Stock Market Valuation Requires a Long-Term View

Includes: DIA, QQQ, SPY
by: J.D. Steinhilber

U.S. stocks are trading at their cheapest valuations (based on earnings and book value) since the early 1980s. Foreign stocks are even cheaper, having fallen to valuations not seen since the mid-1970s. Several broad equity indexes now trade at less than book value. Of course, in this sort of environment—characterized by an economic and market freefall with no end in sign, and a fear of an unstoppable debt deflation spiral—most investors have shunned traditional valuation measures.

This recent comment from analyst John Mauldin is typical: “Valuation is the wrong concept in a secular bear market. When earnings start to stabilize, then we can talk about fair value.” The most bearish analysts posit the application of a “trough” P/E multiple of 8x—10x on “trough” earnings (currently $28.75, which includes all the write downs taken in the financial sector), which produces a target of approximately 300 for the S&P 500.

If you have any faith in our ability to get through this crisis, this analysis is terribly flawed. Ultimately, the value of the stock market is not determined by a single year’s earnings, but rather its long-term, normalized earnings power, which is approximately represented by the mid-point within the upward-trending channel shown below. At present, the bottom of the channel is approximately $50. In the past 60 years of stock market history, the lowest multiple of bottom-of-channel earnings has been 10x, which occurred briefly in 1974 and again in 1982.

This projects a potential worst-case target of 500 on the S&P 500. I am skeptical the market will get that low, largely because of the extreme difference in the interest rate environment. Both the 1974 and 1982 lows were characterized by high risk-free rates (i.e. the T-bill rate). In 1974, an investor could earn 8% on T-bills, and in 1982, 12%. Today, the return on such absolute safety is under 0.5%.

[click to enlarge]