Friday, March 30, 2007

Dendreon (DNDN) Odds Less Long

It is the kind of headline Wall Street loves: Dendreon's stock price doubled in a single day, Friday, March 30, 2007. But the back story is long and I have an amusing incident of my own to tell. I'll also give some guidelines for pricing the stock given the current facts.

I've owned a tiny amount of DNDN for a couple of years; I've followed it for closer to four years. I know that most promising drugs fall by the wayside long before they can be sold; that is one reason the drugs that are approved and marketed cost so much. There are three stages of clinical trials a drug must go through to get FDA approval: Phases I, II, and III. Phase I is mainly to make sure the drug is not an obvious poison, but you would be surprised how many drugs get knocked out in that round. Even if a drug passes Phase II and Phase III, it may not be approved by the FDA, or it may be sent back for more trials. Dendreon's Provenge for prostate cancer did not "meet its endpoints." That is, goals were agreed to with the FDA, and when the statistics were run Provenge did not meet those goals.

So analysts did not like Dendreon. Investors did not like Dendreon. You could buy the stock cheap. Of course Dendreon's management said that they had reason to believe Provenge would be an exception to rules. After all, it was proven safe and it prolonged men's life for a significant period of time (4 weeks). But we all all used to taking management optimism with a grain of salt.

We all knew the independent advisory panel would make its decision of Thursday, March 29th. Wednesday the price of DNDN rose to over $5 per share. I decided to cut my risk by selling 25 to 50% of my stock on Thursday morning. But when I woke up Thursday morning trading had been halted. In my experience trading halted on an individual stock means I've lost most of my money. Well, I knew the risks and I had gambled and lost.

By my figuring the panel had a 50% chance of tossing Provenge out, about a 30% chance of demanding more studies, and about a 20% chance of recommending approval. When you take the potential value of the stock if the 20% comes in against the losses of the other 80%, the stock priced out at (by my reasoning) $12 per share. Of course you can get whatever end price you want by changing the ratios and the values for given outcomes.

Lucky for me, and all Dendreon loyalists, the panel decided to recommend Provenge despite some reservations. So the halted trading made me a richer man.

How should we price DNDN now? There is still some risk that the FDA will disagree with their independent advisory panel and not approve Provenge. I would say that given the entire picture the risk is small but significant. Depending on how conservative you want to approach the issue, you could run it as anywhere from 10% to 50%.

But the real value of the stock is not simply in giving Provenge to one very specific type of prostate cancer patient. Dendreon has set itself up to be a major pharmaceutical company. It built a major manufacturing facility already, rather than waiting for approval. It is ready to produce and market Provenge.

We are going to see plans for a lot more clinical trials of Provenge. The way the FDA works, you need to prove the drug for each type of cancer. In Dendreon's case they can expand to a variety of prostate cancers, but they can also go after all the other cancers. What they have is not so much a drug as a process. Essentially, they activate the human immune system to suppress cancer. This is something everyone's human immune systems are doing all the time. You can look at cancer as an immune system failure. Dendreon has learned how to teach the human immune system to do its job better.

Given the complexity of the human body and its biochemistry, doctors will use Provenge and its sister drugs in combination with other types of cancer therapy.

So how much is Dendreon worth? Even if the FDA changes its mind and requires more tests, the intellectual property Dendreon has created is substantial. Dendreon will have to raise more money, which will dilute the value of current stock. The stock won't become worthless, but the downside risk is substantial.

The upside potential is even more substantial. I see no reason why, at the upper end of the scale, in 5 years Dendreon cannot have the $15 billion market capitalization of a cancer-drug company like Biogen Idec (BIIB). But giving it a current market cap where it might be in five years if we are very lucky does not make sense.

Given the entire picture I would be comfortable with a market cap of $1.5 billion before the final FDA decision and closer to $2 billion if final approval is given. Respectively $18 to $24 per share. But this is no stock for the faint-of-heart. The downside risk is substantial.

Never forget that once a drug starts being given to tens of thousands of people, issues may show up that did not appear in the Phase III trials.

Balance your risk. When you get lucky, don't assume your luck will hold. Figure out the odds as best you can, and play the odds.

For more information you can see my Dendreon page.

Tuesday, March 27, 2007

Oracle on BEA and SAP

Many companies' managers shy away from talking about their competitors during analyst conferences. The analysts follow suit, usually, though mostly everyone knows who is being talked about using terms like "the competition," and "the other guys."

Not Oracle. Following the lead of Larry Ellison, they name their competitors and openly discuss the game of competing. Take the recent (March 20, 2007) Oracle analyst conference on fiscal Q3 2007 results (see my summary). Oracle had a great quarter, but that was not crowing enough.

BEA makes "middleware" or application servers. Basically, this is complex software that sits on top of the operating system (usually Linux or Unix) and allows applications (Web services, for instance) to work efficiently for large numbers of users. Some say that it is unnecessary: a good operating system should handle the job. Still, billions of dollars worth of this software is sold every year. According to Oracle spokespeople, BEA's revenues from middleware grew 8% this last year. Oracle, a late-comer to the middleware game, had a growth rate of 82% and now has higher middleware revenues than BEA. Another big middleware competitor is IBM.

SAP, on the other hand, is known for its ERP (enterprise resource planning) software, and it makes related software like CRM (customer resource management). Oracle bought Seibel, the big CRM player. Combining Seibel's capabilities with Oracles database prowness and making a unified suite of ERP/CRM products, Oracle is now Number 1 in CRM market share, grew this segment by 57% over the year, and is looking to eventually take the number 1 ERP spot from SAP. Given the two-decades of grumbling I have heard about SAP, this is not impossible.

Oddly, there was little mention of Microsoft, a favored recipient of Oracle bashing. In the CRM/ERP market Microsoft is a relatively new entrant and is mainly targetting businesses smaller than Oracle and SAP provide services for. Of course that has usually been Microsoft's manner of entry: start with very small businesses, improve and learn, migrate up the gravy train as those small businesses grow.

It is also notable that Oracle is targetting specific industries (like finance and oil) with ERP/CRM products designed for those industries. Many small players like CDC Corporation (CHINA) have been successful with this strategy, but Oracle can immediately go after the largest corporations.

Oracle, once just a database company, has clearly succeeded in building a suite of products that work well together and meet the needs of medium to large businesses.

I do not currently own Oracle stock (or BEAS or SAP), but I do own a bit of Microsoft (MSFT).

Monday, March 26, 2007

Critiquing the Federal Reserve

With hindsight it is easy to critique the interest rate policies of the Federal Reserve. If they had raised rates earlier in the 1990's there may not have been quite so much of a shock from the bursting of the Internet bubble. If they had not sent rates so ridiculously low in 2002 to 2004 then we would not be suffering now from all the turmoil in the housing market. Some folk even think the nation would have not suffered through the Great Depression if the Federal Reserve had been on the ball, both to dampen enthusiasm in the late 1920's and to flood the country with liquidity after the bubble burst.

I was critical of the Federal Reserve for excluding the huge asset class known as stocks from its analysis in the late 1990's. Being the caretaker for the economy as a whole means you should watch every asset class for inflation and every sector for its ability to impact growth both long run and short run.

I was critical of the extremely low interest rates we saw a few years ago. Recall that the recently ended housing boom began with a rise in house values at a time the rest of the economy was in a recession, a strange anomalie. Interest rates did not need to be that low to revive the rest of the economy. Once the prices of houses started to escalate sharply the Federal Reserve should have started a gradual rise in interest rates. This would have smoothed the curve for the class of people who took out adjustable rate mortgages (ARMs). It probably would have prevented prices and the pace of construction from overshooting. It would also have begun fighting non-construction inflation earlier in the economic cycle.

The Fed lately always seems to be compensating, or overcompensating, for earlier mistakes. Jacking up interest rates every consecutive meeting in 2005-2006 might have been necessary to fight inflation, but it created a huge class of people whose re-adjusting ARMs have led to personal despair and widespread anxiety about the housing market.

So where do we go from here? While inflation is a long term danger, the greater danger right now is the readjusting of ARMs. The Fed caused rates for ARMs to be artificially low. They should take some responsibility for the mess.

If I were on the Fed (fat chance) I'd be arguing for two consecutive half-point cuts. This is despite the fact that I'm not an easy money guy. It is an intensive care treatment for victims of the Fed's past mistakes. It would mean that ARMs coming due during this period would reset at much more reasonable levels for the fools that bought the instruments without thinking seriously about the implications.

Then I would pause and watch. If, despite the cuts, the economy moves into a recession or a very-low growth period, more cuts would be warranted. I think it is more likely that following my 1% cut advice the housing market and mortgage markets would stabilize and the economy as a whole would grow at a good pace. If that happened, then quarter point rises once per quarter would be justified to fight long term inflation.

But I'm not the Fed. Based on the past behavior of Fed, I think it will tend to do nothing until it has to do something too fast, and then it will overdo. It will act like a stupid cow, and then it will jerk and stampede. Its job is to smooth things out, but the wrong people (subject to the wrong pressures from the wrong friends) have been assigned to that job.

Tuesday, March 13, 2007

Applied Materials (AMAT) Slowdown

Applied Materials (AMAT) makes the equipment that makes electronic semiconductor devices (chips). In a sense they compete in the most basic industry of the modern electronics industry.

In AMAT’s last analyst conference they reported a 10% sequential decrease in quarterly revenues for the quarter ending January 28, 2007, but they were still up 23% from the quarter ending January 28, 2006. Asked to explain themselves, they said the main factor causing the sequential decline was decreased spending by display manufacturers. A seasonal slowdown in demand for chips also meant that semiconductor manufacturers were able to slow the pace of equipment purchases.

That makes sense given reports from the rest of the electronics industry. While large flat-panel displays are becoming commodity items, partly this was due to remarkable price reductions going into the holiday season of 2006. This cut into profit margins at retailers like Walmart, Best Buy and Circuit City; it also cut into the margins of the manufacturers of the displays, from high-end players like Sony to low-end players that are running full tilt trying to gain market share in this booming market. If margins are low one elective expense to cut back on is production machinery. Since demand is expanding it is not a matter of purchases of machinery going to zero. Instead excess capacity is minimized and machines are purchased only when they are really needed to meet new demand.

The same basic scenario is true of semiconductor makers. In the spring of 2006 there looked like there might be supply constraints. The people who put the chips into the boards that go in everything from cell phones to toys beefed up their inventories rather than risking being unable to meet demand in the holiday season. But then with the holiday build finished, they looked at 2007. The outlook was unclear. Even with demand good they were able to cut inventories. This meant chip manufacturer’s inventories started climbing (some more than others). So no need to purchase machines to make more chips. Again, it is not that demand went to zero. If you look at their capital budgets they kept spending; at their analyst conferences they all gave numbers for further capital spending in 2007 (which analysts use to try to predict the future for firms like Applied Materials).

A key to understanding the chip equipment makers is the transition question. Right now the leading edge is getting ready to change from the 65 nm process to the 45 nm process. But most cutting edge firms, like AMD, are still transitioning production to 65 nm. Many chips are still produced on older processes like 90 nm or larger (nm is nanometer; less nanometers mean smaller chips or more circuits on the same sized chip). Equipment makers are constantly getting ready for the next big thing, selling the cutting edge stuff, and continuing to sell the less expensive older machines. Older machines get sold off to foundries that make low-end chips.

What are the general categories of equipment that AMAT makes? Deposition processes add a thin, sometimes single-atom, layer to a substrate. Planarization makes silicon perfectly flat. Etching removes unwanted materials to create a circuit pattern. Inspection and measurement systems assure quality throughout the fabrication process. Ion implanting changes the electrical characteristics of a substrate. Thermal processes are used to change physical and electrical properties of substances and for some deposition processes. Cleaning equipment is needed to assure perfect inputs to the various other fabrication processes.

One of AMAT’s competitors is Varian Semiconductor (VSEA), which sold about $794 million in equipment in 2006. Its stock price is today over $50 per share, almost double what it was a year ago. Varian specializes in ion deposition.

AMAT sold $9.1 billion of equipment in 2006 and had profits of over $1.7 billion or $1.17 per share. Even with the Q4 sequential decrease they made $403 million in net profit.

AMAT can continue to grow as the semiconductor industry grows. With a price to earnings ratio of 15.75, or a 6.35% earnings return if you buy at today’s stock price, it is a pretty safe bet. It also yields over 1% in dividends.

With upside potential driven at base by growing demand for electronic goods in developing nations, the only downside risk in the short-to medium run for AMAT would be a global recession. Around April the consumer electronics manufacturers will need to place their bets. They will try to predict demand just right, balancing the risk of not having enough product with the risk of overestimating demand. Those decisions will be passed down the chain to chip makers and eventually to chip equipment makers like AMAT.

Saturday, March 10, 2007

Altera (ALTR), FPGAs and CPLDs

At its November 6, 2006 analyst conference Altera reported revenues of $341.2 million for its 3rd quarter ending September 30th. It predicted 4th quarter revenues would decline 2 to 5%. Then when it reported Q4 revenues at $317.4 million at the February 13, 2007 conference, they were down a whopping 7%. A further decline of up to 4% in revenues was given as guidance for the first quarter of 2007. Should we panick? Is Altera sliding into oblivion? Is this 2001 all over again?

Altera makes semiconductor chips and specializes in FPGAs (Fully Programmable Gate Arrays) and CPLDs (Complex Programmable Logic Devices). All these devices can be contrasted to ASICs, which are Application Specific Integrated Circuit. Supose you have an application, perhaps processing some information inside a cell phone or a hard disk drive. You decide what the semiconductor chips have to do (what their logic structure is) in order to achieve your goals. With an ASIC you then design the layout of the logic that will be in silicon. To create the ASIC chips you go through the standard manufacturing process of using masks, doping, and etching to create batches of chips incorporating your design. FPGAs and CPLDs approach the same problem by giving engineers sets of general solutions that are arrays of possible logic states. You program the FPGA to have the same logic that you would build into an ASIC. So the end products should function identically. Why not use all FPGAs, or all ASICs? There are cost advantages to ASICs once a design is nailed down and if production is on a large scale. But when an engineer sits down to design an ASIC, she could just reach into her parts box, pull out an FPGA, and program it instead. So FPGAs go from idea to finished product much faster. If you want to fix an error, or do an upgrade, often all you have to do is reprogram the FPGA. With an ASIC you have to redesign the circuits, make new masks, etc. Speed and flexibility are what FPGAs are about, but they are also great for prototypes and specialty products with low production runs.

Because of this Altera's competitors are not just other FPGA and CPLD manufacturers; they are ASIC manufacturers as well.

While prices of all stocks fluctuate, there is no way a company like Altera, in the price range it has been in lately (52 week high $22.32, 52 week low $15.54), can be compared to the revenueless Internet stocks of yesteryear. In fact even in the relatively lousy 4th quarter Altera had net income of $99 million, up 43% from its Q4 of 2005. It is in a competitive business, but not that many companies are competing in programmable silicon, so profit margins are healthy.

Altera constantly invests in the future: its R&D runs about $60 to $65 million per quarter. Part of the recent revenue decline is due to customers reducing inventory, getting into cautious positions for 2007 sales. 39% of revenue was in sales to the communications industry, which continues to deploy not just new models of cell phones, but whole new types of products like WiMax and 3G phones and their attendant base stations.

Who competes head-to-head with Altera? Actel, Atmel, Lattice Semiconductor, NEC, Quicklogic, and Xilinx. Actel had $48.2 million in Q4 2006 revenues and has FPGAs as its principal products. Atmel primarily makes microcontrollers, though it does have several FPGA lines; it has been struggling the last few years. Lattice Semiconductor is focused on FPGAs and had revenues of $61.8 million in Q4 2006. NEC is a huge company which derives only a small amount of revenue from FPGAs. Quicklogic’s web site is worth a look: they openly and cleverly challenge Altera and Xilinx on their home page. Their Q4 revenue was $7.7 million, plunging 25% year-over-year. And then there is Xilinx, the main competition. You can find out more about XLNX in my analyst conference summary for this company, which claimed 70% of the global FPGA market in 2006.

The only real short-term threat to Altera’s profitability is a global recession. In the long run their competitors are a threat, but Altera also has the potential to eat into Xilinx's market share. Overall PLD (programmable logic device) usage is certain to grow.

Wednesday, March 7, 2007

Adobe (ADBE): Past, Present, and Future

Adobe is a well-known software company that is included in the NASDAQ 100. I don't own, and have never owned, its stock. I have been tempted to buy it, but it has never seemed to be the kind of bargain I look for. Which means that other investors appreciate its future potential to keep the price . Despite not owning it or having a client willing to pay to have it researched, I follow Adobe and produce summaries of Abobe analyst conferences. Adobe is a big enough company that you can learn a lot about software and IT trends by watching it. Like most people I use some Adobe software. I still use PageMill, though I am transitioning to Dreamweaver. I use Acrobat Reader. I used to use Acrobat (which produces the files that Acrobat Reader reads) but have found that the ability to generate Acrobat files from other programs is adequate for my needs.

In the old days (the 1990's) when I ran a small book publishing company (III Publishing) I used CorelDraw and related products rather than the Adobe products because the Corel products were up to the job (at my level of production needs) and cost a lot less. I also learned to use competing products from Macromedia when I ghost-wrote a book on Macromedia products. The Adobe products I will be using the most this year, DreamWeaver and Fireworks, were both developed by Macromedia.

Adobe has the same problems, and opportunities, that you see at most big software companies these days. The markets of developed nations are largely saturated with products, so income comes from product upgrades and penetration into developing nations. Adobe has a vast amount of expertise in house, but the general level of expertise in the world reduces their in-house advantage. Other companies want some of Adobe's market share, revenue, and profit; many are small, but Microsoft clearly wants to get into the graphic design market. Then there are the software-as-a-service companies, including Google. While so far these companies have little market penetration and are going after low-end users (their products seem light-years away from appealing to graphics professional), they are a long-term threat to Adobe. If not to its dominance, then at least to its pricing ability.

Adobe products are expensive. For that reason they are pirated quite a bit. Today Acrobat 8.0 Professional costs $379.99 at Amazon; as much as a serviceable 64 bit computer. The upgrade is only $149.99, for those on that treadmill. The standard Creative Suite is $849.99, which gets you Photoshop, Illustrator, InDesign and some other stuff. Since the main competition was Macromedia, pricing is intact for now.

I thing Adobe will continue to do well. They are on the same innovate-or-die treadmill that Microsoft and other tech companies tread. They have enough money to buy up the best of their smaller competitors. Their main danger of the moment is Google, but that company is very scattered and likely to put its main efforts into clashing with Microsoft and Oracle, not with Adobe. They have potential to branch out of their graphics niche and to benefit from the video-on-Internet boom.

The next Adobe analyst conference is March 20th.

Monday, March 5, 2007

Fighting the Stock Market: 2001 Over Again?

I became a technology stock research analyst by accident. I was talking to a big-time investor back in 2000. I made indexes for technology books back then; one of my clients was Microsoft Press. He wanted to invest tens of millions of dollars in Linux startups. I told him I did not think it was a good idea. The Linux startups (see Red Hat) were already priced as if they had defeated Microsoft in the field of battle, yet they had made no profits and had little to show in sales. I explained that Linux was not ready for prime time, and more to the point, it was free. How much money could be made on a free product? I said that most of the switching to Linux would come from UNIX sellers, which meant that IBM and Sun would be hurt far worse than Microsoft.

I owned no stocks in 1999. Today I own Microsoft but not Red Hat. But today is not 2000. If someone want's to buy Red Hat today, and was a client, I would make up a report of the downside and the upside. Today Red Hat has substantial sales, makes a profit, and has a Price to Earnings ratio that is steep but not much out of line with its growing prospects. It has diversified around Linux. It has a bunch of competitors, including Novell (NOVL). On the other hand Sun, while it is not out of the woods yet, at least has a realistic stock price. It had substantial profits in 1999, but its stock price was still more than an order of magnitude higher than what its actual numbers could support.

Military historians often attribute the loss of battles and wars to generals who were still stuck in the tactics or strategies of the past. Remember the investors who failed to panick early enough in 2000-2001? Or rather, they should have analysed the companies in their portfolio, rather than buying into a bunch of bull and having to time their panick.

Now some people are acting as if it is 2001 and dumping tech stocks due to the China panick of last week. Take a good look at some of the stocks they are selling. They are profitable and have attractive PE ratios. They are still growing. Their main sources of growth are often Asian markets that are growing strong. Yes, a recession in the US would hurt their earnings. No, none of them looks like the next Microsoft. But they are worth something. And if there is no recession this year, they are going to be worth substatially more than people think today when 2007 is over.

Look at the numbers. Take into account what management has to say by reading my summaries of analyst conferences. Diversify; prefer substance to rhetoric. You'll beat not just the small investors who don't do their homework, but plenty of the professionals as well.

Friday, March 2, 2007

Microsoft Windows Vista Business 64-bit Impressions

Mostly these days I do analytic work. I still write computer programs on occasion in Visual Basic, or PHP for a web page, or something like C++ if necessary. One thing I guessed wrong as an analyst was how quickly the computing world would convert to 64-bit desktop computing. I watched people in the 80's argue that 16-bit computing would not catch on, and argue in the 90's that 32-bit computing was a waste because most data was text characters in 8 bits. In both cases the transitions took place much more quickly than many predicted. People who wrote 8 bit or 16 bit programs midway through the transitions, confident their work would be good for years, found that within two years operating system support for their old programs had been suspended. I watched the Red Cross invest in an 8 bit computer system, pay millions to implement it on a nationwide basis, and then have to abandon it less than two years later.

But 64-bit computing has come slowly and probably won't become dominant in 2007. Mostly this can be blamed on Microsoft, though practically every software maker and hardware manufacturer has dragged their feet. Intel planned to trap most of us in 32 bit computing for the foreseeable future by shifting to Itanium processors for high-end 64 bit computing while leaving us with overheating Pentiums for the desktop. Fortunately AMD stepped in and gave us processors that could run either 64 bit or 32 bit, the brilliant Opteron and Athlon 64 designs. But for those wanted to move to 64 bits, just having a processor was not enough. Linux was available for people who are maniacal geeks. Microsoft brought out a 64 bit version of XP and a beta 64 bit Vista in 2005/2006, but they were clearly for experimenters. I was working freelance for Microsoft up until the end of 2006, but it was just one of my duties, and I could not rationalize allocating time to hassling with test products. Plus I did not believe Vista would be delayed for so long. In March of 2006 I bought a Systemax computer sporting an AMD Athlon 64 X2-3800+. It was blazing fast even with XP, and Vista was due real soon.

I bought the Business edition of Vista in February this year. You can't just buy the 64 bit version, you have to buy the standard package and then send to Microsoft for the 64 bit install disk. Which means Microsoft wants all those ultra-fast 64 bit AMD based computers to run at 32 bits, for now. Sad.

I started the shift on February 21 by backing up files. You can't upgrade directly from 32 bit XP to 64 bit Vista. So I did a clean install. It went without a hitch.

Then I went to install Windows OneCare, which I have been using. You get to use it on three computers for one price and it works great. Only Microsoft does not have a version that works on 64 bits! Those stock options must have made their workers rich and lazy.

Microsoft Visual Studio 2002, Office, and Adobe Studio 8 installed without a hitch. These are 32 bit programs and they run without any problems except that Outlook won't save my email-access passwords. There may be a fix for that, I have not checked yet.

Partly because when I press the "Help and Support" button, I just get an error message. That in itself indicates this program is not quite ready for consumers.

Also, I can't print anything. HP has not made a Vista driver for my relatively new HP LaserJet 1020. My ancient Brother MFC 9700 does not have a driver available either, but it is just about ready for the scrap heap anyway.

Napster installed easily and runs great. My sound card did need a driver from Creative Labs; they had done it! Vista configured itself for my network without going through the long process that was needed with XP. That is a big improvement. I transferred my data and settings from an older computer with Windows Easy Transfer. That failed the first time but went smoothly the second time I tried it.

So now I am set up to write and run 64 bit programs that should be good for the next decade. I can't recommend 64 bit versions of Windows to consumers or businesses yet, but if a few more drivers become available it should be safe to test the waters. Hopefully by 2008 all this will be ironed out and 64 bit Vista will be standard on new computers.

I used