Description

Description

Owens Corning manufactures asphalt roofing shingles, insulation for residential and commercial construction and fiber-glass composites for a variety of industrials applications.

Insulation

• #1 player in the US with ~21% market share and 40% share of the fiberglass insulation segment

• Primary usage in residential and commercial buildings for either thermal or acoustic applications

• In 2013 –

  • US & Canada Residential repair and remodel – 20%

  • International – 16%

  • US & Canada – New Residential Construction – 40%

  • US & Canada – Commercial & Industrial – 24%

• YTD margin performance has been ~5% vs. historical average of mid-teens and peak margins north of 20%

• Industry is operating at 65% capacity utilization with 75% on the operating plants and 90% on the operating lines

• OC has 13 plants in NA. 2 are shut down and 3 are operating below full capacity while 7 plants are operating near full capacity

Composites

• OC manufactures glass reinforcements, technical fabrics and specialty glass mats

• Industry end market exposure: construction 35%, transportation 28%, industrial 14%, consumer 17%, wind 6%

• OC has 37 composite plants in 15 countries – 3 in Brazil, 2 in Russia, 3 in India, 5 in China and 3 in Mexico

• Demand is driven by global GDP growth

• The industry was suffering in previous years from increased supply coming from China that outpaced demand

  • In 2014 industry capacity utilization crossed the 90% mark and industry pricing went up for the first time in several years

Roofing

• #2 manufacturer of asphalt roof shingles in NA

  • Market share: GAF 40%, OC 22%, CertainTeed 22%, Tamko 10%, Other 6%

• Market has consolidated over the years - In the 70s 90% of the market was controlled by 16 companies, 13 in the 80s and 10 in the 90s. Today there are four key players

• Margins have varied between low digit to 30% in the last decade or so

• Asphalt shingles are ~70% of the US market where metal is second with 14% share

  • Asphalt is the main raw material and its costs comprise 35-50% of the price of a shingle

  • Asphalt is a byproduct of refining crude oil (bottom of the barrel). Historically, asphalt prices correlated with crude prices with an R² of 60-70% with several months lag.

• 75% of demand is from re-roof demand

  • Roofs last ~18 years (OC now says 19.5)

  • In the last 20 years industry demand was ~140msqft (2014 will be 107msqf)

• 25% of demand is new construction and storm demand

Thesis

Owens Corning (OC) has been a serial underperformer in the past several years. As the US housing industry has been recovering from a low base, investors have been trying to play this residential theme by purchasing OC and the company has consistently failed to deliver. The Insulation segment is tied to housing starts and has significant upside if the US market returns to a normalized 1.5m starts/year run-rate. Furthermore, the asphalt shingle segment is exposed to residential repair and remodel which is also expected to benefit from a housing recovery. Initially, the composites segment was problematic as it was impacted by cheap Chinese competition and the effects of the slowdown in Asia and Europe. However, in recent years and mainly in 2014, the roofing segment was a significant underperformer.

My thesis resolves around the roofing segment which is 60% of group EBIT. Insulation should perform with housing starts and the pricing environment is favorable (15% price increase in the channel for Jan. 15). The composite segment is coming off a good year, as the company was able to push pricing for the first time in a long time. Furthermore, competition in this segment is less intense than it was historically and industry capacity utilization is high.

2007-2009

I believe that 2015 is shaping up to be very similar to 2009. Therefore, it is important to understand what happened during that period. The asphalt shingle industry has gone through a major transformation in recent years. The merger between GAF and ELK in Feb of 2007 increased the industry concentration to where the top 4 manufacturers control 90% of industry volumes. For comparison, in the 70s – 90% of the market was controlled by 16 companies, in the 80’s it was 13 and in the 90’s it was 10. Asphalt shingle manufacturers were making LSD margins prior to the merger. However, in 2008, companies decided to significantly increase price due to the spike in crude/asphalt prices and the new industry structure. The significant price increases pushed in 2008 resulted an EBIT margin of ~10% for OC that year compared to 2% the prior year.

2009 was one of the best year for the asphalt shingle industry. OC’s margins were up from 10% in 2008 to 28% in 2009. This was a result of stable shingle pricing and a sharp decline in asphalt prices. Further, the economic collapse in 2008 resulted in a sharp decline in oil prices . Oil that averaged ~$120/barrel in Q2 and Q3 of 2008 fell to ~$60/barrel in Q4 2008 ,~$45/barrel in Q1 2009 and averaged ~$60 a barrel in 2009. The collapse in oil prices caused asphalt prices to decline as well. The asphalt PPI was down ~17% in 2009 while roofing flux (asphalt used for roofing) was down ~40%.

While asphalt prices were falling significantly, OC, as the only public manufacturer, was managing expectations during its conference calls in late 2008 and early 2009 in order to maintain its pricing –

Q4 2008 call – “It's worth noting that asphalt prices have not declined at the same dramatic rate as crude oil. With the impact of stimulus on highway paving and infrastructure spending and with the spring storm season approaching, the balance between supply and demand for asphalt could continue to remain tight.”

Q1 2009 Call – “…We have taken significant actions since 2007 to improve the profitability of our business. We have achieved improvements in our production processes, including energy efficiencies, and reduction in the raw material cost of our shingle design. We have also reduced overall manufacturing fixed costs. In addition, we have launched new product lines, improved our mix, and grown our accessories business.”

“…We also saw an upside from the consolidation of the industry which could lead to a more profitable market and elevate our overall performance. The results that we are seeing today are the virtuous combination of good execution by our team and an industry that is responding well to a challenging market”

“…The concern was kind of heightened in the industry or at least inside Owens Corning that a growth in government spending could put additional pressure on asphalt demand during the summer, and that we could see a sizable run-up in asphalt cost heading into the summer.”

Q2 2009 Call –

“We did start to see asphalt costs going up in the second quarter, although that didn't really come through our books in the second quarter. I think, we're starting to see that more in our cost structure late second and we'll see that coming into the third quarter”

“And then I think predictably, we begun to see asphalt cost begin to creep up through the second quarter and that will start coming through our results here late in the second quarter in what we produced. But we'll start to see those increased asphalt costs now coming through our results in the third quarter and probably trail into the fourth quarter depending on how long the winter – the summer driving season and the summer paving seasons and the other things that drive seasonality would be.”

“…We are seeing, asphalt coming through the business at higher levels than what we saw in the earlier part of this year”

During the Q2 2009 conference call, management kept managing expectations to keep pricing firm (see above) but was also forced to explain how margins increased to 33.6% from 21.7% in the previous quarter and from 7.8% in Q2 2008. They no longer could explain this increase due to internal initiatives like they did in Q1. This resulted in a very candid disclosure which basically summarizes my current thesis -

“So, we had some benefit from continued price stability and a relatively tame asphalt environment combined with some volume build from the first quarter to the second.”

“it's hard to imagine that it's only a year ago that oil prices were peaking around $140 a barrel and we were in the environment that we were in last summer. That was really the period of time where I think as a Roofing manufacturer we felt we were really staring down the barrel of a very, very difficult and very difficult market dynamic for us which was prices had not been strong in the industry through 2007 and the early part of 2008, and we were facing dramatic run-ups in our raw material cost with asphalt being a key derivative of crude oil. There was a fair amount of pricing activity through the second and third quarter of last year to recover the run-up in asphalt cost. The real run-up in asphalt costs lags for us because you have to look at when oil goes up and then when the asphalt comes out of the refinery and then we'll bring it into our operations and then when it goes from our operations into our shingles and then we sell them into the market. So, we saw very elevated levels of asphalt cost coming through our business all the way really through the end of last year. We started to get a little bit of relief in the earlier part of this year probably related to oil prices falling in the later half of last year, as well as the seasonality typically our asphalt is a little bit less expensive in the winter than it is in the summer.

So, it's kind of a long winded way of saying that oil is not way up at the elevated levels we saw last summer, which is one of the reasons why we have seen margins widen out based on a stable pricing environment.

As the company stated above The key for the significant margin improvement was stable pricing and dropping crude/asphalt pricing.

The Opportunity

Pricing

2014 was a very difficult year for the asphalt shingle industry. In fact, the past several years have been very tough from a demand perspective. Since 1996 – the industry has averaged 136msqft of shingles and in 2014 the industry will produce 107msqft (the lowest level in that period). There are many reasons for the low demand environment including no significant storms, the effects of pull forward of demand due to previous storm activity, lower compensation from insurance companies and tougher lending standards. Further, 2013 was also a low demand year with 111msqft and 2014 forecasts at the end of 2013 called for 2014 demand of ~120msqft (still a low level of demand compared to historical standards). Instead of the industry being up 8% it ended up being down 4%. This led to an inventory build by manufacturers and distributors and to significant discounting to move inventory. The discounting was initiated by Tamko and was followed by all producers causing significant margin pressure across the board. Current demand expectations for 2015 are low, thus inventory risks for 2015 are low as well. Also, for the investment thesis to play out we don’t need to see any demand improvement.

(96 – 144, 97 – 137, 98 – 144)

The pricing pressure in 2014 has led industry participants to change the messaging and try to stop the discounting ahead of 2015. To clarify, in the winter months demand for asphalt shingles is lower and asphalt prices drop due to less paving demand. This allows manufacturers to sell product at a discount to distributors who stockpile ahead of the busy spring selling season. This proposition is worthwhile to distributors only if the pricing environment is stable/increasing otherwise they are stuck with high priced inventory. This year OC has communicated to investors, customers and competitors that they will not be discounting in the winter months –

Q3 2014 Call – “We would love to see an environment where we can avoid the incentives in the first quarter and continue to price our products where they are today at least early in the year”

BECN – the only public shingle distributor also touched on this in their FQ4 Call –

“Manufacturers also had a very difficult year as you're close to the industry, you know, they weren't able to get any price for the most part. If anything, prices have been going down from them, as evidenced by our COGS, to some degree. I'm sure they're very interested in trying to stabilize the market and gain some price. It's very difficult to say what the winter buy is going to be. You would think that it wouldn't be as deep as last year, but that's just an opinion I have”.

Furthermore, in a discussion I had with a Tamko regional manager on 12/18/2014 he stated that “For the first time in 12 years I’ve been in the business I haven’t seen any discounting this winter so far”.

There is no doubt 2014 was a shock year for the industry and it resulted in renewed discipline. Another example of the shock the industry experienced is that GAF laid off 100 people in November/December.

Shingle manufacturers experienced significant pressure in 2007 and 2008 due to the sharp rise in asphalt prices, and eventually increased prices in 2008. This experienced helped foster the discipline they demonstrated by keeping prices stable in 2009. I believe that the shock the industry faced in 2014 will result in manufacturers showing discipline in 2015. Therefore, pricing in 2015 should be at least stable (if not up). Thus far we’re seeing evidence of that discipline.

Asphalt

The second part of the thesis relates to falling asphalt prices. In fact, we’re starting to see a significant drop in asphalt roofing flux prices in certain regions (Texas, Gulf Coast, Tennessee, and the Mid Con) somewhere between 5-14% in December. These are encouraging signs as December prices are usually stable or trending higher and I am expecting further significant drops in other regions.

Just like in 2008/2009 OC is trying to manage expectations of its customers and competitors regarding the potential benefits of lower crude/asphalt –

OC Q3 Call – “Right now, asphalt costs are above where they we're last year. Even though we've seen oil come down, we expect to maybe get a little bit of relief from the inflation we're seeing. And that might be something that would help us avoid further margin deterioration as we head later into the year with lower volumes. But we do not expect to have a really inexpensive asphalt environment in the first quarter”.

OC’s message was well heard among distributors and BECN is also trying to manage pricing expectations (as mentioned before – distributors lose money in a deflationary environment) -

BECN Q4 Call – “Asphalt pricing has been somewhat in a slight upward pattern even in the June timeframe when oil dropped. The latest asphalt bulletin that I have in November shows at close to $600 a ton... even with oil dropping significantly. Now, we'll see what will happen in the next few months if that will tail down. There's been a lot of discussion about more decoupling now, asphalt versus a barrel of oil than in the past, especially with some of the low light crude that they're pulling out of the shale fields, but I'm certainly not an asphalt expert. There's no doubt that the graph I'm looking at shows asphalt pricing as moving up in the last year”

Even though the pattern of managing expectations that we are seeing now is very similar to what we saw in 2008-2009 – I visited the decoupling of asphalt / oil prices arguments -

1. Usage of more light crude vs. heavy crude results in less asphalt supply

• It is impossible to find relevant data to refute this claim but from anecdotal conversations with energy analysts it seems that the light crudes coming from the Shales have been mainly replacing medium crudes and are being blended with heavy Canadian crudes. In fact, the Flanagan South line from Chicago to Cushing that just started operating is expected to significantly increase the flow of heavy crudes into the US. Heavy crude refiners such as Lyondell are expecting an incremental 400k b/d of heavy Canadian crudes to find their way into the US next year.

2. Installation of cockers by refineries reduced the supply of asphalt

• Again, anecdotally there have been more cockers installed but from conversations with asphalt experts from Poten Partners and ASA it seems that the installation hasn’t had a significant effect on industry supply.

However, there has been a significant decrease in the demand of asphalt in the past decade or so. The majority of the decline as seen in the chart below relates to a demand drop. In fact, roofing flux demand (25% of asphalt demand) is down ~30% in this period while road and construction work is also significantly down (see chart below).

Supply has definitely declined but it is a function of weak demand more than anything else. Unless we see a long term highway bill and significant demand for asphalt there is ample supply available and prices should decline with crude. In fact, with asphalt prices where they are refiners are making more money selling asphalt than they are selling gasoline (per Poten Partners). We also started to see significant declines in paving and roofing flux prices in California, Texas, Tennessee, Midcon and other barge pricing – indicating that prices are not stable and are not decoupling from crude. Moreover, recent conversation (12/23) with Don Wessel from Poten Partners (the author of Asphalt Weekly) reiterated my view – Don is expecting prices to significantly fall next year and he said “2015 is starting to shape up like 2009” and sees ample supply of asphalt. He also stated that most of the large manufacturers blend paving asphalt with their roofing flux thus allowing for a greater mix of supply of asphalt.

Valuation

In 2014 OC will do ~1.75bn in roofing sales and ~235m of EBIT (13.4% margin). According to my estimates COGS are 76.6% of sales (SG&A is 10% - similar to the corporate margin). Industry estimates for the cost of asphalt range from 35% of sales (~45% of COGS) to 50% of sales (65% of COGS). To stay on the conservative side I am assuming that asphalt costs are 35% of sales equaling $613m in 2014.

If we assume that asphalt costs are reduced by 10% and they are only applied to Q2-Q4 based on the companies’ commentary (even though I believe that Q1 will see some impact as well) –

613m * 70% (Q2 – Q4 contribution of sales) = $420m

420m * 10% (decline in asphalt prices) = $42m

42m * 70% (tax effected) = $29.4m

$29.4m/118m (Diluted shares outstanding) = 25c

With crude down ~50% since the summer a 20% drop in asphalt price seems a reasonable assumption. Consensus estimates for OC are not assuming any tailwind from crude. In fact, all the sellside analysts I spoke with talked down the possibility for a potential tailwind citing OC’s decoupling thesis. Consensus is estimating $2.30 EPS for 2015  If we add 50c due to the asphalt tailwind at 18x P/E (all comps trading at 20x. OC currently trading at 16x) my PT is $50

Catalyst

1. Further reduction in asphalt pricing

2. Earnings - progresssion of margin expansion

3. More information about stable pricing