What the world pays to emit carbon, and what it does not

A carbon price attaches a cost to each tonne of carbon dioxide released, on the principle that the damage emissions cause should be reflected in their price. Two main instruments do this directly. A carbon tax sets the price and lets the quantity of emissions adjust. An emissions trading system, or ETS, sets a cap on quantity and lets the market find the price. The practical question for anyone reading the headlines is narrower and more useful: across the world, how much of what we emit actually carries a price, and how large is that price.

How much is covered

The share of global greenhouse gas emissions covered by a direct carbon price has risen steadily. Around 2010 the figure was close to 7 percent. It now stands near 30 percent, spread across roughly ninety implemented carbon taxes and trading systems. That growth is real, but the headline coverage figure can mislead in two ways. Coverage counts whether an emission falls under a pricing scheme, not whether the price is meaningful. And it conceals large gaps: many of the highest-emitting sectors and several large economies still price little or none of their output.

• Coverage measures reach, not adequacy; an emission can be priced at a few dollars and still count as covered.
• Carbon taxes and emissions trading systems coexist, sometimes within the same jurisdiction for different sectors.
• Compliance markets, where prices are set by policy, are distinct from voluntary carbon markets, where buyers offset by choice and prices run far lower.

How much is paid

Among priced emissions, the global average direct price has been estimated near the low tens of dollars per tonne of carbon dioxide in recent years. That average hides extreme dispersion. The European Union ETS has at times traded above 100 dollars per tonne, while large volumes elsewhere are priced in single digits. Less than 1 percent of global emissions face a price above 100 dollars. The voluntary market is lower still, with average offer prices for offsets often in the range of a few dollars per tonne. So a single global number describes very little; the distribution is the story.

The gap with the social cost of carbon

The social cost of carbon is an estimate of the economic damage caused by emitting one additional tonne of carbon dioxide. Central estimates in recent literature sit above 100 dollars per tonne, with a wide range reflecting assumptions about discounting and damages. Set against an average direct price in the tens of dollars, the implication is straightforward: prices in force are, on average, well below the level that would internalise the estimated damage. The gap is not uniform. A few jurisdictions approach or exceed common social-cost estimates for parts of their economy, while most price far under them.

Why distribution and per-capita emissions matter

Carbon pricing intersects with population and equity in ways a single price obscures. Per-capita emissions differ by more than an order of magnitude across countries, so a uniform price falls very differently on different populations, an unevenness that also shapes the demographic pressures traced in our brief on fertility decline and below-replacement birth rates. Within countries, energy and fuel costs take a larger share of lower-income household budgets, which makes the distributional design of a carbon price central to whether it is durable. This connects to the same accounting we set out in our brief on household consumption and Earth-system pressure, where per-capita consumption, not headcount alone, drives the underlying emissions. Revenue recycling, returning carbon revenue to households as dividends or tax cuts, is the main lever used to address that incidence.

A carbon price is best read not as one number but as a distribution: who is covered, at what level, and on whom the cost finally falls.

What these figures do not establish

Coverage and price are inputs, not outcomes. They describe what is charged, not how much abatement results, which depends on how firms and households respond and on the other policies in place. Estimates of both coverage and the social cost of carbon carry meaningful uncertainty and are revised as data and methods improve. Voluntary-market prices in particular reflect offset quality and demand, not the cost of damage. This brief summarises published figures and adds no original estimation; its conclusions are bounded by the sources cited and the dates of their publication. Readers may also find context in our broader work, summarised on the symposium home page and in the editorial method.