That crucial question is at the heart of every retirement financial plan, especially in today’s era of low rates that virtually guarantee that bonds will provide a dismal long-term return. Stocks will have to do the heavy lifting of funding your retirement, in other words. So if you assume too high a return for stocks, you will run out of money during your retirement; assume too low a return and you deny yourself the standard of living you have earned.
The occasion to revisit this crucial question is the release this past week of new research from the Pew Charitable Trusts on the rates of equity return that are assumed by state pension plans. That study shows that the median assumed return has dropped over the last decade from 8.0% to 7.4% in the latest survey.
Read: Retired and facing a stock market downturn? Here’s what to do — and not do
Pew’s researchers add that even this 7.4% return is probably too high, and I heartily agree. For this column I review the predictions implied by eight valuation indicators that have the best record forecasting the stock market’s 10-year return. (For a fuller discussion of these indicators’ track records, consult a Wall Street Journal column I wrote a year and a half ago.)
• Household equity allocation. Specifically, this indicator is the average portfolio allocation that U.S. households commit to stocks, as I discussed in greater length in a recent column;
• The price/book ratio, the ratio of the S&P 500’s SPX, +1.01% by per-share book value;
• The price/sales ratio, the ratio of the S&P 500’s by per-share sales. This indicator was the focus of a recent MarketWatch story;
• The dividend yield, the percentage that S&P 500’s dividends per share represent of the overall index;
• The cyclically adjusted price/earnings ratio championed by Yale University’s Robert Shiller;
• The so-called “q” ratio that derives from research conducted by the late James Tobin, the 1981 Nobel laureate in economics. The ratio is calculated by dividing market value by the replacement cost of assets;
• The traditional P/E ratio. I calculated this ratio by dividing the S&P 500 by its trailing year’s earnings per share;
• The so-called Buffett Indicator, which is the ratio of the total value of equities in this country to gross domestic product. It is so named because Warren Buffett, CEO of Berkshire Hathaway BRK.A, +0.71% BRK.B, +0.71%, suggested in 2001 that is it “probably the best single measure of where valuations stand at any given moment.”
To calculate what each of these indicators is forecasting for the next decade, I constructed an econometric model that most closely fit the relationship between its historical readings and the S&P 500’s subsequent 10-year return at each step along the way. I then fed into that model the indicator’s current reading. The forecasted 10-year nominal return that is implied by these eight indicators ranges from a high of 3.5% annualized (for the traditional P/E ratio) to a low of minus 5.4% (for the price-to-sales ratio).
Pretty depressing, isn’t it? Even the most optimistic of these forecasts is barely a quarter of the 13.6% annualized total return the S&P 500 produced over the last decade.
What about low interest rates?
One rebuttal to these sobering projections is that the stock market deserves to appreciate at a faster rate when interest rates are as low as they are today. But the historical data do not support this rebuttal.
Consider what I found when I added the 10-year Treasury yield as an input to each of the econometric models I constructed for the above eight indicators. In no event did this addition increase the model’s forecasting power, at least at standard levels of statistical significance. In any case, furthermore, my PC’s statistical package found that higher Treasury yields were associated with higher subsequent stock market returns, not lower.
This is just the opposite of what the bulls are assuming when arguing that low interest rates justify higher projected stock market returns, of course. And it is not particularly surprising, once you stop to think about it, since stubbornly low interest rates indicate that the markets expect future economic growth to be no better than anemic.
I discussed this at greater length in a Retirement Weekly column a month ago. As explained to me in an email from Nicholas Bloom, an economics professor at Stanford University and co-Director of the Productivity, Innovation and Entrepreneurship Program at the National Bureau of Economic Research, “interest rates are an excellent predictor of long-run growth potential, and their moribund level [today] reflects the markets’ expectation of sustained low future growth.”Reality check
As a further reality check on the low projected returns over the next decade, I next calculated the stock market’s future return using a bottom-up, fundamental model based on projected sales growth, inflation, and dividend yield. When doing so, I come up with a projection of the market’s nominal 10-year return of 4.5% annualized, and 2.6% after inflation.
My calculations were as follows:
•Sales growth: 0.9% annualized after inflation. My rationale is this: If we assume that corporate profit margins stay at their current inflated level, which is a generous assumption since those margins are so much higher than their historical average, then the stock market’s future growth will be a function of sales growth. And it’s hard to see how sales can grow faster than the economy as a whole, which the Congressional Budget Office will grow at a 1.8% annualized rate above inflation over the next decade. And from this rate we must subtract an estimate of GDP growth that does not come from publicly traded corporations—such as entrepreneurial startups, private equity, venture capital, and so forth. Following research from Robert Arnott, founder and chairman of Research Affiliates, I subtract 0.9 of an annualized percentage point.
• Dividend yield: 1.9% annualized
• Inflation: 1.7% annualized (based on projections from the Cleveland Fed)
To be sure, mine is a simple model. I offer it here solely to show that the projections of the eight indicators with good long-term records aren’t that far out of line with a completely different approach to forecasting the market.
It’s worth noting in this regard that Arnott’s firm employs a far more sophisticated fundamental model than my simple one, and it is currently projecting an annualized 10-year return of just 2.4% nominal and 0.3% inflation-adjusted.
The bottom line?
There no doubt are other models out there that project more robust equity returns over the next decade. But it is very sobering indeed that 10 valuation approaches are each projecting very low returns between now and 2029.
You therefore might want to consider adjusting your retirement financial plan to include lower projected equity returns. If you’re wrong, you’ll be pleasantly surprised, and that’s a far better outcome than being too optimistic and discovering that you have run out of money.