I’ve been resisting the temptation to post about the recent market turmoils. After all, mindful investors ignore market gyrations whether they’re caused by scary-sounding global pandemics or anything else. Mindful investing is the same regardless of the market’s volatility:
- Buy a moderately diversified set of low-cost stock index funds and hold them for the long term (at least 10 years).
You’ll note that this summary doesn’t mention bonds. Mindful investors don’t have much use for bonds, at least at the super-low yields they’ve offered for the last half-decade.
So, I’m puzzled by the many recent news articles and posts citing the ongoing correction as a “perfect” example of why investors need some bonds in their portfolios. This graph of Portfolio Visualizer returns data for the last two months explains why some people are crowing about bonds right now. Stocks are represented by the S&P 500 and bonds by the U.S. 10-year Treasury bond.
I’ve also included here a “combined” portfolio consisting of 60% stocks and 40% bonds because it’s the most commonly mentioned simple portfolio for individual investors. Stocks have had a dismal start to 2020, and bonds have so far proved to be a fantastic safe haven. Does the Mindfully Investing logic supporting stock-heavy portfolios have a big hole in it where the bonds should be? When the stock market is convulsing, are we mindful investors making a terrible mistake by holding zero bonds?
The short answer to both questions is no. Mindful investors should take comfort that all of the problems inherent with investing in super low-yield bonds remain. If anything, most of the problems with bonds have gotten worse over the last month of market mayhem. In this post, I examine two major claims for why bonds were “essential” during the recent market correction, and why these claims are still unconvincing to an investor with a mindful perspective.
1. Bonds Rise When Stocks Fall
Probably the most common reason people buy bonds is that they expect them to mitigate stock losses in a combined portfolio. The combined stock/bond portfolio is rooted in the fundamentals of diversification, where investors seek assets that zig when their other assets zag. The returns for 2020 so far are indeed a good example of bonds fulfilling this function in a combined portfolio. And historical data provide further support for this idea. This graph shows how bonds performed in years when stocks had negative returns based on Aswath Damodaran stock (S&P 500) and 10-year U.S. bond data.
Over the past 92 years, annual stock returns were negative in 25 cases or about 27% of the time. Putting bonds aside for a second, that’s a good reminder that negative annual stock returns are a relatively uncommon occurrence, and that’s why mindful investors like to invest in stocks. But getting back to bonds, in only three of these 25 down years for stocks were bond returns also negative. That’s a pretty impressive track record for bonds.
However, comparing 100% bonds to 100% stocks is not particularly realistic. Almost no one recommends 100% bonds for any investor¹ because most of the time, long-term bond returns have been dismal relative to stocks. So, what happens if we instead examine the historical performance of the standard 60/40 portfolio in years when stocks posted negative annual returns? Here’s that graph.
The solid blue line in this graph shows a theoretical one-to-one relationship between the returns of two portfolios. If a portfolio’s returns fall below that solid blue line, that means it performed worse than the 100% stock portfolio. Although the stocks in the 60/40 portfolio dragged down its overall performance to some extent in negative stock years, the 60/40 portfolio beat the 100% stock portfolio (all the dots are above the solid blue line) in all of these historical cases. The dotted blue line on the graph shows the best-fit for the historical data. Interestingly, this best-fit line indicates that the 60/40 portfolio mitigates stock losses by about half. For example, the dotted line intersects the point where stocks decline by 20% (horizontal axis) and the 60/40 portfolio declines by about 10% (vertical axis). That seems like pretty good mitigation of stock losses. But to be clear, when stock returns are negative, the 60/40 portfolio usually won’t produce positive annual returns.
Assets that mostly move together are “positively correlated”, while those that tend to move in opposite directions are “negatively correlated”.² We’ve seen some pretty strong evidence that annual bond returns are negatively correlated with annual stock returns. But looking at annual data gives a somewhat false impression of the dynamic nature of stock/bond correlations. The claimed negative correlation between stocks and bonds has varied substantially over time, and it’s only in the last 20 years that the correlation has been negative. And as shown in this Portfolio Visualizer graph of two long-lived Vanguard mutual funds (VFINX for stocks and VBFMX for bonds), even in the last 20 years, stocks and bonds were only mildly and sporadically correlated in the hoped-for negative direction.
Given this history, no one can guarantee that bond returns will be negatively correlated with stock returns under all future market conditions. The current relationship between these assets could reverse and start to mimic the 1990s. And 2018 recently reminded all of us that bonds and stocks can both have negative returns in the same year.
Although bonds in a combined portfolio have a good historical track record of loss mitigation, that mitigation is partial (about half in the case of the 60/40 portfolio). And the variations in stock/bond correlations suggest it would be unwise to expect that degree of mitigation for every future market condition. Overall, I give the bond proponents half-credit for this claim.
2. Rebalancing With Bonds Can Beat Stocks
One question with diversified or combined portfolios is whether to “rebalance” them. Given that different assets produce different returns over time, the proportions of assets in a combined portfolio can drift significantly over time. Rebalancing involves periodic resetting of a combined portfolio back to its original proportions to correct for this drift. This is often purported to create a “rebalancing bonus” of increased long-term returns.
My past examinations of rebalancing show that rebalancing stock/bond portfolios often fails to produce a rebalancing bonus as compared to non-rebalanced portfolios, particularly when investing over many years. And researchers have found that the rebalancing bonus is mostly a product of the combined luck of when you choose to rebalance and what the markets are doing at the time.
Nonetheless, I recently saw a graph from 1986 to 2020 showing that a portfolio of 100% stocks underperformed a combined stock/bond portfolio that was rebalanced twice a year. I tried to recreate that graph using Portfolio Visualizer and got somewhat different results. Here’s my graph showing the growth of a $10,000 initial investment. The blue line is 100% stocks (S&P 500), the orange line is a 60/40 portfolio rebalanced quarterly³, and the red line is 100% bonds (U.S. 10-year bond).
Even though stocks still beat the 60/40 portfolio by my calculations, the rebalancing of the 60/40 portfolio produced an annualized return of 9.5%, which was pretty close to the annualized return of 10.4% for the all-stock portfolio, and much better than the 6.8% return from the all-bond portfolio. However, basing conclusions on this one time-period is pretty misleading because it coincides almost exactly with the entirely unprecedented 37-year bond bull market.
For a wider perspective, we can examine how often a rebalanced 60/40 portfolio outperformed a 100% stock portfolio by looking at annualized 5-year rolling returns using the Damodaran data going back to 1928. Because these are annual return data, my calculations assume annual rebalancing of the 60/40 portfolio. This graph shows the difference between 5-year annualized returns for the two portfolios over time.
When the blue line is in positive territory, it means that the 100% stock portfolio had higher returns than the 60/40 rebalanced portfolio. The orange hatching shows the reverse situation when the rebalanced 60/40 portfolio outperformed 100% stocks. It turns out that most of the time the 100% stock portfolio has outperformed a rebalanced 60/40 portfolio. Most of the exceptions include at least one recession, although stocks weathered many other recessionary periods better than the combined portfolio. So, it’s reasonable to say that a rebalanced stock/bond portfolio sometimes outperforms an all-stock portfolio, but it’s more the exception than the rule. Stocks had superior long-term returns performance relative to the 60/40 portfolio, even during the nearly 40-year bond bull market. So, I give bond proponents zero credit for the claim that a rebalanced stock/bond portfolio beats an all-stock portfolio.
Other Claims About Bonds
Most of the other claims I’ve seen recently about bonds are the same recycled stuff that’s been around for years, which I’ve covered in past articles on bonds.
For example, the idea that bonds offer decent returns with less volatility is refuted by the fact that long-term bond returns have been historically half that of stocks, and reduced volatility does not equal reduced risk. Further, the best predictor of total returns over a bond’s duration is the bond’s current yield, and yields are incredibly low right now. The 10-year U.S. bond currently yields less than 0.8% (an all-time low), while my online savings account is still paying 1.5% in annual interest. So, saying it’s a good idea to lock in a 0.8% total return for the next 10 years is absurd, regardless of how low bond volatility may be. You can’t fund your retirement with low volatility.
Another recycled claim is that bonds provide regular income for retirees. But of course, most retirees will find their nest eggs entirely insufficient to fund retirement on less than 1% income. Again, locking in a miserable return for 10-years is way worse than putting that money in a lowly savings account and looking for better options when interest rates and bond yields start to normalize.
Conclusions
The supposed benefits of bonds all seem to circle back to emotional, not rational reasons. Mitigating stock losses sounds great, but if over the long term, stock returns handily trounce bonds, what’s the point? It seems the point is to feel better when the stock market crashes.
However, I submit that few advisers or individual investors know for sure exactly how they will react to the next stock market crash. We’ve seen that the standard 60/40 portfolio has historically halved the magnitude of temporary negative stock returns. So, if stocks drop by 40%, while the beloved 60/40 portfolio drops by 20%, is everyone so sure they won’t panic sell at -20% anyway? Every green soldier wonders whether they might be paralyzed by fear in combat, and none of them knows for sure until the battle starts. Some portion of the investing population will undoubtedly breakdown and sell some or all of their 60/40 portfolio at -20%, or even -10%.
There are tons of folks on Reddit and Twitter that already sound close to panic, even though the stock correction has been relatively mild so far. Having a stock/bond portfolio in no way inoculates you against emotional decisions in an uncertain future. Many people will benefit from practicing mindfulness or similar calming techniques to avoid emotional decisions, regardless of what portfolio they hold.
And if that’s true, all the benefits of bond investing evaporate, because they all assume that reducing volatility is the only way to handle our emotions. If you have a hammer, everything looks like a nail. For finance folks, it seems that every problem needs a financial solution, even when a mental or emotional solution might work much better.
1 – The main exception is that some advisers recommend 100% bonds for investors that are deep into retirement. But I generally disagree with that notion for many of the same reasons that I’m writing about today.
2 – And if two assets move randomly to each other, they are “uncorrelated”.
3 – Quarterly rebalancing gave me the highest returns for the 60/40 combined portfolio.
