Week Ending 17.11.2017
- Australian wage growth is at historic lows, but business spending could provide support in the coming year.
- Low inflation appears likely to persist. It supports spending, but this is offset by high household debt and the mix of living expenses versus the structure of the CPI.
- US inflation is heavily skewed towards house rental costs and is likely to keep inflation low, even when wages rise.
- The impact of the CPI on portfolios is mostly indirect, but nonetheless pronounced.
There was much teeth-gnashing after the release of the weak Australian wage growth numbers. The index registered a 2% increase for the year, with the private sector at 1.9%, behind the public sector at 2.4%. This is after the impact of the increase in the minimum wage, which is estimated to have added 0.2% to wage growth in Q3. The retail sector, typically impacted by minimum wages, only came in at 1.6% for the year.
A notable feature is the narrow band across all industry sectors, ranging from 1.7% in culture, recreational and health jobs, to 1.2% in professional services. Even in construction, the rise was muted, notwithstanding the healthy growth in state based infrastructure spending and the still active housing pipeline.
The most optimistic view on wages is that the trend is lagging the improvement in employment and upbeat business sentiment. Business conditions this week were judged to be the best since 2008, based on the NAB survey. Ironically, this may well reflect the muted wage tone with little by way of labour cost pressure.
There is more than sentiment in the positive tone from business, with cash flow on a distinctly upward trend in the past six months. Those that believe the Australian economy can surprise on the upside expect corporations to put this to work, with the resulting private investment spending flowing through to labour market conditions.
Business cash flow (net balances)
This source of growth is indeed required given the household sector is unlikely to come to the party. Ironically, if business investment does pick up, the RBA also may be expected to turn to a hawkish rate stance, which would keep the household sector in check.
There has been much debate in the US on the low growth in wages there, given unemployment is now at 4.2%. Weak labour productivity and lack of a formal wage system are assessed to be the main hurdles to employee compensation. A similar decline in productivity here came at the same time as the fall in terms of trade (export versus import prices) reducing national income growth. With the upturn in the terms of trade in the past year, wage growth should pick up, but remain well below the levels of the 2000s when commodity prices accelerated. The October unemployment rate also ticked gently lower this week at 5.4%, though the net new jobs added was a measly 3,700. Somewhere around 5% labour conditions are likely to be tightening enough to see private sector wages head in the right direction.
One anecdote that attracted a comment in the labour data was a rise in the employment numbers in Queensland, attributed to net migration to the state as households seek more affordable housing.
If wages do indeed pick up into 2018, the missing piece for the RBA will be inflation. We have frequently commented on the problems with the structure of the CPI as a measure for living expenses, but one could argue it is a reasonable benchmark for the RBA.
The CPI has been reset to reflect changes from the household expenditure survey and some adjustments to the way housing is calculated. The new weights that will be incorporated in the December quarter CPI are as follows:
CPI weights as at September 2017
The housing component is made up of rent, which has been upweighted in this series given the rise in private dwelling rental as a proportion of occupied homes from 28.8% in 2011 to 30.2% in 2016. Rental costs have risen slightly above the overall price index over this five-year time frame at 20.2% compared to 16.3% for the CPI. New dwelling purchases (established housing sales are not included) have fallen as a component of the CPI, as the price impact has lagged, due to the sharp rise in the proportion of low value apartments from 18.2% to 34.7%.
For those that don’t rent or buy new homes, a considerable proportion of the housing CPI is therefore not in their mix. Conversely, high consumers of education and childcare (up 56% from 2011 to 2016), insurance and financial services (up 33%) and healthcare (up 19%) would be experiencing relatively higher living costs. The lowest rises in price over the five-year period were in clothing (3.6%), transport (3.9%) and the winner was communications, where costs rose by only 2.1%, (bearing in mind this is unit costs, not consumption level).
As an aside, the designated sector ‘furnishings, household equipment and services’ includes childcare and hairdressing (which comprise 3% of the 9.4% weighting). One can only muse on the decision to add these to a mostly durable goods-based component!
The ABS tracks the living costs of cohorts of the community. These are segmented into four groups. Those that rely on government support are shown separately as well as combined into the PBLCI segment. This provides a guide for the federal budget in terms of planning social welfare expenditure with the data showing that these two cohorts have been/had to be constrained in their spending growth compare to employees and particularly, self-funded retirees. As the government transfer and aged pension group increases as a proportion of the total population, this also acts as a significant drag on aggregate household spending and is, along with the high household debt, another reason consumption growth is much lower than in the past.
Average weekly expenditure. Change from 16th series (2011) to 17th series (2016)
- The purpose of this analysis is directly applicable to the equity market. Stocks with a reliance on consumption are likely to remain under pressure, unless they can find product and service segments that are capable of gaining share from the constrained hip pocket
It is worth noting the difference between the CPI in the US and Australia, which highlights one of reasons the US rate remains stuck at low levels. Housing costs (sometimes designated as shelter) is nearly 43% of the CPI (the Australian equivalent is 29%). This component comprises rent as well as owner-equivalent rent. The US therefore includes all households as though they were renting, whereas the purchase of a house is considered investment, not spending.
The healthcare cost weight for the US is also 3% above those of Australia; no surprise for those that have tracked the long-standing debate on this topic.
Naturally there must be offsets where the Australian weight is higher. Locally, we allow for a much higher allocation of spending on food, but more important are alcohol and tobacco, which add about 9% to the weighting on these goods compared to the US. Tobacco prices are between three to six times higher here than in the US, depending on the state tax regime. The other differences are in recreation and financial services, where Australian weights are 2% higher.
The overall point is that for the CPI to meaningfully increase in the US, housing rents must rise. Given that the homebuilding sector in the US is less constrained by local restrictions, it is unsurprising the CPI is low. Locally, wages have a relatively higher impact.
- The CPI has an important role across most investment portfolios. It can be used as benchmark (which we believe is far from ideal as financial asset prices may not follow inflation in the short to medium term). Fixed income is invariably sensitive to the CPI and some securities are linked to inflation. In the equity market, utilities, infrastructure and real estate can have a CPI formula in the revenue growth. Low cost pressures help margins, but also dampen pricing power. Companies with new products or enhanced value products and services are likely to make super-normal profit.
Fixed Income Update
- The US treasury yield curve flattens over the last three months.
- In light of recent outflows in US high yield, we examine the market today compared to ten years ago.
The shape of the US yield curve has flattened further in the last three months, led by a rise in yields on the front end, as the likelihood of the Federal Reserve Bank raising interest rates again in December gets priced into markets. The yield on the 2-year treasury is now at 1.70%, which is 0.42% higher than where it was trading three months ago, and at a nine-year high. Yields on 10-year treasuries have only increased by 0.19% over the same period. The probability of a rate hike in December is priced at 97%, up from 85% at the end of October.
Further, volatility in the Treasury market as measured by the MOVE index, has fallen to new record lows this month, down from 72.5 points at the start of the year to just 44 points last week, and currently at 47. The long-term average of this index is at 100 over a 30-year period. These low levels of volatility, the flat shape of the yield curve and the mismatch between the Fed and the market’s expectations for rate rises next year have us concerned that the long end of the curve is vulnerable to a sell off. Others oppose this view, pointing to low inflation, demographics, high debt levels and demand for treasuries from pension funds. Fund strategies that are benchmark aware with a high weighting to US duration will be most affected in the event of a shift higher in yields and a steepening of the curve.
- We prefer funds that have a flexible mandate and are unconstrained in their approach including duration positioning.
US Yield Curve Flattening
Last week saw the biggest outflow in eight weeks from high yield bonds, with $2.5 billion leaving the asset class over the last month, causing spreads to widen by around 40bp. Given the outflows, the largest two high yield ETFs have fallen to their lowest level in seven months. In the lead up to this, we have been observing the contraction in credit spreads in the high yield market and have had concerns on valuations. The average credit spread on high yield securities is trading inside of 400bp, which are levels not seen since January 2007.
However, according to a US based fund manager we met with this week, which specialises in the high yield sector, the composition and fundamentals of the market is very different now to ten years ago. Their reasoning includes:
- In 2007 the default rate in HY was at 2.25% vs today at 1.07%;
- There has been an improvement in company fundamentals and more favourable economic conditions;
- In 2007, 50-60% of new issuance proceeds were used to fund leveraged buy-outs compared to 65% of new issuance proceeds now being used to refinance existing debt;
- The size of the high yield market has grown from $0.7tr in 2007 to $1.3tr in 2017. The benefit here is that it means more diversified industries and less contagion in the event of a credit event in one sector;
- The composition of ratings in the market has also changed. There is now a higher proportion of BB rated companies versus ten years ago, with a rise from 37% to 48%. CCC rated companies have fallen from 20% to 14% since 2007;
- There have been ratings improvements, with the upgrade to downgrade ratio at 1.4x, which is its highest since 2013; and
- There has also been an increase in international demand for US high yield, increasing flows and liquidity.
Composition of Ratings in the high yield sector
While this fund manager expects an extension of the current credit cycle and believes that the tightening has been for good reason, they do have concerns over recovery values in the event of default and the tight pricing of the lower rated part of this market, i.e. CCC rated securities. Confirming their thoughts on contagion, the weakness in the high yield market has been broadly contained to the telecommunications sector. The telecoms within the index have risen on average by 1.2%, compared to the average of the whole index at 0.40%.
- Within our recommended funds, there is limited exposure to the high yield sector through the JP Morgan Strategic Bond Fund that is in our capital growth models. The holdings in this fund are diversified across a number of better quality companies. Positioning is opportunistic with no required mandate to hold high yield, with JP Morgan reducing exposure throughout this year. We are confident in their ability to effectively manage this part of the portfolio, and comfortable with the overall exposure that investors have to this sector.
- Harvey Norman (HVN) reported their financial year to date sales and they surprised on the upside.
- DuluxGroup (DLX) reported a good full year result, however, headwinds are emerging and its valuation looks relatively full.
- Domain Holdings (DHG) has partially demerged from Fairfax (FXJ) with an attractive outlook in the next few years accompanied by a similarly lofty share price.
Harvey Norman (HVN) reported year to date sales, up 4.8% in Australia (4% like for like). The group has pulled away from any commentary on these releases, though most construe that the legacy of the housing cycle is evident in demand for furniture and whitegoods. HVN may have benefited from some disruption at the Good Guys (now part of JB Hi Fi) where it has changed the business model.
Margins expectations should be met, given these positive sales and lack of store openings that historically resulted in subsidies or cannibalisation of sales from other stores.
The stock is trading at a modest 11X PE multiple and an attractive 7% forward yield. It therefore has found a place in yield funds. The potential for capital management given large franking pool remains an option, though the board has shown little appetite to endear itself to shareholders.
The stock trades in a wide and volatile range. This is a classic case where an astute fund manager is the appropriate judge of an investment opportunity rather than a direct holding through these cycles.
Harvey Norman (HVN) performance relative to the ASX 100
DuluxGroup’s (DLX) full year result was typically solid, with underlying profit growth of 7%, alongside a 10% increase in its full year dividend. The company’s core paints division (which represents over 2/3 of overall earnings) was again the key driver, while margin growth (largely on cost out initiatives) was a feature in its smaller Selleys and garage doors and openers divisions.
Dulux’s paints business has been a consistent performer for an extended period of time now, no doubt assisted by the strong conditions in the housing market. While the spike in new housing development is certainly a positive tailwind, the division is much more exposed to the maintenance and home improvement market. This itself has been given a boost from house price appreciation over the last few years, encouraging existing owners to renovate their property. A strong brand name and a good relationship with Bunnings has further helped with market share gains in this time, consolidating its leading market position.
While DLX’s track record since it demerged from Orica has been impressive, the medium term outlook is more clouded. The housing cycle has now likely peaked and rising titanium dioxide prices (a key raw material in paint manufacture) are likely to crimp margins in the next 12 months. DLX is less reliant on the more cyclical elements of the housing industry (such as apartment development), however it still likely faces a more subdued earnings growth environment in the next few years, reflected in consensus forecasts of low single-digit earnings growth in this time. Despite this, the stock’s premium to the industrials sector has increased through this year, leaving it vulnerable to a derating should it disappoint.
DuluxGroup P/E and Premium to ASX 200 Industrials
Also linked to the residential property market is classifieds company Domain Holdings (DHG), which made its market debut after completing a demerger from Fairfax (FXJ). Unlocking shareholder value via the separation had been a central part of the investment thesis of FXJ, particularly with a similarly listed comparable company in REA Group (REA) that had enabled analysts to estimate an implied value of the standalone business. Following the separation, the two stocks will appeal to different types of investors, as it has carved out a part of the high growth component of FXJ’s earnings base.
While REA is a useful starting point in valuing DHG, the two do have their differences; DHG is focused on the domestic market, while REA has a presence in international markets; REA has greater scale and thus earns higher margins; DHG still has some (small) revenues from print advertising (in The Age and Sydney Morning Herald), which should be expected to decline over time; and finally, DHG has developed an “agent ownership model”, which is effectively an arrangement that DHG has with real estate agents whereby they share in the benefits of selling more premium listings.
What may be a surprise to some is that the real estate classifieds are not necessarily tied to the strength of the housing market; a hot market can actually be a headwind for listings if auction clearance rates are high and properties take less time to sell. Instead the key driver of growth is market share gains and improvements in yields. There is a reasonable argument to be made that yields can improve for the foreseeable future (and more so compared with the other key online classifieds categories of cars and jobs), given the duopoly status of the online market, the current low proportion of premium listings and finally, the fractional cost of advertising relative to the sale price of property.
The outlook for DHG looks relatively bright and it should be one of the few large cap companies on the ASX that can sustain double-digit earnings growth in the medium term. With the stock commencing trading on a forward P/E of approximately 40X, much of this good news is arguably already factored into its current valuation. Achieving market share gains and margin expansion relative to REA is a likely requirement given that it is also trading at a healthy premium to its more established peer, which has itself made significant capital gains in the last year.
With FXJ retaining a 60% shareholding in DHG following the separation, it should be noted that this investment still comprises the core part of its current valuation; based on DHG’s current share price, this equates to just under 80% of FXJ’s total value. The remaining businesses – newspapers, radio and digital video streaming service Stan (a 50/50 jv with Channel Nine) – are thus currently ascribed little value by the market; perhaps not surprising given they are predominantly made up of structurally challenged, ‘old media’ assets. The upside potentially resides in M&A activity should media reform eventuate, while patient value-focused shareholders may see an opportunity given the support that DHG will provide to the share price and a relatively attractive yield.