The P/E ratio is a fundamental measure of any security's valuation, indicating how many years of current profits it takes to recoup an investment in the stock. The current S&P500 10-year P/E Ratio is 35.08. This is 77% above the modern-era market average of 19.6, putting the current P/E over 1 standard deviation above the modern-era average. This indicates that the market is Overvalued. The below chart shows the historical trend of this ratio. For more information on this model's methodology and our analysis, keep reading below.
P/E ratios are a cornerstone of fundamental stock valuation analysis, and are most commonly looked at for individual firms. The P/E ratio is (as the name suggests), a ratio of a stock price divided by the firm's yearly earnings per share. The implied logic here is that a mature firm (with no capex investments) returns all profits to shareholders via dividends. The P/E then becomes a measure of how many years it will take the investor to earn back their principal from the initial investment. For example, if you buy 1 share of ACME Co for $100, and ACME consistently makes profits of $10 per-share, per-year, then it follows that it would take the investor 10 years to earn back their original $100 investment.
P/E is (unless otherwise stated) calculated using the last reported actual earnings of the company. Let's look at another example - one where we expect future earnings to grow. Imagine TechCo was founded 5 years ago, and their earnings per year (per share) have been $0, $1, $1.50, $2, and $5. Let's also assume that TechCo's current share price is $100, just like ACME in the prior example. Because the most recent earnings-per-share for TechCo is $5, that means TechCo's P/E ratio is $100/$5 = 20. The message here is that, at current earnings, investors in TechCo will theoretically get their money back after 20 years. This is twice as high as ACME -- but why? If it takes twice as long for TechCo to make profits as it does for ACME, why is their stock valued at the same price? The answer is obviously the growth rate of TechCo's profits. TechCo is a new company, and has been growing profits very quickly over the last 5 years, clearly investors expect that to continue. This is why high-growth companies tend to have very high P/Es - the market has very high expectations for their future results (relative to current results).
The same analysis can be done to the entire stock market. By adding up the price of every share in the S&P500, and comparing that to the sum of all earnings-per-share generated by those companies, you can easily calculate the P/E ratio of the US stock market.
Below are both the total S&P500 aggregate value, and aggregate earnings. Figure 1 shows the data on a normal scale chart, Figure 2 shows the same data on a logarithmic scale, which highlights the relationship between the two more clearly. Note that all data here is adjusted for inflation.
Figure 2 above shows a clear relationship between price and earnings. Just by eyeballing that chart you can see that both have steadily risen over time, and that S&P500 price tends to stay (very roughly) 10x-20x larger than yearly earnings. Chart 2A shows only the last twenty years of data, which highlights the conspicuous divergence of this trend in the prior year, where S&P price has gone up, but earnings are dropping precipitously. Let's observe this relationship between Price and Earnings explicitly by charting the P/E ratio, below.
Figure 3 above shows the standard calculation of the S&P500 RE ratio over the prior century, in light blue. Since this is a measurement of current price divided by most recent annual earnings, the calculation is subject to high volatility caused by peaks and troughs in the business cycle. For example, in mid 2008 at the nadir of the financial crisis, S&P500 earnings across the board fell to almost 0. Despite stock prices also going down significantly, this caused the market P/E at the time to rise over 120, off the scale of the chart.
For that reason, rather than use the current P/E, when doing long-term analysis it is more useful to use the Cyclically Adjusted Price Earnings (CAPE) ratio. This is very similar to the regular P/E, but rather than using the most recent earnings data, the CAPE ratio looks as current Price divided by the average earnings over the prior 10 years. The CAPE ratio is shown in Figure 3A below in dark blue, and largely follows the same trend as the current PE ratio with a lot of the volatility smoothed out.
Going forward we will be using the same CAPE data as above (Figure 3A, dark blue), but only from 1950 onwards. The intention here is to have a historical period from which we can compare current data.
Figure 4 shows the same CAPE data as Figure 3A, but with overlays showing the modern-era average (arithmetic mean) P/E value of 19.8 (baselined as 0%), as well as horizontal bands showing +/- standard deviation bands. As of January 22, 2021, the S&P500 P/E ratio is 77% higher than its modern era average. By this valuation, the market is Overvalued (see our ratings guide for more information). To fall back to the modern era average, the S&P500 would need to return to around $2,200.
Below is the same chart, showing only the data from 2000 to present, to display more recent detail.
And finally, let's look at how this data corresponds to S&P500 performance.
This final chart shows two important ideas:
First, the main line shows the standard S&P500 since 1950, but color coded according to the standard deviation bands in Figure 4. I.e., when the S&P500 was more than 1 standard deviation below its P/E average (such as in 1950, and again during the mid-70's to mid-80's) the chart is colored dark green, signifying undervaluation and a buying opportunity. While this clearly demonstrates that the P/E valuation model is correlated with S&P500 returns (by definition) in the long run, it also shows that the value is limited in trying to time the market for ideal entry/exit points. For example, this model would have shown that the early 1980's was severely undervalued and a great buying opportunity, and also that the late 90's were severely overvalued and a good exit spot. However the model missed the tech crash of the early 2000's, where even at the lowest point of that crash the market was still overvalued according to this model.
Second, take note of the dotted line in Figure 6. This shows the price level of the S&P500 if it were continuously valued at the modern-era P/E (CAPE) average of 19.8. Note that it over the last year or two it has risen sharply as corporate earnings have increased in line. This suggests that while the market is overvalued, real corporate earnings have been increasing steadily over time (i.e., the economy is getting increasingly efficient) which does help justify increased stock prices.
The primary case against using historic PE as a valuation metric is the idea that PE ratios ought to be getting more expensive over time. Here is one cogent example of this argument. There is no arguing that the CAPE ratio has risen over time, and particularly since ~2000 when tech/growth stocks have becoming increasingly dominant in the S&P500. We agree that it doesn't make sense to compare today's market to the 1800's, and so this criticism is primarily why we do not use data prior to 1950 in this model.
The below table cites all data and sources used in constructing the charts, or otherwise referred to, on this page.
|SP500 Price||Yahoo! Finance S&P500 Daily Close Values|
|Historical Corporate Earnings||Publicly available S&P500 monthly data, aggregated and published by Robert Shiller at Yale;|
|Current Quarter Estimated Corporate Earnings||Estimates of current quarter SP500 aggregate earnings. [Click Additional Info > Index Earnings]|