Even in today's ebullient stock market, most investors do pay at least nodding attention to the valuation of individual stocks. But one of the key measurements of valuation -- the Standard & Poor's 500 index -- has become so distorted it is now seriously misleading. Yet almost no one recognizes this.
For many decades the valuation of "the market" has been measured by the S&P 500, which was a useful yardstick because individual stocks typically sell at higher or lower price/earnings ratios relative to the market, depending on whether the businesses behind the shares have faster or slower earnings growth, and higher or lower profitability, than the index.
Thus, at any given time, one could look at a stock and determine whether it was cheap or expensive relative to the market.
Over the years, superior companies such as Coca-Cola, Hewlett-Packard and Merck usually have sold at p/e ratios 20% to 50% above the S&P 500, while companies achieving spectacular growth, such as Microsoft, have had p/e premiums of 40% to 80%.
Underachievers such as International Paper and Niagara Mohawk usually have sold at p/e ratios 40% to 50% below the S&P 500. Over many years,when a stock moved outside of its normal valuation range, such deviation usually provided a useful buy or sell signal.
However, the effectiveness of the S&P as a valuation yardstick has been destroyed by the distortion that has developed during 1998-'99, when a few super-popular companies moved up to unprecedentedly high valuation levels, making the overall index truly a four-foot yardstick.
To understand this, look at the S&P 500's average p/e ratio, and break down the 500 companies between the 25 with the highest market caps and the remaining 475, and look at median or typical p/e ratios. The accompanying table is revealing. (For comparison, all of these numbers are based on trailing earnings.)
The popularity of Microsoft, Cisco, Lucent, America Online, Wal-Mart, the major pharmaceutical stocks and a few other favorites has produced a huge gap between the average p/e ratio of the overall index and that of the typical stock, as indicated by the median.
The only other time such an anomaly has occurred was with "the Nifty Fifty" in 1972-'73. The aftermath was devastating for the most popular stocks.
But the key lesson is that comparing the valuation of any stock to the assumed "market multiple" of 28 and making judgments based on such a comparison is dangerous, because the more realistic market multiple is around 21.
Yet Wall Street keeps saying "this stock usually trades at a 25% premium to the S&P 500, but its current multiple of 32 represents only a 14% premium, so it's a good buy." But is it a good buy at a premium of 52% to the median p/e ratio of the 500 S&P companies?
Carpenters can't build sound houses using grossly distorted yardsticks; nor can investors build sound portfolios that way.