News

The latest round of quarterly results from Australian lithium miners provides a very clear picture of the widening cost dispersion across the global lithium resource base. Greenbushes reported quarterly cash production costs of approximately A$418/t, keeping the operation firmly at the low end of the global hard-rock lithium cost curve. Pilgangoora increased quarterly production to a record 267.6 kt while reducing unit FOB operating costs to A$539/t, down 12% quarter on quarter. Wodgina, with all three processing trains operating continuously, lifted quarterly production to 102 kt of SC6 at an FOB cost of around A$674/t. By contrast, Mt Marion’s SC6 FOB cost rose to A$1,029/t as recoveries declined and lower-grade contact-zone ore accounted for a larger share of feed. Bald Hill, which has only recently restarted and remains in ramp-up, reported costs of approximately A$1,498/t SC6. Even within Australian spodumene, unit costs now span a range from roughly A$400/t to A$1,500/t. More importantly, the lowest-cost producers are not standing still. Pilgangoora continues to lower unit costs through scale expansion, operational optimisation and infrastructure investment, while Kathleen Valley plans to use underground ramp-up and expansion to reduce steady-state operating costs to around A$880/t from FY30 onward. Low-cost assets are extending their advantage, mid-cost producers are using capital expenditure and scale to move leftward on the cost curve, while assets in ramp-up or with weaker resource characteristics continue to face materially higher marginal costs. The key question is whether high-cost supply at the right-hand end of the lithium cost curve will automatically exit and re-enter the market as prices change, in the same way as in more traditional commodities. A conventional commodity cost curve implicitly assumes that high-cost capacity responds relatively directly to price. When prices fall below cash costs, high-cost mines reduce output or shut down, tightening supply and providing support to prices. When prices recover above marginal costs, idled capacity restarts and additional supply limits further upside. The right-hand end of the cost curve therefore represents more than simply “high cost”. It also acts as an automatic stabiliser for the market. Some Australian lithium mines still broadly follow this logic. Assets such as Bald Hill and Finniss already have mines, concentrators and infrastructure in place, and can re-enter the supply system through restarts once market conditions improve. These operations are relatively typical examples of price-responsive marginal supply: they exit when economics deteriorate and return when economics recover. The problem is that an increasing share of global lithium supply at the right-hand side of the cost curve is no longer determined by price alone. In Zimbabwe, lithium concentrate may already have been produced, but its ability to reach the international market remains dependent on export quotas, local processing requirements and other policy conditions. In Jiangxi, some lepidolite mines may again be economically viable at prevailing prices, but they still cannot generate stable supply without completing mining-right adjustments, environmental approvals, safety permits or mineral classification changes. In parts of Africa, production at the mine site does not necessarily translate into immediate supply to China because cross-border transport, road conditions, port capacity and shipping schedules determine when concentrate actually reaches the market. The problem is even more pronounced for some greenfield projects. Even where theoretical costs appear competitive, higher prices cannot quickly translate into additional physical supply if financing, offtake, infrastructure and construction conditions have not been fully secured. For these projects, supply is no longer simply a function of price. More accurately, supply is determined by a combination of price, permitting, policy, logistics, financing and project execution. This creates an important but often overlooked consequence. Over the longer term, global lithium competition is indeed shifting from capacity expansion toward cost competition. The structural advantages of low-cost brines and high-quality Australian spodumene will remain intact, and industry concentration is likely to increase further. But greater dispersion along the cost curve does not necessarily mean greater supply elasticity. In fact, a meaningful share of new high-cost capacity is not shut down or restarted solely in response to price. Its availability depends instead on permitting progress, cross-border transport capacity, export policy and project execution. These assets raise the marginal cost of global lithium supply, but they do not move freely in and out of the market as prices change. The right-hand side of the lithium cost curve is therefore being: extended and pushed upward, without necessarily becoming flatter. Conventional logic suggests that the more high-cost projects there are, the more supply can be activated when prices rise, making the supply curve flatter. But if these projects are simultaneously constrained by non-price factors, an important divergence emerges: Nominal capacity is becoming more flexible, while deliverable supply is not. This is why simply aggregating announced global capacity can materially overstate the lithium market’s true supply responsiveness. A project may exist in a capacity model, but that does not mean its tonnage can be converted into spot-market supply within a few months of a price increase. The same issue exists on the downside. In theory, a high-cost mine should shut down quickly once prices fall below cash costs. In practice, the decision is much more complicated. Mine closures themselves are costly. Labour losses, contractor demobilisation, equipment maintenance, inventory management and the eventual cost of restarting all mean that shutting down is not a costless option. As long as cash flow remains manageable, some producers may prefer to absorb temporary losses rather than suspend operations immediately. The exit lag may be even longer for some Chinese mines and strategically important resource projects. This means that the cost curve can tell us who is losing money, but it cannot tell us who will stop producing next month. There is a time lag between cost deterioration and actual supply exit. That lag is one of the most important variables that a static cost curve tends to miss. If higher prices cannot quickly bring new supply into the market, while lower prices cannot quickly force high-cost supply out, then over certain price ranges the short-term lithium supply curve may be much steeper than the static cost curve would suggest. Higher prices do not immediately create large volumes of new supply. Lower prices do not immediately remove large volumes of existing supply. When supply volumes cannot adjust quickly in the short term, the market has to rely more heavily on inventories and demand to rebalance. If demand is stronger than expected and supply cannot respond immediately, inventories fall first, followed by tighter spot conditions and higher prices. If demand is weaker than expected while supply remains steady, inventories build rapidly and prices must fall further to restore balance. This may also be one of the structural reasons why lithium prices exhibit higher long-term volatility than many mature metals. Lithium volatility is often attributed to the relatively small size of the market, rapid demand growth, increasing financialisation and limited inventory transparency. All of these factors matter. But a deeper issue may lie in the supply structure itself: Lithium lacks a sufficiently effective short-term supply buffer. In more traditional commodities, marginal capacity can absorb part of a demand shock. Stronger demand activates higher-cost supply; weaker demand forces higher-cost supply out. In lithium, however, a significant share of marginal supply is constrained by policy, permitting, logistics and construction timelines. When demand changes, supply does not adjust at the same speed. Inventories move first. This creates a clearer transmission mechanism: Demand change → inventory change → changes in tightness or surplus expectations → price adjustment. Supply adjustment often comes last. Lithium price volatility is therefore not driven only by sentiment and expectations. Part of it is embedded in the structure of the supply side itself. The latest Australian quarterly results also suggest that the competitiveness of low-cost assets needs to be understood more broadly. The value of Greenbushes and Pilgangoora is not simply that their costs are low. More importantly, their supply is highly deliverable. Mature operations already have established mines, concentrators, roads, power supply, port access, customers and financing structures. Their advantage lies not only in their position on the cost curve, but also in the certainty of their supply. A genuinely high-quality lithium resource should therefore possess two characteristics: low cost and high deliverability. Going forward, evaluating a lithium project solely on a per-tonne LCE cost basis will be insufficient. Analysts also need to consider supply deliverability and the speed at which production can respond to price changes. A project may look competitive on paper, but if it remains constrained by permitting, infrastructure or policy, its real supply value may be materially lower than its position on the cost curve suggests. This also means that future global lithium resource models should not rely on a one-dimensional cost curve alone. A more complete framework should incorporate three dimensions: cost, probability of delivery and price-response speed. Low-cost, high-deliverability assets form the core of the industry. High-cost but high-deliverability assets can still perform the traditional role of marginal supply. Low-cost but low-deliverability assets may have significant resource value, but they are unlikely to become effective short-term supply. High-cost, low-deliverability assets are the most vulnerable to being marginalised during the next phase of industry consolidation. A truly high-quality resource is not simply “cheap”. It is one that can be converted into physical supply consistently, reliably and on time. This also means that the cost curve still determines long-term winners and losers, but it cannot by itself define the short-term price floor. Prices can temporarily fall below the full cost of a large number of projects because producers do not exit immediately. Conversely, prices can remain above marginal costs for an extended period while new projects still fail to enter the market quickly because of permitting, logistics, financing and construction constraints. A more complete framework for lithium pricing should therefore be: Supply and demand determine direction; costs determine long-term survivability; inventories determine short-term tightness; and supply elasticity determines the magnitude of price volatility. The latest Australian lithium quarterly reports may appear to be simply another set of production and cost disclosures. In reality, they reveal a deeper shift in the competitive structure of the global lithium resource sector. Low-cost assets are continuing to extend their scale and efficiency advantages. Mid-cost assets are using expansion and process improvements to move leftward on the cost curve. High-cost assets, meanwhile, are increasingly constrained by non-price factors. The conclusion is not that the cost curve has become irrelevant. It is that the cost curve needs to be interpreted differently. It still tells us which assets are more competitive and which producers are more likely to survive over the long term. But it is becoming less capable of answering another critical question on its own: How quickly will supply actually respond after prices change? Over the long term, what is truly scarce in lithium is not the resource itself, nor nominal capacity. What is scarce is supply that can enter the market at low cost, reliably and on time, and that can genuinely adjust when prices change. The most important question for future lithium supply analysis is therefore no longer simply: Which tonne has the highest cost? It is: Which tonne will actually disappear when prices fall, and which tonne can genuinely return when prices rise? The difference between those two determines the market’s true supply elasticity — and helps explain why lithium may continue to exhibit structurally higher price volatility than many more mature metals. SMM Lesley Yang yangle@smm.cn

India has approved a Rs 2,814 crore (US$296.52 million) scheme to establish domestic lithium and nickel processing capacity and reduce critical mineral import dependence.

When Bahlil Lahadalia told the Minerba Convex in Jakarta on 2 November that Indonesian coal, minerals and LNG must no longer pass "through transit countries", the instruction read like a shipping rule. For coal and nickel it is not. Cargoes from Kalimantan and Sulawesi already sail directly to China and India; what passes through Singapore and Hong Kong is the contract, the invoice and the trading margin attached to them. The order is therefore an attempt to change who sits on the paper, and its effect depends less on logistics than on how far Jakarta is prepared to reach into the commercial structures behind US$24.48 billion of coal exports and US$16.49 billion of ferroalloy exports in 2025. The trigger was gas, not coal. On 24 October Indonesia cut pipeline gas exports to Singapore by 31 BBTUD, about 8% of the 409.55 BBTUD flowing from Natuna and Sumatra, and Bahlil has since said that part of the 148 LNG cargoes Singapore received from Indonesia were re-traded rather than consumed. Singapore's government rejected the claim, noting that more than 50 Singapore-based companies trade LNG from sources including Indonesia, and that while these traders make the commercial decisions, the cargo usually moves directly from source to destination without passing through Singapore. The direct-export instruction followed within days and carried the dispute from gas into the minerals and coal trade, where Singapore's role is the same: a hub that books the cargo, not a port that handles it. In coal, that role is reportedly substantial. ESDM Director General of Minerals and Coal Tri Winarno estimates that 50 to 60 million tonnes of Indonesian coal is traded through Singapore, while physical shipments go overwhelmingly to India, which took 100.2 million tonnes in 2025, and China, which took 81.8 million tonnes. The gap between where coal lands and where it is sold is precisely what the government regards as leakage, yet its size remains an estimate: neither ESDM nor DSI has published the share of coal sold to Singapore-based buyers. CORE Indonesia economist Yusuf Rendy Manilet makes the same point: the issue is where profit is recorded, not the shipping route, and the government should reconcile Indonesian export data against importing countries' records. The revenue figures in circulation reflect the uncertainty. His own estimate puts the gain at Rp0.5 trillion to Rp3 trillion a year in tax and royalty; another estimate warns that a US$1 per tonne discount demanded by wary buyers would cost the state about US$141 million; President Prabowo Subianto has said DSI identified US$5 billion of potential value from price differences across three commodities in two and a half months. Estimates that differ by an order of magnitude say more about the absence of buyer-side data than about the policy itself. Nickel is more exposed, because its trade structure is more concentrated. Processed nickel exports reached US$40 billion in 2025, and Finance Ministry Decree No. 32/MK/BC/2026 widened HS 7202.60 so that ferronickel, including nickel pig iron (NPI), now falls under DSI. About 96% of NPI export value ends up in China, but a substantial part of the trade is contracted through trading entities in Hong Kong and Singapore. Whether the order bites in nickel therefore turns on a question it does not address: whether Hong Kong, a Chinese jurisdiction where the cargo does not land, counts as a transit country. If it does, the order reaches much of the ferronickel trade; if only Singapore is in scope, its reach narrows considerably. What the order adds to existing machinery is less than it appears. Under Government Regulation No. 24/2026, DSI becomes the sole export intermediary for coal, palm oil and ferroalloys from 1 January 2027, and it already verifies contracts before export, cross-checks related parties and matches final invoices against export proceeds. Sales through offshore buyers are already visible to it. The real change would be to turn a monitored variable into a prohibited one, which sits uneasily with DSI's assurance that it will not take ownership of cargoes or displace existing market participants, and with the President's description of DSI in August as a processing gateway rather than an exporter. As of 5 November, no regulation or ministerial decree had given the instruction legal form. National Economic Council chairman Luhut Binsar Pandjaitan has said Indonesia has no reason to quarrel with Singapore, and the government has stepped back before: in June, the plan for a state export monopoly was scaled back to price monitoring. The one area where the order would be genuinely new lies outside DSI. Nickel matte, MHP, nickel sulphate and stainless steel are not covered by the current regime. If the instruction is written into law and applied to them, it would become the first export-governance tool to reach the battery-materials chain, a larger step than anything it implies for coal or ferronickel. SMM View SMM considers a disruption to physical coal or nickel supply from Indonesia very unlikely: the cargo already moves directly to end markets, and the instruction concerns the buyer of record rather than the route. The exposure lies in contracts. Given DSI's continuity commitments, the President's framing and the de-escalation from DEN, SMM's base case is that the instruction is absorbed into DSI's existing buyer and related-party checks rather than enforced as a ban on third-country buyers. That view would change if ESDM issues a written instrument naming Hong Kong alongside Singapore, or extends the rule to nickel intermediates outside DSI. Until then, the decisions that matter for both markets are DSI's 2027 operating model and the reference-pricing methodology it applies to coal and NPI.