UBS thinks it’s meaningful:
APRA focuses on new ‘limit’ on interest-only, down to 30% share (now ~40%) APRA announced additional macroprudential policy tightening for housing. The most binding constraint was a new cap on the flow of interest-only (IO) home loans to only a 30% share, which is well below the decade average share of ~40%. This is likely partly indirectly aimed at investors where IO has been a very high ~60%+ share for many years; albeit owner-occupiers have also lifted to a significant ~23% share.
Greater scrutiny of LVR’s >80% on IO; now must be ‘comfortably’ <10% y/y cap APRA also told ADI’s to “place strict internal limits” on IO loans with an LVR >80%, & ensure strong justification >90%. Meanwhile, the investor credit growth cap was kept at 10% y/y, but tightened with lenders told to “comfortably remain below” the target.
Implications: RBA not hiking to contain housing…still using macro-pru instead Looking forward, we expect home loans to slow in coming months, after Jan-17 hit a >2-year high of 16% y/y – as investors surged 27% y/y. While this will slow housing credit growth moderately (now 6.4% y/y; owner-occupier 6.2%, investor 6.7%), investors at 9% annualised in the last few months will have to retrace. This assumes investors will bear the brunt of the adjustment. For instance, under a scenario where there is no change to 1) the ~50-50 split of total home loans between investors & owner-occupiers; & 2) the 23% share of owner-occupiers on IO; to meet the new 30% IO system cap, the share of investor loans on IO would ~halve from ~60% now to ~37% (or the share of owner-occupier on IO can also fall). This policy change will probably (indirectly) crimp household cashflow, as lenders continue to re-price IO loans, and some borrowers switch to principal & interest repayments. Overall, we flagged further macroprudential tightening as the RBA’s first response to strong house prices, rather than hiking rates. While these changes are less constraining than lowering the 10% cap to 7% (as we expected), we believe this could have a meaningful impact on demand (and price growth) over time, supporting our slowing housing outlook and our RBA on hold view (until at least mid-18).
Jonathon Mott worries about the impact on sentiment:
APRA announces a tightening of interest-only lending APRA has announced further measures to reinforce ‘sound lending’ for residential mortgages. This follows its first round of tightening in Dec 2014 and is in reaction to the accelerating boom in the Sydney and Melbourne housing markets. New measures include: (1) Limit the flow of new Interest-Only (IO) lending to 30% of total housing lending. This compares to spot levels of ~40%; (2) Strict internal limits must be applied for any IO loans >80% LVR; (3) Justification must be provided for IO loans >90% LVR; (4) Review serviceability to ensure buffers are appropriate for the current conditions. (5) Investment Property Loan (IPL) credit growth should be “comfortably” below 10%.
How will the banks react? We see this as a step by APRA (and the Council of Financial Regulators – APRA, RBA, ASIC & Treasury) in an attempt to take some of the heat out / de-risk the housing market. Given this tightening only applies to new lending (front-book), repricing the back-book would not be justifiable, in our view, especially given the pre-emptive repricing the banks have undertaken over the last few weeks. However, we would expect the banks to substantially reduce the discounts offered on new IO loans, allowing customers to choose if they want to pay a premium for an IO loan relative to Principal and Interest (P&I). The impact on Bank NIM should be minor.
Credit growth will slow. Impact on housing ‘Animal Spirits’ is uncertain If the banks reduce IO lending from 40% to 30% of the flow, we would assume around half of these customers (ie, 5%) would move to P&I, while around 5% would likely leave the market (eg, high LVR, multiple investment property speculators who are forced to deleverage or sell). This would reduce new mortgage lending flow by ~$15bn p.a., and would reduce total system housing credit growth by ~1% over the next year. From the banks’ perspective, this is manageable (~1% hit to NPAT). What is harder to gauge is the impact on the housing market ‘animal spirits’ if the marginal buyer is removed from the market. Will this take the heat out of the Sydney & Melbourne auction market? Could this lead to a change in sentiment towards housing?
Overall a positive move for Financial Stability with more to come if needed APRA has ‘turned up the dial’ on the overheated housing market, in our view. If these moves don’t work to slow Sydney & Melbourne house price inflation, further policies may be implemented such as higher capital requirements for riskier loans (eg, IO/IPL) and ATO tax return income verification the most likely options, in our view.
Deutsche sees it as marginal:
APRA announces a series of measures to curb housing market risks APRA today announced a series of measures aimed at controlling the risks in the housing market. It has written to ADIs outlining several restrictions on housing lending, including limiting the flow of new interest-only lending to 30% of total new lending. The 10% growth cap on investor lending was maintained. The regulator also indicated that it expects tighter controls on higher-risk categories of lending, although it did not prescribe explicit limits.
Interest-only lending currently contributes ~40% of the majors’ mortgage approvals, and so the 30% limit at face value will have some impact. However, we expect that a large proportion of borrowers should be able to switch to P&I loans instead, and so we expect the overall impact on loan growth to be modest. We have already incorporated a decline in housing loan growth over the next 2 years for the sector (5% in FY18 and 4.5% in FY19). Interest-only lending a key area of focus; comprises 40% of majors’ flow In addition to the 30% limit, APRA also said that it expects ADIs to implement strict internal risk limits on interest-only loans with LVRs over 80%. Also, ADIs should only extend interest-only loans with >90% LVRs where justification exists. Based on APRA stats, interest-only loans contributed ~40% of mortgage approvals in the 12 months to December 2016. While this would suggest the 30% limit will have some impact on the level of credit growth, this should be at least partially offset by the migration of borrowers towards P&I loans (particularly owner-occupiers who account for ~40% of interest-only lending).
No change to the 10% investor loan growth cap The existing 10% cap on growth in investor loans was unchanged today. APRA said that this level continues to provide an appropriate constraint, and that while it wanted to moderate growth in this category of lending, it was cognisant of the upcoming supply of newly constructed housing that needed to be funded. We think it is sensible to avoid disruption to this market.
Serviceability assessments will continue to be monitored In APRA’s letter to ADIs, the regulator also highlighted it will continue to focus on lenders’ serviceability assessments. While no quantitative requirements were set out today, it said it will monitor closely the trend in lending to borrowers with low income buffers, as well as other elements in lenders’ assessment criteria.
Potential benefit to margin We see two potential benefits from today’s announcement: i) further improvement in the quality of flows; and ii) potential for upside to margins if banks elect to further reprice their interest-only mortgage books. Currently there is a 13-17bps gap between interest-only and P&I loans on average, and this could increase further.
Credit Suisse agrees with me that falling IO can be offset by rising PO loans:
Modest credit growth impact could be mitigated by new interest-only flows now being directed into P&I: We see APRA’s additional macro prudential measures as presenting a modest additional constraint on the pace of credit growth. With December 2016 quarter flows of bank-originated interest-only mortgages 38% of total (majors 40%, regionals 31%) this represents, say, a 10% decline in mortgage growth. Accordingly, system mortgage credit growth of 6.4% becomes pro-forma, say, 5.9% on this basis over time (i.e. 90% of the previous pace of growth); with mortgages representing 62% of the total credit stock; 5.0% total system credit growth therefore becomes, say, 4.6% pro-forma over time. This impact could also be mitigated by new mortgage flows that would have been interest-only applications now becoming principal & interest loan applications. Banks may choose mortgage re-pricing (as well as changes to other terms & conditions) to implement this regulatory required adjustment.
Morgan Stanley thinks more tightening will be needed and says sell banks:
Recent regulatory jawboning on further stretched Australian housing conditions have been followed up with a cautious extension of the MacroPru toolkit. A 30% cap on interest-only mortgages will help slow conditions (and household cash flows), but we believe hikes to risk weights will follow.
First of a Two-Step Further Tightening: On the surface, APRA’s “additional supervisory measures to reinforce sound lending practices in an environment of heightened risk” seem cautious and could struggle to have a sustained impact on investment property lending (IPL)growth, or the rise in household leverage. That said, there was clearly an upswing of investor activity over summer,and we see appetite being tempered by the banks’ recent repricing (+23-38bp for investors) and a rationing of the interest-only (IO) mortgages used by ~65% of investors. Looking further out we see it as very likely that the next (more permanent) round could be through increases to risk weights for investor lending, potentially as soon as mid-year (see Australia Macro+: Supervising a Housing Slowdown (15 Feb 2017).
Another Tool to Slow Growth: System IPLgrowth has accelerated from 3.9% to 8.6% on a 3m-annualised basis over the last 12 months to February 2017 (including a 9.7% ann. surge in December), pushing towards APRA’s existing 10% cap. No one likes drivingunder the speed limit for long, but rather than looking for fines by lowering the cap, APRA has asked lenders to limit the share of IO mortgages to 30% (from 39% currently), which should help them remain ‘comfortably’ below the 10% benchmark.
Coinciding with Cracks in the Apartment Cycle…: Of more interest is that these measures continue to build as the cycle starts to show signs of stress, particularly for off-the-plan apartments. Recentheadlines span potential distress (see ‘Brisbane apartments offered at 39pc discount in disputed fire sale’, Australian Financial Review,29 March 2017) and poor returns for new-builds more generally (see ‘Off-the-plan buyers seeing losses and lacklustre growth’, Domain.com.au,28 March 2017). While price growth in established Sydney and Melbourne markets remains buoyant,apartment prices and rents look to be showing signs of oversupply (see Australia Macro+: Australian Housing: Stressing the Foundations, 19 Oct 2016).
….And at a Time of Peak Cycle Valuations: Housinghas been the key driver of the industrial activity cycle over the past three years. 45% of ASX 200 market capitalization is in some way linked to housing,and our forecast slowdown in construction and price growth drives our caution on domestic industrial earnings. Combine this with a market that is stretched on valuation at 15.8x (12mf PE) and with ‘real economy’ stocks (Industrials-ex-Financials)even more expensive at 19.5x, we expect further upward moves in the index to be tempered.
Macro Implications: While this looks to be a cautious further step by APRA, we note that it comes alongside a material repricing of investor mortgages and will potentially be followed by higher risk weights (see Australia Banks: More Capital Build, Except ANZ, 13Feb 2017). The market may also underestimate the degree to which Australian mortgage payments ‘step-up’ when rolling from IO-phase to P&I (at least +30% and often +50-70%, depending on tenors and headline rate). At an aggregate level, we calculate that a 10ppt fall in the share of IO mortgages to the new cap would reduce household free cash flow by around A$3bn (0.26% of income), with the average 14bp of mortgage hikes over the past fortnight absorbing another A$2bn (0.15%). This increases the burden on a consumer seeingno income growth (average wages were -0.5% in 2016),and we again reiterate our caution on the growth outlook, while seeing the market as too hawkish on the prospect of RBA hikes over the next 18 months.
Market Implications: The Banks sector has enjoyed a stellar 1QCY17 thanks to a broader positive valuation pulse in global financials,as well as rotation within the ASX 200 ‘yield’ cohort. Valuations are at peak, whilst growth seems hard to come by,and we believe taking an UW position from these levels would be a prudent move. Across Housing-Linked, we retain very limited exposure to developers, discretionary retail and building materials – all negatively exposed to an environment we see of slowing activity and volume.
Citi is also cautious on banks:
APRA announces much anticipated macro-prudential changes – Today APRA has announced that it has written to all ADIs (banks) outlining further macroprudential measures to slow activity in parts of the housing market.
…with interest-only mortgage lending now firmly in the firing line – The key new measure that APRA has introduced is that all ADIs need to limit new interestonly lending to only 30% of total new lending. This will require the Major Banks to implement meaningful initiatives to slow growth given new business flow is likely to be currently running at ~40% (Figure 1).
Existing measures appear to have been tightened as well, but not publicly – APRA has provided scope to tighten both the investor growth cap and serviceability standards, and we think are likely to do so. ADIs now need to remain ‘comfortably below’ the 10% investment lending cap and serviceability metrics are set at ‘appropriate levels for current conditions’.
We expect a host of changes which will impact banking revenue as a result
– Depending on how APRA implements these proposals, we expect a number of possible market changes (Figure 2), including:
– Further interest only re-pricing to slow growth. In addition, ADIs and mortgage brokers encouraging switching to Principal & Interest loans where possible.
– Slower mortgage credit growth, particularly for investors where interest-only has become prevalent. – Even more intense competition (i.e. higher discounts) on Principal & Interest owner-occupied loans.
…but too early to determine overall impact or determine winners and losers – These measures are likely to further fragment the already dislocated mortgage market. This could be both positive and negative for revenue for the sector and this will become clearer over the next 6-12 months as changes begin to take effect.
…leaving us remaining cautious on the sector – It is clear APRA wants to tighten their macro-prudential measures, but just not in a public manner. This announcement seems designed to limit the risk of unintended downside consequences in certain segments of the housing market, whilst offering flexibility to adjust certain settings as required. The recent rally in share prices leaves us remaining cautious on the sector.
But Goldman says buy:
Interest only lending to be restricted to 30% of new mortgage flow APRA today announced another wave of macroprudential measures, introducing limits to cool interest-only mortgage growth. APRA expects banks to immediately limit the flow of interest-only mortgages to <30% of new residential mortgage lending (vs. c. 40% currently). Within this, banks must place strict internal limits on the volumes of interest-only lending at LVRs >80%, and “ensure there is strong scrutiny and justification of any instances of interest-only lending at an LVR above 90%.”
APRA has not amended the 10% investor mortgage growth speed limit. Restraining lending to higher risk segments APRA has also stressed that all banks must continue to restrain lending growth in higher risk segments (e.g., high LVR loans), and to ensure all serviceability metrics are up to date with recent changes. APRA also notes that it has been “monitoring the growth in warehouse facilities provided by ADIs” and would be concerned if these were growing at a materially faster rate than a bank’s own housing loan portfolio. This would impact some non-bank lenders which have recently been growing strongly. Limited volume impact vs. GSe and offers repricing opportunities We think the new measures will have an immaterial impact on our system mortgage volume forecasts given (i) we already forecast a significant slowdown in investor approvals (Dec-17 investor approvals to fall 12% on pcp, 40% below their previous peak) and therefore system mortgage credit growth (growth to fall from 6.4% as at Jan-17, down to 4.1% by Sep-18; refer to top Margin Exhibit) and (ii) our system credit growth forecasts are relatively insensitive to assumptions around the new flow of loans (annual new flow of mortgages only represents c.15% of total mortgage credit outstanding). Furthermore, we think the regulatory pressure to slow interest-only lending potentially provides the banks with further repricing opportunities and estimate that every 15 bp increase in the rate on interestonly mortgages adds c. 3 bp / 2% to our coverage’s FY17E NIM / EPS. Earnings upgrades will continue to support banks; prefer ANZ Supported by potential for further repricing opportunities and a still benign credit environment, we think the risk to bank earnings remains to the upside. Therefore, with the sector trading at close to one standard deviation lower versus the industrials on a PER basis (refer to bottom Margin Exhibit), we still think the sector offers good relative value and continue to prefer ANZ (CL-Buy).
It remains too little, too late for me. Sell property and banks.
