Climate change: putting a price on carbon Kurt Van Dender, Johanna Arlinghaus, Luisa Dressler, Florens Flues and Michelle Harding are with the Centre for Tax Policy and Administration, Organisation for Economic Co-operation and Development Introduction Reducing the risk of very costly consequences of climate change will require strong reductions of emissions of greenhouse gases. CO2 is the major greenhouse gas, and to a very large extent is the result of combustion of fossil fuels: emissions from the use of coal, oil and gas accounted for 69% of global anthropogenic greenhouse gas emissions in 2010 (IEA, 2014). Including CO2-emissions from the combustion of biofuels would push these shares even higher, by around 10%-point. Cutting CO2-emissions from energy use hence is a core element of climate policy. Abating CO2-emissions by enough to appreciably reduce the expected costs of climate change necessitates policy measures beyond the ones currently in place. For example, to limit global temperature increase to around 2 degrees Celsius in 2100, the target maintained in the UNFCCC negotiations, global CO2-emissions will need to fall by between 40 and 70% of current emissions by 2050, and they should be close to zero by 2100. Realising such deep cuts in greenhouse gas emissions requires, among other things, a transformation of the energy basis upon which modern economies rely and adopting less energy-intensive patterns of economic activity. This requires government policy, as private initiative alone cannot be expected to result in sufficient emissions abatement. This is because emission abatement is a public good: the benefits of abatement are shared by everyone but, in the absence of policy, abatement efforts are not or at least insufficiently individually profitable to ensure the level of effort required from the public point of view. A wide array of policies can contribute to emission abatement, eg. energy efficiency regulations, support for research and development of low carbon technologies, and putting a price on carbon. Pricing carbon has particular appeal in that it incentivises emitters to cut emissions as cheaply as possible. For this reason, carbon pricing is widely seen as an indispensable part of an effective and cost-effective climate change policy package. For example, the partners of the International Tax Dialogue – the OECD, the World Bank, the IMF, CIAT (Inter-American Center of Tax Administrations), the European Commission and the Inter-American Development Bank – affirmed at their global conference in Paris in July 2015 that they “share the view that while pricing CO2 emissions alone may not be sufficient to address climate change, it is an essential part of the solution.” Against this background, this article asks two questions: how do countries currently price carbon (what is the effective carbon rate, ECR), and how do ECRs compare to the minimum rate needed (set at €30 per tonne of CO2 for this article) to provide incentives for emissions cuts in line with the 2 degrees Celsius temperature increase target? In short the answers, for 41 countries representing 80% of world energy use in 2012, are as follows: 60% of CO2-emissions from all energy use in the 41 countries are not subject to an ECR at all, 30% are subject to a rate between zero and €30 per tonne of CO2, and 10% to a rate above €30 per tonne. Hence, 90% of emissions are priced below the low end estimate of the climate cost of CO2-emissions, being €30 per tonne, and 60% of CO2-emissions are not priced at all. In road transport, 46% of CO2-emissions in the 41 countries face a rate of more than €30 per tonne of CO2-emissions. Outside of road transport, the ECR is zero for 70% of emissions, and 96% of emissions from energy use are subject to an ECR of less than €30 per tonne of CO2. Price-based or market-based policies As noted in the introduction, policy instruments that ‘make polluters pay’ are very effective at cutting pollution at relatively low costs, because they align polluters’ incentives with the social interest while giving polluters the ability to select the cheapest abatement options; see OECD (2015a). In the context of climate change, this means that the ECR should be aligned with the marginal cost of climate change (in short, the ‘climate cost’) of CO2-emissions. Estimating the climate cost of CO2-emissions is difficult given uncertainties over the climatic and economic processes involved, and the long term over which these processes will play out. As a consequence, available estimates cover a range of values. This paper uses €30 per tonne of CO2 as a lower end estimate of the climate cost (Alberici et al., 2014). Price-based or market-based policies can take the form of taxes or emissions trading systems. The ECRs in this article are the sum of prices put on CO2-emissions from energy use through (a) emissions trading systems (permit prices), (b) carbon taxes, and (c) specific taxes on energy use; see Figure 1. Components of the ECRs can, and often do, equal zero. The base considered is all CO2-emissions from energy use in 41 OECD and G20 countries. Specific taxes on energy use include excise taxes levied on the consumption of energy and a range of other, quantitatively less important, taxes (value-added tax is not included since it is usually not specific to energy). Many of the specific taxes may not have been introduced to curb carbon emissions, but they do put a price on CO2 and therefore are included in the ECRs. The OECD Publications Taxing Energy Use – A Graphical Analysis (2013) and Taxing Energy Use – OECD and Selected Partner Economies (2015a) provide full country-level detail on explicit carbon taxes and all other specific taxes on energy use that impose a price on carbon emissions. They cover 41 countries, including all OECD countries and Argentina, Brazil, China, India, Indonesia, Russia and South Africa. The OECD estimates of carbon prices resulting from emissions trading systems are new. The following emissions trading systems are included in the calculation of effective carbon rates: the Beijing Emissions Trading System (China), the California Cap-and-Trade Program (United States), the Chongqing Emissions Trading System (China), the European Union Emissions Trading System (which operates in 31 countries, of which 23 are OECD member countries), the Guangdong Emissions Trading System (China), the Hubei Emissions Trading System (China), the Korea Emissions Trading Scheme, the New Zealand Emissions Trading Scheme, the Québec Cap-and-Trade System (Canada), the Regional Greenhouse Gas Initiative (RGGI, covering nine north-east and mid-Atlantic US states), the Saitama Prefecture Target Setting Emissions Trading System (Japan), the Shanghai Emissions Trading System (China), the Shenzhen Emissions Trading System (China), the Swiss Emissions Trading Scheme, the Tianjin Emissions Trading System (China) and the Tokyo Cap-and-Trade Programme (Japan). The systems operate in 29 of the 41 countries and cover approximately 13% of total CO2-emissions from energy use in the 41 countries included in the analysis. Figure 1. Illustration of composition of effective carbon rate Coverage Effective carbon rates are calculated for all energy use in 41 countries, which together account for around 80% of world energy use and of global CO2-emissions in 2012, the most recent year for which detailed energy use data are available; see Figure 2. In 2012, the OECD share in CO2-emissions from world energy use equals 35% and that of the seven non-OECD countries 44%. The share of the latter is expected to grow rapidly, along with their share in world output: on the basis of current membership, the OECD’s share in global output may be expected to decline from 62% in 2013 to 43% in 2050 (Johanssen et al., 2013). Effective carbon rates in 2012 – an overview Effective carbon rates, averaged at the economy-wide level (on a weighted basis by the quantity of emissions subject to each individual carbon rate), differ strongly between countries, ranging from approximately €105 per tonne of CO2 to just over zero when biomass emissions are included in the CO2-emissions base (the maximum is €127 when biomass emissions are not included). In 14 of the 41 countries (12 countries excluding biomass emissions), the economy-wide average ECR is less than €30, a low-end estimate of the climate cost of one tonne of CO2-emissions. In all countries, the ECR consists predominantly of specific taxes on energy use other than carbon taxes. In some countries, explicit carbon taxes are a significant component of the ECR, notably in countries with above average ECRs. The prices of emission trading permits, again weighted over the entire base, only lead to a small increase in total ECRs. Continued extension of coverage of emissions by specific carbon pricing mechanisms and raising the rates has the potential to significantly increase the weight of these specific mechanisms in the ECRs, as well as increasing the level of the ECRs. Economy-wide effective carbon rates hide considerable differences in rates across sectors. The average ECR in road transport is much higher than in other sectors. The ECR on non-road transport emissions is low on average – less than €30 in 37 countries (36 countries when biomass emissions are excluded), and less than €5 in 14 countries (13 countries when biomass emissions are not included). This is because specific taxes on energy use are higher in transport than in other sectors. Carbon taxes and emission permit prices have the potential to bring more balance to ECRs across sectors, and can help increase transport rates where necessary. This would better align carbon prices with climate costs and would improve their cost-effectiveness, but would also require considerably higher carbon taxes or permit prices than currently applied in most countries. Figure 3 shows the shares of CO2-emissions from energy use in the 41 countries that are subject to specific ECR intervals. The left panel applies to all emissions in these countries, the middle panel to road transport emissions, and the right panel to all emissions except road transport CO2-emissions from energy use. Considering all energy use (left panel), 60% of CO2-emissions from energy use in the 41 countries are not subject to an ECR at all, 10% are subject to a rate between zero and €5 per tonne of CO2, 20% to a rate between €5 and €30 per tonne of CO2, and 10% to rate above €30. Hence, 90% of emissions are priced below the low end estimate of the climate cost of carbon of €30, and 70% of emissions are priced at a rate of less than €5, implying there is hardly any policy-driven price incentive to reduce emissions. The middle and right panels of Figure 3 show large differences in ECRs between transport and non-transport emissions. In road transport (middle panel), only 2% of emissions in the 41 countries face a zero ECR, and 3% of emissions face a rate larger than zero but less than €5 per tonne of CO2-emissions; 48% of emissions are subject to a rate of between €5 and €30, and 46% of emissions face a rate of more than €30 per tonne of CO2-emissions. These relatively high rates may not have been introduced to mitigate CO2, but still provide an incentive to reduce the tax burden by cutting CO2-emissions. Excluding emissions from road transport (right panel of Figure 3), for 70% of emissions, the ECR is zero, for 11% of emissions it is above zero but below €5, for 15% of emissions it is between €5 and €30, and for 4% of emissions it is at least €30. While the treatment of CO2-emissions from combustion of biomass can have a large impact for some countries, the differences are less pronounced at the level of aggregation considered in this article. The share of zero rates is 4%-point lower in the case where biomass emissions are excluded than in the case where they are included (as in Figure 3) for non-road emissions and for total emissions, but it remains high at 66% and 56% respectively. Concluding remarks Making polluters pay for CO2-emissions from their energy use can be achieved through taxes on energy use (both carbon and other specific taxes on energy use) and through emissions trading systems. The total price on CO2 resulting from the combination of these mechanisms is presented in this paper as the effective carbon rate. The OECD has estimated the ECR for 41 countries, including all OECD countries and Argentina, Brazil, China, India, Indonesia, Russia and South Africa. The evidence on ECRs leaves no doubt that carbon pricing policies are not being utilised to their full potential. Where stringent alternative policies are in place, this means that CO2-emissions abatement policies are likely to be more costly than necessary. Where alternative policies are lacking or weak, it means that current policies do not reflect the climate cost of CO2-emissions. In either case, increased reliance on carbon pricing will allow more ambitious climate policy and better economic outcomes. Increases in ECRs are needed to reach climate goals, and will often engender domestic co-benefits, eg. by reducing air pollution or raising valuable public revenue. At present, excise taxes are the main drivers of ECRs. Carbon taxes and emissions trading systems currently contribute only modestly to effective carbon rates, because of their limited sector and geographical coverage and their comparatively low rates. If coverage continues to expand and rates increase, carbon taxes and emissions trading systems will play a more prominent role in ECRs in the future.
Effective Carbon Rate (€ per tonne of CO2)
... increased reliance on carbon pricing will allow more ambitious climate policy and better economic outcomes


Figure 2. Global CO2-emissions from energy use (2012) Source: OECD calculations based on IEA Energy Balances
Figure 3. Share of CO2-emissions from energy use subject to ECRs of zero, zero to €5, €5 to €30, and more than €30, for all emissions (left), road transport emissions (middle), and all emissions except road transport emissions (right), 2012 –biomass emissions included. Source: OECD calculations based on data from Taxing Energy Use 2015: OECD and Selected Partner Economies (OECD, 2015a) and on emission trading systems.