The Bank Stress Test We Really Need
Your normal weekend editor is away today.
Which has turned out to be a good thing.
Because I discovered this week, I have a whole lot to say when it comes to Aussies banks.
I mean, who knew something as crucial as a credit-driven economy could take up so many digital words…
Banks make less money
This week has been all about bank profits.
Or the lack of them.
With all the four pillars of Australia’s banks reporting their second year of falling profits, people have been asking questions, like, are the good times over for Aussie banks?
Let’s make it clear though, they still made a lot of money.
They just didn’t make as much money as they did in 2017.
But as I explained yesterday, lowering the availability of credit for houses will do that.
If we’ve learnt anything from the past two years, it’s just how much our banks rely on ever-growing house prices.
Nonetheless, in a bid to soften the blow of profits falling 7% compared to last year, ANZ announced they had ‘stress tested’ their assets.
I’ll be honest, I couldn’t think of a less believable claim.
For starters, most banks have blinkers on when it comes to something that could impact their loan book.
Take this for example.
ANZ — under their own vague assumptions — reckon that if house prices fell around 30% for a couple of years, that would equate to a total $2.7 billion loss to their loan book over three years…
Oh, but some of that would be recouped by claiming on lenders insurance.
Not to derail this, but if ANZ did make claims on mortgage insurers, that would lead to ‘systemic’ issues in the financial marketplace…
But back to ANZ’s self-analysis.
They also reckon that double-digit house price falls would mean we have unemployment climb from 5.5% to 10.5% over three years.
However, ANZ aren’t the only bank claiming to have gone through their loans with a fine-tooth comb.
CommBank tried a similar thing earlier in the year.
They estimate that a 31% fall in house prices would see a $4.09 billion loss to their loan book over a similar three-year period.
CBA also add, that their worst-case scenario is that the unemployment rate would need to be 11% for that to happen.1
Duck and cover
And this brings me to why I highly doubt any bank is creating their own standards for a ‘stress test’.
The past 12 months is a great example of this.
Unemployment has sat between 5% and 5.3%.2
Yet house prices fell…double-digit falls in Melbourne and Sydney in fact.
A Banking Royal Commission forced banks to air their dirty laundry…and bank profits have fallen because they now need to set aside hundreds of millions in retribution costs.
The regulatory body APRA suddenly decided to enforce the rules it had put in place four years earlier.
APRA’s decision to finally enforce these rules reduced how many interest-only loans a bank could make. Or how many investor mortgages it could issue. Plus it meant that banks had to adhere to the old deposit to loan ratios they’d ignored for so long.
APRA enforcing these rules contributed to bank profits falling this year.
The point is, none of these events were predicted by banks in 2017.
Yet, these very factors have significantly reduced the fall in profits we’ve been chewing over this week.
Their own vague stress test criteria is nothing more than a warm and fuzzy PR statement.
It has the same effectiveness as the ‘duck and cover’ solution to calm people’s nerves when a nuclear attack was a possibility during the cold war.
It’s impossible to believe their own internal stress test would see their loan books lose a little either side of a billion each year, if the worst was to happen.
Banks rarely lead the pack when it comes to calling market events. Often that’s because the employees that produce this data are shackled to an internal view.
The thing is, it isn’t what the banks reveal that’s the problem.
It’s all the things they don’t tell us that really matters…
Masking the real problem
The bank can stress test their assets all they want.
The problem isn’t that.
It’s what they don’t tell us that is far more telling to how vulnerable ‘their loan book is’.
As we continue to discuss their earnings, we keep getting tidbits about how small bad debts are.
The Reserve Bank of Australia even created a pretty picture to reassure us how small bad debts are in the Aussie banking system…
Reserve Bank of Australia — Aussie banks 90+ day bad debts
Source: Reserve Bank of Australia
This chart here, shows us all the people in Australia that are more than 90 days behind on their mortgage.
That is, a home loan that hasn’t seen a repayment — in full or partial — for three months.
As you can see, it’s barely at 1%…
…nothing like in the early 90s.
Then the media champion this, saying it’s proof we have nothing to worry about.
But what about the rest?
I mean the debts the banks internalise…
Just how many people are facing financial hardship?
Go on, guess.
And I do mean guess.
Because the truth is, we don’t actually know.
You see, Australian banks must report to APRA when a loan is 90 days in arrears with no payment.
Once a loan goes this long without a payment, it’s unlikely the bank is going to see any money for it.
And that’s what we can see in the chart above.
What are the other risks?
On the surface, this gives officials, banks, and the media something to cheer about.
They — and other analysts — will trot out this figure as a way of saying that Aussies aren’t in financial distress.
However, what about the data you can’t see?
See, Aussie banks are only required to make public the loans that went bad after 90 days…
But all those loans that haven’t seen a payment for a month, or two? Well that doesn’t have to be reported to the authorities…
Nor does the number of people making partial payments either.
If a person can’t make their full repayments, they are referred to the financial hardship team.
And the financial hardship department is more than likely to work with a customer to get some payment.
After all, cents on the dollar is better than total default. If people are willing to make some payment rather than none, the bank will work with that customer for as long as possible.
Banks do everything they can to internalise this information. A debt handled internally is far more likely to be paid in full — albeit at a slower rate — than if it is shipped off to a debt collector.
However, once a debt approaches the three-month mark without payment, it’s not far from being an official ‘default’ to be sold to a debt collector.
Banks generally try to avoid this. Not because of their good nature however. Once a loan is sent to a debt collector, the banks gets less for it.
So it’s in the banks’ interest to work with a customer.
And it’s the exact information the banks rarely feel like sharing with the public.
But here’s where it gets dangerous for investors.
We have no idea how many people are in financial distress.
Because the banks don’t publish that information.
They will advise in their yearly financial statements the percentage increase or decrease of loans in arrears.
But they never disclose the number of people or the total dollar value amount of how many Australians aren’t making regular mortgage repayments.
Which means investors don’t actually know how many bank loans have gone bad…
…and that’s the real stress test we need to see.
That will tell us just how healthy Aussie banks really are.
Until next time,