FREE NEWSLETTER
TIME TO PLAY MARKET strategist. Trying to figure out what sort of U.S. stock returns we can expect over the next 10 years? Nobody knows for sure, of course. But we can at least think about it in a reasonably logical way—by using what some folks call the Bogle method.
What’s that? In a 1991 article for the Journal of Portfolio Management, Vanguard Group founder John Bogle—who died in January 2019—laid out a relatively straightforward method for estimating stock returns. He subsequently revisited the idea a number of times, including in his book Common Sense on Mutual Funds. Here’s what the Bogle method tells us about today’s expected stock returns:
Dividends. This is the one component of future returns that’s known for sure. For buyers of the S&P 500 today, the starting dividend yield is some 1.3%, which is very low by historical standards. Over the past 50 years, the S&P 500’s yield has averaged 2.8%.
Earnings. To the dividend yield, Bogle would add expected earnings growth. Assuming 2021’s corporate profits come in as expected, the S&P 500 companies will have notched average annual growth in earnings per share of 7.3% over the past five decades, or 3.3% a year after adjusting for inflation.
I’m inclined to assume future earnings growth will be similar, but there are no guarantees. Even over decade-long stretches, earnings growth has been all over the map. The S&P 500’s earnings per share grew an inflation-adjusted 2.4% a year in the 1970s, fell 0.7% a year in the 1980s, grew 4.7% in the 1990s, slid 1.9% in the 2000s and soared 8.7% in the 2010s.
I was surprised by this variability. Over the past 50 years, there’s never been a single year when real (inflation-adjusted) GDP grew by 8.7%, let alone notching that rate for an entire decade. Similarly, in the 2000s, when earnings per share shrank by 1.9% a year, after adjusting for inflation, there was only one year when real GDP was negative.
Still, over the past five decades, real GDP grew 2.7% a year, similar to the 3.3% growth rate in inflation-adjusted earnings per share. There may be a short-term disconnect between the two because, say, a recession causes a big hit to reported corporate profits, but GDP and corporate earnings should roughly track each other over the long haul.
Based on the difference in yield between 10-year Treasury notes and 10-year inflation-indexed Treasurys, bond investors expect 2.5% annual inflation over the next decade. If we assume earnings per share grow 3.3 percentage points faster, we’re looking at 5.8% nominal growth in earnings. Add the 1.3% dividend yield, and that puts us at 7.1%.
This is what Jack Bogle called the stock market’s investment return. But is this what investors will get? That brings us to the Bogle method’s third factor.
Speculation. For investors to earn 7.1% a year from the S&P 500 over the next decade, the S&P 500 stocks would need to finish the 10-year period with the same price-earnings (P/E) ratio that they collectively have today.
Suppose the S&P 500 ends the year at 4700, close to Friday’s close. That means we’d be at a P/E ratio of roughly 25, assuming reported earnings for 2021 come in as expected. (A nerdy aside: At year-end, the S&P 500’s P/E will appear higher because, at that juncture, companies won’t yet have reported their 2021 earnings.)
Will the S&P 500 be at a P/E of 25 at year-end 2031? Your guess is as good as mine. Consider the range of possibilities:
Feeling queasy about the stock market’s prospects? Let me toss in even more uncertainty. Over the past 50 years, corporate earnings per share have benefited from widening profit margins, declining corporate tax rates and an end to dilution. Profit margins could narrow and corporate tax rates could rise, hurting earnings growth over the decade ahead. On the other hand, it seems dilution, which is caused by companies issuing ever more shares, is no longer a big issue, thanks to stock buybacks. Those buybacks may ensure inflation-adjusted earnings per share grow faster than the 3.3% I assumed above.
Where does all this leave us? I think we can draw two key conclusions. First, the Bogle method suggests we should ratchet back our return expectations for the next 10 years. It seems unlikely that the S&P 500 will notch 10% a year, and it could return far less. That’s the bad news.
The good news: Even if U.S. stock returns are grim, investors should still make more than they would with bonds. Yes, the ride for stocks will be rougher than that for bonds. But that rougher ride should be rewarded—even if it isn’t richly rewarded.
Jonathan Clements is the founder and editor of HumbleDollar. Follow him on Twitter @ClementsMoney and on Facebook, and check out his earlier articles.
Want to receive our weekly newsletter? Sign up now. How about our daily alert about the site's latest posts? Join the list.
Thanks Jonathan, interesting, but it seems to me that from a practical point of view, the amount of the return is of no consequence. Its kind of like how will the Mets do next year. From an asset choice perspective, it is always equities having the best return, and owning very low returning bonds only to reduce volatility and/or act as a source of funds in a down market. TINA, there is no alternative, trumps everything else. What interests me is which market segment to overweight, and vice versa.
For example, commodities do well in periods of inflation, but not so well if central banks are restricting economic growth to control that inflation. Higher interest rates help financials, however the yield curve is actually flattening, which is worse.
Out of curiousity, my understanding is that buybacks have replaced roughly half of dividends over time. For that reason, I’ve long assumed Bogle’s equation was missing the 2% or so impact of buybacks that happen each year (highly variable thought that is.) Depending on how accurate his equation has been historically(?), it sounds like dilution was not factored in either, so perhaps the two offset and Bogle’s equation still “works”?
Of course, animal spirits (speculation) is a huge part of realized stock returns, which means any outlook has a huge range of error. I’ve been hearing about reduced forward earnings expectations for almost as long as I’ve been investing (even in 2009/2010 as the long bull run started), but I’m sitting at a healthy 9.05% CAGR over 26 years (8.4% if you dollar-weight it, which makes sense as my AA has grown more conservative over time). This seems like it’s right in line with historical returns, given a portfolio that started at 100% equities and now is about 73%.
Regardless of the answer, I just don’t see value to factoring in long range forecasts; the unknowns are too great, and black swans too common. I just rebalance when needed to control risk, and let it ride.
Maybe I missed this, but with the Fed printing money like they are, I feel that action has inflated many asset classes over and above any fundamental legacy metrics.
I have no idea how that will end; hard or soft landing…
So, is this the method that Vanguard uses to generate its Asset-class return outlooks? That’s what I use to periodically tweak my allocations (among the six Vanguard index funds that I hold). So, on that basis, I’ve tilted a bit more toward international and value, especially in my Roth accounts, by adding a bit of VEMAX (emerging markets) and VSIAX (small cap value). Core holdings are VTSAX (total stock market) and VTIAX (total international stock), with also some VGSLX (REIT) and VFIAX (S&P500).
“Market Perspectives: December 2021“
I’m not sure how Vanguard generates its asset class forecasts, but I suspect it’s using a more sophisticated computer model (not that you should assume that greater sophistication means greater accuracy).
Jonathan what is your forecast for Ex-US? Total International? Developed markets? Emerging Markets? Thanks, Dave
I don’t have access to the data needed to produce a comparable forecast for international stocks. But based on valuations and a firm belief in long-run reversion to the mean, I think the outlook for overseas markets — both developed and emerging — is bright for the decade ahead.
I don’t understand the whole P/E discussion because all that really matters is how much I can sell the stocks for the day I need cash. (And the tax rate if gains.)
I was a huge fan of Jack Bogle. I often wondered if dividend growth should have been factored in…starting at 1.3%, but the dividend yield on investment will rise over time.
The growth in dividends is effectively captured by also incorporating earnings per share growth. In other words, Jack could have got the same result for the market’s “investment return” by looking at current dividend yield plus dividend growth as he got by combining dividend yield plus earnings growth. This is the basis for the so-called Gordon Equation:
https://humbledollar.com/money-guide/gordon-equation/
In effect, the Gordon Equation is the same as the Bogle method, except it doesn’t consider changes in price-earnings ratios.
Dividends would be down as stock buybacks have increased as you mentioned in the dilution comment.
And, of course, the hope is that would result in inflation-adjusted earnings per share growth that’s faster than the 50-year 3.3% average. But set against that is the risk of narrowing profit margins and higher corporate tax rates.
Non Us stocks seem to have a higher dividend yield and since they have a lower PE ratio I would have thought they’d have a lower risk of multiple contraction. So probably by this formula they should be likely to outperform US stocks. However I’ve been thinking this for several years now, and so far I have been wrong…🤔
I agree — but I, too, have been wrong!