Good day, ladies and gentlemen, and welcome to the Third Quarter 2012 Owens Corning Earnings Conference Call. My name is Chanel, and I'll be your operator for today. [Operator Instructions] As a reminder, this conference is being recorded for replay purposes. I would now like to turn the conference over to Mr. Thierry Denis, Director of Investor Relations. Please proceed.

Thank you, Chanel, and good morning, everyone. Thank you for taking the time to join us for today's conference call in review of our business results for the third quarter of 2012. Joining us today are Mike Thaman, Owens Corning's Chairman and CEO; and Michael McMurray, Chief Financial Officer. Following our presentation this morning, we will open this 1-hour call to your questions. [Operator Instructions] Earlier this morning, we issued a news release and filed a 10-Q that detailed our results for the quarter. For the purposes of our discussion today, we've prepared presentation slides that summarize our performance and results for the third quarter and 9 months ended September 30, 2012. We will refer to these slides during the call. You could access the slides on our website, owenscorning.com. We have a link on our homepage and a link on the Investors section of our website. This call and the supporting slides will be recorded and available on our website for future reference. Please reference Slide 2 before we begin, where we offer a couple of reminders. First, today's presentation will include forward-looking statements based on our current forecasts and estimates of future events. Second, these statements are subject to risks, uncertainties and other factors that could cause our actual results to differ materially. Please refer to the cautionary statements and the risk factors identified in our SEC filings for a more detailed explanation of the inherent limitations of such forward-looking statements. This presentation and today's prepared remarks contain non-GAAP financial measures. Reconciliations of GAAP to non-GAAP are found within the financial tables of our earnings release. Consistent with our historical practice, we have excluded items that we believe are unrepresentative of our ongoing operations to arrive at adjusted EBIT, our primary measure to look at period-over-period comparison. We exclude significant nonrecurring items, such as the impact of the restructuring actions discussed in our most recent earnings call. We believe that adjusted EBIT is helpful to investors for comparing core results from period-to-period. We adjust our effective tax rate to remove the effect of quarter-to-quarter fluctuations, which have the potential to be significant in arriving at adjusted earnings and adjusted earnings per share. In the third quarter, we have utilized an effective tax rate of 25%, in line with our anticipated annual effective tax rate on adjusted earnings for 2012. For those of you following along with our slide presentation, we will begin on Slide 4. Now opening remarks from our Chairman and CEO, Mike Thaman, who will be followed by CFO, Michael McMurray. Mike will then provide comments on our outlook prior to the Q&A session.

Thanks, Thierry, and good morning, everyone. We appreciate you joining us today to discuss our third quarter 2012 results. Owens Corning revenue in the third quarter was $1.3 billion, down 12% compared with the same period last year. Adjusted EBIT was $81 million, down from $177 million 1 year ago. Before I discuss the key drivers contributing to our business results in the quarter, let me make a few observations about the year-to-date and the overall environment in which our businesses are operating. This has been a challenging year in terms of visibility to demand, whether in the U.S. roofing market or the impact of slower-than-expected growth across many of the global markets for our Composites business. We're disappointed in our third quarter financial results and the need to lower our full year earnings forecast. Despite the change in our near-term outlook, we remain positive with regard to the strength of our businesses. These are market-leading businesses, operating in well-structured industries. We have a strong brand, technological leadership and manufacturing and supply chain strength to serve our valued customers around the globe. As we disclosed earlier this month, we now anticipate full year adjusted EBIT in the range of $280 million to $310 million. This earnings outlook reflects a weaker environment for both our Roofing and Composites businesses this year. I will discuss the specific outlook for each of the businesses in a few minutes. Each of our businesses is making progress against the strategic objectives we've set. Roofing is a great business with a strong market-leading position. This business has delivered great margins in recent years. We believe it is well positioned to sustain that performance. Our Insulation business is benefiting from an improving housing market. We are narrowing losses and progressing towards positive earnings. We are proud that…

Thanks, Mike, and good morning, everyone. I'm excited to be back with all of you, now as the CFO of Owens Corning. Before I begin, I wanted to acknowledge the disappointment of our investor and analyst communities in our recent outlook revision and assure you that we share in your disappointment. The management team understands the importance of being a reliable resource to our investors, who does not take the need to revise our guidance lightly. As Mike mentioned earlier, our third quarter results were impacted by weak near-term market conditions in roofing and composites. As previously announced, we have revised our full year 2012 adjusted EBIT expectation to a range of $280 million to $310 million, with the primary uncertainty attributed to roofing volumes. Now let's start on Slide 5, which summarizes our key financial data for the quarter. You will find more detailed financial information in the tables of today's news release and the Form 10-Q. Today, we've reported third quarter 2012 consolidated net sales of $1.3 billion, down 12% compared to the same period 1 year ago. Our Insulation business grew by 5% on improved demand. Net sales in our Roofing business were down 27% on lower sales volumes, and net sales in our Composites business were down 7%, primarily due to foreign currency translation. In a moment, I'll review our reconciliation of items to get to adjusted EBIT, our primary measure to look at period-to-period comparisons. Adjusted EBIT for the third quarter of 2012 was $81 million compared to $177 million in the third quarter of 2011. Adjusted earnings for the third quarter of 2012 were $39 million or $0.33 per diluted share compared to $110 million or $0.90 per diluted share in 2011. Depreciation and amortization expense for the quarter was $89 million, including $14…

Thank you, Michael. We remain positive on the strength of our businesses and our ability to respond to their markets. Our Roofing business has delivered great margins in recent years, and we believe it is well positioned to sustain that performance. Our Insulation business is benefiting from the improving housing market and achieved profitability in the quarter for the first time in 4 years. Our Composites business has taken decisive actions that will position it for a stronger 2013. I appreciate your continued interest in Owens Corning. And we'll turn the call over to Thierry to lead us in the question-and-answer session. Thierry?

Thank you, Mike. Chanel, we're now ready to begin the Q&A session.

[Operator Instructions] Our first question comes from the line of Stephen Kim from Barclays.

I wanted to ask you a question first regarding Composites. Thierry, Mike, I think you made -- Michael rather, I think you made a reference to additional rationalization. Previously, you had given a guidance, I think, for $70 million, and in your release today, you highlighted that you were substantially complete with that. So I just wanted to get a little granularity on this comment about additional rationalization, how much more. And what do you mean by substantially complete?

Okay, Stephen. This is Mike. Let me make a few comments, then I'll turn it over to Michael. I think we're talking about 2 different things on the call today, and I want to make sure that we clarify those. We have our overall asset strategy, which deals more with our fixed assets regarding the facilities in Europe that we're shutting down, the expanded melter in Russia, the new melter in China -- I mean the new melter in Mexico and also, the new plant in China, which is now loaded. I think we feel very good about the fact that we made a lot of progress in the third quarter on that agenda. We got our really very important significant plant in Amarillo, Texas went through a rebuild, which also hurt the quarter, and we're really looking now through the rearview mirror at a lot of the assets actions that we needed to take and feel like the next couple of years are focused on really matching the right products to the right plants with the right customers at the right margins and optimizing our asset mix, optimizing our cost position and really trying to drive performance of the business. We also talked about inventory curtailments and the need to further curtail inventories in the third quarter based on demand weakness. I think Michael's comments were more specific to that issue, and I'll have him address that. Michael?

Thanks, Mike. Yes, Stephen, as you know, as we talked about in our second quarter call, we entered the year with Composites inventory being a little bit ahead of where we want it to be, and we've had an active program, both in the second and third quarter to curtail inventory and take it down to go where we wanted to get by the end of the year. I actually think we did say in the second quarter call that goal was on the order of $70 million. Because of the weakness that we have experienced in the third quarter, a large part of it being in our Roofing business, it's forced us to curtail or take additional curtailments in the fourth quarter. I would tell you, as we're progressing through the fourth quarter, the target that we've set out is that we're largely on track to meet that goal.

Okay. That is very helpful, although I didn't hear a number, but okay. Let me switch, if I could, to Roofing. Roofing is -- I think as you -- I think your comments are appreciated that perhaps you didn't catch the trends in volume as quickly as maybe you should have, and I appreciate your candor there. I do think that your main issue with Roofing though is margins, and particularly as it relates to whether or not discipline can hold in the industry in the winter months to avoid what we had earlier this year, which was a lot of volume-related discounting. I know that you have talked about the fact that you have a commitment to a discipline in that regard. But also I believe, in the past, you mentioned that you can't control what others in the industry will likely do and that historically, you have needed to match what some competitors do. I guess, I was looking to see if you had a stronger comment in that regard that can give investors a little bit more comfort that, even in a difficult winter environment, we're not going to see a repeat of what happened in the beginning of this year. Is there anything more that you could add there in terms of color?

Sure, Stephen, I'd be happy to address that, and in fact, let me start by agreeing with your opening comment, which is there's a bit of a turmoil in our guidance related to the business based on our lack of ability to effectively forecast demand. I addressed that in my comments. We're looking at that and taking that very seriously. In fact, when you step back from the year, the more important issue, on an ongoing basis, is less related to our ability to forecast quarterly demand numbers, much more related to what is the long-term reasonable margin expectation for the business. This year, in the first quarter, I think you characterized it the way we have, which is we sold an awful lot of shingles at deep discounts, and in doing that, we gave our distribution partners an incentive to want to buy a lot of shingles and try to carry low-cost inventory deep into the year. Now winter buys have always made some sense in the Roofing business. They've been around for a while, and I'd characterize it as the winter buys they've tried to accomplish 3 objectives. The first objective was to help the manufacturers level-load our production a bit. It's a seasonal business. There's more demand on the roof in the second and third quarter. It's nice to be able to give your employees year-round employment. It's nice to have -- to produce some shingles in the fourth quarter and the first quarter of the year. And so I think the manufacturers have some incentive to level-load their facilities, although we've been very clear, it's not an operating leverage or a fixed cost business, it's a materials conversions business, so you can go too far in that thinking about trying to level-load your facilities. And that…

Our next question comes from the line of Mike Wood from Macquarie Capital.

Can you confirm whether or not -- the intercompany sales decline, was that primarily related to Composites sold into your Roofing? And was there any destocking that happened internally given the sudden decline in roofing demand?

I'll let Michael correct me if I get any of this wrong, but I'll take that one, Mike. Most of our intercompany sales are, in fact, sales from our Composites division to our Roofing division. So by and large, what you see in the intercompany line tends to be transfers from our Composites business, which carries production responsibility for the manufacturing of the glass. We transfer at a market price, but Composites also carries the asset, so Composites has clear alignment in terms of return on capital goal, which is a very important goal for us. So what you would see in that is intercompany decline related to that transfer. On a consolidated basis though, the inventories -- if we saw destocking or not, it wouldn't show up in the inventory number because, in fact, we pool the inventories on a consolidated basis. So clearly, the Roofing business is going to want to carry the amount of inventory that's appropriate to its production outlook to the extent they don't take mat or glass from our Composites business, that's going to end up on the Composites balance sheet. Composites has its balance sheet goals so they're going to take curtailment in order to try to reduce those inventory levels on their side of the balance sheet. So that's how that would work internally. Michael, did I get anything wrong there?

Nice job, Mike.

And on the Insulation segment, is there anything unusual impacting the incremental margin in that business? Or another -- said another way, should we expect the incremental margins to slow from where they were in this past quarter?

Yes, I mean I would say we continue to feel comfortable with this 50% operating leverage guidance. So in the first quarter, I think we were subjected to a bit of criticism because we came in below that goal, and at that time, we explained timing of demand, timing of shipments and timing of production can change that result from quarter-to-quarter. I think if you look at what we've said in the second quarter, we were well above that goal and tried to take little or no credit for that, saying that timing of demand, timing of production and timing of shipments can affect that. I think that's kind of where we are in the third quarter. We're in the high-60s year-to-date. That's still going to be affected not just by our U.S. new construction business, but also the performance of some of the non-new construction businesses, like our business in Canada, our engineered business, our business in Asia. I think we would probably be reluctant to get more aggressive about that operating leverage goal until we got to a point in time where we felt like we were getting significant and repeatable margin-enhancing pricing. That certainly is something we believe will happen through this cycle. It's typically happened through the Insulation cycle, and today, obviously prices are way below average prices for the last couple of decades, certainly well below the kinds of prices we saw in the mid-2000s. So I think you could see some movement in the rate of operating leverage once it becomes a price and leverage story. Today, it is still predominantly a leverage story, although Michael did report that sequentially we felt positively about what we saw in pricing from the second to third quarter in virtually all of the businesses inside Insulation.

Our next question comes from the line of Garik Shmois of Longbow Research.

First question just on Insulation revenue growth, up mid-single digits. But when you lag 2Q housing demands into the third quarter, I think we were expecting a little bit more on the top line. I was just wondering, I know there's some mix involved, but maybe if you could provide a little bit more color on what was happening on remodeling and maybe in the international markets and how much those maybe have contributed to an offset with the new housing growth in Insulation.

Sure, Garik. Happy to do that. Thanks for the question. I think if you look at, what is it, Slide 9 in our investor presentation, we put up that little pie chart in the lower left-hand corner, and we do that on a quarterly basis so you can kind of see the evolution of the individual segments. I know we do have some analysts and investors who kind of back-calculate what that means in terms of revenue by quarter and look at the changes in revenue. So we disclose that. We don't disclose the actual revenue numbers, and I'd be clear about that. These are pretty broad estimates, so while we think directionally, we're giving you pretty good information here, if we were to try to actually quantify revenue for each one of these segments, it's pretty hard for us to do. We estimate it based on channels and product line. I think what you'd see is that actually the new residential construction portion of our business grew pretty well in line with what housing starts did. You have to remember that in that pie chart, we report U.S. and Canada. Canada today is pretty significant because Canada hasn't really gone through a housing cycle the way the U.S. has, so it's a much larger percentage of that pie chart. Canada has been relatively flat as we've seen the U.S. recovery, so it's slowing down the rate of improvement in the new residential construction slice because it's flat and the U.S. is growing 20-plus percent. I think what you'd see is the other segments are relatively flat. Repair and remodel has not been terribly robust. I think that's probably a byproduct that we had a mild winter last winter, so we didn't have a lot of people coming out of the winter saying, "I need to do something to insulate my home." We haven't yet seen cold weather this year. So I don't think you have people coming home and feeling like their house is not comfortable and wanting to try to take an action to improve the comfort of their home. So through time, we believe the repair and remodel market should be a single-digit growth market. I don't think it's necessarily quite, at this time, growing at those levels. But as we work through the numbers and look at the different segments, we certainly are confident that we're seeing the U.S. new residential construction translate into volume for that segment of our business in the way we're reporting revenue.

Okay. And then I guess, just switching to Composites. The question is, in 2012, you had some price weakness, obviously, in the segment, particularly as you entered into the annual contracts with excess inventory at the end of last year. You're working on inventory right now, but if you could provide maybe your thoughts on how you're going to be managing these pricing negotiations for 2013 and the way your outlook would be for pricing next year in Composites.

Yes. I think it's a very good question, and that's a real-time question. So that's obviously something we're working right now because that is in a very important fourth quarter negotiation for us. I've talked in the past that one of the key considerations that impacts Composites pricing is kind of rate of change of the market. When we see markets that are growing turn negative, when we see markets that are growing quickly suddenly go to a no-growth state, because of the fixed cost nature of Composites manufacturing and because of the way melters operate, typically the manufacturers would overrun the market, create inventories and that tends to cause a downward pricing cycle. I think we saw that last year. This is about the time where Europe began to slow pretty dramatically kind of mid-third quarter. I think the Chinese exporters to Europe, European manufacturers, everyone was running at a run rate for Europe that was well above what we saw kind of in the last 5 or 6 months. Inventories were created, and we got into a more aggressive pricing environment at the end of last year because of the rate of change of demand in the composites market. Today, demand is fairly weak but the rate of change is fairly stable. So while we're seeing weak demand, we don't think it's unexpectedly weak demand or anything that's particularly surprising. And as a result, we would believe we're in a more stable price environment today than we are -- than we were last year. I think our hope going into next year is we would see some places maybe where we had a little bit of pricing opportunity. We've had some nice new product launches, which could give us an opportunity to get some additional margin by creating…

Our next question comes from the line of Michael Rehaut, JP Morgan. Michael Jason Rehaut - JP Morgan Chase & Co, Research Division: First question on Composites, I think in your opening remarks or I believe it was yours as opposed to Mike's, but talking about longer-term margins maybe not getting to that double-digit rate, and I believe you used the term 2013 and beyond. Going back to the Investor Day and I think earlier this year with the turnover of the asset network, certainly that there was a goal of 13% to even 15%, I believe. Once you reach that, 75% of the network low cost. So has there been any change in the thinking there? Or is this more of just the double digits really won't be achieved due to the lower demand, which would negatively impact utilization and pricing in perhaps the near to medium term?

Thanks, Michael. And I looked back at my script and I think I read it correctly. But if not, I do want to make sure that I clarify it, if I put any confusion. Your comment was that I had said that it put double-digit margins at risk in 2013 and beyond. What I actually said is the way to the attainment of our double-digit margins beyond 2013. So obviously, that statement leaves open the possibility, we could certainly attain them in 2014. We haven't given that guidance yet but we tried to write the language in such a way that it was clear. We certainly think the attainment of double-digit margins is something we can see in our crystal ball. We don't think we have enough volume as we go into next year. We can put a stake in the ground and reconfirm previous guidance we had given, which said that we would get it next year. Now, I haven't gone back and run these numbers, but I would tell you if this year had been 5% growth in demand and we were anticipating 5% growth in demand next year, I'd be pretty confident telling you that we would still expect double-digit margins next year. So by and large, the change in outlook is just the difference in demand outlook. And if you compound those numbers, in effect 3% this year and 3% next year makes next year's market about the same size as we anticipated this year's market to be at 5% growth, and the next 5% growth is going to be a market that might not appear until 2014, if we stay at kind of at 3% growth rate. So I think it's really a volume story. The cost side, the asset side actually, the reduction in…

Yes. Mike, thanks. I mean I think the first thing that I'd point out that really since 2009 and since things had begun to recover, we've been pretty disciplined from a cost perspective and from a headcount perspective. We've actually been keeping headcount flat year-on-year. The big thing that's impacted this year is the variability as we've gone through the year, is largely incentive compensation and so -- and pension. And as we look at this year versus next year, from an overall run rate perspective, I would expect corporate expense to be largely consistent and where we hopefully might see a little bit of variability is in incentive compensation.

Our next question comes from the line of Adam Rudiger, Wells Fargo Securities.

You've already talked a bit about Roofing, so I don't want to try to rehash it all. But I was just curious about what happened at the beginning of the year that was so different than other years and really why the visibility this year was less than in prior years.

Yes. I think it's a fairly simple story, and we've talked about it but let me reiterate it, which is we just saw discounting levels coming out of the fourth quarter last year in terms of these winter buy incentives that are offered by manufacturers to distributors to encourage distributors to stock up inventories in the first quarter. The level of the incentives became so large, offered by Owens Corning and offered by our competition, that our distributors really had an incentive to not just buy in a bit more inventory but they actually kind of trade in shingles and to try to take a very long position in shingles early in the year and then carry that position through the second and third quarter and try to get as deep into the year as possible with a big inventory of low-cost shingles. It created a lot of distortions in the market, from our estimation, none of which were particularly good. So the discipline issue will be, in fact, that the new normal or in fact, our position, which is we think winter buys are a nice incentive for our distributors, destocking some inventories where we can pass on some reduced asphalt costs through and give our distributors some incentives to help us keep our plants running and be ready for the spring selling season. That certainly is a sensible position, we believe, for us to take with our customers. And if that we can sustain that position competitively, we think that's going to be the best position for our business and for our customers.

Okay. And then going back to Composites for a second, in the 10-Q I think you actually mentioned there was a favorable mix of Composites demand this quarter and then there were obviously some negative offsets. But what was the favorable mix that you cited?

One of the things -- and I mean this at a very detailed level within our product line, but the Roofing product line that we manufacture in Composites is a product called wet chop, which is a short fiber that goes into something that almost resembles a papermaking process to make fiberglass paper or fiberglass mat, which becomes the guts of a shingle. So almost think of it as we manufacture fiberglass pulp, to continue the paper analogy. Now we have very, very low-cost focused assets that are very efficient at making this product. As a result, we sell that product at an average price that's well below the price of a number of our other glass reinforcement products, which you'd be more familiar with than some of the products that we feature kind of on our investor website, like rovings, like Type 30, like dry-used chop, but we make very nice margins on them. When we're reconciling top line and reconciling revenue, if we lose a lot of tons of wet chop and mat, which we did in the quarter, you're actually going to see that your average selling price improves a bit because that would be negative to the mix because it's a low-selling-price product. It's not a low-margin product, though. So you get a little different answer if you reconcile earnings versus reconciling revenue, and that's the mix issue that we've put into the 10-Q.

Our next question comes from the line of John Kasprzak, BB&T. John F. Kasprzak - BB&T Capital Markets, Research Division: How much of your Composite business is now tied to just the roofing market in total in North America?

Well, if you look at the pie chart that we put out in our investor deck, which is Slide 10, you can see that we detail about 11% of the market is either U.S. and Canadian new residential construction or U.S. and Canadian residential repair and remodeling. Those numbers track pretty well with the mix of what roofing would see in terms of what's repair and remodel, what's new construction. So that would be sales by Composites to our Roofing business, sales by Composite on a merchant base into other roofing manufacturers. But it would also -- could include sales by Composites to other building materials manufacturers. So for instance, we sell glass into the toilets and bath and shower market, so a lot of fiberglass goes into the manufacturing of bathroom fixtures. We would put that glass into that segment where we show that 11%. We sell some glass into ceiling tiles, in the facings of ceiling tiles, which goes into the construction market. We would also allocate some portion of those sales into that market. So that's a little bit broader market driver than just roofing. So in this disclosure, we're saying in total with 11% of Composites revenue in the end market, I think you should assume that materially, that's being driven by Roofing, but Roofing is not 100% of that. John F. Kasprzak - BB&T Capital Markets, Research Division: Okay, great. The second question is do you think the fact that repair and remodel spending, in general, has been sort of soft again this year and we're -- but we're seeing new residential construction improve, is that having an effect on your product mix at all?

Well, I mean for sure, if I take that directly to the Roofing question, I'm not sure whether you're asking that as it relates to product mix in Composites or whether you.... John F. Kasprzak - BB&T Capital Markets, Research Division: Well, Roofing and Insulation I guess, specifically, yes.

I think specifically to Roofing where, obviously, we think 75% of that business is repair and remodel, and in fact, 15% of that overall end use market is sales of commercial and industrial products, which are primarily asphalt and other products. So what you would tend to think of as our Roofing business, which is shingles, 75% out of 85% of our shingle sales are really to repair and remodel. We certainly believe that the repair and remodel dynamic today is being affected by the tension and the challenges that we see in housing in the U.S., probably more impacted by home prices, level of equity in homes, people's feeling about their house as a decent place to invest, whether they're underwater on their mortgage, whether they actually have equity in their home, which would qualify them for a home equity loan, which might allow them to pay for a reroof job. If all those -- if the answer to all those questions are no, no, I'm not confident on my house; no, I don't think prices are going up; no, I don't have equity in my home, I'm going to try to find a way to patch my roof or do something that causes me to not put more capital into my home. When you get into an environment where people are feeling better about increasing home values, better about the stability of the equity in their home, you see a mortgage market begin to improve as appraisals begin to improve, which might underpin or support refinancings or home equity loans, those are the things that we would say are critical precursors towards seeing an improvement in the reroof market. We've shown -- at our Investor Day, we've put out a slide where we break the shingle market into…

Our next question comes from Bob Wetenhall of RBC.

From Slide 13, you're talking about EBIT margins of 15% at 1.5 million starts. I was curious, would you guys be willing to say what EBIT margins would be at 1 million starts in this segment? And also, what's going on with your market share position in Insulation?

Okay. I can get you close to what we would've said about EBIT margins, and then we can follow up on this, because I don't want to get too pinned on that number. But we've been pretty comfortable saying we think the business would produce $100 million of EBIT on 1 million housing starts. Whereas the last time through, at a 1 million housing starts, the business was about breakeven. So that's -- in terms of mix of business, improvement of the other businesses inside our Insulation segment, as well as cost reductions and operating leverage in our residential business, we think we've moved that kind of break-even point by about $100 million. Today, the Insulation business is doing about $350 million or $400 million a quarter as you see from the third quarter. So we would see some top line growth as we move from 750,000 starts or where we're in this year to 1 million. But if you say it's on relatively flat revenue just because I don't want to give kind of a revenue estimate. We were on 1.6 billion. $100 million on 1.6 billion would be about 6% margin. So we're -- we would say that, that guidance is short of double digits at 1 million starts. And then our guidance, which is we should be back to historical norms of 15% at 1.5 million, that would suggest that we do have pretty good operating leverage as we go from kind of mid- to maybe a little bit better than mid-single digits margins at 1 million. And the next 500,000 starts should be pretty high calorie, and that would be because we'd start to see capacity utilization in the insulation market get tighter, scarcity of insulation become a theme in the market, and we'd expect to see that the pricing curve would start to bend a little bit more in our favor than what we would expect in the early part of this curve. So we're seeing positive price now. But we're multiple price increases away from even getting back to historical averages, and we're going to continue to move pricing as quickly as we can, consistent with the growing market.

And market share trends right now in Insulation?

Market share is pretty stable. There is obviously -- that is also a competitive market, and there is competitive intensity as people try to position themselves for the recovery and make sure that they're aligned with the right kinds of customers and with the right partners. We feel very good about our mix of customers. We feel very good about our relationships, and I would say our primary goal as it relates to market share would be to try to find ways to help our customers grow faster as opposed to trying to figure out how we move a lot of market share around in the market we like. We like the relationships we have. We like the mix of customers, and our goal with the EcoTouch launch, our goal with what we're doing on EnergyComplete, our goal with specification selling and other things is to try to create business opportunities for our customers so that they potentially grow faster than the market. But we're satisfied with the customer mix we have.

So you're not really seeing any movement there is what you're saying?

No, I mean we, obviously, see competitive bidding where there is efforts to move share. But for the most part, I would say, share positions in that business actually have probably been stable over the last decade. There's been 4 of us that have been really 90% of that market now going back a full decade, and while you could see some movements in share over that period of time, it's generally been a market structure that's pretty similar to what we see today.

That's helpful. One final question in Composites. What percentage of your revenues are based on spot markets sales for delivery within the next 3 months versus longer-term contracts?

Yes, I would characterize it a little differently than the way you characterize it. But let me try to give you some thought on that. We have some customers, I would characterize them as distribution customers, and also some customers who have applications that are not quite as sensitive to input glass in terms of how that impacts their yields and the economics of the composites that they manufacture. And I would say those customers tend to buy at prevailing market prices, maybe different than spot prices. So it's not bid necessarily the way you would think of a spot price in a classic financial market. There are some switching costs. There are some marketing costs and other costs associated with switching to a new glass supplier. But if you can't sell them at a competitive, prevailing market price, they have liberty through the year to be able to move share position. I'd say in broad terms, that's probably less than 1/2 of our overall business mix and probably customers who really rely on us to give them some stable pricing so that they can price their end-use composite applications to their customer at stable pricing, where our product is integral to their design, it's integral to their yield, it's integral to their manufacturing process, and it's very important that we take a long-term view together with them to make sure that we keep them competitive and that we're working on technology developments and process development to keep them competitive. That's probably a bit more than 1/2 the business. I think that's a very broad generalization and have we been when -- geography by geography, market by market, you'd find pretty wide variations different from that broad characterization. But I think for the overall market -- the overall business kind of a $1.8 billion to $2 billion, that's probably a pretty good characterization.

That concludes the Q&A session. Now I'd now like to turn the call back over to management for closing remarks.

Well, thank you, Chanel, and thank you, everyone, for your ongoing interest in our company. Obviously, I think, you've heard from the tone of my comments and also Michael's comments that it's been a challenging year for us. We're not terribly happy with our financial results. But I think more broadly than that, we're not terribly happy with our dilutive forecast and guide expectations around financial results. As we look at actual management actions and strategy, we feel like we're still getting the right thing done this year. We feel we have the right strategies in our businesses. We're focused on getting operating leverage in our Insulation business and being a constructive part of the recovery in housing as an industry leader. We're focused on managing margins and volumes in our Roofing business in a way that we continue to sustain the kinds of margin performance that we saw in 2010 and 2011 and maybe reverse this near-term trend we're seeing in 2012 related to some margin compression that we think started earlier in the year. We're very focused on getting the right asset base in composites. I think we've been talking about that now for 2 years. Two years ago, we talked about it in the future tense. 3 or 4 quarters ago, we were talking about in the present tense, and it's my fond hope that by the end of this year we'll be talking about that in the past tense and that as we look ahead to 2013, what we're talking about is having the right inventory levels with the right asset base and the right customers around the world and that we start to get nice positive operating leverage in Composites. We're heading into next year certainly thinking that there's no reason why we shouldn't see improvement in all of our businesses, and that's certainly the challenge that we've given our leadership team. It's certainly the challenge we've given people of Owens Corning. We're very proud of the people of Owens Corning's hard work this year. People are working very, very hard to try to get the right things done in the business and to make sure that we are focused on doing the right things for our shareholders in the near term and the long term. We'll get over this little speed bump in terms of getting through 2012 and making sure that we can deliver on those promises in 2013 and beyond. Thank you very much for your time. Bye-bye.

Ladies and gentlemen, that concludes the presentation. Thank you for your participation. You may now disconnect. Have a great day.