Lithium price crashes again – but this investment bank says 'Keep calm and carry on' buying ASX lithium stocks

Fifteen months ago, the lithium trade was in tatters. The benchmark lithium carbonate contract traded on China's Guangzhou Futures Exchange (GFEX) had bottomed at RMB64,320/t, while Australian 6% spodumene concentrate ("SC6") had touched US$585/t three weeks earlier – for over two years the ASX lithium sector was a place where capital went to die.

Then the market turned – hard. Carbonate rocketed 227% and spodumene an incredible 403%. In the space of a few months, every producer on the ASX was making money again. The trigger? Anti-involution measures taken by Beijing to halt a race to the bottom in prices due to a glut of domestic supply.

Today, investors in the sector are likely having flashbacks because prices are tumbling again, with SC6 plunging 6% on Monday. It's hard to believe that lithium's up-cycle – one that bulls were betting had years to play out – is over in less than 12 months.

There are two things worth knowing before you decide. The first is that both of the latest shocks to hit the market are about information, not tonnes – nothing that's happened in the last fortnight has removed a kilogram of supply or a kilogram of demand from the physical market. Secondly, the most important lithium price for local producers like PLS Group (PLS), IGO (IGO), Liontown Resources (LTR) and Elevra Lithium (ELV) isn't the one that's crashed.

Major investment bank Macquarie believes the market has misread the latest signals. China is simply flexing its anti-involution muscles again, and the longer-term impact for the lithium price could well be positive. In this article, I'll break down the latest developments and the price action to help investors understand whether it’s time to panic or to buy the dip.

Two shocks, no tonnes

The first shock landed on Thursday 3 September, and it wasn't policy at all. A major Chinese lithium consultancy expanded the scope of its weekly inventory survey, which mechanically lifted the reported level of visible lithium carbonate stocks. To a market watching a single number, it looked like inventory had suddenly ballooned. Macquarie's read is unambiguous: the reporting methodology changed, the physical market did not. Carbonate fell 3.2% on 3 September and a further 5.3% the next day, the second of those a sharp afternoon sell-off (given the lithium market remains substantially opaque and China-centric, it’s common for news to disseminate slowly, and for the market to react over more than one session).

The second shock was the one that broke the market on Friday, and it was buried in a story that wasn't about lithium. At 11.56pm Beijing time on Thursday 10 September, Caixin reported that China's Ministry of Industry and Information Technology (MIIT), working with the National Development and Reform Commission and the State Administration for Market Regulation, had frozen approvals for new power and battery energy-storage system (BESS) projects pending a review that concludes at year end.

The detail that mattered was the escalation. Earlier reports, carried by Cailianshe on 5 September and syndicated through Reuters on the 7th, had described a pause covering energy-storage batteries only. Caixin's version extended it to power batteries, named the ministries involved, and revealed the freeze had been running quietly since mid-May. Bloomberg picked it up on Friday morning – and it seems that’s the “news” that sparked the broader market reaction. Carbonate opened at RMB139,980/t and went straight through the floor, trading as much as 9.4% lower at RMB128,080/t intrasession. It recovered to RMB132,340/t by Monday's close – 37% below its 13 May peak.

Latest lithium demand and supply factors

Beyond the inventory accounting shift and the policy freeze, let's catch up on the latest demand and supply factors impacting the lithium market.

The demand data underneath Beijing's move on BESS is genuinely soft: China commissioned 21.8GW – 58.6GWh – of new-type energy storage in the first half of 2026, down 18% and 16% respectively from a year earlier, with the number of new projects roughly halved. It’s one of the key demand engines of the entire 2026 lithium bull market, and it is decelerating.

And there's a structural reason for it: In February 2025 the NDRC and the National Energy Administration scrapped the rule requiring renewable projects to be paired with storage, replacing a mandated pipeline with market-based demand. Several provinces have introduced capacity compensation to fill the gap, but a storage project earns its keep on the gap between what it pays for power when it charges up and what it gets when it sells that power back – and that gap has been shrinking. The thinner it gets, the harder it is to justify building the next one.

Nor is the supply side helping: Chile shipped 24,100 tonnes of lithium carbonate in August, up 42.5% on a year earlier – and while the volume heading to China actually slipped 5.25% month-on-month, Chile's lithium sulphate exports to China more than doubled over the same month. Feedstock arriving in a different form.

More material is arriving, but not all of it is durable: spodumene plants have come back from maintenance and raw material is arriving at ports. Zimbabwe is exporting into a deadline (the concentrate ban still starts in January 2027), but producers there are now lobbying for an extension. Though some of the concentrate arriving from there today looks like stock being cleared ahead of the cliff rather than durable new supply.

Jianxiawo is still producing nothing: back in August the hold-up at CATL's massive swing producer was a safety permit, with a September or October restart expected. The permit has since come through, but the mine is now waiting on environmental approval and the base case has slipped to the December quarter. Forecasts for what it contributes this year have roughly halved, to somewhere near 55,000 tonnes of lithium carbonate equivalent – on its own, enough to absorb most of the surplus the market had pencilled in for 2026.

Is the medicine worse than the disease?

Here's what the sellers appear to have forgotten. The bull market they're busy unwinding was built by exactly this kind of intervention.

Anti-involution – Beijing's campaign against the self-destructive price wars that have hollowed out one Chinese industry after another – is what turned lithium in the first place. Through the second half of 2025 it showed up as mining licence reviews across Jiangxi, environmental approvals pulled at CATL's Jianxiawo operation, and a blunt message to producers that volume at any price was finished. Carbonate went from RMB64,320/t to RMB210,580/t. In July this year, Beijing widened the campaign again with a consumption tax on lithium-ion batteries and solar cells.

The battery capacity freeze is the next instalment, and the numbers explain why it was needed. Around 100 new battery projects were signed between January and July carrying ¥446.9bn of disclosed investment, and the capacity behind them – roughly 2,608.5 GWh – is about 1.5 times everything China actually produced in 2025. The ten largest manufacturers account for some 70% of those plans. Six of them, CATL and BYD among them, have been ordered to file three-year strategic plans, and Beijing is building an early-warning system to track utilisation across the industry.

Morgan Stanley, reported by Bloomberg, expects the crackdown to reward whoever is already running their plants hard – CATL managed 95% utilisation in the first half of 2026 against an industry average of 65%. Licence reviews, consumption tax, capacity freeze: it's the same instrument each time, pointed at whichever part of the chain is destroying value fastest.

The 2025 measures restricted lithium supply, which is unambiguously good for the lithium price. This one restricts battery manufacturing capacity, which sits on lithium's demand side – and that is precisely why markets have sold off on it. But consider that freezing approvals for new factories removes no cells from production and no storage projects from the grid. Approved and under-construction plants proceed, consumer and raw-material projects are exempt, and not a kilogram of 2026 or 2027 lithium demand disappears. What has been removed is a wave of future overbuilding that would have driven cell prices, and eventually cathode margins, into the ground.

The takeaway: Lithium's customer gets healthier, not smaller.

Macquarie: ‘Keep calm and carry on’

Against that, Macquarie's note released yesterday, 14 September – titled Keep calm and carry on – argues the market has drawn the wrong conclusion from both shocks. The broker retains Outperform ratings on each of IGO, PLS, LTR, and ELV. Its arguments run as follows.

The volatility is a futures phenomenon, not a physical one: GFEX carbonate fell roughly 16% between 31 August and 11 September while spodumene was holding above US$2,000/t. The two prices moved in different worlds, and only one of them is where ASX earnings come from. (Note: Macquarie’s "holding above US$2,000/t" lasted until the close of trade Monday).

The early-September crash was a measurement change: the expansion of the inventory survey lifted reported visible stocks without altering the underlying balance. Macquarie doesn't believe it changes the market fundamentals.

The capacity controls have been misread: China runs a two-speed energy storage market, with tier-one cell manufacturers operating near full utilisation while many tier-two and tier-three players sit below 50%. Macquarie's channel checks point to regulatory intervention aimed at curbing further expansion of low-quality, inefficient capacity rather than restricting end-market storage deployments – and it suspects the policy intent has been lost in translation as a demand-side restriction. It should be read as anti-involution (positive) rather than demand destruction (negative); it is the same policy that rebuilt producers' margins in the first place.

None of this is a surprise to them: Macquarie flagged energy storage inventory build and renewed surplus concerns as a key risk for the first half of calendar 2027 back in August. What's playing out is what the broker expected to play out, arriving earlier and a bit louder than anticipated.

The miners are keeping more of the profit than the processors: the concentrate market remains persistently tight, and while inventories may gradually rebuild as Zimbabwean supply returns, current conditions still support concentrate pricing.

The equities imply a worse world than the one that exists: on Macquarie's implied pricing, LTR screens as the most expensive at around US$1,350/t SC6, followed by IGO at roughly US$1,200/t, PLS at about US$1,150/t and ELV as the cheapest at roughly US$1,000/t. At yesterday’s close of US$1,965/t, spodumene is still trading at close to double the bottom of that range.

IGO is the broker's preferred exposure, on attractive free cash flow yields across a range of price scenarios and no major project commitment – 15% and 10% in FY27 and FY28 on base-case forecasts, and still 7–9% even at US$1,500/t spodumene. Macquarie flags a clearly defined cash sweep or distribution mechanism at the TLEA joint venture as a potential re-rating catalyst.

Up more, down less

Is the market too obsessed with volatile carbonate futures pricing, as Macquarie suggests? ASX producers sell concentrate, not carbonate. Picture the chain: Australian miners dig the rock and ship it to Chinese processors, who refine it into the battery-grade carbonate the futures market prices.

Our miners sit at the front of that chain, and for most of this cycle they've kept the bigger share of the profit. That's what Macquarie means when it says value is accruing upstream. Over the last two sessions the gap has started to close, and that's now the single most important thing to watch. Because when carbonate falls faster than concentrate, the squeeze lands on the Chinese converter in the middle – and squeezed converters eventually buy less spodumene. Yesterday’s big fall in SC6 looked like the day that arithmetic caught up.

The big question for shareholders in ASX lithium producers is where the price of concentrate lands. For scale, the last bear market – November 2022 to June 2025 – took spot lithium carbonate down 90%. Concentrate is currently down in the mid-thirties from its May peak. If yesterday's close of US$1,965/t is the low of the current down-cycle, then IGO, PLS, LTR and ELV are still earning comfortably above their cash costs and, on Macquarie’s implied price estimates, undervalued at their current share prices.

Now consider that on Macquarie's spot-price scenario, free cash flow gets thin across the sector once capex commitments are taken into account. Macquarie has PLS's Pilgangoora all-in sustaining cost at US$923/t in FY27, ELV's group cash cost at US$1,033/t and LTR’s Kathleen Valley at US$1,113/t, and IGO's share of Greenbushes runs at roughly US$305/t. No problem there with spot spodumene at US$1,965/t.

Now work in capital spending – e.g., PLS is putting A$1.0 billion into the ground in FY27 against A$1.0 billion of operating cash flow, LTR A$472 million against A$295 million – and PLS Group’s real free cash flow is more like 1% in FY27 and FY28, LTR is at –4% then 14%, and ELV is at 5% then –1%. So, the risk here isn't that today's spot price doesn't work. It's that ASX lithium producers’ capital is committed for years and the price is set daily in Guangzhou.

If yesterday's close isn't the low, the value scales begin to tip at LTR’s implied US$1,350/t, and down to ELV at roughly US$1,000/t – arguably still a very decent buffer. Consider that Macquarie is forecasting SC6 to average US$2,000/t in FY27, before dipping to US$1,827/t in FY28.

Conclusion: where's the floor?

Investors in ASX lithium stocks spent the better part of the last year being paid by Beijing's willingness to intervene, then spent Friday selling because Beijing intervened again. Anti-involution built this bull market by keeping marginal lithium in the ground, and this time it keeps marginal battery capacity off the drawing board – the same instrument, aimed at the same low-quality end, one step further down the chain.

In my last article I pondered the ceiling on medium-term lithium prices: the level where storage projects stop making financial sense, and where sodium-ion becomes cheap enough in the applications that suit it to displace enough tonnes of lithium to matter. Two independent research houses, working from different evidence, arrived at the same cap – somewhere near RMB200,000/t. Spot carbonate had already peaked at RMB200,500/t in May, which corroborates the level rather than forecasting it.

Given recent developments, we may now be probing the floor instead. Is it the current ~RMB132,000/t for carbonate futures, or Macquarie’s already-breached US$2,000/t for concentrate? Momentum says not yet – but momentum shifts quickly in this market, and bulls and bears emerge faster than you can say “supply-demand imbalance”.

This article first appeared on Livewire's sister site Market Index on Tuesday 14 September 2026. For full references see that article.