|Shares Out. (in M):||1,400||P/E||0||0|
|Market Cap (in $M):||32,500||P/FCF||0||0|
|Net Debt (in $M):||0||EBIT||0||0|
|Borrow Cost:||Available 0-15% cost|
Market Cap: $32.5Bn
Share Price: $23.17, 16x $1.47 2014 “core” EPS
Corning (GLW) is a misunderstood, over-valued business which has used misleading accounting (“core earnings”) and a “too good to be true” acquisition to obfuscate the massive deterioration in corporate earnings power / fundamentals in its largest business. We see >40% downside from these levels over the next 12 months as investors realize they are capitalizing the value of hedges and paying >20x underlying earnings (well north of replacement value) for a company that primarily sells commodity products into an oversupplied market. Our target price is $13, based on 12x (historical multiple) underlying earnings power of ~$1.00 (consensus at $1.50+ in 2015), + the value of hedges ($1.00). Sentiment is very bullish (with multiple recent upgrades), and given the large capitalization and minimal short interest, we believe GLW is a compelling short.
Corning operates in 5 reportable segments:
Display Technologies: manufactures liquid crystal display (“LCD”) glass for flat panel displays, used primarily in TV’s, tablets and smartphones
Optical Communications: manufactures carrier network and enterprise network components for the telecommunications industry
Environmental Technologies: manufactures ceramic substrates and filters for automotive and diesel applications
Specialty Materials: manufactures products that provide more than 150 material formulations for glass, glass ceramics and fluoride crystals to meet demand for unique customer needs
Life Sciences: manufactures glass and plastic labware, equipment, media and reagents to provide workflow solutions for scientific applications
In addition, Corning owns 50% of Dow Corning Corporation, a JV with DOW. Corning recognizes its share as Equity Income in the Income Statement.
We will focus here on the Display Technologies segment, which represents 40% of sales and 70%+ of net income, and is grossly inflated by the Company’s “core earnings” metric. We don’t have a differentiated view on the other businesses, with the exception of Specialty Materials (mostly Gorilla Glass), where we see downside to street estimates, but that has a limited impact on earnings power.
Corning reports “core earnings” which is used by the sell-side / most investors instead of GAAP and includes standard adjustments for one-time charges and other items. In addition, because the Display business transacts 100% of revenues in yen (due to historical market convention), “core earnings” uses a yen peg instead of the actual quarterly average rate to translate earnings (currently the peg is at 93). This is designed to match cash flows with earnings, as they generally have hedged most or all of their revenues at the peg rate (100% of 15 hedged at 97, 80% of 16 at 99, and 70% of 17 at 99). As the yen has depreciated dramatically, “core earnings” reporting:
Has inflated the true earnings power of the business ex hedges
Has masked the deterioration of the fundamentals of the business
Consensus EPS for 2014 is $1.47, with $4.3bn of Display revenues at a peg rate of 93 yen / USD. The company has stated in the past a 1 yen move is a $6.5mm impact to net income per quarter with the yen > 100. From 93 to 118, the EPS impact is 25 x 6.5 x 4 = $650mm of net income, on 1.4bn shares = 46c of EPS. Calculated another way, marking the $4.3bn of revenues from 93 to 118 is $900mm of revenue reduction, which flows 100% to PBT as all Display costs are in USD and KRW. At a 30% tax rate, the impact is roughly the same as management guidance, $650mm of net income.
The incremental synergies from Samsung Corning Precision (more on that below), growth in other segments net of Specialty (largely due to acquisitions), and the share buyback accretion are offset by the continued decline in profitability in the Display business (in addition to resetting the yen), netting us to $1.00 of underlying earnings power going forward.
Deterioration in Fundamentals:
In addition to inflating earnings, the devaluation of the yen has masked the true decline in the underlying business fundamentals. Corning provides quarterly sequential changes in ASP’s (e.g., “decline low single digits”) and changes in LCD glass volumes (e.g., “flat to down slightly”). In prior years, Glass ASP’s have historically declined mid-single digits quarterly (on a sequential basis), in excess of costs, given the structural oversupply in the market (more on that later). Throughout 2014, as the yen imploded, Corning’s ASP declines have moderated, implying a stabilization of the business. This has been a focal point of the bull thesis and is often cited by management on recent calls and conferences.
However, these ASP’s are given in local currency (yen), while all of their customers, and the entire TV supply chain, transacts in dollars. In addition, all of Corning’s costs are in USD and KRW, making yen local currency declines irrelevant. ASP declines in USD have not stabilized and continue to deteriorate. Once the yen stabilizes, we believe ASP declines in local currency will return to historical levels. Ironically, the devaluation of the yen has boosted the stock as management pre-released “better than expected ASP’s” before the Barclay’s conference in December as the yen hit new lows.
Display Competitive Landscape:
Corning has 30%+ net income margins (adjusted for the yen), while its customers (panel makers), TV / smartphone manufacturers, and retailers of TV’s have single digits when times are good and negative earnings in the down cycle (with some obvious exceptions). Corning (45% market share) has 2 main competitors, Asahi Glass (25% share) and Nippon Electric Glass (25%), both based in Japan. Corning, Asahi, and NEG manufacture and sell glass substrates to panel makers (Innolux, AUO, Samsung Display, LG Display), who sell displays to TV / smartphone / PC manufacturers (Samsung Electronics, Apple, Sony, Lenovo) who ultimately sell to end consumers through retailers / online.
The glass market has been in structural oversupply for the last few years, due to limited demand growth and continued supply growth. Given the relative screen sizes, the explosive growth in smartphones is less important than TV units and screen size growth (over 50% of glass demand is from TV’s), which has been flat to up mid-single digits. Importantly, as glass gets thinner and thinner, more glass is produced for a given capacity, similar to a shrink for semiconductors (costs decline 8-10% a year generally).
The glass market is characterized by:
Limited barriers to entry: LG Chem and China Inc are both new entrants in the Glass market. They currently make less advanced glass (past generations, thicker glass) but are catching up. Producing glass really just requires capital and limited expertise and time to ramp up. So there is limited incentive to pull back on capacity to rationalize the market, as improved economics will only incent additional competition / supply
Capacity cannot be converted to anything else: capacity must be mothballed (25% of Corning’s capacity is currently mothballed) and cannot be put to another productive use (e.g., transitioning DRAM to NAND), again limiting incentive to reduce capacity
Much higher margins than anywhere else in the supply chain: panel makers have operating margins at 5% in up cycles vs. Corning at 30%+, so they are the target of everyone else in the chain
Limited ability to differentiate product: cannot compete on product specs, only on price or dedicated supply partnerships (e.g., building a new plant next to a large panel customer)
Supply increases structurally over time: as glass gets thinner, supply is effectively increased (similar to a shrink in semis), so more units from the same capacity (8-10% annually)
Asahi and Nippon are not able to be fully rational: both want to fill their utilization given uncertain financial positions. Panel makers won’t let either fail because then Corning would have too much power in the market, and there would be new entrants in this case
Corning announced at the Barclays conference in December that “The company is currently operating all of its on-line capacity to meet its contractual demand and expects to maintain its current share of the LCD glass market.” Corning has 25% of its capacity mothballed, which can be restarted quickly (with some additional capital required). Q4 (current quarter) is the seasonally the strongest quarter. Averaged over the year, we estimate that Corning has been 85% to 90% utilized (excluding the mothballed capacity), with NEG and Asahi running below that. We think the entire market is currently ~15% oversupplied. Taking Corning’s total capacity into account, we believe the market is running at ~75% utilization and will be at nearly 70% after NEG’s China plant comes online in 2H15 (see below).
Demand for glass is segmented between TV’s (60%), desktop monitors (15%), smartphones (15%), and tablets (10%). At best, we think demand grows mid to high singles, in line with natural “shrink” and consensus estimates, which suggests pricing declines should be in line with this year. We forecast desktop monitors flat, tablets down 10%, and smartphones up 20%. To hit the mid to high singles, TV’s would have to grow 10%+ (mid singles units and mid singles area growth). We think there is real risk to TV demand, although we do not incorporate that into our numbers. 2014 was a very strong year for TV’s, given the World Cup driving the replacement cycle. Despite this, pricing declined 15%+ in yen (nearly 25% in USD).
So in this scenario, supply and demand grow in line and the market remains ~15% oversupplied. However, NEG announced in early 2014 plans for a new plant in Xiamen, China next to AUO, which will add 5% to total industry glass capacity once completed. They broke ground on construction in November http://www.yichiglass.com/Candle%20holder/Candleholder_1769.htm http://global.ofweek.com/news/China-s-largest-LCD-substrate-glass-factory-breaks-ground-21384. The plant is scheduled to begin operations in 2H15, with a second phase coming online in 1H16. Construction and equipment will cost roughly 70bn yen ($600mm), and they expect sales of $350mm annually once fully operational. We believe there is additional downside to our estimates, as this new plant combined with weakening TV demand could drive utilization well below current rates and lead to sharp declines in pricing (has been high single digits or even low double digits in single quarters in the past during oversupply situations).
This NEG plant implies the replacement value for Corning’s capacity is roughly $6.5Bn (including mothballed capacity, Corning’s capacity is ~11x this new plant, and the mothballed capacity and some of Corning’s other facilities are older and would require additional capital to be modernized). $6.5Bn replacement value for 70% of earnings of a $32Bn company seems a little off to us.
Corning Display sells a commodity product (glass) into an oversupplied market with limited demand drivers and a concentrated set of customers (panel makers), leading to intense pricing competition. It is not a surprise then that Samsung was looking to exit this market.
Samsung Corning Precision:
On 10/22/13, Corning announced that it had acquired Samsung Display’s 43% share and the remaining minority interests in its JV, Samsung Corning Precision (SCP). In return, Corning provided Samsung with $1.9bn of convertible preferred securities, and Samsung invested a further $400mm in additional convertible preferreds. In total, Corning paid $2.2bn to acquire $350mm of net income ($100mm of synergies offset the $100mm of incremental interest expense from the Samsung convertible preferreds, so $350mm is the underlying net income), representing 6.3x earnings. The synergies from this deal, combined with the change in accounting (consolidated now), has made year over year comparisons difficult and obfuscated the deterioration of the fundamentals in the business.
In addition to the upfront payments, the parties entered into a contingent consideration arrangement, which is dependent on the sales and net income performance of SCP through 2017. In Q3-14, Corning booked a $77mm gain on this contingent consideration, reflecting reduced expectations for future earnings from SCP. In addition, 6 months ago management told us they were 60% hedged for 2017; 3 months ago they told us 65%; and recently they have been telling some investors 70%. The CFO bragged at the Barclays conference in December “we are effectively hedged for the next three years at 99”. As the numerator (hedges) is not changing, perhaps the denominator (expected sales in yen) is falling off a cliff?
Corning’s management team is well regarded by most sell side analysts and investors. The SCP deal seemed too good to be true, they are buying back stock, they are growing their non-Display businesses, and core earnings appear to be healthy and growing. After several tough years, things seem to have stabilized.
We do not share this view. We think management is deceptive in the way they characterize their results, their business, and the industry landscape. For instance, for Q3-14, Corning reported adjusted gross margin of 45%, despite guiding to 46% on the Q2-14 call (“We expect gross margin to be 46% driven by Display and Gorilla Glass”). On the Q3-14 call, the following exchange occurred, incredibly:
<Q - Steven B. Fox>: … And then secondly, I was just curious around gross margins, seemed like it came in a little bit weaker than the 46% peg for the quarter but you're still targeting 46% for this quarter. Can you walk through some of the puts and takes with that, Q3 and Q4? Thanks.
<A - James B. Flaws>: On the gross margin, I don't think we came in weaker than our own expectations.
<A - Wendell P. Weeks>: Yeah, right.
<A - James B. Flaws>: So it may be weaker than what some analysts expected.
In addition, while we are generally huge fans of management teams who repurchase their own shares, repurchasing shares at peak valuation (even when including capitalizing hedges) does not seem very prudent to us, and in fact destroys shareholder value.
Asahi Glass or Nippon Electric Glass consolidating / exiting the Glass market
Stronger than expected TV demand cycle
Accretive acquisitions in other segments
Corning has used misleading accounting and disclosed metrics to obfuscate the significant deterioration in its Display business. After marking its earnings to market, GLW earnings power is $1.00 vs. consensus at $1.58 in 2015. Investors are clearly capitalizing the hedges and, without realizing it, are paying over 20x earnings and well above replacement value for a largely commodity business in an oversupplied market. We see >40% downside from current prices and limited upside from here.
Management has stated they will adjust the yen peg for this year (2015), and the CFO has hinted multiple times publicly the peg will be set at 99, where they are mostly hedged for the next 3 years (15-17). This will reduce consensus estimates and highlight the sensitivity to the yen (not all sell-side estimates currently account for those and the ones that do mostly get it wrong).
We believe once the yen stabilizes, ASP declines will likely revert to mid-single digits from current low-single digits, highlighting the deterioration of the business, the structural imbalance between supply / demand. Weakness in the TV cycle or the NEG China plant opening could further pressure pricing.
Given the expected weakness in tablets this year, we believe the Specialty Materials segment (Gorilla Glass) could materially underperform expectations, which is important for sentiment as it is supposed to be a growth driver going forward.
|Entry||04/22/2015 12:47 PM|
Thanks for the interesting write-up. I have a few questions:
1) What is the best way to track changes in glass pricing?
2) If barriers to entry are so low, why has no one credibly entered the market given the amazing returns these guys have enjoyed?
3) Why has glass pricing deteriorated so aggressively over time within a highly consolidated market?
4) What, if any, impact will the fixed market share agreements that Corning has signed with the vast majority of its display glass customers have on the market structure and/or pricing?
5) Do you think the impact of the JPY is poorly understood by owners of this stock and/or sell-side analysts? If not, what's the catalyst here?
Thank you in advance for your thoughts.
|Entry||04/23/2015 01:21 PM|
1) By tracking changes in display panel pricing. Glass pricing has historically followed, given it is one link further removed in the chain from the consumer. When demand for TV’s, PC’s, tablets, smartphones, etc. gets weak, you see that in panel prices first as TV manufacturers cut orders, and then panel manufacturers cut orders on glass. In April, panel prices started to roll for the first time in a year. There are clear signs that consumer electronics demand is falling, given the strong dollar and weakness in European and emerging market consumers.
2) They have not enjoyed amazing returns – it is a capital intensive business and they overbuilt supply. ROIC’s are not high. It is our view that if somehow the market ever gets tight, there will be new entrants in the future, most likely China Inc. Given the oversupplied markets and deteriorating fundamentals, we would not expect any new entrants any time soon.
3) Glass supply grows at essentially ~10% annually per year excluding any new tank capacity. This is similar to a shrink for most foundry/logic/memory players. To use the DRAM analogy, Samsung has shrunk to 20nm, producing more bits per wafer. At the same time, they boosted their flagship smartphone from 1gb to 2gb to 3gb, providing a faster, better product at a comparable cost to historical given they are producing more bits per wafer, which has high fixed costs. So ASP’s decline in line with costs in a steady state, as the new bits get soaked up, earnings grow over time.
In glass, the equivalent shrink is that glass gets thinner in each successive generation, so more glass square footage can be produced for a given tank. Glass production similarly has high fixed costs, reducing the cost per area of glass, which can be passed on in pricing. However, for glass, density demand does not increase on a per unit basis. More glass is not better for any consumer electronic device – in fact, less glass is better. Average TV size has grown in recent years, but much less than 10% (3-4%), so this helps. But you need unit growth to keep the market in balance. Roughly 25% of the industry tank capacity is currently offline – this market will never get into balance, even if no capacity is built for a decade. And given the high fixed costs, players are incented to move units at pricing down to marginal costs.
This is why Gorilla glass is such a bad business – every year, they come out with a new Gorilla that is twice is strong. Apple says thanks very much, we will take half as much glass per phone.
4) We don’t think these have any effect – pricing is negotiated every quarter for every contract and volumes are not enforceable.
5) We think the order of magnitude is not understood and there is also an element of complacency among those who do understand. There is no way people think this is actually 22x earnings and think that is a good investment. We have heard from sell-side analysts 2 lines of reasoning: 1) the CFO is really smart, he will figure out how to fix the yen issue by 2018 and 2) well no one else seems to care, so why should I?
This will end poorly. The CFO can hedge the yen today for 2018…at 115. There is no magic fix – earnings power has imploded during a really strong TV cycle. We’ll see what happens this year.
Our view on the catalyst is pricing / volumes will be materially worse than expected sometime in the next few quarters when the end market demand rolls through. The story that “the fundamentals are better” will go away, and people will go back and think about what they actually own. There is no valuation support here with real earnings sub $1.00 and continuing to deteriorate. At that point we believe the stock trades into the mid teens, closer to our estimate of fair value.
Perhaps the CEO, CFO and other executives read our write-up…I would recommend people take a look at the wave of insider sales in February.
|Subject||Re: Thoughts on results?|
|Entry||07/09/2015 12:11 PM|
It's clear that inventories are elevated heading into a period of slowing end demand for TV's, PC's, tablets and smartphones. That's generally not a very good combination. An inventory correction is coming. Panel prices are rolling. It's only a matter of time.
This article from Digitimes is only available to subscribers, but you can see the headline here: http://www.digitimes.com/topic/displays/display_panel/
Jul 3, 11:13
Korea panel makers have seen their reserves reach a three-year high in the first half of 2015, and have conservative views for demand in the second half of 2015, accroding to a media report.
The same report says that Samsung and LG revised TV shipments for the year from +20% to +5% for 2015 due to weak demand. Samsung is GLW's largest customer and LG is probably second largest.
AUO reported June sales this week, down 10% yoy, and Q2 is down 3% vs. Q1 (consensus was up low single digits for the QoQ). Large panel shipments were down 13.7% YoY and 7.1% QoQ. GLW guided display glass volumes up low single digits QoQ in Q2. In Q1, AUO reported sales down 10% QoQ and GLW volumes were down low single digits. AUO shipments of large panels were down 8.9% QoQ.
|Subject||Share buyback announcement|
|Entry||07/16/2015 12:06 PM|
GLW announced a $2bn share repurchase program that expires in December 2016: https://www.corning.com/worldwide/en/about-us/news-events/news-releases/2015/07/corning-announces-2-billion-share-repurchase-program.html
In December, GLW announced a $1.5bn repurchase program over the same time period, and repurchased $500mm in Q1. Perhaps they repurchased $1bn in Q2, which would be ~0.5c upside in the quarter, depending on pricing and timing. On street numbers, they should earn over $3bn of core earnings over the next 1.5 years and they will pay out $1bn of dividends. FCF is close to core earnings (given the gains from the hedges are real cash gains as they expire) and therefore they should have $2bn of FCF to repurchase stock, consistent with this announcement, excluding any future M&A (of course they can take on some more debt as well).
We valued the hedges separately at ~$1.5bn in our valuation. And given our view that GLW is materially overvalued here at $19.50, it is value destructive for them to repurchase stock, vs. buying assets priced closer to fair value via M&A or boosting the dividend.
The timing of the announcement is interesting, 2 weeks before earnings...they are clearly trying to prop up the stock and of course use promotional language in the release.
|Subject||supportive points to the short thesis|
|Entry||09/18/2015 02:47 PM|
this has been a good call. i have also been short this stock for a while. something that was not directly addressed in the report is why Samsung sold its stake in its JV to GLW to at such a low valuation. Samsung believed that the business was only worth 6x EPS. in addition to everything written above, I believe there are four other reasons:
1) the shift to thinner glass creates a secular oversupply dynamic for the industry. on GLW's last call, the company projected a bullish tone about how they expected the move to thinner glass to further moderate pricing declines. The fact is that they would actually need to get higher pricing with thinner glass to maintain gross margins. This is because the way they make thinner glass is by effectively increasing throughput. Therefore, the move to thin significantly increases capacity for the industry. Since most costs are fixed, the company will actually need to raise pricing per unit to maintain gross margins which is unlikely to happen.
2) Samsung is developing a competitor to gorilla glass. see this: http://www.gforgames.com/gadgets/samsung-gorilla-glass-alternative-tortoise-48826/
3) The average life of a TV is about 7 years. The strength in 2014 was therefore likely a replacement cycle of the strong year in 2007 for TV shipments and there is already evidence of that replacement cycle fading not only in China but other geographies.
4) OLED TV's do not need LCD glass and may eventually take more share of the TV market.