The way the UK counts carbon is bad for the climate and for growth
To make the UK’s energy transition a success, we must stop counting our successes as a cost to the world
This is Post Haste, our home for swift thoughts, responsive takes, and first attempts at solving problems. Want to write for us? Email editor@britishprogress.org
By David Lawrence (@dc_lawrence) and Pedro Serôdio (@pdmsero)
The British state’s official system of climate accounting attempts to bend the laws of science. It effectively invents a new molecule: “CO2UK”, emissions that are produced on British soil, and pretends these are the only kind of emissions that can contribute to climate change.
This kind of navel gazing is bad for everyone. It means we fail to pursue the policies that are most impactful for fighting global climate change, and we also prevent productive economic activity from taking place inside Britain.
Inside the guidance
The 2020 Green Book Review made it mandatory for every project affecting emissions to carry a carbon cost in its economic case. The responsibility for specifying how was delegated to what is now the Department for Energy Security and Net Zero (DESNZ), whose supplementary guidance tells civil servants which emissions to count and what values to apply.
That guidance does not consider counterfactuals: it does not ask where activity would go if it did not happen in Britain. Likewise, official business cases do not consider whether emissions would be produced elsewhere.
In fact, it’s even worse: because Britain’s electricity grid is one of the cleanest in the world, refusing new projects on the grounds of climate can often be worse for global emissions, as it forces activity to more polluting grids. Civil servants end up scoring comparatively green projects as a net cost to climate.
If a factory or a data centre will be built either in Britain or abroad, and Britain’s grid is cleaner, then building it here lowers global emissions because the UK has led the world in increasing the carbon efficiency of its grid. But our own appraisal methodology penalises the UK for it. If the activity moves to a country with a dirtier grid instead, Britain’s books might look greener, but the planet – and the communities affected by climate change – are worse off.
The pitfalls of territorial accounting
The standard argument for why we should only consider our own emissions is that other countries have their own carbon budgets, and will keep to them. However this is demonstrably false, on both counts. Many countries have not signed up to the Paris Agreement, and many more do not keep to carbon budgets as religiously as we do.
This means we have built a system that can score an increase in the world’s emissions as a success, provided the extra carbon is released somewhere we have decided not to count. We’re simply pretending the environmental damage doesn’t exist provided we’re not responsible for it. The countries that would take the activity are not bound by anything resembling our carbon budgets or our appraisal rules, so when the emissions leave our books they enter no one else’s. We are the only player keeping score, and we have written the rules so that we lose.
In the current context, territorial accounting measures how principled we are, not how impactful. Activity we refuse to allow reappears somewhere with looser rules and a dirtier grid, and more emissions may be the end result.
We are not suggesting that we should throw the rulebook out of the window and approve everything. But we do need to do a better job of accounting for counterfactuals when the activity will happen anyway somewhere else. And we should stop pretending we can’t see it.
Building a road in Britain does not affect the number of car trips in Italy, so the displacement risk is low and territorial accounting is the correct approach. The problem occurs when activity is mobile, where what we turn away will be built somewhere else. Data centres are the clearest case: compute is useful on a global scale, so a UK refusal is close to a straight transfer abroad. For activity of that kind, a more discriminating approach is now a necessity rather than a refinement, particularly if the UK is not to fall hopelessly behind the frontier.
Since the 2020 review, this approach has been applied across government. Every project that affects emissions must carry their cost in its economic case, valued against our domestic carbon budgets and the net zero target.
DESNZ’s guidance requires analysts to consider emissions changes ‘domestically or internationally,’ but then prices any such change using carbon values calibrated exclusively to UK abatement costs — making counterfactual displacement impossible to capture. The framework has no mechanism to express the benefit of activity happening on a cleaner grid rather than a dirtier one.
The framework structurally cannot express the benefit of activity moving to a cleaner grid. Emissions that move abroad simply drop out of the calculation and we pretend they don’t exist. A project that would cut global emissions by happening in the UK rather than in a country with a grid that relies much more on coal, for example, is logged as a pure cost; a decision that increases emissions for the world and worsens the problem of climate change is counted as an environmental benefit.
Consider the counterfactual
We are not making a special plea for British industry. Proper carbon accounting does not always favour the UK. For instance, a data centre in France, with its overwhelmingly nuclear-based grid, would produce fewer emissions than one built here.
If this is the case, our appraisals should take this into account. The point is not that we must “always favour Britain.” The point is to compare against the actual real world counterfactual, instead of pretending the activity, and its emissions, simply vanish when we say no.
When it comes to data centres, our research finds that a data centre built in Britain emits 70% less than the same facility in Saudi Arabia, 43% less than in the United States, and 31% less than in Germany. Demand for compute is global and mobile: a project Britain turns away will get built anyway and will more than likely happen in a country with a dirtier grid.
Over 25 years, the same data centre would emit eight times as much in Saudi Arabia as in the UK, four times as much in the US, and twice as much in Germany. By any sane accounting of the impact on climate change, attracting that investment here is one of the most environment-friendly things Britain could do, both because it lowers global emissions compared with the alternative, and because it rewards the UK’s transition by making its investment in green technology pay off.
But our appraisal rules perversely turn that benefit to the world into a cost. The cleaner our grid becomes, the more Britain becomes the best place to build energy-intensive industry. And yet our system of accounting penalises exactly this.
We have taken one of the most positive consequences of the energy transition, the ability to use energy productively without destroying the climate, and redefined it as environmental damage. We count our successes as a cost to the world.
 | A guest post by
|