Microsoft (MSFT) Windows™ is a popular operating system, averaging a 26.4% share over the last 12 months, but when it comes to number of devices in use, it pales in comparison to Android OS, which averages ~45.1% over the same period.  So as not to offend the Apple (AAPL) base, iOS has averaged ~18.1%, and OS X (Now Mac OS) holds a 5.5% share, so technically Apple OS’s (combined) have a 23.6% share, just a bit behind Windows.  There are other operating systems, particularly Linux, that are used in a wide range of hardware systems, and a number of, for lack of a better name, proprietary operating systems sponsored by companies, for example Chrome OS, the operating system that runs the Chrome  browser, or Samsung’s (005930.KS)  Tizen OS that runs its smart TVs. 
Way back in 2008, HTC (2498.TT) released the Dream smartphone, the first to use the Android OS, competing with Symbian, owned by Nokia (NOKIA.FI) (49.8% share) and Blackberry OS, owned by RIM (BB), with a 16.6% share.  Within 3 years Android became the OS with the largest share and has maintained that position since[1].  The ability to maintain such leadership is based on the fact that the Android kernel is open-source and can therefore be downloaded and modified with paying Google (GOOG). However there are many Google apps that run under Android (Google Maps, Gmail, Google Playstore, etc.) that have to be licensed. 
Because Android is so popular, it is surprising that any company would undertake a project to develop a proprietary OS.  Estimates range from 5 to 10 years to develop the kernel and components for a large team of skilled engineers, with specialties in embedded systems, security, and assembly language.  Cost estimates range from ~$1b to over $10b, with the user interface alone running close to $100m.  This prices out most companies, but instead of developing a proprietary OS from scratch, companies can modify the Android  or Linux kernel in order to create an OS that they believe is best suited to their products and affords some differentiation.
That is exactly what Amazon (AMZN) is thought to be doing.  While the Vega OS is based on the Linux kernel, it is being built from the ground up, allowing it to be highly customized to Amazon hardware, similar to the way Samsung’s Tizen OS is based around Linux rather than Android.  As noted below, the control Amazon gains, including a modern user interface and potentially deep AI integration, is likely far more so than could be achieved through continued Android development, but it takes considerable time to build an application development ecosystem, with no guarantee that it will be successful.  Samsung has been successful with Tizen in the smart TV space, but as the leader in that genre, it has a distinct advantage, although its applications are still a bit limited, even after almost 10 years of development.
Amazon will have to bear the continued development cost and maintenance of the Vega OS system and find ways to encourage developers to hop on board once the OS is officially released (expected some time this year), all of which can be an albatross if little momentum is developed.  It’s a high risk, high reward game that only a few players could afford to play, but Amazon does have over 500m Alexa devices in operation, over 200m streaming sticks, e-readers, home security systems, and lots of smart speakers to eventually connect through Vega OS, so at the least, they have at least a shot over the next few years to gain enough share to make it into the top 5.


[1] Note that in terms of active device share, that crossover did note occur until 2017.

Apple (AAPL) has changed its healthcare focus from a meager “Project Quartz” to a more meaningful and robust “Project Mulberry”, including AI agents to collect and process the data that Apple devices collect about you.  This is not the ‘secret stuff’ brands collect, like what OS you are using, what device you are on, your search results (if you let them), and almost everything about what you have bought[1], but more the data that you allow Apple to collect by using the Apple watch, your iPhone, your AirPods, and even some 3rd party applications.  This is ‘health’ data, that includes sleep patterns, steps, calories, heart rate, weight, and a variety of other metrics about your bodily functions.
The objective is to provide Apple users with information that will make them healthier and more fit, but Apple, even before the platform is available, has made the upgrade to AI agents and an integration with Apple Intelligence, to make that information more ‘real-time’, personal, and meaningful.  The agents are the scavengers that will poll your Apple devices for the health information they collect and bring it to Apple Intelligence for monitoring and evaluation.  It is thought that Apple will not only offer you evaluations of your nutritional and sleep habits but could even offer camera-based assessments of your workouts and access to educational videos, put together by internal and external health experts. 
While the range of detail is thought to delve into physical therapy, mental health, and even cardiology, the initial focus is thought to be nutritional, with monitoring and alerts leading to personalized health advice based on your data, although there has been talk of AI-based mental health counseling and chronic disease predictive analysis.  As one might expect, Apple’s focus seems to be on the ‘user experience’, the part of the Apple persona that allows them to charge a premium for their products, but Apple is certainly not the first to go in this direction in this new age of AI.  Google’s  (GOOG) Fit is a similar collector of personal health data through Android’s Health Connect.  This platform allows permitted 3rd party apps to supply and collect data that feed the Google Fit app, but is more a collector, aggregator, and visualizer than an advice tool, although Google is currently working to integrate that data into its other health related services, with a tie-in to reference ‘reputable sources’ on YouTube.
Amazon (AMZN) also has a health program, but its focus is more oriented toward B2B with the Amazon Pharmacy supplying information on medications and interactions and the Amazon Clinic and One Medical able to set up virtual video or text sessions with clinicians (some on staff) that can evaluate conditions, make diagnoses, and prescribe medication for relatively common illnesses.  There are also companies like Noom (pvt) or MyFitnessPal (pvt) that are more specific to food and calorie management but given the enthusiasm for Ai that seems rampant across the health sector, we expect almost every health related application to leverage AI to stay competitive.
There are a few caveats here, particularly HIPAA regulations which regulate any health information that is maintained or transferred.  Entities involved must encrypt health data, limit access, perform risk assessment, maintain audit trails, breach notifications, and take ‘reasonable steps’ to prevent access to or disclosure of patient information.  HIPAA is difficult enough to understand and maintain, but adding AI to the mix opens everything up to new legal questions, many of which have yet to reach the courts and as liability becomes a potential issue when health-related advice is being given, we expect many new court cases that will not only focus on the potential liability of poor or incorrect data, but will include questions of algorithmic bias, inadequate software testing, and the fact that Ai systems are essentially ‘black boxes’ that make it impossible to derive where or how an AI arrived at a particular diagnosis or conclusion. 
Smart lawyers will not only include site owners but also those who wrote the algos that run them, looking for biases that could cause hallucinations, errors in judgement, or flawed diagnoses based on poor human vetting.  When Ai developers are called into court to defend issues like what data was included in an AI’s training or what process was used to draw a conclusion, high level math will not be how they are judged by a jury, so while Apple jumps into the fray to provide a positive health experience through Project Mulberry and Apple Intelligence, its not like Wikipedia, where you take things with a grain of salt.  Healthcare decisions affect people’s lives, as some can be significantly influenced by the information given by Ai healthcare.  There are good and bad doctors, and sometimes doctors make mistakes, which is why malpractice insurance exists, but will there be malpractice insurance for an application that gives incorrect advice or misdiagnoses an ailment or mental condition?


[1] IP Address
Device Type & Model
Operating System
Device Identifiers (trackers like AAID, IMEI
Screen Resolution
Installed apps (some)
Browser type & version
Cookies (optional)
Browsing History (Optional)
Location Data (Optional)
Referring websites
App usage
Contacts & Calendar (Optional)
Photos & Videos (Optional)
In-app purchases
Search queries (Optional)
Social Media Activity
Shopping activity
Form submissions
Wi-Fi network name
Data usage
Bluetooth data
Sensor tracking
Accelerometer & Gyroscope data
Ambient Lighting data
‘like’ data
DNS lookups
…To name a few.

In the early days of streaming services, the ability to choose content was a valuable resource for viewers and there were no reference points as to the relative value of such services, but as streaming services have proliferated over the years, price points have become more visible as battles for hit content continue to push streaming service providers to spend more for content creation, eventually finding that those costs must be passed on to subscribers.  During 2020 and 2021, the COVID pandemic pushed consumers toward almost any kind of streaming content, and monthly bills became secondary to the insatiable need for entertainment, but as the pandemic subsided and subscribers faced the icy fingers of inflation for the first time in many years, they started to think a bit more about what it was they were paying for and how much value it actually had, now that they were no longer tethered to their screens.
As streaming subscribers, we found ourselves in the same situation, subscribed to a number of streaming services, some of which we could not remember why or when we subscribed.  Some of that indulgence was due to billing, as direct billing through Paypal (PYPL) or similar services made monthly charges more abstract and bundling under Roku (ROKU) or other services almost hid them all together.   Eventually the day comes when subscribers sit down and decide that they must ‘figure it out’ and dump those streamers that provide lesser value. 
That said, we were curious to see what the monthly cost of streaming services has done over the last 10+ years, which led us to put together this data on streaming pricing.  The most meaningful timeframe is the period from 2019 to 2023, which includes the year preceding COVID and roughly a year (mid 2022 to current) after the pandemic.  The overall change in price among the major streaming platforms shown here during that period has declined by 10.5%, with HBO (WBD) leading the charge down with a 33.3% decline in aggregate price.  The combined Netflix (NFLX) prices (all tiers) was down 18.8% but Amazon (AMZN) Prime and Hulu were both up over the period by 15.4% and 15.0% respectively, while Disney (DIS) was up 14.3%.  While the composite numbers are helpful, the trends are far more obvious when looking at the charts. 
We do note that the 10.5% composite price decline seem across all categories over the 5 -year period is primarily a result of Netflix’s price decreases (all categories) this year, which puts its pricing below its initial 2007 pricing, which should indicate the increasingly competitive nature of the streaming business.  For those streamers that go back far enough, the 10-year data shows Amazon Prime up the most (81.8%), Hulu up 15% and Netflix up 8.3%, so if you are looking for price stability over time, it would seem Netflix is the choice.  Of course, much in the streaming world is based on content offering, but no matter how you look at content, any evaluation would be subjective, which makes it moot in the ‘Fun with Data’ genre, but Figure 7, which shows top 5 streamer subscribers and y/y growth, indicates that the overall subscriber count continues to grow, albeit a bit down from the 2020 – 2022 years.  Again, that would seem to indicate that subscribers are still shelling out cash for streaming services but streamers themselves are facing a far more competitive world than in early and pre-pandemic years.
We note that the subscriber data, in most cases, does not come from the steaming services themselves, so we consider the data somewhat questionable, particularly that of Amazon Prime, but it does give some visual representation of the trends.

​Five large Chinese companies have indicated that they would begin delisting procedures to remove their shares from the New York Stock Exchange, tacitly over disputes with the SEC over audit information that has been ongoing.  While the China Securities Regulatory Commission has stated that the choice to delist was based on each company’s own business decisions, they have all complied with the rules and regulatory requirements in US capital markets, although the agency also indicated that it was keeping channels open with ‘relevant overseas regulatory agencies”, indicating that there is some ongoing dialog.  Given the increased tension between the US and China in recent weeks, the timing does seem a bit pointed.
The companies involved, Sinopec (600028.CH), China Life (601628.CH), Chalco (601600.CH), PetroChina (601857.CH), and Sinopec Shanghai Petrochemical (600688.CH) will begin the ADR delisting process this month with some maintaining Hong Kong and Chinese exchange listing.  As we have previously noted China Mobile (941.HK), China Unicom (762.HK), and China Telecom (728.HK) were delisted last year based on a previous administration’s administrative order “Addressing the Threat from Securities Investments That Finance Communist Chinese Military Companies” that was left intact by the Biden administration, which was separate from the audit conflict that is the root of the current delistings. 
Pundits are split on whether this would help or hurt the prospects for a deal that might satisfy the US audit rules, but it seems that the current environment is not one under which China is going to be making many concessions, especially if it means opening up books that might reveal connections that could spur further anti-China sentiment as we head into the November elections.  So far no Chinese CE companies have been forced to or voluntarily delist, but the HFCAA (The Holding Foreign Companies Accountable Act) , which contains the rules in contention has implications for a number of Chinese CE companies and the new delisting announcements only add to the tension.
 
 

​Amazon (AMZN) and Cartier (CFR.SIX) have filed a suit against both a company selling items on Amazon and an influencer who helped to promote the products, for selling jewelry and accessories that illegally carried the Cartier trademark.  The suit, filed in the US District Court of the Western District of Washington, alleges that a Chinese entity known as YFXF (Amazon Selling Account name) and a number of ‘John Does’ sold counterfeit products on the Amazon site that bore the Cartier logo, without any license from owner.  In 2017 Amazon opened the Amazon Brand Registry, a free service to any right’s owner that has a government-registered trademark (even those without an Amazon relationship) that uses machine learning to allow brands to search and report potentially infringing products using image-search software.  With more than 700,000 brands in the program Amazon states that those brands are finding and reporting 99% fewer suspected infringements than before joining.
Amazon also employs a system by which every brand in a second service known as Transparency, applies a 2D code to every product they manufacturer, which allows the company, customers, and law enforcement to identify the authenticity of any enrolled branded product.  By the end of 2021 23,000 brands were participating which covered 750m product units, and an additional service, known as Project Zero, allows enrollees to directly take down counterfeit products on Amazon stores.   Amazon and Cartier have verified that the items were counterfeit and therefore violate the terms of the selling agreement that was signed by YFXF.  The suit asks the court to require that the defendants provide an accounting for all amounts due to Cartier and pay such fees including damages, attorney fees, and interest.  Unfortunately, the owner of the Chinese entity known as YFXF has not been named in the suit, along with 10 unnamed defendants, who were involved with or worked for the offending company, all of whom are living in China, which might make it difficult to gain access to the company records, but we have to give Amazon credit for trying, even though we know it was to prove to Cartier that it takes its legitimate sellers seriously.
 

The big guessing game in the CE space is whether the break in TV panel prices will give TV brands the ability to offer substantial discounts during the holidays and stimulate TV set sales, which have been and are expected to be lackluster.  Component shortages, inventory prices, transportation bottlenecks, and TV panel prices all play into such calculations, and while each brand has their own issues, we took a look at the US CPI (un-seasonally adjusted TV index) and how that relates to actual aggregate TV panel prices to get some understanding of where we are in the cycle, and the possibility for a better than expected holiday season.
The simplest comparison was the CPI TV Index vs. the aggregated price of TV panels since January 2020, as shown in Fig. 1.  As can be seen, the TV CPI index lags the index by about a month, with TV panel prices peaking in July and the TV CPI peaking in August, however the TV CPI did not follow the rising TV panel prices seen in the latter half of 2020.  It was not until the beginning of 2021 that the TV CPI began to track TV panel price increases and while the more recent TV CPI data better reflects the change in TV panel prices that began in September, we expect the TV CPI will take longer to readjust to lower panel prices.   TV panel prices have declined 35.3% from their peak while the TV CPI has declined less than a percent, and while this is not a real comparison as the bases are so far apart, at least from a visual point of view, they seem to be tracking more closely over the last three months than in the last 18 months.
But when it comes to consumers, TV panel prices and the TV CPI are far less relevant than what they see when walking into Best Buy (BBY) or are checking TV prices on Amazon (AMZN).  While underneath the loose budgets that consumers will try to hold to when viewing possible TV or other CE choices lies wage inflation/deflation, a smattering of knowledge, much brand hype, and a desire not to buy something that is overpriced, if something looks like a good bargain, it attracts attention.  In order to see whether TV panel price reductions and TV CPI decreases have made an impact on ‘real’ TV prices, we took a simple survey.  We traced the pricing for the 10 top selling TVs at Best Buy to measure where they currently stand relative to their high and low points.  Since Best Buy does not provide historic data, we used pricing on Amazon to trace pricing, and while this was more often 3rd party pricing, rather than Amazon’s, it does give a reasonable representation of the pricing cycles of these TVs.
Just to clarify the data set, the top ten ‘Best Sellers’ at Best Buy consisted of five Samsung (005930.KS) TVs, two LG (066570.KS) TVs, and three Insignia TVs, which are Best Buy’s house brand.  Of the 10, three were 2021 models, five were 2020 models, and two were 2018 (!) models, with one of the LG models being an OLED TV.  In terms of TV size, there were two 75” and two 65” TVs, with the rest with one each (75”, 58”, 55”, 40”, 32”, 24”), with only two being 4K TVs, five being UHD resolution, and the remaining ones HD.  The average price for the ten models was $585 on Best Buy, although leaving out the OLED TV brings that average down to $450.   Prices on Amazon were typically equal to or higher relative to Best Buy, with the average on Amazon 8.9% above Best Buy, but much of the differences came from the two 70” models, which were 29.2% and 66.3% higher on Amazon, likely because they are in relatively short supply.  
What was most interesting was that while the Amazon prices were down 22.5% from their high points on average and up 17.8% on average from their low points, the current prices of the ten sets on Amazon were 8.9% higher than the current Best Buy prices and the current Best Buy prices were only 0.2% higher than the average lowest price point (historically) on Amazon.  If nothing else, it points to the fact that the ‘best sellers’ at Best Buy are better deals than even the lowest historic prices on Amazon.  While the data proves the point that it is certainly better to shop for bargains on-line, particularly among Best Buy’s ‘top sellers’, than to walk into a store and wander through aisles of TVs with a salesman, it does not give any indication as to how much discounting is being done across the broad spectrum of TV sets sold in the US, despite the ‘previous price’ listed for each set at Best Buy.  We believe those ‘previous prices’ should be ignored, considering some of those sets were over two years old and have seen many short pricing peaks and valleys that had little to do with actual consumer prices. 
We also looked at the 10 most and least expensive models offered at Best Buy, however, most were not available on Amazon and price tracking data was therefore not available but just in passing, the average price of the 10 most expensive TVs at Best Buy was $17,155, with the top of the range set by a Samsung 98” 8K set at $60,000 and the least expensive a Samsung 85” 4K TV at $5,500.  At the other end of the spectrum the 10 lowest priced TV sets at Best Buy averaged $121.30, with the most expensive being a $160 Toshiba 32” TV and the least expensive being an Insignia 19” TV at $80.
All in, it will take a much wider sample to see whether the impact of TV panel price decreases will be able to work their way to consumers by the end of the year.  As set producers built inventory against a fear of continuing component shortages, they built in months of higher cost inventory that still has to be worked through or written off at the end of the year.  With three months of lower cost TV panel inventory, the cost average will come down, but given the multitude of other factors, particularly raw materials, semiconductors, and logistics, we expect much of the effects of TV panel price reductions will be seen in 1Q, making the post-holiday period a more likely place to be looking for real TV bargains.  TV panel producers are already seeing the effects of the price declines and have begun to lower utilization rates, which will continue to pressure y/y comparisons, but the bigger question is whether TV set brands will keep any margin they gain, or pass it on to consumers to move units.  Given the margin issues that have faced TV set brands for much of this year, we expect the former but hope for the latter.

​While many manufacturers complain about component shortages and transportation issues making it difficult for product to reach retail shelves before the holidays, Nintendo (7974.JP) has made a very definitive statement about its plans to reduce shipments from an expected 30m Switches in fiscal 2021 to 24m, despite strong demand in Japan, said to be at a record level.  Nintendo’s President Shuntaro Furukawa stated that “It is not at all that the sales peak has passed, it is that the production volume cannot be produced in full and the production plan is unclear.”
Accordingly, the company notified its suppliers of the reduction, which is based on a lack of silicon based components that have already limited supply in Japan, where stores are resorting to lotteries to manage the lack of supply. Retail store comments have indicated that volumes received are less than half of what was expected, causing sales in September to decline by 37% y/y, following two months of y/y sales declines.  As a check, we looked at the price of the 6.2” LCD Switch on Amazon (AMZN), which can be seen in Fig. 2, a good indication that it is hard to come by.