Marginal Abatement Cost (MACC)
Faced with a climate target, the first question is not whether to cut emissions but which cuts to make first, and what each will cost. Some measures — swapping to LED lighting, insulating a building — actually save money. Others, like green hydrogen for heavy industry, cost a great deal per tonne avoided. Line them all up from cheapest to dearest, and you have a single picture of the whole decarbonisation job: what to do, in what order, and at what price.
That picture is the marginal abatement cost curve, built from one number repeated across every option: the cost of cutting one more tonne of emissions.
The marginal abatement cost (MAC) is the cost of cutting one more tonne of CO₂e with a given measure, net of savings. A MAC curve ranks measures cheapest first; read to a target it shows which to use, and the last measure’s height is the shadow price of carbon.
Definition — the Cost of the Next Tonne
The marginal abatement cost is the cost of abating (reducing) one additional tonne of greenhouse gas emissions through a particular measure, expressed per tonne of CO₂e and net of any savings that measure delivers. “Marginal” means it is the cost of the next tonne, not an average — and for many measures that cost changes as you deploy more of them. A single MAC is one number for one option; its power comes from calculating it consistently across every option a company, sector, or country could use.
Stack those measures up, ordered from the cheapest cost per tonne to the most expensive, and you get a marginal abatement cost curve (MACC) — the familiar step chart where each measure is a block whose width is how much it can abate and whose height is its cost per tonne. Made famous by consultancy and government “abatement cost curves”, the MACC turns a messy list of options into a single, ordered map of the abatement available and what it costs.
Keep the two terms straight. The marginal abatement cost (MAC) is a per-tonne cost for one measure. The marginal abatement cost curve (MACC) is the ranked collection of those costs across many measures. The MAC answers “what does this option cost per tonne?”; the MACC answers “in what order, and at what cost, do we hit the target?”.
How to Read a MAC Curve
A MACC is read from left to right — cheapest first. Each measure’s width is the emissions it can remove; its height is the cost per tonne. Measures below the zero line save money; measures above it cost money. To reach a target, you move along the curve, deploying measures in order, until the cumulative width equals the reduction you need.
| Measure (cheapest first) | Marginal abatement cost | Effect |
|---|---|---|
| LED lighting retrofit | −£60/tCO₂e | Saves money |
| Building insulation | −£20/tCO₂e | Saves money |
| Efficient motors | −£5/tCO₂e | Roughly breaks even |
| Solar PV | +£25/tCO₂e | Costs money |
| Heat pumps | +£55/tCO₂e | Costs money |
| Green hydrogen (hard-to-abate) | +£200/tCO₂e | Costs money |
Illustrative figures. Reading this list: the cheapest measures pay for themselves, so they come first; you then work down the curve until cumulative abatement meets the target. The cost of the last measure you need — say £55/tCO₂e if heat pumps are where you stop — is the marginal cost of that target. See a full, interactive curve built from your own measures in the Marginal Abatement Cost Calculator.
The Formula: Net Cost per Tonne
The marginal abatement cost of a measure is, in essence, its net annual cost divided by the emissions it avoids each year:
MAC = ( annualised cost of the measure − annual savings it delivers ) ÷ annual abatement (tCO₂e)
The subtlety is in “annualised”: a measure’s upfront capital is spread over its lifetime using a capital recovery factor, which depends on the discount rate and the asset’s life — so the discount rate chosen has a large effect on where a measure lands on the curve. The abatement in the denominator is measured in tonnes of CO₂e, with each gas converted by its global warming potential (methane at 29.8 [GreenCalculus gwp.CH4_fossil.ar6_100 · IPCC AR6 WGI Ch 7 Table 7.SM.7 (2021) — AR6 GWP-100 · v2026.203] and nitrous oxide at 273 [GreenCalculus gwp.N2O.ar6_100 · IPCC AR6 WGI Ch 7 Table 7.SM.7 (2021) — AR6 GWP-100] times CO₂ over 100 years) and summed as CO₂e. The full annualisation is set out in the marginal abatement cost methodology.
Negative-Cost Measures and the Efficiency Gap
The most striking feature of almost every MACC is that a chunk of it lies below zero: measures whose energy savings more than repay their cost, so their net cost per tonne is negative. Efficiency improvements are the classic example. On paper these are not just cheap but profitable — which raises an obvious question: if they save money, why haven’t they already been done?
The gap between what looks profitable on a curve and what actually happens — the “energy-efficiency gap” — has real causes: upfront capital constraints, split incentives (the party who pays is not the one who saves), transaction and hidden costs, behavioural inertia, risk, and the discount rate assumed. Critics argue that negative-cost estimates often omit these frictions, overstating the free lunch. The lesson is not to ignore negative-cost measures, but to treat the curve as a guide to be interrogated, not a promise — the barriers are exactly what a good abatement strategy has to overcome.
MACC and the Carbon Price
The MACC is where abatement economics meets carbon pricing. If you draw a horizontal line at a given carbon price, every measure whose bar sits below that line is worth doing — its cost per tonne is less than the price of emitting. Raise the price and more of the curve becomes worthwhile; lower it and less does.
Read the other way, the curve sets the price. The height of the MACC at your target level of abatement is the marginal cost of hitting that target — which is exactly the shadow price of carbon. This is why companies use a MACC to inform an internal carbon price, and why policymakers use it to gauge the carbon price a target implies. The curve and the price are two views of the same thing.
Building a MACC, and Its Limits
There are two broad ways to build one. An engineering (bottom-up) MACC costs each measure individually from technology data — transparent and intuitive, the style of the well-known consultancy curves. A model-derived (top-down) MACC reads the marginal cost at each level of abatement out of an economic model, capturing interactions the bottom-up version misses. Both are useful; they answer slightly different questions.
A MACC is a powerful summary, but it is a static snapshot with known limits. Measures interact — doing one changes the cost and potential of another — so the blocks are not truly independent or additive, and the ordering can shift. Costs and potentials are uncertain and change over time as technologies mature. And the curve hides the frictions behind negative-cost measures. Used as a conversation-starter and a prioritisation aid it is invaluable; used as a precise, fixed cost of a target it will mislead.
Common Confusions
- Confusing MAC with MACC. The MAC is one measure’s cost per tonne; the MACC is the ranked curve of many measures.
- Confusing it with the technology’s total cost. The MAC is a cost per tonne abated, net of savings and relative to a baseline — not the sticker price of the technology.
- Confusing it with the social cost of carbon. The MAC is the cost to abate a tonne; the social cost of carbon is the damage from emitting one. They are different curves entirely.
- Treating negative-cost measures as free. Real barriers — capital, split incentives, hidden costs — stop many “profitable” measures from happening.
- Reading the curve as fixed and additive. Measures interact, order matters, and costs and potentials change over time.
- Equating the curve’s height with a carbon tax. It is the marginal cost (the shadow price) at that target; the policy price a government sets is a separate choice.
Frequently Asked Questions
Marginal abatement cost (MAC) is the cost of reducing one additional tonne of greenhouse gas emissions with a specific measure, expressed per tonne of CO₂e and net of any savings the measure delivers. “Marginal” signals that it is the cost of the next tonne, not an average, and for many measures it rises as more is deployed. It is the building block of a marginal abatement cost curve, which ranks measures from cheapest to most expensive per tonne to show how — and at what cost — a target can be met.
A marginal abatement cost curve is a chart that ranks abatement measures from cheapest to most expensive per tonne of CO₂e. It is usually drawn as a step chart: each measure is a block whose width is the emissions it can abate and whose height is its cost per tonne, with money-saving measures below the zero line and money-costing measures above it. Reading it from left to right shows which measures to prioritise and what it costs to reach a given level of reduction. Consultancies and governments popularised the tool as the “abatement cost curve”.
From left to right, cheapest first. Each measure’s width is how much it can abate and its height is the cost per tonne; measures below zero save money, those above cost money. To reach a target, deploy measures in order until their cumulative width equals the reduction you need — and the cost of the last measure required is the marginal cost of that target. Drawing a horizontal line at a carbon price shows which measures are worth doing at that price: everything whose block falls below the line.
Because their savings outweigh their costs. Energy-efficiency measures — better lighting, insulation, efficient equipment — often cut energy bills by more than they cost to install, so their net cost per tonne of CO₂e avoided is below zero. On the curve they appear profitable. The puzzle is why they are not already universal, and the answer is a set of real-world frictions: upfront capital constraints, split incentives, hidden and transaction costs, behavioural inertia, and risk. Because of these, “negative cost” on a MACC rarely means genuinely free, and good estimates account for the barriers.
They are two views of the same thing. The height of a marginal abatement cost curve at a given level of abatement is the marginal cost of reaching that level — which is exactly the shadow price of carbon for that target. Equally, drawing a horizontal line at a carbon price shows which measures are worth doing: any whose cost per tonne falls below the price. This is why a MACC is used to inform an internal carbon price and to estimate the carbon price a climate target implies.
Need someone who does this? 6 carbon accounting & inventory providers in our directory →