Playing in recent IPOs is usually a sucker’s game . Numerous academic studies (.pdf) document the underperformance of the average IPO. To make things even worse, unless you are a large institutional investor chances are your only opportunity to buy one of the few great companies that comes down the pike will come in the open market following the first-day pop, after much of the easy money has been made. However, sometimes a new issue is poorly understood by the market and/or becomes public at an inopportune time, giving opportunistic investors a chance to buy a stake in a promising growth business at a discounted price. Alpha & Omega Semiconductor Ltd. (AOSL) appears to be one of these cases.
AOSL is a supplier of power semiconductors, primarily power MOSFETs and power management ICs, primarily to the global PC and consumer electronics industry. The company’s primary customers include virtually all the major PC ODMs and OEMs: ASUSTeK, Dell (DELL), HP (HPQ) and Samsung (GM:SSNLF) as well as Compal, Hon Hai, Foxconn, Quanta, TPV Technology/AOC and Wistron. Although AOSL’s headquarters are in Silicon Valley, the company has an unusual trans-national structure where a Bermuda holding company owns a Cayman Islands subsidiary which in turns owns subsidiaries in the U.S. (R&D, HQ), Taiwan (primarily sales & marketing) and China (manufacturing and in-house packaging and test). AOSL is a fabless company, which means it contracts out the manufacturing of its products to 3rd party foundries, but does own some of its own packaging and test capacity.
This structure enables AOSL to leverage world-class U.S. design talent, ultra-efficient Chinese manufacturing, packaging and test and service the PC Island kingdom of Taiwan with an aggressive local presence. Importantly, this structure also allows the company to pay only 10-15% in taxes. Founded in 2000, AOSL has consistently outgrown its primary competition such as Fairchild Semiconductor (FCS), International Rectifier (IRF) and Vishay (VSH) and should report around $300M in sales for its fiscal year ended 6/30/2010. The company went public in late April 2010 at $18.00 per share, and soon traded down to below $15.
At just under $15, AOSL appears to offer investors that rare trifecta: rapid growth, a favorable industry backdrop and a valuation Benjamin Graham could love. Although power MOSFETs are generally considered to be a low margin commodity business, the company has managed to enter the market from scratch and rapidly gain share, posting a five year revenue CAGR (for the year ending 6/30/10) of 23.7% compared to the much weaker growth of Fairchild (2.3% expected for five years ending 12/31/10), International Rectifier (-5.4% expected for five years ending 6/30/10) and Vishay (3.3% expected for five years ending 12/31/10). Industry contacts indicate the company continues to execute and win new designs, so the above-market growth is expected to continue for the foreseeable future.
As for the industry backdrop, AOSL is primarily tied to the global PC market and to a lesser extent the consumer electronics market. The PC semiconductor market is currently very strong, as industry participants struggle to meet high demand. The economic rebound, secular drivers such as Windows 7 and iPads and inventories that were slashed during the downturn are combining to fuel rapid growth. While the pace should slow down, barring a double-dip recession, the PC market is forecast to maintain strong unit growth in 2011.
For some reason, AOSL trades at a valuation more befitting a slow or negative growth company. Deutsche Bank led the IPO and recently published an initiation report calling for $1.70 in EPS for FY 2011. My estimate is for $1.87, but whether the P/E multiple is 8.7x or 7.9X the multiple is very cheap for any company, let alone one with AOSL’s kind of growth and balance sheet (around $4.75 of net cash). Admittedly, Fairchild and Vishay trade at similar multiples, but they are much slower growing and more capital intensive as they own their own manufacturing capacity. For investors more attuned to cash flow, I estimate AOSL will do $62M in EBITDA and $37M in free cash flow in fiscal 2011. This is a FCF yield on forward projected enterprise value of 18%. To demonstrate AOSL’s historical capital efficiency, note that the company raised $30M in a venture capital round led by blue-chip venture firm Sequoia Capital in 2006, and basically never touched the money. Sequoia did not sell any shares in the IPO.
Why So Cheap?
All of this begs the question: Why are investors getting such a seemingly attractive opportunity? I believe the valuation discount exists for three reasons: (1) The SOX has sold off due to fears of a semiconductor cycle correction or even a double-dip recession; (2) AOSL had the poor timing to go public right as the SOX was selling off during the May volatility, causing investors who bought the IPO hoping for a quick pop to sell; and (3) investor perception that the company is in a low-margin, deeply cyclical commodity business. Investors need to form their own opinions about (1), but all the evidence I’ve seen points to a continued robustness in the PC market and long lead times for MOSFETs. As for (2), I think AOSL’s poor timing is the genesis of the current opportunity. IPOs are largely sold to mutual funds that own a large number of stocks and don’t really know their individual stocks in detail, so I am usually happy to buy what they are selling and vice versa. I also believe that AOSL suffered from the fact that there were no published analyst research reports or earnings estimates to rely on during the sell-off in May. (3) is true to an extent, but AOSL must be doing something right to take so much share in the past 10 years. I suspect that they have a competitive advantage that lies in their design prowess, ability to do business with all of the major ODMs as a pseudo-local Taiwanese company and efficient fabless operations. These advantages are not as easy to explain in a 30 minute IPO roadshow meeting but should not go away overnight.
As long as we don’t experience a double-dip recession that negates the forecasts of PC market growth, AOSL should outperform the market and the SOX as the market recognizes its above-industry growth and dirt cheap valuation. I am hoping that AOSL performs like DIOD (another commodity semiconductor supplier who rapidly gained market share) performed from 2004 to 2007.
Disclosure: Long AOSL