A Visit to the Valuation Sanitarium

Most people will agree that this is a wacky market. Internet stocks appear to be held up by antigravity devices, leading many of us to continue to question the market's sanity.

Meanwhile, existing market metrics and ratios that we like to use to gauge if a stock is expensive or cheap have gone by the wayside: Neither price-to-cash nor price-to-book value ratios help us to put a value on information companies, as the intellectual property is not on the balance sheet. That's why growth investors can't look at balance sheets, the province of value investors; instead they must look at price-to-earnings, or P/E, ratios to gauge valuation.

Fair enough, but what happens when there ain't no "E," creating P/E ratios that approach infinity? In these cases, we must turn to price-to-sales ratios as the sole metric available to us. But for a hot young Internet company, even these can be in the stratosphere, with some companies priced at 10 times sales, 25 times sales, even hundreds of times their annual sales. As Crazy Eddie said, "These prices are insane!"

Or are they? Upon closer examination, there appears to be a sweet spot in the market where investors seem comfortable valuing companies that sport revenue growth of plus or minus 100% annually at multiples of between 10 and 25 times annualized sales.

It all relates to growth, and the market's belief in a company's ability to sustain that growth over time. A stock is nothing more than the cumulative future earnings power (actually, the dividend-paying capability) of a company, discounted back to today. The P/E ratio is a quantitative metric that is determined by the relationship between a company's long-term growth rate, and a hypothetical discount rate. In periods of high inflation, the discount rate of future earnings is high, so P/Es are low. In periods of benign inflation, like today, the discount rate is much lower, and investors will pay P/E ratios that may equal or exceed a company's growth rate.

Growth vs. Hypergrowth

In the old days (that being 18 to 36 months ago!), a company was considered "high growth" if it could sustain 40% annual revenue growth. This was considered the fastest rate of growth that a management could execute in a bricks-and-mortar business world. And for growth of this magnitude, the stock market paid a premium, say a P/E range of 30 to 50 times earnings.

Since then, new, more scaleable business models have emerged, ones that don't require building up either physical infrastructure or a direct sales force in proportion to revenue. Instead they merely require adding products, R&D staff and channel partners. The market now accepts that growth of 100% can be sustained for several years in the early stages of a truly hot Web franchise.

These new businesses mean that the sweet spot for growth in the stock market has now soared to between 80% and 120%. Companies in that range are being awarded valuations ranging between 10 and 25 times sales, depending on their revenue growth and margin assumptions.

Margins are still discretionary (meaning a company is spending a lot of money on marketing or building an infrastructure that is underutilized), and expenses are nowhere close to "normal" yet. But if a company continues to deliver on this high growth, and the valuation manages to stays constant (two big ifs, I'll admit, but no one said this was easy), the investor might be rewarded with annual returns of 80% to 120%.

But the advent of these hypergrowth companies means that expectations have ratcheted upward accordingly: Anything growing slower than 80% is considered dull and at risk that the world will pass them by. To put it in perspective: Today, "normal" growth is considered a whopping 15%, while Ford (F) is growing at a mere 4%. On the flip side, anything growing much faster than 120% is considered unsustainable and therefore suspect.

So, am I saying that a company at 10 times sales is cheap? In today's market environment and with today's discount rate, yes, but it can change quickly. Here's how you figure out whether a hypergrowth company is overpriced, underpriced or priced just right.

The New Investing Math

Let's do some (easy) math. If you take a stock's price-to-sales ratio and divide it by the company's after-tax profit margin (earnings to sales), you get the price-to-earnings ratio. If you assume that in today's market, a normal P/E is one times the growth rate (a big assumption, but not a bad one), then you can substitute the growth rate for the P/E. If you know, or can estimate, any two of the above, you can figure out what value the market is placing on the third.

Many Internet companies are losing money, so margins are negative, but in many cases the losses are discretionary, so they are nowhere near their "model earnings." But you can make some assumptions, based on the business model, of just what kind of margins the company will eventually have -- and that is what I believe the market is doing today.

Here are some interesting examples: Broadcom (BRCM), whose revenue is running at about $400 million annually, sports a $10 billion market cap and thus is valued at 25 times sales. Long term, the company should be able to make 20% after taxes, so its implied P/E is 125 (25 divided by 0.20) and therefore its implied growth rate is 125%. Because the company is currently growing faster than that, it can be considered reasonably priced.

If you assume Ask Jeeves (ASKJ) can do a $5 million quarter sometime this year, then it would be selling at 60 times its run-rate sales of $20 million. But the math tells us that even if it can make 33% after-tax margins, it had better be able to show sustainable 200% growth. Ask Jeeves has yet to report its second-quarter numbers, but with its first-quarter revenue totaling only $1.1 million, its valuation already appears out of whack. I say that this already looks rich.

Taking another case, Amazon.com (AMZN) doesn't look so stretched when you consider that its $21 billion market cap divided by its $1.2 billion run-rate revenue provides a valuation of 17 times sales. That's well within the comfort zone of 10 to 25 times. But margins make all the difference. At best, Amazon is looking at a business model producing 5% after tax. Mathematically, that makes its implied P/E, and therefore estimated growth rate, well north of 300%, outside the realm of "sustainable" growth and faster than its current revenue run rate.

If you're smart, you can find a company with a valuation going from 10 times sales to 25 times sales from higher margin assumptions, faster growth, etc. But be warned, if bad things happen, watch out below. A lower growth-rate assumption lowers the return outlook substantially. More importantly, the valuation can quickly drop from 10 times sales to as little as two times sales.

Experienced technology investors know all too well about this end of the equation: Recall what happened when N2K stopped growing? The shares went from 33 to 5 when quarterly revenue flattened at $10 million. Its valuation at the time dropped to around two times sales. But then its growth rebounded and the shares bounced back to 16, or about five times sales, before N2K merged with CDNow (CDNW). Now that's what I call whiplash.

>To order reprints of this article, click here: Reprints