Popular Market Indicators That Work…Until They Don’t
Some in the financial media have been warning that noteworthy indicators—namely the CAPE ratio and the Fed Model—are signaling that stocks are too expensive. The question is whether these models work.1
The CAPE ratio shows the S&P 500 trading at 35 times the previous decade’s earnings. That’s the third highest the indicator has been since the 19th century, and it’s even higher than it was in 1929 just before the Great Depression. This seems to be sending a glaring signal that large-cap stocks are too pricey.
The problem is that elevated CAPE ratios have not been particularly good at predicting secular bear markets. The CAPE ratio was also near its current level in July 1997, when the gauge passed its 1929 peak shortly after Federal Reserve Chairman Alan Greenspan warned of “irrational exuberance” in the market. But the S&P 500 went up +28.34% in 1998 and +20.89% in 1999, and has delivered annualized returns of +8.3% for investors who owned stocks long-term from January 1998 to December 2023.2
In an August 2014 New York Times op-ed, Robert Schiller wrote that “the United States stock market looks very expensive right now. The CAPE ratio, a stock-price measure I helped develop – is hovering at a worrisome level.” Yet 2014 was essentially the midpoint of the longest bull market in history, which ran from March 2009 until the Covid-19 pandemic. And even then, the 2020 bear market was fueled not by stretched valuations but by the shutdown’s effect on earnings.3
I certainly do not mean to admonish the CAPE ratio as an indicator, as it offers investors useful data to consider. The point is that the CAPE ratio has not always been good at predicting down markets because investors must weigh far more than one data point when making investment decisions about the future. The CAPE ratio also uses decade-old earnings data, which is great for historical comparisons but tells us virtually nothing about what lies ahead (which is exactly what we want to inform our investment decisions). Factors like technological innovation (AI), productivity gains, expected inflation and interest rates, and global economic growth trends can greatly impact a company’s future earnings. A stock’s price is influenced by what investors are willing to pay for expected earnings, meaning that other variables like sentiment, growth forecasts, innovation, and management play key roles. The CAPE ratio doesn’t account for any of these factors.
Then there’s the Fed Model, which compares the earnings yield on stocks to bond yields, namely U.S. Treasuries. When the spread is tight or bond yields are higher than earnings yields, stocks are viewed as expensive. In the current environment, the earnings yield on the S&P 500 is approximately 1% higher than the 10-year U.S. Treasury bond yield, which makes stocks look relatively unattractive. But much like the CAPE ratio, the Fed Model has been dead wrong in the past. In November 2007, the Fed Model showed stocks as their cheapest relative to bonds in the model’s history. The Global Financial Crisis came next.
I can also remember when investors were embracing the model of comparing stocks’ dividend yields to bond yields as a signal of whether stocks were cheap or expensive. But again, this model failed at key times—it showed a negative (sell) signal in the late 1970s and never subsequently issued a buy signal. For investors who committed to following this model, it would have meant selling out of the market in the 1970s and never buying back in. The annualized return of the S&P 500 from 1980 to 2023 was +11.87%.
Bottom Line for Investors
Investors are constantly searching for ways to ‘value’ the market. Valuations help us determine when stocks are cheap and therefore attractive, or when they’re expensive and should be avoided. But valuations alone should not direct an investment strategy. If they did, and investors relied on the CAPE ratio, Fed Model, or other metrics as proxies for when to invest over the past few decades, it would have meant being out of stocks for long stretches of time. For long-term growth-oriented investors, that’s a mistake.
Instead, investors should zoom out and think more about the expected return of the stock market over long periods in history, which tends to be about a 6% to 8% return above the risk-free rate (U.S. Treasurys). Instead of focusing on what models tell investors to do in moments in time, it’s better to invest over long periods to generate wealth steadily through compounding—not to try and gain an edge by timing the market.
Disclosure