First, perspective. Over a very long period of time stocks have returned 8.1% a year, or 6.6% a year net of inflation. That is for the period from 1802 through 2012 according to Professor Jeremy Siegel of the Wharton School of the University of Pennsylvania. From the end of 2012 through the end of 2018 the S&P 500 Index with dividends reinvested returned 12.1% a year, or 10.6% net of inflation. That is according to data from Bloomberg. Looking at things from day to day we see stocks bouncing up and down, but from a long run perspective they have definitely gone up. While there certainly will be downturns in the future, if the past is any indicator, the overall trend is decidedly up. If you have a long horizon, staying invested in the stock market is a rather good strategy.
Now, to the point. Still, it is natural to worry. Is there currently reason to believe a downturn is likely in the near future? Should we be selling stocks now in order to put our money into T-bills or money markets? That is always a difficult question because it brings with it another question: When will it be time to get back into stocks? It is not just one question but two. You don’t want to stay in money markets for the long term. You might stay ahead of inflation by a little but not by much. Stocks have beaten T-bills by over 10% per year, on average, over a very long period of time. The question, then, is when to buy and when to sell. Unfortunately, it is very difficult to time the market so as to buy low and sell high. The tendency is to let emotion get the better of us. What is needed is an objective rule. If we had such a rule would now be a time to sell?
We at Northwest Criterion have studied the issue. We used over 60 years’ worth of stock market and T-bill data and looked at a class of rules for switching between stocks, as represented by the S&P 500 Index, and T-bills. We compared the earnings yield (which is the reciprocal of the P/E, i.e., E/P) of the S&P 500 to the yield on T-bills. When the earnings yield on stocks got low enough relative to the T-bill yield, we sold stocks and bought bills. When the earnings yield on stocks got high enough, we sold bills and bought stocks. We tried quite a number of rules of this form, each rule being determined by the switching levels. Every single rule we tried was at least as good as staying long the S&P 500 the whole time. The risk was never increased. The return was often increased, and the ratio of the return to the risk was always increased.
However you choose your switching levels, the results show an improvement. Based on this study our recommendation would be to sell stocks (or specifically, the S&P 500 Index) and buy bills if the earnings yield on the S&P 500 exceeds the T-bill yield by one percentage point or less. Buy stocks when the earnings yield exceeds the T-bill yield by three percentage points or more. This rule definitely would have reduced risk during the period studied. If you had owned stocks the whole time, never switching, you would have made an average annual return of 10.8%, but you would have experienced declines of over 20% on six separate occasions. If you had followed this rule, you would have reduced your annual return slightly, to 10.2%, but you would have experienced only two declines of over 20%.
What is the current situation? According to Bloomberg the P/E of the S&P 500 Index is 18.4 as of February 21, 2019, for an earnings yield of 5.4%. The yield on the 3-month T-bill is 2.4%. That is a spread of three percentage points. At that spread the rule says you should buy stocks if you do not already own them. Significantly, every rule that switched to T-bills at a spread of three percentage points or more did worse than the rule just suggested. Now is not the time to switch to T-bills.
We don’t know whether the stock market is headed for a downturn in the near future, but we believe that for the best long-term results, it is right to be long stocks at the current levels. How much you should have invested in the stock market depends on your own risk tolerance, however. How to think about that is another issue.
The results of the study discussed are linked here.