Semiconductor firm Altera (ALTR) is well positioned to show strong growth in revenue within the next 2 to 3 years. Because of the high gross margins enjoyed by the duopoly which it shares with Xilinx (XLNX), a large portion of the revenue increase will flow to the bottom line, with a predictable effect on share prices. The company has excess current assets, and a strong commitment to returning cash to shareholders.
Digital integrated circuits come in three broad categories: processors, memory and logic. Processors and memory typically rely on standard architecture, while logic requires customization. Logic devices come in three types: PLDs (programmable logic devices), ASICs (Application specific integrated circuits) and ASSPs (application-specific standard products).
Altera and Xilinx design and market PLDs. PLDs include FPGAs (field programmable gate arrays), CPLDs (complex programmable logic devices), and programmable SoCs (system on chips).
Both companies assert that technological factors favor PLDs and specifically FPGAs over ASICs and ASSPs, and both seek to exploit this advantage by replacing or displacing the weaker technology. While there has been discussion of whether this displacement is in point of fact taking place, Altera notes that both they and Xilinx have advanced two slots in the semiconductor hierarchy since 2008, from 6th and 4th to 4th and 2nd, respectively.
The size of this opportunity is huge. From Altera's 2012 10-K:
Based on publicly available data and information derived from Gartner Dataquest, an independent research firm, we estimate that the PLD market was approximately $4.5 billion in 2012. We also estimate that the combined ASIC and ASSP market in 2012 was approximately $101 billion, comprised of the ASIC market at approximately $21 billion and the ASSP market at approximately $80 billion. Due to the low priced nature of some very high volume ASIC and ASSP applications, including mobile phones and certain consumer applications such as PC-related equipment, video games and portable media applications, we do not believe that the entire ASIC and ASSP market is available for displacement by PLDs. As such, we estimate that the PLD accessible ASIC and ASSP market was approximately $50 billion in 2012 ($13 billion in accessible ASICs and $37 billion in accessible ASSPs) and that this represents significant PLD growth potential. (emphasis added)
Growth has been very uneven. There is a long cycle involved in designing the chips, developing software and training end-users in the programming, securing orders and ultimately delivering products. What I see is a cycle of R&D investment followed (with considerable delay) by ramping production and large profits based on operating leverage.
Gross margins run in the 60% to 70% area. As revenue increases in an up cycle, it blows past relatively stable R&D and SG&A expenses and pours down to the bottom line.
The bullish hypothesis here relies on the displacement of ASICs and ASSPs, for growth, and on operating leverage, for rapidly increasing profits. Mr. Market generally provides PE multiple expansion as a kicker when profits increase dramatically.
Working the Numbers
The technological case is complex, and further detailed analysis is beyond the scope of this article, and the skills of the author. I refer readers to Altera's 11/18/2013 Analyst day presentation, which contains a wealth of information.
Altera provides a business model, as well as guidance for 2014, in the November 2013 presentation.
As a method of projection, I take 2014 R&D guidance of $433 million, increase it at a growth rate of 5% for two years, and then divide by 20%, which is the center of guidance for the ratio of R&D to revenue. This procedure estimates 2016 revenue at $2,386 million, to which I apply a 33.5% operating margin (also from guidance) to develop earnings of $799. Taxing that at 12.5%, and dividing by 323 shares outstanding, EPS works out to $2.16.
Altera has $8 per share of current assets above the 2:1 ratio I consider optimal, and above what is required to operate the business in a prudent and orderly manner. Applying a PE of 17 to this growth situation, and adding $8 excess current assets, I arrive at a target price of $44 per share, by the end of 2016.
Capital appreciation, from a recent price of $31.87, if the scenario works out, will be 11.5% per year, to which one can add the dividend, currently at 1.9%.
Duopoly with Uneven Growth
As a way of understanding the trends in this business, here's a ten year revenue table with growth percentages, both year over year and since 2003:
Management mentions the 2010 episode fondly in recent presentations, and they sound like they're looking for something similar to develop 2 or 3 years from now. If that occurs, operating leverage will create large profits, as it did in 2010 and 2011.
It should be noted that on a ten year basis ALTR is gaining share vs. XLNX. It's a seesaw battle, with both sides claiming victory. They're competing on performance, not on margin. It's important, when looking at a duopoly, to be aware of what is the basis of competition. Both companies have a business model that calls for high gross margin, due to the heavy and ongoing cost of R&D.
Doing Something for Shareholders
Altera pays a dividend, and has increased it for over 6 consecutive years, most recently by from 10 cents to 15 cents quarterly. Management explicitly states their intention to be shareholder friendly in this regard, as well as by buybacks, which are to be done opportunistically. I interpret that to mean selectively, a good thing, since this is the type of stock that can become over-valued at times.
This stock is suitable for dividend investors who are more focused on growth than immediate income. Buying at recent prices, an investor has credible reasons to anticipate increased dividends and eventual share price appreciation.
Intel 14 nanometer FinFET
In August last year, Intel (INTC) and Altera announced an agreement whereby Altera will be using Intel as a foundry at the 14 nanometer node, which is the next battleground. In the press release, Altera presents the benefits in no uncertain terms.
"Altera's FPGAs using Intel 14 nm technology will enable customers to design with the most advanced, highest-performing FPGAs in the industry," said John Daane, president, CEO and chairman of Altera. "In addition, Altera gains a tremendous competitive advantage at the high end in that we are the only major FPGA company with access to this technology."
Altrera's most recent presentation utilized Intel's charts to make the case:
Altera in effect has been saying that they are there fustest with the mostest, and announcing victory in a battle that has yet to be fought. Xilinx CEO Moshe Gavrielov not surprisingly holds a contrary opinion:
So Intel has great technology. TSMC is absolutely the best foundry in the world and the best foundry in the world manifests itself through the process technology, the service levels, the business terms, capability. We had options and we looked at all of the options in each of the generations that we decided to continue with TSMC that has definitely proven itself with 28, it's proving itself 20, and we expect to have an incredibly competitive product offering at the next node. I believe that this Altera quoting Intel and then Intel quoting Altera and this vicious cycle of patting yourself in the back is very amusing but there was a similar set of claims at 28 which we've basically blown out of the water, 20 we feel very confident. (Emphasis added)
Without knowing the personalities involved, it's hard to read into this. I will say, that if the product is as good as Altera/Intel say it is, sarcasm and disparaging words aren't going to change things. Taiwan Semiconductor (TSM) is a formidable competitor to Intel in their battle to gain share in the foundry business.
Ultimately, if the displacement opportunity is as large as Altera claims on their 10-K, there will ample business for both members of this duopoly, and their rivalry will be much less important than their joint commitment to high gross margins.
Strategy and Tactics
The scenario developed above will take several years to play out, and is not a slam dunk. Taking a starter position and adding to it (or reducing it) as the price moves and additional information becomes available should be workable. Quarterly earnings and conference calls should be monitored, for progress on increasing revenue and for more information on the extent to which PLDs are replacing ASICs and ASSPs. The objective would be to bring the position to full size while in possession of good information that Altera is succeeding in gaining market share against the industry, while winning the battle at 14 nanometers against Xilinx.
The shares are optionable, and I own the Jan 2015 25 LEAPS calls. I do not plan to sell covered calls against this position, out of concern that share prices may advance suddenly if and when favorable developments make the news.
Role in Portfolio
Dividend growth comes in various flavors. Capital appreciation has always been a major consideration for me when investing in equities.
There has been some discussion to the effect that US companies in the aggregate are not investing in growing their businesses, but are more focused on financial engineering and cost cutting. Capex and R&D play second fiddle.
While retaining the dividend growth strategy, I'm changing the focus of my portfolio to include companies that are investing in their own future. Altera came up on a screen that searches dividend growth prospects for undervaluation based on R&D, capex (net of depreciation), excess current assets, acquisitions and buyback activity, weighting R&D and capex most heavily.
In this case, I believe the increased R&D expense in recent years is a tell for future growth, and money well spent.