First published in Aluminium International Today, September/October 2026, pages 77–79. Reproduced with the editor’s permission.
Six years after the Covid-19 pandemic began, Sergey Belskiy examines how regulation, sanctions, and bank control rebuilt the aluminium trade.
When I wrote the article “The dark forest of Covid-19” for this magazine’s May/June issue in 2020, I named four trends I believed would outlast the crisis: growing protectionism, tightening competition, deepening dependence on institutional finance, and forced digitalisation. Six years of shocks later I can report what the forest did to us.
Three of these trends hardened into the structure of the aluminium trade itself. The fourth, digitalisation, became the base for the verified figures, certificates, and attestations. Together they rewrote what an aluminium transaction must contain to exist – and split the classical trader’s job apart (Fig 1).
Regulation now precedes commerce
Historically, trading was business first, regulations last: partners agreed on price and delivery, signed the papers, and cleared customs. Compliance was an afterthought. For aluminium entering the European Union (EU) today, the sequence is reversed – and the reversal is written into law.
Trimmed before launch by the Carbon Border Adjustment Mechanism (CBAM) Omnibus Regulation (EU) 2025/2083, the CBAM Regulation (EU) 2023/956 entered its definitive regime on 1 January 2026.
Above 50 tonnes a year, covered goods can only enter the EU through an authorised CBAM declarant. This is not a tariff – without the status, the metal cannot lawfully cross. Embedded emissions are declared annually and verified by an accredited verifier, or default to values set intentionally high and rising yearly.
For aluminium, the mechanism carries an imbalance: only direct emissions are priced. Indirect emissions from electricity, for the most power-hungry of the base metals, stay outside the net. While this exclusion lasts, the border can barely tell hydro-smelted metal from coal-smelted. The regulation has scheduled a future review, and the day that window closes, competitive positions will be reordered overnight.
Certificate sales open on 1 February 2027. The first declaration is due 30 September 2027 for 2026 imports. From 2027, declarants will hold certificates against half their embedded emissions, and EU producers’ free allocations will run out by 2034.
In December 2025, the European Commission proposed extending the mechanism to some 180 downstream steel and aluminium products from 2028. My conclusion, and the thesis of this article, is that the regulatory design of an aluminium transaction now precedes its commercial design (Fig 2).
The old sequence
Compliance arrives at the end
The new sequence
Origin became a legal status
Where metal came from used to be a line on shipping documents. For aluminium, it is now a legal status. For Russian metal, that status is written by four jurisdictions and two exchanges, each enforcing different rules.
Russia produces roughly 5% of the world’s primary aluminium, with Moscow-based Rusal being the largest individual producer outside China.
Under US sanctions, no US citizen or company may provide warranting or physical-settlement acquisition services for Russian aluminium produced on or after 13 April 2024.
The UK prohibits acquiring exchange warrants for metal produced on or after the same cut-off. The EU moved a year later. Council Regulation (EU) 2025/395, enacted on 24 February 2025, phased in the prohibition in two stages: first, an import quota of 275,000 tonnes to 26 February 2026, then a residual 50,000 tonnes for contracts concluded before 25 February 2025.
From 31 December 2026, the ban is absolute. Switzerland mirrored it through its Ukraine ordinance in May 2025, three months behind Brussels.
The exchanges turned law into warehouse practice. The LME and CME stopped warranting newly produced Russian metal in April 2024, and since 25 February 2026 the LME will not warrant Russian aluminium in an EU warehouse without a sanctions-compliance attestation. The production date on the certificate of analysis decides whether a tonne goes on warrant.
In addition, origin can be lawfully transformed. Crossing one border for processing allows the metal to leave the press shop with a newly declared origin – one that current aluminium regulations cannot reach.
Steel is already subject to an input-tracing rule, and until aluminium follows, the real test is run by banks and exchanges, not customs. Your metal may be excellent and your buyer willing, but without origin evidenced to the standard of the regime and its banks, there is no transaction. Therefore, the chain of custody became part of the price (Fig 3).
Banks only finance what they control
The commodity-finance failures of 2020–21 cost the banks dearly, and they answered as banks do: fewer names, harder collateral, and structure instead of trust. The uncommitted transactional line gave way to borrowing bases, collateral management agreements, and controlled payment waterfalls.
A credit committee barely asks who the borrower is – it asks what the bank controls, at which moment, by which document.
Financing moves to the front of the deal. Title transfer points, document flows, custody, insurance assignments, settlement mechanics: all engineered before the term sheet, because today they are the term sheet. If you assemble a transaction commercially and bank it afterwards, it returns restructured or declined – months gone either way. The deals that close are designed to be financed from the first draft.
The trader, unbundled
Set these forces against the old division of labour. The trader took title, carried the price risk and earned the spread. In exchange, the producer gave up the customer, margin, and any say over the metal. The bargain was fair while the trader’s book was the only bridge across price, geography, and credit. No longer: the work that makes a modern aluminium transaction executable – regulatory shape, origin evidence, financing structure – is not done from a trading book.
Unbundle the trader and two professions appear. One is genuine risk-taking: carrying stock, running a position, and absorbing basis – and for a producer without the balance sheet for price, credit, and currency risk, that book remains the product. The other, connecting a producer to a distant buyer, assembling the finance, standing behind the paper, needs none of the producer’s title, customer, or margin.
There is a Japanese precedent for this model. The trading houses – the sogo shosha – were founded in the Meiji era to take Japan’s foreign trade back from foreign merchants who then handled nearly all of it. By the postwar decades they carried most of the country’s exports and imports. Speculation was never the point. Their trade was organisation: market access, structuring, finance, logistics, and information.
Move the sogo shosha role onto today’s European market and one feature of the old house goes: the position has turned from an asset into a liability.
A party holding title holds a conflict of interest – bank counsel and sanctions officer both see it. Call it the European sogo shosha: the trading house’s architecture, delivered by a party that takes no title and runs no book.
Swiss law has held the form ready for over a century: the commission agent, the Kommissionär of Articles 425 to 438 of the Code of Obligations, contracting in its own name but for the principal’s account. The allocation of interest stands in statute, not in somebody’s pitch. Title and margin never leave the producer: it obtains customer and chain financing directly through its sogo shosha.
From that seat, one party assembles what none of the old chain could alone: regulatory shape, origin evidence, financing architecture, and execution – every relationship and basis point staying where they were created (Fig 4).
The commission agent here is not an administrator picking up papers behind the producer. Transactions die between disciplines, not within them – where one competence ends, and nobody sees the deal whole.
The agent unifies production, trading, finance, regulation, and supply chains into a single operation, executing them simultaneously, not in turn. A single competence in a single point is what the market lacks – producers first of all.
Out of the forest
A producer fights hard against alternative suppliers from other regions – every element of the structure works for its position. Direct access to the end consumer: the relationship stays with the producer.
Financing the whole supply chain earns an additional premium and releases working capital. Risk does not vanish – it is placed piece by piece: price on the exchange, buyer credit with the insurer, and currency with the bank.
Service to the consumer’s gate adds competitiveness that price alone cannot buy. Together they make the producer not merely resilient – they let it grow. That is the philosophy of the commission agent today. The philosophy of a new sogo shosha.
The 2020 article made one bet: the trends would outlast the shock and harden. The bet paid, with interest – they converged on the trader’s function and took it apart.
What holds a modern aluminium flow together, and keeps its value with the producer, is an architecture assembled before the commercial terms.
The dark forest of 2020 was never a storm to be waited out. The trade re-formed inside it – and walked out on the side of the one who creates the value.
Published in Aluminium International Today, September/October 2026, pages 77–79.
The PDF is generated from this page, so it always carries the current version and the date the inputs were verified.