Expected return of the balanced portfolio

The “60/40” – the portfolio made of 60% equities and 40% bonds – has been the successful paradigm for balanced portfolios for several decades now. It’s sound “diversification 101”: to combine the return-generating equities with the diversifying, hence risk-reducing, bonds.

The main reason for the 60/40 popularity, however, is that it worked, and for a long period of time: over the past 91 years, the 60/40 portfolio returned 8.1% annualised, not too far behind the S&P’s 9.5% annualised, yet with a volatility that was 40% lower than S&P’s.   What’s more, over the same long period, there were only four years when both stocks and bonds lost money: 1931, 1941, 1969 and 2018. Not bad indeed!

Yet, we have to raise a red flag: for all its successful track-record, the 60/40 portfolio is highly unlikely to work in the foreseeable future. Which is why, in the remainder of this short paper, we will describe the fragility of the 60/40 portfolio given the current economic and financial situation, and propose a possible solution.

The Problem

To clear possible misunderstandings, we are not questioning the principle behind the equities/bonds balanced portfolio: combining lowly-correlated assets classes, ideally even negatively-correlated, results in a diversified portfolio, no doubts about that.

So, having said where the problem of the 60/40 portfolio is not, where is then? In fact, what worries us is not the proportion between equities and bonds, be it 60/40 or similar numbers. Our concern is in the combination of the two asset classes themselves, hence it applies to any balanced portfolio made of only equities and bonds. The way we see it, the fragility of equities/bonds balanced portfolios comes from the present low level of yields, both in equities and bonds. Low levels of yields have, historically, been fairly reliable predictors of low returns over the following 7 to 10 years. Hence, simply put, the weakness of the 60/40 portfolio is that it invests in two asset classes that can be expected to deliver very low returns - not much better than zero returns, actually. Not a thrilling prospective.

But how do we reach the conclusion that long term expected returns are that low? Let’s use the U.S. as an example. 

Expected returns for U.S. 10-year bonds

As a proxy for the U.S. 10-year bond we can use an ETF, the IEF (7-10 year). The yield to maturity is 1.8%. This means that, as long as the U.S. government pays back what it has borrowed, at maturity of the bond portfolio underlying the IEF an investor will have earned a nominal return before inflation of 1.8% p.a. Not a penny more, not a penny less. Current expectations for average inflation in the U.S. during the next 7-10 years are 1.4%. Therefore, the expected annual real return for the bond component of the U.S. 60/40 is 1.8% nominal return less 1.4% inflation, or 0.4%.

Expected returns for U.S. equities

The nature of equities make the calculation of their long term returns more challenging than bonds’, but the concept is the same: for equities, too, we need to estimate the current yield. We start from the two things we know: 1) how much money the companies on average are making, hence how much money we are going to receive from them (that is, the future cash flows) and 2) the price we must pay today to receive those cash flows (that is, the current stock price). We can then use this information to calculate the market implied discount rate, that is the interest rate that, applied to today’s stock prices, will yield the future cash flows. Such interest rate is the equities yield, what we were looking for. Figure 1 below shows the historical relationship between the implied discount rate at the point of investment and the subsequent 10-year average real return. The relationship is not perfect, but it is pretty good: since January 1987, the implied discount rate, or the price paid for current cash flows, has explained 79% of subsequent real return.

Today, the market implied discount rate on U.S. stocks, calculated as described above, is a meager 3.7%. Based on the historical relationship illustrated in Figure 1 above, the 3.7% implied yield translates into an expected real return from U.S. equities of zero per cent.   And it is this miserable return expected over the long term that, as we mentioned above, is the problem for equities/bonds balanced portfolios, such as the 60/40.

But, wait: what about the admirable track record of the 60/40 portfolio, over more than nine decades? Shouldn’t such a strong and consistent performance refute, unequivocally, the criticism expressed so far in this paper? This is indeed a valuable objection, which we address by looking at the market environment in which said performance was achieved: that performance, we argue, was achieved in very favorable market environments, which are unlikely to repeat in the foreseeable future.

To see how past market conditions were instrumental in the 60/40’s performance, and how they differed from today’s, we’ll break the 91 years’ period in two subperiods: 1929-1980 and 1980-2017.  

For the 1929-1980 subperiod, bonds – by and large – performed finely their diversifying role: whenever equities fell from cyclical peaks central banks cut rates, allowing rising bond prices to offset, at least partially, the losses from equities. However, crucially, central banks could cut rates because rates were relatively high to start with. Compare that situation with the current one, where rates are at a rock-bottom level. True, in principle they could go even lower, but the balance of probability is unquestionably skewed in favour of higher rates, rather than lower; particularly as a likely shift to expansionary fiscal policy (from a possibly exhausted monetary policy) might well ignite inflation. Therefore, in the current market environment, falling equities would be unlikely to be cushioned by rising bonds: if anything, losses from bonds would likely compound losses from equities.

It should be noted that both bonds and equities, given their unusually low yields, are very sensitive to even small revisions of market expectations. If for example, expectations for the 10-year average cash rate shifted up by 1%, bonds’ prices would drop by about 7.5%. If lower risk appetite caused the market implied discount rate on equities to shift up by 1%, equities’ prices would be expected to fall by 20%.  

As for the 1980-2017 subperiod, market conditions have been even more favourable, in fact ideal: both equities and bonds enjoyed a strong tailwind of falling inflation expectations, which allowed central banks to slash the interest rate on cash, thus making both bonds and equities far more attractive than cash, allowing them to generate strong long-term returns. True, the secular tail wind was interspersed by cyclical shifts in expectations for cash returns, inflation and economic growth, as well as episodes of geopolitical uncertainty, which resulted in investors’ oscillating risk appetite between ‘risk on’ and ‘risk off’ phases. These, however, in hindsight turned out to be just blips on a clear secular trend of falling inflation and interest rates. Therefore, we can see as current lowermost rates environment make the 1980-2017’s scenario essentially unrepeatable.   Hence, again, losses for the equities/bonds balanced portfolio.

The Possible Solution

The main conclusion from what said above is that the foreseeable future will likely punish the buy-and-hold 60/40 portfolio, rather than handsomely rewarding it the way it has done lately over almost four decades: in other words, “sitting on market beta” won’t work anymore.

The possible solution, then, is straightforward: true active management. This means a range of actions for the investors who aim at returns higher than the 0%, or thereabouts, markets are expected to offer.

Rebalancing between equities and bonds is a good starting point: market volatility, either causing bonds to fluctuate less than equities or, even better, causing them to correlate negatively with equities, will add to the 10 year return, provided the investor is disciplined in switching between the two asset classes. That is, to sell bonds and buy equities when “blood is running in the streets” and do the opposite when euphoria for stocks turns into mass hysteria.   

Looking at the widest possible opportunity set is another characteristic of the true active investor: this means looking globally (and certainly beyond the U.S. market, at the moment particularly expensive), as well as considering other asset classes, such as commodities or inflation-protected bonds (such as TIPs in the U.S.), that offer better protection against inflation.

Lastly, and by no means less important, the “staple activity” of the active investor: the selection of the best stocks, industries and themes, offering higher expected returns and better protection against rising inflation expectations.

Conclusion

True active management is, in our opinion, the inescapable way forward from the current low level of expected returns for both equities and bonds. Simple. But not easy, since active management is essentially a zero-sum game, clearly discriminating winners from losers. The tide is going out – to borrow the metaphor from Warren Buffett – and we’ll soon see who’s been swimming naked.