Texas Instruments (TXN) is a major player in the semiconductor business, which constitutes roughly 96% of the firm's revenues. In 2011, the company was ranked third in semiconductor revenue, behind only Samsung Electronics and Intel (INTC), with 4.5% market share.
The company produces a diverse set of semiconductor devices, from system-on-chip products that are used in smartphones and tablets to microcontrollers for the most low power of embedded systems. In what appears to be a broad sector rotation shift in favor of semiconductor firms, should TI be your horse in the race? Let's take a look:
OMAP: Smartphones And Tablets
TI develops a product under the "OMAP" (Open Multimedia Applications Platform) brand. These are SoCs (system-on-chip) built on ARM Holdings' (ARMH) core designs, and they have seen fairly wide adoption in the tablet and smartphone industry. For instance, the Samsung Galaxy Nexus, the Amazon (AMZN) Kindle Fire, Barnes & Noble (BKS) Nook Tablet, and the Research In Motion (RIMM) BlackBerry Playbook all use an OMAP chip.
However, it's important to make a number of observations. First of all, IDC predicts that Google's (GOOG) Android-based smartphones will see their market share peak this year (with roughly 61% market share). This isn't too troubling since it's a vast majority, and secondly it is predicated on a strong proliferation Microsoft's (MSFT) Windows Phone platform around which there are a number of uncertainties.
More seriously, though, is the heated competition in the Android space (since TI does not supply applications processors for either Windows Phone or Apple's (AAPL) iPhone). Intel threw its hat in the ring with its "Medfield" platform that proved to be quite competitive against current ARM chips, Qualcomm's (QCOM) Snapdragon S4 is dominating the CPU benchmarks (and winning Microsoft's heart as the SoC of choice in the current and upcoming Windows Phones), and NVIDIA (NVDA) is growing in strength with its SoCs finding homes in Google's Nexus 7 tablet and Microsoft's "Surface" (and the rumors have it that NVIDIA's Tegra 3 will take TI's place in the next "Kindle Fire" iteration).
Another point of concern is that TI's SoCs do not significantly differentiate themselves in the smartphone/tablet SoC front. NVIDIA designs its own graphics cores while using the off-the-shelf ARM CPU cores, and Intel and Qualcomm use their own CPU cores with the latter also using custom graphics cores. So there will certainly be some competitive concerns going forward as TI's smartphone/tablet SoC's are the least differentiated of the heavy hitters'.
My thesis appears to be bolstered by the results in the most recent quarter in which revenues in the division called "Wireless" (this includes the OMAP) were down 39% year-over-year and 8% sequentially and that this division posted an operating loss of $51M compared to the year ago period of an operating profit of $82M. I believe this is due to fairly significant market share loss in light of strong competitors (note: NVIDIA's recently released earnings call confirms they are taking significant share in this space). Going forward, I believe competitive pressures will not only hurt market share, but will also shrink margins.
Embedded, Analog, and Other
Aside from the more well known SoCs for smartphones and tablets, TI has a very diverse lineup of other products including microcontrollers and DSPs for embedded products categorized as "embedded processing". Further, the company categorizes each of analog and "other" devices separately.
It's unsettling to note that on a year-over-year basis in the most recent quarter, embedded processing revenues were down 15% year-and operating profit was down 64%. The company maintained that the decline in revenue was a result of a decline in communications infrastructure products sold, but that revenue from automotive applications increased. This decrease is cause for alarm as operating profit decreased much more sharply than revenues did.
The analog division is the largest by revenue for the company, and in fact, the company saw a 13% increase year-over-year and a 7% increase sequentially in this division. Operating profit was down 2% sequentially, but up 30% year-over-year. It seems, however, that the year-over-year revenue growth was primarily due to the acquisition of National Semiconductor.
Finally, the "other" which includes calculators, custom ASICs, and DLP products saw a 4% year-over-year revenue decline coupled with a 32% year-over-year profit decline. However, the sharp operating profit decrease was due to acquisition and restructuring charges, so I am not convinced that this division of the company is particularly in trouble.
So it seems that the company's non-Wireless divisions are slowing down as well, implying that there's more going on to TI's disappointing Q2 than simply competition in the phone and tablet space. It is likely a combination of competitive pressure coupled with macroeconomic problems that brought about declines in the other, less well known areas that TI services.
The Company's Valuation Is The Real Problem
While the company seems to be running into headwinds in each of its divisions, it is likely to recover from them due to a strong product lineup overall most likely being hamstrung by macroeconomic conditions. However, what I do think that investors should be skeptical about is the valuation of the company. TI currently trades at 21.58 times past earnings (versus an industry median of roughly 15). Further, the company trades at 2.5x sales and has a PEG ratio of 2.16, all implying that the company is expensive relative to its peers, especially as most of its divisions have shown revenue decreases year-over-year. The large debt position of $4.7B against the cash position of $2.33B also does not inspire confidence.
In addition, the company's dividend of $0.68 per share (roughly 2.4% on an annualized basis) is decent for a tech company, but in this area there are, again, better values.
For growth options in the semiconductor sector with a focus on the widely loved smartphone and tablet markets, NVIDIA trades at roughly 17x past earnings and has a strong $3B cash position against a roughly $9B market cap. Qualcomm trades 20x past earnings and has a net cash position of $15B against a market cap of $105B. While Intel's success in the mobile arena is not yet assured, I would say that Intel's net cash position of $7B, coupled with a strong history of share buybacks and dividends, would also be a potential play in the space at 11x past earnings.
Texas Instruments is still one of the largest semiconductor companies by revenue and by market capitalization. While I think the company itself is solid, I do not believe that the current price represents a particularly good value relative to peers in the semiconductor business. The one sector that is likely to be considered "hot" and used as the justification for a growth-stock-like price-to-earnings, Wireless, is threatened by strong competition in the space, and as such, I would have a difficult time recommending the company at its current prices and would wait for a better entry point before opening a long position.