Tianma Microelectronics (000050.CH) is rumored to have picked up orders from at least one smartphone customer for its 5.5” OLED displays at Mobile World Congress recently.  The panel ordered is said to be a 1080 display and would be Tianma’s 1st commercial OLED product.  They have been sampling a lower resolution model (1280x720) for a number of months.  Tianma current produces their OLED samples on a Gen 4.5 pilot line which has limited commercial production capabilities, but is scheduled to open phase 1 of its Gen 6 OLED fab in September of this year, although we have heard that the fab could open as early as May.

Unfortunately the order remains unconfirmed, and if true will likely be dependent on the Gen 6 line’s start date and ramp, but in any case, if this order is real, it would represent a significant step forward for Tianma’s OLED development and commercialization project.  Tianma has a~1% share of the overall display market on a m2 basis, but has a ~3% share of the small panel display market (dollars), and has been a supplier of small panel LCD displays to the Chinese market for a number of years, which gives them the relationships with mainland smartphone vendors that would help them sell an equivalent OLED display.  More than likely, Tianma will not be making much money on the order, if it goes through, as the new fab will likely be running at a low yield rate, but the idea is to get their foot in the door on OLED and even a small, money losing order from a recognizable Chinese smartphone manufacturer will go a long way toward building that business.

We believe LG Display has ordered a Gen 10+ lithography tool for the new OLED display fab it is constructing in Paju.  We believe the tool is a Nikon (7731.JP) FX-101S series model that will be adapted to the larger size.  The current Nikon Gen 10 lith tool supports substrate size of 2880x3130mm (~9m2) while we believe the LG line will run 3300x3340mm (~11m2).  The smaller tool has a TAC time of 70 seconds per sheet, which would be a bit higher for the larger substrate, which comes to ~360 60” TV panels/hour or 263k/month (3.15m/year).

Of course these estimates are for 60” TVs at 100% yield, and will be discounted, especially as the plant ramps production, but it gives investors the understanding of the capabilities and capacity of these tools.  That said, the typical lead time for backplane lith tools is between 14 and 18 months, although the size modified units might take a bit longer.  If LG placed the order in February, the expected delivery date would be between April and August 2018.  Our timeline for this fab has been phase 1 of the project scheduled for June 2018 which is a bit optimistic relative to the tools construction time, but still within reasonable parameters.  While Nikon is hard-pressed to meet its delivery schedules as it is, the fact that this will likely be the largest lith tool ever built (and likely the most expensive) in the display space, they might have the incentive to speed up the process.  Perhaps LGD has also given them some financial incentives to bring the tool in more quickly?

We noted earlier this week that LG had received 40,000 pre-orders for its flagship G6 smartphone during the 1st four days of its pre-release availability in South Korea.  According to subscription data, the pre-order rate has continued at ~10,000/day through yesterday’s actual release date, although no actual sales data has been generated yet.  Consumer distraction over the final impeachment of South Korea’s president might take some focus off of the initial sales, but this is a make or break product for LG’s mobile business, which has not seem much success over the last few years.  If the phone is not successful in the company’s home market, it will likely not do well elsewhere, and will put LG’s mobile business at significant risk.  Stay tuned.

On Tuesday we noted that the newly appointed Director of the White House Trade Council (a newly created position), Peter Navarro, berated Samsung Electronics (005930.KS) and LG Electronics (066570.KS) in reference to their competition with Michigan based Whirlpool (WHR), citing how uncompetitive it is of the South Korea giants to move facilities from one country to another after losing anti-dumping cases started by Whirlpool.  "I am amazed by how hard it is for our companies to compete on a level playing field. This heartland of America icon is grappling with a practice called country hopping. Two of the South Korean competitors, LG and Samsung, simply move the production to another country each time Whirlpool wins anti-dumping cases against them.  Such country hopping has happened twice to Whirlpool already. LG and Samsung have moved from China to Vietnam and Thailand. This is the kind of trade cheating that must be stopped. It undermines the whole international order, even as it puts thousands of Americans on the unemployment line. It imposes millions of dollars of losses on companies like Whirlpool."

As we noted previously, Professor Navarro did not mention the US anti-dumping duties of 52.5% on Samsung and 32.1% on LG that were imposed in December as a result of a petition from Whirlpool, based on the concept that the two competitors were selling washing machines at lower than ‘normal’ prices in the US.  So what shows up in the Korean press?  “Samsung mulls establishing US plant for home appliance”, resurrecting a 2/6/17 headline “Samsung Electronics to build first home appliance plant in US with production capacity of 2 million units”, which had specified Alabama as the state under discussion for the new plant.

Today’s story, which came from the head of the Samsung home appliance unit stated, “We are in internal talks to build a plant there, as part of our plan to secure a mid- to long-term strategic foothold…."We will make the details public after making specific decisions.”  This appeasement to the Trump administrations whining about unfair competition is similar to LG Electronics ‘plans to build an appliance manufacturing plant in Tennessee’ that also made the headlines on 3/1/2017.

So what will actually happen?  Likely something will be built by both companies in the US; an assembly plant, a distribution center, or similar operation, with significant publicity by the companies, indicating how they are compliant with the new administration’s “America First” program, and the administration flaunting the ‘hundreds of jobs created’ for American workers in massive photo-ops, but did anything really change?  Will Samsung make the same product for the same price here that it does in Mexico?  When you look at hourly compensation costs (2013 data shows Mexico hourly manufacturing worker compensation costs at 19% of US worker costs), it’s hard to imagine that the appliances made here will have the same price profile as those made south of the US border.  So jobs are created, politicians are photographed, speeches are made, and the US consumer will pay higher prices and likely a piece of the lost government revenue from the tax breaks given to ‘incentivize’ Samsung and LG to localize their production.

Politics aside, complaining about competition, tariffs, taxes, and the other mechanisms that the US seems to have adopted to combat the fact that high volume manufacturing is a cost sensitive game, misses the point.  The US is among, if not the most innovative country when it comes to product development and creation, and while the near-term focus is always on putting people back to work at any means, the real focus is developing the workforce to increase its ability to create and innovate, not try to do what others can do more cheaply.  Retrain, refocus, reeducate, and above all make sure will educate a potential workforce toward doing new things in new ways, rather than going back to try to ‘rebalance’ markets that now operate on a global scale.  The US consumer will be happier, the US worker will be happier, and we will be the driving force in the technology world that we have been in the past.  Yes, it’s a very competitive world, and some countries don’t always play fair, but complaining and bullying won’t fix the endemic problem, only transforming the workforce from one oriented toward a world economy that thinks like we are still in the 1960’ or 70’s to one that thinks like its 2030 will push us ahead of the competition. JOHO

The US Departments of Justice and Commerce announced that they had imposed a fine of $1.192b on Chinese telecom supplier ZTE (763.HK) for violating US sanctions against North Korea and Iran by buying products from US companies such as Qualcomm (QCOM) and Micron (MU) and reselling them to both sanctioned countries.  The DOJ indicated that while ZTE, the 4th largest telecom company worldwide and the 2nd largest in China, tried to conceal its activities, it has agreed to pay the fines, which are the largest ever levied for sanction violations.

"Imposing the fine corresponds to send the world a warning that the game is over," US Commerce Secretary Wilbur Ross said. "Countries that ignore economic sanctions and export control laws will not be able to avoid punishment and suffer from the most severe consequences. Under the leadership of US President Trump, we will aggressively enforce strong trade policies for two goals -- US security and the protection of American workers."

However the timing of the DOJ release follows the deployment of the THAAD missile defense system in South Korea, which has drawn the ire of China, who says it could be used against the country, and speculation that further fines on Chinese companies (particularly Huawei (pvt)) might follow seem to indicate that the US administration will continue to put economic pressure on China to have them back away from their tacit defense of North Korea.  It would seem that the pit bulls are in the ring and baring teeth.  No blood has been drawn yet, but the battle has just started.

Daily speculation as to the specifications for the upcoming iPhone 8 is a given with artist renderings (based on little information), model names, screen sizes, chassis types, button configurations and an endless stream of internal components, but does Apple (AAPL) actually have a definite product line for the iPhone 8?  Probably not, and we would expect that Apple has yet to go through any design verification processes concerning the potential variants.

As we have noted previously, Apple is expected to release at least three variants for the tenth anniversary of their iPhone product line.  Speculation that one of the variants will be based on OLED screen technology has lit a fire under panel producers, and led to a variety of press releases and/or ‘leaks’ indicating how much capital said producers would be spending on the technology.  What is most relevant however, is the experience that each panel producer has toward the development and production of small panel OLED displays, and the timelines associated with the increase in small panel OLED capacity.

Based on our comprehensive OLED industry model, we believe that Samsung Display (pvt) and LG Display (LPL) currently account for 90.7% of raw OLED capacity (Feb 2017) and 92.7% of utilized (available) OLED capacity, which leaves only modest capacity for other producers.  Given that much of LG Display’s OLED capacity is oriented toward large panel (TV) production, the numbers are skewed even more toward Samsung Display and as Samsung builds out its A3 and L7 fabs, we expect their share to increase, peaking in 1Q 2018.  But with all the speculation and ‘increased spending’ toward OLED that hits the press each day, Apple has little choice but to use Samsung Display at least in the first year that they might make the shift, and is likely still evaluating various display modalities, both for LCD and OLED before locking down any specific variant details. 

Other producers have promoted accelerated build-out schedules for OLED projects, both large and small, but we believe there are no producers other than Samsung Display that have the expertise in small panel OLED manufacturing to be a primary supplier in the iPhone 8 timeline.  We note that a number of panel manufacturers have running OLED lines, most of which are producing samples, but matching Samsung’s costs and efficiency are above the capabilities of most.  Further, we would expect Apple to want more than average specs for such displays, and potentially some sort of ‘flexibility’ in the design for a 10th anniversary variant, making it more difficult for new producers to compete with Samsung Display’s established flexible resources.  We would expect that Apple is inundated with OLED display samples and designs from a myriad of producers, but when it comes down to a very high profile product, taking a chance on a small or inexperienced producer would seem far too risky.  As Apple gets closer to a design lockdown and an eventual product spec lockdown, we will have a better idea of what to expect from such a display, and panel producer talk about OLED capacity expansion will be pushed out another year for all but Samsung Display.  We certainly don’t doubt that other OLED small panel producers will eventually be able to enter the market, but investor expectations for alternate OLED suppliers should be tempered with the understanding that a board funding allocation or an agreement with a regional government is a far cry from gaining a very demanding customer who has been working with the industry leader who has years more high volume small panel OLED manufacturing experience than the rest of the industry.  It will happen eventually, but not on the timeframes that alternate suppliers dream about.

Japan based LED producer and packager Nichia (pvt) announced that the Düsseldorf Court of Appeals has affirmed the company’s YAG[1] Patent for white phosphors used in LEDs and ruled that four products produced by Taiwan based Harvatek (6168.TT) infringe on that IP.  The patents in question were granted by the European Patent Office in July 2015, and Nichia immediately began enforcement against Everlight (2393.TT) and eventually Harvatek.  Nichia was granted an injunction ruling against Everlight after a vigorous court battle.  The win in the Harvatek case dismisses the appeal and leaves no recourse for further appeal in that venue, although Harvatek can ask a higher court for the right to appeal further.  

While Harvatek says the products involved represent only a small share of its total revenues and will not impact its business or customers, Nichia, as the world’s largest LED packager is an IP bulldog concerning its proprietary LED patents, and continues to litigate with a number of other LED producers and distributors that sell white LEDs using their YAG white phosphor, with the latest, against EBV Electronik GmbH (AVT) filed yesterday.  In most cases, the litigation is followed by a license agreement but details in the Harvatek case are not yet available.



[1] Yttrium aluminum garnet

California based ink-jet printing company Kateeva (pvt) has doubled its manufacturing space in order to facilitate the production of its OLED ink-jet encapsulation tool and potentially to build out production of an ink-jet tool for OLED stack materials.  We have mentioned Kateeva a number of times as the primary producer of large format ink-jet tools that are used to encapsulate OLED materials by depositing alternating organic and inorganic layers that keep oxygen and water from reaching sensitive OLED stack materials.  The Kateeva process has been used to replace glass encapsulation as flexible displays become a focus for OLED panel producers.

Typically, OLED encapsulation is done with thin glass and an epoxy or glass frit sealer with desiccant to absorb any water that might be present.  However flexible displays represent a bigger challenge to encapsulation and a number of materials and methods have been proposed, tested and put into production to meet the goal of keeping flexible OLED devices from degrading through exposure to air and water.  A number of OLED manufacturers are using ink-jet encapsulation at the back end of their OLED production lines to speed up the encapsulation process given that it can have as many as 10 successive layers of material, which take a substantial amount of process time to apply using other methods, leading to production bottlenecks.  Ink-jet printing, which is done in a Nitrogen atmosphere can substantially improve OLED production TAC time. 

While TFE is likely a lucrative business for Kateeva, the ultimate goal of ink-jet printing is OLED material patterning, which would allow a substantial reduction in OLED stack material usage.  In a typical deposition reactor, OLED materials are evaporated and deposited on a substrate through a fine metal mask that sets the pattern for the materials.  Unfortunately the materials are also deposited on the walls of the deposition chamber and on the mask itself, which substantially reduces the material usage efficiency and increases the downtime for chamber cleaning.  Material usage efficiencies as low as 25% can be part of the evaporative deposition process.  Ink-jet deposition applies the materials directly in a precise pattern, reducing material waste and increasing material efficiency to levels above 90%, a key cost savings for OLED producers.

There are problems however with ink-jet printing for OLED stack materials.  The solvents needed to allow the materials to pass through high speed ink-jet heads can change the characteristics of the materials themselves, and issues like solvent drying time or bleed between sub-pixels have kept ink-jet stack material processes in the development stage for years.  While Kateeva is likely adding capacity for the manufacturing of their encapsulation tools, the eventual adoption of ink-jet printing for deposition would be a milestone in the development of OLED manufacturing.

Rumors that Sharp (6753.JP) will be increasing its investment in OLED are circulating, giving investors hope that the company can participate in Apple’s (AAPL) expected transition to OLED technology for their iPhone line, yet only days ago, Sharp indicated that it was going to focus on the nascent 8K TV market after agreeing to build a new Gen 10.5 LCD fab in China.  Has Sharp changed its mind or is management tossing the spaghetti against the wall to see what sticks?

Sharp’s new owners have made a big acquisition, more than likely on the general assumption that it would be good for the number one worldwide assembler to have its own display production capabilities.  But the problem is, they already do have those capabilities, as they own Innolux (3481.TT), who has a 15% capacity share in the display space.  If you add in Sharp, they control over 20% of display capacity, only slightly below LG Display’s (LPL) 21.8% share and substantially above Samsung Display (pvt), AU Optronics (AUO) and BOE (200725.CH).  That is a lot of capacity to control, especially as the combined group has fabs that cover all substrate sizes from Gen 1 to Gen 10, and while Foxconn has indicated that they want to use a portion of that capacity to resurrect the Sharp brand, that still leaves them with lots of production assets, some of which (more Sharp) are being underutilized.

Historically, Foxconn has left Innolux to manage itself and develop its own customer base, while still supplying ~28% of Foxconn’s display needs, while Sharp has been mismanaged by its previous owners for many years.  Now, Chairman Terry Gou, has to decide how things will work for the larger entity and what direction it will take.  Thus far, it seems that the new management has taken a few bold steps.  First, cutting out a big customer or two (Samsung and Hisense (600060.CH)) from its display customer list to allocate capacity to the Sharp TV brand where he set a target of 10m units this year and 20m next, and second, announcing the commitment to building a Gen 10.5 fab in China that will focus on 8K TVs.  All of this was done while shaking the US president’s hand and committing to building something in the US, and expanding a meager OLED effort in order to stay in the sights of Apple, who is expected to convert some iPhone products to the technology.

As any good chairman, Mr. Gou is doing his best to appeal to as many factions in the display space as possible, but there comes a time when actual decisions become more important than rhetoric, and it would seem that Foxconn is still unwilling to give specifics, in similar fashion to the new administration in the US.  Mr. Gou’s philosophy of “we serve where there is a customer” has served him well given his company’s position as Apple’s primary assembler and his net worth of $7.5b US, but its time to lay out just what you intend to do with this massive display asset base and make a commitment as to the direction of those assets.  If it’s going to be focused on TV, than say so, don’t use 8k as a selling point.  If its going to be OLED, then make a real commitment, not just enough to give people the impression that you are in the game.  If you want to use all the assets for building Sharp/Hon Hai (2317.TT)/Foxconn brands, so be it, but take the shot now.  Samsung did it with small panel OLED when no one thought it could work (it did), LG is committed to OLED for TV for the next few years, and everyone assumes that TVs are going to continue to get bigger and cheaper and are building Gen 10+ fabs to compete.  Sharp’s mistakes are their mistakes, it’ time to decide, and maybe make a mistake or two, but you can’t please everyone. JOHO

Hacking is on the rise worldwide, but in South Korea it continues to get easier.  Various hacking services are now easily found on private message boards in South Korea, and price lists are available as if one was shopping for shoes.  The use of Bitcoin, which hides the buyer’s identity have increased the volume of offered hacking services and have helped cybercriminals build more sophisticated business models that offer tools or services to less sophisticated users that can wreak havoc by attacking websites with a variety of products.

If you would like to bypass antivirus software that is used in South Korea, like Naver (035420.KS) Vaccine or V3 (053800.KS), such tools can be had for 10,000 KRW ($8.69 US) with the source code selling for $130.  Need zombie PC for a DDoS[1] attack? You can buy them for $0.26/PC in South Korea, a bargain as they cost $0.43/PC in most Asian countries, $0.61/PC in the US, and $0.87/PC in Europe, which means you can buy 10,000 zombie PCs, enough for a sophisticated DDoS that can do serious damage to a company website, for $2,600.  Zombie sellers advertise by showing videos with the number of PCs they have control over, indicating their place in the cybercriminal subculture, although it is easy to get ripped off by such criminals who don’t deliver and disappear after they receive their bitcoin payment.

Want to hack smartphones?  Let the seller know the number and they will send a text that installs malware allowing you to advertise, control the phone, or examine its contents for passwords, payment accounts or other smartphone numbers that can be used to expand the malware’s reach to hundreds or thousands of phones.  Such systems are also used to gather the zombie PCs mentioned above.  Open the wrong e-mail and your PC becomes another zombie available for sale, and you would likely not even know it.

Of course, the South Korean National Police Agency knows it is happening, but the private message boards that are used to communicate prices and services are hard to locate and trace, and can disappear and reappear on a moment’s notice, while servers used for such purposes are usually housing in countries that do not have the capabilities or the desire to control such issues, and you thought that e-mail from the former Prime Minister of Namibia urging you to accept the $7 million dollars owed to you by a long lost uncle was annoying?  How about someone using your computer to crash a website or using your PC camera to rob you when you are not home, or just ransoming your computer for a few hundred dollars?  If a country as computer savvy as South Korea cannot control such issues, it might be time to look for a place in the woods…



[1] Distributed Denial of Service