More on Housing Finance

As H&H said earlier today, the 5609 dataset from the ABS has been updated for June. We actually need to wait until tomorrow to get the whole story on housing credit because the 5609 dataset has a sister dataset 5671 which contains the investor data.

In the meantime here are some drill-downs on the owner occupier data.

Firstly total owner occupier finance by state, which does not include refinancing. Notice the medium term trends for all states are holding.

Construction of new dwelling by state, which clearly shows Victoria building boom continues

Purchase of new dwellings.

Purchase of existing dwellings.

And finally re-financing.

So it appears that owner occupier finance towards new dwellings is up in NSW and Victoria, while credit towards existing housing is relatively flat or falling across all states. Queensland continues to look atrocious and once again takes the wooden spoon. I am hoping to see a new “stock on market” report from SQM shortly which should help in determining how this latest data will translate into price movements.

Latest posts by __ADAM__ (see all)


  1. Why is that in Gavins graph the blue line is below the 2008 minimum while none of the above graph suggests this is the case?

  2. indo: The blue line isn’t quite “below the 2008 minimum”, although the purple one is.

    But the explanation is that lending (measured by value) has been divided by GDP in order to make long-term comparisons meaningful.

  3. Hmmn.. Think in future we need to hold off updating the charts until media release released!

    Anyway, we live fast changing times.

    With regard to the housing finance results, you might wish to consider the NUMBER of loans approved excluding refinancing as apposed to the VALUE.

    When one does that, there has been a slight lift for the month. Certainly, not enough to call an end of the downturn, and indeed it seems the ABS revised their May numbers down.

    That’s what we have been following for many moons now. It enabled us to call the downturn early on in 2010 when others were still talking prices to the moon.

    There is some logic following the number as apposed to the value. Because the number captures the volume of demand.


    • >With regard to the housing finance results, you might wish to consider the NUMBER of loans approved excluding refinancing as apposed to the VALUE.

      Actually I am more interested in the amount of new money being created in the banking system than I am about numbers in these stats. The numbers of loans are useful for the reasons you mention but doesn’t give you the “debt stimulus” for the economy. As I stated this morning

      The credit dollar value is all you need to predict trends in prices.

      • For louis ofcourse the numbers are more important because in a way by proxy it represents the demand for his services. Value is immaterial in his case.

      • Your analysis on the correlations between value and pricing is very good. However, in theory at least, it could suffer from a skew. That skewness being fewer buyers taking out larger loans; presumably as they are buying at the top end of the market. Or vice versa.

        You believe this skewness might be negligable? Maybe. But if we are to take the results at face value, there were more buyers in the market in June compared to previous month. Perhaps with the value falling, these buyers were weighted towards buying middle to lower end real estate.

        Anyway, just a thought. Not folding the envelope either way.