Picture the scene: you are the finance director of a mid-sized cement manufacturer, and you have just finished the arithmetic. Your company sits just below the threshold that would require it to hold a compliance account in the regional emissions trading scheme. It produces real carbon. It would pay real money for real offsets. And yet the market that supposedly prices those offsets is, for your company, a closed room with no handle on your side of the door.
That is not an accident. It is architecture.
The room where the price is made
Wholesale carbon markets, whether structured as cap-and-trade schemes like the EU Emissions Trading System or as compliance-linked offset registries, are built around one specific kind of participant: the large, regulated installation. A power station. A steel mill. A refinery above a statutory tonnage threshold. These entities hold compliance accounts directly with the registry, which means they can receive, transfer, and retire offset credits in a way that satisfies a legal obligation. The price you see quoted for a tonne of carbon dioxide equivalent is, almost always, the price two of these entities agreed upon.
Everyone else is watching through glass.
The wholesale market's liquidity concentrates among a surprisingly small number of counterparties. In the EU ETS at maturity, fewer than a thousand installations account for the overwhelming majority of traded volume. That number matters: it tells you this is less a broad market than a club with a strict door policy. Brokers and intermediaries exist, but they charge for access, impose credit requirements, and set minimum trade sizes that can run to tens of thousands of tonnes. A company emitting forty thousand tonnes a year, which sounds substantial in ordinary conversation, is a rounding error to a desk that normally clears in blocks of two hundred and fifty thousand.
How the plumbing actually excludes you
Think of the market's internal structure less like a stock exchange and more like the hydraulics of an old municipal water system: the pressure is real, the flow is genuine, but the connection points were installed for a different era and a different class of building.
The first layer is the registry. Offset credits, whether under a voluntary standard like Gold Standard or Verra's Verified Carbon Standard, or under a compliance mechanism, must sit in a registry account before they can be transferred. Opening and maintaining that account requires legal documentation, sometimes a fee, and in compliance markets, a demonstrated regulatory obligation. A company with no compliance obligation in the relevant jurisdiction simply cannot hold a compliance-grade account. It can hold a voluntary account, but voluntary credits and compliance credits are not the same commodity. The price difference between them can exceed fifty percent, a gap that dwarfs most companies' offset budgets.
The second layer is the intermediary chain. Because most non-regulated entities cannot access the registry directly, they must go through a broker or a carbon finance house. That intermediary buys in bulk, packages credits, and sells smaller lots at a markup. The markup is not trivial. Industry practitioners have described spreads of ten to twenty percent on small retail-equivalent transactions, against spreads of under one percent on large institutional trades. The emitter who most needs affordable offsets to manage a modest liability ends up paying the highest per-unit cost. That is not an inefficiency waiting to be corrected; it is the market working exactly as designed, just not for them.
The third layer, and the one least discussed, is counterparty risk management. Wholesale desks that do allow smaller participants to transact typically require either prepayment or a credit facility. Prepayment locks up working capital. A credit facility requires a banking relationship and a balance sheet that many smaller manufacturers, logistics companies, or agricultural processors simply do not have in the form the desk demands. The market is not hostile to these companies exactly. It is indifferent in a way that amounts to the same thing.
Two companies, one price, very different outcomes
Consider two plausible companies. Hartley Industrial, a large glass manufacturer, holds a compliance account and emits just over four hundred thousand tonnes annually. When it needs to cover a shortfall, its treasury desk calls a broker, agrees a price close to the screen price for a block of one hundred thousand tonnes, and settles within three days. Total transaction cost: roughly half a percent above spot.
Then there is Morven Logistics, a road freight company that has committed to its customers to offset the emissions from their supply chains. It generates around sixty thousand tonnes of scope three liability per year and wants to retire verified credits against that figure. No compliance obligation, no registry account, no existing broker relationship. The retail carbon platforms it can access charge a price that, once the markup and platform fee are included, runs about thirty-five percent above the same underlying credit Hartley bought. Morven is buying the same tonne of avoided deforestation. It is paying a materially different price for it, and it has no mechanism to close that gap.
The market is not lying to Morven. It is just not designed for Morven.
What people misunderstand about the voluntary tier
A persistent misconception holds that the voluntary carbon market is a parallel, accessible version of the compliance market, open to anyone with a cheque and a conscience. In practice, the voluntary market has its own concentration problem. Verra and Gold Standard between them hold the majority of issued voluntary credits, and issuance on those registries is dominated by project developers and aggregators who sell forward to large corporate buyers under multi-year offtake agreements. By the time a credit reaches the spot market, it has often already been committed.
The spot voluntary market that smaller buyers actually encounter is a secondary or tertiary market. It is composed of credits that large buyers have passed on, credits from older project vintages that are less attractive for quality-sensitive claims, or credits that carry some question about additionality that a sophisticated buyer's legal team flagged. None of that is visible in the headline price.
And additionality, the core concept that a credit represents emissions reductions that would not have happened anyway, is genuinely harder to verify when you are buying through four layers of intermediary. The emitter with a compliance obligation and a direct registry account can trace a credit's provenance in minutes. The small buyer on a retail platform is trusting the platform's due diligence, which varies considerably. That should bother policymakers more than it apparently does.
The structural consequence nobody advertises
So ask yourself: if the published carbon price is not your price, what exactly are you managing toward?
The market's design reflects its origins. Cap-and-trade systems were built to price carbon among large industrial emitters, and they do that reasonably well. The problem is that policy ambition has expanded faster than market architecture. Corporations of every size are now making net-zero commitments that require offset procurement, and the infrastructure has not been rebuilt to serve them. The result is a system that prices carbon accurately for perhaps five percent of the entities now obligated to care about it.
What remains is a price signal that is real and a market that is genuinely liquid, but only for a class of participant that represents a fraction of the emitters who now need access to it. The cement company doing the arithmetic in its finance department is not being paranoid. The liquidity exists upstream of where they are allowed to stand.
The practical implication is blunt: if you are a smaller emitter, or a company with voluntary rather than compliance-grade obligations, the published carbon price is not your price. Your price is whatever the intermediary layer adds to it. Understanding that markup, and what drives it, is more useful than watching the exchange screen. Until the registry infrastructure catches up with the policy ambition, that spread is where the real climate finance story lives.