Debt levels may be high, but low interest rates mean that households are no worse off today than they were in the years up to 2003 in terms of servicing a mortgage

Among the comments in yesterday's story about the current low levels of New Zealand bankruptcies, was the observation that our debt-to-income (DTI) ratio is close to 170%.

The implication is that a DTI measure is a better indication of household economic stress than bankruptcy levels.

And that is fair enough.

But there is a broader perspective behind the DTI data.

The raw data is available on the RBNZ website, here.

This source also gives data for the average mortgage rate, which can be added to the standard RBNZ chart perspective like this:

As mortgage interest rates fell, the household debt-to-income ratio has risen.

Low interest rates allow buyers to bid up asset prices while keeping their servicing costs "under control".

This DTI is the relationship between a 'stock' (household debt levels) and a 'flow' (household incomes).

But it suffers from the same problem that "median multiples" suffer from. They are measuring an economic relationship, but not one that relates to how households actually assess their budgets.

The core affordability assessment is between take-home pay and the cost of making the regular payments (both 'flows'). That is quite different to relating take-home pay to an asset price, or a total debt load.

Serviceability is the key here.

And the RBNZ's data also includes that: "Servicing as % nominal disposable income" (sic).

So, what do we have here? It clearly shows 2016 serviceability at the same levels that existed in the decade plus between 1991 and 2003. The anomaly was 2004 to 2010.

This supports the idea that in 2017 household financial stress is not elevated.

This RBNZ data is not available regionally. Nor by income deciles. Both these aspects may well paint a different picture to the overall, national perspective.

We welcome your help to improve our coverage of this issue. Any examples or experiences to relate? Any links to other news, data or research to shed more light on this? Any insight or views on what might happen next or what should happen next? Any errors to correct?

We welcome your comments below. If you are not already registered, please register to comment or click on the "Register" link below a comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current Comment policy is here.



If I could rephrase the title, it should be "financial stability of the nation entirely dependant on record low interest rates". Wow

Boom. I'd add to that, increases in house prices are largely reliant on ever lower interest rates.

The central banks of the world reached for their text books during the GFC and said lower interest rates increases demand and investment. But in reality, all it did was boost asset prices and largely unproductive asset classes such a property. Net effect is lower growth as the debt burden has to be eventually repaid, inflated away or defaulted on somehow.

The other effect of low interest rates and the consequent increased propensity to borrow is a massive cash injection into the economy. Last financial year we bunged 28 billion of new money (through private sector debt) into our economy - one dollar in nine of our gross turnover (GDP). A significant moderating of that, never mind it actually stopping would pretty much break the economy with some very serious consequences: asset values collapsing, high unemployment and immigration flows reversing.
The B of E are quite concerned and their debt ratios are less than ours.

With all of this massive amount of cash flowing into our economy you would think we couldn't have it better.
What is really going on?
"During the past nine years, the percentage increase in unemployment, underemployment and underutilisation has been two to three times higher than the increase in employment.
Let's take those one at a time.

For the whole of New Zealand, over that period, the number of people unemployed has increased 49 per cent.

Then there are the underemployed. Those are the people who are employed for fewer than 30 hours a week and would like to be working more hours. Their number has increased 61 per cent.

And there are the underutilised. This is the grouping of people who are unemployed, underemployed and in the potential labour force. They have grown by 38 per cent.

Welcome to the precarious proletariat, aka, the precariat; the growing group of second-class citizens who struggle to get a decent job, a decent wage, a decent life."
Really interesting article BTW

Those sound like GFC numbers. What are the numbers like since 2013 ( to try wash out the effect of the GFC ) ?

I found your comments interesting. So I went back and looked at the Statistics NZ HLFS data on this.

Taking them one at a time:

The number of people unemployed has risen from just before the GFC to today from 93,600 to 139,100, a rise of 45,500. In the same period, the number of people employed full time has risen by 325,300 to just over 2 mln for the first time ever.

The number of people unemployed and underemployed rose by 95,000 from Q1 2008 to Q1 2017. The number of people employed (full and part time) in the same timeframe rose by 400,000.

So yes, the "precariat" has risen, but those employed have risen by more than four times as much.

I am not minimising the issues of people who struggle to find a well-paid full time job. But in this case focussing on the % changes of them without looking at the much. much larger numbers who do get such jobs, is a distorted view.

You can argue that the gains haven't been enough, but it is not right to imply that we have gone backwards in any way. Besides, since Q1 2008 our full-time unemployment rate is unchanged at 3.6%, while our part-time unemployment rate has risen from 6.4% to 8.6%. It is people wanting part-time work who are finding it harder to get, not people who want full-time work. And further, our participation rate has risen from 67.7% to 71%, indicating more people want to work - and are succeeding.

Regionally, Wellington has the highest employment rate, but the most gains - and by far - have been made in Auckland employment.

Thanks David, you are quite correct but it does raise a big question mark over the supposed need for the level of low skilled immigration and highlights the numbers of our people that have become marginilised; non participants almost. That side of of it really concerns me; I know it's their own fault in many ways but it doesn't help when you've had decades of an open handed welfare policy then compound the dependency problem by undercutting the availability and rewards of work with mass low skilled immigration. I just can't imagine what these figures will be like if/when we have a significant recession with this big increase in the labour force.
That section is copied from the Herald article so not "my figures" exactly, they are correct as far as I can tell a little "cherry picked" though.

As a rule of thumb we (society) accept an unemployment rate of under 10%

Using your numbers it requires a 48% increase in unemployment to achieve an increase in the employed number of 19%

Visualize a plane that can fly no higher, engine running flat out and the fuel gauge is broken. Honestly, this party is over. The hangover sets in once global interest rates normalize at the same time as our next cyclical downturn!


The Q is, a) why would interest rates "normalise? and b) what is normal? Japan has been at almost 0% for what? 3 decades with no end in sight? (yes no end in sight)

Going back to a) this assumes a grow for ever exponentially on a finite planet scanerio, only works if you are a flat earther.

"no end in sight" this is only being achieved with in effect stagnation of our global economy and cash injections by every central bank that matters to keep it that way.

The real hangover is when US shale oil output drops (and when is an interesting Q) and the mbpd output starts its inevitable and irreversible decline to 0 by about 2050. The downturn then will be forever.

"The Q is, a) why would interest rates "normalise? and b) what is normal?"


The trillions of $$s printed out of thin air and dished around all over the world will be collected back albeit slowly through interest rate incremental increases ( slowly and smoothly) without shocking world economies ( obviously banks and mutual funds have also used the money in the same asset classes as everyone else) so enough time must be given to dispose and get out from these into something else like Shares and Bonds ..Janet Yellen let that cat out of the bag in her testimony in the congress lately ...
Ever asked why the dow jones and S&P500 indexes are so overvalued when all the fundamentals and indicators show they are well overpriced...yet The CBOE Volatility Index (VIX) made a new low of 9.30%, indicating record low levels of stock volatility.

So there is some hidden confidence in the market and The old book of fundamentals has been locked in a closet until the effects of the GFC is slowly watered down to maybe another mild recession.

Steven - it is a good question, in fact the major question that borrowers need to consider. Best guess as to where is normal now is probably the RBNZ's estimate of what the neutral OCR rate is (not too hot not too cold) - yesterday the RBNZ stated that they see that at 3.50%. That's good news in that they saw it at around 5.25% 10yrs ago before the GFC, and so now a good 1.75% lower.

So what does that mean ? A normalised floating mortgage rate might up around 7.50% if bank funding costs remain the same, and the popular 2yr rate around maybe 6.5%. However, when rates rise I suspect deposit funds will return to banks and their funding costs will reduce somewhat so I suspect we're probably talking closer to 7% floating and 6.00% 2yrs. How long to get to neutral is the other big question, but also if they actually need to move past that to "tight conditions" aka 2003-2006, where at those levels and based upon past relativities, rates would be closer to 8-9%. Of course we're a long way from a move to neutral, and nothing to suggest tight conditions past that yet - that would required some surprise late spurt in inflation that no one can see in the near-medium term future currently, but then, economic surprises aren't that uncommon and its the risk that borrowers need to be mindful of longer-term - the danger is foremost for those that claim a central bank won't take rates to levels that hurt, when in fact that 9 times out of 10 its central banks doing so that triggers a recession (and that includes 2003-06 here as it was the 8.25% OCR that triggered our one, not the GFC).

This is really interesting, so should we stop nagging older owners who bought at lower prices but carried similar serviceability stresses

Does New Zealand include student debt in its measures of household debt. If not why not, as comparisons to others including Australia are simply understating our problem.

Yes it does.

Thank you David. The numbers are there once one looks deep enough.

Where do you find the numbers. I found student debt (from statistics NZ) and the total non bank personnel loans (RBNZ) but that can't be where the figures are hiding because the student loans are well over 15 billion but loans in that category are only 4.6 billion.
I know that student loans and advances to beneficiaries are listed as a government asset but where is the contra showing up?

Another line should be drawn on the chart. Home ownership rates . In 1991 73.8 percent , 2001 68.8.percent 2006 66.9 percent 2013 64.8 percent, 2017 62.0 percent. What is household debt , and what provides the largest percentage of household debt . In percentage terms those households with mortgage debt is decreasing, whilst those who are mortgage free is also decreasing. If the stock of household debt is being used without adjusting for the change in those holding the 'stock' then what are the true adjusted numbers.

Could be the 70,000 new arrivals each year dont want to buy and are happy to rent?

The data that will be interesting, which we do not have yet, is how much the debt servicing % will increase with interest rate increases. I suspect that interest rates will have a higher impact than in the 2004 to 2010 period. It's a matter of wait and see (assuming interest rates continue to climb).

They are measuring an economic relationship, but not one that relates to how households actually assess their budgets.

Well yes, but if you actually put that into practice, do households really assess their housing costs budgets into two components (mortgage interest to be paid and principal to be repaid)? Do we make an assumption that households only consider interest payments (therefore their stress levels are no different to some point in time in the past)? I would think that most households think about Total debt to be repaid and the serviceability time comparison is somewhat of a red herring, much like propaganda used by central banks and property publications.

Repayment deficiencies seem to be rising (or at least are the highest they have been in this quarter);

RBNZ define as "Loans where actual repayments were less than scheduled repayments. Records the amount that actual repayments were less than scheduled repayments (interest and principal)."

David, I get what you're saying. However, the issue is not whether 2017 is currently in financial stress, but rather, what is the risk of heading that way.
The RBNZ FSR May 2017 had this to say

"Many borrowers are estimated to be vulnerable to higher mortgage rates (figure A2)".

It goes on to point out that 9% of all current mortgage debt would have mild (2%) or severe stress (7%) if interest rates hit 7% to even service their loans. 7%, the recent average - would see $24b in mortgage debt under stress. It's worse for recent borrowers. That's a pretty big system risk!

DTI is an accountig nonsense because it compares a liability "D" with cashflow "I". Ask any accoutant, you should never mix a "Profit & Loss" statement with the "Satement of Financial Position"
This is shown spectacularly well by the 1st graph in this article, there is NO correlation between Debt and Income (because it misses "E" = expense, mostly interest in this case and "A" asset).
To put it more simply, let's say we agree that a DTI of 5 is good. What happens if interest rates double ? Well the "D"ebt and the "I"ncome are still the same, so the DTI hasn't changed... so we're all good ? Yeah right, if interest rates double we are in deep deep recession

Vote for the many (most kiwis), or for the few (heavily leveraged specuvestors).

thank you David Chaston - great outlay of a confusing issue, nicely presented

In my line of business I know a huge amount of people. I believe this article and the stats in here are not reflecting what is actually going on out there. A lot of people are struggling week to week and many dual income families are only just scraping through. Wages aren't moving but the cost of housing, rents and living is going up. As the banks turn the money taps off, interest rates rise and people's credit tighten, we will see things change in the next couple of years.

Saw an interesting chart yesterday showing real wages were significantly higher in 1978, which could be one factor in the higher home ownership rate.

Like to see that. Can you supply a link?

I have this link currently:

I am trying to track down the original source, so will return and post what I find.

I understand it's Graph 5.4 in “The New Zealand Economy: An Introduction” by Ralph Lattimore and Shambubeel Eaqub.

I would put it another way a false sense of security as wages are stagnant high debt levels given low interest rates....need a slow down in the economy and job loses like the engineering firm hit the wall and then you will see mortgagee sales go up. There is a reason why banks have bought in the LVR measures as they are nervous about extreme house price growth....reverse gear is starting to take hold.