Australia & New Zealand Banking Group
ANZ continues to shed unwanted risks in its institutional (large corporate) loan book with the result total risk-weighted assets decreased $2bn over the quarter.
As we saw in CBA’s 2H18 result in early August, credit quality in Australia remains excellent but ANZ’s derisking of its institutional loan book is also contributing. The specific provision charge in 3Q was the bank’s lowest since 2014, due partly to elevated writebacks and recoveries in the book.
ANZ retains an overlay for the retail sector. Home loan arrears were flat over the quarter and the pockets of stress remain confined to WA, especially Perth.
ANZ referenced slowing system growth in home lending, increased price competition, increased capital intensity and tighter credit conditions. As at end-June, YTD APRA system growth was 4.1%, down from 5.3% over the pcp. ANZ continues to emphasise owner-occupier principal & interest loans because they are less risky and more conducive to building a long-term relationship with the customer than investor loans. ANZ’s total Australian home lending portfolio grew at just 0.4 times system in the June quarter, presumably to avoid irrational price discounting.
Interest-only switching to principal & interest remains manageable with no material addition to arrears despite the start of repayments. The volume of switching ahead is set to decline.
Wholesale funding costs remain elevated, compressing interest margins.
For some time we have viewed our StocksInValue Model Portfolio weight of 4.5%+ in ANZ as too high for an ex-growth sector facing worsening margin pressure. At the time of writing, ANZ’s closing price of $29.64 ANZ was trading on 12.6 times consensus FY19 EPS but historically the trading multiple has peaked around 13 times.
Figure 1. Excellent credit quality in ANZ’s book
Figure 2. Mortgage switching is currently manageable and set to abate
After the stock went ex the $2.31 fully franked final dividend last week, it was trading around 13 times FY19 consensus EPS downgraded by ~1.5% since the result. We think this is the peak multiple on which CBA should trade in today’s era of slow and declining lending growth, intense price competition in mortgages, elevated funding costs, shrinking interest margins and falling returns on equity. The average trading multiple since the GFC is 13.4 times but CBA should trade on a lower multiple now after the end of the post-GFC home lending boom. Mortgage repricing is required for consensus upgrades from here.
Within the StocksInValue Model Portfolio, we have reduced weighting in both the ANZ and CBA as part of a strategy to reduce our banking exposure to create room for other stocks with more growth and the potential for consensus upgrades. We are now down to a total banking exposure of 9.3%.
Despite consensus downgrades to FY19 and FY20 earnings, CBA shares had rallied since the result because it was not as bad as some feared. There were predictions the final dividend would be rebased lower, so the board’s confidence in raising the final dividend by one cent to $2.31 per share were taken as a positive surprise. As shown in the charts below, CBA has plenty of capital and the payout ratio is only in the middle of the board’s target range, ie. not aggressive at the top of the range. We never thought CBA would cut its dividend and we also now think ongoing increases of one cent or better per half are likely given bad debts expense should normalise only gradually. This is despite the divestment of some 8% of group earnings (wealth management, mortgage broking, life insurance), as the remaining 92% of earnings are the most capital-generative. In this result CET1 was flat over the year with 32bp of organic capital generation offsetting the Austrac fine and adverse APRA adjustments. Post-divestments, CET1 is 10.7%. Capital generation should remain solid given slow loan growth and management expects to be ready to apply for partial withdrawal of APRA’s $1bn punitive capital impost a year from now.
Management guided to another year of elevated legal, regulatory and compliance costs in FY19 but said these would eventually fall. Without inflation in this part of the cost base, 2H18 pre-provision earnings would not have fallen 6%. FY20 should see positive jaws driven by some normalisation of these costs. By FY20, conduct concerns and costs are unlikely to dominate sentiment on this stock.
As at ANZ, switching from interest-only to principal & interest is proving to be a non-event for bad debts. Management said switchers get a year’s warning and while there is a noticeable uptick in arrears after switching, this eventually normalises to the book average with no material transition to impaired status. Switching is however negative for interest margins.
CBA is committed to ongoing leadership in digital, data and analytics. This has been a winning approach for the group since 2005. Ignore predictions large banks will cease to exist in 10 years’ time. In contrast large banks will eventually use technology to materially lower their wage costs (though it will take time and the upfront costs are heavy).
What’s not in the share price? Potential sources of revisions to consensus FY19 EPS
|Interest margins||Likely –ve to neutral. Price competition in mortgages is “intense”, offshore wholesale funding costs will keep inching higher, mortgage repricing of a moderate degree is likely once political pressure abates. Deposit pricing is a source of +ve surprise. Longer funding maturities will help stabilise NIMs|
|Home lending growth||Likely –ve. CBA guided to 4% pa system growth (was 5.5% yoy in June, so has further to slow) and is deliberately at less than system to avoid irrational discounting|
|Business lending||Neutral to marginally +ve. System growth has slowed to barely marginal though CBA talked about success it was having in the category|
|Non-interest income||Neutral to marginally +ve. Probably stops falling but upside will have to come from volatile markets income, which would require global interest rate volatility – unlikely unless inflation surprises|
|Operating costs||Neutral. Guidance is for another year of elevated legal, compliance and regulatory costs with meaningful attention to the operating cost base pushed out to the medium term. Management’s discussion of cost reduction potential was vague and a bit disappointing|
|Bad debts||Another +ve surprise in this result but only because analysts wrongly expected incremental normalisation higher. Unlikely to surprise again and the market wants pre-provision growth more than this. The surprise was due to one-offs: non-recurrence of a single-name UK provision in 1H18 and lower overlays. Mortgage arrears continue to inch higher from a low base, with outer metro areas now the problem. Perhaps bad debts now does start to normalise higher. Pre-provision earnings fell 6% in 2H18.|
Below are the key charts which tell the story of CBA’s FY18 result and its journey through a turbulent year.
Figure 3. CBA’s response to its turbulent year
Figure 4. CBA’s strong position in digital
Figure 5. CBA’s strong capital position supports the dividend
Figure 6. Payout ratio is not excessive, which also supports the dividend
Figure 7. Interest margins fall in the second half
Figure 8. The heavy cost to shareholders of CBA’s failures
Figure 9. Historically low bad debts expense continues
Figure 10. Home loan arrears rising off a low base
National Australia Bank
After last week’s rally, NAB was trading on 12 times FY19 consensus earnings, which is full without being stretched. The June quarter trading update revealed similar trends to ANZ and CBA: funding cost pressures, intense price competition in mortgages, solid asset quality, and elevated legal, regulatory and compliance costs. NAB’s fees-for-no-service scandal exposed in recent days at the Royal Commission, amongst other matters, will cause an additional conduct-related expense provision, of an amount NAB is currently unable to determine, in 4Q. This could put further pressure on the capital position, now the lowest of the peer group, and renew the market’s concerns about dividend sustainability. NAB remains the only major bank which might cut its dividend; our base case is the dividend reinvestment plan will be discounted and underwritten before that happens.
3Q18 cash earnings fell ~1% from the 1H18 quarterly average as 1% revenue growth, due to solid growth in SME lending and New Zealand, was offset by increases in investment spending, consistent with guidance, and loan impairment expense.
Net interest margin only “declined slightly”, a good outcome given elevated funding costs and mortgage price competition. NAB might be managing its margins better than peers, possibly because it is underweight mortgages relative to peers and is growing in SME lending, where there is less price competition.
Expenses rose ~2% on the 1H18 quarterly average due to higher compliance costs, investment spend and depreciation and amortisation consistent with the restructuring strategy. NAB seems to be on track to deliver expense growth within its 5-8% guidance range, alleviating some of our concerns about execution risk and a cost blowout.
Asset quality metrics remained broadly stable. Gross impaired assets fell to $1,585m from $1,609m on 31 March and 90 days past due plus gross impaired assets remained steady at 0.71% of gross loans.
Balance sheet trends were mixed. CET1 fell to 9.7% on 30 June from 10.2% on 31 March due to payment of the interim dividend but also elevated SME loans growth of ~2% over the quarter. Business loans are more capital-intensive than mortgages, so NAB’s soft capital ratio is a barrier to faster lending growth to SMEs. This is frustrating given NAB’s strong capabilities in digital business banking.