The price-to-earnings (P/E) ratio can be a lousy indicator of value

This benchmark can be deceptive. Using a more accurate measure of stock values will help you avoid common investment mistakes.

Issue: August 1, 2001

ByJeffrey B. Osher

Investors never want to buy an overvalued stock, nor do they want to sell an undervalued stock that has upside potential. Perhaps the single most widely used factor in assessing whether a company is cheap or pricey is the price-to-earnings ratio, or P/E. However, there are extreme risks and value traps associated with investing in a company solely on its P/E ratio. Many of these mistakes are made because of the pervasive acceptance of the P/E ratio as a benchmark for value. To avoid this trap, it is essential to understand what exactly a P/E reflects and why it’s a rearview indicator that often misguides investors.

P/E ratio defined

A P/E ratio is simply the latest closing price of a stock divided by the last four quarters of earnings per share. So a company with $2.00 of earnings per share over the last year that closed at $50.00 per share has a P/E of 25. Intuitively, the P/E of a stock represents what an investor is paying for $1.00 of earnings per share. So in the above example, an investor is paying $25.00 for every $1.00 of earnings. In the past, it was rare to see a tremendous variation in earnings, so analyzing the P/E gave a snapshot of whether a company was expensive relative to its sector or the market. We can’t stress enough that a low P/E does not always represent value and a high P/E does not always mean a stock is expensive. These are misconceptions that could lead to very costly blunders.

P/E trap

When discussing a new stock acquisition with one of our clients, we often hear the question, “Is the stock cheap, what is its P/E?” This train of thought relegates the future earnings of a company in importance to its trailing earnings.

If we look at two semiconductor companies, the P/E trap becomes clearer. At the close of trading on June 6, 2001, Intel (Nasdaq: INTC) was $29.82 per share and had trailing earnings of $1.45 per share, resulting in a P/E of 21. International Rectifier (NYSE: IRF), another chip company, closed at $62.01 per share with $2.59 of earnings per share, which resulted in a higher P/E of 24. Looking at the price or the P/E of International Rectifier, it would seem that buying Intel would be a more value-oriented strategy.

However, if an investor looks forward to the earnings expectations for the upcoming year, it becomes clear that International Rectifier represents a much better value with superior growth. For the next 12 months, Intel is expected to earn only $0.56, an earnings contraction of 61%. On the other hand, International Rectifier is expected to see its earnings fall slightly from $2.59 to $2.36 per share, a decrease of only 9%. Both companies are wonderful semiconductor companies suffering through the cyclical trough in this current cycle. Nonetheless, on next year’s earnings, an investor is buying International Rectifier for a P/E of 26, and Intel for a P/E of 53. Suddenly Intel does not look like such a good bargain despite its relatively low P/E ratio.

The above example illustrates Wall Street’s anticipatory ability to price in future events. Rarely over the past 10 years has Intel traded at such a low P/E on its trailing earnings, and an investor looking at its price per share and the 65% decline from the highs, or specifically the P/E, would think he or she was stealing Intel. Intel’s P/E is so low today because it will become extremely expensive as the lower earnings are reported over the following months. Often, P/Es head lower before the earnings actually head down. This is often called a P/E contraction.

Generally, P/E contractions occur when Wall Street anticipates earnings declines and consequently drives a stock price lower in anticipation of those earnings, hence the lower numerator in the P/E. The opposite is a P/E expansion, where a stock will head higher in anticipation of stronger earnings, which on the surface can make a stock look expensive.

Surge in earnings

Another example of this fallacy in valuing companies on P/Es is with the company that will experience a surge in earnings over the coming year(s). This surge could result from a new product, an improving regulatory environment, a major contract win or a cyclical upturn, which is what Intel and most of the chip sector will likely see in 2002. Many times, a company has a high P/E relative to the market and may seem expensive on the surface, but actually represents good value. Looking at two pharmaceutical companies, Barr Labs (NYSE: BRL) and Eli Lilly (NYSE: LLY), it becomes clear how high P/Es can be cheaper than similar companies with lower P/Es, making for more prudent investments.

Barr closed on June 6 at $75.45 per share. Wall Street’s consensus estimate for earnings per share in 2001 is $1.56, ballooning to $3.51 in 2002. These numbers give Barr a lofty P/E of 48 for 2001, but a very reasonable P/E for 2002 of 21, significantly below the pharmaceutical sector’s estimated 2002 P/E.

Lilly should see its earnings expand modestly from 2001 to 2002, with earnings going from $2.82 to $2.99 per share. With Lilly’s closing price of $86.99, their P/E only drops from 31 to 29. However, on 2001 numbers, it would appear that Lilly is the better buy. This assumption would not take into account the future growth that is expected from Barr’s generic Prozac. Though we believe the generic drug space will be a very profitable sector, and one where investors should have exposure, we are not making a specific recommendation on either company. Instead, we are using this example to demonstrate how disparate earnings growth can lead to the P/E trap.

More accurate measure

We believe a ratio that is a better snapshot of a company’s “value” is its P/E-to-growth ratio, or P/E/G. This number is found by dividing the P/E by the expected average annual earnings per share growth over a certain period of time, typically 3 to 5 years. By accounting for earnings growth, investors can get a better gauge on what they are paying today for future earnings, which matter much more than past earnings. The lower the P/E/G, the better value the company presents. It also allows investors to examine the relative value of two companies in completely different businesses.

Looking at several companies provides good insight into the usefulness of the P/E/G ratio versus the traditional P/E ratio as a means for determining value. Before looking at an example, it is important to understand that there are a myriad of factors to be considered and researched to ensure that the future growth numbers are attainable. There are also several sectors where P/E/G is less useful, such as banking, real estate and energy, where net assets are more indicative of value. Overall, for a value benchmark, the P/E/G provides better insight than the traditional P/E.

P/E/G example

Coca Cola (NYSE: KO) is a dominant company with a wonderful franchise. We would assert that it is not a wonderful investment. When comparing it to Calpine (NYSE: CPN), an independent power producer, the flaws of comparing P/Es instead of P/E/Gs becomes apparent. Coke closed on June 6, 2001 at $46.75 per share. Its 2000 earnings were $1.45 per share, giving it a P/E of 29. The 20 Wall Street analysts that cover Coke are projecting a long-term growth rate of 13%, which gives Coke a P/E/G of 2.20. The lack of growth is reflected in the earnings estimates for 2001 and 2002, which are $1.61 per share and $1.84 per share, respectively. The mediocre growth is highlighted when comparing Coke’s numbers to Calpine’s, but is magnified by its P/E/G.

Calpine closed on June 6, 2001 at $43.73 per share with 2000 earnings of $1.05 per share, giving Calpine a P/E of 34, slightly higher than Coke’s 29. Twenty analysts also cover Calpine and are forecasting an average growth rate of 34% over the next 5 years, resulting in a P/E/G of only 0.69, drastically below that of Coke’s. Earnings are expected to climb to $1.81 per share in 2001 and to $2.34 per share in 2002. Should both company’s earnings come in as forecasted, Calpine will be a much better investment in terms of value despite its higher current P/E. With a P/E of 34, Calpine would seem “more expensive” than the market when compared to the 29 P/E of the S&P 500. Using the more detailed ratio, the P/E/G, the conclusion can be drawn that Calpine is actually trading at a discount to the S&P 500, or the market’s P/E/G of 1.5.

Focus on P/E/G

Most investors are taught that the P/E ratio is the best way to judge how rich a stock is. We think it is very important to see through this erroneous assumption and focus more on the P/E/G ratio when determining a stock’s value. P/Es fluctuate, and many times a P/E is low because earnings will likely deteriorate in the future, and conversely there are a plethora of companies with high P/Es that will see earnings surge in the future, bringing their future P/Es down dramatically. As investors get pickier about what they are willing to pay for stocks, we believe, for strong returns, it is indispensable to scrutinize a company’s P/E/G, not its trailing P/E.