By Mark Hulbert
Stocks don’t always beat bonds, no matter how long you hold them.
That is the disturbing conclusion of one of the most important research projects completed over the last month. It’s alarming because, if we can’t count on stocks to beat bonds no matter how long the holding period, we need to rethink what otherwise has been the bedrock principle underlying virtually all of our financial plans.
The research in question was conducted over the past four years by Edward McQuarrie, a professor emeritus at the Leavey School of Business at Santa Clara University. Soon after retiring from university teaching in 2016, he told me in an interview, he embarked on the painstaking process of creating and updating a database of U.S. stock and bond returns back to 1793. He said he had no idea it would take him four years to complete, but it did. He finished the project earlier this year and in late March posted on the Social Science Research Network a summary of his findings .
A full discussion of his findings deserve many columns. But in this, my regular monthly review of the latest Wall Street research, I want to focus on what arguably is the most explosive of McQuarrie’s findings.
The accompanying chart, above, summarizes McQuarrie’s data. The way to interpret the chart is to remember that, if the odds of stocks beating bonds go up with holding period, the columns would get higher as holding period lengthens.
But notice that this is the case for only one of the three sub-periods since the late 1700s. Over the 1794-to-1862 sub-period (represented by the red columns), in contrast, the odds that stocks will outperform bonds go down as holding period lengthens. And in the 1863-to-1942 sub-period (represented by the green columns), there is no trend: Stocks’ odds of beating bonds are no greater with a 50-year holding period than with a one-year holding period.
Only in the period since World War II (the blue columns) has it been the case that stocks’ odds of beating bonds grow monotonically with holding period. Yet it is this most recent period alone that most financial planners and analysts use to extrapolate the past into the future. It’s dangerous to simply ignore the rest of U.S. history, according to McQuarrie.
The future is uncertain
Clearly, U.S. stock and bond history is far more complicated than most of us, until now, have thought. McQuarrie believes we are lulling ourselves into a false sense of security when we repeat, mantra like, that the long-term always bails out stocks.
To be sure, stocks beat bonds by enough in the decades after World War II so that, when averaged over the entire period since 1793, stocks do come out ahead of bonds — by 1.9 annualized percentage points.
But there are two crucial aspects to this long-term average that you need to keep in mind.
The first is that it is a lot lower than the equity premium that has been found by other historians focusing on shorter historical periods.
The second is that even this reduced equity premium is of small consolation to investors whose entire investment careers took place during the 150-year period in which there was no equity premium (or a negative one).
The corollary is that, even when focusing on a period as long as 226 years, conclusions about what the future will be like are profoundly uncertain.
“No one knows,” McQuarrie concludes, “and that is the real point of the new history. The new data show that results for the 20 century did not generalize to the 19 century; there is no reason to expect them to generalize to the 21 century either.”
He adds: “Sometimes both stock and bond investors do poorly. … Sometimes both assets enjoy strong returns. … But many other permutations are possible. … Both stocks and bonds can disappoint over an arbitrarily long interval, even as either asset can bestow handsome rewards.”
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at firstname.lastname@example.org .