{"id":86093,"date":"2026-05-18T14:19:34","date_gmt":"2026-05-18T13:19:34","guid":{"rendered":"https:\/\/www.lse.ac.uk\/granthaminstitute\/?post_type=news&p=86093"},"modified":"2026-05-18T14:19:35","modified_gmt":"2026-05-18T13:19:35","slug":"defining-residual-emissions-and-why-that-matters-for-corporate-net-zero","status":"publish","type":"news","link":"https:\/\/www.lse.ac.uk\/granthaminstitute\/news\/defining-residual-emissions-and-why-that-matters-for-corporate-net-zero\/","title":{"rendered":"Defining residual emissions and why that matters for corporate net zero"},"content":{"rendered":"\n

Thousands of companies have now committed to reaching net zero. Many have set science-based targets, published transition plans and made pledges that stretch to 2050. But there is a fundamental challenge at the heart of corporate climate strategies: almost none of them can say with precision what the ‘net’ in net zero means for them — because there is no agreed definition of the emissions they expect and need to be neutralising when they get to net zero. This is not a paradox to resolve, so much as a reality to manage, explain Josh Burke, Injy Johnstone, Leo Mercer and Paul Zakkour.<\/p>\n\n\n\n

In a new report, What Are Residual Emissions?<\/path><\/svg><\/a>,<\/em> we survey the scientific literature, interrogate the assumptions embedded in climate models and review the leading corporate standards. Our central finding is an inconvenient truth: no consensus definition of residual emissions exists. Corporate net zero commitments are increasingly under the microscope, as new research<\/a> suggests there is little evidence that adopting a long-term net zero target leads to large or immediate emissions cuts, or to broad changes in climate governance — and the absence of a consensus on this definition carries real risk for the integrity of these commitments.<\/p>\n\n\n\n

However, there is a tractable solution. We propose a more dynamic approach that resolves the definitional deadlock. By following this approach, companies can develop a carbon dioxide removal (CDR) strategy proportionate to their abatement ambition and timeline, and, most importantly, revise it as the evidence changes.<\/p>\n\n\n\n

A term without an enduring definition<\/strong><\/h2>\n\n\n\n

The phrase ‘residual emissions’ entered the climate lexicon through the Fifth Assessment Report<\/path><\/svg><\/a> of the Intergovernmental Panel on Climate Change (IPCC) in 2014 and it has since become ubiquitous in IPCC reports, national net zero strategies and corporate sustainability frameworks alike. Yet the IPCC never formally defined the term, and neither has anyone else in a way that has endured since.<\/p>\n\n\n\n

We find definitions that variously describe residual emissions as those that are more expensive to abate than to neutralise, emissions that are technically or socially hard to eliminate (the emissions that society has decided, for now, it is unwilling to give up, such as from long-haul aviation or beef consumption), and emissions that simply remain at the net zero balance point. As Lund et al.<\/path><\/svg><\/a> outline, residual emissions are “discursively constructed” — continuously brought into being through the claims of actors navigating mitigation demands.<\/p>\n\n\n\n

This is not merely a semantic quarrel. The definition a company adopts determines how much it is permitted to neutralise through CDR rather than eliminate through abatement. A permissive definition inflates the residual category; a stringent one compresses it.<\/p>\n\n\n\n

Residual emissions are model outputs\/artefacts<\/strong><\/h2>\n\n\n\n

Much of our analysis concerns how residual emissions are generated inside integrated assessment models (IAMs) — the computational frameworks that underpin IPCC climate scenarios and, by extension, the science-based targets that companies are encouraged to adopt.<\/p>\n\n\n\n

The key insight is that residual emissions are model outputs, not policy inputs. They emerge from cost-optimisation exercises. IAMs calculate the cheapest pathway to a given temperature target, and whatever emissions remain once cheap abatement options are exhausted are treated as residual. The level of residuals is therefore highly sensitive to assumptions about carbon prices, future technology costs including abatement and CDR, and the discount rates<\/a> applied to future expenditure. Change those assumptions and the quantity of residuals changes too.<\/p>\n\n\n\n

This creates a genuine risk of circular reasoning when IAM outputs are translated directly into corporate guidance. Sectors labelled ‘hard to abate’ in model scenarios — aviation, agriculture, heavy industry — become institutionalised as residual in policy frameworks, not because they are intrinsically impossible to decarbonise, but because the model was built with certain technological and economic assumptions baked in. We are appropriately cautious here. Today’s residual emissions are not necessarily tomorrow’s.<\/p>\n\n\n\n

What the standards say and don’t say<\/strong><\/h2>\n\n\n\n

We reviewed the leading corporate net zero frameworks including the Science Based Targets initiative (SBTi), ISO 14068-1 and the UN High-Level Expert Group recommendations. Despite differences in language, most converge on the view that residual emissions are whatever remains after all technically and economically feasible abatement has been implemented, and that they must be neutralised — not offset — through carbon removal.<\/p>\n\n\n\n

Yet the quantitative guidance varies considerably. SBTi implies a 5–10% residual share for most companies, ISO allows sectoral ranges reaching up to 28%, and Race to Zero anchors on an economy-wide 10% residual target. These ranges are not trivially different. For a large industrial company, they could imply wildly divergent CDR procurement strategies and capital expenditure plans.<\/p>\n\n\n\n

All frameworks agree on one principle: purchasing carbon credits cannot substitute for abatement that is technically and economically achievable. The sequencing also matters. But without explaining further what this means or how to achieve it, it can leave the impression that such framing is empty and performative.<\/path><\/svg><\/a><\/p>\n\n\n\n

While this implies moving up the marginal abatement cost (MAC) curve in a linear fashion, a MACC, which measures and compares the financial cost and abatement benefit of individual actions, is unable to show the dynamic costs of abatement and what happens after you invest in terms of the learning rates, spillovers and network effects that cascade through an economy. Some actions taken today with seemingly high static costs can have low dynamic costs, as was the case with solar photovoltaic technology. The same dynamic logic must be applied to abatement options across the economy.<\/p>\n\n\n\n

The implication is that one should not only pick what looks cheapest today but rather ask “what actions, taken today will minimise the future cost of mitigation, both today and into the future<\/path><\/svg><\/a>?” The view that seemingly expensive investments today result in lower costs in the future is broadly akin to the theoretical work of Vogt-Schilb et al.<\/path><\/svg><\/a> Although it might seem counterintuitive, they find that in certain circumstances — such as achieving a carbon budget — it makes sense to implement some expensive options before exhausting the abating potential of the cheapest options if their potential is higher and their inertia is great.<\/p>\n\n\n\n

For an organisation developing a roadmap to decarbonisation, this generally implies two things. First, start with the low-hanging fruit that already pays for itself. Second, identify which high static-cost items have the strongest dynamic potential and invest in them. Viewing a MACC in this way may drive corporate decarbonisation strategies to move in a non-linear way, bringing CDR into their portfolios at a much earlier stage even when cheaper abatement options exist. This is no longer theoretical. Microsoft, Stripe and Google have all procured CDR at scale to address emissions across their value chains. This approach may be further supported if outside-of-inventory interventions<\/path><\/svg><\/a> (such as research and development [R&D] or catalytic investment in early-stage mitigation technology) are considered a commensurate action in addition to or alongside target fulfilment. As long as companies are transparent about their actions, adopting this approach may drive further corporate interventions into critical — but underfunded — areas such as CDR R&D.<\/p>\n\n\n\n

A practical way forward<\/strong><\/h2>\n\n\n\n

Rather than waiting for a definitional consensus that may never arrive, we propose a pragmatic alternative. Instead of asking companies to identify a fixed category of ‘residual emissions’, they should classify their emissions along two dimensions — how difficult they are to abate, and over what timeframe abatement might become feasible. This produces three working categories of emissions:<\/p>\n\n\n\n