Their earnings have sunk since the GFC, yet even weaker bases underpin the S&P 500’s returns. Aussie stocks thus offer better prospects.
Overview
Over the past decade and more, the total (incorporating the reinvestment of dividends) returns of the All Ordinaries and S&P/ASX 200 indexes have underperformed the S&P 500 Index. According to Roger Montgomery, dividends – specifically, Australian companies’ allegedly overly-generous payments, retention of insufficient earnings and restriction of capital expenditure, which, he asserts, cause shareholders’ returns to fall – are to blame (see Dividends do not make a good company – Part 1, 6 November 2023 and “Why the ASX 200 has gone nowhere in 16 years,” Firstlinks, 8 November 2023).
In Dividends Aren’t a Bane – They’re a Boon (20 November 2023), I comprehensively disproved Montgomery’s assertions. Neither in the U.S. nor in Australia are dividends a bane of investment life. Quite the contrary: they’re a boon and bulwark of long-term returns.
Why, then, have Australian shares underperformed? Shane Oliver (Australian shares at new record highs – is it sustainable? 17 July) nominates a range of factors. “The near 15 year period of underperformance since 2009 reflects:
- “payback for the huge outperformance of the 2000s on the back of the mining boom;
- the slump in commodity prices from 2011;
- the lagged impact of the surge in the $A into 2011;
- relatively tighter monetary policy in Australia for much of the post-GFC period;
- fear more recently that rate hikes will hit Australia harder due to more indebted households and Australia’s relatively expensive property market;
- worries about tensions with China and the slowing Chinese economy;
- and a low exposure to tech stocks – with tech stocks propelling U.S. shares in the pandemic and more recently with AI excitement.”
Most of these points – particularly the last one – seem plausible. Yet Oliver, like Montgomery, asserts rather than analyses: he doesn’t attempt to demonstrate logically that commodity prices, exchange rates, monetary policy, etc., affect Australian equities’ returns; still less does he show empirically that over the past 10-15 years they’ve crimped returns. Accordingly, neither individually nor jointly are his reasons: none, as I’ll demonstrate, digs to the root of Australian equities’ recent underperformance.
That’s partly because they’re Oz-centric; they thereby downplay or ignore key developments in the U.S. Yet posing the question “why have Australian equities underperformed American ones?” is necessarily to ask: “why have American shares outperformed Australian ones?”
Oliver continues: “some of these factors have waned … But some still remain (and they) could still push U.S. and hence global shares up more than Australian shares. So … it is probably too early to be confident that the underperformance is over even though Australian shares offer better medium-term prospects.”
As I’ll also show, in this respect Oliver’s right. Indeed, the All Ordinaries Index’s long-term as well as its medium-term prospects exceed the S&P 500’s. But my explanation differs greatly from Oliver’s.
In this article, I calculate and compare the total returns of the All Ordinaries and Standard & Poor’s 500 indexes over various (including all five- and ten-year) periods over the past 50 years. I uncover four key results:
- The total return of one index relative to the other is neither random nor permanent; instead, it’s cyclical.
- The duration the All Ords’ current relative underperformance is the longest of the past 50 years; but compared to previous stretches its magnitude is mild.
- Following the method devised by John Bogle, I partition these indexes’ total returns into three components – and thereby identify the source of the All Ords’ underperformance and the S&P 500’s outperformance: earnings growth (U.S.) and the lack thereof (Australia).
- I also identify and quantify three weak bases of the S&P 500’s outperformance since the GFC: the rising reliance of earnings growth upon share buybacks, and consequently high debt to equity and cyclically-adjusted P/E (“CAPE”) ratios.
In short, since the GFC the All Ords’ total returns have lagged the S&P 500’s mostly because Australian equities’ CPI-adjusted earnings have fallen – and American stocks’ earnings have soared.
Australian equities’ earnings per share (EPS) have lagged partly because debt-financed share buybacks don’t (as they do in the U.S.) underpin them; accordingly, Australian debt-to-equity and CAPE ratios are comparatively low, and over the past 20 years the debt-to-equity ratio has fallen.
Conversely, American stocks’ EPS has zoomed, certainly to a significant degree and perhaps to a considerable extent, because S&P 500 companies have repurchased colossal quantities of their own shares. They’ve also used large amounts of debt to finance the buybacks; as a result, their debt-to-equity and CAPE ratios aren’t merely relatively high: since 2005 they’ve trended upwards.
From these results I draw two fundamental conclusions. Firstly, American equities have generated appreciable rewards since the GFC – but at current prices they also entail considerable risks.
Secondly, Australian equities, not despite but because of their underperformance over the past 10-15 years, aren’t merely more robustly financed; they’re also more conservatively priced and thus offer superior medium- and long-term prospects (see also Why the S&P 500’s five-year prospects are poor, 27 December 2023).
Data
I’ve analysed the data originally compiled by Robert Shiller for his book Irrational Exuberance (Princeton University Press, 1st ed., 2001) and updated regularly since then. Over the periods which I’ve considered (the 50 years since 1974), these include monthly closing levels of the S&P 500 index, Standard & Poor’s monthly calculations of the index’s earnings and dividends, and the U.S. Bureau of Labor Statistics’ quarterly estimates of the Consumer Price Index (which Shiller has extrapolated into monthly data).
I’ve also analysed monthly closes, earnings and dividends of the All Ordinaries Index (obtained from S&P since 2005 and the ASX from 1974 to 2005), as well as the Australian Bureau of Statistics’ quarterly estimates of the Consumer Price Index (which I’ve extrapolated into monthly observations). With these data I’ve replicated Shiller’s method of calculating total CPI-adjusted returns.
Four Intervals over the Past 50 Years
Figure 1 plots the growth on a monthly basis since January 1974, including and assuming the reinvestment of dividends and net of the Consumer Price Index, of investments of $1 in portfolios that mimicked the All Ordinaries and S&P 500 Indexes.
These results ignore taxes, rebalancing-transaction and other costs (which would be considerable); they also exclude the value (which would be appreciable) of Australian franking credits since the commencement of the dividend imputation system 1987.
The portfolio that replicated the All Ordinaries Index grew to $19.80 in June 2024; that’s a compound annual growth rate (CAGR) of 6.1%. Each $1 invested under the same assumptions in a portfolio that’s replicated the S&P 500 grew to $33.73 in June of this year; that’s a CAGR of 7.2%. Over the past half-century, particularly during the Dot Com Bubble and since the GFC, the S&P’s CPI-adjusted total return has outperformed the All Ords’.
Figure 1: CPI-Adjusted Accumulation of Capital per $1 Invested, All Ordinaries and S&P 500, Monthly, January 1974-June 2024
Figure 2 plots these series since January 2000. Each $1 invested during that month in a portfolio that’s mimicked the All Ordinaries Index grew to $3.48 in June 2024; that’s a CAGR) of 5.2%. Each $1 invested in a portfolio that’s replicated the S&P 500 grew to $3.19 in June 2024; that’s a CAGR of 4.9%.
Since the turn of the century, the All Ords’ CPI-adjusted total return has slightly but continuously outperformed the S&P 500’s.
Figure 2: CPI-Adjusted Accumulation of Capital per $1 Invested, All Ordinaries and S&P 500, Monthly, January 2000-June 2024
From the mid-1970s until the late-1990s, the two investments compounded virtually in tandem. During the Dot Com Bubble of the late-1990s, the S&P 500 greatly outperformed the All Ords. But then the bubble burst and Australia’s mining boom occurred; as a result, for almost a decade (a period which included the Global Financial Crisis) the All Ords outpaced the S&P 500.
Since 2011, however, the American index’s total return has exceeded its Australian counterpart’s. Since then, and particularly since 2020, Australian equities have significantly underperformed their very long term (since 1974 and 2000) averages (Figure 3); if their returns subsequently revert to these long-term means (which, I’ll show below, is reasonable to expect), then their CAGRs will rise.
In contrast, since the GFC American stocks have greatly outperformed their long-term averages; unless their returns permanently exceed 9% per year, which I strongly doubt, during the years to come they’ll recede towards their long-term means – which since 2020 has seemingly begun to occur.
Figure 3: CPI-Adjusted Total Returns (CAGRs), All Ordinaries and S&P 500, Four Intervals to June 2024
All Five- and Ten Year Intervals over the Past 50 Years
How do Australian and American equities’ rolling five-year and ten-year CAGRs compare? From January 1974 to January 1979, the All Ordinaries Index’s CPI-adjusted CAGR was -2.54% and the S&P 500’s was -2.47%. The Australian index thus underperformed its American counterpart by -2.54% – (-2.47%) = -0.07%. Percentages below 0% thus denote underperformance and percentages above 0% indicate outperformance. I repeated this exercise for all subsequent five-year CAGRs (February 1974-February 1979, … June 2019-June 2024) as well as all ten-year CAGRs; Figure 4 plots the results.
Figure 4: Performance of the All Ordinaries Relative to the S&P 500 (CPI-Adjusted Total Returns (CAGRs)), Monthly, January 1984-June 2024
Four results from the five-year series are most noteworthy:
- The performance of one index relative to the other is neither random nor permanent; instead, it’s cyclical.
- The All Ords outperformed from May 1979 to August 1984, March 1987 to May 1989 and January 2002 to March 2011.
- The S&P 500 outperformed from September 1984 to February 1987, December 1988 to December 1993, April 1996 to August 2001 and since November 2011.
- The duration All Ords’ current underperformance vis-à-vis the S&P 500, which exceeds 10 years, is the longest of the past 50 years; but compared to previous stretches (which averaged -6.5%, exceeded -10% and approached -15%) its magnitude (average of -4.6% and maximum of -9.4%) is comparatively mild.
The Source of the All Ords’ Cyclical Underperformance
John Bogle, in his book The Little Book of Common Sense Investing (John Wiley & Sons, revised and updated 10th anniversary ed., 2017), emphasises “the remarkable, if essential, linkage between the cumulative long-term returns earned by U.S. business – the annual dividend yield plus the annual rate of earnings growth – and the cumulative returns earned by the stock market …” (p. 10).
He divides stocks’ returns into three parts: “(1) investment return (enterprise), consisting of (a) the initial dividend yield on stocks plus (b) their subsequent earnings growth (together, they form the essence of what we call ‘intrinsic value’) and (2) speculative return, the impact of changing price/earnings multiples on stock prices” (p. 15).
“Look,” he says, “at the record since the beginning of the 20th century. The average annual total return on stocks was 9.5%. The investment return alone was 9.0% – 4.4% from dividend yield and 4.6% from earnings growth.” The difference between the total return and the investment return “of 0.5% per year arose from … speculative return. (It) may be a plus or a minus, depending on the willingness of investors to pay either higher or lower prices for each $1 of earnings at the end of a given period than at the beginning” (pp. 10-11).
What underpins the speculative return? “The price/earnings (P/E) ratio measures the number of dollars investors are willing to pay for each $1 of earnings. As investor confidence waxes and wanes, P/E multiples rise and fall.” Bogle acknowledges that changes of interest rates affect the market’s P/E.
Whether it’s rates or some other reason, “when greed holds sway, very high P/Es (and positive speculative returns) are likely … When fear is in the saddle, P/Es are typically very low (and speculative returns become negative)” (p. 11).
Speculative returns – positive and negative – punctuate some periods more than others: “to be sure, stock market returns sometimes get well ahead of business fundamentals … But it has only been a matter of time until, as if drawn by a magnet, they ultimately return to the long-term norm, although often only after falling well behind for a time, as in the mid-1940s, the late-1970s and the 2003 market lows. It’s called reversion (or regression) to the mean” (p. 13).
“In our foolish focus on the short-term stock market distractions of the moment,” warns Bogle, “investors often overlook this long history. When the returns on stocks depart materially from the long-term norm, we ignore the reality that it is rarely because of the economics of investing – the earnings growth and dividend yields of our corporations. Rather, the reason that annual stock returns are so volatile is largely because of the emotions of investing, reflected in those changing P/E ratios” (pp. 13-14).
He doesn’t stop there: “while the prices we pay for stocks often lose touch with the reality of corporate (profits and dividends), in the long run reality rules. So, while investors seem to intuitively accept that the past is inevitably prologue to the future, any past stock market returns that have included a high speculative stock return component are deeply flawed guides to what lies ahead” (p. 14; italics in the original).
In Index funds’ key flaws – and how we overcome them (23 October 2023) I replicated Bogle’s approach but improved his method: rather than disaggregate one-year results into their three components, I partitioned the S&P 500’s five-year CAGRs into their three components. In that article, I analysed the S&P 500’s return over various intervals from 1871 to 2023; here, I compare the All Ordinaries and S&P 500’s returns over key intervals from 1974 to 2024. Figure 5 summarises the American results.
Figure 5: Components of the S&P 500’s Average, Five-Year, CPI-Adjusted Total CAGR, Four Intervals since 1974
As time has passed, the average five-year total CAGR has varied; so too have its components. Most significantly, during each period earnings are by far the biggest contributor; dividends contribute a much smaller portion, and the CAGR’s speculative component – changes of the P/E ratio – is the most variable.
Moreover, since 2000 the increase of the speculative component has contributed an ever greater portion of the total return.
Figure 6 summarises the Australian results. As in the U.S., so in Oz: over time the All Ords’ average five-year total CAGR has varied; so have its components. In particular, the speculative component has contributed an ever-greater share of the total return.
Figure 6: Components of the All Ordinaries’ Average, Five-Year, CPI-Adjusted Total CAGR, Four Intervals since 1974
Yet the compositions and changes in the two countries differ fundamentally. In the U.S., dividends are a constant but minor contributor to total returns; in Australia, they’ve been a major and the most consistent contributor.
Why, particularly since the GFC, has the All Ords underperformed the S&P 500? One key reason is that earnings haven’t contributed to its total return. Quite the contrary, they’ve crimped it – either marginally (since 2009) or significantly (by three percentage points since 2020).
Causes of the All Ords’ Cyclical Underperformance
Over the past half-century, the All Ordinaries Index’s total return has underperformed the S&P 500 Index’s. Since the turn of the century, however, the All Ords has marginally outperformed the S&P 500. Over shorter intervals, the Ords’ relative total return has been cyclical: at some points since 1975, it’s markedly underperformed, and at others it’s significantly underperformed. In particular, its continuous but modest underperformance since 2011 is the longest since 1975.
Why since the GFC have Australian equities underperformed? Unlike in the U.S., where earnings have continuously been the biggest contributor to the S&P 500’s total return, in Australia since the GFC earnings have crimped the All Ords’ total return. Figure 7, which plots each index’s CPI-adjusted earnings, substantiates this sharp discrepancy. In order to render the two series comparable, I’ve rescaled each with an identical origin ($1.00). Net of CPI, the S&P’s earnings grew to $3.37 in June 2024; that’s a CAGR of 5.2%. In sharp contrast, the All Ords’ shrank to $0.84; that’s a CAGR of -0.5%.
Figure 7: CPI-Adjusted Earnings, All Ordinaries and S&P 500 Indexes (January 1974 = $1.00), January 1974-June 2024
During the 20 years to the mid-1990s, the Australian index’s earnings mostly kept pace with their American counterparts. During the Dot Com Bubble, however, the S&P’s earnings vaulted ahead of the All Ords’ – and when the bubble burst they crashed to a level equivalent to the Ords’. From then until the GFC, earnings in both countries rose dramatically – in Australia, as a result of the mining boom and to an all-time high in January 2008.
Then came the GFC. The S&P’s earnings collapsed more than 90% – from $2.30 in May 2007 to just $0.18 in March 2009. The decrease of the All Ords’ earnings was much milder but lasted much longer: 48% from $2.07 in January 2008 to $1.07 in July 2013. Yet since the GFC the S&P 500’s earnings have zoomed: they rose to $3.07 on the eve of the COVID-19 crisis (December 2019), fell one-third during the crisis (to $2.05 in December 2020), but rose strongly thereafter. They reached an all-time high of $4.02 in December 2021; by June 2024, they’d receded to $3.47.
In sharp contrast, the All Ordinaries’ earnings flat-lined between 2013 to 2020, collapsed 78% during the COVID-19 crisis (from $1.36 in August 2019 to $0.30 in April 2021) recovered all of its loss and rose to $2.15 in June 2022, and plunged more than 60% over the next four years (to $0.84 in June 2024).
Why have the S&P’s CPI-adjusted earnings risen so strongly? Why have the All Ords’ stagnated and fallen? Figure 8, which plots data collected by Standard & Poor’s, provides a major reason: buybacks of shares by the companies comprising the S&P 500 Index have skyrocketed.
Figure 8: Buybacks, Billions of $US, S&P 500 Companies, Annualised, January 1990-March 2024
Buybacks increased from a CPI-adjusted rate of ca. $200 billion per year from 1990 to 2005 to more than $800 billion per year on the eve of the GFC. The Crisis crushed the pace of buybacks to less than $200 billion in 2010, but its effect was temporary: by the eve of the COVID-19 panic (i.e., during the four quarters to January 2019), buybacks exceeded $1 trillion per year. In the year to April 2021, they crashed below $600 billion – and in the year to January 2022 again exceeded $1 trillion. Most recently (12 months to January 2024), S&P 500 companies have repurchased $825 billion of their stock.
Buybacks, it’s crucial to understand, artificially inflate earnings per share; they thereby exaggerate the growth of earnings.
Consider a simple example: if during a given year a company earns $100 of Net Profit After Tax (NPAT) and has 1,000 shares on issue, its earnings per share (EPS) is therefore $100 ÷ 1,000 = $0.10 per share. If during the next year it repurchases 100 of its shares and its NPAT increases 5% to $105, its EPS becomes $105 ÷ 900 = $0.116. And if in a third year it again repurchases 100 shares and its NPAT rises 5% to $110.25, its EPS becomes $110.25 ÷ 800 = $0.138. During these three years its NPAT has compounded 5% per year, but its EPS has compounded more than twice as rapidly – by 11.3% per year.
In “Siegel vs. Shiller: Is the Stock Market Overvalued?” (Knowledge at Wharton, 18 September 2018), Jeremy Siegel attributed an astounding 71% of the S&P 500’s long-term, CPI-adjusted earnings growth – total growth of 3.5%, of which buybacks contributed 2.5% – to buybacks. Other studies assign to repurchases of shares a much lower but still significant effect: they underpin ca. 30-40% of the S&P 500’s long-term EPS growth.
Figure 9: Share Buybacks, S&P 500 Companies, Annualised and Cumulative as Percentages of U.S. GDP, January 1990-March 2024
Since 1990, the companies comprising the S&P 500 have repurchased a CPI-adjusted average of $457 billion per year of their shares, and an astounding cumulative total of $17.7 trillion.
That’s obviously a colossal amount, and Figure 9 puts it into context. On average since 1990, these companies’ annualised repurchases of their shares have been the equivalent of 2-4% of America’s Gross Domestic Product (GDP). Cumulatively, their buybacks are equivalent to almost 45% of America’s current GDP.
How has this occurred? On average, S&P companies have financed these repurchases partly with cash but largely with debt.
Figure 10, which plots the average debt-to-equity ratio of all companies (not just those who’re components of the All Ords or the S&P 500) in Australia and the U.S., corroborates this assumption. (The International Monetary Fund originally collected these data, and the St Louis branch of the U.S. Federal Reserve has collated them.)
Figure 10: Debt-to-equity Ratios, Australia and the U.S., Quarterly, September 2005-January 2024
In recent years, the companies comprising the All Ordinaries Index have repurchased a comparatively trivial quantity (ca. $A10-15 billion worth) of their shares per year. That’s equivalent to less than 0.5% of a given year’s GDP. For this and other reasons, Australian companies’ average leverage is relatively low; specifically, their debt-to-equity ratio is presently 44% and over the past 20 years has trended downwards.
In the U.S., in contrast, leverage is very high: over the past 20 years it’s trended upwards, on the eve of the COVID-19 crisis it exceeded 100% and presently exceeds 80% (see also Sirio Aramonte, “Mind the buybacks, beware of the leverage,” Bank of International Settlements, Quarterly Review, September 2020).
Consequences of the All Ords’ Underperformance and the S&P 500’s Outperformance
The cyclically-adjusted P/E ratio (“CAPE”) devised by John Campbell of Harvard University and Robert Shiller of Yale, is the pre-eminent measure of equity valuation. In “The Many Colours of CAPE” (Yale ICF Working Paper No. 2018-22), Farouk Jivraj and Shiller “investigate the efficacy and validity of CAPE from several different perspectives. First, we … find that CAPE consistently displays economic and statistical significance far better than any of its peers. Second, we explore alternative constructions of CAPE … (and) find that original CAPE is still best when comprehensively and fairly reviewing the other proxies …”
Figure 11: CAPE Ratios, All Ordinaries and S&P 500 Indexes, Monthly, January 1974-June 2024
Figure 11 plots the CAPE ratios of the All Ordinaries and S&P 500 indexes since January 1974. The course of the S&P 500’s CAPE is well-known. It averaged 10.2 from 1974 to 1979, sank to a generational low during the recession of the early-1980s (6.6 in July 1982), for the next decade rose steadily – and reached its all-time high (44.2 in December 1999) during the Dot Com Bubble shortly before and after the turn of the century. Then the bubble burst and CAPE halved to 21.2 in February 2003.
Before the GFC, CAPE fluctuated within a narrow range (reaching a pre-GFC maximum of 27.6 in May 2007) and crashed during the crisis (halving to 13.3 in March 2009). Since then it’s mostly risen: before the COVID-19 panic it rose as high as 33.3 (January 2019), sagged as low as 24.8 during the crisis (March 2020) and recovered all of this loss and more (38.0 in August 2021). In June of this year, it was 35.5 – among its highest levels of the past half-century.
Australian investors rarely compute and analyse the All Ords’ or ASX/S&P 200’s CAPE ratios (see, however, Everything the mainstream says about earnings in wrong, 13 March 2024 and How low could stocks go in 2023? 14 November 2022). Before the GFC, the All Ords’ CAPE broadly paralleled the S&P 500’s: it halved during the chaos of the Whitlam years (halving from 9.0 in January 1974 to its all-time low of 4.9 in December 1974), rose steadily to 12.9 in November 1980 and tumbled as low as 7.1 (July 1982) during the recession of the early-1980s.
The Ords’ CAPE then zoomed to 24.2 in September 1987 – the eve of the Crash of 1987 – and by February 1988 plummeted as low as 12.8. It meandered within a narrow range, falling as low as 11.1 in January 1991, before rising steadily (but not nearly as high as the S&P 500’s) to a maximum of 23.8 in March 2000. It suffered relatively little when the Dot Com Bubble burst, sagging as low as 16.6 in March 2003, and then rose as high as 27 in May 2007. During the GFC it halved to 11.2 in March 2009.
In sharp contrast, since the GFC the All Ords’ and S&P 500’s CAPEs have increasingly diverged. In particular, the S&P 500’s has risen considerably and the All Ords’ modestly; the American index’s valuation has thus risen relative to the Australian’s.
From August 2009 (15.4) to January 2019 (17.1) the All Ords’ ratio fluctuated without trend; it then rose to 19.2 on the eve of the COVID-19 panic (January 2020), fell as low as 13.8 during the crisis (March 2020) and rebounded as high as 22.2 in August 2021. Since then it’s mostly fallen – to as low as 17.1 in October 2023; in June 2024, it was 19.4. Unlike the S&P 500’s CAPE, whose current value (35.5) greatly exceeds its mean since 1975 (21.7), the All Ords’ current CAPE barely exceeds its mean since 1975 (16.2).
Why does this matter? The CAPE ratio is strongly mean-regressing: the higher it rises, the lower, on average, is the index’s subsequent medium-term (five-year) and long-term (ten-year) total return, and vice versa.
I sorted relevant data in Shiller’s dataset by CAPE. For each month since January 1974, I also calculated the S&P 500 Index’s total CPI-adjusted return (CAGR) during the previous five years and the next five years. I then divided the data into five equal (by number of observations) segments. Quintile #1 contains the 20% of observations with the lowest CAPEs, Quintile #2 the next 20%, … and Quintile #5 the 20% of observations with the highest CAPEs. Finally, I computed each quintile’s average CAGRs for both the previous and subsequent five years. I then repeated the exercise for 120-month (10-year) periods, and both exercises with the Australian dataset. Table 1 and Table 2 summarise the results.
Table 1: Total Returns (CPI-Adjusted CAGRs), S&P 500 Index, Stratified by CAPE Ratio, January 1974-June 2024
In both countries, the higher is the CAPE ratio during a given month, the higher, on average, has been the total return (CAGR) during the past five and ten years – and the lower is the average CAGR during the next five and ten years.
For the All Ordinaries Index, whose CAPE is currently 17.9 (which straddles the boundary between Quintiles #4-5, highlighted in red in Table 2), that’s reassuring: under current conditions, purchasers of Australian equities can expect CPI-adjusted CAGRs of ca. 7% per year during the next five years, and 4.8% per year over the next 10 years.
Table 2: Total Returns (CPI-Adjusted CAGRs), All Ordinaries Index, Stratified by CAPE Ratio, January 1974-June 2024
Why since the GFC has the S&P 500 outperformed the All Ords? Its CAPE has reached a level (presently 35.5, within Quintile #5, highlighted in red in Table 1) exceeded only by its pre-COVID-19 high and its all-time high during the Dot Com Bubble.
That’s disquieting: under current conditions, buyers of the S&P 500 can expect CPI-adjusted CAGRs of just 0.2 % per year during the next five years, and losses of 0.9% per year over the next 10 years.
These results must, of course, be interpreted cautiously. The entries in each cell of Table 1 and Table 2 are averages. Underlying them are ca. 120 observations, and these observations vary considerably. It’s possible, of course, that during the next five and ten years the S&P 500 will generate above-average results. Yet investing is about odds, and unless it’s different this time the odds don’t favour above-average results.
Over the past 50 years the medium- and long-term total returns of the All Ordinaries and S&P 500 indexes have regressed to their historical means. The consequences of the All Ords’ past underperformance of the S&P 500 thus include future outperformance. Similarly, the costs of the S&P 500’s past outperformance of the All Ords thus include future underperformance.
Implications
Since its formation in 1999, Leithner & Company has purchased reasonably priced and underpriced securities. These equities, almost all of which have been Australian, have comprised an average of one-half of it portfolio. Over the past decade, and particularly since the COVID-19 panic, we’ve been well aware of Australian stock indexes’ underperformance of the S&P 500 – but for various reasons, including those detailed in this article, we haven’t been tempted to shift a portion of our holdings to the U.S.
In Australian shares at new record highs – is it sustainable? (17 July), Shane Oliver wisely states that “it is probably too early to be confident that the underperformance (of Australian equities relative to the S&P 500) is over…” He correctly adds (but doesn’t attempt to demonstrate) that “Australian shares offer better medium-term prospects.”
Australian equities are hardly without risk: their CPI-adjusted earnings have long been stagnant, the “speculative” share of their total return (Figure 6) is high and their PE (as opposed to CAPE) ratio has recently reached an all-time record. Yet American equities are even riskier: colossal buybacks have magnified their leverage and greatly exaggerated their earnings, and their valuation (measured by the CAPE ratio) is sky-high.
I don’t know when – or even if – Australian indexes’ underperformance relative to their historical averages and the S&P 500 will end. In truth, nobody does. What I do know, however, is that the All Ordinaries Index’s total, CPI-adjusted returns vis-à-vis the S&P 500 have been cyclical – and that each index’s CAPE ratio has regressed strongly towards its long-term mean. On these bases (see also Figure 3), it’s reasonable to infer – not despite their recent underperformance but because of it – that if past is prologue then the Ords’ medium- and long-term prospects exceed those of the S&P 500.
I hope that this conclusion, and the data and analysis which underlie it, heartens owners (and their advisers) of reasonably-priced Australian equities. I also hope that it chastens those who, in effect, blindly urge Australians to purchase heavily-leveraged and unduly-expensive American stocks.
Figure 12 plots a portion of the data from Table 1 (p. 16) of Credit Suisse’s Global Investment Returns Yearbook 2023 Summary Edition. The Yearbook’s authors are Elroy Dimson (Chairman of the Centre for Endowment Asset Management at Cambridge Judge Business School, and Emeritus Professor of Finance at London Business School), Paul Marsh (Emeritus Professor of Finance at London Business School) and Mike Staunton (Director of the London Share Price Database, a research resource of London Business School).
Figure 12: Top-Five Developed Nations’ CPI-Adjusted Equity Returns (CAGRs), 1900-2022
Australian equities haven’t continuously outperformed. As my analysis of all five- and ten-year intervals since 1974 indicates, they’ve cyclically out- and underperformed – and since the GFC and particularly the COVID-19 panic, have underperformed. Why? The S&P 500’s earnings have outstripped the All Ords’ earnings. Why? Not least, it seems, because S&P 500 companies have repurchased colossal quantities of their shares – and borrowed heavily in order to do so. Consequently, they’ve become much more leveraged and (judging by their CAPE ratio) overpriced than Australian stocks.
Sensible reasons might induce an Australian investor to purchase some (or an index of) American stocks at current prices. However, and particularly if an advisor is counseling such an investor, a clear appreciation of risks should accompany the advice and the decision. “The dumbest reason in the world to buy a stock is because it’s going up,” Warren Buffett told The New York Times Magazine (1 April 1990).
Similarly, advocating American stocks, particularly American technology stocks, simply because their returns have recently exceeded Australian stocks is at best naïve, probably irresponsible and at worst very foolish.
Whether they’re Aussies or Yanks, what long-term return can investors reasonably expect? On 22 March 2019, The Sarasota Herald-Tribune quoted Buffett: “stocks are a decent way to make 6-7% annually.” The historical returns in Figure 3 and Figure 12 corroborate this expectation; so do the prospective five- and ten-year returns of the All Ordinaries Index at its current CAPE ratio (Table 2). Buffett therefore cautions: “anyone who expects to make 15% (long-term) … is living in a dream world” (also recall Figure 4).
“Now, maybe you’d like to argue a different case,” Buffett challenged bulls at the apex of the Dot Com Bubble (“Mr Buffett on the Stock Market,” Forbes, 22 November 1999).
“Fair enough. But give me your assumptions. If you think the American public is going to make 12% a year in stocks, I think you have to say, for example, ‘Well, that’s because I expect GDP to grow at 10% a year, dividends to add two percentage points to returns, and interest rates to stay at a constant level.’ Or you’ve got to rearrange these key variables in some other manner. The Tinker Bell approach – clap if you believe – just won’t cut it.”