Carbon pricing is working — just not where it needs to

This article is part of a series. You're reading part one.

There is a story told with great confidence in climate policy circles: raise the price of carbon, and the fossil fuel industry will eventually stop producing. The logic is clean. The economics, it turns out, are not.

Carbon pricing has achieved something real. Gas-fired power plants across Germany, the Netherlands, and the UK are facing genuine commercial pressure. Some have closed. The reason is not coincidence — it is that EU carbon allowance prices, now around €65 per tonne, are close enough to the threshold that makes a power plant uneconomical to operate. For the downstream energy sector, the EU Emissions Trading System is doing its job.

But upstream — the gas fields, the liquefaction terminals, the offshore platforms — something very different is happening. Production continues largely undisturbed. Not because operators are ignoring carbon pricing, but because the carbon price required to change their decisions is not €65 per tonne. It is closer to €700 to €1,400 per tonne, depending on the asset. That gap is the subject of this article.

Carbon pricing operates at the wrong margin. It binds downstream, where carbon intensity is high. It remains infra-marginal upstream — changing the accounting of assets, but not their behaviour.

The decision that carbon pricing needs to reach

To understand why this gap exists, you need to think about how exit decisions actually work in the energy industry. A gas field operator does not gradually dial down production as costs rise. The decision is binary: continue operating, or permanently abandon the asset. That is an irreversible choice. Once you close a field and begin decommissioning, you cannot easily reverse course.

This means there is a threshold — a specific carbon price — at which the economics of continued operation tip from positive to negative. Below that threshold, the operator continues. Above it, the operator exits. Researchers call this the shadow carbon value: the minimum carbon price that makes immediate abandonment the rational choice.

The critical insight from recent modelling work is that this threshold is not determined by carbon prices alone. It is set by the interaction of three forces: the current gas price (which drives revenue), the irreversibility of the abandonment decision itself, and the value of potential future uses of the asset — most importantly, the option to retrofit it for carbon capture and storage.

Each of these forces pushes the exit threshold upward, in some cases dramatically.

Why the threshold is so high

The first force is simple arithmetic. The Scope 1 emissions from upstream gas extraction — the CO₂ released during drilling, compression, and processing — are typically between 0.025 and 0.04 tonnes per megawatt-hour of gas produced. At €65 per tonne, that translates to a carbon cost of roughly €1.60 to €2.60 per MWh. Against a gas revenue of €35 per MWh, carbon pricing represents about 5 to 7 per cent of operating cost. It is present in the accounts. It does not govern the decision.

The second force is more subtle. Because abandonment is irreversible, operators are not indifferent between shutting down today and shutting down tomorrow even if expected cash flows are identical. The option to wait has value. You might be wrong about future gas prices. A cold winter, a supply disruption, a policy reversal — any of these could restore profitability. Closing the field extinguishes that possibility permanently.

The mathematics of this option value, formalised in real-options theory, add a substantial premium to the exit threshold. In practice, this irreversibility premium raises the required carbon price by anywhere from 20 to 50 per cent above the simple break-even level.

The third force is the most counterintuitive. As carbon prices rise, gas assets become more valuable as potential CCS sites, not less. A mature North Sea platform with existing pipeline infrastructure has a post-abandonment future that becomes more attractive precisely when the carbon price is high. The option to convert the asset — rather than scrap it — has positive value, and that value increases as the carbon price rises. This means a higher carbon price actually delays some exits by making the asset more valuable to keep alive for future conversion.

The result: for a mature North Sea gas field, the modelled exit threshold is approximately €727–€950 per tonne CO₂ depending on the analytical framework. The current EUA price of €65 is not even in the same conversation.

The upstream–downstream asymmetry

This brings us to the most important finding, and the one with the most direct policy implications.

Gas-fired power generation and upstream gas production are both subject to the EU ETS. But their exposure to carbon pricing is completely different, and that difference is what drives the asymmetry in decarbonisation dynamics.

A combined-cycle gas turbine emits approximately 0.35 tonnes of CO₂ per megawatt-hour of electricity generated. At €65 per tonne, that is a carbon cost of €22.75 per MWh. Against typical clean spark spreads — the margin between electricity revenue and gas costs — of €10 to €20 per MWh, that carbon cost is decisive. It tips the balance. It is why power plant retirement decisions are live commercial questions in every major European market right now.

An upstream gas producer, by contrast, emits approximately 0.025 to 0.04 tonnes per MWh of gas produced under Scope 1 accounting. The carbon cost at €65 is €1.60 to €2.60 per MWh — a rounding error against revenues. The exit threshold for that producer is not €65. It is an order of magnitude higher.

This is not a subtle quantitative difference. It is a structural one. The EU ETS, as currently designed with Scope 1 upstream coverage, is appropriately calibrated for the power sector and structurally miscalibrated for the upstream extraction sector. The instrument is working where it was designed to work. It was never designed to work upstream.

Power plants are closing because carbon pricing has crossed their exit threshold. Gas fields are not closing because carbon pricing has not come close to crossing theirs. These are not failures of the same policy. They are consequences of a single policy being applied across two fundamentally different economic contexts.

Market regimes and the timing problem

The gap between carbon prices and exit thresholds is not static. It responds to market conditions in ways that matter for policy.

During periods of energy market stress — the 2022 gas price spike being the obvious example — gas revenues rise sharply while carbon prices, which are partly anchored by policy credibility, rise more slowly. In the crisis regime, the gas–carbon correlation turns negative: the two prices decouple. The net effect is that the irreversibility premium on exit decisions actually increases during energy crises, because the volatility of the operating spread widens.

This creates an uncomfortable dynamic: the moments when political pressure to act on fossil fuels is highest — energy price crises — are precisely the moments when the economic case for upstream gas production is strongest, and when carbon pricing is furthest from its effective threshold. Policy credibility and market dynamics work in opposite directions.

The glut regime tells a complementary story. When LNG supply is abundant and gas prices are depressed, the gas–carbon correlation is weak and positive. Carbon pricing has more relative weight. But in a glut, the operators most under pressure are LNG exporters and high-cost marginal producers — not the low-cost, long-life fields that drive the bulk of European supply.

What this means for capital allocation

The gap between policy signals and exit thresholds has a direct and underappreciated implication for capital allocation.

Transition risk analysis — the assessment of how climate policy will affect the value of fossil fuel assets — typically assumes that carbon pricing will progressively make those assets uneconomical. It is embedded in scenario analyses, stress tests, and ESG-adjusted discount rates.

But if carbon pricing does not enter the operator's exit decision under any plausible near-term price trajectory, transition risk for those assets is being assessed against a signal that does not affect behaviour. Capital markets are pricing stranded-asset risk using an instrument that is, for upstream producers, behaviourally irrelevant.

The implication is not that upstream gas assets are safe from transition risk. It is that the risk channel is different from the one most analysts are modelling. The relevant risks are demand destruction from downstream decarbonisation, supply competition from LNG, and regulatory change — not the ETS price mechanism working through production economics.

This matters for investors: the transition risk that terminates upstream gas production is more likely to be a demand-side shock than a carbon price signal. That changes how you model it, stress-test it, and price it.

Three things that would actually change the calculus

None of this is an argument against carbon pricing. It is an argument for precision about where carbon pricing works and where it does not. Three things could change the upstream calculus.

The first is Scope 3 coverage. The emissions from burning the gas — combustion by the end user — dwarf Scope 1 extraction emissions by a factor of fifteen to twenty. If upstream producers were liable for those downstream combustion emissions, the effective carbon intensity would transform the economics overnight. A carbon cost of €65 per tonne applied to full lifecycle emissions would make upstream exit decisions immediate and obvious. This is politically and legally complex, but it is the direct mechanism that would make carbon pricing upstream-relevant.

The second is a sharp step-change in EUA prices. The modelling suggests exit thresholds around €700 to €1,400 per tonne for conventional upstream assets. No current policy scenario projects EUA prices in that range within the decade. But for LNG liquefaction facilities — with their higher Scope 1 intensity — the threshold is closer to €115 to €200 per tonne. These assets are within reach of more aggressive policy scenarios, and they should be treated as genuinely at risk in transition planning.

The third is supply-side regulation that does not work through price signals at all: production licensing limits, mandatory decommissioning timelines, well-specific regulation. These instruments bypass the threshold problem entirely by setting administrative rather than economic exit conditions. They are currently underused in the European policy toolkit.

The harder lesson

The energy transition is advancing more unevenly than most policy frameworks assume. Downstream, the EU ETS is working. The power sector is decarbonising, not smoothly but directionally. Carbon pricing has crossed the threshold.

Upstream, the instrument has not crossed that threshold, and under current design it is unlikely to do so within any policy-relevant timeframe. Production will continue until demand destruction, financing constraints, or direct regulation forces the issue.

This creates the structural asymmetry the modelling makes precise: consumption adjusts, but production does not — at least not through the carbon price channel. The result is not an orderly, price-led transition. It is a prolonged mismatch between demand-side adjustment and supply-side persistence, with the attendant risks of price volatility, delayed capital reallocation, and continued upstream investment in assets whose long-term commercial case is genuinely uncertain.

Carbon pricing is doing what it was designed to do. The problem is that the design stops at the wellhead.

Closing the gap between climate ambition and upstream exit decisions requires either extending the reach of carbon pricing into upstream economics — through Scope 3 liability, much higher prices, or direct regulation — or accepting that supply-side decarbonisation will require instruments that the current policy toolkit does not contain.

That is not a comfortable conclusion. But it is the one the numbers support.

illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.

See how the companies in your sector perform on sustainability. On illuminem’s Data Hub™, access emissions data, ESG performance, and climate commitments for thousands of industrial players across the globe.