The chart above illustrates how dividend yield, a simple measure of valuation, has been a good predictor of long-term returns. The chart’s black line tracks the S&P 500® Index’s dividend yield (dividend divided by price) as of the end of each calendar year dating back to 1957.
The chart’s gold line shows the index’s 10-year annualized total return. The chart aligns the two series to show for each year-end both the dividend yield and the annualized total return for the 10-year period that followed. For example, the dividend yield at the end of 1957 was 4.4% (left scale) and the annualized total return for the ten years that followed was 12.8% (right scale).
The chart shows the strong correlation between dividend yield and the 10-year returns that follow.3 In general, when dividend yields have been high, so have returns for the following ten years. And, when dividend yields have been low, low returns have followed.
So, what does this mean for returns going forward? Year-end dividend yields have ranged from 1.1% to 5.4% since 1957. The S&P 500® Index’s dividend yield at the end of 2017 was 1.9%. This indicates that returns over the next ten years are likely to be below average.
Below average does not necessarily equate to bad, however. For example, at the end of 2005 the dividend yield was 1.8% and the annualized 10-year total return that followed was 7.3%. That’s below average, but not bad. Nor does the current dividend yield necessarily equate to a negative 10-year return. Dating back to 1957, negative 10- year returns have been extremely rare with only two occurrences. Those two periods began with dividend yields of just 1.3% and 1.1%, respectively. But, in addition to low beginning dividend yields, another significant event contributed to the negative returns: both periods ended during the bear market that coincided with the worldwide financial crisis of 2008-2009.
Our stock market forecasting model, which incorporates dividend yield along with other measures, currently points to a 10-year annualized return range of approximately 4.0% to 5.3% for the S&P 500® Index. Our model is conservative, but appropriately so, in that it assumes that stocks end the period at a valuation, relative to earnings, slightly below to slightly above the long-term averages. Of course, the valuation level ten years out is impossible to predict. Should the index’s end-of-period valuation be higher than our assumption, then returns will also be higher, and vice versa. It is also important to mention that we have yet to adjust our model for the positive effect that the new tax law will have on corporate earnings.
Nevertheless, our model’s forecasted return range is the lowest in many years. Therefore, it will be important for investors to take actions to improve their results. One action we took in 2017 was to increase our recommended target allocation to international equities, which have more attractive valuations and the potential benefit of rising currency values relative to the U.S. dollar. A second action that can enhance returns is to maintain your target asset allocation by adding to your equity portfolio when prices have declined and trimming it back after substantial rallies.
Bonds continue to be an important part of a diversified portfolio. Our outlook for bond returns is little changed from one year ago despite the rise in yields on shorter-maturity bonds over the past year. We currently forecast that annualized total returns over the next ten years for portfolios of U.S. Treasury and corporate bonds with 1-10 year maturities will average just below 3.0%. Our model assumes that interest rates will gradually rise and that spreads (the difference in yield between bonds with credit risk and U.S. Treasury bonds) will widen. Should yields rise as we anticipate, our future expected returns will rise with them.