Its not all about income
The thirst for yield has been the rallying cry for investors around the globe over the last year or so. As central banks reduce (or have reduced) cash rates, we are witnessing a global shift towards higher yield equities and bonds. In the specific case of Australia, why put your money in the bank when you can own it and get a higher payout, especially after tax?
With respect to equities, there is plenty of academic evidence to support the notion that dividend yields are an important (or indeed the most important) component of total returns over time1. With equity dividend yields globally generally higher than bond yields, the attraction of an investment strategy anchored on income generation is easy to comprehend. This is particularly the case for investors who require some form of regular income stream to be paid from their assets, for example part or full retirees.
However, yield hungry investors would do well to take stock of their real investment objectives before making the headlong plunge into rapidly appreciating high yielding stocks and bonds. Income is but one (important) component of the total return equation, but it is not the only component. Ultimately it is total real return and the way in which it is delivered that matters and not the components of that return (tax differentials aside).
Maximising total real money weighted returns is the real objective for all investors and best achieved through:
- Reducing downside volatility of the total portfolio to reduce “sequencing risk”;
- Biasing the portfolio towards those assets which offer better real total returns commensurate with the investors risk objectives.
In particular, observed dividend yields on stocks are in some part a manufactured outcome that is the result of real cash flow, earnings, payout level, leverage and the costs of leverage. An overt focus on headline distribution yield can blind investors to a significant change in real underlying risks of the investment such that a safe, stable dividend may end up leading to a significant capital loss. Investors would do well to remember the folly of those who chased yield into listed property trusts and structured credit products prior to the GFC.
This paper is an analysis of the characteristics of portfolios skewed towards dividend yields and a review of alternative portfolios that delivering real money weighted outcomes to investors, particularly those in the drawdown phase where income is especially relevant.
It is our contention that valuation is the key factor in determining which asset classes offer better longer term risk adjusted real returns and in reducing downside volatility of the total portfolio. Yields offer some insight into valuation but are not the determinant of valuation or future performance and therefore should not be overemphasised in the portfolio construction process. In particular the significant outperformance of higher yielding Australian equities (and high yield credit) has increased the risks of capital loss to those investors.
Why the focus on income?
There are a number of reasons why income is appealing to investors, in particular the ongoing need for cash flow to meet spending needs such as in post-retirement. The current emphasis on income generation is relatively easy to understand from a behavioural perspective in light of the fall in yields around the world from traditional yield areas (e.g. cash and bonds). The chart below shows the collapse in real bond yields in the US and Australia.
Chart 1: US and Australian 10 year Government bond yields (1998-2013)
Source: Global Financial Data, Schroders
Not unsurprisingly, this collapse in real bond yields has meant that yield has become scarcer in the “safer” assets such as cash and bonds. As investors have reacted to the “improving state of the global economy” pumping of liquidity into the global economy via the various Quantitative Easing strategies in place around the globe, the push into higher risk assets has grown. However, while credit bore the initial onslaught of liquidity (somewhat rightly given the risk-adjusted yields available at the time), this has gradually broadened out into higher yielding stocks as investors seek some form of “safety” in the chase for yield. This is evidenced in the strong outperformance of higher yielding stocks in Australian equities and the broader market response to QE in the charts below.
Chart 2: Performance of high yield stocks in Australia relative to market
Chart 3: S&P500 and various QE programmes
Source: UBS, Morgan Stanley Research. Higher yielding stocks are definedas ASX 100 large caps with higher than average yield.
In particular, Chart 4 shows that equity yields have surpassed bond yields only rarely in the last 30 years. Consequently, we contend that many investors consider higher yielding equities a better longer term investment (possibly with less real risk) than bonds.
Chart 4: Australian equity vs 10 year Government bond yields
Source: Global Financial Data, Schroders, ASX200
What’s wrong with the emphasis on income?
In emphasising the relative differential between equity and bond yields, it is worth putting this in a longer term context. The current state of affairs, where equity yields now exceed bond yields, while unusual in the context of the last 30 years, is not out of historical norm. In fact it is only in the last 30 years as central banks moved to fight inflation aggressively and disinflationary forces have been in play that equity yields have fallen below long term bond yields. Chart 5 shows equity yields generally higher than bond yields, however now, like bond yields, near their historical lows.
Chart 5: Australian equity vs 10 year Government bond yields
Source: Global Financial Data, Schroders, ASX200
Investors would do well to remember that equity yields can (to a point) be manipulated as with the yield on any structured investment (e.g. managed funds , guaranteed funds, structured products, CDO’s, property and other trusts)2. When there is a thirst for yield, supply will expand to meet that demand. Ultimately however, it is real underlying cash flows that matter. The “manipulated” yield is essentially a smoothed version of the underlying earnings based on qualitative decisions by management about payout ratios.
Chart 6 shows that while dividend yields in the US have been somewhat stable through time, the earnings yield has been significantly more volatile and ultimately dividend yields are just a smoothed, manufactured result of the underlying earnings. In fact, the earnings yield has had double the standard deviation of the dividend yield throughout this period.
Chart 6: US earnings yield vs dividend yield
Source: Global Financial Data, Schroders
Clearly for yields to be high in the long term on any asset class there will be a need to generate the cash flows to support those yields. Consequently, yield can serve as a useful valuation yardstick. However, it is not in and of itself the final measure of valuation, particularly given its ability (and track record) to be manipulated.
More pertinent questions that investors should ask are:
- To what extent are current dividend yields correlated with the future real performance of the asset class in aggregate?
- Does a focus on higher income securities within a specific asset class bias a portfolio towards potentially more negative capital outcomes?
Asset class level analysis
Firstly, on the basis that it is real returns that matter, investors should consider real dividend yields not nominal yields. Chart 7 shows the historical relationship between real yields on Australian and Global equities, bonds and cash.
Chart 7: Real 3 year rolling yields: 1926-2011
Source: Global Financial Data, Schroders, 1926-2011.Bonds – Australian 10 year Government bond yields.Other data as per appendix.
Chart 7 also shows that real yields have varied quite substantially (no surprise if nominal yields haven’t and inflation has). In addition the relationship between asset class real yields has also varied substantially in the last 30 years relative to the prior period.
As noted, our contention is that it is total real return that matters, not nominal or real yield. As such, Charts 8 and 9 show for Australian equities (at an overall asset class level) how the real dividend yield at a particular point in time has correlated with future 1 and 10 year real returns.
Chart 8: Australian Equities Real Dividend Yield vs 1 Year Future Real Return Minimum, Maximum and Average ( 1969-2013)
Chart 9: Australian Equities Real Dividend Yield vs 10 Year Future Real Return Minimum, Maximum and Average (1969-2013)
Source: Global Financial Data, Schroders,1900-2011
On a 1 year future view, not surprisingly there is almost no correlation between average future return and current dividend yield. However, on a longer term basis somewhat more surprisingly there is still almost no correlation between current dividend yield and future returns for most yield deciles with the exception of the uppermost 2 or 3. Note that these yield deciles correspond with real yields on Australian equities in the order of 7.5% or higher, well in excess of levels today. In fact, the current real yield on the Australian equity market corresponds with decile 2 in the Charts above.
Our conclusion from the above is that yield is one component of the total return and one potential measure of valuation. However, it is not the measure of valuation and it is only in the extreme that yield and valuation potentially correlate.
To further illustrate this point, Charts 10 and 11 show the same data divided into quartiles. However, in Chart 10 shows the entire data set since 1969 for when P/E data is available, and Chart 11 shows those periods where P/E’s were greater than their long term average of 16.5 (but using the same quartile definitions).
Chart 10: Australian Equities - Dividend Yield vs Future 10 Year Real Return - All Observations . 1969 – 2013
Chart 11: Australian Equities - Dividend Yield vs Future 10 Year Real Return - P/E > Average. 1969 - 2013
There are 98 observations in each of the quartiles in Chart 10. Chart 11 had respectively 82, 51, 34 and 6 observations for each quartile.
Source: Global Financial Data, Schroders, 1969-2011
Quite clearly, it is valuation that is the dominant driver of future returns, not yield. In fact, those periods where yields were high, but valuations greater than their long term average showed the worst future 10 year real returns. At present, the Australian market is somewhat above its long term average P/E.
Focusing solely on dividend yield or income from a portfolio construction perspective can lead to sub-optimal portfolios that offer potentially lower total returns and potentially higher risk. This is clearly not maximising the objective function for investors requiring income where lower total volatility (particularly downside volatility) and maximising real total return outcomes is more important.
Stock level analysis
While the above presents an asset class view of yield and its impact on returns, of greater relevance to investors right now is what characteristics a yield biased portfolio within an asset class might reveal. In the case of bond portfolios, this is relatively straightforward. For an individual bond, the yield is known at the time of investment and, short of a default, this will not change at the whim of the market or management (assuming the bond is held to maturity). Chart 12 shows for various credit ratings the current spread, range of spread and “rock bottom spread”. Rock bottom spread is our analysis of the minimum required return in order to offset default risk.
Chart 12: Credit Spreads and Rock Bottom Spread. As at 17 May 2013
OAS = Option adjusted spread
Source: OAS to LIBOR data sourced from City Velocity (Citi Fixed Income Index – AusBIG Index – Corporate Rating Category) and Barclays Global High Yield Index. Min and max range indications have been trimmed by 5% at both points to remove extremes. Rock Bottom Spread per Schroders calculation using average term of relevant City Velocity data set and Barclays Global High Yield at 6 years average term using Ba3 average rating.
As one would expect, the lower the credit quality the higher the yield. Clearly in this case the real question for investors then becomes is the yield increment sufficient to justify the increased risk of default (read as the rock bottom spread). This is a more objective assessment, however at least the “price” for default risk can be assessed.
However, with respect to individual equities, there is no “hold to maturity” return from the dividend as there is no maturity. Consequently, the capital value at the time of sale has a very important bearing on the total return and will contribute significantly to overall risk. Chart 13 highlights this experience for holders of A-REITs/LPT’s over the last decade.
Chart 13: Income and capital return in A-REITs: 2000- 2013
Source: Schroders, Global Financial Data
While the “income” level may have looked attractive prior to 2008, quite clearly there was an increasing level of risk in these vehicles that was being masked by the apparent stability of the income yield. As of today, investors are still down in nominal terms on their investment of 13 years ago. This emphasis on distribution yield in an asset where there is the ability to manufacture yield by varying the payout ratio and changing the capital structure (and hence risk) means that the chase for yield in and of itself as a portfolio construction approach can materially increase risks.
This high payout ratio thematic has been particularly important as headline yield has totally overshadowed valuation in the short run. Chart 14 shows the payout ratio vs. P/E for ASX200 stocks.
Chart 14: Payout ratio (December 2014) vs PE (using Consensus December 2014). Calculated as at June 2013.
Source: Schroders, IRESS.
It is clear from the above that those stocks with higher payout ratios have in general been rated more highly by the market. More specifically, in Australian equities the thirst for yield has driven capital into the financial services sector given the relatively high payout ratios, dominance of domestic earnings (and hence franking credits) and the high proportion of profits made by financial services businesses in Australia. It has also supported defensive industrials and REITs. However, the pure size of listed financial services relative to industrials dictates that this is where a disproportionate amount of market capitalisation has been added.
In the case of banks, it is our view the risks in using the sector as a defensive income proxy are not fully appreciated. Credit growth since the early 1990’s has clearly been explosive. An extended period in which policymakers disregarded inflation led to consistently falling interest rates and double digit credit growth. As displayed on Chart 15 the vast majority of growth has been in the housing sector, whilst credit growth in other sectors has remained relatively stable as a percentage of GDP.
Chart 15: Credit to GDP
Source: RBA, ABS, Schroders
We are now at a point where household leverage and house prices are high, and the primary justification for the stability of our banking system is the quality of household balance sheets. This is clearly a circular reference, as property values are the dominant source of household wealth and values are supported by the quantum of credit.
The value of banking sector equity against total loan assets is flat as a percentage of total assets, suggesting that banks are just as highly geared as they have been through the past 30 years. Simplistically, in our opinion the banking system as inherently more unstable than it was prior to the credit boom for a few simple reasons:
- The nominal value of the system is massively larger and has expanded at a far greater rate than the revenues and incomes.
- The process of credit growth promotes increasing wealth inequality.
- Lower inflation has made mortgage loans far longer duration assets. Given that equity has not expanded relative to the size of the loan book and profits are currently shouldering bad debt charges which are at historically low levels versus the size of the book, we necessarily conclude that the banking system cannot be lower risk despite assertions of participants to the contrary.
Chart 16: Australian Banks Market Cap to GDP: 1990 - 2013
Chart 17: Equity to assets – Australian banks : 1990 – 2013
Source: Factset, Schroders. As at March 2013.
It is expected return on capital to remain the primary driver of stock performance in the long run. The impact of quantitative easing and artificial easing of the cost of capital is allowing stock market values to inflate in the short run. This will only be justified in the longer run if the return on capital of underlying businesses remains stable, justifying a higher price to book multiple. Economic reality dictates that profitability will eventually be bid down to match the artificially lowered cost of capital.
As return on capital is falling fastest in cyclical sectors which are being impacted more quickly by both falling demand and competition from cheap capital, it is likely that returns in these sectors may shift to lower levels faster than defensive peers. However, given returns in areas such as financial services are already at levels which are substantially higher than the real economy, the impact of any mean reversion in these returns, particularly when combined with financial leverage, will be severe.
Low cost resource businesses and cyclical industrials with sensible financial leverage should provide better long term income potential than is reflected in current dividend yields. This is due to the gradual repositioning of businesses to match lower long term demand growth, which should allow reinvestment levels to reduce and payout ratios to increase. The mathematics of growth being a function of the rate of return and the percentage of earnings retained will remain in place.
Chart 18: Free cash flow yield of major resources: 2008 – 2020(f)
Source: Macquarie Securities, April 2013
Analysis of alternative portfolios
Prior papers3 have discussed the very long term outcomes of a typical investment approach. The first step in analysing potential longer-term portfolio outcomes is to consider the upper and lower bounds of any reasonable investment strategy. For Australian equities (the best performing developed equity market in the long term) this has been circa 7% p.a. real, while for a more diversified balanced/growth portfolio long term real return outcomes have been in the order of 5% p.a.
Charts 19 and 20 show the rolling 10 year and longer term average returns to a selection of alternative investment strategies, namely 100% in Australian equities, a “balanced” strategy (30% global equities, 30% Australian equities, 30% bonds, 10% cash), a conservative strategy (12.5% global equities, 12.5% Australian equities, 45% bonds, 30% cash) and 100% cash.
Chart 19: Rolling 10 year real returns Return - All Observations
Chart 20: Average real returns % pa .(1900-2011)
Source: Schroders, Global Financial Data, ASX 200 and other data series as per appendix
As noted, longer term equity returns for Australian equities have been slightly in excess of 7% p.a. real, balanced just over 5%, conservative 3% and cash 0.6%. Chart 20 also shows the 30 year averages which, given the strong disinflationary focus of central banks globally, and the high starting point for bond yields and inflation, has resulted in substantially better real returns over the last 30 years.
From the perspective of an investor requiring some form of drawdown over time, longer term outcomes of between CPI +1[(-2) was not found] (being cash-like and in line with the RBA’s target) and CPI+5% (as a growth balanced strategy) are the likely lower and upper bounds of any long term investment strategy.
Recognising that a pure Australian equity portfolio is likely to be too volatile for any representative drawdown analysis, Charts 21 and 22 compare two alternative investment strategies from the perspective of an individual entering the drawdown phase, being the balanced and conservative balanced strategies (or approximately CPI+5% and CPI+3%). In particular, the distribution of outcomes for the length of the drawdown period, where drawdowns commence anytime 1930 through to 1990 is examined .
Chart 21: Distribution of drawdown outcomes 1930 - 1990
Chart 22: Distribution of drawdown outcomes 1930 -1990
Source: Global Financial Data, Schroders. Assumes initial drawdown of 8% of lump sum and indexed with CPI thereafter. Conservative balanced is 12.5%, Australian equities is 12.5%, Global equity is 12.5%, Australian bonds is 45%, Cash is 30%. CPI+ portfolio is a theoretical return series equal to CPI+ not an actual managed strategy.
There are three observations we would make from the above analysis:
- By reducing the volatility of returns relative to inflation, the outcomes are on average relatively similar when considering the entire sample period. However, the volatility of outcomes is massively reduced with the CPI+ options as the sequencing risk is better managed.
- The CPI+ “portfolios” deliver not only more consistent outcomes but on average better outcomes than the balanced or conservative balanced solutions.
- The incidence of severely negative outcomes in the drawdown phase is removed through reducing the sequencing risk.
Of course, managing to a CPI+ outcome requires a very different investment approach to the fixed asset allocation approach of a traditional balanced or conservative portfolio. In order to generate lower volatility inflation relative outcomes, particular attention has to be paid to valuation and likely risk-adjusted returns of the underlying assets in the construction of the portfolio.
It is clear from the above that an overt focus on yield when constructing portfolios is unlikely to be an appropriate strategy for maximising the total real money weighted outcomes for individuals. In particular, a strategy that emphasises lower volatility of total real returns will lead to better and more consistent outcomes across member cohorts.
Bearing in mind that the upper and lower bounds of any long term investment strategy (without leverage) are in the order of CPI+1% and CPI+5%, then it seems reasonable that a more appropriate group of strategies for those seeking consistent, inflation adjusted returns would be those that sit within this range and specifically target lower volatility, inflation relative outcomes.
It is our contention that valuation is the key factor in determining which asset classes offer better longer term risk adjusted real returns and in reducing downside volatility of the total portfolio. Yields offer some insight into valuation but are not the determinant of valuation and therefore should not be overemphasised in the portfolio construction process. This is particularly the case now, where the “price of yield” has increased substantially.
Bibliography of Prior Research Pieces on Objective Based Investment Strategies (and related topics) from Schroder Investment Management Australia Limited referred to in this paper. Copies available from Schroders:
2007, 2008, 2009, “CPI+5 White Paper”;
January 2009, “It’s about risk, not return”;
April 2009, “What price complexity”;
August 2009, “Keeping it simple, back to the future for Asset Allocation”
February 2011, “Complexity Adding Value”
August 2011, “Post Retirement – Time to Focus on the Endgame”
September 2011, “Life Cycle Funds – Just Marketing Spin”
March 2012, “Why SAA is Flawed”
April 2012, “Asset Allocation - How flexible do we need to be?”
May 2012, “Understanding the Journey to Retirement”
October 2012, “Risk Parity – No Free Lunch”
November 2012, “Avoiding the valuation traps in Strategic Asset Allocation”
February 2013, “Searching for the Holy Grail in Asset Allocation”
Data sources used in this paper
Data sourced from Global Financial Data:
Australia Consumer Price Index
Australia Total Return Bills Index
Australia 10-year Government Bond Return Index
GFD World Return Index
Australia S&P/ASX 200 Accumulation Index
S&P 500 Total Return Index (w/GFD extension)
1For Example: Merton H Miller and Franco Modigliani, 1961, “Dividend Policy, Growth, and the Valuation of Shares”, Journal of Business 34, 411-433. Fama, E. F., & French, K. R. (1988). Dividend yields and expected stock returns. Journal of Financial Economics, 22(1), 3-25. Fama, E. F., & French, K. R. (1989). Business Conditions and Expected Returns on Stocks and Bonds. Journal of Financial Economics, 22(1), 3-25. Goyal, A., & Welch, I. (2003). Predicting the Equity Premium with dividend Ratios. Management Science, 49, 639-654
2For a great illustration of this point, see “Fooling Some of the People All of the Time, A Long Short Story”, By David Einhorn.
3Prior papers published by Schroder Investment Management Australia Limited: 2007, 2008, 2009, “CPI+5 White Paper”; January 2009, “It’s about risk, not return”; April 2009, “What price complexity”; August 2009, “Keeping it simple, back to the future for Asset Allocation”; February 2011, “Complexity Adding Value”; August 2011, “Post Retirement – Time to Focus on the Endgame”; September 2011, “Life Cycle Funds – Just Marketing Spin”; March 2012, “Why SAA is Flawed”; April 2012, “Asset Allocation - How flexible do we need to be?”; May 2012, “Understanding the Journey to Retirement”; October 2012, “Risk Parity – No Free Lunch”; November 2012, “Avoiding the valuation traps in Strategic Asset Allocation”; February 2013, “Searching for the Holy Grail in Asset Allocation”
Opinions, estimates and projections in this article constitute the current judgement of the author as of the date of this article. They do not necessarily reflect the opinions of Schroder Investment Management Australia Limited, ABN 22 000 443 274, AFS Licence 226473 ("Schroders") or any member of the Schroders Group and are subject to change without notice. In preparing this document, we have relied upon and assumed, without independent verification, the accuracy and completeness of all information available from public sources or which was otherwise reviewed by us. Schroders does not give any warranty as to the accuracy, reliability or completeness of information which is contained in this article. Except insofar as liability under any statute cannot be excluded, Schroders and its directors, employees, consultants or any company in the Schroders Group do not accept any liability (whether arising in contract, in tort or negligence or otherwise) for any error or omission in this article or for any resulting loss or damage (whether direct, indirect, consequential or otherwise) suffered by the recipient of this article or any other person. This document does not contain, and should not be relied on as containing any investment, accounting, legal or tax advice. Schroders may record and monitor telephone calls for security, training and compliance purposes.