Returns from the stock market: What to expect
(MoneyWatch) On July 12, 2013, the S&P 500 finished at a record close of 1,680, a 148 percent rise (not including dividends) from its 677 closing low on March 9, 2009. After withdrawing hundreds of billions of dollars from stock funds in recent years, investors are now pulling assets out of safe bond funds and using the cash to buy stocks -- at much higher prices and much higher valuations than they were selling them for not too long ago. Individual investors are, once again, making investment decisions based on the rearview mirror. They should not. When markets are setting new highs, it's just as important to keep a balanced perspective and adhere to your investment plan as when markets are setting new lows. With that in mind, let's take a look at the market's current valuation and what implication that has in terms of what returns you can reasonably expect.
Smart investors know that while there's little to no evidence that you can successfully use valuations to time the market, valuations do matter. And they matter a lot in terms of predicting the mean expected return you can expect from your portfolio. Higher valuations predict lower returns and vice versa. And since valuations matter, there's no way to decide on an asset allocation plan without forecasting returns. Unfortunately, even investors who actually have investment plans, fail to estimate returns when making their asset allocation decision.
- What to worry about when investing
- The new normal: Did Bill Gross get it right?
- Beware bad financial journalism
Smart investors also know to avoid the mistake of treating the expected return as anything more than the mean of a very wide dispersion of potential outcomes. There's simply no way to intelligently design a plan without having a forecast of returns.
Not long ago, I discussed the evidence on using Robert Shiller's P/E 10 ratio (also called the CAPE ratio) to estimate returns. On July 15, 2013, the P/E 10 stood at 24.6. Cliff Asness' study on the P/E 10 found that when it was between 21.1 and 25.1 the real return averaged just 0.9 percent. When it was above 25.1 (within spitting distance of the current level), the real return over the following 10 years averaged just 0.5 percent -- virtually the same as the long-term real return on the risk-free benchmark, one-month Treasury bills, and 6.3 percent below the market's long-term real return.
That example provides us with two lessons. The first is that just because the market has historically provided a real return of close to 7 percent, it doesn't mean we should use the historical return to project the future. Again, valuations matter. The second is that you shouldn't treat the mean as the only possible outcome. For example, when the P/E 10 was between 21.1 and 25.1, the best 10-year real return was 8.3, and the worst was -4.4 percent. When the P/E 10 was above 25.1, the best 10-year real return was 6.3 percent, just 0.5 percent below the historical average, and the worst was now -6.1 percent. As you can see, as valuations rise, the mean not only falls, but the best returns are lower and the worst returns become even worse.
It's important to understand why there's such a wide dispersion in outcomes away from the mean. The reason is that valuations change. Consider the following evidence presented by Gary Miller and Scott MacKillop in the July 2013 edition of Financial Advisor. In order to test for time periods that would allow them to test the predictive value of the 10-year normalized earnings yield, as an indicator Miller and MacKillop looked for time periods where the beginning, ending and average P/E 10 were all the same. They found just two such periods since 1926.
The first period was March 1927 through July 1954. The P/E 10 at the beginning of the period was 14. It was 14.1 at the end of the period, and the average annual P/E 10 during this time was also 14. That translates into an earnings yield of 7.1 percent. If the earnings yield is a good predictor of future stock market returns, we would expect to see annualized real returns of about 7.1 percent over the relevant time period. The actual return during that period was 7.1 percent.
The second time period was March 1961 through January 2010. The P/E 10 at the beginning and end of the period was 19.6, while the average annual P/E 10 through these years was 19.4. This translates into an earnings yield of 5.1 percent. The actual return during that period was 5.0 percent. As long as the beginning and ending valuations are the same, the P/E 10 does a great job of predicting returns. However, valuations drift, and they can drift a lot.
With the current P/E 10 at 24.6, we get an earnings yield of 4.1 percent. However, because the P/E 10 is based on the 10-year trailing earnings and real earnings tend to grow at about 1.5 percent a year, we need to multiply the 4.1 percent yield by 1.075 (1.5 percent x 5 -- since we are dealing with a 10-year period the average lag is five years). That brings us to an adjusted earnings yield of 4.4 percent -- which is well below the 6.8 percent historical figure. And that return depends on valuations remaining unchanged. It's important to understand just how much valuations can vary. In 1920, the P/E 10 fell below 5, and in 1999 it rose to almost 45.
The wide variation of the P/E 10 demonstrates why it's so important that a financial plan incorporate the potential for such variation -- and that you consider ahead of time what actions you will take to prevent your plan from failing to achieve its goals should valuations fall from current levels.
The bottom line is that those investors making decisions based on the historical long-term real return to stocks of 7 percent are highly likely to be disappointed. While there have been periods when valuations were as high as they are today and investors earned even greater returns, the evidence suggests that this is highly unlikely. Those investors relying on such returns will likely end up with portfolios that are well below the level needed to meet the spending objectives set in their plan.
Finally, it's important to understand that this post is not meant as market-timing recommendation -- using the current valuation as a signal to get out of stocks. Remember that while stocks have an expected real return of just 4.4 percent (using an adjusted Shiller earnings yield), with expected inflation at less than 2.5 percent, you have to extend the maturity of an investment in Treasury bonds to about 10 years to earn any real expected return.
Image courtesy of Flickr user Tax Credits.