Michael Janda| ABC| 13 December 2022
Depending on how high interest rates go — and how bad the latest property bust gets — it could go down as among the worst decisions in Australian financial regulatory history.
No — unlike the media pack tearing shreds off Phil Lowe — I’m not talking about the Reserve Bank of Australia governor’s repeated statements that he expected rates would stay near zero until at least 2024.
I’m talking about a much-less publicised action, before the pandemic, that allowed tens — if not hundreds — of thousands of borrowers to act on Lowe’s comments by borrowing much more than they could really afford.
This is what went down.
In the few days immediately after the re-election of Scott Morrison’s Coalition government in May 2019, the bank regulator — Australian Prudential Regulation Authority (APRA)— revealed its plan to ditch an interest rate floor on mortgage serviceability tests.
More on serviceability tests, the floor and the mysterious beast that is APRA later.
For now, suffice it to understand that this change meant people could borrow more than before the change.
And removing the floor meant that the maximum amount home buyers could borrow would keep rising with every interest rate cut.
So, does it make any sense whatsoever for APRA to have removed that floor just as the Reserve Bank started cutting rates again in 2019?
Yes, it does … if you’re a bank, mortgage broker, property developer, real estate agent or anyone else who makes money from the buying and selling of homes, or the issuing of debt to do so.
In 2019, Australia — and particularly the two most-expensive markets of Sydney and Melbourne — were at the depths of a property funk that had started when APRA, ASIC and then the Hayne banking royal commission had all tightened the screws on lending standards.
If people are allowed to borrow less, it stands to reason they can’t pay as much for a property, and prices fall.
Before APRA lifted its floor, banking and property heavyweights were demanding action.
Then-recently re-elected treasurer Josh Frydenberg publicly met with then-APRA chair Wayne Byres within days of the government’s 2019 re-election, and just a day after the regulator had announced its plans to ditch that cap.
Despite the timing of the meeting, there is speculation in some quarters of the financial sector that APRA removed the floor at the then-treasurer’s behest.
Even if he didn’t urge APRA to remove the floor, Mr Frydenberg praised the regulator’s move as a “positive development that will continue to spur lending growth across the economy”.
That it most certainly did.
Home lending roughly doubled between when APRA removed the floor and when it increased the buffer.
Lots of other things happened during that period, such as a pandemic that shut down large parts of the economy, resulting in massive government and Reserve Bank stimulus, including the HomeBuilder scheme and record low mortgage rates.
However, there is no doubt that removing the rate floor on mortgage serviceability tests boosted lending.
It also allowed people to borrow amounts that they’d struggle to afford if rates returned even to fairly average levels from over the past couple of decades.
Serviceability, buffers and the maximum you can borrow
The serviceability test is at the core of applying for a home loan.
It’s the bank’s way of deciding how big a loan you can afford to service and repay.
To do the test, the bank needs to know all your income and expenses to calculate how much money you have left over each week to service a loan.
However, there is one more factor to consider: what interest rate you will be tested on. Obviously, the higher the rate, the more you will be paying in interest and the smaller the maximum loan size you can afford to service.
In Australia, most loans are taken out on variable rates, and even fixed-rate loans generally last no more than four or five years.
So, it would be extremely risky to base the maximum amount a bank was willing to lend you on the current interest rate.
If you could only just meet your repayments and then the RBA raised rates next month, you’d be on a fast track to default.
That’s why there’s always been a buffer above the current interest rate to ensure most borrowers can cope with rising repayments.
At the moment, that buffer is 3 percentage points above current mortgage rates, until late last year it was 2.5 per cent.
However, between December 2014 and July 2019, that buffer was 2 percentage points, but with a 7 per cent interest rate floor.
The floor was exactly as advertised on the sticker: an absolute minimum interest rate that had to be applied by banks when deciding the maximum amount they could lend someone.
The rationale for the floor was simple.
At the time it was introduced, the cash rate had already been at 2.5 per cent for well over a year and the RBA’s measure of typical discounted variable mortgage rates was just above 5 per cent.
A 2 per cent buffer meant borrowers were being tested about whether they could afford repayments at rates just above 7 per cent.
However, the cash rate was about to fall further and, without a floor, the serviceability test would fall with it.
By May 2015, the cash rate was 2 per cent, by August 2016 it was 1.5.
By then, discounted variable rates were around 4.5 per cent. Under a 2-percentage-point buffer without the floor, those borrowers would have been tested on a 6.5 per cent mortgage rate.
If that sounds reasonable to you, consider these facts.
The average discounted variable mortgage rate since the RBA’s data began in June 2004 has been 5.59 per cent, with the highest rate at 8.96 per cent in July and August 2008, just before the global financial crisis.
Discounted variable mortgage rates were 7 per cent or above in 40 of the 222 months for which this data presently exists: that is, 18 per cent of the time.
On recent history, a mortgage serviceability rate floor of 7 per cent already seems dangerously low — there was roughly a one-in-five chance of exceeding it in any given month, worse odds than Russian roulette.
However, removing the floor made the risks much worse.
During the pandemic, the RBA slashed the cash rate to a record low of 0.1 per cent.
The RBA’s indicator discounted variable rate for owner-occupiers fell below 3.5 per cent, meaning the test for someone on that rate would be tested to see if they could cope with a mortgage rate of less than 6 per cent.
However, many borrowers were getting variable rates down to 2.5 per cent, and even below, meaning some could have been tested against mortgage rates as low as 5 per cent, underneath even the average discounted variable rate over the past 18 years.
Horse bolted, gate shut?
APRA only increased that buffer to 3 percentage points at the end of October last year, when the property boom that low rates and government HomeBuilder stimulus had sparked was just starting to fizzle out anyway.
Now the cost of that boom is becoming all too apparent as surging interest rates expose those who borrowed more than they could prudently afford — and the banks that lent the money.
The RBA has increased interest rates by 3 percentage points since the start of May.
That means even those very recent borrowers subject to the higher serviceability buffer have seen it totally eroded away. And the RBA said last week that it still “expects” to raise rates further.
Those who got their loans over the previous couple of years on a 2.5 percentage point buffer are now living an experiment.
Their monthly repayments are already higher than what they were tested to be able to afford.
Put another way, without other changes in their circumstances, their serviceability equation — of income minus expenses — no longer leaves enough money over to meet their minimum mortgage repayments.
Many will have borrowed below their maximum limit, increased their income or got far enough ahead on their loans to see them through, for the time being at least.
However, there’s no doubt others who are already straining, and possibly failing, to pay mortgages they were never really able to afford.
How many will become apparent as the months roll by, and will rise along with any further increases in interest rates.
That’s not even factoring in those who might have stretched the truth on their application to get a bigger loan, possibly encouraged by their banker or broker, as regularly revealed in the UBS “liar loans” survey.
An additional irony of APRA’s move late last year to increase the size of the buffer is that it’s now making it even harder, in many cases impossible, for struggling borrowers to refinance their loans to get a more-competitive interest rate that might relieve some financial pressure.
They simply don’t meet the serviceability test anymore for the same-sized loan they were granted just one or two years ago.
When interest rates had not risen in about a decade, having a buffer might have seemed like a waste of time.
However, the lower rates got below historical norms, the bigger that buffer should have become to protect against a reversion to mean (average), something that happens quite a lot in economies and on financial markets.
Instead, by removing the floor, the regulator effectively did the exact opposite, taking the same gamble as the RBA governor that interest rates would stay abnormally low for quite some time.
An APRA review overdue?
The regulator that watered down those protections — at the time of greatest need for them — surely has a lot of questions to answer.
APRA rarely does though, at least via the media.
Unlike Philip Lowe — who continues in his role facing regular questioning from the media and the parliament — the man who had run APRA since 2014, Wayne Byres, stepped down on October 30 this year, almost two years early.
In his only television interview during eight years in charge of Australia’s prudential regulator, I had just over five minutes to ask him some of those questions.
Depending what happens to Australia’s housing market from here — and to the people who borrowed huge sums over the past few years to break into it — he and the others who ran the regulator may soon find they have quite a few more questions to answer.
Could a review into APRA be next in line, once the RBA review wraps up early next year?