From the always excellent Jonathon Mott at UBS:
Sydney and Melbourne House prices accelerate
House prices in Australia have been on a steady upward course over the last five years since the RBA cut rates from a peak of 4.75% down to current levels of 1.5%. Since this time house prices in Sydney have risen 76%, Melbourne by 50%, while the rest of Australia has been more subdued, up 12%. While there was evidence the housing market was taking a break around the end of 2015, two RBA rate cuts in mid-2016 have reinvigorated the housing market. Despite record housing starts Sydney house price growth has accelerated to 18.8% (Y/Y) and Melbourne prices have accelerated to 14.4%. Recent data from Westpac shows 70% of mortgage drawdowns since September 2016 have occurred in NSW & VIC, while ABS data shows 46% of drawdowns in NSW are for Investment Property. This is leading to calls to address ‘the housing affordability crisis’ especially from First Home Buyers (FHB).
Rapid house price inflation since the RBA’s rate cuts in mid-2016 has been associated with ongoing household leveraging. While income growth has been subdued at 3%, mortgage credit growth is growing at 6.4% p.a which has driven household debt to disposable income to a record 187%. This has now pushed Sydney median house prices to 12.2x disposable income (Demographia data)
Importantly, recent growth has been driven by renewed investment property lending (IPL) activity. IPL credit growth has now hit 6.6% growth (Y/Y) while during the last three months it has been growing at 8.8% annualised. Owner Occupied housing credit growth has continue to be strong but has not re-accelerated, running at 6.3% (Y/Y) and 5.4% annualised over the last three months.
Upgrading around house price assumptions
For 2017 our Economics team had been expecting a moderation of National house price growth to ~5% (Y/Y), given record housing supply/completions and tighter Chinese capital flows. However, the spike in year-to-date price growth (already >3% nationally) leads us to upgrade our 2017 forecast to 7% y/y, again driven by Sydney and Melbourne. The recent out-of-cycle rate hikes, plus the increasing likelihood of tighter macro-prudential policy, suggest house price growth should slow in coming months.
Sensitivity to higher rates
A key metric of housing affordability (for new borrowers) is the mortgage repayment share of income (assuming principal & interest) for a new 80% LVR loan on a median-priced home. We estimate mortgage repayments have lifted to ~27% of income, the highest level since 2011 (i.e. housing affordability has deteriorated sharply). This compares to the long-run average of 23% since 1978. Importantly, with record low interest rates, repayments are nearing extreme historical tipping points where prices fell. Indeed, if mortgage rates rose by only 100bp, this measure of affordability would approach the worst on record (similar to 1989 and 2008). As a result, we think the RBA will be quite slow to hike the cash rate, especially amid renewed out-of-cycle mortgage rate increases by the Major and Regional banks.
Housing activity – a correction, not a collapse
Nonetheless, we stick to our forecast for a ~30% peak-to-trough drop in the quarterly pace of dwelling commencements, with a forecast decline to 200k in 2017 and further drop to 180k in 2018. This leads to GDP-basis real dwelling investment slowing to ~flat y/y in 2017, and a 7% y/y decline of in 2018, which subtracts close to ~½%pt y/y from real GDP growth.
Australia has already been expecting a sharp correction of new housing activity (see our 81-chart deep dive report for details… Housing outlook: how much of a downturn?), given increasing indications of tighter credit conditions and developer caution. Indeed, our forecast for commencements approaches prior housing cycle downturns, which have historically been triggered by RBA rate hikes (see Figure 7). Overall, our view of a ‘correction, but not a collapse’ (which is arguably still a ‘soft landing’ for the broader economic impact) has been supported by the recent trend of residential building approvals. Although approvals have dropped back from a record high trend of ~240k in the last couple of years, down to ~210k in recent months, they are importantly showing some signs of stabilising (rather than quickly collapsing down to a long-run ‘normal’ level of closer to 150k).
Increasing calls to address the ‘housing affordability crisis’
As house prices continue to rise there is growing disillusionment in elements of the community (especially first home buyers) who are blaming these pressures on Investment Property buyers and foreigners. These issues are now being debated on an almost daily basis on the front page of most newspapers and media channels.
This has led to increasing pressure for the Government to respond. In a recent radio interview (Ray Hadley on 2GB, 13th March) with Federal Treasurer Scott Morrison the vast majority of the interview was focused on Housing Affordability. While hinting that a major focus will be on “freeing up supply” he stated “Those answers will be delivered in the Budget” (due in May).
The RBA had previously been pushing back on the extent of momentum in housing prices but now appears more convinced of the recent re-acceleration. There has been no indication from the RBA of a preference to tighten Monetary Policy to reduce house price inflation and the risk of instability. However, RBA Assistant Governor, Michele Bullock suggested further Macro Prudential tightening may be considered, stating
“The early experience suggests that, while the resilience of both borrowers and lenders has no doubt improved, the initial effects on credit and some other indicators we use to assess risk may fade over time. We are continuing to monitor their ongoing effects and are prepared to do more if needed”.
This is consistent with recent comments from Wayne Bryes, Chairman of APRA who said in a speech on 10th February
“We are not complacent, however, as recent months have seen a pick-up in the rate of new lending to investors… But, at least for the time being, the benchmarks that we communicated – including the 10 per cent benchmark for annual growth in investor lending – remain in place and lenders that choose to operate beyond these benchmarks are under no illusions that supervisory intervention, probably in the form of higher capital requirements, is a possible consequence. If that is encouraging them to direct their competitive instincts elsewhere, then that’s probably a good thing for the system as a whole.”
Potential policy options
There have been a number of proposals to address the “housing affordability crisis”. We believe the most likely potential options could be:
Potential Political Responses
(1) Eliminate Negative Gearing – There is a growing chorus of people who believe negative gearing should be removed entirely (existing transactions being grandfathered) or only allowable for new properties (Labor’s policy). The view being that negative gearing provides an unfair advantage to property investors relative to owner occupied upgraders and first home buyers. We believe such a policy change is unlikely near term as it has been rejected on multiple occasions by the Liberal government and is a key policy of the Labor opposition.
Implications for the banks – We believe this would lead to a sharp reduction in Investment Property credit growth. The marginal buyer in the market could be sharply reduced. There is a risk that such a policy may lead to many investment property owners wanting to sell and a large supply of housing coming onto the market. Some impact on the broader economy would be likely.
(2) Capping negative gearing to a dollar deduction or number of properties – We believe a policy of capping negative gearing to a set number of investment properties (perhaps two or three) or a dollar value deduction cap is more likely. Given the Government has stated on several occasions that many people who use negative gearing are middle income earners, capping deductions would not impact these constituents. We see such a proposal as more likely.
Implications for the banks – We think this is likely to be more moderate as only the most leveraged investors would be removed from the market. However, investment property credit growth and house price inflation would be likely to slow. This may be partially offset by an increase in owner occupied credit growth.
(3) Reducing Capital Gains Tax (CGT) discount from 50% to 25% – We see this as the most likely political response by the Federal Government. Given the yields on most residential properties are low (around 3% in Sydney and Melbourne) new property investors are more reliant on future capital gains. By reducing the CGT discount from 50% to 25% this would reduce the incentive to gear into the housing market and would increase revenue for the Government over time (exiting purchases would likely be grandfathered). Another alternative would be the reintroduction of CGT indexation, although we see this as less likely. A differentiation of CGT by asset class is possible, but would probably be deemed too complicated.
Implications for the banks – Likely to be moderate as the most marginal buyers may be removed from the market. Some slowdown in IPL credit growth would be likely.
(4) Allowing Superannuation to be used by First Home Buyers for a deposit – This proposal has been floated numerous times over the years, with the thought being to allow first home buyers to access another pool of savings for their deposit and to compete with Investors. However, we believe if FHB were allowed to access Superannuation for a deposit it would only fuel house price inflation. Further, this savings pool would inevitably be leveraged leading to accelerating credit growth. We believe allowing FHB to access their superannuation would effectively lead to a direct transfer of retirement savings from younger FHB’s to downsizing baby-boomers and would do little to nothing to address affordability.
Implications for the banks – Seen as positive near term as it would lead to increasing owner occupied housing credit growth.
(5) Stamp duty reductions for FHB’s – Has been proposed as a way to help first home buyers into the market, especially if they are purchasing new properties or under a price cap. This policy has been announced in Victoria for properties costing less than $600,000, with stamp duty discounts for properties up to $750,000. We believe such policies could potentially be rolled out in other states, however state budget positions means this may be unlikely
Implications for the banks – Unlikely to be material.
(6) Higher stamp duty on foreigners – We see further increases in stamp duty for foreigners as likely, especially in New South Wales and Victoria. This would help local purchasers compete and raise additional tax revenue for the state governments. It has also been used extensively overseas (eg Hong Kong). Finally, foreigners don’t vote.
Implications for the banks – Unlikely to be material.
(7) Attempts to accelerate housing supply/reform – As stated above this is appears to be a key policy in the upcoming Federal Budget. However, many of the rules, regulations and taxes that inhibit the increase in supply of housing are state based not Federal. As a result we believe such policies are unlikely to be effective near term.
Implications for the banks – Unlikely to be material.
Monetary Policy Response (1) Hike rates – This seems like the most obvious way to stem house price inflation in Australia. However, until recently the RBA has been pushing back against the extent of house price growth and has only recently become more convinced prices are again picking up. That said, we believe the RBA is unlikely to hike rates near term given low levels of CPI, with the RBA’s commentary indicating a preference for more Macro Prudential initiatives by APRA.
Implications for the banks – While an increase in interest rates would theoretically be positive for the banks’ NIM it would likely be negative for sentiment and consumer confidence. Fears of further hikes and the implications on the credit cycle would come into play. As a result it is unlikely banks would perform well despite a modest earnings boost from higher rates.
Macro Prudential responses
(1) Reducing the cap on IPL credit growth from 10% to ~7% – Although there have been no official comments from APRA on reducing the cap on Investment Property credit growth, this was alluded to by both Wayne Byres and Michele Bullock (RBA). A cap of around 7% would still be twice the level of household income growth and a logical step to reduce the risk to financial stability. We see this as a likely outcome in coming months.
Implications for the banks – Would lead to a reduction in investment property credit growth which is unlikely to be offset by owner occupiers. However, the banks would be likely to increase IPL interest rates to both slow credit growth and insulate earnings. As a result we do not believe this would be a material earnings driver, although it could increase negative sentiment towards the banks.
(2) Sydney and Melbourne postcodes limits on IPL credit growth or high LVR lending – Given the vast majority of house price inflation in Australia is concentrated in Sydney and Melbourne policies which target these cities are possible. This would be done most easily on a postcode or municipality basis. This may include limits on IPL credit growth within these postcodes or LVR caps. These policies have been introduced in New Zealand given the rapid house price growth in Auckland. However, it is difficult to tell how effective they have been given ongoing interest rate cuts by the RBNZ.
Implications for the banks – Similar to reducing the cap on IPL, we believe these policies would lead to a reduction in investment property credit growth and be partially offset by repricing. However, we are unsure if all the banks’ systems would be sophisticated enough to have differentiated pricing at a geographical level. It could also increase negative sentiment towards the banks.
(3) Tightening lending standard and documentation to include provision of tax returns – We believe mortgage misrepresentation is systemic in Australia with our recent Evidence Lab study showing 28% of mortgagors stating they were not factually accurate in their application. This was especially evident through the mortgage broker channel. Importantly, 18% of the people who were not factually accurate and used a mortgage broker stated they overstated their income.
One of the most unusual elements of the mortgage application process in Australia is that ATO tax returns are not required to be provided as proof of income. Instead pay slips and Group Certificates from employers are required. Given systemic mortgage misrepresentation, we believe this is no longer effective and it is likely APRA will begin to require the banks to check tax returns as a form of income validation (tax returns are required in most countries). This would also allow the banks to see all deductible debt and other tax deductions claimed by the applicant which could be factored into mortgage serviceability calculations.
Implications for the banks – Likely to lead to a reduction in both investment property and owner occupied credit growth as mortgage application process is tightened.
(4) Increase capital requirements for IPL – This is a potential option for APRA but we believe would most likely be used as a last resort. That said it was one of the Basel 4 proposals where the borrower was materially dependent on the rental income to service the loan. However, given discussions around Basel 4 continue to be pushed out this is now less likely. Alternatively, APRA may consider using a Counter Cyclical Capital buffer (despite credit growth being below trend) as has been used in Hong Kong.
Implications for the banks – This would be a clear negative for the banks and put further pressure on their capital position at a time when they are required to accumulate additional capital to reach their ‘unquestionably strong’ benchmarks. While the banks would reprice their IPL books to offset some of this impact, share prices would reach negatively, in our view.
Impact on REIT sector
We have back-tested Macro Prudential measures (10% investor credit growth limit, foreign stamp duty) to get a sense of the possible impact to REITs (Mirvac, Stockland and Lend Lease) as well as the impact to house prices. We have also examined the impact of macro-prudential measures on the Auckland market for comparison purposes.
Our analysis found the 10% investor credit growth cap and extra stamp duty impost on foreign buyers correlated with c. 2% underperformance in the 60 days leading into and 30 days after implementation. While fairly small, there is a risk that negative sentiment grows.
In 2015 SGP/LLC/MGR underperformed the REIT sector 10-20% for most of the year after being impacted by concerns over growing apartment supply, settlement risk and further macro-prudential regulation. These concerns were ultimately unrealised and the stocks have since partially re-rated. Looking forward, we don’t expect the initial ‘sticker shock’ to be as pronounced.
For the past 3 years national house price growth has range traded between 6-13% p.a. growth. The 10% investor credit growth cap had some success keeping house price growth around 8% p.a. for a 6mth period, even in the face of two successive rate cuts. Compared to the current run-rate of 13%, a slow down to 5-6% should be welcomed in our view as it elongates the cycle and may put less pressure on the RBA to raise rates.
By comparison, in Auckland, New Zealand an LVR crackdown from 80% to 60% and the introduction of an investor capital gains tax has cooled house price growth from 27% y/y to 6% despite a substantial 7 rate cuts over the 18 month period.
All sensible enough. What do I think that they will do? Cut the investor lending speed limit to 6%, just below where it is now, which will slow the market but crash nuthin’.