Is there a housing bubble?

Rismark International

In an excellent post, finance whiz Christopher Joye explains that all those foreign hedge funds who are actively shorting Australian bank shares are themselves taking a bath.

He deflates the claim of “US investment legend” Jeremy Grantham that Australia’s housing market is a time bomb with house prices 7.5 times family incomes. The hedge funds believe the housing bubble will burst and threaten the balance sheets of Australian banks.

Mr Joye, who is the MD of Rismark International, points out that the home price to disposable income ratio in Australia was actually only 4.6 in the June quarter 2010, a figure generally in line with the average since 2003.

He points out that unlike other estimates:

Rismark’s national home price-to-disposable income ratio includes dwellings in all regions (ie, not just capital cities), all property types (ie, not just detached houses), and the ABS’s quarterly measure of average disposable household incomes (ie, not just average weekly earnings), which captures income earned from all areas (eg, labour and investment income) and reflects the fact that there is typically more than one income earner per household.

Here Mr Joye explains how the international hedge funds are taking another bath because they’ve bet that interest rates in Australia are going to fall.

9 Comments on “Is there a housing bubble?”

  1. Joseph says:

    Alan, you may want to subject this analysis to more scrutiny before describing it as excellent, I would instead describe it as deliberately misleading. The first flaw is in comparing a mean with a median. Since Australian income distribution is far from equal the difference between the two is substantial. The second flaw is in using ‘disposable household income’. What does this include exactly? From a cursory enquiry this appears to be post tax, does a figure of $95k of post tax ‘disposable’ income per household sound right when an average wage in Australia is around $50-$60k? The third flaw is in the graph you post, you may note that neither axis goes through the origin, if it did you would notice that the best fit line would imply houses were free at an income level of $27k.

    The income of the median household comes from wages. There is perfectly good data on median house prices and median wages. It is entirely reasonable to divide one by the other and compare internationally.

    The methods used by Joye are standard techniques of those who wish to mislead with statistics and all done with an agenda of trying to convince that black is white. Australian houses are very expensive by international standards, no amount of statistical trickery will change this. It is unsurprising that someone in the real estate industry makes these arguments. What I am a little surprised by Alan is that you appear to have fallen for them.

    • Alan Davies says:

      Joseph, I’ve been following Chris Joye for a while and like the RBA and the finance industry have been increasingly impressed by his analysis, so I have’t linked to him lightly.

      That’s not to say I should assume he always gets everything right. The points you raise are important. Since they’re largely methodological issues that I’m not in a position to answer definitively (not easily anyway), I’ve written to Chris seeking his response.

      In the meantime, in relation to the graph note that it is over the period from 1993 to 2010 i.e. the first observation shows the disposable income/price combination in 1993 and the last one shows the same combination in 2010. I can’t see that the idea of extending the line of best fit to zero income makes much sense – clearly there’s an income “floor” below which people can’t buy into the housing market. Re your point about disposable income vs average wage, note that the “95K disposable income” refers to household income. Re the average vs median issue, I recall that he explained this a long time ago but I don’t know where (unfortunately there doesn’t seem to be an archive on his site) but hopefully he will explain that himself. You might want to have a glance at this which shows he is well aware of this issue.

      • Joseph says:

        My comment regarding the graph is just one of a number of criticisms you could make of it. Quoting an R-squared for a relationship between 2 series that tend to increase over time is nonsense and any statistician should know it. I can regress house prices against my daughters height and get a good fit. The point I make about the line not going through the origin is simply to illustrate that the relationship is meaningless. An often quoted relationship on house prices internationally is that they have tended to be around 3x income. Joye’s relationship suggests it is minus $161k + 6 x income. This cannot possibly be accurate for any extended period, the relationship is spurious.

        Regarding average v median, I don’t think we need a response from Joye I can probably guess what he will say – we should divide average by average which is effectively what the link you posted suggests. Now let’s imagine that Andrew Forrest’s income was $100m, does he live in a house worth $460m? I suspect not. If we are considering housing affordability for a typical Australian we should look at the median.

        There are a couple of things you could usefully get from Joye, firstly what exactly is included in personal disposable income, the treatment of tax, mortgage interest, income into superannuation accounts, government stimulus payments I think all come into play. It is complicated and, unless you are a government statistician, confusing. Secondly, how do you reconcile the ABS data on household income done on a bottom up survey approach with the top down national accounts based approach he uses.

        An easier way to determine whether 4.5x income or 6x or 7x income is more plausible is to play the game of guess your neighbours’ incomes. Estimate the value of a neighbour’s house, divide by 4.5 and then see if this coincides with your guess of the their DISPOSABLE income, in the vast majority of cases it won’t.

  2. Michael says:

    Without being able to point to anything specifically wrong with Mr Joye’s analysis, I still feel something is amiss. but if valuations are to be believed the houses in my street have been appreciating at about 18 to 20% a year for the last five years. So presumably three things could account for this:
    1. They were undervalued.
    2. There is a bubble driven by momentum or cheap credit etc..
    3. We are in a period of high inflation.
    What am I missing? There doesn’t appear to have been any significant increase in amenity.

  3. Michael says:

    A general point to add is that the data being used is spread across the whole country. How relevant is this pooling of data? What’s going on in a regional town in Western Australia doesn’t bear much relationship to inner city flats or vice versa. When people are concerned about house prices, they are concerned about house prices in a fairly narrow band of places they would actually be willing to live. I doubt there is a big percentage of the Australian workforce that is really mobile across the country as they exist in the dreams of market fundamentalists. This is the same technique people like Bob Birrell use. Stats are pretty malleable when you can arbitrarily draw the boundaries.

  4. Alan Davies says:

    Hmmm…..some good points here that give me pause for thought. No sign of Chris Joye so far unfortunately (the Melbourne Urbanist can get the likes of Edward Glaeser and Tim Harford to comment but not, perhaps, Chris Joye!).

  5. YB says:

    Joye’s excellence? Pass the salt.

    I’d suggest reading the thorough evisceration of Joye’s statistical prestidigitation by Atlas at the blog Critical Influence first.

  6. YB says:

    No worries, great job on the blog.

    While Mr Joye may be a well-informed commentator, his extensive efforts to correct / innovate (!) the housing price data sets are vaguely puzzling (Ockham’s Razor and all that jazz).

    Perhaps he’s simply seeking to “materially improve the quality of information and analytics available on the world’s biggest yet one of the least understood asset-classes – residential housing” as it says on his Rismark website. For what it’s worth it would seem that his company also directly benefits from the ongoing appreciation of the Australian housing market.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s