A watershed year for the commodity trade

As we're closing out 2022, let's look at one of the most significant events in commodity markets this year. In March, Chinese nickel producer Tsingshan Holdings Group had to cover a massive short position at a loss, causing a meltdown at the London Metals Exchange (or LME). The resulting short squeeze highlights how the traditional derivatives exchange model can actually add systemic risk, instead of limiting it. I’ll explain here why the existing system — in place since 1865 when CBOT launched futures — is due for an update and why digitalising the physical commodity markets is crucial.

Futures exchanges manage risk through brokers, also known as Futures Commission Merchants (FCMs). These brokers request performance bonds or margin balances to ensure traders can meet their financial obligations. Traders must post cash collateral, called the initial margin, when buying or selling futures. FCMs make money from collecting fees and interest on customer cash balances. Most traders are customers of a broker that is a clearing member of the exchange, which guarantees the trader's obligations to the exchange and provides additional protections against default. When a clearinghouse — the party between buyers and sellers — experiences a default, the risk is usually shared by all clearing members through a guaranty fund. ​

Nickel Pig Iron

​ In March, Chinese nickel producer Tsingshan Holdings Group had to cover a massive short position at a loss and caused a dramatic intervention at LME. Tsingshan is a private Chinese stainless steel producer who through innovation in metallurgy, became a steel powerhouse and the world's largest nickel producer. Nickel is used primarily in alloys such as stainless steel and increasingly in batteries. It is sold in various forms, with the purest being high-grade (Class 1) nickel traded on the LME and stored in their warehouses worldwide.

But many end-users don’t need such refined nickel. Stainless steel production accounts for over two thirds of nickel demand, up from one-third in the past three decades. Buyers who use nickel in ferroalloys like stainless steel smelt it with iron. Thus a product consisting of 25-35% nickel and 65-75% iron will work fine. Tsingshan pioneered a process called Rotary Kiln Electric Furnace (RKEF) to produce low-cost nickel pig iron (NPI) using nickel ore from laterite, rather than the refined nickel sold on the LME. This NPI contains 10-12% nickel mixed with iron, making it attractive for stainless steel producers who can pay the same price (or slightly less) for NPI as they would for traditional nickel, while also getting a bonus of iron. The innovation revolutionized the stainless steel industry and the mines that produced laterite. Laterite production in Indonesia — who have the largest nickel reserves in the world — went from 130,000 tonnes in 2015 to an estimated one million tonnes in 2021.

NPI weighed the nickel price down in the 2010s, but the market picked up again in 2018 as the markets priced in the need for nickel in EV manufacturing. Now NPI is good for stainless steel, but it’s of no use for modern battery cathodes, whose production requires the type of refined nickel sold on LME. In March of 2021 however, Tsingshan announced a deal to sell 100,000 tonnes of nickel matte (essentially smelted nickel ore with lower iron content) to two large Chinese battery makers. The company managed to further refine the NPI they’d been producing from Indonesian laterite into a product with a 75% nickel concentrate, making it suitable for battery usage. The announcement caused an 8% drop in the nickel price.

The London Metals Exchange

Most of the world’s nickel — even refined concentrate — isn’t sold on LME. It’s a futures exchange where traders trade six ton lots of nickel that they have no intention of producing or receiving. This creates a benchmark for private parties to use when making deals for the delivery of nickel and other metals. The LME, which is over 145 years old, is somewhat of an old-boys club and benefits from its monopoly status in certain markets like nickel. When I started working in M&A in London in 2009, I received a tour on LME and remember how small I thought that floor was given its importance in the global metals market that influences almost everything we touch.

Metals producers use futures exchanges like LME to forward sell their anticipated production through short positions, effectively locking in a price they would receive for the metal it plans to produce before the contract’s delivery date. If the price falls, the company can close the trade at a profit. A rising price doesn’t matter to a short producer either, because any price changes on the futures market should offset price changes in the value of the inventories (as long as the traders can meet their margin calls).

Tsingshan wanted to increase production dramatically by manufacturing nickel matte for electric vehicle batteries. The company planned to produce 850,000 tons of nickel in 2022, an increase of 40% in a year. The obvious consequence of this additional nickel hitting the market would be a fall in the price of nickel. However, not everyone shared this pessimism about prices. Hedge funds were buying nickel contracts, betting on the electric vehicle boom. Glencore supposedly also had a position on the LME that would benefit from rising prices. In the beginning of 2022, it had already taken ownership of more than half of the available nickel in LME warehouses.

You probably realized by now that LME did not accept the pig iron nickel Tsingshan produces, or even the matte nickel they’re selling to battery makers. Exchange delivery nickel has to be the refined product and that nearly all comes from one place, Norilsk in Russia.

Norilsk is the world’s northernmost city, built around a series of high-grade nickel sulfide deposits in Siberia whose ore contain concentrations of nickel, copper, cobalt, platinum and palladium. A higher-grade ore, with a nickel content typically above 2 percent, is crushed and smelted to remove any impurities. The smelted material then goes to a refinery where it is processed into pure nickel. The Russian company that operates the mine and smelters, Nornickel, makes and sells briquettes and cathodes of a purity that make them suitable for delivery to LME warehouses to fill open contracts.

The short squeeze

Everything changed when Russia invaded Ukraine. While Russia’s nickel exports haven’t been targeted by sanctions, U.S. and European buyers have nonetheless sought alternatives to Russian sources. Nickel’s price moved sharply higher in the week after Russia’s invasion. This significantly hurt Tsingshan’s short position. Remember, when prices move up, traders like Tsinghshan who have sold futures contracts face margin calls; they must put up more cash to cover potential losses. The short position didn’t raise any eyebrows among traders at first, because it belonged to one of the largest stainless steel manufacturers in the world. Tsingshan could probably get their hands on some refined nickel at delivery time. But the invasion would make that a lot more difficult. If Tsingshan didn’t have the goods, they had to try and casually close the contracts out by buying nickel on LME. All traders know that there’s no reason for the world’s biggest nickel pig iron supplier to be buying nickel on LME, unless they’re short and can’t cover their position.

The moment they realized this, traders started pouncing. It’s free money. On March 7, nickel’s price began its staggering ascent, surging from $30,000 a ton to more than $50,000. LME brokers and their clients were hit with margin call after margin call. Several large brokers got margin calls of close to $1 billion each over the course of the day.

Nickel soars to record high

Tsingshan was unable to meet its margin calls due to a lack of available cash and credit. Its banks and brokers, including JPMorgan Chase, BNP Paribas, and Standard Chartered, were also affected as they had taken the other side of Tsingshan’s contracts and offset those by taking short positions on the LME. They now had to pay margin calls without receiving margin from Tsingshan. In response, some of these brokers rushed to buy back nickel contracts, causing the price of nickel to increase further in a short squeeze. This cycle intensified the difficulties for Tsingshan and its brokers.

By now, the entire nickel industry was in crisis. The LME assembled its Special Committee that has the power to issue emergency rules for the market and ultimately suspended trading. The price was frozen, below the record high but still at $80,000 a ton. At that point, Tsingshan wasn’t the only nickel company that was struggling — just the largest. Many producers, traders, and users of nickel with short positions on the LME were facing margin calls many times larger than they were prepared for. At that price of nickel, the brokers themselves wouldn’t be able to pay their margin calls. Four or five of the brokerages that are LME members would have failed, a shock that could have devastated the global metals industry.

LME made an unprecedented decision. It decided to cancel a day worth of trading. Roughly $3.9 billion in value, according to Bloomberg. Exchanges sometimes cancel trades when technology glitches or “fat fingers” cause one-off mistakes, but it’s extremely unusual for an exchange to cancel whole sessions of trading after the fact. The last time LME did this was during the 1985 “Tin Crisis”, when a cartel of producers collapsed. The decision meant traders wouldn’t need to pay margin calls on the basis of the $80,000 nickel price. Effectively, it artificially lowered the market to the moment when prices closed on the 7th of March, at $48,078.

Shortcomings laid bare

To show you how this is also a failure of the traditional derivatives exchange model, I will use a simple example to zoom in on LME’s shortcomings in this debacle.

A trader — let’s call him Tsingshan — has $2b of collateral and puts on a $3b short in a contract trading at $1 (so 3 billion contracts). The markets start to go up in price and eventually they're at $1.33. Now Tsingshan has:

position size: 3b contracts = $4b
collateral: $1b

Tsingshan has a huge position now and is fairly leveraged — $4b notional, 4x leverage. Another 25% move and Tsingshan is underwater; each 25% move beyond that costs about $1b. What exchanges in such a situation do, is reach out to the trader and warn them to top up — sending a maintenance margin call. But Tsingshan has at least 1 business day to top up. If it's a Friday evening, then they have at least 3 days or 72 hours.

Unfortunately, a war is unfolding in Ukraine and commodity prices are going through the roof. By the time the next business day ends, there have been 3 days of unusual high volatility. A 25% move bankrupts Tsingshan. A 50% could cost the exchange or whoever is backstopping $1b, if the customer cannot put up more collateral. This could happen in 3 days! And in fact, it did in March of this year. The price of LME nickel rose a whopping 250% in less than 24hrs!

​There are really two problems here. First, the exchange margin calls based on time, not on price. When Tsingshan reaches 6x leverage, it's time to issue a margin call. Because the goal of the margin call is to ensure that before Tsingshan runs out of collateral, the trader either

(a) tops up their collateral
(b) reduces their position

Once the contract reaches a price of $1.66, trader will fully run out of collateral; any move beyond that and trader goes negative account value. The goal is for Tsingshan to deleverage before the contract reaches $1.66. Time is arbitrary here. The contract could move 1% in 3 days, or it could move 150%. What matters is the price. You have to close down before contracts hit $1.66.

Say that closing down Tsingshan’s position — buying back 3b contracts — would have a 10% impact. This means that, if Tsingshan isn't going to top up, you have to start reducing their position before the contract hits around $1.50 — adding 10% impact to that would move markets to around $1.66. So in this hypothetical situation, you have to begin liquidating Tsingshan's position when the contract hits $1.50 completely, independently of when that happens. If a week later the contract is at $1.35, it's fine. But if a minute later it's at $1.50, it's time to start closing down Tsingshan. This is how liquidations should be done. Exchanges should begin deleveraging a position as soon as it's running too low on collateral, independent of time. Sometimes one day is too long.

Which brings me to the second problem. This happened just before the weekend. The exchange wasn't open over the weekend and so the fastest they could margin call was 72 hours. On Monday, 7 March, the market unraveled quickly, with prices rising by a stunning 66% to $48,078. On Tuesday the price went vertical, soaring $30,000 in a matter of minutes and just after 6 a.m. the price of nickel passed $100,000 a ton.

Souce: Bloomberg | Bloomberg

Nickel’s 250% price spike in less than 24 hours plunged the industry into chaos, triggering billions of dollars in losses for traders and leading LME to suspend trading for the first time in 3 decades. And worst of all, they canceled trades. As explained above, this could have been prevented if margin calls were made sooner. This was not some black swan event. Margin calls are delayed regularly, because the interests of brokers (FCMs) are not always aligned with the exchange. They could call their customer — who may be the largest nickel producer in the world representing a significant part of income — and ask to top up the margin. Tsingshan could have replied that they’re good for it, but cannot make payment immediately. By not forcing margin payment, this broker is now risking the guarantee fund i.e. other customers' funds. It is all based on trust and individual decisions. Moreover, regulators didn’t realize the size of the over-the-counter positions, having wrongly assumed that looking at the positions on LME was enough — ignoring privately negotiated over-the-counter (otc) derivative and physical commodity trades.

Pioneering new infrastructure

FTX’s proposal to CFTC earlier in the year addressed the first issue and would be groundbreaking for futures markets. They were proposing a direct clearing model built around real-time risk technology, similar to that currently used by spread betting platforms such as IG Index, Robinhood and E*Trade. This was seen as a threat to traditional market infrastructure players such as central counterparty clearing houses (CCPs) and FCMs, as FTX was proposing to risk manage each underlying client directly. Some CCPs are already calculating margin requirements on a near real-time basis. However, they typically only request additional collateral at certain times of the day and often only send these requests to clearing brokers, who may not pass them on to (important) clients.

In the proposed model the relationship is directly with the clients, meaning that the clearing house would be responsible for ensuring that they have sufficient collateral on a real-time basis and cannot rely on brokers to provide credit i.e. asking margin upfront, instead of extending credit. The new model would ensure that positions are closed out if collateral held is insufficient, instead of simply requesting additional collateral if the risk increases beyond a certain threshold. This real-time margining prevents investors from accruing greater and greater losses. In the case where there is a large liquidation and markets are moving down too quickly to successfully close the position before bankruptcy, a backstop liquidity provider system would step in. Liquidity providers who have opted in to the system will internalize the position, taking over the whole obligation and collateral. They would do this before the account actually goes bankrupt, so they have a chance to successfully manage the position. Interestingly, this could introduce competition for future exchanges—something even FTX’s critics admitted would be a positive development.

This new clearing model thus fosters competition, democratizes derivative trading and protects smaller investors from accumulating debts they cannot afford. CME, despite its strong criticism to FTX’s proposal, unscrupulously followed suit later by applying to CFTC for an FCM license themselves — likely inspired by rising interest rates and the lost revenue from brokers collecting margins. Meanwhile, FTX perversely became the victim of the risk that they correctly identified to CFTC, by not swiftly deleveraging a losing position of a key customer—they allowed their sister company, hedge fund Alameda Research, to trade entirely outside their own liquidation rules.

The other issue laid bare by the LME nickel squeeze — otc trade in physical commodities and their derivatives — can be addressed in several ways. Some argue for regulation, in the same way that Enron was addressed by Sarbanes-Oxley and after the 2008 financial crisis we saw Dodd-Frank and Basel III. Javier Blas pointed out that G7 countries could for example set up a register of international oil transactions, as they agreed to do in 1979. This was never implemented because oil traders were vehemently against it. Wider regulation is probably not the answer, as commodity trade is global and would require a multilateral approach — unilateral government interference is what usually leads to windfall opportunities for traders.

The only way I believe you can improve this is by creating a digital exchange with an otc environment (dark pool) — enabling traders to move volumes around the globe openly or privately, while aggregating all pricing & volume information for the sake of risk management. This approach would ensure self-regulation and make commodity trade more efficient in the process. Moreover, it would address the fact that otc markets often fail to demonstrate resilience to market disturbances and become illiquid and dysfunctional at critical times. In periods of high volatility with a lot of noise, price discovery breaks down and traders exit the market — leading to severe supply chain disruptions.

Vital role of commodity traders

The proper functioning of commodity markets is of critical importance to mankind. Most countries, including the Netherlands and the majority of other OECD countries for example, are not self-sufficient. They depend on breadbaskets for food—countries like the US, Brazil or Ukraine—that produce more crops than they consume domestically. They depend on mining powerhouses, like Australia, Chile and Russia, for metals like nickel, cobalt and copper that are crucial for EVs and the clean energy transition (IEA estimates that the market for such metals could increase 30-fold by 2040). As a result, any supply volatilities — climate change, war — can have a dramatic impact on pricing and the global economy. It is vital that commodities move efficiently from places where there is surplus to where there is deficit. This is what traders do best, but offline physical markets do not create the systemic resilience we need. Plus, digital derivative markets could use an update as well, as LME has shown us. In fact, digitalising physical trade and real-time margining goes hand in hand — without real-time benchmarks from the physical commodity markets, it will be hard to price derivatives in this new clearing model.


At Vosbor, we have developed an exchange for the global trade in agricultural commodities. We have launched in beta with the leading groups in the industry, determined to facilitate the transition of international commodity trade to a digital environment that benefits all participants in the supply chain. We are not out to disrupt anyone — in fact, we want to add more traders and liquidity. Nor do we intend to change the relationship-based business, but rather help everyone capture more value. Soon, we will offer commodity derivatives and will increase liquidity with financial investors. We plan to expand into other markets like metals, as trade in bulk commodities is very similar across the board (why our beta users Cargill and Glencore are successful in both agriculture and metals). We are lowering the cost and risk of trading. Together we are going to make commodity markets more efficient, robust and accessible to all.

Read more articles