If you have heard “stocks are really cheap,” or “stocks are too expensive right now,” it is probably in reference to valuations—widely followed calculations that compare a stock or an index’s price to some aspect of their underlying businesses. Financial professionals have long touted their predictive powers because—in theory—they hint if stocks are overvalued or undervalued and primed to reverse course. However, Fisher Investments’ reviews of market history show they don’t actually predict stocks’ direction.
In Fisher Investments’ reviews of financial commentary, we find price-to-earnings (P/E) ratios usually attract the most attention. There are three main kinds: The trailing P/E (price to the past 12 months of earnings), the forward P/E (price to analysts’ estimates of earnings in the next 12 months) and the cyclically adjusted P/E, or “CAPE” (price to the trailing 10-year average of inflation-adjusted earnings). We will focus on the first two here.
Generally speaking, people view P/Es above the long-term average as expensive and those under it as cheap. Yet this framework doesn’t do much good when trying to identify market inflection points. Exhibit 1 shows US stocks’ trailing P/Es and their long-term average alongside the four most recent bear markets (including 2022’s). As you will see, there is no set level at which a bear market begins—nor any inherent level where the market becomes overvalued. Similarly, there is no magic P/E threshold where bear markets must end.
Exhibit 1: Bear Markets Begin and End at High, Middle and Low Trailing P/Es
While P/Es tend to rise during bull markets and fall in bear markets, their inflection points are clear only in hindsight. Moreover, as seen above, P/Es wiggle plenty along the way, in part because earnings fluctuate. A common mistake is thinking only prices affect P/E, but, in reality, they can fall when earnings rise and vice versa. In our view, considering why a P/E is moving is important.
Even P/Es’ longer-term trends within bull and bear markets aren’t straight up and down. For instance, trailing P/Es often rise early in a new bull market, then fall for a spell before rising again in the bull’s more mature stages. In Fisher Investments’ view, this is because stocks are a leading economic indicator while earnings lag. In most cases, earnings continue falling in the early months of bull markets. Meanwhile, prices rise, lifting P/Es and thus making stocks appear pricier. When earnings catch up as the bull matures, P/Es start falling even as stock prices still rise. Stocks may then appear “less pricey,” but their actual prices are much higher.
This brings up another weakness of P/Es: They are backward-looking. The numerator reflects stocks’ performance, which doesn’t predict future returns. Meanwhile, the denominator—the prior 12-months’ earnings—reflects previous economic activity, which has no effect on future results. Fisher Investments’ reviews of market history suggest stocks lead the economy, pricing in likely earnings growth 3–30 months ahead. To see this, consider one of the most extreme examples: March 2009, when stocks reached the low of the 2007–2009 global financial crisis and bear market.[i] At the same time, trailing P/Es hit an astronomical 110—much higher than their 20.6 long-term average.[ii] The recession’s end was still three months away, and earnings still had further to fall—after all, economic growth drives earnings, not the other way around. But stocks had bounced high off March 9’s low in anticipation of this recovery. Earnings continued falling through Q4 2009, then rebounded quickly—sinking P/Es to 17.4 by the end of February 2010.[iii] Over the same stretch, stocks rose a whopping 66%.[iv] While 2009 is an extreme example, it represents a repeat phenomenon. Stocks typically price the eventual economic and earnings recovery before it actually happens, in our experience.
Fisher Investments recognizes that there isn’t strong evidence forward P/Es are predictive, either. Here, too, the numerator reflects past performance—stocks don’t predict stocks. Meanwhile, their denominator represents analysts’ earnings estimates for the next 12 months, but analysts tend to be very late to shift their predictions, staying both bullish and bearish for too long. This can distort forward P/Es’ signals early in bear and bull markets alike, as shown in Exhibit 2.
Exhibit 2: Bear Markets and Forward P/Es
It is also important to consider why forward P/Es move during a cycle. As bull markets mature and sentiment improves, investors become willing to pay increasingly larger premiums for future earnings. Fisher Investments’ reviews of market history suggest this is a normal part of sentiment’s evolution, often lasting a while before a bull market ends. Similarly, in a bear market, panic selling can last a while—deflating P/Es. Therefore, relatively “expensive” stocks can always become more expensive and “cheap” stocks can always get cheaper.
Valuations themselves might not tell much about markets’ direction, but extreme changes—and investors’ reactions to them—can provide a decent snapshot into sentiment. A rapid spike late in a bull can indicate euphoria, while a fast dive late in a bear might indicate sentiment has overshot to the downside. Monitoring investors’ reactions is important, as it might hint at the degree to which investors are dismissing the moves—giving a better gauge of the gap between sentiment and reality. Valuations are not timing tools, but assessing their rapid movements in context with sentiment can help assess whether surprise power lurks to the upside or downside, in Fisher Investments view.
Investing in securities involves a risk of loss. Past performance is never a guarantee of future returns. Investing in foreign stock markets involves additional risks, such as the risk of currency fluctuations. The foregoing constitutes the general views of Fisher Investments and should not be regarded as personalized investment advice or a reflection of the performance of Fisher Investments or its clients. Nothing herein is intended to be a recommendation or a forecast of market conditions. Rather it is intended to illustrate a point. Current and future markets may differ significantly from those illustrated herein. Not all past forecasts were, nor future forecasts may be, as accurate as those predicted herein.
Sources:
[i] Source: FactSet, as of 09/13/2022. S&P 500 price level, 10/09/2007–03/09/2009.
[ii] Ibid. S&P 500 trailing-12-month P/E, monthly, 12/31/1990–09/13/2022.
[iii] Ibid. S&P 500 earnings-per-share and trailing-12-month P/E, monthly, 12/31/2007–02/26/2010.
[iv] Ibid. S&P 500 Index price return, 03/09/2009–02/26/2010.