Thursday, August 21, 2008

This site has been moved.

Sorry - just one post so far this week, but I have been busy migrating this blog to

I've been working on new insights into Wells Fargo - since I found them to have been naughty with their numbers - much before the Wall Street Journal picked up on it.....

Please look at my older article on Fannie, Freddie and Wells Fargo. I will resume posting over the weekend.

Sorry for the trouble.


Tuesday, August 19, 2008

The next ISRG ?

Intuitive Surgicals needs no introduction. The company's daVinci surgical robot is allowing the company to grow at a heady rate, and added to this, ISRG makes a gamut of disposable tools [which serve as the "working end" of the robotic arm used in surgery] which need to be "renewed" for every surgery. The tools provide ISRG a stream of revenue in a manner that is similar to Gillette's blades. ISRG's revenues are almost at a billion dollars a year run-rate, and it sports a market cap of over $10 Billion. The largest company that exclusively focuses on medical devices is Medtronic [MDT]. MDT's market value was at $10 billion - back in 1995 - the same year that ISRG was founded.

While there is still a lot of money to be made - in ISRG, there always is room for newer disruptive technologies in the field of MIS [minimally invasive surgery]. The two companies of today - which are often compared with ISRG - albeit with ISRG from about a half a dozen years ago, are: HNSN - Hansen Medical and STXS - Stereotaxis. Hansen1 develops, manufactures and sells a new generation of medical robots designed for accurate positioning, manipulation and stable control of catheters and catheter-based technologies. Stereotaxis2 focuses on the cardiac surgery market, and enables physicians to complete more complex interventional procedures by providing image guided delivery of catheters and guidewires through the blood vessels and chambers of the heart to treatment sites.

STXS has an extensive array of patents covering their magnetically guided catheter systems. HNSN has an equally impressive array of patents protecting their Sensei system. After a review of both patent portfolios, I'm confident that STXS has an edge over HNSN. Plus, STXS's system has a huge operational advantage - their catheters are substantially slimmer than HNSN's comparable ones. This is due to the fact that STXS's catheters have a magnetic tip, and are guided by a magnetic field surrounding the patient - while HNSN's guidance wires are bundled with the catheters [and hence thicker]. STXS's system has one huge disadvantage - the room in which the system is located - needs to be custom built - with magnetic shielding. Also, operational procedures similar to those taken by MRI facilities [no ferrous/ferric/magnetic stuff anywhere close to the system or the room it is in] need to be taken into account. In short, once a STXS magnetic guidance system is in place, it is cost-prohibitive to even think of moving it.

HNSN's most recent quarterly numbers were hampered by "sales lumpiness", but the company fell short of their goal for quarterly system sales [system sales drive the sale of disposables]. STXS has the edge in system sales for now - despite the fact that their systems are a lot more difficult to deploy. But, STXS had their own missteps - where they had to go back and re-engineer their magnetic irrigated catheter [it needs to be re-approved by the FDA]. Optimistically, this will contribute to revenues in calendar Q1-2009.

 STXS Q2, 2008STXS Q2, 2007HNSN Q2, 2008HNSN Q2, 2007ISRG Q2, 2001ISRG Q2, 2000
Gross profits6.5M3.5M1.08M0.781M6.1M1.6M

From the table, once can see similarities in the numbers....... ISRG had a reverse 1 for 2 split in July 2003 - which means that there was 100% dilution from the point of time that I chose to pick in 2001 for ISRG. STXS has been smart with their money, and got a $18Million payment in royalty from Biosense Webster - their partner in the manufacture of ablation catheters. Looking at numbers from other 10Q's filed by ISRG, ISRG extracted much more in sales - for the amount of expenditure on R&D - even in the early stages of the company.

The bottom-line:
1. There is no clear winner here in the battle for becoming the next ISRG. In fact, the next ISRG is ISRG - as it has another half a decade of growth left in it.
2. Between STXS and HNSN, my pick is STXS - for their technology has been through 10,000 cardiac procedures - some so complex that they were not attempted before STXS's Niobe entered the fray.
3. HNSN can use the fact that they are more easily deployable - to get an edge over STXS, but at the end of the day, the thinner, more routable STXS catheters will be able to perform surgeries that HNSN's Sensei cannot touch.
4. Both companies are burning money now, and will probably continue to do so for another three years. So, there is no urgency to get into either stock.

Operating rooms are rarely re-located. Neither are MRI labs. So, STXS's biggest negative is probably not a big deal in most cases.

0. I have no positions in either stock. May initiate in the near future.
1. From Hansen's latest 10Q.
2. From Stereotaxis' latest 10Q.
3. ISRG's 2001 10Q.

Thursday, August 14, 2008

Intel Can Grow again. Really......

Back in the early 1980's Andy Grove famously got Intel out of the much larger [in revenues] memory business, and put all of Intel's eggs in one basket and propelled INTC to the #1 spot in the semiconductor industry. But lately, there have been chinks in INTC's armour. For starters, the perennial also-ran in the CPU market - AMD had a temporary lead in the high-end [Opteron, multi-core]. But Intel's response was swift and decisive. Before long, AMD was back in the cellar and for now, Intel need not worry too much - about microprocessors and mobility.

While INTC is very good when it applies to their core business, they have had a less than stellar record - when you look at their acquisitions and other ventures. Even under Andy Grove's watch, Intel started diworseifying [sic] into the memory market. At first it was NOR flash [now divested to Numonyx]. Then they tried getting into programmable logic [Altera eventually bought most of that unit], and in 1999, they bought Level 1 Communications in one of their largest ever acquisitions. INTC bought smaller companies like Shiva to add to this unit - which was eventually packaged and sold to Marvell in 2006.

By far, Intel's worst investment till date is probably in the Intel/Micron Flash unit - which till now has cost upwards of $2 Billion. I wonder what Intel thought they'd get out of the NAND flash market, but one need not wait much longer to be convinced that it no longer is a market that INTC needs to be in.

Much of this is obvious when one looks at the most current 10Q of INTC's where for the six months ending June 28, 2008, net income attributable to the Digital Enterprise Group [includes Microprocessors and Chipsets] was $3.473 Billion; Mobility's net income for the same period was $2.417 Billion, and "all other" had a negative net income of $1.573 Billion. For the quarter ending June 28, 2008, the net income numbers for each of these units was roughly half of the six month numbers - at $1.710 Billion for Digital Enterprise, $1.251 Billion for Mobility and negative $706 Million for "other".

So, for Intel to rock and roll again:
a. Go back to basics. Andy Grove was right - get rid of anything that has to do with memory. Every dollar invested here will end up as a dollar of "impaired assets" - sooner rather than later.
b. Focus on mobility and microprocessors.
c. Stay ahead of AMD.
d. Get "junk" like Numonyx off the balance sheet.
e. Good job buying back $15 Billion of stock. Get authorized to buy back more......


Disclosures: Long INTC.

Tuesday, August 12, 2008

nVidia. I refuse to believe that the worst is over yet.

There have been a lot of road-kills in the graphics business [Chips & Technologies, S3, Cirrus Logic, Tseng Labs come to mind], which is always cyclical, and "owned" by a player or two. For now, nVidia is one of the big two, and will continue to be - till the next major breakthrough in graphics processing creates a new winner [or a new old winner].

I had an opportunity to listen to nVidia's mea culpa conference call - after the earnings announcement that fell short, but wasn't as bad as expected. While this is fine and dandy, it just seems too "perfect" to me, and I refuse to believe that the long-term effects of this admitted snafu will last only one quarter. Why am I being pessimistic ?

a. NVDA came in closer to the low-end of their own guidance of $875M to $950M in revenues [I'd have thought that they had an excellent idea by mid-July as to what they would probably announce].
b. I believe in the cockroach theory. If you see a roach, there are a hundred below the floor-boards that you do not see.
c. In the semiconductor industry, even the best systems cannot stop the "propagation" of a mistake into other areas.
d. nVidia's cycle-time - from when they start designing a chip, to getting it fully qualified at Microsoft [to work with the OS and the graphics API] - to the time the company recognizes revenue from the product is anywhere from twelve to eighteen months [though occasionally, I have heard the company venture smaller numbers (which is possible for re-designs and incremental improvements)].
e. The announcement of a stock buy-back means nothing. Implementation thereof is key [see Linear Tech in my previous article].

Let me be very clear that I believe in Jen-Hsun Huang and his resilience. In fact, NVDA has been through bankruptcy protection once and emerged triumphant [which very few semiconductor companies have done in the past].


Monday, August 11, 2008

The other lucrative niche in semiconductors.

Since I wrote the Altera vs. Xilinx article, I have had multiple requests asking me what the other niche in semiconductors is -that has generated gross margins in excess of 60%...... the answer - Analog/Mixed Signal chips. Devices that perform functions like analog to digital and digital to analog conversion, power supply regulation, DC to DC conversion, etc. There are four companies in this space, but I will focus on the two companies that are better managed, and have pristine financials.

They are Linear Technology Corp. (LLTC) and Maxim Integrated Products (MXIM.PK). Maxim is in the process of re-stating earnings - as they did some options-backdating-shenanigans, and hence, need to go back all the way to 2005, and redo the numbers that they reported. In the most recent conference call, MXIM's CFO said that they would re-file all of the 10Q's and 10K's [going back to 2005] - by the end of 2008. Hence, there is more risk in buying MXIM's stock [and hence a greater capital appreciation potential too].

Getting to the reason for this article, LLTC has historically sported gross margins in excess of 70% and Maxim, in the high 60's. This is only part of the reason for their exemplary numbers.

LLTC's and MXIM's chips are small. They have a few hundreds of devices on average, and some have less than twenty. In a business where Intel and nVidia cram 20 Million plus transistors into a CPU/GPU, this is refreshingly different. The average selling price per LLTC chip was $1.57 in Q3 2008 [ended March 31, 2008] vs $1.67 in Q3 2007 [ended April 1, 2007]. So, these are cheap, ubiquitous chips - found everywhere - primarily in consumer, industrial, automotive and military/space. Yet, they had gross margins of 77.5% in Q3 2008 vs 77.8% in Q3 2007. Moreover, both companies have a culture wherein engineers constantly ask themselves if they can perform the same function - or perform the same function with more precision - with fewer devices.

Since these chips are small, one does not need a $2 Billion fab and $10 Million dollar mask sets to make the chips that they design. In fact the process that they use to make their chips date back a decade [0.25u for the curious types - when Intel is pushing 0.025u (u=1/1,000,000 of a meter)]. So, they can make their chips using depreciated equipment; test their chips using legacy testers - making LLTC a free-cash-flow machine [I sound like a cheap internet "home business" advert].

In the last reported quarter, Maxim had record revenues. They could not say much else. So, let us focus on LLTC's numbers. LLTC had revenues of $298M of which $231M was gross profits for the quarter ended March 31, 2008 [the last 10Q I could pull from the SEC]. Revenue increase was 17% yoy, and costs increased 18% yoy - meaning gross profits increased 16% yoy from $198M in 2007. Net income was $99M in 2008 vs. $98M in 2007. While this looks bad, it really isn't - as LLTC issued some $1.7 Billion of Senior securities [I am not going to discuss the +'s and -'s of issuing debt, but LLTC did and at a very attractive coupon rate of 3% for $1Billion and 3.125% for $700M, and ALL of it was used to re-purchase their stock]. EPS increased from 32c/share to 44c/share fully diluted. The company paid a dividend of 21c/share in 2008 for the qtr. vs. 18c/share in 2007 for the corresponding quarter.

There were 26% fewer shares for Q3 2008 vs. Q3 2007. The company used the entire debt facility mentioned in the prior paragraph to purchase their shares. If you want to know the gory details, pull up this file from the SEC.

So, in conclusion:
1. LLTC and MXIM participate in the highest margin niche in the semiconductor sector.
2. LLTC sells for an attractive valuation, plus the company pays a dividend of 84c/share/year.
3. They make cheap chips that are used everywhere. But LLTC and MXIM produce them cheaper.
4. They use legacy technology and equipment to produce chips - keeping a lid on capital expenditures.
5. LLTC has been aggressively buying back their stock.
6. MXIM may look risky, and is risky, but will probably do well - as soon as the uncertainties around the stock evaporate by the end of the year.
7. I used the words pristine financials for MXIM too - as they have over a billion in cash and no debt, and a very high gross margin business that generated record sales.

Notes: I currently have no positions in either stock, but after writing this article, I might buy into either on Aug 13, 2008.

Sunday, August 10, 2008

Altera vs. Xilinx. Who is better ?

In the semiconductor business, there are two niches that have had gross margins in excess of 60%. One of them is programmable logic. I worked for ten years in companies that wanted to emulate the success of Xilinx and Altera. It took me ten years [and a $35 million dollar judgement against my last Programmable Logic employer] to figure out that this duopoly will own the programmable logic [FPGA and PLD] till the end of time [with a small market-share conceded to Lattice, Actel, QuickLogic, Atmel and Cypress Semi].

ALTR and XLNX make chips that can be programmed to perform any function you would want a chip to perform. Meaning, they make "blank slate" chips that can be configured to implement any logic function that you can dream of. In fact, Intel uses huge boards made of tens of these chips to make sure that their logic works - before they get their CPU's fabricated into silicon. Smaller ASIC companies have become smarter over the last decade and make prototypes of their chips [using programmable logic] before they start mass-production.

The elegance of this solution is obvious - when you take into consideration - the amount of $$$ it takes to design a chip. By the time you design, commit to a mask-set, run a few wafers through TSMC, UMC/whoever, get the chips back and test them, you'd be out $10M. The only drawback to this approach is that the programmable chips run slower than a custom designed silicon solution, and one could not possibly mass-produce using the prototyping chips - as whatever gains you might have in time to market, are erased by the expense for these chips - that can cost up to $1000 a pop. If you use a PL chip, all you do is to erase its original configuration, load a new one, and repeat till you succeed [easy eh ??]. Yes, there are exceptions to every single statement that I made here [which is what makes analyzing semiconductors - interesting].

Companies that tried to compete with the two big boys include Intel, Motorola, AT&T, AMD. Needless to say, they either failed or divested their business. Actually, AMD deserves credit for creating the first PAL [through their MMI division - now part of LSCC]. The reason why all these brand-name companies tried to get into PL, is the gross margin number that I mentioned in the first sentence of this article. The reasons as to why they failed are many-fold, but this article is about which of the companies is the better investment - looking forward the next two to five years.

Both Xilinx and Altera recognize the fact that the business that they are in [however profitable it may be], is mature. Any attempt by either company to enlarge their SAM [served available market] always resulted in decreased gross margins - this is due to the fact that all programmable logic chips can be replaced by an ASIC. In fact, both ALTR and XLNX have a path that they offer - which will allow one to go from a "changeable" programmable logic chip to a "fixed" ASIC, but they both do it for a reason [Bapcha at Gmail if you want to know why]. They both generate prodigious amounts of cash, and pay a small dividend that they easily earn - as they recognize the fact that they are in a niche in semiconductors that is not expanding at the rate that it was in the last decade.

Looking at the prior quarter's numbers, Altera had revenues of $359M and gross margins of 67%. Xilinx's numbers were $488M and 64% respectively. Their revenue growth rates were 12.5% and 9.4%. ALTR bought back 15% of their shares vs. 7.26% for XLNX [yoy fully diluted]. EPS was 32c/share in 2008 and 22c/sh in 2007 for ALTR. XLNX's earnings were 30c/share in 2008 and 28c/sh in 2007 [last reported quarter]. Dividend was raised by ALTR from 4c/sh to 5c/sh yoy/qtr. XLNX went up from 12c/sh to 14c/sh.

Before looking at the numbers, I thought that my decision would be easy, but it sure wasn't. While Xilinx has always had an edge over Altera in terms of market share, profitability, and earnings predictability [historically], ALTR is growing faster, has slightly better gross margins, and most significantly, better employee morale. Xilinx might be a better stock for total returns as it pays a significantly higher dividend, has a CEO in place who is determined to contain costs, but a work-force with a lot lower employee morale.

I started writing this article, wanting to pick Xilinx over Altera, but I have to pick the real growth stock among the two - ALTR. The market seems to agree with me - rewarding ALTR with higher multiples than it does for Xilinx. When the semiconductor market heats up, both these stocks have traded at PS numbers in the double digits [but that was when they were real growth stories].

Bottom-line: ALTR for growth.


Notes: I have NO positions in all of the stocks mentioned. Statements about employee morale constitute my opinion. Both companies have employees that are happy and vice-versa. Both companies have significant software investments, but it is a means to an end [selling chips].

Wednesday, August 06, 2008

Amgen needs to start paying a dividend.

Amgen thinks that it is a scrappy little biotech. They sure are a biotech company, maybe even a little scrappy, but definitely not little. In fact, their growth numbers look anemic [they need a shot of EPO :-)]. Amgen was a duopoly for fifteen years - during this period, they sold two drugs - Epoetin Alfa [aka EPO, Epogen, Procrit], and Neupogen [Filgrastim]. Keeping things simple, EPO helps the body build red blood cells and Neupo, white blood cells.

When Amgen had no drugs [in the early 1980's] they sold the rights for dialysis, diagnostics, non-human and overseas [except Japan, which went to Kirin], rights for EPO to Johnson & Johnson (JNJ), and attached an often disputed royalty clause based on EPO's sales. It was an agreement a tiny biotech had to sign - in order to get their development efforts funded [Genentech (DNA) made a similar agreement with Roche in the 1980's - Roche still owns 56% of DNA; and Chiron (now part of Novartis) made a similar deal with Ciba-Sandoz]. The reason for the disputes from my point of view, is that both companies grossly underestimated the amount of money EPO would bring in.

Amgen was smart - before the patent on EPO ran out, they added on a sialic acid group to EPO, and EPO+Sialic Acid (Darbopoetin) was christened NESP [Novel erythropoiesis stimulating protein]. Similarly, AMGN added a PEG group [polyethylene glycol, OK, it really is monomethoxypolyethylene glycol] to Neupo and called it Neulasta. Of course AMGN wanted more money for NESP and Neulasta - than for EPO and Neupo respectively. The first to balk was Britain's NHS. Canada followed, and even medicare started asking tough questions. Then JNJ claimed that NESP was a derivative of EPO and hence JNJ had rights to NESP too.

OK - so now, the reason I went through all of the above is to show that AMGN is now in the big-leagues. They now have nine drugs [the four compounds mentioned above plus Erbitux account for 90% plus of AMGN's sales]. While AMGN faces competition, there is no real generic threat.

Looking at AMGN's balance sheet, they are sitting on $9 Billion in cash - about a half of which is outside the USA - which if repatriated will cost the company dearly, so the company clearly states that if the $4.5 Billion were repatriated to the US, it would be taxes at AMGN's marginal tax rate [the highest], plus local taxes in the country from which the money is repatriated. The bottom-line is that the company cannot use this money directly to issue dividend checks [but there are ways around this].
AMGN repurchased 8.8M shares for $537M in the first three months of 2007 [they still have $6.8 Billion in repurchase authorizations]. Till now, this has been AMGN's preferred way of returning cash to stockholders.

Given that that stock chart for the past five years looks like:

AMGN needs to start paying a dividend. This will also boost investor confidence in the fact that they will be around for a lot longer.

Disclosures: Long JNJ. No positions in AMGN, DNA, AVE, NVS, WYE.