Current PE using trailing 12 month earnings is higher than at average market peak.
Pundits who claim stocks are not overvalued are using meaningless noise of PE spikes that occur because earnings collapsed in a recession.
When a bull market has run for a few years, a declining PE can be a sign prices are about to roll over. They could be now?
On a historical chart of PE using trailing 12 month earnings ("TTM"), the current level does not look particularly high. Most of the high points come when stock prices are low, but earnings have collapsed even more during a recession. Understanding valuation requires putting the PE in context.
The above chart of PE confuses the viewer because of the tendency to assume market highs occurred when the PE hit a high. The reality is often different.
The chart above cuts off two of the spikes in PE since the scale only goes to 35, but it shows all the meaningful stock market peaks.
PE has averaged 17.11 at the 18 stock market peaks. The PE for March 2014 of 18.60 is a fraction above the PE where the market peaked in October 2007.
Let's be clear on where the numbers come from to calculate the PE. I am using "as reported" earnings from the S&P statistical service since it has a much longer record than "operating" earnings. I am not using any forecasted earnings. So the March PE uses the latest 12 months earnings available, which is 100.20 through December 2013. The price is the average of the daily closing prices during the month. The average was 1863.52 for March. To compare apples to apples, the PE for each month on the chart uses the price for that month and the earnings from 3 months earlier. Data prior to the official beginning of the S&P 500 is from Cowles and Associates Common Stock Indexes.
Using the average monthly price occasionally gives an unexpected result. For example, in 2000, the month with the highest average price was August rather than April, when the highest daily close occurred. So the red dot marking the high in the chart above is plotted in August 2000.
Another unexpected result on the 2000 high was that the PE peaked in April 1999 and declined for 16 months before the market peaked. The PE declining for a few months before stock prices peaked also happened for the peaks in: 1902, 1906, 1909, 1916, 1937, 1956 and 1973. Going into these eight peaks, stock prices were moving higher while earnings were rising faster. Apparently, exuberance drove up stock prices ahead of the good earnings, and when the stronger earnings came, the more slowly rising prices were just a transition to falling prices. So the downtrend in PE from 19.1 in November 2013 could mean the slower rising stock prices of the last few months are a transition to a bear market.
Here is a closer look at the data since 1970.
The stock market peak in October 2007 was a bit unusual in that the market peaked while the PE appeared to be near its trough. On the front end of the trough, earnings were rising faster than prices in recovering from the 54% drop in earnings during and after the 2001 recession. On the back end, earnings started falling several months before stock prices peaked and fell faster than prices after the peak. So the PE continued rising after prices peaked.
If you are using volatile 12 month earnings to calculate a PE, it becomes very important to understand what part of the cycle the economy and stock prices are in. The difference between data being noise and meaningful is understanding the context.
Using a smoothed, more stable version of earnings, as Robert Shiller does with a 10 year moving average, or as I do with single exponential smoothing, makes it easier to interpret PE.
I use real stock prices and the smoothed real earnings in the chart above to construct what I call PEses shown in the chart below.
In this chart, market tops are much more likely to coincide with tops in the PEses. The March PEses of 45.38 is substantially above the average at which bull markets peak and is higher than all but two market peaks. This measure of valuation is useful in evaluating the expected return after dividends and inflation over the next 15 to 20 years, but is a very poor timing tool. While it is possible the stock market is putting in a major top, this measure of valuation could only show that with hindsight.
In context the current PE using 12 month trailing earnings or a version of smoothed earnings shows the stock market (NYSEARCA:SPY) to be substantially overvalued and at risk of significant loss. The PE by itself does not warn when that loss will begin.
Disclosure: I am short SPY. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: There is no guarantee analysis of historical data their trends and correlations enable accurate forecasts. The data presented is from sources believed to be reliable, but its accuracy cannot be guaranteed. Past performance does not indicate future results. This is not a recommendation to buy or sell specific securities. This is not an offer to manage money.