Louisiana-Pacific Corporation (LPX) Q3 2019 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by, and welcome to the Third Quarter 2019 Louisiana-Pacific Corporation Earnings Conference Call. [Operator Instructions]. I would now like to hand the conference over to your speaker for today, Mr. Aaron Howald, Director of Investor Relations. Please go ahead, sir.

Thank you, Sydney. Good morning, everyone, and thank you for joining us today to discuss LP's financial results for the third quarter of 2019. My name is Aaron Howald. I'm LP's Director of Investor Relations. I'm joined today by Brad Southern, LP's Chief Executive Officer; and Alan Haughie, our Chief Financial Officer. As we have done in the past, we are hosting a simultaneous webcast in addition to this public conference call. The webcast can be accessed at our website, www.lpcorp.com. We have also provided a presentation with supplemental materials to which we will refer during this morning's comments. And finally, we have filed our dark 10-Q and 8-K this morning with some supplemental information. I want to remind all participants on the call about forward-looking statements and our use of non-GAAP financial information during our discussion. I will refer you to Slides 2 and 3 of the accompanying presentation for more detail. The appendix attached to the presentation has some necessary reconciliations that have been supplemented by the Form 8-K filing we made this morning. Rather than reading those statements, I incorporate them herein by reference. Now let me turn the call over to Brad.

Thanks, Aaron, and thank you all for joining us this morning. I'll begin today's call with a few highlights from the third quarter and an update on our transformation, particularly with regard to our growth and efficiency initiatives in the context of the current market environment. I will then turn the call over to Alan for a more detailed look at our financial results. First, I want to acknowledge and welcome Nicole Daniel, who recently joined LP as our Senior Vice President, General Counsel and Corporate Secretary. Nicole earned her juris doctorate from Indiana University and has both an impressive legal background and a wealth of senior leader leadership experience. Welcome to LP, Nicole. We are glad to have you on the team. I am pleased to discuss our performance relative to a market that remains challenging, with sluggish housing starts and stagnant OSB pricing. Specifically, I will talk about 4 highlights for the quarter. First, we are ahead of pace to achieve our long-term targets for growth and efficiency. Second, we hit our growth targets for SmartSide Strand, setting records this quarter for sales and for safe operations. Third, we significantly rationalized our OSB production while idling Peace Valley and curtailing OSB production in our Siding mills. And finally, we generated excellent operating cash flow. The difficult housing and price climate I discussed during Q2's earnings call continued into the third quarter. Single-family starts were a little over 3% compared to Q3 of last year. However, on both a year-to-date and trailing 12-month basis, starts were lower than last year. Despite these headwinds, we continue to make progress towards our transformation goals, which I will discuss in more detail shortly. As shown on Slide 5, we grew SmartSide Strand revenue 13% to $213 million, which is another quarterly record. We continue to grow SmartSide Strand revenue and volume at rates meaningfully above that of the underlying market. For OSB, prices were down 35% and volume was down 14%. Our business responded, delivering the lowest OSB cash cost of production in more than 2 years while also increasing our value-add Structural Solutions mix. All businesses continued to improve operating efficiency measured as OEE and gain on ground on sourcing savings, achieving $28 million in year-to-date efficiency improvements for LP as a whole. I'll put that in context relative to our longer-term targets in a moment. Finally, we reduced finished goods inventory by $29 million, which made a significant contribution to the $59 million of cash we generated from operations in the quarter. Alan will walk you through the impacts of this on our EBITDA and the other elements of our capital allocation strategy in more detail. So I won't steal his thunder, other than to point out that we completed the accelerated share repurchase in Q3, bought back a further $42 million worth of shares on the open market and declared a quarterly dividend of $0.135 per share here. Now I'll discuss our transformation in more detail and review our progress towards our long-term goals. The chart at the left of Slide 6 shows U.S. single-family housing starts and SmartSide Strand revenue on a trailing 12-month basis compared to last year. Despite an uptick in Q3, starts are down 3% over the past 12 months from 890,000 to 863,000. Over the same time frame, we grew SmartSide Strand revenue by 10%, demonstrating continued progress on the central theme of our transformation. The table to the right shows EBITDA generated by our transformation initiatives on a year-to-date basis. SmartSide growth has generated $14 million of EBITDA so far in 2019, even including a significant increase in sales and marketing expenditures. Improvements in OEE and savings from strategic sourcing and logistics optimization have added a further $11 million for a total transformation EBITDA impact in Siding of $25 million. Alan will share more detail in a few minutes, but the net result of this is a 19% EBITDA margin for Siding for a quarter in which we removed OSB production from its mills and significantly reduced finished goods inventory. In addition to sales growth, the Siding business set another record by ending Q3 without a single OSHA recordable injury. Safety is a core value at LP. It's about making sure every employee returns home safely to their families at the end of the workday. I want to commend the Siding business for an overall good quarter. The OSB market did continue to deteriorate in Q3. Our average selling price for OSB in Q3 was down 35% compared to Q3 of last year. OSB sales volume was down 14%, with our decision to idle Peace Valley being the largest single contributor. The OSB business responded to these market headwinds with focused operating discipline and outstanding cost control, resulting in the lowest cash cost of production in two years and the lowest cash conversion cost in 5 years. As a result, despite the lowest OSB market prices in 4 years, the business essentially broke even for the quarter. It rarely feels like a win to breakeven, but let me briefly put that in context by comparing the results to the last quarter, during which we experienced similar OSB pricing. In Q3 of 2015, OSB prices were essentially identical to Q3 of this year. We made about 100 million square feet less OSB this quarter than in 2015. But remember that in 2015, we were still operating Dawson Creek as an OSB mill, and we were still running Peace Valley. These 2 mills represent a bit over 200 million square feet less capacity available in Q3 of 2019. So while capacity was down by 200 million square feet, production only fell by about half that amount. We have said in the past that OEE improvements would be like finding a mill within our existing network. We are seeing that impact in this quarter. In terms of product mix, we produced more of our higher value-add Structural Solutions products in 2019 than in 2015. You would expect this to increase both our average sales price and our cost of production. Prices were, in fact, higher this year, but costs were actually lower. In other words, compared to 2015, we produced less total volume with a richer mix that earned us a higher price and with efficiency improvements that would have resulted in lower actual production cost. This comparison to 2015 demonstrates the transformation of the OSB business through growth of Structural Solutions, OEE improvements, operating discipline and relentless cost control. The OSB team is executing on our strategy, and it shows in their results. So far in 2019, the OSB business has contributed $7 million of EBITDA from growth and $14 million from efficiency for a total of $21 million towards our transformation goal. When added to the $25 million from Siding and the $3 million from EWP in South America, we have achieved $49 million of transformation EBITDA impact year-to-date. Our target for 2021 is a cumulative $165 million. We are 1/4 of the way through the 3-year period, and we have achieved 30% of our goal. We still have work to do, but I'm proud to report we are ahead of pace. I've talked about our business results at a high level. Now I'll talk about our products and link them to our strategic transformation. Our products provide a range of solutions with challenges faced by home builders and homeowners. Our SmartSide family of strand siding products are extremely durable and attractive and are available across a broad product offering, including lap, panel, trim and soffit. Our Structural Solutions portfolio of products are key components to a structurally sound and durable home. TechShield Radiant Barrier can increase the energy efficiency of a home. WeatherLogic helps protect against water intrusion. Legacy flooring provides industry-leading strength and stiffness. FlameBlock helps slow spread of a fire should one occur. Homebuilder site labor scarcity is a major impediment to housing growth. Our value-added products address this ongoing labor shortage by combining multiple steps into a single product system. For example, WeatherLogic eliminates the time, cost and mess associated with house wrap. SmartSide is easy to work with and install. Pre-finished siding comes with an extremely durable factory finish, send a homebuilder or remodel the step of painting after installation. Our strategy is focused on growing these type of products organically and through M&A. Last quarter, we announced the acquisition of PSPI, now known as LP Green Bay, our entry into prefinished strand siding. So it had no impact on Q3 results. Last month, we acquired a second pre-finishing facility, now dubbed LP St. Louis. Our investment in Entekra and off-site prefabricated construction solution was motivated by the labor scarcity constraints impacting housing growth. Entekra has begun operations at the new automated facility in Modesto, California. We are early in the start-up curve, but happy to report that market interest and acceptance of this product and solution is being validated as more product comes available. As we transform ourself into a provider of value-added and differentiated solutions, we are positioning ourselves to grow and thrive despite the ebbs and flows of the housing cycle. That is why we rebranded ourself LP Building Solutions. Our transformation is all about pushing ourselves and our products to solve more problems and add more value. Of course, our improvements in efficiency and growth above underlying market growth does not happen in the vacuum. We are transforming our ability to deliver top-tier returns to our shareholders and value-added solutions to homebuilders and homeowners because of the commitment and dedication of all 4,800 LP employees. We have talented and high-performing teams driving every aspect of our strategic transformation, and I'm immensely proud of their dedication and commitment. The current market may be sluggish, our transformation is anything but. With that, I'll turn the call over to Alan for a detailed review of our third quarter results.

Thanks, Brad. In addition to reviewing the consolidated results for the quarter, I'll be providing high-level revenue and EBITDA bridges between this year and last year for the Siding and OSB segments and briefly updating you on the progress of our capital allocation plan. Throughout my prepared remarks, I'll be referencing specific pages of our earnings presentation, which has been posted on our Investor Relations website. So first, let's move to Slide 7 for a review of the third quarter, starting with the consolidated income statement. Net sales fell year-over-year by $134 million. The combination of lower OSB prices, which were down on average of 35%; and commodity OSB volumes, which were down 14%, account for $164 million of the revenue decline. However, this was at least partly offset by the 13% growth of SmartSide, which added $25 million to the top line. Gross profit fell by $138 million principally due to the flow-through of the OSB price decline. The benefits from productivity and efficiency improvements were partly offset by lower fixed overhead absorption as a result of downtime and a quite significant reduction in finished goods inventory, more of which in a moment. Selling and administrative costs of $58 million increased by $7 million over the prior year. And as with the first and second quarters, the principal driver of the increase is our continuing investment in sales and marketing, consistent with our growth strategy. Other charges and credits of $9 million comprised of $5 million impairment charge and $4 million of severance charges, $2 million of which relates to the curtailment of our OSB mill in Peace Valley, British Columbia and $2 million for reorganizations within our corporate offices. Nonoperating income and expense of $5 million includes net interest charges of $4 million and other sundry items. And net interest expenses are higher than last year due to lower interest income as a result of significantly lower cash balances, given almost $600 million of share buyback activity over the last 12 months. So after a tax provision of $3 million, we generated $2 million of net income. Removing the impairment and severance charges and normalizing the tax rate results in non-GAAP net income of $10 million for the quarter and non-GAAP earnings per share of $0.08. The accelerated share repurchase initiated in the first quarter was completed early in the third, resulting in the delivery of a further 4.4 million shares in addition to the 12 million shares delivered at inception, and this results a diluted weighted average share count of 122 million for the third quarter earnings per share calculation. Now the capital allocation plan established in February of this year was in part made possible due to LP having a core business capable of generating healthy EBITDA irrespective of OSB prices, and for the most part, irrespective of housing growth. I'm, of course, referring to the Siding business and SmartSide, in particular. The tables on Slide 8, showing revenue and EBITDA by segment, help illustrate this point. The first three columns show the sequential quarterly results so far in 2019, with the final column comparing the third quarter of 2019 to the third quarter of 2018. So over the first three quarters of this year, Siding revenue has grown from $236 million in the first quarter to $238 million in the second and to $259 million in the third. In particular, SmartSide Strand revenue was 7% higher in the second quarter than in the first and then 7% higher again in the third quarter than in the second. And admittedly, this includes some benefit from the March price increases and seasonal demand patterns. The quarterly EBITDA for Siding has also risen sequentially even after a growing investment in marketing and a diminishing EBITDA contribution from OSB produced at the Siding mills. Siding is the only segment to exhibit this trend of consistent sequential quarterly revenue and EBITDA growth this year, which is why its share of total company EBITDA has risen from 74% in the first quarter to 87% in the second and to 96% in the third. In short, the Siding business is the reason LP has remained profitable during this down cycle in OSB prices. But before discussing Siding and OSB segment performance in more detail, I'll briefly cover EWP in South America. EWP's third quarter revenue of $105 million is $4 million lower than last year due primarily to adverse pricing, which flow directly through to EBITDA, which is consequently $5 million lower than prior year. Revenue in our South American business was impacted by the weakening of the Chilean peso against the dollar, which dampened the benefit of local volume increases. And the okay economic climate also drove prices and hence, EBITDA slightly lower by $1 million. And as a reminder, we now allocate a significant portion, about 75%, of our hitherto unallocated SG&A costs to the businesses to further drive line management accountability. We reconcile the quarterly detail for 2018 in an 8-K filed with the SEC in February, and it is against these recast 2018 numbers that all EBITDA comparisons are based. Slide 9 shows the third quarter revenue and EBITDA for Siding. The bars in the waterfall represents the year-over-year EBITDA impacts. Where applicable, the corresponding revenue changes are shown in orange text below the EBITDA bars. And we've grouped the growth in SmartSide Strand, the market investment and the efficiency savings together under the heading Transformation Impact in order to demonstrate our progress. And it's these same numbers that were summarized on a year-to-date basis earlier on Slide 6. Now there are quite a few items to cover on this waterfall, but I'll start with the overwhelming highlight, which is, of course, SmartSide Strand's year-over-year growth of 13%. I mean Brad has already said it, but I'm going to say it again: we had another quarterly record for SmartSide revenue. However, as disclosed in the appendices to the earnings deck, volume grew by 14% and gross prices by 1%. Rebates of around 2%, including some catch-up from the first half of the year, brought net pricing to a negative 1% compared to the third quarter of last year. This is in stark contrast to the second quarter of this year, in which volumes were flat to the prior year, with prices 3% higher. Therefore, on a year-to-date basis, SmartSide revenue has grown 10%, comprising 7% volume growth and 3% from pricing. This Smartside growth generated $7 million of incremental EBITDA, which was almost offset by $5 million of incremental year-over-year marketing spend. And I must emphasize that our investments in marketing is a long-term enabler for growth and should not be expected to correlate with growth in the period in which it is spent. Having said that, the marketing costs were at least offset by efficiency savings of $5 million generated through a combination of OEE improvements and strategic sourcing gains. As a reminder, during the third quarter of 2018, our Dawson Creek mill in British Columbia was producing OSB; whereas this year, it is ramping up production as a newly converted Siding mill and not producing any OSB. As a result of this conversion, we bore $3 million of unrecovered labor and overhead this quarter compared to 2018, which is an improvement on the $9 million incurred in the first quarter and the $6 million incurred in the second. The curtailment of OSB production in Siding reduced segment revenue by $12 million and EBITDA by $4 million, although the EBITDA loss was mostly price. I'm now going to address the bar labeled absorption in some detail, given its size and its impact on siding EBITDA for the quarter. In the third quarter of 2018, the Siding segment was producing about twice as much OSB as it was directly selling to customers while transferring the rest to the OSB segment at standard cost. With the removal of this production, as part of our mill optimization, the direct costs in the Siding mills have, of course, been saved. But there remains about $4 million of fixed costs that are no longer being funded by the OSB segment. And overall, this has a positive impact on LP's profitability because not only have we shifted OSB production to lower-cost mills, but the same $4 million shows up as a saving for the OSB segment. The remaining $3 million of the $7 million labeled as absorption is a natural consequence of the significant reduction in finished goods inventory. Any period during which a manufacturing operation sells more than it produces, will result in a capitalization into inventory of less fixed costs than it incurs. The impact on earnings is pronounced enough to call out this quarter because of the magnitude of the inventory reduction and because in the third quarter of last year, Siding sold less than it produced and so it overabsorbed costs last year. I should stress that this is a temporary phenomenon caused by the drive to reduce inventory and generate cash. So let's summarize the current status of the Siding business. The Dawson Creek mill is up and running. Siding is no longer producing OSB; has ample capacity for further growth, which should, in turn, deliver operating leverage; and is reporting its highest quarterly EBITDA this year while incurring its highest marketing spend of the year. We've directly entered the prefinished market with the recent acquisition of facilities in Greenway and St. Louis while posting another revenue record quarter for SmartSide. The Siding EBITDA margin is a respectable 18% for the quarter without the benefit of OSB production. So if we then exclude the cost of the Dawson ramp-up, which is by definition, temporary, the EBITDA margin would have been closer to 20%, which is our stated long-term target for the segment after all. Slide 10 shows the third quarter revenue and EBITDA waterfall for the OSB segment. As Brad said, first to breakeven performance for our OSB business is a remarkable achievement given the current OSB price environment and stagnant year-on-year housing. And this is truly a testament to the ongoing transformation underway within this business, and I want to walk through the controllable elements by which we delivered this result. After all, the drop in OSB prices accounted for the entirety of last year's EBITDA. And embedded within this $122 million of adverse pricing is a drop of 33% in Structural Solutions prices, which compares favorably with a 41% drop in commodity prices, testament to the power of our ongoing portfolio of value-added products. Commodity OSB volumes were down 14% year-over-year, and we took 97 downtime days in the quarter, 90 of which were market-related. Compared to the second quarter of this year, we collapsed 2/3 of our downtime into Peace Valley by idling it, with the remainder taken in Maniwaki. This approach is far more efficient than sprinkling downtime across our system and was a major contributor to our outstanding cost performance in a sluggish market. However, the combination of the lower commodity volumes and prices reduced EBITDA by $134 million. Therefore, in the absence of the offsetting action, which I'm about to describe, the OSB segment would most likely have reported an EBITDA loss of $11 million. So the transformation impact call-out for OSB includes $2 million of EBITDA due to increased volumes of our value-added Structural Solutions, and it also includes the EBITDA benefit resulting from closing Peace Valley, which is saving the company $2 million of fixed costs per month. Along with commodity OSB, we also produced TechShield Radiant Barrier at Peace Valley. We are growing Structural Solutions, but we won't cling to that volume at the expense of profit. And if there was a subset of our Structural Solutions portfolio that we were willing to let go, it was the high-cost TechShield manufactured at Peace Valley then shifted a great expense to the U.S. West Coast. As a result, we jettisoned $8 million of revenue with no loss of EBITDA and realized the $4 million of savings from the closure. Brad compared the benefits of improving OEE to earthing a hidden mill within our network. Arguably, it is the improvements in OEE and the ability to absorb volumes at our remaining mills that, in part, enabled the closure of Peace Valley. We are therefore including the savings, which can be thought of as accelerating or monetizing OEE as part of our transformation. And on the subject of OEE, the third quarter of 2018 was something of a high point. So there was minimal year-over-year improvement this quarter, although the benefits from the improvements over the first half of the year have been maintained. Now I covered the absorption impact that arose from the reduction of finished goods inventory in Siding when I discussed that segment. While this same phenomenon also impacted OSB EBITDA, though the distinction between absorption and downtime is rather blurry. Nonetheless, the combined impact of downtime and finished goods inventory reduction cost the OSB segment $7 million of EBITDA. This was, of course, partly offset by the $4 million of costs transferred, if you will, from Siding. And as I mentioned, when discussing Siding, this $4 million item is therefore awash to LP as a whole, which is rather academic for OSB because unlike Siding, which we -- as we know, is in growth mode, the OSB business remains focused on operating efficiency. The result of their efforts, much like in the second quarter, is cost savings of $7 million net of inflation. The result of all these efforts is a $10 million of EBITDA recovery, resulting in an EBITDA loss of just $1 million in an extremely harsh pricing environment. EBITDA for the first 9 months, therefore, is a positive $4 million. Our third quarter and year-to-date cash flows are summarized on Slide 11. Aside from EBITDA, a major driver of our $59 million of operating cash flow in the quarter was the $31 million of inventory reduction. This was almost entirely the result of a $29 million reduction in finished goods. However, during the third quarter of last year, although not shown on the slide, we increased finished goods inventory by $8 million. I'm stressing this point because it is the divergence of finished good movements of $37 million, in other words, $29 million inflow this year compared with $8 million outflow last year, that drove the absorption hit to EBITDA of about $10 million across Siding and OSB. From a cash flow perspective, this was unquestionably the right thing to do despite the negative impact on EBITDA and earnings in the quarter. On a year-to-date basis, we've now generated $58 million of cash from operations. As a reminder, we generated $54 million in the second quarter, which largely offset the $55 million outflow in the first quarter. Now when we introduced our capital allocation plan in February, we guided to a range of operating cash flow for 2019 and described the sensitivity of our cash flow guidance to OSB prices. We also stated our goal of generating $165 million of additional EBITDA from growth and efficiency by 2021. After adjusting for labor inflation and taxes, this $165 million should produce about $100 million of incremental annual cash flow over 3 years. Our progress on this transformation in 2019 is and will continue to be reflected in improved operating cash flow. For example, should the Random Lengths 7/16 OSB price ultimately average $180 per thousand square feet for 2019? We believe that our cash operations will exceed the $80 million we modeled at that price. As mentioned earlier, the accelerated share repurchase was completed in early August, after which we bought $42 million worth of shares on the open market at an average price of $24. We ended the third quarter with $318 million in cash, resulting in net debt of basically 0. And when including $350 million of undrawn revolver, we have over $650 million of liquidity. From a share buyback perspective, we have $158 million remaining of the $600 million authority outlined in our capital allocation plan for 2019. And I fully expect that when we report our full year numbers, I'll be announcing the completion of that plan. So before I turn the call over for Q&A. Slide 12 provides some limited guidance for 2019 and beyond. We are reaffirming all guidance, but further improving our capital spending guidance. On the second quarter call, we indicated a range of $160 million to $170 million for the year, and we now believe our capital spending will not exceed $160 million. And with that, I will open up the call for Q&A. Operator?

[Operator Instructions]. And our first question comes from George Staphos with Bank of America.

And congratulations on the progress. Brad, Alan, if you could talk a little bit quickly on Siding pricing and realization. And you'd mentioned that there is a swing in rebates from 2Q to 3Q, which explains a lot of the change in price realizations. Could you comment at all as to whether there's any change within mix, whether you're responding to any competitive activity? It wouldn't appear to be, at least from your formal remarks on volume. But for what it's worth, realization, a little bit lower than we were modeling. I just want to say that if there's anything else beyond rebates in that. And then I had a quick follow-on.

Yes. George, mix was an issue in the quarter. We did have a higher proportion of our sales go into our retail channel, which tends to be more skewed to our panel product, which is a little lower pricing than our lap and trim. So there was definitely a mix component. Alan talked about the rebate catch-up. And then I would say from other discounting and sales incentives, they were pretty normal for what we're doing at this time of year. But mix definitely played a component in it, along with the rebates.

So it would be fair to say that, on the one hand, you had a little bit more in the way of retail because of just what's happening in terms of starts versus, I guess, DIY and repair/remodel activity. Would that be what was driving that? And it sounds like you weren't -- you didn't need to respond significantly, if at all, measurably to any increase in competitor capacity in the siding market, which obviously has been an ongoing question this year for your business.

George, I would say, as far as competitive situations as it relates to lap and trim, that usually isn't something that is priced at point of sale. So it's not like we feel that in a quarter. It's more reflective of what we have to do on the rebate side. So we address the competitive situation through rebate and incentives that show up kind of -- no, they show up in our net pricing that we report but more of a subtraction from our gross price than immediate type of a response to competitive situations. And just the original part of your comments, I'd -- it's true that -- well, first of all, we've had really good growth at retail this year. And then -- and relative to what we're seeing in overall housing growth, there's a little bit of a skew towards selling more into retail relative to what we've done in the past from a growth standpoint.

And our next question comes from Chip Dillon with Vertical Research.

Yes. This could be, I guess, either Brad or Alan, you mentioned, I think, if I heard you right that if we assume the 1 90 North Central price, that cash from operations might actually exceed the $80 million you originally modeled, which obviously, attest to the efficiencies you've gained. The question I had, though, is we've seen in recent months a pretty big divergence between Southeast and Western Canadian OSB prices from that benchmark. So I didn't know if you could add some color to how maybe that might make it more of a challenge or maybe it won't as the case might be.

Well, I would say from a guidance standpoint, we haven't got -- we're sticking with what Alan has said. We haven't got to the point where we're differentiating between the regions. But no question, Chip. There's been that divergence, especially prior to the last couple of weeks of price movement. And if you think about the capacity, especially that we took out of our Siding mills that are usually selling into that North Central region, there's been -- it's been less competitive than in the regions that we had lower pricing. And so -- and I think some of these recent moves around capacity coming out of the market is going to have pricing across the regions even out a little bit more to what we've seen in the prior 3 or 4 months. But from us being refined enough to make that distinction on the guidance based on regional differences, we're okay sticking with what we said before.

Okay. That's super helpful. And then just one quick follow-up. You now, I guess, are not making -- or at least in the third quarter, OSB in the Siding segment, which I assume is reflective largely of Dawson making it up its learning curve. But if you could talk a little bit about how we could see the OSB production in that segment evolve as you do your next conversion. Obviously, I would imagine that you would continue to make OSB at your Siding mills as you -- until you fill that mill up with Siding demand.

Yes. First of all, let me back up and just get us all grounded on our capabilities for OSB production in Siding. So we can currently make OSB at 3 locations: Hayward, Swan and Dawson. That's where we can make OSB competitive. I guess technically, we could make it at all the other mills, but in reality, those are the 3 places that we put OSB capacity. But with that said, especially on a delivered basis, when you look at our current network of OSB mills in our OSB business, those would be the 3 highest-cost places to make OSB in our current system overall. So as we looked at rationalizing production this year as an absolute need for what we needed to do, and we ran a network optimization. And that's what led us to the decision to remove the OSB from those operations so that we could put the volume there into lower-cost OSB assets like Sagola. So that is the removal of OSB volume from our Siding is in response to our current market conditions. Obviously, we would rather be running OSB in those facilities when OSB pricing allows us to do so. So when we convert the next mill, it's a little bit depending on what that mill is, but we would certainly aspire to having OSB flexibility at the converted mill; and if not, retaining in the rest of the SmartSide system so that we could balance production across the whole system.

And our next question comes from Mark Connelly with Stephens.

Brad, recent stories are blaming old people for not moving out of their houses. And it seems to me that if that's the new trend, that it simply means that the market needs more low-priced homes than high-priced homes and not enough is getting built. So if that is the new trend, how does that affect your profitability in OSB and Siding? I assume you take a modest hit on square footage, but would there also be a material shift in your Siding demand if houses got smaller?

Well, certainly, if the house is smaller, it's going to use less OSB and Siding, that's by definition.

But is your average house materially bigger right now? I mean is it going to be a material hit to you if the average house in America gets bigger? Where do you sit on that curve now?

Maybe just clarify the question a little bit.

Well, I'm just -- I'm really just trying to get a sense of how big your average siding start is now. Is it materially bigger than the starter home? Or -- I'm just trying to get a sense of this, if this is going to be a big issue.

Got you. Okay. So I would say our -- the average home that SmartSide goes on day is bigger than the starter home. We typically start -- I mean I'm making a real generalization, but typically, starter homes are vinyl. And so as you move up the value equation on home size, that's where we tend to play more competitively from a hard siding standpoint. So yes, I agree that -- I mean this is just my experience, but I would agree that the average SmartSide home is larger than -- well, maybe not above average, but it's certainly not a starter home, typically a starter home. So if the market was to swing to a heavy -- a lot bigger proportion of smart -- of starter homes, then the size of the home would be smaller, and we would need to be more competitive there against vinyl. And the way you get competitive against vinyl, not necessarily is always about price but by offering a product offering that can be beneficial to a homeowner in a starter home and thus, the move into prefinish and some of the other specialty SKUs that we're working on in our SmartSide business.

That makes sense. Just a related question, in general, are the siding margins that you experienced in repair and remodel significantly different? You talked about rebating. I'm just wondering, do you spend materially more marketing in the R&R channel than you do in the new home channel?

We definitely spend more in the R&R channel than in the new construction channel or the retail channel, without a doubt. It's more of a consumer sell than what we see in new home construction. But I will say also, there's probably a little richer mix because there's going to be much more lap and trim compared to panel. So we do benefit on a margin and pricing standpoint because of mix. And as we move into prefinish and get -- and able to retain the incremental margin associated with a prefinished solution, which we want to play really well in repair and remodel, the opportunity -- and we look at repair and remodel as a higher cost to serve but also higher margin at the end of the day because of the prefinished opportunities that exist there.

And our next question comes from Ketan Mamtora with BMO Capital Markets.

First question, Brad or Alan, can you talk a little bit about sort of your OSB order books? And what you see in terms of inventory in the channel for this time of the year?

Yes. Sure. We're out of 2 to 3 weeks, which is normal for us, keeping right now. As everybody on this call knows, we had a really nice rally a couple -- the last couple of weeks in OSB pricing. The reason -- one of the -- from a demand standpoint, the reason for that rally was inventories were very low 2 or 3 weeks ago. And I think there was a scramble to rebuild those as some of the production came off-line. And so there has been some return to normalcy in inventory in the channel, but I would say it's currently still below normal for this time of year. But because the distributors and dealers are just playing it so close to the vest up to 2 or 3 weeks ago, there's been a little bit of scrambling in an attempt to rebuild inventory levels.

Got it. That's very helpful. And then just one other question on the Engineered Wood business, can you talk about how you are thinking of the business long term? I mean year-over-year EBITDA through the first 9 months is still down. There's new capacity that's coming into the market in the U.S. South. Just talk about kind of -- is this business really core to LP?

Well, it's core to LP. As I've mentioned before, the distributor base that we sell our EWP into are the same distributors, the same customers for Structural Solutions and for our Siding business. And so -- and it does make the portfolio offering that we have to the channel more relevant by being in the EWP business. But as I've also said in the past, that's no excuse for not earning the cost of capital. And so we continue to work on that within the business. You know, Ketan, we've done a lot around SG&A reduction. We've done a lot around pricing in that business, and we continue to work within our mills with our OEE initiative to drive efficiency. So we do definitely have a target of return -- that business getting to where it's returning the cost of capital. And I -- and we've made good progress against that goal, but there's still work to do.

And our next question comes from Mark Weintraub with Seaport Global.

A couple of real quick ones. First, if you could just let us know -- since you're not making OSB in the Siding segment anymore, at least for now, in the fourth quarter and the first half of this year, what would the transfer to OSB segment have been in those periods? It was $4 million in the period just ended.

Given as being working backwards through time, $1 million less each quarter. So about $3 million in the second and $2 million in the first, something like that.

Okay. And the fourth quarter of last year?

I'll have to get back to you on that one.

Okay. Sure. And then second, are the finished goods inventory now at levels where you want them to be in the Siding business?

Yes. Yes.

Great. Third, can you update us on the rebates? Are we still seeing -- is it a fair place to expect we're still up like those 2%-type rebates? Or have market dynamics changed at all since the -- in the last few weeks, months?

Well, I mean, market dynamics haven't significantly changed. I mean the rebate -- the majority of the rebate in the third quarter was actually a sort of a catch-up to a year-to-date basis. So the pricing and the rebate impact, if you take Q3 as a discrete quarter, was a little better than it looks.

And our next question comes from Sean Steuart with TD Securities.

A couple of questions. Are you at a point where you can provide any guidance on 2020 CapEx plans and maybe some detail on specific discretionary initiatives?

Yes. I mean, it will be lower than the projected maximum of $160 million in 2019. At the moment, I would say a very safe number to work on for 2020 is about $140 million.

And would there be any exciting conversion capital in that number?

Nothing significant in there. No.

Okay. Second question, there was reference in the notes to a $5 million impairment to a Canadian facility that you expect to sell within the next year. Any context you can provide there?

Not yet for confidentiality reasons. So I would say it's a small facility that we're looking at that we're most likely going to disposal.

And our next question comes from Steve Chercover with D.A. Davidson.

So I think maybe Ketan touched on why prices are down in Engineered Wood. But on a percentage basis, profits are down more than revenues, which I think is surprising since presumably the cost for OSB used for web stock is down a lot. Can you help us reconcile that?

Well, I think I would say that the component of total raw materials cost in EWP represented by web stock is small enough that, that impact was not surprising to us. There's a significant amount of raw materials other than that web stock. And so while the cost reduction of the OSB is helpful to our I-Joist manufacturer, it has less of an impact on the other components of the EWP business.

Okay. So is it fair to say that it really didn't run that well in the quarter?

I'm sorry. I didn't catch that. Can you repeat the question?

Operationally, I mean. Clearly, you've made some good progress in OSB with the lowest production cost in two years. But you couldn't say the same thing for EWP. Is that accurate? That it ran well?

It is accurate that we cannot say we've made as much progress in EWP as we have in OSB. And I will say, just to kind of dive another level of detail, we still -- the LSL production in EWP is a very volume-sensitive mill. It's a big mill, has a lot of fixed costs. And so as we -- the mix component of that product into our EWP business can have a pretty big impact on overall profitability for the quarter. Our EW -- our I-Joist and LVL business remains okay, but we do have quarter-to-quarter swings in LSL production that can really impact earnings. So I'm saying when it looks funky, Steve, it's usually a mix change that's associated with LSL.

Okay. And then just one on the transformation in OSB. I mean you basically found, to paraphrase you, about a mill's worth of capacity thus far, and we still got another 2 years to go. You think there's another mill to be found?

There's a lot of opportunity. There is a lot of opportunity in there, Steve. And we -- and as we've talked about in the past, there's minimal capital associated with funding that mill. I mean there's some capital that's associated with it. It's primarily really just running really well. And so all you're really doing is adding variable costs. So we are really focused on our growth in OSB being a result of OEE, and then there's still a lot of room to answer your question.

And our next question comes from Paul Quinn with RBC Capital Markets.

Just a high-level question on SmartSide, if you could break down that end-use demand between home and, I guess, outside the home. And then within the home, what portion is new home? Or what portion is R&R?

So what we said for a long time, about 40% -- we estimate about 40% SmartSide volume goes into new home construction. And that leaves 60%, obviously. 100 minus 40 is 60. About -- we've got 20% to 30% of the remainder going into retail, into the shed segment and then 20% to 30% into repair and remodel in some form or other. And those are -- we don't have direct visibility into that because it goes into distribution, but that's our estimate on where the product ends up.

Okay. And then as Dawson's in start-up, maybe you could give us an idea of timing for the next mill conversion. Does that take us out at least through 2021?

Yes. Paul, I think we'll be talking probably more specifically -- be in a position late next year to talk specifically about location. And then, as Alan mentioned in the CapEx guidance, we would currently think we'll begin spending capital on the conversion in 2021 for a late-2021, early-2022 production.

And we have a follow-up question from Mark Weintraub with Seaport Global.

One quick one on marketing expense. I think you mentioned that you're kind of at a high point seasonally. What type of expectations in terms of the trajectory for marketing should we be thinking about in Siding?

In general, continued increases. I mean the marketing is supporting our drive into repair/remodel and prefinishing. So as we've said, we are continuing to grow Siding and continuing to invest in that growth. We have some sufficient confidence in it, but we believe the returns on these marketing dollars will be excellent.

Okay. And then lastly, in the slide at the end, you talk about returning about 50% of cash above and beyond what's needed to sustain the core business and growth, Siding. Can you give us -- is there any sense you can give us as to how to think about what that number to sustain the business and grow Siding is?

Yes. Certainly. And I'm -- it's somewhere -- it's about $120 million to $130 million a year, was our rough estimate that included necessary maintenance and the relevant elements of growth capital.

And we have a follow-up question from Chip Dillon with Vertical Research.

Yes. Just an update on your plans for buybacks. You had mentioned, I think a year or so ago, when you first started talking about your new capital allocation that you would eventually incur a meaningful amount of net debt. But as long as you had a revolver in place, you were good to do that. So I just was curious if you're continuing to -- what -- kind of some guidance in terms of how much you might be looking to buy back in the next quarter to next year?

Well, that's a little -- pushing me to say a little more than I'm willing to say at the moment, Chip. So we will be completing the $600 million buyback, obviously and...

By year-end?

Yes. Yes. And thereafter, we have -- so when we release our fourth quarter results, that's the point at which I'll give further clarity on how we see -- what we see our next tranche of development being. So it's a little early for me to come out with that right now.

All right. That's super helpful. And just a quick follow-up. Of the -- that means you basically have $120 million left for the fourth quarter. Was there any reason you couldn't have been buying back stock up until now? I'm not sure what your like lockups are or quiet periods are.

No. Not really. Other than post the last earnings call. But I will say this, and 2 days ago, the share price was $29 change and $30 or so. I don't consider the difference between $30 yesterday and $24 that we bought our shares back at -- over the third quarter as being a meaningful difference. I still consider the stock to be significantly undervalued. And the -- our buyback activity is not in any way related to an assessment of the share price. I don't consider the share price to be high.

And I'm not showing any further questions at this time. I will now turn the call over to Aaron Howald, Director of Investor Relations for any further remarks.

Okay. Thank you, everyone. This concludes our earnings call for Q3. We look forward to speaking with you again sometime in February to discuss our fourth quarter results. Back over to you, Sydney.

Ladies and gentlemen, this concludes today's conference. Thank you for participating. You may now disconnect.