If past performance is no guarantee of future results, then how can I estimate what will be the future return of my entire portfolio for the particular risk level I am taking?
John C. Bogle, in his 1999 book “Common Sense on Mutual Funds” emphasizes an Occam’s Razor approach to investing, particularly when you index your portfolio to broad market indices. He looked at all 10-year periods starting in 1926, and found all stock market returns for each were attributable entirely to three components: the initial dividend yield, the average growth rate in earnings, and the change in the P/E ratio. The first two are based on fundamentals, but the last one is determined by speculation.
While it is possible the US stock markets could have a long period of zero or negative returns, what would be the cause? The most likely one is the speculative change to the market’s P/E ratio. If the current Shiller P/E 10 measure is 27 and the adjusted mean is somewhere around 19 or 20, reversion to the mean could certainly cause a period of lower returns, as it did in the 1970s. For that decade, Bogle’s fundamentals forecast for initial dividend yield and growth rate in earnings were 3.4% and 9.9% respectively, but due to P/E compression during the period of -7.6%, the predicted total return was only 5.7%. The actual return for the period was 5.9%. Pretty close!
The “Japan crowd” takes it further, though. Their argument is one of secular stagnation. As defined by the Financial Times, it is “a condition of negligible or no economic growth in a market-based economy.” I do not agree with those who argue it is happening in the US and in Europe. First, there is no comparison when considering the economic diversification and wealth of natural resources that the US has versus Japan. While the decline in US population growth is certainly a cause for concern, it is not enough alone for there to be a real similarity. Another difference is the economic globalization brought on by the dominance of multinational corporations. Also, if you index and hold both domestic and foreign market weights, you have a lot of diversification between different economies, even if one major economy stagnates. So, is this even a concern for long-term investors? Yes, a large meteor could wipe out the whole planet, but then we won’t be here to argue about the returns then, will we?
One can never accurately predict what changes in the stock market will be determined by speculative forces. That leaves the fundamentals. Critics of fundamental forecasting point to Japan, and its decades of stagnant growth, as one good reason we cannot assume economic growth and positive returns in stock markets. While the argument seems to be in accord with what we know about the high correlation between corporate profit growth and GDP for the last 100 years, the difference is over whether perpetual economic growth can be assumed into the future. This is a rather fatalistic view of the world, in my view. Japan’s situation was really more about its high starting P/E valuations than secular stagnation on an economic level. The economic factors were also Japan-specific like declining consumption, a declining workforce, and restrictive labor laws. That is not to say those factors cannot happen here, but the point is that investors with a long-term focus, who hold indexed global portfolios, with a healthy amount of bonds, can make confident assumptions with long-term forecasts of returns, even when those returns include the possibility of secular stagnation. If that problem were to become a global phenomenon, the pain would be unavoidable for all of us. In that case, what is the point of the argument?
Rick Ferri publishes a 30-year forecast each year. I like that it is a 30-year forecast because it respects the very volatile nature of returns in the short-term, while giving long-term investors a way of estimating expected real returns for every asset class that generates real returns. This is why you will not find gold or commodities in the forecast.
Finally, let’s say you have eight different market slices, such as Large Cap Blend, Small-Cap Value, REITs, International Total Stock Market, Emerging Markets, Intermediate Treasuries, Intermediate Corporates, and TIPS, each with its own 30-year expected real return. Even if the 30-year returns end up being off for a few of them (perhaps one or two will be higher and one or two will be lower) the overall portfolio return has a pretty good chance (no guarantees!) of meeting your overall long-term real return estimate. This is a good reason for diversifying using multiple approaches, such as asset type (stocks and bonds), factors (small and value), and geography (domestic and foreign). Calculate the portfolio real return by multiplying the percentage for each asset class by the real return, then adding the total after doing the same calculation with each asset class in your portfolio. For example, if US Large Cap is 10% of your portfolio, look up the real return for it, which for 2015 is 5%. So, the US Large Cap portion of the portfolio’s total expected return is 0.5% (0.1 x 0.05 = .005).
As for the other approaches, even if you do not prefer factor tilting, be sure to include at least some foreign exposure (20-50%), and ignore short-term returns. If you do these things, you will greatly improve the chances your portfolio’s long-term real return target will be met.