Carbon Pricing

Core Concepts

What carbon pricing does

Carbon pricing is an attempt to solve a market failure. When a company burns fossil fuel or emits greenhouse gases, the costs of that pollution — damage to future climate stability, to agriculture, to health — are borne by people who have no say in the transaction. The emitter pockets the profit; everyone else absorbs the harm. Carbon pricing puts a monetary value on that harm and charges it to the emitter.

In theory, once emissions carry a price, firms and households adjust their behavior: they invest in efficiency, switch fuels, or reduce consumption. The goal is not to ban emissions but to make cleaner alternatives comparatively attractive.

Two main designs have been deployed at scale.

Carbon taxes

A carbon tax sets a fixed price per tonne of CO₂ (or equivalent). The government charges that price when fossil fuels are sold or combusted. The quantity of emissions is left to the market: emitters reduce as much as is profitable at that price, but there is no guarantee of a specific emissions total.

The most compelling feature of a carbon tax is price certainty. Firms know what emissions will cost a year from now, which aids investment planning. Research comparing carbon taxes and emissions trading systems finds that tax-based systems may produce more predictable behavioral responses precisely because their price signal is stable.

Sweden's experience is the longest-running data point. Sweden introduced its carbon tax in 1991 at what was then a modest rate, integrating it into a broader tax reform that reduced income taxes as compensation. Today Sweden levies the highest carbon tax rate globally at SEK 1,190 per metric ton of CO₂ — approximately US $126. Over three decades, Sweden achieved a 27% decrease in greenhouse gas emissions between 1990 and 2018, with particularly clear effects in transportation and industrial sectors.

British Columbia launched a carbon tax in 2008 — one of the earliest subnational implementations in North America. The tax was designed to be revenue-neutral: revenue returned to households and businesses through reduced income taxes and dedicated credits for low-income households. Early empirical assessments found aggregate emissions reductions of 5–15%, including roughly 8% reductions in gasoline consumption in private vehicles and 4% in manufacturing.

Cap-and-trade

A cap-and-trade system works differently. Rather than fixing a price, the regulator fixes a quantity — an annual cap on total emissions across covered sectors. The cap is divided into permits (allowances), each representing the right to emit one tonne. Firms that cut emissions below their allocation can sell surplus allowances; those who need more must buy them. The market sets the price.

The theoretical advantage is environmental certainty: the cap defines the maximum total emissions, so the environmental target is guaranteed regardless of how high prices rise. Cost efficiency follows from trading, which in principle routes abatement to wherever it is cheapest, regardless of which firm does it.

The EU Emissions Trading System (EU ETS), launched in 2005, is the world's largest cap-and-trade program. It covers electricity generation, steel, aviation, chemicals, and other sectors — more than 40% of the EU's total greenhouse gas emissions across 31 countries. Firm-level causal analysis finds that regulated manufacturing firms reduced CO₂ emissions by 14–16% with no detectable contraction in economic activity, suggesting that firms absorbed carbon costs through efficiency improvements rather than by cutting production.

Allowance allocation: free vs. auctioned

Within cap-and-trade, how allowances are distributed matters enormously. The two main approaches are free allocation (given to firms at no cost) and auctioning (firms must buy allowances).

Free allocation comes in two variants: grandfathering assigns allowances based on a firm's historical emissions, while benchmarking assigns them based on industry-average efficiency standards. Grandfathering is politically easier but rewards historical high emitters and can function as an implicit subsidy — less-productive, high-emission firms stay in business longer than they otherwise would.

Auctioning is economically more efficient. When revenue from allowance sales is recycled to reduce other taxes (e.g., income or capital taxes), the GDP cost of the policy falls by roughly 33% compared to a free-allocation system. But early cap-and-trade programs almost universally started with free allocation to reduce industry opposition. The EU ETS followed this path, with the 2008 revision establishing auctioning as the default from 2013 onward, progressively replacing grandfathering.

The Social Cost of Carbon

Both mechanisms require some sense of how damaging a tonne of CO₂ actually is. The Social Cost of Carbon (SCC) is the standard answer: an estimate of the total present-value harm caused by emitting one additional tonne of CO₂ today — accounting for effects on agriculture, health, sea level rise, and more.

The optimal carbon tax rate, following standard welfare economics, should equal the social marginal damages from that additional tonne. In practice, estimating that number is deeply contested.

SCC estimates have risen substantially over the past decade. High-discount-rate estimates moved from roughly $9/tonne to $40/tonne; low-discount-rate estimates rose from $122/tonne to $525/tonne. The spread is not primarily about new climate science — it is about value judgments. The major modeling disagreements center on:

• Discount rate: How much do we value harm to people decades from now relative to people today? A higher discount rate produces a lower SCC.
• Catastrophic tail risks: How much weight should be given to low-probability but civilization-scale outcomes?
• Distributional equity: Should damages to poor populations count more? If so, the global SCC rises considerably.

These are not technical disputes that more data can resolve. They are normative choices embedded in numbers that look authoritative.

---

Compare & Contrast: Carbon Tax vs. Cap-and-Trade

Dimension	Carbon Tax	Cap-and-Trade
What is fixed	Price	Quantity (emissions cap)
What varies	Quantity (emissions response)	Price (allowance market)
Environmental certainty	Low — no guaranteed total	High — cap limits total
Price certainty	High — firms know cost in advance	Low — prices fluctuate
Investment planning	Easier (stable signal)	Harder (volatile prices)
Revenue	Yes, directly	Yes, if allowances are auctioned
Average effectiveness (meta-analysis)	~6.4% reduction	~9.7% reduction
Political economy	Labeled as a "tax" — often harder to pass	Can be obscured as a "market mechanism"

Price volatility and design fixes

A fixed cap creates price volatility, which is one of cap-and-trade's persistent weaknesses. When the economy contracts (as in the 2008–2009 financial crisis), emissions fall and the surplus of allowances crashes permit prices — sometimes to near zero — removing the investment signal for clean technology.

Regulators have developed several mechanisms to manage this:

• Price floors and ceilings: A floor prevents permits from becoming worthless; a ceiling prevents compliance costs from becoming prohibitive. Together with banking, they produce flatter price trajectories over time.
• Banking and borrowing: Firms can bank unused allowances for future periods, or in some systems borrow against future allocations. This intertemporal flexibility reduces overall abatement costs by allowing firms to equalize their marginal costs across time.
• Flexible caps: Output-based allocation adjusts the total supply of allowances in response to economic activity, dampening price swings — though at the cost of diluting the environmental signal.

---

Worked Example: British Columbia and Sweden

Two different designs, two different contexts, similar directional results.

British Columbia (2008)

• Revenue-neutral by design: every dollar collected in carbon tax was returned via income tax cuts and targeted credits.
• Broad coverage: approximately three-quarters of all provincial emissions were subject to the tax.
• Low-income households: a dedicated Low Income Climate Action tax credit was designed to exceed the average carbon tax payment for lower-income households — after 2010, low-income households were net beneficiaries of the policy package.
• Results: 5–15% aggregate emissions reductions, with stronger effects in transportation (roughly 8% decline in private gasoline consumption) than in industry.

Sweden (1991–present)

• High rate, selective coverage: the tax targets transport fuels and household heating. Industrial energy use receives exemptions.
• Revenue recycling through income tax reduction, not direct dividends.
• Achieved approximately 6–9% reductions in transportation emissions attributable to the carbon tax specifically, with overall greenhouse gas emissions falling 27% from 1990 to 2018.
• The high rate (US $126/tonne) co-exists with sectoral exemptions for industry — a political economy concession that limits coverage.

The contrast between BC and Sweden illustrates a recurring tension in carbon tax design: broad coverage at a lower rate, or high rates applied selectively? Both achieved results, but the choice shapes who pays, how much revenue is raised, and which emissions are actually addressed.

What both cases confirm: Carbon pricing works in the direction the theory predicts. Behavior responds to price signals. The magnitude varies by sector, design, and price level. Neither system operates at prices close to the full social cost of carbon — and both still fall short of what Paris-aligned pathways require.

---

Common Misconceptions

"Carbon pricing destroys the economy"

This is the most common objection and is not supported by the evidence. The major meta-analysis of 21 schemes found emissions reductions of 5–21% with no statistically significant negative effects on aggregate GDP growth. British Columbia maintained a competitive economy through its carbon tax years. The EU ETS produced firm-level emissions reductions with no detectable contractions in revenues or employment at regulated firms.

The distributional impacts are real and require attention (see below), but aggregate economic harm is not the established finding.

"Carbon pricing causes carbon leakage everywhere"

Carbon leakage — firms moving production to jurisdictions with no carbon price — is a theoretically coherent risk. But empirical studies of the EU ETS find no evidence of widespread leakage: regulated firms in 31 countries showed increases in revenues and fixed assets with no significant impact on profits or employment, and did not relocate production. The most recent work using OECD data finds some leakage in specific sectors, but not at the scale often assumed. The risk is real for certain emissions-intensive, trade-exposed (EITE) industries — which is why those sectors receive special treatment in most systems.

"Carbon pricing is inherently regressive and hurts the poor most"

Partly true — and partly the wrong frame. Carbon pricing is regressive on the consumption side: lower-income households spend larger shares of their budgets on energy and transport, so the same carbon price is a larger burden relative to their income. An IMF study found carbon pricing incidence was regressive in all 19 EU countries examined.

But the overall distributional outcome is determined by how revenues are recycled — not by the carbon price itself. A lump-sum dividend to all citizens converts carbon pricing into a progressive policy. BC's Low Income Climate Action credit made lower-income households net beneficiaries. Carbon pricing is a tool: regressivity is a design choice, not an inherent feature.

"If carbon pricing works, we don't need anything else"

Current carbon prices are nowhere near the levels required to drive the transformational changes that Paris-aligned pathways require. Estimates of climate-consistent prices are in the range of $50–200+ per tonne by 2030; most existing schemes are well below this. Meta-analytic evidence confirms 5–21% reductions from existing systems — meaningful, but insufficient for the pace of decarbonization needed. Carbon pricing at current levels must be complemented by technology investment, regulatory standards, and behavioral measures.

There is also a policy interaction problem: complementary policies like energy efficiency standards or renewable mandates can undermine carbon pricing by generating permit surpluses that crash allowance prices, as research in Nature Climate Change documents. The full policy portfolio needs to be designed coherently, not assembled piecemeal.

"Voluntary carbon offsets are an equivalent alternative to direct emissions reduction"

For most currently traded offsets, this is not supported by evidence. Approximately 87% of carbon credits purchased by the 20 largest corporate offset buyers from 2020–2023 carry high risk of not providing real, additional emissions reductions. Additionality — the requirement that a reduction is beyond what would have happened anyway — is conceptually essential and practically very hard to verify. Offsets should not be treated as equivalent to direct decarbonization.

---

Boundary Conditions

Carbon pricing is not sufficient at current price levels

The mechanism works — but the prices deployed so far are too low to close the gap with climate targets. This is a political constraint, not a technical one.

Leakage risk is real for specific sectors

While large-scale leakage has not materialized in EU ETS data, emissions-intensive, trade-exposed industries (steel, cement, aluminium, fertilizers) face genuine competitive pressure when regional carbon prices diverge. Multiple OECD countries have granted full or partial exemptions to these industries — a political economy concession that reduces coverage and environmental effectiveness.

Regressivity in developing economies works differently

The assumption that carbon pricing is regressive does not travel universally. In low-income countries, carbon pricing can exhibit progressive incidence because poor households cannot afford carbon-intensive goods in the first place. The design implications for a high-income economy and a low-income economy are therefore quite different.

The CBAM creates new equity tensions globally

The EU's Carbon Border Adjustment Mechanism (CBAM) charges importers for the embedded carbon content of goods entering the EU, targeting cement, steel, aluminium, fertilizers, electricity, and hydrogen. The mechanism entered its definitive phase in January 2026, requiring importers to purchase and surrender CBAM certificates priced at EU ETS allowance rates.

The CBAM has real environmental rationale — it reduces incentives for carbon leakage by leveling the competitive playing field between EU producers and importers — and there is evidence it creates spillover effects that accelerate carbon pricing adoption elsewhere. But it also disproportionately burdens developing countries, particularly African exporters, by reducing their competitive access to EU markets before they have the institutional capacity or financial resources to rapidly decarbonize their export industries. Its WTO compatibility remains legally uncertain.

Carbon offset markets are not self-regulating

The voluntary carbon market contracted by 56% in 2023, reflecting growing recognition of quality problems. The systemic failure of additionality verification is not a fringe concern — it is documented in peer-reviewed research and has triggered coordinated federal oversight responses. Reforms such as the ICVCM Core Carbon Principles represent an attempt at governance, but implementation and enforcement remain inconsistent. Until additionality can be reliably verified at scale, offset claims should carry significant epistemic discounting.

Public support is not guaranteed by good design

Research in Nature Climate Change finds that carbon tax rebate programs have limited impact on public support for carbon pricing. Making the policy economically fair does not automatically make it politically popular. Public opposition often reflects distrust of government, concerns about competitiveness, or preference for different climate instruments — not simply distributional concerns. This matters for policy durability: Canada's consumer carbon tax, once a model of dividend-based design, was eliminated in March 2025.

---

Quiz

1. A carbon tax sets a ____, while a cap-and-trade system sets a ____.

Answer

A carbon tax sets a fixed price (per tonne of CO₂). A cap-and-trade system sets a fixed quantity (an emissions cap). In each case, the other variable is determined by the market.

---

2. Why is the Social Cost of Carbon so contested? Name two sources of disagreement that are fundamentally normative rather than purely technical.

Answer

(1) Discount rate selection: how much moral weight to give to harms experienced by future generations relative to present ones — a value judgment embedded in a number. (2) Weighting of catastrophic tail risks: how much probability weight to assign to low-probability, civilization-scale climate outcomes — again, a normative choice about risk tolerance, not just a statistical question.

---

3. A company claims it has achieved net-zero emissions by purchasing carbon offsets to cover 100% of its remaining emissions. What single question most directly tests whether this claim is credible?

Answer

Whether the offset credits are additional — i.e., whether the emissions reductions they represent would not have occurred anyway without the carbon revenue. Additionality is the core requirement for an offset to represent a genuine reduction. Without verified additionality, the offset is not equivalent to a direct emissions cut.

---

4. In a high-income country, carbon pricing tends to be regressive when assessed on the consumption side. Why might this pattern be reversed in a low-income country?

Answer

In low-income countries, poor households may not consume significant amounts of carbon-intensive goods (energy, motor fuel) because they cannot afford them. The higher-income households who do consume these goods bear the carbon cost. This inverts the regressivity pattern relative to high-income countries, where all households consume energy but low-income households spend a larger share of their budget on it.

---

5. The EU Emissions Trading System co-exists with renewable energy mandates and energy efficiency regulations. What unintended consequence can this interaction create?

Answer

Complementary policies that reduce emissions — such as renewable mandates or efficiency standards — can depress the demand for emissions allowances, generating a permit surplus and crashing the carbon price. A near-zero allowance price removes the investment incentive for clean technology in covered sectors, undermining the purpose of the ETS. This is the policy interaction problem: the full portfolio must be designed coherently, not just layered together.