According to the Taiwan Stock Exchange, Foxconn (2354.TT) Chairman Terry Gou has pledged 430m Foxconn shares as collateral for a loan from a three-bank consortium.  The transaction had a closing value of $1.117b US with expectations that the pledge could generate between 50% and 70% of the securities’ value.

We note that Mr. Gou, aside from his Chairmanship of Foxconn and Hon Hai (2317.TT) (he started the company in 1974), the world’s largest contract manufacturer, is also a personal investor in the Sharp Gen 10 Sakai LCD plant, and through a recent acquisition, is the de facto head of Sharp (6753.JP).  As is well known, foreign companies that sell their products in the US are under intense pressure to shift manufacturing back to the US.  President Trump has stated that various import taxes would be imposed on companies that don’t make such changes, pushing a variety of foreign companies to make noise about shifting manufacturing operations to the US.  Mr. Gou has pledged support for such a concept, but has also committed to the same in China, the biggest CE market worldwide.

We doubt this transaction signals the commitment of Mr. Gou’s resources to a particular location at this time, particularly as the Trump administration has yet to detail actual specifics about a tariff, and is facing opposition from a variety of US companies that have substantial operations overseas, but it does put at least some wheels in motion toward a decision by Gou held companies.  We believe it would be difficult to remain competitive should Foxconn decide to build an LCD fab in the US, despite the relatively high level of automation, with a more likely solution being some sort of highly automated assembly operation, but regardless of the near-term, Mr. Gou has not built a $6b empire by giving in to threats or coercion, and will likely make the final decision based on the absolute financial benefits of both locations and the concessions made by each government.  In either case we would expect both sides to claim victory…

Cree (CREE) indicated last week that it was terminating its agreement with Infineon Technologies AG (IFX.GR) concerning the sale of its Wolfspeed Power and RF Division.  The termination was based on the national security concerns of the Committee on Foreign Investment in the United States (CFIUS), which is an inter-agency committee that reviews the national security implications of foreign investments in US companies.  CFIUS is chaired by the Secretary of the Treasury, Steve Mnuchin (since last week) a former Goldman exec, hedge fund manager and film producer (strangely produced and acted in the recent Warren Beatty film, “Rules Don’t Apply”), and has representatives from 16 departments including State, Commerce, and recently Homeland Security.  We believe the transaction issues were based on both the raw material used by Wolfspeed for its device production, which is silicon carbide and Gallium Nitride, and the proprietary nature of some of its products, which are likely used in a number of defense related projects.   CREE will receive a $12.5m termination fee to soften the blow and will integrate Wolfspeed back into its growth plans.

The committee which operates according to Section 72 of the Defense Production Act (1950) and the Foreign Investment & National Security Act of 2007 whose key provision is the consideration of transactions that impact critical infrastructure as well as energy assets and critical technologies, with regard to the relevant country’s compliance with the US and multilateral counter-terrorism, non-proliferation, and export control regimes[i].  All transactions by a foreign entity must be reviewed by CFIUS with the focus whether funds or technology from a US business might be transferred to a sanctioned country.  The committee has 30 days to review a submission and either authorize the transaction or begin an investigation, which can take up to 45 additional days, at which point a decision must be made as to whether the deal is permitted or not. According to a 2006 review, “

CFIUS, which reported its 2014 review last February 2016, noted that in 2014 it reviewed 147 transactions (classified details), a follow-up investigation on 52 of those, with one transaction rejected and 12 withdrawn, with a total value of $2.84t.  This was the highest number of reviewed transactions since the Committee’s formation in 2009 and of the 358 reviewed transactions between 2012 and 2014, we calculate that 19% had China as the acquiring country.  We note that only one transaction has ever gone to the president for a final decision in the past, with last year’s block of Germany based Aixtron’s (AIXG.GR) takeover by a Chinese consortium the second.



[i] ‘‘(I) the transaction threatens to impair the national security of the United States and that threat has not been mitigated during or prior to the review of a covered transaction under paragraph (1); ‘‘(II) the transaction is a foreign government controlled transaction; or Deadline. VerDate 14-DEC-2004 15:32 Jul 30, 2007 Jkt 059139 PO 00049 Frm 00004 Fmt 6580 Sfmt 6581 E:\PUBLAW\PUBL049.110 APPS06 PsN: PUBL049 PUBLIC LAW 110–49—JULY 26, 2007 121 STAT. 249 ‘‘(III) the transaction would result in control of any critical infrastructure of or within the United States by or on behalf of any foreign person, if the Committee determines that the transaction could impair national security, and that such impairment to national security has not been mitigated by assurances provided or renewed with the approval of the Committee, as described in subsection (l), during the review period under paragraph (1); or ‘‘(ii) the lead agency recommends, and the Committee concurs, that an investigation be undertaken”

According to the most recent Harris Reputation Ratings Poll, Samsung’s brand has fallen from 3rd place in 2015 to 49th place in 2016, primarily due to the Note 7 fiasco.  In 2015 Samsung (005930.KS) was #3, ahead of both Amazon (AMZN) and Apple, with the 2012 #13 ranking being the previous low and the 1st year of inclusion in the poll, which measures the reputations of the most visible companies in the US, as perceived by the general public.  The 2016 results show Amazon as #1 and Apple as #5, with Wegmans (pvt) a $7b+ regional supermarket chain as #2, Publix Supermarkets (pvt) as #3, and Johnson & Johnson (JNJ) and #4.  Samsung is followed by #50, the US Postal Service, which gives you some idea of where public sentiment lies in the US.

Outside of TV, movies, and books, zombies are fiction, however if you walk down any street (our favorites are 3rd and 5th Ave. in NYC) you will see hundreds, if not thousands of zombies.  Not the ones that have decayed body parts or oozing wounds, but those who are looking down at their smartphones as they walk on sidewalks and streets.  These zombies, who are the bane of us who do not live with a smartphone in their hands at all times, are prone to walk into humans (luckily not biting), or more importantly across streets without looking up.  New York streets are ripe with bike messengers who have little regard for your texts about who had diner at which fancy restaurant with who, and less regard for your safety.  Taxicabs and delivery trucks don’t seem to care about anything other than getting through a busy interception so the zombie population, we are sorry to say is at serious risk.

That said, the Netherlands is taking steps to protect its zombie population and while signs abound, the county is beginning to install ground-level traffic systems in the town of Bodegraven.  These “light lines” are designed to attract the attention of pedestrians who are looking down at their devices and will indicate whether it is safe to cross.  While much local traffic is on bicycle, the system is adaptable to vehicular traffic.  We understand the next development will be a dome over the city that will eliminate most sunlight, helping zombies read their smartphones more easily...

As Apple continues to pursue facial recognition for its upcoming iPhone line, it seems to have spent a mere $2m (US – unconfirmed) for a small Israeli start-up RealFace (pvt) (formerly known as Pickeez) that specializes in recognition of facial characteristics.  The small firm has an app (Pickeez) that automatically chooses the best photos from any platform where the user might have images stored, but the ultimate application being part of the potential facial recognition that Apple is expected to use to eventually replace fingerprint identification.  The RealFace website is no longer accessable.

Apple has made a number of acquisitions in Israel relating loosely to facial recognition, with the $345m acquisition of PrimeSense (pvt) in November 2013, a company that makes 3D sensors, which help facial recognition understand depth.  Facial recognition will use existing camera technology to eliminate the space needed for a fingerprint recognition button on smartphones, where non-display real estate is considered a liability.

Innolux (3481.TT) has purchased $16.3m worth of equipment from Japanese Photomask producer Hoya (7741.JP) which we believe are destined for their F8 fab in Kaohsiung, Taiwan.  This fab is a mixed use facility, producing a-Si and LTPS LCD displays and small quantities of OLED panels.  While we cannot make the assumption that the masks are destined for a particular line, as the expanded OLED line is not expected to be in operation until early 2019, we would assume they will be used for existing LCD production.

Hoya was the original developer of photomasks and we believe remains the leader in the field today. These masks are used as the ‘masters’ during the process of transferring highly detailed circuit patterns onto wafers or other substrates during most semiconductor processes.

Xiaomi (pvt), the former darling of the Chinese smartphone brands, is said to be revamping its supply chain in order to allocate component purchases to a number of new suppliers, although they are expected to use use their internally developed Pinecone V670 CPU in a number of new models to be released next month.  Among the potential new suppliers was Truly (732.HK), a Hong Kong based display producer that has been in production since the early 1990’s.  Truly operates a Gen 4.5 OLED line that produces both rigid and flexible OLED displays for smartphones, wearables, and automotive applications, and is building a 2nd Gen 4.5 line of 60,000 sheets/month, although we believe the current fab also produces LCD based displays.  Down the road, we believe Truly is considering the construction of a Gen 6 OLED fab but we do not believe final plans or financing has been determined to date and also produces PM OLED[i] devices on a Gen 2.5 line.

Xiaomi, which was once the upstart Chinese smartphone supplier that pushed aside major brands to take the number one sort on the mainland, now faces competition from OPPO (pvt), Huawei (pvt), Vivo (pvt) and others, and has fallen to 4th place in the Chinese market.  Truly supplies, to varying degrees, displays to a wide variety of Chinese and overseas smartphone customers, so it would seem strange that Xiaomi would believe adding Truly would further differentiate their product unless they were able to garner a specially designed product from Truly, in particular an OLED product.



[i] Passive Matrix OLED – Different from current “Active Matrix” OLED which have a more sophisticated pixel addressing system.

On Feb 7 we noted that not only had LG Display (LPL) made a decision concerning the construction of a large format fab, but had already begun construction on the shell.  Today Digitimes reported that LG Display was constructing a Gen 10.5 TFT-LCD fab in Paju, South Korea.  The Digitimes article goes on to state that the 1st phase of the project will be slated for completion in February 2018 and goes on to say that the factory will begin converting the Gen 10.5 factory to OLED by 2020.

We believe that some of this information is incorrect and that the facility is being built as an OLED production plant, however, as almost every OLED display production line has a backplane line, producing what is essentially the electronics that controls the display at the sub-pixel level, there might be some confusion as to the ultimate purpose of the fab.  To simplify matters, we note below some of our expectations for the fab from our previous note:

We believe the fab will be based on oxide backplanes, at least for the large panel displays, and will likely have a capacity of 15,000 sheets/month (two phases), but we believe the configuration of the lines is still under development as there are a number of Gen 10 size configurations.  We expect a 2880 x 3080 substrate size, as it lends itself well to both 60” and 65” panels, but if small panel displays are also being produced, the substrate size might be different across the fab, or on a line by line basis. 

As there has been no OLED fab using such large substrates, the deposition equipment will have to be custom designed.  We believe the preliminary design work is being done by YAS (pvt), a Korean tool vendor that is known for linear deposition tools that allow for a variety of source scan modes, and given the unusually large size of the substrate, this will be a significant consideration.

Our OLED industry model and specifically the LG Display P10 fab were conservative, assuming a Gen 9 mixed use (OLED & LCD) fab.  Our assumption of 15,000 sheets/month is based on that mixed use assumption.  If the fab is fully dedicated to OLED production, the ultimate capacity could be larger than 15,000 sheets/month, but we maintain that assumption until we can confirm different plans from LGD. Our original estimation of the completion of phase one of P10 was for June 2018, which seems to be in line with construction plans, but we would not be surprised to see some delays in start-up, given the unique nature of the fab. 

While initially, Apple (AAPL) will be forced to use Samsung Display (pvt) for its OLED mobile device production, this new fab, should it follow the path of both rigid and flexible OLED lines, would become an alternative source for small panel flexible OLED by late 2018, and would help LG Display maintain its primary display provider position with Apple regardless of their transition to OLED displays.

We believe LG Display is still making decisions concerning the fab, such as the exact format, as we believe there is a chance that there will be other line configurations aside from the Gen 10.5 lines, and as such we have built into our model two Gen 10.5 lines and a single Gen 5 line for small format OLED production.  This would give LG Display the option to add small panel OLED capacity without having to convert existing LCD facilities at a later date, and could be used for specialized products outside of the normal OLED smartphone.  Regardless of the final formats, the new fab will give LG Display the capacity to further expand its OLED TV display production although we note that OLED remains a relatively small portion of LG Display’s total capacity.

Charging your mobile device is a pain, no matter whether it means carrying around a charger and search airports for AC outlets, or forgetting to plug it in before going to sleep, and while wireless chargers are become a bit more commonplace, you still have to remember to put the device on the charge pad.  Hopefully, that all will change over the next few years as a ’new’ technology will make charging almost a thing of the past.  No, not a perpetual motion generator, or an infinite life battery, but a way to make a whole room become a charger.

As we have mentioned in the past, Chinese appliance giant Haier (600690.CH) demonstrated a completely wireless TV at a show we attended a few years back, and while the product was never commercialized, we were quite intrigued with the concept, which is called magnetic resonance.  This concept uses highly tuned magnetic coils as ‘transmitters’ and similar coils tuned to the same resonant frequency as receivers, and as such, can transmit power across open spaces.  The concept is being adopted by a number of smartphone manufacturers, with Apple (AAPL) being rumored as one of the new participants.  The technology, which was used to power the Haier demo years ago, could be placed in a room, making that room into a huge charger.  Any device with a coil tuned to the transmitting frequency would receive power anywhere in the room. 

A small private Watertown, Massachusetts firm, Witricity (pvt), has been championing the concept, particularly for charging hybrid and fully electric automobiles, but the obvious connection to mobile devices of any type have helped the company find a an interesting group of investors, including Haier, Foxconn (2354.TT)), Intel (INTC), and Toyota (7203.JP), aside from a number of venture firms.  We expect the large scale adoption of the technology will have some challenges, particularly with those that are concerning that the ‘broadcasting’ of power in an unconfined space could cause problems similar to those alleged to be caused by smartphones, but power levels for standard devices are within safety standards that regulate all electro-magnetic devices including smartphones and wireless routers.  Imagine being able to charge a number of mobile devices just by leaving them in a particular room for some time, not having wires connected to your TV, not having to replace pacemaker batteries, or charging your electric car just by putting it in the garage!

While this is not the first time we have seen the Chinese press extol the virtues of the Chinese display industry and belittle the South Korea dominance of the market currently, a number of recent articles have pointed to the idea that China will become the largest display producing region by the end of 1919.  The battle, which is the way the Chinese government looks at the competition, will be won with spending ($42.7b US for new capacity), although they still expect China to remain behind South Korea in the OLED space, where they admit the Chinese OLED industry is ~3 years behind.

We expect a slightly different path, although with the same result, as we expect that on a raw capacity basis, China will overtake South Korea in 3Q 2020 rather than the end of 2019.  This will be accomplished by the phase two investments for a number of ultra-large format fabs, and the continued buildout of a number of small panel LCD and OLED lines.  Other than a meltdown in display pricing, and this would not necessarily slow Chinese fab expansion, we believe this is an inevitable conclusion, as China’s ultimate long-term goal has been to reduce or eliminate China’s dependence on outside display suppliers.  While this goal is important to the Chinese government, panel producers also understand that the Chinese market is not the worldwide market, and they must learn to effectively compete globally as well as in China.  This means matching or beating foreign competitors on not only price, but on quality, which has been an issue in the past.  As Chinese display firms become more experienced, we see the quality issue becoming less important, but proving that to end users can be more time-consuming than expected.

We do note that we have seen a number of comments in the Chinese press expressing concern that the Chinese display industry might be heading into a period of over-supply, based on the government’s aggressive financing of display capacity projects.  We note the same concern a bit down the road, as industry growth slows and panel size increases moderate somewhat, but the articles expressing such concerns remain buried below those proclaiming South Korea’s ultimate defeat.