Major city mortgage stress by suburb

Via Martin North today come some cool mortgage stress maps. Sydney:





And Cambra:

Let the envy begin.


  1. thomickersMEMBER

    shame inner melbourne isn’t dark blue. I know a dude who has 8 properties around the city centre and very stressed.

  2. People who decided to live in those dire suburbs have only got themselves to blame.
    As the great lifter, Smokin Joe said: Get yourself a better job. He did. And a better postcode.
    Make mine a cuban

    • Ballarat and Toowoomba. Don’t go there after dark in about 2 years time.

      Actually don’t go there in the daytime now, either

  3. Mining BoganMEMBER

    Hands up all that looked at their area first and gave a little fist pump and yessss…

  4. JosephSMEMBER

    Not much blue in Adelaide. Though it looks like this is a nominal number per post code? You would think a ‘x in stress per thousand’ measure would make more sense, otherwise a suburb might show up blue simply due to higher density of households…

    • blindjusticeMEMBER

      First off Martin North is a legend. But yea its nominal, even if it were mapped to an ABS statistical area the nominal would then be comparable mapped like that. Mapped by postcode isnt a good method for nominal like you say x per or % would do it. The deeper you drill down into this kind of data the better you can model it. Geography type: Statistical Area Level 2 (SA2) (2016)
      Name: Virginia – Waterloo Corner
      Code: 402021035
      Map search it here
      Has a median household income of $949 per week, Median mortage repayments of $1603 per MONTH.So 42.2% of income on the mortgage. Thats mortgage stress for sure but the map isnt showing it. Now I dont know if Martin strips out the renters household incomes or those on welfare (any datasets i can access disallow this) to improve the model. It would be awesome to model asking prices/sales prices in statistical areas against the income per a typical household formation couple (census data allows this) as you can use the income per age group guided by the average age of a first time buyer and other profiles etc. Then you can model stress by purchaser group per statistical area……….dunno maybe Martin already does this, but through his surveys?

      ps some of my data is outdated – from the 2016 census as i`m just using publicly available data – Martins is up to date or is using public data adjusted for how the economy has changed since the census (i.e x multiplier for wage growth/interest repayments etc). Anyway the man is a total legend

      • Martin doesn’t do averaging because it gives false readings. DFA does households. Qualative

      • Denis413MEMBER

        As others have pointed out, it’s a nominal measure only (i.e. not relative to anything). Needs to be expressed as a % for each post code, or at the very least, provide how many households are in each postcode so we can ascertain if the suburb is in trouble, or if it just so happens to be a large household, thus, having more of a likelihood of having households under stress.

    • Tassie TomMEMBER

      I agree about the “nominal number per postcode” thing. I suspect that a lot of the blue shading is in postcodes that have more people than the average postcode. They are large because there used to be a sparsely scattered population, but now that the urban sprawl has been spreading into them but their borders haven’t been adjusted they have a larger population. More population = more people in mortgage stress, all else being equal.

      Still, great graphs!

  5. Denis413MEMBER

    Does a link exist to help understand how the mortgage stress values per suburb were calculated?

    • Yep, would be helpful

      Also should state the sample size per city if north wants to be taken seriously.

      Tards dont care though, they just read the headlines

      • innocent bystanderMEMBER

        like that headline the other day, yesterday? some suburbs have dropped their price 40%, I drilled and drilled thru the articles – no data to support it whatsoever.

  6. RussellMEMBER

    Keep watching those graphs, love to see a before and after in 6 months time. These stress levels will seem mild.

  7. Love your work, Martin. However, postcodes are not of standard population size. e.g., postcode 4350 serving some 100,000 people in Toowoomba and the surrounding area.

    So Melton Vic may or may not deserve its status as a dark blue and the small inner city postcodes likely understate the STRESS households are under as their balance sheets deteriorate and a larger and larger proportion go into negative equity.

    You do NOT want the agony of negative equity. Ten years after the GFC, ~20% of US households with mortgages are still in negative equity. Do you have a spare decade to just throw away?

    Don’t Buy Now!

    • Interestingly, people who didn’t buy at the bargain prices of 2009 have thrown away a decade.


      • Not unless they did nothing with their money. Choices. Something those who bought recently wont have any of.

    • Yeh maybe he should plug in ABS statistical areas- shapefiles etc are avail on their website. Also some postcodes can include two distinct and seperate areas, don’t ask why but set up for postage not population research. Average punter won’t care though. Would allow him to plug and play a bunch of other mappable abs stats though which is always fun.

  8. and the economy just started to slow down. no job loses yet, just some OT and Commissions loss. Obviously not as easy to refinance and consolidate car, holiday, reno and home loan into new single home loan plays big part too.

  9. Every time this comes around I make the same comment. Camden (2570) has eight times as many households as Lindfield (2070), so it is obvious that the absolute number of households in mortgage stress is going to be higher in Camden than in Lindfield. The picture would be far more meaningful if it showed the number of households in mortgage stress divided by the total number of households in each postcode.

    • True 2570 covers a vast range of suburbs, from old established Camden South to new build youngens in Oran Park.

    • innocent bystanderMEMBER

      you can point this sort of stuff out endlessly. they ignore it.
      then the question is: why?

    • Canberra suburbs are tiny and each postcode would be probably 1/4 the size of Perth urban postcodes (for example).

      So Cambra is always gonna be green / yellow.

  10. 2155 and 2154 Kellyville, Rouse Hill and Beaumont Hills were darker blue on previous maps, what happened? Or have the colours and number of stressed households changed so that now Campbelltown and Liverpool have way more households stressed?

    • those are the growth areas. so I was told until about end of 2017 start of 2018. growth in stress perhaps.

  11. kannigetMEMBER

    2605 consists of 3 suburbs.
    Curtin, population 5200 with about 32% purchasing
    Hughes, Population 3000 with about 33% purchasing
    Garran population 3200 with about 27% purchasing

    Thats roughly 3600 people living in a residence involved in a PPOR mortgage

    so what, about 2000 mortgages? looks like a nasty situation if between 1000 and 1500 mortgages are in stress…..

    Some of the other suburbs are worse still, 2611 is a large areas but most of it is pine forest or farmland.