In a previous post, I discussed how to find the appropriate P/E ratio of a company based on current earnings and I gave a rough estimate for appropriate P/E ratios if the company’s earnings are increasing or if they are decreasing. In this article, I will address historical average P/E ratios in the stock market.
Please take a look at the graph of historical PE ratios vs. the price of the DJIA over the last 100 years. The most obvious conclusion we can draw from the chart is that there are distinct cycles in the market: bear markets with decreasing P/E ratios are followed by bull markets and increasing P/E ratios. Another obvious fact we can see is that for most of its history, the average P/E on the S&P 500 has been between 10 and 20. Also, the stock market’s returns have been greatest following periods of low P/E ratios than in periods following high P/E ratios.
Unfortunately, these trends do not bode well for our near-term performance. There are two likely possibilities in the near term (5-10 years): stock prices will fall, perhaps precipitously, or stock prices will neither fall nor rise. In the first case, P/E ratios would also fall quickly, and by the time they fall to about 10 on average, we would want to be fully invested. In the second case, P/E ratios will gradually decline as profits increase but stock prices go nowhere.
I am not an expert on forecasting the future, but there are forecasters who predict both of the above possibilities. John Mauldin, in his book Bullseye Investing (an excellent book by the way), predicts what he calls a ‘muddle-through economy.’ Mauldin sends out a free weekly newsletter to which everyone should subscribe.
Richard Russell, on the other hand, in keeping with Dow Theory, argues for a standard bear market with falling stock prices. Russell, by the way, is ranked as the second best market-timer by Mark Hulbert since 1980. The best market timer is a system based solely on the date.
Anyway, future stock returns do not look good at this point, at least for the averages. If we look at individual stocks, though, there are still some that are good values. Unfortunately, that list is small, and I can find few great values. Even if the market goes against us (or does not move at all), we can do fine. I do not think that we should anticipate average returns exceeding 10% per year over these next few years. That being said, an overvalued market is a stockpicker’s market. Whereas Sir John Templeton could buy 100 cheap European stocks in 1939, including many in bankruptcy, and make fabulous profits on almost all, we will have to pick and choose carefully. There are still companies that are undervalued, with a potential for giving us outsized profit.
As investors concerned with value, we must remind ourselves that a bear market is a good thing. This is where time diversification benefits us. As stock prices fall (or, as they do not move but profits increase), and P/E ratios fall, we must gradually increase our allocation of assets to stocks. Thus, as value gradually returns to the market, we should gradually sell bonds (or use cash) to buy more stocks.
I am not a market timer. I can think of very few times when it actually made sense to be out of the market. In 1929 it would have been advisable. But in 1937, 1966, and 2000 there were still some fairly valued stocks. For example, we would have been fine buying Sysco [[SYY]] in 2000, but not Cisco [[CSCO]].
Disclosure: I own no shares in any of the companies mentioned. See the disclosure policy.
[This was originally published October 2005 . The market’s performance since then validates my belief that market timing is not worth the effort.]