Mortgage stress numbers seem implausible

Another month has gone by and mortgage stress continues to grow, according to Digital Finance Analytics as reported in The AFR:

“Many households in stress have less hours worked than they want, and no income growth. Bigger mortgages by first time buyers and equity drawdowns are lifting repayments as lending standards ease and more interest-only loans are being made”…

According to the below table, around nine in 10 households across Stanhope Gardens and Bidwill are mortgage stressed:

Mortgage stress by suburb

These numbers are implausible.

I do not understand how mortgage stress can continually have risen since the pandemic began.

Mortgage rates have tanked, as illustrated clearly in the next chart from CoreLogic:

Australian mortgage rates

Cheaper mortgages

Household disposable income has risen sharply on the back of massive stimulus, experiencing its strongest annual rise since 2008:

Real household disposable income per capita

Real per capita household disposable income has broken a decade of stagnation.

The ratio of debt repayments to household disposable income has fallen to a 17 year low, according to the Bank for International Settlements:

Australian household debt repayments

Australian household debt repayments have fallen to a 17 year low compared to disposable income.

Finally, Australian households have built up a massive war chest of savings, which augers against the “stressed” theme:

Australian household savings

Australian households saved a record $200 billion in the year to March 2021.

Therefore, the aggregate data points to declining financial stress since the pandemic began.

What am I missing?

Unconventional Economist


  1. MathiasMEMBER

    Broken spirits?

    He must be doing something funky with the Mortgage Stress formula.

    I dunno. I dont even watch his stuff anymore. Havent watched a single one of his shows in many months.


    Is there a way to separate for high income households? I’ve a hypothesis that much if the disposable income boon is for higher income households who already have a good handle on the mortgage/household costs. And those mortgage stressed are those who were already house-poor and remain so. Ie. They’re not saving the money from the overseas trip they can’t make, because they were never going anyway.

    Also those better off (ie. With better LVR, income security) are actually in a position to negotiate/take advantage of cheaper rates. Thus those who were already battling with poor LVRs and credit scores would struggle to renegotiate with current lender, let alone find another loan.

    Average variable rate is 3.94% according to finder. Seems quite high. Need from 60-80 LVR to get the sub 2% fixed-rates.

    • Bingo! Many white collar workers in Sydney are back on the maximum bonuses of $50-300k this year – with mortgage rates lower, and discretionary spending restricted, the wealth effect is extremely concentrated in the 10%. I’d say the mortgage stress figures aren’t plausible, but you could equally argue that 90% of the working population are no better off than a few years ago – except for their house and share prices, which will always stay permanetly high, and their incomes are steadily rising with no other cost of living pressures apparent over the next decade, so yeah…. no worries! Equity mate!

    • No one pays standard variable rate unless it is bridging finance. Standard variable rate is only a reference point for applying your negotiated discount on.

  3. I mean that’s on average right? Maybe some specific areas are in greater stress but the overall the picture is different.

    • mrsnrubMEMBER

      Yes. A quick google says there are 3,800 suburbs in Australia. The table presents data for the 11 worst suburbs for mortgage stress, or <0.3% of suburbs that very clearly aren't randomly sampled.

      What reason is there to expect that these 11 suburbs in any way reflect the aggregate?

      I don't think you can conclude that mortgage stress is rising based on that table… I mean you wouldn't conclude that mortgage stress is on average at 75% Australia wide. And equally, the aggregate data doesn't make the table implausible.

      • From my understanding don’t they survey 50,000 people? This means about 13 households per suburb. At an aggregate level, numbers would be okay and at a segment level I.e. different regions also should be okay. The issue is that when you apply the survey at too fine a level the variance would be quite large and hence unreliable to make too much inference from.

        • mrsnrubMEMBER

          I’m not questioning the survey methodology.
          Even if they surveyed everybody in the country – why would the 11 worst suburbs be representative of the whole country?

          That may be what you mean by applying it too finely – I am calling this another sample of the population essentially. I don’t think the variance is the only issue in doing that. Seems to me whoever made that table has taken a (probably) good sample, and cherry picked a biased sample. Then drawn a conclusion about the whole country. I’m not surprised it doesn’t match other data that would be expected to represent the country.

  4. Know IdeaMEMBER

    Maybe those specific suburbs are exceptionally attractive to purchasers who are highly susceptible to experience mortgage stress?

  5. Perception. Few people feel unstressed with a massive mortgage, that sense of obligation and “what if” scenarios always running through their heads. Affordability and spreadsheets tell one story, but its unemotive and logical. People are irrational and emotional and a big farK!n mortgage freaks people out.

  6. Do we have an analysis of what is included in “household disposable income” & do we have a breakdown of how it has increased so much?
    I don’t know anyone who would say their disposable income has increased massively.
    What am I missing?

  7. Jumping jack flash

    “Household disposable income has risen sharply on the back of massive stimulus, experiencing its strongest annual rise since 2008”


    “Finally, Australian households have built up a massive war chest of savings, which augers against the “stressed” theme:”

    Yea, nah!
    At worst its all fake numberwang. At best its early-access super (and possibly other stimulus) that’s still being hoarded and added to existing money saved up for paying massive deposits for houses. Remember, while deposits are a percentage of prices, as prices approach infinity, so do deposits.

    But on the off chance that it is real and people actually do have massive wads of cash in the bank and more disposable income, then that’s great because the global CPI kick is coming, but we needn’t be afraid of it. Just endure it and wait for the wage increases as a result of it. Everything will turn out fine.

    Even though Scomo and Joshy-boy stuffed up our stimulus rollout, we should all be ok considering all this cash everyone supposedly has…

  8. brett sensman

    A per capita measurement says nothing of the distribution, if all that increase is in the top 10% by wealth it won’t move the level of mortgage stress at all, given they are in the bottom 30? 50? % of income predominantly.
    Maybe go find a number for the % of households that have increased rather than the per capita average.

    • Martin_DFAMEMBER

      Correct – is a distributional question. (Which few want to talk about).

  9. No one has ever questioned DFAs modelling. I do. For example, it is INCOME based and makes no allowance for access to assets.
    Don’t be bashful about interviewing Mr North over the apparent discrepancies!

    • innocent bystander

      I did, back in the beginning. It didn’t add up to me, the survey or the modelling (as opaque as the details were) so I discounted the analysis.
      On top of that the long winded videos that seem more about ego than anything else I gave DFA a miss.
      In the beginning the DFA summary suited MB’s themes so I don’t think they questioned the process too much.

  10. Martin_DFAMEMBER

    Pretty simple. The gross averages other cite do not sufficiently differentiate between different household positions. While some are doing well, others are not. The main rise in stress this time was rental stress, thanks to end of protections, plus lifting rents. And finally, we look at total outgoings, in incomes, not just the mortgage – as incomes are flat, and costs rising, the bigger loans are biting many. No changes to the DFA formula.

    • BabundaMEMBER

      What is your threshold for “stress”? I assume it is some % of income rather than a subjective

      • Martin_DFAMEMBER

        Its a cash flow measure – total money in, compared with total money out. If negative then in stress. Might have access to credit cards or savings, but over time, negative cash flow will bring problems unless there is a change. If with mortgage, and negative then in mortgage stress. Have to hunker down, and spend less.

  11. Your graphs and interpretation need some addressing here.

    Your last graph which takes data from the last release or looks more like midway through last year – consider the latest release and view it on the ratio for a clearer less bias picture. Latest ABS data as of June 2021 (yours is till 2020) shows it has collapsed straight back down.

    Again – house hold income stagnated between 2010-2020, house prices more than doubled – yet repayment ratios went down 17% to 14% – that right there is self negating clap trap. Literally impossible.

    Your first graph looks impressively precipitous declines – but the reality is that in all those falls the most is around 1.5% – which is on the scheme of things – very little change.

    Finally your indices on the left for the second graph – real house hold disposable income shows a rise of $2,000 / year – with housing price rises of as this very website has pointed out 25%.

    So $2000 increase in wages – is $38 / week – seriously not much when monthly repayments have gone from $4,400 / month on a million dollar loan to $5,500/ month.

    Thats why.

  12. ‘”Therefore, the aggregate data points to declining financial stress since the pandemic began”

    You’ve got the answer to your question right there in that statement Leith – “aggregate data”

    Dis-aggregate / segment the data to expose the marginal mortgagees who are suffering, remembering that there is a long end of the tail on very modest incomes living close to or beyond their means. I’m sure Martin can do this for you.

    Also bear in mind cost of living has increased for those of us who live in the real world: groceries, transport, maintenance, utility bills have all gone up, and at a greater clip in the last 12 months.

  13. BabundaMEMBER

    Leith you’re missing nothing except maybe an understanding of DFA’s methodology which I suspect is very shonky. The bottom line is that it hasn’t been this cheap to get into a mortgage for a very long time. Because interest rates. This is why prices are booming everywhere. Every other data point is completely at odds with these “stress” numbers. North clearly has a political agenda and one has to wonder if his numbers are now pure propoganda

  14. My kids have researched buying a house to live in. They’ve (& they’re friends) have all 100% decided NO. TFH, not gonna do it even though we can’t leave AUS for UK, USA ETC, they’ve all decided property is toooo big a risk!!!