A few years later, Shiller and Campbell (1988) showed that variations in dividend yields do not forecast the growth rates of future cash flows, but do forecast future market returns. Recognizing that dividends are a poor measure of a company’s cash flows, Shiller and Campbell used a ratio of real (net of inflation) market price relative to 10-year average of real earnings—which they called the cyclically adjusted PE, or CAPE, ratio—to reach the same conclusion. Almost three decades later, Shiller received the 2013 Nobel Prize in Economics, largely in recognition of his research on the long-horizon importance of valuation.
The Gordon Growth Model, which assumes constant returns and constant growth rates, is an oversimplification, but because valuations cannot grow indefinitely, Shiller’s logic applies: valuations should predict either future cash-flow growth or future returns, or a mix of the two. If valuations do not forecast future growth rates, they must forecast future returns, and indeed they do; the log of CAPE offers a correlation with subsequent 10-year and 20-year stock market returns that is even stronger than −80 percent in the 90 years covered by the Ibbotson data, and a still-impressive −58 percent in the 135 years spanned by the Shiller data.
CAPE is many value investors’ North Star, guiding them as a beacon, revealing the market as either cheap or rich. Some investors follow this signal with religious fervor as a one-stop shop in all markets, at all times. The zealots should recognize that, like the blind men disagreeing about the nature of an elephant (one measuring the leg, another the trunk, another the tusk, another the ears), CAPE gives us an incomplete picture. CAPE, like all financial measures, is only one imperfect measure. A cult-like adherence to a single valuation metric can lead to missed opportunities at best, and at worst, can be value destroying. As we have clearly demonstrated, however, ignoring CAPE is not a sensible choice. We recognize CAPE is an imperfect market-timing tool, at best, but supported by a strong empirical and theoretical background.
Rob Arnott is the founder and chairman of Research Affiliates. Vitali Kalesnik is director and head of equity research. Jim Masturzo is senior vice president of asset allocation.