With uncertain end market demand in the technology sector, coupled with competitive pressures, one wonders whether the recent outperformance of Texas Instruments (TXN) and Altera (ALTR) may pullback. Despite managements of both companies lower guidance, investors nevertheless remain confident about fundamentals. While the consensus is a "hold" for TI, it is a "buy" for Altera.
From a multiples perspective, TI, however, is the cheaper of the two. It trades at a respective 12.8x and 14.8x past and forward earnings while Altera trades at a respective 14.3x and 18.4x past and forward earnings. In addition, TI also offers slightly less volatility and a dividend yield that is around 130 bps higher than its competitor at 2.2%. With Xilinx having a dividend yield at 2.3%, Altera investors may shift to another semiconductor that offers less risk.
At the third quarter earnings call, Altera's CEO, John Daane, noted some of the company's strengths:
Our success in 40-nanometer, with over 65% market share, continues in the 28-nanometer node. We shipped over $2 million in Stratix V revenue in the third quarter, a similar ramp to Stratix IV during the same time period, and we estimate our 28-nanometer market share to be about 70% thus far. Within the quarter, we shipped Stratix V GT devices, with industry-leading 28 giga bit per second transceiver capability, extending our high end FPGA leadership. We also announced our fourth and fifth 28-nanometer families, name Cyclone V SoC and Arria V SoC, with embedded dual ARM Cortex-A9 microprocessors, as part of our broader embedded initiative, and released the industry's first Virtual Target for processor software development on these products.
Forbes magazine named Altera as one of the world's top 100 most innovative companies, for our culture of innovation that drives business success. It is through innovation that we deliver PLDs to replace ASICs, microprocessor solutions for the embedded market, and complex intellectual property blocks to replace ASSPs, opening a combined opportunity space, 10x larger than the total PLD market. Our goal is to continue to grow at a rate of twice the semiconductor industry.
Yet, the company is facing weakness broadly on all fronts, save U.S. military. Clients are reducing inventories due to uncertainty in end market demand, largely the impetus for expectations being reduced to a 16% sequential decline. Even still, semiconductors are benefiting from strong holiday demand for smartphones and tablets. As for Altera, specifically, the company is gearing itself for strong returns through wireline and wireless infrastructure build-outs that represent nearly one half of revenue. Add the fact that Altera is gaining market share in PLD from Xilinx with a ramp-up of 40 nm and you have some upward potential.
Consensus estimates for Altera's EPS are that it will decline by 6.8% to $2.32 in 2011, decline by 12.5% in 2012, and then grow by 18.7% in 2013. Assuming a multiple of 20x and a conservative 2012 EPS of $1.97, the rough intrinsic value of the stock is $39.40, implying 5% upside. If the multiple were to decline to 16.5x and 2012 EPS is 10.3% below consensus, the stock would fall by 20%. As TI trades at a much lower multiple, the bear case is concerning.
TI is now anticipating EPS and revenues as low as $0.21 and $3.19B, respectively. The tech company has had impressive design wins from Toshiba, Samsung, and Google (GOOG). As customers cut back on inventory, however, margins face substantial risk.
Consensus estimates for TI's EPS are that it will decline by 19.1% to $2.12 in 2011, decline by 1.4% in 2012, and then grow by 23.9% in 2013. Assuming a multiple of 15.5x and a conservative 2012 EPS of $2, the company is roughly at fair value.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.