l'union fait la force
Buffett on Technology

Format for Printing

Format for printing

Request Reprints


By stocksure
May 8, 2001

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!


I think the point is that companies like Coke and Gillette have pretty stable earnings streams and growth trends, thus the relative certainty of future earnings warrants higher multiples. There is far more uncertainty with tech companies, and thus the risk to principle is higher. In a way, it is irrational for people to reward that uncertainty with higher multiples.

As much as my prior posts and articles (not to mention my profile, the magazines on my desk, and, well, just about everything else) provide ample evidence suggesting the contrary, I've become increasingly partial, as of late, to Buffett's (and apparently Gates') view on the proper value of certain types of technology companies. It all comes back to Buffett's thought-provoking claim that (pardon my over-reaching attempts to be gender-neutral) an investor should only put his/her hard-earned money into a given company, at a given price, if he/she would be willing to buy said company in its entirety at that price, should the necessary funds be available to him/her. Looking at the tech universe from this particular angle can be an eye-opening experience, one that may leave many investors forever unable to look at the long-term prospects of a number of their favorite companies with the same degree of confidence that they once afforded them.

Take Juniper, as an example. I've been a fan of this company for quite a while, since a point in time when its IP core router market share was a fraction of what it is now. And I'm confident that, for the next couple of years, Juniper will continue to rapidly grow its share of this nascent, rapidly growing market, and, by the time that all's said and done, may even end up usurping Cisco's leadership position in it. But what about Juniper's status eight years from now? Or even five years from now, for that matter? For such extensive durations, I don't think even Scott Kriens himself could predict, with any degree of certainty, how Juniper will hold up. It's quite possible that, by 2005, Avici, HyperChip, Procket (Tony Li's company), or some yet-unformed startup set to be created by a caffeine-addicted insomniac of a Cal-Tech student will have done to Juniper what Juniper did to Cisco. Perhaps, by 2008, not only will this have happened, but also some other startup will have done to one of these companies what they did to Juniper. Or perhaps not.

The point is, none of us know; yet you have to know, in order to be comfortable with investing in Juniper at its current price. The company currently trades at 84x its trailing pro forma earnings, and has a book value of $2.63/share. As confident as one might be about how Juniper will fare during the next couple of years, it has to be quite obvious to even the most starry-eyed optimist (OK, maybe not George Gilder, but he really doesn't count) that Juniper isn't going to justify its present valuation merely via the profits that it's set to generate between now and 2004 or 2005. If an investor is clueless about how well Juniper might be doing in 2008, yet is depending on the company's 2008 earnings to help justify the price that he/she paid for the company's shares, he/she isn't engaging in investing, but rather speculation.

There are plenty of other examples scattered throughout this awe-inspiring yet oft-treacherous universe, of course. For example, given the breakneck pace at which companies are leapfrogging each other in the DWDM transport equipment market, how long can one count on Ciena being a leader in this field? Or, how confident can an investor in Sun be about the permanence of the company's present stature within the high-end server market? And how much of a guarantee does someone who has put the equivalent of a week's paycheck into Checkpoint have that the firm will still be a major player in the firewall software market in 2007? When it comes to such companies, I definitely don't blame a (relatively) conservative investor such as Buffett for not showing much interest.

I know that, by now, I've scared the hell out of plenty of people who have read my prior work, and are now wondering if I've officially migrated over to the dark side of the NASDAQ, destined to join Warren and his friends in investing solely in insurance companies, soft-drink manufacturers, and other such long-standing staples of the global consumer culture. Well, not exactly. As I've already said, I agree with the commentary provided by Buffett and Gates as it applies to certain types of technology companies. I do see plenty of exceptions to the rule, so many that, in the cases of industry leaders, they may be the rule rather than the exception. Generally, they break down into one of three categories:

1. Tech companies that really aren't tech companies. These are firms whose success primarily depends on things like brand name, economies of scale, and consumer preferences, and thus shouldn't be adversely affected by any technological trends that may come their way, but are unfairly classified as technology companies due to their utilization of products developed by companies that are. AOL and Yahoo! are examples, as are Dell's PC and laptop businesses. It doesn't matter if Cisco develops data center switches superior to the ones Yahoo! currently buys from Alteon (Nortel), or if AMD one-ups Dell's sole CPU supplier, Intel, in the microprocessor wars - the companies can switch vendors without compromising their competitive advantages one iota.

2. Tech companies that have proprietary locks on the markets that they serve. Let's assume for a moment that the Justice Department's jihad against Microsoft proves successful, and that the company's operating systems and applications operations are split up into separate, publicly traded businesses. Somehow, given that the overwhelming majority of the OS division's revenues are bound to come from the PC and laptop markets, I wouldn't be too nervous about making future earnings/cash flow projections for the next ten years, even if the business does require billions in R&D outlays, and involves product life cycles of roughly 2-3 years. The reason for this, as I'm sure many here would be quick to point out, is that, as long as PCs and laptops continue to be sold, Microsoft will continue make a mint in OS sales, given the company's proprietary control over the source code of the de facto operating system standard for the PC and laptop industries.

Meanwhile, patents can work just as well as source code, with Rambus and Qualcomm serving as two excellent examples. In each case, the company possesses an intellectual property portfolio whose utilization happens to be essential for technologies whose preeminence within a given market is unlikely to be challenged for the next ten years or so. Of course, after that, contenders may very well emerge, with significant market share deterioration possible a couple of years later; and by year fifteen the strength of the current patent positions of the two companies, with regards to the previously mentioned technologies, will begin to erode. However, if you happen to conduct DCF analyses of Rambus and Qualcomm, utilizing a 12% initial discount rate and a 15% terminal rate, there's a very good chance that over 80% of the value one expects the company to create will be generated by year 12. And, as those who follow these two companies know, neither of them happens to be standing still with regards to the development of new technologies for the markets that they currently target. Granted, counting on Qualcomm to be behind some next-generation mobile communications standard that hasn't been invented yet, and will begin to be implemented around 2010, requires just as much blind faith as counting on Juniper to have the most popular router on the block in 2008; but with such faith only necessary after having reached a point where one's intrinsic value requirements for a given company are all but satisfied, it's not quite so intolerable.

3. Tech companies that offer a degree of security due to the sheer breadth of their product line for a given market. Unlike Juniper, Checkpoint, or Sun, which depend on just a handful of different products to drive the lion's share of their respective revenue bases, companies like Applied Micro Circuits and International Rectifier rely on, literally, scores of them, spread out over dozens of different sub-sections of multi-billion dollar markets. Thus, even if some run-of-the-mill garage startup develops an OC-192 framer that's superior to Applied Micro's, or if another one comes up with a more cost-effective IGBT for controlling the fuel injection systems of diesel engines, the companies will still be able to flourish without much trouble, as long as the markets they serve continue to see rapid growth.

So, in short, I would say that I don't find what Buffett and Gates have to say about the investment potential of technology companies -- or rather, the construction of their comments by others -- to be completely disagreeable. My disagreements, where they exist, relate to the extent to which the sentiments that they express are often applied, whether or not this extent happens to represent a proper comprehension of the aforementioned comments. To state that certain types of tech companies may not be suitable investments due to the difficulties involved in estimating their long-term future cash flows appears to be poignant commentary. To state that, relative to the Cokes and Gillettes of the world, it's too difficult to estimate the future cash flows of all companies that are usually classified as belonging in the tech sector, appears to be an excessive generalization.