In the last two months, semiconductor stocks have experienced a ~16% correction in equity valuation, and with more reasonable valuations, we recommend investors begin compiling a potential list of industry takeout targets. According to the most recent GCA Semiconductor Sector Report, 10 years ago, there were 130 public semiconductor companies; at the end of 2018, there were 72; and we think in the next 3-5 years, there will be half as many publicly listed companies versus current levels.
Thus, we view those companies which have efficiencies such as technical knowledge, cost, or scale may be takeout targets by larger industry participants in the future. Many semiconductor executives have often complained about industry valuations, suggesting they remained too high for serious M&A consideration. With many semiconductor companies on our list, well below the median EV/Sales takeout multiple of ~3-3.5x, we believe there could be more potential deals announced in 2019 and beyond.
However, we would note that recent increased tensions with the U.S./China trade dispute, along with Chinese regulators not approving Qualcomm's (QCOM) acquisition of NXP Semiconductors (NXPI), more than likely reduce historical takeout multiples due to the added risk of not receiving Chinese regulatory approval.
We have created a screen identifying semiconductor companies trading below ~$6B market capitalization, with NTM EV/Sales multiples at or below ~3x, with NTM Forward P/E multiples <15x. Those that met both criteria have been identified as potential future takeout targets. Within our list of 10 companies which met requirements, we believe investors should look closer at Lumentum (LITE) and Rambus (RMBS).
Our Methodology for Identifying Potential Semiconductor Takeout Targets
Historical data regarding semiconductor takeout multiples provided by the GCA Semiconductor Sector Report, and semiconductor merger and acquisitions announced in 2019, suggest the median takeout multiple based on the next twelve months' (NTM) Enterprise Value (EV)/Sales is ~3-3.5x. However, within semiconductor subsectors, multiples can vary to <2x for automotive/testing/manufacturing focused companies, or ~6-7x for highly technical datacenter technology.
Source: GCA Semiconductor Sector Report
Nvidia's (NASDAQ:NVDA) announced acquisition of Mellanox Technologies (NASDAQ:MLNX) for $125/share (~5.5x NTM EV/Sales) seems potentially at risk based on the significant gap between shares currently trading at $110.50, equaling a 13% discount to the announced acquisition price. Infineon's (OTCQX:IFNNY) acquisition of Cypress Semiconductor (NASDAQ:CY) (~4x NTM EV/Sales) could see potential U.S. CFIUS issues given the company's exposure to military programs, as well as being the last major U.S. supplier of SRAM technology. SRAM is a key component to many data center devices such as routers and switches. The only other major U.S. SRAM supplier, besides Cypress, was Integrated Silicon bought in early 2015 by the Chinese.
Given the potential downward pressure on historical median takeout multiples for the broad semiconductor industry, as well as the added risk to deals closing due to potential regulatory headwinds, we have created a screen which identifies semiconductor companies currently trading below ~$6B market cap, with NTM EV/Sales multiples at, or below ~3x, with NTM Forward P/E multiples <15x. Both Cypress and Mellanox were trading below 3x EV/Sales prior to press reports of a potential acquisition, or plans for strategic review, which provides a limited data set backtesting of our methodology.
Grinder Capital Potential Semiconductor Takeout Targets
Utilizing our methodology, we have identified 10 potential takeout targets within various subsectors of the semiconductor industry, which meet both criteria. Companies with NTM EV/Sales and NTM Forward P/E multiples equal to or less than requirements were: Canadian Solar (CSIQ), Diodes (DIOD), ChipMOS (IMOS), JinkoSolar (JKS), Lumentum, MagnaChip (MX), Rambus, SunPower (SPWR), Synaptics (SYNA), Tower Semiconductor (TSEM), and Vishay (VSH).
Secondary potential takeout targets that only met one of the two requirements were: Amkor (AMKR), Cabot Microelectronics (CCMP), Cirrus Logic (CRUS), DSP Group (DSPG), FormFactor (FORM), Himax (HIMX), ChipMOS Technologies, nLIGHT (LASR), MACOM Technology (MTSI), SolarEdge (SEDG), and Vivint Solar (VSLR).
However, while our screener may suggest a lower probability of future takeout, it doesn't work in absolutes. Thus, certain companies with strong technology may only meet one requirement but could still be attractive to larger companies based on the subsector.
We have provided investors a copy of our most recent screen (below) from which we identified the companies which either met one or all requirements and those that met neither. We would highlight that based on the subsector, our screener may suggest a company may not fit our model for a potential takeout target but could still be viewed valuable to partners, competitors, or other industry participants. These names are typically associated with the datacenter. One such name that we believe could become a future takeout target, but did not meet either requirement is Vicor Semiconductor (VICR). Vicor is focused on distributing power effectively in the data center, as well as developing advanced products targeting GPUs.
Quote from VICR 1Q19 Earnings Transcript
While Q2 demand for Advanced Products remains weak, our penetration of servers, supercomputing, and AI accelerators is gaining momentum with major design wins for NBMs and Lateral Power Delivery solutions entering production in the second half of this year. We're also seeing early traction for our Vertical Power Delivery systems. Owing to their superior power density, lateral and Vertical Power Delivery solutions are the solutions of choice for high-performance demanding processor applications, but figuratively AI accelerators.
At the occupational level, the transition from 12-volt to 48-volt power distribution is gaining momentum in the cloud and the automotive market segment. In data centers, [for] years after Google's pioneering initiative to convert from 12-volt racks to 48-volt racks, other hyperscalers did not follow through. However, confronted with the necessity to power GPUs and other 48-volt loads, bastions of legacy 12-volt power distribution systems have recently started to crumble and decide to convert to 48-volt within a few years.
Power distribution infrastructure developments in data centers and the cloud reflect corresponding developments in automotive where, for similar reasons, GPUs and AI for autonomous driving and general vehicle [occupational] requirements. 12-volt legacy energy storage and 12-volt power distributions are giving way to 48-volt. This set the stage for broad adoption of power distributional architectures, our conversion topologies, control systems and packaging technologies that Vicor invented and comprehensively patented over the last 15 years.
As our Advanced Products gain broad adoption in the data center and automotive markets, we continue to expand the performance gaps that sets Vicor apart from so-called competitive solutions, at the point-of-load, in AC front ends and in complete power systems. With the rollout of 4G power modules across our Advanced Product families, we're expanding our power density advantage to at least 2x, and in many instances for certain classes of products, as much as 5x that of the closest so-called competitive product."
Source: VICR 1Q9 Transcript
Given the company's focus on highly technical datacenter aspects, such as power distribution for GPU accelerators, GPUs for autonomous driving and AI, and general vehicle requirements, we wouldn't be surprised if the company received a takeout offer in the ~5-5.5x NTM EV/Sales multiple. However, at this time, based on utilizing our takeout screener, it did not meet the requirements to be listed as a potential target.
Source: Sentieo, Grinder Capital
Two companies which met both requirements and possess what we believe is technical know-how or IP that might interest a larger competitor or partner could be Lumentum and Rambus.
Lumentum is a leading provider of optical and photonic technology. We anticipate significant consolidation within the optical industry in coming years, but see Lumentum as a company which could offer technical know-how and additional product lines to an acquirer with customer overlap, which could offer a high level of operational efficiencies and scale. The company also operates in some consumer industries which may offer value to sell-off in a future transaction or could work with a company that also offers consumer-based products. According to the company's latest 10-K, its key customers in 2018 were Apple (AAPL) at 30% of sales, Huawei at 11%, and Ciena (CIEN) at 11%. Cisco (CSCO) was 12.4% of 2017 revenues, but was less than 10% in 2018.
Source: Lumentum Holdings 2018 10-K LITE 10-K (2018)
With a more reasonable valuation following the double impact of slowing sales of Apple products and the U.S. ban on Huawei, we believe that Lumentum is now properly priced for a potential suitor to become more aggressive in looking at a potential acquisition. With a current NTM EV/Sales multiple of 1.9x and a Forward P/E of ~9x, the current valuation is more than likely assuming a significant downturn in the company's future opportunities. We disagree, as datacenter speeds continue to increase and the move towards photonics increases, we believe Lumentum will remain a leader. The Apple business will likely ebb and flow, but has allowed the company to broaden its base from previously primarily only serving communications customers.
Assuming a median NTM EV/Sales takeout multiple of 3x, in line with broad semiconductors, the company, if acquired in the future, could potentially offer upside from current levels of ~55-60%.
Rambus is a more niche-based IP company. Rambus still holds a significant patent library which could be monetized through patent sales or utilized by an acquiring company to differentiate their products. Currently, Rambus still works within the memory business but has expanded into mobile encryption, ASIC development, advanced technology development, and its lighting business. Both the lighting and encryption businesses were acquired prior to the IP litigation setback, and to date, we don't believe either have achieved the targets laid out by prior management teams regarding growth opportunities and contribution. Thus, with several management teams in the last 7-8 years, little growth in equity valuation, and a large patent/technology library, Rambus seems like a perfect activist target or future acquisition target.
Rambus has a concentrated customer base per its latest 10-K filing.
"Our revenue is concentrated in a few customers, and if we lose any of these customers through contract terminations or acquisitions, our revenue may decrease substantially. We have a high degree of revenue concentration. Our top five customers for each reporting period represented approximately 49%, 55% and 63% of our revenue for the years ended December 31, 2018, 2017 and 2016, respectively. For 2018, revenue from Broadcom and NVIDIA each accounted for 10% or more of our total revenue. For 2017 and 2016, revenue from Micron, Samsung and SK Hynix each accounted for 10% or more of our total revenue. We expect to continue to experience significant revenue concentration for the foreseeable future."
Source: Rambus, Inc. 2018 10-K
However, we believe many of its current customers could benefit from technical innovation within Rambus, as well as continue to utilize and develop new products which incorporate some form of Rambus technology.
Assuming NTM EV/Sales takeout multiple of ~4x, which is near the low end for historical IP-based acquisitions, the company, if acquired in the future, could potentially offer upside from current levels of ~40%.
With more reasonable industry valuations following the recent pullback in semiconductor stocks, we see a much more favorable environment for further industry consolidation. While our takeout screener may capture a broad set of potential opportunities, we think that each must be looked at on an individual basis. We provided a more in-depth view on both Lumentum and Rambus, but also recognize companies that may have not met all screener requirements could still be potential takeouts in future periods due to technical expertise in the datacenter, such as Vicor Semiconductor.
Disclosure: I am/we are long VICR. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: We also own shares of Lumentum (LITE) and Rambus (RMBS) in one of our investment strategies (Long/Short). We own shares of ChipMOS Technology (IMOS) in our International Value Strategy. We are short Mellanox (MLNX) shares in personal accounts as well as our Long/Short strategy.
Investors should utilize our semiconductor takeout screener as a tool in their own overall research process. Grinder Capital is not recommending investors buy or sell securities based on this report. Investing involves risk.