​We have mentioned flexible OLED panel producer Royole (pvt) a number of times over the years, first as the winner in the race to produce a commercial foldable OLED display (although it was a developer’s kit), beating small panel OLED leader Samsung Display (pvt) and Samsung Electronics (005930.KS), who now dominate the space, and later as the company repeatedly touted itself as the leader in China’s foldable OLED development’ and its leader as a genius who beat the leaders at their own game.  The company was able to raise over $1.1b from Chinese investors, venture firms and regional and state-owned banks and filed an S-1 with the SEC for an IPO in 2020 at a valuation of ~$6b.  Unfortunately the documents seem to indicate that very few of the company’s foldable phones had actually been sold and raised questions about the company’s Gen 6 flexible OLED production line, said to have cost a fraction of typical OLED fab costs due to its ‘unique’ processes, which eventually caused IPO plans to be scrapped.
The company continued its development plans with the release of a 2nd generation foldable later in 2020 and announced plans to build another flexible OLED fab ($2.4b) in Qingdao in cooperation with a number of local state-run organizations and file papers for listing on the Chinese STAR board where requirements were a bit less onerous.  Unfortunately the company did not meet those requirements and the STAR IPO was rejected as rumors of missed payments to employees and suppliers began to surface.  There was considerable talk of potential bailout partners for a few months late last year and early this year but by May Royole management issued a notice to many employees requesting that they take long (3 month) vacations, which did little but to stimulate workers to protest to the local government.  Some workers were paid missing wages by the local government but it seems the handwriting was already on the wall and no bailout appeared, leaving creditors hanging as to when or if they were to get paid as the company continued to flounder.
Recently a number of those creditors filed redress documents against the company and further investigation led to the discovery that there were a number of pending disputes between the company and construction companies, service providers, public relations firms, and advertising companies, which forced a review of the company’s standing by the courts.  As of the end of last month the courts ruled in favor of the plaintiffs and froze what came to $5.3m in assets held by a company holding entity, essentially preventing the company from access to any capital, potentially signaling the end to the company’s operations, at least until the courts can sort out the complete asset and creditor lists.
As we had noted a number of times, we have doubted Royole’s abilities from the start, as we could find few who had any customer relationships with the company, and once we were able to see paperwork from the IPO filings, it became obvious that there had been considerable speculation as to the company’s prospects when early funding was made, especially around the time of the aborted STAR listing, which placed a $2+b valuation on the company.  It was not that we doubted the ability of a small company to find a shorter path to building a better mousetrap than its infinitely larger brethren, but that the massive hype behind the company’s founder’s ability to create a process that cost a small percentage of the manufacturing cost of flexible OLED production, without any details ever being disclosed, seemed a bit of artifice, which it seems it was.  While we don’t know what the courts will ultimately decide, we expect Royole will likely be shut down, a sad day for unicorns everywhere given the hopes originally placed on the company’s success, but sooner or later Royole had to pay the piper….

​In a recent interview, Mike Hack, a VP of Business Development at Universal Display (OLED), indicated that regardless of whether UDC or another company was able to develop a commercial blue phosphorescent OLED emitter material, they would still have to obtain a license from UDC, which we assume, would entail paying royalties.  The conversation came about as Samsung recently purchased the IP assets of blue TADF developer Cynora (pvt), without actually purchasing the development team or other assets.  This implied that Samsung was either protecting its own development program by trying to cover a large swath of potential TADF IP, or they were trying to keep another party from acquiring same.
Additionally, the CEO of Samsung Display indicated that the company was working on the development of its own ‘blue’ , which was a bit unusual in that most emitter materials are developed by companies such as UDC, along with customers who tend to provide parameters and leave the actual chemistry to others who have background in material science.  We are still not convinced that the statements made by SDC were quite what they seemed and more likely SDC is involved in the development of a blue emitter material with a partner, although perhaps a bit more deeply than in the past, but those statements opened the door to further questions about who will be the first to commercialize a blue emitter and how it might be licensed.
US patent law is about as complex and arcane as law can be and requires very specific knowledge that can require understanding both science and the nuances of language, but we expect Mr. Hack was referring to UDC’s ownership of Utility patents that cover the use of heavy metals in organic phosphorescent materials and their use in a light-emitting device.  The metals involved are Iridium, Osmium, Platinum and others, so if someone were to find another way to produce a phosphorescent emitter that did not use heavy metal ligands, they might be able to bypass the material side of the IP, but if they were to use a phosphorescent material to produce a light-emitting device, they would have to license the IP as Mr. Hack suggested and while such a new material could be produced and sold by other than UDC, its use in any device would require licensing. 
We are the first to admit we are not patent attorneys or organic material scientists, but after reading hundreds of patents relating to organic materials we have some knowledge of how the system works.  For reference…
There are requirements that must be met to apply for patent protection, and here is where things get fuzzy.  The first requirement is it has to be ‘patentable subject matter’, which seems obvious until you ask what is patentable subject matter, and that is somewhat open to legal interpretation.  Some of the things that are not patentable are natural phenomena, abstract ideas, printed matter, and business methods, however the last item is no longer part of the ‘no’ list as new rulings have allowed certain business processes.  Additional requirements are the patent has to be novel, it has to have utility, it has to be non-obvious, and it has to be subject to ‘enablement’, with each of those being subject to interpretation.  In terms of just ‘utility’ or the ability to garner a ‘utility patent’  the subject has to be a process, a machine, a composition of matter, a manufacturing process, and an improvement over existing process or IP.  If that doesn’t prove that IP is a subject that could lend itself to an almost infinite amount of conjecture, and if the US patent courts are any indication, it does.
That said, while the specific IP relating to Mr. Hack’s comments were not given, the fact that every commercial OLED manufacturer (or at least those producing full color OLED displays with phosphorescent materials, which is most of them) licenses UDC’s device  and material patents, is a good indication that the vast UDC IP portfolio covers all aspects of using phosphorescent materials in a light-emitting device and that regardless of who is the first to commercialize a blue phosphorescent emitter, license negotiations will occur with UDC. 

OLED TV, continues to grow as a percentage of total TV sales and shipments, but in the ‘premium’ TV market, which, in theory, consists of OLED TVs, 8K TVs (all types), LCD TVs with Quantum Dots, which would include all with Mini-LEDs., OLED TVs face a more difficult share growth path as overall LCD capacity is vastly larger than  that of OLED TV capacity and the inclusion of a quantum dot layer does not require dedicated fabs, as does OLED TV production.  Given that LG Display (LPL) is the primary producer of OLED TV displays, with Samsung Display (pvt) recently joining with its QD/OLED offerings, maximum capacity for OLED TV is limited to the existing capacity of LG Display and Samsung Display, which is 190,000 Gen 8.5 sheets/month between the two.
While we expect Samsung Display to increase its QD/OLED capacity over the next 18 months, overall OLED capacity seems static for the remainder of 2022 and at least the first half of 2023, if not the entire year.  Estimates for display equipment spending overall have come down for this year as high inventory levels and declining panel prices have slowed expansion plans across the industry, and while spending for LCD will come down by 28.5% y/y spending for OLED (both large and small panel) is expected to increase by 21.9%.  Both categories are expected to decline next year but return to growth in 2024 for both categories, while in 2025 LCD equipment spending is expected to decline while OLED equipment spending is expected to increase as shown in Figure 1.  We are a bit skeptical as to by how much LCD and OLED equipment spending will decline next year and could foresee a scenario where some of the equipment spending currently estimated for 2024 and 2025 could find its way back into 2023, but for now the outlook for 2023 display equipment spending continues to look rather grim.
Along with the decline in spending this year and next comes the decline in TV set shipments overall, which have been declining yearly after peaking in 2020 during the early days of the COVID pandemic.  Both Samsung (005930.KS) Electronics and LG Electronics (066570.KS), the leaders in TV set sales, are expected to see declines in shipments this year, leading to a unit volume decline in set units this year of ~3%.  While we expect much of the unit decline will fall to generic LCD TV sets, the decline has put pressure on the overall TV market, including OLED TVs.  With estimates of ~10m units for WOLED panel shipments this year, we believe that the potential for LGD to miss this target has increased, particularly given that the negotiations between Samsung and LG Display for the purchase of ~2m WOLED panels this year seem to have evaporated. 
While Samsung Display’s QD/OLED shipments will be modest this year, likely under 1m units, they will also impact what might have been potential WOLED panel shipments from LGD, which adds to the possibility of a shortfall.  On the positive side of the ledger, we expect Sony (SNE) to increase its presence in the OLED space, although they will be a QD/OLED customer, along with Samsung Electronics, which could lessen the positive impact of the Sony OLED expansion this year.  Recently there have been rumors that utilization rates have dropped at LG Display’s E4 fab, the larger of LGD’s South Korean OLED fabs, which could indicate that like LCD panel producers, LGD is trying to work down some excess inventory.  With 79 days until the unofficial start of the holiday season a slowdown in WOLED production does not bode well for the full year target of 10m units, however as long as the potential shortfall is not greater than 2m units, there will be some, albeit small, growth in the OLED TV business this year. 
The bigger risk is that LGD continues to produce at a rate that gets it near or to its target of 10m panels as this will create an even larger inventory issue going into 2023.  We expect production to decline by over 18% q/q in 3Q and by ~15% in 4Q, along with a modest (6.8%) increase in demand over the worst case scenario, which would create a 13.0% panel excess.  Given a safety level of 10%, the best case scenario would set the stage for a better 2023, while the worst case scenario would create a ~27% excess in panel stock, making 2023 another difficult year.  Much will rest on LGD’s production decisions and brand demand, but we expect TV brands are ordering on an almost day-to-day basis to try to gauge how consumer demand is flowing.

​China’s Sichuan province was hit with a magnitude 6.8 earthquake in a relatively remote area near the Tibetan border, causing 66 deaths, 250 injuries, and the evacuation of over 5,000 people from the area.  This follows the lockdown of the Sichuan Provincial capital Chengdu, a city of ~21m, last Thursday, in order to try to control a COVID-19 outbreak and the closure of many factories in August due to power blackouts caused by the record heat and lack of rain, causing hydroelectric power systems to fail to meet typical power requirements.  Despite the quake, Chengdu remains on lockdown.
There are a number of LCD and OLED fabs in Chengdu, particularly those of BOE (200725.CH) and Tianma (000050.CH).  We don’t have information on the status of these fabs due to the lockdown, however given that Chengdu is ~125 miles from the quake’s epicenter, we expect the damage to the city was relatively light, although power was lost in Chengdu and Chongqing nearby for some residents.  A quick look at display facilities in Chengdu show that ~8.6% of BOE’s LCD capacity is produced in Chengdu fabs while 18.4% of the company’s OLED capacity is produced in the local area, so it would seem that local shutdowns, if they were to or have fully closed, would be relatively minor unless sustained for an extended period.  BOE has made no public comment about either the effects of the lockdown or the earthquake to date.

Given that the headlines over the Labor Day weekend were abounding with comments on the discounts available due to overstocked items in the CE space and beyond, and that the shortages and transportation bottlenecks are rapidly ending, we thought it appropriate to check Samsung’s (005930.KS) Mini-LED/QD Tv pricing to see if such discounts were apparent in that TV space.
While it was only 22 days since we last checked prices, it turns out there was a bit more price movement than we might have expected, although we know the inventory situation in the TV space overall entered the period at a higher than normal level.  There was considerable disparity between Samsung’s 2021 TV line[1] and the new 2022 line.  In terms of price movement, only 5 sets in the 2021 line saw a decrease in price, while 25 saw price declines in the 2022 line, indicating in which model years the excess inventory was held.  Further, the average period decline for the 2021 line was -0.5% while the average period decline for the 2022 line was 6.3%. 
More telling is the fact that the average decline from the original release price for the 2021 line was 27.6% while the average price decline from the original price for the 2022 line was 21.8% but the 2021 decline took place over 474 days, while the 2022 decline took place over 146 days, a rate of decline almost twice as fast as last year..   While we hesitate to speculate on where that would put 2022 Mini-LED/QD TV prices by next year, with 17 2022 models at their lowest price since release and only 10 at their lowest out of last year’s line, the trend toward lower Mini-LED/QD and QD only TVs seems set, at least into the holidays.
All in, it seems Samsung is certainly a bit more aggressive as to pricing for its Mini-LED/QD line this year, and with offerings from TCL (000100.CH), LG Electronics (066570.KS), Sony (SNE), Philips (pvt), Konka (000016.CH), Huawei (pvt) and Hisense (600060.CH) either available or announced this year, as opposed to Samsung, TCL, and LG last year, it is not surprising that Samsung is competing a bit more aggressively this year.  To get an idea of the competition and range of prices we put together a list of currently available Mini-LED TVs and prices.  We note that we cannot certify that all models shown have Mini-LED and quantum dots and we not that all models vary as to other features.


[1] Our data includes only Samsung’s Mini-LED/QD and QD Only TV lines and does not include Samsung’s generic LCD TVs.

Remember Satphones, those bulky devices that show up as global communication devices that allow soldiers, CIA operatives, and zombie apocalypse holdouts to communicate with others, regardless of their location?  They seem to be in resurgence as Chinese smartphone brand releases its new Mate 50 line that has the capability to connect to China’s Beidou Satellite System for emergency SMS, although the phone does not have 5G service due to trade limitations placed on the company by the US.  The satellite network will allow for simple messaging, GPS, and emergency notifications but will be limited as to data speeds (2G rates most likely).
While ~86% of the global population has cellular coverage and less than 25% have 5G coverage, and while estimates vary, actual cell coverage is limited to between 10% and 15% of the world’s land mass, which means there are a lot of places where your cell phone is just a piece of expensive hardware that does little or nothing.  But the addition of satellite communication capabilities opens the communications grid to an almost infinite number of locations that have been inaccessible to mobile communications..  There are satphones that can be bought by us regular folk, and they are not as expensive as one might think (between $700 and $1,700 typically), but must include a voice/data plan, which can become expensive depending on use and at relatively slow data speeds.  You can buy a base station that allows a computer or cell phone to link to a satellite network, but those cost between $3,000 and $18,000.
While the Huawei satellite hook-up is actually of relatively little use for other than messaging, it predates Apple’s (AAPL) iPhone 14 announcement (tomorrow) that is rumored to include some form of satellite communication capabilities.  There was supposed to be a similar system for the iPhone 13 but Apple and whoever the satellite carrier was to be were unable to reach an agreement before last year’s release.  There is no information yet about the extent of Apple’s involvement with satellite networks, but if the iPhone 14 even hints at such a capability, there will be a renewed interest in the world of satphones and satellite communication networks , and while the cost might be high, it certainly would be a plus if you could whip out your iPhone on K2 and send a few pictures to the folks back home, that is if the phone is not completely frozen (The operating temperature of an iPhone 13 is 32⁰ to 95⁰, in theory). 
The real benefit however would come from the potential for new silicon development, particularly a combination of existing positioning services and other satellite functions.  Mediatek (2454.TT) has completed the first 5G non-terrestrial network connection test, paving the way for such potential silicon, and while plans for new satellite networks offer the potential to accommodate the distance restrictions that mmWave 5G and 6G face due to absorption by buildings, trees, and other objects as a satellite link would eliminate those restrictions and the potential density of base stations needed.  

Over the last few months we have been researching the potential for the advancement of automation in the semiconductor space, meaning the ability of a semiconductor fab to self-correct as silicon moves down the process line, with the ultimate purpose being to increase yield.  As part of the research we have found that such a system, and even systems currently in place, would not be limited by a lack of data necessary to make such corrections in-process, but quite the opposite.  In most fabs, and even in much less complicated production processes, there is too much data, a result of smaller and more accurate sensors that can be placed in more locations with better sensor measurement capabilities and faster data collection networks.
We have found that while there are some data standards that span a number of semiconductor processes, it is the job of process data engineers to look at data generated from each piece of equipment in order to analyze where the line is facing issues, and this could mean data from many sensors on each process tool.  As data collection sensors have improved and collection points increased, the amount of data that needs to be reviewed has become overwhelming, to the point that it can be a hindrance rather than a help.
One vivid example was shown to us concerning a relatively simple process in the CE world, solder reflow inspection on PC boards, a process where the board is pre-heated, solder paste[1] added, and then cooled, which attaches components to the PC board.  While reflow is a basic process that has evolved over the years, typical problems that occur during reflow, such as component shift or delamination of the board, can not only cause the board to be pulled off the line for further inspection (and cost), repair, or to be scrapped entirely, but can become a source of serious problems when the boards reach customers in products.
In many cases PCB boards are inspected visually as they pass through a line, although humans are easily distracted or fatigued, leading to the use of AOI (Automated Optical Inspection) tools that use image sensors to match a board’s captured image to a stored template.  However even AOI tools are subject to errors, particularly false defects, which can be caused by light reflection or other optical interference, which pushes more sophisticated PCB lines to use tools that perform a number of measurements in addition to template matching as the boards pass through the production line. 
One example we saw was a system that was evaluating PCB boards that were ~8.3” x 6.3”, a bit more than half the size of a sheet of paper.  The tool measured height, width, and area at over 3,400 points on the board and checked for x/y offset (components that were placed incorrectly during reflow), generating 17,000 data points for each board.  The data was fed to SPC (Statistical Process Control) software, which, in theory, is supposed to correlate the data, allowing a process engineer to track back where defects were being generated in order to make corrections to the line. 
What was most surprising was that in many cases, when a board was flagged by the system, the process engineer let the board remain in the line, under the theory that the final AOI and SPC review would pull the board before it was shipped.  This defeated the purpose of the entire system as the flagged board continued to be processed, despite the fact that it had defects and would either need repair which could entail removing later process steps, or could be scrapped entirely, and if the defect was not clarified and the reflow process fixed, it could continue to affect subsequent boards.   When asked why the engineer let the flagged board go through, the comment was that there are so many data points that the chance for false negatives continues to increase with the number of measurement points, so rather than pull the board they left the determination up to the final inspection engineer, essentially defeating the purpose of the data collection early in the process line.
While this was an example using a relatively simple process and tool set, producing semiconductors is far more complex, with memory producer Micron (MU) stating that it takes more than a thousand process and measurement steps to produce modern chips.  To put just the feature size of those structures into perspective, think of a chip die, typically about the size of a fingernail, being blown up to the size of a football field.  Take one blade of grass from that field, cut it in half, cut it in half again, and cut it in half a third time.  That’s the size of one transistor, essentially one bit of storage, out of the 8 billion on a typical memory chip, so the amount of data generated by tools to make sure things are where they should be on semiconductors is enormous, more than all the data collected by Facebook (FB), Amazon (AMZN) and all the other social media companies, although over 80% of it is not ever looked at by process engineers.
Most semiconductor process engineers are only interested in data when there is an ‘excursion’ in the semiconductor fabrication process, a nice way of saying ‘an error’, as these can cause millions in losses considering the number of die on each wafer and the number of wafers processed each hour.  Some semiconductor customers require (military, automotive) that the data be stored for up to 15 years, even if it has never been reviewed, but ~90% of the available semiconductor process data is less than two years old according to T&M companies, which means the amount of data being generated during the last few years has increased dramatically.
So the question then becomes whether it is actually helpful to add sensors, data collection points, and sample rates if most of the data is never going to be used?  The fear that at some point the data might be useful pushes the industry to collect everything, even if it seems unnecessary, ‘just in case’, but does it really help to improve yield, the ultimate goal, if most of the data is ignored?  Not really, so improving sensors, adding data collection points, and increasing data collection network speeds all sound good on paper but the increases in process data volumes don’t seem to lead to large gains in productivity and yield, so where is the problem? 
The problem is that humans are not interested in the data as a whole, only if it solves an immediate problem, while computers are designed to process data, especially large amounts of data, and can be ‘trained’ to spot even the smallest anomaly.  This is essentially what SPC software is supposed to do, help a human engineer spot a problem and  trace it back to its source, a process that even in the PCB example above, would entail the engineer having to look at the data from a number of tools in order to spot the root cause of the error, something that computers can do far more easily than humans, and if we trusted the computer, we could actually let the computer make the necessary adjustments to the process tools without getting involved.
This is the goal of semiconductor foundries; to fully automate the process using a self-correcting system that could make corrections on a tool-by-tool or process by-process-basis, leaving the human engineering staff to come up with better and more efficient process steps.  Unfortunately there is a problem, and that comes back to the data.  Much of that data is collected at the tool level and tool manufacturers want users to be able to use that data through a platform specific to that tool while the next tool vendor wants the user to use his platform for the data, leading to lots of data that is not uniform, especially as tool vendors and the industry overall is just coming to the conclusion that the amount of data is so vast that AI and machine learning systems are the only way in which the data can be utilized.  This leads to the conclusion that the data must be uniform and organized to avoid the need to ‘clean’ the data or reformat same before it can be processed, which puts bottlenecks into the potential feedback loop.
There are some tools that are self-correcting, such as ASML’s (ASML) EUV tools that allows the tool to measure and compensate for sub-nanometer inaccuracies that occur as the scanner operates, feeding back that measurement data directly to actuators at the reticle stage, which compensate for those minute changes that would otherwise become potential excursions.  ASML is the first to admit that the stresses that their systems work under, such as the fact that the system must move the wafer position to within a quarter of a nanometer 20,000 times each second,, must be checked, compensated for and adjusted with each movement for the system to work correctly.  In such a system there is no human intervention, which forces engineers to trust that the data is being correctly analyzed and the compensation increases the tools yield.
That said, the goal of a unified data set and a master AI system for a semiconductor fab that could be self-correcting is a lofty and far-off goal, but the challenge is more to bring uniformity to fab data and reduce the number of proprietary data ‘silos’ that exist in fabs currently.  It is an enormous task, but one that must be addressed as the complexity of semiconductor processing continues to increase and as the data volumes continue to increase, we expect fab spending for software systems designed to work with disparate systems will increase, while the industry reaches for standards that will shift that focus from data compensation and presentation to generating AI systems that are able to look across a fab, find production bottlenecks and compensate for this issues by adjusting tool parameters anywhere in the fab.  Might not be in our lifetime, but we will keep digging in coming weeks to see if anyone is on track to making a dent in all that data and whether there is any hope of a unified solution that could make the vision of an automated fab a reality.
Note:
FOUP – Front Opening Unified Pod – Wafer Carrier
FDC – Fault Detection & Classification
EUV – Extreme Ultraviolet – Lithography tool
 


[1] Solder paste is a mixture of solder powder and a flux.  The flux holds the components in place, prevents oxidation, and allows the melted solder to flow.  The solder itself completes the connection between the component and the traces (think of flat wires) on the board.

“Did I do that?” was the catch-phrase of Steve Urkel, a character played by Jaleel White for the sitcom ‘Family Matters’ that ran between late 1989 and mid-1997.  As the annoying protagonist, the Urkel character typically made a mess of whatever situation in which he appeared, either by saying or doing something foolish, absurd, or stupid.  When confronted he feigned surprise and stated “Did I do that?”    Then there is the time when, after a night of bar hopping, you came home and began a rant on Twitter (TWTR) about how “…that stupid Barbara T. was a skank for talking to the guy YOU were talking to at the bar, and how she was always in EVERYBODY’S FACE, and was FAT and couldn’t get a guy even if SHE WAS NAKED!”, and then you fell asleep, only to wake up the next morning to realize that instead of Brenda T., who was the skank that you were mad at, you put ‘Barbara T. in the Tweet, with Barbara T. being your boss.
The frantic texts and calls to Twitter about deleting the Tweet came to no avail and the meeting the next day with HR and Barbara T. did not go well, leaving a mangled career path and the swearing off of all drinks with umbrellas, but the Tweet remained as a reminder that restraint is a complex emotion and not an easy one to foster, with or without alcohol, but those days are soon to be over…for a price.  It seems that Elon Musk, in his attempt to exit his $44b offer for Twitter, seems to have pushed management to get off the fence about a way for users to retract their Tweets, that is, if they are awake and if they pay $5.00/ month.
Twitter is ‘testing’ its newly developed ‘edit’ button, that would allow users of Twitter Blue, essentially a paid subscription to Twitter that is ad free and has a few ‘custom’ features, but most importantly will allow the ‘edit button to undo any Tweet for 30 minutes, giving one time to reconsider the impact of their latest Tweet.  Of course, the user needs to be conscious and coherent enough to realize, “yes, it was you who wrote that…” and not to make it worse by editing the original rant, but the subscription is a requirement and the Tweet will be marked as edited, although that seems a small price to pay to keep your friends, family, or job from firing or ghosting you forever.  But there is another catch, and that is as the app stands now, the modified Tweet will have a history attached “to help protect the integrity of the conversation and create a publicly accessable record of what was said…” according to Twitter, which would then open the original Tweet up to those taking the time to check the time-stamped history.  Perhaps once this highly requested addition is added to Twitter users will request another ‘level of editing’ that could wipe out the original (for an additional fee?) or a ‘Did I do That?’ button that holds all Tweets for 24 hours if your smartphone smells alcohol.  The possibilities are endless… 

During the last week there have been a number of stories in the trade press indicating that inventory levels ‘in certain products’ have returned to normal after brands cut orders last months, although we have seen little or no data indicating that such has occurred.  We expect there is a considerable amount of ‘promotion’ in such statements, as suppliers try to entice brands back to placing orders by hinting that others might be doing so, with much of the focus being on notebooks, which saw a shipment decline of 15.4% q/q and a decline of 13.0% y/y in 2Q, while notebook display shipments declined by 25.8% q/q and declined 22/1% y/y during the same period.
Our assumption would be that if inventory levels for notebooks are now returning to normal (at least for some brands and product types) there should be a bit more stability in the retail market as to pricing, however pricing notebooks is a difficult task as there are an almost infinite number of variations as to type (laptop, Chromebook, 2 in 1, etc.), display size (13.3”, 15.6”, 17.3”, etc.), memory, drive size, and a host of other characteristics, which makes price comparisons difficult.  That said, we went to Best Buy (BBY) and listed the ‘most popular’ notebook models and cross referenced that list against Amazon (AMZN) pricing for the same models.  If we could not find an exact model match on Amazon, we eliminated that model, but we were able to come up with a list of 10 models that were on the Best Buy ‘Best sellers’ list that were also represented on Amazon.  We note that in some cases the reason some of the Best Buy models do not appear on Amazon is that they are either too new, too old, or have an unusual configuration, which throws a bit of a question as to how Best Buy determines the ‘Best seller’ list, but that is a question for Best Buy.
All of that said, the average discount from list price for all 10 models at Best Buy was -26.7% and in comparison the average discount from the highest price on Amazon was -18.4% and the change from the lowest price on Amazon was a slight increase of 0.2$, with the detail being all but two models were at or below their lowest price on Amazon, which, as a single datapoint would indicate that recent discounting to move inventory has not quite ended.  For reference, and this was a bit of a surprise to us, all but one notebook model saw higher current pricing on Amazon than at Best Buy, with an average difference of +14.1% higher on Amazon.  We had assumed that Amazon pricing would have been equal to or lower than Best Buy, but that was not the case, although the basis for the lest was Best Buy’s ‘Best sellers’ which implies that list is attractively priced.  That said, Amazon tends to be lower priced for such items in most cases, unless the Amazon pricing comes from 3rd party sellers, rather than Amazon itself, which was the case for a number of models.  These are anecdotal data points that are, at the least, helpful in checking direction and momentum.
All in, our conclusion is that it is a bit early to be calling for a return to what  has been ‘typical’ ordering for notebook brands and CE products generally, and while we expect seasonality will help to stimulate some additional ordering as of late August early September.  Typical 3Q notebook panel shipment seasonality (5 year average) calls for an 8.3% increase q/q, although that same 5 year average for 2Q was +13.8% while the industry saw a 25.8% decline, so we only use those seasonal averages as reference.  There have been a few signs that the rapid increases in inflation might have leveled off a bit but as an incentive to consumers who deal with ‘non-seasonally adjusted’ prices for essentials, expensive CE products are likely not high on the list of things they ‘need’ to purchase over the next few months, so we see a more gradual demand recovery that could easily extend into 2023. Especially given the aggressive demand cycle the industry went through in late 2020 and early 2021 as a result of COVID. 
Bringing inventory levels back to normal is only half of the problem, with the other half being the stimulation of a new demand cycle, and that we expect will take lower prices.  Whether that comes from a disinflationary environment next year, or a desire for CE brands to dump high cost inventory before year-end (a practice not unfamiliar to many CE companies, particularly those outside of the US where the focus is on asset quality over sales growth), will determine how the holiday season plays out.  While we expect some cost relief across the CE supply chain over the next few months, we expect that the holiday season this year will be mediocre for most CE products.

​We have spent a considerable amount of time in our notes on Samsung Display’s (pvt) ‘new’ Quantum Dot/OLED display technology, with much of that time concerning the technology itself and how it differs from other TV/large panel display technologies, but that technology oriented perspective gives short shrift to the application of that technology in Samsung Electronics’ (005930.KS) first commercial OLED TV offerings since SDC (likely with parent Samsung Electronics) decided that WOLED (OLED with color filter) was not practical for the mass production of TVs back in 2013.  While LG Display (LPL) has developed a production process and market for WOLED displays and flaunted the popularization of OLED TVs in Samsung’s face, Samsung has concentrated on RGB OLED displays, which are limited to sizes below typical TV size range.
As SDC perfected the use of quantum dots as a replacement for the color filter in a large panel setting and developed a production line for same, we begin to look at QD/OLED as a competitive product against a number of other TV display modalities, such as LCD, Mini-LED LCD, Mini-LED LCD with quantum dots, WOLED, and micro-LED, and to that end we have to look at how the sets using QD/OLED are priced.  In fact, while there will be a small segment of the buying public that will buy a QD/OLED TV because it is new, (those who answer surveys with “I am always the first to try new technology”) the ability of the technology and its application to be a viable competitor to other TV display technology is the key to success, no matter how much is spent on marketing, and that will have a great deal to do with how it is priced.
There has been much in the marketing literature for QD/OLED about ‘perceived’ specifications, separate from hard metrics, which leads us to believe that Samsung is ready to do battle with other TV modalities, and is passing through the initial release stage where price is almost irrelevant.  In fact, Samsung itself brought to our attention the fact that its two QD/OLED models have just been put on sale, with the 65” model being discounted from $2,999 to $2,099 for Labor Day, a 30% discount, and the 55” model reduced from $2,199 to $1,699, a 22.7% reduction.  This follows a previous price reduction from $2,999 to $2,599, so the reductions from the initial offering prices are substantially higher.  (See our note of 7/11/22 for details).  Samsung is also offering phone and tablet  trade-ins to be applied to the reduced price, and no-interest financing of the purchase price in 4 installments or monthly over four years (4 year option comes to $43.75 and $35.42 monthly) and delivery in 5 days, or pick-up (Best Buy) in 3 days.  Amazon is a few dollars lower.
All of this comes down to the fact that the competition in the OLED space has now been stepped up, although to pin down where the QD/OLED sets would fall against other 55” and 65” TVs is a task that involves classification by specifications and price across a multitude of TV brands ranging in price (4K) from $250 to ~$3,600.  Given that Samsung Display’s capacity is relatively limited with one fab in production for these displays, the question we ask is why is Samsung competing on price this early in the product cycle?  The obvious answer is to sell more QD/OLED TVs, but with near-term capacity relatively limited, we see this more as a way to bring the technology into the eye of the general public rather than the cognoscenti that have a vested interest in staying close to the TV space. 
Samsung is very good at building momentum behind products where they are the exclusive supplier (flexible OLED, foldables, LTPOP, etc.), but in this case the difference between Samsung’s QD/OLED product and other large panel OLED displays is a bit more subtle, forcing Samsung to more down the price curve a bit faster than they might if there were no other large panel OLED TV competitors.  That said, Mini-LED LCD and QD/Mini-LED TV have given some extra life to the premium TV market and in order to maintain their TV leadership role across the globe, Samsung must continue to broaden its product portfolio.  Adding another category here, at what is becoming a reasonable price point, is a move that will make sure they have appeal to every potential TV buyer, regardless of the price range or quality level.