CCUS context

The UK has committed a sizeable budget to carbon capture. In October 2024, the Government set out up to £21.7 billion over 25 years for the first two clusters, the East Coast Cluster and HyNet, and in December 2024, the Northern Endurance Partnership reached financial close on the country’s first CO2 transport and storage network. Construction on Teesside is now underway.

Most of the public argument about CCUS is still about the capture plant: which solvent, what capture rate, and how high the cost per tonne. If you are deciding whether a UK capture project will actually be built, and when, that is the wrong thing to watch.

The binding constraint is neither the capture technology nor the cost of capture. It is the shared CO2 transport and sequestration network, and the contractual chain wrapped around it. Capture is the mature part. The pipeline, the store, and counterparty risk among emitters determine delivery. Three dependencies, set out below in order of how binding they are, explain why.

The CO2 transport and storage network has to exist before a CO2 capture plant is worth anything

A capture plant with nowhere to send its CO2 is a stranded asset. That sentence sounds obvious, and it is exactly the point that is often skipped in discussions of current technology.

Post-combustion amine capture is not the difficult part here. It is a mature technology, deployed at commercial scale, sitting at the top of the technology readiness scale. You can buy it. What you cannot buy off the shelf is a route to a permanent store. On the East Coast Cluster, that route is a 145 km offshore pipeline to the Endurance saline aquifer, roughly 1,000 metres below the seabed, plus the onshore gathering and compression to feed it.

Shared transport and storage have an awkward economic shape. It is capital-heavy and has to be built ahead of the demand that justifies it. Moreover, it must be sized larger than any single emitter requires, so that future emitters can connect. No individual capture project can carry that cost, and no individual project will build a pipeline and a subsea store on the chance that others follow. This is a textbook coordination problem, and an unsupported market does not solve it on its own.

There is also a timing mismatch on top of the cost. The network spends its capital years before the emitters that connect later contribute any revenue, so the early years are deliberately underutilised by design. Someone has to fund that gap and carry the risk that the later emitters never arrive. A merchant developer cannot price that exposure into a single project, which is precisely why it falls to a regulated structure rather than to the market.

That is why the Northern Endurance Partnership needed a government-backed, regulated revenue model to reach financial close at all. The honest implication for any single capture project is uncomfortable: your timeline is hostage to infrastructure you do not own and cannot control. The capture plant is a procurement decision. The network it depends on is not.

The real bankability question is counterparty risk, not capture risk

The second dependency is contractual, and it is where most business cases are quietly optimistic.

In a cluster, every emitter’s economics depend on every other emitter and on the transport and storage operator, all of which must perform as planned. The CO2 has to flow, the store has to accept it, and the shared cost has to be spread across a connected set of projects. That interface is the real risk, and it has very little to do with whether the capture technology works.

The UK business model architecture is built specifically to allocate that risk. The Industrial Carbon Capture contract is a contract for difference, with a 15-year term and a strike price set to cover both capital and operating costs, negotiated bilaterally for the first projects before any move to competitive allocation. The transport and storage network sits on its own regulated model. The November 2025 update to the business models exists because these interfaces are genuinely hard to paper over.

Here is the part that does not show up in a single project’s spreadsheet. If one anchor emitter slips, the shared infrastructure’s unit economics move for everyone connected to it, because the fixed cost of the network is now spread across less volume. That is why the network’s revenue cannot sensibly be tied to the tonnes that happen to flow in a given year. It has to be underpinned by availability, so the operator is paid for standing ready rather than only for what passes through the pipe, with volume risk deliberately pushed elsewhere in the chain. The question that follows is who holds it, and on what terms.

That cross-chain exposure is the real question of bankability. It is also why the first clusters could not be financed as ordinary merchant projects and instead needed bespoke, state-backed contracts.

Once it is funded, delivery is gated by sequence and supply chain

Suppose the network is committed and the contracts are signed. The third dependency is the least glamorous and, right now, the most active: getting it built.

The East Coast Cluster is a useful gauge because it is public. By March 2026, its main contractors had awarded more than 230 sub-contracts worth £1.5 billion to UK suppliers, including around £500 million of new awards through competitive tender since the third quarter of 2025. The first major equipment for Net Zero Teesside Power arrived at Redcar Bulk Terminal in April 2026, with start-up expected in 2028.

At this stage, the binding constraint is not whether amine capture works. It is engineering and construction capacity, skilled labour, equipment lead times, and the sequencing of a very large set of interdependent works. First-of-a-kind infrastructure carries real schedule risk, and schedule risk on a shared network is shared by everyone waiting to connect. None of that is a technology problem.

A worked example

Considering these dependencies for the East Coast Cluster confirms this pattern is clean. The network came first and needed de-risking: the Northern Endurance Partnership reached financial close in December 2024, not because the pipeline was technically novel but because someone had to underwrite the coordination problem. The contracts came next: the Industrial Carbon Capture model and the regulated transport and storage model allocate the interface risk between emitters and the network. Delivery came third: £1.5 billion of supply-chain contracts and a 2028 start-up date now turn on sequencing and EPC capacity. At no point in that chain is the capture technology the thing in doubt. This is the live UK case, and it is publicly documented.

What this means for you

For investors screening UK CCUS projects, the diligence order is the wrong way round in most decks I see. Capture readiness is the box that is easy to check and rarely fails. Spend your scrutiny on the transport and storage counterparty, the cross-chain default and curtailment terms, and the construction schedule. A confident capture-cost figure attached to a project with an uncertain route to store is an optimistic model, not a bankable one.

For industry leaders in hard-to-abate sectors, your decarbonisation date is set by your cluster’s transport and storage availability, not by your capture vendor’s datasheet. If your site is not on a Track-1 or Track-2 cluster, that connection and its timing should be at the top of your plan, ahead of solvent selection.

For policymakers, the £21.7 billion and the layered business models are doing the right job: de-risking a coordination problem the market cannot solve alone. The lesson the first clusters already teach is one of sequence. Build, or at least firmly commit, the transport and storage backbone ahead of the capture plants that depend on it. The infrastructure is the enabling constraint, so it has to lead.

The conclusion

A CO2 capture plant is a procurement decision. A CO2 transport and storage network is a coordination problem, and coordination problems determine whether clusters are actually built. The UK’s first two clusters did not reach financial close because the capture technology finally worked. They reached it because the state absorbed enough of the network and counterparty risk to make the chain bankable. That is the part worth watching as the next wave of projects comes forward.

PS — the capture-cost ranges get all the attention. The IEA puts capture at roughly USD 15–25 per tonne of CO2 for concentrated streams such as gas processing, rising to USD 40–120 per tonne for dilute streams like cement and power. Yet a 2025 analysis by Agora Industry and the Öko-Institut put end-to-end costs in the EU at €105–280 per tonne. My team’s work has shown that the capture cost can fall between these ranges, depending on the host plant and CO2 capture cost. Notably, most of the gap between those numbers is in CO2 transport and storage infrastructure, as well as the system around CO2 capture, which is where the delivery risk lives. Capture is rarely where these projects are won or lost.