Using Price Multiples Wisely

Our first stop in learning how to value stocks is traditional measures such as the price-to-sales (P/S) or P/E ratios. Although these measures do have some advantages—for example, they're very easy to compute and use—they also have some significant pitfalls that can lead the unwary investor to fuzzy conclusions.

The most basic ratio of all is the P/S ratio, which is the current price of the stock divided by sales per share. The nice thing about the P/S ratio is that sales are typically cleaner than reported earnings because companies that use accounting tricks usually seek to boost earnings. (Firms can use accounting tricks to boost sales, but it's much less frequent and it's easier to catch.) In addition, sales are not as volatile as earnings—one-time charges can depress earnings temporarily, and the bottom line of economically cyclical companies can vary significantly from year to year.

This relative smoothness of sales makes the P/S ratio useful for quickly valuing companies with highly variable earnings, by comparing the current P/S ratio with historical P/S ratios. Motorola, for example, takes special charges so often that they're no longer very special, and, as a result, it had negative net income in three of the five years between 1998 and 2OO2. With such spotty earnings, a P/E ratio isn't going to help us very much. But over that same time period, sales haven't jumped around nearly as much, which makes the P/S ratio useful. In mid-2003, Motorola's P/S ratio was about 1.0, close to a five-year low, which made the stock look relatively inexpensive compared with where it traded in the past (see Figure 9.1).

However, the P/S ratio has one big flaw: Sales may be worth a little or a lot, depending on a company's profitability. If a company is posting billions

Motorola

1998

1999

2000

2001

2002

Price/Earnings Price/Sales

NA 1.3

119.7 3.2

34.9

1.2

1.1

NA 0.7

In sales, but it's losing money on every transaction, we'd have a hard time pinning an appropriate P/S ratio on the shares because we have no idea what level (if any) profits the company will generate. We can see the drawbacks of using sales as a proxy for value in the marketplace every day.

Retailers, which typically have very low net margins—that is, they convert a relatively small percentage of every dollar of sales into profits—tend to have very low P/S ratios. The average grocery store, for example, had a P/S ratio of 0.4 in mid-2003, whereas the average medical device firm had a P/S of around 4.3. The reason for this huge difference isn't that grocery stores happened to be dirt cheap—it was that the average grocery store had a net margin of 2.5 percent, whereas the average medical device firm had a net margin of 11 percent. A grocer with a P/S of just 1.0 would look ridiculously overvalued, but a medical device manufacturer with the same P/S ratio would be an absolute steal.

Therefore, although the P/S ratio might be useful if you're looking at a firm with highly variable earnings—because you can compare today's P/S with a historical P/S ratio—it's not something you want to rely on very much. In particular, don't compare companies In different industries on a price-to- sales basis, unless the two industries have very similar levels of profitability.

Another common valuation measure is price-to-book (P/B), which compares a stock's market value with the book value (also known as shareholder's equity or net worth) on the company's most recent balance sheet. The idea here is that future earnings or cash flows are ephemeral, and all we can really count on is the net value of a firm's tangible assets in the here-and-now. Legendary value investor Benjamin Graham, one of Warren Buffett's mentors, was a big advocate of book value and P/B in valuing stocks.

Although P/B still has some utility today, the world has changed since Ben Graham's day. When the market was dominated by capital-intensive firms that owned factories, land, rail track, and inventory—all of which had some objective tangible worth—it made sense to value firms based on their accounting book value. After all, not only would those hard assets have value in a liquidation, but also they were the source of many firms' cash flow. But now, many companies are creating wealth through intangible assets such as processes, brand names, and databases, most of which are not directly included in book value.

For service firms, in particular, P/B has little meaning. If you used P/B to value eBay, for example, you wouldn't be according a shred of worth to the firm's dominant market position, which is the single biggest factor that has made the firm so successful. Price-to-book can also lead you astray for a manufacturing firm such as 3M, which derives much of its value from its brand name and innovative products, not from the size of its factories or the quantity of its inventory.

Another item to be wary of when using P/B to value stocks is goodwill, which can inflate book value to the point that even the most expensive firm looks like a value. When one company buys another, the difference between the target firm's tangible book value and the purchase price is called goodwill, and it's supposed to represent the value of all the intangible assets—smart employees, strong customer relationships, efficient internal processes—that made the target firm worth buying.

Unfortunately, goodwill often represents little else but the desperation of the acquiring firm to buy the target before someone else did, because acquiring firms often overpay for target companies. Be highly skeptical of firms for which goodwill makes up a sizable portion of their book value. The P/B may be low, but the bulk of the B could disappear in a hurry if the firm declares the goodwill as "impaired" (in other words, the firm admits that it grossly overpaid for a past acquisition) and writes down its value.

Price-to-book is also tied to return on equity (equal to net income divided by book value) in the same way that price-to-sales is tied to net margin (equal to net income divided by sales). Given two companies that are otherwise equal, the one with a higher ROE will have a higher P/B ratio. For example, over the past five years, Nokia's P/B ratio has averaged about 14, whereas rival mobile-phone giant Motorola's has averaged 3.1. One major reason for this difference is that Nokia's average ROE of 29 percent over the same period left Motorola's average ROE of 3 percent in the dust. The reason is clear—a firm that can compound book equity at a much higher rate is worth far more because book value will increase more quickly.

Therefore, when you're looking at P/B, make sure you relate it to ROE. A firm with a low P/B relative to its peers or to the market and a high ROE might be a potential bargain, but you'll want to do some digging before making that assessment based solely on the P/B.

A caveat: Although P/B isn't terribly useful for service firms, it's very good for valuing financial services firms because most financial firms have considerable liquid assets on their balance sheets. The nice thing about financial firms is that many of the assets included in their book value are marked-to-market— in other words, they're revalued every quarter to reflect shifts in the marketplace, which means that book value is reasonably current. (A factory or piece of land, by contrast, is recorded on the balance sheet at whatever value the firm paid for it, which is often very different from the asset's current value.)

As long as you make sure that the firm doesn't have a large number of bad loans on its books—see Chapter 17 for more on banks and bad loans—P/B can be a solid way to screen for undervalued financial firms. Just remember that financial firms trading below book value (a P/B lower than 1.0) are often experiencing some kind of trouble, so you'll want to investigate just how solid that book value is before stock investing.

Now "we come to the most popular valuation ratio, which can take you pretty far as long as you're aware of Its limitations. The nice thing about P/E Is that accounting earnings are a much better proxy for cash flow than sales, and they're more up-to-date than book value. Moreover, earnings per share results and estimates are easily available from just about any financial data source imaginable, so it's an easy ratio to calculate.

The easiest way to use a P/E ratio is to compare it to a benchmark, such as another company in the same industry, the entire market, or the same company at a different point in time. Each of these approaches has some value, as long as you know the limitations. A company that's trading at a lower P/E than its industry peers could be a good value, but remember that even firms in the same industry can have very different capital structures, risk levels, and growth rates, all of which affect the P/E ratio. All else equal, it makes sense to pay a higher P/E for a firm that's growing faster, has less debt, and has lower capital reinvestment needs.

You can also compare a stock's P/E to the average P/E of the entire market. However, the same limitations of industry comparisons apply to this process as well. The stock you're investigating might be growing faster (or slower) than the average stock, or it might be riskier (or less risky). In general, comparing a company's P/E "with industry peers or "with the market has some value, but these aren't approaches that you should rely on to make a final buy or sell decision.

However, comparing a stock's current P/E with its historical P/E ratios can be useful, especially for stable firms that haven't undergone major shifts In their business. If you see a solid company that's growing at roughly the same rate with roughly the same business prospects as in the past, but it's trading at a lower P/E than its long-term average, you should start getting interested. It's entirely possible that the company's risk level or business outlook has changed, in which case a lower P/E is "warranted, but it's also possible that the market is simply pricing the shares at an irrationally low level.

This method generally works better with more stable, established firms than with young companies "with more uncertain business prospects. Firms that are growing rapidly are changing a great deal from year to year, which means their current P/Es are less comparable to their historical P/Es.

Relative P/Es have one huge drawback: A P/E of 12, for example, is neither good nor bad in a vacuum. Using P/E ratios only on a relative basis means that your analysis can be skewed by the benchmark you're using.

So, let's try to look at the P/E ratio on an absolute level. What factors would cause a firm to deserve a higher P/E ratio? Because risk, growth, and capital needs are all fundamental determinants of a stock's P/E ratio, higher growth firms should have higher P/E ratios, higher risk firms should have lower P/E ratios, and firms with higher capital needs should have lower P/E ratios.

We can see why this is true intuitively, without breaking out any equations, by thinking about the basics of valuation—the three big factors that affect value are the amount, timing, and riskiness of a firm's future cash flows.

Firms that have to shovel in large amounts of capital to generate their earnings run the risk of needing to tap additional funding, either through debt (which increases the risk level of the company) or through additional equity offerings ("which may dilute the value of current shareholders' stake). Either way, it's rational to pay less for firms with high reinvestment needs because each dollar of earnings requires more of shareholders' capital to produce it.

Meanwhile, a firm that's expected to grow quickly will likely have a larger stream of future cash flows than one that's growing slowly, so all else equal, it's rational to pay more for the shares (thus, the higher P/E ratio). On the flip side, a firm that's riskier—maybe it has high debt, maybe it's highly cyclical, or maybe it's still developing its first product—has a good chance of having lower future cash flows than we originally expected, so it's rational to pay less for the stock.

When you're using the P/E ratio, remember that firms with an abundance of free cash flow are likely to have low reinvestment needs, which means that a reasonable P/E ratio will be somewhat higher than for a run-of-the-mill company. The same goes for firms with higher growth rates, as long as that growth isn't being generated using too much risk.

A few other things can distort a P/E ratio. Keep these questions in the back of your mind "when looking at P/E ratios, and you'll be less likely to misuse them.

Has the Firm Sold a Business or an Asset Recently? When you're looking at a P/E ratio, you must be sure that the E makes sense. If a firm has recently sold off a business or perhaps a stake In another firm, it's going to have an artificially inflated E, and thus a lower P/E. Because you don't want to value the firm based on one-time gains such as this, you need to strip out the proceeds from the sale before calculating the P/E. In late 2OOO, it looked as though Oracle had a ridiculously low P/E based on the past four quarters' earnings— until you dug into the numbers and saw that the company had booked a $y billion gain by selling part of its stake in Oracle Japan. Based on operating earnings, the stock wasn't all that cheap.

Has the Firm Taken a Big Charge Recently? If a firm is restructuring or closing down plants, earnings could be artificially depressed, which would push the P/E up. For valuation purposes, it's useful to add back the charge to get a sense of the firm's normalized P/E.

Is the Firm Cyclical? Firms that go through boom and bust cycles—semiconductor companies and auto manufacturers are good examples—require a bit more care. Although you'd typically think of a firm with a very low trailing P/E as cheap, this is precisely the wrong time to buy a cyclical firm because it means earnings have been very high in the recent past, which in turn means they're likely to fall off soon. For cyclical stocks, your best bet is to look at the most recent cyclical peak, make a j udgment whether the next peak is likely to be lower or higher than the last one, and calculate a P/E based on the current price relative to what you think earnings per share will be at the next peak.

Does the Firm Capitalize or Expense Its Cash Flow-Generating Assets? A firm that makes money by building factories and making products gets to spread the expense of those factories over many years by depreciating them bit by bit. On the other hand, a firm that makes money by inventing new products—drug firms are the classic example—has to expense all of its spending on research and development every year. Arguably, it's that spending on R&D that's really creating value for shareholders. Therefore, the firm that expenses assets will have lower earnings—and thus a higher P/E—in any given year than a firm that capitalizes assets.

Is the E Real or Imagined? There are two kinds of P/Es—a trailing PIE, which uses the past four quarters' worth of earnings to calculate the ratio, and a forwardPIE, which uses analysts' estimates of next year's earnings to calculate the ratio. Because most companies are increasing earnings from year to year, the forward P/E is almost always lower than the trailing P/E, sometimes markedly so for firms that are increasing earnings at a very rapid clip. Unfortunately, estimates of future earnings by Wall Street analysts—the consensus numbers that you often read about—are consistently too optimistic. As a result, buying a stock because its forward P/E is low means counting on that future E to materialize in its entirety, and that's usually not the case.

The PEG is an offshoot of the P/E ratio that's calculated by dividing a company's P/E by its growth rate. The PEG is extremely popular with some investors because it seeks to relate the P/E to a piece of fundamental information—a company's growth rate. On the surface, this makes sense because a firm that's growing faster will be worth more in the future, all else equal.

The problem is that risk and growth often go hand in glove—fast-growing firms tend to be riskier than average. This conflation of risk and growth is why the PEG is so frequently misused. When you use a PEG ratio, you're assuming that all growth is equal, generated with the same amount of capital and the same amount of risk.

But firms that are able to generate growth with less capital should be more valuable, as should firms that take less risk. If you look at a stock that's expected to grow at 15 percent trading at 15 times earnings and another one that's expected to grow at 15 percent trading at 25 times earnings, don't just plunk your money down on the one with the lower PEG ratio. Look at the capital that needs to be invested to generate the expected growth, as well as the likelihood that those expectations will actually materialize, and you may very well wind up making a very different decision.