Until recently, the expected long-term total return for equities seemed obvious. Since 1871, U.S. stocks had outgained inflation by an annualized 6.9%. When Wharton professor Jeremy Siegel extended that analysis to 1802, the result barely altered, registering 6.7%. That figure closely resembled the average stock market gain of 20 non-U.S. countries during the 20th century.
Thus, equities figured to earn 6% to 7% per year, after inflation. The evidence urged the conclusion. While stocks are of course highly risky in the short term, over time their returns were, to cite Siegel’s phrase, “remarkably durable.”
That premise of persistent 6%-plus stock market returns, when measured over long periods, has since been threatened. Additional work from the international-stock investigators trimmed their estimate to an annualized 4.3%. And in “Stocks for the Long Run? Sometimes Yes, Sometimes No,” which scrubbed and updated the U.S. equities database, Edward McQuarrie of Santa Clara University cut the pre-World War II estimate for real equity performance to 5.4%. That essentially matched what bonds had delivered.
As detailed in McQuarrie’s paper—and summarized in my article—the combination of lower stock market returns and somewhat higher bond market totals (which have also been revised because of further research) scotched the assumption that equities invariably outgain fixed-income securities. Before World War II, stocks managed that feat on only about half the rolling 30-year periods.
McQuarrie’s objection
Should you care? McQuarrie believes so. As he points out, the advice to hold stocks for the long run, as popularized by Professor Siegel, relies heavily on the “stationarity” of equity returns. The logic supporting that advice was:
- It is true that the future will not repeat the past.
- It is also true that investment conditions evolve. Industries come and go, as do government policies, interest rates, and corporate profitability.
- However, knowing that things will be different in the future is not particularly helpful, because profiting from that realization requires that one predict how things will differ. Good luck with that.
- Consequently, the best forecast for future events is what already occurred. The more stable the historical results, the more reliable that estimate.
- Because long-term real stock market returns, across both time and countries, have been extremely stable, the expectation that over several decades they will earn a real return of at least 6% annualized, while also beating bonds, is quite sensible.
McQuarrie’s paper undermines that argument by challenging the stationarity presumption. When examining both the full U.S. record and the experience of other countries, he maintains, equities do not outperform so consistently. The apparent norm is, in fact, a case of modern American exceptionalism.
The revised history
For McQuarrie, U.S. investment history consists of three stages:
Three investment cycles
(U.S. Markets, McQuarrie's Formulation)
Source. Edward McQuarrie.
Initially, equities and bonds traded wins. Equities then triumphed spectacularly and unprecedentedly for four decades, before the horse race once again resumed. (Strictly speaking, equities have also outdone bonds during the third period, but the contest has been much closer than during the previous decades.)
The counterargument
The rebuttal to this interpretation comes naturally. The affairs of other nations are only moderately relevant to this country. What happened when buggy whips were popular, even less so. A reader amusingly phrases the latter contention:
“New data reveals that mortality rates for leg amputations from 1793 to 1920 far exceed what was previously reported. Therefore, we should adjust the expected mortality rate for leg amputations going forward. Some may ask, ‘Have not advancements in medical knowledge, equipment, medicines, and hygiene greatly reduced the risk of dying after surgery?’” Not, however, the data collectors.
Obviously, his contention is technically incorrect. McQuarrie certainly has contemplated such issues—but has been unconvinced. After all, some aspects of human existence have changed less than others. Modern medicine, transportation, and communications far exceed Alexander Hamilton’s experience. On the other hand, Mr. Hamilton would immediately almost understand today’s courtrooms and political discussions.
Which side Is correct?
The debate about the utility of the additional investment data is therefore practical rather than theoretical. Some 19th-century experiences remain pertinent, while others have become outdated by either technology or custom. What about stock and bond market performances? Where do they rate in that framework?
On that issue, the academic community and investment professionals tend to diverge. When projecting future investment returns, both parties extrapolate from the past. In that, they are alike. Where they differ is with data selection and adjustments. For most finance professors, forecasts are best conducted with as much data as possible and as little tinkering. In contrast, investment professionals tend to pick their spots. When sifting through the history, judgment is required.
My sympathies land somewhere in the middle. On the one hand, American investment markets have indeed hugely changed. The 19th century was a cesspool of stock price manipulation, fraudulent offerings, and investment panics. What’s more, the Federal Reserve system had not yet been created. Consequently, bad times all too frequently became worse owing to the lack of a lender of last resort.
That said, as even the dissenting reader granted, McQuarrie’s evidence is wonderfully useful for challenging complacency. By historical standards, modern American equity returns have been exceptional. They are unmatched within this country’s history, and with few equals outside it as well. There are many reasons for their success, and many intelligent and thoughtful observers who believe that such prosperity—give or take—can continue. (Warren Buffett, for one.)
But the optimists could be wrong. The case for the ongoing dominance of stock is less overwhelming than we once believed. That observation bears consideration, especially for retirees tempted by advice that they invest heavily in equities. Such counsel is not necessarily wrong, depending upon individual circumstances (in particular, wealth levels), but it often is coupled with the implicit assumption that stocks will inevitably beat bonds over the long term.
Maybe. If not, though, retirees do not get a second bite at the financial apple. That lesson is very much worth pondering.
Note: When I sent Professor McQuarrie the reader’s comment, he forwarded me the following response:
“My take is in the article’s title. Sometimes stock returns will soar far above bond returns, as after the war. That outperformance can be sustained for decades. Other times stocks will lag bonds, for decades. There’s no rhyme or reason to it, and in all likelihood, no predictability over the individual investor’s limited time horizon of several decades.
“As for your reader’s clever riposte, here is my redo: ‘The rate of death from disease and epidemics stayed at a relatively high and constant level from 1793 to 1920. Then advances in modern medicine fundamentally and permanently altered the trajectory ... or so it seemed until COVID-19 hit in February 2020.’”
John Rekenthaler has been researching the fund industry since 1988. He is a columnist for Morningstar.com and a member of Morningstar's Investment Research Department. The views of the Rekenthaler Report are his own. This article is general information and does not consider the circumstances of any investor. Originally published by Morningstar and edited slightly to suit an Australian audience.
Try Morningstar Premium for free