« May 1998 | Main | October 1998 »

August 17, 1998

Internet Investors: Beware of the Proxy Valuation

“If you can’t be with the one you love, honey, love the one you’re with.”
     -Stephen Stills, Love the One You’re With

As the market begins to grapple with the lingering effects of a seven year bull market, the financial crisis in Asia, and the undeniable shortfall in second quarter earnings, the ground underneath our feet just doesn’t seem as strong as it once did. For those of you investing in Internet stocks, gauging the solidity of the earth below is even more difficult given the daunting distance from the bottom of your shoes to the actual surface of the earth. Perhaps everything is fine…At presstime, the Internet stocks are in fact acting quite resilient. However, if recent market skittishness has caused you to reassess, make sure that you’re not overly exposed to the risks of the “proxy” valuation.

Any financial academician will tell you that the only proper way to value a stock is to predict the long-term cash flows of a company, discount those cash flows back to the future, and then divide by the number of shares. As this is much easier said than done, many practitioners often shortcut the process using tools that serve as a “good enough” proxy for cash flow. One good example is the price to earnings ratio. It’s not a perfect measure of cash flow, and there are many loopholes between the two, but as John Maynard Keynes said “I would rather be vaguely right, than precisely wrong.”

In certain emerging markets, particularly ones that are capital intensive, the cost of market entry is so high that even market leaders lose money for several years before eventually turning profitable. This presents a bit of a dilemma for the typical market investor as his/her standard proxy tools are irrelevant due to the fact that the variables used for computing valuation -- earnings, earnings growth, cash flow per share – may all be nonexistent. Rather than throw in the towel, innovative investors must turn to new proxies based upon variables that are indeed measurable. Simply make some assumptions that tie the proxy back to standard valuation tools and you are on your way.

The cable television and cellular telephone industries owe a great deal to the proxy valuation. The cash flow required to build these communication infrastructures was so high that most of the market players performed quite poorly when evaluated using standard valuation tools. However, many optimistic investors “knew” that these companies would eventually reach economies of scale that would then lead to profitability. Therefore, with no earnings to measure, these investors grabbed hold of any variable they could.

In the cable television era, the variable that was most commonly used was “homes passed”. Divide the market value of the average cable company by the average number of homes who could potentially subscribe to the service to calculate the value per home-passed. This proved to be a useful tool for valuing cable franchises, and you could always rationalize the value by calculating a rough estimate of what the lifetime value of a single customer should actually be. Other variables that proved popular in both cable and cellular were “price-per-sub (subscriber)” and Earnings before Interest, Taxes, Depreciation, and Amortization – also known as EBITDA (more on this later).

The Internet is currently going through a similar stage. Investors, who are rightfully optimistic about the future and potential of the Internet, are anxious to invest in companies who are clearly many years away from profitability (perhaps many, many years). Not to be shut-out for lack of a valuation tool, these investors have also created proxies that are tied to the variables that we happen to be able to measure. Some popular ones include market capitalization per subscriber, market cap-to-subscriber, market cap-per-unique visitor, market cap-to-page view, and the most popular of all -- market cap-to-revenues.

While abstract proxy valuation tools are indeed risky, there is no reason to belittle those that use them. Keep in mind that while these tools are inaccurate, they still offer a distinct advantage over the alternative of holding one’s finger in the air. Additionally, the advanced risk that is apparent in these situations is typically offset by a greater potential upside. Columbus would have never discovered America if he had waited around for someone to invent the Global Positioning Satellite. Many investors in both the cable and cellular industries were handsomely rewarded for betting big and betting early, and the same reality has already been proven true on the Internet.

However, as emerging markets begin to mature, the dangers of proxy valuation become more apparent. These proxy risks can take several forms. First consider symmetry risk which flows from the fact that not all “variables” are created equal. When you base your valuation on cash flows, it is safe to assume that a dollar of cash at one company is equal to a dollar of cash at another. However, as we move down the proxy continuum to earnings, and subscribers, and visitors, and page views, the likelihood that we are still comparing apples to apples diminishes. Is an E-Trade customer worth the same as a CDNow customer? Does a page view on Yahoo equal a page view on GeoCities? It’s hard to say.

A second risk is assumption risk. After proxy valuations begin to stick, we begin to treat them as fact. For instance, we might divide companies’ market capitalization by its customer base and decide that a customer is worth $X. However, we should really take a look at the cash flows that are likely to come from that customer and then determine if the overall valuation makes sense. Becoming too comfortable with the proxy and its assumptions can be dangerous. For many years, cable investors were convinced that EBITDA was a great proxy valuation and that cable companies should be worth as much as 15-20 times EBITDA. Unfortunately, it turned out that ignoring depreciation was a major mistake, since cable systems needd constant upgrades. The proxy was poor, and those that came to believe in it eventually suffered.

Another reason to be skeptical of proxy valuations is arbitrage. If Wall Street comes to believe that customers, or visitors, or page views, or even revenues are uniquely valuable (regardless of profitability), than entrepreneurs are likely to rush to market with companies that can achieve these targets quite handily, but may have little chance at producing real value in terms of cash flow. With no focus on costs, it is easy to reach non-financial targets. This is the great thing about cash flow-based valuation, it’s hard to sweep costs under the rug.

To highlight this point, consider the absurd example of a Web-based company whose core service is to sell dollars for $0.85 each. This company could obviously achieve record visitors and page views at its Web Site. Revenue growth would easily set records, and it is quite conceivable that sales could reach into the billions within the first few quarters of operation. Apply even the most conservative Internet price-to-revenue multiple to this franchise and we are talking about a multi-billion dollar market cap. Perhaps you are questioning how this high-flying company will ever turn profitable? You are obviously forgetting that with traffic like this, the potential for advertising and targeting will be tremendous!

Clearly, there are a few Internet companies that do not deserve this type of ridicule. Yahoo and E-Trade have already proven profitability, and many investors are quite confident that Amazon can achieve similar results over time. Additionally, even younger companies such as Ebay, a San Jose auction house, are achieving record revenue growth without compromising profitability. However, these are the exceptions. Investors should increasingly evaluate the potential long-term profitability of Internet businesses before they invest. If the market continues to slip, optimism will be replaced with skepticism, and the companies that are most dependent on proxies will be the ones that fall furthest.

On the margin, it certainly appears that arbitrage is in full swing. We are witnessing an increase in companies that file for IPO’s with losses that are as large as (if not larger than) revenues. Recognize that this means that expenses are over twice that of revenues—economics that are actually worse than my company that sells dollars at a discount. Additionally, it is rumored that certain Internet CFOs are pushing investor’s to look at EBITDAM. The M represents marketing, and is an attempt to get Wall Street to ignore what has become the single biggest expenditure for Internet startups. This only makes sense if you truly believe that marketing costs will one day go away, which should be considered unlikely. Perhaps we should make it easier and skip straight to EBE (Earnings Before Expenses)… but that looks suspiciously similar to price-to-revenues, doesn’t it?

Posted by Bill Gurley on August 17, 1998 at 08:00 AM | Permalink | Comments (0) | TrackBack

DISCLOSURE: The information contained Above the Crowd has been obtained from sources believed to be reliable but is not necessarily complete, and its accuracy cannot be guaranteed. Any opinions expressed herein are subject to change without notice. The author is a general partner of Benchmark Capital, a venture capital firm in Menlo Park, Calif. Benchmark Capital and its affiliated companies and/or individuals may have economic interests in the companies discussed herein. © J. William Gurley 2005-2006. All rights reserved.