9 Carbon price paths

9.1 Introduction

From the terms of reference of the Committee:

There is considerable uncertainty associated with future international climate negotiations, and consequently with future emission prices. By basing the cost-benefit analysis of public measures on uniform assumptions with regard to the future prices of greenhouse gas emissions, everyone preparing calculations for public investment projects will apply the same assumptions in such respect. This will contribute to projects being dealt with in a consistent and comparable manner. The NOU 2009: 16 Green Paper; “Global Environmental Challenges - Norwegian Policy”, recommends that a carbon price path be included in the cost-benefit analysis circular of the Ministry of Finance, thus making this mandatory for all central government cost-benefit analysis. The Expert Committee shall assess the said recommendation and propose potential guidelines for the pricing of greenhouse gas emissions in view of two alternatives; one carbon price path that reflects current expectations with regard to future prices in the EU ETS, and one path that supports the 2-degree target supported by Norway.

Public investments and projects will often have an impact on greenhouse gas emissions1. Transportation projects or new investments in the defence sector may be good examples of this. Increased emissions result in undesired global climate effects and thus impose a cost on other economic agents, or on society. Projects that result in emission reductions will, correspondingly, entail a benefit. All changes in greenhouse gas emissions should be valued for cost-benefit analysis purposes.

Whilst Chapter 4 on changes in real prices focused on how environmental goods can be valued and subjected to real price adjustment for cost-benefit analysis purposes, the present Chapter will specifically address greenhouse gas emissions. A stable climate differs from many other environmental goods, principally, in its global nature and in the possibly irreversible and potentially serious effects of major changes. Emissions have the same impact on the atmosphere irrespective of whether these take place in Norway or elsewhere on Earth. The focus on international agreements and cap-and-trade systems is one of the reasons for discussing greenhouse gas emissions separately, and not together with other environmental goods, in the present Report. This is also in line with the terms of reference.

Because the detrimental effect of greenhouse gas emissions (the social cost of carbon) is independent of the geographic location of such emissions, a cost-efficient climate policy will involve all agents facing one identical price for such emissions. In a theoretical scenario where one joint carbon price is applicable worldwide (i.e. to all emissions), it would be appropriate to apply (the best estimate for) the social cost of carbon for cost-benefit analysis purposes in Norway. No such global carbon price exists at present, and slow progress in the international negotiations under the UN Framework Convention on Climate Change suggests that an international agreement resulting in one global carbon price is unlikely to be a realistic prospect anytime soon.

The absence of a global market for greenhouse gas emissions therefore implies that sufficient mechanisms are not in place to ensure that the marginal cost of reducing emissions (the marginal abatement cost) equals the marginal cost to society of increased emissions (the marginal social cost of carbon). Consequently, the abatement cost varies between sectors and across borders, whilst the global marginal social cost of carbon will always be the same. It may in practice be difficult to estimate the marginal social cost of carbon. It is therefore not obvious what calculation price for greenhouse gas emissions should be applied for cost-benefit analysis purposes in Norway.

A general cost-benefit analysis principle is that calculation prices for private goods should be identical to producer prices. This results in overall production efficiency when the private and the public sector are considered as a whole. However, we will see that carbon prices in Norway vary considerably between different economic agents. Large parts of Norwegian emissions fall within the scope of the EU Emissions Trading System (the “EU ETS”) for businesses, where the carbon price is determined in the cap-and-trade system. Other emission sources are taxed, at tax rates that vary quite significantly between sectors, whilst for example the agricultural sector is exempted from carbon pricing. For public investments, however, it would be reasonable to envisage that all projects should be examined on the basis of identical calculation prices for greenhouse gas emissions. If the same effects are priced differently in different project alternatives, no meaningful profitability comparison is feasible. The current tax structure means that it is not possible to ensure that calculation prices for private goods are identical to producer prices whilst, at the same time, public projects are examined on the basis of identical carbon prices.

NOU 2009: 16 Green Paper; Global Environmental Challenges – Norwegian Policy, provides a comprehensive and thorough account of principles for an appropriate Norwegian climate policy. Chapter 13 of the NOU 2009: 16 Green Paper; Key recommendations and implications for regulatory frameworks, etc., recommends, inter alia, that “central government should formulate a carbon price path to be used for central government cost-benefit analysis purposes and to provide guidance for the regulation of greenhouse gas emissions … The carbon price path should be included in the cost-benefit analysis circular of the Ministry of Finance, and thus be made mandatory for all central government cost-benefit analysis” (page 131). The terms of reference explicitly request the present Committee to consider the above mentioned recommendations from the NOU 2009: 16 Green Paper. Moreover, the terms of reference request the Committee to propose guidelines for the pricing of greenhouse gas emissions in view of the EU cap-and-trade system and the two-degree path.

The Committee will in the present Chapter discuss the issues raised in the terms of references with regard to determining a calculation price for greenhouse gas emissions for cost-benefit analysis purposes, with a focus on the recommendations from the NOU 2009: 16 Green Paper. A number of important topics within the climate area will be discussed rather briefly or not at all in the following. The Committee refers to the NOU 2009: 16 Green Paper for a more general approach and discussion of the climate challenges and Norwegian climate policy.

The structure of the Chapter is as follows: In Chapter 9.2 we briefly discuss the background to the pricing of greenhouse gas emissions, before presenting, in Chapter 9.3, an overview of Norwegian climate commitments and which carbon prices one can observe in Norway today. Chapter 9.4 discusses the current international cap-and-trade systems, as well as the two carbon price paths referred to in the terms of reference. In Chapter 9.5, we discuss principles of relevance to choosing which carbon price path should be use for the cost-benefit analysis of public projects. Chapter 9.6 provides a brief description of the use of carbon price paths in Norway and some other countries, whilst 9.7 and 9.8 present the assessments and recommendations of the Committee, respectively.

9.2 Background to the pricing of greenhouse gas emissions

Greenhouse gas emissions contribute to global warming and climate change. The basis for regulating the emission of CO₂ and other greenhouse gases is that these indirectly impose costs on the international community that the emitter does not take into consideration when making decisions. The emissions therefore represent a global externality (see Chapter 2 for a more detailed description). The “Stern Review on the Economics of Climate Change” (Stern, 2006), which was commissioned by the UK authorities, prepared by Sir Nicholas Stern and published in October 2006 (hereinafter referred to as the “Stern Review”), provided a thorough review of economic aspects of the climate challenge and drew considerable attention. The report refers to climate change as the “greatest market failure the world has seen”.

The authorities have several policy measures at their disposal that may remedy this market failure. One way of getting economic agents to internalise the negative external costs they impose on others is to introduce an emission price that corresponds to such cost. Setting the same price for all greenhouse gas emissions will result in a cost-efficient climate policy, under the assumption of well-functioning markets. There are two main methods for introducing such a price: A tax on greenhouse gas emissions or a market for trading in allotted emission allowances (see Box 9.1). The present Chapter will refer to both taxes and allowance prices as “carbon prices”.

The authorities may also influence emissions through other policy measures, like direct regulations (orders/prohibitions and quality standard requirements), subsidies (for e.g. the development of new technology) and the dissemination of information. In principle, although not always in practice, a marginal cost (or shadow price) of the abatement measures can be calculated in respect of such policy measures as well. In some situations, such policy measures can work well (for example if households have inadequate information), but it can also be the case that implications become difficult to ascertain when several policy measures are in use, and one policy measure may have a negative impact on the effects of another. In such case, the overall cost of achieving emission reductions may be unnecessarily high.

There are also other measures for preventing climate change apart from emission reductions. The technology carbon capture and sequestration seeks to capture CO₂ from, for example, coal power plants and sequestrate it instead of releasing it into the atmosphere. Geo-engineering is concerned with alternative methods for reducing temperatures on Earth (like, for example, making more clouds or removing greenhouse gases from the air). The Committee does not embark on a detailed examination of other climate-related policy measures than allowances and taxes.

9.3 Norwegian commitments and carbon prices in Norway

9.3.1 Norwegian commitments and targets

The UN Framework Convention on Climate Change (the “Climate Convention”) and the Kyoto Protocol are the key binding international legal instruments within the climate area internationally today; see Box 9.2. The ultimate objective of the Convention is to stabilise greenhouse gas concentrations at a level that would prevent undesirable anthropogenic interference with the climate system. Norway has supported a target of limiting the increase in the global mean temperature to two degrees Celsius relative to pre-industrial levels.2

Norway’s commitments under the Kyoto Protocol (1998) play a key role in Norwegian climate policy. A basic premise of the Kyoto Protocol is that those industrialised countries which have ratified the agreement (the so-called Annex B countries) are required to hold a number of UN-approved emission allowances matching each country’s total national emissions during the commitment period. In other words, the countries are subject to a national allowance requirement. If the total emissions of a country exceed the emission allowances such country has been allotted through the agreement, the authorities in the said country will have to acquire additional emission allowances.

In addition to the national emission commitments of Norway under the Kyoto Protocol, Norway also participates in the EU Emissions Trading System for businesses (the “EU ETS”); see Box 9.3. Norwegian enterprises subject to allowance requirements under the EU ETS (the “allowance requirement sector”3) must hand over allowances matching their emissions to the Norwegian authorities. The allowances submitted by such enterprises are used to meet the national allowance requirement under the Kyoto Protocol.4

Norway also has a long-term climate target that is not embedded in international agreements. In Report No. 34 (2006-2007) to the Storting; Norwegian Climate Policy (the first “Climate Report”), the Government formulated the following target (the “30-percent target”):

Norway shall until 2020 make a commitment to reducing global greenhouse gas emissions by an amount corresponding to 30 percent of Norway’s emissions in 1990.

In the same document, the Government also presented a domestic emission reduction target for 2020 as follows:

Based on the analysis of measures carried out by the National Pollution Control Authority5, the sectoral climate action plans, as well as existing policy measures, the Government is of the view that it is realistic to seek to reduce emissions in Norway by 13–16 million tonnes of CO₂ equivalents relative to the reference path as presented in the National Budget for 2007, when forests are included. This would imply that about half, and up to two thirds, of Norway’s total emission reductions are effected nationally.

In addition, the Government called for Norway to be carbon neutral by 2050, i.e. that Norway shall ensure (global) emission reductions corresponding to Norwegian emissions in 2050.

In the Agreement on the Climate Report (2008) (the “Climate Settlement”), the Socialist Left Party, the Labour Party, the Centre Party, the Conservative Party, the Christian Democratic Party and the Liberal Party agreed “that the interval from the Government’s Climate Report can be expanded to 15–17 million tonnes of CO₂e. … This would imply that about two thirds of Norway’s total emission reductions are effected nationally.” The parties also supported the 30-percent target. Moreover, agreement was reached on the following:

As part of a global and ambitious climate agreement under which other industrialised countries also make major commitments, the parties agree that Norway should commit itself to achieving carbon neutrality no later than 2030.

In the wake of the Conference of the Parties in Copenhagen in 2009, Norway also submitted the following expansion of the 30-percent target to the UN:

As part of a global and comprehensive agreement for the period beyond 2012 where major emitting Parties agree on emission reductions in line with the 2 degrees Celsius target, Norway will move to a level of 40 per cent reduction for 2020.

On 25 April 2012, the Government submitted a new Climate Report, Report No. 21 (2011-12) to the Storting; Norwegian Climate Policy (2012). The following is stated on page 14: “The climate targets as specified in the Climate Settlement remain unchanged.”

Proposition No. 111 (2011-2012) to the Storting, submitted on 15 May 2012, states the following (page 74):

The Government will establish, within the scope of the EU guidelines on state aid, a scheme for the reimbursement of indirect allowance costs in Norway, cf. the guidelines being drafted pursuant to Article 10a No. 6 of the revised EU ETS Directive. The Government will revert on how a Norwegian scheme shall be defined after the contents of the EU guidelines have been clarified.

9.3.2 Carbon prices in Norway

In 2012, about 70 percent of Norwegian greenhouse gas emissions will be subject to a carbon price. The prices faced by economic agents in Norway can be divided into two categories:

Source Ministry of Finance

• CO₂ tax: CO₂ tax is payable upon the purchase of petrol, mineral oil and natural gas/LPG. In 2012, about 35-40 percent of Norwegian greenhouse gas emissions were liable to tax. Emissions from the Norwegian petroleum sector are additional thereto (about 25 percent), and are both subject to allowance requirements and liable to tax. (The majority of domestic aviation activities, as well as some uses of natural gas, are also both liable to tax and subject to allowance requirements.)

• Allowance price: Major parts of onshore manufacturing industry fall within the scope of the European cap-and-trade system for enterprises (the “EU ETS”). From 2012, greenhouse gas emissions from the aviation sector are included in the cap-and-trade system. The portion of emissions subject to allowance requirements will be just below 40 percent of total Norwegian emissions in 2012 (inclusive of the emissions from the Norwegian petroleum sector). From 2013, the cap-and-trade system will be further expanded to encompass more manufacturing industry enterprises (including major parts of process industry). The price of the EU allowances is determined within the cap-and-trade system. Enterprises subject to EU ETS allowance requirements may also make use of the Clean Development Mechanism (“CDM”); see Box 9.3.

One will note from Chart 9.1 that there are major variations in carbon prices in Norway. The CO₂ tax on petrol is the highest, at NOK 384 per tonne of CO₂ (NOK 0.89 per litre) in 2012. The process industry has been exempted from taxes and allowance requirements up and until 2012, but becomes part of the EU ETS from 2013. Agriculture, fisheries and hunting are exempted from tax, and are not included in the EU ETS either. The price of EU allowances is recorded in the chart, at NOK 61 per tonne of CO₂e. Additional information on the CO₂ tax structure in Norway is available in Chapters 2 and 4 of Direct Taxes, Indirect Taxes, Customs Duties 2012 (2012).

A differentiated tax structure, like that illustrated by the chart, will increase the economic costs of reducing emissions, and is therefore, generally speaking, not cost-efficient. However, a desire to protect specific industries may lead to a conscious political choice to deviate from the principle that the price structure should be cost-efficiently designed. In Norway, we have for example noted that the agricultural sector is exempted from tax on CO₂ emissions. Traditionally, the prices of greenhouse gas emissions in energy-intensive industries exposed to international competition have been lower than in other industries. The latter may reflect carbon leakage considerations. An implication of a deeply integrated world economy is that measures which reduce emissions in Norway or Europe, may result in increased emissions outside Europe. It may thus be argued that industries particularly exposed to such leakage should be subject to less stringent climate regulations. Bye and Rosendahl (2012) divide the most frequently discussed carbon leakage effects into two categories:

• The energy market effect works via the market for fossil fuels. Climate policy in a group of countries will normally reduce their consumption of oil, gas and coal, which will typically also result in lower prices for these goods internationally. Countries that do not pursue a corresponding climate policy may thus increase their consumption of these fuels.

• The competition effect works via the markets for energy-intensive goods exposed to international competition (for example metals, cement and chemical products). If climate policy results in higher production costs in energy-intensive sectors, the profitability of enterprises in such sectors will deteriorate, and increase correspondingly in countries without a climate policy.

Generally speaking, the amount of carbon leakage is inversely related to the size of the geographic area introducing a higher carbon price. Macro-based analyses indicate 10-30 percent carbon leakage upon the introduction of a unilateral carbon price for large geographic areas like the OECD or the EU (i.e. that each tonne of CO₂ emission reduction due to climate policy in a specific area is accompanied by a 0.1–0.3 tonne emissions increase outside such area) (Bye and Rosendahl, 2012). For smaller areas, like each of the Scandinavian countries, leakage is much higher, and studies indicate that it may reach 60–90 percent (Nordic Council of Ministers, 2012). Micro-based studies tend to generate higher carbon leakage estimates than do macro-based studies (Nordic Council of Ministers, 2012). For energy-intensive industries like cement, iron and steel, such studies indicate 40–50 percent leakage even for large regions (Nordic Council of Ministers, 2012).

We will revert in more detail to criteria for the realisation of climate policy targets and principles for the valuation of greenhouse gas emissions in Chapter 9.5. Reference is also made to Chapter 9 of the NOU 2009: 16 Green Paper.

9.4 International carbon price paths

9.4.1 Current international cap-and-trade systems

Norwegian discussions on current international cap-and-trade systems tend to focus on the EU cap-and-trade system for enterprises. However, other systems are in existence: The Kyoto Protocol cap-and-trade system enables countries to engage in the purchase and sale of allowances between countries with commitments under the Kyoto Protocol. Such allowances are termed AAUs; see the discussion in Box 9.4. In addition, smaller cap-and-trade systems have been established, or are emerging, in the United States, Canada, New Zealand, Japan and Australia.

The allowance price under the EU ETS for enterprises, both for the current year and for the years until 2020, is determined in the market on a continual basis. The allowance price is quoted on a daily basis. There are several market places for trading in EU allowances, including, inter alia, the European Climate Exchange (ICE ECX), NASDAQ OMX (formerly Nord Pool), and the Nordic power exchange (Nord Pool Spot). Spot prices and futures prices are quoted for both EU allowances (“EUAs”) and for approved allowances from CDM projects (“CERs”).

Chart 9.2 shows developments in the spot price of EU allowances (“EUAs”) and CDM allowances (“CERs”) since 2009. Prices have fluctuated considerably. The highest EUA quote during that period was on 2 May 2011 at EUR 17.03 euro per tonne, whilst the lowest quote was on 4 April 2012 at EUR 6.08 euro per tonne. Prices increased throughout 2009 in the wake of the rebound after the financial crisis in 2008, and contracted significantly again throughout 2011, primarily as the result of renewed cyclical downturn.

Source Reuters EcoWin

Political signals and pronouncements have, in addition to macroeconomic factors, energy prices and temperature fluctuations, had a major impact on the EU cap-and-trade system at times. The organisation of the cap-and-trade system and the determination of allowance volumes are key influences on what price clears the market. Expectations of changes in such factors may therefore affect allowance prices. One example of this having a major impact was the EU decision in the summer of 2011 not to increase, for the time being, emission reduction ambitions towards 2020, which resulted in a steep decline in prices. Germany’s decision to bring the phase-out of its nuclear power plant forward to 2022 had the opposite effect. A third example of market reactions to political statements came on 20 December 2011, when the price increased by more than 30 percent because an EU Parliament committee voiced its support for a draft legislative supplement to withhold “a significant” portion of allowances from the cap-and-trade system to bring about more scarcity and a higher allowance price (Wall Street Journal, 21 December 2011, page 4). Alberola and Chevallier (2009) study how the allowance price (during the first phase, 2005-2007) was influenced by various political decisions, especially the decision not to permit the saving of allowances from the first to the second phase.6

Although both EU allowances and CDM allowances have thus far been approved as EU ETS allowances, the CDM allowances have consistently carried a lower price tag than the EU allowances. The price difference may be caused by certain restrictions on the use of CDM allowances (see for example Mansanet-Bataller et al. (2010), which examines this price difference and its underlying causes in more detail).

9.4.2 Future carbon prices

The EU ETS and market prices for the submission of allowances

Future international carbon prices are subject to considerable uncertainty. Global economic developments and the outcome of international climate negotiations in coming years may have a significant effect on the price of EU allowances. However, both economic theory and empirical studies may give us some indications as to how prices will develop in future.

The futures price is the price under a binding contract between two parties for the purchase/sale of an emission allowance at a fixed future delivery date. Enterprises encompassed by the EU ETS are entitled to save allowances from a given year to future years. If we assume that the market is in equilibrium, investors must be indifferent between purchasing an allowance “now” (at the spot price) and purchasing an allowance for delivery in a later year (at the futures price).7

Futures prices for allowances under the EU ETS are quoted for a number of years ahead. Currently observed futures prices therefore provide a basis for estimating future EU ETS spot prices, all else being equal. As an estimate of allowance prices for the period for which no futures prices are quoted, it is possible to use the current spot price extrapolated at an interest rate that reflects the opportunity cost of capital (on the margin) of the investors in the cap-and-trade system (with the same risk).

Model simulations and studies of the cap-and-trade system can also be useful for purposes of obtaining information about future prices. The work relating to Climate Cure 20208 (2010) involved the preparation of a separate sub-report on future allowance prices (Climate Cure 2020, 2009). Three different scenarios for EU ETS allowance price developments until 2020 were presented due to a high level of uncertainty, based on calculations made by Statistics Norway and PointCarbon; see Table 9.1.

The report also discusses allowance prices after 2020, but emphasises that uncertainty in that respect is so high that it chose to start out from prices that were compatible with the two-degree path (see below), rather than explicitly preparing its own allowance price estimates. Climate Cure 2020 assumes a carbon price of EUR 100 per tonne from 2030 onwards, based on the calculations from Point Carbon.

In practice, future allowance prices under the EU ETS are subject to considerable uncertainty. Prices will be influenced by the negotiations relating to a new commitment period under the Kyoto Protocol, which are to be completed by the end of 2012. Ambitious targets may, for example, signal higher prices in future (although it is unclear to what extent the negotiations concerning the second commitment period will have a direct impact on the EU targets).

The current EU emission target is to reduce emissions by 20 percent in 2020 relative to the emission level in 1990. Whether the target should be increased to 30 percent has been discussed internally within the EU for a long time. Low carbon prices for a protracted period of time, both as the result of the financial crisis and as the result of other climate measures introduced by the EU, have given rise to a concern that necessary restructuring desired by the EU is being implemented too slowly. In addition, very low prices may undermine confidence in the cap-and-trade system. One implication of the low prices is that it may be politically more feasible to increase the target, i.e. to lower the allowance cap. In late July 2012, the European Commission proposed that part of the allowances to be auctioned before 2015 be withheld in order to increase the market allowance price. Whether the allowances will be withheld temporarily (and auctioned later) or permanently is under discussion.

The future prospects of the market for CDM allowances are also subject to considerable uncertainty. In 2011, the EU decided not to accept the introduction of additional CDM allowances into the EU ETS from CDM projects registered by the UN after 2012, unless such projects originate in one of UN-designated least developed countries (“LDCs”) or in countries with a bilateral emissions reduction agreement with an EU country. The decision is expected to result in a considerable reduction in the overall demand for CDM allowances after 2012, because the EU countries have thus far accounted for the majority of the overall demand for CDM allowances and because a very small number of CDM projects have thus far been registered in the LDCs.9

The decision at the Conference of the Parties to the UN Framework Convention on Climate Change in South Africa in December 2011 concerning a new commitment period under the Kyoto Protocol for the EU countries, Norway and certain other countries (not clarified as per the date of the present Report) may, on the other hand, contribute to sustained demand for CDM allowances; see Box 9.5. Potential new types of mechanisms that may replace or supplement the CDM, and that may generate larger allowance volumes, are also being examined. It was agreed, at the Conference of the Parties to the Climate Convention in South Africa in 2011, to commence work on a new sectoral mechanism (by sectoral is meant that the mechanism will encompass entire sectors rather than individual projects). The commencement of such work was an important negotiation outcome for the EU. This suggests that one can expect more allowances from countries without commitments to become available in the longer run. Such supply increase may hold back international carbon prices, but may also result in the EU permitting the expanded use of such allowances, as well as introducing a stricter emission target.

Chart 9.3 shows the market price of EU allowances as per 7 September 2012 for the delivery of allowances in December each year for the years until 2020. Average annual nominal allowance price growth, as measured in Euros, is about 5.4 percent between now and 2020.

Market prices as per 7 September 2012 for the delivery of allowances in December each year until 2020. Euros per tonne of CO₂. Nominal prices.

The two-degree path

The Intergovernmental Panel on Climate Change (IPCC, 2007) estimates that stabilising the concentration of greenhouse gases in the atmosphere at about 450 ppm will result in an expected increase in the mean global temperature of two degrees Celsius relative to pre-industrial times.10 It is an overarching objective of Norway to contribute to the climate negotiations organised by the UN resulting in a comprehensive climate agreement that ensures developments in conformity with the two-degree target. Realising this target requires, according to the Intergovernmental Panel on Climate Change, global greenhouse gas emissions to be reduced by 50-85 percent by 2050, relative to 2000. By the two-degree path is meant the measure cost path (the carbon price path) defining, at any given time, the upper cost limit associated with the global measures necessary to realise the two-degree global warming target.

A number of organisations and groups of researchers have performed model-based calculations as to which carbon price paths are required to realise the two-degree target. A joint feature of these calculations is that they highlight the necessity of a future price that is considerably higher than suggested by the EU ETS allowance prices, both spot and futures prices. However, the resulting paths are critically dependent on the assumptions underpinning the analyses. Different estimates with regard to population growth, economic growth, technological developments, substitution opportunities in the event of price differences, as well as electricity consumption, are some of the elements that will influence the calculations. How large a portion of global emissions is facing any given carbon price is of particular importance to the price path. A cost-efficient approach would require all emissions to face the same price. It is debatable how realistic this is, but it remains the traditional approach adopted by most analyses. However, some studies have sought to estimate the cost of not covering all emissions. The authors tend to find a large increase in costs if adopting a non-efficient approach; see Hoel et al. (2009, p. 98) and the NOU 2009: 16 Green Paper (footnote 5, p. 32).

The Italian research initiative International Center for Climate Governance (ICCG, 2011) has prepared an overview of a number of model computations of carbon price paths from 2008–2010 that are consistent with stabilising the concentration of greenhouse gases in line with the two-degree path (450 ppm).11 The paths vary in their assumptions as to which geographic regions participate, what type of climate policy is pursued, rules governing the cap-and-trade systems, expected GDP and population growth, technological improvements, energy types (for example use or non-use of nuclear power), etc. In Table 9.2 below, we have summarised some of the findings from recognised models. In addition, we include the ICCG mean and standard deviation for all model computations.

Mean and standard deviation are reported for all model computations presented by the ICCG (not the selection reported by the Committee). As far as the computations of Paltsev et al. (2009) are concerned, we report the mean across four model simulations with different assumptions as to technological progress. “Model” refers to the name of the model used in the calculations. The ICCG does not specify the base year of the various prices.

The models used in the studies included in the above table are so-called “Integrated Assessment Models” (integrated economy-energy-environment models). These integrate general economic equilibrium models with a detailed modelling of energy demand and production, as well as the pollutant emissions associated therewith. Some of the models also model costs associated with increased greenhouse gas concentrations. These models are used in the analysis of issues within energy economics and environmental economics, such as estimating the costs associated with various climate policy ambitions.

The mean estimated price in 2020 is EUR 43 at 2012 prices (about NOK 323 at an exchange rate of NOK 7.5 per euro). Corresponding figures for 2030 and 2050 are EUR 68 and EUR 235, respectively. There are substantial differences between the estimated price paths, especially for the distant future. The standard deviation in 2050 is EUR 169. All price paths feature a fairly steep increase in the real price over time. The average increase from 2020 to 2030 is 4.6 percent per year, whilst the increase from then until 2050 is about 6.4 percent per year.

The International Energy Agency (“IEA”) publishes thorough analyses of the energy markets on a regular basis. Its most recent published computations (as per the date of the present Report) are presented in the World Energy Outlook 2011 (IEA, 2011). The analysis presents different global climate scenarios. Their “450 ppm scenario” assumes that the EU carbon price will increase from USD 45 per tonne in 2020 to USD 95 dollar per tonne in 2030 and USD 120 dollar per tonne in 2035 (about NOK 260, 550 and 700, respectively). In the United States, Canada, Japan, South Korea, Australia and New Zealand, the price starts out at a somewhat lower level, before converging to the price calculated for the EU in 2035. For China it is assumed that the carbon price in all sectors will increase from USD 10 per tonne in 2020 to USD 65 per tonne in 2030 and USD 95 per tonne from 2035. For Brazil, India, Russia and South Africa, a carbon price in 2035 of USD 95 per tonne is assumed, but in these countries such price applies to the power sector and to manufacturing industry only. Moreover, it is assumed that many other developing countries will also put a price on major parts of their emissions. The IEA does not provide any carbon price computations beyond 2035.

The findings from computations generated by comprehensive, long-term models of the type presented above are debated extensively. Much of the debate on climate models in economic literature has been concerned with the discount rate. For projects and investments with long lifespans, it is especially important how costs and benefits at different points in time and, in particular, in the distant future are compared and traded off against each other. This will influence what is held to be an optimal emission target, but also the price paths for realising specific targets. In Chapter 5, the Committee discusses issues relating to the discount rate. In addition, a debate has arisen with regard to dealing with potentially catastrophic climate effects. This is discussed in Chapter 8.

9.5 Principles for the valuation of greenhouse gas emissions in cost-benefit analysis

9.5.1 Introduction

In order to reduce greenhouse gas emissions to what is, in global terms, an optimal level, marginal emissions should carry a price tag corresponding to the global economic cost associated with such emissions (hereinafter referred to as the “social cost of carbon”). If such price applies to all economic agents and to all emissions, and no agents have market power, the behavioural decisions of all will result in the marginal abatement cost (the cost of reducing emissions by one unit) being equal to the social cost of carbon. However, it is not particularly realistic, as mentioned above, to expect such a system to be established in the foreseeable future. Consequently, the pricing of greenhouse gas emissions for cost-benefit analysis purposes is a more complex issue.

The principles that should underpin the pricing of greenhouse gas emissions for cost-benefit analysis purposes therefore depend on which questions one would like the cost-benefit analysis to answer. These may depend, as we will note below, on both preferences and policy, including the current national and international climate policy situation and how it develops. It will, in particular, be decisive whether emissions from a project will result in increased global emissions or whether the increased greenhouse gas emissions are offset by emission reductions elsewhere.

We will here present and discuss various principles for the valuation of greenhouse gas emissions for cost-benefit analysis purposes. Two of these are based on the two-degree path and international carbon prices (EU ETS allowance prices), respectively, cf. the terms of reference of the Committee. In addition, we deem it relevant to discuss carbon prices derived from declared or actual climate policy in Norway at present. Once a principle for choosing calculation prices for greenhouse gas emissions for cost-benefit analysis purposes has been decided on, one needs to apply (best estimates) for such prices for all years of the analysis period. As discussed in Chapter 3, real prices (i.e. prices adjusted for inflation) are what should be used for analysis purposes.

As a first step towards discussing the principles behind various calculation prices, we will outline some carbon pricing principles. We will start out from the discussion in the NOU 2009: 16 Green Paper on how greenhouse gas emissions should be priced in Norway, based on various objectives and preferences.

9.5.2 Carbon pricing principles and findings from the NOU 2009: 16 Green Paper

The NOU 2009: 16 Green Paper discusses various potential objectives for Norwegian climate policy. One such objective is for Norway to comply with its international agreements and commitments at the minimum possible cost. However, formal compliance with commitments is not tantamount to making a contribution to reduced global greenhouse gas emissions. The NOU 2009: 16 Green Paper notes that carbon leakage issues (see Chapter 9.3.2 above) and challenges relating to the CDM mechanism may imply that actual emission reductions are less than would be credited under an agreement (see Chapter 9.5.5 below).

Another objective for the authorities may be for Norway to comply with its Kyoto Protocol and other commitments through measures that result in actual global emission reductions. The NOU 2009: 16 Green Paper concludes that this seems to be the most realistic objective, and its discussion is therefore based thereon.

The NOU 2009: 16 Green Paper also discusses whether the authorities may have other objectives beyond those resulting from international agreements and commitments. The first Climate Report (see Chapter 9.3.1) specified that it was realistic to aim for two thirds of Norway’s contribution to global emission reductions in 2020 to take place domestically. The NOU 2009: 16 Green Paper briefly discusses the extent to which the said objective can be held to guide the use of policy measures in Norway.

The NOU 2009: 16 Green Paper concludes that taxes and tradable allowances are suitable policy measures for the regulation of greenhouse gas emissions. If the objective is to limit Norway’s contribution to global emissions, and carbon leakage is disregarded, it will be cost-efficient to apply the same carbon price in all sectors. As large parts of Norwegian emissions fall within the scope of the European cap-and-trade system, this should define the “reference price”, and emissions from other sectors should be subject to tax at a rate corresponding thereto. Furthermore, the NOU 2009: 16 Green Paper notes that since the price of greenhouse gas emissions for enterprises subject to allowance requirements will vary from day to day, it would however be particularly challenging to charge the same prices for greenhouse gas emissions both within and outside the allowance requirement sector.12

Carbon leakage considerations may suggest that sectors where major leakage is expected might merit a lower carbon price. However, the NOU 2009: 16 Green Paper finds that the “combination of inadequate information about technological possibilities and relocation costs and the paucity of effective policy measures mean that the environmental gains from a policy aimed at carbon leakage seem highly uncertain”, and therefore recommends that all sectors be regulated with equal strictness.

The NOU 2009: 16 Green Paper also discusses separate domestic emission reduction targets. The following is stated at the beginning of such discussion:

A domestic target implies that the authorities regulate domestic emissions more strictly than is required to achieve a cost-efficient solution. Any global agreement that reduces global emissions by an amount sufficient to materially reduce the risk of major and irreversible detrimental effects, will result in major domestic emission reductions as the allowance price increases, cf. the discussion in Chapter 4. Hence, the arguments in favour of domestic targets cannot rely exclusively on the need for domestic reductions to comply with a strict international agreement, because such reductions over time will also result from a cost-efficient policy. A domestic target must instead mean that domestic emission reductions shall be implemented earlier than would be the case under a cost-efficient policy.

Two possible reasons for having such targets are then discussed:

• One argument is that domestic enterprises and households are not paying sufficient heed to the fact that the carbon price will increase over time. Such a line of argument is based on the premise that the authorities have better information than private agents about which emission reductions should be implemented now. The report notes, however, that it seems less plausible that domestic emitters will generally fail to take note of expected long-term price developments in a situation where a large (and growing) portion of Norwegian greenhouse gas emissions is regulated via the EU cap-and-trade system.

• Another argument in favour of domestic targets may be that such targets make it easier to establish an international agreement. Early adoption of climate regulations may give rise to a form of positive externalities, in the same way as when an individual country devotes itself to important R&D measures. The report notes, however, that the effect of such initiatives on climate regulations in other countries is difficult to measure.

In the event that the authorities have such objectives, the NOU 2009: 16 Green Paper recommends that the domestic objective be formulated as a price target, and not as a quantitative target, in view of the uncertainty about costs. Moreover, it is recommended that the authorities announce how strict regulations will be over a certain period, thus relieving emitters of the risk of sudden changes in regulations. The NOU 2009: 16 Green Paper recommends, irrespective of whether the target is formulated as a price or a quantitative target, that tax be used as the policy measure (for enterprises in the allowance requirement sector this implies an additional tax imposed on top of the allowance price, such as to make the sum of these equivalent to the tax level outside the allowance requirement sector).13

In subsequent sections we will discuss in which situations and for which climate policy objectives the various carbon price paths should be used as calculation prices for cost-benefit analysis purposes. This is also discussed in the NOU 2009: 16 Green Paper. It is there noted that finding the optimal calculation price in a small, open economy is, generally speaking, simple if an international price exists and one does not value national emission reductions higher than global ones: “The international price should be used as the calculation price for all domestic decisions. With a global climate agreement resulting in a joint international carbon price, the international price will represent the cost of Norwegian greenhouse gas emissions, and thus the basis for all domestic use of policy measures” (page 90).

The NOU 2009: 16 Green Paper goes on to note that the decisive factor in the valuation of greenhouse gas emissions, in the absence of such a global agreement, is how Norway meets its emission commitments on the margin. For Norway this has, until 2012, been through internationally available allowances (primarily CDM allowances). The argument is that Norway is a price taker in the cap-and-trade system, and that the cheapest way of reducing greenhouse gas emissions is therefore to implement all measures in Norway that carry a cost below the allowance price, and thereafter to effect any additional emission reductions through allowance purchases abroad.

In subsequent sections we will discuss in which situations and for which climate policy objectives each carbon price path will apply. We will first discuss whether current producer prices for greenhouse gas emissions could or should also be used for public cost-benefit analysis purposes.

9.5.3 Differences in actual prices, identical calculation prices?

A general cost-benefit analysis principle is that calculation prices for private goods should be identical to producer prices. This results in efficient resource allocation between the private and the public sector. A project that features private goods only will thus be subject to the same profitability assessment in both the private and the public sector.

No further corrections shall, as noted in Chapter 7 of the NOU 2009: 16 Green Paper, be made in cost-benefit analysis in respect of externalities when such externalities are adequately reflected in producer prices (through taxes or allowance prices). In other words, in the presence of externalities (like greenhouse gas emissions) that are internalised in private decisions through the use of taxes (or cap-and-trade systems), the public sector cannot improve on the market solution by using calculation prices that differ from producer prices inclusive of tax.

Hence, whether a cost-benefit analysis shall disregard the taxes applicable in the various sectors, and instead apply one joint carbon price path for the valuation of emissions in all sectors, depends on whether one believes that current taxes adequately reflect external costs. One may argue, on the one hand, that the tax structure chosen by the authorities is in actual fact carefully “thought through” by the authorities and that, for example, carbon leakage considerations suggest that the differentiated taxes seek to meet the requirement that producer prices should adequately reflect external costs. In such case, these prices should be used as calculation prices for cost-benefit analysis purposes. It should be noted that a challenge under this approach is that information about future policy is required to prepare estimates for future carbon prices in Norway.

It may, on the other hand, be argued that the current tax structure is the result of political prioritisations that have nothing to do with the external cost of greenhouse gas emissions, and therefore should be excluded from cost-benefit analysis. In such case, the current tax structure should not be used for cost-benefit analysis purposes. It would in that case be preferable to discuss the implications of the measure in other policy areas in other ways suitable in the analysis context. For example, relevant distribution effects should be discussed separately, cf., inter alia, Chapter 2.3 and Chapter 3 on distribution.

If one uses calculation prices that differ from producer prices, one will have to estimate the actual overall changes in greenhouse gas emissions resulting from a project. Thereafter one will need to multiply ultimate net changes in emissions by the calculation price used, less any tax or allowance price already paid in respect of the emissions. The global and ultimate changes in greenhouse gas emissions resulting from an investment, as effected through market input-output mechanisms, can often be very difficult to calculate, and it will normally be necessary to introduce simplifications. If producer prices are used instead, the probability calculations can be based on observable prices.

9.5.4 Global marginal social cost of carbon

If we use a calculation price based on the marginal social cost of carbon; i.e. the global cost of emitting one additional tonne of CO₂e, cost-benefit analysis will measure the net willingness to pay for a project, provided that the greenhouse gas emissions from such project are not offset by other measures – and thus actually impose global climate-related damage. In addition, it is implicitly assumed that the willingness of persons in both Norway and other countries to pay to avoid climate change shall be taken into account in the assessment of Norwegian projects.

However, estimating this social cost of carbon is convoluted and complex. Firstly, the social cost of carbon depends on future emissions. If one does not succeed in curtailing emissions, it is likely that the marginal social cost of carbon will be higher. Other issues relate to the discount rate (see Chapter 5), risk aversion and so-called “tipping points” or catastrophic events (see Chapter 8). In addition, there are challenges associated with how to interpret the global willingness to pay as a measure of the welfare effect of preventing damage, since the poorest part of the world’s population will have little scope for signalling its interests via its willingness to make monetary payments. A particularly thorny issue is, for example, the valuation of lives lost in poor countries.

A recent attempt at estimating a global marginal social cost of carbon was carried out at the behest of the authorities in the United States in 2009 (Interagency Working Group on Social Cost of Carbon, 2010). The estimates are strictly dependent on the discount rate used. The point estimates for 2010 (the first year presented by the working group) vary between USD 4.7 and 35.1 at 2007 prices (as per 2010). See also Bell and Callan (2011) for a discussion of this and other estimates.

9.5.5 Necessary cost of realising a given emission target

A physical emission target means that Norwegian emissions are subject to a binding cap, and that it is not permissible to exceed such cap.14 If a project or measure results in emissions, emissions in the remainder of the economy must therefore be reduced correspondingly. Alternatively, the question may be posed as follows: What is the net willingness to pay for the project, provided that emissions must be reduced correspondingly elsewhere?

It should be noted that the global marginal social cost of carbon (discussed in 9.5.4) is not of relevance to this question, because it is assumed that total emissions will not increase, thus implying that the project will not give rise to any change in global damage. This question may then be differentiated on the basis of what emission target is being considered. The stricter is the emission target, the higher will be the abatement cost in the economy to begin with, and the higher will be the cost of reducing emissions elsewhere.

Necessary cost of realising the two-degree target

At present, there is a consensus that the international climate negotiations shall proceed on the basis of the two-degree target. Norway has, as previously mentioned, expressed its support for this target. The target necessitates, according to the Intergovernmental Panel on Climate Change (IPCC, 2007), global emission reductions of 50–85 percent by 2050, relative to the 2000 level. The reduction in overall emissions from rich countries must be 25–40 percent by 2020 and 80–95 percent by 2050, relative to the 1990 level.

Several attempts have been made, as we noted in Chapter 9.4, at calculating what prices are required to enable the world to realise this target. If one chooses to apply such prices for cost-benefit analysis purposes in Norway, it implies that one poses the following question: What will be the net willingness to pay for the project, provided that emissions from the project are offset by emission reductions elsewhere in the economy, in a situation where Norway’s domestic greenhouse gas emissions are determined by a hypothetical international agreement in conformity with the two-degree target?

In other words, the valuation of greenhouse gas emissions we use to answer such a question will be determined by the costs necessary to comply with a given emission target (in this case Norway’s “share” under a hypothetical agreement), and not what it will cost the world if one does not realise such target. The latter question concerns the social cost of carbon; see 9.5.4. Only if it turns out that the two-degree target implies an optimal trade-off between the social cost of carbon and abatement costs, will the prices implied by the two-degree path and the prices implied by the estimated marginal social cost of carbon be identical.

One may envisage several reasons for using a price path for Norwegian cost-benefit analysis purposes that is premised on the two-degree target. Firstly, the above question will be of interest if we are in a situation where an international agreement in line with the said target does actually exist, and where the Norwegian authorities intend to comply with such agreement. In that scenario, such a carbon price will represent the actual and necessary cost of offsetting the emissions from the project elsewhere in the economy. In such a scenario it is also quite conceivable that this price path will be reflected in international cap-and-trade systems, and that the Norwegian tax level will be calibrated with such target in mind. In such case it will be appropriate to apply this price path for cost-benefit analysis purposes.

No such agreement exists for the time being. Another reason for applying such price path for cost-benefit analysis purposes might be that Norwegian authorities would like to act as if such an agreement existed. It may, for example, be that Norwegian authorities would like to send political signals with regard to credible climate policy by acting as if Norway was encompassed by such an agreement, or because they believe that it may contribute to other countries also choosing to act in the same manner. In such case on should also change the tax level (and not only calculation prices for cost-benefit analysis purposes) in line with a hypothetical agreement of this nature.

Thirdly, it is conceivable that the authorities may wish to analyse potential projects as if such an agreement existed, even though taxes and other policy measures are not calibrated to realise any such target in practice. This may be interesting as a hypothetical exercise, but will imply that project assessments are made on the basis of principles that are not in conformity with the other practical policies pursued by the authorities.

It should be noted that, as long as no international agreement exists, the basis for using prices in line with the two-degree path must either be defined national targets, or assumptions to the effect that a group of countries that includes Norway will face an allowance price resulting from analyses of the type presented in Table 9.2.

Necessary cost of meeting existing emission targets

Although much remains to be decided about Norway’s next commitment period under the Kyoto Protocol, as per the date of the present report, it has been established that Norway will also after 2012 be seeking to comply with an international emission commitment (until and including 2017 or 2020). In addition, objectives have already been formulated with regard to the limitation of domestic greenhouse gas emissions by 2020.

A question of obvious interest from a cost-benefit analysis perspective is therefore what is the net willingness to pay for a project, provided that emissions will have to be reduced correspondingly elsewhere to meet existing binding emission targets?

If “existing binding emission targets” is here taken to include climate policy considerations that have not been incorporated, or that are in the process of being incorporated, into practical policy, one will be faced with the question of the legitimacy of such targets, and of whether calculation prices shall be based on said emission targets or on the policy measures in actual use. Consequently, additional clarification as to the meaning thereof will be required.

International allowance price as calculation price

In a situation where Norway is subject to an international emission commitment, Norway (i.e. the Government of Norway) will have the option of purchasing allowances from abroad. If the binding emission targets of the authorities only quantify Norway’s contribution to global emission reductions, without any additional requirements as to where and how this is to be achieved, the international allowance price represents the marginal opportunity cost of emission reductions applicable to Norway as a country. As long as the Government of Norway is in a position to purchase EU allowances (or similar), the abatement cost of any project implemented in Norway may be measured by using the international allowance price as a yardstick.

If Norway would like to reduce emissions where the cost of doing so is the lowest, irrespective of location, it would imply that all measures with a cost below the allowance price are implemented in Norway, and that any abatement measures on top of that are effected via the purchase of allowances in the international market. Such a policy means that the international allowance price will reflect the price necessary to realise existing emission targets.

However, one important objection may be raised in this regard. If the systems for the measurement and control of emission reductions abroad are inadequate, it may mean that the purchase of international allowances does not represent actual emission reductions. In such case the international allowance price will not represent the appropriate opportunity cost of national emission reductions either. Economic literature has observed that this may be a problem as far as CDM allowances are concerned. Hagem and Holtsmark (2008) have, for example, discussed problems relating to leakage and compliance with the additionality condition (the requirement that the project triggering the emission reductions would not have been implemented had it not been for the income from the sale of allowances) when using the CDM mechanism. However, the Committee is not aware of any criticism to the effect that the purchase of EU allowances does not represent actual emission reductions.

If an international allowance price is used as the calculation price, one will also have to examine how such price will change over time in real terms. The EU allowance price changes on a daily basis, and the uncertainty associated with international negotiations and European climate policy is significant enough to make it challenging to project allowance price information from the market 20–30 years into the future, cf. the discussion in Chapter 9.4.2. There is also considerable uncertainty with regard to the period up to 2020; the market for trading in allowances for delivery in the years close to 2020 is fairly thin, and the Committee is not aware of whether any market for the delivery of allowances after 2020 does exist.

Necessary cost of domestic emission reduction targets

It was assumed above that the binding emission targets of the authorities only concern the quantity of Norway’s contribution to global emission reductions. If the authorities pursue a climate policy that involves additional real restrictions, such additional restrictions will influence the cost of offsetting the greenhouse gas emissions of a project through reductions elsewhere in the economy.

The Climate Settlement (see Chapter 9.3.1) contains wording to the effect that two thirds of the emission reductions in 2020 shall take place domestically. If this is implemented, it will increase the marginal cost of compensating for the greenhouse gas emissions of projects. In such case the international allowance price cannot, as a matter of course, be taken to represent the marginal abatement cost. If the domestic abatement requirement is just met to begin with, the domestic emissions of the project will have to be compensated by domestic emission reductions, irrespective of whether these are more expensive than paying the international allowance price. If a “binding emission target” defines requirements with regard to both the contribution to global emission reductions and the domestic portion thereof, the domestic marginal cost, and thus the calculation price, will probably exceed the international allowance price.

One challenge in basing calculation prices on politically defined targets is precisely the fact that it requires the authorities to formulate credible emission targets for a sufficiently long time horizon. An issue arising in this context is what weight to attribute to a domestic emission target if one observes that current climate policy is not in conformity with such target. It may in Norway be argued that the Government currently has a domestic emission target for 2020 (cf. Chapter 9.3.1), but no such target for subsequent years. When the central government working group Climate Cure 2020 examined policies and measures for realising such target, it indicated that the domestic marginal cost was NOK 1,100 – 1,500 per tonne of CO₂e (Climate Cure, 2010). If it is assumed that no additional measures will be implemented within the allowance requirement sector, the calculations from Climate Cure 2020 showed that the necessary marginal cost will be significantly higher. As long as actual taxes are not adjusted in line with the domestic emission reduction target, it is not obvious that such target can be considered, for the time being, to constitute a real, binding restriction on Norwegian greenhouse gas emissions.

9.6 Current use of carbon prices in cost-benefit analysis in Norway and abroad

9.6.1 Use of carbon prices in Norway

The Cost-Benefit Analysis Guide of the Ministry of Finance (2005) includes a brief discussion of dealing with greenhouse gas emissions in a cost-benefit analysis context. The following is stated on page 24:

Global environmental problems are linked, inter alia, to greenhouse gas emissions. In such cases national targets will often be related to international commitments. A national commitment can be handled through a tax system that reflects the economic costs of compliance therewith. The environmental taxes should in such case be considered an established fact for cost-benefit analysis purposes. If the commitment can alternatively be met through the purchase and sale of allowances internationally, the analysis can instead be based on the international allowance price. Alternatively, one may use model computations, etc., to estimate the value of an allowance. It is, as a general observation, important to take into account the possibility that calculation prices may change over time as the result of changes in international environmental targets.

The practical use of carbon prices and carbon price paths in the analysis of public investments in Norway is more clearly outlined in the various sectoral cost-benefit analysis handbooks. The Norwegian National Rail Administration has recently revised its guide from 2006. It follows from Method Handbook JD 205 Cost-Benefit Analysis for the Railways (Norwegian National Rail Administration (2011), pages 81-82) that no greenhouse gas emission estimates are included in the analysis of new railway investments in Norway, whether in direct operations or in the production of the propulsion energy. The reason given for this by the Norwegian National Rail Administration is that it purchases certificates of origin for electricity and that payment for these “is used to develop additional production capacity at a named hydroelectric power plant, thus implying that the indirect contribution to increased demand for polluting power is eliminated” (page 82).

The Norwegian National Rail Administration has informed the Committee that it assumes, for purposes of most analyses, that emissions in the construction phase and in the production of rolling stock have in the long run been captured by taxes and regulations. In other words, that the environmental costs are implicit in the estimated construction costs and capital costs of rolling stock, respectively. Moreover, the Norwegian National Rail Administration states that it will, during the course of 2012, be revisiting the assumption that the purchase of certificates of origin makes no contribution to increased demand for polluting power.

For other means of transport (and diesel-powered trains), the Norwegian National Rail Administration uses estimates from the Climate Cure 2020 medium scenario for the EU ETS allowance price; see Table 9.1. Global air pollution rates per vehicle kilometre are based on the said prices per tonne, as well as on emission factors for different means of transport based on calculations from the former National Pollution Control Authority.

The Norwegian Public Roads Administration conducts climate analyses using the EFFECT estimation software. In the most recent version (as per the date of the present Report); EFFECT 6.4, the Climate Cure 2020 medium scenario has also been used: For the years until and including 2015, a fixed unit price of NOK 210 per tonne is used. For the years between 2015 and 2030, a unit price is estimated by interpolation between the specified years and their appurtenant values. For years from 2030 inclusive, a fixed unit price of NOK 800 per tonne is used. The Norwegian Public Roads Administration also notes that it estimates emission costs associated with construction, emission costs associated with maintenance changes (primarily altered quantities of bitumen and electricity), as well as emission costs associated with changes in traffic emissions.

The use of carbon prices is also of relevance outside the transportation sector. The sectoral guide for cost-benefit analysis in the defence sector (Ministry of Defence, 2010) does, for example, encourage the use of market data for carbon prices for purposes of valuing greenhouse gas emissions.

9.6.2 Recommendations on the use of carbon prices in other countries

The recommendations on the pricing of greenhouse gas emissions abroad vary from country to country:

• Sweden adheres to the transportation sector guidelines in ASEK 5 (2012). These recommend using a calculation price of SEK 1,080 per tonne for short-term investments (the current petrol tax) and SEK 1,450 per tonne for long-term investments (the net present value of the petrol tax, real price adjusted on the basis of GDP per capita). One is also recommended to prepare sensitivity estimates based of SEK 3,500 per tonne.

• The United Kingdom has changed its method for the valuation of greenhouse gas emissions. Prior to 2009, the estimated global social cost of carbon was used, but one has now switched over to pricing in line with the necessary marginal cost of meeting long-term domestic emission reduction targets in conformity with the EU Climate and Energy Package. Until 2030, different prices are used for different sectors, depending on whether these are within or outside the EU ETS. The calculation prices for sectors encompassed by the EU ETS shall reflect the expected market price of allowances, whilst prices for other sectors are in line with national emission targets. The UK authorities have specified which prices to use until and including 2100. However, preparation and presentation of sensitivity analyses is also encouraged; see Table 9.3.

• HEATCO, an EU project for harmonising the valuation of transportation projects (HEATCO, 2006), recommends using identical prices for all analyses in Europe. The guidelines are based on the social cost of carbon, as well as on abatement costs. It is emphasised, in particular, that future emissions will involve a higher social cost of carbon than current emissions, which indicates, all else being equal, an upwards sloping price path (reference is made to Watkiss et al. (2005a)). HEATCO presents, on the basis of computations made by Watkiss et al. (2005b) for the Department for Environment, Food and Rural Affairs in the United Kingdom, a table recommended for use in cost-benefit analyses in Europe (see Table 9.4).

9.7 The assessments of the Committee

In order to ensure efficiency in production, the same prices must be applied to private and public decisions. Hence, the same applies to calculation prices for cost-benefit analysis purposes. If the taxes on greenhouse gas emissions had been such as to adequately adjust for the externalities associated with greenhouse gas emissions, subject to any domestic targets and carbon leakage considerations, these producer prices should also have been used for cost-benefit analysis purposes.

At present, carbon prices in Norway differ so much between sectors and between different types of fossil fuels that it is difficult to justify such differences by carbon leakage considerations. Consequently, the current differentiated taxes are not well suited as a basis for establishing calculation prices for greenhouse gas emissions for cost-benefit analysis purposes. Instead, the Committee recommends that one joint carbon price path be used in the cost-benefit analysis of public decisions across all sectors.

What is the correct calculation price for greenhouse gas emissions depends on what question one would like the analysis to answer. It will in many cases be appropriate to perform calculations using different calculation prices, which calculations will in such case address different issues. This applies, in particular, to analyses of projects and measures which are motivated by climate considerations or which have major emission implications. However, since the climate issue will often be only one of the many considerations reflected in a cost-benefit analysis, it is necessary to choose one price path to represent the main alternative.

A decisive question when choosing a calculation price for cost-benefit analysis purposes is whether emissions from a project will result in higher global emissions, or whether the greenhouse gas emissions increase will be roughly offset by emission reductions elsewhere. If the emissions from a project result in a corresponding global emissions increase, the relevant calculation price is the global marginal social cost of carbon. If the emissions increase is offset by reductions elsewhere in the economy, the marginal abatement cost is the relevant basis for determining the calculation price. Which abatement cost paths should be applied in such cases depends on the context:

Two-degree global warming target: If an international agreement in conformity with the two-degree target exists, or if Norwegian authorities have expressly stated that they will act as if such an agreement exists, this may be held to define a limitation on Norwegian emissions. The marginal abatement cost inferred from such target should therefore be used as the calculation price of greenhouse gas emissions for cost-benefit analysis purposes.

Domestic targets: If the authorities have adopted, and comply with, national emission reduction targets, calculation prices may in principle be based thereon, as these would represent a limitation on domestic emissions. At present, the Norwegian authorities have outlined objectives with regard to domestic emissions in 2020; see Chapter 9.3.

International allowance prices: If Norway is subject to an international commitment, like the Kyoto Protocol, or if Norway has of its own accord defined targets that are met through the purchase of allowances, increased greenhouse gas emissions in Norway will be offset by reduced emissions elsewhere under the cap-and-trade system. At present, Norway is subject to the Kyoto Protocol, which specifies an emissions cap for its member states. Access to an international cap-and-trade system may also be interpreted to mean that the Norwegian authorities have the option of realising emission reductions at a fixed international price.

An international agreement in line with the two-degree target is not a realistic short-term prospect. The Committee is therefore of the view that it would not be appropriate to use price paths inferred from the two-degree target as a main alternative in the present situation. Elements like the highlighting of efforts to combat climate change and other non-priced effects must be assumed to be of importance to decision makers’ assessment of projects. However, these and similar considerations should be discussed separately.

The Committee finds, against this background, that if the authorities have binding domestic emission targets, then both producer prices (taxes) and calculation prices for public cost-benefit analysis purposes should be inferred from such targets. Climate Cure 2020 (2010) has computed different paths based on various assumptions towards 2020.

If Norwegian targets are instead related to overall global emissions caused by Norway, and Norwegian emissions are subject to an international cap-and-trade system, the calculation price for greenhouse gas emissions should be based on expectations with regard to the international allowance price.

The recommendation not to use prices inferred from the social cost of carbon or the two-degree target is based on the assumption that increased emissions from individual projects will not result in increased global emissions – only in changes to where those emissions are taking place. It is important to bear this in mind both when planning analyses and when interpreting the (analyses) findings. In particular, calculation prices determined in this manner are not suited for evaluating how ambitious a global climate policy should be, since the abatement cost appears to be considerably lower than the social cost of carbon (or the necessary marginal cost of realising the two-degree target) over the next few years.

By international allowance prices are primarily meant the prices faced by the Norwegian authorities when purchasing allowances. During the first commitment period under the Kyoto Protocol, this has predominantly been the price of CDM allowances. This price has been closely correlated with the price of EU allowances. The latter has a stronger institutional foundation than the market for CDM allowances, whilst the market for CDM allowances, or allowances from other/new mechanisms, including the second commitment period under the Kyoto Protocol, faces a more uncertain future. Doubt has also been expressed with regard to the actual magnitude of the emission reduction effect from CDM allowances. The Committee recommends, against this background, that market expectations concerning the price of EU allowances be used as the basis for formulating a carbon price path that reflects the international allowance price.

The uncertainty associated with future international climate agreements is so high that the Committee is of the view that it is challenging to arrive at a definite position with regard to allowance prices beyond 2020 on the basis of information from the current cap-and-trade system. International allowance prices will also be uncertain, and highly susceptible to political decisions and signals, between now and 2020. The two-degree target has a very strong position in the international climate negotiations and is expressly incorporated into Norwegian policy. Although internationally binding agreements in relation to this target have yet to be concluded, the Committee takes the view that a realistic scenario is for the price of EU allowances to approach the two-degree path over time. If an international allowance price path is used, the Committee recommends that such path be based on the EU ETS allowance price as far as concerns those years for which futures prices are quoted, before gradually approaching a path reflecting the two-degree target, cf. Table 9.2.

The Committee also recommends, against the background of a number of conflicting considerations in the determination of calculation prices for greenhouse gas emissions, that sensitivity analyses be carried out, using different principles in establishing calculation prices for emissions, especially in those cases where the cost-benefit analysis is particularly sensitive to different carbon price paths. Such computations may be of considerable benefit to decision makers (the authorities).

9.8 Summary recommendations

Based on the discussion in the present Chapter, the Committee makes the following recommendations:

• The current differentiated tax and quota structure for the private sector is not suitable for use in cost-benefit analysis. A joint carbon price path should be applied for purposes of cost-benefit analysis.

• The appropriate calculation price for greenhouse gas emissions depends on what question one would like the analysis to answer. The Committee adopts the assumption that the authorities are subject to binding emission limitation targets, thus implying that increased emissions in one location necessarily have to be compensated by reductions elsewhere. The Committee recommends, on this basis, that the calculation price for greenhouse gas emissions be based on the marginal cost of emission reductions (the marginal abatement cost). If there are no binding emission limitation targets, the carbon price path should, in principle, be based on the marginal social cost of carbon instead.

• If the authorities are subject to binding domestic emission reduction targets, the calculation prices should be derived from the constraints resulting from such targets. Climate Cure 2020 (2010) has calculated a number of such paths towards 2020.

• If binding Norwegian targets are related to the contribution to total global emissions caused by Norway, and Norwegian emissions are subject to an international cap-and-trade system, the calculation price for greenhouse gas emissions should be based on expectations as to the international allowance price. From the various allowance prices in current international trading systems, the Committee recommends the use of the EU ETS allowance price. The path should be based on market expectations of future allowance prices. For years in respect of which no prices are quoted, the price path should over time approach an assumed two-degree path based on internationally recognised model computations.

• For projects where the cost-benefit analysis is particularly sensitive to different carbon price paths, it will be useful to prepare sensitivity estimates assuming a two-degree path for all years.

If the national or international political situation changes, such as to make new climate targets binding on the Norwegian economy, it is the marginal abatement cost given these new targets that should form the basis for the main joint calculation price alternative for greenhouse gas emissions.

If Norway finds itself, in future, in a situation where the authorities are not subject to binding emission reduction targets, thus implying that emissions increases in one location cannot be assumed to imply emission reductions elsewhere, the carbon price path in cost-benefit analysis should, in principle, be based on the marginal social cost of carbon.

The specific paths should be prepared by the Ministry of Finance in consultation with other affected ministries.

9.9 Bibliography

Agreement relating to the Climate Report (2008). (In Norwegian only. Norwegian title: Avtale om klimameldingen (2008).) Labour Party, Socialist Left Party, Centre Party, Conservative Party, Christian Democratic Party and Liberal Party. Available at http:// www.regjeringen.no/Upload/MD/Vedlegg/Klima/avtale_klimameldingen.pdf

Alberola, E and J. Chevallier, J. (2009). European carbon prices and banking restrictions: Evidence from Phase I (2005-2007). The Energy Journal 30(3), pp. 107-136.

ASEK 5 (2012). Economic Principles and Estimates for the Transportation Sector: ASEK 5. (In Swedish only. Swedish title: Samhällsekonomiska principer och kalkylvärden för transportsektorn: ASEK 5.) Swedish Transport Administration.

Bastianin, A., A. Favero and E. Massetti (2010). Investments and Financial Flows Induced by Climate Mitigation Policies, 2010.013 Note di Lavoro.

Bell, R. G and D. Callhan (2011). More than Meets the Eye: The Social Cost of Carbon in U.S. Climate Policy, in Plain English, Policy Brief, Environmental Law Institute, World Resources Institute.

Bosetti, V., C. Carraro, R. Duval, A. Sgobbi, and T. Massimo (2009). The Role of R&D and Technology Diffusion in Climate Change Mitigation: New Perspectives Using the Witch Model. FEEM Working Paper No. 14. 2009; CMCC Research Paper No. 63.

Bye, B. and K. E. Rosendahl (2012). Carbon Leakage: Causes and Policy Instruments. (In Norwegian only. Norwegian title: Karbonlekkasje: Årsaker og virkemidler.) Samfunnsøkonomen No. 1/2012.

Cancún Agreements (2011). United Nations Framework Convention on Climate Change.

Climate Cure 2020 (2009). Assessment of future allowance prices. A report from the central government working group Climate Cure 2020. (In Norwegian only. Norwegian title: Vurdering av framtidige kvotepriser. En rapport fra etatsgruppen Klimakur 2020) Report. TA No. 2545/2009.

Climate Cure 2020 (2010). Measures and Policy Instruments for Meeting Norwegian Climate Targets Towards 2020 (In Norwegian only. Norwegian title: Tiltak og virkemidler for å nå norske klimamål mot 2020) Report. TA No. 2590/2009.

Cost-Benefit Analysis Guide. (In Norwegian only. Norwegian title: Veileder i samfunnsøkonomiske analyser) (2005). Ministry of Finance.

Department of Energy & Climate Change (2011). A brief guide to the carbon valuation methodology for UK policy appraisal. Department of Energy & Climate Change.

EU Directive 2009/29/EC (2009). Directive 2009/29/EC of the European Parliament and of the Council, L 140/63.

Guardian (2011). Traders condemn EU's 'Mickey Mouse' carbon market after botched trading statement, downloaded 9 February 2012. Available at: http://www.guardian.co.uk/environment/2011/jan/21/emissionstrading-eu [downloaded February 2012].

Hagem, C., and Bjart Holtsmark, (2008). Does the CDM scheme have a future? (In Norwegian only. Norwegian title: Er det noen fremtid for CDM-ordningen?) Samfunnsøkonomen No. 5, 2008.

Hagen, K. P. (2010). Market-Oriented Environmental Policy Management Methods (In Norwegian only. Norwegian title: Markedsorienterte styringsmetoder i miljøpolitikken), NTNU, Concept Report No. 24.

Hamilton, K., M. Sjardin, M. Peters-Stanley and T. Marcello (2010). Building Bridges: State of the Voluntary Carbon Markets 2010, Report from Ecosystem Marketplace and Bloomberg New Energy Finance.

HEATCO (2006). Developing Harmonised European Approaches for Transport Costing and Project Assessment, Deliverable 5, HEATCO Project under the 6th EU Framework Programme.

Hoel M. and L. Karp (2002). Taxes versus Quotas for a stock pollutant. Resource and Energy Economics, Vol. 24, pp. 367-384.

Hoel, M., M. Greaker, C. Grorud and I. Rasmussen (2009). Climate Policy Costs and Design. A Survey of Some Recent Numerical Studies, Nordic Council of Ministers.

IEA (2011). World Energy Outlook 2011. International Energy Agency.

Interagency Working Group on Social Cost of Carbon (2010). Technical Support Document: Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order 12866.

International Center for Climate Governance (ICCG) (2011). Fondazione Eni Enrico Mattei (FEEM) and Fondazione Giorgio Cini, Italy. The table is available upon request. Its reports (“International Climate Policy and Carbon Markets) are available at http://www.iccgov.org/publications/publications-2.htm.

IPCC (2007). Summary for Policymakers. In: B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, L.A. Meyer (red.), Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom, and New York, NY, United States.

Kyoto Protocol (1998). Kyoto Protocol to the United Nations Framework Convention on Climate Change, Kyoto, Japan.

Mansanet-Bataller, M., J. Chevallier, M. Hervé-Mignucci and E. Alberola (2010). The EUA-sCER Spread: Compliance Strategies and Arbitrage in the European Carbon Market, Post-Print halshs-00458991, HAL.

Ministry of Defence, (2010). Cost-Benefit Analysis Guide for the Investment Activities in the Defence Sector. (In Norwegian only. Norwegian title: Veileder i samfunnsøkonomiske analyser for investeringsvirksomheten i forsvarssektoren.)

Nordhaus, W. (2010). Economic Aspects of Global Warming in a Post-Copenhagen Environment, PNAS 2010 107 (26), pp. 11721-11726.

Nordic Council of Ministers (2012). Carbon leakage from a Nordic perspective. Report TemaNord 2012:502.

NOU 2009: 16 Green Paper; Global Environmental Challenges – Norwegian Policy (In Norwegian only. Norwegian title: Globale miljøutfordringer – norsk politikk). Ministry of the Environment.

Norwegian National Rail Administration (2011). Method Handbook JD 205, Cost-Benefit Analysis for the Railways, Version 3.0.(In Norwegian only. Norwegian title: Metodehåndbok JD 205, Samfunnsøkonomiske analyser for jernbanen, versjon 3.0) July 2011.

Paltsev, S., J.M. Reilly, H.D. Jacoby and J.F. Morris (2009). The Cost of Climate Policy in the United States. Report, MIT Joint Program on the Science and Policy of Global Change.

Proposition No. 1 (2011-2012) to the Storting; Direct Taxes, Indirect Taxes, Customs Duties 2012 (In Norwegian only. Norwegian title: Skatter, avgifter, toll 2012). Ministry of Finance.

Proposition No. 111 (2011-2012) to the Storting; Supplementary Apportionments and Reallocations in the Fiscal Budget 2012. (In Norwegian only. Norwegian title: Tilleggsbevilgninger og omprioriteringer i statsbudsjettet 2012)

Report No. 34 (2006-2007) to the Storting; Norwegian Climate Policy (In Norwegian only. Norwegian title: Norsk klimapolitikk). Ministry of the Environment.

Report No. 1 (2011-2012) to the Storting; National Budget 2012. (In Norwegian only. Norwegian title: Nasjonalbudsjettet 2012). Ministry of Finance.

Report No. 21 (2011-2012) to the Storting; Norwegian Climate Policy (In Norwegian only. Norwegian title: Norsk klimapolitikk). Ministry of the Environment.

Stern, N. (2006). Stern Review on the Economics of Climate Change. HM Treasury, United Kingdom.

Tol, Richard p. J., (2009). The Feasibility of Low Concentration Targets: An Application of FUND, Papers WP285, Economic and Social Research Institute (ESRI).

Watkiss, P., T. Downing, C. Handley, R. Butterfield, (2005a). The Impacts and Costs of Climate Change. Final Report September 2005. Commissioned by the European Commission. Part of the project Modelling Support for Future Actions – Benefits and Cost of Climate Change Policies and Measures. ENV.C.2/2004/0088.

Watkiss, P., D. Anthoff, T. Downing, C. Hepburn, Chr. Hope, A. Hunt, R. Tol (2005b). “The Social Cost of Carbon (SCC) Review – Methodological Approaches for Using SCC Estimates in Policy Assessment. Final Report November 2005.

Weitzman, M. L. (1974). Prices vs. quantities. Review of Economic Studies 41(4), pp. 477–491.