population grew at a rate of 2.1 per cent during the year to September 2009.  With an addition of almost 452,000 residents to the population during the last year where do you think they are going?  They go where the jobs are… and where are the jobs?  The jobs are where 64 per cent of the population is, in the capital cities.  More accurately, most jobs are in Sydney and Melbourne and to a lesser extent Brisbane and Perth.

Add to the mix the fact that unemployment has peaked much lower than forecast and continues to improve, the Australian resources sector is once again warming up, mortgage arrears remain well under 1 per cent and, as a nation, we remain undersupplied in housing by more than 178,000 dwellings.  With all this in mind, the outlook for Australian home values doesn’t appear to be anything like a 42 per cent correction.  Our view is that home value growth is likely to moderate, tracking household income growth which is likely to be circa 5 per cent over the coming year.

About Tim Lawless

Tim heads up the RP Data research and analytics team, analysing real estate markets, demographics and economic trends across Australia

Connect with CoreLogic

Enter your email address to subscribe to our e-newsletter, and have new posts delivered via email. You can also connect with CoreLogic on social media.

12 Responses to Housing bubble? Not here.

  1. Nathan Webb June 18, 2010 at 2:22 pm #

    The gist of your argument – that Australia has a much higher level of urbanisation compared to other countries – is based on flawed demographic statistics. When comparing Sydney and Melbourne to other cities, you aren’t comparing apples with apples.

    For the USA, a “city” is more akin to our Local Government Area. According to that definition, New York has a population (from memory) of around 2 million. In comparison, the New York Metropolitan Area has a population in excess of 20 million, making it one of the largest urban conurbations in the world. Likewise, the City of LA has a population of around 4 million, but the LA Metro area has more than 17 million people, made up principally of 4 counties. These two metro areas alone make up 11% of the population, more than the total of the top 20 “cities” that you referenced in your blog.

    Likewise, the London metro area is estimated to be almost 14 million (http://en.wikipedia.org/wiki/London), or 28.5% of the total population, which is much more than Sydney’s contribution.

    A fairer (but fundamentally flawed) comparison, using Australia’s LGA populations, puts Australia’s top 20 “cities” at around 25% of the total population, which is *less* than that of the UK.

    In other words, your argument sits on flimsy statistics. Personally, I agree with the idea that different countries *choose* to spend money on different things. In Australia, we choose to spend more money on living in a desirable location than other countries do. So a comparison of San Fran house prices to Sydney house prices is much better than (as the HIA did) comparing Sydney to a surburban wasteland like Atlanta or Dallas-Fort Worth. But even these comparisons are not very nice for Australians. House affordability rankings regularly show Sydney and Melbourne near the bottom, compared to SF, London, New York, Honolulu, etc…

    For a company that prides itself on accuracy, this statistical bias is disappointing to see.

    I would be very interested in seeing a far more robust assessment that, instead of cherry-picking population figures, actually looked at key world-cities, and compared incomes to house values from a long historical (e.g. 50 year) time-frame. Has there been a divergence in the last 30 years, as the property bears claim? If so, why, and what does that say about any possible correction?

    • RP Data Research June 18, 2010 at 4:22 pm #

      Hi Nathan

      Thanks for your comments.

      The data we have sourced is from relevant Census Bureau data. New York for example has a population of 8,274,527 however, of the 50 largest cities none of the others detailed are within NY. In California, LA comes in with a population of 3,834,340 with other California cities including: San Diego (1,266,731), San Jose (939,899), San Franciso (764,976), Fresno (470,848), Long Beach (466,420), Sacramento (460,242) and Oakland (401,809). Of those detailed only Long Beach could be considered part of the Los Angeles Region.

      In the UK London is detailed as having 7.172 million residents whilst Birmingham is the second largest city at 970,892.

      I can assure you nothing has been cherry picked and we have used the best data available which we can find which is why the best data we had for the UK was from the 2001 Census. We believe the facts remain that most other countries are extremely decentralised whilst Asutralia is not and this clearly has an adverse impact on housing affordability within capital cities.

      As for the long-term analysis unfortunately reliable Australian housing data is not available pre-1980 and international housing data is very difficult (and costly) to come by. We would love to be able to do that sort of analysis however, the data availability and acquisition costs to undertake such an analysis are extremely prohibitive.

      • abner June 19, 2010 at 3:16 am #

        The US Census Bureau also defines a “primary census statistical area” (PCSA) as a defined metropolitan area that is not a component of a more extensive defined metropolitan area. PCSAs give a more accurate view of the population of contiguous urbanised areas than the politically-shaped boundaries of individual cities. The top 20 PCSAs have a combined population of over 136 million (45% of the national population); the top 50, over 188 million (62%).

      • RP Data Research June 19, 2010 at 7:50 am #

        I take your point and we possinbly should have used a PCSA. If the top 50 cities account for 62% of the US population they still account for a smaller proportion of the total US population than the 8 capital cities in Australia. With four Australian cities accounting for 55% of the population

        The point is with the population so centralised across such a small area of land it creates upward pressure on prices. Until jobs are sent outside of the capital cities long-term housing demand in capital cities is likely to remain strong.

        We believe the Government could start by decentralising the public service, in most instances they don’t need to be in the CBD and renting or owning in the CBD is high cost (a high cost to tax payers).

        Secondly, incentives should be offerred to businesses to decentralise perhaps a lowering of the company tax rate. To go along with corporate incentives, residents should also be incentivised to move outside of the capital cities. Some state governments do offer incentives but without the jobs to support a move out of the capital city demand is unlikely to be strong.

        Jobs first and then the human capital should follow. But we can’t keep sticking more people in our capital cities as Governments have proved that they are unable to provide the necessary infrastructure and are also unable to provide affordable housing.

      • Nathan Webb June 19, 2010 at 11:29 am #

        I understand where you sourced your data, and I’m not questioning the accuracy of it. I am questioning your statistical analysis skills. If you wanted to do a fair comparison (although rather meaningless) you should have used the Australian LGA population figures. However, does anyone care that Brisbane (LGA) has a population of 1 million, while Sydney’s is only 178,000? What does that even mean? What does that say about population density? What does that say about demand for housing for both LGAs? What does it say about centralisation?

        The answer to all of these questions is the same: Nothing.

        Alternatively, you could have sourced metropolitan area population statistics (http://en.wikipedia.org/wiki/List_of_urban_areas_by_population). I also deal with statistics as part of my job, and a key step that I do is asking myself if the results match up with my intuition. In this case, does it match with your intuition that LA, with 3.8 + 0.5 = 4.3 million people, has the same population as Sydney?

        But at any rate, it still tells you nothing about population density, decentralisation, housing demand, supply issues, etc…

        Cherry-picking was the wrong word, as it implied intention. However you have used an inappropriate comparison.

  2. Brad Gould June 19, 2010 at 12:03 am #

    I agree with what you are saying regarding Australia having a centralised population, making demand and prices higher in the capital cities. However eventually as the capital cities become more and more unaffordable and demand shifts to more regional areas there will be a few problems being 1. there needs to be jobs in these regional centres to attract people to start with; 2. there is insufficient infrastructure to cope with increased population levels in regional areas; and 3. when people actually do move from the city areas to regional areas, it’s because the property values in the cities have become unaffordable, which will mean demand will gradually shift from the cities which will then in turn lower values in the city (i.e. supply and demand, if demand decreases due to unaffordability then prices will drop until they reach a more affordable level again). I also think it is a little foolish to say that just because Jeremy Grantham is from another country with a different population makeup that he is wrong, and assuming that he hasn’t done his research properly and is only applying rules from different countries property markets to Australia. Don’t forget that he forecast the GFC, he runs a company that provides INTERNATIONAL investment advice, so it is his job to understand international markets, his company has an office in Sydney and he also didn’t only comment on Australia’s property market he also said that Britain’s property market is also in a bubble. Finally even if the shortage of supply remains which should theoretically keep values going up, it all comes back to affordability. You agreed that houses in capital cities are unaffordable which will shift the demand to regional areas, and even if the jobs are there it’s all relative, pay levels are lower in regional areas meaning that they can only afford Houses at lower values. Trust me I know, I live in regional Australia and just because House values are lower doesn’t mean that we are all flushed with cash because they are supposedly more affordable.

  3. OzExpat June 19, 2010 at 1:46 am #

    This posting is just plain misleading and is relying on completely irrelevant facts. Yes it’s true that most Americans don’t live in the big cities, but the big cities/metro areas in the USA have MUCH larger populations.

    So how does this explain Australian city house price to income ratios being justifiably higher than New York City, San Francisco, Los Angeles, Dallas/Fort Worth, Houston, etc. etc. There is no justification..hence the definition of a bubble!

    • david webster July 29, 2011 at 8:26 pm #

      Explanation is simple. A nation of speculators that has dumped using their brains to make a living in favor of becoming either a salesman [best in RE] selling a $3 product for $10 or why not $50 or a speculator. Other than that a small portion will dig holes and sell the dirt to smarter countries. A cultural Chernobyl this country has become, no doubt.

  4. David June 21, 2010 at 12:04 am #

    I don’t understand how population data makes any difference to claiming that there isn’t a property bubble.

    Where is the relationship?

    Look, I wouldn’t factor in regional areas being in the bubble. They are just irrelevant outliers or at least bundle or compare regional with regional. The bubbles are in the metro areas. I’m seeing properties being flipped within 1 month of being bought and then resold. This is not rational, its property trading with a hot potato. There will be a lot of burnt hands.

  5. tomas edward June 25, 2010 at 11:46 pm #

    Graham has rightly used the word ‘bubble’. Ours property policy is much different than that of USA.But, I think other people should come up with optimistic thoughts.

  6. Rob June 28, 2010 at 3:21 pm #

    The statement that the 20 largest cities in the country, in terms of population, accounted for just 10 per cent of the population is not directly comprabale to Australia because they define a city differently over there. For example in Boston they only count the people living in the actual CBD with the result that the population is given as under 1 million. However, the population of ‘greater Boston’ – which includes all the satellite towns immediately around the city – is 4.5 million. In Australia we count Parramata and Liverpool and Penrith and so on as being part of ‘Sydney’ while in the USA they would be counted seperately. In reality Boston and Sydney are about the same size.

  7. Francisco June 30, 2010 at 11:42 am #

    One very simple way to measure urbanisation is public transport. In Australia public transport is very poor because population density is too low for the size of the city. You can get around the other larges cities you mentioned without a car. Try that in Australia and it will take you a few hrs to get anywhere using public transport.

Leave a Reply to david webster Click here to cancel reply.

Notify me of followup comments via e-mail. You can also subscribe without commenting.