Since I first published my bullish article on Marvell (MRVL) here, the stock has risen by 14.5%. The Street still remains bullish on the stock and rates it near a "buy." Based on my multiples analysis, DCF model and review of the fundamentals, I still find room for appreciation Marvel, in addition to Broadcom (BRCM) or Intel (INTC).
From a multiples perspective, Intel is the cheapest of the three. It trades at a respective 11.4x and 10.5x past and forward earnings with, by far, the highest dividend yield at 3.1%. Broadcom and Marvell, meanwhile, trade at a respective 12.4x and 13.9x forward earnings. To put this into perspective, consider that Intel is valued at roughly two-thirds of its historical average PE multiple.
On the fourth quarter earnings call, Broadcom's President and CEO Scott McGregor noted solid results:
"Broadcom performed well in the December quarter, with somewhat stronger revenue than we anticipated. Our quarterly revenue was $1.82 billion, above the guidance provided in December at our Analyst Day in New York.
At the outset of the year, we established a plan to outgrow the industry while delivering results within our target operating model. I'm pleased to report that we delivered on this goal. Broadcom's annual product revenue increased almost 9% year-over-year, significantly better than the overall industry, which grew in the low single digits.
2011 non-GAAP product operating margin came in at 20.9%, within our targeted range of 20% to 22%. As a result, Broadcom delivered record cash flow from operations of more than $1.8 billion, driving record cash balances of $5.2 billion."
The company is able to drive up margins through outsourcing chips from foundries and has solid diversification to hedge against weakness in any one particular segment. Broadcom has further shown strong execution in takeover activity, which has helped the semiconductor improve scale. Dune, Innovision, Teknovus have all meaningfully added to the company's product diversity. With that said, it is concerning how the company's top five customers have accounted for no less than one-third of business over the last four years. Delays in the Nokia EDGE may make buyers weary about future business.
Consensus estimates for Broadcom's EPS forecast are that it will decline by 4.2% to $2.77 in 2012 and then turn around to grow by 10.8% and 7.5% in the following two years. Modeling a three-year CAGR of 4.5% for EPS and then discounting backwards by a WACC of 9% yields a fair value figure of $42.46, implying 12.7% upside.
By my calculations, Marvell has similar upside. It is the leading supplier of chips in HDDs, but much of its success hinges on uncertain R&D. On the other hand, its strong relationship with Western Digital (WDC) helps improve the sustainability of the revenue streams. In regard to the Thailand flooding, investors stand to significantly gain from the irrationally myopic outlook of the market. Bottom line: secular trends are strong in end market demand, especially given the explosion of data demand. And the firm has an early-mover advantage in handsets for the Chinese market. With that said, design losses at Research in Motion (RIMM) will hamper value creation. Free cash flow is also expected to decline by around $330M to $777M in 2013. Non-storage markets are further particularly vulnerable to competitive pressures.
Consensus estimates for Marvell's EPS forecast are that it will decline by 24.4% to $1.24 in 2012, decline by 5.6% in 2013, and then grow by 28.2% in 2014. Of the 32 revisions to estimates, all have declined for a net change of -3.9%. Assuming a multiple of 16x and a conservative 2013 EPS of $1.14, the rough intrinsic value of the stock is $18.24, implying 11.4% upside.