Exchanging Ideas on Climate
National Round Table on the Environment and the Economy
Exchanging ideas on Climate

Common menu bar links

Achieving 2050: A Carbon Pricing Policy for Canada (Technical Report)

Previous -- Next

4.0 The Essential Elements of Design and Implementation: Broad and Unified Pricing Over The Long-Term

To achieve Canada?s emission reduction targets at least cost will require a carbon pricing policy that transmits a uniform price across the economy as broadly as possible and that is robust over the long-term.

Our Getting to 2050 report left open the question of the core design features of the preferred carbon pricing policy. This chapter focuses on that question and demonstrates that the critical issue for selecting the preferred policy instrument is not simply about choosing between carbon taxes or cap-and-trade systems, but rather how to design a unified carbon pricing policy to deliver least-cost reductions in the long-term, while meeting the government?s emission reduction targets. But the NRTEE?s research also concludes that in order to deliver these reductions, addressing uncertainties will be central to the carbon pricing policy. This conclusion indicates the need for equal consideration of the design and implementation of the carbon pricing policy. To achieve the goal of attaining the emission reduction targets at least cost, this section identifies two essential objectives for carbon pricing policy:

1. The carbon pricing policy must be designed to apply a uniform carbon price across all emissions and jurisdictions; and,

2. A robust policy must be implemented to send a long-term price signal while being responsive and adaptive to changing circumstances through time.

Finally, this chapter provides an overview of the broad structure of a carbon pricing policy that best meets these objectives and delivers cost-effective reductions. The carbon pricing policy consists of three distinct policy elements and also an implementation strategy for institutions and processes to manage the policy over time and adapt it as required. The rest of the report then addresses how the details of the design of this carbon pricing policy can meet the complementary goal of minimizing adverse impacts through policy design.

Table of Contents

Adobe PDF version (3.07 Mb)

Achieving 2050

4.1 The Need for a Unified Carbon Pricing Policy

An essential objective for carbon pricing policy should be to apply a uniform carbon price with broad coverage over the Canadian economy. Both economic and political dimensions are important for unified pricing policy: the emission price should be unified across emissions and across jurisdictions within Canada.

4.1.1 Unifying the Carbon Price Across Emissions19

  • Carbon pricing is most efficient when it applies a common price across all emissions. Design must ensure that carbon pricing is economy-wide, capturing as many sources of emissions as possible.
  • Although some emissions sources are impractical to include in a carbon pricing policy, such as process, landfill and agricultural emissions, failure to address these emissions raises the costs of a carbon pricing policy. Other mechanisms, such as regulations, are necessary to deal with such emissions.

The extent to which all emissions are included under a unified pricing policy is a critical issue for policy design. This chapter first demonstrates that broader, more uniform coverage generally leads to more cost-effective pricing policy. Yet unified pricing across emissions is a challenge; many current and proposed pricing policies in other jurisdictions do not have uniform coverage. The European Union Emissions Trading Scheme (ETS), the Regional Greenhouse Gas Initiative (RGGI) and other initiatives do not include all sectors of the economy. In this section, it is also showed that part of the reason for less than full coverage is that many emissions are administratively and practically more challenging to include under a carbon pricing policy. Finally, the efficiency gains possible from including these ?hard-to-reach? emissions are demonstrated.

Trade-offs between coverage, stringency, and targets

To be effective and efficient, a carbon pricing policy must aim to include as many sources of emissions as possible. Broader coverage results in greater emission reductions at a lower carbon price, since it implies more opportunities for cost-effective reductions. If sectors or regions are exempted from policy, more stringent policy (with a higher price of carbon) is required if emission reduction targets are still to be achieved. Additionally, more expensive, emission reductions must then come from sectors included under the policy, raising the costs of the policy.

On the other hand, sector-specific policies are often more politically acceptable to regulated entities. The current tendency in provincial, national, and international climate policy has been to exclude emissions that are perceived to be politically challenging, at least at first. This approach means that large industrial emitters tend to face carbon pricing while transportation, light manufacturing, households and buildings remain somewhat exempt, despite accounting for significant amounts of emissions. Sector-specific pricing policies may raise greater concerns about the potential for rent-seeking by the sector, raise questions about the ability of regulators to maintain objective and independent oversight, and reduce political and public confidence in the system. Sectoral policies also move away from a unified price signal, resulting in economic inefficiencies.

NRTEE modelling analysis supports the idea that broader, more uniform coverage is more efficient. Decreasing coverage of the policy means that the price of carbon and the costs of the policy must increase if targets are to be achieved. If only large industrial emitters are included under pricing policy, the price of carbon must rise 2.25 times higher than if the unified price was applied across all emissions in Canada. Medium-term impacts would be particularly acute, with total capital, operating and maintenance and energy expenditures rising in 2020 to $26 billion above the unified case. Indeed, if lower prices are imposed on households and transportation sectors, there is a risk that 2020 and 2050 emission reduction targets cannot be achieved. Analysis using the CIMS energy-economy model shows that even a price of $600 / tonne on large industrial emitters alone would likely not result in the targets being reached. Further, reductions appear insensitive to even higher carbon prices, so additional reductions are uncertain.

Further, if certain sectors face a much lower price for emissions than others, an inefficient portion of economic activity ? and emissions ? will shift into these sectors over time, making it more expensive and difficult to achieve long-term emission reduction goals. For example, the analysis suggests that if large emitters, such as the electricity sector, face a higher relative price signal than households, households will reduce their use of electricity and switch to fossil fuels. These actions can raise household emissions and reduce the effectiveness of emission reductions in the electricity sector.

Failure to broaden the scope of carbon pricing leads to higher costs

The analysis summarized in Table 1 highlights the importance of addressing all greenhouse gas emissions, including emissions from upstream oil & gas, agriculture and landfill waste as part of a carbon pricing policy. The table compares the required emission price to achieve 2020 and 2050 targets if these emissions are not included and if complementary regulations are used to address them. It demonstrates that the carbon price per tonne is cheaper if all sectors are included through complementary regulations. If these sectors are not addressed, other sectors must make up the difference in order to meet Canada?s targets. To achieve this outcome, higher carbon prices will be needed in the sectors that are within the pricing policy.

Note that the high emissions prices under some of these scenarios are unlikely. At emission prices of $500 or $775, the response of the energy system is uncertain; the CIMS model is incapable of representing newly innovated technologies that would likely result at such a strong price signal. Further, such a price would pose a formidable political challenge. As such, the prices in Table 1 should be taken only as a directional indicator of the benefits of broadening the coverage of the carbon pricing policy through complementary regulations.

Challenges of including all emissions under a carbon pricing policy

Complete and direct coverage of all emission sources within a carbon pricing policy is challenging. Sectors of the economy with emissions that are difficult to include under a carbon pricing policy have been identified, but can nevertheless provide cost-effective emission reductions through other means. Complementary regulations (discussed in Section 7.2) or offsets (discussed in Section 5.4) can be used to address these emissions.

  • High upstream oil and gas well venting and flaring (estimated at 65.7 Mt CO2e in 2005, or approximately 9% of Canada?s emissions). Venting and flaring, to dispose safely of uneconomic quantities of gas, results in the direct release of large amounts of greenhouse gases. These emissions are difficult to quantify, because they come from numerous sources in remote conditions, but all estimates show them to be very large.

  • Pipeline combustion (estimated at 10.1 Mt CO2e in 2005, or approximately 1% of Canada?s emissions). Transport of oil and natural gas by pipeline, especially natural gas, creates significant combustion and fugitive emissions. Transmission firms may track some of the gas combusted, as this is no longer available for sale, but the amounts used to run gas actuated auxiliaries and fugitive leaks are extremely hard to measure, as they are from many thousands of sources in very remote conditions.

  • Landfill gas (estimated at 28 Mt CO2e in 2005, or approximately 4% of Canada?s emissions).

    Landfills release significant amounts of methane from anaerobic decomposition of organic waste by bacteria. Landfill gas can be trapped and combusted as a flare, converting it from methane to carbon dioxide, a weaker greenhouse gas. Because these methane emissions are not from combusted market fuels and are difficult to quantify, a carbon pricing scheme will have no direct effect.

  • Agricultural emissions (estimated at 56.6 Mt CO2e in 2005, or approximately 8% of Canada?s emissions). A significant portion of Canada?s GHG emissions come from enteric fermentation in the stomachs of sheep and cattle (25.0 Mt CO2e), manure management (8.6 Mt CO2e), and agricultural soil management (23.0 Mt CO2e). Agricultural emissions can be reduced through changes in land use and agricultural practices. Because these emissions are not from combusted market fuels and are spread all over Canada from virtually millions of sources, a carbon pricing policy by itself would have no effect on them.

4.1.2 Unifying Carbon Pricing Policy Across Jurisdictions Within Canada20

  • Carbon pricing policies are currently being developed and implemented in an uncoordinated approach at the federal, provincial and regional levels, resulting in policy fragmentation across Canada.
  • A nationally harmonized carbon emissions price is more economically efficient than a patchwork of regional prices set through federal, provincial, territorial or regional policies

A key issue for Canadian carbon pricing policies is reconciling federal, provincial and regional carbon pricing policies and approaches. Just as the issue of coverage illustrates the importance of unifying carbon pricing policy across sectors and emitters, the issue of fragmented policies illustrates the importance of unifying carbon pricing policy across jurisdictions. The issue of governance is challenging, as federal and provincial governments share jurisdiction for regulating carbon emissions. At the same time, the federal government has jurisdiction over border adjustments and international trade issues. In addition to the federal Turning the Corner plan which proposes an intensity-based emissions trading system, British Columbia and Quebec have implemented forms of carbon taxes, Alberta has developed a provincial emissions trading system, and British Columbia, Manitoba, Ontario, and Quebec plan to participate in the Western Climate Initiative regional cap-and-trade system, along with seven US States.21

Economic efficiency of fragmented, regional carbon pricing policy approaches

To provide a sense of the economic trade-offs between fragmented regional and coordinated national approaches, a fragmented policy scenario using the CIMS model was explored. Results of this illustrative modelling suggest that a coordinated national approach to carbon pricing, where the carbon price is unified across jurisdictions, tends to be more cost effective than a fragmented regional approach.

Under the fragmented policy scenario, each region represented in the model (BC and the territories, Alberta, Saskatchewan, Manitoba, Ontario, Quebec, and the Atlantic region) was modelled separately.22 For each region, the carbon prices required to meet that region?s share of Canada?s emission reduction targets for 2020 and 2050 was determined. These carbon prices thus represent the pricing approach each region could take independently to reduce its emissions by 20% in 2020 and 65% by 2050 from its business as usual trend. The required price trajectories are shown in Figure 3. These price trajectories were compared to the Canada-wide fast and deep price signal. Total emission reductions across Canada are thus the same for both the unified and fragmented scenarios; the distribution of emission reductions, however, is quite different, as regions with lower-cost reductions achieve greater reductions, and regions with higher-cost reductions achieve fewer. For example, Alberta?s carbon price would be about $400 / tonne CO2e in 2020 versus $150 / tonne CO2e for a unified carbon price.

The fragmented policy scenario illustrates the increased economic efficiency of a national approach. Figure 4 compares a weighted average of the varying regional price trajectories to the emission trajectory required for a national program to achieve national emissions reductions of 20% by 2020 and 65% by 2050 relative to 2006 levels. While both approaches result in the same amount of overall emission reductions, the fragmented regional approach requires an emission price almost 25% higher than the national approach. The modelling analysis suggests that achieving reduction targets under the fragmented approach has significantly higher overall costs of abatement than under a unified approach. A nationally unified approach has lower costs for Canada because it allows for the least expensive emission reductions overall to be achieved rather than requiring a specific amount of reductions in each region. These lower costs result in greater economic efficiency for achieving the national target and thus more cost-effective policy.

Economic costs of the fragmented policy scenario

These results have four important implications for consideration in designing a Canadian carbon pricing policy:

First, the overall cost of carbon pricing policy is reduced where a common carbon price is applied across Canada. A fragmented, province-by-province approach will increase the overall costs to Canadians of reaching Canada?s emissions targets (modelling suggests an increase in costs of $40 billion). In terms of economic impacts, the GDP costs of this fragmentation relative to an efficient unified policy are 7% greater than the unified approach in 2020, 20% in 2035, and 7% in 2050. While provinces can have a strong role in implementing carbon pricing policy, a unified carbon price improves economic efficiency, and reduction targets are achieved at lower cost.

Second, some provinces (such as British Columbia and Alberta) would face significantly higher costs of achieving their share of Canadian emission reduction targets if they were to act independently. Indeed, if it acts independently, Alberta will have difficulty achieving 20% reductions by 2020, even at prices of $400 per tonne. Similarly, achieving 65% emission reductions in Alberta alone by 2050 will require very high emission prices in that province.

Third, while a national approach is more cost-effective overall, it has implications for distributional effects. Effectively, the economic efficiency gains that would be achieved through a coordinated approach result from less expensive reductions from one region replacing more expensive reductions in another. Regions like Manitoba and Quebec would abate more emissions, and regions like Alberta and BC would abate fewer. On the level of firms, some emitters will thus be relatively better off under a national policy, while some will be relatively worse off, compared to a fragmented policy.

Fourth, without a national system, provinces have less incentive to implement stringent provincial policy. They can benefit from emission reductions in other jurisdictions without imposing costs on local industry and households (i.e. there is an incentive to ?free-ride?). National policy is therefore more likely to achieve the medium and long-term targets for the country as a whole, in the most cost-effective way.

Administrative complications of regional policy variation

The fragmented policy approach will also result in additional, harder-to-quantify costs as compared to a coordinated national approach. Policy variation among regions, not only in terms of the price of emissions, but also with respect to the rules for compliance, will complicate business planning for firms with inter-provincial operations. Firms will face higher transaction costs from compliance with multiple regional policy regimes. Variation in carbon pricing policies among regions could increase uncertainty as to the long-term durability of these policies. A coordinated carbon pricing policy regime would be more credible over the long-term, and thus provide clear incentives for firms to invest in low-carbon technologies with longer life-cycles.

Inter-provincial leakage could also be an issue. Large discrepancies between the stringency of policy in different regions could result in industry relocating from regions with higher carbon prices to those with lower prices.

Finally, some smaller provinces may have insufficient administrative capacity to develop and implement effective provincial carbon pricing policy. A national approach would ease the administrative burden and could help share costs.

4.2 The need for robust, long-term carbon pricing 23

? Given the long time periods required for the decarbonization of the Canadian economy, carbon pricing policy implementation must pay particular attention to issues of uncertainty. It must balance ensuring policy adaptability with providing a durable, long-term price signal.

A second essential element of carbon pricing policy is that it must send a price signal to the economy that is durable and credible over the long term. The NRTEE?s research suggests that two elements are essential to create such a policy. First, the carbon pricing policy must provide investors with policy certainty, making clear the ?rules of the game.? Second, policy must be adaptable through time in order to manage uncertainty and respond to changing circumstances. While policy adaptability and policy certainty are essential elements for any carbon pricing policy, there are trade-offs between the two criteria. If a policy has clearly been designed to be flexible or changeable at some future time, uncertainty as to the future nature of the policy follows. On the other hand, an attempt to fix policy in advance would imply a failure to adapt to new information, such as evolving climate science or the policies of Canada?s trading partners. Effective carbon pricing policy needs to find a balance between adaptability and certainty ? it should be adaptable to changing and unknown future circumstances, but certain enough to transmit a durable, long-term carbon price signal to the economy upon commencement.

4.2.1 Policy Certainty to Influence Long-term Investment Decisions

Firms and households routinely manage risk and uncertainty when making investment decisions. Yet uncertain climate policy raises additional risks. It raises the cost of capital and alters investment decisions. Policy uncertainty increases incentives to delay investments in carbon-reducing technologies in order to wait for additional information or clearer policy commitment from governments. Firms and households also tend to avoid making investments that could disadvantage them as an early mover, especially if they believe they might be forced to further reduce emissions under a policy, or that policy could change to a less stringent one in the future.

The NRTEE?s research suggests that a clear communication of a government?s long-term commitment to a pricing policy is critical to achieving low cost reductions aligned with the GHG targets. If consumers believe government might backslide, or soften pricing policy as a result of political pressure, the policy?s effectiveness is reduced. In Figure 5, two scenarios are presented, one where investments are made with complete confidence in the carbon pricing policy and one where there is no confidence. With a lack of confidence, there is a lower level of overall investment resulting in much lower emission reductions. This conclusion is supported by studies that have shown that under uncertainty, a higher price on carbon is required to trigger investment in low-carbon emission technologies and that adoption of different electricity generation technologies can vary substantially depending on investors? perceptions of carbon price uncertainty.24 With confidence in the carbon pricing policy, investments are made that reflect the future value of carbon and so better long-term technology choices prevail.

Expected future prices of carbon emissions influence investment decisions in the present. As a result, effective policy must be implemented to clearly and consistently communicate the nature of a carbon pricing policy. Policy certainty therefore suggests that the carbon pricing policy will be maintained and is defined through time. Essentially, policy certainty ensures the carbon price signal is not diluted by uncertainty about the permanence or longevity of carbon pricing policy.25

4.2.2 Policy Adaptability, Given the Multiple Sources of Uncertainty

While the issue of policy certainty illustrates the importance of addressing uncertainty from the perspective of firms and households, the issue of policy adaptability highlights the significance of uncertainty from the perspective of policy-makers. Substantial sources of uncertainty complicate the design of a carbon pricing policy. Key sources of uncertainty include:

  • The stringency and timing of climate policy to be implemented by Canada?s major trading partners;

  • The urgency of emission reductions, as dictated by evolving climate science;

  • The effectiveness of policies;

  • The costs of policies;

  • Economic activity; and

  • The impacts on stakeholders.

Because of these sources of uncertainty, governments face risks in implementing domestic carbon pricing policy, and must therefore take these issues into account. If, for example, Canada was to implement policy independent from its trading partners, it could subject Canadian industry to heightened competitiveness concerns. If policy was set too stringently, and emission mitigation costs were unexpectedly high, the Canadian economy could suffer disruption.26 Similarly, if short-term Canadian emission reductions were too shallow, Canada might be forced to move toward more aggressive reductions in the future that may cost the economy more than if the emission reduction targets had been set higher in the present. These risks can be reduced if the policy is designed to be adaptable and flexible. Policy adaptability would therefore allow a policy to respond to new information in the future and help ensure it remains effective and economically efficient.

Principles of adaptive management can be applied to climate change policy. Such principles revolve around the notion that policies should respond to changes over time and make explicit provision for learning.27 Basically, they seek to accommodate uncertainty within the policy framework. An adaptive management framework relies on monitoring results of policy experiments to test the impacts of different policy and management approaches on complex systems. This approach can be useful given the complexity of the energy system and the uncertainties described above. Explicitly incorporating such adaptive mechanisms as automatic policy reviews and adjustments, regular and transparent reporting requirements, clear evaluation and revisions, into policy design can enable learning.28

4.2.3 Balancing Policy Certainty and Adaptability

While policy adaptability and policy certainty are important objectives for a carbon pricing policy, there are trade-offs between the two goals. If a policy has clearly been designed to be flexible or changeable at some future time, uncertainty cannot be avoided. On the other hand, fixing policy for the long-term implies that it cannot adjust to new information. The NRTEE?s research suggests that a carbon pricing policy should instead strive toward balancing adaptability and certainty; it should be adaptable in future time periods, but transmit a robust price signal to the economy. Achieving this balance is really about governance: the design of institutions and processes associated with implementing and managing a carbon pricing policy over time. Uncertainties from the policy makers? perspective can be managed through this adaptive approach.29

4.3 A Carbon Pricing Policy for Canada 30

The essential objectives for cost-effective carbon pricing policy have been identified above. To meet these objectives, a policy that includes three design elements as well as an implementation strategy to ensure the carbon pricing policy can be managed over the long term, is required. An overview of this carbon pricing policy, which is the focus of the separate Advisory Note, is set out below. Design options for these elements are explored in detail in subsequent sections of this report.

4.3.1 Design of Three Policy Elements

The NRTEE Advisory Note recommends a carbon pricing policy based on three main elements to meet the high-level, essential design objectives set out above. The carbon pricing policy includes:

1. A unified carbon price across all emissions through a single national cap-and trade system. Detailed design options are explored in Section 5.0. This unified pricing instrument would include:

  • A cap for large emitters.31 This covers approximately 51% of emissions. By setting a maximum carbon price, desirable elements of carbon taxes can be included to enhance price certainty and contain costs while ensuring the cap on emissions works efficiently; and
  • A cap for the rest of the economy. This covers approximately 36% of remaining emissions in buildings, transportation and light manufacturing. A cap would be applied at a point in the fuel distribution chain to those that distribute or import fuel, thereby limiting the number of trading entities while broadening coverage throughout the economy.

2. Complementary regulations and technology policies. Complementary regulations can further improve the cost-effectiveness of the carbon pricing policy by ensuring all low-cost emission reduction opportunities are achieved. Emissions from agriculture, waste, and upstream oil and gas can be difficult to include under a cap-and-trade system directly. Other complementary regulations are necessary to ensure the carbon price reaches these sectors of the economy, supplemented by targeted technology policy to address market barrier issues. This approach would include regulations for some of the remaining hard to reach emissions. Detailed design options for these complementary policies are explored in Chapter 7.0.

3. International carbon abatement opportunities. As carbon prices rise, there will need to be more awareness of how the domestic carbon costs align with that of major trading partners. Also, as carbon costs rise and additional units of reductions become more expensive, a disproportionately high amount may be spent on relatively few additional reductions. For these reasons, access to reductions outside of Canada are included in the carbon pricing policy, again to ensure Canada?s emission reduction targets can be achieved at least cost. Further discussion of international purchase opportunities is found in Chapter 6.0.

4.3.2 Governance and Implementation of an Adaptive Carbon Pricing Policy

Implementation is at least as important as policy design. Institutions and processes to manage the policy over time are essential to implementing a durable, long term carbon pricing policy. Higher level recommendations for the design of these institutions and processes are also developed as part of an implementation strategy for the carbon pricing policy. Detailed implementation options are explored in Chapter 9.0.

Previous -- Next