Thursday, January 25, 1996
MARKET CLOSE

INDEX:

I. Market News: Government Debt Worries Stop Market Rally
II. Heroes: Activision, CompUSA, RMI Titanium, Forefront Group, Electroglas
III. Goats: Electronic Arts, Loewen Group, NetFRAME, MySoftware
IV. Investment Perspective: The Digital World---Part Three
V. Calendar: Friday's Economic Events

MARKET CLOSE

DJIA: 5216.83 -26.01 (-0.50%)
S&P 500: 617.03 -2.93 (-0.47%)
NASDAQ: 1035.95 -7.51 (-0.72%)

MARKET NEWS

Even though almost no one believes the U.S. would really default on its government debt, Moody's warning that it's reviewing billions in government bonds for a possible rating downgrade sent shivers through the investing community. Just the threat of default is enough to roil the markets, and yesterday's stock rally went for naught today as interest rates crept back up.

HEROES

Activision <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:ATVI)") else Response.Write("(NASDAQ:ATVI)") end if %> bounced hard today, rising $1 5/8 to $11 5/8 after reporting fiscal third-quarter earnings in line with consensus estimates. The old home-video-game-turned-CD-ROM-developer had been smashed with the so-called edutainment stocks after a number of negative developments fell on the industry; then they were topped by a skeptical article in Barron's which pushed these stocks way off their September highs. Activision was upgraded after the earnings were announced by Alex Brown, meaning the analyst there either had low-ball estimates, was privately worried they would not make their numbers, or just wanted to generate some business.

CompUSA <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:CPU)") else Response.Write("(NASDAQ:CPU)") end if %> was another tattered technology-related stock that experienced a renaissance today, up $3 1/2 to $35 3/8. The news was that earnings were way ahead of consensus estimates, topping the analysts' best guesses by almost 20%. CompUSA's focus on the sophisticated computer buyer as opposed to the first-time buyer (like semi-competitors Best Buy and Circuit City do) paid off in spades. CompUSA did not resort to the frenzied discounting many of its peers in the industry did, relying on customers to pay up for quality and knowledgeable service. The management has long signaled that the stock is undervalued, buying shares privately as well as starting a buyback last month with their free cash flow.

RMI Titanium <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE:RTI)") else Response.Write("(NYSE:RTI)") end if %> continues to benefit from the fact that the ex-Soviet Union has run out of natural resources to dump on the world markets, surging $1 3/4 to $12 1/4 days after it announced that it was adding surcharges to its prices because supplies were getting a little tight. This must be music to Jim Rogers ears, as he was pounding the table on titanium in the most recent issue of Barron's, part of its annual Roundtable. Although Fools sometimes poke fun at the errant "Investment Biker," scoffing at this return is another matter altogether. If titanium supplies really do become tight, heavily-leveraged RMI Titanium stands to benefit significantly.

Forefront Group <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:FFGI)") else Response.Write("(NASDAQ:FFGI)") end if %> was up $1 1/8 to $8 1/8 today as more investors bought the stock's central story---off-line online. The company develops software that saves customers online time by downloading Web content and "caching" it on the hard drive, saving the end-user the time it normally takes to have the graphics pop up. With most users logging on at 14400 baud and below, assistance in shortening online time makes sense. Investors here, however, need to know there is a heck of a lot of competition here in a product that ASDL, ISDN and cable modems potentially make obsolete in two to three years. A highly speculative Internet play---the very kind people buy into as they ignore the companies that make the building blocks of the Internet (see today's Investment Perspective).

Don't you hate when you look at a stock, think to yourself, "Gosh, that one is really cheap," file that away and forget about it? Happened last night when we glanced at Electroglas <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:EGLS)") else Response.Write("(NASDAQ:EGLS)") end if %>, still pretty bruised after the market fell out of love with anything associated with semiconductors. Electroglas makes semiconductor equipment, specifically wafer probes used to detect bad chips. Its competition is limited to Japanese firms, with KLA Instruments reselling Tokyo Electron's probes in its equipment in the U.S., for instance. Wafer probes increase throughput, a fancy semiconductor term for output, by pulling bad wafers out of the batch. As the product increases productivity, it was really no surprise when this company smashed earnings by more than 10%. The stock was up $2 7/8 to $22 1/4.

GOATS

Even as Activision enjoyed a good day, competitor Electronic Arts <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:ERTS)") else Response.Write("(NASDAQ:ERTS)") end if %> was off $3 1/8 to $22 1/4 after it sounded the warning about next quarter. The company has experienced delays with its leading sports titles on Sony's new PlayStation, which will take until next fiscal year to be unveiled. The fourth quarter does not look as good as it previously did for Electronic Arts. PlayStation addicts who have to wait a few months for hockey and football bashed the stock today. Electronic Arts makes video games for cartridge and computer markets, focusing far too much on the hostile cartridge segment for its own good.

Funeral home owners Loewen Group <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:LWNGF)") else Response.Write("(NASDAQ:LWNGF)") end if %> got hosed today, down $5 11/16 to $18 5/8, after a Mississippi court found that the company had committed a breach of contract. The company needs to put up a $625 million bond as it appeals the publicized jury award of $500 million to Gulf National Corp., which had originally sought only $26 million in damages. This is causing Loewen to mull over the possibility of bankruptcy, something that scared the pants off of analysts, investors and debt raters, who all cut down the shares in one way or another. It appears that acquisition-crazed Loewen opted not to buy a cemetery from Gulf after it signed the papers---something the company would certainly like to bury now. This brings to mind Texaco's bankruptcy in the 80s due to a very similar suit, where the once-proud stock was driven down. That said, anyone who picked up Texaco during the uncertainty made a heck of a lot of money, plus a series of stellar special dividends (about $7 cash a share plus another $1 of a special preferred issue).

Just to prove not all stocks with the word "net" in them go to the Moon, we have NetFRAME <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:NETF)") else Response.Write("(NASDAQ:NETF)") end if %>, down $1 to $5 1/4 today. The company posted earnings $0.03 a share below consensus expectations of $0.04, a sizable percentage shortfall. This after the thinly traded shares rallied last week when mentioned on CNBC's Buy, Sell or Hold (just to tell you how thinly traded they are). NetFRAME lost no opportunity to blame the government in its press release, or more specifically the government shutdown, for its poor earnings. NetFRAME makes multiprocessor servers (a subject we touch on in the Investment Perspective below).

MySoftware <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:MYSW)") else Response.Write("(NASDAQ:MYSW)") end if %> took investor money today, losing $4 9/16 to fall to $6 5/16 after the company missed its numbers for the most recent quarter. Volpe Welty, which helped bring the company public, took the opportunity to cut its rating on the shares to "neutral" from "strong buy." The numbers were 33% below expectations, causing the stock to set a new 52-week low. Product sales in retail stores and direct mail have been disappointing for its sales and marketing software. Management's solution? Become an Internet company. They bought Mediatech today, a publisher of "Internet products" like Live Markup and Live Markup Pro. Both products have the dubious distinction of being sold only over the Internet.

INVESTMENT PERSPECTIVE:

The Digital World---Routers, Highways and Railroad Ties

In the first two parts of this series, I have emphasized that the Internet is all about distribution. The Internet is a new mode of distribution for products, services and information. That people fixate on the products, services and information to the extent that they fail to recognize the means of distribution, suggests that they're ignoring valuable investment opportunities. That which is essential remains invisible to their eyes. The voices that decry the Internet as "speculative" and "faddish" fail to apprehend the deep connections between the digital world and the last two major "revolutions" in distribution that changed the world landscape---cross-country railroads and the interstate highway system.

Mary Meeker makes the connection between the Internet and the roll-out of the railroad system in Morgan Stanley's "Internet Report," likening the situation that confronts investors today to that of market aficionados at the close of the Civil War in 1865. The expansion of the railroad system from coast to coast was initiated by two dozen capital-intensive start-ups that sought to dominate regional rail traffic by laying track wherever they could. These companies were viewed with the utmost importance by the Street, evidenced by the fact that as late as 1896, the first Dow Jones Industrial Average included railroads as eleven of its twelve components.

The distribution made possible by these railroad companies is what was revolutionary; suddenly, you could slaughter hogs in Chicago and deliver them to New York in hours. The railroad system first and foremost allowed the relatively cheap transport of industrial goods in mass quantities. Agricultural producers could now congregate in climates friendly to their crops or livestock without dramatically limiting their markets. Small and large industrial concerns in New England enjoyed the rapid addition of a Western consumer base, which was virtually impossible before the railroads. Finally, those with the wherewithal to purchase tickets could journey in relative safety to visit relatives, travel that previously had been arduous at best.

The analogy to the highway system is even more dramatic. The interstate highway system was the brainchild of the Eisenhower administration and had as its goal the quick movement of military forces across the continental United States. This construction had the salutary effect of opening up the nation's roads to trucking firms, allowing 18-wheelers to challenge the monopoly of the iron horse. More competition drove the prices for the transport of goods even lower, allowing businesses to make money in new regions. In addition, millions of Americans took to the roads in automobiles, "paving the way" for the automobile industry's rapid expansion.

The "Information Superhighway" is an appellation oft grafted to the Internet by politicos and paparazzi who want to seem hip. In spite of the ignorance of the majority of those who use it, the metaphor sticks. The reason the analogy works is because it emphasizes the revolutionary character of the digital world to those familiar with the history of the highway system. The cognoscenti could as easily say the "Information Super-railroad," but that seems to lack the same dramatic flair. The connection is simple; just as the railroad system and the highway system brought the price of delivery of goods down, so does the Internet. Retail operations can open overnight with no overhead other than storage, mailing products to consumers across the world that otherwise would have not been able to visit their store. Providers of services such as advertising and consulting can suddenly maintain cheap and constant contact with clients without being hamstrung by huge long distance bills. And just as the highway system caused the mass consumer adoption of automobiles, the mass consumer adoption of some kind of Internet platform cannot be far behind.

Now, if you had been around in 1865 or the early 1950s, investing in the railroad or construction firms that were building the highway would have been dicey business. The railroads were dicey because they needed to consume a heck of a lot of capital to lay that track, constantly going back to investors for more money. Construction firms are an even dicier lot, since very few are public and even those that are have spotty earnings because of their reliance on commodity products, high-wage labor and expensive equipment to get the job done. No, investing in railroads or construction firms would not have made anyone a lot of money at the start. But what would have?

A savvy investor at either time would have done better by investing in the companies that were selling to the railroads or to the construction firms, as the sudden demand for thousands of miles of rails and asphalt made it a seller's market for those products. The analogy here to the Internet is very revealing. Internet Service Providers (ISPs) suffer the same problem as the railroads; in order to expand their network in the face of negative cash flow, they need to keep coming back to the Street to sell more stock. This process dilutes shareholder equity in a way reminiscent of biotechs, destroying the possibility of long-term capital gains. Investors in companies like UUNet Technologies <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:UUNT)") else Response.Write("(NASDAQ:UUNT)") end if %> have to pay for a lot of track before these companies start to get enough money to fund their own expansion.

How fast are the ISPs expanding their network? There were about 1000 Internet hosts in 1984 compared to over 7 million today---shocking compound growth. America Online's AOLNet plans to expand its POPs 30% by March and 60% by September, according to industry reports, meaning that it will need a lot of network "ties" to build its own network in the coming weeks and months. The fact that AOLNet is one of the Group of Six dominating the Internet's backbone, and currently owns about 300 POPs, this translates into a significant amount of Internet equipment.

So which companies are making the railroad ties for the Internet? Surprisingly few. Most suppliers to ISPs, specifically to the Group of Six (see yesterday's article), are "sole-source" suppliers. What this means is that PSINet only buys Cisco 7000 routers in order to build their Points of Presence (POPs), the dial-up number that you connect to (sometimes called a "node"). A look at a typical PSINet POP is very instructive as to who is making the railroad ties:

The connection starts with the backbone, normally a T3 line that moves incredibly fast. PSINet's T3s are made by ADC Telecommunications <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:ADCT)") else Response.Write("(NASDAQ:ADCT)") end if %>---the ADC Kentrox Data Smart T3/E3 DSU. The T3 connects to an ATM Switch made by Cascade Communications <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:CSCC)") else Response.Write("(NASDAQ:CSCC)") end if %>. From this switch, T1 and T3 lines shoot off to connect large users who achieve direct access through networking software like Windows NT or Netware. For other users, the ATM Switch hooks into a POP Distribution Router made by Cisco Systems <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:CSCO)") else Response.Write("(NASDAQ:CSCO)") end if %>, with lines shooting off to analog modems, Ethernet networks and an Ascend Communications <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:ASND)") else Response.Write("(NASDAQ:ASND)") end if %> Router to provide service to ISDN and Local Area Networks (LANs). This set-up also requires a server, with the most popular made by Sun Microsystems <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:SUNW)") else Response.Write("(NASDAQ:SUNW)") end if %>. The server is where the information that users want is stored and is just as crucial as the connections in the whole distribution picture.

So we have four major names here---ADC Telecommunications, Cascade Communications, Cisco Systems, Ascend Communications. A look at one of UUNet's hubs yields essentially the same data: Cisco 7000 Routing Engine, ADC T3 Plus DS-3 Bandwidth Manager, Cascade 9000 Customer Access Switch and Cascade 9000 Backbone Switch. These companies are smokin', to coin a phrase.

ADC Telecommunication has the most competition of the four, something that keeps it profit margins in the 9.9% range. The company has increased revenues at a solid 30% clip over the past few quarters with earnings growth maintaining about the same pace. Cash-rich and debt-free, the shares trade at a premium valuation of 34 times trailing earnings, giving it a PEG of 1.07 (see the Fool's School article, "The Fool Ratio, AKA PEG) and a YPEG of 0.77 (see MF DowMan's Fribble on the YPEG). The shares look pretty fairly valued to a Fool's eye, but given that the company is on the fast track of the Information Super-railroad, surprises have been consistently to the upside. T3 lines are the backbone of the Internet, but like any body, you only need so many backbones, which puts a ceiling on growth. ADC also competes pretty heavily with companies like Tellabs <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:TLAB)") else Response.Write("(NASDAQ:TLAB)") end if %>, lacking the command of market share that many of its partners down the network chain enjoy. The perception that Tellabs is taking market share from ADC accounts for Tellabs premium PEG of 1.30.

Cascade Communications enjoys a decided lack of competition---which partially accounts for its beating estimates by 20% the last two rounds. The valuation seems nothing short of extreme, with a PEG of 2.53. However, given that the company has a hammerlock on the market for the switches it provides the ISPs, these are pretty cool railroad ties it is pumping out. The company commands margins of 18.9%, a testament to the stranglehold it has on its niche. Looking forward, given its 50% estimated growth rate, the shares seem much more fairly valued with a YPEG of 0.97. Volatile, yes. . . but hold it for a year and if it makes its numbers, there is still some upside room, despite the heart-stopping 107 multiple. Revenues are growing like wildfire, with 575%, 274%, 242% and 140% over the past few quarters.

Cisco Systems has seen revenues ramp up as fast as Cascade Communications, although not to the same degree, as it has about 10 times more in trailing sales. Cisco is a monster of a company, though, continuing to make strategic acquisitions (eating up TGV Software yesterday) in order to boost crucial market share across various product segments. This sucker is the Internet equipment/software version of a Tyrannosaurus Rex. The trade figures that I have seen list Cisco as the sole-source supplier of 80% of the routers that are used for POPs. The wide 24% profit margins fuel the massive growth that this company has seen in the past few years, allowing it to make all of its acquisitions without taking on debt and delivering profound growth to its shareholders. It does get some competition from 3Com <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:COMS)") else Response.Write("(NASDAQ:COMS)") end if %> and Bay Networks <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:BNET)") else Response.Write("(NASDAQ:BNET)") end if %> for LAN stuff, but in the Internet segment, it reigns supreme. The PEG is an uninspiring 1.08 but the YPEG is 0.64, suggesting the company trades at 24 times forward earnings, with a growth rate of 33%. Some upside? Maybe.

Ascend Communications commands a 90% market share among the ISPs---something that railroad tie companies only fantasized about. This is one of the reasons Ascend can command fat 20% profit margins and a massive multiple on trailing earnings; this is a company in the catbird seat. Revenues for Ascend in the last four quarters have risen 199%, 247%, 289% and 333% as it has sold its Routers to a selection of the 300-plus providers of Internet service, from MCI Communications to Joe's Bar and Internet Hook-Up. Ascend's valuation brings to mind Cascade, with a PEG of 1.92 and a YPEG of 1.17. How much should you pay, though, for a 90% market share in a fast-growing segment?

Servers are a trickier subject for a couple of reasons. First off, Sun Microsystems <% if gsSubBrand = "aolsnapshot" then Response.Write("(NASDAQ:SUNW)") else Response.Write("(NASDAQ:SUNW)") end if %> is the run-away favorite in the group, although it is rife with competition from major powerhouses like Digital Equipment <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE:DEC)") else Response.Write("(NYSE:DEC)") end if %>, Compaq Computer <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE:CPQ)") else Response.Write("(NYSE:CPQ)") end if %> and Hewlett Packard <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE:HWP)") else Response.Write("(NYSE:HWP)") end if %>. The competition is fierce and a clear winner is not likely to emerge, unlike with the other companies we've discussed.

How should an investor build an Internet "tie" portfolio? To be honest, I am not really sure. I am seriously considering what Mary Meeker at Morgan Stanley suggests and buying Ascend, Cascade and Cisco to be safe, adding whichever of the T3 providers seems to have the highest quality (lowest multiple relative to growth, reputation and highest market share are the most important considerations for me). The possibility of adding a server company also seems attractive, particularly if I were convinced that the Internet was going to do for the personal computer what the highway system did for cars. (Qualms about cheap Internet access make me a little wary here and I am still percolating possibilities, which will probably show up in another installment of this series.) The clearest one-stock Internet infrastructure holding is Cisco---hands down. The company is the powerhouse in Internet "rails," the online equivalent of Bethlehem Steel.

The money in the beginning of the digital revolution, just like in the personal computer revolution, will be made in hardware. The real profits from software/services and content, although potentially nearer than most believe, are at the very least a few years off. Just as in the early years of PCs, the hardware seems to be where the money is going to be made in the coming months. The profit margins and market shares of many of these companies comes as close to guaranteeing this as a papal blessing.

TOMORROW: The Digital World: ISPs, The New Biotechs

CALENDAR: Friday's Economic Events

---November Wholesale Trade (10:00)
---Treasury announces size of 52-Week Bills (2:30)
---Weekly Commercial/Industrial Loans (4:15), tentative

Byline: Randy Befumo (MF Templar)