From the always excellent Vimal Gor today:
Markets came into July braced for higher volatility, led by global yields. Despite sell-off attempts in both US Treasuries and German Bunds, yields ended the month significantly off their recent highs. Adding to this were positive data releases from the US, Europe and China, all absent of any sign of demand-pull inflation, resulting in the perfect environment for risk-seeking behaviour once again.
Whilst yields exhibited some volatility during the month, the response in risk assets was muted and largely benign. Relative to the overall move in US yields in the month, the moves in the US dollar were outsized, led by its continued year-long slide against the Euro. Perhaps this was the beginning of the summer of carry that was so rudely interrupted in June. Perhaps we do have the ingredients for another risk squeeze, led by strong growth and accompanied by disinflation, which keeps yields anchored. But the market reactions at the end of June and start of July give us some indication as to just how tightly wound the system might be. When volatility has been crushed, relatively small volatility spikes have the ability to create disorderly reactions across markets. Performance over the month was strong with all flagship funds outperforming their benchmarks.
From cuts to on hold…
The last few Newsletters have been focused on China and its impact on the global economy and markets. This month I wanted to look at the Australian economy given all the recent clamour for rate hikes and discuss how we are seeing things currently. We have had the view for a while that the next move from the RBA would be rate cuts, most likely in February and May 2018. This view was based on an emerging housing slowdown in late 2017, driven by the Macro Prudential measures, announced in April, which were aimed at cooling investor activity in certain property markets, and improving overall financial stability. These measures were introduced against a backdrop of low inflation, low wage growth, and significant excess capacity in the Australian economy, so it was easy to build a dovish case for the RBA. So far our view has been correct as we have already witnessed the MacroPru measures having some effect on the pace of house price inflation and contributing to a declining building approval outlook, albeit both from very elevated levels. In fact, our read of the internal view at the RBA at the time was that if it wasn’t for financial stability concerns, rates would already be lower.
However, the Australian economy is also influenced by many things un-Australian. For example, despite escaping the worst of both the global financial crisis and the European sovereign crisis, Australian interest rates have been dragged into unusually low territory alongside those of other developed markets. And with China its number one trading partner, Australia’s miners and commodity exporters received a freebie as China pursued “reform” of its coal and steel industries in 2016 in order to avoid potential waves of mass defaults in those sectors. The Australian economic backdrop also looks healthier today than at the start of the year solely due to the strength of demand from China. What looked to be a clear deterioration of Australia’s terms of trade at the beginning of the year was sharply halted then reversed when the regulatory brakes were loosened on the Chinese financial system. Chart 1 depicts just how closely the fortunes of Australia are tied to levels of economic activity in China.
I have explored the China macroeconomic situation in recent newsletters and I won’t dwell on it this month, except to say that there is credibility behind the promise of stability in a year of leadership reshuffle. And it is largely this view of stability in China that is the reason we have taken out the two rate cuts in H1 2018, and now forecast unchanged cash rates for 2018.
Whilst certain indicators such as consumer confidence and wage growth remain below RBA’s expectations, continuing tailwinds from China in the form of improving prospects for Western Australia and Queensland, as well as stronger business sentiment and employment metrics contribute to an overall improved outlook. On balance, these factors allow the RBA to continue to run at the current 1.5% cash rate for longer. In fact, Governor Phil Lowe may well get to two years into his term as Governor (September 2018) without touching rates.
But to be clear we do not buy in to rate hikes in the coming 12 months. The economic readjustment process is still embryonic and ongoing and conflicting signals in the labour market and sentiment are likely to leave more room for caution from the RBA. For example, why is improving employment failing to lead to higher wages? Why are businesses confident while consumers remain cautious? And how well are Macro Prudential measures on housing likely to work?
Has the Philips Curve gone AWOL?
A number of countries have pushed through levels traditionally considered to be full employment and seen only tepid wage growth. In fact against the long held faith in the Phillips curve, Germany and New Zealand have experienced falling unemployment and falling wage growth and inflation. Only in the US has the large fall in unemployment driven a rise in wage growth, although at a far lower rate than historically.
This isn’t the first time that the Philips Curve seems to have deserted us. The expected relationship between inflation and employment also broke down during the stagflation of the 1970s. The issue is that the curve is a Keynesian demand driven concept and doesn’t fit the current world we live in which is being driven by the supply side of the economy.
The RBA is not alone in facing this ‘lack of wage pressure’ dilemma. This global problem has many suggested factors, including globalisation, technology, lower productivity and even job security, but labour market composition must also play a significant part. Luckily the RBA has the luxury of being able to wait and see how other employment markets resolve this before they need to act.
The RBA has been cautiously optimistic on employment growth for 2017. More recently, we’ve started to see the unemployment rate turn down, and is now closer to 5.5% than 6%. But we have to view this slight drop in the unemployment rate in a structural context, as there has been a major shift in the composition of the workforce over the last decade. Whilst around two thirds of current jobs are fulltime, only one third of jobs created in the last 5 years have been fulltime. This is a really worrying trend, and can be explained if we look where the jobs have been created, largely in the services sector.
The service sector while providing employment is very ‘harsh’ in its treatment of part time workers, many of which are not part time by choice. The HILDA survey indicates around 25% of part time workers are seeking on average an extra 14 hours per week. This is a massive amount of excess capacity which is obviously one of the reasons why wage growth is so sluggish.
This increase in the proportion of part time jobs has led to the RBA increasing its focus on the underemployment rate, as well as the unemployment rate, as a measure of excess capacity. As can be seen in Chart 3 below, whilst unemployment has turned down this year underemployment remains near its highs. We thinks it’s likely that the RBA would need to see unemployment down at what they think is ‘full’ employment before they would start to contemplate hikes. This level is likely still around 5%, although experience in the US suggests their NAIRU (non-accelerating inflation rate of unemployment) may be lower than previous cycles. There is no target for underemployment but that would also have to be significantly lower than here.
When pondering if better employment growth will eventually ever kick into wages – that is, lags are just longer these days – RBA Governor Dr Lowe observed in a recent speech it “can’t be completely discounted” but was of low probability. Rather the most likely outlook is that “wage growth picks up gradually”, meaning the RBA will not have a wage or inflation problem for some time to come. Remember that inflation targeting central banks were set up in a time, post the 1970s and 1980s, where their biggest job was to rein in inflation in strong economic recoveries through hiking rates. Whether the central banks mandates are fit for purpose any more is a valid discussion that should be, but won’t, be happening. But clearly inflation is not this recovery’s problem anywhere, and discussions still focus on how to generate but quell inflation. Simply put, there is no urgency to respond to strengthening employment with rate hikes, as would have happened in the past.
One positive note though is the recent Fair Work Commission (FWC) decision to increase the minimum wage by 3.3%. Nearly 25% of employees are covered by awards and the RBA estimates this increase will directly add 0.5% to the Wage Price Index. Indirectly it will also have an impact on workers not covered by awards but in related jobs. The RBA’s guess on this one is another 0.25% to 0.5% on WPI.
So on the (tenuous) link between employment and wage growth, the RBA clearly comes down on the side of high caution. Whilst recent employment numbers are encouraging they need to do a lot more heavy lifting to shift the focus to rate hikes.
Buoyant business, cautious consumer
The second unusual dynamic facing the RBA and the economy is the high divergence between the consumer and business sectors. The NAB business survey shows business conditions back to the frothy pre-GFC levels. Yet, the Westpac Melbourne Institute Consumer Confidence Index remains below the long term average. Normally they move in lockstep, but the current divergence has never been greater nor has persisted for this length of time.
Unsurprisingly the answer to this divergence largely lies in the wages discussion. Consumers are also wage earners who have not seen the benefits of an improving economy come through in higher real wages. Real disposable incomes have even fallen slightly over the last five years. Businesses have captured the economy’s growth through higher profits and have not needed to share this with employees. Again this is a global phenomenon and this low labour share of GDP is the driving force towards more populist political parties and is a subject I have covered at length in past newsletters.
Obviously low wage growth and therefore consumption is a problem for the RBA. Given consumption is around 60% of GDP it is difficult, if not possible, to see them reaching their 3% GDP target if consumers are only getting 2% wage growth. Business may be benefitting from strong conditions, driven by the construction and services sectors, but eventually the very workers whose subdued wages are helping their profits will reduce their consumption. This recently led to comments by the RBA Governor encouraging workers to be “prepared to ask for larger wage rises. If that were to happen it would be a good thing”, even as the RBA staff are handed sub-inflation pay rises.
In order for this sentiment gap to close it will involve not only an increase in consumer confidence but a pullback in business conditions. Again this will lessen any urgency from the RBA to hike rates should wages start to slowly turn.
MacroPru and its bite…
For a home owner living in Sydney, or Melbourne, the last few years have been excellent; not only has the weather been good but they have been ‘blessed’ with massive house price gains. In case you’ve not fully appreciated it, here’s what has happened to house prices in the capital cities in the last 5 years.
Of even greater concern to the authorities, who are tasked to monitor financial stability, is the rise and prevalence of investors, not only crowding out first home owners but rapidly pushing up outstanding interest only loans. Last year the flow of new investor loans was above 50% of total new loans, well over the roughly 33% average, and this flow drove the percentage of investor loans to c.40% as shown in Chart 9 below.
Late 2014 was the first call to arms for regulators, as investors were climbing to 45% of total outstanding loans. For a while, MacroPru Round 1 worked as can be seen in the pullback in Chart 9 above. Measures brought in then were a soft cap of 10% growth on investor lending and stress testing a 7% interest minimum floor on serviceability of loans. However rate cuts in early 2016 reignited the animal spirits and by March this year APRA introduced a far more rigorous Round 2 of MacroPru.
So let’s recap these Round 2 measures. Importantly, they are hard caps this time:
1. The flow of new interest only lending to 30% (from a 50% run rate)
2. Strict internal limits on interest only lending to loans above 80% LVR
3. The soft cap of 10% for growth in lending to investors becomes a hard cap with the expectation it will be comfortably below.
4. Tightening on serviceability assessments of borrowers These measures have flowed on to far stricter implementation by individual commercial banks, to both meet the requirements above but also meet behind the scenes pressures from APRA.
The commercial banks responded as the regulator had wanted a Round 2 of MacroPru, but they also used it as an excuse to jack up rates and increase margins. These responses have included:
1. The four majors increasing interest only mortgage rates by between 30bp and 35bp
2. Lenders introducing stricter LVR limits for investors. These limits are from 60% to 90%, but in general have fallen from 90% to 80%. This is partly to ration loans but also due to stricter eligibility criteria.
Collectively we are seeing APRA force the banks to ration credit to property investors. The rough expectation is that if investors are 40% of new mortgages and 70% of investors take out interest only, then banks should land below the 30% cap. Banks will be stricter on interest only for owner occupiers and given the wide gap between interest rates on interest only and principal and interest mortgages, the incentive to choose the former is greatly diminished.
In other (coordinated?) measures, recent state budgets saw NSW increase stamp duty on foreigners from 4% to 8% and Victoria introduce a new 7% land transfer duty on foreigners (on top of an existing, like Queensland, 3% surcharge).
These measures mean there will be a similar, and potentially sharper, impact on both activity and pricing in the housing sector than we saw in 2014, and this was the reason for our expectation for further rate cuts.
Impact on housing activity and house prices
APRA and the RBA have been explicit in saying that MacroPru is targeting financial stability and not house prices, as they have to. The RBA in particular would like to see an extended period where household debt and house prices increase slower than wages, the opposite of the last five years. With tepid wage growth that effectively means no growth in either. That is, MacroPru is partly designed to increase the resilience of the economy to future shocks by lowering debt to income levels.
Recent data released showed the buffers that mortgage holders have, as lower interest rates over the last few years have seen two-thirds of mortgage holders get ahead of their payments. This leaves round a third with no buffer, as shown in Chart 10 below.
Also the household savings rate has been rapidly falling.
Typically the savings rate declines with rising wealth. Animal spirits mean investors have little interest in saving when rates are low and asset values are rising, therefore it is often a warning sign of a boom whose bust will soon follow. Also of concern is that the recent HILDA survey also showed that between 30-40% of young home owners actually increased their debt in the last year. That is, people are accessing equity in their home to fund living expenses.
Given all these headwinds it should not come as a surprise therefore if the second half of 2017 looks different for housing than the first. Approvals are falling, construction is peaking, and a massive amount of new supply is coming on to meet the excess demand. Investors have had around a 35bp hike with potentially more to come.
House price growth will slow and potentially turn negative and, via the wealth effect, consumption will be affected. However with a stronger business outlook, on the back of China, this weakness should be able to be absorbed by the wider economy and seen for what it is – a necessary levelling out. There is no doubt there will be front page headlines of non-completions, pulled projects and mortgage stress. In the short term this may knock confidence further, but given the pull from China there is too much momentum elsewhere in the economy to cause a significant impact on growth. Regulators should be confident they will help engineer the long overdue soft landing in housing.
Housing therefore will give the RBA reason once again to keep interest rates on hold but we should look for a sentiment impact in late September/October as the spring season gets into full swing.
Where’s the anchor?
A recent internal RBA discussion paper came to the conclusion that the new normal for long term neutral real cash rates is 1%. Put in the 2.5% inflation target and that is a nominal cash rate of 3.5%. ‘Neutral’ means a rate where monetary policy is neither providing stimulus or restraint. The 1% conclusion the RBA arrives at is consistent with the US where the Federal Reserve “dots” congregate around 3% in the long term, or 1% above their target inflation rate.
The RBA is likely to provide more detail around how they arrived at this rate in the months ahead. What the forecast did do though was highlight the differing views in the market around this level. The assumption of this rate is a major input into any long term forecasting models, whether for economic forecasting or for asset valuations.
So what determines a neutral real rate of interest? The two commonly quoted sources are population and productivity. These drive real growth and therefore real interest rates. Whilst productivity has been lower for some time, population has been growing at around 1.5% for a number of years. If you assume some rebound in productivity then a real rate of 1% looks low.
However there are a number of other factors at play. The global interplay between savings and investment is also important. Whilst a country can run a persistent imbalance, through a current account surplus or deficit, globally savings and investment must balance. An excess of global savings relative to available productive investments drives down rates. If the investment outlook is weak globally then surplus savings will look for growth opportunities in countries like Australia, driving down our interest rates. The currency should also strengthen which would also have a dampening impact on rates. This had a big impact on Australian rates through 2011-14 when Europe was in trouble and was a driver of Australian cash rates down to the current negative real rates.
Another factor recently highlighted by RBA Deputy Governor Debelle is risk aversion. Increased risk aversion leads to higher savings and less investment, lowering neutral rates. Importantly market rates will widen to policy rates, as seen in the widening below between the RBA cash rate and mortgage rates since the GFC. Less than half of the move from 3% to 1.5% by the RBA reached mortgage holders.
Therefore to maintain a similar market rate as pre GFC a lower policy rate is required. This, combined with an estimated fall of 1.5% in potential growth due to lower labour force growth and productivity, has seen the RBA arrive at a 1% neutral real interest rate.
Term real rates, as measured by 10yr Treasury Indexed Bonds, have been below 1% since 2014, so the RBA estimate is close to market estimates and was not a surprise. However a number of models used for asset valuations still had higher real rates in them and may need repricing.
It will be interesting to further read how the RBA calculate it so we can keep track on how this dynamic estimate may change in the months and years ahead and therefore potentially drive changes in monetary policy.
On hold, with reason for caution
In summary, we have removed our call for two cuts in 1H 2018, due to an improved economic outlook driven by continued strength and stability in the Chinese economy. However, we think it is too early to call for hikes, and the lack of wage growth recovery and growing gap between business and consumer sentiment are reasons for the RBA to err on the side of caution. Moreover, we have yet to see the true bite of Round 2 of MacroPru on the housing market. Should these measures prove too successful in curtailing housing, rate cuts may come back into play. On balance, if rates were to move in the next twelve months, cuts are more likely than hikes.
I remain comfortably of the view that the next move remains down as China slows and a falling terms of trade adds to all of the other headwinds described.