Complacency kills – the ticking time bomb for Aussie banks


In the late 1980s at the peak of the property bubble, the Imperial Palace in Tokyo was worth the equivalent to the entire state of California. Greater Tokyo was worth more than the whole United States. The Japanese used to joke that they had bought up so much of Hawaii that it had effectively become the 48th prefecture of Japan. Japanese nationwide property prices quadrupled in the space of a decade. At the height of the frenzy, Japanese real estate related lending comprised around 41.2% (A$2.5 trillion) of all loans outstanding. N.B. Australian bank mortgage loan books have swelled to 63% (A$1.7 trillion) of total loans.


Sensing the bubble was getting out of control, the Bank of Japan went into a tightening rate cycle (from 2.5% to 6%) to contain it. Unfortunately it led to an implosion in asset markets, most notably housing. From the peak in 1991/2 prices over the next two decades fell 75-80%. Banks were decimated.

In the following two decades, 181 Japanese banks, trust banks and credit unions went bust and the rest were either injected with public funds, forced into mergers or nationalized. The unravelling of asset prices was swift and sudden but the process to deal with it took decades because banks were reluctant to repossess properties for fear of having to mark the other properties (assets) on their balance sheets to current market values. Paying mere fractions of the loan were enough to justify not calling the debt bad. If banks were forced to reflect the truth of their financial health rather than use accounting trickery to keep the loans valued at the inflated levels the loans were made against they would quickly become insolvent. By the end of the crisis, disposal of non-performing loans (NPLs) among all financial institutions exceeded 90 trillion yen (A$1.1 trillion), or 17% of Japanese GDP at the time.

The lessons are no less disturbing for Australia. Don’t be surprised to hear the authorities and local banks champion stress tests as validity that we are safe from any conceivable external shock. The November 2018 Reserve Bank of Australia minutes revealed that the next rate move is likely up but the board is happy to sit on its hands because housing is slowing even at 1.5% cash rates.

With US rates heading higher, our banks are already facing higher funding costs because of our reliance on overseas wholesale markets to fund mortgage lending. Japanese banks have 90%+ funding from domestic deposits. Australia is around 60-70%. Our banks need to go shopping in global markets to get access to capital. Conditions for that can change on a dime. External shocks can see funding costs hit nose bleed levels which are passed onto consumers. When you see the press get into a frenzy over banks passing on more than the rate rises doled out by the RBA, they aren’t just being greedy – a large part is absorbing these higher wholesale funding costs.

What about America? Who could forget former Goldman Sachs CEO and US Treasury Secretary Hank Paulson tell us how robust US financial institutions were right before plugging $700 billion to rescue the crumbling system? US banks such as Wells Fargo, Citi and Bank of America (BoA) have been reducing mortgage exposure relative to total loans outstanding. Yet each received $10s of billions in TARP (bail out funds) courtesy of the US taxpayer.

By 2009 the Global Financial Crisis (GFC) had turned over 16% of Bank of America’s residential mortgage portfolio into either NPLs, mortgage payments over 90-day in arrears or impaired (largely from the shonky lending practices of Countrywide (which BoA bought in 2008). Countrywide’s $2.5bn acquisition price turned out to cost BoA shareholders a further $50bn by the end of the clean-up. Who is counting?

Oh no, but Australia is different. Residential property prices in Australia have had a far steadier rise over a longer period – a 5-fold jump over 25 years – meaning our local banks should be less vulnerable to external shocks. There is an element of truth to that, although it breeds complacency.

Property loans in Australia as at September 2018 total A$1.653 trillion. 82% of those loans are made by the Big 4 banks. Interest only loans are around $500 billion of that. As a percentage of total loans outstanding in Australia, mortgages make up 65%. The next is daylight, followed by Norway at around 40%. US banks have cut overall property exposures and Japanese banks are now in the early teens. Post GFC, US banks have ratcheted back mortgage exposure. They have diversified their earnings through investment banking and other areas. You can see this below.


The advent of interest only loans has helped pushed property prices higher. NAB notes in its latest filing that 29% of its mortgage loan book is in interest-only form. The RBA expects $120 billion of interest only loans resetting to principal & interest (P&I) each year to 2020 which will hike monthly mortgage repayments to jump 30-40%. If investors were up to the gills in interest only mortgage repayments, adding one third to the bill will not be helpful. This is before we have even faced a bump in wholesale finance rates due to market instability. Look at the way that GE – once the world’s largest company in 2000 – is being trashed by the credit markets as they seek to reprice the risk attached to the $111bn in debt after a credit downgrade. This is a canary in the coalmine issue.

We also need to consider what constitutes a bubble in property. Sensibly, affordability makes the strongest argument. At the height of the bubble, the average central Tokyo property value was around 18.2x income. Broadening this out to greater Tokyo metropolitan area this was around 15x. This figure today is around 5x. Making arguments that ever higher levels of migration will keep property buoyant is not a sound argument as affordability affects them too.

Back in 2007, Sydney house prices were 8x income. In 2017 Demographia stated average housing (excluding apartment) prices are in the 13-14x range. The Australian Bureau of Statistics notes that 80% of people live in houses and 20% on apartments. Only Hong Kong at 19x beats Sydney for dizzy property prices.

In 2018, Australia’s GDP is likely to be around A$1.75 trillion. Our total lending by the banks is approximately $2.64 trillion which is 150% of GDP. At the height of the Japanese bubble, total bank lending as a whole only reached 106%. Mortgages alone in Australia are near as makes no difference 100% of GDP.

Balance sheets are but snapshots in time. If we look at our current bank exposure to mortgages, it is easy for analysts to paint rosy pictures. Banks’ shareholder equity has quadrupled in the past 16 years. Prosperity and record bank profits should give us comfort. Or should it? We need to understand that the underlying tenets of the Australian economy are completely different to that of a decade ago.

At the time of Global Financial Crisis (GFC) Australia’s economy was lucky to get away broadly unscathed. We carried no national government debt and were able to use a $50 billion surplus to prime the economy through that period of turmoil. Many countries were not so lucky. Our fiscal stewardship leading up to the crisis allowed economic growth to remain in positive territory soon after. Now we have $600 billion debt and charging the national credit card with all of the promises so aggressively that we should expect $1 trillion of debt in the not too distant future.

Australian banks are highly leveraged to the mortgage market. It should come as no surprise. In Westpac’s full year 2018 balance sheet, the company claims around A$710 billion in assets as “loans”. Of that amount, according to the latest APRA data, A$411 billion of lending is ‘real estate’ related. Total equity for the bank is A$64.6 billion. So equity as a percentage of property loans is just shy of 16%. If Australia had a nationwide property collapse (we have not had one for three decades) then it is possible that the banks would face significant headwinds.

What that basically says is if Westpac suffered a 16% decline in the value of its entire property loan book then it would at least on paper appear in negative equity, or liabilities would be larger than assets. Recall in 2009 that BoA had over 16% of its residential loan portfolio which went bad. It can happen. CommBank is at a similar level. ANZ and NAB are in the 20% range before such a hypothetical situation would be triggered. See the chart below. Note how the US banks stung by the GFC have bolstered balance sheets


Of course the scenario of a housing collapse would imply that a growing number of borrowers would have to find themselves under mortgage stress and default on payments. It also depends on the portfolio of the properties and when those loans were written. If the majority of loans were made 10 years ago at 40% lower theoretical prices than today then there is lower risk to solvency for the bank if it foreclosed and dumped the property.

Although if we look at the growth in loans since 2009, the Australian banks have been making hay while the sun shines. As it stands, the likes of Westpac and CommBank each have extended mortgage loans to Aussies to nearly as much as BoA has to Americans. That said the American banks, so stung by the GFC, have become far more prudent in managing their affairs.


It goes without saying that keeping one’s job is helpful in paying the mortgage. If you were a two income family and one of you lost your job, it is likely that dining out, taking fancy overseas holidays, buying new cars (which have been awful this year) and so on will go on the backburner. Should those actions swell to a wider number of mortgage holders, the economic slowdown will exacerbate in a downward spiral. Even your local coffee store may be forced to close because $4 is just cash you and others might not be able to spend. Boarded up High Streets were everywhere in America and Europe post GFC.


The following chart shows the negative correlation between housing prices and unemployment rates. US unemployment doubled to 10% when Lehman collapsed. Housing prices took heavy hits as defaults jumped. It is not rocket science.


On the other hand, Australia’s unemployment curve remained below 6% for around two decades. Even with GFC, jobless numbers never got out of hand. Our housing prices only suffered a mild dip.

We can argue that a sub-prime style mortgage crisis is highly unlikely. But it does not rule the risk out completely. To have that, mortgage holders would need to be in arrears on monthly payments, their houses would need to be in negative equity and banks would be required to take asset devaluations.

An ME Bank survey in Australia found only 46% of households were able to save each month. Just 32 per cent could raise $3000 in an emergency and 50 per cent aren’t confident of meeting their obligations if unemployed for three months.

According to Digital Finance Analytics, “there are around 650,000 households in Australia experiencing some form of mortgage stress. If rates were to rise 150 basis points the number of Australians in mortgage stress would rise to approximately 930,000 and if rates rose 300 basis points the number would rise to 1.1 million – or more than a third of all mortgages. A 300 basis point rise would take the cash rate to 4.5 per cent, still lower than the 4.75 per cent for most of 2011.”

Do you know how many homes NAB has under repossession on its books at the latest filing? Around 277. Yes, Two hundred and seventy seven. Out of 100,000s. Recall BoA had 16% of its loan portfolio go bang in 2008?

If we think about it logically, examining the ratio of total assets to shareholder equity (i.e. leverage), the Aussie banks maintain higher levels than the US banks listed below did in 2008. Were total asset values to suddenly drop 7% or more ceteris paribus, Aussie banks would slide into a negative equity position and require injection.


Human nature is conditioned to panic when crisis hits. Sadly many of our middle management class have never experienced recession. They are in for a rude shock. As for depositors note that you should be focused on the return “of” your money, not the return “on” it.

As Mark Twain once said, “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so!



The financial health of Millennials

Changing Balance Sheet across Generations

The St Louis Fed has published a report on Millennial balance sheets, comparing them to Gen Xers. The average value of total assets was lower among millennials than Gen Xers. Millennials held an average of $162,000 of assets relative to Gen X’s average of $198,000. The report also found that Millennials held a slightly higher level of total debt, at an average of $72,000 compared to Gen X’s average of $67,000. However the composition was markedly different – average student loan levels surged from $4,200 for Gen X to $14,700 for millennials. In short, millennials’ average asset position is lower, while they hold slightly more debt, which leads to an average net worth of $90,000 for millennials and $130,000 for Gen X.

In closing the report notes,

We observe that millennials have been going to school longer and delaying major life events. Thus, it makes sense that they hold lower levels of assets. They have had less time in the labor force, and a smaller share of them have moved out on their own, which contributes to the lower levels of residential assets. However, they have shown a higher propensity to save for retirement and to avoid credit card debt.