Wednesday, June 5th, 2019
The four-year bank earnings downgrade cycle rumbled on after WBC’s interim result, with downgrades of up to 7% to the bank’s FY19 and FY20 earnings per share by broking analysts even though expectations were already low before the result. Analysts focused on weaker housing and SME loan growth, interest margin pressure from price competition in mortgages and switching from higher-margin interest-only loans to lower-margin principal & interest loans, falling fee income, billion-dollar customer remediation costs, elevated legal, regulatory and compliance costs, rising retail borrower arrears and pockets of negative equity, loss of earnings from business divestments, potential adverse changes to negative gearing policies and reduced prospects for buybacks and special dividends due to regulatory uncertainty about capital.
Then the surprise federal election result heralded five positive catalysts in quick succession:
- The demise of Labor’s policy to end negative gearing for existing properties after a start date has improved investor sentiment on housing
- Refunds of surplus franking credits will continue to be allowed. This does not increase bank valuations but it does improve sentiment on bank stocks, which are widely held for franked yield
- The banking regulator APRA removed the minimum 7% interest rate threshold for assessing mortgage serviceability and now proposes to allow banks to assess applicants at their prevailing rate for new mortgages plus a 2.5% buffer. This increases maximum borrowing capacity by ~9%. Before last week’s announcement, we expected some kind of policy relaxation here given the 7% threshold had outlived its usefulness and was not relevant across all types of loans
- RBA Governor Philip Lowe said the central bank’s board would consider the case for reducing the cash rate at the 4 June meeting. This would further increase borrowing capacity for new mortgage applicants and reduce debt-servicing costs for existing borrowers on variable rates
- RBNZ Deputy Governor Geoff Bascand said the central bank would consider modifying the plans it announced in December (for heavy extra capital imposts on Australian banks operating in New Zealand), including flexibility on timing, a reconsideration of whether or not hybrid debt would be included in the measurements, in addition to the ultimate level of capital required. The RBNZ was willing to consider banks’ submissions seeking the possibility of other forms of capital, such as hybrid debt, being included in Tier 1 capital and was open to Australian banks’ arguments the original proposed increase was too much. Less regulatory Tier 1 capital should mean higher ROEs.
These added to the bipartisan policy proposal in the election campaign for government to underwrite insurance on certain mortgages with loan-to-valuation ratios of over 80%. This signalled the political system’s view it had become too hard to get a mortgage and underwriting criteria should be relaxed.
The central argument for increasing bank exposure is the multi-year earnings downgrade cycle for the sector is close to bottoming, so banks are likely to trade at the upper ends of their earnings multiple ranges and valuations are therefore headed higher. Several reasons justify this argument.
First, although we do not expect a return to boomtime rates of growth in home lending and house price appreciation given property prices and household debt are still high, we do think average house prices will bottom within nine months – with the bullish case being sooner – and this will see a return to modest growth in investor lending (currently zero growth). Real estate agents report a jump in buyer confidence and average auction clearance rates in Sydney and Melbourne rose above 60% last weekend. We also see a moderately faster pace of home lending from current rates below 6% pa. Banks have blamed less demand for mortgages as one main reason for the slowdown, so lending growth should improve as this effect unwinds.
We think house prices will bottom when buyers see value in housing again and population growth driven by immigration absorbs pockets of oversupply. New housing supply is tapering, evidenced by weakness in building approvals, so the ratio of demand to supply should evolve more favourably or less negatively, supporting a diminishing rate of house price depreciation in coming periods. In NSW, the ratio of housing demand to supply has already risen from 19-year lows.
Figure 1. NSW housing demand to supply ratio
Source: Credit Suisse
Figure 2. Apparent decline in rate of house price depreciation
Source: Macquarie Equities
Stimulus in the form of tax cuts should be backdated to 1 July and increase borrowers’ ability to service and receive approval for mortgages.
Reserve Bank rate cuts, which seem imminent, would reduce deposit margins because cash account and term deposit rates are so low they can hardly be cut further. However, we expect banks will use an RBA rate cut(s) to reprice existing mortgages by not passing on all of the cut. This was an important support to interest margins in bank first halves as price competition in mortgages weighed on margins.
We have noticed a trend towards more short-termism in some analyst research on banks and not enough credit for the generally solid execution of core strategies by major bank management teams. Once remediation costs have peaked, banks will cycle depressed earnings, organic capital growth should start to overtake remediation costs, and this should support sentiment on the sector. In WBC’s case, this should come sometime in FY20, when the bank will provide for its costs in remediating customers of its aligned advisers. Provisions for remediating customers of salaried advisers are being taken in FY19.
Meanwhile, WBC continues to reduce its underlying operating costs to offset the lack of revenue growth but is still investing in its strong customer franchise. Guidance is for a 1% reduction on the FY18 cost base.
The domestic yield curve is inverted, signalling the policy cash rate is too high. RBA rate cuts would help restore a positive slope to the Australian yield curve, a situation associated with a positive performance by bank shares – and also undervalued shares in general.
The RBA recently cut its 2019-20 GDP growth forecasts from 3.00% to 2.75% and forecast inflation would take longer to return to its target range. Growth was 2.3% over the year to the December quarter. Even if there is another downgrade to 2.50%, banks still face a benign mix of no imminent hard landing and policy stimulus. Meanwhile, the buoyant trade surplus increases private saving of foreign currency and Australian dollar depreciation supports domestic firms in the tradables sector.
WBC has significantly derisked by reducing the proportion of interest-only loans to 31% of its mortgage book. While this is still well above peers, the ratio is down from a peak of 50% two years ago.
Stable unemployment remains central to this thesis. Ongoing increases in funding for education, healthcare, aged care and disability support should drive further employment growth in these sectors, admittedly temporarily offset by lower employment in dwelling construction.
Substantial constraints on earnings, ROEs and valuations remain. In addition to the list of negatives at the start of this section there are also:
Structural reductions in fees to rebuild customer trust. This was a negative theme of CBA’s 3Q19 trading update
Banks are still transitioning from widespread use of household expenditure benchmarks to tougher individual expense verification, so there is more pressure on lending growth to come here.
However, we consider these detractions from earnings and returns are largely factored into consensus estimates after four years of downgrades.
Our valuations factor in 5% upgrades to consensus FY20 EPS for each major bank. This is a deliberately intermediate stance while we assess the outlook for loan growth and loan impairment charges after the five positive catalysts above. A return to housing loan growth of ~4% in FY20 and FY21 with impairment charges flat at ~13bp of gross loans would see average upgrades of up to 8% to some of the more bearish FY20 and FY21 EPS forecasts in the market. Overall we see more upside to our valuations than downside, especially if there is a broad improvement in consumer and business confidence and bad debts stay lower for longer.
On dividends, we expect ANZ will hold its dividend steady at 80 cents per half and refrain from another buyback until it has more clarity on APRA and RBNZ capital requirements. After NAB recently cut its dividend, attention has turned to WBC as the next bank which could also cut. Given the bank’s adequate capital position, at this stage, we think discounted and/or underwritten dividend reinvestment plans are more likely given WBC appetite to distribute its flow of franking credits. Share issuance from the currently discounted DRP is pressuring EPS growth but in our view, it would take a material surge in bad debts expense for capital to come under sufficient pressure for a dividend cut to be necessary to protect the balance sheet.
CBA is not in the model portfolio because it remains overvalued in our view, probably due to expectations of a franked capital return in the 2H of CY19. Similar demand for franking credits sent Woolworths’ share price above intrinsic value ahead of its recent off-market buyback.
Clime Group owns shares in CBA, WBC, ANZ and NAB.