AN ASSESSMENT OF
THE ECONOMIC AND
ENVIRONMENTAL IMPLICATIONS
FOR CANADA
OF THE KYOTO PROTOCOL
ANALYSIS AND MODELLING GROUP
NATIONAL CLIMATE CHANGE PROCESS
November 2000
November 29th, 2000
David Oulton John Donner
Co-Chair Co-Chair
NAICC-CC NAICC-CC
Suite 600, 55 Murray Street 10th Floor, 9915 - 108th Street
Ottawa, Ontario K1N 5M3 Edmonton, Alberta T5K 2G8
Dear David and John:
On behalf of all the members of the Analysis and Modelling Group (AMG), we are pleased to
forward its report An Assessment of the Economic and Environmental Implications for
Canada of the Kyoto Protocol. This report summarizes the analytical approach, the
assumptions, the results and the main learnings of the AMG work over the past two years.
The report was prepared in accordance with the mandate approved by the NAICC-CC in
mid-1998. This mandate directed the AMG to conduct an integrated analysis of the economic,
social, health and environmental implications of the Kyoto Protocol.
All AMG members agree that this document is an accurate reflection of the analytical work to
date. While there remain some differences of interpretation in certain areas, there is general
consensus on the main learnings.
In the Main Learnings, several issues for further analysis are identified. Moreover, the AMG
has acquired important insights about the analytic and consultative processes. If required by
NAICC, the AMG will provide its recommendations on both the substance and structure of
future analytic effort.
Best regards,
Charles Lester Neil McIlveen
AMG Co-Chair AMG Co-Chair
Newfoundland and Labrador Dept. of Mines and Energy Natural Resources Canada
On behalf of:
Shelley Murphy
British Columbia Ministry of Employment and Investment
Grant Hilsenteger
Alberta Environment
David Hanly
Saskatchewan Energy and Mines
Wilf Falk
Manitoba Bureau of Statistics
Than Nguyen
Ontario Ministry of Energy, Science and Technology
Thomas Storring
Ontario Ministry of Finance
Roger Corbeil
Minist�re des Ressources naturelles du QuŽbec
Claude SauvŽ
Minist�re de lÕEnvironnement du QuŽbec
George Foote
Nova Scotia Department of Natural Resources
Anna Soininen
Dept. of Resources, Wildlife & Economic Development, Northwest Territories
Stephen McClellan
Environment Canada
Jay Barclay
Environment Canada
RŽjean Casaubon
Natural Resources Canada
Nick Marty
Natural Resources Canada
Table of Contents
ACKNOWLEDGMENTS ................................................v
EXECUTIVE SUMMARY ............................................... vii
Analytic Approach................................................. ix
Main Learnings .................................................. xiv
Concluding Remarks ..............................................xviii
1. INTRODUCTION ................................................1
2. OPTION PACKAGES AND PATHS ..................................5
2.1 Issue Table Options ...........................................5
2.2 Option Packages .............................................5
2.3 The Paths...................................................7
3. ANALYTICAL FRAMEWORK ....................................11
3.1 Overview..................................................11
3.2 The Models ................................................12
3.3 The Framing Assumptions .....................................17
3.4 Understanding the Results......................................20
4. ENERGY - TECHNOLOGY MODEL RESULTS ......................25
Key Learnings ..............................................25
4.1 Context of the Canada Acts Alone Results .........................27
4.2 Results: Canada Acts Alone ...................................29
4.3 Context of International Scenario Results . .........................38
4.4 Results: Path 2, Kyoto Tight and Kyoto Loose .....................39
4.5 Results: Path 4, Kyoto Tight and Kyoto Loose .....................42
4.6 Sensitivities ................................................44
5. MACROECONOMIC RESULTS ...................................49
Key Learnings ..............................................49
5.1 Introduction................................................50
5.2 Context of Results ...........................................51
5.3 National Results.............................................52
5.4 Sectoral Results .............................................59
5.5 Provincial Results............................................60
5.6 Competitiveness.............................................62
6. A COMPLEMENTARY VIEW: ANALYSIS USING CASGEM ..........67
Key Learnings ..............................................67
6.1 Background ................................................67
6.2 Results: Path 4, Canada Acts Alone .............................69
6.3 Results: Path 4, Kyoto Loose ..................................71
6.4 Results: Path 3, Canada Acts Alone..............................73
6.5 Comparison to Results of Other Models ...........................75
7. ENVIRONMENTAL AND HEALTH IMPACTS ......................77
Key Learnings ..............................................77
7.1 Introduction................................................78
7.2 Approach .................................................78
7.3 Observations and Results ......................................80
7.4 Qualitative Assessment........................................86
7.5 Future Work ...............................................87
8. CONCLUSIONS .................................................89
9. ANNEX I: MOVING FORWARD - STAKEHOLDER PERSPECTIVES
(EXECUTIVE SUMMARY) BY CCEAF .............................95
10. BIBLIOGRAPHY ...............................................101
List of Charts
CHART 2.1 Issue Table Options for the Roll-up ...............................6
CHART 2.2 Extent of Permit Trading in Path 3 and Path 4 ........................8
CHART 2.3 Summary of Path by Policy Direction ..............................9
CHART 3.1 The Roll-up Framework.......................................11
CHART 3.2 The Modelling Structure.......................................12
CHART 3.3 EIA Case - Percent Change from the Base Case ....................19
CHART 3.4 Elements to Assess Economic, Environment and Social
Costs and Benefits ...........................................21
CHART 4.1 Paths and Scenarios Examined in Micro Economic Analysis ............ 25
CHART 4.2 National GHG Emission in 2010, Canada Acts Alone .................29
CHART 4.3 Marginal Costs of GHG Reduction, Canada Acts Alone ............... 30
CHART 4.4 Investment and Net Cost by Path, Canada Acts Alone ................30
CHART 4.5 End-Use Energy Consumption by Fuel in 2010, Canada Acts Alone ...... 32
CHART 4.6 Sectoral Shares of GHG Reductions in 2010, Canada Acts Alone ........ 32
CHART 4.7 Electricity Sector GHG Emissions in 2010, Canada Acts Alone .......... 33
CHART 4.8 Capture and Storage of CO2 in 2010, Canada Acts Alone ............. 33
CHART 4.9 Electricity Sector Energy Consumption by Fuel in 2010, Canada Acts Alone 34
CHART 4.10 Transportation Sector Energy Consumption by Fuel in 2010,
Canada Acts Alone ..........................................35
CHART 4.11 Provincial Emissions - BAU 2010................................36
CHART 4.12 Regional Shares of GHG Reduction in 2010, Canada Acts Alone ........ 37
CHART 4.13 Emission Per Capita, Canada Acts Alone ..........................37
CHART 4.14 Regional Costs, Canada Acts Alone ..............................38
CHART 4.15 National GHG Emissions in 2010, Path 2 Scenarios ..................39
CHART 4.16 Investments and Net Costs, Path 2 Scenarios .......................39
CHART 4.17 End-Use Energy Consumption in 2010, Path 2 Scenarios .............. 40
CHART 4.18 Sectoral Shares of GHG Reductions in 2010, Path 2 Scenarios .......... 40
CHART 4.19 Regional Shares of Reduction - 2010, Path 2 Scenarios ............... 41
CHART 4.20 National GHG Emissions - 2010, Path 4 Scenarios...................42
CHART 4.21 Investments and Net Costs, Path 4 Scenarios .......................42
CHART 4.22 End-Use Sectors Energy Consumption - 2010, Path 4 Scenarios ........ 43
CHART 4.23 Sectoral Shares of GHG Reductions - 2010, Path 4 Scenarios .......... 43
CHART 4.24 Regional Shares of Reduction - 2010, Path 4 Scenarios ............... 43
CHART 5.1 Paths and Scenarios, Informetrica Analysis .........................51
CHART 5.2 Direct Impact on Income Sectors 2000 TO 2020 (CIMS, MARKAL) .... 53
CHART 5.3 Direct Cost Impacts on GDP (CIMS, MARKAL) ...................54
CHART 5.4 GDP Canada (CIMS, MARKAL) ...............................55
CHART 5.5 Inflation (GDP Deflator) Impact (CIMS, MARKAL) ................. 56
CHART 5.6 Disposable Income per Household (CIMS, MARKAL) ............... 57
CHART 5.7 Balances - Savings Impact (CIMS, MARKAL) .....................58
CHART 5.8 Changes in GDP by Sector 2000-2012 (CIMS, MARKAL) ........... 59
CHART 5.9 Changes in GDP by Sector 2013-2018 (CIMS, MARKAL) ........... 60
CHART 5.10 Provincial GDP Impact 2013-2018
Quebec and East (CIMS, MARKAL) ............................61
CHART 5.11 Provincial GDP Impact 2013-2018
Ontario and West (CIMS, MARKAL)............................ 61
CHART 5.12 Measures of Competitiveness...................................62
CHART 5.13 Total Factor Productivity (CIMS, MARKAL) ......................63
CHART 5.14 Change in Exports by Sector 2013-2018 (CIMS, MARKAL) .......... 64
CHART 5.15 Change in Unit Cost by Sector 2013-2018 (CIMS, MARKAL) ......... 64
CHART 6.1 What CaSGEM Predicts ......................................68
CHART 6.2 Paths and Scenarios for CaSGEM Analysis ........................68
CHART 6.3 Channels for Emission Reductions................................69
CHART 6.4 Emissions and Use of Fossil Fuel in Electricity Production, Path 4CA ..... 69
CHART 6.5 Impact on Source of Energy, Path 4CA ...........................70
CHART 6.6 Largest Sectoral Reductions, Path 4CA ...........................70
CHART 6.7 Provincial GDP, Path 4CA.....................................71
CHART 6.8 Comparison of Energy Prices and Use, Path 4CA....................75
CHART 6.9 Comparison of Fossil Fuels in Electricity Production, Path 4CA ......... 75
CHART 6.10 Comparison of Sectors with Largest Decline in GDP, Path 4CA ......... 76
CHART 6.11 Comparison of Provincial GDP 2013-2018, Path 4CA ................ 76
CHART 7.1 Paths and Scenarios for EHI Analysis .............................78
CHART 7.2 Base Case CAC Emissions ....................................80
CHART 7.3 Changes in Selected CAC Emissions for Models and Analysis Paths ...... 81
CHART 7.4 Required Reduction in CAC Emissions by Region to Achieve
CWS for PM and Ozone ......................................82
CHART 7.5 Maximum Ambient Air Concentration Change ......................83
CHART 7.6 Physical Impacts in 2010, Path 2CA (MARKAL) ...................84
CHART 7.7 Estimated Co-Benefits ........................................84
Acknowledgments
In undertaking the "roll up" analysis, the Analysis and Modelling Group (AMG) has benefitted
enormously from the advice and assistance of a large number of individuals and organizations
from the research, consulting and stakeholder communities. The professionalism, dedication
and cooperative spirit exhibited by these individuals and organizations were critical to the
completion of this extraordinarily complex task. With apologies to anyone inadvertently
omitted from the list, the AMG wishes to express its appreciation to the following:
¥ Richard Loulou, Amit Kanudia and their colleagues at HALOA.
¥ Mark Jaccard, John Nyboer, Alison Bailie and the staff at M.K. Jaccard and
Associates/ERG.
¥ Carl Sonnen and his associates at Informetrica Ltd.
¥ Jeremy Rudin, Philip Bagnoli, Aled ab Iorwerth, Tony Peluso and Yazid Dissou from
the Department of Finance modelling team
¥ Doug Bruchet, Al Howatson, Abha Bhargava, Morgan Macrae, Stacey Schorr, Erin
Down, Brian Guthrie, Stephen Dobson and the rest of the Climate Change Economic
Analysis Forum team
¥ Peter Sol and Nick Macaluso of Environment Canada
¥ The Issue Table liaisons, in particular Jack Belletrutti, Darcie Booth, Matthew Bramley,
Graham Campbell, John Dillon, Chris Feetham, Michel Francoeur, Rick Hyndman,
John Lawson, Tony Lempriere, Bob Redhead, Randolph Seecharan and Ed
Wojczynski
¥ Cathy Wilkinson of Global Change Strategies International Inc., Angelo Proestos of
Cheminfo Services Inc., Robert Caton of Alchemy Consulting Inc., Jacek Kamininski
of ARM Consultants and Mike Moran of Environment Canada.
¥ All the members of the Environmental and Health Impacts Subgroup and the
Stakeholder Advisory Group
¥ John Sargent of the Department of Finance and the Tradeable Permits Working Group
¥ Philippe CrabbŽ of the University of Ottawa
¥ AndrŽ Plourde of the University of Alberta
¥ Ian Hayhow of Natural Resources Canada
¥ Former AMG members Warren Bell, Lynn Berthiaume, Trevor Dark, Duncan
Ferguson, Tony Harras, Linda Hutchinson, Luis Leigh and Dean Stinson O'Gorman
Two special acknowledgments are warranted:
¥ The former co-chairs of the AMG, Sue Kirby of Natural Resources Canada and Dan
McFadyen of Saskatchewan Energy and Mines, whose vision and leadership were
instrumental in establishing the framework in which we worked
¥ The AMG Secretariat, Catherine Roberts and Nancy Roberts of Natural Resources
Canada, who, in addition to their substantive contributions to the analytic effort,
accomplished miracles of organization to keep the process on track
Executive Summary
In April 1998, federal, provincial and territorial Ministers of Energy and Environment launched
the National Climate Change Process (NCCP), a wide-ranging inquiry into the feasibility and
implications of CanadaÕs Kyoto Protocol commitment. The centrepiece of the NCCP was the
establishment of a number of Issue Tables and Working Groups to address and make
recommendations concerning various dimensions of the challenge posed by the Kyoto
commitment.
The Analysis and Modelling Group (AMG) is one of these working groups. It is comprised of
officials and analysts representing two federal departments (Natural Resources Canada and
Environment Canada) and the governments of the provinces and territories. The AMGÕs
principal task is to undertake the so-called Òroll upÓ Ð the integrated assessment of the
economic and environmental consequences, for Canada, of achieving the Kyoto target. In
undertaking the roll up, the AMG was instructed to use as the key inputs the options,
associated analysis and other insights developed by the Issue Tables Ð in particular, from those
such as Agriculture, Buildings, Electricity, Forest Products, Industry, Municipalities and
Transportation which focussed on specific sectors Ð and by the Tradeable Permits Working
Group (TPWG). Further, the AMG was to conduct its analysis in a transparent, step-by-step
process to ensure broad stakeholder review of the results.
This report describes the analytic approach developed by the AMG, presents the main results
and findings, notes the limitations of the analysis and suggests areas for further research. The
purpose of this executive summary is to provide the highlights of the report. Before doing so,
however, it is important to spend a few paragraphs to situate the report and its findings Ð to
indicate what they are and what they are not.
The AMG roll up has been usefully described as Òrange finding.Ó Its primary objective is to
provide policymakers with Òorder of magnitudeÓ guidance on some fundamental issues related
to the achievement of the Kyoto target, including:
¥ the economic implications of different broad policy approaches, such as using various
combinations of a suite of specific measures and a major economic instrument - the
consequences of requiring each sector to achieve a common target versus an overall
national target;
¥ the potential benefits and costs of greater access to the Kyoto flexibility mechanisms;
¥ the sectoral and regional distributions of emissions reductions and costs of achieving the
target;
¥ the degree to which CanadaÕs competitive position might be affected by the
achievement of the Kyoto commitment;
¥ the relative importance of the co-benefits of greenhouse gas (GHG) mitigation, in
particular those related to improvements in air quality; and
¥ areas or actions which appear to offer large potential for emissions reductions and,
conversely, areas or actions for which the potential seems to be limited.
It should be clear from the above that the roll up results should not be viewed as a plan of
action. The resolution of the findings is too coarse and many of the major assumptions are
too speculative for such an interpretation. Three examples underscore this point. First, both
CanadaÕs decision, and probably that of the United States as well, to ratify the Kyoto
Protocol will obviously be predicated on the outcome of negotiations on the structure of the
various flexibility mechanisms (most importantly, an international emissions trading system,
Joint Implementation (JI) and the Clean Development Mechanism (CDM)). At time of
writing, there is little information concerning the outcome of these negotiations. Thus, the
assumptions in the AMG roll up concerning the likely shape of the Kyoto mechanisms and
the response of CanadaÕs trading partners, in particular the United States, to them are no
more than educated guesses. The analysis will need to be refined as more information
concerning the negotiations becomes available.
The second example relates to the estimated impacts on industry. Although the various
models all indicate that there will be considerable variability in the impacts across
industries, there is not agreement on which industries will be most affected. The differences
in the rankings reflect different assumptions about the capacity of particular industries to
adjust and different emphases on the importance of the impacts on suppliers, customers and
competitors on a given industry. Although the AMG has made some progress in examining
the competitiveness issue, further industrial analysis, perhaps focussing on specific
measures, will be required to gain more confidence in the industry impacts.
The third example relates to the application of a domestic emissions trading system. The
Tradeable Permits Working Group made significant progress in examining the principles
and features associated with such a system. It did not, however, agree on a method to
allocate the permits. The roll up required that some assumption (for example,
grandfathering, auction with or without recycling of proceeds) be made concerning
allocation. As noted in the report, the method Ð allocation of permits to households Ð
chosen by the AMG, solely for analytic reasons, results in inflationary pressures. The point,
however, is that each allocation method will produce its own unique set of consequences.
Until there is a proposed approach to allocation, the AMGÕs findings concerning the
implications of a tradeable permit system should and must remain preliminary and
provisional.
Analytic Approach
The analytic approach to the roll up has three main components: a series of policy
packages, referred to as ÒPathsÓ; a set of framework assumptions, the most important of
which are the international scenarios describing the likely shape of the Kyoto mechanisms
and the response of the United States to them; and a modelling structure within which to
examine systematically the various path-scenario combinations or other sensitivities. Each
component is described briefly below.
The Paths
The National Air Issues Coordinating Committee (NAICC) directed the AMG to examine
five policy packages or Paths. Although somewhat complex in terms of their specification,
the Paths are differentiated by different degrees of reliance on specific measures and
tradeable permit systems and by the imposition of sectoral versus national targets. The Issue
TablesÕ options are present in all Paths, either explicitly or implicitly, in the sense that the
underlying actions are available for selection if they meet certain cost effectiveness criteria.
The five Paths are:
Path 0: the integrated summation of the Issue TablesÕ options.
Path 1: each sector achieves a minus 6 percent target using the TableÕs options,
supplemented by a tradeable permit system for electricity generation and a motive
fuel tax in transportation.
Path 2: an optimized approach in which the minus 6 percent target is established
nationally and measures and actions are taken in order of cost effectiveness. As
one of the options proposed by the Tradeable Permits Working Group (TPWG), a
tradeable permits system is established jointly for large emitters (essentially
electricity generation and large industry which account for roughly 35 percent of
total emissions). Other sectors employ specific measures based on lowest abatement
costs.
Path 3: employs the same permit system as in Path 2, but the joint cap is fixed at 6
percent below 1990 levels. All other sectors employ the specific measures used in
Path 1.
Path 4: based on the alternative TPWG proposal, a tradeable permit system is
established over the largest portion of the economy as practical. The resulting
system covers about 85 percent of emissions. The remaining sectors, such as
landfills, agriculture and non-combustion emissions related to oil and gas
production, use specific measures to achieve the required emissions reduction.
As can be seen, the Paths move, more or less progressively, through increasing reliance on
a major economic instrument to achieve the Kyoto target. As well, Paths 1 and 3 explore
the implications of an equivalent target for each sector while Paths 2 and 4 examine,
through different mechanisms, a broad national target and measures and actions based on
cost-effectiveness.
The Kyoto Scenarios
A simulation analysis of this complexity requires a large number of external assumptions
each of which must be carefully considered in light of available evidence. The most
important of these are the Kyoto Scenarios.
The Kyoto Protocol incorporates a number of flexibility mechanisms that allow countries to
discharge a portion of their obligations internationally. At the time of writing, the form of
these mechanisms is not well articulated and is the subject of ongoing international
negotiations. Nonetheless, it is important to understand the implications of potential
outcomes of these negotiations for CanadaÕs capacity to meet its Kyoto target. Equally
important is the likely response of CanadaÕs trading partners, in particular the United States,
because of the repercussions, principally through trade, on Canadian economic activity.
In order to cover a range of responses three scenarios were developed. In the Canada Acts
Alone Scenario, Canada is assumed to be the only country to undertake its Kyoto
commitment. This Scenario was developed primarily to show the impact of Canadian
emissions reduction policies in isolation from the potential effects that may occur as a result
of the policies of other countries. It is not, admittedly, a likely scenario.
The two remaining Scenarios were developed from a 1998 study by the Energy
Information Administration (EIA) of the U.S. Department of Energy. The EIA study
analysed several cases in which the United States achieved its Kyoto target using different
combinations of domestic and international permit trading. Two such cases, referred to in
this report as Kyoto Loose and Kyoto Tight, were selected.
Kyoto Loose posits a situation in which, by 2010, there is a well-established international
permit trading system with low transaction costs, buy-in by some developing countries and
plentiful JI and CDM opportunities. Under such circumstances, the EIA study suggests that
the U.S. can discharge 75 percent of its obligations internationally at an international permit
price of C$24 per tonne of CO2. This outcome approximates the Clinton AdministrationÕs
preferred position.
By contrast, Kyoto Tight assumes that the international permit trading system is not as well
developed, with consequently higher transactions costs, that participation by developing
countries is limited and that CDM and JI opportunities are constrained. This results,
according to the EIA study, in the U.S. discharging only one-third of its obligations
internationally, meeting the rest through domestic action. The resulting permit price is C$58
per tonne of CO2. Kyoto Tight can be interpreted as the minimum conditions for U.S.
ratification.
The primary purpose of using the EIA study is to obtain a credible range of international
carbon prices to incorporate in the domestic analysis. An additional benefit is that the study
also provides a comprehensive analysis of the assumed U.S. actions on energy prices,
energy imports and economic activity with which to frame the domestic analysis.
The Modelling Structure
To evaluate the various path-scenario combinations systematically, the AMG linked
together a number of specialized models available from the private sector or within
government into an overall modelling structure. The approach is to use the outputs of one
set of models as inputs to the others. The three sets of models are:
Micro models: These models evaluate the direct impacts - required investment, changes in
energy flows and GHG emissions reductions - associated with path-scenario combinations
under the deliberately imposed assumption that economic activity remains largely
unchanged. Two energy-technology models Ð the Market Allocation Model (MARKAL)
and the Canadian Integrated Modelling System (CIMS) - were selected for this purpose to
capture different modelling perspectives. MARKAL is an optimizing model that provides
the lowest financial cost solution to the achievement of a constraint, such as a stipulated
emissions target. CIMS, by contrast, incorporates evidence concerning actual consumer and
producer experience and allows for responses to the proposed policy that take into account
non-financial considerations. Other things equal, MARKAL should produce a lower cost
outcome than CIMS. MARKAL and CIMS also differ in their approach to energy pricing.
The former employs marginal cost pricing so that the price of carbon is determined by the
last tonne abated. CIMS assumes average cost pricing, so that only the average cost of all
the emissions reduced is incorporated in the price. The distinction is not particularly
important for the pricing of oil products or natural gas. It is, however, crucial to the pricing
of electricity. The emissions profile and other circumstances in agriculture are so different,
compared to other sectors, that special models maintained by Agriculture Canada were
employed and their results were combined with the CIMS and MARKAL results.
Macro models: Macro models evaluate the overall economic consequences - changes in
economic activity, employment, trade and competitiveness and government balances - from
a policy shock. Two models were also employed for this purpose - the Informetrica Model
(TIM) operated by Informetrica Ltd and the Canadian Sectoral General Equilibrium Model
(CaSGEM) developed by the Department of Finance. Both models contain considerable
sectoral detail and can produce results at a provincial level. They differ, however, in
approach. TIM focuses on the adjustment process as the economy responds to the new
policy and, therefore, allows for transitional under or over-employment of capital and
labour. CaSGEM employs a general equilibrium framework and focuses on the long-term
result of the policy change after all adjustments have taken place. CaSGEM is also more of
a hybrid model, incorporating both micro and macroeconomic features. It can provide,
therefore, a fully integrated solution covering both the changes in economic activity and the
emissions consequences of those changes. Thus, while CaSGEM does incorporate some of
the results from the micro models, it is probably best to think of it as providing an
complementary view to the complex of MARKAL/CIMS - TIM results.
Environmental and health models: To address the environmental and health consequences
of the policy options, in particular the changes to health resulting from the reduction in
criteria air contaminants associated with the GHG policies, the AMG relied on a suite of
models maintained by Environment Canada. The spreadsheet model estimates the changes
in atmospheric pollutants Ð SOX, NOX, VOC, particulates and carbon monoxide Ð
associated with the various path-scenario combinations, using the energy change results
from the micro models as input. Several specialized models then translate these changes in
emissions levels into changes in ambient air quality. These results in turn provide input to
the Air Quality Valuation Model (AQVM) which estimates both the physical impacts on
health Ð mortality risk, hospital visits, etc. Ð and assigns a monetary value to these impacts.
The latter are based on estimates in the health literature concerning willingness-to-pay for
the avoidance of such risks.
In sum, the basic approach is to use the micro models to aggregate the direct impacts of the
various path-scenario combinations and then to use these results as the inputs to both the
macro and environmental and health models. To accomplish this sequencing, the AMG
imposed the assumption on the micromodelling approach that economic output (i.e., tonnes
of steel, volumes of chemicals, production of oil and gas) does not change. The limited
exception to this constraint concerned some aspects of transportation, such as kilometres
driven, because the policy intent of the measures was, in fact, to reduce this item.
The constant output assumption has generated considerable concern from the stakeholder
community and requires some further explanation. The constraint was not imposed because
the AMG believes that economic activity will be unaffected by GHG reduction policies Ð it
is almost axiomatic that it will. The reasons for the constraint are twofold. First, the energy-
technology models do not provide a good representation of the total economy and the many
complex linkages among sectors. They concentrate, quite appropriately, on energy
intensive processes, industries and activities. The macro models, by contrast, incorporate all
the major linkages and can address trade, competitiveness and fiscal and monetary policy
issues. An industry which uses little energy may be largely unaffected by the direct impacts
of the emissions reduction policies. It may, however, be greatly affected by the impact that
those policies have on either its suppliers or its customers. The macro models capture these
important second-round effects.
The second reason for the fixed output constraint is that the roll up analysis is
extraordinarily complex. If only the final impacts were reported, it would be impossible for
stakeholders to ascertain whether the initial measures had been incorporated correctly or
what their consequences might be. Keeping the output constant for the micro analysis
reduces the complexity and allows an interim Òreality checkÓ on the results before moving
on to the macro analysis.
Main Learnings
The AMG was asked to address the question Òwhat are the economic and environmental
consequences, for Canada, of achieving the Kyoto target?Ó While not a definitive answer,
the analysis provides some important insights into this question. These Main Learnings are
as follows:
¥ At the national level, attainment of the target results in sustained, long-term,
negative economic impacts. In the long run, the reduction in gross domestic product
(GDP), relative to the business-as-usual case, ranges from 0 to 3 percent depending
on the path-scenario combination.
It is important to provide perspective on these estimates. For example, a reduction
in GDP of 3 percent in 2010 means that, over the decade, the economy will grow
by about 26 percent instead of 30 percent as projected in the reference case. This is
equivalent to the loss of roughly one yearÕs growth, or, viewed in absolute terms, in
2010, the loss in annual economic output of approximately $40 billion (or $1100
per capita).
¥ The overall GDP impacts vary over time. Initially, economic activity increases
modestly in response to increased investment in emissions reducing technologies.
Thereafter, however, higher production costs, deterioration in competitiveness and
lower incomes combine to reduce GDP below business-as-usual levels. The
analysis also suggests that the adjustment process may not be completed until after
2010.
¥ The provincial GDP impacts are generally within 1.5 percentage points of the national
average impact. The relative ranking of each province typically varies by Scenario. In
the Canada Acts Alone Scenario, Newfoundland, Prince Edward Island, Quebec and
British Columbia are less affected, relative to the national average, whereas Ontario and
Saskatchewan are more negatively affected. The impact on Alberta is close to the
national average. The results for Nova Scotia, New Brunswick and Manitoba vary
between the studies.
In the International Scenarios, Newfoundland, Prince Edward Island and British
Columbia remain in the Òless affectedÓ category and are joined by Manitoba.
Saskatchewan remains in the Òmore affectedÓ category, which now includes Alberta
and New Brunswick. By contrast, OntarioÕs GDP impact is smaller than under Canada
Acts Alone, becoming close to the national average. QuebecÕs position is largely
unchanged across the Scenarios, although one study indicates that its GDP would be
consistently higher than in the business-as-usual case. For Nova Scotia, one study
indicates impacts that are greater than the national average, while the other suggests the
opposite.
¥ The findings suggest the potential for substantial variability in GDP impacts across
industries. Unfortunately, it is not possible to identify unambiguously the sectors
that would be negatively or positively affected, since industry variation is not
uniform across models and Paths.
¥ The greatest potential for emissions reduction appears to reside in the electricity
generation sector Ð between 40 and 60 percent of the reduction. Two actions Ð
capture and storage of CO2 in aquifers in Alberta and Saskatchewan and enhanced
interprovincial hydro-electricity trade, in particular from Manitoba and Quebec to
Ontario Ð account for the bulk of this potential.
¥ Policies to reduce greenhouse gases will both reduce energy consumption and
encourage switching from more to less carbon-intensive fuels. All of the analysis
suggests some declines, relative to business-as-usual, in oil product and coal
consumption. Interestingly, natural gas consumption also declines largely because
the capture and storage of CO2 and enhanced hydro electricity trade reduce the
attractiveness of gas-fired electricity generation. Were these options not fully
available, natural gas demand would increase.
¥ The industrial sector, particularly the oil and gas industry, and the transportation
sector face significant challenges in achieving large emissions reduction.
¥ Were Canada to act alone in achieving its Kyoto target, the marginal cost of
abatement in 2010 could range from $40 to $120 per tonne of CO2. Were these
costs to be incorporated in energy prices, gasoline prices would increase by 13 to
35 percent, natural gas prices (for residential use) by 30 to 75 percent, and average
coal prices by 300 to 800 percent.
¥ The outcome of the negotiations concerning forestry and agriculture sinks is an
important factor in the cost to Canada of achieving the Kyoto target. According to
one estimate, a pessimistic assumption concerning this outcome, the effect of which
widens the gap by about 20 percent, results in an increase in the marginal cost of
abatement from $57 to $100 per tonne of CO2. This result also suggests that the
Canadian emissions abatement cost curve, constructed from the analysis and
insights of the Issue Tables, becomes increasingly steep as the target is approached.
¥ The analysis supports the contention that competitiveness Ð measured by changes in
productivity - will be adversely affected by the achievement of the Kyoto target.
The impact is somewhat attenuated if CanadaÕs trading partners are also committed
to attaining their targets and Canada has access to flexibility mechanisms. However,
under this Scenario, CanadaÕs trading partners will also face reductions in economic
activity, with adverse consequences for Canadian export performance. At the
national level, the net impact of these two forces is to reduce the GDP loss by about
0.5 percentage points.
¥ The analysis strongly supports the conclusion that moving from individual sector
targets to an economy-wide target will achieve the desired objective at significantly
lower cost. Moreover, sector specific emissions targets do not distribute the
economic burden evenly across sectors.
¥ Based on preliminary modelling of some characteristics of emissions permit trading,
this instrument can be viewed as cost-effective mechanism for achieving emissions
reductions in the industrial and electricity sectors. However, the analysis to date
underscores the importance of the design of such an instrument, in particular the
permit allocation mechanism. Each allocation method carries with it different
distributive effects on the economy.
¥ Measures and actions to achieve the Kyoto target will also result in the reduction of
sulphates, ozone and other atmospheric pollutants. These reductions will lead to
ancillary benefits from improved air quality and improvements in human health.
These co-benefits are immediate, local and reasonably certain and can make a
significant contribution to the attainment of clean air goals as enunciated in the
Canada-Wide Standards Initiative.
¥ The analysis indicates that the societal benefits of these improvements in human
health are in the range of $300 to $500 million per year. Most of this value derives
from reduced risk of mortality. These societal benefits represent the value that
Canadians place on these co-benefits, as determined by estimates of their
willingness-to-pay to achieve these avoided impacts. They are not comparable to
the GDP impacts noted above. These estimates do not cover the full spectrum of
pollutants and do not include sulphate reductions in western Canada.
¥ Although increased reliance on the international mechanisms to reduce greenhouse
gases lowers the cost of achieving the Kyoto target for Canada, it also reduces the
domestic clean air benefits. The analysis suggests that this reduction in societal
benefits is on the order of $200 million per year.
Analysis is most useful when it succeeds in resolving issues. Even when it fails to do so,
however, it is still valuable if it points to gaps in our understanding. The AMG has
identified the following areas where future analysis should yield useful insights:
¥ Much greater effort is required to measure welfare benefits and costs. Attention
should focus on the welfare implications of transportation and Òlife styleÓ change
initiatives.
¥ Some progress has been made in identifying competitiveness impacts, but a much
more sophisticated understanding of this issue is required. In particular, there is a
need to evaluate the importance of the so-called ÒleakageÓ issue: the possibility that
some industries will lose market opportunities and investment to developing
countries not subject to emissions reduction targets.
¥ Much more analysis of the implications of the various approaches to the design of
an emissions trading system is required. The suggested priority area is the allocation
mechanism.
¥ More province-specific analysis is needed. The current macroeconomic models
develop provincial impact estimates by distributing the national results according to
the industrial mix in each province. This approach is reasonable if the characteristics
of an industry are similar across provinces, but questionable if this is not the case.
More refinement of this assumption is required.
¥ The assumptions concerning how the United States might address the Kyoto
Protocol requirements are too simplistic to comprehend the complex ways in which
that countryÕs climate change policies could affect both CanadaÕs economy and our
policy options.
¥ The analysis, to date, has not comprehensively modelled how CanadaÕs other
trading partners might respond to their commitments under the Kyoto Protocol.
¥ Despite considerable progress, the quantitative co-benefits analysis requires further
development.
Concluding Remarks
In focussing on the substance of AMG analysis, this executive summary has perhaps given
insufficient attention to how the task was accomplished. The AMG deliberately created a
transparent, interactive process involving the stakeholder and modelling communities.
Largely through the multi-stakeholder workshops organized by the Climate Change
Economic Analysis Forum (CCEAF), the various phases of the analysis were presented,
reviewed and critiqued. The dialogue generated by these discussions and the network of
expertise established via this process are important achievements, providing a solid basis for
further cooperative analytic effort as Canada develops its strategy on climate change.
In the above context, it is important to obtain feedback from the stakeholder community on
the process. Therefore, at the request of the AMG, CCEAF surveyed the stakeholder and
modelling communities concerning their views on the analytical process and suggestions
for the future. The results of that survey are contained in an annex to this report. The results
provide a useful counterpoint to the main body of the report and will be of considerable
value to the AMG as it prepares its recommendations to NAICC concerning the priorities
for and organization of the post-JMM analysis.
In addition to this report, the AMG has generated a large number of studies focussing on
the microeconomic, macroeconomic and environmental and health phases of the roll up
analysis. These, in turn, are supported by numerous reports, prepared by CCEAF and
others on specific topics. The AMG believes that this research will form a valuable base
for further climate change analysis.
Chapter 1
Introduction
In December 1997, in an international agreement known as the Kyoto Protocol, Canada
agreed to reduce its greenhouse gas emissions by 6 percent on average, relative to 1990 levels,
over the period 2008-2012. At the same time, other industrialized countries made similar
commitments, albeit for differing percentages, with an overall average reduction of 5.2 percent.
Shortly after the conclusion of these negotiations, First Ministers, meeting in Ottawa, instructed
federal, provincial and territorial Ministers of Energy and Environment (Joint Ministers) to
develop a process to achieve Òa thorough understanding of the impact, the costs and the
benefits of the ProtocolÕs implementation and of the various implementation options open to
Canada.Ó
In April 1998, the Joint Ministers announced a process for the establishment of the National
Climate Change Process (NCCP). The centrepiece of the process was the creation of fifteen
Issue Tables, comprised of experts from governments and the stakeholder community, to
provide advice on various aspects of the implementation strategy. Some of the Issue Tables
were to address cross-cutting issues such as public education and outreach, enhanced voluntary
action and the Kyoto mechanisms.1 Other Tables - Agriculture, Buildings, Electricity, Forest
Products, Industry, Municipalities and Transportation - were asked to focus on the
development of options to achieve at least the 6 percent reduction within the sector they
represented.
The April 1998 announcement also created the Analysis and Modelling Group (AMG), a
federal, provincial and territorial organization with the overall responsibility for the quantitative
analysis of the economic and environmental consequences of the potential avenues to achieving
CanadaÕs Kyoto target. This responsibility included the development of a reference outlook to
frame the work of the Tables and guidance to the Tables on critical methodological issues. The
AMGÕs central task, however, is to conduct the integrated analysis of the economic, social and
environmental implications for Canada of achieving its Kyoto commitment. This is referred to
as the roll-up analysis.
1
Kyoto mechanisms include International Permit Trading, Joint Implementation (JI) and the
Clean Development Mechanism (CDM).
The primary objective of this report is to present the findings based on the results of the roll-up
analysis undertaken by the AMG. In essence, the roll-up involves the evaluation of the
integrated impact of the Options Packages analyzed by the Issue Tables, along with other
ÒpathsÓ for the attainment of the Kyoto objective specified by the National Air Issues
Coordinating Committee (NAICC). The analysis also considers the impacts, under different
Òscenarios,Ó concerning the likely shape of the Kyoto flexibility mechanisms and the consequent
response of CanadaÕs trading partners, in particular the United States.
To undertake this analysis, the AMG has created an integrated modelling structure which links
together a number of detailed energy-technology (microeconomic), agriculture, macroeconomic
and air quality-health models. Two energy-technology models, reflecting different modelling
perspectives, optimizing versus behavioural, have been employed. The agriculture emissions
were modelled using the Canadian Economic and Emissions Model for Agriculture maintained
by Agriculture Canada. The macroeconomic analysis was undertaken with an econometric
model. A fourth model, using general equilibrium principles, which combines elements of the
energy-technology and the macroeconomic models provided another perspective to the
analysis. Environmental and Health Impact (EHI) modelling was undertaken using Environment
CanadaÕs Air Quality Valuation Model. Also, working largely through the Climate Change
Economic Analysis Forum (CCEAF), the AMG has developed a set of framing assumptions on
which to base the analytical efforts. These assumptions help to define key variables such as the
concept of the Kyoto mechanisms, the response of the United States and impacts on
international energy prices.
This analysis should not be construed as a plan to achieve the Kyoto target because:
¥ the international negotiations to define the Kyoto mechanisms are too fluid to allow
much more than educated guesses concerning the likely outcome. As well, the
responses of CanadaÕs trading partners, particularly the United States, which are
predicated on the outcome of the negotiations, are uncertain.
¥ the reference case, although developed with a high degree of consultation, is one view
of the future. The results depend on the assumptions incorporated in that outlook.
¥ the modelling structure does not have sufficient resolution to analyze specific policies.
¥ there are a large number of inter-locking assumptions. While it is believed that these
assumptions are reasonable, it is clear that changing any one of them will alter the
results, possibly in a major way.
¥ the analysis assumes a certain timing and penetration of the measures; variations from
this could have a significant effect on some of the results.
This analysis employs the best available models with sound analytical groundings. The objective
is to provide a plausible range of outcomes and to provide insights on, and general guidance
concerning the following issues:
¥ the relative economic impact of two of the more significant policy approaches - a
combination of specific measures and the use of a major economic instrument.
¥ identification of options that appear to offer significant emissions reduction potential and
conversely, those where the potential seems to be limited and/or costly.
¥ the implications of applying common sectoral targets as opposed to an overall national
target.
¥ the benefits Canada might derive from access to the international flexibility mechanisms
to meet, at least, a part of its commitment.
¥ key areas of uncertainty, such as the future degree of inter-provincial trade in the
electricity sector, the amount of carbon that can be absorbed by sinks and the potential
for capture and storage of CO2, for which different assumptions can dramatically alter
the results.
¥ information concerning the potential distribution of the regional and sectoral costs and
benefits and their economic impact of achieving the national target.
¥ the potential health impacts associated with the reduction of air pollutants that may result
from a GHG reduction strategy.
The rest of this report is organized as follows:
¥ Chapter 2 reviews the options packages developed by the Issue Tables and the Paths
used in the analysis.
¥ Chapter 3 provides an overview of the analytic approach. It includes descriptions of
the modelling structure employed in the analysis, and of the framework assumptions.
As well, there is a discussion pertaining to understanding the results and some caveats
concerning the results.
¥ The microeconomic (energy-technology) results for the various path-scenario
combinations are presented in Chapter 4. These results are the direct implications
associated with the policy packages, on the assumption that the level and composition
of economic activity is largely unaffected by their introduction. The microeconomic
estimates from Chapter 4 and the options packages from the Issue Tables provide the
input to the macroeconomic and environmental and health impacts of Chapter 5 and 7
respectively.
¥ Chapter 5 presents the results of the macroeconomic analysis. The macroeconomic
analysis examines the consequences, at the national, sectoral and regional levels, for
economic performance, employment, trade, competitiveness and government finance of
the various path-scenario combinations.
¥ Chapter 6 provides a complementary view of the macroeconomic aspects of the
analysis using a general equilibrium model developed by the Department of Finance.
¥ Chapter 7 explores the environmental and health implications of the some of the path-
scenario combinations. The analysis largely focuses on the changes in criteria air
contaminants (SOX, NOX and particulates) associated with the various GHG reduction
approaches. It provides estimates of the physical effects and the value of these changes
on human health and environment. The chapter also provides a qualitative assessment
of the implications of GHG reductions on other pollutants.
¥ Chapter 8 summarizes the main findings and conclusions from the analysis, notes its
limitations and suggests the most important areas for future research.
¥ Annex I contains the executive summary of the report prepared by CCEAF, at the
request of the AMG, on the views if the analytic process to date and the priorities for
future climate change analysis.
¥ It is recognized that there is a wealth of information contained in the modelling reports
and that analysts will continue to use this information to gain further insights, refine the
issues and examine the uncertainties. The background studies and detailed reports from
the modellers are listed in the bibliography.
Chapter 2
Option Packages and Paths
This chapter describes the development of the Issue TablesÕ options to reduce GHG emissions,
which are the cornerstone of the roll-up analysis, and will define the analytical paths requested
by the NAICC.
2.1 Issue Table Options
As noted earlier, the NCCP announcement created 15 Issue Tables, including several
specialized task groups, consisting of over 450 experts from a broad cross-section of
government, industry, the academic community, environmental groups and non-governmental
organizations. The objective of the Tables is to develop options, examine and analyze their
impacts, costs and benefits, within their sphere of expertise, to reduce GHG emissions. Tables
representing the main sectors of the economy were requested to develop options to achieve at
least a 6 percent reduction in GHG emissions, below the 1990 levels, in their particular sector.
The options, as presented in the Issue TablesÕ Option Reports, are the foundation of the AMG
roll-up analysis. For this analysis, the most important of these Option Packages were
developed by: Buildings; Municipalities; Industry; Electricity; Transportation; Agriculture; and
Forest Products. In addition, the roll-up analysis explores suggestions from the Tradable
Permits Working Group (TPWG).
2.2 Option Packages
The Option Packages reports proposed a wide range of measures and actions2 such as
regulations, information programs, financial incentives and partnerships. The AMG worked
closely with the TablesÕ representatives to ensure consistency of the analysis, without changing
the meaning and intent of the Option Packages.
In the case of the Industry and Forest Product Tables, although a large number of actions were
examined, these tables were unable to develop measures that would motivate the
2
A measure is a combination of a specific GHG reduction action and the policy instrument
that implements it.
actions. With the assistance of the AMG, a set of generic options,3 including enhanced
voluntary action, cogeneration and subsidies was constructed.
In total, 106 measures were used in the
roll-up analysis (Chart 2.1). About one
third of the measures relate to Buildings,
which covers both the residential and
commercial sectors. The Municipalities
Table proposed eight direct measures
and five enabling measures.4 In the roll-
up analysis, these measures were
assigned to the commercial and
transportation sectors. As noted above,
the Industry and Forest Product Tables
agreed with the generic measures as
estimated by the AMG. Underlying these
generic measures are about 100 other
Chart 2.1
Issue Table Options for the Roll-Up
Table
Number of
Remarks
Options
Buildings 34
Municipalities 8
Industry 3
Electricity 2
Transportation 48
Agriculture 10
Forestry 1
Total 106
A and B packages. Exceeds target.
Included landfill gas. Meets target.
Includes generic measures. Does not meet target.
Does not meet target without permit trading.
Meets target.
Meets target, if soil sinks included.
Included with Agriculture.
actions that were examined by the Industry sub-Tables.
The Electricity Table provided two specific measures. Without endorsing specific targets for
reducing GHG emissions, the main recommendation of the Electricity Table was that the most
efficient way to achieve substantial GHG reductions would be to set emission caps and allow
firms to meet targets by trading emission permits. It also proposed a 2.5 cents per KWh
subsidy for renewable technologies. This Table also undertook a review of actions available to
reduce greenhouse gas emissions in the sector, such as fuel switching, expansion of interprovincial
trade and the capture and storage of CO2 in deep aquifers. This research was used
to inform the roll-up analysis.
The Transportation Table proposed a large number of measures with the majority being
described Òmost promisingÓ and Òpromising.Ó However, measures in these two categories were
not sufficient to reduce emissions to meet the Kyoto target in the transportation sector.
Therefore, on the advice of the Transportation Table, four Òless promisingÓ measures were
added to meet the target. In addition, the Transportation Table undertook an extensive analysis
of motive fuels taxes, but did not suggest this option as part of a general package.
The Options Package from the Agriculture and Agri-Food Table includes 10 measures. These
generally pertain to farm management practices (no-till, summerfallow and grazing strategies)
3
The Industry Challenge, ERG/M. J. Jaccard and Associates, November 1999
4
These measures include education and suasion.
and agro-forestry activities. Several of these measures are predicated on the inclusion of soil
sinks in the final version of the Kyoto Protocol. The Afforestation measure, recommended by
the Forest Products Table, is included with the agriculture and other sector.
The TPWG provided advice on the extent of the coverage for the sectors that may participate
in an emissions trading systems.
2.3 The Paths
The AMG roll-up analysis is designed to explore a range of policy directions to achieve the
Kyoto target. The main issues to be explored are the attainment of sector targets versus an
economy-wide target and a set of specific measures versus a major economic instrument. The
NAICC specified five approaches, referred to as Paths, to understand these issues.
The Paths assess various combinations of specific measures, sector targets and emission
trading.
Path 0: Tables Measures. This Path is the aggregation of the all measures, including
the four Òless promisingÓ Transportation measures, identified in the Options Packages
by the Issue Tables. This aggregation is not a simple summation because of the
interaction among the measures. It was apparent that the industry sector could not
achieve its target, therefore, only the generic measures, costing up to $75 per tonne of
CO2 equivalent, were included. Since it was clear that this Path would not achieve the
target it is referred to as Path 0.
Path 1: Sectoral Targets. Each sectorÕs emissions (electricity, industry, residential,
commercial, transportation and agriculture) for the period from 2008 to 2012, are
limited to 6 percent below the 1990 levels. With the exceptions noted below, the
Tables measures were employed to meet the target. In the Industry sector, generic
measures were implemented up to a cost of $300 per tonne of CO2 and the four Òless
promisingÓ measures from Transportation were replaced by a motive fuels tax, sufficient
to meet this sectorÕs cap. Consistent with the recommendation of the Electricity Table,
sector-based emission trading is incorporated as a measure in the Electricity sector.
Path 2: Permit Trading for Large Emitters and an Economy-wide Target. The
main feature of this Path is that it replaces the sectoral targets with an economy-wide
target of 6 percent below the 1990 level, allowing the selection of a least cost set of
measures. All of the TablesÕ measures are available for selection by the residential,
commercial and transportation sectors based on their cost. This Path examines one of
the proposals from the TPWG. The tradable permit system is applied to the Large
Final Emitters (LFE), which are defined as the Electricity sector and major industries,
such as pulp and paper, iron and steel and cement.
Chart 2.2
Extent of Permit Trading in Paths 2, 3 and 4
Percent of Emissions Covered
Paths 2 and 3 Path 4
Oil Sands 100 100
Gas Pipelines 100 100
Petroleum Refining 100 100
Electricity 95 100
Chemicals 80 100
Pulp and Paper 80 100
Smelting and Refining 80 100
Cement 80 <100
Iron and Steel 80 <100
Other Industries 55 <100
Gas Processing (CO2 only) 50 to 70 50 to 70
Commercial 0 100
Residential 0 100
Road Transport 0 100
Air Transport 0 100
Rail Transport 0 100
Other Emissions 0 100
Landfills 0 0
Oil and Gas Processing (CH4 only) 0 0
Oil Processing (CO2 only) 0 0
Total Coverage 35 85
Provided by the Tradable Permits Working Group
The extent to which these sectors and sub-sectors are allowed to use tradable permits
was provided by the TPWG (Chart 2.2). About 35 percent of total national emissions
are covered by the trading system. In this Path, some sectors reduce their emissions by
more than 6 percent, while other sectors do not.
Path 3: Permit Trading for Large Emitters and Sector Specific Targets. The
coverage of the trading system is the same as Path 2. However, the LFE are capped at
6 percent below their 1990 level. Other sectors, not included in the emission trading
system, such as residential, commercial, small firms and transportation, meet their
respective 6 percent target using the Tables measures defined in Path 1, including
measures valued up to $300 per tonne of CO2 for small industry and a motive fuels tax
in Transportation.
Path 4: Broadest Practicable Permit Trading. This Path reflects the second major
approach from the TPWG. The emission trading system, which caps the emissions at 6
percent below the 1990 level, is expanded to include all sectors for which permit
trading is administratively feasible. The sub-sectors excluded from permit trading for
administrative reasons are: carbon dioxide from conventional oil processing, methane
from oil and gas processing, small firms and emissions from landfills. The sub-sectors
achieve their target reductions through the Tables measures or generic measures costing
up to $300 per tonne of CO2. In this Path, about 85 percent of total emissions are
covered by permit trading.
The operationalization of these Paths, for analytic purposes, involves a considerable amount of
detail, which can obscure the policy direction that the Path is designed to examine. Chart 2.3
provides a summary of the policy intent of each Path, across the two dimensions: specific
measures and permit trading
and sectoral versus economy-
wide targets (optimization). As
Chart 2.3
the analysis moves through the
Summary of Paths by Policy Direction
Paths, defined measures are
reduced, and permit trading is
expanded to cover a larger
segment of the economy. Path
1, and to a lesser extent Path 3,
represent approaches in which
each sector is required to
achieve a common target. By
contrast, Paths 2 and 4
examine less restrictive
Emission Trading Measures Optimization
Path 0 No All No
Path 1 Electricity only All Others Electricity only
Path 2 Large Emitters All Others Economy-wide, no sector targets
Path 3 Large Emitters All Others
For Large Emitters only, Sector
targets for all others
Path 4 Broad as Practical Remainder
For Broad as Practical, with an
Economy-wide target
approaches in which least cost
solutions are sought to achieve
the Kyoto target. The main
distinction between Path 2 and Path 4 is that, in the former, the lowest cost measures are
selected, whereas in the latter, higher energy prices, owing to the cost of the permits, cause the
action underlying the measures to occur. While the direct impacts of Paths 2 and 4 are
generally similar, the question of who bears the cost of GHG reduction may have sectoral
implications in the macroeconomic analysis.
Agriculture and
Sinks Models
Net GHG reductions / removals
net criteria pollutants reductions
required private and public investment
energy savings
ÒmicroÓ competitiveness impacts
Economic Models of Canada
by Sector and Province
Energy / Technology Models
by Sector and Province
GDP, employment, trade,
ÒmacroÓ competitiveness,
government finance impacts
Air Quality
Health and
Environment Model by
province
health and
environment
impacts
Inputs for AMG Report
on Impacts and Implications
Options Packages
Reference Case
Assumptions
Simulation
Assumptions
International
flexibility
mechanisms
US/World
Scenarios
Chapter 3
Analytical Framework
3.1 Overview
The roll-up is designed to capture
Chart 3.1
many energy, economic and The Roll-up Framework
environmental dimensions of the Paths Issues Tables
to the achievement of the Kyoto target.
This task requires a comprehensive
and systematic analytical framework.
The framework constructed by the
AMG comprises two components
(Chart 3.1).
An integrated system of energy,
agriculture, macroeconomic and
environmental-health models to
analyze systematically the Paths. The
approach uses specialized energy and
agriculture models to evaluate the
direct consequences of the
investments, energy prices, the level
and mix of energy and the resulting
emissions reduction in each sector and
each Path. These results were then
used to assess the impacts on the whole economy and on environment and health.
A set of assumptions to frame the domestic analysis. These assumptions, which were
largely developed through extensive stakeholder consultations, include a reference case
against which to estimate the impacts, considerations of the likely shape of the Kyoto
mechanisms, the extent to which they may be used by Canada and the consequent
responses of CanadaÕs trading partners, particularly the United States.
The Options Packages from the Issue Tables provide the principal input to the energy-
technology models, which in turn generate direct investments, fuel cost savings and fuel
mix. The direct investments and savings are inputs to the macroeconomic models which
also use information from the Options Packages. The fuel use estimates from the energy-
technology models are inputs to the EHI models.
3.2 The Models
Chart 3.2
The Modelling Structure
Energy Technology (micro) Models
(1) Optimisation (MARKAL)
¥ Lowest financial cost to achieve specified reduction
(2) Behavioural (CIMS)
¥ Select options or assume degree of penetration based
on consumer and behavioural preferences
Agricultural Model
(CEEMA)
GHG emission reductions, private and
public investment energy savings
+
Criteria pollutant reductions
Macro-Economic Models
(1) CGE (CaSGEM)
¥ Long-term implications after adjustment
(2) Econometric (Informetica)
¥ Adjustment path
Air Quality, Health &
Environment Model
(AQVM)
Competitiveness, employment, trade, government finance impacts,
health & environmental impacts,
The modelling structure (Chart 3.2) comprises four parts: energy-technology
(microeconomic), agriculture, macroeconomic and environmental and health impacts.
A complete description of these models is contained in the detailed reports from each of the
modellers.
Energy-Technology Models
The energy-technology or microeconomic models examine the choices that individual
economic agents, such as consumers and businesses, make in the purchase and use of
energy. These models focus on the influence of various factors, such as energy prices and
government policies on the choices that the individual agents make. The models integrate
the implications of combining multiple GHG reducing measures. Summing the separate
impact of measures, in isolation, may not yield consistent results. For example, a measure
to improve lighting efficiency and a measure to reduce the use of coal in electricity
generation will not have the same effect when combined than when treated separately. The
energy-technology models are designed to represent these interactions.
These models simulate the impact of changes in policy or market conditions against a
reference case. The main outputs consist of changes in energy use, energy prices, GHG
emissions, the investment cost and potential cost savings resulting from the policy change.
While the models provide the cost of GHG abatement, they do not assign the payment of
that cost to any particular part of the economy. These results are important to identify the
direct effects of GHG reduction policies. Also, the investments and savings are inputs to
the macroeconomic analysis, which assess the impact of those investments and monetary
flows on the whole economy. The change in energy mix, provided by the energy-
technology models, is a primary input to the EHI modelling.
There are two distinct perspectives in energy technology modelling. The first, known as
optimization, assesses options solely on the basis of financial cost. Given a specific
objective, such as a stipulated emissions target, an optimizing model will choose the
measures in order of lowest financial cost. By contrast, a behavioural model incorporates
evidence concerning consumer preferences (i.e., less tangible non-financial considerations,
such as quality or the type of technology) in determining which option to select or the
degree of penetration of a particular technology in the market place. In order to ensure that
these two perspectives receive full expression in the analysis, the AMG has selected two
models for the microeconomic analysis: the optimizing Market Allocation Model
(MARKAL) operated by HALOA Inc. from McGill University and the behavioural model
Canadian Integrated Modelling System (CIMS) developed the Energy Research Group
from Simon Fraser University.
MARKAL is an optimization model that integrates production, trading, transformation and
the end-use of various energy forms. It computes a regional equilibrium based on the long-
term, least total cost for the entire system. It assumes that the markets are fully competitive.
It also assumes that all agents (producers and consumers) minimize their own long-term
costs under the following conditions:
¥ Each agent has perfect information of all other agents over the model horizon of 40
years.
¥ Each agent adopts the long-term view to optimize its financial cost.
¥ All agents use the same discount rate.
¥ Electricity is priced using marginal costs, reflecting a fully competitive market.
By contrast, CIMS is a behavioural model that uses competitive attributes tempered with
the prospective behaviour. Behaviour may be the result of financial or non-financial
factors, such as quality. However, the perception of cost, by the purchaser, is the prime
determinant of the market share of a given technology; that is, technological characteristics
give the relative position of one technology to another and behaviour gives the market
share of the technologies. CIMS is also an equilibrium model that integrates decisions
regarding specific technologies, in the five major sectors, and decisions concerning energy
prices, supply and demand. The CIMS producers and consumers:
¥ Make decisions on limited information about the future.
¥ Make decisions for reasons other than financial cost.
¥ Use different discount rates reflecting their payback period and their perceived risk.
¥ Use electricity prices based on average costs, reflecting a cost of service approach.
The different pricing philosophies have important implications for the results, particularly
for electricity. In the CIMS model, the price is determined by the average cost of
producing electricity. The end-users pay an incremental price for electricity which just
covers the total cost of investments to reduce greenhouse gases. In MARKAL, electricity
producers set the price to reflect the marginal (incremental) cost of producing output. The
marginal cost price generated by MARKAL is significantly higher than the average cost
price determined in the CIMS model. In the MARKAL model, users of electricity are
charged a price as if the cost of all emission reductions are as costly as the last emission
reduction action implemented. The result of marginal cost pricing is that the electricity
producing sectors receive extra profits because the additional revenue is greater than the
total cost of emission reduction policies. In reporting results for MARKAL, the extra
profits are assumed to stay with the electricity producers and are reflected as a cost to the
energy end-use sectors.
Both MARKAL and CIMS have a similar regional breakdown: six provinces and a
combination of the four Atlantic provinces into one region which were subsequently
estimated for the individual provinces by the AMG. The economic sectors, represented in
the models, are: electricity, upstream and refining, industry, residential, commercial and
transportation. The industrial sector comprises 10 major industries. MARKAL is
particularly strong in modelling the electricity generation industry, whereas CIMSÕ strength
is its representation of the industrial sector. Both models include all GHG except hydro-
fluorocarbons. However, neither model incorporates the non-energy emissions from the
agriculture sector.
The strength of both models is the detailed representation of energy technologies such as
space heating, industrial processes and electricity generation. The modelsÕ databases
describe the technologies in terms of installed capacity, unit cost and efficiency on regional
and sectoral basis. Options for future improvement, such as retrofitting or replacement of
existing technologies and fuels are included and used by the model, when economically
viable.
Agriculture Models
The Agricultural scenarios were analyzed using the Canadian Economic Emissions Model
for Agriculture (CEEMA) which consists of two major sub-models.
The first component of CEEMA is the Canadian Regional Agricultural Model (CRAM).
CRAM is a static model that optimizes the net benefits to producers and consumers. The
model integrates all sectors of primary agriculture on a regional basis. Supply response is
determined by the relative profitability of alternative crops. The model allows for both
inter-provincial and international trade in primary and processed products. It can also
incorporate the effects of government programs and policies.
The GHG emissions component of CEEMA links the activity levels generated by CRAM
to emission coefficients, in order to estimate emissions of CO2, CH4 and N2O from primary
production resulting from changes in land use and management practices. Emissions from
the production of farm inputs, food processing and off-farm transportation are also
estimated. The coefficients from the CENTURY model5 were used to measure the rate of
soil carbon sequestration.
Macroeconomic Models
Macroeconomic analysis integrates consumption, investment, production and trade
decisions in the whole economy. This analysis captures not only the interaction among
industries, but also the implications for changes in producer prices, relative final prices and
income. Also detailed are government fiscal balances, monetary flows, interest and
exchange rates. The AMG decided to use an econometric model which incorporates the
frictions and rigidities of the economy and therefore, provides insights to the adjustment
path. The Informetrica Model (TIM), provided by Informetrica Ltd of Ottawa was selected
for this analysis.
TIM analyzes the macroeconomic impacts at the national level and then distributes these as
sector effects across the provinces. For goods producers, the national model provides detail
to distinguish impacts provincially. Service industry output is then determined by the
requirements of the provincial goods producers, population and household incomes. In this
analysis, the regional differences in electricity generation and other energy production are
captured by detailed analysis that is reflected in the energy-technology models.
TIM is a dynamic econometric model of the Canadian economy. It has interdependent
relations between demand, industrial performance, cost of production and price formation.
It represents the spending of households, business and non-business. The effect of GHG
5
A model that predicts the accumulation of soil carbon.
reduction policies on business investment, consumption and other sources of demand
directly affect the operations of the supplying businesses. Their requirement for materials
and services indirectly affects all other businesses. Consequent changes to the income of
labour and business along this chain of supply induces further spending to provide
multiplier effects. The induced spending is also sensitive to consequent changes to unit
costs of production, and to Kyoto-related changes such as taxes, subsidies and the trading
system, as these are reflected in prices. Producer selling prices are detailed for the industries
represented in the model, which are used to determine the prices of final demand. TIM
incorporates input-output tables to link final demands to industrial output, thereby
representing the interdependence of industries and to determine selling prices used in the
calculation of final demand. To reflect changes in fuel-using technology, the results
reported by the energy-technology models are used.
TIM models these effects as an ÒadjustmentÓ over time and, as such, the system is not
always in equilibrium. As a result, there may be periods of under/over-capacity, higher
unemployment and imbalances in government current account and other-sector savings.
TIM has 750 categories of final demand and represents 133 industries at a provincial and
territorial level. It also has an international component to account for exports and imports,
which are detailed for approximately 100 commodities.
For the roll-up analysis, TIM uses the results from the energy-technology models, in
particular capital investment, energy savings and emission permit prices, as input to
calculate the impacts on economic activity, competitiveness, trade and government fiscal
position. Complementary, and consistent impacts on current spending are drawn from the
Issue Tables.
To develop a complementary macroeconomic view, the AMG used a general equilibrium
developed by the Department of Finance (Economic Studies and Policy Analysis
Division). This economic model includes some of the energy-technology features of CIMS
and MARKAL and some of the macroeconomic features of TIM, although not to the same
level of detail.
The Canadian Sectoral General Equilibrium Model (CaSGEM) is a static,6 computable
general equilibrium model of the Canadian economy. Relative price changes are the main
factor causing behavioural changes. Therefore, the elasticities - the relationship between
price changes and the changes in supply or demand - are critical to the functioning of the
model. The model is made up of 51 sectors which, together, produce 59 different goods
and services. The public sector collects taxes and transfers money in addition to providing
6
A static model produces a single result after the adjustment period when the model has
reached equilibrium. It does not provide any insights as to when this equilibrium may
occur.
goods and services. International trade is also represented. In the model, consumers
interact with producers through the relative price mechanism until the price-quantity
relationship is in equilibrium. Taxes are collected and redistributed through the government
part of the public sector. The model assumes perfect competition. Its output includes: fuel
prices and demand, gross domestic product and the marginal cost of GHG abatement.
Environmental-Health Models
The Environmental and Health Impacts were examined in both quantitative and qualitative
terms, using a number of approaches and analytical tools.
The quantitative assessment involves the following steps:
the use of a spread sheet model that defines the relationship between energy use and criteria
air contaminant (CAC) emissions to estimate changes in CAC for the reference case and
the Paths analyzed.
these results are used to estimate changes in ambient concentrations of pollutants (air
quality).
the Air Quality Valuation Model (AQVM) estimates the physical health and environment
impacts associated with the changes in air quality and estimates the economic value of the
effects that would be avoided by reducing emissions.
AQVM evaluates the costs and benefits of proposed climate change initiatives on air
quality. It is based on the damage function approach in which projected reductions in the
ambient air concentration of certain air pollutants are used to estimate reductions in various
human health and environmental effects. Economic values are then applied to these
physical impacts, largely using the concept of willingness-to-pay, based on wage
differential studies. Benefits estimates, produced by AQVM, provide a view of the value
that individuals place on avoidance of these negative health and environmental impacts.
The qualitative assessment examines the relative environmental and health impacts across
the Paths and Scenarios.
3.3 The Framing Assumptions
The analysis of the Paths requires a context, which is provided by the framework
assumptions. The two principal components of this framework are a reference case and a
set of views regarding the shape of the Kyoto Mechanisms with the consequent reactions of
CanadaÕs principal trading partners. These elements are discussed below. There are other
framing assumptions required which are germane to specific phases of the analysis. These
are discussed in later chapters with the corresponding results.
The Reference Case
A reference case provides a policy-as-usual base against which to evaluate the impact of
the Paths. In May 1999, the National Air Issues Coordinating Committee on Climate
Change (NAICC) requested that the AMG provide this reference case. The AMG decided
to update the previous reference case7 incorporating new methodologies and data, as well
as insights from the Issue Tables and other stakeholders.
The new report, CanadaÕs Emissions Outlook - an Update (CEOU), was released in
December 1999. This report projects that, in a Òpolicy-as-usualÓ environment, GHG
emissions will rise from 601 Megatonnes (Mt) of CO2 equivalent in 1990 to 764 Mt in
2010. For a consistent evaluation of the Paths, all the economic models were calibrated to
the assumptions and results of the CEOU.
CanadaÕs Kyoto target is to reduce its GHG emissions to 565 Mt in 2010, 6 percent below
the 1990 level. Based on this projection, the required reduction, in 2010, is 199 Mt which is
26 percent below the reference case. The CEOU did not include estimates of agriculture
and forestry sinks because the estimates are contingent upon the outcome of the
international negotiations regarding the accounting for land use changes and forestry in
GHG inventories. At this time, there is considerable uncertainty regarding the definition of
forestry sinks and the inclusion of agricultural sinks. However, for analytic purposes an
estimate of agriculture and forestry sinks is required. Based on advice from the Sinks and
Agriculture Issue Tables, it is assumed that the negotiations will permit the use of 10 Mt for
Forestry sinks and 6 Mt for Agricultural soil sinks. Under these conditions, the emissions
in 2010 are reduced form 764 to 748 Mt, hence the gap is reduced to 183 Mt.
The Scenarios
The Kyoto Protocol incorporates a number of flexibility mechanisms which allow countries
to discharge a portion of their obligations internationally. At this time, the form of these
mechanisms is not well articulated and is the subject of future international negotiations. It
is important to understand the implications of potential outcomes of these negotiations.
Equally important is the likely response of CanadaÕs trading partners, in particular the
United States. The United StatesÕ response will, because of the strong trade links, have
repercussions on Canadian economic activity. In order to cover a range of responses, three
Scenarios were developed by the AMG. The first Scenario, Canada Acts Alone, assumes
that Canada meets its Kyoto target solely by domestic emission reduction actions and no
7
Natural Resources Canada, CanadaÕs Energy Outlook 1996 to 2020. April 1997.
other country effects any emission reduction initiatives. Export prices and export demand
are maintained at the same level as in the CEOU. This Scenario is designed to show the
impact of Canadian GHG policy in isolation from the potential effects which may occur as
a result of implementation of GHG policies in other countries. The two International
Scenarios assume that GHG reductions can be achieved using different degrees of the
application of international mechanisms.
The AMG requested CCEAF to
develop two scenarios of international
actions that would represent a plausible
range of the flexibility mechanisms.
CCEAF recommended that the study
prepared by the Energy Information
Administration (EIA) of the U.S.
Department of Energy,8 form the basis
of the Scenarios. The EIA study
analyzed eight cases for the U.S. to
meet its Kyoto target. Each case was
based on different proportions of
Chart 3.3
EIA Cases
Percent Change from the Base Case
Kyoto Loose Kyoto Tight
2005 2010 2020 2005 2010 2020
Change in GDP 0 -1 <-1 <-1 -2 <-1
Change in inflation 0 1 1 2 3 2
Natural gas export prices 1 0 13 1 18 38
Natural gas exports -1 -1 3 0 7 9
Crude oil export prices (light including
synthetic; and heavy including bitumen)
-1 -4 -7 -2 -11 -9
Crude oil exports 0 0 1 -1 -1 0
Average electricity export prices 2 20 30 23 49 45
Coal prices
- End use
- Import price
11
2
150
3
260
3
170
2
385
2
385
1
Carbon Price (US $1996/per metric ton) N.A. 67 99 N.A. 163 141
CO2 Price (CDN$1996/metric ton) N.A. 25 36 N.A. 60 52
domestic and international permit
trading. CCEAF recommended two cases from the EIA study, defined in terms of the
scope and stringency of the Kyoto mechanisms, to model the International Scenarios
(Chart 3.3).
Kyoto Loose9 posits a situation in which there is a well established permit trading
system with low transaction costs, buy-in by developing countries and plentiful JI
and CDM opportunities. Under these circumstances the U.S. can discharge
75 percent of its obligation internationally. The resulting permit prices is C$(1995)
24 per tonne of CO2. Kyoto Loose approximates the Clinton AdministrationÕs
preferred position.
Kyoto Tight assumes the permit trading system is not as well developed, with
higher transaction costs. Participation by developing countries is limited, CDM and
JI opportunities are constrained. This results in the U.S. discharging only 33
8
The Impacts of the Kyoto Protocol On U.S. Energy Markets and Economic Activity ,
October 1998.
9
The terms Kyoto Loose and Kyoto Tight, developed by CCEAF, are intended to reflect
the degree of flexibility in the Kyoto Mechanisms. These Scenarios were developed from
the EIA cases: 1990 +24 percent and 1990 +9 percent respectively. The EIA reference
case indicated that U.S. GHG emissions would be 33 percent higher than the 1990 level.
The U.S. target is 7 percent below the 1990 level.
percent of its obligations internationally, meeting the remainder through domestic
actions. The permit price is C$(1995) 58 per tonne of CO2. CCEAF argues that
Kyoto Tight represents the minimum conditions for U.S. ratification.
The primary purpose of using the EIA study is to obtain a credible range of international
carbon prices to incorporate into the analysis. An additional benefit of this study is that it
provides a comprehensive analysis of the U.S. actions on energy prices, energy imports and
economic activity, all of which have important implications for CanadaÕs achievement of
the Kyoto target. For consistency in its analysis, the AMG decided to use the major results
from the EIA. Some of the more important energy inputs are provided in Chart 3.3. Of
particular note is that world oil prices in 2010, decline by 4 percent in Kyoto Loose
compared to the reference case, and by 11 percent in Kyoto Tight. The EIA does not
expect Canadian crude oil exports to be affected.10 By contrast, natural gas prices rise by
18 percent, and imports from Canada increase by 7 percent under Kyoto Tight, as natural
gas displaces coal in electricity generation in the U.S. In Kyoto Loose, there is little impact
on the natural gas prices and imports. The AMG did not use the EIA electricity imports,
from Canada, as they did not appear to be consistent with the electricity pricing
assumptions. The AMG decided that electricity trade would be about the same as the
reference case and that no new plants would be built solely for electricity exports from
Canada. The EIA study also provided some information on the macroeconomic impacts,
which indicated that there would be a reduction in Gross Domestic Product of 1 to 2
percent.
3.4 Understanding the Results
The following chapters of this report present the estimates of the economic, environmental
and health impacts. Before presenting a detailed description of these results, it is critical to
understand what is implied by each of the results.
In general, the results reported reflect the change in activity from a reference point referred
to as the business-as-usual case (BAU). The results are, therefore, to be viewed as
incremental to the activity that would have occured in the BAU. As well, the results are
typically reported for 2010, as this is the mid-point in the Kyoto commitment period (i.e.,
2008-2012).
There was some divergence of opinion among the AMG regarding the impact of the
Kyoto Scenarios on the level of energy exports, particularly crude oil. There is a view
that the imposition of abatement costs on the Upstream sector may render some supply
sources uneconomic. Alternatively, the cost of abatement is likely to be less than some
of the recent low price shocks, which had no discernable impact on energy exports. This
area requires more detailed analysis.
The section attempts to provide the reader with guidance on a set of fundamental issues
related to the AMG analysis. The elements are portrayed in Chart 3.4, which provides an
overview of the results generated by the microeconomic, macroeconomic and
environmental and health impact analyses.
The first element is the direct costs of
abatement. These were estimated by the
micro-models (energy technology) Ð
MARKAL and CIMS Ð and are
summarised in Chapter 4. These direct
costs consist of the increase in
expenditures on capital and labour (i.e.,
the real resource costs) to undertake the
new investment to be incurred in
response to the measure. In some cases,
there may be decreases in expenditures.
For example, an investment in a more
energy-efficient technology will result in
lower expenditures on energy consumption. These increases and decreases are added
together to obtain the total abatement cost. Since all expenditures, or avoided expenditures,
do not occur in the same year, it is necessary to discount those that occur in the future, since
a dollar spent or saved ten years from now does not have the same value as a dollar spent
or saved this year.
Chart 3.4
Elements to Assess Economic, Environment and
Social Costs and Benefits
GHG Emission
Reduction Benefits
Not Estimated
CAC Emission
Reduction Benefits
Estimated
GHG Mitigation Measures
Net Cost
CAC Mitigation Measures
Net Cost
Not EstimatedEconomic (fixed Outputs)
Estimated
-------------------------------
Welfare
Not Estimated except for
(Changing Outputs)
transportation losses
Economic Impact
Indicators
GDP, etc.
Estimated
Welfare Costs (or Social Costs) = private costs + externalities (e.g., ÒwelfareÓ loss or gain due to change in demand)
Ideally, the analysis of the direct costs of abatement should include both financial
expenditures and non-monetary impacts (i.e., welfare costs). The latter includes the value
of time spent (e.g., to make a trip), other impacts on the environment and changes in the
level or quality of service being provided.
While both microeconomic models provide estimates of cost of abatement, there are
variations in what these costs actually represent. In MARKAL, all costs are actual costs of
resources incurred by the consumers or producers. In CIMS, the costs are those perceived
by consumers and producers, taking into account consumer product preferences, attitudes
to risk, financial costs, relative to other product attributes, and time preferences.
The exclusion of welfare cost estimates means that macroeconomic results may not
represent the net social cost to Canada of realising the Kyoto target. MARKAL estimated
some welfare costs associated with the increased motive fuels tax. CIMS did not explicitly
estimate any welfare costs. There may be other welfare losses or benefits occurring in the
microeconomic modelling, but these impacts are not captured by this analysis.
Both models also provide an estimate of the marginal cost of abatement (i.e., the change in
total costs when one more unit of CO2 is reduced). Marginal costs Ð reported as the cost
per tonne of GHG saved Ð are useful to consider in addition to total costs because they
reflect the ÒstrainÓ on the economy by reducing another unit of GHG from any given level.
The reader should be aware that the cost per tonne multiplied by the GHG saved does not
yield the total cost.
The costs reported in the micro analysis are partial in nature in two different senses. First,
they are the initial costs incurred by industries and consumers to reduce GHG. They can
be considered direct costs in that they do not include the Òsecond round effectsÓ on
industries covered by the macroeconomic analysis. In this initial impact, the non-energy
output (e.g., industrial manufacturing output) remains constant. However, Òsecond round
effectsÓ effects are included in the costs for the energy supply industries. Second, the costs
estimated by the microeconomic models do not include all the intangible costs (a
component of total social cost) such as the value of time, changes in the quality of services
provided, or the environmental and health benefits. The environmental and health benefits
are examined in Chapter 7.
The second element is the macroeconomic analysis. The micro-modelling results provided
a basis for undertaking the macroeconomic analysis, summarised in Chapter 5. The
resource costs and fuel-related costs are inputs to the macroeconomic analysis.
Results from the macroeconomic analysis focus on impacts across the economy and, to the
extent that they capture second-round effects, such as changes in industrial output, are more
complete than the microeconomic analysis. The macroeconomic results provide information
on several indicators, the most important of which are gross domestic product (GDP),
employment, total factor productivity, inflation, imports and exports and government
finance.
GDP, as a measure of the economyÕs total income or production, cannot be compared with
welfare. However, since GDP provides information on changes in final demand for goods
and services, it is sometimes interpreted as a measure of general well being. While the
macroeconomic impacts account for changes in output through sectors of the economy and
the overall economy, they do not reflect changes to consumer utility. In other words,
changes in the level of GDP do not account for the relationship between distribution of
income and overall social welfare. Moreover, the GDP results do not account for the direct
implications for the economy of climate change, nor for the local climate implications that
follow from reducing the emissions of other pollutants that are associated with GHG
emissions. Therefore, comparisons of changes in GDP to welfare changes (i.e., benefits or
losses) are not appropriate.
The economic impacts of implementing packages of measures to reduce GHG emissions
can be expressed as changes in GDP. It is important to bear in mind that these impacts are
presented as Òa reduction of x percent in GDPÓ and are relative to the GDP underlying the
BAU. For example, a reduction in GDP of 3 percent in 2010 means that, over the decade,
the economy will grow by about 26 percent instead of 30 percent as projected in the BAU
(97 percent of 130 percent). This is equivalent to the loss of about one yearÕs growth. It
does not mean that the annual growth rate for GDP declines by 3 percent.
The environmental and health impacts (EHI) associated with the GHG mitigation actions
constitute the third element. The fuel mix changes from the micro models are inputs for
calculating the environmental and health benefits. Like the micro- and macroeconomic
analyses, the EHI impacts are changes from the BAU. The EHI results are presented in
both physical (i.e., changes in emissions levels) and quantitative (i.e., financial) terms.
The benefit assessment comprises two parts Ð the primary benefit and the secondary or cobenefit.
The primary benefit relates to the contribution that the GHG mitigation actions
make to an overall improvement in the climate and the resulting impacts on the
environment, economy and society. The report does not provide a valuation of these
benefits. Such benefits are very difficult to estimate as they are uncertain, future, global in
nature and contingent on others acting.
The secondary or co-benefits include those associated with the impact of the proposed
GHG reduction measures on the more Òconventional pollutantsÓ such as NOX, SOX,
volatile organic compounds and particulates. These emissions lead to deteriorating air
quality and can have negative health and environmental impacts. Therefore, measures to
reduce GHG that also reduce these other emissions usually yield positive benefits, but
occasionally negative co-benefits may occur. In essence these co-benefits can be viewed as
a ÒbonusÓ to the primary reason for undertaking these measures to reduce GHG emissions.
The EHI analysis provides a valuation of the co-benefits. The co-benefits are more readily
measured as they are local in nature and less contingent on actions by others.
The financial impacts resulting from the EHI analysis reflect a change in societal welfare.
As such, they can be compared to the costs of GHG emissions abatement, but not to the
impact on GDP. However, caution should be taken in comparing the EHI results in
Chapter 7 with the microeconomic results in Chapter 4. The EHI analysis does not take
into account the potential for realising the same benefits through non-GHG reducing
measures. If more analysis were undertaken to fill-in such information gaps, a comparison
could be made between the microeconomic costs and the monetized EHI co-benefits.
Uncertainty
Climate change analysis is a complex issue with many uncertainties. The principal areas of
uncertainty include:
¥ The reference case assumptions, especially its economic growth, energy prices and
the projection of GHG emissions.
¥ The assumptions used in this analysis, particularly regarding agriculture and soil
sinks.
¥ The timing of the implementation of Kyoto-related policies.
¥ The nature of the international mechanisms.
¥ The degree of interprovincial electricity trade.
¥ The sensitivity of Canadian imports and exports to changes in foreign prices.
Chapter 4
Energy-Technology Model Results
This chapter provides the direct results at
the national, sectoral and regional levels
Chart 4.1
Paths and Scenarios
for the two energy-technology models.11
Examined in Micro Economic Analysis
These results include the GHG emissions
reduction, the associated cost and energy Canada Acts Alone Kyoto Tight Kyoto Loose
consumption for various path-scenario Path 0 X
combinations. Not all of the Path 1 X
combinations were explored because Path 2 X X X
only limited additional insights would be Path 3 X
provided. The combinations analyzed Path 4 X X X
are shown in Chart 4.1. All five Paths
were analyzed in the Canada Acts Alone
Scenario. Paths 2 and 4 were fully
examined under the two International Scenarios. Both sets of results from the energy-
technology models (CIMS and MARKAL) are shown.
Detailed results from both sets of analysis with the energy-technology model are available
in individual reports. The results with CIMS model may be found in the report, Integration
of GHG Emission Reduction Options Using CIMS, prepared by Energy Research Group /
M.K. Jaccard and Associates. The results with the MARKAL model are contained in the
report, Integrated Analysis of Options for GHG Reduction with MARKAL, prepared by
HALOA, Inc.
Key Learnings
¥ There is a clear relation between the level of the flexibility available in an emission
reduction policy and the total cost of emission reduction efforts. It can generally be
shown that moving from individual sector emission targets to a cross-sector
emission target will allow the desired objective to be achieved at a lower national
cost.
11
The results for the Agriculture sector, which were modelled separately by the Agriculture
and Agri-Food Issue Table, are also included.
¥ Greenhouse gas emissions are reduced through improvements in efficiency that
reduce fossil fuel use and through fuel switching away from more carbon-intensive
fuels. The relative impact on consumption of different energy sources depends on
the size of the emission target and the availability of less carbon-intensive fuels and
efficient technologies. For example, coal and oil product uses tend to switch to
natural gas, thereby reducing the consumption of coal and oil products in all Paths,
relative to BAU. Somewhat surprisingly, domestic demand for natural gas also
declines. The reason for this is that increasing natural gas demand, from switching
to natural gas in some areas, is offset by improvements in efficiency of the use of
natural gas. Changes in specific assumptions about availability of technology or
trade opportunities could change this result. If capture and storage of CO2 from
coal-fired generation and to a lesser extent, enhanced hydro-electricity trade were
unavailable then greater fuel switching to gas-fired electricity generation could
increase the demand for natural gas.
¥ In order to achieve cost-effective emission reductions in Canada, different sectors face
opportunities or challenges to reduce emissions. The industrial sector, particularly oil
and gas producers, and the transportation sector seem to face the greatest challenges.
By contrast, there appears to be lower-cost opportunities in electricity generation,
related to the availability of CO2 capture and storage, and interprovincial trade.
¥ Both energy-technology models suggest similar emission reduction patterns across the
provinces. In all Paths in which the national objective is attained, emissions in Alberta
and British Columbia remain above 1990 levels. All of the other provinces, to varying
degrees, reduce their emissions below 1990 levels and in most cases by more than 94
percent of those levels. However, the direct cost pattern is somewhat different. For
most provinces, the direct costs decline when an optimized policy approach (i.e Paths 2
and 4) is employed. For Alberta and Saskatchewan, however, the costs increase. This
is largely related to the additional costs associated with CO2 capture and storage.
¥ For Canada, access to the Kyoto mechanisms significantly lowers the direct costs of
reducing emissions, and changes the emission burden across sectors and provinces.
¥ Further work is needed on international scenarios, particularly in understanding
Canadian vulnerability to changes in world and North American commodity prices as a
result of an international climate change agreement.
Sensitivity analysis on the results suggest that the costs of emission abatement in Canada,
derived from the analysis and insights of the Issue Tables, rise gradually to within about 50 Mt
of the Kyoto target. Emissions abatement costs then begin to rise at a much faster rate. This
result suggests that increases in CanadaÕs emission gap will disproportionately increase the total
costs of reducing emissions. It also suggests that access to international flexibility mechanisms,
the outcome of international negotiations on agricultural and soils sinks, the availability of low
cost emission reduction technologies and assumptions in the reference case will all have impacts
on the national, sectoral and regional costs of emission abatement.
4.1 Context of the Canada Acts Alone Results
In order to interpret the results it is important to understand the context in which they were
derived. There will be differences between the results given by the two models, mainly
attributable to the differing philosophies of the models as discussed in Chapter 3. The AMG
chose these two models specifically to examine the impact of the different model philosophies.
Other differences will occur as a result of the unique model structure and set of assumptions for
each model. The most important differences may arise from the treatment of the following
issues:
The Post-2012 Period. The two energy-technology models treat this time period
differently as a result of the different structure of the two models. For the optimizing
model, MARKAL, the Kyoto commitment of 565 Mt was held constant, whereas for
the behavioural model CIMS, the marginal cost of abatement, for 2010, was held
constant. This may cause the GHG emissions in CIMS to be higher than the Kyoto
target after 2010.
Average vs. marginal cost pricing. As mentioned in Chapter 3, the price of
electricity is determined differently in the CIMS and MARKAL models. In the CIMS
model, the price is determined by the average cost of producing electricity. In
MARKAL, energy producing sectors set the price to reflect the incremental cost of
producing the last unit of output. The result of marginal cost pricing is that energy-
producing sectors receive extra profits because additional revenue is greater than the
total cost of emission reduction policies.
Inter-provincial electricity trade. MARKAL assumes that the electricity market is
freely traded between provinces, whereas inter-provincial trade is restricted in CIMS.
This implies that MARKAL allows for new inter-provincial transmission while CIMS
does not. As a result, CIMS shows a higher level of emissions in the electricity sector
since some hydro-electricity would not be available to displace fossil fuel power
generation.
Refinery Output. In the CIMS model, it is assumed that the Canadian refinery
industry remains competitive in the North American market; thus, if Canadian demand
for refined product declines, any surplus can be exported. By contrast, in the
MARKAL model some portion of this industry will not be competitive, meaning that a
decline in demand may cause a reduction in refinery capacity. It may be expected that
CIMS would produce a higher level of emissions under this assumption.
For the purposes of the AMG analysis, the following assumptions are common to the CIMS
and MARKAL work:
Fixed non-energy sector output. This assumption maintains the output of all non-
energy sectors of the economy at reference case levels. The AMG deliberately
imposed this constraint on the energy-technology models because it felt that a
macroeconomic model would better capture the effect of energy changes on all sectors
of the economy. The only exception is in the transportation sector where road travel
demand is allowed to change. This exception was made because many of the TablesÕ
measures were aimed specifically at reducing travel demand.
Fossil Fuel Production. In the Canada Acts Alone cases, fossil fuel production was
held to reference case levels. This assumption was made to isolate and understand the
direct effects of GHG abatement strategies. This assumption places greater demands
on energy producers to reduce emissions.
Energy Exports. The production of fossil fuels is maintained at the reference case
levels; consequently, exports vary depending on domestic demand. For example, if
domestic crude oil demand declines, then exports will rise by the amount of the decline;
in other words, the surplus is sold to the U.S. to respect the constant production
assumption.
Energy Export Prices. These have been maintained at reference case levels. The
reason for this assumption is twofold. First, Canada would not be expected to have any
effect on world oil prices, and second, without directly modelling the North American
natural gas market, supply and demand in the U.S. and the resulting effect on prices are
uncertain.
The variations to these assumptions for the microeconomic analysis of the International
Scenarios are provided in Section 4.4; for the macroeconomic analysis, in Chapter 5. The
sensitivity analysis presented in this chapter will be compared to the MARKAL results based
for Path 2 Canada Acts Alone.
4.2 Results: Canada Acts Alone
National Results
GHG Emissions
In the reference case (BAU),12 GHG
emissions are projected to reach 748 Mt in
2010, provided that the assumption of 16
Mt for Forestry and Agricultural sinks is
included in the final text of the Kyoto
Protocol (see Chapter 3). The required
reduction is 183 Mt to reach the Kyoto
target of 565 Mt (Chart 4.2). Each of the
Paths was intended to reduce emissions to
achieve this target. However, Paths 0 and
1 did not attain this level. Paths 2, 3 and 4
achieved the targeted reduction by design.
Chart 4.2
National GHG Emissions - 2010
Canada Acts Alone
Mt CO2
800
700
600
500
400
CEOU Path0 Path1 Path2 Path3 Path4
Kyoto Target: 565
CEOU
CIMS
Markal
Path 0, which is an aggregation of all Issue Table measures, does not achieve the Kyoto Target.
While some of the Issue Tables could not provide sufficient measures to meet the emission
target, those that were submitted achieve about two-thirds of the necessary reductions. One
reason that Path 0 did not achieve the emission target was that the Electricity sectorÕs preferred
method of achieving emission reductions, emission pricing, was not included. The emission
reductions in Path 0 fall short of the target by 44 and 54 Mt for CIMS and MARKAL
respectively.
In Path 1, each sector is expected to meet a Ò6 percent below 1990 emissionsÓ target. The
overall reduction target was not achieved in Path 1 because the Upstream and Refining sector
had insufficient measures to meet its emission target, even at a cost of $300 per tonne of CO2,
and high growth in emissions related to oil and gas production.
Costs
Both models provide estimates of the long-term cost of abatement on a unit and total basis.
The unit costs reflect the marginal cost13 of abatement, which can be interpreted as the price of
an emissions permit.
12
The terms reference case and BAU are used interchangeably.
13
Marginal cost is the cost of the last unit of GHG abated. CIMS uses this term
synonymously with shadow price.
The 2010 marginal costs of abatement for
Chart 4.3
Marginal Costs Of GHG Reduction:
each Canada Acts Alone Path are
Canada Acts Alone
($ per tonne)
provided in Chart 4.3. No costs are
Electricity Industry & Residential & Transportati Fuel Tax
Upstream Commercial on censt/ litre
MARKAL
shown for Path 0 since the measures were
Path 0
imposed. For Path 1, the only relevant
Path 1 23 5 70 16
Path 2 57 57 57 57
marginal cost is for the electricity sector.
Path 3 32 32 84 19
Path 4 49 49 49 49
Of greater interest are the results for the
CIMS
Path 0
optimized Paths (2 and 4). MARKAL
Path 1 30 10 50 12
Path 2 120 120 120 120
indicates that the economy-wide Path 3 110 110 20 50 12
Path 4 120 120 120 120
abatement cost is about $57 per tonne of
CO2, whereas the CIMS estimate is much
higher at about $120 per tonne of CO2.14
The slightly lower results for Path 3 and 4 occur because the coverage of the emission trading
system excludes several high abatement cost sub-sectors. Also of note, in Paths 1 and 3, is the
motive fuels tax increase required to achieve the target in Transportation. The additional tax is
16 to 19 cents per litre in MARKAL and about 12 cents per litre in CIMS. To put this in
perspective the current federal excise tax on gasoline is 10 cents per litre and the combined
federal and provincial level of taxes is, on average, 32 cents per litre.
The total investment and net costs for the
Chart 4.4
Investment and Net Cost by Path
five Paths are shown in Chart 4.4. In
Canada Acts Alone
(NPV10 $Billions)
order to interpret these results some
Path 0 Path 1 Path 2 Path 3 Path 4
explanation is required:
Investment
CIMS 151 139 106 121 106
MARKAL 91 77 22 71 30
¥ They are incremental to the Total Net Cost
CIMS 4161 4546 45
MARKAL 46 52 14 36 20
reference case. This is
GHG Reduction (Mt CO2)
important when comparing
CIMS 141 157 187 192 190
MARKAL 128 164 181 182 182
apparently large cost differences
between models and across
Excludes welfare costs.
Paths.
¥ The investments include the real
14
In Paths 2 and 4, both models attempt to estimate the price of carbon which just causes
the target to be reached, i.e., the price required for the last tonne to achieve the target to
be abated. In estimating this "marginal cost" price, MARKAL accounts only for direct
financial costs. CIMS, by contrast, also takes into consideration less tangible costs which
are based on observed consumer behaviour. Although such intangible costs are relevant
to determine consumers reaction to policy initiatives, they do, by their nature, contain a
large element of subjectivity. The CIMS modelling team has suggested that were these
behavioural parameters to be eliminated from their analysis, the strict "financial cost"
estimate of the marginal abatement cost might be about $80 to $90 per tonne of CO2.
resource costs associated with capital formation and operation and maintenance;
they do not include any welfare costs.15
¥ The net cost is determined by comparing the total investments required in the BAU
case to the investments for GHG reduction measures in each Path, less the resulting
energy cost savings generated by those measures.
¥ The investments and savings are accumulated over a 20 year period and expressed
in present value terms, discounted at 10 percent (PV10).
¥ All investments are net of the undepreciated values after 2022 and are expressed in
1995 Canadian dollars.
The incremental investments range from $106 to $151 billion for CIMS and $22 to
$91 billion for MARKAL. The PV10 of the net costs associated with these reductions vary
from $41 to $61 billion for CIMS and $14 to $52 billion for MARKAL.
In terms of investment requirements, Paths 0 and 1, in which the measures are imposed, are
more costly. Path 3, which allows some emissions trading, is slightly less costly. Paths 2
and 4 show the lowest investment cost because the models select the least-cost suite of
measures that will meet the emissions target. The incremental investment is significantly
smaller in MARKAL.
The required investments can give rise to savings in energy costs. The net costs capture the
offsetting impact of investments and savings. On a net basis, both models conclude that
Path 1 is the most costly because each sector has an imposed target and is forced to use
higher cost measures than are available elsewhere in the economy. By contrast, an
optimized approach such as Path 2 and 4, in which an aggregate target is used and the most
cost-effective measures are selected, appears to offer a significantly lower cost solution.
Paths 2 and 4 attain this result by reducing the required emission reduction in high-cost
sectors, such as upstream oil and gas, and to a lesser extent, transportation, and increasing it
in the low-cost sectors, particularly electricity. An interesting contrast between the models
is the similar costs displayed by CIMS for Paths 2, 3 and 4, as compared to the noticeable
increase in MARKAL for Path 3. The MARKAL result is expected because imposed
measures, which are not selected in Paths 2 and 4, are introduced in Path 3, thereby
increasing investments. In CIMS, the measures are selected in these three Paths, which
suggests that the behavioural attributes of CIMS cause the cost of the imposed measures,
particularly transportation, to be less than the marginal cost of abatement.
15
MARKAL did provide an estimate of welfare cost related to transportation, CIMS did
not. However, there is some uncertainty regarding the inclusion of welfare costs by
Transportation Table.
End-use Energy Consumption
In all Paths, total end-use energy
consumption is about 30 percent lower
than the reference case (as shown in
Chart 4.5). MARKAL predicts higher
energy savings than CIMS, a result of
higher penetration of energy-efficient
technologies. The change in the fuel
mix is more interesting. Electricity
demand declines slightly, relative to the
BAU. Natural gas shows a modest
decline in CIMS, but a significant
decline in MARKAL, up to 40 percent,
in Paths where CO2 capture and storage
Chart 4.5
End-Use Energy Consumption by Fuel - 2010
Canada Acts Alone
PJ
12000
10000
8000
6000
4000
2000
0
CEOU
CIMS
Markal
CIMS
Markal
CIMS
Markal
CIMS
Markal
CIMS
Markal
Path0 Path1 Path2 Path3 Path4
Electricity Gas RPP Coal Biomass Other
is used, since coal remains viable as a generation fuel source in Alberta and Saskatchewan.
Refined petroleum product (RPP) demand decreases substantially, about 30 percent. Both
models suggest significant reductions in gasoline and diesel consumption due to reduced
demand and efficiency improvements especially in Paths 0, 1 and 3.
Electricity accounts for about 19 percent share of end-use energy in the BAU, rising to
about 25 percent in all Paths. In MARKAL, natural gas has a similar share of end-use
energy as the reference case (32 percent). However, the CIMS results indicate that natural
gas rises to about 40 percent. The share of RPP declines from a 40 percent share in BAU to
about 35 percent in the MARKAL results and 30 percent in the CIMS results.
Sectoral Results - Canada Acts Alone
This section provides the emissions reduction by sector. The analysis covers seven sectors,
namely: electricity, upstream and refining, industry, residential, commercial (including
Municipal waste), transportation and agriculture/other. The agriculture/other sector is
treated in the same way for all Paths.
Only the results for Paths 2, 3 and 4,
which attain the Kyoto target, are
displayed.
The electricity, upstream and
transportation sectors, which represent
about 60 percent of the BAU emissions,
account for over 75 percent of the
emissions reductions for all Paths and in
both models. Both models rank
electricity as the largest contributor to
Chart 4.6
Sectoral Shares of GHG Reductions - 2010
Canada Acts Alone
(percent of the gap)
Path 2 Path 3 Path 4
CIMS MARKAL CIMS MARKAL CIMS MARKAL
Electricity 43 57 43 40 43 55
Upstream & 10 12 10 14 10 14
Refining
Industry 757 575
Residential 5 5 0 3 5 5
Commercial6 6 4 4 6 6
Transportation 24 9 31 27 24 10
Agriculture/ 6 5 6 6 6 6
Other
emission reductions in Paths 2, 3 and 4. In MARKAL, the upstream and refining sector
typically ranks second in Paths 2 and 4, whereas CIMS ranks transportation as the second
largest contributor. In Path 3, both models rank the contribution in the same order:
electricity, transportation and upstream. (Chart 4.6).
Electricity provides large contributions to emissions reductions because there are lower cost
options, relative to other sectors, such as capture and storage of CO2. Transportation
provides significant reductions due to fuel savings and the upstream and refining sector is
assumed to have high cost generic measures.
Electricity Sector
Measures in the electricity sector work in opposite directions. The efficiency measures, in
end-use, tend to reduce demand, while fuel switching to electricity tends to increase
demand.
In 2010, the BAU projection for GHG
emissions is 119 Mt. The emissions
reductions projected in Path 3 are
similar for both models (68 Mt). For
Paths 2 and 4, the CIMS and
MARKAL results differ: 67 Mt for both
Paths for CIMS and about 95 Mt for
MARKAL (Chart 4.7).
In Paths 2 and 4, CO2 capture and
storage, in the deep aquifers of western
Canada, is an important abatement
strategy in Alberta and Saskatchewan
which allows the continued use of coal-
fired generation. In these Paths, the
capture of CO2 reaches about 27 Mt in
CIMS and 43 Mt in MARKAL (Chart
4.8). In both models, capture and
storage of the emissions represents
about 45 percent of the total reduction
in this sector. In MARKAL Path 3,
only about 2 Mt is reduced in this
manner, a result of the lower marginal
cost of abatement for the Large Final
Emitters through the permit system,
Chart 4.7
Electricity Sector GHG Emissions - 2010
Canada Acts Alone
120
100
80
60
40
20
0
MtCO 2
1990: 95
25
- 46
- 77
- 47 - 46
- 74
BAU Path2 Path3 Path4
BAU
CIMS
Markal
% change from 1990
Chart 4.8
Capture and Storage of CO2 - 2010
Canada Acts Alone
(megatonnes)
Path 0 Path 1 Path 2 Path 3 Path 4
MARKAL 0 2 43 2 41
CIMS 0 0 27 24 27
which induces more gas-fired generation in western Canada at the expense of coal.
The fuel mix to generate
electricity changes, depending on
the reduction strategy chosen.
Overall, the amount of fuel to
generate electricity is lower in all
Paths, relative to the BAU (Chart
4.9). In both CIMS and
MARKAL, nuclear energy
consumption remains unchanged,
while hydro increases by up to 5
percent from the reference case.
In all Paths, CIMS tends to
predict less coal-fired generation
than MARKAL and
consequently the amount of
Chart 4.9
Electricity Sector
Energy Consumption by Fuel- 2010
Canada Acts Alone
PJ
4500
4000
3500
3000
2500
2000
1500
1000
500
0
Path2 Path3 Path4
CEOU
CIMS
Markal
CIMS
Markal
CIMS
Markal
Hydro Nuclear Coal Gas RPP Wind Biomass
capture and storage is lower. Conversely, CIMS expects more natural gas-fired generation.
In both models, the consumption of refined petroleum products (RPP) declines in all Paths.
Renewable fuels, particularly wind and biomass, do not achieve a very large penetration,
typically less than 5 percent, in any Path, including Path 0 where low-emitting electricity
generation is subsidized.
In all Paths the electricity sector is expected to achieve substantial reductions in GHG at
relatively low cost. Typically, this sector provides 20 to 60 percent of the total reductions
required to meet the target.
Upstream and Refining Sector
CIMS indicates reductions in the upstream and refining sector to be about 18 Mt for the
Paths. MARKAL suggests greater reductions (24 Mt). In 2010, the reference case
projection is 146 Mt.
Industrial Sector
Detailed results of industry sub-sectors, such as Pulp and Paper and Iron and Steel, can be
found in the reports from the microeconomic modellers.
From the reference case level of 115 Mt, all Paths and both models indicate a reduction of
about 13 Mt for the industrial sector as a whole.
Total energy consumption is projected to decline from about 4000 PJ in the reference case,
to about 3800 PJ for CIMS and somewhat lower, 3500 PJ, for MARKAL in all Paths. The
use of RPP shows the largest decline across all Paths for both models. Natural gas remains
about the same as the BAU in the CIMS analysis, but is slightly lower in MARKAL. Both
sets of energy-technology model results show the use of wood and hog fuel in the industrial
sector declining. There is little variation in fuel mix or consumption among the results for
the different Paths.
Residential and Commercial Sectors
The combined residential and commercial sectors, including municipal waste, generate 106
Mt of emissions in the BAU. The analysis indicates significant reductions for these sectors,
in both models for all Paths. Paths 2 and 4 show the highest level of reduction at 27 Mt
and 22 Mt for CIMS and MARKAL, respectively. Path 3 shows the lowest level of
reduction, at about 14 Mt for both models. The majority of these reductions are obtained
through the measures proposed in the Options Packages.
The reference case predicts total energy consumption at about 2,600 PJ, which is
dominated by electricity and natural gas. Total consumption drops to about 2,200 PJ in
CIMS and about 2,450 PJ in MARKAL. While the fuel shares do not vary substantially,
there is a slight trend away from RPP and electricity use in both models. Natural gas
increases its share marginally, but overall consumption is generally lower than in the BAU.
Transportation Sector
In this sector, both models indicate that emissions decline from the reference case levels of
197 Mt to about 140 Mt in Path 3. This Path incorporates the 44 Òmost promisingÓ and
ÒpromisingÓ measures, as well as the increase in motive fuels tax of 12 cents to 19 cents per
litre over current levels. Paths 2 and 4 show smaller reductions, 43 Mt and 24 Mt for
CIMS and MARKAL, respectively.
For CIMS, the energy consumption
declines sharply form 2,800 PJ in the
reference case to about 1,700 PJ in Path
3, and to about 1,900 PJ in Paths 2 and
4. The pattern is similar in MARKAL,
but the reductions are lower, with fuel
consumption being between 2,000 PJ
and 2,400 PJ (Chart 4.10). The full
amount of this reduction occurs in RPP
use. In the CIMS analysis, natural gas
makes a slight penetration, less than 1
percent, into transportation energy
consumption.
PJ
3000
2500
2000
1500
1000
500
0
Chart 4.10
Transportation Sector
Energy Consumption by Fuel- 2010
Canada Acts Alone
CEOU
CIMS
Markal
CIMS
Markal
CIMS
Markal
Path2 Path3 Path4
Electricity Gas RPP
As many of the measures proposed by the Transportation Issue Table are relatively high
cost, few of them are included in the lower-cost Paths (2 and 4). Therefore, the
transportation sector has difficulty achieving significant emission reductions in either of
these Paths.
Agriculture/Other Sector
This sector comprises non-energy- related emissions from agriculture, land use, propellants,
hydro-fluorocarbons (HFC) and anaesthetics. The agriculture sector accounts for almost
90 percent of emissions from these sources, which were 66 Mt in the reference case (after
an allowance for potential soil sinks of 6 Mt). More detail of the agriculture sector can be
found in the report, Analysis of Strategies for Reducing Greenhouse Gas Emissions from
Canadian Agriculture: Technical Report to the Agriculture and Agri-Food Table.
As a result of the options proposed by the Agriculture and Agri-Food Table, agricultural
emissions in 2010 are expected to be 8 Mt less than in the BAU. Three areas comprise
most of this reduction: increase in no-till, decrease in summerfallow and grazing
management.
The reduction for the other components (propellants, HFC and anaesthetics) is achieved
through permit trading as no measures were proposed.
Provincial Results - Canada Acts Alone
Chart 4.11
Chart 4.11 provides the provincial
Provincial Emissions -BAU 2010
distribution of GHG emissions,
before adjustments for sinks, in
the reference case. Alberta and
Ontario each account for about 30
percent of CanadaÕs emissions.
Quebec accounts for about 12
percent, and British Columbia and
Saskatchewan for about 10
percent each.
Mt CO2
0
50
100
150
200
250
Alberta and Ontario provide the highest
Chart 4.12
Provincial Shares of GHG Reductions - 2010
share of GHG reduction in all Paths.
Canada Acts Alone
(percent of the gap)
AlbertaÕs share is about 35 percent of
the total reductions. The Ontario share
Path 2 Path 3 Path 4
CIMS MARKAL CIMS MARKAL CIMS MARKAL
Nfld1 2 1 212
varies between 25 and 35 percent
PEI0 0 0 000
NS 544354
(Chart 4.12). In Alberta, high levels of
NB3 3 3 333
Quebec10 6 11 9 11 6
Ontario 25 30 26 36 25 30
Manitoba3 3 3 4 33
Sask10 11 10 6 10 11
reduction are attributable to the capture
and storage of CO2 from electricity
Alberta 34 35 34 30 34 35
BC 858885
production. In Ontario, the reductions
are concentrated in the transportation
and electricity sectors. There are minor
variations between the models
regarding which province provides the
most emissions reduction. In Path 3, CIMS shows Alberta with a higher level than
Ontario, whereas MARKAL shows the reverse. In Paths 2 and 4, both models show
Alberta with the highest share of emission reduction. Quebec, British Columbia, Nova
Scotia and New Brunswick have about a 4 percent and a 3 percent share, respectively.
Manitoba has the lowest share. These results reflect the levels of GHG in each province;
those provinces with higher emission levels tend to provide the larger share of emission
reductions.
In Paths 2 and 4, both models indicate
that Alberta and Saskatchewan have the
Chart 4.13
Emissions per Capita
(tonnes per person)
Canada Acts Alone
highest emissions per capita, 52 and 40
tonnes in CIMS and 49 tonnes and 31
Path 2 Path 3 Path 4
tonnes in MARKAL. These estimates are Nfld
BAU
22
CIMS
14
MARKAL
15
CIMS
14
MARKAL
14
CIMS
14
MARKAL
14
much higher than those for other PEI 15 12 12 13 13 12 12
NS 21 13 18 14 20 13 18
provinces, whose emissions per capita NB 27 16 15 17 17 16 15
tend to be in the range of 10 to 20 tonnes, Quebec
Ontario
12
16
10
13
10
12
9
12
9
11
9
12
10
12
because of Alberta and SaskatchewanÕs Manitoba
Sask
20
57
15
40
16
31
15
40
15
39
15
40
16
31
relatively small population and relatively Alberta 70 52 49 51 52 52 50
BC 16 13 13 13 12 13 13
high emissions in the upstream oil and gas
and electricity sectors. Both Alberta and
Saskatchewan show increased per capita
emissions in Path 3 for MARKAL, whereas most other provinces experience a decline. This is
due to much less capture and storage of CO2 occurring in Path 3 compared to other Paths.
Although the CIMS results show less variation across the Paths than those from MARKAL, the
increase in Path 3 is not apparent in CIMS as the level of capture and storage of CO2 is
retained in this model (Chart 4.13).
The costs identified for each province should not be viewed as a cost, or in some cases
revenue, to the provincial jurisdiction. They simply reflect the private and government
investments and savings that occurred in each province.
The PV10 of the net costs varies widely among
Chart 4.14
Provincial Costs
provinces (Chart 4.14). In CIMS, the cost
Canada Acts Alone
(NPV10 $1995 billions)
changes across the Paths are small as most
Path 2 Path 3 Path 4
measures are retained as a result of the
CIMS MARKAL CIMS MARKAL CIMS MARKAL
Nfld 0.8 -3.7 0.8 -3.3 0.8 -3.4
PEI 0.2 0.1 0.2 0.1 0.2 0.2
marginal cost of abatement in the model being
NS 1.1 0.8 1.2 0.8 1.1 1.0
NB 0.9 -0.3 1.0 0.0 0.9 0.0
Quebec 8.7 -0.2 11.0 10.4 8.7 1.2
Ontario 18.0 8.9 16.1 18.7 18.0 11.8
Manitoba 1.9 -0.7 1.4 -0.1 1.9 -0.7
higher than the cost of most measures. This
also explains the Path 3 results, which are not
Sask 3.0 1.4 2.7 1.3 3.0 1.7
Alberta 6.3 6.3 7.6 2.4 6.4 5.2
BC 3.5 1.0 4.4 5.8 3.5 2.8
consistently higher in all provinces. Ontario,
Manitoba and Saskatchewan show smaller
Excludes welfare costs.
costs in Path 3, while British Columbia,
Alberta and Quebec show higher costs.
In MARKAL, Ontario and Alberta show the highest costs across all Paths, except for Path 3.
This Path indicates a decline for Alberta because there is less capture and storage of CO2. The
higher cost in Path 3 is particularly noticeable in Quebec because of the imposed measures,
mostly in transportation. Newfoundland and, to lesser extent, New Brunswick, show Ònegative
costÓ in the MARKAL cases, reflecting the export of electricity. Nova Scotia and Prince
Edward Island have relatively consistent costs across all Paths and both models.
4.3 Context of International Scenarios Results
In the International Scenarios, the following changes in assumptions have been made to the
Canada Acts Alone Scenario:
Fixed output. This assumption is relaxed for oil and gas production and electricity
generation. If domestic demand for fossil fuels change then production is allowed to
change.
Energy Exports. The percentage increase in energy prices and oil and natural gas
export levels were provided by the EIA study (Section 3.3). Total oil and natural gas
production then becomes the sum of exports and domestic demand.
The Post-2012 Period. Both models acquire a sufficient amount of permits such that
the Kyoto target is maintained.
The cases will be distinguished using CA for Canada Acts Alone, KT for Kyoto Tight and KL
for Kyoto Loose. In the CA Scenario, all the reductions are achieved domestically. With the
introduction of international emission trading, not all of the reductions need to be attained
domestically because the lower amount of domestic reductions is offset by the purchase of
international permits. However, in all cases the Kyoto target is achieved.
4.4 Results: Path 2 Kyoto Tight and Kyoto Loose
National Results
GHG Emissions
In Path 2KT, the permit cost ($58 per tonne of CO2) causes domestic emissions to rise as
fewer reductions are achieved through measures. The increase in emissions is 30 Mt and 7
Mt for CIMS and MARKAL
respectively, relative to CA (Chart
4.15). In CIMS, about 85 percent of the
reduction is achieved domestically,
compared to over 95 percent in
MARKAL. KT in MARKAL is
similar to CA because the international
price of permits is almost the same as
the marginal cost of abatement in CA.
For CIMS, more permits are purchased
because the permit price is considerably
lower than the marginal cost in CA.
Nonetheless, measures are still used for
GHG abatement, although their
penetration is lower than in CA.
Costs
In Path 2KL, domestic emissions rise by 50 Mt for CIMS and 80 Mt for MARKAL
compared to the Path 2CA results. The CIMS domestic share of reductions in emissions
becomes about 70 percent and the MARKAL figure is 55 percent. This is a result of the
lower permit cost assumption in Path 2KL of $24 per tonne of CO2, which is significantly
less than the marginal costs projected by both models.
Chart 4.15
National GHG Emissions - 2010
Path 2 Scenarios
400
450
500
550
600
650
700
Path2-CA Path2-KT Path2-KL
CIMS Markal
Mt CO 2
Acquired by international permitsKyoto Target: 565
Both models indicate that the net costs
Chart 4.16
for the two international Scenarios are
Investments and Net Costs
Path 2 Scenarios
lower than CA. The MARKAL net
(NPV10 $billion)
costs indicate that there is a larger Path 2 CA Path 2 KT Path 2 KL
surplus in KT than in KL (Chart 4.16). Investments
CIMS 105 63 46
Larger cost variations are seen in CIMS, MARKAL 22 24 17
a result of the much higher marginal cost Permit Purchases
CIMS 7 5
of abatement compared to MARKAL MARKAL
Net Costs
6 9
($120 vs $57 per tonne of CO2) in the CIMS 44 12 -7
CA Scenario. The total cost of the MARKAL 14 -27 -18
permits in both models is about the same Excludes welfare costs.
for both international Scenarios. This
apparently anomalous result is due to the
assumption that permits are being purchased up to 2020 and the figures indicated are PV10
of these purchases, compared to the single year (2010) of emission reductions.
The MARKAL net costs indicate that there would be surplus in both KL and KT, even
after the purchase of permits. This surplus is created by not implementing the high cost
measures, by higher priced electricity exports and by fuel savings in the transportation
sector.
Energy Consumption
In 2010, Total end-use energy
consumption is projected to be slightly
higher than the CA results in both KT
and KL (Chart 4.17). This is due to the
exclusion of higher cost measures,
particularly in transportation.
Sectoral Results
In Path 2KT, the CIMS results track the
CA Scenario only in the
residential/commercial sector because
there are low cost measures available in
this sector (Chart 4.18). Both the
electricity and transportation sectors
have a smaller share of emission
reductions compared to CA, because
the marginal cost of reductions in these
sectors is higher than the international
Chart 4.17
End-Use Sectors Energy Consumption - 2010
Path 2 Scenarios
PJ
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
Chart 4.18
Sectoral Shares of GHG Reductions - 2010
Path 2 Scenarios
(percent of the gap)
Path 2 CA Path 2 KT Path 2KL
CIMS MARKAL CIMS MARKAL CIMS MARKAL
Electricity 43 57 32 55 27 23
Upstream & 10 12 15 13 4 11
Refining
Industry 7 5 4 4 3 3
Residential 5 5 4 4 3 3
Commercial 6 6 5 6 5 5
Transportation 24 9 18 9 15 7
Agriculture/ 6 5 6 5 6 5
Other
International 0 0 16 4 27 43
Permits
permit price. Permits provide about 16 percent of the reductions.
The MARKAL analysis for KT is very similar to the CA results. This is expected since
the marginal cost of abatement is similar in both cases.
For CIMS, in Path 2KL the same trends as the KT case are evident, but with smaller shares
of reduction in the electricity and transportation sectors. In CIMS, the GHG reduction
resulting from permit trading is 27 percent of the total reduction. In this case, the
MARKAL results are much different from CA, but resemble the trend exhibited in the
CIMS results. The MARKAL results show even lower shares of reduction in the
electricity, transportation and upstream than either CA or Path 2KT in CIMS. The
MARKAL assessment of 2KL indicates that permit trading accounts for nearly half of the
abatement. This result is expected owing to the low cost of international permits relative to
the domestic only cost of abatement.
Regional Results
In KT, CIMS projects that Alberta
Chart 4.19
increases its share of emissions reduction
Regional Shares of Reductions - 2010
Path 2 Scenarios
relative to the other provinces from 34 to
(percent of the gap)
41 percent relative to CA (Chart 4.19
Path 2 CA Path 2 KT Path 2KL
does not include the contribution of CIMS MARKAL CIMS MARKAL CIMS MARKAL
Atlantic 9 9 5 8 4 4
Kyoto mechanisms). Ontario and the Quebec 10 6 8 6 6 4
Atlantic provinces show the largest Ontario
Manitoba
25
3
30
3
18
3
30
3
15
2
28
3
decrease in their provincial share of Sask
Alberta
10
34
11
35
10
34
12
37
9
30
4
17
emissions reduction. However, the BC
International
8
0
5
0
6
16
4
0
6
28
4
36
overall ranking of the provinces does not Permits
change. In MARKAL, the provincial
shares are very similar to CA.
The KL results for CIMS are virtually identical to KT. A significant difference occurs,
however, in the MARKAL results for KL. Alberta reduces its share relative to CA, from
35 to 27 percent, and Ontario increases its share from 30 to 43 percent. This changes the
relative ranking of Ontario and Alberta, but the other provinces remain about the same.
Ontario now shows the highest level of reductions because the electricity and upstream
sectors in Alberta are achieving their reductions through the acquisition of permits, whereas
some measures are still being used in Ontario.
4.5 Results: Path 4 Kyoto Tight and Kyoto Loose
National Results
GHG Emissions
In Path 4KT, the increase in emissions
is 40 Mt and zero for CIMS and
MARKAL respectively, relative to CA
(Chart 4.20). This is due to the permit
cost ($58 per tonne of CO2). KT in
MARKAL is similar to CA because the
international price of permits is almost
the same as the marginal cost of
abatement in CA. For CIMS, more
permits are purchased because the
marginal cost is lower than CA.
Nonetheless, measures are still used for GHG abatement, although the penetration of the
measures is lower than in CA.
Costs Chart 4.21
Investments and Net Costs
Path 4 Scenarios
In Path 4KL, the emissions rise much (NPV10 $billion)
more, about 60 Mt for both CIMS and Path 4 CA Path 4 KT Path 4 KL
Investments
MARKAL. Compared to the CA CIMS 106 63 46
Scenario, this is a result of the lower MARKAL
Permit Purchases
29 30 23
permit cost assumption of $24 per tonne CIMS 7 5
of CO2, which is significantly lower MARKAL
Net Costs
3 8
than the marginal cost of abatement CIMS
MARKAL
45
20
13
-8
-4
2
estimated by each model.
Excludes welfare costs.
Both models indicate that the
investments and net costs for the two
international Scenarios are smaller than CA (Chart 4.21). Larger cost variations are seen in
CIMS because of its higher marginal cost of abatement compared to MARKAL ($120 vs
$57 per tonne of CO2) when Canada acts alone. In MARKAL the total cost of permits is
lower than CIMS in KT, but higher in KL. This may be attributed to the selling of a small
amount of permits in the MARKAL KT case.
The MARKAL net costs indicate that there is a surplus in KT, but not in KL, CIMS
indicates almost identical results with the international Scenarios of Path 2, with a small
cost in KT and a small surplus in KL
Chart 4.20
National GHG Emissions - 2010
Path 4 Scenarios
400
450
500
550
600
650
700
Path4-CA Path4-KT Path4-KL
CIMS Markal
Mt CO 2
Kyoto Target: 565 Acquired by international permits
Energy Consumption
In 2010, total end-use energy
consumption is projected to be slightly
higher than the CA results in both KT
and KL (Chart 4.22). This is due to the
exclusion of higher cost measures,
particularly in the transportation sector.
Sectoral Results
The CIMS and MARKAL results in
Path 4 are almost identical to those in
Path 2 (Chart 4.23).
The significant difference from Path
2KL occurs in the MARKAL results.
Purchases of permits in Path 4KL
decline to 36 percent of the GHG
reductions compared to 43 percent in
Path 2KL. This results in the electricity,
upstream and transportation sectors
increasing their contributions.
Regional Results
In Path 4, the regional results are the
same as those for the respective models
in Path 2 (Chart 4.24).
Chart 4.22
End-Use Sectors Energy Consumption - 2010
Path 4 Scenarios
PJ
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
Path4-CA Path4-KT Path4-KL
CIMS
Markal
CIMS
Markal
CIMS
Markal
Electricity Gas RPP Coal Wood/Hog Other
Chart 4.23
Sectoral Shares of GHG Reductions - 2010
Path 4 Scenarios
(percent of the gap)
Path 4 CA Path 4 KT Path 4KL
CIMS MARKAL CIMS MARKAL CIMS MARKAL
Electricity 43 55 32 56 26 24
Upstream & 10 14 15 15 14 15
Refining
Industry 7 5 4 4 3 3
Residential 5 5 4 4 3 4
Commercial 6 6 5 6 5 5
Transportation 24 10 18 9 15 8
Agriculture/ 6 6 6 6 6 6
Other
International 0 0 16 0 28 36
Permits
Chart 4.24
Regional Shares of Reductions - 2010
Path 4 Scenarios
(percent of the gap)
Path 4 CA Path 4 KT Path 4 KL
CIMS MARKAL CIMS MARKAL CIMS MARKAL
Atlantic 9 9 5 8 4 4
Quebec 11 6 8 5 6 3
Ontario 25 30 18 30 15 27
Manitoba 3 3 3 3 2 3
Sask 10 11 10 13 9 6
Alberta 34 35 34 37 30 17
BC 8 5 6 5 6 4
International 0 0 16 0 28 36
Permits
4.6 Sensitivities
Both to understand more fully the results and to address the concerns raised by the Issue Tables
and other stakeholders regarding the analysis, the AMG undertook Òsensitivity analysesÓ to
examine the potential scope and impact of using different assumptions in the roll-up analysis.
The results of these sensitivities highlight the implications of assumptions which have a major
impact on the results. They also identify some of the sources of difference between the
MARKAL and CIMS results.
The sensitivities were undertaken by HALOA, Inc. with the MARKAL model and the results
are compared to the Path 2CA. These four sets of results detailed below give an indication of
the sensitivity of the overall analysis to changes in specific assumptions.
Sensitivity One: Carbon Sinks
The assumption that 16 Mt could be sequestered in agriculture soil and forestry sinks
reflects an optimistic outcome for Canada in the international negotiations to finalize the
Kyoto Protocol. This sensitivity was undertaken in two parts using:
¥ The amount that can be sequestered is zero; therefore 16 Mt of emissions reductions
would have to be found elsewhere, the +16 Mt sensitivity; and
¥ Agricultural soil and forests become a net source of emission of 21 Mt, the worst
case estimate of the outcome of the negotiations as suggested by the Sinks and
Agriculture and Agri-Food Tables. In this case 37 Mt of emissions reductions
would have to found elsewhere, the +37 Mt sensitivity.
The results indicate that the marginal cost of abatement rises from $57 per tonne of CO2 in
Path 2CA to $71 and $99 per tonne for +16 Mt and +37 Mt respectively. Total net costs
rise by almost $4 billion for +16 Mt and $10 billion for +37 Mt, 28 and 70 percent
respectively.
In both components of the sinks sensitivity analysis, three sectors - electricity, transportation
and residential - provide most of the additional GHG reductions. In the +16 Mt case these
sectors provide over 90 percent and in +37 Mt they provide almost 85 percent of the
additional reductions.
In electricity, although more power is consumed, the reduction is achieved through more
use of high efficiency coal plants, with between 5 and 13 additional megatonnes of CO2
being captured and injected into aquifers. As well, additional hydro and wind power is
brought on-stream, which increases inter-provincial trade. Electricity sector emissions are
reduced to14 Mt and 9 Mt respectively, with most of the impact in British Columbia,
Alberta and Saskatchewan, although all provinces are affected to some degree.
Transportation achieves some of its reduction by switching to alcohol based fuels, but the
largest portion is due to fewer vehicle kilometres travelled (VKT) and is fairly uniform
across the country. Total reduction from this sector is 5 and 11 Mt for the two parts of this
sensitivity analysis.
In the residential sector in 2010, more efficient heating systems are adopted and there is a
significant switch to wood, 30 PJ in the +16 Mt and 140 PJ in the +37 Mt. Wood use,
using advanced combustion technology that eliminates the associated CH4, represents about
15 percent of the residential fuel in +16 Mt and it approaches 25 percent in +37 Mt.
Revising the sink assumptions produces very high costs and leads to some unusual results
such as more coal-fired generation.
Sensitivity Two: Natural Gas Prices
A 50 percent increase in natural gas prices from the original reference case level was
analyzed as a sensitivity case, with the MARKAL model, for Path 2CA. CO2 capture and
storage was maintained at the level in Path 2CA (43 Mt). The 50 percent increase in
natural gas prices results in a 17 percent reduction in natural gas use in Canada relative to
Path 2CA case. The decline in natural gas use leads to greater use of fuel oil in industry
and the residential sector and causes these sectors to undertake fewer emission reductions.
To meet the Kyoto target, greater emission reductions occur in the electricity and
transportation sectors. Emissions are reduced in the electricity sector despite a slight
increase in carbon intensive electricity from coal and a substantial reduction in electricity
produced by natural gas. Additional emission reductions in the electricity sector are
achieved by nearly equivalent increases in the amount of electricity produced from hydro
power and wind. Wind power generation increases by over 60 percent from the original
Path 2CA. Higher natural prices and other adjustments in electricity production cause
increases in electricity prices across the country.
Sensitivity Three: Capture and Storage of CO2
In Path 2, there was no technological limit of the quantity of CO2 that could be injected into
deep aquifers in western Canada. Since this is a major source of emission reductions the
robustness of the assumption, using MARKAL, was tested in three ways:
¥ The total volume injected was restricted to half the projected amount in Path 2 (cut
from 43 Mt to 21 Mt) until 2012, thereafter there was no limit imposed. In this
case, 21 Mt of GHG reductions must be found from other sources until 2012 - the
21 Mt sensitivity.
¥ The limit is maintained after 2012, such that total injection is limited to 21 Mt. In
this case, 21 Mt of GHG reductions must be found from other sources for the entire
projection - the 21 Mt LTD Sensitivity.
¥ There is no potential to capture and contain CO2. In this case, 43 Mt of GHG
reductions must be found from other sources for the entire projection - the Zero
sensitivity.
The costs for all three sensitivities are presented, but only the 21 Mt case is discussed for
GHG reductions.
In these sensitivities, the marginal cost of abatement rises form $57 per tonne of CO2 in
Path 2 to about $63 per tonne of CO2 in 21 Mt and 21 Mt LTD. The Zero sensitivity
indicates a cost of $68 per tonne of CO2 in 2010. It should be noted that the marginal cost
rises significantly post 2010 in all sensitivities, particularly the Zero, which reaches over
$170 per tonne of CO2 by 2020. The PV10 of the net costs rise marginally in 21 Mt, $160
million (1 percent compared to Path 2). In 21 Mt LTD the net cost is higher, almost $600
million (4 percent) and the Zero is $1.5 billion higher than Path 2 (10 percent).
In the 21 Mt sensitivity, the electricity sector increases its emissions by almost 5 Mt, offset
by 4 Mt in the Transportation sector through fuel switching to alcohol based fuels. Other
sectors provide the balance, generally through fuel switching to hydro, natural gas and
wind.
Since the electricity sector has increased by 5 Mt relative to Path 2, 16 Mt of reductions (21
less 5 Mt) has been achieved within this sector through switching to less GHG intensive
fuels. Compared to Path 2, the principal differences are:
¥ An increase in hydro in British Columbia and Manitoba of about 4 percent.
¥ An increase in wind (45 percent) and natural gas-fired generation (65 percent) in
Alberta and Saskatchewan.
¥ Less coal-fired generation (50 percent) in Alberta and Saskatchewan.
¥ Increased provincial trade between British Columbia and Alberta and between
Manitoba and Saskatchewan.
¥ Natural gas exports decline slightly and crude oil exports increase slightly.
Provinces to the east of Manitoba are unaffected by this sensitivity.
Sensitivity Four: Inter-provincial Electricity Trade
In the MARKAL model electricity is freely traded among provinces, whereas CIMS
adopted the view that deregulation would not proceed quickly and the inter-provincial
flows would remain at the existing levels plus any announced inter-connections. In order
to test the impact of these differing views, electricity trade in the MARKAL model was
constrained to match that of CIMS.
Under this constraint, the marginal cost of abatement rises very slightly, less than $1 per
tonne of CO2. However, the PV10 of the net costs increases by over $1 billion (7 percent).
Electricity emissions increase by just over 3 Mt, as hydro production declines, which is
offset by small decreases in all other sectors, principally through fuel switching. An
additional 2 Mt is injected into deep aquifers. In 2010, interprovincial trade is reduced by
18 TWh relative to the unconstrained case. This represents about 40 percent of the
ÒimportsÓ to Ontario, Alberta and Saskatchewan.
¥ The main changes, relative to Path 2, to electricity generation are:
¥ Total production is larger by 11 TWh (2 percent).
¥ Less hydro is produced, in Quebec (6 TWh) and Manitoba (4 TWh).
¥ More hydro is produced in British Columbia (1.5 TWh).
¥ Gas-fired generation increases in Ontario by 21 TWh, 14 percent of total provincial
consumption. This offsets all of the decline in imports.
¥ Exports between Quebec and Ontario decline by 13 TWh.
¥ Exports between Manitoba and Ontario/Saskatchewan decline by about 5 TWh.
48
Chapter 5
Macroeconomic Results
This chapter provides the impact of the GHG measures on the economy as a whole as
estimated by the Informetrica Model (TIM). The results from the energy-technology models
are the source of sectoral assumptions about the changes in real investment required to put
new technologies into place, which in turn determine reductions and/or changes in fuels
used in the economy. The macroeconomic analysis also estimates permit payments, for
industries covered by permits. For ÒuncoveredÓ industries the investment flows determine the
value of subsidies. The full set of results from Informetrica may be found in the report
Macroeconomic Impacts of GHG Reduction Options: National and Provincial Effects.
Key Learnings
¥ At the national level, attainment of the target is expected to lead to sustained, long-term
negative economic impacts. In the long run, the reduction in gross domestic product
(GDP), relative to the business-as-usual case, ranges from 0 to 3 percent depending on
the path-scenario combination and the microeconomic input (CIMS or MARKAL).
¥ The overall GDP impacts vary over time. Initially, economic activity increases modestly
in response to increased investment in emissions reducing technologies. Thereafter,
higher production costs, deterioration in competitiveness and lower incomes combine to
reduce GDP below business-as-usual levels. The Òadjustment periodÓ is expected to
be lengthy, perhaps as long as 15 years.
¥ Based on preliminary modelling of some characteristics of emissions permit trading, this
instrument can be viewed as cost-effective for the industrial and electricity sectors.
However, the analysis to date underscores the importance of the design of such an
instrument, in particular the permit allocation mechanism. Each allocation method carries
with it different distributive effects on the economy.
¥ Over the long-term, total employment is reduced moderately, but unevenly across
sectors. Transportation tends to show large gains, whereas the energy industries
indicate losses.
¥ The findings suggest the potential for some variability in GDP impacts across industries
between the CIMS and MARKAL inputs.
¥ The provincial GDP impacts are generally within 1.5 percentage points of the national
average impact. The relative ranking of each province typically varies by Scenario. In
the Canada Acts Alone scenario, Newfoundland, Prince Edward Island, Quebec and
British Columbia are less affected, relative to the national average, whereas Ontario and
Saskatchewan are more negatively affected. The impact on Alberta is close to the
national average. The results for Nova Scotia, New Brunswick and Manitoba vary
between the studies.
In the International Scenarios, Newfoundland, Prince Edward Island and British
Columbia remain in the Òless affectedÓ category and are joined by Manitoba.
Saskatchewan remains in the Òmore affectedÓ category, which now includes Alberta
and New Brunswick. By contrast, OntarioÕs GDP impact is smaller than under Canada
Acts Alone, becoming close to the national average. QuebecÕs position is largely
unchanged across the Scenarios, although one study indicates that its GDP would be
consistently higher than in the business-as-usual case. For Nova Scotia, one study
indicates impacts that are greater than the national average, while the other suggests the
opposite.
¥ The analysis supports the contention that competitiveness Ð as measured by changes in
productivity Ð will be adversely affected by the achievement of the Kyoto target. The
impact is somewhat attenuated if CanadaÕs trading partners are also committed to
attaining their targets and Canada has access to flexibility mechanisms. Under this
Scenario, CanadaÕs trading partners will also face reductions in economic activity, with
negative consequences for Canadian export performance. At the national level, the net
impact of these two forces is to reduce the loss by about 0.5 percent of GDP.
5.1 Introduction
An important assumption of the analysis undertaken with the energy-technology models is
that output is held constant. In the macroeconomic analysis, this constraint is relaxed so
that the induced effects are incorporated.
Many measures proposed by the Issue Tables include changes to current spending,16 which
imply changes to productivity. For ÒuncoveredÓ industries, current spending varies with
16
These expenditures include operating and maintenance improvements that lead to energy
efficiency and productivity gains. They may also imply reductions in productivity as
public and private resources are used to prompt action.
the changes to taxes and subsidies for agriculture, forestry, transportation, buildings, and
municipalities. For those instances the Issue Table estimates were used directly.
The economic assumptions underlying the International Scenarios of Path 2 are provided in a
study by the Energy Information Administration (EIA).17 The EIA results provide estimates of
the impacts on GDP, energy exports and imports and other parameters of meeting the U.S.
Kyoto commitment. The changes in the U.S. economy directly affect Canadian exports, and
through imports, Canadian price formation and interest rates (see Section 3.3).
The TIM results show the effects, both regionally and sectorally, on parameters such as Gross
Domestic Product (GDP), Inflation, Personal Disposable Income (PDI) and some measures on
Canadian competitiveness. In general, the results are shown in percentage variations from the
reference case. In the case of GDP, this means that the size of the economy is x percent smaller
or larger, in a given year, than the reference case (see Chapter 3 Understanding Results).
As with the energy-technology models, Chart 5.1
Paths and Scenarios
not all Paths and Scenarios were INFORMETRICA Analysis
modelled (Chart 5.1). Complete
macroeconomic impacts for both Canada Acts Alone Kyoto Tight Kyoto Loose
MARKAL and CIMS were not Path 0
Path 1
undertaken. However, the cases Path 2
Path 3
X
X
X X
examined provide a reasonable range of Path 4 X
results.
5.2 Context of Results
Some additional assumptions are required to undertake the macroeconomic analysis:
Permits - The TPWG provided several options for permit allocation. The AMG chose
a gratis allocation to households as the most practical for modelling purposes.18
Purchasers of permits acquire them from households at a uniform price. While this
assumption may not be the actual allocation process, it provides a view of the economic
impact.
17
Energy Information Administration, Impacts of the Kyoto Protocol on U.S. Energy
Markets and Economic Activity, October 1998.
18
Modelling the allocation mechanism of any permit system is complex. In the Informetrica
analysis, it was not possible to transfer fully the proceeds of the permit sale back to
households. This incomplete transfer results in a small overestimation of the negative
impact on GDP and disposable income after 2007.
Subsidies The government covers the full cost of the investments associated with the
implementation of generic measures, through subsidies, precisely allocated to those
industries defined by TPWG (see chapter 2).
Fiscal and Monetary policy - is assumed not to change. Taxes are assumed to remain
consistent with the BAU, except for the additional motive fuels tax.
5.3 National Results
Direct Investments
The investments required to implement the measures affect other spending, which may have
positive or negative effects. The extent to which current spending, transfers and productivity
effects are incorporated into the Paths depends on the extent of the inclusion of the Issue Tables
measures.
Information campaigns for the enforcement of speed limits for example, require additional
spending for labour and increase associated purchases of goods and services. These resources
are usually purchased by governments. It should be noted that when government spends for
goods and services, supply to meet the spending is often produced by business (e.g., advertising
campaigns).
As a direct effect on resource requirements in the economy, introduction of major new capital
implies a restructuring of demand, leading to reductions in operations for some industries and
increases for others. Prominent in this regard is urban transit. In Paths where expansion of the
transit system occurs, it needs to be recognized that this implies an increase in transit riders, at
the expense of spending for parking, repairs and other non-fuel expenses related to the
operation of a vehicle.
In the CIMS-based results, current government spending is about equal across each variant of
Path 2 (Chart 5.2). Path 3 shows a large negative number, due to government revenue from the
incremental motive fuels tax. Path 4 shows virtually no government expenditures because the
permit system substitutes for government expenditures for education, enforcement and other
stimuli to reduce GHG. Household non-durable spending is consistent across all the CA Paths.
It drops to almost zero in Path 2KT and becomes income in Path 2 KL. This is because of the
income to the households from the permits. Business operating expenses are consistent across
all Paths.
Chart 5.2a
Chart 5.2b
Direct Impact on Income Sectors 2000 to 2020
Direct Impact on Income Sectors 2000 to 2020
CIMS
MARKAL
$(1997) billions
$(1997) billions
Current Spending
Current Spending
Govt spending less taxes 18.7 16.2 16.2 -119.2 1.1
Govt spending less taxes -67.7 16.0 18.5 -210.7 7.5
Household spending 37.6 -22.6 4.3 35.7 37.6
Household spending -33.4 -117.8 -19.6 23.8 -33.4
Business op expenses -17.7 -20.3 -17.7 -19.1 -21.0
Business op expenses 6.7 7.6 7.6 -7.9 6.9
Investment Spending
Investment Spending
Government 21.7 63.3 69.9 34.5 22.0
Government 4.3 4.3 4.2 26.7 4.0
Household 205.2 61.1 64.8 208.7 205.3
Household 71.9 -0.1 1.1 159.3 72.1
Business 166.8 106.7 125.3 163.8 166.6
Business 28.3 16.8 16.8 60.3 26.1
Total 393.7 231.2 260.0 411.0 393.9
Total 104.6 21.0 22.0 246.3 102.2
Government investment spending is highest in Path 2KL and KT. Paths 2 and 4CA are the
lowest and about equal, while Path 3CA is somewhat higher because of the imposed measures
in that Path. Household investment is fairly consistent across the CA cases and much lower in
the international Scenarios. Business investments are consistent across the CA cases, but lower
in the international Scenarios.
The MARKAL-based results, for current government spending, show a similar trend to CIMS,
except in Path 2CA. The only similarity between the models in current household spending is in
Path 3CA and, to a lesser extent, in Path 2KL. CIMS indicates generally negative figures, for
the business operating expenses, whereas the MARKAL based results are generally positive in
sign.
Total investment spending shows similar trends at the macro-level between the models, but as
would be expected the MARKAL-based results are generally much lower.
Investment
pct impact
Chart 5.3a
Direct Cost Impacts on GDP
CIMS
percent change from BAU
5
4
3
2000 2005 2010 2015 2020
Note: 2CA and 4CA are practically identical
0
Path Path Path Path Path
Path Path Path Path Path
2CA 2KL 2KT 3CA 4CA
2CA 2KL 2KT 3CA 4CA
pct impact
Chart 5.3b
Direct Cost Impacts on GDP
MARKAL
percent change from BAU
5
4
3
2
1
2000 2005 2010 2015 2020
-1
2CA
2 KL
2 KT
3 CA 4 CA
2 CA
2 KL
2 KT
3 CA 4 CA
Direct investments have been aggregated for all sectors. Additional spending for new
capital, operations of government and business is proportionately doubled (1 to 3 percent) in
the CIMS-based CA cases compared to that of the international Scenarios, although this
difference narrows after 2005 (Chart 5.3).
All Paths show a substantial positive impact on GDP (plus 2 to 4 percent) up to 2005. This is
due to the investments required for GHG abatement actions. As expected, Path 3CA shows
the highest impact and Path 2KL and KT the lowest. After 2005, there is a decline, but these
investments continue to be positive. The gap between Path 3CA and the international
Scenarios turns to widen.
By contrast, the MARKAL-based results show virtually no change for Paths 2KL and KT
and a less than 1 percent increase for Paths 2 and 4CA. Path 3CA shows a similar trend to
the CIMS-based results, but about 1 percentage point lower.
Gross Domestic Product
It is expected that there would be an increase in Canadian demand and production early in
the implementation of policies to meet the Kyoto target. This is due to the demand for
production from actions being larger than the reduction from demand for fuels, or from the
negative effects that may follow from increased unit costs of production (Chart 5.4). There
would be lasting negative effects as long as the real resource costs of an action exceed the
resource saving of fuel reduction.
Chart 5.4a
GDP Canada
CIMS
percent change from BAU
-4
-3
-2
-1
0
1
2
3
2000 2004 2008 2012 2016
pct impact
2CA 2KL 2KT 3CA 4CA
Chart 5.4b
GDP Canada
MARKAL
percent change from BAU
-4
-3
-2
-1
0
1
2
3
2000 2004 2008 2012 2016
pct impact
2CA 2KL 2KT 3CA 4CA
In the near term, there appears to be little distinction between the Paths in the CIMS-based
results, and the economy is expected to be 1 to 2 percent larger. For this period, the view that
access to international permits leads to less economic disturbance is not clear as investment
levels are similar. However, over the longer term GDP is expected to be 1 to 3 percent less
than the reference case with the KL and KT Scenarios tending towards the more positive end
of the range, indicating that international actions are more beneficial than domestic actions alone.
In the MARKAL-based results the GDP increase in the early years is barely visible, largely
because the investments are substantially lower than CIMS. However, the long-term effects are
similar, but somewhat muted compared to CIMS, except in Path 3CA where GDP loss is
greater.
Inflation
Chart 5.5 shows the change in inflation, compared to the BAU, for each of the Paths
analyzed.
As a result of the motive fuels tax, both models indicate a spike in inflation (2 to 3 percent
higher than BAU) in 2003 for Path 3CA. The spikes in both models and all Paths in 2008,
are due to the introduction of permit trading, although the MARKAL increase is less
because of lower permit prices in Paths 2 and 4CA. These inflationary spikes are likely a
result of the allocation method used in the modelling. The permits have been allocated
gratis to households which, in turn, sell them to GHG emitters. Alternative allocation
methods may yield different results. It is important to identify these kinds of ÒspikeÓ risks
when designing an allocation system.
The main finding is that, regardless of option, or of whether Canada acts alone or in
concert, there is some appreciable risk that price increases will be accelerated. Parallels to
the OPEC ÒshockÓ years are inexact because, in that instance, the supply price of fuels was
raised, whereas in these instances, it is either unchanged or, in the Scenarios, is reduced (at
least for oil). During the period after the OPEC shock (1974) it has been suggested that the
lasting damage may have been induced by the monetary policy, in which interest rates
rocketed upwards. In this analysis, a real interest rate rule has been maintained, so that
nominal interest rates rise modestly during the next decade, and sharply when the spikes
occur.
Early monetary policy action is expected to be modest. If the large action is concentrated in
years after 2010, supply-constrained sector inflation is a high risk, but not included in these
results.
Chart 5.5a
Inflation (GDP Deflator) Impact
CIMS
percent change from BAU
3
2
Chart 5.5b
Inflation (GDP Deflator) Impact
MARKAL
percent change from BAU
3
2
0
0
-1
-1
2000 2004 2008 2012 2016
2000 2004 2008 2012 2016
2CA
2KL
2KT
3CA 4CA
2CA
2KL
2KT
3CA 4CA
pct impact
pct impact
1
Personal Disposable Income
The CIMS-based results suggest that there is little that distinguishes the Paths in their effect on
average income of households through most of this decade (Chart 5.6). The BAU indicated
substantial growth of disposable income between now and the end of the decade. In the longer
term, the international Scenarios appear to have the least negative effect, but are still negative.
There is not much difference between one domestic scenario over another, although it should be
noted that additional motive fuel tax, as in Path 3CA, would have transitory negative effects on
real income.
The CIMS analysis indicates that, for Paths 2CA and 4CA, that total spending, fuel reduction,
and permit payment levels are about the same. They are distinguished only by a smaller fiscal
thrust in Path 4CA, and the relative price effects of a broad-based trading system, as compared
to other trading schemes. In the TIM results, disposable real income per household is about the
same. The difference is in the number of households resulting from the two Scenarios.
Maintaining the household formation at the reference case levels may have missed some insights
into the effects of disposable income.
The effects on income distribution may vary more sharply among the options, but there are no
distinctive signals in the CIMS-based analysis.
Similar trends are evident in the MARKAL-based results, except for Path 3CA which indicates
a continuous and relatively large erosion of Personal Disposable Income, compared to other
Paths.
Chart 5.6a
Disposable Income per Household
CIMS
percent change from BAU
-10
-8
-6
-4
-2
0
2
2000 2004 2008 2012 2016
pct impact
2CA 2KL 2KT 3CA 4CA
Chart 5.6b
Disposable Income per HouseholdMARKAL
percent change from BAU
-10
-8
-6
-4
-2
0
2
2000 2004 2008 2012 2016
pct impact
2CA 2KL 2KT 3CA 4CA
Employment
In Path 2CA, employment increases modestly overall (about 1 percent over the period 2000 to
2012) in the CIMS analysis. Over the longer term employment is reduced in all Paths.
However, there are some major sectoral differences. The increase is dominated by
transportation services (20 percent) followed by construction (6 percent). This is due to the
heavy investments and operating costs that are expected for urban transit systems. Consumer
goods and services also show an increase, 0.7 percent. The resource-based industries show a
decline of 1.5 percent. Path 4CA shows similar trends. These results are not reflected in the
MARKAL analysis, although showing similar overall trends, but with no sector experiencing
substantial gains nor losses.
In Path 2KL, both models indicate that total decline in employment is small (0.4 percent) with
no sector showing strong gains, since the urban transit systems are not built. The resource
based industries show a similar decline to Path 2CA.
In Path 2KT, total employment in the CIMS-based results increases very slightly (0.1 percent)
with strong gains in transportation services (10 percent) and construction (3.5 percent). As
with the other two cases, resource industries show a decline (1.7 percent). MARKAL shows a
modest decline in employment early in the projection and modest increase in the latter part, with
no sector showing large gains nor losses.
In the CIMS-based Path 3CA, employment decreases very slightly (0.1 percent) over the
period 2000 to 2012. There are, however, some major sectoral differences. Electrical and
electronic component (-8 percent) and metallic mineral and products (-6 percent) show the
most significant declines. The most significant increases are in Transportation services (20
percent) and construction (5 percent). MARKAL shows a similar overall decline, but only the
recreation sub-sector shows appreciable decline (6 percent). Transportation shows similar
gains to CIMS. In the longer-term both the CIMS and MARKAL-based analyses show
employment declining by about 2 percent.
Balances
Relative to the BAU, incomes are reduced, but current spending is either higher or little changed
(Chart 5.7). This follows from a reduction in the average saving of households in all of the
Paths. That is, consumers standards of living are ÒprotectedÓ by reduced saving. Government
spending remains at reference case levels, except for directly related expenditures - such as
planning, enforcement, and education programs - that increase modestly. However, these are
not present in Path 4CA. This outcome may be compared to the period 1990 to 1999, in
which the average personal saving rate fell from about 10 percent to almost zero, while incomes
were falling or stagnating.
Chart 5.7b
Chart 5.7a
Balances - Savings Impact
Balances - Savings Impact
MARKAL
CIMS
average pct of change as pct of GDP
average pct of change as pct of GDP
2CA 2KL 2KT 3CA 4CA
2CA 2KL 2KT 3CA 4 CA
00-12 13-18 00-12 13-18 00-12 13-18 00-12 13-18 00-12 13-18
00-12 13-18 00-12 13-18 00-12 13-18 00-12 13-18 00-12 13-18
Personal -0.1 -0.1 0.1 0.1 0.2 -0.1 -2.2 -3.2 0 -0.2
Personal -1.1 -2.3 0.1 -0.5 -0.1 -1.1 -1.5 -2.6 -1.1 -2.1
Business -0.1 -0.4 -0.1 0.1 0.1 0.9 1.2 3.3 -0.1 -0.3
Business 0.8 2.2 0 0.7 0.3 2.0 0.9 2.0 0.7 1.6
GovÕt 0.2 0.3 -0.4 -0.3 -0.4 -0.2 0.5 -0.3 0.1 .03
GovÕt 0 -0.5 -0.4 -0.8 -0.3 -0.7 0 -0.6 0.2 -0.2
Foreign 1.3 1.0 1.1 0.7 1.1 -0.2 1.6 1.3 1.3 1.0
Foreign 0.1 0.3 0.3 0.4 0.2 -0.5 0.8 0.6 0.1 0.3
The magnitude of the Current Account deficit is increased throughout and in all Paths due to
additional recourse to foreign borrowing. Foreign borrowing, to finance Canadian investment,
is a feature of the entire last century. A sustained increase, equivalent to 1 percent of GDP,
would be notable in currency markets and among international economic policy observers.
The effects on government balances are the least significant among all the saving systems. For
the most part, the subsidies provided by governments to stimulate emissions reduction do not
have significant impact on the balances. The incremental motive fuel tax revenues of Path 3CA,
because they are not recycled, protect government balances, and could compensate for the
damage to revenues caused by negative effects on private incomes.
5.4 Sectoral Results
The sectors for the macroeconomic analysis are somewhat different from the energy-technology
model sectors, a reflection of the differing model structures. However, the groupings have been
arranged to resemble the energy-technology models as closely as possible. The sector results
are shown as average changes over two time periods:
2000 to 2012 (the end of the first commitment period) and 2013 to 2018.
2000 to 2012
There are only minor sector distinctions between Paths 2CA and 4CA (Chart 5.8). The CIMS
view is that the capital costs, fuel saving and permits trading are the same for these two cases.
They are distinguished by slight variations in: current government spending and transfers; the
relative price differences owing to the price increases for oil, gas and electricity in Path 4CA;
and the allocation effect of permit payments to a larger number of sectors in Path 4CA than in
Path 2CA.
Chart 5.8a
Change in GDP by Sector 2000 to 2012
CIMS
-4
-2
0
2
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
Average pct Impact
Resource-based goods Durable and Investment goods
Business related services Consumer goods and services
Government and social services
Chart 5.8b
Change in GDP by Sector 2000 to 2012
MARKAL
-4
-2
0
2
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
Average pct Impact
Resource-based goods Durable and Investment goods
Business related services Consumer goods and services
Government and social services
Resource industries show the most severe negative effects in all Paths. Within this sector,
mining, chemicals and the agriculture/food industries are the most affected. Energy producersÕ
output is moderately affected in the CA cases, but significantly reduced in the international
Scenarios. Forest products are little affected.
Other sectors are little affected and some, notably durables in the early years, indicate an
increased GDP. Either directly, or through induced income effects, these sectors benefit from
increased domestic demand caused by actions to meet Kyoto targets.
Path 3CA, in the CIMS analysis, represents the most damaging case. In 2010, there are some
severe effects on iron ore mining (-30 percent), rubber and plastic products (-15 percent),
leather goods and clothing (-15 percent), iron and steel (-20 percent), non-ferrous smelting (
12 percent), automotive assembly (-13 percent), non-electrical machinery (-15 percent), truck
assembly (-30 percent), and electronic products (-17 percent). Positive effects for the
construction of roads and urban transit are related to the implementation of the urban transit
measures.
The coal industry is the one industry that will face significant deterioration. The size of the
industry is reduced by from 40 to 50 percent in Path 3CA. This occurs early in the projection,
given the CIMS view that reductions in this fuel would be sharply curtailed immediately. Apart
from direct effects on thermal coal, related to electric power, metallurgical coal would be
unfavorably affected in the CA cases as Canadian unit costs of production would rise relative to
those of competing international suppliers.
The MARKAL based results show a similar trend across all Paths except Path 2CA, but the
effect is more muted. In Path 2CA, durables, business services and consumer goods and
services are all mildly negative in this period, compared to CIMS being positive. This is due to
the differences in the investment levels between the models.
2013 to 2018
In the longer term, effects are generally more severe and widespread across in the industrial
sector (Chart 5.9). Most of the reductions in GDP have been realized by 2012. However,
further erosion to output is expected after 2013. Over the long-term, the resource industry
Chart 5.9a
Change in GDP by Sector 2013 to 2018
CIMS
-8
-6
-4
-2
0
2
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
Average pct Impact
Resource-based goods Durable and Investment goods
Business related services Consumer goods and services
Government and social services
Chart 5.9b
Change in GDP by Sector 2013 to 2018
MARKAL
-8
-6
-4
-2
0
2
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
Average pct Impact
Resource-based goods Durable and Investment goods
Business related services Consumer goods and servicesGovernment and social services
faces further declines in GDP, in the range of 4 percent. Durables face a similar decline in
GDP. These trends are evident in both CIMS and MARKAL
5.5 Provincial Results
The provincial results are presented using the averages for 2013 to 2018. Early in this decade,
the pattern of effects includes positive impacts, followed by varying degrees of decay to
below BAU estimates by the end of the decade. The provincial impacts are a reflection the
sectoral impacts.
For the period 2013 to 2018, the results indicate lasting, long-term effects in both models for
the most provinces (Charts 5.10 and 5.11). The more moderate negative effects of
International Scenarios are reflected across the regions, compared to impacts based on
domestic action only, Central Canada shows the largest improvement between domestic and
international actions. Ontario is somewhat more vulnerable than Quebec, owing to the positive
effects of QuebecÕs hydro electricity generation. The impact on Ontario is due to its trade-
sensitive manufacturing industries. Saskatchewan and Alberta are most affected in the long term
because of the dominance of resource-based industries in those provinces. Among Atlantic
provinces, Nova Scotia and New Brunswick appear most vulnerable. Newfoundland and
P.E.I. are less affected due to the high level of influence of government spending, which remains
unchanged by assumption. Effects on British Columbia and the Territories are generally smaller
than the Canadian average because of the relatively small impacts on the forestry products
industry, which maintain exports at the BAU levels, on the assumption that exports to North
Asia are relatively unaffected..
Chart 5.10a
Provincial GDP Impact 2013 - 2018
CIMS
QuŽbec and East
4
2
Chart 5.10b
Provincial GDP Impact 2013 - 2018
MARKAL
QuŽbec and East
4
2
-4
-4
-6
-6
P2CA P2KL P2KT P3CA P4CA
Nfld
P2CA P2KL P2KT P3CA P4CA
Nfld
PEI
NS
NB
Que
PEI
NS
NB
Que
Average pct Impact
Chart 5.11a
Provincial GDP Impact 2013 - 2018
CIMS
Ontario and West
P2CA P2KL P2KT P3CA P4CA
Average pct Impact
Ont Man Sask Alta BC&Territories
Chart 5.11b
Provincial GDP Impact 2013 - 2018
MARKAL
Ontario and West
-6
-4
-2
0
2
4
P2CA P2KL P2KT P3CA P4CA
Average pct Impact
Ont Man Sask Alta BC&Territories
Average pct Impact
0
-2
-2
4
2
0
-2
-4
-6
The most notable different between CIMS and MARKAL occur in Newfoundland and
Manitoba which show modest ÒgainsÓ (i.e., the provincial GDP is higher than the reference
case) in some Paths, particularly 2KT and 2KL in MARKAL, while CIMS shows a negative
effect. This is mainly due to electricity exports.
In the Canada Acts Alone Scenario, Ontario accounts for more than one-half of all lasting
negative effects on the Canadian economy. In the International Scenarios, this province
accounts for about one-third of the impacts. Alberta, Quebec, and British Columbia follow next
in importance, in terms of their effect on the Canadian economy.
These regional and provincial distinctions would likely be more pronounced if the analysis were
to account for inter-provincial migration. As a general rule, it would be expected that the labour
force and population would respond to relative impacts on jobs, but this channel of influence is
not operational in these results.
5.6 Competitiveness
Competitiveness is a multi-dimensional
Chart 5.12
issue, involving factors such as
Measures of Competitiveness
productivity, export performance, fiscal
climate and technology. The AMG
Plourde Hirshhorn
approach to competitiveness is based on
Change in real GDP per capita Extent of competition from foreign
firms
two reports.19 In these reports various
Change in total factor productivity Significance of price competition
measurements of competitiveness were
Change in imports and exports Expected impact of GHG policies on
foreign competitors
defined (Chart 5.12).
Change in real unit costs
Change in net investment flows
It is more difficult to quantify the
measurements used by Hirshhorn, these
will be discussed qualitatively. Where
results are available for PlourdeÕs measurements, impacts on competitiveness will be quantified.
The extent of competition from foreign firms depends on the scenario under consideration. In
CA, Canada will be at a definite disadvantage, since it will be undertaking abatement measures
alone. Therefore, costs arising from GHG reductions will be higher in Canada. In KL and KT,
it is assumed that other countries will also reduce GHG emissions. In this instance, the relative
19
AndrŽ Plourde; Competitiveness and the National Implementation Strategy: Elements
of a Framework of Analysis, commissioned by the AMG; and Ronald Hirshorn:
Characteristics of the Canadian Economy and the Implications for Climate Change,
commissioned by Industry Canada.
costs of GHG abatement will be important. As there are no directly comparable data, this
aspect is difficult to determine. However, there is an indication from the EIA study, used to
develop the international Scenarios, that the impact on the U.S. economy would be similar to
that projected for Canada. Should the U.S. implement GHG reduction policies, its economy
would be smaller, hence opportunities for exports may be curtailed somewhat.
In this analysis, the effect of price competition is reflected in the analysis at the domestic level.
At this time it is difficult to infer the international consequences for prices since the GHG policies
of other countries such as the United States, are not well articulated at this time.
The change in real GDP per capita will decline in all Paths, especially after 2005.
The level of Total Factor Productivity (TFP)20 is reduced, in the CIMS-based results, in all
cases over the early years, when substantial incremental capital is being added while fuel cost
reductions are being introduced (Chart 5.13). The incremental capital is directed at GHG
reductions and does not change output. In the longer term, fuel savings Òcatch upÓ with the
increase in capital to stabilize the level of reduction from the direct effects. It is not expected
that in the indirect and induced effects of TIM would alter this.
An important indication, among the domestic policy sets, is that there appears to be no sharp
distinction in competitiveness between the CA cases. The selection of one or another of these
Paths would probably be based on grounds other than the impact on productivity potential.
Acting in concert with other countries (KL and KT) appears to be less damaging.
Chart 5.13a
Total Factor ProductivityCIMS
percent change from BAU
-4
-3
-2
-1
0
1
2000 2004 2008 2012 2016
pct impact
2CA 2KL 2KT 3CA 4CA
Chart 5.13b
Total Factor Productivity
MARKAL
percent change from BAU
-4
-3
-2
-1
0
1
2000 2004 2008 2012 2016
pct impact
2CA 2KL 2KT 3CA 4CA
20
TFP is defined as Total Output divided by an index representing Labour and Capital.
The MARKAL-based results show a more gradual decline over the projection for all Paths
except Path 3CA. This Path is similar to the CIMS outcome.
In the CIMS-based analysis, in all cases exports decline (Chart 5.14). As expected, Path 3CA
shows the steepest decline for all sectors while Path 2KL has the least impact. Business-related
services and Consumer goods and services are generally the most affected sectors. The impact
on Resource-based industries and Durable and investment goods is more modest. These
declines are attributable to the shrinking of the export market, although the international
Scenarios show smaller impacts.
Chart 5.14a
Change in Exports by Sector 2013 to 2018
CIMS
percent change from BAU
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
0
Chart 5.14b
Change in Exports by Sector 2013to 2018
MARKAL
percent change from BAU
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
0
Resource-based goods Durable and Investment goods
Business related services Consumer goods and services
Resource-based goods Durable and Investment goods
Business related services Consumer goods and services
Average pct Impact
-2
-2
Average pct Impact
-4
-6
-8
-10
-4
-6
-8
-10
The MARKAL analysis shows similar trends, although there is almost no impact on exports in
Path 2 KT.
The change in unit costs mirror the change in exports (Chart 5.15). Path 3CA shows the most
severe increases, approaching 8 percent in some sectors and Path 2KL the least impact, with all
sectors in the range of 1.0 to 2.5 percent. Consumer goods and services and Durable and
investment goods have the highest increases across all Paths. Resource-based industries have
the lowest increase.
Chart 5.15b
Change in Unit Cost by Sector 2013 to 2018
MARKAL
percent change from BAU
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
8
Resource-based goods Durable and Investment goods
Business related services Consumer goods and services
Average pct Impact
6
4
2
0
-2
While there are some differences in the affected industries in MARKAL, particularly in Path
3CA, where consumer goods and business-related services are more affected than in CIMS,
the trend is similar. However, in Path 2CA
Average pct Impact
Chart 5.15a
Change in Unit Cost by Sector 2013 to 2018
CIMS
percent change from BAU
Path 2CA Path 2KL Path 2KT Path 3 CA Path 4 CA
8
6
4
2
0
-2
Resource-based goods Durable and Investment goods
Business related services Consumer goods and services
there are no discernable effects in MARKAL.
Canadian competitiveness will be affected the
most in the CA cases as there would be no
cost ÒpenaltiesÓ to CanadaÕs trading partners
due to GHG reduction measures. In the
international Scenarios the picture is less clear
because there are little data to establish the
relative effects of TFP and unit costs among
the trading partners. To expect no effect
would be unrealistic. It would appear that the
areas most vulnerable, based on change in
exports and unit costs, are business-related services, particularly finance and insurance and
consumer goods and services, particularly the leisure industries.
Another important issue that the macroeconomic analysis has not been able to capture is the
potential for leakage of investment to countries that have lower, or no costs, to industry under
the Kyoto Protocol. This leakage involves the diversion of investment that would otherwise
occur in Canada, but is diverted by the costs that Canadian industry faces to meet Canada's
Kyoto commitment. The effect is likely to be specific to a certain set of industries that are most
vulnerable to international price competition. The diversion of investment from Canada leads to
the potential migration of certain sub-sectors of the Canadian economy to other countries.
While difficult to quantify, it is a risk to the Canadian economy. Analysis of this issue is an area
for future work and should be done on an industry specific basis.
Chapter 6
A Complementary View:
Analysis using CaSGEM
This chapter explores a complementary view to the Informetrica analysis using a computable
general equilibrium model. The nature of this model is such that some input assumptions are
different from those used by the energy-technology and econometric models. The full results for
CaSGEM can be found in the report, A Computable General Equilibrium Analysis of
Greenhouse Gas Reduction Paths and Scenarios, prepared by the Economic Studies and
Policy Analysis Division of the Department of Finance.
Key Learnings
¥ There are long-run economic costs to implementing the Kyoto Protocol.
¥ When implemented in a cost-effective way (e.g., Path 4), these costs are neither
inconsequential nor enormous.
¥ Cost-effective implementation implies very uneven effects across sectors and regions.
¥ Using sector-specific emissions targets would add to the overall cost of implementing
Kyoto without reducing the variation in economic impact across sectors.
¥ ÒTechnicalÓ assumptions have a large bearing on the overall impact and its distribution
across sectors and provinces.
6.1 Background
Computable general equilibrium (CGE) modelling combines the micro and macro perspectives
in a single model. That is, the CGE approach simultaneously models input choices (including
energy) and changes in sectoral and aggregate economic activity. Consequently, it is not
possible to use all of the output of energy-technology models like CIMS and MARKAL as
input into a CGE model. Thus the CGE analysis should be seen as a complement to the Issue
Tables-CIMS/MARKAL-TIM results.
The model used to undertake the CGE analysis was the Canadian Sectoral General Equilibrium
Model (CaSGEM), a model that was developed and is maintained by Finance Canada.
CaSGEM simulates the Canadian economy using representative consumers and producers. It
describes the complete economy as an aggregate of 51 sectors producing 59 goods and
services, most of which are traded internationally.
CaSGEM is a static model which means
Chart 6.1
that there is only one time period in the
What CaSGEM Predicts
model. Thus CaSGEMÕs simulations of a
GHG-reduction policy should be
interpreted as showing the long-run
impact of a policy change after the
economy has completed the (possibly
lengthy) adjustment to the new policy. The
model assumes perfect competition. In
Chart 6.1, Region A shows the economy
Value of variable, e.g., GDP
Time path with
policy change
Adjustment Completed
year
Impact
B
C
A
Time path without
policy changePolicy Implemented
before a policy change has been
introduced, and Region C shows the
economy after the adjustment to the new
policy has been completed. Unlike TIM, CaSGEM does not describe the transition of the
economy from the introduction of the policy to the time at which the economy has fully adjusted
to the new policy (Region B in Chart 6.1). Note that the long-run growth rate of GDP (and
associated variables) is assumed to be unchanged by the policy, but the change in the level of
GDP is permanent.
As shown in Chart 6.2, CaSGEM was used
Chart 6.2
Paths and Scenarios
to simulate three path-scenario
For CaSGEM Analysis
combinations:
¥ Path 4 Canada Acts Alone
Canada Acts Alone Kyoto Tight Kyoto Loose
¥ Path 4 Kyoto Loose
Path 0
Path 1
¥ Path 3 Canada Acts Alone
Path 2
Path 3 X
Path 4 X
X
It is assumed that all tax rates will be
unchanged except for the fuel tax changes
specified in Path 3. In a dynamic model like
TIM, this could alter the governmentÕs
budget balance in each year of the
simulation, which translates into a change in the government debt at the end of the simulation
period. As CaSGEM is a static model, the government cannot accumulate debt and reduce it
later because there is no Òlater.Ó To implement AMGÕs request to leave tax rates unchanged,
per capita (i.e., lump sum) transfers to consumers were adjusted to ensure that the government
budget remained balanced.
6.2 Results: Path 4 Canada Acts Alone
The centrepiece of Path 4 is a tradeable
permit system that covers most of the
Chart 6.3
Channels for Emission Reductions
economy.21 The simulation of Path 4
under the Scenario Canada Acts Alone
(CA), leads to a 0.8 percent decline in
CanadaÕs Gross Domestic Product
(GDP), relative to what it would
otherwise be. This may appear to be
small, but the change would be
permanent, so it is not inconsequential.
Moreover, the precise figure depends on
a number of ÒtechnicalÓ assumptions, as
discussed below.
Consumer reductions
GDP Effect
Share Effect
Intensity Effect
These GHG emissions reductions are accomplished through four different channels, as shown in
Chart 6.3.
The most important of these channels, the
intensity effect, suggests that at least some
sectors can fairly readily replace GHG-
intensive inputs with less emitting ones. In
this particular simulation, much of this
reduction in intensity is occurring in
electricity production. Chart 6.4 shows
the reduction in emissions from coal and
natural gas predicted by the model for that
sector and the reduction in fuel use. The
distinction between emissions and fuel use
for coal is caused by the introduction of a
new technology to capture and store
Chart 6.4
Emissions and Use of Fossil Fuel in Electricity
Production Path 4CA
percent change vs BAU
Fuel Reduction in
Emissions
Reduction
in Use
Coal -87 -55
Natrual Gas -18 -18
21
Details on the assumed coverage of the tradeable permits system can be found in the
CaSGEM report. Measures that reduce non-fuel-related emissions in upstream oil and gas
and agriculture were also imposed in all path-scenario combinations.
emissions from coal-fired generating stations in Alberta and Saskatchewan. This technology
implies that coal can continue to be burned without emitting CO2 into the atmosphere.22
While the use of coal and natural gas in electricity production is reduced sharply Ð accounting
for almost a third of CanadaÕs CO2 reduction Ð the decrease in electricity production is a much
more modest 2 percent. The reduction in emissions intensity for that sector is more than
70 percent.
Chart 6.5 shows the impact of Path 4CA
on fossil fuels, which are responsible for
about 81 percent of CanadaÕs current
GHG emissions. The table shows both
how much more expensive they become Ð
because of the permits that are needed for
their combustion Ð and the degree to
which their use is curtailed.
The substitution effect accounts for
considerably less of the reductions than
does the reduction in the emissions
intensity. However, the impact of the
GHG-reduction policy falls very unevenly
across sectors. The sectors with the
largest declines in activity (as measured
with value added, or GDP) are shown in
Chart 6.6.
Some of the very uneven sectoral impact
is due to the assumption that Canada Acts
Alone. This puts Canadian producers of
industrial chemicals, steel and cement at a
disadvantage with respect to their foreign
competitors. The picture changes Ð but
not completely Ð when other countries
also meet their Kyoto target.
Chart 6.5
Impact on Source of Energy Path 4CA
percent change vs BAU
Source Price to Domestic
Purchaser Use
Coal 212 -52
Gasoline 15 -7
Diesel 18 -10
Fuel Oil 26 -18
Natural Gas 35 -32
Electricity 7 -1
Chart 6.6
Largest Sectoral Reductions Path 4CA
Sector Percent change
in GDP vs BAU
Gas Pipelines -22
Steel -12
Cement -12
Industrial Chemicals -11
Petroleum and Coal Products -8
Pulp and Paper -7
Non-ferrous Smelting -7
Secondary Metal -5
Air Transport -5
Coal Mining * -4
* Includes coal mimed by electricity producers for use in Canadian power plants
22
In all of the results reported here, if capture and storage of CO2 is economical, the amount
of emissions that are stored is taken from the corresponding run of the CIMS model.
Since the industrial make-up of each
Chart 6.7
Provincial GDP Path 4CA
provinceÕs economy is different, it might
percent change vs BAU
be expected that the variations in sectoral
Province Including Electricity Excluding Electricity
Sector Sector
impacts to be parallelled by wide
Newfoundland -0.4 -0.7
PEI -0.7 -0.4
variations among provincial economies.
Nova Scotia -2.8 -0.6
New Brunswick -2.3 -0.8
Chart 6.7 shows that differences do occur
Quebec 1.2 -0.8
among provinces, especially when the
Ontario -1.8 -0.7
Manitoba -0.8 -0.7
electric power sector is included in the
Saskatchewan -1.7 -1.1
Alberta -1.1 -1.6
calculations.
BC and Territories 0 -0.8
Canada -0.8 -0.9
Electric power generation (and hence, its
GDP) increases in provinces where
electricity is produced with low-GHG technologies such as hydro-electricity, at the expense of
provinces that burn fossil fuels to generate electricity.
Surprisingly, the impacts on the electric power production and coal consumption appear muted
in Alberta, even though most of the electricity generation is coal-fired. The reason is that
capture and storage of CO2 is assumed to be available and more economical in Alberta and
Saskatchewan than switching to other fuel sources. This technology is assumed to be
economical provided that it costs more than $38 for alternative methods of abating CO2 . Since
the cost of a permit to emit one tonne of CO2 in Path 4CA is found to be above $44 in this
simulation, electricity producers in Alberta and Saskatchewan find it cheaper to capture and
store their emissions rather than buy a permit allowing them to burn coal without capture and
storage.
Note that in this simulation, it is assumed that additions to hydro-electric capacity can be
installed in some provinces, and that inter-provincial trade in electricity can increase
substantially. These assumptions, along with the assumption of capture and storage, play an
important part in the results. Without those assumptions, the total impact of the policy package
on the economy would be larger, and the regional distribution of the results would change.
6.3 Results: Path 4 Kyoto Loose
In this international Scenario, other countries also implement the Kyoto agreement and trade
permits internationally. As discussed in Chapter 3, the permit price in Kyoto Loose is assumed
to be $24 per tonne of CO2.
CaSGEM is a model of the Canadian economy only. To address international scenarios, it
requires detailed input on prices and economic activity in foreign countries. The requisite input
to CaSGEM was taken from a multi-country CGE model, G-cubed. The CaSGEM report
includes an introduction to G-cubed. 23
This international Scenario has two main advantages for Canada:
¥ Since other countries are complying with Kyoto, Canadian exporters will be placed at
less of a competitive disadvantage: many of their foreign competitors will also be
spending money on GHG reduction or on permits.
¥ The portion of the gap that must be met by domestic actions is reduced, and Canada
can now satisfy its commitment at a lower permit price ($44 vs. $24 per tonne of CO2).
There are also two principal disadvantages to Canadians under this international Scenario:
¥ Since CanadaÕs trading partners will also be reducing GHG emissions, their economies
will contract. This will provide smaller markets for Canadian exporters.
¥ Furthermore, Canada is a net exporter of crude oil and natural gas. The international
demand for these two fuels may decline as other countries reduce their GHG emissions.
This in turn may drive down the price that Canadian exporters receive for their oil and
gas.24
The net effect of all these forces, under Path 4KL, is to reduce CanadaÕs GDP by 0.3 percent
from its business-as-usual case. This represents a smaller impact than Path 4CA. Furthermore,
Canadians buy enough permits from abroad to account for just under one-third of the difference
between BAU emissions and the Kyoto target.
The impacts of Path 4KL on the various sectors differ from those in Path 4CA. The overall
impact on cement, electricity, industrial chemicals, pulp and paper, and steel is more muted
under the Kyoto Loose Scenario as compared to when Canada Acts Alone. Furthermore,
capture and storage is no longer a worthwhile proposition in Alberta and Saskatchewan as it is
now cheaper to buy permits than to capture and store CO2.
23
G-cubed was developed at the Brookings Institution. The version that is used at Finance
Canada includes a fully specified block for Canada that was developed by Philip Bagnoli.
A full description of the G-cubed model can be found in W. J. McKibbin and P. J.
Wilcoxen, ÒThe Theoretical and Empirical Structure of the G-Cubed ModelÓ, Journal of
Policy Modelling, 1999.
24
This is at variance with the expectations of the EIA analysis (see Chapter 3)
6.4 Results: Path 3 Canada Acts Alone
Path 3 is also simulated under the assumption that Canada alone fulfils its commitment under the
Kyoto Protocol. The difference between this Path and Path 4 is that the permit-trading regime
is restricted to a smaller number of industries and is assumed to complement a number of
sector-specific measures which are aimed at helping each sector meet its own Kyoto target.
In this Path, only the electric power and industry sectors participate in a permit-trading scheme.
Other sectors adopt measures proposed by the various Issue Tables. These measures were first
simulated using the CIMS model; the results were then input into CaSGEM using CIMSÕ
calculations of each measureÕs incremental capital expenditure and energy reductions.
The treatment of the transportation sector differs somewhat. As specified in Path 3, it faces a
tax on diesel fuel and gasoline to discourage their consumption.25 In addition, a number of fuel-
saving measures are implemented that are paid for by the emitters or, as is the case for much of
the investment in new transportation infrastructure, by government.
In the simulation of Path 3CA, the Canadian economy contracts by 0.9 percent with respect to
its BAU case. The economic contraction is only marginally greater than in Path 4. There are
two different causes, working in opposite directions, for the differences between the two Paths.
The first difference is the way that compliance is distributed across sectors. In Path 4,
compliance with the Kyoto target is distributed across sectors in a cost-effective way because
almost all sectors of the economy face a uniform price on emissions.26 This leads those sectors
with the least costly abatement options to do most of the abatement. In Path 3, the policy
measures are not chosen with overall cost-effectiveness in mind.
The second difference is that the simulation of Path 3 incorporates measures from the Tables.
They proposed a number of low-cost GHG-reducing actions that were not forecast by
25 According to the CIMS model, total gasoline use falls by about 40 percent while that of
diesel fuel in transportation falls by 25 percent relative to BAU.
26 Uniform pricing of emissions should mimimize the aggregate cost of the additional
resources (e.g., capital equipment) that are needed to reduce emissions. This is the
conception of ÒcostÓ that was used in the energy-technology models to describe the
impact of compliance on sectors other than energy producers and transportation. In a
model like CaSGEM, minimizing the aggregate resource cost should also mimimize the
impact on aggregate GDP.
CaSGEM under Path 4.27 Furthermore, the policies that are applied on a sector-by-sector
basis are assumed to be perfectly targetedÐ in other words, the measure is as effective and no
more costly than planned. Among other things, this means that government subsidizes only
additional expenditures that result directly from the new policy.
Again, the net effect of these two opposing influences is that Path 3CA is only marginally more
costly than Path 4CA. As shown in the CaSGEM report, when the inexpensive sources of
abatement that are assumed in Path 3, but not in Path 4, are removed, Path 3 is noticeably more
costly than Path 4. The upshot is that Path 3 is comparable in cost to Path 4 only if a
policymaker can induce people to reduce emissions in ways that they would not otherwise do
under a price instrument such as permits. Moreover, despite the use of sector-specific emission
targets, the variation in the economic impact of the policy package across individual sectors is
about the same in Path 3CA as in Path 4CA.
Besides the greater impact on GDP, the major differences between the results of Paths 3 and 4
are:
¥ Diesel fuel and gasoline use declines significantly more from BAU under Path 3.
¥ Because of these declines, consumers and the transportation sector (excluding air
travel) reduce emissions significantly more (and in the case of transportationÕs sectoral
GDP, the contraction is greater) than under Path 4.
¥ The electricity sectorÕs emissions fall less under Path 3 than under Path 4 (a reduction of
46 percent in Path 3 compared to 49 percent in Path 4).
As in Path 4, the price of permits is above the $38 per tonne of CO2 that makes capture and
storage economical. CO2 emissions reduction from coal burning plants in Alberta and
Saskatchewan is, therefore, a significant contributor to overall GHG reductions.
27
There may be a number of reasons why CaSGEM and the Tables have different views on
the resource cost of reducing emissions. CaSGEM is based on historical responses to
price changes and may be unable to predict the ease with which new technologies are
adopted in a world where GHG-emissions are more costly. Or, the Tables may
underestimate the costs (i.e., the risk inherent in new technologies, the preference of
private vehicle travel over public transportation) borne by consumers and producers who
adopt new technologies or practices.
6.5 Comparison to Results of other Models
The introduction to this chapter noted that
the quantitative results differ across the
models. This section explores this further
by comparing some key results from
CaSGEM to those produced by the other
models. To focus the comparison, all of
the results are taken from simulations of
Path 4 under Canada Acts Alone.
However, certain assumptions,
particularly for oil and natural gas exports,
are different from those used in the TIM,
CIMS and MARKAL models. In
Chart 6.8
Comparison of Energy Prices and Use Path 4CA
percent change vs BAU
CIMS MARKAL CaSGEM
Price to Purchaser
Coal 800 300 212
Gasoline 35 15 15
Diesel 40 15 18
Fuel Oil 27
Natural Gas 70 45 35
Electricity 7
Domestic Use
Coal -55 -45 -52
Gasoline -30 -15 -7
Diesel with gasoline with gasoline -10
Fuel Oil -18
Natural Gas -6 -19 -32
Electricity -4 -17 -1
CaSGEM these exports are expected to decline, whereas in the TIM analysis they are
expected to remain flat for crude oil and increase for natural gas.
One way to summarize the energy-technology aspects of Path 4CA is to consider the simulated
impact on energy prices and use. Chart 6.8 compares the results obtained from MARKAL,
CIMS, and CaSGEM. The changes in price and energy use are broadly comparable across the
models. The predicted changes in natural gas use vary the most across models.
In all three models, the electricity sector
plays a central role in Path 4CA. Chart
6.9 contrasts the predicted impact of Path
4 on the use of fossil fuels across the three
models. The CaSGEM results lie between
those of CIMS and MARKAL. In all
three models, coal use declines sharply.
CIMS shows the largest decline in coal
use, MARKAL shows the smallest.
Natural gas use rises in CIMS, but falls in
MARKAL. CaSGEMÕs prediction lies
between those two; there is a decline in
natural gas use, but it is not as dramatic as
the decline predicted by MARKAL.
Chart 6.9
Comparison of Fossil Fuels Used in Electricity
Production Path 4CA
percent change vs BAU
Fuel CIMS MARKAL CaSGEM
Coal -63 -51 -55
Natural Gas 11 -35 -18
It is possible to summarize the
macroeconomic aspects of Path
4CA by comparing the impact on
industries. Chart 6.10 shows the
industries with the largest declines
in GDP in the simulations from
TIM and CaSGEM.28 The models
have differing industrial structures
and close correlation is unlikely.
As expected, there is a closer
correlation with the
TIM/CIMS/MARKAL
combination, although two
industries (coal mining and natural
gas distribution) are common to all
three models.
Chart 6.11 contrasts the provincial GDP results across the models. It shows that there is very
little coherence in the results across TIM and CaSGEM. Comparing TIM/CIMS to CaSGEM
Chart 6.10
Comparison of Sectors with Largest Decline in
GDP Path 4CA
INFORMETRICA
CIMS MARKAL CaSGEM
Largest Decline Consumer Electronics Coal Mining Gas Pipeline
2nd Largest Coal Mining Gas Distribution Steel
Truck & Bus Body
Assembly
Truck & Bus Body
Assembly Cement
Miscellaneous Transport
Equipment Iron Ore Mining Ind Chem
Beverage Manufacturing Consumer Electronics Pete & Coal
Iron Ore Mining Refineries Pulp & Paper
Leather Goods Manufacture Electronics, Computers &
Office Machinery
Non-ferrous
Smelting
Telecommunications
Equipment Beverage Manufacturing Secondary Metal
Garments & Clothing
Manufacture
Miscellaneous Transport
Equipment Air Transport
10th Largest Gas Distribution Electric Power Utilities Coal Mining
shows that GDP losses would occur in 8
of 10 provinces. The major difference
being Quebec, which TIM/CIMS expects
a loss of GDP, but CaSGEM expects a
gain. The TIM/MARKAL comparison
with CaSGEM shows the same sign on
GDP in four provinces. There are many
sources for these discrepancies, including
the different results for the national
sectoral GDP impacts that underlie the
provincial results, and the particular
treatment of the distribution of electricity
production used for the CaSGEM results.
Moreover, certain assumptions regarding
Chart 6.11
Comparison of Provincial GDP Path 4CA
2013 - 2018
percent change vs BAU
TIMS - CIMS TIM - MARKAL CaSGEM
Province
Newfoundland
PEI
Nova Scotia
New Brunswick
Quebec
Ontario
Manitoba
Saskatchewan
Alberta
-0.5
-1.3
-1.3
-0.4
-1.5
-2.8
-1.5
-2.4
-1.9
0.5 -0.4
0.3 -0.7
0 -2.8
-0.2 -2.3
0 1.2
-0.5 -1.8
-1.0 -0.8
-0.7 -1.7
0 -1.1
0.2 0
-0.8
BC and Territories -1.2
Canada
natural gas and crude oil exports are different from TIM.
Despite these differences in the quantitative results, the qualitative conclusions stated at the
beginning of the chapter generally apply to the results of all of the models.
28
To make this comparison, the results from TIM were aggregated up to the 51 sectors that
are represented in CaSGEM.
Chapter 7
Environmental and Health Impacts
This chapter of the report focuses on the environmental and health impacts (EHI)
associated with the mitigative actions to reduce GHG under the various Paths and
Scenarios. It is part of the overall assessment of the economic, environmental and social
impacts associated with implementation of the Kyoto Protocol. The full results for EHI can
be found in the document The Environmental and Health Co-benefits of Actions to
Mitigate Climate Change.
Key Learnings
¥ The analysis indicates that there are co-benefits associated with the actions to
mitigate GHG. Relative to the benefits from GHG emissions reduction, co-benefits
are immediate, local and reasonably certain. The net present value of these cobenefits,
discounted at 10 percent, is between $2.7 billion and $4.5 billion. These
estimates should not be compared to the changes in GDP. There may be
similarities between the co-benefits and the costs identified elsewhere in this Report.
However, the estimates from the energy-technology models exclude welfare costs
and the co-benefit assessment is incomplete. For this analysis they should not be
compared.
¥ The EHI assessment, may be described as range-finding. The results should be viewed
in conjunction with the framework and the assumptions on which they are based.
¥ The co-benefit estimates indicate that the benefits under the Kyoto Loose Scenario
would be less than those for the Canada Acts Alone scenario.
¥ The assessment indicates that actions to mitigate climate change can make a important
contribution to Canada-Wide Standards (CWS). It achieves most of the NOX
reduction, some of the SOx and makes minor contributions to PM2.5 and VOC.
¥ The estimates tend to be conservative, since they do not cover all of the pollutants and
omit the contribution of sulphate reductions in western Canada.
7.1 Introduction
The early part of this chapter focuses on the cost-effectiveness. The benefits comprise two
parts: the primary benefits; and the secondary or co-benefits. The primary benefits relate to the
contribution that these actions make to an overall improvement in the climate and the resulting
impacts on the environment, economy and society (prepared by the Science Impacts and
Adaptation Group (SIAG)). The secondary or co-benefits include those associated with the
impact of the proposed GHG initiatives on the conventional pollutants, such as oxides of sulphur
and nitrogen, volatile organic compounds (VOC) and particulates.
This assessment is critically dependent on both the framework and assumptions on which it is
based. The assessment framework is both qualitative and quantitative. The qualitative portion is
based on expert opinion. The quantitative assessment is based on a sequential modelling
framework that begins with inputs from the energy-technology assessment (Chapter 4). The
energy-technology models define the input fuel demands from which the changes in air emissions
are estimated.
The three path-scenario combinations that
were assessed for the EHI analysis - Path Chart 7.1
Paths and Scenarios
2CA, Path 2KL and Path 3CA - provide For EHI Analysis
a reasonable range of results (Chart 7.1).
The physical and monetary analysis of co-Canada Acts Alone Kyoto Tight Kyoto Loose
Path 0
benefits is mainly focused on sulphates Path 1
and ozone for which there is fairly good Path 2
Path 3
X
X
X
data and modelling. By focusing on a Path 4
selected subset of pollutants the level of
uncertainty is reduced.
7.2 Approach
Broad Framework
The framework for the EHI analysis relies on the work undertaken by the Tables, the
SIAG and the AMG. The Issue Tables were asked to carry out an analysis of the
environmental and health impacts outlined in the guidelines supplied by the AMG.29 In
addition, the Tables estimated and reported changes in emissions of criteria air
contaminants (NOX, VOCs, SOX and PM). These assessments are included with the
29
Framework for the Analysis of Environmental and Health Impacts, Analysis and
Modelling Group, February 16, 1999.
AMGÕs overall assessment of options. The AMG estimated the clean-air ancillary, or cobenefits,
of GHG mitigative actions and provided a broad qualitative assessment of the
environmental and health impacts.
It was not possible for the SIAG to estimate the primary avoided impacts resulting from the
implementation of the Kyoto Protocol. Nonetheless, it was requested to describe the
potential known impacts of climate change on CanadaÕs physical environment.
Quantitative Assessment
Climate change actions, in addition to reducing greenhouse gas emissions, can also lead to
reductions in conventional pollutants such as CO, SO2, NOX/VOCs, and particulate matter.
These other pollutants lead to deteriorating air quality and can have negative health and
environmental impacts. Therefore, climate change measures to reduce GHGs which also
reduce these other emissions can yield positive co-benefits from the improvement in air
quality. These co-benefits can be viewed as a ÒbonusÓ and should be considered in the
assessment and selection of strategies for reducing GHGs.
The quantitative assessment involves the following four steps:
¥ Estimating changes in conventional air pollutants, known as criteria air
contaminants (CAC), from the baseline;
¥ Estimating changes in ambient concentrations of pollutants (air quality)
corresponding to the CAC emission changes;
¥ Estimating the physical health and environment impacts associated with the changes
in air quality; and
¥ Estimating the economic value of the effects that would be avoided by reducing the
emissions.
Qualitative Assessment
Climate change actions, in addition to reducing greenhouse gas emissions, could have a
range of environmental impacts. The EHI framework categorizes these potential impacts as
being related to atmospheric, aquatic and terrestrial effects. Some of these impacts can be
quantified in either physical or monetary terms, and are the subject of the quantitative
assessment. Other impacts are less amenable to quantification due to incomplete
information on the potential implications. These are examined in a qualitative manner.
7.3. Observations and Results
To fully interpret the results of the analysis of CAC and co-benefits it is important to
understand the baseline economic and policy assumptions. The CEOU was used as the
reference case, with some EHI specific inclusions and exclusions:
¥ Tier I vehicle emission standards are included, Tier II30 are not.
¥ Lower sulphur regulations for gasoline are included, but not for diesel.
¥ The proposed Canada Wide Standards31 are not included.
In general, policy pronouncements that are not yet fully articulated could reduce the size of
the impact on CAC and co-benefits. The assumption on Canada-Wide Standards is the
most significant.
The reference case growth rates for four
of the seven CAC examined are shown
in Chart 7.2. These four contaminants
were selected because of their relative
importance to air quality, and these
pollutants are the most affected by the
Chart 7.2
Base Case CAC Emissions
Percent Change from 1995
30%
25%
20%
15%
intervals from 2000 to 2020. With the
-5%
-10%
exception of NOX, which shows minor
declines until 2010, all other
contaminants increase. By 2020, PM2.5,
SOX, NOX and VOC are respectively
24 percent, 20 percent, 10 percent and 5 percent higher than the 1995 levels. Based on the
assumptions in CEOU, the overall trend is upward. However, the sector and regional
trends vary significantly from the totals. Highlights of these trends are provided below.
2000 2005 2010 2015 2020
PM2.5 SOx NOx VOC
In the reference case, the largest percentage increases in NOX and VOC are in the
electricity sector (27 percent and 43 percent) due to increased natural gas generating
capacity. Increases in SOX, mainly from the upstream oil and gas (33 percent), petroleum
refining (29 percent) and base metal smelting and refining (14 percent) are somewhat offset
by declines in the electric power generation sector. All emissions from transportation are
actions to reduce GHG emissions. The
Percent Change
growth estimates are shown in five year
10%
5%
0%
30
Tier I and Tier II are U.S. EPA emission reduction programs for light duty vehicles.
31
Canada-Wide Standards are national standard for ozone and PM2.5. By 2010 PM2.5 is to
average 30Fg/m3 over a 24 hour period. Ozone is to average 65 ppb over an 8 hour period.
expected to decline due to improvements in fuel quality and emissions intensities: SOX (-15
percent), NOX (-20 percent), VOC (-42 percent) and PM2.5 (-17 percent).
In the reference case, the largest NOX reductions are expected in Quebec (-23 percent),
Manitoba (-11 percent) and the Atlantic Region (-10 percent), with significant increases in
Alberta (+16 percent). The largest SOX increases are in Saskatchewan (37 percent), British
Columbia (31 percent), Ontario (20 percent) and Alberta (19 percent), whereas emissions
decline significantly in the Atlantic Region (-27 percent). The biggest increases in PM2.5
are expected in Alberta (27 percent) and Ontario (20 percent).
Several sensitivity analyses were conducted on the reference results to examine the
implications of changes in key forecast parameters:
¥ Significantly reduced sulphur levels for Canadian motor diesel fuel to match U.S.
EPAÕs proposed levels for on-road vehicles (15 ppm) and planned potential levels,
from 500 to 15 ppm, for off-road diesel vehicles; and
¥ Tier II emission standards applied to CanadaÕs light duty gasoline vehicles.
¥ Reductions in the sulphur content of diesel fuels lower transportation sector SOX
emissions by an additional 30 percent relative to the BAU, and lead to a total
reduction of transportation sector SOX emissions of 41 percent by 2010 relative to
1995 levels (vs. 15 percent in the BAU). By contrast, the impact of the Tier II
vehicle emissions program suggests more modest reductions in NOX and VOC
emissions of 8 percent and 5 percent relative to the base case forecast in 2010.
Relative to 1995 levels, transportation sector NOX and VOC emissions decline by
27 percent and 45 percent (vs. 20 percent and 42 percent in the BAU).
Chart 7.3 provides a sample of the
changes in CAC in 2010 for both
energy technology models across three
of the Paths and Scenarios analyzed.
All Paths and Scenarios examined show
significant declines in NOX (ranging
from 9 percent to 19 percent) and SOX
(ranging from 14 percent to 22 percent).
Generally, the reductions are larger for
MARKAL; also, the Canada Acts
%Change vs Base Case (2010)
Chart 7.3
Changes in Selected CAC Emissions
for Models and Analysis Paths
(2010 vs BAU)
MARKAL MARKAL MARKAL CIMS
P2 CIMS P2 P3 CIMS P3 P2KL P2KL
0%
-5%
-10%
-15%
-20%
-25%
Alone Scenarios have larger reductions
than the International Scenarios. There
are slight declines in other CAC, with
the exception of PM2.5 which shows increases in some Paths, largely due to increased
residential biomass use in the MARKAL results. These results mask significant sectoral
SOx NOx
and regional variations in changes in CAC emissions. A detailed comparison of these
results is provided in the EHI Report.
In June 2000, CCME Ministers
Chart 7.4
endorsed Canada-Wide Standards
Required Reduction in CAC Emissions by Region
to Achieve CWS for PM and Ozone
(CWS) for particulate matter and ozone.
(percent change from 1995 levels)
Previous analyses of options for these
standards had estimated the changes
Regional Targets PM2.5 SOX NOX VOC
Atlantic (Nova Scotia) 0 0 -30 -30
required in emissions by province in
Quebec -20 -20 -35 -35
order to achieve these standards. The
Ontario -30 -30 -45 -45
Manitoba 0 0 0 0
required reductions in CAC emissions
Saskatchewan 0 0 0 0
Alberta 0 0 -20 -20
by region to achieve the CWS for PM
B.C. 0 0 -10 -10
and ozone are shown in Chart 7.4.
Generally, the results indicate that the
reductions from any of the climate
change options are insufficient to
achieve all of the VOC targets and most of the PM targets. However, the options
sometimes exceed or provide a portion of the required reductions for SOX and NOX.
Further details on this comparison are contained in the EHI Report.
The Air Quality Valuation Model (AQVM) was used to estimate the physical and
economic health and environmental benefits associated with the Paths and Scenarios
examined. To improve the transparency of the AQVM results, it was shared with
stakeholders to provide a better understanding of the operating characteristics of this model.
The physical effects associated with the various Paths and Scenarios were estimated using
the concentration-response functions in the model. The model then uses market and non-
market information to convert these physical impacts into monetary terms - the most
controversial of these being the values used to estimate reductions in the risk of death. The
central estimate in the model of $4.1 million, for the value of a statistical life, implies that
the Òaverage individualÓ would be willing to pay approximately $400 to reduce their risk of
premature mortality by 1 in 10,000.
Data limitations made it impossible to estimate benefits for the full set of pollutants. The
analysis examined impacts from three pollutants -sulphate aerosols (SO4)32, ground-level
ozone, and sulphur dioxide (SO2). A range of human health impacts and a small group of
environmental impacts, associated with these pollutants was analyzed. Although it is
32
SO4 or sulphate aerosol, comes primarily from the oxidization of sulphur content in
fossil fuels upon combustion. Major combustion sources containing sulphur include:
coal, gasoline and diesel fuel, heavy and light fuel oil. Processing related sources of
sulphate include calcium sulphate in cement processing and oxidization of sulphur in
the smelting of crude grades of mineral ore.
expected that these pollutants should provide a good sense of the order of magnitude of
total benefits, the impacts from other pollutants were not included in this analysis.
The only measure of particulate matter (PM) used in this analysis was SO4. This has the
potential to understate the estimated impacts since it does not include many of the other
elements of the complex mixture composing PM. However, as many concentration-
response functions were developed based on reported sulphate aerosol measurements, it is
uncertain to what degree results are underestimated.
Because of limitations on the air quality data, estimates for SO4 are only available for
central and eastern Canada. Therefore, there is a potentially serious omission of possible
benefits from cleaner air in western Canada.
The co-benefits reported are an estimate of the societal benefits that could be realized in
various regions of Canada, resulting from the avoided non-market health and
environmental impacts provided by the analysis. These societal benefits represent the value
which Canadians place on these co-benefits, as determined by estimates of their
willingness-to-pay to achieve these avoided impacts. In the context of EHI analysis,
willingness-to-pay is the maximum amount of money or other goods a person is prepared to
forgo in order to avoid the negative health or environmental outcome.
The changes in ozone range from less
Chart 7.5
than 1 ppb-day per year in
Maximum Ambient Air Concentration Change
Newfoundland (MARKAL Path 2) to
180 ppb-day per year in Southern
Ontario ( MARKAL Path 3). These SO4 “g/m 3 SO2 ¥g/m3
ozone index changes, measured at air MARKAL Path 2 CA 0.76 3.97
quality monitoring stations correspond to CIMS Path 2 CA
MARKAL Path 3CA
0.46
0.76
2.38
4.01
a maximum 0.5 ppb change in mean MARKAL Path 2 KL 0.57 3.01
daily peak hour ozone.33 This is
roughly equivalent to 5-10 percent of
changes required by Canada-Wide
Standards. Chart 7.5 shows the
maximum changes in SO4 and SO2 by
Path from SOX reduction, interpolated from analysis using the acid deposition and oxidant
mechanism.
33
This takes into account the relative local emission contribution to local ozone
formation, so that the maximum changes in the index do not necessarily occur where the
maximum base index values occur.
Chart 7.6 shows estimates of the
Chart 7.6
Physical Impacts -2010
avoided physical impacts associated
MARKAL Path 2 CA
with MARKAL Path 2CA in 2010. To
Atlantic Quebec Ontario Western Total
Canada
Mortality 5 45 95 0 140
Hospital Admissions 5 50 110 0 160
put these physical impacts into
perspective - total air pollution is
(Respiratory/Cardiac)
Emergency Room Visits 10 125 275 1 410
estimated to cause about 5,000
Chronic Bronchitis (Adults) 15 155 325 0 490
Acute Bronchitis (Children) 200 1,950 4,300 0 6,450
premature deaths per year in Canada.34
Restricted Activity Days 3,000 31,000 78,000 1,000 115,000
Asthma Symptom Days 6,000 65,000 145,000 500 220,000
Days with Acute
The avoided death estimate under this
Respiratory Symptoms 20,000 210,000 480,000 2,000 715,000
Path is about 3 percent of this number.
Attempts have been made to ensure the more serious effects are separated from
These estimates do not include any
the more numerous effects, to eliminate double counting
Non-health impacts have not been tabulated in physical terms, but are included
in the calculation of monetary impacts
contribution from sulphate reductions in
western Canada.
The physical impacts were simulated for two years: 2010 and 2020, and then monetized to
yield economic estimates of the benefits in those two years. A linear interpolation from
2000 to 2020, was then performed to derive the net present value and annualized stream of
benefits over the period. Chart 7.7 presents a comparison of the estimated economic cobenefits
associated with the Paths that were examined. These benefits exclude any
contribution from sulphate reductions in western Canada.
The ranking of the Paths from most to
Chart 7.7
least co-benefits follows the order of the
Estimated Co-benefits
(96 $ million)
Paths ranked by SOX emission
reductions. This is to be expected, since
Net Present Value Annualized
7 percent 10 percent 7 percent 10 percent
the primary contribution to total benefits
comes overwhelmingly from SO4,
MARKAL Path 3 CA 6,200 4,500 580 520
MARKAL Path 2 CA 6,000 4,300 550 500
which is a secondary pollutant formed in
MARKAL Path 2 KL 4,500 3,200 410 370
CIMS Path 2 CA 3,800 2,700 350 320
the atmosphere from primary SOX
emissions. One important consequence
of these results is that the level of overall
uncertainty on the total estimates is
reduced, since SOX emissions are fairly
straightforward to model.
Comparing results from MARKAL, Path 2KL yields significantly less domestic cobenefits.
This is to be expected, since the purchase of international permits under the KL
Scenario reduces the required level of domestic action. These domestic mitigation activities
produce local CAC reductions, and hence improved air quality and the realization of more
34
This estimate is derived from the results in: The Effects of the Urban Air Ambient
Pollution Mix on Daily Mortality Rates in 11 Canadian Cities, R. Burnett, S. Cakmak
and J. Brook, Canadian Journal of Public Health, Volume 89, Number 3.
EHI co-benefits for Canada. This points to the trade-offs involved in the pursuit of
international GHG emissions trading programs.
The existence of thresholds has not been resolved.35 Therefore, a sensitivity analysis was
performed on this key parameter. The results show that when thresholds were applied to
SO4 and ozone, the economic benefits declined by about 50 percent.
A useful benchmark by which to evaluate these results is to compare them to other analyses
of co-benefits that have been made internationally. A number of developed and developing
country studies on this issue have been performed recently. In 2010, these analyses
indicates a range of $3 to $5 per tonne of CO2-equivalent. These are average benefits and
should not be compared with marginal costs from the energy-technology models.
However, this range is comparable to similar studies conducted in the US.
The co-benefits estimates should not be compared with changes in GDP, but there may be
similarities between the co-benefits and the energy-technology model costs. However,
those cost estimates exclude welfare costs and the co-benefit assessment only addressed a
portion of the benefits. Thus for this analysis they should not be compared.
There is a legitimate concern about Òdouble countingÓ benefits from improving air quality.
Caution must be exercised in establishing the policy baseline. The EHI analysis uses the
CEOU as its policy baseline. These assumptions are fairly stringent in that they only
included future policies that were well articulated. In this analysis, the clean air co-benefits
are attributed to GHG mitigation measures, since there are no future clean air policies in the
BAU. For example, GHG co-benefits that are created through reductions in SO2 cannot
also be attributed to new acid rain actions. The reasoning is that climate change mitigation
actions, reducing SO2 emissions, shifts the SO2 emissions baseline downward so that new
emissions reductions caused by acid rain action must be referenced to the new, lower, SO2
emissions baseline. Care must be taken with air issue analyses regarding the policy
baseline so that the same emission reduction benefits are not counted twice. It is important
to note that the assumptions surrounding the policy baseline impacts both costs and
benefits.
Uncertainties affect the results in three categories:
¥ Policy uncertainties, such as the implementation of Canada-Wide Standards for PM
and ozone;
¥ Translation of Issue Table measures and options into the energy-technology models
in the context of future technology; and
35
A threshold is an ambient concentration level for a given pollutant below which negative
health or environmental impacts do not occur.
¥ Inaccuracies arising from the chain of analysis in deriving quantitative EHI results.
The focus is on the third category since uncertainty surrounding policy and future
technologies are discussed elsewhere.
Qualifications for key EHI uncertainties are noted below:
¥ SOX estimates from energy use are reasonably accurate since sulphur content in
fuels is well known.
¥ NOX and VOC changes are uncertain, but are of lesser importance to the EHI
benefit estimates (benefits from ozone reduction are small relative to those related to
sulphate reductions).
¥ Ambient air quality analysis was based on an interpolation of complex photochemistry
modelling.
¥ Uncertainty regarding environmental and health impacts from changes in ambient
air quality are explicitly demonstrated as uncertainty ranges in the AQVM
estimates.
The inability to accurately forecast future technologies was identified as having a
significant effect on the EHI analysis. The absence of impacts from particulate matter in
western Canada implies that benefits are missing for part of Canada. The use of SO4 only,
as a particulate measure, obscures the potential for worsened air quality that could come
from biomass combustion, as indicated in the MARKAL Path 2 results.
7.4 Qualitative Assessment
The qualitative assessment is focussed on Paths 3 and 4 as they represent the most
interesting comparison from the point of view of a strategic policy assessment. The key
points from the relative comparisons are:
The relative emphasis on electricity sector reductions in Path 4, lead to less coal use and
consequent decreases in acid deposition and sulphate aerosol particulate matter. However,
increased use of large-scale hydro electric development may lead to habitat fragmentation
and biogenic release of mercury.
The relative emphasis on reductions from transportation and industry, in Path 3, will likely
lead to lower NOX and VOC emissions implying less ground level ozone formation. This
indicates improved urban health benefits and decreased ozone impacts on agricultural
productivity.
From the point of view of complementing the quantitative EHI analysis, the qualitative
assessment lead to the following key points:
¥ The possible extensive penetration of biomass combustion, as expected in the
MARKAL results, could potentially lead to urban particulate matter problems.
Policy-makers need to consider standards for alternative fuel use in order to ensure
that there is a GHG benefit to alternative fuels without increasing harmful
particulate emissions.
¥ CanadaÕs use of Kyoto flexibility mechanisms, as implied in Kyoto Loose and
Tight Scenarios, will likely mean fewer beneficial EHI impacts in Canada.
¥ Long-range transport of heavy metals (mercury, lead, cadmium), into the Arctic
could decrease as a result of less combustion of fossil fuels. Although not all heavy
metal deposition is from Canadian sources, decreased Canadian reliance on fossil
fuels for combustion would lead to some decline in heavy metal deposition in the
Arctic.
7.5 Future Work
This is the first time that such a detailed assessment of environmental and health impacts
has been undertaken in Canada. It clearly represents both a milestone and starting point. At
the August CCEAF workshop, stakeholders voiced their concerns that insufficient
resources and time have been allocated to this area. These comments clearly support the
need for additional research and development into the environmental and health impacts
associated with changes in air pollution. In particular:
Historical CAC Data. Improvements can be made in both coverage and accuracy
of the base data. Information is currently collected on a voluntary basis,
consideration should be given to mandatory provision of this information
CAC Projections. Technological change is an exogenous input into the CAC
model. Greater emphasis should be placed on trying to estimate the potential for
technological change.
Air Quality Models. The weakest link in the analytical framework is the
methodology used to estimate air quality changes. A library of model results was
examined and used to estimate the changes in air quality associated with the GHG
actions. It was not possible to run these detailed models because of time and
resource constraints. In the future, consideration should be given to creating
simplified versions of these air quality models to facilitate their use in policy
analysis.
Air Quality Valuation Model. The most significant improvement that could be
made to the AQVM is to expand the number of environmental impacts which the
model can examine.
Integrated Policy Analysis. Concerns about potential double counting and the
implications of climate change measures on air quality highlights the usefulness of
more integrated analysis of both climate change and clean air policy options.
Chapter 8
Conclusions
In April 1998, federal, provincial and territorial Ministers of Energy and Environment launched
the National Climate Change Process (NCCP), a wide-ranging inquiry into the feasibility and
implications of CanadaÕs Kyoto Protocol commitment.
The Analysis and Modelling Group (AMG) is one of a number of Issue Tables and working
groups that were formed to address the various dimensions of the challenge posed by the
Kyoto Protocol. The AMGÕs principal task was to undertake the so-called Òroll upÓ of the
options, associated analysis and other insights developed by the Issue Tables. The AMG
conducted its analysis in an open, transparent, step-by-step process to ensure broad
stakeholder review of the results.
The analytic approach to the roll up considered, among other important components, a series of
complex policy packages, designed by the National Air Issues Coordinating Committee, and
referred to as ÒpathsÓ and framework assumptions.
The roll up results provided in this report should not be viewed as a plan of action. The results
would be best described as a Òrange findingÓ exercise. They provide policymakers with a
directional guidance on some fundamental issues related to the achievement of the Kyoto
Protocol.
It is important to note that implementing the Kyoto agreement in Canada would likely be
done through a ÒpackageÓ of measures. Some elements of that package would be aimed at
reducing emissions. Other elements would be aimed at reducing the uneven impact of
emissions-reductions measures across sectors and regions. The analysis presented here
considers chiefly measures to reduce emissions; the overall impact of a complete policy
ÒpackageÓ has yet to be analysed.
To evaluate the various path-scenario combinations systematically, the AMG linked together a
number of specialized models available from the private sector or within government into an
overall modelling structure. The basic approach was to use micro models to aggregate the
direct impacts of the various path-scenario combinations and then to use these results as the
inputs to both macro and environmental and health models. The macro models incorporate all
the linkages and address trade, competitiveness and fiscal and monetary policy issues.
While the quantitative results from the models (CIMS, MARKAL, TIM and CaSGEM)
differ, the qualitative conclusions apply to all four sets of results and the key messages are
generally consistent across the different analytical frameworks
Main Learnings
The AMG was asked to address the question Òwhat are the economic and environmental
consequences, for Canada, of achieving the Kyoto target?Ó While not a definitive answer, the
analysis provides some important insights into this question. These Main Learnings are as
follows:
¥ At the national level, attainment of the target results in sustained, long-term,
negative economic impacts. In the long run, the reduction in gross domestic product
(GDP), relative to the business-as-usual case, ranges from 0 to 3 percent depending
on the path-scenario combination.
It is important to provide perspective on these estimates. For example, a reduction
in GDP of 3 percent in 2010 means that, over the decade, the economy will grow
by about 26 percent instead of 30 percent as projected in the reference case. This is
equivalent to the loss of roughly one yearÕs growth, or, viewed in absolute terms, in
2010, the loss in annual economic output of approximately $40 billion (or $1100
per capita).
¥ The overall GDP impacts vary over time. Initially, economic activity increases modestly
in response to increased investment in emissions reducing technologies. Thereafter,
however, higher production costs, deterioration in competitiveness and lower incomes
combine to reduce GDP below business-as-usual levels. The analysis also suggests that
the adjustment process may not be completed until after 2010.
¥ The provincial GDP impacts are generally within 1.5 percentage points of the national
average impact. The relative ranking of each province typically varies by Scenario. In
the Canada Acts Alone Scenario, Newfoundland, Prince Edward Island, Quebec and
British Columbia are less affected, relative to the national average, whereas Ontario and
Saskatchewan are more negatively affected. The impact on Alberta is close to the
national average. The results for Nova Scotia, New Brunswick and Manitoba vary
between the studies.
In the International Scenarios, Newfoundland, Prince Edward Island and British
Columbia remain in the Òless affectedÓ category and are joined by Manitoba.
Saskatchewan remains in the Òmore affectedÓ category, which now includes Alberta
and New Brunswick. By contrast, OntarioÕs GDP impact is smaller than under Canada
Acts Alone, becoming close to the national average. QuebecÕs position is largely
unchanged across the Scenarios, although one study indicates that its GDP would be
consistently higher than in the business-as-usual case. For Nova Scotia, one study
indicates impacts that are greater than the national average, while the other suggests the
opposite.
¥ The findings suggest the potential for substantial variability in GDP impacts across
industries. Unfortunately, it is not possible to identify unambiguously the sectors that
would be negatively or positively affected, since industry variation is not uniform across
models and Paths.
¥ The greatest potential for emissions reduction appears to reside in the electricity
generation sector Ð between 40 and 60 percent of the reduction. Two actions Ð
capture and storage of CO2 in aquifers in Alberta and Saskatchewan and enhanced
interprovincial hydro-electricity trade, in particular from Manitoba and Quebec to
Ontario Ð account for the bulk of this potential.
¥ Policies to reduce greenhouse gases will both reduce energy consumption and
encourage switching from more to less carbon-intensive fuels. All of the analysis
suggests some declines, relative to business-as-usual, in oil product and coal
consumption. Interestingly, natural gas consumption also declines largely because the
capture and storage of CO2 and enhanced hydro electricity trade reduce the
attractiveness of gas-fired electricity generation. Were these options not fully available,
natural gas demand would increase.
¥ The industrial sector, particularly the oil and gas industry, and the transportation sector
face significant challenges in achieving large emissions reduction.
¥ Were Canada to act alone in achieving its Kyoto target, the marginal cost of abatement
in 2010 could range from $40 to $120 per tonne of CO2. Were these costs to be
incorporated in energy prices, gasoline prices would increase by 13 to 35 percent,
natural gas prices (for residential use) by 30 to 75 percent, and average coal prices by
300 to 800 percent.
¥ The outcome of the negotiations concerning forestry and agriculture sinks is an
important factor in the cost to Canada of achieving the Kyoto target. According to one
estimate, a pessimistic assumption concerning this outcome, the effect of which widens
the gap by about 20 percent, results in an increase in the marginal cost of abatement
from $57 to $100 per tonne of CO2. This result also suggests that the Canadian
emissions abatement cost curve, constructed from the analysis and insights of the Issue
Tables, becomes increasingly steep as the target is approached.
¥ The analysis supports the contention that competitiveness Ð measured by changes in
productivity - will be adversely affected by the achievement of the Kyoto target. The
impact is somewhat attenuated if CanadaÕs trading partners are also committed to
attaining their targets and Canada has access to flexibility mechanisms. However, under
this Scenario, CanadaÕs trading partners will also face reductions in economic activity,
with adverse consequences for Canadian export performance. At the national level, the
net impact of these two forces is to reduce the GDP loss by about 0.5 percentage
points.
¥ The analysis strongly supports the conclusion that moving from individual sector targets
to an economy-wide target will achieve the desired objective at significantly lower cost.
Moreover, sector specific emissions targets do not distribute the economic burden
evenly across sectors.
¥ Based on preliminary modelling of some characteristics of emissions permit trading, this
instrument can be viewed as cost-effective mechanism for achieving emissions
reductions in the industrial and electricity sectors. However, the analysis to date
underscores the importance of the design of such an instrument, in particular the permit
allocation mechanism. Each allocation method carries with it different distributive effects
on the economy.
¥ Measures and actions to achieve the Kyoto target will also result in the reduction of
sulphates, ozone and other atmospheric pollutants. These reductions will lead to
ancillary benefits from improved air quality and improvements in human health. These
co-benefits are immediate, local and reasonably certain and can make a significant
contribution to the attainment of clean air goals as enunciated in the Canada-Wide
Standards Initiative.
¥ The analysis indicates that the societal benefits of these improvements in human health
are in the range of $300 to $500 million per year. Most of this value derives from
reduced risk of mortality. These societal benefits represent the value that Canadians
place on these co-benefits, as determined by estimates of their willingness-to-pay to
achieve these avoided impacts. They are not comparable to the GDP impacts noted
above. These estimates do not cover the full spectrum of pollutants and do not include
sulphate reductions in western Canada.
¥ Although increased reliance on the international mechanisms to reduce greenhouse
gases lowers the cost of achieving the Kyoto target for Canada, it also reduces the
domestic clean air benefits. The analysis suggests that this reduction in societal benefits
is on the order of $200 million per year.
The AMG recognizes that there are many uncertainties regarding projections of this nature,
which have been noted earlier. In order to test the robustness of the ÒrangeÓ various sensitivity
analyses were conducted. The three more important areas are forestry and agriculture sinks,
CO2 capture and storage and inter-provincial electricity trade. Of these three, the most
dramatic effect is given by the forestry and agricultural sinks. This is because of the assumption
that 16 Mt of CO2 could be sequestered at no cost. This would also be equivalent to increasing
the ÒgapÓ by 16 Mt. This particular sensitivity underlines the importance of the assumptions
used in the analysis and the significance of the reference case.
Analysis is most useful when it succeeds in resolving issues. Even when it fails to do so,
however, it is still valuable if it points to gaps in our understanding. The AMG has identified the
following areas where future analysis should yield useful insights:
¥ Much greater effort is required to measure welfare benefits and costs. Attention should
focus on the welfare implications of transportation and Òlife styleÓ change initiatives.
¥ Some progress has been made in identifying competitiveness impacts, but a much more
sophisticated understanding of this issue is required. In particular, there is a need to
evaluate the importance of the so-called ÒleakageÓ issue: the possibility that some
industries will lose market opportunities and investment to countries that have lower, or
no costs, to industry under the Kyoto Protocol.
¥ Much more analysis of the implications of the various approaches to the design of an
emissions trading system is required. The suggested priority area is the allocation
mechanism.
¥ More province-specific analysis is needed. The current macro-models develop
provincial impact estimates by distributing the national results according to the industrial
mix in each province. This approach is reasonable if the characteristics of an industry
are similar across provinces, but questionable if this is not the case. More refinement of
this assumption is required.
¥ The assumptions concerning how the United States might address the Kyoto Protocol
requirements are too simplistic to comprehend the complex ways in which that
countryÕs climate change policies could affect both CanadaÕs economy and its policy
options.
¥ The analysis, to date, has not comprehensively modelled how CanadaÕs other trading
partners might respond to their commitments under the Kyoto Protocol.
¥ Despite considerable progress, the quantitative co-benefits analysis requires further
development.
In addition to this report, the AMG has generated a large number of studies focussing on
the microeconomic, macroeconomic and environmental and health phases of the roll up
analysis. These, in turn, are supported by numerous reports, prepared by CCEAF and
others on specific topics. The AMG believes that this research will form a valuable base
for further climate change analysis.
Annex 1
Climate Change Economic Analysis Forum
MOVING FORWARD: Stakeholder Perspectives
August 31, 2000
Climate Change Economic Analysis Forum
Executive Summary
The CCEAF Moving Forward: Stakeholder Perspectives report is a collection of viewpoints
from the various parties involved in the National Climate Change Process (NCCP). It was
prepared separately from the AMG report. The main objectives of the ÒMoving ForwardÓ
project are to characterize the analytical dimensions of "unfinished business" and future research.
In particular, the project aims to identify and describe:
Analysis that would have been useful to do, had sufficient time and resources been available;
¥ Data gaps discovered through the NCCP;
¥ Sensitivity analyses that would contribute valuable future insights to existing work; and
¥ Process improvements for future consideration.
Based on the reports from CCEAF workshops from 1998 to 2000, and structured interviews
with stakeholders and modellers, key messages for "moving forward" have been grouped
around the following six themes and needs.
1. Framing the Analysis. There is a need to establish clear objectives, questions
and decision contexts to guide future analysis.
Analysis and modelling choices need to be driven by the questions and decisions they
are intended to inform. Although the overall goal of the exercise - to examine the costs,
benefits and impacts on Canada of implementing the Kyoto target - was clear,
objectives became fuzzier at more detailed levels. Issue Tables were sometimes
unclear about their goals, and modellers were sometimes unclear about what they were
to deliver. Given the involvement of such a large and diverse group of stakeholders,
strong co-ordination was required. For example, if the Tables had known the
composition of the Paths and Scenarios before they commenced their work, their
proposed options might have been different.
The use of decision structuring, and other techniques of decision analysis, will enable
future analysis to be more sharply defined. This in turn will reap benefits in terms of
matching model strengths to questions, and sequencing inputs required from various
sources.
2. Institutional Capacity-Building. There is a need to ensure that Canada has a
long-term capability to address future analytical challenges.
Given that climate change mitigation will be a long-term process with many uncertainties
along the way, and that Canada's actions to mitigate will similarly be a long-term affair,
there is a concern about the institutional capacity to address the issue over the short and
longer term. Repeated comments were made about the time constraint (two years) for
an undertaking as ambitious as the current NCCP. Modellers need opportunities,
outside the constraints of severe deadlines, to understand data inputs and the structures
of other models. Some individual modelling frameworks require further development,
while an integrated modelling capability is required to treat with consistency the various
analytical components. It was recognized that the process developed much "human
capital" through networked relationships, and that preserving this capital is essential in
building a long-term analytical capability.
3. Iterative Processes. There is a need to engage stakeholders and modellers in
parallel, iterative processes of mutual understanding.
If analysis and modelling are viewed as means to the end of enhanced understanding,
then stakeholders and modellers need a "back and forth" process. An early choice of
modelling frameworks, with preliminary runs provided by modellers, would have
generated considerable learning for stakeholders and modellers. Modellers working
together with Issue Tables would have provided natural, continuous "bilaterals" for
resolving issues such as technology representation, data adequacies, and sectoral
aggregation. A peer review process for the Tables may have improved the quality of
the options proposed for the roll-up.
4. Data Adequacy. There is a need to ensure that a long-term, co-ordinated
process is established to fill data gaps, guided by the value of information in
decision contexts.
The process provided an opportunity to identify, and in some cases address: data
issues, missing or uncertain data, or data in ÒwrongÓ formats. An important next step is
addressing the data gaps identified by the analysis and modelling process to date.
However, the acquisition and maintenance of data is costly; it must be guided by the
value of the information for expected future decisions. Overall, there is a need to
improve the reliability level around the inputs used at various stages of analysis.
5. Coping With Uncertainty. There is a need for analysis that supports the
development of strategies that will be robust against a range of uncertain
outcomes.
Uncertainties pervade the climate change issue. Leaving aside the "science"
uncertainties, there are many arising from the fields of technology, economics, policy
and international agreements. Difficulties are compounded when uncertainties must be
carried forward from one analytical framework to another. Sensitivity analyses and
scenario methods are two of the major tools for dealing with uncertainty that have been
explored in this process. Their use could be extended, and other tools for dealing with
uncertainty might also be explored.
6. Communicating Results. There is a need to ensure that communication of
these results clearly portrays the level of uncertainties and notes the extent to
which the results are preliminary.
Some important issues need to be resolved in communicating results to a wider group.
There is a need to determine what should be communicated and what is ready to be
communicated. Distinguishing between data and information is extremely important
when communicating results. Presentation of summary results will have to be done in a
manner that clearly identifies the level of uncertainties, and correctly portrays the
preliminary nature of these results. It was suggested that Workshops, Press Releases
and other means of professionally communicating the results should be used.
Post-JMM
In addition to addressing the NCCP process to date, it is important to look beyond the October
JMM. The analysis completed to date represents an important first step. Many gaps need to
be filled. Future analysis and modelling should focus far less on broad impacts on the national
economy, and much more on detailed study of individual promising policy measures capable of
taking Canada to its Kyoto target and beyond into commitment periods post-2012. It is
especially important to begin immediately detailed design work for a major economic instrument
such as domestic emissions trading.
99
Bibliography
It is anticipated that these documents will be available on the NCCP website
(http://www.nccp.ca).
Canada. Department of Finance, Economic Studies and Policy Analysis Division. 2000. A
Computable General Equilibrium Analysis of Greenhouse Gas Reduction Paths and
Scenarios.
Canada. National Climate Change Process. Analysis and Modelling Group. 2000. ÒAtlantic
Disaggregation Working Documents.Ó
Canada. National Climate Change Process. The Analysis and Modelling Group. 1999.
CanadaÕs Emissions Outlook: An Update.
Canada. National Climate Change Process. Agriculture and Agri-Food Climate Change Table.
2000. Reducing Greenhouse Gas Emissions from Canadian Agriculture.
Canada. National Climate Change Process. Buildings Issue Table. 2000.
Commercial/Institutional Sector Options Report.
Canada. National Climate Change Process. Buildings Issue Table. 1999. Residential Sector
Options Report.
Canada. National Climate Change Process. Credit for Early Action Table. 1999. Report of the
Credit for Early Action Table.
Canada. National Climate Change Process. Electricity Industry Issues Table. 1999. Options
Paper.
Canada. National Climate Change Process. Enhanced Voluntary Action Issue Table. 1999.
Options Report.
Canada. National Climate Change Process. Environmental and Health Impacts Subgroup of
the Analysis and Modelling Group. 2000. The Environmental and Health Co-benefits of
Actions to Mitigate Climate Change.
Canada. National Climate Change Process. Forestry Products Table. 1999. Options for the
Forest Sector to Contribute to CanadaÕs National Implementation Strategy for the Kyoto
Protocol.
Canada. National Climate Change Process. Industry Table. 2000. Overview Report.
Canada. National Climate Change Process. Industry Table. 1999. Petroleum Product
Refining Plant-Level Analysis of Costs and Competitiveness Implications of Greenhouse
Gas Emission Reductions.
Canada. National Climate Change Process. Industry Table. 1999. Options and Analysis
Paper for the Industrial Chemicals and Synthetic Resins Sector.
Canada. National Climate Change Process. Industry Table. 1999. Aluminum Industry
Options Paper.
Canada. National Climate Change Process. Industry Table. 1999. Cement and Concrete
Options Paper.
Canada. National Climate Change Process. Industry Table. 1999. Industrial Minerals
Options Paper.
Canada. National Climate Change Process. Industry Table. 2000. Iron & Steel Plant Level
Analysis and Options Paper.
Canada. National Climate Change Process. Industry Table. 1999. Semi-Fabricating Metals
Options Paper.
Canada. National Climate Change Process. Industry Table. 1999. Upstream Oil and Gas
Industry Options Paper.
Canada. National Climate Change Process. Industry Table. 1999. Other Manufacturing:
Technical Opportunities for Reducing Greenhouse Gas Emissions. Draft.
Canada. National Climate Change Process. Industry Table. 2000. Greenhouse Gas Options,
Policy and Measures for the Canadian Transportation Equipment Manufacturing
Industry.
Canada. National Climate Change Process. Kyoto Mechanisms Table. 1999. Options Paper.
Canada. National Climate Change Process. Municipalities Table. 1999. Munipalities Table
Options Paper.
Canada. National Climate Change Process. Sinks Table. 1999. Land-Use, Land-Use
Change and Forestry in Canada and the Kyoto Protocol.
Canada. National Climate Change Process. Technology Table. 1999. Enhancing Technology
Innovation for Mitigating Greenhouse Gas Emissions.
Canada. National Climate Change Process. Tradeable Permits Working Group. 2000. Using
Tradeable Emissions Permits to Help Achieve Domestic Greenhouse Gas Objectives.
Canada. National Climate Change Process. Transportation Table. 1999. Transportation and
Climate Change: Options for Action.
Climate Change Economic Analysis Forum. 1999. ÒFinal Project Report: Kyoto Mechanisms.Ó
Report prepared for the AMG.
Climate Change Economic Analysis Forum. 2000. Moving Forward: Stakeholder
Perspectives. Report prepared for the AMG.
Energy Information Administration. U.S. Department of Energy. 1998. The Impacts of the
Kyoto Protocol on U.S. Energy Markets.
Energy Research Group/ MK Jaccard and Associates. 2000. Integration of GHG
Emission Reduction Options Using CIMS.
Energy Research Group/ MK Jaccard and Associates. 2000. Modelling of Energy /
Technology Actions and Measures for Reducing Greenhouse Gas Emissions in the
Industrial Sector (The Industry Challenge). Prepared for the Analysis and Modelling
Group.
HALOA, Inc. 2000. Integrated Analysis of Options for GHG Reduction with MARKAL.
Hirshhorn, Ronald. 1999. Characteristics of the Canadian Economy and the Implications
for Climate Change. Report prepared for Industry Canada.
Informetrica, Ltd. 2000. Macroeconomic Impacts of GHG Reduction Options: National
and Provincial Effects.
Plourde, AndrŽ. 1999. Competitiveness and the National Implementation Strategy:
Elements of a Framework Analysis. Report prepared for the Analysis and Modelling Group.