Smartphone shipments in China saw a large jump from a weak February, ↑65.6% m/, and while this is below the March 5 m/m average of 109.8%, the more telling metric is the March y/y shipment increase of 65.9%, far exceeding the 5 year average of ↓8.5%.  Thus far, all three months this year have seen very strong y/y increases, although we temper that given that the COVID-19 pandemic had its greatest effect on smartphone shipments in China in 1Q of last year.  That said, absolute shipment numbers are also higher by 17.9%, 50.3%, and 27.1% when compared against 2019 shipments, so the trend of negative smartphone growth in China seems to have ended for at least the 1st quarter of this year.

5G Smartphone shipments in China were 27.5m units in March, back to January levels and at the highest share of total phone shipments sine 5G shipments began in China, which was in mid-year 2019., and the share of 5G smartphones relative to all types shipped continues to increase.  While the 5G share of new models remains in a slow but steady uptrend.  In all 1Q trends in the Chinese smartphone market have been positive on both a relative basis (against last year’s COVID-19 impact) and on an absolute basis, which makes the data that much more positive.  This differs somewhat from inventory building we have seen with a number of other CE products, but as this data represents product shipped and not sold through, we can only surmise that brands and dealers are not building appreciable inventory and that the demand increase is real and will continue into 2Q.

​In Mid-December we noted that Taiwan based FitTech (6706.TT) had recently been officially added to the Apple (AAPL) mini-LED supply chain as a provider of mini-LED testing tools.  FitTech stated recently that it has received ‘significant’ orders for equipment for probing, testing, and sorting mini-LEDs, with shipments scheduled until September of this year and expects sales of such equipment to increase from 20% - 30% of revenue last year to 40% - 50% this year, with over 75% of 1Q sales from same.  While the company mentioned that Chinese and Taiwanese LED manufacturers have been expanding production capabilities for mini-LEDs, we expect demand, or at least perceived demand, is being driven by the supposition that Apple will announce a new iPad Pro this month with a mini-LED display, which will begin production in early May.
While Apple tends not to be a first mover when it comes to new technology consumer products, they are certainly one when it comes to doing early stage R&D and even manufacturing process R&D, whether by themselves, through acquisition, or with partners, as we have noted a number of times in reference to mini and micro-LEDs.  However the idea of Apple releasing a mini-LED product this year is not nearly as much of a stretch as one might think given that mini-LED technology is merely an extension of existing LED backlight technology.  Of course there are issues that are specific to mini-LEDs given their small size and large numbers relative to standard LED backlights, but they are manufactured in the same way (MOCVD) as standard LEDs, allowing a supply chain that is similar, if not the same as traditional LED backlight production.  This allows Apple to use existing relationships and suppliers for any mini-LED product, which reduces risk, leaving only the cost of mini-LED backlight arrays as a point of contention.
Fittech could be seeing a general increase in interest for mini-LEDs from LED producers or a more directed focus from suppliers or OEMs that are expecting to be included in the production of an upcoming Apple product using mini-LEDs.  That said, Samsung Electronics (005930.KS) and Chinese TV brand TCL (000100.CH) have already released mini-LED TV product, with TCL on its 2nd generation, and LG Electronics (066570.KS) is expected to at least announce a similar product this year, although they will be careful not to overstate the technology relative to OLED.
What mini-LEDs do for all TV and CE brands that are looking for product differentiation at the high-end of the market, is give each their own twist on the technology for which there are no set standards.  This allows each to vary the number of LEDs (usually the more the finer contrast control) and other characteristics to give marketing a point at which to boast about their particular product, a most important characteristic in an increasingly competitive market.  If FitTech is any indication, the momentum is just beginning.

​Chinese panel producer Viusionox (002387.CH) has obtained an order from Honor (pvt) the company recently separated from original parent Huawei (pvt).  The order worth ~$136m is for flexible display and touch screen components.  The displays and other components are expected to be produced this year at Visionox’s Gen 6 OLED line in Hebei and is expected to have a significant impact on the company’s unconsolidated sales for this year.  Honor is no longer associated with Huawei, who remains under US trade sanctions and has seen vastly reduced smartphone shipments over the last few months.
 

​As we noted on 3/30/21, Corning (GLW) raised glass substrate prices ‘moderately’ in 2Q, a result of increased raw material, energy, and logistics costs, which Corning had been absorbing until the price change.  As noted, typically glass prices decline each quarter so this was a seminal event, but given the commonality between display glass suppliers, of whom there are few, it is logical to expect that others would follow.  Last night in Japan, AGC (5201.JP) announced a revision to its forecast for 1H results that indicated a 6.6% increase in sales from the previous forecast, an 88.8% increase in operating profit, all of which would result in sales growth of 22.2% y/y instead of the previously expected 14.6% and a 313% y/y increase in operating profit.  AGC also raised full year 2021 sales expectations by 7.8% and operating profit expectations by 60.0%, indicating a slight increase in sales expectations for 2H with an accompanying increase in operating profit, albeit at a lesser arte than in the 1st half.
AGC and NEG (5214.JP) both experienced issues at glass facilities recently, with AGC’s Gumi glass plant experiencing an explosion in late January and NEG suffered a power outage last December, both of which put additional pressure on an already tight substrate glass market.  However the loss of capacity was not enough to offset the increase in glass prices at all three major glass substrate suppliers, with NEG raising guidance earlier this month.  While certainly a positive for glass producers, with the glass substrate is usually among the highest single cost component of displays, particularly large panel displays, these component price increases continue the push toward higher consumer CE prices, with panel producers now passing on such increases as lower cost component stock is diminished.  Made worse by shortages of silicon (foundry) components, such end user price increases will become increasingly more commonplace going forward, and unless the world sees an unprecedented financial boom, eventually such price increases will curtail demand.

​According to South Korean trade press, it seems that Samsung Electronics (005930.KS) is going to enter the OLED TV space this year, but not the QD/OLED set that affiliate Samsung Display (pvt) is developing (2022 more likely), but sets based on OLED panels produced by South Korean rival LG Display (LPL), the dominant supplier of large panel OLED displays.  The two companies have said to have met voluntarily under the auspices of the Korean government and agreed on the framework with final details to be ironed out in the near-term.  According to the source, Samsung and LG Display have committed to 1m units for this year and 4m units next year.  Given that LGD’s expectations for total OLED panel production this year is between 7m and 8m units, the Samsung deal makes Samsung suddenly a large customer (quick estimate ~13%), and an even larger one next year if the deal plays out (~30%).
Samsung Electronics and LG Electronics have made some rather scathing comments about each other over the last few years while Samsung Display and LG Display a bit less, but this is certainly the first time that we can remember that Samsung has purchased OLED panels from LGD.  We expect Samsung’s TV division understands that as Samsung Display winds down its LCD large panel production and begins scale production of its QD/OLED product, which could take time to produce acceptable yields, the division needs to keep itself as exposed to the premium TV market as possible.  Samsung’s quantum dot LCD TVs certainly maintain a slot and the company will be building out its mini-LED TV offerings, but as the price of OLED TVs declines, especially relative to LCD TVs, Samsung must continue to feed customers new and different high-end TV product.
While justifying the potential QD/OLED TV would be relatively easy for Samsung’s marketing, given that Samsung has championed the use of quantum dots in the TV space for years and is the leader in small panel OLED production (“The Best of Both Technologies…”), but we expect would have a harder time justifying a pure OLED TV, especially one based on LGD’s WOLED technology, which Samsung has trashed many times over the years.  Of course, they could choose to ignore the past (likely) and market the new OLED line as “Better than the Original…” by adding some bells and whistles not common to other OLED TV sets (likely), but in any case the real take away here is that regardless of stance or ‘face’, if there is a better way to generate profits for Samsung’s TV division, they will do it.   IF LCD large panel prices continue to rise and OLED panel prices rise less, that spread is what matters and Samsung will likely tell new product marketing to stop thinking about how to sell the new line rather than dwelling on what has happened in the past.  Samsung Electronics has stated that “The official position is that it is not true because it has not been confirmed.  However with regard to securing parts, it is the basic policy to review competition without restricting them.”

Disclosure with Chinese companies can be a bit problematic and rarely do you see monthly sales or shipment reports from Mainland companies, but Truly (0732.HK) is a Chinese panel producer that lists on the Hong Kong exchange, which, as in Taiwan, requires monthly sales disclosures.  This allows for a comparison with results from Taiwan panel producers as noted above.  Truly is a small panel producer, with Gen 2.5 and Gen 4.5 lines producing at a stated capacity of 15,000 sheets/month.  While this does not compare directly with AU Optronics or Innolux, Hannstar similarly produces primarily small panel product.
Truly saw an increase of 23.2% in m/m sales for March, very similar to Hannstar, after m/m declines in January and February, again similar to Hannstar.  While the incremental gain was similar in March, Truly saw only a 3.1% increase on a y/y basis, while Hannstar saw over 100% y/y growth, a function of what we believe is Truly’s maxed out capacity.  As a relatively small player in a market that is increasingly dominated by large panel producers, Truly is one of those producers that lives on the razor’s edge but cannot break out of its niche without capacity expansion.  More likely the company will be absorbed by a state-owned entity when panel prices are in cyclical decline, but until then it gives anecdotal evidence as to the relative performance of panel producers in the very short-term, unusual for Chinese companies.

All three major Taiwanese panel producers reported strong panel sales in March, not surprising given the panel price increases that have been instituted over the last year.  AU Optronics (AUOTY) reported March sales of NT$ 30.72b ($1.08b US), ↑20.0% m/m and ↑49.8% y/y which included a 9.9% sequential increase in area shipped, leaving the rest of the m/m gain attributed to price increases and mix.  AUO no longer reports large or small panel shipment numbers.  Innolux (3481.TT) reported March sales of NT$ 30.99b ($1.09b US), ↑22.3% m/m and ↑57.5% y/y.  Innolux shipped13.36m large panels in March, ↑15.5% and ↑36.5% y/y, with small panel shipments of 29.87m, ↑19.3% m/m and ↑33.7% y/y.
Hannstar (6116.TT), who specializes in small panel production, saw sales of NT$ 3.03b ($106.4m US), ↑27.3% m/m and ↑132.7% y/y. with small panel shipments of 31.3m units,↓6.7% m/m and ↑3.5% y/y.  Hannstar shipped 525,000 large panel units in March, ↑176.3% m/m and ↑212.5% y/y, which was the largest number in the company’s history and more than it shipped during the entire 4th quarter last year.  Given the rapid increase in large panel prices, we expect Hannstar pushed out as many large panel units as capacity would allow, which added to March profitability despite the decrease in small panel sales.
While product and panel type mix would certainly have an influence on panel revenue, we expect that panel pricing and demand were both responsible for the March sales increases.  Panel pricing has been finite in that we can get a good read on panel price movement in a wide variety of sizes, but demand is far more difficult to quantify, and the continuing litany of stories we hear about double and triple ordering and exclusive deals being made in both the semiconductor and the display space add to that lack of transparency.
That said, we expect external (meaning what it would be without double or triple ordering) demand was sequentially better than February, as the Chinese New Year celebration ended and an aggressive vaccine roll-out and an additional monetary stimulus in the US gave consumers in these, the two largest CE markets, a chance to get out and spend.  That new-found optimism as to the world’s ability to maintain some control over the COVID-19 pandemic seem to encourage brands to hold their shipment and share expectations steady for the 2021 year, while the fear of shortages pushed brand buyers to keep an aggressive stance with suppliers as low-cost inventories diminished.  In the mind of CE buyers, price is less important than supply, which continues to feed panel price and semiconductor price increases.
We have just begun to see the results of those aggressive panel price increases, with a recent round of CE end product price increases, which we expect will continue as long as semiconductor foundry and panel production utilization remains high.  The problem we see however, is not whether consumers will rebel against higher CE device prices now, but rather when price increases continue after demand slows and component price increases flatten.  As the CE space builds inventory to ‘protect’ itself from shortages, that high cost inventory will take months to work down, keeping end-user prices high while component prices are flat or declining.  We see the next few quarters as a motorcycle jump over the Grand Canyon.  It's looking good from the jump point, but just a few inches short and it’s a long way down…

Industry Associations are usually good sources of data, albeit usually with a positive bias.  The data below is compiled by the ECIA (Electronic Component Industry Association), an organization whose members are leading component manufacturers, reps, and distributors.  Among other functions, the organization polls its members monthly and quarterly to ascertain the state of the industry.  While much of the data is for paid members, the ECIA does occasionally revel some of the survey data.  The March survey saw the sales sentiment among members increase substantially, indication a strong surge in demand, with an increase of between 26 and 27 points to 158 in March, against 132 in February (100 or above equals ‘growth’), which is the highest point in the last 18 months.  While expectations for April growth are not as high as March, the number will still be near 140, a positive sign for the component industry.
The ECIA also indicated survey results in reference to the m/m increase in lead times for certain components and the y/y increase in semiconductor sales by region.  While we have focused on MLCCs, a subset of the capacitor market, which itself is a subset of the passive component market, it seems that a rising tide lifts all boats, as can be seen from the substantial lead time increase for all passives.

​Xiaomi (1810.HK) sold the most TVs in China last year, coming 8 years after it entered the TV market, and and following the same metric in 2019..  Known for their ‘value,’ Xiaomi TVs continue to widen their reach and after entering the Indian TV market in 2018, captured the greatest TV share last year.  All of that said, Xiaomi is not immune to the rapidly rising cost of TV components, particularly LCD panels, and has now joined many other TV brands by notifying customers that it would be raising TV prices.. 
The release indicates that the cost of materials (panels and chips) has been rising and is expected to continue to rise, which will force the company to ‘adjust’ the retail price of its TVs accordingly.  The company states that it will take full advantage of the supply chain to ‘ensure’ that after the price adjustment, the product will still be very cost effective., and then goes on to list some of the company’s latest achievements.  No mention as to when or by how much retail TV prices will rise, but it seems quite inevitable that as older component inventory dwindles those increases will occur.

​In a note we posted in January, we mentioned the problems associated with Wi-Fi in the city, and for that matter in many places.  This affects almost all applications that move data from a source, such as a gaming or streaming content server to your laptop, smartphone or TV.  Even with the most current version of Wi-Fi, which is 802.11.ax, also known as Wi-Fi 6, there are those times when instead of watching “The Mandalorian” you are watching a rotating disc or similar object, or your Zoom call freezes while your Wi-Fi router fights for a connections.
It’s not just that your kids are sucking up bandwidth with games or other bandwidth hungry applications or that your wife is watching “Bridgeton” on her smartphone, but your router must also contend with any others that are within range, including your neighbor’s router, which is also contending for access to Wi-Fi 6 channels.  They are not ‘stealing’ your Verizon (VZ) or Optimum (ATUS) bandwidth, but on a technical basis both you and your neighbor are vying for the same Wi-Fi channels, which are limited in size.  If the demand for Wi-Fi bandwidth exceeds what a channel can provide, someone gets dropped and the disc starts spinning.
Wi-Fi 6 uses two frequency bands, 2.4GHz and 5GHz, and while like 5G spectrum, the 2.4GHz band is able to reach further but is slower, while the 5.0 GHz band is faster but has a smaller reach.  In theory the Wi-Fi 6 specs allow a maximum speed of 9.6GHz in either band, but on a practical level that is commonly closer to 1Ghz. to 2Ghz.  This limits the total bandwidth that can be accessed by your router, even if it is one of the relatively new Wi-Fi 6 routers, but the FCC actually did something right last year and allocated another band, 6GHz to what is called Wi-Fi 6E to give a bit more bandwidth to users.  In fact, the new band actually quadruples the bandwidth available to Wi-Fi, so there will be less contention for full speed data transfer requests, but there is a catch.  In order to utilize Wi-Fi 6e, as the new version is called, you need to buy another router…
After a while buying new routers to take advantage of higher potential Wi-Fi speeds can get expensive and a bit frustrating, however Samsung (005930.KS) has come to the rescue.  The company has joined up with Mediatek (2454.TT) who is providing its MT7921 AU chipset to Samsung’s 8K QLED TV line which, as of yesterday, now includes the ability to receive Wi-Fi 6E signals.  While this doesn’t sound like much, the addition of the 6GHz spectrum opens up the equivalent of 28 80Mhz wide channels that will ease the contention that many Wi-Fi systems work under and increase realized data speeds.
As with all technology there are ‘catches’.  Again similar to 5G (mmWave vs Sub6) as the frequency of the usage band increases, it has a lesser broadcast reach and can more easily be blocked by obstructions such as thick walls or floors.  In order to compensate for these limitations, the ‘newest’ routers (which you will have to buy to use the WiFi 6E Samsung TV) can ‘meld’ low and high bandwidth when blockages occur, and as a portion of the 6Mhz spectrum is still reserved for licensed users, you cannot use your Wi-Fi 6E bandwidth outside, as you might interfere with the traffic of a licensee.
All in, the adoption of Wi-Fi 6E in a TV set is a good thing and one that will likely become common within the next year, at least for high-end TVs, but will be preceded by replacing current Wi-Fi 5 or even Wi-Fi 6 routers with those that are compatible with Wi-Fi 6e, and those are in short supply currently.  Samsung seems to be looking a bit ahead, but being first to be certified “6e” is a big deal in the CE marketing game.  No mention as to whether there would be any change in price (not likely) as the chipset remains the same, but the Samsung QN800A (8K) series sells for between $3,500 and $6,500, while the QN900A (8K) series sells for between $5,000 and $9,000 depending on screen size.