How the 60/40 portfolio outperforms
In Critics of the 60/40 portfolio have their blinkers on (27 September), David Thornton notes that “the age-old 60/40 portfolio,” comprised of 60% stocks and 40% bonds, “has copped a growing chorus of criticism lately.” On 15 October 2021, for example, Goldman Sachs asked: “is the 60/40 dead?” Barron’s was unequivocal. On 4 November, it titled its lead article “The 60/40 Portfolio Is Dead. Here’s How Advisors Are Replacing It” (see also “The Year That Broke the 60/40 Portfolio,” The Australian Financial Review, 30 June 2022).
It’s telling: these and other detractors apparently accept no obligation to substantiate their claims with evidence – or even, in one instance, with coherence. They content themselves with idle assertions; in sharp contrast, its strongest defenders rigorously analyse valid and reliable data. My analysis reconfirms others’ – and thereby demonstrates that recent criticisms are misconceived, exaggerated, incoherent, or simply incorrect. The 60/40 portfolio doesn’t preclude losses. It does, however, reduce the frequency and severity of short-term negative returns, and thereby generates more consistent long-term results, than an all-stock portfolio.
These crucial attributes, particularly during bear markets and financial crises and including the year to 30 September – its worst 12-month result since the Great Depression – have enabled the 60/40 portfolio to match and outperform all-stock portfolios over long-term periods. It certainly isn’t dead, and it’s not even broken; it’s been bruised, but its prospects remain robust.
Leithner & Company provides an example. Although its portfolio since 1999 has been more 50/50 than 60/40, its returns have equaled and for long stretches outperformed the All Ordinaries Index (visit our website for details). So ignore the naysayers: 60/40 has long been and today remains, as Thornton rightly concludes, “a compelling proposition” for sensible investors.
The 60/40 portfolio
Why should most investors, and all sensible ones, allocate approximately 60% of their portfolios to stocks and 40% to bonds? Contemporary mainstream finance theory provides one justification. Investopedia summarises a very large and complex field: “In theory, a 60/40 mix allows you to maintain balance in your portfolio … It’s designed to minimise risk while generating a consistent rate of return over time, even during periods of volatility.” As we’ll see, this rationale should be amended to read “… especially during periods of downward volatility.”
There’s nothing sacrosanct about these weightings. As Benjamin Graham wrote in The Intelligent Investor:
We have suggested as a fundamental guiding rule that the investor should never have less than 25% or more than 75% of his funds in common stocks, with a consequent inverse range of between 75% and 25% in bonds. There is an implication here that the standard division should be an equal one, or 50-50, between the two major investment mediums (see also Shun stock-pickers - choose investors).
Two key preliminary questions – and an implication
At first glance, it might seem that nothing in investing could be easier than a traditional, “balanced” portfolio that comprises 60% stocks and 40% bonds. But upon reflection two difficult sets of questions appear: which stocks – domestic or foreign, “growth” or “value,” large or small caps, etc. – should the portfolio hold? And which bonds – domestic or foreign, corporate or sovereign, investment-grade or junk, short or long duration? The answer to these questions depends upon a variety of factors, not the least of which is the investor’s willingness to risk long-term mediocre returns or even losses.
The more conservative (and, dare I say it, sensible) is the investor, the more her portfolio’s allocation to equities should be weighted towards domestic, large-cap and “value” stocks, and the more its bond component towards shorter-duration Australian Commonwealth bonds (or, for DIYers and SMSFs, term deposits).
The more aggressive is the investor, the more his portfolio might be weighted towards foreign, small-cap, and “growth” stocks, and corporate, longer-duration, foreign and sub-investment grade (“junk”) bonds – with the caveat that the totals returns of “aggressive” bonds and equities tend to be positively correlated. But the aggressive investor (“speculator” is more apt) should beware: in the investment marathon the tortoise typically beats the hare (see, for example, The myth of the small cap premium and Do tech stocks really outperform their value counterparts?). Unless they’re lucky, which in aggregate they aren’t, aggressive investors typically lose, and sometimes they lose grievously; the 60/40 portfolio will likely mitigate their losses, but it probably won’t prevent them. The problem in these cases isn't the bottle (60/40 portfolio); it's the wine (overconfident asset selection).
The key implication is that a 60/40 portfolio, or something that resembles it, suits most investors – from the most conservative to the more aggressive.
The crucial underpinning
Whatever it's weightings of bonds and stocks, how can a portfolio that’s weighted primarily towards stocks but nonetheless contains a substantial percentage of bonds reduce the frequency and severity of low or negative returns, and generally generate consistent returns? How, under certain conditions, can it outpace an all-stock portfolio?
The key assumption is that bonds’ and stocks’ total (capital gain plus dividend or distribution) returns are, generally speaking, uncorrelated.
This means that the return of the portfolio’s bonds doesn’t affect the return of its stocks and vice versa; it also means that there’s no third factor that jointly influences the returns of both stocks and bonds. If, for example, during the most recent 12-month period the equities portion of a 60/40 portfolio has generated an above-average return, then it’s just as likely that the bonds’ return has been above-average as below-average.
In this central respect, critics of the 60/40 portfolio often misconstrue its rationale. A recent critic writes: “The last nine months or so have tested to destruction the reassuring idea that when one of these two asset classes falls the other tends to rise.” This detractor asserts that the viability of the 60/40 portfolio presupposes that bonds and stocks be inversely correlated. It’s great if they are, but it’s not necessary: the 60/40 portfolio posits merely that these two classes of assets be uncorrelated.
To test its key underpinning, I’ve made two middle-of-the-road (that is, neither overly conservative nor aggressive) and greatly simplifying assumptions: first, the equities portion perfectly mimics the S&P 500 Index; second, its bonds portion consists exclusively of U.S. 10-year Treasury bonds.
These assumptions enable me to analyse data compiled by Robert Shiller (see Irrational Exuberance (Princeton University Press, 1st ed., 2001). For each month from January 1871 to September 2022, I’ve calculated bonds’ and stocks’ total (that is, capital growth + dividends or payments of interest) 12-month return; and for each rolling 60-month period, I’ve correlated the bonds’ and stocks’ total returns. Figure 1 plots the results.
Figure 1: Correlation of Bonds’ and Stocks’ 12-Month Returns, Monthly Data and Rolling Five-Year Intervals, January 1871-September 2022
These data corroborate the key assumption which underlies the 60/40 portfolio: over almost 150 years, the average correlation over rolling five year periods between the S&P 500’s and the 10-year Treasury’s 12-month total return has been 0.02.
If bonds’ and stocks’ returns were perfectly positively correlated, such that one asset’s positive return perfectly predicted the other’s, then their correlation would be 1.0. If the returns were perfectly inversely (that is, negatively) correlated, then one asset’s positive return would perfectly predict the other’s negative return, and their correlation would be -1.0. The average correlation is, for all practical purposes, 0.0; hence these two series’ returns are uncorrelated.
It’s vital, however, not just to acknowledge but to emphasise that history and present reality are complex and messy. Although the average correlation of 12-month returns over rolling five-year periods since the 1870s has been practically zero, during specific intervals it has seldom been zero: often it’s been positive, and occasionally it’s been strongly positive.
This result demonstrates that recent critics of the 60/40 portfolio err in additional ways. One doubter writes: “first on my list (of truisms) is the foundational belief that dividing your portfolio between shares and bonds will always (italics added) smooth your investment journey.”
Figure 1 demonstrates that, although on average they are uncorrelated, the correlation of stocks’ and bonds’ total returns has been erratic over time; by implication, the 60/40 portfolio won’t – indeed, nothing will – invariably protect your downside.
Yet detractors can be forgiven for assuming that the total returns of bonds and stocks are normally inversely correlated: as Figure 1 shows, this has been the case for the past 20 years. Indeed, never before 2000 had the correlation been so consistently negative for so long.
Without exception since April 2001, the correlation between 10-year Treasuries’ and the S&P 500 Index has been negative – and often strongly negative. That’s been a boon for the 60/40 portfolio: when stocks crashed during the Dot Com Bust and the Global Financial Crisis, bonds’ returns rose; and when bonds’ returns swooned (2013-2014), stocks soared.
Some truisms – and strengths of the 60/40 portfolio
One recent criticism of the 60/40 portfolio is simply incoherent. It’s bizarre that it should be entitled Now is the time to myth-bust investment truisms (3 October). A truism, says one dictionary, is “an undoubted or self-evident truth;” according to another, it’s “a claim that is so obvious or self-evident as to be hardly worth mentioning, except as a reminder or as a rhetorical or literary device.” It’s hardly worth mentioning – and is thus a truism! – that an undoubted or self-evident truth is the polar opposite of a myth. In this context, it’s essential to restate two truisms about the 60/40 portfolio:
- because bonds generate negative total returns much less frequently than stocks, the 60/40 portfolio reduces the risk (compared to a 100%-stock portfolio) of a negative return;
- because bonds’ returns fluctuate less than stocks’ returns, the 60/40 portfolio will generate more consistent returns than the stock-only portfolio.
Figure 2 reconfirms these self-evident truths. During 12-month periods since January 1874, the S&P 500 has generated an average total 12-month return of 11.0%. The 10-year U.S. Treasury bond has returned 5.2% and the 60-40 portfolio 8.7%. Stocks don’t merely generate the highest and bonds the lowest average return: stocks’ returns also fluctuate most and bonds’ returns vary the least. The standard deviation of the 10-Year Treasury’s 12-month returns is 5.9%, the 60-40 portfolio's is 11.7% and the S&P 500’s is 18.9%.
Figure 2: Four Truisms About All-Stock, All-Bond, and 60-40 Portfolios' Returns, Rolling 12-month Periods, January 1872-September 2022
Another truism: bonds generate negative returns comparatively infrequently (that is, during 10.0% of the 12-month intervals since January 1872), the 60-40 portfolio more often (21.4%) and the S&P 500 most regularly (26.5%). Moreover, when bonds’ return is negative, their average 12-month loss is relatively modest (average of -3.4%) whereas the S&P 500’s loss is much more consequential (11.6%). In this respect, too, the 60/40 portfolio occupies a middle position: during 21.4% of the 12-month periods, it has generated a negative return; and when it has, its loss has averaged 6.5%.
These truisms refute the 60/40 portfolio’s critics
Table 1 expresses as compound annual growth rates (CAGRs) the returns of all-stock (S&P 500 Index), all-bond (10-year Treasury) and 60/40 portfolios over six intervals to September 2022. Significantly, these intervals include data for the year to 30 September – one of the 60/40 portfolio’s worst results on record (details appear in the next section). It demonstrates that:
- Throughout various intervals over almost a century, the 60/40 portfolio has generated reasonable returns.
- Over periods of five and 10 years, a stock-only portfolio (S&P 500 Index) has outperformed the 60/40 portfolio.
- Over periods of 15 years or more, however, the 60/40 portfolio has effectively matched or outperformed the all-stock portfolio.
Table 1: Total Returns (Compound Annual Growth Rates), Three Portfolios over Six Intervals
Why has the 60/40 portfolio underperformed during the past five and 10 years? The answer is two-fold: first, the S&P has generated a much higher CAGR during these intervals than it has since 1926; second, 10-year Treasuries have generated very poor (compared to their 96-year CAGR) results.
Why has the S&P 500 recently outperformed compared to its long-term CAGRs? A comprehensive answer would be extended and complex. For my purposes, a very simple and concise one will suffice: until this year, the ten years since 2012 contained just one bear market – which lasted barely one month and was thus by far the briefest on record. This interval also included one recession (according to the National Bureau of Economic Research, the arbiter of such things, it also ranked among the shortest in history). But no long and severe bear market, recession, bursting bubble, or economic crisis has marred the last decade.
Investors during previous intervals weren’t nearly so fortunate – and it’s reasonable to wonder whether investors in the next decade they’ll be so lucky. During severe bear markets, bursting bubbles, financial crises, and long and deep recessions (and one depression), stocks were mostly crushed and bonds largely shone – in relative and sometimes in absolute terms – and the 60/40 portfolio thereby outpaced the all-stock portfolio.
The 15 years to September 2022, for example, contained the GFC, and during the GFC stocks were crushed and bonds greatly outperformed; as a result, during this period the 60/40 portfolio’s CAGR matches the S&P 500 Index’s. Similarly, the 20 years to September encompassed not just the GFC but also the Dot Com Bust – during which the S&P floundered and bonds again outperformed. And the 96 years from September 1926 included the Great Depression, the long recession and bear market from the mid-1970s to the early-1980s, the Dot Com Bust, and the GFC. During these events, stocks were mostly pole-axed and bonds largely outperformed.
The Most Recent 12-Month Result Is Highly Atypical – and Likely Won’t Persist
According to The Australian Financial Review (“The Year That Broke the 60/40 Portfolio,” 30 June),
For nearly 40 years, the traditional 60/40 portfolio of stocks and bonds has weathered all market conditions to deliver reliable returns to investors – rewarding those that kept it simple. But not this year. The 60/40 has had a shocker as stocks are down but bonds – which are meant to provide a buffer in down markets – … are down even more. “The big challenge of the last 12 months is it’s been the first time bonds and equities have gone down at the same time – there’s no place to hide,” Morningstar (said).
Morningstar isn’t quite right. Figure 1 indicated that until the past 20 or so years bonds’ and stocks’ total returns have often been positively correlated. And Table 1 demonstrated that the 60/40 portfolio has always – including the five years to 30 September –delivered reasonable and reliable returns. But it’s certainly true that the 12 months to 30 September have been very bad.
Figure 3 quantifies how bad. The 60/40 portfolio’s total return during the 12 months to 30 September (-18%) is worse than the -16% in 1974; only the Great Depression (-17% in 1930 and -25% in 1931) produced even lower results.
Figure 3: Total Return, 60/40 Portfolio, 12-month Periods to September 1926-2022