WBC - Westpac

[Snoopy]

According to the comment at the end of this article, my total $100m paedophile fine settlement could be some way out:

https://www.finextra.com/newsarticle...undering-rules

"23 million breach occasions @ A$17 - A$21 million penalty per breach occasion is A$391 trillion - A$ 483 trillion."

To put that into perspective, total shareholder funds on hand at EOFY2019 was $65 billion. So potentially WBC could be fined over:

391/0.65 = 600

times the amount of shareholder funds on the balance sheet! That sounds like crazy stuff that would immediately collapse the bank. Hyperbolic surely?

My best guess so far is that Westpac will be facing a fine of between $A20m and $A483,000m

Can I narrow that down a bit further?

https://qz.com/1754928/australias-we...ng-pedophilia/

"Austrac’s current record fine was issued against Commonwealth Bank for A$700 million. But its number of violations was less than 2% of Westpac’s 23 million total."

So let's extrapolate Westpac's potential fine based on their errant transaction level.

$700m / 0.02 = $35,000m. Hmmmm

"Overall, Austrac found 12 customers who together had sent more than 3,000 payments worth around $340,000 to child abusers in the Philippines and elsewhere in Southeast Asia."

That is a rate of 250 incidents per rogue customer.

Of the 23m breaches noted at Westpac, then that would imply 92,000 rogue customers.

Using an A$17-A$21 million penalty per rogue customer (a punishment rate I just made up), I get an overall fine of $1,564b to $1,932b. That several orders of magnitude larger than the $0.700b CBA fine. But the banking environment has changed in the Covid-19 environment. A fine of materially more than $700m at this time could cause WBC and, as a consequence, the rest of the banking system in Australia to become unstable. Would a court risk that? Or would the legal system carry on down their path regardless of the economic consequences of any fine imposed?

SNOOPY

[Snoopy]

WBC
	Annual Net Impaired Asset Expense (A)	Total Impaired Loan and Credit Commitment Provision (B)	(A)/(B)	Total Loans (impairment (B) included) (C)	(B)/(C)	EBT (before impaired asset exposure) (D)	(A)/(D)
FY2019	$794m	$3,913m (*1)	20%	$718,683m	0.5%	$10.543m	7.5%

(*1) Includes a $980m increment from the adoption of accounting standard AASB9 (on Financial Instruments, including impairment). AASB9 includes a new 3 level impaired debt assessment classification as part of the newly adopted ECL (Expected Credit Loss) model.

Banks must hold capital to support their loans. For Westpac the minimum capital required is calculated as 8% of total 'risk weighted assets'. The concept of 'risk weighted assets' is that some assets are easier to sell, in the event of a loan going bad, than others. So it is good for a bank to stack their loan book with loans that are relatively easy to liquidate in the event of the customer being unable to repay the loan at some time in the future.

How to measure the ease with which a loan can be liquidated? If we define a standard loan with a risk weighted adjustment factor of 1....

(aside question: don't ask me why standardised loans have a RWA of 0.93 in the table below _ I would love to find out!)

...and an 'another category loan' to have an RWA factor of 0.25, that means the 'another category loan' is less risky. To reflect this lower risk, the bank is happy to hold four times the dollar value of the 'another category loan' in place of one times the dollar value of 'standardised loan'.

From p13: https://www.westpac.com.au/content/d...ember_2019.pdf (except the last column that I have calculated)

Westpac averaged Risk Weighted Assets EOFY2019 ($m)

Asset Categories	Risk Weighted Assets {A}	Exposure at Default {B}	Averaged Risk Weighted Factor {A}/{B}
Corporate	74,807	139,173	0.54
Business Lending	35,470	54,570	0.65
Sovereign	2,068	90,960	0.023
Bank	8,339	28.761	0.29
Residential Mortgages	131,629	559,018	0.24
Australian Credit Cards	5,089	17,541	0.29
Other Retail	12,395	15,951	0.78
Small Business	16,090	33,365	0.48
Specialised Lending	55,262	65,553	0.84
Securitized	5,749	25,774	0.22
Standardised	20,966	22,512	0.93
	-------	-------
Total	367,864	1,054,178	0.35

If we look at the largest category of loans, 'Residential Mortgages', we can see that the RWA value is 0.24. Does this mean that a 'Residential Mortgage' is only about a quarter as risky as a standard loan? Unfortunately the answer isn't so simple, because under the Basel 3 convention, the riskiness of a residential loan is tied to the amount of equity that the homeowner has in their home.

Loan to Value Band	RWA Factor
Below 50%	20%
50%< <60%	25%
60%< <80%	30%
80%< <90%	40%
90%< <100%	50%
>100%	70%

One way to interpret that 24% RWA figure for residential mortgages at EOFY2019, is that it is equivalent to having the outstanding balance yet to be repaid on each mortgage lying between 40%-50% of the value of the residential property (work backwards in the table above aligning the AWF figure with the home equity held). Of course it is extremely unlikely in practice if all residential mortgages held by Westpac fall into this single loan category. It is likely that Westpac hold residential mortgages in all of the above 'variable equity' categories. The big 'capital holding risk' to Westpac is to ask the question "How many property loans are in the 0-10% equity bracket?" Because if the property market turns sour, it is those customers that will largely determine how much the risk adjusted loan balance base increases. The increase/decrease in the 'risk adjusted earnings base' then determines how much extra/less capital the bank must hold.

SNOOPY

[Snoopy]

From p13: https://www.westpac.com.au/content/d...ember_2019.pdf (except the last column that I have calculated)

Westpac averaged Risk Weighted Assets EOFY2019 ($m)

Asset Categories	Risk Weighted Assets {A}	Exposure at Default {B}	Averaged Risk Weighted Factor {A}/{B}
Residential Mortgages	131,629	559,018	0.24

If we look at the largest category of loans, 'Residential Mortgages', we can see that the RWA value is 0.24. Does this mean that a 'Residential Mortgage' is only about a quarter as risky as a standard loan? Unfortunately the answer isn't so simple, because under the Basel 3 convention, the riskiness of a residential loan is tied to the amount of equity that the homeowner has in their home.

Loan to Value Band	RWA Factor
Below 50%	20%
50%< <60%	25%
60%< <80%	30%
80%< <90%	40%
90%< <100%	50%
>100%	70%

Here are the results of a 'stress test' I have performed on Westpac residential property loans. The aim of the test is to answer the question: How much incremental capital need Westpac hold should residential property prices decline by 10%?

LVR Loan Position at EOFY2019

LVR %	0< <60	60< <70	70< <80	80< <90	90< <100	> 100	Total
Residental Mortgage Balance	$257,596m	$83,853m	$100,623m	$67,082m	$22,361m	$9,503m	$559,018m
RWF	5/6*0.2+1/6*0.25=0.2083	0.3	0.3	0.4	0.5	0.7	N/A
Risk Weighted Mortgages	$53,657m	$25,156m	$30,196m	$26,833m	$11,181m	$6,652m	$162,675m

The actual given risk weighted figure for residential mortgages was $136,689m (Westpac Pillar 3 Report December 2019 p10). That means my calculated figure of $167,675m is too high. The two most likely reasons for this discrepancy are:

1/ I have assumed more mortgages are in the 50-60% LVR range than is really the case. This cannot be easily checked because my 50-60% LVR is only a linear estimate and the real figure which WBC would know has not been publicly released.
2/ I have used the Australian LVR figures for all mortgages (Results Presentation FY2019 p93) when I know the NZ figures (Results Presentation FY2019 p129), and probably others outside of Australia, are more conservative (Results Presentation FY2019 p129) .

Nevertheless it is the change in the Risk Weighted Mortgages as a result of a 10% reduction in the value of retail property, not the absolute value that is of interest. The change in Risk Weighted Mortgages is less sensitive to any calculation errors than the absolute values, so I will continue.

A change in house prices does not change the value of a mortgage. But it does change the leverage that the house owner has on their property. I can approximate a 10% market decline by moving all the leveraged mortgages up one bracket, as in the adjusted table below.

LVR Loan Position at EOFY2019 (but with Imposed 10% Property Market Crash)

LVR %	0< <60	60< <70	70< <80	80< <90	90< <100	> 100	Total
Residental Mortgage Balance	$214,663m	$42,933m	$83,853m	$100,623m	$67,082m	$31,864m	$559,018m
RWF	5/6*0.2+1/6*0.25=0.2083	0.3	0.3	0.4	0.5	0.7	N/A
Risk Weighted Mortgages	$44,714m	$12,880m	$25,156m	$40,249m	$33,541m	$22,305m	$178,845m

The difference between the risk weighted mortgage figures under the two scenarios is:

$178,845m - $162,675m = $16,170m

The amount of incremental bank capital, notionally needed to support the increase in leverage of mortgage holders, (caused by properties sinking in value) is 8% of the incremental total:

0.08 x $16,170m = $1.3billion

Note: From "Final Basel III Modelling: Implementation, Impact and Implications" by Ioannis Akkizidis and Lampros Kalyvas" page 14.

"As from 1st January 2015, the minimum Common Equity Tier 1 (CET1) was set (under Basel 3) at 4.5% while (total) Tier 1 was 6%, implying that the additional Tier 1 should be at most 1.5% of the total RWA. The minimum total capital (TC) requirement would remain unchanged in comparison to Basel II at 8%."

SNOOPY

[Snoopy]

The difference between the risk weighted mortgage figures under the two scenarios is:

$178,845m - $162,675m = $16,170m

The amount of incremental bank capital, notionally needed to support the increase in leverage of mortgage holders, (caused by properties sinking in value) is 8% of the incremental total:

0.08 x $16,170m = $1.3billion

0.08 x $16,170m = $1.3billion

This figure is not the same as the amount of new capital needed. The amount of new capital required could be less as the total does not take into account any 'buffer capital' the bank may already have on the books in anticipation of such a residential property downturn. Mortgagee sales are another way to reduce the amount of 'bank support capital' needed.

Of particular interest is how this 'potential increased capital requirement' squares off against the provision set aside under the newly adopted ECL (Expected Credit Loss) method of booking 'Expected' bad debts.

From

https://www.westpac.com.au/content/d...ember_2019.pdf

and on page 13, the 'Regulatory Loss for Non-defaulted Exposures' is on the books at $1.088billion for 'residential mortgages'. That is

$1.3b -$1.088b = about $200m

IOW $200m would be a ballpark figure if we had to guess any extra provisioning required should the residential property market fall by 10%. That doesn't seem much of a gap to bridge when you talk about the kind of money these 'big four' Aussie banks deal with. It could even be less (or non-existent) if some of those loans that now exceed the value of the underlying property are counted as 'in default'. ($1,642m is the amount put aside for expected losses for defaulted and non-defaulted exposures). For current Westpac shareholders, I think that is encouraging.

SNOOPY

[peat]

IOW $200m would be a ballpark figure if we had to guess any extra provisioning required should the residential property market fall by 10%. That doesn't seem much of a gap to bridge when you talk about the kind of money these 'big four' Aussie banks deal with. It could even be less (or non-existent) if some of those loans that now exceed the value of the underlying property are counted as 'in default'. ($1,642m is the amount put aside for expected losses for defaulted and non-defaulted exposures). For current Westpac shareholders, I think that is encouraging.

SNOOPY

some strong provisioning coming through today Snoops, a lot more than 200m , I'm reading 1,430m plus 140m in insurance.
Share price didn't react much though.

[Snoopy]

some strong provisioning coming through today Snoops, a lot more than 200m , I'm reading 1,430m plus 140m in insurance.
Share price didn't react much though.

I was investigating 'Residential Mortgages' only Peat. The "up to $200m" was only for that. I was also working on a 10% drop for residential property from the EOFY2019 position. If Westpac are modelling more than that as a drop for residential, then my "up to $200m" will go up.

Of course there are many more kinds of lending that Westpac do, that can go sour. I wonder what provisioning Westpac is setting aside for those?

Still in the background we have the on going "Fee for no service" saga on customer superannuation balances which I am not sure was fully provisioned for. Next, the as yet unspecified fine the courts are expected to hand down to Westpac, for facilitating some customers engaging in paedophilia paid for via Westpac platforms. The market at least knew about these two, even if the final quantum of expense is unknown.

What quantity of the write-downs today are 'new' and relate directly to the loan book?

'Insurance', in the form of Westpac's in house product, including 'mortgage repayment insurance' seems to be only offered to Australian customers. Well, it has never been offered to me as a Westpac customer in NZ anyway. Could the extra insurance write off have some connection to the just wrapping up Australian fire season?

SNOOPY

[peat]

Could the extra insurance write off have some connection to the just wrapping up Australian fire season?

SNOOPY

yes it is exactly that.

Its hard to understand how they were over $30 not that long ago.
Historic PE of under 8 now
I've been glancing through Grahams Intelligent Investor and just wondering if any of these Pillars are now meeting his requirements

[macduffy]

yes it is exactly that.

Its hard to understand how they were over $30 not that long ago.
Historic PE of under 8 now
I've been glancing through Grahams Intelligent Investor and just wondering if any of these Pillars are now meeting his requirements

Perhaps, but the E will have to be assumed to be a lot lower now.

[percy]

What do you think their forward PE will be.Under 12 ?

[bottomfeeder]

Making a provision for fines of over 1 bill, now seems a fait accompli. Company telling everyone that that is what they are prepared to pay. Surely they could have done better with that one. Perhaps said difficult to quantify, but making a provision of $500k, pending the final outcome. What dunderheads, commercially sensitive information should not be leaked.