Teck Resources Limited (TECKB) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by. Welcome to Teck Resources Modeling Workshop Conference Call. [Operator Instructions] This conference call is being recorded on Thursday, April 2, 2020. I would now like to turn the conference call over to Fraser Phillips, Senior Vice President, Investor Relations and Strategic Analysis. Please go ahead.

Thanks, Elena, and good morning, everyone, and thanks for joining us for Teck's Modeling Workshop Conference Call. Before we begin, I would like to draw your attention to the Caution Regarding Forward-Looking Statements on Slide 2. This presentation contains forward-looking statements regarding our business. This slide describes the assumptions underlying those statements. Various risks and uncertainties may cause actual results to vary. Teck does not assume the obligation to update any forward-looking statement. We'd also like to point out that we use various non-GAAP measures in this presentation. You can find explanations and reconciliations regarding these measures in the appendices. This session is being recorded, and a playback will be available on the Investors section of our website by tomorrow. Well, again, thanks for joining us in these somewhat unusual, obviously, and difficult times. Unfortunately, COVID-19 has necessitated a change in format for the workshop, as you're obviously aware, from our in-person meeting that we had planned to hold to a conference call. Actually, we contemplated whether to postpone the workshop until perhaps the world will normalize. But as you may recall, we already postponed it once last fall, and all the materials were prepared after a lot of hard work by the team. So we decided to proceed with the conference call. To the extent people aren't able to listen to it all, well, as I said just a moment ago, it will be recorded, and all the materials, of course, posted to the website. Like all of you, I'm sure we're working from home, and everybody's dialed in remotely. So we'll anticipate everything will go smoothly. But if not, just bear with us and we'll sort out any hiccups. This next slide shows our agenda for this morning. We're going to kick off with base metals, starting with a presentation on base metals pricing and concentrate contracts by Mike Schwartz. He's our Director of Market Research. Next, James Woeller, Manager of Strategic Initiatives and Business Analysis for our base metals business, will take us through modeling of the base metals business itself. We will then turn to the modeling of our steelmaking coal operations with Ryan Podrasky. Ryan's our Director of Finance and Operating Excellence for the coal business. And then energy modeling with Rob Sekhon, who's our Director of Finance for our energy business. And Glenn Burchnall, our Director of Energy Marketing and Logistics will also be available for questions. We will then look at corporate income statement and balance sheet items with Crystal Prystai. Crystal is our VP and Corporate Controller. And then finally, Thomas Cheung, our Director of Tax, will walk us through our income and resource taxes. We're going to have -- unfortunately, it's just not going to be interactive all through the piece, but we will have Q&A after each presentation. And then at the very end, conclude with a final Q&A to the extent there are any questions remaining. And we'll wrap all of this up by noon Eastern Standard Time. Each time we are ready for questions, the operator will provide some instructions on how to join the queue. Before we start, I just want to provide some context. We will not be providing any new information or guidance on the call. The purpose of this workshop is to help you use our public disclosure to model our business. To do this, we'll use our reported results and other public information sources available to you. With that, let's get started with Mike Schwartz on base metals pricing and concentrate contracts. Mike, over to you.

Thanks, Fraser. Good morning, everyone. Operator, if we can start at Slide 3, please. So today, I will cover some of the basics of concentrates and their terminology that we use in the concentrate market, and then cover the workings of a copper concentrate contract in invoice, and then we'll sort of top off with some zinc and lead. I'll also cover some of the other costs that go into the sale of a ton of concentrates. On Slide 4, the first thing you need to understand for those of you that are new to the mining business is how we get from mine to concentrate. Now this is a basic flow sheet that covers most of what happens in a typical sulfide mine, whether it's copper, zinc or lead. There's some obvious differences, but for the purposes of this discussion, we're going to try to understand some of the terms that we will be using throughout this presentation. Starting at the top left, we have the ore that comes out of the ground. And for copper, that tends to grade about 0.5% to 1% copper contained in the ore. Zinc grades tend to be a bit higher at 5% to 10%. And while primary lead mines are relatively rare, the grades of lead in zinc or silver concentrates tend to be about 2% to 4%. Again, this is just for illustration purposes only. The ore is then crushed, put through a SAG mill and/or a ball mill reducing the particle size, then it flows through a flotation circuit to separate out the desired minerals. There's often a regrind and recirculation to capture or increase recoveries. And then the separate materials are sent either to a tailings or through filtration, where the water is removed. What comes out of the filtration process is the concentrates. In zinc concentrates, we typically get around 50% metal contained in concentrate grade. In lead, it is slightly higher, and in copper, the concentrate grade tends to average between 23% and 29% copper contained, with the majority of the other elements in all 3 concentrates being sulfur and iron. When we sell concentrates, we sell them as wet metric tons with about 7% to 10% moisture. A dry metric ton is the calculation that removes the water from the equation. The water itself is necessary to prevent the concentrate from dusting during storage and transportation, and whereas too much water in the concentrate will cause agitation in the hold of a vessel during shipping and with sufficient momentum can actually capsize the vessel. So we have to be careful with the amount of moisture that we put into the concentrates when we ship them. From the ore grade -- sorry, you can often hear these terms interchange during the discussion with miners, but as a simple rule, the terms on the right is the concentrate flow, ore grade out of the mine, to concentrate grade out of the mill, to a shippable wet metric ton to a calculated dry metric ton that removes the moisture and all the other mineralization in the iron, sulfur, gold, et cetera, you arrive at the contained metal component of the concentrate. As we continue through this discussion today, we will see how we get to the next step, which is the payable metal for each of the concentrates. You should also know that companies will report differently on these terms. Some companies report production as contain metal, some report as payable metal and some companies that like to see analysts really work for the money, will report equivalent metal, where you have to calculate back to the actual content of the metal and the concentrate. Next slide. So to start off today, we're going to look at a typical concentrate -- copper concentrate contract, and we'll go into a little bit more detail in copper initially than we will for lead and zinc. But what we're going to cover here will be very similar to lead and zinc. On Slide 6. As I said, a typical copper concentrate can range in copper content from 23% to 29%, with some concentrates guiding -- grading much higher. Over the last 2 decades, there has been a decrease in the overall concentrate grade across the industry, where 30% used to be a typical concentrate. Today, 25% to 27% is much more typical and concentrates as low as 18% to 20% are actively traded in the market. That being said, if the copper content of the concentrate is below 30%, the smelter will typically pay for 96.5% of the value of the copper, subject to a 1% or 1 unit deduction, which means that if the concentrate grade is 27%, the smelter would pay for 27% minus 1 unit or 26%. At 30% to 38%, the deduction is essentially removed and the buyer pays for 96.65%. And above 38%, the buyer pays for 96.75%. These payments are based upon agreed to and -- or agreed upon future unknown quotational periods based on the copper price on the LME exchange. Next slide. We note that in most copper concentrates, there is also present economic values of gold and silver, and miners will get paid for their value. If the content of the silver in a copper concentrate is below 30 grams per dry metric ton, not the copper content, but the total weight of the concentrate, then the smelter is unwilling to pay for the silver as he's less likely to be able to economically recover in the process. However, above 30 grams per dry metric ton, the smelter might be willing to pay 90% of the value of the silver contained in the concentrate. Similarly, for gold content, anything under 1 gram of gold per dry metric ton typically does not get paid for. Above 1 gram per dry metric ton, payments can range from 90% to 98% payable on the gold, depending on the gold grade, with higher gold grades commanding a higher payable percentage. These terms are all negotiable and will differ between regions, customers and metal content, but they can be a significant revenue source to both the minor and the smelter. When miners report their costing in their financial statements, the revenue from the precious metal stream are usually used as an offset to the C1 or reported cash costs. So the next slide. We have the treatment charge. This is looking back historically over benchmark annual settlements, what the smelters will charge the miner to process their material. Historically, this was the payment made to the mine -- made by the mine to a smelter or processing facility to turn the concentrates into metal and return the metal to the miner. Today, most custom concentrates are sold directly to the smelter with no intention of the mine ever receiving the metal back. The smelters pay for the payable content of the metal, less the treatment charge, and then sell the refined anodes or cathodes themselves on the open market. So when we talk about treatment charges, for a miner, a low treatment charge, discount off the metal contained, is good. And for a smelter, a high treatment charge, what they can charge the miner to process, is good. So miners like high metal prices and low treatment charges, and smelters like high metal prices but are more concerned with high treatment charges. Looking back over the years, their reported annual concentrate treatment charges were negotiated. Previously in copper, there used to be some price participation in the contract with the smelter and miner would share in drastic price movements on the LME, but that has not been in the standard contract for well over a decade. In copper, there is also a refining charge. In copper, as opposed to lead and zinc, copper smelters do exist without refineries attached to them. So the refining charge is there to capture some of the cost that the smelter has to pay to the refinery to convert blister into capital. On Slide 9, looking back over history, the relationship between the annual contract and the spot terms in the market. Spot terms usually traded a premium to the miner and a discount to the smelter. Miners have historically been willing to give up or pay a higher treatment charge to ensure that they have secure homes for a large portion of their annual production. Annual and midterm contracts are usually negotiated between October and December and cover a 1-year period. Midterm contracts used to be more prevalent. They're not so much anymore. They were just offsetting with a June settlement date rather than a December settlement date. So there will be annual contracts from June to June as opposed to January to January. Spot contracts refer to any contract that is not annual. Some of these contracts could be for one-off parcels delivered in a short period of time, giving the miners some flexibility on the production schedule. But some spot contracts can also be for future periods in time. For example, a miner can go into the spot market and get a quote delivering a parcel in 6 months' time. Negotiating parties may also enter into contracts over multiple delivery periods over a period of time. For example, 3 parcels to be delivered, one each in March, June or August. All of these would be considered spot tonnes and would come with very different valuations and maybe combined into one set of terms or multiple sets of terms. So when you read the spot assessment in a metal bulletin or CRU publication, note that there's a lot of work that goes into the analysts trying to bring everything they hear in the market back into a single number. And understand that their view on the spot market is an assessment of various things they have heard and very little is actually verifiable. Unlike the open-outcry ring at the LME that gives massive liquidity with immediate and accurate price discovery, treatment charges or spot treatment charges can be very vague. On Slide 10, I'll walk you through a typical copper concentrate calculation. And we'll use these components on the top here. So using a $5,600 per tonne copper price, a $17 per troy ounce silver price and a $1,600 per ounce gold price and using a typical concentrate that contains 26% copper, 150 grams per dmt of silver and 2 grams of gold per dmt, this is sort of a typical concentrate. The copper payment for the metal is made at 96.5% with a minimum deduction of 1 unit. So 26% times 96.5% would equal 25.09%. But with the minimum deduction of 1 unit, the single unit is calculated here. So 25% of the copper is payable. If the copper grade was 35%, at 96.5%, it would be 33.77%, which is less than the 1% minimum deduction. So in this case, the percentage payable would be used. So 26%, we deduct 1 unit, pay 25% at a copper price of $5,600. This works out to $1,400 per dry metric ton of payable copper in the concentrate. With silver, above 30 grams per tonne and pay 90% of the silver content, we have 150 grams per tonne, times 90%, 135 grams are payable. That works out to about 4.3 troy ounces, times the silver price of $17 an ounce. And we have $73 of silver value in 1 dry metric ton of concentrates. Similarly with gold, we would pay at the low end of the scale. 90% when the grade is over 1 gram, and that works out to about $96 per dmt. This means that the total payable value of the metals in the contract works out to $1,569.10 per dry metric ton. On Slide 11, we take the payable metal, dry metric tons and then we calculate the deductions or treatment charges that we would have to pay to the smelter. We start with the base treatment charge of $62 per dry metric ton, then we calculate the refining charge, which is expressed in cents per pound of payable metal. A dry metric ton is 2,204.6 pounds, times that by 25% from the previous slide, gives us 551 pounds of payable metal for the refining charge. So working through this on the silver and gold, we come out to total deductions of $98.24. So our final invoice per dry metric ton of concentrates is USD 1,470.86. Based on the value of the payable metal content versus the invoice value, this would make up about -- treatment charges would make up about 6% of the total payable value of the metal, not including any of the other free metals that are not paid for by the smelter and recover. Also, the smelter can assess penalties for certain elements that are above the threshold in the concentrates. These could include arsenic, zinc, fluorine, cadmium, antimony, mercury, anything that the smelter feels is deleterious to their production process and is a cost to them to remove from the process. We'll not get into those here because they're all negotiated on a mine-by-mine basis, but in your modeling, if you see high levels of these elements in the assay, expect the mine to be penalized for them. On Slide 12, we can just look at a various number of different scenarios under treatment charges and price scenarios and how the example that we've illustrated here would fit on that curve. Moving to Slide 13. We can now look at zinc concentrates. The zinc concentrate contract is essentially very similar to copper with a few minor variations. Slide 14 shows that a typical concentrate grade higher than copper and rather than deduct 1 unit below the cutoff grade, zinc deducts 8 units or 8%. Zinc smelters also pay for only 85% of the content of the metal. And this goes back to historic recovery rates in zinc smelters, which were typically lower than that of copper smelters. Today, this is just part of historic calculation, and as my previous boss used to tell me, whether the payable is 95% or 85%, it is just math, and it becomes part of the negotiation of the overall value proposition of the concentrate. In zinc concentrates, there are not usually significant amount of gold left according to the concentrates, but silver can be present. However, it is much more difficult for a zinc smelter to recover the silver out of the zinc concentrate, so the payability of silver is much different than copper. So in silver, below 93 grams per dmt, there is no payability, and above 93 grams per tonne, there's a 3 ounce deduction and the smelter will pay 70% of the remaining ounces. Slide 15. Unlike copper, mostly, smelters are integrated with their refinery, so there is only a treatment charge and no refining charge. In zinc, the treatment charge is negotiated each year and set to a basis price. In the past, when price participation was included in the contract, this basis price was used as a point of reference from where price participation would be calculated. And we're going to get into that in a bit. Slide 16. Similar to the copper, the TC deduction from payable on the value of the concentrates' price participation is escalated or de-escalated based on the movement of the LME price over or under the basis price, which is why you see, historically, the benchmark TC would move around a lot more than it did in copper. Slide 17. This is a historical price participation calculation, and using a basis price of $2,200, and this is also negotiated at the time of the contract. So the basis price would move depending on when the contract was negotiated. Where and when prices are above or below $2,200, an escalator is added, and below, a de-escalator is taken off. On the upper side, there will be $0.05 per dry metric ton for every dollar move above $2,200 and below $2,200, a de-escalator of, I think, it's $0.02 per dry metric ton is given back by the smelter. Next slide. So looking at our payability calculation on the zinc concentrate contract using a zinc price of $2,400, and again, a silver price of $17, we're selling a zinc concentrate today with 50% zinc content and 5 ounces per dmt of silver. In zinc, we pay for 85% with a minimum deduction of 8 units. At 85%, a 50% zinc concentrate would calculate to 42.5%, which is higher than the minimum deduction of 8, so we'll go with the 8 units. Therefore, 50% less 8 is 42%, times the $2,400 per dry metric ton, gives a payable metal value of $1,008. Okay. And for those of you playing at home today, at $0.83 zinc prices, that number drops below $770 per dry metric ton. Silver, we have 5 ounces, so it's payable at 70%. And after we deduct the first 3 ounces, you get 2 ounces at 70%, gives you $17. Sorry, times $17 gives you the $23.80. So the metal payable in this example is $1,031.80. On Slide 19, we move into the treatment charge. So there's no refining charge in zinc, just treatment charges. At $245, on a $2,400 zinc payment, we'll add in, for discussion purposes here, an escalator. So at $2,400 zinc with $200 above the $2,200 basis price that we started the contract with. So for $0.05 for every $1 above gives us an extra $10 to the smelter per dry metric ton. So the total deductions for the mine are $255, giving us a payable of $776, which works out to about 25% of the value of the concentrate in the treatment charge. On Slide 20, we could show the various price participation -- or prices and treatment charges for zinc and where our example would settle in. Moving on to Slide 21. We'll look at the lead contract, which is very similar to zinc. Slide 22. Lead grades, as I suggested, were higher than zinc concentrates, closer to 60%. So less than 60%, we can deduct 3 units. And above 60%, it is pay 95%. The silver payment normally kicks in at 50 grams per dmt. On Slide 23, we look back at some of the historic lead settlements. Lead, there also used to be price participation historically but has been mostly moved from the current contracts. On Slide 24. Looking back at historic lead treatment charges, you can see the volatility in the lead TCs. Some of the spike that you're seeing there in 2019/2020 had to do with the U.S. trade war that was imposed on U.S. exports, which really affected concentrates from Doe Run and from Red Dog going into China. Lead TCs can also be differentiated between high and low silver concentrates, but we won't get into that here today. Next slide. Looking at a typical lead payability table at an LME price of $1,800 and a silver price of $17, we modeled a concentrate that contains 54% lead and 455 grams of silver per dmt. At 54%, lead at 95% payable, defaults to the minimum deduction of 3 units, times $1,800 a ton, gives us $918 of metal value. We add in the silver, which again hits the minimum deduction, that gives you $221 of silver value per dmt and a total payable metal value in the concentrates of $1,139. On Slide 26, we bring in the deductions of the treatment charges. The payable discount -- is discounted by the treatment charge, and in this case, in lead, there is a refining charge for the silver. So the smelter will charge the miner a refining charge for the silver, which can be up to $1.50 per troy ounce of payable silver. There can also be refining charges for gold if it is insufficient quantities and if it's negotiated into the contract. This gives us $189.50 to the smelter, and leaves the mine with $949.50. In this case, about 17% of the payable revenues are going to the smelter, not including any of the free metal components from the previous page. Slide 27 gives you a plot on the graph of where that contract would sit on historic tables using various different treatment charges and prices. And then Slide 28, before we close off, we want to cover some of the other costs and revenues that impact our invoicing. On Slide 29, if we go back to our copper invoice example, and I apologize, we updated the copper price on those slides, but missed this one. So the mine number that I've got there should be $1,480 from all of the previous slides, not $1,684. So that's the invoice number that we got. The final invoice number from the copper invoice, but that's not the final revenue. $1,480 does not go directly into the pocket of the miner, as the miner is usually responsible for getting the product to the customer or at least to their port. So depending on how the concentrates are moved from the mine to the customer, there are additional charges that have to be taken into account. From the mine to the port, they may be trucking or rail or possibly pipeline. And depending on how the pipeline costs are apportioned, that cost to go to the port could be -- that cost to get to the port could be between $0 to $45 per wet metric ton. Remember, we have to pay the railway to move the water, but we can't charge for it. Port costs can run around $20 per wet metric ton, again, depending on whether the mine owns the port or it's a third-party port. And shipping costs, depending on where the material is going or whether the mine has negotiated long-term shipping contract or COAs with the shipping line, oftentimes, you'll hear us talk about CIF Rotterdam parity or CIF main Japanese port parity, where the miner and smelter will negotiate the shipping terms based on the average cost of going to Rotterdam or to a main Japanese port. Anything outside of that cost would be a cost to the smelter. Next slide. In zinc, using our invoice value of $777 per dry metric ton, there are storage costs. For example, at Red Dog, we incur additional storage costs for the 9 months that our concentrate set up at the port, waiting for shipping season to start. Whereas at Antamina, we have a pipeline on our own port, but we incurred dewatering costs after the slurry comes down over the pipeline. Red Dog has long-term multiyear agreements with shipping lines, and they ship mostly in fully chartered vessels, whereas a Chilean miner or a trader in the spot market would have only part of a holding vessel and may be subject to higher market terms. And then where is the cargo going? How far? What port? Is that to be transshipped into barges or smaller vessels to get to the customer? And who incurs these costs? All of these have to be considered, and all of them are part of the negotiation process for a contract. On Slide 31, we also want to talk about quotational periods. From month-to-month, prices can be very volatile, and such, exposure to these prices can be significant. Smelters are in the process -- in the business of processing material and capturing free metal where they can. With only $100 of treatment and refining charges to work with, a $200 per tonne move in the copper price from 1 month to the next could be disastrous to the smelter. So all contracts are done with unknown quotational periods. These can be monthly averages, averages over a period of time, over several months, but they are agreed to ahead of time for a period of time in the future that is agreeable to both parties. They are priced either off the LME or the SHFE, the Shanghai Futures Exchange, for copper, and the LBMA or the COMEX or some other index for precious metals. This way, both buyer and seller can agree to a price for selling and buying at the end of the contract. Smelters typically look for QPs, or quotational periods, 2 to 3 months after delivery so they can match the physical refined sales with their concentrate purchases. Invoices will be sent out provisionally by the mine on shipment or delivery and then adjusted later to the agreed-upon quotational period in the future. So when you hear companies talk about provisional price adjustments, this is where this takes place. You provisionally invoice, you deliver, your quotational period is 2 to 3 months later, and that is the price that's agreed upon. Next slide. The last thing we want to cover on the piece of revenue, and this accrues only to the smelter, are the refined metal premiums. And premiums are for metal sold to customers, which usually cover the cost of transportation, warehousing, financing and alloying of the product into different shapes. These premiums are usually charged to the end customer and are over and above the LME price of the day. Smelters will also sell any of their byproducts, including acid from their sulfur capture, specialty metals, or precious metals they can recover and they don't pay for in the invoice as well as any residues or granulated products. That's where the smelter gets their revenues. And that, in a big nutshell, is how we value a copper concentrate contract across most base metals. And Elena, you can now open it up to any questions.

[Operator Instructions] Our first questions is from Dan Day with Bioresource.

Fraser, it's actually with B. Riley FBR. I worked with Lucas Pipes. Just curious how the concentrate kind of pricing method differs from any like SX-EW producers? Like, I guess, is that just cutting the smelter out totally and they just sell cathode from the SX-EW?

Yes, that's correct. So an SX-EW producer would be leaching copper through the ore directly into a copper liquor that would go directly into the refining process and produce a copper cathode. That would then be sold, just like a smelter would sell, with a premium on top of the copper cathode, over and above the LME. There would be no smelting or refining costs because that would be all built into the actual SX-EW operation.

There are no further questions registered at this time. So I will turn the meeting over to Mr. Phillips.

Okay. Thanks, Elena. Thanks, Mike, for -- that presentation was great. We appreciate that. But at this point, I think we'll do is go on with James, looking at the base metal operations and putting a model together in those. James, over to you.

Yes. Thanks, Fraser, and good morning, everyone. Starting on Slide 3, I'll be covering the modeling of our copper and zinc business units and providing commentary on items to be aware of in our report. We will focus specifically on examples for Highland Valley, Red Dog and Trail, but the methods can be applied more broadly to our other copper assets as well. Jumping to Slide 5. For our copper operations, we report tonnes milled, grades and recoveries as well as resulting production for copper, and we report production for our main byproducts. We do not report on some of our less material byproducts. A reminder that Teck reports production on a contained metal basis, which differs from some of our partners, which can be on a payable basis. Current and 3-year production guidance is provided by operation, but as noted in our recent release, we have suspended guidance given the high degree of uncertainty associated with the current COVID-19 situation. We left this information in just to show the format of the guidance. For production forecasts, we suggest you project the milling rate, grade and recovery. Head grades, and therefore, production will fluctuate depending on mine sequencing and ore availability. Mill throughput can also fluctuate primarily due to ore hardness. Reserve tonnes and grades as well as remaining current mine lives of our operations are reported in our Annual Information Form as well as resources that support ongoing mine life extension and expansion studies at several operations. On Slide 6, to build an annual revenue model, you start with sales. We recommend that you assume our sales are equivalent to production for the full year. We report sales on a contained metal basis, so you'll apply payable terms to get payable metal. You should use your own price forecast for expected metal prices, and our reported revenues are net of any applicable treatment and refining charges, and therefore, you will deduct these. Mike provided information earlier this morning on the details of copper concentrate terms and how to calculate the applicable deductions and charges. We recommend you use a forecast for typical industry terms in your calculations. Let's look at an example on Slide 7. At Highland Valley, we report on copper and molybdenum production and sales. For copper, you would take your forecasted copper sales, apply a payable assumption based on industry terms and convert it to arrive at copper payable pounds. You would then calculate treatment and refining charges, where I have used the combined unit TC/RC as stated by CRU on a per pound basis, but you could also calculate it using standard treatment and refining charges separately per dry metric ton of concentrate and per pound of payable copper. Taking payable copper multiplied by price and deducting their total treatment and refining charges gives you estimated copper revenues in U.S. dollars, which you would then convert to Canadian dollars. Use your own forecast for the Canadian to U.S. dollar exchange rate. That gives us estimated copper revenues. On Slide 8, you would take a similar approach for molybdenum. Using your forecasted molybdenum sales, applying a payable assumption based on industry terms, multiplying by the price and deducting an estimated molybdenum discount. I should note that the molybdenum payable and discount are negotiated terms that have been left blank for this example, and therefore, molybdenum revenues are slightly overstated. The combination of copper and molybdenum revenues would typically be greater than 95% of revenues at Highland Valley. The difference would mainly be due to silver and gold byproducts, which we do not report on a site-by-site basis. Turning to Slide 9. For gross profit, you would take your forecast for revenue and subtract your forecast for total cost of sales. Total cost of sales is comprised of: operating costs; transportation costs; royalty cost, if applicable; and depreciation and amortization. Over to Slide 10. We report the components of total cost of sales for each site in our financials. We also report or you can infer various cost drivers, including tonnes mined, ore milled and copper produced and sold. Operating costs will vary based on tonnes mined and milled, consumable prices, haul distances as well as production and sales volumes and other factors. Presented here are a few different ways you can consider back calculating unit costs to use as projections, but there is, and you should expect to see, variability in these numbers for the reasons I mentioned. On Slide 11. In our quarterly reports, we provide a reconciliation of cost of sales to total and net cash unit costs for the copper business unit, and we provide annual guidance ranges for unit costs. A few things to note on these reconciliations. Firstly, byproduct revenues are inclusive of both byproducts, for example, molybdenum in Highland Valley and Antamina; as well as co-products, for example, zinc in Antamina. And byproduct cost of sales include costs allocated to byproduct production at the sites as well as cost recoveries associated with our streaming transactions, which I will discuss on a coming slide. We back out noncash and onetime charges from the reported cost of sales and smelter processing costs and convert to a unit basis per pound copper sold to arrive at total cash unit costs. We then deduct cash margins for byproducts to arrive at net cash unit costs. On Slide 12, it's the same as the previous slide but converted to a U.S. dollar basis. Moving to Slide 13. We don't report site-by-site unit costs. However, you can use our reported gross profit before depreciation and amortization to calculate a unit margin for each of the operations. Taking the difference between the average copper price and unit margin, you can then back calculate a unit cost measure for each site. However, it's important to note that adjustments for items like inventory write-downs and any labor settlement or strike costs, which are backed out of our reported cash unit costs, will skew the comparison and therefore need to be adjusted for. I have done this on a business unit basis, but it's usually possible to adjust for these items on a site-by-site basis using our disclosures. There may still be some minor differences due to realized versus average pricing as well as rounding. Slide 14 brings us to our first of 2 slides on other considerations for the copper business unit. A reminder that Antamina processes both copper-only and copper-zinc ores. When calculating production, copper grades apply to the total ore milled, while zinc grades apply only to the copper-zinc ore. The amount of copper zinc ore will fluctuate year-over-year, but the ore mix must equal the reserve mix over the mine life, and the grades over the remaining mine life should average out to the reserve grades. Turning to Slide 15. We have 2 precious metal streams where we receive an upfront cash payment for the delivery of future gold and silver production. At Carmen de Andacollo, 100% of gold production is delivered to Royal Gold, for which we receive 15% of the average gold price. At Antamina, 100% of our share of payable silver is delivered to Franco-Nevada, for which we receive 5% of the spot silver price. Because of these transactions, our revenues do not include gold and silver for these sites, and the payment we receive for delivery of gold and silver is netted off of our cost of sales. Our interest in Antamina is also subject to a 1.667% net profits royalty on free cash flow, which gets reported under royalty within cost of sales. That concludes our copper operations modeling. Skipping to Slide 17 and over to zinc. Similar to copper, current and 3-year production guidance is provided by operation. But as noted previously, we have currently suspended guidance and have left this information in to show the format of our normal guidance. We also include guidance for refined zinc production as well as mined and refined lead. For forecasts at Red Dog, you would project the milling rate, grade and recovery. Head grades, and therefore, production, will fluctuate depending on mine sequencing and ore availability. Mill throughput can also fluctuate, which is mainly a result of ore hardness. Reserve tonnes and grades as well as remaining current mine lives of our operations are reported in our Annual Information Form as well as resources that support ongoing mine life extension studies at Red Dog. Turning to Slide 18. To build an annual revenue model, again, you start with your sales forecast. We recommend that you assume our sales are equivalent to production for the full year, but note that Red Dog production guidance can be refined by our sales guidance for the following quarter in each quarterly report. We report sales on a contained metal basis, so you'll apply payable terms to get payable metal. You should use your own price forecast for expected metal prices. And as with copper, our reported revenues are net of any applicable treatment charges, and therefore, we deduct these in the calculation. Let's look at an example on Slide 19. At Red Dog, we report on both zinc and lead. For zinc, you would take your forecast of zinc sales, apply a payable assumption based on industry terms and convert it to arrive at zinc payable pounds. You would then calculate treatment charges based on expected benchmark terms. I have simplified things here and have not included price participation, but Mike covered this in detail earlier. Historically, treatment terms on some contracts were averaged over the preceding few years, which was referred to as bricking. This is no longer the case and realized treatment terms during 2020 would be expected to reflect the combination of 2019 and 2020 terms, in line with our seasonal sales pattern, where concentrate shipped last year but registered as sales in 2020 would generally be under 2019 treatment charge terms. Taking payable zinc multiplied by price and deducting treatment charges gives you zinc revenues in U.S. dollars, which you would then convert to Canadian dollars using your own forecast of the exchange rate. That gives us estimated zinc revenues. Turning to Slide 20. We do the same thing for lead, using industry payable terms and treatment charges for lead concentrates. This gives you estimated lead revenues. Together, presented on the calculated revenue line, zinc and lead revenues would typically be approximately 95% of revenues at Red Dog. The difference would primarily be due to silver byproduct revenues, which we do not report. On Slide 21. For gross profit, you would take your forecast for revenue and subtract your forecast for total cost of sales. Total cost of sales is comprised of: operating costs; transportation costs; royalty cost, which at Red Dog is the NANA royalty; and depreciation and amortization. Turning to Slide 22. We report the components of total cost of sales for each site in our financials. We also report or you can infer various cost drivers. Operating costs will vary based on tonnes mined and milled, consumable prices, haul distances as well as production and sales volumes and other factors. Presented here are a few different ways you can look at back calculating unit costs to use as projections. But again, there is, and you should expect to see, variability in these numbers for the reasons I mentioned. Royalty costs at Red Dog relate to our operating agreement with NANA and are currently 35% of net proceeds. The royalty percent increases by 5% every 5 years to a maximum of 50% in 2032, with the next increase in late 2022. The royalty itself is calculated on an adjusted cash flow basis and is, therefore, volatile year-over-year as well as quarter-over-quarter. For your calculations, you should take revenues, subtract operating costs, transportation costs and capital costs, including capitalized stripping and multiply this by the applicable rate. On an annual basis, this should be fairly close, but because of timing impacts, can still be quite different on a quarterly basis. On Slide 23. Similar to copper, we provide a reconciliation of cost of sales to total and net cash unit cost for the zinc business unit, and we provide annual guidance ranges for these unit costs. There are a few things to note for zinc: firstly, we need to adjust both revenues and costs for Trail intersegment adjustments; secondly, Antamina is reported within our copper business unit, so will not show up here; and lastly, byproducts include both lead and silver. We backed out noncash and onetime charges from the reported cost of sales as well as the NANA royalty to arrive at an adjusted cash cost of sales. On Slide 24, we then put this on a unit sold basis and add smelter processing charges to calculate the total cash unit costs. Deducting cash margins for byproducts gives us net cash unit cost per pound of zinc sold. Slide 25 is the same, but converted to U.S. dollars, which is how we report and provide our zinc unit cost guidance. Turning to Slide 26. At Red Dog, net cash unit costs vary significantly throughout the year, in line with our normal seasonal sales pattern as sales of Red Dog lead, our main byproduct, are typically completed in the third and fourth quarters, driving lower costs during these quarters and higher costs in the first and second quarters. Presented here is an average of sales volumes for the past 5 years, and we recommend you use similar sales profile assumptions in your projections. Slide 27 brings us to Trail operations, which seeks to maximize profitability from treatment charges, free metal and byproduct production and optimizes its feed sources to achieve this. On the modeling side, Trail produces zinc, lead, silver and gold, which are reported and which we pay for in concentrates, and a basket of other products, including specialty metals, chemicals and fertilizers, which we do not report on and which are generally not paid for in concentrates. We will also review different ways of looking at concentrate and operating costs. On Slide 28, production and sales of refined zinc, lead, silver and gold typically account for at least 80% of Trail's revenues. Production of lead, silver and gold vary with availability and concentrate feed sources and as we seek to optimize these feeds. To forecast revenue, you would take your forecasted production volumes, assume sales is equal to this, apply your metal price forecast and would arrive at an estimated revenue from reported metals. A forecast for other revenue should then be added, which relates to our unreported products and pricing premiums. This approach would suggest an other revenue figure of just over $300 million if you averaged it over the past several years. This would get you a total revenue forecast. Turning to Slide 29. Gross profit for Trail differs from our mining operations as it also includes concentrate purchase costs. Concentrate purchase costs are comprised of payable metal terms and applicable freight costs net of any treatment charges on the purchase concentrate. On Slide 30. We also report zinc and lead concentrates treated, so you can estimate concentrate costs using a forecast of concentrate treated and your forecast for industry treatment terms. A reminder that concentrate costs at Trail include intrasegment revenue paid to Red Dog. As a check on concentrate purchase costs, if we look at this as a percentage of revenues from reported metals, as we calculated on Slide 28, we have historically seen a range of 75% to 80%, depending on prevailing metal prices and treatment charges. Trail's operating costs show an upward trend here, which primarily relates to the Waneta Dam sale, which took effect in the second half of 2018. 2019 was the first full year of operations with the contracted power rate following the sale. Presented here are a few different ways you can consider back calculating unit costs to use as projections. There is, and you should expect to see, variability in these numbers. And finally, on Slide 31, Red Dog sells approximately 30% of its zinc production to Trail, which is equivalent to approximately 60% of Trail's zinc feed. Red Dog also sells some lead concentrate to Trail. Payment is based on standard payable terms and treatment charges for the concentrate but on consolidation of our zinc business unit results, zinc and lead revenues at Red Dog are eliminated against concentrate purchase costs at Trail. And with that, I'll turn it over for any questions.

[Operator Instructions]

We certainly have time for questions, and we're running a little ahead of where we thought we might be.

We do now have a question from Carlos De Alba with Morgan Stanley.

Just a quick question. So it was mentioned that the production grades or the mining grades should average over time the reserve grade. But given that reserves are an economic measure and the reserves would move with prices, is this -- how can we kind of reconcile that? Is the price assumed in the reserve calculation move over time or the cut-off grades move over time? How does that affect the calculation that you suggested earlier?

Yes, that's a good question. Obviously, we do update our reserves on an annual basis and report them on an annual basis and update pricing in those reserve calculations accordingly. So that comment was really just general modeling guidance, but you will see those numbers change and move through time.

[Operator Instructions] There are no further questions registered at this time, so we'll again turn the meeting over to Mr. Phillips.

Okay. Thanks, Elena. We are running well ahead of time, which is good. That's fine. But to remind everybody, if you do have questions or want to go over something again, feel free to definitely ask questions. But with that, I guess, where we're at now is steelmaking coal operations. And so Ryan, I'll turn the floor over to you.

Okay. Thank you, Fraser, and good morning, everyone. I'll begin on Slide #3 with our steelmaking coal strategy, followed by about 14 slides that will provide the details on how to model the coal business unit and some additional color on our financial results. The 4 key overarching coal strategic pillars are: safe and sustainable and productive operations; second, the business unit has the potential to produce approximately 27 million tonnes for decades; thirdly, maximizing and sustaining strong cash flow, focusing on our new cost reduction program and targets as well as sustaining our improved productivities; and fourthly, executing on digital transformation and innovation that will deliver exceptional results, which will be focused on and delivered through our RACE21 program, which we've announced previously. So moving on to Slide 4. In 2019, our 6 steelmaking coal operations in Western Canada, you can see on the map there, are geographically concentrated in the Elk Valley, produced 25.7 million tonnes of coal, with sales of 25 -- with sales of 25 million tonnes. The majority of our sales are to the Asia Pacific region, with lesser amounts going primarily to Europe and the Americas. Our long-term production capacity is approximately 27 million tonnes, and we have total proven and probable reserves of 840 million tonnes of steelmaking coal. It is important to point out that we closed Coal Mountain operations in the second quarter of 2019, and we will be moving our Cardinal River operation in Alberta to closure in Q3 of this year. The reduction in production capacity from these closures will be made up by our other Elk Valley mines. I'll go into more detail in a few slides. In December 2019, we announced a long-term rail agreement with CN for shipping of our steelmaking coal from our 4 BC operations between Kamloops and Neptune Terminals and other west coast ports. The agreement runs from April 2021 to December 2026 and will enable us to significantly increase shipment volumes through an expanded Neptune Terminal. West-bound rail service for the mines located in the Elk Valley is currently provided by the Canadian Pacific Railway company, CPR, which we have a 10-year agreement that does expire in 2021. CPR transports a portion of these west-bound shipments from Kamloops, B.C., and then interchanges the trains with the Canadian National Railway company for further transportation to the west coast. CN also provides rail service for Cardinal River mine to the west coast. Looking at our pricing mechanisms that impact the average realized price on Slide 5. This slide compares our sales that are linked to the index versus fixed price sales as well as our sales mix, our product mix and factors impacting our average realized steelmaking coal price. Just wanted to point out effective April 1, 2017, a change in pricing methodology occurred for our quarterly contract sales from a negotiated quarterly benchmark to an index-linked pricing mechanism based on the average of key premium steelmaking coal price assessments. Quarterly price sales represent approximately 40% of Teck sales, with the balance of Teck sales priced at levels reflecting market conditions when sales are actually included. Lower grade semi-soft coals and pulverized coal injection, or PCI, pricing continued to be negotiated on a quarterly benchmark basis. Substantially, all of our revenues from our sales of our coal products were derived from sales to third-party end users, mostly of which are steelmakers. In 2019, sales to Asia accounts for approximately 80% of our annual coal sales volume, which was a record high, and that was mainly due to increased sales volume to areas with the greatest demand growth, such as India, and then reduced sales to Europe due to the impact of steel production curtailments implemented in July and disclosed in Q3 last year. Looking at our sales mix, approximately 40% of our sales are based on the quarterly contract price, which is based on the last 3 months with a 1-month lag, and approximately 60% based on shorter than the quarterly price mechanism, including spot, which is the average monthly price when sales are concluded. Looking at our product mix that impacts our average realized price on Slide 6. Approximately 75% of all coal we produce is high-quality hard coking coal, although the percentages can vary from period to period. We also produce lesser quality semi-hard coking coal, semi-soft coking coal and PCI coal products, which in aggregate accounted for almost 25% of our annual sales volume in 2019. A byproduct of our steelmaking coal production is thermal coal, which accounted for approximately 2% of our total coal sales volume in 2019. On a U.S. dollar basis, historically, over the past 10 years, we have realized an average price of approximately 92% of the quality differential to the quarterly benchmark price. These factors impacting price is reflective of our product mix, timing of our sales, direction and underlying volatility of daily price assessments, spreads between various qualities of steelmaking coal and arbitrage between FOB Australia and CFR China pricing. This now brings us to a simplified annual model for our steelmaking coal operations on Slide 7. You can see a 3-year history of our actual reported numbers in the table throughout the next few slides. Note that the math does not work exactly at every stage as we are presenting a simplified version, but you should get really close. To start here to build an annual revenue model, you start with sales. We recommend that you assume that our sales are equivalent to production for the full year. We provide production guidance for the following year in our Q4 report annually. Note that this can be refined by our sales guidance for the following quarter in each quarterly report. To calculate the average realized price for the year, you would need to start with your own forecast for the average annual benchmark price. The benchmark price, as reported by industry news is a quarterly benchmark, in line with the majority of the contracts, which we are now on a quarterly basis as well. You then multiply your price forecast by your assumption or your average realized price percentage of the benchmark price. Based on our previous slide, we would expect an average of 92% historically. We report in Canadian dollars, so you then convert to Canadian dollars using your own forecast for the Canadian to U.S. dollar exchange rate. Following that, you can then calculate a revenue forecast by multiplying sales by average realized Canadian -- by the average realized price in Canadian dollars. In addition, you will then need to add royalties that are paid to Teck from our joint venture partner, Poscan. I'll go through this calculation on the next slide. So moving on to Slide 8, where I'll walk through the royalty impact from Poscan. Just a little bit of context, on February 11, 2019, we agreed with Poscan to a reopener in the Greenhills joint venture agreement, and that was to increase the royalty paid by Poscan in respect of its 20% share of Greenhills coal production. Teck Coal is the manager and operator of Greenhills and takes 80% of all the coal produced in Greenhills. Poscan takes the remaining 20% and pays a quarterly royalty based on the price achieved for Greenhills coal sales. For context, as noted on the table showing a 3-year historical view, as a result of the new agreement, the royalty we received increased by $74 million from $21 million in 2018 to $95 million in 2019. At current exchange rates, meaning the average FX in 2019 throughout the modeling tables in this deck, a USD 10 per tonne change in the coal price would increase or decrease the royalty by CAD 6 million. The new royalty is calculated quarterly and it amounts to approximately 2% of our total sales, and the agreement remains in effect until December 31, 2022. Looking at Slide 9, a simplified view of how to model unit costs. We provide annual guidance ranges for our unit cost. Typically, we provide ranges for both our coal site cost of sales and coal transportation costs annually. But some of these would be our combined coal costs for total cash unit costs and that -- and is reported in the table in the Steelmaking Coal section of our quarterly press releases. To calculate our total unit cost of sales per IFRS, you'll make an assumption for our depreciation and amortization unit cost and add that to our total cash unit cost. You can refer to our history over the past 3 years of making your assumption. We would recommend taking an average of the last 3 years for depreciation and amortization for modeling purposes. Moving on to Slide 10. We look at our total annual costs. Here, you simply multiply each category unit cost by our sales and add them together. Note that we also have some royalties in coal that relate to Cardinal River and are disclosed in our AIF, but they are relatively a small component. You can see $20 million there in 2019 and $17 million in 2018. In addition, as discussed previously, on my first slide and second slide, Cardinal River will move to closure in 2020, and we recommend removing Cardinal River's royalties post-2020. This brings us to Slide 11 to walk through gross profit. Here, simply you will take your forecast for revenue and total cash costs in the previous slides and calculate the difference for gross profit before depreciation and amortization. Moving on to Slide 12. And what we're talking about on this slide is unit cost of sales, which has been broken out into various buckets to show what makes up our costs in 2019. Transportation costs make up approximately 29% of our unit costs; Cardinal River royalties, 1%; you see depreciation and amortization, 23%; and operating costs at 48%. Our operating costs breakdown in the table on the right-hand side of the slide is split between the key cost drivers that make up close to 50% of our total unit cost. It's important to note, we would expect some of these percentages to change in a high versus low coal price environment. For example, contractors and consultants would decrease or increase based on the nonroutine scope of work required in a typical year. Also repairs and maintenance parts will vary depending on the timing of component change-outs on our mobile fleets as well as our replacement strategy. The largest component in our energy spend is diesel, and diesel costs account for 14% of our operating costs and other energy accounts for 3% as an example, in 2019. Finally, this brings us to our last modeling slide, shown a view of all-in sustaining cash cost view in a typical year. We've got 3 years of history there. You'll simply take total cash cost plus capitalized stripping and our sustaining CapEx to get an all-in sustaining cost for the coal business unit. Our capitalized stripping in 2020 guidance is $350 million to $400 million. We recommend to use this amount for capitalized stripping for modeling purposes. In addition, you will also find our annualized capitalized stripping amount in our Q4 release and guidance. I'll go into more detail regarding our sustaining capital spend on the next slide. So next slide here on Slide 14. I want to spend some time highlighting other considerations for steelmaking coal modeling. The steelmaking coal business unit is starting to see the benefits of the company-wide cost reduction program that we announced in Q3 2019. We expect spending to decrease by approximately $100 million of operating cost reductions through 2020 compared to 2019. Please note that this amount excludes capital reductions. Looking at our coal capital, our long-term run rate for sustaining CapEx is approximately $6 per tonne, and that is to reinvest in our mobile equipment fleet. On average, our 5-year major enhancement spend is expected to be between $145 million to $180 million per year, mostly related to increasing and maintaining our long-term production capacity of 27 million tonnes. The largest investment is related to increasing plant capacity at our Elkview operations, which is our $9 million project; and secondly, the development of Castle at our Fording River operation. We continue to implement the water quality management measures required by the Elk Valley Water Quality Plan, which is an area-based management plan that was approved in the fourth quarter of 2014 by the British Columbia Administrative Environment. That plan establishes short, medium and long-term water quality targets for selenium, nitrate and sulfate to protect the environment and human health as well as a plan to manage calcite formation. Over the next 4 years, from 2021 to 2024, we plan to invest approximately $350 million to $400 million of capital to further increase water treatment capacity to 90 million liters per day by the end of 2024. In addition, during the same period, we plan to spend approximately $85 million in capital on source control and calcite management and up to $90 million on tributary-specific treatments. Capital spending in 2020 on our water treatment plan is expected to be approximately $290 million. The majority of that plan spend relates to the completion of our Fording River active water treatment facility and Elkview Phase 2 Saturated Rock Fill. Capital spend is expected to be similar in 2021 as of 2020 and then significantly decreased through to 2025. Operating costs associated with water treatment were approximately $1.30 per tonne in 2019 and are expected to increase gradually over the long term to about $3 per tonne range as additional active water treatment facilities and SRFs, Saturated Rock Fills, become operational. After 2024, ongoing capital cost for construction of additional treatment facilities are expected to average approximately $2 per tonne. Note that all these details, all water values can be found in our Q4 MD&A. Continuing on to Slide 15. We expect quarterly cost of sales to fluctuate quarterly with higher cost of sales in the second and third quarter typically, when our operations are scheduled to complete major plant maintenance outages. Looking on the slide there for mine life and Teck interest in clean coal. Our Fording River operation at the north end of the Elk Valley is about 20 years -- 29 years of mining at 265.2 million tonnes. We got Elkview operations at 36 years at 256 million tonnes. Our Greenhills operation is about 50 years of mines at 236 million tonnes, and our Line Creek at approximately 15 years at 58.3 million tonnes. And then RACE21 piece, as a note, 2020 targets are not included in our 2020 guidance. Moving on to Slide 16. This is the last slide here, and I just wanted to go through a summary on how the view is setting up for strong long-term cash flows. Our strip ratio is decreasing over the next 4 years. You can see on the chart in the right there, and that is mostly due to our Elkview operation, and our future strip ratio is on par with the historical average. Strategically, we are replacing high-cost tonnes with low-cost tonnes, and that's Cardinal River coming to closure, which will be made up by our Elkview operation expansion in 2020. We are investing, as I mentioned earlier, in RACE21 technology and digital transformation, which will lower our operating costs and increase EBITDA. And lastly here, increasing our Neptune Terminal nameplate capacity to greater than 18.5. This will help lower our operating costs as well as increasing our logistics chain flexibility in the future. This concludes our coal business unit model items. And with that, I'll turn it over to the operator for any questions.

[Operator Instructions] Our first question is from Gordon Lawson with Paradigm Capital.

With the current divergence between Australia and Chinese coal prices, how would you recommend we estimate our benchmark price to use for modeling purposes? And what is Teck currently doing to take advantage of these differences?

In regards to the best estimate, the prices we use for the average realized price, which is 92% of the benchmark price. The difference there is usually made up of the volatility in prices, mostly timing, also product mix, which can be due to our mine sequence changes as well as mine plans, and then any spreads or deltas between the semi-hard coking coal price and hard coking coal price. But for modeling purposes, we'd use 92% of the benchmark.

And Gordon, that premium of the Chinese coals is significant has been. But when they come in buying seaborne coal, if you look at what they're actually paying, it's a lot closer to that seaborne price. So there's not a huge advantage in that from our perspective. The other thing is what are we doing to take advantage of it to the extent we could, it's worth keeping in mind that we're -- last year, we're about 10% or a little bit below of our -- 10% of our revenues were -- our coal sales, excuse me, were into China, and that's going down. So it's come down over the years. Right now -- or more recently, it's been because the sales to India have been rising. So it's essentially a secular kind of change in our distribution. So our exposure to China, in that regard, is getting smaller as it is.

The next question is from Tony Robson with Global Mining Research.

Thanks again to Teck for putting on the workshop. A lot of companies that are harder to model that should be following you guys, I'm afraid don't. So well done to Teck. Two points, please, in terms of the met coal. One, is there any commercially sensitive reason why you can't break out the high-quality hard met coal to the lower quality given semi-hard, semi-soft and PCI is, as you say, 25% of the mix. Would that not help the analyst, a, get better forecast for revenue, and b, maybe overcome the quarterly issues about what discount or whatever that Teck is getting relative to the market? So that's my first question. The second question is depreciation rates. Is there anything better than just using history? I use, for example, unit of production measures. But does unit of production or a straight line or declining balance for depreciation get us better guidance going forward?

I think from the depreciation question, we expect the 3-year average historical would be the best average rate used for depreciation and amortization going forward for modeling purposes. In regard -- Fraser, do you want to jump in on -- in regards to the price?

Yes, right. And Tony, I think the short answer to your question is yes, we do think that the mix -- exact mix of that remaining 25% is commercially sensitive in terms of what our overall product offering is. And I just feel -- because I think, as far as we can tell, anyway, there's nobody that -- none of our competitors that provide that sort of detail and so we don't, and it is largely for commercial reasons. I think the best we can do is -- I think anybody can do is use a percent of benchmark because of the lack of breakout and then watch that list of things that we put on that slide that Ryan talked to, just watch some of those in terms of directionally and try and vary the percentage perhaps quarter-to-quarter if we see big changes. One of the obvious red flags or signals to watch what's going on or think about our realization is when we get big moves in pricing. So for -- back in 2017, with spikes to over $300 a tonne that was in terms of the spot price or the index, we don't sell a lot of volume and nobody does really at those very high prices. And so something like that happens, then we're not going to get an arithmetic average in the quarter, and that's going to, typically, in that case, mean that our percent of the benchmark is lower.

Okay. It would help as much, but we...

But we're -- it is -- we do think it's commercially sensitive in terms of the specific product mixes.

Okay, understood. A follow-up on that. The 75%, 25% mix, perhaps you could talk to what extent that split depends upon market conditions and how much you want to wash the coal to -- relative to the coal quality you're actually mining. So how much leeway is there in that 75%, 25% split, please, on a quarter-to-quarter basis?

It will depend on mine plans and sequence changes and what we're producing in that year.

And I don't it depends on -- Sorry, Tony. I don't think it has -- I don't think it depends so much on washing the coal, as Ryan says, what we're mining in the reserves at the time.

Okay. And just very, very briefly, how much would it change quarter-to-quarter? What's the last couple of years in your history?

Quarter-to-quarter is -- go ahead, Fraser.

Go ahead, Ryan.

I was just going to say from quarter-to-quarter, it's pretty consistent. It's more on the annual plan in the year. The one thing that can change quarter-to-quarter is just a small byproduct we produce as thermal, which is immaterial on our sales.

There are no further questions registered at this time. So I'll turn the meeting back over to Mr. Phillips.

Okay. Thanks, Elena. We remain well ahead of schedule, which is fine, about half an hour now or so. Next up, we'll go on to energy with Rob Sekhon. Rob, over to you.

All right. Good morning, everyone. I'm Rob Sekhon. I'm the Finance Director with the Energy business unit. I'll be walking you through the modeling slides for energy. We also have Glenn Burchnall on the line, who heads up our oil marketing logistics. He'll be able to answer any specific pricing or logistics questions a little later. Turning to Slide 3. You'll see our 2019 energy P&L developed from information in our Q4 disclosures. The model I'll be walking you through is really the netback model, but also you have the netback ties to the P&L first. The main items in revenue include our proprietary revenue netted against crown royalties. We also have some nonproprietary sales from time to time, but the net impact of these sales should be minimal. On the cost of sales side, we have the cost of diluent, operating costs, transportation, and depreciation and amortization. As you can see, our gross profit before depreciation and amortization for 2019 was $144 million. On Slide 4, we have the same $144 million gross profit before depreciation and amortization on the left. On the right, you have the operating netback table, which shows the Fort Hills performance on a unit of bitumen sales basis. The operating netback for 2019 was $11.85 per barrel. And if you multiply this by the number of bitumen barrels sold in the period, you get to the same $144 million gross profit before depreciation and amortization. The important thing to note here, as well as the rest of the presentation is that the type of barrel you use in your calculations is critically important. As you will see, you will need to get familiar with blended barrels, diluent barrels and bitumen barrels. Before getting into the actual model, on Slide 5, you will see a high-level schematic of our operations. This will allow you to visualize what goes into each of the line items in the operating netback. The large blue box at the top is the Fort Hills mine, where the bitumen is produced. Fort Hills is managed through a limited partnership and operated by a Suncor affiliate. Each of the partners takes ownership of their bitumen as it enters the East Tank Farm, which is a partnership between Suncor and local First Nations and operated by Suncor. Teck acquires diluent from the Edmonton area and ships it to the East Tank Farm on the Norlite pipeline. Diluent and the bitumen is then blended at the East Tank Farm facility. The blended bitumen is then shipped to Hardisty on the Wood Buffalo pipeline system. We have a 425,000 barrel tank at Hardisty to manage inventory. Our product is either sold at Hardisty via pipeline or rail loading. We also send approximately 10,000 barrels per day to the U.S. Gulf Coast on the Keystone pipeline. You will notice the color coding. The light blue or teal represents all the items that factor into the bitumen price realized in the operating netback. This will include all our sales net of diluent costs, which includes the cost of the product and the cost of store and transport and blended diluent. The dark purple represents the transportation cost and the netback, while the blue box in the top would be the site operating costs. On Slide 6, you have the actual model. I'll be walking through each of the bars on the following slides, but I did want to highlight a few things here. The left side of the chart represents the revenue piece in U.S. dollars and also converted into Canadian, starting from WTI and taking you to blended bitumen price realized at USD 45.20 per barrel that you see in our disclosures, essentially the price we get for our product. Important to note that the left side is on a blended bitumen basis. The right side is the operating netback in Canadian dollars and on a bitumen basis. The middle diluent bar is probably the trickiest as you can't really take CAD 7.76 shown here and multiply it by any volume to get diluent cost, but I will walk you through a fairly simple way to calculate the cost of diluent a little later. So on Slide 7. To calculate the revenue, you would multiply blended bitumen volumes sold by the blended bitumen price realized. This slide addresses the volumes. You can assume bitumen production equals bitumen sale. However, you will need to convert the bitumen to blended bitumen using the blend ratio. Typically, you need 25% to 30% diluent for every bitumen barrel produced. Another way to look at it is that 1 barrel of blended bitumen contains about 75% to 80% bitumen and 20% to 25% diluent. For 2019, we needed 31% diluent for every barrel of bitumen or 1 barrel of blend contained 77% bitumen and 23% diluent. All volumes can be found on Page 59 of our Q4 disclosure. On Slide 8, we start looking at the individual components of our bitumen price realized at USD 45.20. Remember, this is multiplied -- this multiplied by the volumes and converted to Canadian will give you the revenue for the period. This slide talks to WTI, which is fairly straightforward. Most sales contracts assume the average WTI for the month of delivery. As you know, WTI is publicly available. Slide 9 is important as you need to factor in our sales profile. As you know, we ship approximately 10,000 barrels per day to the U.S. Gulf Coast, which usually attracts better pricing. You can see here, our sales at Hardisty attracted a WCS differential of USD 12.76 per barrel, while our U.S. Gulf Coast sales were closer to WTI at a discount of just under $1 per barrel. Our weighted average differential was $10.02 for the year, which highlights the benefit of our diversified marketing strategy. I'll talk to the specific mechanics of WCS on the next slide. So on Slide 10. As mentioned on the previous slide, you will need to know how the WCS differentials are determined for both our Hardisty and the U.S. Gulf Coast deliveries. For Hardisty deliveries, WCS is a volumetric average determined on the first 9 to 11 trading days in the month prior to deliveries. In the example here, for October deliveries, the WCS was settled based on transactions from September 3 to September 13. Any WCS settlements after the trading period are considered spot. Although small, this could be used for sales after the trading window, including managing any post apportionment sales triggered by the Enbridge mainline. The spot price could be better or worse depending on market conditions at the time. For sales to the U.S. Gulf Coast, the WCS differential is determined or negotiated on each separate transaction. A question, I believe, a lot of you are struggling with: Are WCS prices available publicly? Yes, they are. They are available in the public domain on platforms like Bloomberg and Platts. However, you will need to be careful as these usually quote current price for WCS or WCS differentials and do not take the timing we just discussed into account. Another good source is Oil Sands Magazine. Link is provided on the slide. This will provide current and historical pricing. It allows you to dig into the pricing as well. This is also a good source for all pricing impacting oil sands, including WTI, WCS and condensate or diluent pricing. So on Slide 11, we look at market and quality adjustments. Our FRB product typically sells at a slight discount to WCS. The USD 1.81 would also capture any spot sales, which could have a positive or negative impact, as I mentioned previously. So on the next slide. This summarizes the revenue. Once you account for the FX, you would take the CAD 59.97 and multiply it by the blended bitumen barrels you calculated earlier using the blend ratio. Total proprietary product revenue was CAD 961 million for the year. So on Slide 13, we'll take a look at the diluent or condensate. There are 2 parts here, the cost of the product itself and also the transportation and blending costs associated with the diluent. To calculate the cost of the product, you need the volumes first. Again, you can use the blend ratio we talked about earlier. In 2019, we needed 3.8 million barrels of diluent. The price of the product itself, which is typically referred to X C5+, usually trades fairly close to WTI, but could vary depending on the market. Certainly, it's trading at about a $10 or $11 discount to WTI, which isn't typical. Again, this is available in the public domain. After converting to Canadian dollars, the cost of the product was CAD 266 million in 2019. As for the associated transportation and blending, you will need to make your own assessment based on historical performance. In 2019, this amounted to CAD 56 million for a total of CAD 322 million for diluent. You can assume about 70% of the transportation and blending is fixed. On Slide 14, you'll find the operating netback table. The bitumen price realized is essentially our product revenue less its cost of diluent reported on a bitumen sold basis. So royalties, the key thing to remember is that we are in the pre-payout phase and will remain in this phase for the foreseeable future. We also have a royalty slide in the appendix, but there is quite a bit of information on the Government of Alberta website as well. Moving to post-payout, we'll depend on things like oil prices, capital costs, et cetera. Transportation costs include all pipeline storage and rail loading costs, downstream of the East Tank Farm, including the Keystone pipeline to the U.S. Gulf Coast. Operating costs represent the cost of the Fort Hills -- cost at the Fort Hills site to mine and process a barrel of bitumen. So on Slide 15, it's -- looking at the entire model again. As we've discussed, you can take the CAD 59.97 and multiply it by the blended bitumen barrels sold to calculate the revenue. For diluent, the calculation is found on Slide 13. And finally, the operating netback table on the right to calculate on a bitumen basis in Canadian dollars. So on Slide 16, this is just another way to look at the same chart, but it highlights why the volumes are important, whether it be blended bitumen, diluent or bitumen. So Slide 17, I won't go through these in detail, but it just provides the summary of the key inputs and drivers for modeling Fort Hills. And then finally, on Slide 18, other things to consider when modeling Fort Hills. The Government of Alberta curtailment came into effect January 1, 2019, and will continue through 2020 with the option for the government to end them early. However, you did see our recent press release last week. I mentioned the move to one train operation in 2020, which results in production being lower than the curtailment level. The details of our 2020 production and operating costs can be found in the news release. Fort Hills will go through a major turnaround every 5 years, where the annual production will dip by about 5% to 10% -- sorry, 5% to 8%. The first turnaround is currently scheduled for 2023, but this could change. We've provided you with our long-term guidance already for production, operating cost and capital, and this hasn't changed. As we've disclosed previously, we see some debottlenecking potential at Fort Hills. But as you can imagine, we aren't looking to spend any money in this environment. So with that, we'll open it up for questions for myself or Glenn.

[Operator Instructions] The first question is from Alex Hacking with Citi.

My question is on the WCS differential. Even before the current collapse in oil prices, that WCS differential is quite -- was quite volatile. I guess I'm not so familiar with energy markets. Could you maybe describe what are the fundamental drivers of the WCS differential? And what was causing some of the volatility in 2018 and 2019?

Yes. Actually, it's Glenn Burchnall here, Alex, and thank you very much for that question. The WCS differential, you're absolutely right. It really does move around it. It's quite volatile. It's mainly based on the overall quality of Canadian heavy crude and how that crude competes on a North American market. You have to remember that Canadian crude for the most part is landlocked. We're only reliant on both the Canadian and U.S. markets. Those markets can consort offshore crude, particularly the U.S. Gulf Coast, which is the largest market in the world for crude oil and heavy crude as well. So Canadian crude does compete against other sources into the U.S. Gulf Coast. We did see a particular amount of volatility back in late 2018 and early 2019. And the reason for that was that there was a lot of new production that came on from Canada that was trying to access markets. It was competing with other crude types as well as there was restrictions on the ability to secure pipeline capacity. So ultimately, because of those factors, we saw that differential really widened out. And then when the Alberta government brought in its curtailment program, heavy barrel -- heavy supply was reduced, and we're able to start to access markets again. And overall heavy crude gained and started to really get into high demand mainly because of a loss of production out of Venezuela and Iran. So we saw basically a real snapback of differentials in the first couple of months of 2019. So overall, there -- we do compete, but it is based on the quality and trying to access markets.

And there are no further questions registered at this time. So I'll turn the meeting back over to Mr. Phillips.

Thanks, Elena. Thanks, Rob and Glenn. That was great. So next, we're going to go on to look at the corporate items, both the income statement balance sheet -- income statement, excuse me, and balance sheet. And for that, we're going to go to Crystal Prystai. And Crystal, over to you.

Thanks, Fraser, and good morning, everyone. My presentation is going to focus on income statement items that are presented below gross profit, with the exception of income taxes, which will be covered by Thomas Cheung. I'm also going to do a refresh on how QB2 is presented in our financial statements now that we've had almost a year of reporting QB2 subsequent to the partnering transaction with Sumitomo Metal Mining and Sumitomo Corporation. So turning over to Slide 3, you can see the agenda for my section. We have green boxes around the items that I'll cover in my presentation, as you can see, everything sort of below gross profit. I'm going to provide you with an overview about how you can think about modeling some of these other operating and financing costs, focusing the discussion particularly on other operating income and expense with modeling of settlement pricing adjustments, modeling stock-based compensation to the extent possible. And then I'll take you through the finance expense and the go-forward composition, including QB2 capitalized interest. And as I mentioned at the end, I'll take you through the QB2 presentation of financial information throughout our various financial statements. I would note before I get started, we are reviewing how we present our adjusted earnings calculation, and we may be making some changes to that going forward. On Slide 4, you can see a summary of the last 5 years and the 5-year average of our general and administration, exploration, research and innovation, and asset impairment line items. From a modeling perspective, our general and administration costs include our Vancouver head office and offices that we have in various locations, including Santiago, Calgary, Spokane and Beijing. For G&A cost, a reasonable basis to consider in your model would look something like the 2018 spending, not as low as the average in the last 5 years, but not as high as 2019, particularly in light of our cost reduction program. For exploration costs using this 5-year average, it's relatively consistent with the last 2 years on a reasonable basis for you to be thinking about modeling going forward. Research and innovation costs includes our RACE21 spending that is not capital in nature as it doesn't go on our balance sheet as well as some other research and innovation costs that we have relating to water and our technical services function. 2019 is a good basis for modeling that line going forward, but I would encourage you to look to our RACE21 guidance that we provided at the end of 2019, and that will be updated on a quarterly basis going forward, just to ensure you're picking up that piece correctly. Asset impairments, obviously, not an item that can be modeled and any impairments that we have or reversals of those impairments are included in our adjusted earnings calculation. Moving to Slide 9 (sic) [ Slide 5 ]. You can see there an excerpt of our other operating income and expense -- financial statement notes from our 2019 annual financial statements. This is a financial statement note that we include every quarter. It contains consistent information. So you can see the statement breakdown by line item in our financial statement notes. And of the items noted on the slide, I'm going to speak in more detail to the settlement pricing adjustments and how you can think about those after Mike's presentation. And then on the share-based compensation, I'll provide more detail and take you through lastly the general items. And again, as I mentioned at the beginning, we are reviewing how we present our adjusted earnings calculation. We may make some changes to the calculation going forward, and some of the items that I speak to are already currently adjusted in that calculation. On Slide 6, you will see -- we've provided you with a simplified settlement pricing adjustment model. We present this each quarter for your reference. And I'd encourage you to refer to that when you're thinking about settlement pricing adjustments. But just to add a bit more to Mike's presentation and give you a bit further context as to why we have pricing adjustments and how they're presented. And then I'll take you through some things to think about if you're trying to model and not using our simplified model on the graph there. As Mike mentioned, we do issue a provisional invoice on our concentrate sales. They are priced 2 to 3 months after the sale. Typically, we would recognize revenue on concentrate loading based on a forward price at that date. We don't adjust our revenue subsequent to that date, it's locked in. Any change in the price between that provisional invoice and settlement, those are mark-to-market adjustments. We record them in our other operating income and expense as settlement pricing adjustments. Each quarter, we disclose the volume of unpriced metal concentrates for copper and zinc to enable you to model the price adjustments. You'll need to look at the change in price from the end of the last period to the end of the current period and apply that to those outstanding volumes. There will be some variance that you won't be able to capture as a result of timing of the sales and the settlement and the fluctuation of prices during the period. Typically, that won't be significant unless there is a lot of price volatility like we are seeing in the month of March. But using the information we provided in our disclosure should give you a reasonable basis to model on, and you could confirm that against the simplified model that we are showing on the slide there. On Slide 7, we have also included a simplified model for you to think about when you're trying to model our share-based compensation to try to make it less complicated. It is a difficult thing to model, but we do provide information in our disclosures should you want to go into the detail of doing the calculations yourself. We have 2 types of share-based compensation to consider: stock options and share units. And within share units, we have DSUs, RSUs, PSUs and PDSUs. When you're thinking about stock options, the stock compensation expense is determined per option on grant using a Black Scholes calculation, and it's recognized evenly over the vesting period of 36 months. This is a fairly steady number. It's around $4 million to $5 million a quarter or $15 million to $20 million a year. Modeling of the compensation expense relating to the different types of units is more complicated because there is a vesting component as well as a mark-to-market component based on our share price. In our financial statements, on an annual basis, we do provide you with the vested and unvested units and the vesting period, and you can look to the share price at each period end to estimate should you want to go into that detail. On a quarterly basis, we also provide you with information on grants and grant prices in vesting and Black Scholes information as well in our equity note to our financial statements. Otherwise, the simplified model on the slide will provide you with the reasonable estimates to use if you don't want to go into all of that other detail. Turning to Slide 8. The items listed comprise the remainder of that other operating income and expense line on the income statement. For each line item, there is information on these items by year on the slide as well as a driver to assist you in modeling these items where possible. So I'll take you through each of these items listed there. Starting with environmental costs. This includes the cost of remediation work for our closed operations, and it also includes the change in our decommissioning liabilities for these sites relating to the quarterly change in the discount rate that we used to discount these liabilities. So in periods where you see larger expenses on that slide, these arise as a result of changes in the discount rate and increases in expected costs of ongoing remediation work at our closed operations. We are reviewing this item for a possible adjustment in our adjusted earnings calculation going forward. Care and maintenance costs, as you can see on the slide, they increased in 2019. That was as a result of the closure of Pend Oreille and Coal Mountain during 2019, as Ryan mentioned. With these operations closed for the full year in 2020 as well as the closure of Cardinal River in the third quarter of 2020, we can expect that these costs will increase in 2020. So as we go through the year in 2020, that will be a reasonable estimate for you to use in modeling going forward. But 2019, with some increase is a good proxy for -- going forward rather than using 2018 or years prior to that. Social responsibility and donations, that's generally a steady amount. You can use 2019 as a basis for that. Gains and losses on sale of assets. We adjust for that in our adjusted earnings calculation currently. So there is nothing for you to consider in your models there. Commodity derivatives. This relates to our zinc and lead forward swap program that we have in place and is intended to match the shipping season and seasonality of sales at Red Dog. It also includes an embedded derivative that we have in our road and port contract at Red Dog that's connected to the zinc price. All of these items fluctuate with commodity prices, both lead and zinc, and difficult to model and are -- is something we are considering in our adjusted earnings calculation going forward. Take-or-pay contract costs. Those relate to the QB2 power purchase agreements. You can use 2019 as a proxy for modeling. And then the last 2 lines there, Waneta Dam sale, obviously, that was a one-off item in 2018. And so nothing for you to model there. And the Other line tends to be fairly steady, so you can use an average or 2019 as your basis for modeling. On Slide 9, you can see an excerpt of our finance expense note disclosure from our Q4 2019 financial statements that gives you a breakdown of our finance expense. This note is included in our quarterly financial statements each quarter for your reference. And I'm going to take you through how you can think about this expense line in 3 different buckets: interest on borrowings, general items and capitalized borrowing costs. But before I move into the modeling aspect, I just wanted to note a few things here on the balances just for your reference. Debt interest decreased significantly in 2019 as a result of the reduction in our outstanding notes with the redemption of the 8.5% notes back in June of 2019. You can see the interest in 2019 showing on the advances from Sumitomo Metal Mining and Sumitomo Corporation, referred to as SMM/SC on the slide. These do not eliminate on consolidation, and the interest is recorded there in finance expense. Going forward, the interest on the QB2 project financing will also be included. Interest on lease liabilities also increased in 2019. That was a result of the adoption of the new lease accounting standard under IFRS on June -- I'm sorry, January 1, 2019. There was no restatement required of prior period, so you can see the increase there, and that's the reason for that. I'm now going to go into the various buckets to give you a bit more detail on how you can think about that from a modeling perspective. Starting on Slide 10 with interest on borrowings. Here, we have outlined the driver of each of the categories of interest on borrowings and the source of the information that you need to model these items all available within our financial statements. Starting with debt interest. For your reference, our weighted average interest rate on borrowings was about 5.9% at the end of 2019. That's information that we disclosed again in our financial statement notes. Our debt outstanding is also outlined in our financial statement notes in the debt note. It's broken down by the various debt maturities that we have. So you can look to that for the debt outstanding, multiply that by the rate, and that will give you your debt interest information. The advances from SMM/SC, that is its own line on our balance sheet, so you can look to that for the balance outstanding. And the interest is calculated at LIBOR plus a margin of approximately 3%. And so again, you can check your calculation against the finance expense that's been recognized to see if you're in a reasonable range. Interest on our lease liabilities can be modeled using information in our leases note and our finance expense note again to check your calculations in our financial statement information. Beginning in Q1 of 2020, you will find the details of the QB2 project financing in our debt note, including the amount outstanding. And as noted on the slide, the rate to consider there is LIBOR plus a margin of about 1.6% annually. On Slide 11, you can see the general items included in our finance expense and how you can think about each of these items. For letters of credit, if you were to go back on Slide 9, I should have mentioned this, letters of credit and standby fees decreased significantly in 2019. This was a result of our return to investment-grade credit ratings and the release of a number of the letters of credit we had against our power purchase contracts and again, some transportation services agreement. So 2019 is a better proxy for modeling the letters of credit and standby fees going forward rather than looking at years prior to that when we were not investment grade. Accretion on our DRP liabilities, that's driven by the discount rate on our decommissioning liabilities. It's outlined in our provisions note to the annual financial statements. You can find that information there. And then both interest on our retirement benefit plans and the Other line item within finance expense were not material for you to think about in your modeling. On Slide 12, the last component of our finance expense is an offset to finance expense and relates to our capitalized borrowing costs. If you look to the trend on the bars on the right side of the slide, you can see that through 2016 and 2017 and into the first half of 2018, we had an increase in the amount of borrowing costs capitalized. And that relates to the Fort Hills project construction where we had commercial production on June 1, 2018. And then you can see the drop-off in the third quarter of 2018 and then starting to ramp up from there with the sanctioning of QB2 in the fourth quarter of 2019 -- or sorry, of 2018. And so going forward, the largest component of our capitalized borrowing cost will obviously be on the QB2 spending. And as mentioned in our investor presentation yesterday, there is about USD 3.9 billion remaining on QB2 spending in addition to amounts already capitalized that we use for our borrowing cost calculation. All of the project financing interest and the interest on the advances from SMM and SC are capitalized and then a portion of our other debt interest would also be capitalized. If you're thinking about that calculation, you take the total capitalized amount for QB2 times the weighted average borrowing rate disclosed in our annual financial statements. As I mentioned previously, it's about 5.9% at December 31, 2019. That will give you the QB2 portion. We are currently capitalizing approximately $65 million of borrowing costs on QB2 each quarter, and I expect that will increase by around $10 million a quarter through 2020 for QB2. The net finance expense on our income statement will remain fairly constant as the project financing interest will be fully capitalized. And the borrowing cost rate of capitalization, just want to note, will decrease once we hit commercial production of QB2. This commercial production date is as defined under IFRS. So it is not necessarily first production, but it's when we have the asset available for use as intended by management, which will be disclosed at some point in the future. In addition to QB2 capitalized interest, you should add around CAD 10 million to CAD 15 million of interest capitalization a quarter relating to our other capital projects. Turning over now to Slide 13. You will see an excerpt of our nonoperating income and expense note. The items on this slide and in this line item on our income statement are items that are typically adjusted for in our adjusted earnings calculation, including foreign exchange, gains and losses that we may have on any debt redemption. Previously, we had a revaluation of a debt prepayment option. We no longer have that. That related to our 8.5% notes that have been redeemed. So that should have 0 balance for the foreseeable future. The other line item includes the remeasurement of the ENAMI dividend liability, which I'll speak to when I go through the QB2 section. So those line items, there's not really anything for you to think about from a modeling perspective. We do largely eliminate those in adjusted earnings. So finally, I'm going to do a quick refresh now on how QB2 is presented in our financial statements following the partnering transaction with Sumitomo Metal Mining and Sumitomo Corporation. This presentation has been included in our financial statements since Q1 of 2019 when the transaction closed, so it's a good time now to do a recap after a year of having this presented in our financial statements. So starting on Slide 15. Just as a reminder before I take you through this, we did conclude on the closing of the partnering transaction that we still retained control over QBSA. So that means we continue to consolidate 100% of QB2 in our financial statements. On the balance sheet, we've only shown the liabilities and equity aspects because that's where there are more relevant items for you to think about. But on the asset side of things, what I would note is that we do include 100% of the project spending within our property plant and equipment balance on the balance sheet. And if you are looking for our share of spending and amounts net of project financing, we do provide this disclosure in our MD&A each quarter. So I'd encourage you to look there for that detail. It will be found in our Capital Expenditure section of the MD&A. So turning over back to that same slide, looking at the liabilities and equity. 100% of the project financing will be presented within our debt line. It won't be its own line on the balance sheet, but you will be able to find details in our debt note to the financial statement starting in Q1 of 2020. And again, I'd refer you back to the MD&A disclosure if you're looking for our share of spending in 2020. The advances from SMM and SC, as I mentioned, are shown as their own line. These are not eliminated on consolidation. They are external loans. So they are presented separately and not within our debt balance. The shareholder loans that Teck makes down into QBSA are eliminated. These are considered intercompany loans, and so on consolidation, they don't show up on our external financial statements. SMM and SC's cumulative interest is shown as a component of our equity. It's referred to as attributable to noncontrolling interest. We do have other entities that we do not own 100% of. So this is not exclusively the SMM/SC interest, but the majority of the balance is. And if you're looking for more details, I would refer you to the noncontrolling interest financial statement note in our annual 2019 financial statements. That will give you the split of noncontrolling interest relating to our other entities. On Slide 16, you can see the income statement. We will pick up 100% of the revenues and expenses for QB2 once operational. And Sumitomo and ENAMI share the profit, it will be shown at the bottom of the income statement as flagged in that green square box Turning to Slide 17. You can see the various items on our cash flow statement and how they'll present -- be presented once QB2 is operational and how the funding and spending is reflected at this time. Within operating activities, 100% of the profit will be shown there along with depreciation once we are operational. 100% of project spending is shown in investing activities. Again, I would draw your attention to the MD&A disclosures in our capital expenditures table for a breakdown of what our cash funding is on a net basis. And then lastly, in the financing activities, you can see 100% of advances, and equity contributions and repayments are presented in each separate line item labeled accordingly so you can find that information there. And then the last slide for me, Slide 18. Just to give you a quick refresh of ENAMI's ownership interest in QB. Many of you will be aware of this. This slide was presented in our investor deck when we sanctioned the project back in December of 2018. ENAMI owns a 10% non-funding interest in CMTQB. They do not fund QB2 development costs. They are entitled to Board representation. Until shareholder loans are fully repaid to SMM/SC and Teck, ENAMI is entitled to a minimum dividend based on net income that approximates 2% to 2.5% of free cash flow. Thereafter, ENAMI receives 10% of dividends and free cash flow. We do record the discounted value of this dividend as a liability on our balance sheet. We do remeasure it when there are significant changes and inputs that drive the amount payable, like the copper price and interest rates on the shareholder loan. So that's something that we adjust for, as I mentioned. It's presented in nonoperating income and expense. And that's everything for me. I'm happy to take any questions that you may have.

[Operator Instructions] The first question is from Tony Robson with Global Mining Research.

Sorry, I actually withdrew my question. Star 2, clearly, did not work. Apologies. I have no question.

[Operator Instructions] There are no questions registered at this time. So I will turn the meeting back over to Mr. Phillips.

Okay. Thanks, Elena. And thanks, Crystal. That was great. So that brings us to our final section. And that is income and resource taxes. And for that, we're going to turn to Thomas Cheung. Thomas, over to you.

Good morning, everyone. My pleasure to be here. I'll begin on Slide 3, income and resource taxes. In my presentation today, I will first provide you with a basic overview of the tax regimes, including the effects of tax rates on operating profits in each jurisdiction that we're operating in. After that, I will discuss some main influences on our overall effective tax rate, illustrating how the distribution of operating profits in each jurisdiction contributes to the overall effective tax rate for Teck as a whole. I make a distinction between tax and operating profit versus our net profit. This is because items below operating profit, such as finance and nonoperating expenses, are generally tax-effected at a lower rate, as these items are not deductible in computing resource taxes. Later, I will highlight how in times of lower operating margins relative to our finance and nonoperating expenses, our effective tax rate may skew upwards, possibly to extreme proportions. Teck provides this general guidance for modeling purposes in a normal operating environment and overall annual effective tax rate between 35% and 37% on net profit. This range comprises both income and resource taxes. Please note that while this is our annual guidance, quarterly results may fluctuate outside of this range. I'm now on Slide 4. Canadian tax in 2020. This is a quick snapshot of the corporate tax regime we have in Canada. For our B.C. coal and copper mining business, the all-in effective tax rate is about 36.5%. Note that resource taxes, such as the B.C. Mineral Tax, is deductible for income tax purposes. This is a point that I'll come back to in that it's generally true for all the countries we operate in. Resource taxes paid are deductible for income tax purposes. For our energy and trail operations, the effective tax rate is only the Canadian corporate income tax rate of 27%. This is because there is no resource tax applicable to these operations. This is also the rate applicable to finance and nonoperating expenses as these are not deductible for resource taxes. On capital gains from one-off transactions, such as from the sale of the Waneta Dam in 2018, only half of the gain is taxable. Hence, the effective tax rate is 13.5%. As our Canadian operation is significant relative to the rest of our worldwide operations, you can see how our tax rate guidance generally reflects these rates. We're now on Slide 5, Teck's Canadian income tax pools. As you are aware, we regularly note in our MD&A that we are currently not cash taxable for income tax purposes in Canada. This is the case because of the very significant income tax pools we have as highlighted on this slide. First, we have net operating losses, which can be written off immediately against any income or gains. Next, we have Canadian development expenditures, which can be deducted at 30% per annum on a declining balance basis. Lastly, we also have allowable capital losses that can be used only against capital gains. In 2018, we were able to use capital losses to shelter a significant portion of our capital gain on the sale of Waneta without utilizing too much of our net operating losses or other tax pools. I'm on Slide 6, B.C. Mineral Tax. The BCMT is a 2-tier tax system calculated on a mine-by-mine basis. The first tier is a minimum 2% tax on net current proceeds, which is generally our gross profit. The second tier is a maximum 13% tax on net revenues, which is the tax on the excess of cumulative revenues over cumulative operating and capital costs. In other words, after all costs are recovered. The first tier tax is creditable against second tier tax, and we effectively pay at the higher of the 2 taxes. For all our operating B.C. mines, we're currently in the 13% tax regime. Finance and nonoperating expenses are deductible from B.C. Mineral Tax purposes. And for financial statement or accounting purposes, we treat BCMT as an income tax expense. I'm on Slide 7. The Alberta Oil Sands Royalty, which applies to our Fort Hills mine. Similar to the B.C. Mineral Tax, this is a 2-tier or 2-phase system with a minimum pre-payout royalty on gross revenues and a maximum post-payout royalty on net revenues after recovery of operating capital costs. Fort Hills is currently in the pre-payout phase. The royalty is calculated on a progressive linear scale based on WTI prices in the range noted. Any royalties paid is deductible for income tax purposes. Note, however, that for financial statements or accounting purposes, this royalty is not treated as income tax expense, but is netted against revenues. I'm now on Slide 8, Peruvian tax in 2020. The applicable corporate tax rate on earnings is 29.5%, with a withholding tax on any dividends paid at 5%. Turning to Peruvian resource taxes, there are 2. The Special Mining Tax and the Modified Mining Royalty, which are applicable to our Antamina mine. Both are computed on the same progressive sliding scale on operating margin. On a combined basis, these taxes start at 3% on the first 10% of operating margin, and then the rate increases at every 5% increment of operating margin, topping out at 20.4%. For new projects, such as Zafranal, one can apply for a tax stability agreement with the Peruvian government to stabilize the income tax rate up to 15 years. However, the stabilized rate will be at 2% premium or 31.5%. I'm now on Slide 9, Chilean taxes. There are corporate income taxes at 2 levels. The first level of tax is 27% on taxable income. The second level of tax, also known as withholding tax, is 8% and it's also computed on the same base as the first level tax, making the total taxes on full repatriation of Chilean earnings at 35%. As a footnote, however, for financial statement purposes, we have not been accruing withholding taxes as we expect to be reinvesting our earnings in Chile in the foreseeable future. However, for modeling purposes, you may wish to consider this. The specific mining royalty is calculated on operating margin based on a progressive sliding scale. It starts at 5% for the first 35% in operating margin, and then it increases at each 5% increment, topping out at 14% overall. The special mining royalty is treated as income tax expense for financial statement purposes. I'm on Slide 10, U.S. tax in 2020. For our Red Dog mine, the combined statutory federal and Alaska state corporate income tax rate is 25%. However, due to special income tax deductions, namely the percentage depletion allowance and the FDII deduction, our U.S. income tax burden is much lower. The percentage depletion allowance is calculated at 22.5% -- 22% on adjusted net revenues, capped at 50% of taxable income. The new FDII deduction is an export incentive that effectively reduces the federal tax on portions of our profit from export sales of Red Dog concentrate. The withholding tax rate on dividends paid out in the U.S. is 5%. The Alaska mining license tax is based on 7% of operating profit, and it's also eligible for similar percentage depletion allowance. This mining license tax is deductible for income tax purposes and is treated as income tax expense for financial statement purposes. On Page -- on Slide 11, we have a simplified illustration of U.S. taxes in 2020. As you can see on the right column, the all-in effective tax rate on repatriated U.S. earnings is 22%. I'm now on Slide 12. Here, I am summarizing all the applicable effective statutory tax rates on operating profit in each jurisdiction we operate in. As you can see, we have a wide range of effective tax rates, which depending on our distribution of operating profits across these jurisdictions for any given quarter or year, our overall consolidated effective tax rates can fluctuate. In a normal operating environment, we expect the weighted average annual rate on operating profit to be around 34% to 36%. I'm on Slide 13. Using the effective statutory tax rates from the previous slide, I will now illustrate using these 2 scenarios how our overall consolidated effective tax rate and operating profit is influenced by the distribution of operating profits. You can see that it produces a range of 34% to 36% on a weighted average basis under these 2 scenarios. The way to look at this table is under each column or operations, multiply the statutory tax rate, which is A, by the assumed percentage of operating profit or loss, which is B1. The consolidated column to the far right is the weighted average of all of these separate jurisdictional tax rates. As you can see, depending on the distribution of profits or loss, the overall consolidated effective tax rate can range from 34% to 36% and, sometimes, outside of this range in a particular quarter. I'm now on Slide 14. This slide illustrates the relationship between different operating profit levels, finance and nonoperating expenses, and our overall effective tax rate on net profit. As our heading suggests and also illustrate on the chart to the right, the general rule is that our overall effective tax rate on net profit rises as operating profit falls. Assume that our effective tax rate on operating profit is taxed at 35%. This is the range from the previous slide. In scenario 1, even at high operating margins, finance and nonoperating expenses can skew our rates upward by at least 1%. You can see the effective tax rate flattened out at about 36% as we go from low to higher levels of operating profit. However, at lower operating margins such as in scenario 2 and 3, we see the effective tax rate skew upwards in a very steep manner. Slide 15 is our last slide. Coming back to our general guidance for an annual overall effective tax rate for Teck, here is a chart showing what our actual overall effective tax rate has been for the past 4 years on normalized net profit. After adjusting for nonrecurring or unusual items, such as the one Waneta Dam sale in 2018, which was subject to a lower capital gains tax rate, and the significant impairments in 2019, largely in our energy BU, we see that our overall effective tax rate has been consistently around 36%. This chart also shows what -- for the past 4 years, what our current tax rate has been, ranging from 19% to 33%, and on the average, about 25%. This concludes my presentation today, and I will try to answer any of your questions.

[Operator Instructions] There are no questions registered at this time. So I will turn the meeting back over to Mr. Phillips. I'm sorry, there are no questions registered at this time, Mr. Phillips.

Hello, operator? [Technical Difficulty] I'm hearing you at this time. Can you hear us?

Yes. Now I'm hearing you. I do apologize, I believe there must have been a technical error there as I was trying to speak, but there are no further -- we do actually now have a new question. And it's just been canceled as well. So there are no further questions registered at this time, Mr. Phillips.

Okay. So that does take us through the presentations. And thanks to Thomas for the last one, and to all the presenters. We certainly have time if anybody does have any additional questions. We could pull one more time on any of the subjects because everyone's on, we'd be happy to take them.

[Operator Instructions] And we do have a question from Carlos De Alba from Morgan Stanley.

Great. I just wanted to refer back to Slide 15 of the operating -- of the corporate slides, the corporate income statement and balance sheet items that Crystal presented. So on that slide, there are 2 line items for the Sumitomo, I guess, participation in QB2. One is advances and the other is -- and that is on liabilities. And the other is on equity, the Sumitomo interest, equity interest. How can we differentiate those? So more precisely, are all future contributions from Sumitomo going to be considered as advances? I'm just a little bit confused there.

Sorry, Fraser, do you want to take that one?

No, go ahead.

Okay. I was just going to say that, no, they will be coming via -- sorry, the funding coming through from SMM and SC will be split between loans and equity contributions going forward. I don't have that split off of the top of my head, but we can get it through Fraser's team to circulate back out to you. And so you will be able to see those items split going forward. Fraser, do you have anything else to add?

Carlos, it'll depend on the -- it depends on the thin cap rules in Chile. So about -- ultimately -- we'll check the exact split. But ultimately, you're limited to 75% debt. So you have to put in 25% in equity and 75% in debt in terms of any advances that are put down into the operating entity, either by ourselves or Sumitomo.

Okay. And the -- I guess, you're going to try to maximize up to the 70%. Those will be incurring interest expenses for the -- for QB2 legal entity in the future and being paid back to the equity partners, correct?

Yes. So anything that's advanced -- yes, any advances of Sumitomo and ourselves are essentially interest-bearing loans.

Right, okay. Although your part of -- gets eliminating consolidation. Okay.

The next question is from Brian MacArthur with Raymond James.

I have 2 questions. I guess, my first one is for tax. Just looking at deferred taxes versus current taxes. Is there any way you can give us a better guidance going forward of exactly what you think a good deferred tax rate? I mean I get it. It depends on which country you're in. But is there any more guidance you can just give on that because you show on Slide 16, it does move around a fair bit -- or Slide 15.

Yes. The best we can do is what I've shown you here, showing you what the average has been in the range. It's difficult to really predict the split between current and deferred taxes because these are -- deferred taxes are mainly on timing differences and timing differences can vary from more than 1 year to 10, 20 years. So it's hard for us to give you kind of a composite kind of defined range. But what I can show you on this chart is that we do pay cash taxes for resource tax purposes, and that's mainly what we pay in Canada. And then in the other jurisdictions, we pay cash taxes for income and resource taxes.

Great. My other question goes back a little bit to Tony's question earlier on the call where we talked about 75-25 for the hard coking coal versus PCI and semi. Just going forward, once we get rid of Cardinal River, does that not change as you get more Elkview material coming in?

Yes, we will get a higher realized price for Elkview's hard coking coal versus Cardinal River. But we still recommend, for modeling purposes, to use 92% of the benchmark price.

It's also, Brian, that the split is -- longer-term split based on the 75% based on the reserves. And so you can see that some of the numbers that Ryan, as you already know -- but some of those numbers that Ryan presented in terms of reserve life at Cardinal, you take Cardinal out in terms of relative size, it's not going to change that calculation. Or not much.

The next question is from Tony Robson with Global Mining Research.

In terms of your Canadian income tax pools, the cumulative losses, if you like, some companies refer -- well, it also gets -- they talk about how much of -- they're talking about pretax earnings that would shield, some talk about the actual tax that would save. So the $2.6 billion, could you provide some clarity, please, on where you're talking about, if that's actually the tax that gets saved or the pretax income that shields?

Thank you. That $2.6 billion is the amount of losses, and it's the pretax number.

And there are no further questions registered at this time. So I will turn the meeting back over to Mr. Phillips.

Okay. Thanks, Elena, and thanks again to all the presenters. And thank you to all of you for joining us. Hopefully -- not an ideal format, obviously, but under the circumstances, I hope you found it useful that the presentations will be -- are being recorded and it'll be up in the website, along with the materials. You can refer to them there. And certainly, if you've got questions arising from those or more questions, don't hesitate to get in touch with myself or Lori or Ellen, and we'd be happy to arrange to take you through some of this or answer any specific questions. So with that, thanks again. Stay safe and be healthy, and we'll talk to you all soon. Thanks very much.

Thank you. The conference has now ended. Please disconnect your lines at this time, and we thank you for your participation.