Avoided emissions occupy an uneasy place in climate discourse. They are increasingly invoked by startups, investors, and policymakers as evidence of climate impact and yet they describe something that, by definition, cannot be directly observed, audited, or measured in the same way as emissions already released into the atmosphere. Unlike a company’s carbon footprint, which accounts for emissions that have already occurred, avoided emissions attempts to quantify a parallel reality: a world in which a technology was not developed, a product was not adopted, or a system remained unchanged. In practice, avoided emissions can be calculated both retrospectively, based on how much of a solution has actually been sold or deployed, and prospectively, to estimate the potential impact of technologies that have yet to reach scale. This article focuses on the latter. Not because forward-looking avoided emissions is the only valid use case, but because they are the least intuitive and the most influential in shaping investment decisions and transition narratives.
If climate action is ultimately about transforming systems rather than merely shrinking balance sheets, then understanding what could be prevented matters. But when impact hinges on assumptions about markets, behavior, and adoption pathways, the line between disciplined modeling and optimistic storytelling becomes thin. As avoided emissions migrate from academic frameworks into investment decks and corporate claims, the central question is no longer whether they are useful but whether they can be used responsibly.
Measuring Impact After the Fact vs. Before It Happens
For decades, climate accountability has been built around a simple premise: emissions are counted after they occur. Corporate carbon footprints, national inventories, and regulatory disclosures all rely on tracing greenhouse gases that have already entered the atmosphere and assigning responsibility for them. This approach, known as attributional accounting, has the virtue of clarity. Emissions can be measured, verified, and compared across time. But it also has a blind spot. It tells us little about whether a company, technology, or investment is actually changing the trajectory of the system.
Avoided emissions emerge precisely in this gap. Rather than asking how much a company emits, they ask how much worse emissions would have been without a particular intervention. The accounting logic shifts from ownership to influence, from static inventories to dynamic change. In methodological terms, avoided emissions rely on intervention accounting and, where possible, consequential lifecycle assessment frameworks designed to capture how decisions ripple through markets, supply chains, and behavior.
This distinction is more than semantic. A low carbon footprint does not necessarily imply a positive climate contribution. Conversely, a solution that introduces emissions today may still reduce overall emissions if it displaces a more carbon-intensive alternative at scale. Traditional carbon accounting struggles to accommodate this logic, because it is designed to allocate responsibility, not to evaluate transformation.
Carbon footprints remain essential for accountability and reduction targets. Avoided emissions, by contrast, are increasingly positioned as a tool for understanding societal progress toward decarbonization. At the core of the debate is a concept familiar to economists but less so to auditors: avoided emissions describe a counterfactual world. They depend on assumptions about what technologies would have been used instead, how quickly markets would have evolved, and whether regulatory or behavioral changes would have occurred anyway. In other words, they quantify impact not by observing reality, but by modeling its alternative. The question, then, is not whether avoided emissions are “real” in the same sense as reported emissions. It is whether their assumptions are explicit, their methods disciplined, and their uncertainty acknowledged.
The Baseline Problem: Choosing the World That Never Was
The headline number is rarely the most important part of an avoided-emissions claim. The real signal lies in how the baseline was constructed, how often it is revisited, and how openly its limitations are disclosed. In a field built on counterfactuals, credibility is not achieved by eliminating judgment, but by making it visible.
Every avoided-emissions claim rests on a single, deceptively simple question: what would have happened otherwise? The answer to that question, the reference or baseline scenario, is where avoided emissions either gain analytical credibility or collapse into speculation.
In theory, the baseline represents the most plausible alternative reality in which a solution did not exist. In practice, defining it requires a series of judgments about markets, technologies, regulation, and behavior. Would users have adopted an incumbent technology, an emerging alternative, or no solution at all? Would efficiency gains have been mandated by regulation anyway? Would the carbon intensity of electricity grids, materials, or transport have improved over time regardless of the intervention? Each assumption pushes the result in one direction or another.
Methodological guidance generally discourages heroic baselines. Best practice favors “average market solutions” rather than worst-case alternatives, precisely to avoid overstating impact. Yet even this restraint leaves room for interpretation. Markets are rarely homogeneous. Baselines vary by geography, sector, and time horizon. A technology that displaces a coal-heavy solution in one region may merely replace a relatively clean alternative in another. Aggregating these realities into a single reference scenario can obscure as much as it reveals.
The baseline problem becomes even more pronounced when avoided emissions are projected into the future. Many solutions, particularly in energy, buildings, and infrastructure, have long lifespans. Over decades, the emissions intensity of the baseline is likely to change due to policy, innovation, or economic shifts. Locking in today’s baseline risks exaggerating future benefits; updating it too aggressively risks erasing legitimate contributions.
This is where avoided emissions diverge most sharply from traditional carbon accounting. A carbon footprint assumes the world as it is. Avoided emissions require an explicit theory of how the world evolves. That theory may be grounded in data, but it is ultimately an interpretation of uncertainty.
Uncertainty, Rebound Effects, and the Myth of Precision
Avoided emissions are often presented with striking numerical confidence. Charts extend decades into the future. Impact is expressed to the nearest ton of CO₂. Probability-adjusted totals are summed, discounted, and graphed as if the future were merely a delayed dataset. The precision is reassuring and largely illusory.
Unlike historical emissions, avoided emissions are exposed to layers of uncertainty that compound rather than cancel out. Market adoption rates, technology performance, policy shifts, energy-system evolution, and consumer behavior all interact over time. Each assumption may be reasonable in isolation; together, they create a range of possible outcomes far wider than a single headline figure suggests. Yet many avoided-emissions claims collapse this range into a point estimate, giving an impression of certainty that the underlying models cannot support.
One of the most persistent blind spots in this process is the rebound effect. In its simplest form, rebound occurs when efficiency gains lower costs, leading to increased consumption that offsets some of the emissions savings. But rebound effects extend beyond individual behavior. They can emerge indirectly, when cost savings are spent elsewhere in the economy, or systemically, when efficiency reshapes entire markets. These effects are well documented in economic literature, yet remain difficult to quantify in practice. As a result, they are frequently acknowledged in footnotes rather than integrated into impact estimates. Ignoring them tends to bias results in a single direction: upward.
This is where uncertainty management should replace false precision. More rigorous approaches do not pretend to predict the future exactly. Instead, they test how sensitive outcomes are to key assumptions, explore best and worst-case scenarios, and communicate results as ranges rather than absolutes. In these models, uncertainty is not a weakness to be hidden, but a property to be disclosed.
For an audience accustomed to audited financials and verified emissions inventories, this can feel uncomfortable. But the alternative, treating modeled futures with the same confidence as measured pasts, risks eroding trust in climate metrics altogether. If avoided emissions is to play a serious role in guiding capital, policy, and innovation, they must abandon the myth of precision and confront uncertainty head-on.
Why Avoided Emissions Matter
The climate transition is not primarily a problem of measuring what already exists. It is a problem of deciding what to build, scale, and finance next. Avoided emissions, for all their imperfections, attempt to answer that forward-looking question.
Traditional carbon accounting excels at assigning responsibility for today’s emissions. It is far less capable of distinguishing between activities that merely operate more efficiently within a high-emissions system and those that actively shift the system’s trajectory. Avoided emissions address this gap by focusing on substitution, displacement, and systemic change. They help explain why two companies with similar footprints may have radically different implications for future emissions and why some high-emitting activities may still be aligned with decarbonization if they enable broader transformation.
This does not make avoided emissions a substitute for emissions reductions, nor a justification for delay. They cannot neutralize a carbon footprint, offset ongoing pollution, or replace science-based reduction targets. Their roles are different. Properly used, they function as a directional signal: a way to assess where future emissions are likely to be reduced, under what conditions, and with what risks.
The challenge is governance. As avoided emissions migrate from academic frameworks into investment memos, corporate reports, and policy narratives, the temptation to oversimplify is strong. Large numbers travel well. Ranges and probabilities do not. Yet credibility will not come from bigger impact claims, but from more disciplined ones: claims that expose their assumptions, acknowledge uncertainty, and resist being used as shortcuts to climate legitimacy.
References
[01] Guidance on Avoided Emissions, World Business Council for Sustainable Development
[02] The Avoided Emissions Framework, Mission Innovation
[03] Full Climate Impact Assessments, Mission Innovation
[04] Estimating and Reporting the Comparative Emissions Impacts of Products, World Research Institute
[05] How to measure systemic impact, Planet A Ventures
[06] Input-output models and increasing economic output, The Scottish Parliament
[07] Net Zero Initiative 2020-2021 Final Report, Net Zero Initiative
