9.3% of the US population wear contact lenses and 9.0% of the US population have type 2 diabetes.  What does this have to do with technology?  A Korean research team has figured out a way to measure blood sugar levels by embedding grapheme and metallic nano-wires into soft contact lenses, with the added function of also measuring inter-ocular pressure, a monitor for glaucoma.
Typically type 2 diabetics have to draw blood and test with indicator strips as many as 5 times/day in order to maintain correct glucose levels.  This is an inconvenient and sometimes painful process, albeit essential to keeping diabetes under control.  There have been a few commercial soft contact lenses that have attempted to take similar measurements using the eyes tears as a glucose measurement source, however the sensors needed limited users’ vision and have been unsuccessful.  The research team at the Ulsan National Institute of Science and Technology in Koreas has developed a transparent grapheme sensor that measures both blood glucose and inter-ocular pressure (high intraocular pressure can lead to glaucoma) and therefore does not limit the user’s visual range.
Does this mean you have to wear a hat containing batteries to power the sensors in your eye?  No, given that the embedded wireless antenna both sends data and can receive power from a wireless source, there is no need for internal power.  While still in the early stages of testing (live rabbits), the lenses continued to provide data even when modified or exposed to foreign substances in the eye.  Finally, an IoT concept that has a truly practical and commercially appealing value, rather than one that saves the user from having to get out of a chair to do something.  Think of the ramifications, should the testing and trials be successful, the data from your eyes, aside from being sent to your doctor, can also be sent to your refrigerator, which will say, “Dave, I am afraid I can’t open the refrigerator door right now.  Your blood sugar is over 200 and that chocolate cake must stay in the refrigerator.  Why don’t you ask drawer #7B to give you a rice cracker?”  OK, maybe a little over-the-top, but even though it’s still in the early stages, such a device would have a very big impact on a sizable proportion of the population, and would help us to believe that IoT is less a marketing tool and more of a possible help to humanity.  

LED Street lights have been evaluated ad nauseam, especially by suppliers eager to help municipalities replace their halogen or mercury-based street lighting systems.  There is little, other than upfront cost, to deter government buyers from replacing the higher maintenance and shorter lifetime existing lighting with more efficient and longer lifetime LED lighting.  There are questions about light pollution and color choice from consumers, but the data (as long as actual lighting meets manufacturer specs) is pretty uniform and continues to improve as better luminaire designs reduce glare and other issues.  That said, there isn’t a day when we don’t see headlines about IoT or connected appliances, lighting, doorbells, or security systems, let much of the LED street lighting that is going into communities do not share information or even generate information.  Given the bean-counter approach that many local governments have to public works projects, one would think that they would be interested in such functions to make ongoing improvements to street lighting generally, or at least to protect their jobs.
What holds back the idea of ‘smart’ streetlights, especially as they are in a unique position to count or identify consumers, generating trend information for managers, or less sophisticatedly reporting maintenance issues?  Networks and power.  We believe smart street lighting has a share under 5%, as local and even city sized programs are unfamiliar with the network technology needed to connect and collect data from smart street lights, along with concerns (incorrectly so) about the extra energy needed to power street lamp data collection.  In most cases, the extra power needed for sensors and data transmission is easily offset by the application of such data to make the street lights operate more efficiently.  Remote location streetlights can be dimmed when there is no traffic and returned to full brightness when approached by a moving vehicle or pedestrian, or daylight sensors can adjust brightness according to actual conditions, rather than by a 24 hour clock.
Networks for data collection however can be a hindrance to local governments, who are not used to such tools and usually don’t have the expertise to maintain such networks once they are installed by providers.  Then there is the data, and the need to collect, store, and interpret information that can be as diverse as weather, traffic, breakdown detection, and essentially any other sensor information that is included in the luminaires.  Small towns are not used to big data, and the cost of the network, data storage, and the analysts needed to read and gain insight from such vast quantities of information likely look overwhelming to local government officials, many of whom have no experience in such matters.
But, as the world of IoT continues to expand, the cost/unit for data collection will continue to decline, and point-to-point wireless networking will become less costly, leaving ‘talent’ the only stumbling block to the adoption of smart street lighting.  Taiwan based Lite-On Technology (2301.TT), the market share leader in Taiwan and North America, who ships ~500,000 street lights/year, is expecting to see a 5x increase in the number of smart street lights it ships this year, amounting to ~5% of overall shipments.  A recent Lite-On program in Taoyuan City (Taiwan) that will replace 21,400 mercury vapor streetlights with LED lighting, will be oriented toward a 50% smart street light goal, and other cities in Taiwan have similar programs that specify smart streetlights for certain locations, but a generational change is still necessary for government officials to see past the cost of setting up the systems necessary to collect and use the data.  Gen X officials are probably too old to feel that networks and data are part of life generally, but millennials will not think twice about the concept, especially if you like them on Facebook (FB), so vote young for better street lighting, and make sure not to toss anything from your vehicle…they can see you now…

The recently released LG (066570.KS) G6 flagship smartphone uses an LCD display, indicating that LG Electronics continues to largely ignore OLED display technology.  They have used OLED displays on the LG G Flex (released 10/2013) and the G Flex2, but more to emphasize ‘bendability’, rather than the quality of the OLED display or its conformable characteristics.  As Samsung Electronics (005930.KS) champions ‘flexible’ (read conformed) OLED displays for its mobile devices, and Samsung Display (pvt) is the world leader in conformed OLED commercial production, LG Electronics’ primary supplier, LG Display (LPL) has been focused on the use of flexible OLED for large panel displays, as in their ‘wallpaper’ flexible TV, and has tacitly ceded the rigid OLED space to its competitor.
Now that Google (GOOG) and Apple (AAPL) have and are expected to enter the curved OLED smartphone space, it looks like LGE is going to give in and release a conformed OLED smartphone later this year, as LG Display improves manufacturing yields on its OLED displays.  Not only is LG Display expected to supply such conformed OLED displays to its parent, but Beijing based Xiaomi (pvt), is expected to receive similar displays from LGD for its Mi Note 2 successor, expected toward the end of this year.  LG Display is expected to start delivering such modules toward the end of this quarter, and recently indicated that it is focused on improving its capabilities and yields on its flexible OLED process lines. 
While Samsung Display has been a volume supplier of similar displays to Samsung Electronics and others for some time, the flexible OLED manufacturing timeline is still in its infancy, and LG Display and other potential producers have a better chance of competing with Samsung in the flexible OLED arena than in the rigid OLED display space.  With the need to offset the potential loss of a portion of Apple’s LCD iPhone display business, LG Display is making a concerted effort to bolster its small panel OLED manufacturing, and will spend 70% of its 2017 capex budget on OLED, with a new emphasis on small panel and flexible OLED.  Last year, the OLED capex was focused on expanding the company’s large panel OLED TV capacity, which is now capable of producing ~1.6m units this year.

Shanghai based BOE (200725.CH) has shown an updated 7.8” flexible OLED display at a technology exhibition in China, and while we have seen innumerable flexible displays at shows over the last few years, we have yet to see any such technology applied commercially, including BOE’s.  That said, there have been significant improvements in flexible OLED technology and significant flexible OLED capacity is being built to enable new applications, particularly in the mobile market. But we believe the applications for flexible displays are far more widespread than mobile devices, particularly in the digital signage area.  Being able to create a display that follows a complex surface, especially a display that is actively providing real-time information, is a dream of advertisers everywhere, and applications for public spaces, such as airports or subways are endless, as are the commercial retail applications in brick-and-mortar settings.
The BOE device, and we do not single it out for any reason other than it was just exhibited, is based on an LTPS backplane with an OLED display that is .24mm thick (normal human hair is between 0.06 and 0.08mm), and can be used open as a tablet, or folded to smartphone size.  The company gave no indication as to the timeline for commercialization, although we expect to get a bit more information at the upcoming SID show next month.

AU Optronics (AUO) reported strong 1Q results with net profit up 5.7% q/q and EPS exceeding consensus.  Sales had been previously reported.  AU Optronics no longer has an English conference call, so details are few, other than that the company expects order momentum to increase in 2Q and that utilization rates should remain high.  We will have more when the English translation is released.

Smartphone demand was weak in 1Q, as noted by panel producers in their 1Q results, and smartphone panel pricing declined between 1.5% and 11.1%, which represents a significant change from the last year.  Chinese panel producers have been looking to gain share among domestic smartphone brands and have been offering prices for LTPS based smartphone displays that compare to those for standard a-Si based displays, where typically LTPS has garnered a premium price.  LTPS backplanes are also used for small panel OLED smartphone displays and discounting has been rumored for some LTPS OLED displays, again by mainland panel producers.
Most Chinese panel producers have exposure to the small panel smartphone market, with Infovision (pvt) having the largest share (8.6%), followed by BOE (200725.CH) with a 5.9% share, and Tianma (000050.CH) with a 3.6% share, while China Star (pvt) and Panda (pvt) having little impact.  When compared to top small panel producers such as LG Display (49.2% share), they have a long way to go, and their ability to attract mainland brands is even more important. We are concerned that in cases where demand for a particular display type weakens, as it has in smartphones, price competition will heat up quickly and erase much of the pricing increases seen in the last 12 months.  The Chinese smartphone market is more mature than the Chinese IT market, so we would expect such smartphone discounting to gain share to be a bit unusual in the current environment, but it goes to show how quickly the panel pricing picture can change, and how quickly ‘follow the leader’ pricing can become the norm. Demand levels for IT (notebook and monitor) are less of an issue in China, and the Chinese TV market has a life and characteristics that set it apart from other regions, but the risk of panel pricing changes continues to increase, with the smartphone market foretelling how buyer philosophy can change on a dime.

LG Display (LPL) reported their 1Q results, which were ahead of consensus on an earnings basis.  We summarize the results below and add our comments in red.  On a general basis, the call was a bit more forwardly cautious than we would have expected, and the company continues to hold off making capex decisions as long as possible to better gauge whether there will further panel price increases or positive changes in demand.  In terms of the shift toward OLED, which LG Display champions, the focus was decidedly toward flexible OLED rather than the usual focus on OLED TV.  While we do not think this indicates any change in their attitude toward OLED TV, OLED capex is likely to shift toward small panel, flexible OLED, as the company is one of the few that can seriously compete with small panel OLED market leader, Samsung Display (pvt).  That said, we would not expect too much out of LG Display on the small panel flexible OLED front, as they are still in the early stages of high volume manufacturing ramp and have to balance low yields and negative EBITDA against the need to gain traction against Samsung Display.  All in it was a good quarter, but hopefully not the best of the year.
1Q Results

• Sales down 11% q/q – up 18% y/y – small panel seasonality and lack of additional capacity limit q/q sales – easy comparison vs. last year
• ASP generally up but ASP per m2 down as small panel volume declines – mobile mix from 31% in 4Q to 26% in 1Q confirms this
• Best operating profit ever on strong margins
• Strong margins due to better panel pricing on TV – TV mix up from 38% (4Q) to 43% in 1Q offsets loss of higher ASP/m2 small panel mix shift
• Capacity down 6.1% q/q – Up 5.1% y/y – Closings of Gen 5 and smaller LCD fabs and OLED conversions influencing capacity
• Industry inventories flat q/q – Industry tight for large panel – Size specific but no comment on small panel industry levels

Construction

• E4-2 – Expansion complete in 2Q – A bit behind our original estimates but actual production in-line as we built in an extended ramp schedule
• E5 – Flex line – On schedule for 2H production – Also behind our original estimate (end of 1Q) but we built in an extended ramp schedule here also
• G1 – OLED Lighting – Evaluating 15k production in 2H – We estimated 1/1/2018 for initial 15k production -  2H would be a bit ahead of schedule
• Capex could increase from expected levels – too early to tell – Likely no decision until end of 2Q
• 70% of capex to OLED – focus on small panel/flex this year – Apple (AAPL), Apple, Apple…
• Noted a more conservative view of spending on OLED this year – OLED capex is initially higher/m2 than LCD

2Q Guide

• Area Shipments Flat q/q – Capacity limited – Will remain so in 2Q
• TV Shipments Flat to down q/q – Pulled some 2Q TV demand into 1Q
• ASP by Area – down single digit – small panel seasonality
• Capex – Additional comments
  • Will continue to move from LTPS LCD to OLED but still opportunities in LCD Automotive, Ultra Hi res – Same as most other large panel producers
• LCD Panel Pricing – Demand still shifting to larger panel sizes but potential resistance to higher pricing – Brands unable to be profitable at higher panel price levels

OLED TV

• Optimistic internally about OLED TV profitability being better than expected – exceeding expectations – Hard to evaluate ‘optimism’
• Production ratio 3x – 1H to 5x – 2H – As expected
• 55”, 65”, 77” reached ‘golden level’ – The dollars in my wallet have reached ‘golden level’ – Does that mean I have $5 or $5,000?  This refers to OLED TV panel yield which continues to improve but no specifics were given

General

• “Is high LCD profit margin justified?” – Technical differentiation and reliability justifies it – Good until its not, but it pays for OLED capacity expansion now
• 8K in 2019 – Panel producer perspective – Maybe not the same as consumer perspective after slow 4K content ramp
• What about sluggish smartphone growth? – Growth is limited but hi-end growing and capacity tight – Capacity tight at high end but expanding – move to OLED more important than higher resolution

This was a good quarter for LG Display, but not unexpected.  On a general basis profitability was better than expected.  LG Display should remain at relatively high fab utilization levels (LCD) at capacity reductions limit sales upside and OLED conversions will not replace LCD capacity on a one to one basis.  That said, the shift toward small panel and flexible OLED, while obviously motivated by Apple, will take some time to produce revenue and profitability benefits.  Near-term, much rests on LCD panel pricing because of the limited capacity and the growth of ultra-large TV sizes, where the company expects greater than 10% growth and will expand its offerings in 2H for the holiday season.
LG Display has yet to decide about much of its longer-term construction plans, with such decisions not likely before July/August.  Overall capex, LCD/OLED ratios, Gen 10 fab possibilities and the rate of OLED conversions are all up for grabs, as the company and industry gains have been based more on rising panel prices and less on improving demand as TV brand sales struggle.  Stable large panel pricing should give LG Display at least another quarter of strong results, but the risk of lower panel pricing is higher now than in the previous quarter and the loss of Apple demand has to be offset by increased sales to Samsung as small panel flexible OLED capacity and yields are still at low levels.  Still a good quarter, and in the panel space, you take them when you can and hope the cycle is sustainable…

SNU Precision (080000.KS) a South Korean based producer of OLED deposition equipment, has extended its contract with Truly Semiconductor.  Hong Kong based Truly (0732.HK) is a producer of both LCD and OLED displays through its fabs in Huizhou, China.  The company is currently running a Gen 4.5 LCD and OLED fab, which it is building out (phase 2) with a target date of January 2018.  We believe current capacity at that fab is ~15,000 OLED sheets/month.  Truly is also building a Gen 6 LTPS OLED fab, although the target date for that fab is still undetermined (we have seen dates from 4Q 2018 to 1Q 2020).  We assume that SNU is providing deposition equipment for either the Gen 4.5 or Gen 6 build-outs as their DSP deposition tools are scalable to substrate sizes over Gen 5, as are their thin film encapsulation systems.
Truly is primarily known for its wide variety of relatively small panel displays, although the Truly name is not a brand known for its own products.  They are a supplier to a number of display segments including appliances, automotive, and mobile devices, but have been making a concerted effort in the OLED space, albeit with most of their OLED panels being under 3” and more often mono color displays.  On a gross capacity basis the company currently has a less than 1% share, which will likely remain under 1% even with the potential build-outs by 2020, as other producers also add both LCD and OLED capacity.  Truly has shown a number of flexible displays over the last two years, and the new OLED lines are expected to be able to handle flexible substrates, but we expect they will face the same challenges as other producers when bringing that technology into mass production.  We expect to see newer Truly flexible displays at the upcoming SID conference next month.

Corning (GLW) reported strong 4Q results, driven by display, optical, and specialty materials.  To put the display side in perspective, we note that 1H 2016 was characterized by very weak panel pricing and slowing display volumes, while 2H 2016 saw increasing panel pricing and moderate volumes, which makes GLW’s display results all the more significant, but the key last year, and likely again this year is glass price declines.  The company has indicated that they expect glass price declines to be ~10% this year, and possibly better, which implies 2016 glass price declines of 12% to 13% (~1%/month).  We note that during other industry cycles we believe price declines have peaked at ~2%, so GLW has been seeing very moderate pricing pressure for the last year or so.
Why is this, especially during 1H last year when panel prices were dropping precipitously? Here’s why… the display glass market is a triopoly[1], with Corning, Asahi Glass (5201.JP), and NEG (5214.JP) being the major players.  There are others, such as AvanStrate (Pvt), St. Gobain (SGO.FP), Schott AG (pvt), and a few small Chinese producers, but the display glass business is allocated primarily among the top three, particularly as it refers to large panel display production.    In previous display cycles the three primaries were focused on gaining market share, which meant discounting to attract new customers or increasing penetration at existing ones.  While sometimes successful on a short-term basis, inevitably share changed little long-term, and took a big chunk out of profitability.  After the last display downturn cycle, glass manufacturers changed their focus from share to profitability and reduced capacity expansion to match demand.  This philosophy has changed the display glass supply/demand picture since then, and while Corning still maintains better profitability and a higher share than the others, glass pricing declines have moderated and continue to do so.
This glass price moderation comes during periods when capacity demands have been relatively modest, with average display size increases keeping overall glass demand positive, again particularly on large panel (TV) displays.  With only three major players, a more rational approach to glass capacity expansion has led to glass price declines of less than 1%/month.  This coupled with the need for display producers to be guaranteed a percentage of total capacity from their chosen glass supplier, especially when display utilization rates are high, as they are currently, leads to a strong glass pricing environment, as we see now.
The next question is ‘How long can it continue?”.  Has this been a sea change across the industry or is it just another cyclical bump?  According to most glass producers, better glass pricing is endemic to the industry, but even Corning, the market leader, while optimistic for the year, understands that things do change, and the industry has been operating under a positive panel price scenario for the last three quarters, and panel producers are more interested in keeping utilization rates near 100% to capture that positive panel pricing than they are about material cost[2].  They need glass to keep their lines moving and price is less of an object than guaranteed volume.  We believe this means logically that should the panel pricing environment begin to shift back to negative pricing, we could see some pressure on glass pricing return.  We doubt it would have an immediate impact on display glass producer results, but a sustained or aggressive downward panel pricing trend would imply either a supply situation trending toward oversupply, or a weakening of demand, either on a unit volume basis or a screen size basis.  While the oversupply situation would be far better for display glass suppliers, as volumes would still be high despite weaker glass pricing, a slowing of display panel size increases would imply lower glass volumes and lower glass prices, the worst of the two scenarios.  If the three glass suppliers react by shuttering capacity, the industry will remain tight, even in the later scenario, so the answer to the question of ‘How long can it continue?’ is less a question of panel pricing and more a question of how the glass producers react to it.
But wait, there’s more…Corning’s display glass business is profitable, with display margins topping 30%, but competitor display glass margins, who have broader glass businesses, are lower, and in some cases close enough to breakeven that a return to more aggressive price declines could push results to or near loss levels.  While this is not the case currently, it does put into question any potential glass production capacity increases that might be needed over the next few years.  Corning has glass producing assets that can be converted to or from display glass, a portion of which came from its purchase of the Samsung/Corning JV, but others are less flexible, particularly as float glass melters are built out on a larger scale than the Corning more modular tanks.  This gives Corning a competitive advantage in such situations, something the company is not shy about espousing, but it does not protect the display glass industry from a return to more aggressive negative pricing should the panel producers find themselves in more of a buyers’ than sellers’ market.  We are not predicting same, just pointing out the scenarios, as we have learned over years in the display space that things change on a daily basis which makes predictions good until the next morning.  We have a bias toward a somewhat less benign panel pricing environment for this year, but we do not expect much of an impact on glass pricing until 2018.  As to Corning’s specialty glass business (aka Gorilla Glass, we will have more commentary over the next few days.
Display
                4Q – GLW glass volume down slightly q/q – up low teens y/y
                                4Q – glass prices declined less than 3Q
                                2016 – Glass demand up mid-single digits (in line w. expectations)
                                2016 – GLW glass demand up mid-single digits (in line w. industry)
                                2016 – TV screen size up 1.5”+
                                2016 – Supply chain inventory lean
                              1Q – Expect GLW glass volume up mid-teens y/y – in line w. market – Down q/q on lower capacity and shorter quarter
                              1Q – Q/Q price declines moderate (usually the biggest declines yearly)
                                2017 – Retail glass market up mid-single digits
                                2017 – TV units flat to up 1%
                                2017 – TV screen size up 1.5”
                                2017 – Glass demand up 4% to 5%
                                2017 – IT (monitors, NB) down in units, up in screen size – Net flat
                                2017 – Supply chain inventory lean
                                2017 – Overall glass demand up mid-single digits
                                2017 – GLW glass demand up mid-single digits
                                2017 – Panel capacity will increase during the year
                                2017 – Glass price declines less than 2016 (smallest in last 5 yrs)
                                2017 – Price declines (yearly) could be 10% or better
 
Specialty Materials
 
                                4Q – Sales up 22% (above expectations of high teens)
                                4Q – Record GG volume (rapid GG5 adoption)
                                4Q – GG5 premium pricing
                                1Q – GG5 Premium pricing to continue
                                1Q – Expect high teens growth y/y
                                2017 – Secondary supply chain driven by new products not calendar
                                2017 – Expect Specialty Material growth in 2017 but adoption (GG5) will be key
 
1Q TV panel shipments down q/q but edge up y/y, maybe?
 
According to Taiwan based Witsview, TV panel shipments declined 10.7% q/q in 1Q 2017, but were up 0.4% on a y/y basis.  We note that March is typically the ‘seasonal recovery month’, as TV panel shipments are usually weakest in January and February, after the holiday selling season.  We note also that there tends to be differences between data sources on actual panel shipments, with others indicating an 11.5% q/q decline and a 2.9% y/y decline thus far.  We average sources as the data come in.
 
As can be seen from table 1, Samsung Display (pvt) saw a significant drop in TV panel shipments as it began converting LCD fabs to small panel OLED, and reduced the Korean share to 35.7%.  China continues to gain share at 30.5% in 1Q and will likely surpass South Korea later this year or early next, as China adds capacity and Samsung Display reduces its large panel LCD exposure. 
 


[1] rule or domination by three persons, parties, or entities – (under review)

[2] Not to say they don’t care about glass price, but it is less of an onus than when display pricing is weak

Samsung Electronics’ (005930.CH) Galaxy S8 received very significant pre-orders in its home country of South Korea, indicating consumer acceptance of the new device, yet the press has continued to badger Samsung with ‘red tint’, WiFi issues, and breakability stories, many of which seem to be oriented toward creating a controversy similar to the Note 7 recall that sold lots of newspapers and page views.  Thus far, the ‘red tint’ issue, which we highlighted last week, and the Wifi disconnect problem, have been addressed with software patches, and the breakability issue, well, when you make a device out of glass, regardless of the characteristics of the glass, you have to expect it will be a bit more vulnerable than a metal or plastic case.
Yes, even with the extensive testing Samsung did before the phone’s release, problems were found, as they are with any complex mobile device, but all in, consumers make their choices based on what they like, and if a phone looks good and doesn’t explode, that seems to be enough for most users.  The real test will be worldwide acceptance, as South Korea tends to favor the home team, but at the least, the early results have indicated that the new phone is attractive to consumers.  Of course the repair companies are indicating that they are already seeing cracked cases and warn consumers of the risks, yet they certainly are not turning repair customers away, which tempers their altruistic tone a bit.  No burning pockets, smoldering Tahoes, or burning airline seats is a good thing; maybe the tech press should focus on something else, like the potential for conflict between the country with the largest smartphone and display producers and its unbalanced northern neighbor that sits just a few miles from one of the largest display complexes in the world.  Just a thought.