If company earnings were the most important factor associated with company stock returns, shouldn’t the S&P 500 average annual index performance have reflected the average annual earnings growth of the companies that it comprises? According to Ben Carlson of Ritholtz Wealth Management, over the past 80 years corporate earnings have increased by about 6% annually yet the index has returned 11%, almost double that amount. Even when comparing the returns of the S&P 500 with corporate earnings growth on a ten year basis, it’s hard to see any correlation. From 1930 through 1939, for example, public company earnings declined by 42% in aggregate. (No wonder it was called a Depression!). But over the same period the stock market declined only 1%. In the 1950s corporate earnings rose only 46%, while stocks were up 487% (the best decade for the stock market over the last hundred years). How about the decade beginning in 1970? Earnings climbed 157% but stock prices grew only 77%.

Carlson and most other market watchers assert that the reason is that markets are forward looking and reflect expectations of the future rather than the present. That viewpoint is also supported by regression data indicating that the correlation between stock prices in one year and the following year’s market performance is higher than the correlation between those two factors within the same year. But I prefer to express it in a different way.

Besides company earnings the other factor that impacts stock market performance is what I (and other professionals) call investor sentiment. It’s measureable indirectly through a company stock’s Price/Earnings (P/E) ratio. This multiple indicates the price per share investors are willing to pay for a certain level of company earnings per share. When the stock’s P/E ratio is higher, it means that investors are willing to pay more for the same level of earnings from that company as compared to when its stock’s P/E ratio was lower, and vice-versa. Individual company P/E ratios fluctuate all the time and you can aggregate them to reflect investor sentiment with respect to the overall market at any given moment.

When stock prices rise significantly more than company earnings as they did in the 1950s, you’ll find that the difference comes from the change in investor sentiment. For example, at the beginning of 1950 the Schiller Cyclically Adjusted P/E (CAPE) ratio was about 10.5. At the end of the decade it was over 18.3. And it had nearly reached 20 in 1956. The greater willingness of investors to pay higher premiums for stocks over that period was the reason the market rose so much faster than company performance did.

Unfortunately investor sentiment is at best a lagging indicator of market returns. It does accurately explain why stock prices were not correlated with company earnings over certain periods in the past. But it provides no indication as to how investors might behave in the future. So even if you could predict company performance you will have no clue about future investor sentiment.

There is a trading strategy called Momentum investing. Proponents believe that stocks that are rising in price will continue to do so for some moderately extended period of time, and vice-versa. But that is really a market timing strategy and there is no data I have seen that would support anyone’s ability to consistently outperform the market using it.

It all comes down to this: you can’t predict market movements so don’t bother to try. If you are going to take advantage of all the growth benefits that market investing offers, use diversification and other strategies to keep the volatility at a level you can tolerate while providing you with the growth you need.