This chapter analyses tax-based energy price signals across all forms of energy use. The chapter pays special attention to whether energy and carbon taxes favour low- and zero-carbon energy sources over more polluting options. The chapter also discusses the relative tax treatment of gasoline and diesel in road transport. The chapter annex presents energy tax profiles for all 44 OECD and Selected Partner Economies.
Taxing Energy Use 2019
Using Taxes for Climate Action
Taxing Energy Use
2. Energy price signals through taxes
Abstract
The big picture
Effective tax rates on energy use – i.e. the sum of fuel excise taxes, explicit carbon taxes, and electricity excise taxes, net of applicable exemptions, rate reductions, and refunds (see Chapter 1) – vary widely across the world. Figure 2.1 shows the distribution of these effective energy tax rates across two groups of countries (the OECD and the eight selected partner economies covered in this report) and international aviation and maritime transport. Rates tend to be higher in OECD countries than in the partner economies. International aviation and maritime transport are generally not subject to energy and carbon taxes.
Figure 2.1. The distribution of effective energy tax rates across energy use in OECD and Selected Partner Economies and international transport
Note: Tax rates applicable on 1 July 2018. The energy use is for 2016 and adapted from IEA (2018[1]), World Energy Statistics and Balances. Electricity and heating imports are not shown as the associated primary energy use is not known, and to avoid the double-counting of this energy use. The vertical axis is cut off at EUR 25, but the share of energy use taxed higher is negligible.
There are good reasons to tax different forms of energy use differently. Not all forms of energy use impose equal external costs on society, and revenue-raising considerations can also justify different rates. The distribution of effective energy tax rates is indeed highly skewed. Figure 2.1 demonstrates that most energy use is untaxed, whereas rates vary widely across the energy use that is taxed. However, the remainder of this chapter also shows that tax rates are poorly aligned with the polluter-pays principle. More generally, the distribution suggests a need and a merit to simultaneously review both tax rates and tax bases when discussing energy and carbon tax reforms.1
One reason for the highly skewed distribution is that effective energy tax rates vary widely across countries. Figure 2.2 shows average effective energy tax rates for all countries that are covered in TEU 2019, in addition to energy used in international aviation and maritime transport.
Figure 2.2. Average effective energy tax rates by country
Note: All EU member countries levy electricity taxes, but tax rates are not always discernible in the figure. Tax rates applicable on 1 July 2018. The energy data that is used to calculate the weighted averages is for 2016 and adapted from IEA (2018[1]), World Energy Statistics and Balances. Average tax rates do not include electricity and heating imports, in order to avoid the double-counting of this energy use.
In all countries covered, fuel excise taxes are the largest component in the average effective energy tax rate. In fact, fuel excise is the only specific tax on energy use in several countries, namely Australia, China, Indonesia, Israel, Korea, New Zealand, Russia and the United States. In an increasing number of countries, explicit carbon taxes play an important role as well, as is further discussed in Chapter 3. Electricity excise taxes, levied on electricity consumption by end users, are also relevant, especially in those OECD countries that are also members of the European Union.
Incentives for energy savings
What all specific taxes on energy use have in common is that they increase the final price of the taxed energy products.2 Higher energy prices encourage citizens and businesses to consume less energy. Energy savings can result from energy conservation, e.g. heating less in the winter, or shifting to less energy-intensive forms of economic activities.
Energy savings may also result from energy-efficiency improvements, which reduce the amount of energy needed for a given output (products, services, comfort, etc.). Citizens faced with higher energy prices, may, for instance, choose to better insulate their homes than they would if tax rates were lower. Similarly, businesses facing or anticipating higher energy prices can be expected to invest in research, development and deployment of more energy-efficient technologies, which can lead to energy savings, e.g. for aluminium smelters.
Energy and carbon taxes avoid direct rebound effects (Linares and Labandeira, 2010[2]). Direct rebound effects occur because energy-efficiency improvements decrease the cost of using energy-related products and services, which in turn increases demand for these products or services (Small and Van Dender, 2007[3]). Given that standards and related policy instruments encourage, or require, the uptake of more efficient technologies, but do not make their use more expensive, not all energy-efficiency improvements result in net energy savings. Rebound effects generate consumer benefits, but also additional external costs, and are detrimental to welfare when the marginal social costs of extra demand exceed the marginal benefits. Energy and carbon taxes, on the other hand, do not only affect investment decisions, but also increase the marginal cost of using energy-related products and services; they therefore avoid the rebound effect.
The empirical literature confirms that the demand for energy products decreases as prices rise (Labandeira, Labeaga and López-Otero, 2017[4]). This meta-study additionally shows that consumers adjust their consumption patterns more strongly in the long-term than in the short term.
As a result, countries with higher taxes on energy use can be expected to have a lower intensity of GDP, everything else being equal. Figure 2.3 shows that there is indeed a negative correlation between energy taxes and energy intensity of GDP. For instance, Denmark and Switzerland not only have some of the highest average effective energy tax rates, they are also among the countries with the lowest energy intensity of GDP.
Figure 2.3. Countries with higher energy taxes tend to have less energy-intensive economies
Note: Average effective energy tax rates are calculated as in Figure 2.2. The output-based GDP is for 2016 and from the OECD National Accounts database. International transport is not shown as no GDP data is available. The vertical axis is log-transformed.
The selected partner economies, on the other hand, tend to have relatively low average effective energy tax rates – and more energy-intensive economies. This correlation is not exclusively driven by differences in energy and carbon taxes. For instance, countries with less energy-intensive economies also tend to have less stringent regulatory mandates. There may also be structural reasons for differences in energy intensity (e.g. Ireland’s knowledge-based economy, Norway’s hydro power reservoirs, or Russia’s oil and gas reserves), which are not necessarily the result of energy and carbon tax levels.
Providing incentives for energy savings through taxes generally improves environmental outcomes if the foregone energy use would have created environmental damage.3 Environmental damage varies widely by energy source.4 Figure 2.4 shows, inter alia, that countries at the moment still meet most of their energy needs with combustible energy sources (89%), especially coal and other solid fossil fuels, natural gas, as well as oil products, such as diesel and gasoline. Non-combustible energy sources, namely nuclear, hydro and other renewables such as wind, solar and geothermal only represent 11% of primary energy use.5
Figure 2.4. Effective energy tax rates and their carbon benchmarks for 44 OECD and Selected Partner Economies and for international transport
Note: Tax rates applicable on 1 July 2018. The energy use is for 2016 and adapted from IEA (2018[1]), World Energy Statistics and Balances. The energy base does not include electricity and heating imports, for which the primary energy source is not known. Biofuels are marked with an asterisk as CO2 emissions from the combustion of biofuels are considered zero in the greenhouse gas inventories reported under the UNFCCC (see Chapter 1).
In the absence of carbon pricing policies, final energy prices will not reflect the damage carbon emissions from energy use impose on society, and there is also non-climate damage. Combusting fuels releases CO2 and other potentially harmful substances into the atmosphere.6 Typically, consumers use energy products as long as the private benefit of these products is larger than the purchase cost. This results in wasteful consumption of an energy product if consumers’ marginal private cost is below the consumers’ marginal private benefit – provided that marginal private benefit is lower than the marginal social cost associated with using the energy product.
Differences in effective energy tax rates are generally not proportional to energy products’ carbon contents. Specifically, Figure 2.4 shows that for most combustible fuels, effective energy tax rates on average do not reflect even a low-end carbon benchmark of EUR 30 per tonne of CO2 (OECD, 2018[5]). The figure illustrates this by using a black diamond to indicate the tax rates that would result (in EUR per GJ) if countries were to implement a carbon tax set to EUR 30 per tonne of CO2.
The discrepancy between the low-end carbon benchmark and the average effective energy tax rate is particularly pronounced for coal and other solid fossil fuels. This is disconcerting as coal represents the largest share of primary energy use (see also, Chapter 3). Figure 2.4 also shows that there are two energy categories – diesel and gasoline – that are, on average, taxed at a higher rate than the selected low-end estimate of the marginal climate damage of fuel use would suggest. The climate damage is, however, likely to be higher than the low-end carbon benchmark (OECD, 2018[5]). In addition, there are important non-climate reasons for higher tax rates in the road transport sector – which is where the bulk of diesel and gasoline is consumed.
Combusting biofuels also releases CO2 and other pollutants into the atmosphere. However, if sustainably sourced, biofuels may be carbon-neutral over the lifecycle because the biofuels feedstocks have previously absorbed a similar amount of CO2 from the atmosphere. Unlike in Taxing Energy Use (see Chapter 1), CO2 emissions from the combustion of biofuels are considered zero in the greenhouse gas inventories reported under the UN Framework Convention on Climate Change (UNFCCC). The emissions and sinks from biofuels are there instead accounted for as net changes in carbon stocks included in the annual reporting on Land Use, Land Use Change and Forestry (LULUCF), in accordance with the guidelines of the UN Intergovernmental Panel on Climate Change (IPCC). The figure therefore marks biofuel use with an asterisk.
If countries exclusively relied on the tax system to ensure the sustainability of biofuels, they would tax biofuel use based on the carbon emissions that result from their combustion, and subsidise farmers and foresters that sustainably source biofuels based on the carbon that is absorbed by their activities. However, governments generally favour sustainability standards for this purpose, and do not tax, or exempt, most biofuel use outside the road sector. Explicit carbon pricing through carbon taxes or emissions trading typically do not cover or exempt emissions from the combustion of biofuels.
In sum, tax-induced energy savings or shifts to less polluting energy sources can generally be expected to improve environmental outcomes if the energy not consumed would have come from combustible energy sources. Also, both fossil fuel and biofuel combustion additionally cause non-climate damage.
By contrast, with respect to non-combustible energy sources, it is less clear as to whether tax-induced energy savings improve environmental outcomes.7 However, such energy sources – mainly hydro, wind and solar, as well as nuclear – are, on average, also taxed. The reason is that many countries levy output taxes on the use of the electricity generated by these energy sources. As these energy sources do not emit CO2 at the time of use, their carbon benchmark is zero. Local air pollution impacts are equally negligible.
Incentives for switching to cleaner energy sources
Taxing energy use can shift energy demand towards cleaner energy sources. By taxing combustible sources – which do emit CO2 when combusted – at higher rates than non-combustible sources, energy tax systems can provide abatement incentives in support of decarbonisation objectives, and provide co-benefits such as reduced local air pollution.
Such differential tax treatment – effectively putting a surcharge on the use of combustible energy sources – does for instance make it relatively more profitable to switch from vehicles based on an internal combustion engine to electric or hydrogen vehicles. It can also contribute to the electrification of industrial processes, e.g. by increasing the competitiveness of electric arc furnaces in steelmaking. Note, however, that the combustion surcharge is mainly driven by relatively high taxes on road fuels. Last but not least, a combustion surcharge can help direct private and public resources towards the development of new clean technologies (Acemoglu et al., 2012[6]).
Most countries do indeed tax combustible fuels at higher average effective tax rates than non-combustible fuels, as shown in Figure 2.5. The difference between these two tax rates does, however, vary substantially across countries. The combustion surcharge is largest in Iceland, closely followed by Switzerland. Notice that Iceland comes before Switzerland in this comparison, even though its average tax rate on combustibles is lower. The reason is that, unlike Switzerland, Iceland does not indirectly tax non-combustible energy sources because the country does not levy an electricity tax.8
Figure 2.5. Most countries tax combustibles more than non-combustibles
Note: The weighted average tax rates are calculated based on the tax rates applicable on 1 July 2018 and energy use for 2016 that was adapted from IEA (2018[1]), World Energy Statistics and Balances. Average tax rates do not include electricity and heating imports to avoid the double-counting of this energy use.
In Denmark and the Netherlands – which have among the highest overall tax rates on energy use (Figure 2.2) – the combustion surcharge is negative on the other hand, which is also the case for Brazil. The main reason for this is that these countries levy relatively high taxes on the consumption of electricity. Given that most non-combustible energy sources are used for electricity generation, this increases the average effective tax rate of these energy sources.
The combustion surcharge is a summary measure indicating the extent to which overall energy tax system provides incentives to move to cleaner, non-combustible energy sources. Everything else being equal, countries that levy a higher surcharge on combustible fuels can be expected to have lower carbon-intensity of energy use. In countries that tax combustible fuels more, energy use tends to be less carbon intensive. Figure 2.6 shows that there is indeed a negative correlation between the surcharge and countries’ carbon intensity. Iceland has the lowest carbon intensity, followed by Norway and France.
Figure 2.6. In countries that tax combustible fuels more, energy use tends to be less carbon-intensive
Note: Average tax rates are calculated based on the tax rates applicable on 1 July 2018 and energy use data for 2016 that was adapted from IEA (2018[1]), World Energy Statistics and Balances. Energy-related carbon emissions are calculated based on IEA data. Emissions from the combustion of biofuels are included. Average tax rates do not include electricity and heating imports to avoid the double-counting of this energy use. WAV refers to international aviation; WMA to international maritime transport.
While tax-induced energy price signals partly explain the observed differences in carbon-intensities across economies, energy and carbon taxes are not the only explanatory factors. Against this background, it is worth noting that Iceland and Norway benefit from exceptional endowments with renewable resources (hydropower in both countries and geothermal energy in Iceland).
Combustible energy use is also associated with local air pollution. In this case, however, the link between energy price signals and air emission intensity is not as straightforward as with carbon intensity.9 On the one hand, reducing combustible energy use also reduces emissions from local pollutants. On the other hand, air emissions are not directly related to the energy use, but also depend on the end-of-pipe technology used and can vary with the specifics of the combustion process.10 End-of-pipe abatement technologies may, for instance, substantially limit emissions of air pollutants from coal and biomass plants (Portugal-Pereira et al., 2018[7]). Accordingly, Figure 2.7 shows that the linear correlation between the combustion surcharge and air emission intensity is weak (and it is not statistically significant).
Figure 2.7. There is no clear link between the combustion surcharge and air-emission intensity at the country level
Note: Average tax rates are calculated based on the tax rates applicable on 1 July 2018 and energy use data for 2016 that was adapted from IEA (2018[1]), World Energy Statistics and Balances. Average tax rates do not include electricity and heating imports to avoid the double-counting of this energy use. Air emissions data as reported as reported on OECD.Stat.
Electricity taxes and decarbonisation
Electrifying final energy consumption could contribute to decarbonising energy use (IEA, 2018[8]) – provided that electricity generation itself is decarbonised.11 In the road sector, for instance, which causes approximately 18% of energy-related CO2 emissions (see Chapter 3), the technological barriers to further electrification are falling rapidly. The technologies to meet residential and commercial demand for heating and cooling by using electricity already exist today. In other areas, such as high-temperature industrial heat demand as well as aviation and maritime transport (see Chapter 3),12 electrification is more challenging – but not inconceivable (Lechtenböhmer et al., 2016[9]).
Against this background, it is important to note that many countries do not only tax fuels directly, but also tax electricity. This is particularly common in the European Union where at least some form of electricity taxation is mandated by the EU Electricity Tax Directive.13
Electricity taxes typically do not directly encourage power producers to shift to cleaner sources, and do not provide direct incentives for the decarbonisation of the power sector. The reason is that most electricity taxes are not differentiated by energy source, and hence make all energy sources more expensive irrespective of the carbon content. There are exceptions to this rule, however, such as electricity taxes in Ireland and South Africa. In addition, many countries exempt certain small-scale installations that produce electricity for own consumption from the obligation to pay electricity taxes. To the extent that such exemptions benefit rooftop solar or other cleaner sources, electricity taxes do provide some direct incentives for decarbonisation. Electricity taxes also incentivise reducing electricity use in general. In liberalised power markets, fossil fuel-powered generators are frequently the marginal electricity producer. Energy savings induced by electricity taxes could thus indirectly decrease emissions.14
Electricity taxes may decrease the effectiveness of carbon taxes in achieving emission reductions because taxing electricity use makes switching to electricity less profitable for end users, everything else being equal.15 As a result, electricity taxes, as well as other levies and charges not modelled in TEU (see Chapter 1),16 may discourage the electrification of sectors such as road transport and industry.
Fuel excise and explicit carbon taxes, on the other hand, specifically apply to combustible fuels, making fuels more expensive relative to non-combustible energy sources. Given that fuel excise and carbon taxes are directly levied on the energy product, they also provide direct incentives to increase power plant efficiency – unlike electricity taxes.
In fact, most of the primary energy use associated with electricity generation is not subject to electricity taxes, as shown in Figure 2.1. There are three mechanic reasons for this that are generally not driven by environment policy objectives. First, for most energy products, a substantial part of primary energy is lost in the conversion process – it never becomes electricity. Second, electricity taxes typically do not apply to the own use of the electricity industry and electricity that is lost before reaching end users. Finally, not all countries tax electricity use, and those who do, tend to mainly tax residential and commercial use.
Figure 2.8. Electricity taxes fall largely on residential & commercial users
Note: Electricity excise rates are those applicable on 1 July 2018. Energy use data includes all primary energy use associated with electricity generation in 2016, and is adapted from IEA (2018[1]), World Energy Statistics and Balances. The figure includes electricity imports, but does not show electricity exports to avoid the double-counting of this energy use. The y-axis is cut off at EUR 25, but the share of energy use taxed at a higher rate is negligible.
While the environmental case for electricity taxes is arguably decreasing; their relative importance as fiscal policy instruments might be on the rise. Unlike fuel excise and carbon tax revenue, which could eventually disappear, electricity as a potential tax base appears here to stay.
Countries will be reluctant to give up electricity as a tax base, considering that practically feasible tax bases are rare. In fact, there are a number of countries, the Netherlands and Denmark in particular (see Figure 2.2), where electricity tax revenues contribute substantially to government budgets.
One potential strategy to avoid conflicts between environmental and fiscal objectives would be to temporarily replace pre-existing electricity taxes with carbon taxes in such a way that overall energy tax revenues remain constant. At the beginning, the gradual erosion of the carbon tax base could be mitigated by increasing carbon tax rates over time. Eventually, as the energy system is approaching full decarbonisation, electricity taxes could be reintroduced if so desired.
However, the most productive (and politically expedient) use of carbon tax revenue will differ substantially depending on the local circumstances. This suggests a need and a merit to conduct a broader review of domestic tax and spending priorities before deciding on using carbon tax revenues for a specific purpose (Marten and van Dender, 2019[10]). With respect to tax revenues from road transport, recent OECD work suggests, for instance, that gradual tax reforms, shifting from taxes on fuel to taxes on distances driven, can contribute to more sustainable tax policy over the long term (OECD/ITF, 2019[11]).
Sectoral patterns
Tax rates differ depending on the sector where energy sources are used. Figure 2.9 shows that road transport fuels tend to be taxed at far higher rates than energy use in other sectors, which is true for all 44 countries covered in this report (Annex 2.A.). The discrepancy is particularly pronounced when comparing road transport to international aviation and maritime transport (see Figure 2.1), where energy use is not taxed (see also, Chapter 3). In the figure, international transport is grouped under off-road, and is therefore taxed “on average” because domestic aviation and navigation is sometimes taxed, and so are other forms of off-road fuel use, such as use in railways.
Figure 2.9. Effective energy tax rates across sectors
Note: Tax rates applicable on 1 July 2018. The energy use is for 2016 and adapted from IEA (2018[1]), World Energy Statistics and Balances. The energy base includes all 44 countries as well as energy use in international transport. The energy base does not include electricity and heating imports to avoid double-counting. The figure groups energy categories that represent less than 1% of the horizontal axis into “miscellaneous energy use”.
There are good reasons to tax road transport fuels at higher rates than other fuels. Especially in urban road transport, non-climate external costs associated with gasoline and diesel use can be considerable, e.g. because of congestion and local air pollution (Van Dender, (2019[12]); Teusch and Braathen, forthcoming). Fossil fuel combustion in other sectors does, however, also cause air pollution; coal-fired power generation, for instance, can be associated with substantial external costs (Coady et al., 2019[13]). To the extent that more targeted policy instruments are not feasible, excise taxes can be effective policy instruments to make polluters pay for these externalities.
The figure also shows that diesel is, on average, taxed at higher rates than gasoline in road transport. This effect is, however, largely driven by the United States, where this is indeed the case. In fact, Figure 2.10 shows that only three countries tax diesel at higher rates than gasoline, even on a per litre basis. These countries are Mexico, Switzerland and the United States. Further, the United Kingdom taxes diesel and gasoline at the same rate per litre. All other countries effectively tax diesel less than gasoline.
Figure 2.10. Diesel continues to be taxed at lower rates in most countries
Note: Average tax rates are calculated based on the tax rates applicable on 1 July 2018 and energy use data for 2016 that was adapted from IEA (2018[1]), World Energy Statistics and Balances. New Zealand is a special case because diesel vehicles pay distance-based road-user charges, which are not included in TEU because they affect different behavioural margins than energy taxes.
In principle, taxing diesel at higher rates than gasoline is sound environment policy. Effectively putting a surcharge on diesel makes sense from a climate perspective, considering that CO2 emissions per litre of diesel are higher. In addition, non-climate damage per litre of diesel use tend to be higher than for gasoline use. This damage includes environmental externalities such as air pollution, as well as congestion (Harding, (2014[14]); Teusch and Braathen, forthcoming).
In countries where the diesel discount is higher, diesel consumption tends to be relatively higher as well, as shown in Figure 2.11. The effect is, however, only statistically significant if New Zealand is excluded from the sample. New Zealand is a special case because diesel vehicles pay distance-based road-user charges, which are not included in TEU because they affect different behavioural margins than energy taxes. Notice that consumers’ choice for or against diesel vehicles tends to be influenced by regulatory standards17 as well as vehicle purchase and circulation tax structures, among other factors, which may help explain the relatively weak correlation. The weak correlation is also due to the fact that diesel completely dominates among the heavy goods vehicles – irrespective of whether diesel is taxed at a discount or not.
Figure 2.11. Countries with higher diesel discounts tend to have higher diesel shares
Note: The diesel discount was calculated as in Figure 2.10. The diesel share is calculated as the diesel consumption in the road sector divided by the sum of gasoline and diesel consumption in the road sector, whereby all consumption is expressed in litres.
Compared to the previous edition of Taxing Energy Use, some progress has been made in certain countries; Belgium, Turkey and India in particular (Figure 2.12). Such progress is not necessarily driven by rate changes, but in countries where the diesel discount is positive, may simply be the result of inflation that gradually eats away the diesel discount.
Figure 2.12. The diesel discount is diminishing in most countries
Note: Changes are calculated based on constant 2018 local prices. The comparison excludes USA and Canada as data on subnational taxes is not available for 2015. Columbia and Lithuania are missing because they were not yet covered in the previous vintage of TEU. Argentina is missing because no inflation data was available. Sweden is excluded from the chart as the diesel discount is not comparable across time (the tax bases have changed as a result of a new greenhouse gas reduction obligation system that came into force on 1 July 2018).
In a number of countries, the diesel discount increased, however. In Indonesia both gasoline and diesel are subject to the same ad-valorem tax, i.e. 5%; as in 2015 – the change is driven by pre-tax price changes. Estonia increased taxes on both, but less so on diesel. In Brazil, the main difference is that the effective diesel tax rate was reduced to zero, whereas the gasoline rate remained unchanged. In Mexico, which continues to have the lowest diesel discount (the discount is negative as Mexico taxes diesel at a higher rate than gasoline), both rates increased, but gasoline more so than diesel.
Making progress with the diesel discount is challenging, considering that many governments have long encouraged consumers to buy diesel vehicles. In addition to lower fuel excise taxes, there were also tax purchase-related tax incentives that effectively favoured diesel vehicles, such as those based on CO2 emissions (OECD (2016[15]); Teusch and Braathen (2019[16])). It would therefore appear important to give consumers time to adapt to changes that reverse the effect of flawed existing policy signals. Complementary policies, such as investing in public transport or electric vehicle charging stations, or providing other forms of support that would allow owners of old polluting diesel vehicles to switch to cleaner mobility options might thus be warranted.
References
[6] Acemoglu, D. et al. (2012), “The Environment and Directed Technical Change”, American Economic Review, Vol. 102/1, pp. 131-166, http://dx.doi.org/10.1257/aer.102.1.131.
[17] Allcott, H., S. Mullainathan and D. Taubinsky (2014), “Energy policy with externalities and internalities”, Journal of Public Economics, Vol. 112, pp. 72-88, http://dx.doi.org/10.1016/j.jpubeco.2014.01.004.
[22] Alm, J., E. Sennoga and M. Skidmore (2009), “Percect competition, urbanization, and tax incidence in the retail gasoline market”, Economic Inquiry, Vol. 47/1, pp. 118-134, http://dx.doi.org/10.1111/j.1465-7295.2008.00164.x.
[19] Banerjee, A. and T. Besley (1990), Moral Hazard, Limited Liability and Taxation: A Principal-Agent Model, https://www.jstor.org/stable/pdf/2663347.pdf (accessed on 10 May 2019).
[20] Borenstein, S. and J. Bushnell (2018), Energy Institute WP 294 Do Two Electricity Pricing Wrongs Make a Right? Cost Recovery, Externalities, and Efficiency, http://ei.haas.berkeley.edu/support/. (accessed on 18 September 2018).
[13] Coady, D. et al. (2019), “Global Fossil Fuel Subsidies Remain Large: An Update Based on Country-Level Estimates”, https://www.imf.org/en/Publications/WP/Issues/2019/05/02/Global-Fossil-Fuel-Subsidies-Remain-Large-An-Update-Based-on-Country-Level-Estimates-46509 (accessed on 13 May 2019).
[14] Harding, M. (2014), “The Diesel Differential: Differences in the Tax Treatment of Gasoline and Diesel for Road Use”, OECD Taxation Working Papers, No. 21, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jz14cd7hk6b-en.
[1] IEA (2018), “Extended world energy balances”, IEA World Energy Statistics and Balances (database), http://dx.doi.org/10.1787/data-00513-en (accessed on 16 October 2018).
[8] IEA (2018), World Energy Outlook 2018, International Energy Agency, Paris, https://dx.doi.org/10.1787/weo-2018-en.
[21] Kouloumpis, V., L. Stamford and A. Azapagic (2015), “Decarbonising electricity supply: Is climate change mitigation going to be carried out at the expense of other environmental impacts?”, Sustainable Production and Consumption, Vol. 1, pp. 1-21, http://dx.doi.org/10.1016/j.spc.2015.04.001.
[4] Labandeira, X., J. Labeaga and X. López-Otero (2017), “A meta-analysis on the price elasticity of energy demand”, Energy Policy, Vol. 102, pp. 549-568, http://dx.doi.org/10.1016/j.enpol.2017.01.002.
[9] Lechtenböhmer, S. et al. (2016), “Decarbonising the energy intensive basic materials industry through electrification – Implications for future EU electricity demand”, Energy, Vol. 115, pp. 1623-1631, http://dx.doi.org/10.1016/j.energy.2016.07.110.
[2] Linares, P. and X. Labandeira (2010), “Energy efficiency: Economics and policy”, Journal of Economic Surveys, http://dx.doi.org/10.1111/j.1467-6419.2009.00609.x.
[10] Marten, M. and K. van Dender (2019), “The use of revenues from carbon pricing”, OECD Taxation Working Papers, No. 43, OECD Publishing, Paris, https://dx.doi.org/10.1787/3cb265e4-en.
[5] OECD (2018), Effective Carbon Rates 2018: Pricing Carbon Emissions Through Taxes and Emissions Trading, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264305304-en.
[15] OECD (2016), “Israel’s Green Tax on Cars: Lessons in Environmental Policy Reform”, OECD Environment Policy Papers, No. 5, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jlv5rmnq9wg-en.
[11] OECD/ITF (2019), Tax Revenue Implications of Decarbonising Road Transport: Scenarios for Slovenia, OECD Publishing, Paris, https://dx.doi.org/10.1787/87b39a2f-en.
[7] Portugal-Pereira, J. et al. (2018), “Interactions between global climate change strategies and local air pollution: lessons learnt from the expansion of the power sector in Brazil”, Climatic Change, Vol. 148/1-2, pp. 293-309, http://dx.doi.org/10.1007/s10584-018-2193-3.
[3] Small, K. and K. Van Dender (2007), Fuel Efficiency and Motor Vehicle Travel: The Declining Rebound Effect, https://econpapers.repec.org/article/aenjournl/2007v28-01-a02.htm (accessed on 13 May 2019).
[18] Sterner, T. and B. Turnheim (2009), “Innovation and diffusion of environmental technology: Industrial NOx abatement in Sweden under refunded emission payments”, Ecological Economics, Vol. 68/12, pp. 2996-3006, http://dx.doi.org/10.1016/j.ecolecon.2009.06.028.
[16] Teusch, J. and N. Braathen (2019), “Are environmental tax policies beneficial?: Learning from programme evaluation studies”, OECD Environment Working Papers, No. 150, OECD Publishing, Paris, https://dx.doi.org/10.1787/218df62b-en.
[12] Van Dender, K. (2019), “Taxing vehicles, fuels, and road use: Opportunities for improving transport tax practice”, OECD Taxation Working Papers, No. 44, OECD Publishing, Paris, https://dx.doi.org/10.1787/e7f1d771-en.
Annex 2.A. Energy tax profiles
This annex provides energy tax profiles for all 44 countries covered in TEU 2019. General assumptions are explained in Chapter 1. For country-specific assumptions and more fine-grained data, please consult the online country notes. Notice that vertical axes vary widely across countries, depending on how high tax rates are.
Annex Figure 2.A.1. Effective energy tax rates in Argentina
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.2. Effective energy tax rates in Australia
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.3. Effective energy tax rates in Austria
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.4. Effective energy tax rates in Belgium
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.5. Effective energy tax rates in Brazil
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.6. Effective energy tax rates in Canada
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.7. Effective energy tax rates in Chile
Note: Tax rates applicable on 1 July 2018. The explicit carbon tax shown in dark blue refers to the Green Tax (see online country note for modelling assumptions). Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.8. Effective energy tax rates in China
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.9. Effective energy tax rates in Colombia
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.10. Effective energy tax rates in Czech Republic
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.11. Effective energy tax rates in Denmark
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.12. Effective energy tax rates in Estonia
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.13. Effective energy tax rates in Finland
Note: Tax rates applicable on 1 July 2018, assuming use of common diesel (see online country note). Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.14. Effective energy tax rates in France
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.15. Effective energy tax rates in Germany
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.16. Effective energy tax rates in Greece
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.17. Effective energy tax rates in Hungary
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.18. Effective energy tax rates in Iceland
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.19. Effective energy tax rates in India
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.20. Effective energy tax rates in Indonesia
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.21. Effective energy tax rates in Ireland
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.22. Effective energy tax rates in Israel
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.23. Effective energy tax rates in Italy
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.24. Effective energy tax rates in Japan
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.25. Effective energy tax rates in Korea
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.26. Effective energy tax rates in Latvia
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.27. Effective energy tax rates in Lithuania
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.28. Effective energy tax rates in Luxembourg
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.29. Effective energy tax rates in Mexico
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.30. Effective energy tax rates in the Netherlands
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.31. Effective energy tax rates in New Zealand
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.32. Effective energy tax rates in Norway
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.33. Effective energy tax rates in Poland
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.34. Effective energy tax rates in Portugal
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.35. Effective energy tax rates in Russia
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.36. Effective energy tax rates in the Slovak Republic
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.37. Effective energy tax rates in Slovenia
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.38. Effective energy tax rates in South Africa
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.39. Effective energy tax rates in Spain
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.40. Effective energy tax rates in Sweden
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.41. Effective energy tax rates in Switzerland
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.42. Effective energy tax rates in Turkey
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.43. Effective energy tax rates in the United Kingdom
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Annex Figure 2.A.44. Effective energy tax rates in United States
Note: Tax rates applicable on 1 July 2018. Energy use data is for 2016 and adapted from IEA (2018), World Energy Statistics and Balances. The figure groups energy categories that represent less than 2% of the horizontal axis into “miscellaneous energy use”, which is not always labelled.
Notes
← 1. The skewed distribution also implies that average tax rates have to be interpreted with caution, especially at the level of aggregation shown in the figure. The remainder of this report therefore reports average effective tax rates at lower levels of aggregation. The full distribution of unaveraged tax rates are available as online country notes.
← 2. This holds even if taxes are only partially passed through to end users, e.g. in settings where energy suppliers enjoy market power (Alm, Sennoga and Skidmore, 2009[22]).
← 3. The statement only considers the direct effects of the energy savings on environmental outcomes. There could be environmental damage associated with the production and deployment of more energy-efficient technologies, or the revenue raised by such taxes could be employed for environmentally harmful purposes. Such indirect effects can in principle outweigh the benefits resulting from the energy savings.
← 4. For non-climate damage, other factors, such as end-of-pipe technologies, as well as time and place of use are also important in determining the magnitude of the environmental damage.
← 5. The figure somewhat overstates the contribution that combustible fuels make to meeting countries’ energy needs. The reason is that the figure shows energy use in primary energy equivalents as defined in the IEA World Energy Balances (see also, Chapter 1). This implies that a substantial share of combustible energy use is not available for final consumption as it is thermal waste lost in the conversion process. Notably, when coal is converted to electricity, 62% of the primary energy is lost on average – this “thermal waste” never becomes electricity. Most non-combustible sources, on the other hand, do not cause thermal waste, meaning a large part of the primary energy use shown is available for final consumption by end users. Notable exceptions are nuclear power and geothermal energy, where approximately two thirds of the primary energy use are lost.
← 6. Combusting biofuels also releases CO2 and pollutants into the atmosphere. However, if sustainably sourced, biofuels may be carbon-neutral over their lifecycle because the biofuels feedstocks have previously absorbed a similar amount of CO2 from the atmosphere.
← 7. Notice, however, there may be other environmental damage, for example in relation to nuclear waste, that is not necessarily reflected in market prices – e.g. because of power producers’ limited liability, which may warrant corrective taxation (Banerjee and Besley, 1990[19]). Energy savings incentivised by energy and carbon taxes could thus lead to improvements in environmental outcomes, even with respect to non-combustible energy sources. Also note that there are non-environmental reasons for fostering energy efficiency improvements through taxes, e.g. energy security issues for energy-importing countries (Linares and Labandeira, 2010[2]). By contrast, there are better instruments than energy and carbon taxes to deal with the possibility that some energy users undervalue fuel savings (Allcott, Mullainathan and Taubinsky, 2014[17]).
← 8. Iceland introduced an electricity tax in 2010, but discontinued it at the end of 2015.
← 9. The link is even more indirect with respect to the health and environmental damage caused by the energy combustion, which additionally vary with time, location and population density, among other factors.
← 10. Some countries, notably Sweden (Sterner and Turnheim, 2009[18]), directly tax certain air pollutants, but standards are the more commonly used policy instrument to tackle local air pollution.
← 11. From a social welfare perspective, it is also important that the carbon benefits of electrification are not outweighed by other environmental costs (Kouloumpis, Stamford and Azapagic, 2015[21]).
← 12. Norway is in the process of electrifying all domestic ferries (by 2025) – but electrifying long-distance maritime transport remains a challenge.
← 13. EU minimum rates are low, however, relative to fuel tax minima (on an energy content basis). As a result, the existence of electricity taxes in EU countries is not always discernible in the country profiles (Annex 2.A).
← 14. Note, however, that in many countries that rely on electricity taxes, the power sector is additionally subject to an emissions trading systems, which may further decrease the effectiveness of electricity taxes at triggering net emission reductions. Also note that electricity taxes also have the advantage that they can be levied on electricity imported from abroad. In countries where geographic conditions and transmission infrastructure permit, countries may be hesitant to introduce or raise fuel excise and carbon taxes on inputs used to generate electricity, for fear of putting domestic producers at a disadvantage to foreign producers. Sufficiently high effective carbon tax floors agreed among interconnected countries could address such competiveness concerns and ensure a level playing-field between domestic and foreign electricity producers. This would decrease the need to rely on electricity taxes as second or third best instruments to achieve emission reductions from trade-exposed power sectors.
← 15. Carbon taxes would still be cost-effective, but the emission reduction triggered for a given carbon tax level would be lower.
← 16. In addition, pre-tax electricity prices generally do not reflect the marginal private cost of electricity supply. In particular, pre-tax electricity prices rarely reflect the fact that supply costs vary substantially across time. In addition, many fixed costs associated with the provision of electricity, e.g. for network infrastructure, are recuperated at the margin (though network tariffs that are charged on a volumetric basis). Aligning electricity pricing with decarbonisation objectives will therefore require going beyond energy and carbon tax design (Borenstein and Bushnell, 2018[20]).
← 17. These could be environmentally related, such as emission standards for PM and NOx. The recently introduced stricter emission standards regarding NOx emissions have significantly increased the relative price of diesel vehicles, compared to petrol vehicles – especially for relatively small (cheap) vehicles.