Description

SQM is the leading Chilean producer of lithium.  The stock trades on the NYSE under the ticker SQM, and the market cap is ~$28b.  In early 2018 I posted SQM as a short, which worked out pretty well.  I think it’s an even better short today because I believe that the same dynamics that were at play back then are in play today, bet even more exaggerated.  On top of that, in today’s market I think it’s nice to have a short in the portfolio that is not only liquid (SQM trades ~$200m per day), but is still near 52wk highs with low short interest (~2% for SQM).

Context / 2015-2020 as an analog to today:

Back in 2015 the lithium market went into a deficit and the price of the commodity began climbing and reached a peak in 2018 of between $20k-30k per ton.  By 2017 the market talk was all about the incredible secular drivers of demand, how new supply was coming on at higher marginal costs, and how new supply would be late, low quality, etc.  In reality, secular demand did continue unabated, however looking back, demand actually undershot the sell-side models that I was looking at, at the time.  Meanwhile, despite all the concerns about whether supply projections would actually be realized, 2019 supply actually overshot. 

Commodity cycles:

Back in 2017/2018 Bernstein did some work on commodity prices showing that in nearly every industrial commodity – nickel, copper, zinc, aluminum, iron ore, met coal, thermal coal, etc. – when new project IRR’s rose above one standard deviation from their long-term average, they proceeded to fall back to, and eventually below that long-term average, with the price of the commodity being the primary determining variable.  Makes sense – that commodities go in boom-bust cycle as S/D causes returns to go above their cost of capital, which in turn draws capital in, raising supply, and eventually pushing prices down, and thus returns down to their long-term average, with the long-term average project IRR roughly equating to cost of capital.  Not that this dynamic is going to be super enlightening to anyone, but the Bernstein work put some nice quantitative work and historical data up against the “this time is different / secular demand trumps all” drumbeat that always seems to accompany high prices.

Situation today:

I think today is a similar situation to back in 2018, with COVID exacerbating both supply constraints, and demand drivers.  Prices have skyrocketed from <$10k per ton to ~$70k per ton today.  Ironically, given that the price screams scarcity, lithium is one of the most common commodities in the world, with a reserve life exceeding 400yrs (vs other industrial commodities like nickel, copper and iron ore in the 30-60yr range).  As one operator recently put it in an interview: “there is not a scarcity of lithium in the world.  This is not a commodity that is in short supply”.  For what it’s worth, that same operator said in the same interview: “we’ve seen this before in the lithium cycle…we’ve got signs that lithium is more bubble-like in behavior”.  As a result of these bubbly prices, majors like SQM and ALB, as well as the Chinese, the Australian miners, etc. are all investing in significant capacity expansions.  Even SQM CEO Ricardo Rodriguez doesn’t seem to be suggesting that lithium supply won’t accelerate dramatically – at their recent investor day he said that he expects demand to continue to be at least 90% of supply.  In the commodity world of marginal cost pricing, I believe even that will provide a lot of room for lithium prices to fall.  And speaking of marginal costs, the cost curve- is between $10k per ton and $20k per ton depending on whether you use today’s demand or even out to 2030.  Lots of room in between that range and current spot prices.

Where things go from here:

SQM earnings are largely determined by the spot price of lithium (majority of volumes are either spot or index driven), therefore the stock very closely tracks the price of lithium.  Rather than deriving a specific earnings number and multiple at which to target fair value / covering, I think it’s more practical to think about actions relative to the commodity price.  Most sell-side decks have lithium gently falling and leveling out at $50k per ton as the market moves into a slight surplus.  For what it’s worth, that was the type of forecast back in 2018, and it did not work out that way.  My expectation is that, similar to past cycles both in lithium and other industrial commodities, the commodity price will move more rapidly than expected back towards marginal cost.  I think that will begin to happen soon given we are starting to comp a lot of the supply/demand factors that caused the recent market tightness, and now the global economy is weakening rapidly as supply is set to accelerate.  My base case is that covering the position at a $20k per ton lithium price (despite for historical precedent for an overshoot on the way down just like the way up), will be accompanied by a >50% fall in SQM’s share price.

Additional thoughts on SQM:

SQM seems to operate in a tough political environment, as demonstrated by the recent vote on a proposed new constitution.  While the voting results, which rejected measures that would have included stricter mining regulations, were taken as a short-term positive for SQM by the market, just the fact that this was an issue shows the kind of shareholder-unfriendly environment that the company is operating in.  SQM operates in the Salar de Atacama under a license extended by CORFO, the governmental organization that regulates mining, among other things, in the country.  Back in early 2018 the company renegotiated their license with CORFO – the agreement extended through 2030 and was for increased production in return for greater economics to the government.  At SQM’s recent analyst day, nearly the entire CFO presentation was around how much they pay the government.  Also at the analyst day, the CEO’s presentation centered heavily around a new $1.5b capital project to tackle some of the environmental concerns that were addressed in the recent constitutional proposal.  The third speaker at the investor day was an NYU professor who as far as I could tell was there solely to argue as to why SQM shouldn’t be nationalized.  Rather than an event centered around the company’s strategy for delivering shareholder value, the event seemed like posturing / an opening salvo at a new negotiation with the government.  It seems to me that the most likely scenario will be that CORFO once again increases SQM’s production limit, in return for greater economics and increased environmental spend.  There certainly seems to be a negative pattern emerging here – prices go up, SQM gives up more of their economics in order to expand capacity, that capacity drives prices down.

Risks:

Supply disruptions / delays – fwiw, these were talked about last cycle, and if anything, supply should be getting easier post-COVID.

Faster demand growth – fwiw, comping the rebound in consumer demand for EVs in both developed markets and China, as well as having potentially passed peak gas prices in the near-term, this should be more than baked in.

Favorable negotiation with CORFO – fwiw, this is not the historical precedent, and I don’t see the bargaining power at the company that would lead to this result.

Catalyst

Falling commodity price