I. Market Sentiment
Nickel has been the best performing LME metal so far this year thanks to the Indonesian government’s surprise announcement to move up a ban on the export of nickel ore to January 1, 2020 – two years earlier than planned. Prices surged by 50 percent in early September, hitting over $18,300/ton (8.3 cents per lb), the highest level in five years. Although prices have since settled down in the range of $16,000 – 16,500/ton, the market is finding support from historically low LME warehouse stocks.
Indonesia exported 20 million tons of Ni laterite ore in 2018, all of which went to China for the production of nickel pig iron (NPI), an important feedstock for the production of stainless steel. Indonesia’s Ni ore exports this year are expected to reach 29 million tons (+45 percent), equivalent to about 325kt of recoverable Ni contained. It is not clear how China will manage to cover this potential shortfall in 2020, as laterite ore shipments from the Philippines, China’s other largest supplier, are not expected to increase further due to depleted resources and lower grades.
Refined Cobalt prices also soared during September to above $18/lb (+50 percent) due to Glencore’s announcement that it plans to put its Mutanda Cu-Co mine in the DRC on care and maintenance at year end; effectively removing 27kt of cobalt from the market in 2020 (approximately 20 percent of total global mined cobalt output). Meanwhile, Glencore has been attempting to get the Congolese government to repeal the super profits tax, and higher royalties which came into effect in 2018 with the new mining law. Lower cobalt prices and high costs were cited as the justification for shutting down. Refined Cobalt prices have since retreated to around $17/lb, reflecting expectations of a more balanced market in 2020. In our view, unusually high inventories of raw material and finished goods stockpiled at the mines and throughout the Chinese supply chain will act to offset the shock of the Mutanda mine shut down, at least through the end of next year.
By comparison, prices of other battery metals such as lithium, manganese, and graphite have disappointed over the past year. Lithium, previously a rising star of Li-ion battery (LIB) supply story, now suffers from weak prices for spodumene, lithium hydroxide and carbonate. While demand will undoubtedly pick up as we approach the much-anticipated EV sales boom in 2022-2023, it remains to be seen whether these metals will be in quite so critical short supply as previously thought. Moreover, recent questions have arisen about whether China can maintain the rate of growth in EV sales, given reduced subsidies from the government. While some are predicting that prices for EV metal materials have already bottomed out, our view is that it is still premature to call the bottom, despite the improvement in cobalt and nickel prices in recent weeks.
II. Battery Raw Materials: Too Much or Too Little?
In recent years, investors seemed to be most worried about the risk of running out of critical battery metals to keep pace with explosive demand for EVs. At yet, some now worry that EV sales could fall short of projections, or even materialize more slowly than expected. A recent report from the China Association of Automobile Manufacturers (CAAM) revealed that sales in China of new energy vehicles (NEVs) and hybrid EVs fell by over 34 percent in September due to changes in government incentives to EV car buyers. Under the new policy, EV subsidies are being incrementally phased out, and eventually eliminated entirely, halving the subsidies for NEVs with a range of 400km or more from RMB50,000 to RMB 25,000, and cutting all subsidies to NEVs with a range of less than 250km. A first decline in sales of this magnitude sent shock waves through the global EV industry, raising fresh concerns about the short-term future of EVs without the heavy support of government subsidies. While this may be only a temporary blip in an otherwise long-term growth story for EVs, the dip in sales is nonetheless a wakeup call for emerging new industry that has only known exponential growth up until now.
III. EV Battery Production: Build It and They Will Come
According to industry data there are at least 75-80 Mega (or ‘Giga’) factories in various stages of planning & construction, or already in operation. Located in 11 different countries, China, alone, is home to 53 of these factories. Total capacity by 2028 could be up to 1,800 GWh, compared with only about 500 Gwh of capacity in operation today. Gearing up to supply such a massive network of megafactories in a relatively short time is obviously a tremendous undertaking. By some estimates, the supply of cobalt to the battery electric vehicle (BEV) industry may need to triple by 2030, up from the current level of 98.1k tons per year. Additionally, battery grade Ni supply, in the form of Ni Sulfate, will need to expand ten times from 110kt currently to over 1 million tons per annum. Where will it all come from?
IV. Supply Risks
In terms of supply risks, there are wide ranging views about whether there will be enough critical battery materials to meet projected demand once the EV boom goes into full swing post 2022; the year EV prices are expected to match equivalent models of conventional internal combustion (ICE) automobiles. The five key battery cathode and anode materials include: Nickel (Ni), lithium (Li), cobalt (Co), manganese (Mn), and natural flake graphite. In the near term (2019-2021), most of the battery metals, with the exception of class 1 nickel, appear to be comfortably supplied, or even in surplus, judging from the weak prices over the past 12 months. . and could arise.
Large volume commodity metals like copper and aluminum are clearly abundant enough that they are not likely to be a limiting factor for batteries. Flake graphite and manganese are also viewed as relatively low risk because they are used widely in so many other industries. In the case of graphite, battery use is only about 2-3% of total consumption and even if it does increase 10 times over the next decade, the abundance of graphite in the earth’s crust is quite high, and there good potential for new production in India, Brazil, and East Africa, where projects are already gaining traction. Synthetic graphite can also be substituted for flake in making battery anode materials, though at a higher cost. Lithium and cobalt are often viewed as the two most uncertain of the five metals, though in the case of lithium, supply is expected to outpace demand as it has a diversity of extraction technologies to choose from, and salt brines are abundant enough that production can be ramped up fairly quickly from existing known reserves.
Cobalt is more problematic due to the concentration of supply in a single country, the Democratic Republic of Congo (DRC), where there is a history of political instability and controversial labor practices. Today, 75percent of the world’s cobalt supply is mined in the DRC, so the risks of supply disruptions caused by government policy, or social instability are quite real. Nevertheless, these risks may tend to be short-term supply gaps, as opposed to long-term structural limits to supply. Moreover, from the geological reserves and mining extraction point of view, resources in the Congo are abundant and relatively low cost to mine. Diversifying sourcing away from the DRC, and reducing the amount of cobalt in the battery chemistries are being actively pursued, though it takes time for exploration projects to be developed, and new technologies to be commercialized.
Looking out to 2025, and beyond, it is highly possible there could be short to medium term imbalances and bottlenecks, should the EV market grow as quickly as forecast. Nonetheless, we do not see any absolute limitations on meeting rising demand for battery materials, with the only exception being class 1 nickel in the short term, and cobalt in the medium to longer term.
V. Nickel Supply/Demand Outlook
China’s total imports of Ni ore this year are expected to hit 70m tons (+16 percent y-o-y), nearly 30 million tons of which will come from Indonesia, 40 million from the Philippines, and the remaining balance from several smaller sources. With the Indonesian Ni ore ban coming into effect January 1, 2020, it is not clear how China will manage to cover its shortfall of an estimated 250kt – 350kt Ni units needed to feed its domestic NPI production, expected to reach 584kt (Ni) in 2019. During the last Indonesian Ni ore ban (2014-2016), the Philippines was able to come to China’s rescue, shipping a record 50 million tpa of ore, though it is unlikely that the Philippines can increase tonnage significantly in 2020 due to resource depletion and lower grades. Other sources of Ni ore, from New Caledonia and Guatemala, are only able to supply an additional 4-5 million tpa.
Historically low LME Ni warehouse stocks (2019 has seen a seven-year low) are another factor keeping the Ni market on edge. LME grade nickel (“Class 1 nickel”) is used in the production of stainless steel, plating and alloys, Li-ion batteries (LIB), and a variety of other applications. Although the battery sector only consumes about 4 percentof global Ni production, it’s growth rate is by far the fastest and it is starting to compete with stainless steel and other industries for a share of limited class 1 nickel.
According to the International Nickel Study Group (INSG), the global nickel market was already in deficit by 84kt this year, and is projected to have a deficit of at least 40kt in 2020. Additionally, the temporary closure of the Ramu Nickel Smelter in Papua New Guinea last month for environmental reasons, and gradual nickel ore mine depletion in the Philippines are other factors raising concerns about short-term supply.
It’s important here to distinguish between class 1 and other forms of nickel, such as NPI, ferro-nickel (known as class 2 nickel) because both class 1 and class 2 nickel can be used in the production of stainless steel, whereas, only class 1 nickel is suitable at this time for producing high purity nickel sulfate NiSO₄(H₂O)₆, used in Li-ion batteries.
The rapid growth of BEVs has been singled out for threatening the stainless steel industry’s share of class 1 nickel, but this is not actually the case. Rather it is the substitution of lower-cost NPI, and Ni scrap that has reduced the use of class 1 nickel stainless steel making process; a trend that began well before the high growth rate in BEVs got started. On average, only about 13-14 percentof class 1 nickel is used to make 300 series stainless steel today; the remaining Ni units come from NPI (31 percent), Fe-Ni (17.5 percent), and stainless scrap (37 percent). This substitution has already freed up an estimated 300kt of Ni for utilization in other industries, such as batteries.
The full impact of the Indonesian Ni ore export ban might not be felt until H2-2020, partly cushioned by the surge of Ni ore shipments during H2-2019, and growing stockpiles or ore at Chinese ports. Even then, we do not expect to see massive shortages of nickel that would adversely impact production of stainless steel or even battery manufacturers. Temporary supply bottlenecks can usually addressed by substituting alternate sources of Ni, and maximizing existing production capacity. Should there be any significant longer term shortages in the future, there are multiple alternatives to choose from in terms of production methods and sources of feedstock. For example, Ni matte can be leached in the NPI process, which is well-proven technology, though higher cost. As demand pushes Ni prices higher, alternative processes such as high pressure acid leach (HPAL) become more economically viable. Quite a few HPAL projects are already under development in Indonesia, Australia, and elsewhere.. HPAL is ideally suited to meeting class 1 Ni demand battery chemicals going forward. Until recently, the relatively high capital cost ($50-60k/t) of building HPAL plants has been an impeded development, though Chinese companies are making progress in reducing the capex down to below $20kt, such as in the case of the Tsingshan HPAL project currently under construction in Morowali.
Indonesia’s NPI output in 2020 is expected to reach 535kt (Ni), up by 50% over 2019, overtaking China’s domestic NPI production for the first time. China’s net shortfall of Ni from the ore export ban (between 250 to 350kt) may to some extent be offset fairly quickly by NPI and Fe-Ni produced in Indonesia by Chinese funded smelters. Indonesia already has 14 operating nickel smelters, plus an additional 27 smelter projects under construction, with a total capacity exceeding more than 1m tons (40 percent of global Ni output). Interestingly, massive Chinese investment in the Indonesia has created a new, low cost source of Fe-Nickel and NPI, but has also created a giant competitor in the regional stainless steel market, with Indonesian stainless capacity already topping 4 million tons per year.
French miner Eramet and its JV partner Tsingshan recently reported that Weda Bay Nickel plant in Indonesia would start operation ahead of schedule in the first half of 2020. Weda Bay, a nickel pig iron (NPI) project, is a 9m ton nickel resource, targeting production of 30kt Ni per annum.
Low Ni prices in recent years forced the closure of a raft of higher cost sulfide and laterite Ni mines, including, Ravensthorpe, Mirabela, BCL, Munali, and Savannah, to name a few. Munali, in Zambia, has already restarted production, producing 6,000 tpy Ni in concentrate, while others are making plans to restart operations during 2020-2021, subject to market conditions. Additionally, a number of existing mines are in the process of restarting or ramping up expansions, including Ambatovy (Madagascar), Goro (New Caledonia): Terrafame (Finland), Ramu (PNG), and Meta Nikel (Turkey). In total, these restarts/expansions have the potential to add between 150-200kt of class 1 Ni, suitable for making Ni Sulfate and LIBs.
VI. Cobalt Supply/Demand Outlook
Global mine output of cobalt was 136kt in 2018, and is expected to rise to 153kt in 2019 (+12.5 percent). Refined production was 114kt in 2018, and 127kt in 2019 (+11.4 percent). Cobalt consumption in 2018 111,300 tons, and is expected to rise to 120,000 tons (+7.7 percent) in 2019.
Cobalt is used in numerous applications, though the lion’s share (54percent) goes into the battery industry, which is also the fastest growing end use (+9.5 percenty-o-y) for cobalt. Cobalt prices have been soft for nearly two years and the market is likely to continue to be in surplus until through next year. 65 percent of global cobalt is refined in China, and Chinese companies now control the largest share of mining operations in the DRC. Cobalt is produced mainly as a by-product, or co-product, of copper and nickel mining. Today, 72 percent of cobalt comes from the copper industry, and 26 percent comes from the nickel industry, though the ratio was the other way around not too long ago. The leftover 2 percent comes as a by-product of platinum group metals (PGM) mining (South Africa), or as a primary mined ore (Morocco), which is rare.
After more than ten continuous years of rapid growth, investment new Cu-Co mines in the DRC is showing signs of slowing down, though there are still a number of projects in the pipeline for completion in the next few years. Currently, not all of the cobalt plants are running at full capacity, and many are sitting on large inventories of hydroxide product, such as Ruashi for example. With Mutanda going on ‘care and maintenance’ this might take some of the pressure off some of the smaller mines. Artisanal miners, who make up for about 10-15% of cobalt mined in the DRC, typically adjust the level of their activity as market prices for cobalt go up and down. At the moment, it is believed there are as many as 120,000 artisanal miners working in the Kolwezi area under questionable health and safety conditions.
In the near-term (2020 – 2022) it’s a fairly safe bet that cobalt supply will continue to be enough to meet demand, or be more or less in balance up to 2023. Fast forward to 2025, if demand for EVs really does double as forecast, then demand could outstrip supply by as much as 20%. Looking out to 2030 and beyond, there is little doubt that cobalt supply will need to grow two to three times larger in order to keep pace with BEV demand. This forecast assumes will not be much new mine development between now and 2025, though it’s possible at least a few of the more than 60 known exploration projects may obtain funding and go into production between 2025 and 2030. Moreover, rising prices would likely serve to reinvigorate investment and incentivize the DRC’s informal mining sector to crank up artisanal mining activities.
Cobalt processing capacities at the mine level have expanded significantly over the past decade. Ten years ago, most of the cobalt exported from the Congo was in the form of mixed Cu/Co high grade ores and concentrates, or alloys (alliage blanc). Today, nearly all the cobalt exported from the Congo is in the form of intermediates such as cobalt hydroxide, and to a lesser degree refined cobalt. All or most of the intermediates go to China for downstream processing. A number of Chinese companies, even those without their own captive mine resource, have set up cobalt hydroxide processing facilities in the DRC, securing raw materials from artisanal miners on the open market.
Swiss trader Trafigura recently joined a consortium who are in the process of financing a $450m Cu-Co processing plant at the Mutoshi mine in the DRC where they will produce 16,000 tpy of cobalt hydroxide. It is reassuring that forward-looking projects like this are going ahead despite the weakness in cobalt prices over the past two years.
Meanwhile, China’s domestic infrastructure for processing downstream battery-ready chemicals has developed at lightning speed, with enormous capacity for producing precursors for lithium cobalt oxide (LCO) and other lithium iron battiers (LIBs) such as cobalt tetroxide, cobalt sulfate, cobalt oxide, among other battery chemicals. There are already several hundred processing plants scattered across a dozen or more provinces in China. Even if cobalt demand from LIB manufacturers was to double by the year 2025, as forecast, there is little risk of the world running out of battery materials processing capacity.
Elsewhere in the world, some companies are making progress in developing cobalt resources in less hostile jurisdictions. Jervois Mining (ASX:JRV) is fast-tracking its Idaho Cobalt operations (ICO) in the U.S., where over $120M dollars has already been spent, defining a measured and indicated (M+I) resource of 3.87M tons of ore, grading Co 0.59%, and Cu 0.85%. ICO was acquired by Jervois through the merger with eCobalt Solutions (TSX: ECS) in August this year. Advanced pre-construction is now underway, with most of the permits already in hand. Additionally, Jervois has concluded an initial drilling program at its highly prospective Bujagali and Kilembe Cu-Co project areas in Uganda (part of the M2 Cobalt asset portfolio). Uganda is said to hold the world’s most prolific source of cobalt, with a similar geology to the DRC, but still largely unexplored. There is an existing cobalt producer in Kilembe.
Supply of Ni and Co, along with the rest of the battery metals appear to be relatively well supplied, at least through the next few years (2020 – 2022). They are also well-positioned to keep pace with peak demand when EV sales begin accelerating post-2023. In the longer-term (2025-2030) the supply/demand picture gets a bit murkier, though there is no question that heavy investment in mining resources will be needed to keep pace with projected EV demand.
The mining industry has always suffered from feast or famine cycles. Periods of high prices and peak demand give rise to over-investment, resulting in periods of over-supply and low prices. For investors, getting the timing and capital costs right is critical. What price level is optimal to stimulate adequate investment in new exploration and development of nickel and cobalt resources? Our guess is around $18/lb for cobalt, and $18,000/ton for nickel (the “magic 18s”).
Battery processing and assembly capacity in China and in other countries appear well positioned to meet rapid acceleration of the EV industry, and, in fact, given the very large number of battery manufacturers and processing plants in China, there could even be over capacity which will require consolidation in the future.