Safety alarms scheduled for today. Okay, thank you very much. As always, can I please draw your attention to the cautionary statement and in your own time, please take a moment to read through it. Leading to the agenda for today, Chris Griffith, our CEO, will present on our ESG journey to safe and responsible mining and the operational performance of the company. Craig Miller, our Finance Director, will then present the financial results, and Chris will then continue and take us through the PGM market review and finish with a look at Anglo Platinum's next phase of value delivery. There'll be time for Q&A at the end of the presentation. With that, I hand you over to Chris.
Thanks, Emma, welcome to Anglo American Platinum's 2019 results presentation. Welcome also to our Chairman, Norman Mbazima and John Vice, members of the board. Good morning to you, ladies and gentlemen. I'm proud to be presenting a record set of results. First and foremost, I'd like to remind you that, and emphasize that safe production continues to underpin the way in which we do work. For the first time in the company's history, we've been able to go a calendar year fatal free at our managed operations. Our record performance, that is benefiting from strong PGM fundamentals and prices, as well as the steady operational performance, has led us to firstly maintain our ESG performance, which continues to receive global recognition.
Robust PGM fundamentals led to a 27% increase in the dollar basket price and a 38% increase in the ZAR basket price. Our ROCE increased to 58%. We've a strong balance sheet, increasing our net cash position to ZAR 17.3 billion. We continue to pay industry leading returns with a total of ZAR 14.2 billion of dividends for 2019. The first section will focus on ESG, our approach to safe and responsible mining. The company's core value starts with our strategic focus to eliminate fatalities. In 2019, as I mentioned in the introductory slide, that we've been able to mine an entire year without a fatality in the company's own managed operations.
Tragically though, we did have one fatality at one of our joint ventures, the independently managed Modikwa JV, where Thomas Maluleke lost his life in a fall of ground accident early in the year at Modikwa. Once again, we'd like to send our sincere condolences to his family, friends, and colleagues. We also continue to work with our joint venture partners, to ensure that we can mine fatality free at all operations. Our safety performance continues to improve, we've seen a reduction in our injury frequency rates with a total recordable case frequency rate now down to 2.5 per million hours worked. That's down 53% since 2016. Whilst we're proud of our safety performances for 2019, we continue to pioneer and implement new technologies, digitalization, and modernization at our operations to further improve safety.
As part of our safety strategy, we also continue to focus on the health and wellness of our employees, with a significant reduction seen in tuberculosis and HIV-related deaths. By providing medication to people infected with HIV and TB, there's been a significant reduction in the TB-related deaths from 63 recorded in 2013 to only three last year. We've also managed to reduce our TB incidence frequency rate by 50% since 2016, to well below the South African national average of 567. All of you may remember a couple of years back, the mining industry was known by being at least twice what the national average was. Anglo American Platinum has also surpassed the UNAIDS commitments of 90-90-90 by 2020, where 97% of our employees know their status, 91% of those that are infected are on antiretroviral therapy treatment, and we have a 90% viral load suppression.
We continue to reduce both our waste disposal and our water usage. We've made considerable progress in the management of hazardous and non-hazardous waste to landfill, improving 83% since 2016, with programs and projects in place to be able to achieve a zero waste to landfill ambition by the end of 2020. We've had no material environmental incidents at any operation since 2013. We've also reduced our potable water intensity by 9% since 2016. Now 44% of all water that comes onto our operations is provided through treated effluent and gray water. In response to the global challenge of climate change, we continue to find improvements in both our energy use intensities and our emission reductions. We have reduced our energy intensity by 10% since 2016. We have reduced our greenhouse gas emissions by 7% over the same period.
We've now set ourselves 2030 targets to improve energy efficiencies and to reduce absolute greenhouse gas emissions by 30%, as I mentioned, by 2030, and that's off a 2016 baseline. A large enabler of this significant reduction is the introduction of hydrogen fuel cell mining trucks. These trucks will utilize green hydrogen generated from solar PV farms and will have zero carbon emissions. We are on track to trial the first hydrogen fuel cell truck at Mogalakwena in the second half of this year. We continue to make a number of social and community investments. We've contributed to infrastructure benefits in communities with our Mapela community water project being completed this year. That enables 70,000 community members access to daily fresh water.
We've donated 270 hectares of land to the Rustenburg municipality and local communities, which includes bulk services that we put onto that land, so that this land can now be used for housing by the local government. We continue to ensure that the communities benefit from our mining. Our procurement of local goods and services increased to ZAR 3.8 billion this year, and that's within a local zone of influence, and that includes ZAR 2.4 billion of goods and services that come from our doorstep communities. Our social and labor plans have resulted in 1,200 jobs created in our local areas. We continue to invest in our local communities with ZAR 392 million on social and labor plans and on corporate social investment in the community.
ZAR 227 million has been paid out in dividends through the community involvement in the chrome plant at Amandelbult, but also our Alchemy community share scheme. As you can see from this slide, as a result of our long-term and focused ESG strategies, we continue to receive global recognition in many areas of ESG. From the slide, you can see that we were awarded the first-place ranking by Sustainalytics out of 55 precious metal peers in the global mining sector. We are one of only 325 companies globally that were included in the Bloomberg Gender-Equality Index, and only one of eight companies in South Africa. We've also received the highest ratings on the ISS quality scores for social and environmental performance. Turning now to the review of our operational performance. The business delivered a steady operational performance. Own mine production was stable, up by 1% year-on-year.
However, it is worth noting that we had record performances again from Mogalakwena and Unki mines, up 4% and 5% respectively. This performance is despite a number of headwinds that occurred in 2019. The first of which obviously was, no surprise, was from the Eskom power disruptions, which resulted in 38,000 ounces of PGM mining production lost. We had maintenance on a cracked mill at Mogalakwena, the unprotected strike at Mototolo, and then also a complete rebuild of the rope shovel at Mogalakwena. Refined production, which excludes the tolling and refining of 4E material that was previously POC, and of course, that material we give back to the third parties. Refining performance increased 11% year-on-year due to the strong processing performance, in particular in the second half of the year.
This was despite the power outages that disrupted the process operations, which led to a 89,000 PGM ounces disruption that'll flow over into 2020. The all-in sustaining costs for the company reduced to $293 per platinum ounce sold. That compares to a platinum price of $861 per ounce. Mogalakwena continues to deliver and has had another record performance, as I mentioned a bit earlier, producing 1.2 million PGM ounces, an increase of 4% year-on-year. This was despite the shutdown for maintenance on the cracked secondary mill at the North Concentrator and a significant rebuild of the rope shovel. Going into 2020, greater mining equipment availability, the purchase of additional trucks, and the mining of ore that has already been exposed will result in greater mining performance for 2020. Despite these headwinds, Mogalakwena had the best EBITDA margin in the portfolio of 56% and a ROCE of 55%.
The mine generated ZAR 9.9 billion of economic free cash flow at a negative all-in sustaining cost of $429 per platinum ounce sold, highlighting just the benefit of the basket of metals that we mine at Mogalakwena. Finally, we completed the purchase of the Kwanda and Central mining blocks. Previously, those were exploration rights that are contiguous to Mogalakwena. These have been subsequently incorporated into the Mogalakwena mining right, effectively doubling the Mogalakwena mining right area. You can see just pictorially what that looks like on slide 65 a bit later. Amandelbult is a mining transition and delivered PGM production of 893,000 PGM ounces, an annual increase of 3%. The mine had a difficult Q1, which affected the first half performance, but I'm very pleased to say that the mine had a much better second half, with production up 12% on the first half.
As both immediately available and immediately stoppable reserves have been increasing, we're now starting to see a much greater production performance from Dishaba Lower. The chrome plant capacity was again expanded through the Merensky concentrator this past year, where we constructed a further two chrome modules, which were commissioned in the third quarter, this will result in an incremental 340,000 tons of chrome produced per annum. Amandelbult's EBITDA margin increased to 30% with a return on capital employed of 49%. The mine generated economic free cash flow of ZAR 3.3 billion at an all-in sustaining cost of $390 per platinum ounce sold. In the case of Amandelbult, particularly benefiting from its high rhodium content. Unki mine, as I mentioned earlier, delivered a record production performance.
PGM production was up by 5% year-on-year to 202,000 PGM ounces, as underground mining efficiencies have improved, as well as concentrator throughput grade, higher grade, and higher recoveries. The mine delivered an EBITDA margin of 35% and a ROCE of 24%. Economic free cash flow of ZAR 1.1 billion, due to their strong performance at all-in sustaining cost of $88 per platinum ounce sold. Again, in the case of Unki, demonstrating too their mix of metals that they produce. Perhaps just a final comment on Unki, that despite the challenging macroeconomic environment in Zimbabwe, Unki continues to deliver. Mototolo had a difficult start to the year, with PGM production impacted by a three-week unprotected strike and the subsequent ramp-up. The mine also had a lower built-up head grade as it transitions through difficult ground areas that contain multiple geological features.
The previous period also had the benefit of material from 2017 being treated at Bokoni, and allocated back to Bokoni in 2018. In a like-for-like comparison, Mototolo's production was 9% down year-on-year. That said, it was very pleasing again for them to see a much better second half performance, and H2 was up 26% on H1. Production at Mototolo will normalize for the whole of 2020, as the mine access is now contiguous ground that we put into the rights from Der Brochen, so to the down-dip side of the mine, but also from the Two Rivers area, as agreement was reached with the Two Rivers partners to be able to allow us to mine through the boundary into a block of ground that Two Rivers can't access.
The mine delivered an EBITDA margin of 43%, a ROCE of 45%, and delivered economic free cash flow of ZAR 1.4 billion at an all-in sustaining cost of $73 per platinum ounce sold. From the charts on the left-hand side, you can see the total PGM ounces for joint ventures and the purchase of concentrate was flat year-on-year, this comprised of joint venture production that was down by 4% and total POC that was up by 1%. We had solid production from Kroondal in 2019, we experienced challenges at Modikwa due to equipment availability, much more importantly, due to the stoppages post the fatality in the early part of the year. From the charts on the right-hand side, you can see that the JV and the POC business continue to generate stable margins.
From the mine volume, we generated a 41% EBITDA margin, while the POC and toll business increased its EBITDA margin to 14%. POC and toll generated economic free cash flow of ZAR 1.1 billion from the operations at an all-in sustaining cost of $860 per platinum ounce sold. Refined production, which shows the charts on the left-hand side, you can see increased by 11% on a like-for-like basis due to strong operational performances, again, as I mentioned, particularly so in the second half. This was despite being impacted by power outages, which led to a temporary buildup of work-in-progress inventory, again, as I mentioned earlier, of 89,000 PGM ounces. That increase in inventory should largely be refined in 2020. Sales volumes, depicted by the graphs on the right-hand side, decreased by 1%, but it increased by 7% platinum volumes and 13% for palladium volumes.
That is, of course, on the back of increased refining production. Total PGM sales were down, however, by that 1%, largely as a result of minor metal sales that reverted to more normalized levels after a year of very strong sales in 2018, where we not only sold all of our production, but we also sold down stock of minor metals that we had. I'll hand you over to Craig, who will take us through the financials.
Thank you, Chris. Good morning, everyone. We are today reporting record financial results for Valterra Platinum, benefiting from higher market prices, continued steady operational performance. In 2019, we've seen EBITDA double to ZAR 30 billion. Headline earnings have increased 145% to ZAR 18.6 billion, or ZAR 70.87 per share. While return on capital is 58%. The company's balance sheet strengthened significantly to the end of the year, with ZAR 17.3 billion net cash, a ZAR 14.4 billion improvement from the end of December 2018. On the back of the strong results, we're able to declare ZAR 14.2 billion in dividends for 2019. Today we announce our best ever EBITDA performance.
EBITDA more than doubled to ZAR 30 billion from ZAR 14.5 billion in 2018, benefiting from the higher US dollar palladium and rhodium prices and the weaker South African rand, contributing ZAR 12.4 billion and ZAR 4.9 billion respectively. This was partially offset by the impact of CPI and higher royalties of ZAR 2.8 billion. Our improved operational performance resulted in higher sales of PGMs of ZAR 0.8 billion and base metals of ZAR 0.2 billion. Had we not been impacted by Eskom and the load shedding, which resulted in lower mined and refined volumes, EBITDA would have been approximately ZAR 1.5 billion higher. The impact on the fiscus was ZAR 500 million in lost taxes. Following the strong EBITDA performance, we have also seen an increase in our EBITDA margins.
Our EBITDA margin improved, supported by strong mine-to-market margins from our own mines of 44% and JV mines of 41%, together with a POC and toll margin of 14%. Our total EBITDA margin increased from 20% at the end of 2018 to 32% at the end of 2019. We're less pleased with our 2019 unit cost performance, although we did experience some operational headwinds. Our final unit cost of ZAR 22,294 is within the updated market guidance we provided in December. Underlying unit costs were up 5% year-on-year, with the benefits of increased production being more than offset by the higher than CPI labor and electricity increases we experienced. The unit cost is up 8% when including the ore stockpile drawdown at Mogalakwena and the impact of load shedding.
Our unit cost per PGM ounce is ZAR 10,189, and the all-in sustaining cost of ZAR 293 per platinum ounce sold against a platinum averaged realized price of $861 per ounce. We continue to focus on optimizing working capital levels, which reduced to another record low. Whilst we have seen higher inventory levels as a result of the impact of the power disruptions, this was more than offset by the increase in the customer prepayment, which increased by ZAR 3.2 billion in the period, reflecting the higher PGM prices as well as the weaker rand. This resulted in trade working capital at the end of December, reducing to ZAR 3.1 billion or three days compared to the ZAR 4.9 billion we saw at the end of 2018. We continue to maintain a disciplined capital allocation approach, focusing on value enhancing rather than volume.
Our 2019 CapEx was ZAR 6 billion within guidance. SIB capital of ZAR 4.8 billion was mainly incurred on maintaining our asset integrity and improving safety, investing in additional mining equipment at Mogalakwena, and progressing our SO2 abatement project at the Polokwane smelter. Project capital was ZAR 1.2 billion, focused on low capital, fast payback breakthrough projects, which included continuing with the Tumela 15 East mechanized section at Amandelbult, the new chrome module plant at Amandelbult as well, and the commencement of the copper debottlenecking project at the base metals refinery.
Capitalized waste is ZAR 2.1 billion. 2020 CapEx will increase to between ZAR 6.8 billion and ZAR 7.5 billion to support our continued investment in breakthrough projects and value delivery. The company achieved its strongest ever cash position, ending the year with net cash of ZAR 17.3 billion, compared to net cash of ZAR 2.9 billion at the end of 2018. This is a ZAR 14.4 billion improvement after paying ZAR 5 billion in dividends this year. You'll recall net debt at the end of 2014 was ZAR 14.6 billion. We've seen a ZAR 32 billion turnaround in the last five years. Even if we exclude the customer prepayment of ZAR 9.4 billion, we would still be in a net cash position of ZAR 7.9 billion.
Most pleasing, the chart on the right, you will see that underlying operational free cash flow increased by more than 200% to just below ZAR 17 billion. Given our confidence in the underlying cash generation of the business, the quality of our assets in our portfolio, and our solid operational performance, I'm pleased to announce that we've declared a second half cash dividend of ZAR 11.2 billion or ZAR 41.60 per share. That's equal 100% of our second half earnings.
This dividend comprises of a base dividend based on our 40% payout ratio of headline earnings of ZAR 16.60 per share, and a special dividend of ZAR 25 per share. This is the first special dividend we've declared since 2001. This brings the total 2019 dividend to ZAR 14.2 billion or ZAR 52.60 per share, a 76% payout of full year headline earnings. Finally, underpinning our strategy, we have a disciplined and value-focused approach to capital allocation, with clear prioritization on maintaining the integrity of the assets, continuing our base dividend of 40% of headline earnings, and ensuring a strong balance sheet.
Discretionary capital is then allocated to between fast payback, high margin, value enhancing projects, growth investments that meet our stringent value criteria, and additional returns to shareholders. As a result of this capital discipline, we've declared the special dividend for the second half of 2019. Thank you. I'll now hand you back to Chris.
Thanks, Craig. I'll take you now through a review of the PGM market. The US dollar palladium price, which is shown in red, gained a huge 48% on average in 2019, setting numerous price records along the way. The rhodium price in light blue gained the most, up 73%. The platinum price shown in dark blue, fell on average by 1% during the year. As a result, again, as we mentioned earlier, the dollar basket price, which is shown in beige, increased by 27% on average. With a 9% weakening of the rand, we saw the rand basket price in gray increasing by 38%. This slide shows the diversified basket of the metals that we produce. Firstly, the relative size of the charts reflects the size of revenue, which increased by 33% in 2019, driven by the increasing basket price to reach ZAR 100 billion.
Platinum accounted for 39% in the previous year. This was with a slightly disappointing price performance last year. Its contribution dropped to ZAR 29 billion. This was massively outweighed by the palladium's contribution to the basket price, which is now 40% of the total, and rhodium, which has increased to 18% of the basket, highlighting both their relative but also their absolute contributions to the basket. Overall, we see in the short and the medium term, demand outlook for the metals and the basket of metals that we mine as positive across all the metals that we produce. On the left-hand side, just a reminder, I know most of you know this inside out. As a reminder, platinum is generally shared fairly equally amongst the automotive market, jewelry, and industrial. Generally, we have investment running between 3% and 5%.
This past year, we saw investment demand exceptionally strong at 1.1 million ounces. Industrial demand was strong, we continue to see potential growth. Jewelry demand remained a little bit mixed, did decline a bit in last year. We see the potential again for that to stabilize and to increase again, largely on the back of the market development that we're doing in China, also the growth that we're seeing in India. In the automotive demand, over the last number of years, we've been seeing very strong declines in diesel, light-duty diesel in particular in Europe. What we saw in 2019 was a much smaller demand drop in absolute vehicle numbers. With the size of vehicles increasing, we saw actually a very limited amount of ounce drop in the automotive market, particularly in Europe.
What we're starting to see is increasing demand in the heavy duty market, particularly because of emission regulations coming out of China and India. In the medium term, so that was all what happened last year. In the medium term, though, I think the potential substitution of platinum back into the gasoline autocats will drive strong platinum demand, due to both automotive demand growth, so absolute amount of vehicles, but also the increasing loadings that are coming from emission standards. Then also what we will see is an increasing demand coming from heavy duty vehicles as well, again, as a result of vehicle numbers increasing and also increased loadings. In the two charts on the right-hand side, in both palladium and rhodium, the demand is dominated by the gasoline autocat sector.
While industrial demand for both of those metals seems to be weakening a little. In our view, that's largely just due to response to high prices. Automotive demand is rising, in particular due to the rising emission legislation, in many countries. Largely driven by China and India. Notwithstanding a slight drop in vehicle numbers last year, the increased amount of ounces, due to what I've mentioned earlier, increase in regulation to try and get lower levels of emissions, is driving the demand for both palladium and rhodium. We see the outlook for palladium and rhodium demand in the short and medium term extremely positive over that time period. It's the fundamentals of demand for both palladium and rhodium is what's been driving the underlying increase in price.
Palladium, if you look at the charts on the top in gray, you'll see that the palladium market's been in deficit for some time. We can see that the current material deficit being sustained again this year. Rhodium has been moving steadily from a surplus in 2015 to a deficit in 2019, with a larger deficit expected again this year. With the price of palladium reaching double that of the platinum price in 2019, the concept of replacing palladium in some of the gasoline catalyst converters with platinum has become increasingly likely to occur. Again, not likely to happen in the very short term. Turning now to how Anglo Platinum will deliver the next phase of value. The next phase of value that we'll drive through this business is focused on four key focus areas.
The 1st is around operational efficiency improvements. That's through achieving and beating world benchmark performances, what we call internally as P101, as well as using modernization and digitalization to drive these efficiencies. The next focus area is around breakthrough opportunities, which enables a step change in production or value through new technology deployment known as FutureSmart Mining. The 3rd focus area is all around projects, firstly around fast payback, high margin value accretive projects. Also projects that have a potential for growth expansion. Lastly, around market development, which continues to be a key strategic initiative for Anglo Plat to drive demand for new applications for PGMs. Over the next few slides, I'll just show you some of the examples of the progress that we're already making on these fronts. Firstly, some examples of operational efficiencies which are currently in execution.
They include double benching at Mogalakwena, which is aimed at reducing the amount of waste tons mined and gaining a larger ore footprint. Rope shovel efficiencies at Mogalakwena, enabling cost reductions through higher equipment utilization. The modernization at Amandelbult, which will lead to both productivity and cost improvements. Some examples of breakthrough opportunities include, firstly, the bulk ore sorting at Mogalakwena. During the course of 2018, we were focused on the lab and doing the technical work, and then in the course of 2019, built a big pilot plant. We're currently trialing now this technology, which aims to improve the grade that you put through the concentrator plant. Effectively, just getting more ounces through an existing plant, with only additional mining being done. The next area is around coarse particle flotation.
During the course of 2019, we were working on the lab scale work and also on the flotation technology. We'll continue this in a full-on pilot plant in the first quarter of 2021. The aim of this is to reject coarse, barren particles much earlier on in the process, which unlocks downstream concentrator capacity and also reduces energy and water requirements. Finally, we're deploying the use of extra low profile mechanized mining equipment at Amandelbult's Tumela 15 East section. This is aimed at using a separate mechanized section that will lead to safer and more efficient mining of the steeper narrow reef ore body. The next two slides will show what we're doing on the project side. Firstly, on this slide, we'll focus on the fast payback project, a fast payback, high margin, value accretive project.
This has been the kind of focus for the company for a number of years now. On this slide, you can see a selection of approved projects. They've been listed on the table, and you can see that the project's focus is on maximizing value, not necessarily volume, with our disciplined capital allocation focused on generating strong returns. You can see that by the IRRs on the right-hand side of that slide. The 2nd area on projects is around growth. We continue the project studies on a number of our value accretive growth options at our high-quality assets. The key project study is the potential expansion of Mogalakwena, and we're on track to deliver the project studies, hopefully for the first half of 2021.
We continue to assess both the construction of a third concentrator with increased open cast mining, or the potential to upgrade existing concentrators, combined with potential underground operations that we currently study. If we choose to do the growth through the third concentrator, that will likely take between 18 and 24 months, and that will result in the expansion of around about 500,000 PGM ounces. We have said that we don't want you to hold that to us, because we're doing these expansion studies now and hope to finish those, as I mentioned, by in 2021. The most progress study that we have at the moment is the replacement of Mototolo into the Der Brochen ground, which is now in feasibility study and hopefully should be delivered this year.
That study is focused on establishing the Der Brochen new shaft, replacing Lebowa shaft, which is now coming to the end of its life of mine. The potential expansion at Mototolo, Der Brochen is also being studied. What we're looking at that expansion is to see what new technologies we can use to help us use the existing plant and get more volume through that. Hopefully that has the potential to increase volume from Mototolo, Der Brochen by about 33%. This is a slide that we showed you in 2018 when we started talking to you about some of the future value that we could see from our base operations. At that time, we said that we thought that we could increase the margin, uplift the margin by another five to eight basis points, and that by 2023.
We did say that at that time, that excluded any impact from price or FX or expansion. That was just from the base business, demonstrating that we could uplift the value of this company. In 2019, that middle bar, we already want to say that we're on that journey. We've already delivered two percentage points margin expansion due to the operational excellence that's already increased run times at concentrators, recovery improvements and cutoff grades, as well as the delivery of fast payback projects, such as the commissioning of the new chrome plant at Amandelbult in the course of last year. The expansion at Mogalakwena and Mototolo, Der Brochen could lead to further growth in the EBITDA margin. Then, of course, on top of that, you can put whatever price and FX you think is likely to occur at that time.
For the final two slides of this section, just talk to a little bit about our strategic priority to grow demand for PGMs. Our market activity, as you know, is undertaken globally. In 2019, we invested ZAR 25 million to jewelry development through the PGI, focusing on creating demand in China and India. In China, as you know, we've seen a reduction in jewelry demand over the last number of years, and PGI is working with the industry to change the model and to change the look and feel to a more branded product in China. We're starting to see the increase already in India over the last number of years. We invested ZAR 5 million in investment demand creation through the WPIC. We're promoting platinum as a viable alternative for investors, and in the last couple of years, that's been spread globally.
In addition to that now, the team are focusing very heavily in China, on retail investors and particularly working through the banks. We spent ZAR 6 million on industrial demand generation last year. That includes our latest investment into Li-ion Battery Technologies, which is aimed at launching the next generation of lithium-ion batteries that use PGMs as part of their mix. We've committed a total of ZAR 100 million over five years into AP Ventures.
Which has been able to attract already another ZAR 130 million of additional investment, ZAR 100 million from the PIC. We've now got three additional companies that have joined this venture capital fund, that have an interest in fuel cells, hydrogen, and clean energy. Last slide of this section talks to the momentum that we're seeing around hydrogen and fuel cells. 2019, we think, was an unprecedented year of growth in the hydrogen economy and fuel cells.
We participate through our investment in AP Ventures, which has six portfolio companies focusing on the hydrogen economy value chain. Industry alliances for hydrogen economy and the Hydrogen Council increased its membership to 81 members this year. We've seen growing commitments of deployment of fuel cell electric vehicles at a national level, at a governmental level, with 30 countries now having committed to having over 10 million vehicles with 10,000 refueling stations over 10 years. This has now been known as the 10-10-10 Project. This has been further backed up by the increasing availability of fuel cell vehicles, with now 45 fuel cell models on the market. The key roadblock to the broad adoption of the hydrogen economy and fuel cells has been the relatively high cost of hydrogen and the unavailability of hydrogen refueling station infrastructure. We're seeing progress on both of those.
The cost of hydrogen as a fuel has reduced by 60% now since 2010. Finally, we're seeing hydrogen refueling stations being constructed globally, and that's been up 50% since 2015. You can see that fuel cells, which use platinum in particular, are now starting to become a current day reality. Turning now to our guidance for 2020 and a summary of 2019. Our 2020 guidance for produced, refined, and sold metal, it's very important to stress that this excludes the toll refined ounces as we return that metal to our third parties. PGM production is expected to be between 4.2 million ounces and 4.6 million ounces. Refined PGM production expected to be between 4.2 and 4.7, and that's the same number for sales. We should see, that extra 100,000 ounces is the material that has flowed over as a result of Eskom disruptions last year.
Capital expenditure, as Craig mentioned, will be between ZAR 6.8 billion and ZAR 7.5 billion, with capital waste stripping at ZAR 2.4 billion-ZAR 2.6 billion. Whilst our cost performance was higher than input inflation in 2019, we believe that we can limit our unit cost increase to ZAR 10,600-ZAR 11,000 per PGM ounce produced in 2020. Of course, there are potential headwinds in 2020, with Eskom load shedding being the most apparent potential impact, both on production and costs, but as well as the as yet unquantifiable effects of the coronavirus. To conclude, Anglo Platinum had a record set of results. We had no fatalities at our managed operations and achieved our best ever safety performance. Our focus on the ESG performance is receiving global recognition. We delivered steady production with record performances from Mogalakwena and Unki.
Robust market fundamentals for PGMs have led to a strong rand basket price, that is predicted to continue. We're in a strong financial position, leading us to be able to increase returns to shareholders, including a special dividend this year, totaling ZAR 14.2 billion of dividends in 2019. We have a strategy in execution for the next phase of value. Lastly, ladies and gents, you would have seen the announcement this morning that said that I'm going to be stepping down as the CEO of Anglo Platinum. I think the company's in good shape, with more momentum to come. I think this is an ideal time for both the new succession of leadership in Anglo Platinum. I think it's also a good time for me to look for new opportunities.
I think it's very important to remember that we have a very strong executive management team in place. We have a very strong operational management team in place. The company's in good hands. I would like to thank my executive team, all the management teams at the operation, and every single Anglo Platinum employee for their contribution to what we as a company have been able to do over the last seven years. I would also like to thank the board for their support. I think it's underestimated in the market how much was needed and how much was done by the board standing behind management, particularly in the early days when we had to take on some very, very difficult things.
I think the collective team effort that I'm absolutely convinced of is going to be part of the way we do business going forward in this company. That'll remain. The team effort all the way from the board to the executive management team to the management team at the operations and every single Anglo Platinum employee, your company is in good hands. Thank you. We're happy, Craig, myself, and Emma. You can give me one props for the day. Craig, Emma, myself, and my executive team there in the front are happy to take any questions that you may have. Thank you.
Chris. You go, no, go.
Yeah. You.
Right. Cool. Morning, Chris. Morning, Craig. Morning, team. It's Chris Nicholson here from RMB Morgan Stanley. Clearly, you've presented a slide which is showing widening market deficits for palladium and rhodium. I think with what prices have done, I think certainly surprised many of us to the upside, would seem to be signaling to the producers very strongly now that the market is short of metal, and potentially signaling that we need to bring new supply to this market.
My question to you is whether you still think that it is the right approach, to focus on some of the more operational efficiency type of things that you're looking to do on investing in the market through jewelry, et cetera, or to accelerate supply into this market. I guess the risk of not doing so potentially could be demand destruction at some point if prices remain here. Interested in your view on that. Thank you.
Thanks, Chris. Chris, a couple of comments. The 1st thing I would say is you need to look at the 3 E basket together. I know you certainly do look at that. With platinum, if you exclude the million ounces that we increased demand from investment last year, platinum is still six, seven, 800,000 ounces in surplus. The first thing before we just go and grow the market, there actually is a solution. That solution, before we go and build new mines, that solution is platinum. I think for now, if you look at that. Clearly, that's not going to stop all the deficits, the very first thing that we need to do, of course, that's the easiest and the most quickest ounces to market, will come actually from the platinum surplus.
I think that's the first thing we all need to remember. There is a solution for the problem that we have for OEMs. That solution is platinum. There's a solution for the deficit in market. That solution is platinum. The time to market is much more easily solved with that issue, and that's metal to market. It comes from platinum. Yes, there does seem to be growing increase in demand, and therefore, it does seem that the market wants us to increase production, and that's why we are continuing our studies. I think the worst possible thing that you would want from us is to rush into a project without having done the diligence around that. Clearly, we can't only invest for these kind of prices. You've got to invest, we would think, in a disciplined way through the cycles.
These prices, if you're only investing for these kind of prices, my guess in the next 10 or 20 years, there's going to be times when you're going to be bitterly disappointed. I think we're doing what we need to do. We're not stopping our market development, and we are not only focused on operational efficiencies. I think we said we're equally focused on four key areas. We absolutely have to drive efficiency improvement from our operations for which we've already spent capital. The best example of that is Mogalakwena. Mogalakwena, we grew from 300,000 platinum ounces to 500,000 platinum ounces without spending any material capital. We still see that there's plenty of that kind of potential left in this business for us to pursue. We're going to do that.
In addition to that, so not in replacement of it, in addition to that, we have a focus on both fast payback projects that add plenty of value to the company. We're also thinking about growth. What we're not going to tell you is that we're going to ditch all of our discipline that we've now built up over the last seven years, and now all of a sudden pile into something that we don't yet know the final answers to. One of the things that may seem like it's taking a bit longer for Mogalakwena to deliver that project is actually because of all the choices that we have. If we're going to take, let's say, ZAR 1 billion-ZAR 1.5 billion of capital, we want to make sure it's going to be in the place that makes the best returns.
One of the other challenges that we have at the moment, well, it's a good challenge, is that we have a whole range of these new developments that are coming through that we want to make part of that future solution so we can get the benefit of all of the work that we've done. I know it's a long answer, but I think, number one, I'll summarize that by saying there are solutions to these deficits, and they are platinum. They will also be solved much faster that way. Secondly, we've got a range of areas that we're going to deliver value from.
Only one of those is expansion and putting extra new capital in. There's a whole range of other things, and it's going to be all of those areas that we drive as the next wave of value for the company. We're not about to ditch all the discipline that we've had and just pile into something without knowing the answers and without going through a rigorous process.
Chris at Anglo Plat. Well done on the very excellent results to you and your team. It's very satisfying to see this coming through. Martin Creamer reminded me that you have a reputation of getting out at the top. It's not the top yet, so why not stay a few more years? I think you've answered that question. Thank you for all the help that you've given us, to you and your team over the years, and especially me, and it's really very much appreciated. I just have a question on the dividend policy going forward. I see it's about 66% payout at the moment, or 60% payout at the moment on earnings. I think you said 50%. That excludes the special dividend, I suppose.
Rene, I'll just reiterate it, just in case. We have a base dividend policy, which is 40% payout of headline earnings. You'll recall that we increased that from 30% last year to 40%. The first payout was in February last year at the 40%. We will continue to maintain that as part of our disciplined capital allocation framework. We'll return excess cash to shareholders, the special dividend of ZAR 25 per share being declared. That takes us up to the full year payout, including the special dividend of 76% of full-year headline earnings.
Good morning. I'm Nkateko Mathonsi from Investec. Congratulations from me as well on a good set of numbers. My question is for Craig as well. If you can give us a bit of color on the customer prepayment of ZAR 9.4 billion. My impression was the agreement was for five years. What are your options with that prepayment? Can you roll it forward, or are you now in a position where you need to pay it back?
In terms of the customer prepayment, yeah, it has increased to ZAR 9.4 billion this year, ZAR 3.2 billion up from the end of December 2018. The customer prepayment has some way still to go. We'll start to repay the prepayment in the middle of 2022, and it will unwind a year thereafter. That's what the contract is at the moment.
Hi, it's Patrick Mann from Bank of America Securities. I just wanted to ask a little bit more around the market. Auto sales have been weak, but yet palladium and rhodium prices have been exceptionally strong. You did allude to it saying there's some technical difficulties around the substitution. Maybe can you just tell us what you're seeing so far year to date in 2020? Are you still getting the sense from your customers that they're looking for metal everywhere? Maybe also just around what's taking so long with substitution, and when would you expect it to occur? I think that's kind of what everybody who's looking at the market is trying to figure out. It might be very helpful for us if you can elaborate. Thanks.
Thanks, Patrick. Look, I think what we started seeing at the Mining Indaba this year was for the first time, actually, fabricators say that they are starting to do some work now. That's different from what we've heard before. I think that really has been the first signal that we've seen that there's starting to be some additional work that is being done on substitution. I don't think we should underestimate the kind of pressure that the OEMs, the car manufacturers, have been under over the last number of years. Frankly, whilst it's a big deal in our lives and we're trying to understand what's going on, in an OEM's life, they've had much bigger problems to deal with.
They've had big pressure from society around climate change as to what they're doing in their fleets to change electrification of their drivetrains, either through battery, electric vehicles, fuel cells, or hybrid. That's been a much, much bigger issue for them, is to try and find out how they're gonna come up with models and solutions around electrification that society find acceptable. There's other two big issues that have been on their agenda, and that is around also post the VW scandal, is around real-world driving tests. Now you'll remember that in the VW scandal, on the roads, people were seeing substantially higher emissions than what was being tested in the lab.
Eventually they were regulated and said, "We won't let you sell vehicles that don't meet either 2.1x ," then reducing to 1.5x , and then 1x what the emissions are that you set in a lab. They've had to find solutions for that. Most of those solutions, I think luckily for us, have been PGM solutions, but that's what the technical folk have been busy with. Then in particular in Europe, where many of these OEMs have their presence, they've had also CO2 targets. If you miss the CO2 targets, then you get $ billions worth of fines. The combination of all of those things meant that actually for OEMs, the majority of their technical folk have been doing that work. The price delta between palladium and platinum really hasn't been a big issue.
For a lot of the time, that was only about $70 that a car would cost you more. You're not about to not sell a vehicle because it costs $70 more. I think one of the important parts is, remember, the tide rises for everyone. Everyone's cars are just costing $70 more. That problem is now $200, and I think your point, Patrick, is it's not just about the prices, it's whether you can get all the metal. I think that's what's really starting to make the OEMs pay attention. For the first time, we're actually hearing that they are doing some work around that. Even if they were to start today, everyone, and say, "Okay, we're gonna go for this now," it still takes about 18 months for that to happen. They've got to come up with their mixtures.
They've got to test those mixtures. They've got to put 100,000 mi of that mixture under the clock. They've got to get regulatory approval. Only once they've got all that can they then change those mixtures into their vehicles. I think, again, a long answer, but saying that OEMs are starting to look at the fabricators can confirm that they're seeing that, but not in a major way, though. I don't think we should say this is all now flat taps. I think in the next 18 months to two years is when you're likely to see this. If the scarcity of metal becomes a much bigger problem, I think you're likely to see some of that timeframe accelerated, not necessarily just the price differential of these metals.
I think the important fact is we mentioned earlier there is a solution to that. We all being very clear to our customers, to the fabricators, "Guys, this is not panic stations." There are solutions, and it comes from the suite of metals that we mine. When we see platinum moving across more into particularly the underpans of gasoline vehicles, I think you're going to see demand more equalized to the kind of ratios that we produce.
Good morning. Leroy Mnguni from HSBC. Mine is a follow-up on Rene's question. You talk about paying out excess cash. You probably have enough in your savings to fund the Mogalakwena expansion, give or take. Does that really mean that from here you pay out all the cash you generate because you've held quite a bit back? I'm just trying to understand how we should look at your dividend going forward. Is there any reason for you to accumulate any more cash than you currently have? My 2nd question is on Twickenham. There's got to be a lot of interest given where prices are at the moment. Is there an update on that disposal? Does it maybe make sense for you to run it in the meantime while you're waiting for a buyer?
Go for the cash.
Okay. Yes. Leroy, when looking at the balance sheet, we have ZAR 17.3 billion of cash. ZAR 9.4 billion of that is in the customer prepayment. That's when you strip that out, we're at ZAR 7.9 billion. We're paying out ZAR 11.2 billion in March. We're paying out all sort of the available cash excluding the customer prepayment. In terms of our capital allocation framework, we will continue to reassess what we've generated in the period, what uses we have for that. As we progress in Mogalakwena expansion, that'll be taken into consideration. Certainly, we'll continue to pay out the 40% base dividend of headline earnings. Any excess cash after that, we'll certainly return that to shareholders as we've demonstrated this year.
Do you want to talk about Twickenham?
In terms of Twickenham, we do have a process underway, in terms of the disposal. There has been some interest, we'll follow that process. We're updating the mine plan at the moment, and expect to progress that, in the second half of the year.
I think perhaps the only other thing to add to what Craig said is that the question is why not just run Twickenham? Remember, Twickenham has still got quite a bit of capital to be spent on it, including a concentrator. In our hands, we've got better alternatives than to invest a fairly substantial amount of money that still needs to go into Twickenham. That's the reason we wouldn't want to just start it up, because number one, you can't just start it up. There's still quite a bit of development and capital needed to be spent on a concentrator.
Martin Creamer from Mining Weekly Online. I think I was in Barry Davison's office when there was a substitution. Ford came through and said they're going to move to palladium. It seemed to be easy. The substitution in the days gone past seemed to be quick. We used to say they're going to move from platinum to palladium to platinum. All of a sudden, it seems a big headache. Has the engine changed? I believe there's extra heating now. Why is the substitution such a pain? Can you give me some idea what's happened with your Li-ion batteries? Are you ever going to get into the Tesla battery with some platinum and palladium and platinum with platinum? My last one, what percentage of this enormous dividend goes to communities? Is there any way communities share in this enormous dividend? Thank you.
Okay, Martin. Thank you. We have a couple of answers for that. Okay, let's start. I think the question about such a pain, there's always been a needed to go through a process. I think we have to be careful not to overreact. We could very easily see tomorrow, OEM come out and saying, because of course they realize what such an announcement would do. It wouldn't be unexpected to see an OEM come out and say, "Oh, we've got a big solution and a substitution for palladium, and it's called platinum." We know that. In a hope to try and temper prices I don't think you should be overly concerned if you see such a headline. The reality is that there is a solution available, there is a potential, and it's called platinum.
The 2nd thing is, I think now with the regulatory focus on real-world driving tests, CO2 emissions, and the VW emissions scandal, that is a big deal for them. In all likelihood, you're much more likely to see them being very, very cautious before they switch applications. I think that's why it won't perhaps be done as quickly as it was in the past. In the past, the emission regulations were actually quite easy. You didn't have to be as fine in your mixture and your solution to get the right solution, to get the right emission legislation. Over years, if you look at that NOx, particulate matter, CO2, if you look at that table how that has shrunk over many years, it's much, much more difficult to achieve that.
The combination of they are much more cautious, it is tougher to find the right solution. The fact is, yes, of course, the vehicle engines have changed. To get the right efficiencies, they have to spend more time coming up with the right technical solutions, and that's why they can't just do it overnight. Don't panic if you get an OEM saying, "Hey, we've got a solution for substitution for palladium and rhodium." In all likelihood, that substitution is going to be platinum. Because the fact is that they're actually really, really good metals. They do the job that they're required to do. Are we going to see us replacing Tesla batteries anytime soon with PGMs in there? I think it's really early to say.
I think the majority of our work will still be focused around fuel cells and the hydrogen economy. The green economy, I think, is a much, much better potential for PGMs to form part of that solution than it is for PGMs to be part of batteries. The fact is that PGMs do such amazing things in so many applications. We don't even know where some of the solutions in the future will come from. Who says that they can't be a really cost-effective solution for batteries? That's why we're pursuing that option. I don't think if you asked us to hedge our bets, I think it would be better to expect a greater uptake in the green economy for platinum to be associated with hydrogen and fuel cells. Lastly, what was the last one?
The community.
Yeah, the community. Yes, we have the dividend allocation of ZAR 41.60 to communities will be ZAR 260 million for H2, and ZAR 157 of that will go to them repaying their loans, and just over ZAR 100 million going to the communities themselves. Absolutely, they are shareholders like anyone else, and when we increase our dividends, they get their share.
Shall we quickly see if there's anybody on the line? Okay.
It's Arnold van Graan from Nedbank. Just a quick one from my side. The customer prepayment, can you pre-settle that? Is there a financial benefit to doing that?
No. Look, we're quite comfortable with the customer prepayment as it is at the moment. There's no intention from our side, at this stage, to pre-settle it.
No interest.
There's no interest.
No interest.
No.
There'll be no benefit.
Correct.
Thanks.
We make a bit of money on the interest, because the interest costs us. Are we done?
Sorry. Is anyone Okay. If there are no more questions, I think that concludes our results presentation for 2019. Thank you, everybody.