- Licensing Revenue: licensing royalties from its patents. This year is expected to earn approximately $450 million from royalties (approx. 10% of total revenue)
- OEM Revenue: Selling NAND flash memory components to OEMs (such as handset manufacturers (approx. 30% of total revenue)
- Retail Revenue: Retail products such as USB memory sticks, its line of Sansa media players, and its memory cards for digital cameras and cellular phones. (Approx. 60% of total revenue)
Is SanDisk a Dominant Player in the Markets it operates in?
The answer here is a resounding YES.
NAND Flash Memory Production
According to SEMI ('Fabfutures' and 'Fab Capacity Report') SanDisk is the world's largest producer of data storage memory chips. It manufactures, via a joint venture with Toshiba via its 'megafabs' located in Japan. Fab 4, which commenced operations this September, is said to be the size of five football fields, and the most advanced manufacturing plant of its kind in the world.
Retail - USB and Memory Cards
SanDisk is the dominant player in all markets it has entered. It has approximately 42% market share in North America; In Europe, it doubled its market share in the last 12 months to 28% and indicated that it is experiencing difficulties in meeting European demand; in Asia and Australia market share is 23%.
Retail - Media Players
According to recent data from the NPD Group, SanDisk is ranked a distant 2nd in the US MP3 player market, with approximately 10% of market share, compared to Apple's 73%. The good news however is that whenever Apple sells an iPhone or iPod, SanDisk indirectly earns revenue in the form of royalties from Samsung, who make the chips for Apple's products.
Does it possess an 'Economic Moat' (a durable competitive advantage)?
SanDisk's economic moat is its business model. Based on tight vertical integration, SanDisk's business model has allowed it to remain competitive and protect its margins during the tougher periods.
(1) License Royalties: SanDisk's intellectual property, which it strengthened with the M-Systems acquisition in 2006, is the foundation of the business. With more than 700 U.S. patents and more than 400 Foreign patents, it is the only company in the world that has the rights to manufacture and sell every major flash card format and USB flash drive. Revenues from license royalties account for approximately 10% of total revenue and will amount to around $450 million for 2007. The estimated 2-3% that competing manufacturers pay to SanDisk in licensing provides that much of a buffer to SanDisk's margins and allows it to remain competitive, and continue to be the lowest cost producer.
(2) Manufacturing Capabilities: The captive production from the SanDisk-Toshiba Fabs allow SanDisk to pay significantly lower prices for NAND flash components for its own products than its competitors. Again, this enhances SanDisk's competitive position.
(3) Technological Expertise: In the 2006 acquisition of M-Systems, SanDisk acquired superior Flash technology that has yet to be rolled out. Known as X4, it allows NAND flash manufacturers to produce twice as much memory on the same wafer, at half the cost. It is estimated that this will translate to a 30 percent cost reduction in producing a 1 gigabyte NAND flash wafer. During the recent conference call an answer to one of Craig Ellis' questions revealed that in 2008 the majority of products will still be 2-bit per cell, and not X3. This suggests that SanDisk has chosen to keep their 'big technological guns' - X3 and X4 "in reserve" and to roll them out when the business environment turns for the worst. You could come to the conclusion that the fact that SanDisk's management has not chosen to do so as yet means that they are not really 'feeling the heat' yet – and are confidently competing well.
(4) Retail Operations: With more than 210,000 global retail outlets, Sandisk enjoys a large distribution network, which sells its USB and flash memory cards for cell phones, media players, cameras and video games.
What are the Concerns?
(1) NAND Flash is a Commodity – its pricing is Volatile
SanDisk operates in an extremely price-competitive environment. NAND flash memory is a commodity, and is traded on an exchange, with daily pricing which can be viewed at http://www.dramexchange.com/. While SanDisk is vulnerable to this volatile market, it is less so than its competitors because of its low cost manufacturing capabilities.
(2) 2007 SanDisk's growth was not organic, but mostly a result of the M-Systems acquisition
The SanDisk Q3 press release announced: "Revenue Grows 38% Year-Over-Year", "Product revenue was $919 million in the third quarter, up 36% year-over-year and 28% sequentially" and "License and royalty revenue for the third quarter was $119 million, up 52% year-over-year and 11% sequentially". Impressive numbers, right? Well, not exactly.
As Avishai Ovadia correctly pointed out in his blog (Hebrew), you cannot compare revenue and earnings numbers for this quarter to the same quarter a year ago - it's comparing apples to oranges. This year's Q3 data includes the revenues and earnings from the M-Systems acquisition. Last year's Q3 data does not.
(3) SanDisk's Supply Cannot Meet Demand
I think this came through loud and clear in the last conference call. That demand is strong is clearly a good thing, but when that demand cannot be served, customers seek the product elsewhere – and market share is eroded. Which brings me to my next concern.
(4) Capital Intensive Requirements
In early September, SanDisk and Toshiba jointly opened their latest fabrication facility in Japan - Fab 4 – larger than 5 football fields. According to SanDisk's July 2006 8-K, the cost to SanDisk will be in the vicinity of $1.5 billion (through to the end of 2008). SanDisk have already commenced discussions with Toshiba on constructing a new facility – Fab 5, but as was learned from the conference call, capital expenditure requirements for this plant are not expected to hit the books before the second half of 2009.
(5) Samsung and Other Manufacturers May Aggressively Re-Negotiate Licensing Agreements
Samsung, the second largest producer of NAND flash memory components may grow tired of paying licensing royalties to SanDisk, and may aggressively re-negotiate these contracts in 2009. I am not able to provide a qualified opinion on this issue.
(6) Litigation Risks
Sandisk CEO, Eli Harari received a grand jury subpoena as the US Department of Justice is looking into possible anti-trust violations in the industry .Also, the company, along with 23 other companies, are also being sued in a consumer class action that alleges a conspiracy existed to fix flash memory prices.
(7) NAND flash may become superseded by a newer more superior technology
There is always a possibility that this may occur, but even if a more superior technology was developed, billions of dollars have already been invested in manufacturing plants, cell phone design, etc. and I doubt that the new technology will be embarced. I think the probability of this happening is low, and that NAND flash has become the standard for the mid to long term.
What's the Catalyst? What trends will benefit Sandisk's Business?
The catalyst here is forecast NAND flash growth. According to Gartner Dataquest, current global NAND Consumption is estimated at less than 3 trillion MB. By 2011 however, the entire NAND landscape will be completely unrecognizable with estimated consumption at – wait for it – 33.5 trillion MB. This is mind-blowing growth!
Do these estimates make sense? Consider the following:
- 3 years ago there wasn't a single handset that could use an external memory card.
- Up until this year, only 30% of mobile phones could use an external memory card.
- According to estimates by Sweden's Telefon AB L.M. Ericsson, almost two billion people [predominantly in India and China] are projected to start using mobile phones in the next five years.
- Sandisk's recent conference call revealed that the company shipped more mobile card units in Q3 than all of 2006.
- Apple's iPhone has set the standard with the minimum amount of memory required in the handset. It no longer sells the 4 GB model, only the 8GB one. Handset manufacturers that wish to compete are left with little choice but to do the same. All great news for SanDisk's business.
In short, the major growth driver will be the next generation of multi-functional cell phones. Just as I did with my new Nokia I95, users will realize the potential of these devices (high resolution cameras, mp3 players, GPD capabilities] and will demand significantly larger amounts of memory than we are using today.
Another secondary catalyst to watch for is the trend towards laptops with NAND flash memory drives (called Solid State Drives or SSD). These are more reliable than conventional hard drives with moving parts, and allows for a laptop that is significantly lighter. While some high-end models are already on the market, the costs are still too high to hit the mainstream. If prices come down further, SSD drives for laptops may become a secondary growth driver for Sandisk's business.
So What do the Numbers Say?
SanDisk's Balance Sheet is strong:
- Total Assets are more than 3 times total liabilities
- Benjamin Graham typically demanded that Long Term Debt should not exceed Working Capital (or Net Current Assets defined as Current Assets minus Current Liabilities). Sandisk's Balance Sheet meets this criteria.
- 23% of the current share price is cash or short-term securities ($2.31 billion or $9.73 per share).
- Earning in the past 5 years grew on average by 34%, and analysts' consensus has earnings growth in the next 5 years to be 22.75%
- Quality of Management metrics such as Return on Equity (ROE) and Return on Assets (ROA) are not high. Warren Buffett typically seeks businesses with an ROE that has been consistently greater than 12%. This suggests that the business is able to create value for its shareholders. Sandisk's ROE and ROE are under 5%. This is what you would expect from a business that must continuously plow back its earnings into building new manufacturing plants.
So what's the Business Worth?
According to Thomson / First Call, 12 analysts have coverage on Sandisk, and the target valuation ranges between $45 to $75, with a $65 median. Lazard Capital's Daniel Amir, who in my opinion has a good handle on the company values the business at around $63 a share. Citi's Craig Ellis has a target price of $65
If I performed a discounted cashflow solely on the royalty license revenues (which are in essense free cashflow) with the intial year being $450 million, 10% growth for years 1 to 5, and 3% growth for years 6 to 10, and a discount rate of 10%, I receive a valuation of $42 per share - or the current share price.
A complete valuation of the business, with estimated average 5-year growth of 19%, provides a valuation of around $58 per share. This is a 33% discount to the current share price.
So What do I think?
Why did the price drop 15% on Friday after Thursday's earnings call? I'd love to be able to put my hand on my heart, look you in the eye, and give you a good reason. But I can't. I've never really known how to properly explain the sudden movement of crowds. Expectations not met? Perhaps. Lower Q4 guidance? Who knows? Fortunately, I do not assess a business because of price action. Instead, I try to assess what the free cashflows are going to be in the years to come, and formulate a rough valuation on that basis. This is not an easy thing to do with SanDisk or with this industry.
Usually, I try to look at a 10 year history of a business before making an investment. When you analyze 10-year financials you get a feel how a business has been able to handle tough periods, whether it has been able to protect margins, or whether management has been able to create value for shareholders. Looking at Sandisk's 10-year financials is useless. The environment is changing too quickly. 3 years ago, a mobile market did not exist for mobile memory cards. Today it is Sandisk's leading product. This year there is barely a market for Solid State Drives - but that may quickly change.
This is not a business that Warren Buffett would invest in. He would place it on the 'too hard' pile as it is capital intensive, because of its changing nature, and because of its low Return on Equity.
I however, see value here. The NAND memory market is forecast to experience continued explosive growth and while it is true that the industry is highly competitive, SanDisk is extremely well-positioned to capitalize from this growth. It has demonstrated that it is able to enter new markets quickly, and because of its vertically integrated model is able to quickly establish a dominant position.
If I was to place my bets on one player in this industry, it would be SanDisk, and recent share price weakness presents an opportune entry point.
Please read the Legal Disclaimer.
Disclosure: Avi Ifergan has an interest in SNDK.
Avi Ifergan is the Managing Partner of Israel Value Funds (www.israelvalue.com) an Israel-based investment partnership that follows a disciplined and long term oriented Value Investing approach, with a primary focus on Israeli public companies. Avi is a former equity analyst, corporate advisor and serial entrepreneur. These days he spends his time teaching economics at a major Israeli university and seeking value investing opportunities. He very much appreciates your feedback at avi@israelvalue.com.
3 comments:
Hi Avi,
first I'd like to thank you for sharing - it is a good blog, not much do we have in Israel (Rafi Nelson's is the second i could think of).
Second - how come you didn't write about the P/E of SNDK which is above 100?! this is one of the first rules of value investing - isn't it?
it needs some explanations...
thanks in advance, Lior
Hi Lior,
Thanks for visiting and thanks for your compliments. You are right - It is unfortunate that there are fewer Value Investors in Israel. Perhaps it is a reflection of the infancy of the market - or the temperament of the average Israeli investor.
Rafi Nelson's blog is one of the best Value Investing blogs on the web today. It is unfortunate that it is only in Hebrew. His columns sometimes appear on Globes. I think he is organizing a delegation to Omaha next May for the BRK AGM. I intend to attend so I may join his group.
Regarding the PE: You are absolutely right. I did not refer to any valuation multiple. I am not sure how meaningful mentioning a PE 100 is. On the face of it - it looks extremely overpriced. Personally, I am not a big fan of the PE multiple. for 2 reasons:
(1) On the top part of the equation, the 'P' can often be misleading. In SNDK's case the 'P'is 9.9B. But that's not really the case because there's $2.3B of cash & ST securities sitting on the Balance sheet. If you were to buy the entire company for $9.9B you would be getting back the $2.3 B in cash. The real 'P' is around $7.6B
(2) On the bottom, the 'E' also does not reflect normal operations or free cash flow which counts most. The 'E' needs to be normalized - which is what Buffett tries to do. He adds back all non-cash items - such as depreciation, one-time charges - such as SanDisk's acquisition costs related to FAB 4, and higher than normal R&D costs. SanDisk also entered invested heavily to gain dominance in European and Asian markets - and so marketing and sales expenses were higher than normal.
To asses value I prefer to use Free Cash Flow Yield. This is calculated by dividing Free Cash Flow by the Market Cap. In Sandisk's case this is $400m / 9.9B (or more exactly $7.6B if I remove cash). The answer is a little above 4%. I can compare this return to a long-term bond - which, in my opinion provides a more meaningful measure of value.
Also, as I mentioned in this post - this is not a typical value investment - it is too difficult to forecast cashflows out into the future - the industry landscape is changing too quickly.
All the very best,
Avi
Again, thanks for pointing this out.
Thanks for the serious answer.
Well you are right, it is not a value investment, cause even if i measure it by the things you mentioned, it may be a good company, but not "on sale" like Graham teaches us to find...
I appreciate the honesty, not trying to argue just for arguing.
Thanks again for the Blog.
Lior.
Post a Comment