The Problem with P/E Multiples

We often get correspondence from our subscribers that discusses stocks in terms of a price/earnings (P/E) multiple--especially our higher-rated stocks that trade at high multiples. The letters usually go something like this: "I see you have a 5-star rating on stock XYZ. How can this company be a buy at a P/E multiple of 40 when the market trades at a P/E multiple of 20?"

The reality is that the P/E multiple is just a shortcut method for valuation, and a pretty bad shortcut at that. There are quite a few problems with P/E-based valuations. They include the following.

• Trailing vs. forward P/E confusion: Not all P/E multiples are created equal, and one has to be careful to compare apples to apples. Trailing P/E is calculated based on the trailing 12-month earnings, while forward P/E is based on notoriously optimistic Wall Street estimates. And they are just that--estimates.
• Short-term errors in earnings: Earnings include a whole lot of accounting estimation and opinion, allowing for manipulation or just plain error. Over short time horizons, temporary effects on earnings can generate a large error in trailing P/E. As an extreme example, Global Crossing  recently traded at a P/E of 0.029 based on 2003 earnings because of $24.9 billion of "earnings" resulting from a write-down of liabilities.
• Earnings volatility: Earnings of many companies are volatile, rising and falling dramatically with the economy or other factors. This makes the denominator of the multiple volatile, producing high P/Es during times of depressed earnings.
• Debt: A company with more debt in its capital structure will generally have higher returns on equity, but the earnings volatility will be greater. All else being equal, volatile earnings should be worth less, so a more leveraged company should trade at a lower P/E.
• Growth--A more quickly growing company will typically be valued at a higher P/E multiple because rapidly growing cash flow is worth more than slowly growing cash flow. Some investors try to compensate for this by using a PEG ratio (P/E divided by growth rate) and use a PEG of 1 as a benchmark. But even with this adjustment, the PEG ratio isn't a good basis for comparison across companies because it doesn't include information about how long the company will be growing at that high rate.

Let's focus on the problems of using a P/E on a rapidly growing company. A great historical case where a P/E-based valuation would have led you astray is  Home Depot (HD). The table below contains some of the metrics we've been discussing for the period 1990 to 2003. We calculated both the cumulative return had an investor held the stock to 2003, as well as the return over a five-year holding period.

Home Depot Metrics
Year	Price* ( $ )	Earnings ( $ Mil )	Earnings Growth ( % )	P/E Ratio	PEG Ratio	Return to 2003 ( % )	5-Year Return ( % )
1990	2.88	163	46	31	0.7	21	30
1991	7.48	246	51	60	1.2	14	8
1992	11.2	363	47	66	1.4	11	12
1993	8.78	457	26	40	1.5	15	36
1994	10.2	605	32	37	1.1	15	46
1995	10.6	732	21	32	1.5	16	34
1996	11.1	938	28	28	1.0	18	36
1997	19.6	1,160	24	38	1.6	10	4
1998	40.7	1,614	39	62	1.6	-3	-3
1999	68.6	2,320	44	73	1.7	-15	N/A
2000	45.7	2,581	11	39	3.5	-8	N/A
2001	51.0	3,044	18	43	2.4	-17	N/A
2002	24.0	3,664	20	15	0.8	48	N/A
2003	35.5	N/A	N/A	N/A	N/A	N/A	N/A
* Data as of year-end date each year.

In 1990, Home Depot was a bargain at a trailing P/E of 31. If you'd bought Home Depot in 1990 and held it to the end of 2003, your average annual return would have been 21% over that time period. In fact, in 1991, it was approximately fairly priced at a trailing P/E of 60, with a return to the end of 2003 of 13% per year.

One great example of a stock with similar characteristics today is  CarMax (KMX). CarMax is trading at a trailing P/E multiple of about 30 times earnings--and we think it is a bargain at this price. Because CarMax is a great business and will continue to increase earnings for many years, our discounted cash-flow model values the company at $49, or a trailing P/E of about 47. The price of the company on Mr. Market's voting machine may not rise today, next week, or next month. But we think the success of the company will drive up the earnings and cash flows for many years to come, and eventually all that cash will start to pile up on Mr. Market's weighing machine. When Mr. Market looks at the scale, his greed will take over and the stock price will reflect the value of all of that cash.

We ran a screen in the Morningstar.com  Premium Stock Screener to search for other stocks that we think are good values despite high P/E multiples. We selected companies with a trailing P/E above 30, a five-year average P/E above 30, and a current Morningstar rating of 4 or 5 stars. The examples below are other stocks like CarMax--stocks we like for their long-term growth potential.

 Apollo Group 
Morningstar Rating: 5 Stars
Economic Moat: Wide
Business Risk: Average
From the  Analyst Report: "Online growth has been particularly strong and has provided Apollo with piles of cash. Apollo's accreditation and access to government financial aid programs contribute to its wide moat."

 Microsoft  (MSFT)
Morningstar Rating: 4 Stars
Economic Moat: Wide 
Business Risk: Below Average
From the  Analyst Report: "Microsoft can still generate growth in the mid- to low teens, and as it is still a cash-generating machine, it can afford to reinvest in the business while still giving cash back to shareholders."

 Paychex  (PAYX)
Morningstar Rating: 5 Stars
Economic Moat: Wide
Business Risk: Below Average
From the  Analyst Report: "We think Paychex's strategy of expanding its payroll-processing client base and cross-selling ancillary services to customers will retain its potency."

 Southwest Airlines  (LUV)
Morningstar Rating: 4 Stars
Economic Moat: Narrow
Business Risk: Average
From the  Analyst Report: "We assume compound annual revenue growth in the low teens through 2008. We also expect operating margins to return to double digits by 2005, with the help of even mild relief from fuel prices."

Click  here to run this screen yourself. Note: The stocks mentioned above passed our screen as of Dec. 23, 2004. The results of the screen may change due to daily price fluctuations or other factors. After clicking, you can save the search to use later by clicking the "Save Criteria" button in the bottom right-hand corner of the screen. (You will need to be logged in as a Premium Member to view and save the complete screen.)

A version of this article appeared March 15, 2004.