Regulating Greenhouse Gases in Canada:
Constitutional and Policy Dimensions
Shi-Ling Hsu and Robin Elliot*
Les missions de gaz effet de serre du Canada
ont augment dramatiquement depuis les ngociations
du Protocole de Kyoto en 1997. Cette augmentation a
continu mme subsquemment la ratification du
Protocole par le Canada en 2002. En plus de la
dislocation conomique, les barrires constitutionnelles
la rglementation ont parfois t cites comme
justification la prudence dans la rglementation des
gaz effet de serre. Cet article value de manire
critique les arguments constitutionnels et examine les
considrations de politiques entourant les diffrents
instruments rglementaires qui pourraient tre utiliss
pour rduire les gaz effet de serre. Nous concluons
que la constitution canadienne ne prsente pas de
barrire significative la rglementation fdrale ou
provinciale et que les considrations de politiques
favorisent fortement lutilisation de deux instruments,
soit une taxe fdrale sur le carbone pour imposer un
cot marginal aux missions et la Loi canadienne sur
lvaluation environnementale pour valuer les projets
fdraux qui pourraient augmenter les gaz effet de
serre.
Canadas greenhouse gas emissions have risen
dramatically since the 1997 negotiation of the Kyoto
Protocol, and that rise has continued through Canadas
2002 ratification of the Protocol. Along with economic
dislocation, constitutional barriers to regulation have
sometimes been cited as the reason for caution in
regulating greenhouse gases. This article critically
evaluates the constitutional arguments and examines
the policy
surrounding various
regulatory instruments that might be used to reduce
greenhouse gases. We conclude that the Canadian
constitution does not present any significant barriers to
that policy
federal or provincial regulation and
considerations strongly
two
instruments: a federal carbon tax to impose a marginal
cost on emissions and the Canadian Environmental
Assessment Act to review federal projects that may
increase greenhouse gases.
considerations
favour
the use of
* Shi-Ling Hsu, Associate Dean for Special Projects, University of British Columbia Faculty of
Law. Robin Elliot, Q.C., Professor of Law, University of British Columbia Faculty of Law.
Shi-Ling Hsu and Robin Elliot 2009
To be cited as: (2009) 54 McGill L.J. 463
Mode de rfrence : (2009) 54 R.D. McGill 463
[Vol. 54
465
465
466
469
474
477
482
483
485
489
489
489
490
491
493
495
497
498
499
500
500
505
507
513
515
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
464
Introduction
I. Regulating Greenhouse Gas Emissions in Canada
A. Potential Regulatory Instruments
B. Federal Attempts at Greenhouse Gas Regulation
C. Provincial Experiences with Greenhouse Gas
Regulations
II. The Constitutional Dimension
A. Provincial Jurisdiction
1. Carbon Taxes
2. Cap-and-Trade and Intensity-Based Trading
Regimes
3. Command-and-Control Regimes
B. Federal Jurisdiction
1. A Carbon Tax
2. A Cap-and-Trade or Intensity-Based Trading
Regime
a. Criminal Law
b. The National Concern Branch of POGG
c. The National Emergency Branch of POGG
3. A Command-and-Control Regime
4. The Canadian Environmental Assessment Act
C. Summary
III. The Policy Dimension
A. The Canadian Environmental Assessment Act
B. Cap-and-Trade vs. Intensity-Based Emissions
Trading
C. Carbon Taxation vs. Cap-and-Trade
D. Command-and-Control Regulation
Conclusion
2009]
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
465
Introduction
In a 2007 speech to the Canadian Bar Association, former Alberta Premier Peter
Lougheed warned of an impending constitutional crisis over the regulation of
greenhouse gases. A major constitutional battle was brewing between the federal
government, which faces increasing international and domestic pressure to regulate
the emissions of greenhouse gases, and the government of Alberta, which jealously
guards its provincial prerogative to oversee emissions-producing oil and gas
development.1 Public pressure, in Lougheeds view, was likely to force the
passage of strong federal environmental laws, while the economic forces driving oil
sands development were likely to lead to resistance from Alberta in the form of
conflicting legislation.2
Is there really a constitutional storm on the horizon? Although there is tension
between federal and provincial authority over the regulation of Canadian greenhouse
gases, this tension need not and should not be an obstacle to sensible greenhouse gas
regulation.
I. Regulating Greenhouse Gas Emissions in Canada
Canadas greenhouse gas emissions have risen sharply since 1990, the baseline
year from which the commitments under the Kyoto Protocol to the United Nations
Framework Convention on Climate Change are derived.3 Indeed, Canadas increase
in total aggregate greenhouse gas emissions from 1990 to 2007 was the highest
among G8 nations,4 rising from 596 megatonnes in carbon dioxide equivalents (CO2
1 Peter Lougheed, Address (delivered to the Canadian Bar Association, Calgary, 14 August 2007)
[unpublished], online: Cable Public Affairs Channel
tional battle would centre on s. 92A(1) (Constitution Act, 1867 (U.K.), 30 & 31 Vict., c. 3, reprinted in
R.S.C. 1985, App. II, No. 5). See generally How to Head Off an Oil-Sands Clash, Editorial, The
Globe and Mail
(16 August 2007), online: The Globe and Mail
2 Lougheed, ibid.
3 11 December 1997, 2303 U.N.T.S. 148, 37 I.L.M. 22 (entered into force 16 February 2005) [Kyoto
Protocol].
4 Allan Dowd, Canada Led G8 in Greenhouse Gas Emissions Growth Reuters (23 April 2008),
online: Planet Ark
United Nations most recent data, Canada ranked ninth among Annex I countries with a 21.7 per cent
increase in greenhouse gas emissions from 1990 to 2006, behind Turkey, Spain, Portugal, Australia,
Greece, New Zealand, Ireland, and Iceland (United Nations Framework Convention on Climate
Change (UNFCCC), National Greenhouse Gas Inventory Data for the Period 19902006 (Geneva:
United Nations Office at Geneva, 2008) at 9, online: UNFCCC
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
466
equivalents5) to 747 megatonnes in this period.6 It is now impractical for Canada to
comply with its Kyoto commitment to lower its emissions to 563 megatonnes. There
have been increases across almost all sectors between 1990 and 2007, including
emissions from electricity generation, transportation, petroleum production, mining,
agriculture, waste, and fugitive releases from natural gas production.7 It no longer
makes sense for Canada to unilaterally and immediately cease the upward momentum
of emissions and begin an emissions reduction of more than 25 per cent over the next
three years.8 However, given the direness of the climate change problem, Canadians
must embark upon an effective greenhouse gas emissions strategy. Fortunately, as this
article argues, a number of federal and provincial regulatory possibilities are available
that avoid constitutional confrontation.
A. Potential Regulatory Instruments
include regulation
While the many possibilities for greenhouse gas regulation have been treated
extensively elsewhere, a brief review of potential regulatory instruments will help
frame the discussion in the Canadian context. This part of article outlines the most
frequently discussed types of schemes: command-and-control regulations; cap-and-
trade programs; intensity-based emissions trading; carbon taxes; and regulation under
the Canadian Environmental Assessment Act.9 Alternative means of reducing
information disclosure,
greenhouse gases
government subsidies, voluntary
litigation. A
comprehensive treatment of all such methods, which would involve scores of ideas, is
beyond the scope of this article.
First, greenhouse gas regulation could take a traditional form of environmental
regulation sometimes referred to as command-and-control regulation. This term
typically contemplates some administrative standard that serves as a baseline for
pollution control performance. The standard could be fixed as a specified numerical
initiatives, and common
that mandates
law
regulated under
5 Carbon dioxide equivalents is a common metric used to directly compare emissions from all six
the Kyoto Protocol: CO2, methane, nitrous oxide,
greenhouse gases
hydrofluorocarbons, perfluorocarbons, and sulphur hexafluoride. The metric is used to create a
relative index that is weighted by the heat-trapping effect of emissions of the different greenhouse
gases, in comparison with the effect of a tonne of CO2. For example, since methane has twenty-one
times the heat-trapping power of CO2, emissions of methane are multiplied by twenty-one in
calculating the index (UNFCCC, ibid. at 3, n. 3). See also U.S. Environmental Protection Agency,
Emission Facts: Metrics for Expressing Greenhouse Gas Emissions: Carbon Equivalents and Carbon
Dioxide Equivalents (February 2005), online: U.S. Environmental Protection Agency
6 Environment Canada, Canadas 2007 Greenhouse Gas Inventory: A Summary of Trends,
online: Environment Canada
7 Ibid.
8 See Kyoto Protocol, supra note 3, art. 3, s. 1.
9 S.C. 1992, c. 37 [CEA Act].
467
than defining compliance
in
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
expression of performance, such as in the regulations governing chlor-alkali plants
under the Canadian Environmental Protection Act, 1999.10 These regulations provide
that [t]he quantity of mercury that the owner or operator of a plant may release into
the ambient air from that plant shall not exceed (a) 5 grams per day per 1,000
kilograms of rated capacity, where the source of the mercury is the ventilation gases
exhausted from cell rooms.11 Alternatively, a standard could be linked to industry
practices and could contain keywords that hint at how ambitious the polluter must be
relative to the industry practice, such as the Best Available Technology
Economically Achievable (BATEA) standard.12 While command-and-control
regulatory schemes take on a wide variety of forms, the distinguishing feature of
command-and-control systems is that compliance is determined administratively. This
determination often (but not always) focuses on whether an emitter has adopted the
right technology or industrial practices, or has achieved a level of performance
administratively deemed to be acceptable or attainable.
Second, in a marked break in philosophy from the traditional means of
environmental regulation, cap-and-trade programs have gained popularity as a
regulatory
terms of some
administratively set standard, cap-and-trade programs involve the issuance of
allowances to emitters that permit them to emit a certain quantity of pollution.
Compliance is thus determined solely by whether the emitter has enough allowances
to cover its emissions. Allowances can be traded, and economic theory predicts that
the allowances will flow to their highest and best useto those emitters for whom
emissions reduction would be the most costly. This flow has the effect of
concentrating emissions reductions among those for whom it would be cheapest,
thereby minimizing overall industry compliance costs. Additionally, cap-and-trade
programs are thought to spur innovation because imposing a cost on emissions should
induce emitters to undertake self-interested efforts to reduce their emissions. Cap-
instrument. Rather
10 S.C. 1999, c. 33 [CEP Act].
11 Chlor-Alkali Mercury Release Regulations, S.O.R./90-130, s. 3(1)(a).
12 This was the language in a 2005 plan by the then-governing Liberal Party mandating that new
industrial facilities large enough to be considered large final emitters would, for the first ten years,
have emissions targets based on the emissions rate obtainable by the industry (Notice of intent to
regulate greenhouse gas emissions by Large Final Emitters, C. Gaz. 2005.I.2489 (Canadian
Environmental Protection Act) at 2494-95). What exactly was meant by this terminology is unclear,
though similar language in U.S. statutes suggests that the technology required would lie somewhere
between those technologies and techniques that are commonly available and those that are cutting-
edge. The U.S. Clean Air Act provides that when a new stationary source of air pollution (defined in
the statute as certain criteria air pollutants: 42 U.S.C. 7408(g) (1970)) is constructed or
significantly modified, the facility must achieve the lowest achievable emission rate if it is located
in a heavily polluted zone (ibid., 7503(a)(2)), and must install the best available control
technology if it is located in a less polluted zone (ibid., 7475(a)(4)). The terms heavily polluted
zones and less polluted zones are our own. They reflect the more technical distinctions drawn by
the act, which refers to attainment areas and non-attainment areas. See Shi-Ling Hsu, The Real
Problem with New Source Review (2006) 36 E.L.R. 10095.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
468
and-trade programs in the greenhouse gas context typically involve the issuance of
allowances to emit some quantity of carbon or CO2.
Third, in the wake of concerns about the compliance costs of cap-and-trade
programs, a less effective alternative has emerged, one favoured by the last two
Canadian federal governments: intensity-based emissions trading. Intensity-based
emissions trading involves not hard and fixed caps, but moving caps that seek only to
reduce greenhouse gas emissions relative to the amount of goods produced, or the
greenhouse gas intensity, and not necessarily the absolute amount of emissions.
Under the intensity-based emissions trading programs proposed by Canadian
governments, allowances are issued to emitters on the basis of their productive
output. Thus, any emitter that becomes more efficient in operations will be given
more allowances. Because the cap is dependent upon productive output and can be
ratcheted up by the achievement of productive efficiencies, there is no hard and fixed
emissions cap per se, and no control over the absolute amount of emissions.
Fourth, similar in economic philosophy to cap-and-trade programs, Pigouvian
taxes have long been popular among economists to address large-scale pollution
problems,13 suggesting that a carbon tax may be appropriate. A carbon tax is a
payment based on the actual or anticipated quantity of carbon emissions released into
the atmosphere. In practice, the tax is levied at some point of sale involving a carbon-
based product that is intended for combustion.14 The rationales behind Pigouvian
taxation and cap-and-trade programs are the same: impose a marginal cost on
emissions, and the emitters that can most cheaply reduce emissions will do so. The
difference between taxation and cap-and-trade programs is that a cap-and-trade
program is essentially a quantity instrument, while a taxation program is a price
instrument; taxation programs offer a degree of certainty for emitters that the price of
emissions will stay at a particular level, while cap-and-trade programs attempt to set a
particular maximum level of emissions, but only among those emitters covered by the
program.
13 Pigouvian is meant to describe a tax that would be consistent with Pigous prescription that a
tax equal to the marginal social harm from pollution should be imposed to provide just the right
amount of disincentive for pollution: A.C. Pigou, The Economics of Welfare, 3d ed. (London:
MacMillan, 1929) at 133-37. Taxes that reflected the extent of negative externality thus became
known as Pigouvian taxes. See William J. Baumol & Wallace E. Oates, The Theory of
Environmental Policy, 2d ed. (Cambridge: Cambridge University Press, 1988) at 21-23 (In sum …
the proper corrective device is a Pigouvian tax equal to marginal social damage levied on the
generator of the externality with no supplementary incentives for victims at 23). See e.g. Tom
Tietenberg, Environmental and Natural Resource Economics, 3d ed. (New York: Harper Collins,
1992) (We have shown that as long as the control authority imposes the same emission charge on all
sources, the resulting reduction allocation automatically minimizes the costs of control at 373
[emphasis in original]); Paul A. Samuelson, Economics, 11th ed. (New York: McGraw-Hill Book
Company, 1980) (Economists propose that greater use be made of pricing mechanisms. Taxes are to
be put on firms and industries that put out effluents into the air and ground at 744).
14 See text accompanying notes 57-68.
469
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
Finally, some regulation may be achieved by using an existing federal statute, the
CEA Act. The CEA Act requires an environmental assessment for projects proposed
by a federal authority or receiving financial assistance from a federal authority, for
any sale or lease of federal lands, or for any federal action or allowance that
implicates an area of federal concern identified by regulation.15 The environment is
construed broadly, encompassing air, including all layers of the atmosphere.16 The
CEA Act already plays a powerful environmental role in requiring assessment of
almost all significant federal projects, and it might be deployed in a similar manner in
requiring agencies to consider the greenhouse gas implications of federal projects,
much as they already consider other environmental impacts. This regulatory option is
different from the other options in that it is a procedural one, and not one aimed at
achieving any substantive outcome.
B. Federal Attempts at Greenhouse Gas Regulation
In 2007, under international pressure, Prime Minister Harper dragged the
Conservative Party into the climate change discussion, announcing an intention to
reduce Canadas total emissions of greenhouse gases to 20 per cent below 2006 levels
by the year 2020, and 60 per cent to 70 per cent below 2006 levels by 2050.17 The
Harper plan is an intensity-based emissions trading program that covers most
greenhouse gasemitting industries, including the electricity generation, oil and gas,
aluminum, cement, and pulp and paper industries. Large facilities in existence before
2004 will have 2010 reduction targets of 18 per cent below 2006 levels, with 2 per
cent further reductions annually.18 New facilities (with a first year of operation after
2003) will be required to achieve intensity reductions of 2 per cent annually after the
third year of operation.19 Oil sands facilities coming online after 2012 must install
carbon capture and storage technology.20 As noted above, it is difficult to determine
how much emissions reduction an intensity-based emissions trading program will
actually achieve, because the number of allowances is keyed to productive output. If
there is economic pressure on output, then improvements in productive efficiency will
lead to the availability of more emissions allowances, thereby lifting the ceiling on
emissions.
Government projections of a 20 per cent decrease from 2006 levels by the year
2020 are hard to evaluate, based as they are on a complicated macroeconomic
15 CEA Act, supra note 9, s. 5(1).
16 Ibid., s. 2(1).
17 Environment Canada, Turning the Corner: Regulatory Framework for Industrial Greenhouse Gas
Emissions (Minister of Environment, March 2008) at 7, online: Government of Canada
Framework].
18 Ibid. at 3.
19 Ibid.
20 Ibid. at 11.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
470
model,21 but they clearly incorporate some optimistic assumptions. For example, the
model assumes that by the year 2020, the following will occur: despite the absence of
federal regulation, passenger and freight transportation efficiency, along with some
questionably large gains in automobile efficiency, will reduce emissions by thirty-five
megatonnes from a business-as-usual forecast;22 the East-West transmission grid will
be expanded to transport clean power across Canada,23 a project that will require
considerable inter-jurisdictional cooperation; contributions into a Technology Fund
will somehow generate twenty megatonnes of emission reduction;24 and offsets from
the agricultural and forestry sectorsgreenhouse gasreducing actions that would not
have otherwise been undertakenwill produce almost fifty-five megatonnes of
reduction.25 It may be unduly skeptical to discount these assumptions, but considering
the fanfare with which the federal government announced its intentions, these
assumptions seem like a tenuous foundation upon which to make such specific
claims.
Despite mutual criticism between the Liberal and Conservative parties over
greenhouse gas regulation, the current proposal bears an odd resemblance to a plan
rolled out in 2005 by thenPrime Minister Paul Martin, in that it is an intensity-based
emissions trading program that covers roughly the same set of seven hundred or so
large final emitters and allows contribution to a Greenhouse Gas Technology
Fund to substitute for actually achieving the mandated emissions intensity
improvements.26 If, as opposition parties and environmental groups have argued, the
current plan is insufficient,27 then the previous Martin plan was delusional. The
Martin plan was in large part an intensity-based emissions trading plan for large final
emitters, and the remaining four-fifths of the emissions reductions were projected to
occur as a result of a variety of vague spending programs, such as the Greenhouse
Gas Technology Fund.28 To put it bluntly, the Martin Plan consisted of a modest
21 Environment Canada developed a model called the Energy-Economy-Environment Model for
Canada, which incorporates a variety of economic factors, many of which are global in nature. See
National Round Table on the Environment and the Economy, Greenhouse Gas Emissions
Forecasting: Learning from International Best Practices, online: National Round Table on the
Environment and
the Economy
22 Environment Canada, Turning the Corner: Detailed Emissions and Economic Modelling
(Minister of Environment, March 2008) at 10, online: Government of Canada
23 Ibid. at 11.
24 Ibid. at 6.
25 Ibid. at 11.
26 Environment Canada, Regulatory Framework, supra note 17 at 14.
27 MPs, Environmentalists Slam Greenhouse Gas Targets CBC News (26 April 2007), online:
CBC News
28 For a critical analysis of the plan, see Report of the Commissioner of the Environment and
Sustainable Development to the House of Commons, Chapter 1: Managing the Federal Approach to
Climate Change (Ottawa: Office of the Auditor General, 2006) at 21-25.
471
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
emissions-trading plan and a collection of bald assertions about the effectiveness of
spending money on undefined research projects.
The Martin plan did contain an interesting twist, however: the emissions trading
plan for large final emitters included a safety valve provision that guaranteed that
the price of an allowance to emit a tonne of CO2 would not exceed fifteen dollars
during the 20082012 period.29 Such safety valves are not new to environmental
economics.30 If the safety valve level is low enough, it sets the price of emissions and
essentially creates a carbon tax.31 By most accounts, fifteen dollars per tonne is a low
level, and as noted above, the fact that the program was intensity-based means that
allowances could be plentiful enough to drive the price still lower.32 The interesting
question is why such an elaborate emissions trading plan with a safety valve would be
put in place if the goal was essentially to tax emissions at a maximum of fifteen
dollars per tonne. Could some sort of a tax scheme not be devised to achieve the same
thing, but in a much simpler fashion?
There are two answers to this question, one psychological and one political. The
psychological answer harkens back to the special aversion to all policies bearing the
word tax, especially in Alberta.33 Taxes per se are so unpopular in North America
that economists have argued that a safety valve is a way of introducing a tax-like
mechanism without necessarily introducing the baggage of emissions taxes.34
But there is another interesting aspect to this question: how did the figure of
fifteen dollars per tonne come about? The answer is not, as one might think, that it
represented the acceptable level for those in Albertan oil and gas industries. Fifteen
dollars per tonne would have represented a tax of about six dollars and sixty cents per
barrel of oil,35 and a mere four cents per litre of gasolinea cost that could almost
29 Notice of Intent to regulate greenhouse gas emissions by Large Final Emitters, supra note 12 at
2491; Greenhouse Gas Technology Investment Fund Act, s. 8(5), being Part 14 of the Budget
Implementation Act, 2005, S.C. 2005, c. 30, s. 96.
30 See e.g. Henry D. Jacoby & A. Denny Ellerman, The Safety Valve and Climate Policy (2004)
32 Energy Policy 481; Marc J. Roberts & Michael Spence, Effluent Charges and Licenses under
Uncertainty (1976) 5 Journal of Public Economics 193; Willam A. Pizer, Prices vs. Quantities
Revisited: The Case of Climate Change (Resources for the Future Discussion Paper 98-02, 1997),
online: Resources for the Future
31 Jacoby & Ellerman, ibid. at 481.
32 See e.g. Jacoby & Ellerman, ibid. at 484.
33 Carbon Tax Proposal a Non-Starter in Alberta CBC News (8 January 2008), online: CBC News
34 See e.g. Jacoby & Ellerman, supra note 30 at 484; Willam A. Pizer, Choosing Prices or Quantity
Controls for Greenhouse Gases (Resources for the Future Climate Issues Brief No. 17, 1999), online:
Resources for the Future
carbon tax can be achieved without the baggage accompanying an actual tax at 9).
35 A carbon tax levied on production of a barrel of oil would measure the carbon content on a barrel,
and one could levy the tax against the producer. The carbon content of crude oil is approximately 19.9
metric tonnes per terrajoule, or 0.0199 tonnes per gigajoule. A barrel of oil typically contains 6.1
in
the
losing
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
472
invisibly be passed on to the gasoline consumer. It was not Alberta that insisted on
this safety valve.
The answer is a political one. It can be found in the ridings that the Liberal
imminent federal election:
government was most afraid of
manufacturing-heavy, greenhouse gasintensive ridings
in Southern Ontario.
Pandering to Alberta would have done the Liberal Party no good, but minimizing
defection to the Conservative Party in Liberal ridings was critical to preserving a
Liberal minority government. For example, the Ancaster-Dundas-Flamborough-
Westdale riding is home to the Carmeuse Lime production facility, which emitted
over six hundred thousand tonnes of CO2 in 200436 (about three-quarters of Ontarios
lime production emissions37). That riding saw very close races in 2004 and 2006:
Liberal candidate Russ Powers narrowly defeated Conservative candidate David
Sweet by 40 per cent to 36 per cent in 2004,38 only to have that advantage reversed in
a 2006 loss.39 In the extremely greenhouse gasintensive riding of Sarnia-Lambton,
facilities belonging to Cabot Canada, Imperial Oil, Suncor, BP Canada, TransAlta
Energy, and NOVA Chemicals emitted a reported total of over 4.52 megatonnes of
CO2 in 200640over 6 per cent of all of Ontarios emissions. That riding saw a
similar flip in a tight race, with Liberal MP Roger Gallaway narrowly winning in
gigajoules, so the carbon content of a barrel of oil is typically 0.12 tonnes. The tax is on emissions of a
tonne of CO2, which has a molecular weight of 44, as opposed to carbon, which has the molecular
weight of 12. Emitting 0.12 tonnes of carbon would thus be the same as emitting 0.44 tonnes of CO2.
With a tax of $15 per tonne of CO2, the tax on a barrel of oil would be about $6.60 per barrel. There
are forty-two U.S. gallons of gasoline to a barrel, so that this tax amounted to about 15 per gallon of
gasoline (Bioenergy Feedstock Information Network, Bioenergy Conversion Factors, online:
Bioenergy Feedstock
Information Network
Department of Energy, Weekly All Countries Spot Price FOB Weighted by Estimated Export Volume,
online: Energy Information Administration
36 Environment Canada, 2004 Emissions Data, Table 3: Summary of GHG Emissions by Facility,
online: Environment Canada
37 Environment Canada, National Greenhouse Gas Inventory Report, 1990-2004: Greenhouse Gas
Sources and Sinks in Canada, Annex 12: Provincial / Territorial Greenhouse Gas Emission Tables,
19902004, online: Environment Canada
38 CBC News, Canada Votes 2004, online: CBC News
[Environment Canada, 2006 Emissions Data, Table 3]. Ontarios total CO2 emissions in 2006 were
71.4 megatonnes (Environment Canada, 2006 Emissions Data, Table 2: GHG Emissions by
Provinces/Territories, online: Environment Canada
riding/110/>.
riding/110/>.
473
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
2004 but losing to a Conservative in 2006.41 The Liberal Party did manage to hang on
to their Mississauga South riding, home to Petro-Canada and St. Lawrence Cement
plants, the sources of another 1.72 million tonnes of CO2 emissions;42 but Liberal
incumbent Paul Szabos margin of victory shrunk from eighteen points in 2004 to
four in 2006.43
Politicians are particularly sensitive when talking about greenhouse gas
regulation in Southern Ontario because many of the regions industries, such as
automobile manufacturing (both parts production and assembly), lime and cement
manufacturing, and chemical manufacturing, are vulnerable to trade pressures.
Carbon taxes are particularly unwelcome in an economically distressed environment.
The suffering and high-emitting automotive industry is always nervous about
greenhouse gas regulation. Indeed, some of Ontarios most competitive ridings, such
as the St. Catharines and Oshawa ridings,44 are home to General Motors truck and car
assembly plants. Rather than imposing vehicle fuel efficiency regulations on the
Canadian auto industry, the 2005 Liberal Plan instead entered into a memorandum of
understanding with the industry, calling for a reduction of 5.3 megatonnes per year by
2010.45 This was an unambitious target, given that road vehicles accounted for 135
megatonnes of greenhouse gas emissions in 2005.46
Ontario also produces almost half of Canadas cement,47 40 per cent of which is
exported to the United States.48 In a highly competitive world market, imposing added
costs upon Canadian cement manufacturers might affect their competitiveness,
causing their world market share to fall. A cement industry spokesperson reports that
cement imported from China is only slightly more expensive than that made in North
America. The difference between Canadian cement and Chinese cement landing in
41 CBC News, Canada Votes 2004, online: CBC News
42 Environment Canada, 2006 Emissions Data, Table 3, supra note 40.
43 CBC News, Canada Votes 2004, online: CBC News
44 In 2004, Conservative Colin Carrie won the Oshawa riding by a less than 1 per cent margin (CBC
News, Canada Votes 2004, online: CBC News
and in 2006, Conservative Rick Dykstra won the St. Catharines riding by 0.42 per cent (CBC News,
Canada Votes 2006, online: CBC News
45 Natural Resources Canada, Automakers Agreement to Reduce GHG Emissions, online: Natural
Resources Canada
46 Environment Canada, National Inventory Report, 19902005: Greenhouse Gas Sources and
Sinks in Canada, Table S-3, online: Environment Canada
47 Ontario Ministry of Northern Development, Mines, and Forestry, Cement Production and
Quarrying in Ontario (June 2009), online: Government of Ontario
48 Cement Association of Canada, Backgrounder, Cement [unpublished, on file with author].
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
474
Seattle is about fifteen dollars per tonne.49 Because the cement industry emits
greenhouse gases at the rate of very roughly one tonne of CO2 to one tonne of
finished cement,50 a fifteen dollar per tonne tax on CO2 would exactly offset the
competitive advantage currently enjoyed by Canadian cement manufacturers over
their Chinese competitors. Could this fact have given rise to the fifteen dollar per
tonne safety valve? Certainly, no government official or cement
industry
representative would admit as much, but the coincidence is curious.
The safety valve, then, seems to have been aimed not at protecting Albertas oil
and gas interests, but at protecting Ontario manufacturing interests and addressing the
fear that manufacturing jobs would be lost to the United States, which had no
prospect of greenhouse gas regulation in 2005. But Canadian public opinion, and
Ontarian public opinion in particular, has never been as fearful of greenhouse gas
regulation as federal politicians assume. Greenhouse gasintensive (and supposedly
fearful) Ontario has joined Quebec, British Columbia, and Manitoba in the Western
Climate Initiative, a California-led state and provincial effort to reduce greenhouse
gases.51 The federal government has always seemed to trail public opinion and even
industry opinion on greenhouse gas regulation. While Canadas constitution might
appear to present obstacles to greenhouse gas regulation, closer inspection reveals
that this is not the case. The only obstacles are political, and even these are not
necessarily accurately perceived.
C. Provincial Experiences with Greenhouse Gas Regulation
While greenhouse gas policy has been a political football at the federal level,
provincial governments have largely gone their own separate ways in developing (or
not developing) greenhouse gas policies. In 2002, Alberta announced its plan to
reduce carbon intensity to 50 per cent below 1990 levels by 2020.52 Again, no actual
emissions reduction was required, only an improvement in the rate of greenhouse gas
emissions per unit of output. The non-profit Pembina Institute issued an analysis
showing that the intensity targets were so lax they could have allowed a 72 per cent
increase in emissions by 2020.53
49 Email from Martin Vroegh, Environment Manager, St. Marys Cement Inc., to Patrick OBrien,
(15 July 2008) [on file with author].
50 Ibid.
51 See text accompanying note 68.
52 Climate Change and Emissions Management Act, S.A. 2003, c. C-16.7, s. 3.
53 Pembina Institute, Media Release, An Embarrassing Week for Albertans: Stelmachs Proposals
on Climate Change Will Actually Legislate an Increase in Emissions (9 March 2007), online:
Pembina Institute
475
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
An updated plan was announced in 2007, which called for an interim set of
intensity targets to be met by 2010.54 The government of Alberta also announced that
it would embark upon a program to fund carbon capture and storage, an end-of-pipe
technology that captures CO2 as it leaves the smokestack and pipes it to underground
caverns to be stored in perpetuity.55 Generally sticking with its 2002 plan, Alberta
projected that its emissions in 2050 would be 14 per cent lower than in 2005. Like the
federal government, the Alberta government underscored the fact that the 2050
emissions reductions would be 50 per cent below business-as-usual levels,56 which
certainly sounds like an improvement. But that statistic compares the emissions
reduction with a projected upward trajectory of future emissions growth; the Alberta
government is essentially congratulating itself for diverging from its current
profligacy.
British Columbia and Quebec have implemented carbon taxes levied at the point
of sale, in essence a sales tax on fossil fuels sold in those provinces. This approach
has many administrative advantages, as the wholesale or retail purchase of fossil fuel
is an easily trackable transaction and therefore a convenient enforcement point. In
general, carbon taxes are administratively simpler to design and carry out than any
emissions trading scheme, particularly an intensity-based scheme.
The Quebec carbon tax applies to the distribution within the province of
gasoline, diesel fuel, heating oil, propane, petroleum coke or coal, but not aviation
fuel, marine bunker fuel, hydrocarbons used as raw material by industries that
transform hydrocarbon molecules through chemical or petrochemical processes or
renewable fuel content.57 The carbon tax is administered by the Regie de lenergie,
the provincial energy regulatory agency, which determines the tax rate annually by
[t]aking into account greenhouse gas reduction targets … and the overall financial
investment.58 The actual levy paid by distributors of fossil fuels is determined at the
end of the year by dividing the desired amount of annual financial investment into a
Green Fund by the total amount of carbon emissions,59 then calculating each
distributors share of those emissions, taking into account the carbon content of
different fossil fuels.60 Fossil fuels sold in Quebec are presumed to be intended for
54 Specified Gas Emitters Regulation, Alta. Reg. 139/2007, ss. 3-4; Alberta, State of the
EnvironmentClimate Change: Greenhouse Gas Emissions Intensity, online: Government of Alberta
55 Alberta, News Release, Alberta to Cut Projected Emissions by 50 per cent under New Climate
Change Plan (24 January 2008), online: Government of Alberta
56 Alberta, Albertas 2008 Climate Change Strategy: Responsibility/Leadership/Action (January
2008) at 7, online: Government of Alberta
57 An Act respecting the Rgie de lnergie, R.S.Q. c. R-6.01, s. 85.34, s.v. fuel.
58 Ibid, s. 85.36.
59 Regulation respecting the annual duty payable to the Green Fund, R.Q. c. R-6.01, r. 6, s. 2.
60 Ibid., s. 4.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
476
consumption in Quebec unless otherwise shown by the distributor.61 Quebecs carbon
tax took effect in 2007.62
In 2008, British Columbia enacted a carbon tax of ten dollars per tonne of carbon
emissions (as measured by the carbon content).63 The tax increases by five dollars per
year to thirty dollars per tonne in 2012. For gasoline, the tax amounted to 2.34 per
litre in 2008, set to increase to 7.09 per litre by 2012. Diesel fuel and home heating
oil start at a tax of 2.69 per litre and rise to 8.09 by 2012.64 An important political
element of this plan was the stated intention to make the carbon tax revenue neutral
by somehow returning revenues from the tax to provincial residents and firms.
Forecasted tax revenues seem to allow the B.C. Ministry of Finance to announce
lump sum payments as well as specific cuts in corporate, small business, and personal
income tax rates.65 Notably, the lump sum payments and the personal income tax
reductions are tilted towards lower-income British Columbians to address perceptions
that consumption-based taxes such as carbon taxes and gasoline taxes are
regressive.66
In addition, the British Columbia government has passed an act providing for a
cap-and-trade program that, when it comes into force, will apply to greenhouse gas
emitters within the province.67 Almost all of the specifics of the program have been
left to regulations, which is understandable given the provinces commitment to
participate in the California-led emissions-trading reduction plan, the Western
Climate Initiative, the details of which have not been finalized.68
Manitoba, which has also joined the Western Climate Initiative, announced that it
intends to legislate a commitment to meeting its share of Canadas Kyoto targets: a 6
per cent reduction in greenhouse gases below 1990 levels.69 Unfortunately,
Manitobas plan seems predicated on the same creative accounting employed by the
61 Ibid., s. 5.
62 Ibid., s. 9.
63 Carbon Tax Act, S.B.C. 2008, c. 40, Sch. 1(1), Table 1.
64 Ibid.
65 British Columbia, Balanced Budget 2008, Backgrounder: B.C.s Revenue-Neutral Carbon Tax
(2008), online: Government of British Columbia
66 The Climate Action Credit provides an annual lump sum payment of $100 per adult and $30
per child, increasing in future years. Personal income tax rates will be reduced on the first $70 000 in
earnings (ibid. at 2). The actual determination of whether a gasoline tax is regressive or not is
complicated. For further discussion, see text accompanying notes 208-13.
67 Greenhouse Gas Reduction (Cap and Trade) Act, S.B.C. 2008, c. 32 (assented to 29 May 2008).
68 See Western Climate Initiative, online: Western Climate Initiative
69 Manitoba, News Release, Beyond Kyoto Outlines Manitobas Green Future: Rondeau (21 April
2008), online: Government of Manitoba
477
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
last two federal governments.70 It measures emissions reduction in terms of its
divergence from a business-as-usual baseline. For example, Manitoba credits itself
with 1.1 megatonnes of greenhouse gas reduction for construction of the Wuskwatim
Hydro Generation Project, which will generate electricity for export out of the
province.71 While this hydro project may be a laudable way to meet increasing
electricity demands, it is a bit self-serving to call the construction of a dam an
emissions reduction.
Ontario and Quebec are also jumping on board with the Western Climate
Initiative. In 2008, those provinces entered into a biprovincial memorandum of
understanding, agreeing to agree on a joint cap-and-trade scheme.72 While details are
lacking, a joint initiative of the two most populous Canadian provinces is clearly a
signal of widespread impatience with federal efforts. Ontarios initiative also defies
federal politicians expectations that greenhouse gas regulation would be a political
hot potato in that greenhouse gasintensive manufacturing region.
Curbing greenhouse gas emissions in Canada will obviously be challenging, as it
will be for any industrialized country subject to Kyoto targets. But an overly cynical
treatment of the greenhouse gas problem as a political football and the dubious use of
business-as-usual baseline calculations are surely not helping matters. These tactics,
along with the perception in some quarters that constitutional barriers exist, pose
unnecessary obstacles to the formation of meaningful greenhouse gas regulation.
British Columbia and Quebec have certainly taken a lead in greenhouse gas
regulation; but the magnitude of the greenhouse gas reductions that are required of
Canada necessitates a federal response, and one that is considerably more serious than
any proposed to date.
II. The Constitutional Dimension
The validity under sections 91 and 92 of the Constitution Act, 1867 of legislation
enacted by the federal and provincial orders of government to regulate greenhouse
gas emissions will depend on a number of factors.73 One of these is obviously the
70 Manitoba, Kyoto and Beyond: Meeting and Exceeding Our Kyoto Targets, online: Government of
Manitoba
71 Ibid. at 3.
72 Ontario, Memorandum of Understanding Between the Government of Ontario and the
Government of Quebec: A Provincial-Territorial Cap and Trade Initiative (2 June 2008), online:
Government of Ontario
73 The jurisdictional question addressed in this part of the paper has been discussed in one form or
another by a number of authors already. See e.g. Chris Rolfe, Turning Down the Heat: Emissions
Trading and Canadian Implementation of the Kyoto Protocol (Vancouver: West Coast Environmental
Law Research Foundation, 1998); Joseph F. Castrilli, Legal Authority for Emissions Trading in
Canada in Elizabeth Atkinson, ed., The Legislative Authority to Implement a Domestic Emissions
Trading System (Ottawa: National Round Table on the Environment and the Economy, 1999); Philip
Barton, Economic Instruments and the Kyoto Protocol: Can Parliament Implement Emissions
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
478
precise nature of the legislation enacted. Some kinds of legislation will be easier to
defend than others. For example, there is little reason to doubt that under subsection
91(3), the federal order of government has the power to enact legislation imposing a
carbon tax.74
Another factor is the extant body of jurisprudence governing the scope and
meaning of the various heads of federal and provincial power in sections 91 and 92,
which the two orders of government would rely on in support of their legislation. In
the case of some of the relevant heads of powerParliaments power to legislate for
the Peace, Order and Good Government of Canada (POGG), for examplethe
courts have formulated reasonably comprehensive definitions or tests. In the case of
othersthe provincial legislatures power to legislate in relation to Property and
Civil Rights in the Province, for exampleour understanding of their scope and
meaning is based on a series of decisions rendered over a long period of time that tell
us which kinds of matters come within the head of power and which do not. In
either case, judges often have a good deal of leeway when called upon to apply the
extant jurisprudence in a specific case.
A third factor is the set of analytical tools the courts have created to assist them in
characterizing particular legislative enactments for division of powers purposes, and
the manner in which those tools would be used in the context of challenges to
particular legislative enactments. That characterization processthe determination of
the impugned legislations true matter or pith and substanceis critical to the
outcome of a constitutional attack on division of powers grounds. The parties to the
challenge will each advance one or more characterizations that, in their view, will
improve their chances of obtaining a favourable result. While the tools judges use to
make that determination do serve to constrain the choices available to them in this
regard, those tools are nevertheless sufficiently malleable to leave judges with a great
deal of room to manoeuvre in many cases.
A fourth factor is the attitude that the judiciary will bring to bear on the task of
reviewing the constitutionality of legislation in this area. It is this factor that will
influence the choices judges make in exercising the discretion they have in such
cases. The judicial attitudes that will matter most are attitudes towards Canadian
federalism, both generally and in the specific context of environmental protection,
and, more particularly, attitudes towards the goals underlying attempts to reduce
greenhouse gas emissions. Some judges can be expected to have centralist leanings,
either generally or in this specific context, others to have provincialist leanings, and
still others to be agnostic and therefore receptive to shared jurisdiction in this area.
Trading Without Provincial Co-operation? (2002) 40 Alta. L. Rev. 417; Nigel D. Bankes & Alastair
R. Lucas, Kyoto, Constitutional Law and Albertas Proposals (2004) 42 Alta. L. Rev. 355. It should
be noted that the conclusions reached by these various authors in relation to the specific issues they
considered were far from unanimous.
74 But see Nathalie J. Chalifour, Making Federalism Work for Climate Change: Canadas Division
of Powers over Carbon Taxes (2008) 22 N.J.C.L. 119.
479
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
Receptivity to both orders of government legislating in this area will likely be
enhanced by an acceptance of the importance of the goals underlying such attempts.
The fact that the validity of legislation depends on so many factors means that
confident predictions are difficult to make. While we make a number of predictions in
this article about the likelihood of certain kinds of legislative initiatives being open to
the two orders of government in this paper, we do not wish to be taken as having
committed ourselves unreservedly to those views.
We begin with three general observations about the manner in which the Supreme
Court of Canada has tended to approach the task of reviewing environmental
legislation on federalism grounds. First, the Court has made it clear that the power to
protect the environment does not reside exclusively with either Parliament or the
provincial legislatures. As Justice LaForest put it on behalf of an eight-member
majority in Friends of the Oldman River Society v. Canada (Minister of Transport):
[T]he Constitution Act, 1867 has not assigned the matter of environment sui
generis to either the provinces or Parliament. The environment, as understood
in its generic sense, encompasses the physical, economic and social
environment touching several of the heads of power assigned to the respective
levels of government.75
Justice LaForest in fact went so far as to say that the environment in this broad sense
was a constitutionally abstruse matter which does not comfortably fit within the
existing division of powers without considerable overlap and uncertainty.76
The jurisprudence makes it clear that this connection to heads of power on both
sides of the federal-provincial divide is present even if the word environment is
understood in more limited terms to mean the physical environment alone. Hence, the
courts have upheld both federal77 and provincial78 legislation designed to protect the
physical environment. They have been able to do so in part because of their
willingness to permit Parliament and the provincial legislatures to rely on their
respective jurisdictions over both causes and effects of polluting activities.79 For
example, Parliament can regulate the polluting activities of interprovincial railways
because it has jurisdiction over Railways … connecting [one] Province with any
other or others of the Provinces under paragraph 92(10)(a). It can also regulate
75 [1992] 1 S.C.R. 3 at 63, 88 D.L.R. (4th) 1 [Oldman River].
76 Ibid. at 64.
77 See e.g. R. v. Crown Zellerbach Canada Ltd., [1988] 1 S.C.R. 401, 49 D.L.R. (4th) 161 [Crown
Zellerbach cited to S.C.R.] (upholding the legislation under POGG); R. v. Hydro-Qubec, [1997] 3
S.C.R. 213, 151 D.L.R. (4th) 32 [Hydro-Qubec cited to S.C.R.] (using the criminal law power under
s. 91(27)).
78 See e.g. R. v. Lake Ontario Cement Ltd., [1973] 2 O.R. 247, 35 D.L.R. (3d) 109 (using s. 92(13)
and 92(16) together).
79 The term effects in this context is intended to refer to environmental damage to places, entities,
or activities that is caused by the polluting activity in question.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
480
polluting activities that harm the fisheries80 and the waters of the territorial sea81
because it has jurisdiction over seacoast and inland fisheries and the territorial sea
under subsection 91(12) and the POGG power, respectively. Similarly, it is generally
understood that the provincial legislatures can regulate the polluting activities of the
mining and manufacturing industries because they have jurisdiction over the business
activities of those industries under property and civil rights in subsection 92(13).82
Provincial legislatures can also regulate polluting activities that harm provincial
Crown lands and inland waterways because they have jurisdiction over such lands
and waterways under subsections 92(5) and 92(13), and/or 92(16), respectively.83
The courts willingness to approach the validity of environmental protection
legislation in this manner contributes greatly to the considerable overlap of federal
and provincial legislation in this area noted by Justice LaForest in Oldman River. The
same polluting activities can, in theory, be regulated by both orders of government
one on the basis of its jurisdiction over the cause of those activities and the other on
the basis of its jurisdiction over the entities or places experiencing the effects. For
example, a shipping company whose routes include waters that feed into local
waterways can at one and the same time be subject to federal legislation (enacted
under subsection 91(10)) and provincial legislation (enacted under subsections 92(13)
or 92(16)).84 Only if the provincial enactment can be said to conflict with the federal,
thereby triggering the application of the doctrine of federal paramountcy, will the
shipping company be able to avoid the application of the former. Even under the
Supreme Court of Canadas new approach to the doctrine of federal paramountcy, this
is not an easy hurdle to meet.85
The second observation is that the Supreme Court of Canada has permitted
Parliament to regulate certain kinds of polluting activities under its POGG and
criminal law (subsection 91(27)) powers even though it has had to push the doctrinal
80 See e.g. Northwest Falling Contractors Ltd. v. R., [1980] 2 S.C.R. 292, 113 D.L.R. (3d) 1.
81 All seven of the judges in Crown Zellerbach agreed with this proposition in obiter (supra note
82 See Peter W. Hogg, Constitutional Law of Canada, 2008 student ed. (Toronto: Thompson
Carswell, 2008) c. 30.7, [Hogg, Constitutional Law]. The use of this head of power to sustain
provincial legislation regulating industries such as these is a function of the early jurisprudence of the
Judicial Committee of the Privy Council. See e.g. Canada (A.G.) v. Alberta (A.G.) (Reference Re
Insurance Companies), [1916] 1 A.C. 588, 26 D.L.R. 288 (P.C.) [Insurance Reference cited to A.C.];
Reference Re the Board of Commerce Act, 1919, and the Combines and Fair Prices Act, 1919 (1921),
[1922] 1 A.C. 191 (P.C.) [Board of Commerce Reference].
83 There is no direct authority in support of this proposition that we are aware of, but in our view, it
can be said to be implicit in the approach taken to the division of legislative authority over the
environment in the majority reasons for judgment in Oldman River (supra note 75).
84 This assumes, of course, that the provincial legislation is directed at the protection of the local
waterways rather than at the polluting activities of ships per se.
85 See Robin Elliot, Safeguarding Provincial Autonomy from the Supreme Courts New Federal
Paramountcy Doctrine: A Constructive Role for the Intention to Cover the Field Test? (2007) 38 Sup.
Ct. L. Rev. (2d) 629.
77).
481
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
envelopes governing those two heads of power in order to do so. In Crown
Zellerbach, decided in 1988, the Court upheld the federal Ocean Dumping Control
Act86 on the basis of the national concern branch of the POGG power. It did so in
spite of the fact that, as the dissenting judges pointed out, the matter attributed to
this act (marine pollution), arguably lacked the characteristics required under the
national concern rubric.87 In Hydro-Qubec, decided in 1997, the Court upheld the
toxic substances provisions of the CEP Act88 under subsection 91(27) in spite of the
fact that, again as the dissenting judges pointed out, the provisions looked to be far
more regulatory than prohibitory in nature.89 Taken together, these two decisions can
be said to reflect a willingness on the part of the Supreme Court of Canada to use the
room to manoeuvre that the doctrine in this area leaves them with to afford the federal
government broad authority to protect the physical environment. These cases also
reflect a high degree of sympathy on the Courts part for the goal of environmental
protection.
Hydro-Qubec can also be said to reflect the Courts growing preference for
permitting both orders of government to legislate in furtherance of that goal. Justice
LaForest, who authored the majority reasons in the case, defended his use of
subsection 91(27) to validate the CEP Acts toxic substances provisions inter alia on
the ground that the use of the federal criminal law power in no way precludes the
provinces from exercising their extensive powers under s. 92 to regulate and control
the pollution of the environment either independently or to supplement federal
action.90 This feature of the federal criminal law power differentiates it from the
national concern branch of POGG, which the federal government had advanced as an
alternative basis upon which to sustain the toxic substance provisions. In other words,
by upholding the provisions on the basis of subsection 91(27), the Court would in no
way restrict the ability of the provincial legislatures to enact environmental protection
legislation using the array of weapons available to them.
86 S.C. 1974-75-76, c. 55.
87 Supra note 77. There were three dissenting judges in this case: Beetz, Lamer, and LaForest JJ.
Their reasons for judgment, authored by LaForest J., placed particular emphasis on the significant
negative impact that sustaining the act on the basis of the national concern doctrine would have on
provincial jurisdiction over the area in question (here, environmental protection), arguably the most
important consideration courts are required to take into account when asked to uphold federal
legislation under that rubric. See also Reference Re Anti-Inflation Act, [1976] 2 S.C.R. 373, 68 D.L.R.
(3d) 452 [Anti-Inflation Reference cited to S.C.R.].
88 Supra note 10.
89 Supra note 77. The dissenting judges were Lamer C.J.C., Sopinka, Iacobucci, and Major JJ. Their
reasons for judgment were co-authored by Lamer C.J.C. and Iacobucci J. The test that federal
legislation has to meet in order to qualify as criminal law under s. 91(27) includes the requirement that
the legislation be prohibitory in character. See Reference Re Validity of Section 5(a) of the Dairy
Industry Act, [1949] S.C.R. 1, [1949] 1 D.L.R. 433 [Margarine Reference cited to S.C.R.]. It is worth
noting that the majority in Hydro-Qubec accepted this test as the governing one (supra note 77).
90 Hydro-Qubec, ibid. at para. 131.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
482
The third observation is as follows: the law is clear that the power to enact
legislation in order to implement international treaty or convention obligations
undertaken by the Government of Canada does not fall to Parliament simply because
the legislation has been enacted for that purpose. As Lord Atkin of the Privy Council
put it in the Labour Conventions Reference, [f]or the purposes of ss. 91 and 92, …
there is no such thing as treaty legislation as such.91 Jurisdiction to enact legislation
that implements treaty obligations rests with the order of government that has
jurisdiction to legislate in relation to the subject matter of those obligations. The
federal order therefore cannot claim jurisdiction to enact legislation that regulates
greenhouse gas emissions simply on the basis that such legislation is being enacted in
fulfillment of Canadas obligations under the Kyoto Protocol.92
That said, there is jurisprudential support for the notion that where federal
legislation has been enacted to implement treaty obligations, this fact might assist the
federal governments cause if that legislation were to come under attack on federalism
grounds, at least if the subject matter of the treaty relates to a matter of
predominantly extra-provincial as well as international character and implications.93
That language comes from Justice LeDains majority reasons for judgment in Crown
Zellerbach, in which, as noted above, the Supreme Court of Canada upheld the
federal Ocean Dumping Control Act on the basis of the national concern branch of
POGG. That statute had been enacted in fulfillment of Canadas obligations under the
Convention on the Prevention of Marine Pollution by Dumping of Wastes and Other
Matter,94 and Justice LeDains description of that treatys subject matter as being of
predominantly extra-provincial as well as international character and implications
appears to have been a significant factor in his reasoning. Given that the Kyoto
Protocol clearly deals with a matter fitting that description, there is reason to believe
that federal legislation regulating greenhouse gas emissions would be on stronger
ground than it might otherwise be because of its connection to that treaty.
A. Provincial Jurisdiction
In this part of the article, we consider whether the provincial legislatures have the
requisite constitutional authority to regulate greenhouse gas emissions through the
91 Canada (A.G.) v. Ontario (A.G.) (Reference Re Weekly Rest in Industrial Undertakings Act,
Minimum Wages Act and Limitation of Hours of Work Act), [1937] A.C. 326 at 351, [1937] 1 D.L.R.
673 (P.C.) [Labour Conventions Reference].
92 But see Stewart Elgie, Kyoto, the Constitution and Carbon Trading: Waking a Sleeping BNA
Bear (or Two) (2007) 13 Rev. Const. Stud. 67 at 90-103.
93 Crown Zellerbach, supra note 77 at 436.
94 29 December 1972, 1046 U.N.T.S. 120, U.K.T.S. 1976 No. 43 (entered into force 30 August
1975). The Ocean Dumping Control Act went further in terms of its reach than this convention
required Canada to go. It applied to internal marine waters as well as the territorial sea and other
external marine waters. It did not, however, apply to inland waters (supra note 86).
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
vehicles of (1) a carbon tax, (2) a cap-and-trade or intensity-based trading regime, and
(3) a command-and-control regime.
483
1. Carbon Taxes
The provincial legislatures power to tax is prescribed by subsection 92(2) of the
Constitution Act, 1867 as the power to impose Direct Taxation within the Province
in order to the raising of a Revenue for Provincial Purposes.95 Those terms suggest
that in order for provincial legislation to be sustained on the basis of subsection 92(2),
the legislation must (1) impose a tax, which must (2) be direct, (3) be imposed within
the province, and (4) raise a provincial revenue.
Given the manner in which requirements (1), (2), and (3) have come to be
understood, there is little doubt that provincial legislation establishing a carbon tax of
the kind discussed above would be held to impose a tax that would be both direct and
imposed within the province. The monies paid under such legislation would clearly
be a tax. On the assumption that the tax was levied against consumers of the products
in question in respect of the particular units of those products purchased, as the
carbon tax in British Columbia is,96 that tax would be held to be a direct tax.97 Such a
tax is in the nature of a sales tax levied against consumers, which the courts have long
accepted as direct taxes. And the tax would be held to be levied within the province,
because the only consumers required to pay it would likely be those who either
purchase and consume the product in the province in which the tax is levied, or those
who, as residents of or business-owners in that province, purchase the product
elsewhere and bring it into the province for consumption.98
This leaves us with requirement (4): that the tax be levied in order to the raising
of a Revenue for Provincial Purposes. On the face of it, that language appears to
provide the basis for a challenge to a revenue-neutral provincial carbon tax like the
one imposed by the Legislature of British Columbia.99 Arguably, a tax that is required
by legislation to be revenue neutral, and is advertised as such, has not been levied in
order to the raising of a Revenue. And if the tax has not been levied for that purpose,
can it not be said that the legislation imposing it exceeds provincial jurisdiction under
subsection 92(2)?
This argument would not rest on the text of subsection 92(2) alone. The Privy
Council has given substantive content to similar language in subsection 92(9) of the
Constitution Act, 1867, which authorizes provincial legislatures to legislate in relation
95 Supra note 1.
96 Carbon Tax Act, S.B.C. 2008, c. 40, ss. 8-13.
97 The definition of a direct tax adopted by the courts is a tax that is levied against the very
persons expected to bear the burden of it. For a discussion of this distinction and the relevant
jurisprudence, see generally Hogg, supra note 82, c. 31.2, 31.7.
98 Ibid., c. 31.11.
99 See Carbon Tax Act, supra note 96, s. 3.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
484
to Shop, Saloon, Tavern, Auctioneer, and other Licences in order to the raising of a
Revenue for Provincial, Local, or Municipal Purposes. In Russell v. R., the Privy
Council held that the power of granting licences is not assigned to the Provincial
Legislatures for the purpose of regulating trade, but in order to the raising of a
revenue for provincial, local, or municipal purposes. The Act in question is not a
fiscal law; it is not a law for raising revenue.100 The Privy Council effectively held
that in order to fall within subsection 92(9), provincial legislation has to have been
enacted for the purpose of raising revenue. If that is how subsection 92(9) has been
understood, would subsection 92(2) not also be understood this way?
There is also the decision of the Privy Council in Reference Re Alberta Bills,101
which struck down a tax that the Social Credit government of Alberta imposed on
banks shortly after it came to power in the mid-1930s. In the Privy Councils view,
the real purpose of the tax was not to raise revenue from banks, but to eliminate them
from Alberta, and the legislation imposing the tax was therefore, in pith and
substance, banking legislation rather than taxation legislation. The clear implication
of Alberta Bank Taxation Reference is that even if a provincial tax does raise
additional revenues, that tax will not be sustained under subsection 92(2) if it relates
to a matter falling within a head of power in section 91. This reasoning also suggests
that the phrase in order to the raising of a Revenue imposes a substantive
requirement: only if the real purpose of the tax is to raise revenue will it qualify under
subsection 92(2).
In spite of these arguments for limiting the scope of subsection 92(2), we believe
it unlikely that a provincial carbon tax would be struck down on the grounds that it is
not in order to the raising of a Revenue for Provincial Purposes. While Russell gave
substantive content to the raising of a revenue language in subsection 92(9), the
practical consequence of that interpretation was simply that legislation regulating (or
prohibiting) the retail trade in liquor could not be anchored in that particular head of
power. It did not mean that the provincial legislatures were barred from enacting such
legislation. In fact, in Hodge v. R., decided within a year of Russell, the Privy Council
held that the provincial legislatures could regulate the retail trade in liquor under the
combination of subsections 92(8), 92(15), and 92(16).102 In the Local Prohibition
100 (1882), 7 A.C. 829 at 837, 8 C.R.A.C. 502 [Russell]. See also Ontario (A.G.) v. Canada (A.G.),
[1896] A.C 348 (P.C.) [Local Prohibition Reference] (in which the Privy Council relied on its prior
holding relating to s. 92(9) in Russell).
101 Alberta (A.G.) v. Canada (A.G.) (1938), [1939] A.C. 117, (sub nom. Reference Re Alberta Bills)
[1938] 4 D.L.R. 433 (P.C.) [Alberta Bank Taxation Reference]. The decision in this case has been
considered in at least two other cases, but in neither of these cases did it form the basis of the decision:
C.P.R. v. Saskatchewan (A.G.), [1951] 2 W.W.R. (N.S.) 424, 4 D.L.R. 21 (C.A.), affd [1952] S.C.R.
231; Cosyns v. Canada (A.G.) (1992), 7 O.R. (3d) 641, 88 D.L.R. (4th) 507 (Div. Ct.).
102 (1883), 9 A.C. 117.
485
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
Reference, decided in 1896, the Privy Council also held that the provinces could
prohibit that trade under either subsection 92(13) or subsection 92(16).103
Perhaps most importantly, it is factually untrue that a provincial carbon tax does
not raise revenue. It clearly does raise revenue. The province has simply chosen to
raise revenue in a different manner than it did previously. It would seem meddlesome
to hold that a province that chose to raise revenues by taxing carbon instead of
income could not make
is
constitutionally troublesome, how revenue neutral would a tax have to be to fall afoul
of that rule? And how could the courts be sure that a particular tax would in fact be
revenue neutral? A revenue-neutral carbon tax, which shifts taxation from income
taxes to another source, may reflect a different method of revenue raising, but it
indisputably raises revenues.
In the result, then, it is our opinion that provincially created carbon taxes would
be held to fall within the scope of subsection 92(2) of the Constitution Act, 1867, and
would therefore be upheld as valid.
that change. Moreover,
if revenue neutrality
2. Cap-and-Trade and Intensity-Based Trading Regimes
The validity of a provincially created cap-and-trade or intensity-based trading
regime would depend to a very considerable degree on the form it took, with the
controlling factor being the entities to which the regime applied. If the regime were
limited to business undertakings that the provincial legislatures have the authority to
regulate qua businesses under any or all of subsections 92(5),104 92(10),105 92(13),106
and 92A,107 there is good reason to believe that it would be upheld as valid. If,
103 Supra note 100.
104 The Management and Sale of the Public Lands belonging to the Province and of the Timber and
Wood thereon (Constitution Act, 1867, supra note 1).
105 This section reads:
Local Works and Undertakings other than such as are of the following Classes:
(a) Lines of Steam or other Ships, Railways, Canals, Telegraphs, and other Works
and Undertakings connecting the Province with any other or others of the Provinces,
or extending beyond the Limits of the Province;
(b) Lines of Steam Ships between the Province and any British or Foreign Country;
(c) Such Works as, although wholly situate within the Province, are before or after
their Execution declared by the Parliament of Canada to be for the general Advantage
of Canada or for the Advantage of Two or more of the Provinces (ibid.).
106 Property and Civil Rights in the Province (ibid.).
107 This head of power, which was added to the list of provincial powers by the Constitution Act,
1982 (being Schedule B to the Canada Act 1982 (U.K.), 1982, c.11), is not being reproduced verbatim
because it is a very lengthy one with numerous subsections. For our purposes, what is significant
about it is that it grants to the provincial legislatures power over the development and management of
non-renewable natural resources, forestry resources and electrical energy (ibid.).
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
486
however, it were not so limited, but were made applicable to business undertakings
that fall within federal legislative jurisdiction, there is good reason to believe that the
courts would hold the regime to be invalid, at least in its application to those
undertakings.
It is trite law that the power to regulate business activities in Canada resides
presumptively with the provincial legislatures under subsection 92(13) of the
Constitution Act, 1867, which grants those legislatures exclusive jurisdiction in
respect of Property and Civil Rights in the Province. The theory underlying the
inclusion of business activities within subsection 92(13) can be traced back to early
decisions of the Privy Council, which held that the freedom to engage in the business
activity of ones choice (and to engage in that activity in the manner of ones choice)
is a civil right.108 Hence, legislation that in any way restricts that freedomwhich
all regulation of business activities through licensing and other regimes does to some
extentis presumptively legislation in relation to civil rights. That presumption is,
however, a rebuttable one, and it will be overcome by constitutional grants of
legislative authority to Parliament over the business activities of particular industries.
Hence, it is clear that Parliament has jurisdiction to regulate the business activities of
postal services (subsection 91(5)), shipping companies (subsection 91(10)), those
engaged in seacoast and inland fisheries (subsection 91(12)), banks (subsection
91(15)), savings banks (subsection 91(16)), and interprovincial transportation and
communication undertakings (subsections 92(10) and 91(29)). As a result of judicial
decisions defining the scope of Parliaments POGG power, it is now also clear that
Parliament has jurisdiction to regulate the business activities of those involved in the
aeronautics109 and nuclear power generation110 industries. And, as a result of judicial
decisions defining the scope of subsection 91(2), it is clear that Parliament has
jurisdiction to regulate international and interprovincial trade, as well as to legislate in
the whole Dominion,111 a carefully
respect of [general]
circumscribed source of power pursuant to which Parliament has been able to
legislate in the areas of competition policy112 and trademarks.113
trade affecting
108 See the Insurance Reference, in which Lord Haldane spoke of the federal legislation there at
issue, which regulated large insurance companies, as … depriv[ing] private individuals of their liberty
to carry on the business of insurance (supra note 82 at 595). Later in his judgment, he stated that it
must now be taken that the authority to legislate for the regulation of trade and commerce [in s. 91(2)]
does not extend to the regulation by a licensing system of a particular trade in which Canadians would
otherwise be free to engage in the provinces (ibid. at 596).
109 Johannesson v. West St. Paul (Rural Municipality) (1951), [1952] 1 S.C.R. 292, [1951] 4 D.L.R.
110 Ontario Hydro v. Ontario (Labour Relations Board), [1993] 3 S.C.R. 327, 107 D.L.R. (4th) 457.
111 Citizens Insurance Co. of Canada v. Parsons (1881), 7 A.C. 96 at 113, 8 C.R.A.C. 406 [Parsons].
112 General Motors of Canada Ltd. v. City National Leasing, [1989] 1 S.C.R. 641, 68 O.R. (2d) 512
[City National Leasing cited to S.C.R.].
113 Kirkbi AG v. Ritvik Holdings, 2005 SCC 65, [2005] 3 S.C.R. 302, 259 D.L.R. (4th) 577 [Kirkbi].
609.
487
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
The industries that fall within provincial jurisdiction under this arrangement are
numerous, and include many of the industries that emit large amounts of carbon into
the atmosphere and are therefore good candidates for a cap-and-trade or intensity-
based trading regime, such as oil and gas, manufacturing, mining, forestry,
construction, and intraprovincial truck and bus lines. Moreover, the power of the
provincial legislatures to regulate the business activities of those industries has been
understood broadly by the courts. In particular, that power has been held to permit the
regulation of those activities for a range of different purposes: to protect consumers
from fraudulent dealings; to protect the health and safety of consumers; to establish
quality standards; to ensure adequate supply; and to protect the economic and other
interests of employees.114 It has also been held to permit their regulation for the
purpose of protecting the environment.115 There is every reason to believe, therefore,
that provincial legislation establishing a cap-and-trade or intensity-based trading
regime that is limited in its scope to such undertakings would be upheld as valid.
Would it be open to a provincial legislature to extend the reach of a cap-and-
trade or intensity-based trading regime to include industries that normally fall within
federal legislative jurisdiction, such as aeronautics, international/interprovincial truck
and bus lines, and nuclear power generation? Courts would likely analyze a
constitutional attack on a provincial cap-and-trade program that explicitly includes
one or more such industries by reference to the necessarily incidental doctrine. The
current understanding of that doctrine requires consideration of three distinct
questions: To what extent does the impugned part of the statutehere, the inclusion
in the list of industries to which the cap-and-trade regime applies of the federally
regulated industry in questionencroach on the legislative jurisdiction of the federal
order of government when that part is viewed in isolation? Is the rest of the statute
valid? Given the answer to the first question, is the impugned part sufficiently
integrated into the rest of the statute to profit from that overall validity and thus be
considered valid itself?116
The answer to the first of these questions would likely be that provincial
legislation imposing legally enforceable constraints on the carbon emissions of a
federally regulated industry is a very serious encroachment on federal legislative
jurisdiction over that industry. In fact, there is good reason to believe that the courts
would view such provincial action as an incursion into the core, or basic, minimum
and unassailable content,117 of federal legislative jurisdiction over the industry. That
114 The relevant jurisprudence is discussed in Hogg (Constitutional Law, supra note 82, c. 21,
especially at 21.5-21.13).
115 This interpretation of the power clearly seems implicit in the majority reasons of LaForest J. in
Oldman River, particularly in his reference to the s. 92(10) provincial power over local works and
undertakings (supra note 75 at 68).
116 This current understanding is based on the decision in City National Leasing, supra note 112.
117 That language comes from Beetz J.s reasons in Bell Canada v. Quebec (Commission de la sant
et de la scurit du travail), [1988] 1 S.C.R. 749 at 839, 51 D.L.R. (4th) 161 [Bell No. 2].
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
488
core has been defined in a series of cases dealing with the doctrine of
interjurisdictional immunity118 to include authority over labour relations and other
important aspects of the management and operation of companies doing business
within industries that fall under federal jurisdiction.119 That definition seems more
than broad enough to capture control of operational matters as important as
production processes.
It is also our view that, if the first question were to be answered in that manner,
the courts would hold that the inclusion of a federally regulated industry in the list of
industries to which the regime applied is unconstitutional. There are no cases
involving the necessarily incidental doctrine in which the Supreme Court of Canada,
in answering the first question, has found the degree of encroachment to be so great
as to extend to a core area of federal (or provincial) legislative jurisdiction. Yet it is
difficult to see how, if a court were presented with such a case, it could do anything
other than strike down the impugned part of the statute. If, as the doctrine of
interjurisdictional immunity requires, valid, generally worded provincial legislation
cannot constitutionally be applied in contexts in which such application would extend
the reach of that legislation into a core area of federal legislation,120 it cannot be open
to provincial legislatures to include such contexts in a list of contexts to which their
legislation is to apply. Regardless of how closely integrated into the rest of the (valid)
statute the impugned part might be, the fact that the impugned part encroached on a
core area of federal jurisdiction should render it invalid.
As noted above, the cap-and-trade regime proposed by the Legislature of British
Columbia may be integrated into a regionally defined cap-and-trade system that will
include at least one other Canadian province (Manitoba) and several of the states in
the western United States. Would the fact that the regime has this kind of regional
character render it constitutionally suspect in the eyes of the courts? We do not
believe that it would. While it is true, as noted above, that the regulation of
international and interprovincial trade falls within exclusive federal legislative
jurisdiction under subsection 91(2) of the Constitution Act, 1867,121 a regime of this
nature merely makes it possible for the undertakings governed by the British
Columbia statute to engage in the interprovincial and international trading of
emission allowances if they believe that it is in their interests to do so. The regime is
118 For a discussion of this constitutional doctrine, see generally Hogg, Constitutional Law, supra
note 82, c. 15(8).
119 See Commission du Salaire Minimum v. Bell Telephone Co. of Canada, [1966] S.C.R. 767, 59
D.L.R. (2d) 145; Bell No. 2, supra note 117.
120 For the Supreme Court of Canadas most recent applications of this doctrine, see Canadian
Western Bank v. Alberta, 2007 SCC 22, [2007] 2 S.C.R. 3, 409 A.R. 207; British Columbia (A.G.) v.
Lafarge Canada, 2007 SCC 23, [2007] 2 S.C.R. 86, 281 D.L.R. (4th) 54.
121 Parsons, supra note 111.
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
not directed at the regulation of such trading, as it would have to be in order to be
vulnerable to attack under subsection 91(2).122
489
3. Command-and-Control Regimes
The ability of provincial legislatures to regulate greenhouse gas emissions on the
basis of a command-and-control approach turns on the same considerations as their
ability to do so through the enactment of cap-and-trade or intensity-based trading
regimes. If the legislations reach is limited to industries that fall within provincial
legislative jurisdiction, it will likely be valid. If, by contrast, the legislation also
applies to industries that fall within federal jurisdiction, it will be vulnerable to attack,
at least insofar as its extension to those industries is concerned.
B. Federal Jurisdiction
In this part, we explore the question of whether it is open to Parliament to
regulate greenhouse gas emissions through the mechanisms of: (1) a carbon tax; (2) a
cap-and-trade or intensity-based trading regime; (3) a command-and-control regime;
and (4) the CEA Act.
1. A Carbon Tax
Unlike the provincial legislatures, Parliament has a very broad power to levy
taxes. Subsection 91(3) of the Constitution Act, 1867 authorizes it to legislate in
relation to The raising of Money by any Mode or System of Taxation. There is no
limit on the kinds of taxes Parliament can create under this grant of authority, nor is
there any territorial limit. The only requirements are that the legislation entail taxation
and that it raise money.
Although other bases of jurisdiction for a federal carbon tax have been
suggested,123 we are confident that federal legislation creating a carbon tax of the kind
described above would be upheld under subsection 91(3). Such legislation would
both entail taxation and raise money. If the tax were made revenue neutral, as was the
tax proposed by the Liberals, it would be open to opponents of the tax to challenge its
validity on the ground that it did not raise money. We do not think that such a
challenge would succeed, however, and for the same reasons we do not believe that a
provincially created revenue-neutral carbon tax would be vulnerable to attack on such
a ground.124
122 See Carnation Co. Ltd. v. Quebec (Agricultural Marketing Board), [1968] S.C.R. 238; Manitoba
(A.G.) v. Manitoba Egg and Poultry Association, [1971] S.C.R. 689, 19 D.L.R. (3d) 169.
123 See e.g. Chalifour, supra note 74 (arguing that the criminal law, trade and commerce, and POGG
powers are tenable bases for federal jurisdiction).
124 See Part II.A.1, above.
490
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 54
2. A Cap-and-Trade or Intensity-Based Trading Regime
In our analysis of the constitutionality of provincially created cap-and-
trade and/or intensity-based trading regimes, we argued that such regimes should pass
constitutional muster, provided they are limited in their scope to industries whose
business activities fall within provincial legislative jurisdiction, such as oil and gas,
mining, manufacturing, construction, forestry, and intraprovincial truck and bus lines.
It follows that a federal cap-and-trade or intensity-based trading regime would also
pass constitutional muster if it were limited in scope to industries whose business
activities fall within federal legislative jurisdiction, such as aeronautics, nuclear
power generation, and international/interprovincial truck and bus lines. The more
interesting and difficult question is whether a federal cap-and-trade or intensity-based
trading regime like the plan announced by the Conservative government in 2007,
which reached beyond those industries and brought provincially regulated industries
such as oil and gas, construction, and manufacturing into its regulatory fold, would
survive an attack on federalism grounds. It is to that question that the following
analysis is devoted.
In our view, the federal government could reasonably seek to justify such
legislation on one or more of the following bases: the criminal law power (subsection
91(27)); the national concern branch of POGG; and the national emergency branch of
POGG. Although others have made arguments that a cap-and-trade or emissions-
intensity program could be justified under the trade and commerce power (subsection
91(2)),125 we do not believe that these types of programs can be sustained under this
head of power. At bottom, it seems that the trade and commerce power is intended to
vest the federal government with jurisdiction over economic matters.126 This is
especially true with respect to the second branch of that head of power, the general
trade and commerce branch. The leading case on that branch, City National Leasing,
uses the word economic repeatedly.127 If City National Leasing were not limited to
economic cases, there would be little to distinguish the trade and commerce power
from the national concern branch of the POGG provision.128 While Professor Elgie
has argued that emissions trading serves an economic purpose by seeking to
concentrate compliance costs among those that can reduce emissions at the lowest
cost, there is no denying that both cap-and-trade programs and emissions-intensity
programs have an environmental objective as their core purpose.
125 See e.g. Elgie, supra note 92; Castrilli, supra note 73.
126 For a discussion of the jurisprudence relating to this branch of s. 91(2), see generally Hogg,
Constitutional Law, supra note 82, c. 20.2.
127 The test for the second branch can be found in City National Leasing (supra note 112 at 674,
676). The Court held in that case that the federal Combines Investigation Act (R.S.C. 1970, c. C-23)
satisfied that test and upheld its provisions on that ground. That test was recently affirmed by the
Court in Kirkbi (supra note 113 (upholding provisions of the federal Trade-marks Act (R.S.C. 1985, c.
T-13) on that ground)).
128 Barton, supra note 73 at 445.
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
We now consider three other defensible bases of federal jurisdiction.
491
a. Criminal Law
On the face of it, the highly regulatory character of a cap-and-trade or intensity-
based trading regime would appear to preclude it from being upheld as criminal law,
even if it contained the requisite offence-creating provisions. The Privy Council made
it clear that federal legislation will not qualify as criminal law merely because it
contains offence-creating provisions.129 Criminal law, the Privy Council told us, is
about prohibiting socially harmful conduct, not regulating it. This understanding of
the role of criminal law, and hence of the reach of subsection 91(27), came to be
reflected in the test that the Supreme Court of Canada eventually established for
subsection 91(27), which imposes three requirements: First, the legislation must be
prohibitory. Second, it must provide a penalty for those who violate the prohibition.
Third, it must have been enacted for a public purpose which can support it as being
in relation to criminal law, such as [p]ublic peace, order, security, health, [and]
morality.130
In Hydro-Qubec, the Court upheld the toxic substances provisions of the CEP
Act by a narrow five-to-four margin on the basis of subsection 91(27).131 It did so in
spite of the highly regulatory character of those provisions. Why the Court was
prepared to uphold the provisions on this basis is not entirely clear from the judgment
itself, but a number of reasons suggest themselves. One is the fact that, as noted by
Justice LaForest in his majority reasons, environmental protection does not lend itself
to the creation of broadly defined prohibitions. As he put it: Having regard to the
particular nature and requirements of effective environmental protection legislation, I
do not share my colleagues concern that the prohibition [against releasing a toxic
substance without a permit or in contravention of an interim ministerial order]
originates in a regulation.132 Another reason is the importance the majority attached
to protecting the environment. [S]tewardship of the environment was said to be a
major challenge of our time, an international problem, one that requires action by
governments at all levels, and a fundamental value of our society.133 And a third
reason is the fact that, by upholding the impugned provisions under subsection 91(27)
instead of under the national concern branch of POGG, the Court avoided assigning
the federal order of government exclusive jurisdiction over the release of toxic
129 See Board of Commerce Reference, supra note 82; Toronto Electric Commissioners v. Snider,
[1925] A.C. 396, [1925] 2 D.L.R. 5.
130 Margarine Reference, supra note 89 at 50.
131 Supra note 77. It should be noted that the 2005 Liberal Plan explicitly relied upon s. 91(27).
132 Ibid. at para. 147.
133 Ibid. at para. 127.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
492
substances into the environment.134 As Justice LaForest put it, the Constitution
should be so interpreted as to afford both levels of government ample means to
protect the environment while maintaining the general structure of the Constitution.
This is hardly consistent with an enthusiastic adoption of the national dimensions
doctrine.135
If these were the reasons the majority upheld the CEP Acts toxic substances
provisions under subsection 91(27) in Hydro-Qubec, then they could all be invoked
in support of upholding a federal cap-and-trade or intensity-based trading regime
under that head of power as well. But would they be viewed as strong enough reasons
to do so? We are doubtful. Although the problem of climate change is immediate and
extremely serious, the emission of CO2 does not have the same directness of harm as
the emission of more lethal substances, such as the polychlorinated biphenyls (PCBs)
at issue in Hydro-Qubec, which seem more worthy of criminal prohibition. Also, a
cap-and-trade or intensity-based trading regime permits companies to buy and sell the
right to cause the very environmental harm that the regime aims to control. Finally, it
would be difficult, as a matter of both logic and principle, for the courts to label a
federal cap-and-trade or intensity-based trading regime prohibitory (as they would
have to do to uphold it on the basis of subsection 91(27)) while simultaneously
labelling very similar provincial regimes as regulatory (which they would have to
do in order to uphold those regimes on the basis of subsections 92(5), 92(10), 92(13),
and 92A).
We are well aware that a number of scholarly comments support the use of the
criminal law head of power for greenhouse gas regulations.136 We are also aware that
the CEP Act provision currently used to regulate CO2 emissions is the same one that
was upheld in Hydro-Qubec.137 However, the fact that this provision was upheld in
the context of PCBs and other substances that are truly toxic (or, in Justice LaForests
134 It is the position of the Court that if federal legislation is upheld under the national concern
branch of POGG, the matter of that legislation is foreclosed to the provincial legislatures. See Crown
Zellerbach, supra note 77 at 433.
135 Hydro-Qubec, supra note 77 at para. 116.
136 See e.g. Peter W. Hogg, A Question of Parliamentary Power: Criminal Law and the Control of
Greenhouse Gas Emissions, Backgrounder, No. 114 (Toronto: C.D. Howe Institute, 2008); Bankes &
Lucas, supra note 73; Elgie, supra note 92; Chalifour, supra note 74; Barton, supra note 73.
137 At issue in Hydro-Qubec was s. 11 of the CEP Act. This provision was virtually identical to
what is now s. 64 of the CEP Act (supra note 10), which provides:
[A] substance is toxic if it is entering or may enter the environment in a quantity or
concentration or under conditions that
(a) have or may have an immediate or long-term harmful effect on the environment
or its biological diversity;
(b) constitute or may constitute a danger to the environment on which human life
depends; or
(c) constitute or may constitute a danger in Canada to human life or health.
493
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
words, poisonous138) does not mean that the federal government has a free hand to
bring within its reach any and all substances it considers to be harmful to the
environment. If it were open to the federal government to do that, there would be no
limit to Parliaments jurisdiction over the environment. By the same token, the
implication of upholding a cap-and-trade or intensity-based trading program under
subsection 91(27) is that there would be little if any practical significance left in the
requirement that federal legislation be prohibitory in character in order to qualify as
criminal law; and, as a consequence, there would be very little in the way of
meaningful limits on the scope of federal jurisdiction under subsection 91(27).
b. The National Concern Branch of POGG
Parliaments power to legislate for the Peace, Order and Good Government of
Canada is currently understood to have three distinct branches: (1) the national
emergency branch; (2) the national concern branch; and (3) the gap branch.139 The
gap branch of POGG captures matters over which the Parliament of Canada has
authority to legislate because they cannot plausibly be assigned to any of the
enumerated classes of subject in sections 91 to 95 of the Constitution Act, 1867.
There are in fact very few such matters, and they would certainly not involve the
protection of the environment. Our focus is therefore on the first two branches of
POGG.
judgment in the Local Prohibition Reference; in particular, in the following passage:
The origins of the national concern branch lie in Lord Watsons reasons for
[T]he exercise of legislative power by the Parliament of Canada, in regard to all
matters not enumerated in s. 91, ought to be strictly confined to such matters as
are unquestionably of Canadian interest and importance, and ought not to
trench upon provincial legislation with respect to any of the classes of subjects
enumerated in s. 92. To attach any other construction to the general power
which, in supplement of its enumerated powers, is conferred upon the
Parliament of Canada by s. 91, would, in their Lordships opinion, not only be
contrary to the intendment of the Act, but would practically destroy the
autonomy of the provinces.140
This passage makes it clear that, unlike the gap branch of POGG, the national
concern branch provides Parliament with the authority to legislate in relation to
matters that do have a connection with one or more of the classes of subjects
assigned to the provincial legislatures. The passage also makes it clear, however, that
in the view of the Privy Council, the courts should be loath to uphold such legislation.
Only in relation to such matters as are unquestionably of Canadian interest and
138 Hydro-Qubec, supra note 77 at para. 141.
139 This understanding is reflected in Professor Hoggs discussion of POGG (Constitutional Law,
supra note 82, c. 17).
140 Supra note 100 at 360-61.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
494
importance should they be willing to do so;141 otherwise, the interest in protecting
provincial autonomy should hold sway.
The current understanding of the national concern doctrine reflects a similar
reluctance to permit Parliament to make frequent use of this branch of POGG. This
understanding stipulates that
[Vol. 54
[f]or a matter to qualify as a matter of national concern … it must have a
singleness, distinctiveness and indivisibility that clearly distinguishes it from
matters of provincial concern and a scale of impact on provincial jurisdiction
that is reconcilable with the fundamental distribution of legislative power under
the Constitution.142
The current understanding also suggests that a relevant consideration in making such
an assessment is the effect on extra-provincial interests of a provincial failure to deal
effectively with the control or regulation of the intra-provincial aspects of the
matter.143 The implication of this suggestion is that only when the courts are satisfied
that such a provincial failure would have significant harmful effects on extra-
provincial interests should they be willing to hold that a matter is truly of national
concern.
Another important feature of the current understanding of the national concern
branch is that the consequence of the courts holding that a particular matter is a
matter of national concern, is to render that matter one within exclusive federal
jurisdiction. That matter, including its intra-provincial aspects, is removed in its
entirety from provincial legislative jurisdiction.144 This feature of the doctrine can
only add to the courts reluctance to use the national concern branch as a basis for
upholding federal legislation, particularly in relation to social and economic issues in
which the provinces can be said to have a strong and legitimate interest.145
Could the federal government successfully demonstrate that a cap-and-
trade or intensity-based trading regime of the kind we are considering should be
upheld under POGG on the basis that it dealt with a matter of national concern? The
answer to that question would depend at least in part on how the matter of such a
regime was formulated. Some formulations might serve, at least superficially, to
distinguish the subject matter of the regime from matters of provincial concern. But
there are clearly limits to how creative one can be in the drafting exercise. Bearing
these considerations in mind, we presume that the matter would be formulated in
141 Ibid.
142 Crown Zellerbach, supra note 77 at 432.
143 Ibid.
144 LeDain J., speaking on behalf of the majority in Crown Zellerbach, put this feature of the current
understanding in the following terms: where a matter falls within the national concern doctrine … as
distinct from the emergency doctrine, Parliament has an exclusive jurisdiction of a plenary nature to
legislate in relation to that matter, including its intra-provincial aspects (ibid. at 433).
145 For an excellent example of this occurring, see the majority reasons of LaForest J. in Hydro-
Qubec (supra note 77).
495
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
terms of something like protecting against the harmful effects of global warming by
reducing greenhouse gas emissions on the part of Canadian industry.
There are some factors that would support a finding that the regulation of
greenhouse gases by a cap-and-trade or intensity-based program constitutes a matter
of national concern.146 The fact that the federal legislation would have been enacted in
furtherance of Canadas obligations under the Kyoto Protocol, and would deal with a
matter of predominantly extra-provincial character and implications, would likely
count in favour of such a finding. The harmful extra-provincial effects that would
flow from the failure of provincial governments to regulate greenhouse gas emissions
effectively would also support that finding.
On the other side of the ledger is the fact that the matter of this kind of regime
could be said to lack the required singleness, distinctiveness and indivisibility that
clearly distinguishes it from matters of provincial concern.147 Courts could well find
that this matter is not single or indivisible at all, but simply a combination of a federal
matter (the regulation of greenhouse gas emissions by federally regulated
undertakings) and a provincial matter (the regulation of greenhouse gas emissions by
provincially regulated undertakings). Furthermore, the courts would almost certainly
view the scale of impact on provincial jurisdiction148 inherent in allowing
Parliament to enact a comprehensive cap-and-trade regime for the entire country as
extremely serious, particularly for provinces like Alberta. Finally, and perhaps most
importantly, there is the fact that, if the courts were to uphold such a regime on the
basis of the national concern doctrine, provincial legislatures would be precluded
from regulating greenhouse gas emissions produced by industries such as oil and gas,
manufacturing, and construction. Canadian courts would not look at all favourably
upon such a consequence. On balance, we do not believe that the national concern
branch of POGG would sustain a federal cap-and-trade or intensity-based program.
c. The National Emergency Branch of POGG
The national emergency branch of POGG has its origins in the judgments of Lord
Haldane, written in the early part of the twentieth century. Lord Haldane believed
strongly in the need to restrict the scope of federal legislative jurisdiction in order to
protect provincial autonomy, and that belief led him to construe the POGG power
even more narrowly than had Lord Watson in the Local Prohibition Reference.149 It
was Lord Haldanes position that Parliament could only make use of POGG in
exceptional circumstances such as war or famine, when the Dominion as a whole
was truly imperilled and legislative intervention by the federal order of government
146 Rolfe, supra note 73 at 351; Barton, supra note 73 at 431. Both authors conclude that the
national concern branch of POGG could sustain such a federal regime.
147 Crown Zellerbach, supra note 77 at 432.
148 Ibid.
149 Supra note 100.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
496
was required to save it from disaster.150 That position was rejected in subsequent
cases,151 but the notion that Parliament should be able to legislate in times of national
emergency has remained part of the law and has given rise to what we now refer to as
the national emergency branch of POGG.
The jurisprudence relating to this branch of POGG, in particular the decision of
the Supreme Court of Canada in the Anti-Inflation Reference,152 has generated a body
of doctrine on which the courts would be expected to rely if the federal government
sought to invoke the national emergency branch in support of a comprehensive cap-
and-trade or intensity-based trading regime. In our view, that body of doctrine can be
summarized as follows: (1) the federal government can rely on the emergency branch
both to respond to existing emergencies and to prevent new emergencies from
arising;153 (2) emergencies for this purpose are not limited to those identified by Lord
Haldane in his judgments, but can include economic emergencies such as a high rate
of inflation;154 (3) the courts should be loath to second-guess a decision by the federal
government that an emergency exists or is threatened, and need only be satisfied that
the government had a rational basis for making such a decision;155 (4) the emergency
branch can only be invoked to sustain legislation of temporary duration;156 (5) the
legislation should indicate, in a preamble or otherwise, that it has been enacted for the
purpose of dealing with at least a serious national condition,157 if not a national
emergency; and (6) unlike in the case of the national concern branch, upholding
federal legislation on the basis of the national emergency branch does not preclude
the provincial legislatures from legislating in their own ways to deal with the
emergency in question (assuming they can do so in a manner that respects the limits
on provincial legislative authority under section 92).158
In our view, there is reason to believe that courts applying this body of doctrine
could well uphold a comprehensive federal cap-and-trade regime under the
emergency branch of POGG. The fact that the jurisprudence permits Parliament to act
150 Board of Commerce Reference, supra note 82 at 200.
151 An important decision in this regard was that of the Privy Council in Ontario (A.G.) v. Canada
Temperance Federation, [1946] A.C. 193, [1946] 2 D.L.R. 1 (P.C.) [Temperance Federation cited to
A.C.].
152 Supra note 87.
153 Ibid. See also Temperance Federation, supra note 151.
154 Anti-Inflation Reference, ibid.
155 Ibid. See also Fort Frances Pulp and Paper v. Manitoba Free Press, [1923] A.C. 695, [1923] 3
D.L.R. 629 (P.C.).
156 Anti-Inflation Reference, ibid. See also Canada (A.G.) v. Ontario (A.G.), [1937] A.C. 355 (P.C.)
[Employment and Social Insurance Reference]; Temperance Federation, supra note 151.
157 This can be said to be implicit in the reasons for judgment of the majority in the Anti-Inflation
Reference, from which these words were taken (ibid. at 422). Beetz J., in his dissenting reasons, took
the position that the indication that Parliament was relying on the emergency branch in support of the
impugned legislation had to be unmistakable (at 463), a standard that in his view had not been met
in that case.
158 See LeDain J.s reasons for judgment in Crown Zellerbach (supra note 77 at 432).
497
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
in anticipation of a new emergency arising would serve federal interests in a very
direct way. Moreover, there seems little reason to doubt that an environmental
disaster of the kind that global climate change portends would be held to qualify as an
emergency for this purpose. The posture of judicial restraint that the doctrine calls for
in evaluating the need for legislative action would also serve federal interests well.
The requirement of temporary duration is one that can be met by careful drafting, as
can the need for appropriate signalling. Finally, the fact that upholding such a regime
on the basis of this branch would leave it open to the provincial legislatures to take
whatever steps they consider advisable to reduce greenhouse gas emissions would
make it a much more attractive option to the courts than the national concern branch.
As the reader will have noted, the previous paragraph referred only to a cap-and-
trade regime and not to an intensity-based trading regime. The omission of the latter
was deliberate. Even with a posture of judicial restraint, we think it unlikely that the
courts would consider an intensity-based trading regime, which would permit
greenhouse gas emissions to increase over time, to constitute a genuine attempt by
Parliament to respond to a pending national disaster. It is only the cap-and-trade
option that, in our view, could plausibly be defended on the basis of the national
emergency branch of POGG.
Any suggestion that the federal government was considering the use of the
emergency branch would undoubtedly result in strong opposition from the provincial
governments, who would portray such an initiative as a direct and profound assault
on their ability to devise and implement policies tailored to their respective
economies and populations. However, the federal government could reduce the sting
of that opposition by making it clear that it would only pursue such an initiative if the
provincial legislatures did not take what it considered to be strong enough action over
the course of a prescribed time period. The federal government could also draft its
legislation in such a way as to make its implementation contingent on that condition
being met.
3. A Command-and-Control Regime
Parliaments ability to enact a command-and-control regime under section 91
would turn on the same considerations as its ability to enact a cap-and-trade regime.
If the scope of a federal command-and-control regime were limited to industries
whose business activities fall within federal legislative jurisdiction, it would almost
certainly be valid. Its validity would only be open to attack if its reach extended into
the provincial sphereoil and gas, manufacturing, construction, and so on.
Such an extended command-and-control regime could plausibly be defended on
the same bases as an extended cap-and-trade or intensity-based trading regime:
subsection 91(27); the national concern branch of POGG; and the national emergency
branch of POGG. In our view, the analysis we provided above with respect to the viability
of the latter two sources of jurisdiction in the context of a federal cap-and-
trade or intensity-based trading regime are equally applicable in the context of a command-
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
498
and-control regime. Hence, we believe that such a regime would likely not be upheld
under the national concern branch of POGG, but might well be upheld under the
national emergency branch. Insofar as subsection 91(27) is concerned, our view is
that, while the criminal law power may not sustain a cap-and-trade or intensity-based
trading program, the more prescriptive nature of command-and-control regulation
renders it more likely to be upheld as a valid exercise of federal power. Unlike a cap-
and-trade or intensity-based trading regime, a command-and-control regime does not
permit the companies it governs to buy and sell the right to cause environmental
harm.
4. The Canadian Environmental Assessment Act
The final regulatory option to be considered is the use of the CEA Act to require
federal authorities to consider the greenhouse gas implications of new projects
governed by that statute before approving them. It will be recalled that this statute
calls for environmental assessments of projects that a federal authority is itself
proposing, that a federal authority intends to support financially, that involve the sale
or lease of federal lands, or that implicate an area of federal concern identified by
regulation.159 It will also be recalled that, in making such assessments, review panels
are required to consider any change that the project may cause in the
environment,160 with environment understood as encompassing air, including all
layers of the atmosphere.161 As we discuss below, the CEA Act presents an excellent
opportunity to utilize an existing statute and existing institutions to bring
consideration of greenhouse gas emissions into federal decision-making.
There is every reason to believe that, if challenged, this option would be upheld
as constitutionally valid. In Oldman River, the Supreme Court of Canada made it
clear that it is open to Parliament to require that the environmental implications of
projects that engage areas of federal concern be considered before these projects are
approved.162 The Court also held that, in assessing these implications, the reviewing
bodies are entitled to take all of the projects possible environmental effects into
account. In the course of his majority reasons for judgment in that case, Justice
LaForest considered the example of a project involving the construction of a new
interprovincial railway. In his view, a panel asked to assess the environmental
implications of such a project would be entitled to take into account the impact of the
new line on ecologically sensitive habitats such as wetlands and forests; potential
hazards to the health and safety of nearby communities if dangerous commodities
are to be carried on the line; and the possible economic benefit to those
communities through job creation and the multiplier effect that will have in the local
159 CEA Act, supra note 9, s. 5(1).
160 Ibid., s. 2(1), s.v. environmental effect, (a).
161 Ibid., s. 2(1), s.v. environment. See also ibid., s. 16(1)-(2) (outlining what review panels are
required to consider).
162 Supra note 75.
499
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
economy.163 In fact, he said, not permitting the panel to consider such matters
would lead to the most astonishing results, and it defies reason to assert that
Parliament is constitutionally barred from weighing the broad environmental
repercussions, including socio-economic concerns, when legislating with respect to
decisions of this nature.164
Was it important to Justice LaForests reasoning in this regard that interprovincial
railways are federal undertakings under paragraph 92(10)(a) and therefore, qua
undertakings, within exclusive federal jurisdiction? Would he have taken a more
restrained view of the permissible scope of a federally mandated environmental
assessment if the project in question had been one that fell prima facie within
provincial jurisdiction (such as the dam at issue in Oldman River), with the federal
interest being limited to the impact of that project on an area of federal jurisdiction
(such as the navigability of the Oldman River)? We do not believe so.165 Justice
LaForest did not draw any such distinction himself, as he could well have done given
the nature of the case before him. Moreover, he referred with approval to an
Australian case, Murphyores Incorporated Pty. Ltd. v. Commonwealth of Australia, in
which the High Court of Australia upheld the constitutionality of an inquiry, made
pursuant to Commonwealth legislation, that assessed the environmental impact of the
mining of particular substances by a company seeking permission to export those
substances, even though the mining activity was acknowledged to be predominantly
a state interest.166
We conclude, therefore, that it is open to the agencies of the federal government
to include greenhouse gas emissions in the list of environmental concerns to be
considered by panels constituted under the CEA Act.
C. Summary
In summary, then, it is our view that both orders of government have a relatively
broad array of options available to them under the constitution to deal with
greenhouse gas emissions. The provincial legislatures can levy a carbon tax on
consumers. They can also impose different kinds of regulatory regimes on the main
industrial emitters of greenhouse gases within their respective boundaries, provided
that those regimes are limited in their application to industries understood to fall
within provincial legislative jurisdiction. Parliament, too, can levy a carbon tax as
well as impose different kinds of regulatory regimes on industrial emitters. Its
authority to create such regimes is clearest if the regimes are limited in scope to
163 Ibid. at 66.
164 Ibid.
165 See Meinhard Doelle, The Federal Environmental Assessment Process: A Guide and Critique
(Markham, Ont.: LexisNexis, 2008) at 67-71. Contra Steven A. Kennett, Federal Environmental
Jurisdiction After Oldman, Case Comment, (1993) 38 McGill L.J. 180.
166 [1976] HCA 20, 136 C.L.R. 1.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
500
industries that are understood to fall within federal legislative jurisdiction. However,
it is possible, given the extent and nature of the global climate change problem, that
Parliament could regulate all industrial emitters using the national emergency branch
of POGG. While some have argued that Parliament could regulate greenhouse gas
emissions under the criminal law power, we have our doubts about this line of
argument. Finally, it is open to the federal government to use the provisions of the
CEA Act to assist in its efforts to control climate change.
Given the wide range of available options, the choices our governments make in
this area willor at least shouldbe based primarily on policy considerations. It is
to those considerations that we now turn.
III. The Policy Dimension
Apart from the constitutional considerations, there are sharp policy differences
that render some options considerably better than others. Effectiveness in reducing
greenhouse gas emissions is very much tied into the ease with which regulatory
options can be incorporated into Canadas regulatory infrastructure, and it is here that
the options diverge. A discussion of the main regulatory options outlined in Part I.A
follows.
A. The Canadian Environmental Assessment Act
In terms of climate change action, there is no lower-hanging fruit than the use of
the CEA Act. The environmental assessment process is common throughout the
world, and other countries have used it to challenge greenhouse gasemitting projects
or policies. In the United States, a number of cases have involved administrative
decisions with greenhouse gas implications.167 All of these environmental assessment
cases challenged agency findings that a project or policy would have no significant
impact and therefore that no environmental impact statement was required under
the National Environmental Policy Act of 1969.168 While the results have been mixed,
no court has questioned the appropriateness of a fairly detailed evaluation of the
greenhouse gas impacts of projects or administrative actions. In Australia, a similar
cluster of cases involving the development of coal mines, in which plaintiffs sought
to use the environmental assessment process to force consideration of greenhouse gas
emissions.169 In New Zealand, a number of cases have arisen involving the failure to
167 Los Angeles v. National Highway Traffic Safety Administration, 912 F.2d 478 (D.C. Cir. 1990),
overruled in part by Florida Audubon Society v. Bentsen, 94 F.3d 658 (D.C. Cir. 1996); Friends of the
Earth v. Mosbacher, 488 F. Supp.2d 889 (N.D. Cal. 2007); Border Power Plant Working Group v.
Department of Energy, 260 F. Supp.2d 997 (S.D. Cal. 2003); Natural Resources Defense Council v.
Abraham, 355 F.3d 179 (2nd Cir. 2004).
168 42 U.S.C. 4321 to 4370f.
169 Australian Conservation Foundation v. Latrobe City Council, [2004] VCAT 2029, 140
L.G.E.R.A. 100; Gray v. Minister for Planning, [2006] NSWLEC 720, 152 L.G.E.R.A. 258; Wildlife
501
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
consider net greenhouse gas effects in refusing applications for wind farms.170 The
environmental assessment statutes in all of these countries are very similar to
Canadas CEA Act.
The idea of using environmental assessment to consider greenhouse gas effects is
not new to Canada, but the practice has been spotty. A 2000 report sponsored by the
Canadian Environmental Assessment Agency (CEA Agency) concluded that [t]he
extent to which climate change was factored into each environmental assessment
(EA) varies considerably and that a gap exists between climate change science and
its application to the EA community.171
The CEA Agency, in collaboration with with provincial and territorial agencies,
has published a general guidance document for incorporating climate change into all
environmental assessments, not just those under the CEA Act.172 Perhaps its genesis as
a cross-jurisdictional and cooperative effort necessitates its modest and general scope:
the document states that it is intended as general guidance, to be considered at the
discretion of jurisdictions and regulatory authorities173 and that [t]he consideration
of climate change in environmental assessments is not intended to impose any
mitigation obligations over and above the obligations that will be imposed through
the implementation of the general climate change policies.174 In any case, guidance
documents published by the CEA Agency have received very little deference from
agencies or courts.175
As a formal matter, then, there is little in the way of procedural or substantive
requirements mandating the consideration of a projects greenhouse gas effects in the
environmental assessment process. This does not mean that greenhouse gas
considerations are never taken into account under the CEA Act. In a joint review
Preservation Society of Queensland v. Minister for the Environment and Heritage, [2006] FCA 736,
232 A.L.R. 510.
170 Genesis Power Ltd. v. Franklin District Council, [2005] NZRMA 541; Meridian Energy Ltd. v.
Wellington City Council (2007), No. W31/07 (Environment Ct.); Environmental Defense Society v.
Auckland Regional Council, [2002] 11 NZRMA 492; Environmental Defense Society v. Taranaki
Regional Council (2002), No. A184/02 (Environment Ct.).
171 Canadian Institute for Climate Studies, Climate Change and Environmental Assessment, Part 1:
Review of Climate Change Considerations in Selected Past Environmental Assessments by Rick J.
Lee (2000), online: Canadian Environmental Assessment Agency
172 Federal-Provincial-Territorial Committee on Climate Change and Environmental Assessment,
Incorporating Climate Change Considerations in Environmental Assessment: General Guidance for
Practitioners (2003), online: Canadian Environmental Assessment Agency
173 Ibid. at 1.
174 Ibid. at 2-3.
175 See e.g. Friends of the West Country Association v. Canada (Minister of Fisheries and Oceans)
(1999), [2000] 2 F.C. 263 at para. 22, 248 N.R. 25, Rothstein J. (I do not find the independent utility
principle or the portions of the Guide which may reflect the independent utility principle helpful for
the purpose of interpreting subsection 15(3) of the CEAA).
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
502
panel involving the National Energy Board (NEB), the panel undertook a very brief
discussion of the greenhouse gas effects of the construction of a gas pipeline in
British Columbia.176 The panel noted that the emissions were minor in comparison to
overall emissions on Vancouver Island and that [on] a global scale, any change in
climate or the environment caused by GHG emissions from the Project could not be
defined, measured or described.177 The panel also complained that at the present
time, there are no defined criteria to measure significance in relation to GHG when
considered in a environmental assessment … Had there been detailed policies or
regulations for targets in place, the Panel could have evaluated GHG emissions
against these.178 How, then, is a panel to meet its mandate, set out in sections 20 and
37 of the CEA Act, to determine whether the project is likely to cause significant
adverse environmental effects (SAEEs) as a result of greenhouse gas emissions?
In the absence of any federal or provincial guidance on how to evaluate the
environmental effects of greenhouse gas emissions, the panel considered the project
against the backdrop of federal and provincial initiatives to reduce greenhouse gases
and assessed whether the pipeline would prejudice the ability of Canada to meet its
Kyoto Protocol commitments. It concluded that it would not: [N]ew natural gas
pipeline and energy generation projects have been factored into the outlook. Because
such developments have been incorporated in the outlook, the GSX project should
not compromise Canadas ability to reach our Kyoto target.179 In other words, the
panel concluded that the pipeline was consistent with the then-Liberal federal
government plan for how Canada would meet its Kyoto Protocol commitment. The
panel evaluated the significance of the environmental effects not by any empirical
determination, but by evaluating whether the greenhouse gases were anticipated by
governmental greenhouse gas reduction plans.
Concerning another project, in Pembina Institute for Appropriate Development v.
Canada (A.G.), the Federal Court held that a joint review panel failed to adequately
address the environmental effects of the greenhouse gas emissions resulting from the
proposed Kearl Oil Sands Project, which would emit an average of 3.7 megatonnes of
CO2 equivalents every year over its five-year life, accounting for about 0.5 per cent of
Canadas annual emissions and 1.7 per cent of Albertas annual emissions.180 The
court held that the panel erred in not explain[ing] in a general way why the potential
environmental effects, either with or without the implementation of mitigation
measures, will be insignificant181 and in failing to provide a clear and cogent
176 Joint Review Panel Report: GSX Canada Pipeline Project (Calgary: National Energy Board,
2003), online: Canadian Environmental Assessment Agency
177 Ibid. at 57.
178 Ibid. at 58.
179 Letter From Environment Canada (February 2003) at 5, cited in ibid.
180 2008 FC 302, 323 F.T.R. 297, 80 Admin L.R. (4th) 74. The reference year used in these
calculations was 2002 (ibid. at para. 70).
181 Ibid. at para. 73.
503
is
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
articulation of the reasons behind the Panels conclusion.182 The court remitted the
matter back to the panel for the sole purpose of stating the bases for its conclusion
that the environmental impacts would be insignificant.
The panel responded, in an addendum, that it must give [Alberta Environment]s
endorsement of the target significant weight in its consideration of the adverse
environmental effects of the Project given [Alberta Environment]s role as the
provincial agency responsible for establishing, monitoring and enforcing emission
standards.183 Like the joint review panel that assessed the British Columbia gas
pipeline, the Kearl Oil Sands Joint Panel looked to regulatory programs in place and
decided that the project was in keeping with, or accounted for by, existing regulatory
programs, essentially deferring to governmental agencies that are apparently working
on the problem.
Given the legal void, these panels can be forgiven for struggling with the
determination of the significance of a large, project-specific increase in greenhouse
gas emissions. Making that determination by reference to a regulatory backdrop
seems like a reasonable alternative to throwing up ones hands and concluding that
the greenhouse gas emissions of any single project will have no significant effect in
the global context. However, joint review panels have no basis in law to rely on this
backdrop. Under the CEA Act, the critical determination is whether a project is likely
to cause significant adverse environmental effects.184 Such an inquiry must focus on
the environmental effects themselves, not on whether the project is in keeping with a
provincial or federal agencys grand plan for reducing greenhouse gas emissions. In
fact, environmental assessment is in part meant to act as a check on agency discretion,
bringing to light environmental information that would otherwise be embarrassing or
therefore paradoxical
to use
unfavourable
governmental policy as the reference point for determining what is an SAEE.
Lurking in the background is the much more difficult question of whether the
CEA Act, as currently constituted, can address climate change at all. If, as we argue,
the CEA Act does not permit a determination of environmental impact on the basis of
a projects consistency with legislation or with some governmental plan or policy,
then can the CEA Act do anything to address climate change? The obvious problem is
one that pervades every effort to address climate change: viewed on an incremental,
project-by-project basis, even large projects are insignificant in the context of global
greenhouse gas emissions. While the Kearl Oil Sands Project is unusually large in
to project development.185 It
182 Ibid. at para. 78.
183 Joint Panel Report, Kearl Oil Sands Project, Addendum to EUB Decision 2007-013, Additional
Rationale for the Joint Review Panels Conclusion on Air Emissions (6 May 2008), online: Canadian
Environmental Assessment Agency
184 Supra note 9, ss. 20(1), 37(1) (this standard is found in both sections).
185 Bradley C. Karkkainen, Toward a Smarter NEPA: Monitoring and Managing Governments
Environmental Performance (2002) 102 Colum. L. Rev. 903 at 904-905.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
504
terms of greenhouse gas emissions, it would have represented a mere 0.04 per cent of
the worlds CO2 emissions in 2006.186
The responsible answer to this more difficult question is, of course, that work
must commence immediately on curbing greenhouse gas emissions, even if that work
will not, by itself, make an immediate difference to climate change. The CEA Act,
having been in place for over a decade and having acquired a body of jurisprudence,
is an obvious mechanism for the regulation of greenhouse gases, at least in the
context of federal projects. But because the current CEA Act standard of SAEEs is not
useful in the climate change context, the terms of the CEA Act need amendment to
specifically incorporate climate change concerns. Either a legislative amendment
must adapt this phraseology to climate change, or, by regulation, the phrase must be
defined in terms of what is permitted in the way of greenhouse gas emissions.
A legislative solution would appear to be the cleanest approach to adapting the
CEA Act to climate change concerns. Companion sections paralleling sections 20 and
37 of the CEA Act might provide for a separate process evaluating a project
specifically for its greenhouse gas emissions. For example, a parallel section might
require, in lieu of an SAEE, that any project subject to the CEA Act be carbon
neutral187 or have undertaken reasonable efforts to mitigate greenhouse gas
emissions or conform to some other standard reasonably susceptible of review by a
panel. Standards such as these would shift the CEA Act regime into more of a
substantive statute, calling for substantive outcomes. However, this would be a
necessary consequence of providing agencies and reviewing courts with something
more helpful than the concept of an SAEE. The most important move would still be
to simply require some explicit consideration of greenhouse gas effects for projects
subject to the CEA Act and to provide some guidance for panels and agencies
reviewing projects in relation to such effects.
186 The worlds CO2 emissions totalled 8230 megatonnes in 2006 (Gregg Marland, Bob Andres &
Tom Boden, Global CO2 Emissions from Fossil-Fuel Burning, Cement Manufacture, and Gas
Flaring: 1751-2006, Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratories,
(2009), online: Carbon Dioxide Information Analysis Center
emissions (Pembina Institute for Appropriate Development v. Canada (A.G.), supra note 180). Thus,
the Kearl Oil Sands Projects emissions would only have represented about 0.04 per cent of the
worlds CO2 emissions in 2006.
187 Carbon neutrality is a frequently used term indicating that the greenhouse gas emissions of a
project or action will, by performing offsetting actions, mitigation actions, or both, ensure that the total
greenhouse gas emissions after the completion of a project are lower than before. Offsetting actions
might include planting trees to take up CO2 or capturing landfill gas (a powerful greenhouse gas) that
would not have been captured otherwise. Carbon neutral was the New Oxford American
Dictionarys word of the year in 2006 (Oxford University Press Blog, Carbon Neutral: Oxford Word
of the Year (13 November 2006), online: Oxford University Press
505
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
The problem with a legislative solution is that it may not come to pass for an
unacceptably long time. Since Canadian ratification of the Kyoto Protocol in 2002,
the federal government has failed miserably to enact greenhouse gas legislation. As
discussed above, federal politicians have demonstrated far more interest in tossing
climate change around as a political football than in any genuine effort to address the
climate change problem.
The simpler solution would be, then, to define by regulation SAEEs for federal
projects that involve greenhouse gases. By regulatory fiat, Environment Canada could
decree that any federal project that is not, say, carbon neutral, has an SAEE, and
hence must not be approved or must be justified in the circumstances in order to
proceed.188 The usual objection to such an administrative approachthat it can be
easily undoneseems less persuasive in light of the pressing need to address
greenhouse gas emissions sooner rather than later.
Although the CEA Act is a logical place to start in terms of engaging the federal
government in the regulation of greenhouse gases, it is important to recognize the
limitations of this approach. The CEA Act can only address new projects and does
nothing to bring existing sources of greenhouse gas emissions under control. With
Canada needing a 25 per cent reduction from current emissions to meet its Kyoto
targets, holding firm on the status quo is insufficient. Adapting the CEA Act to include
project review of the greenhouse gas implications is an important part, but only one
part, of a Canadian response to the climate change problem.
B. Cap-and-Trade vs. Intensity-Based Emissions Trading
The cap-and-trade idea is one of the major administrative reforms in the last three
decades, taking most pollution decisions out of the domain of government policy and
placing them into the hands of emitters. The most notable and successful cap-and-
trade program to date has been the U.S. sulphur dioxide (SO2) emissions trading
plan,189 under which most of the fossil fuel-fired electricity generating plants in the
United States were allocated a certain number of allowances and required to have an
allowance for each ton190 of SO2 emitted. The allocation of permits is based on an
historical baseline (a string of years in the 1980s) and is set at a total lower than that
baseline so that some overall emissions reduction is achieved. In its initial phase, the
188 CEA Act, supra note 9, ss. 20(1)(b), 37(1)(b). Significance is not defined in the CEA Act
regulations or guidelines. While all of the provinces have environmental assessment procedures, only
Nova Scotia defines the word significant (Lawrence Environmental, Significance in Environmental
Assessment (2000)at 7, online: Canadian Environmental Assessment Agency
Regulations, N.S. Reg. 26/95, s. 2(1)(l)(i).
189 Clean Air Act, supra note 12, 7651a-7651o (1990).
190 We employ the word ton to indicate the U.S. short ton.
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
506
program imposed a somewhat hard nationwide cap of 8.90 tons of SO2 per year,191
while in subsequent years more facilities have been included and the cap raised
slightly.192 The U.S. SO2 cap-and-trade program reduced SO2 emissions nationwide
from over twenty-one million tons in 1994 to under fifteen million in 2006.193
Partly as a result of the perceived success of the SO2 program, emissions trading
has gained worldwide acceptance as a way to reduce global greenhouse gas
emissions. The Kyoto Protocol explicitly endorses emissions trading, not only
permitting individual countries to achieve their national targets by emissions trading,
but also encouraging trading by and between countries.194 The European Union has
committed itself, in addition to the commitment of its member states, to an emissions
reduction of 8 per cent below its 1990 levels195 and has undertaken an EU-wide
emissions trading program to achieve it.196
While cap-and-trade programs minimize industry-wide compliance costs, they
still impose them. Some proposals have sought to soften the economic blow further
by allocating allowances that are keyed to productivity. As explained in Part I.A,
above, these intensity-based trading programs essentially divide the absolute amount
of emissions by some denominator that has to do with the quantity or value of the
product produced. Greenhouse gas emissions intensity from electricity generation, for
example, would be measured in terms of tonnes of CO2 per kilowatt-hour produced,
so that if more efficient combustion techniques were discovered, boosting the amount
of electricity produced, then the amount of allowed CO2 emissions could be
increased.
The problem with intensity-based programs is that there is no credible way of
knowing what actual greenhouse gas emissions will ultimately be, or even that there
will be any reduction at all. If, for a particular emitter, production efficiency
improvements outpace the rate at which emissions intensity targets tighten, then that
emitter will have a pool of surplus allowances available, which it can sell to other
emitters, relieving them of the need to reduce emissions. A facility that doubles
191 Clean Air Act, supra note 12, 7651b(a)(1) (1990). Special legislative dispensations, however,
have pushed the real cap upwards. See text accompanying note 202.
192 Ibid., 7651b-7651d (1990).
193 U.S. Environmental Protection Agency, Latest Findings on National Air Quality: Status and
Trends Through 2006 (North Carolina: U.S. Environmental Protection Agency, 2008) at 23, online:
U.S. Environmental Protection Agency
194 Kyoto Protocol, supra note 3, art. 6.
195 Ibid. at Annex B.
196 Congressional Research Service, Climate Change: The EU Emissions Trading Scheme (ETS)
Enters Kyoto Compliance Phase by Larry Parker (11 February 2008), online: National Center for
Science Education
Trust, The EU Emission Trading Scheme, online: Carbon Trust
507
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
production and meets a 20 per cent greenhouse intensity reduction target can still emit
60 per cent more than it had originally.197
Intensity-based programs are also economically inefficient for reasons having
nothing to do with environmental effects. The award of emissions allowances on the
basis of productive output amounts to a distortionary output subsidy.198 An output
subsidy creates economic inefficiency by encouraging overproduction, directing
resources that might be used for other valuable goods into production of the
subsidized good. While every industry has an incentive to innovate to increase profit
margins, an intensity-based program creates an added incentive in the form of an
extra source of wealth from productive efficiencies: the award of extra allowances.
Apart from the environmental effects of this distortion, it creates a disadvantage for
other industries.
Given equal initial conditions, it is safe to say that intensity-based emissions
trading is both economically and environmentally inferior to cap-and-trade programs.
C. Carbon Taxation vs. Cap-and-Trade
A more serious policy debate involves a comparison between a carbon cap-and-
trade program and a carbon tax. From a policy perspective, cap-and-trade programs
supposedly create some certainty about the quantity of emissions allowed, while
taxation programs provide some certainty with respect to the price of emissions.
Some environmentalists have therefore called for a cap-and-trade program, on the
reasoning that it is important to control the quantity of greenhouse gas emissions
rather than worry about the cost.199 However, this is a superficial reason to favour
quantity controls over price controls; any quantity can be achieved by price
197 If the intensity target is a 20 per cent reduction, then a facility producing 100 units of output that
doubles its production but improves its efficiency by 20 per cent (emitting only 80 per cent of the
emissions per unit of output) can still emit greenhouse gas emissions equal to 160 units (80 per cent of
200).
198 An output subsidy is a payment keyed to production, so that an extra incentive is provided to
produce the subsidized good. This is distortionary because it draws resources into production of the
subsidized good in excess of what market signals would otherwise call for. For example, subsidizing
production of all kinds of agricultural commodities has provided inexpensive food for consumers, but
it is likely that some agricultural land would have been put to better use. For a discussion of how the
output-based allocation of emissions allowances amounts to an output subsidy, see Carolyn Fischer,
Rebating Environmental Policy Revenues: Output-Based Allocations and Tradable Performance
Standards (Resources for the Future, Discussion Paper 01-22, 2001), online: Resources for the Future
199 See e.g. E&ETV, Climate: Pews Claussen Compares Cap-and-Trade with Carbon Tax
Approaches for Emissions Reduction (On Point, 07/16/2007) (16 July 2007) (quite honestly, Id
rather put my money on the market, which is what a cap and trade does, because there the market sets
the price. The government doesnt set the price) [transcript on file with author].
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
508
mechanisms simply by setting the tax at an appropriate level.200 There are, however, a
number of important differences that separate the two types of programs. The
majority of these differences should give federal and provincial governments cause to
favour a carbon tax program.
First, implementation problems have plagued cap-and-trade programs. As noted
above, while a carbon tax is levied upon any sales transaction with a paper trail, a
cap-and-trade system requires some design. A cap-and-trade program requires a
determination of the level of the cap, which entities are subject to the program, and
above all, how the emissions allowances are to be initially allocated. All of these are
fraught with political peril. Moreover, a cap-and-trade system can only apply to
certain emittersthose that have the resources to monitor their emissions and can
buy and sell emissions allowances,201 but that are small enough in number for
regulatory agency to monitor their compliance.
How allowances would be allocated is a thorny implementation issue. The
traditional and most familiar approach is to give allowances away for free, based on
some historical baseline of emissions, as was done in the SO2 program. The baseline
calculation for what became a complicated formula was to grant fossil fuel-fired
power plants emissions allowances equal to roughly half of the plants average
emissions over a five-year period from 1980 to 1984. But arriving at this rule required
extensive negotiations and was a sobering exercise in rent-seeking. Paragraph
404(a)(3) of the U.S. Clean Air Act provides that utilities in Indiana, Ohio, and
Illinois would receive a special clump of two hundred thousand allowances for the
years 19951999, to be split in proportion to their baseline emissions.202 One would
be hard pressed to find a more naked example of raw political power.
One way around this initial allocation problem is to allocate allowances by
auction, which does away with quarrels over historical baseline rules. Auctioning
allowances also provides significant economic benefits in that the revenues could be
recycled and used to reduce other taxes. The problem with auctioning allowances is
one of political economy. To the extent that allowances are given away for free by
law, lawmakers writing cap-and-trade legislation are essentially printing money for
distribution to appreciative constituents; hence, the inevitable but inelegant marriage
of rent-seekers and lawmakers.
200 In 1974, economist Martin Weitzman showed in a seminal work that it is only the uncertainty
and steepness of the marginal pollution abatement curve that make either a quantity-control scheme
(such as a cap-and-trade program) more or less economically efficient than a price-control scheme
(such as a Pigouvian taxation program): Martin L. Weitzman, Prices vs. Quantities (1974) 41
Review of Economic Studies 477.
201 Jack Mintz & Nancy Olewiler, A Simple Approach for Bettering the Environment and the
Economy: Restructuring the Federal Fuel Excise Tax (University of Ottawa: Sustainable Property
Initiative, 2008).
202 Supra note 12, 7651c(a)(3).
509
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
By ignoring historical emissions entirely, a carbon tax avoids having to deal with
the self-serving appeals in favour of one baseline rule or another. A carbon tax is not
free of political peril, but is harder to finagle. By definition, a carbon tax would have
to be applied to carbon-containing fuels meant for combustion. Any carve-out from a
universal rule would be conspicuously peculiar. For example, trucking industries,
which would be hard-hit by a carbon tax designed to reduce gasoline usage, could
conceivably lobby for an exemption for diesel fuel, but how politically saleable
would such a special dispensation be? Would commuters paying more for gasoline
tolerate such a dispensation? And then why not provide an exemption for the shipping
industry? The slippery slope problems inherent in granting exemptions would make it
more difficult to grant any. By contrast, the ways in which cap-and-trade allowances
have been distributed are not necessarily obvious or free of controversy.
Another implementation issue pertains to the question of whether and how to
incorporate offsets, a way for emitters to generate additional allowances by
undertaking projects that supposedly reduce emissions from some baseline or
business-as-usual path. The Kyoto Protocol has been vulnerable to this form of rent-
seeking. The clean development mechanism, by which a developed country may
finance a low-carbon project in a developing country and in so doing collect credits
towards meeting its Kyoto targets,203 has led to illusory emissions reductions. Far
from achieving any greenhouse gas reductions, the program has mostly been a
boondoggle, subsidizing projects in developing countries that are only undertaken
because of the Clean Development Mechanism program.204 The problem with offsets
is that it is difficult for a certifying authority to ascertain whether the business-as-
usual path is a genuine one or an ingeniously concocted story. For example, a
proposal to generate offsets by lengthening rotations may or may not produce
emissions reductions, as tree rotations may be extended for any number of economic,
regulatory, or ecological reasons. Granting credits under such circumstances is
gratuitous and frustrates emissions reduction objectives.
Perhaps the more salient question is how politically acceptable a carbon tax
would be. The political reality is that the very mention of the word tax in the same
sentence as carbon evokes emotional reactions. One study found that people were
more positively inclined towards a program requiring a payment than one that
involved a tax, even if the programs were substantively identical.205 While the word
203 Kyoto Protocol, supra note 3, art. 12.
204 Michael Wara, Is the Global Carbon Market Working?, Commentary, (2007) 445 Nature 595;
Christa Marshall, Carbon Offsets: The Shoppers Guide to Reducing Carbon Footprints
ClimateWire (11 September 2008) [on file with authors].
205 Edward J. McCaffery & Jonathan Baron, The Political Psychology of Redistribution (USC
CLEO Research Paper C05-4, 15 March 2005) at 18-19, online: Social Science Research Network
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
510
tax is always loaded in North America generally,206 in Alberta the word stirs up
deep-seated resentment stemming from the Lougheed-Trudeau clash in the 1980s.207
Moreover, the carbon tax meets with political resistance because it is thought to
be regressive.208 Gasoline taxes, for example, impose higher transportation costs that
take up a larger proportion of a poor drivers paycheque than that of a rich driver, so
the thinking goes, such that an increase would deprive poorer drivers of more basic
goods than rich drivers.209
This line of thinking, however, seems to be based more on selective anecdote
than on empirical analysis.210 Moreover, the question of whether a carbon tax is
regressive or not is more complicated than is typically presented in public discussion.
Is a carbon tax regressive if the lowest quintile of households is hurt more than the
second-lowest quintile, but the second-lowest quintile is hurt less than the richest
quintile? How many income classifications are needed for analysis? Is elasticity to be
taken into account?211 Do we think about regressiveness in terms of a present
snapshot in time or do we think about the lifetime income or consumption of
206 Edward J. McCaffery & Jonathan Baron, Heuristics and Biases in Thinking about Tax in
David Merriman, ed., Proceedings of the Ninety-Sixth Annual Conference on Taxation, Chicago,
Illinois, November 13-15, 2003 (Washington, D.C.: National Tax Association, 2004) 434.
207 See e.g. Susan Blackman et al., The Evolution of Federal/Provincial Relations in Natural
Resources Management (1994) 32 Alta. L. Rev. 511.
208 Sarah E. West & Roberton C. Williams III, Estimates from a Consumer Demand System:
Implications for the Incidence of Environmental Taxes (2004) 47 Journal of Environmental
Economics and Management 535 at 535 (Most studies suggest that environmental taxes tend to be at
least mildly regressive, making such taxes less attractive options for policy); Shi-Ling Hsu, Carbon
Tax Heuristics and Politics: The Case of the Gasoline Tax (15 April 2008), online: Social Science
Research Network
209 James M. Poterba, Is the Gasoline Tax Regressive? (1991) 5 Tax Policy and the Economy 145
at 145-46 (there is a long-standing view that excise taxes such as the gasoline tax are regressive,
imposing a heavier burden on low-income households than on their higher-income counterparts).
210 The New York Times ran a series of articles on the impact of high gasoline prices on various
individuals throughout the country, highlighting the hardships imposed upon cabdrivers (Compared
to a year ago, I pay $15 more a day in gas, said Miguel Gonzalez, 67, of Queens. I only take home
$100 a day, so thats my lunch and dinner right there), immigrants (Lesly Richardson, 50, a Haitian
immigrant from Brooklyn, nodded in agreement. Thats $100 a week, he said. Thats your grocery
bill), and single mothers (Cindy Wright spoke of the pain high gas prices cause the single mothers
who make up many of the clients at the public health clinic in Torrington, where she is a nurse): As
Gas Prices Go Up, Impact Trickles Down The New York Times (30 April 2006) A24.
211 Regressivity could be measured by different delineations of income, and using a large variety of
different assumptions about how drivers respond. The most careful study of the projected incidence of
a gas tax increase was done by Sarah E. West and Roberton C. Williams III. They estimated separate
demand models for each of five income quintiles and found that under the most severe and simplistic
assumptionsthat gasoline is perfectly inelastic and that people make no adjustments whatsoever to
changes in the price of gasolinethe incidence on the poorest quintiles is not substantially different
from that of the next two higher quintiles (West & Williams, supra note 208 at 551, Table 3).
511
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
individuals?212 It is simplistic to flatly pronounce, as NDP leader Jack Layton has, that
a carbon tax would hurt the poor.213
One way to blunt these critiques is to return or recycle carbon tax revenues in
such a way as to reduce the economic pain of those having to pay the tax or to
redistribute income to the poor.214 As economists generally consider income and sales
taxes to be distortionary,215 proposals to reduce environmental harm by taxation have
the potential bonus of reducing distortionary taxes and increasing social welfare.216 Or
the revenues could even be recycled back to emitters forced to pay the tax, as Sweden
has done with a tax on emissions of nitrogen oxides.217 These revenue recycling ideas
often serve to dull the political sharp edges of taxation proposals.
An important consideration to bear in mind, however, is that the point of a carbon
tax would be to decrease consumption of carbon-emitting activities such that tax
proceeds would eventually decline. The carbon tax should not be oversold, then, as
both an effective and economically painless way to reduce emissions. What carbon
tax proceeds could provide is some temporary aid for the various transitional costs
associated with making the kinds of structural societal changes required to reduce
greenhouse gas emissions.
212 Kevin A. Hassett, Aparna Mathur & Gilbert E. Metcalf, The Incidence of a U.S. Carbon Tax: A
Lifetime and Regional Analysis (2007) National Bureau of Economic Research, Working Paper
13554 [on file with author].
213 Joanna Smith, Carbon Tax Would Hurt Poor, NDP Says Toronto Star (23 May 2008), online:
Toronto Star
214 It should be noted that if emissions allowances are auctioned, cap-and-trade programs can also
raise revenues.
215 See e.g. Lawrence H. Goulder et al., The Cost-Effectiveness of Alternative Instruments for
Environmental Protection in a Second-Best Setting (1999) 72 Journal of Public Economics 329; Ian
Parry, Roberton C. Williams III & Lawrence H. Goulder, When Can Carbon Abatement Policies
Increase Welfare? The Fundamental Role of Distorted Factor Markets (1999) 37 Journal of
Environmental Economics and Management 52; Ian Parry & Wallace E. Oates, Policy Analysis in
the Presence of Distorting Taxes (2000) 19 Journal of Policy Analysis and Management 603.
216 This economic effect, popularly known as the double dividend, is the subject of debate. It has
been argued that environmental taxes increase the cost of goods, such that reducing distortionary
income taxes may not offset the excess burden of the environmental tax. See Lawrence H. Goulder,
Effects of Carbon Taxes in an Economy with Prior Tax Distortions: An Intertemporal General
Equilibrium Analysis (1995) 29 Journal of Environmental Economics and Management 271.
However, it has also been argued that this fails to account for the fact that the income tax system, by
allowing deductions, creates distortions by favoring certain kinds of spending. Thus, if environmental
taxes can reduce income taxes, it can also reduce these distortions: Ian W.H. Parry & Antonio M.
Bento, Tax Deductions, Environmental Policy, and the Double Dividend Hypothesis (2000) 39
Journal of Environmental Economics and Management 67.
217 The nitrogen oxide tax in Sweden is levied upon energy producers but rebated to them in
proportion to energy output (International Institute for Sustainable Development, The Nitrogen Oxide
Charge on Energy Production in Sweden, online: IISD
This would, however, convert the tax into a distortionary output subsidy (Fischer, supra note 198).
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
512
The more direct response to these political and psychological objections is for
policymakers to emphasize the hidden costs and administrative headaches of poorly
designed cap-and-trade programs. It is worth repeating that a cap-and-trade program
and a carbon tax should work exactly the same way in economic theory, but the cap-
and-trade program entails some more difficult design issues, such as determining who
is covered by the program. Cap-and-trade programs may appear to be less expensive
because costs are somehow hidden from view. For carbon taxes, recycled revenues
can be directed towards transitional relief to temporarily assist with capital
expenditures that help with adjustment into a lower carbon-emitting economy. The
most honest and effective aspect of a carbon tax is that it will induce the kind of
widespread changes to consumption patterns that are needed to reduce emissions.
In environmental instrument choice, pollution taxation has played the role of
bad cop to cap-and-trades good cop because taxation always appears to be more
costly than cap-and-trade programs. The obvious but obscured reality is, however,
that environmental progress will always have coststhe question is who bears the
costs, not whether they are borne at all. Among economists, there is a growing
consensus that a carbon tax is a superior means of addressing greenhouse gas
emissions.218 The economic virtue and political downfall of taxation programs is that
they generally present the costs in an open and transparent fashion, while cap-and-
trade programs, if implemented by issuing free, grandfathered allowances, hide
them.
A carbon tax is clearly the most economically and environmentally effective
option to address climate change. The implementation advantages of administering
what is essentially another sales tax over the regulatory infrastructure that would be
needed to design and administer a cap-and-trade program are compelling. While
political shenanigans have saddled cap-and-trade programs with special allocation
perks that frustrate emissions reduction objectives, a federal carbon tax would be
more difficult to sabotage. Because a federal carbon tax would typically be levied on
a transaction like a sales tax, it would require a bit more audacity to write some
blatant giveaway into legislation to insulate or exempt certain industries or
individuals. Taxes are by their nature more universal: they come with a presumption
that everyone pays them. Moreover, even if federal and provincial cap-and-trade
218 Economists Favor Fossil Fuels Tax to Spur AlternativesSurvey, E&E News PM, Feb. 8, 2007.
Nobel Laureate Economist Joseph Stiglitz, a former chief economic advisor to President Bill Clinton,
called for a global carbon tax in 2006 (Joseph E. Stiglitz, A New Agenda for Global Warming,
online: (2006) 3:7 Economists Voice 3
to President George W. Bush, did as well (N. Gregory Mankiw, Raise the Gas Tax The Wall Street
Journal
Journal
(20 October 2006), online: Wall Street
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
programs were to coexist, the practical complexities of such a situation could
undermine the effectiveness of either or perhaps both programs.219
513
D. Command-and-Control Regulation
Because greenhouse gases are a by-product of such a wide variety of activities, a
regulation of the command-and-control type would likely be complex, might take one
of many different forms, and might draw on a wide variety of technologies. For coal
combustion, industry standards might refer to carbon capture and storage technology
or coal gasification,220 or any number of technologies and processes that have come
along in the drive to save coal combustion from obsolescence in a carbon-constrained
world. For natural gas exploration, command-and-control regulation might mandate
techniques to limit flaring, the wasteful initial burning off of natural gas before the
gas stream can be harnessed. For other combustion and industrial processes, a variety
of other technologies and techniques may be possible. Command-and-control
regulation in the context of greenhouse gas regulation would thus be a mandate to
install some emissions-reduction technology or adopt some emissions-reducing
practices, most likely ones that are ascertained by looking at industry practices or
perhaps common industry ideas. It would be impossible to cover all Canadian
greenhouse gas emitters, as there are thousands of smaller emitters that are too
numerous to identify and regulate.
That said, a small number of credible voices have called for command-and-
control type regulation of greenhouse gases simply because immediate and dramatic
governmental action is required. From a policy perspective, there may be
considerable advantage in a blunt but broad instrument, one that might achieve some
deep reductions very soon even if it comes at a high compliance cost. While
economics might theoretically favour cap-and-trade or carbon taxation programs, the
practicalities and politics of such programs may cause a delay that humankind may
not be able to afford. The advantage of the traditional command-and-control type of
regulation is that administrative agencies in developed countries such as Canada
already know how to carry it out. With the kind of market signals that politicians have
recently talked about implementinga modest forty dollars per tonne in the case of
the 2008 Liberal Party proposal for a federal carbon taxlarge-scale structural and
219 Bankes & Lucas, supra note 73.
220 Carbon capture and storage typically involve separating CO2 from other gases in the emissions
process, compressing it to a high density, and then storing it underground or beneath the ocean to
isolate it from the atmosphere (International Panel on Climate Change, IPCC Special Report: Carbon
Dioxide Capture and Storage, Technical Summary by Edward Rubin et al., online: IPCC
[Vol. 54
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
514
cultural changes may not take place in time.221 While a price on carbon is a necessary
condition to greenhouse gas reduction, it may not be a sufficient one.222
Prominent economists such as Jeffrey Sachs, the director of the Earth Institute at
Columbia University, have thus argued for large-scale governmental intervention into
the many technological possibilities that could make a major and near-term difference
in reducing greenhouse gases. For example, carbon capture and storage technology,
which would capture CO2 at its point of emission and pipe and store it underground
without allowing its escape into the atmosphere,223 would require a substantial
amount of government-sponsored research, the construction of pipelines that cross
property boundaries and jurisdictions, and the monitoring of storage facilities to
ensure the CO2 actually stays underground.224 Development and maturation of this
technology is not possible without substantial governmental involvement. It has also
been argued that climate technologies need such widespread and rapid deployment
that uniformity of technology is required to coordinate their worldwide adoption.225 In
light of the difficulty of inducing developing countries to undertake emissions
reductions, agreement upon a single way of doing things may facilitate a fairly large-
scale change in relatively short order.
All of the considerations that favour a command-and-control response are global
in nature and only implicate Canada as one of many developed countries that could
lead by example. Like the United States, however, Canada has some uniquely
favourable conditions for undertaking large, government-supported projects that
could produce global command-and-control strategies: a huge (too huge)
infrastructure for the mining, transport, and combustion of coal; a vast (yet not vast
enough) network of pipelines that could be utilized for CO2 transport; and oil and gas
exploration ventures that might benefit from a means of enhanced recovery using
CO2 as a gaseous pump to extract more oil or gas.226 One pilot project involves the
piping of CO2 captured from a plant in North Dakota to an oil field in Saskatchewan
to increase production from the oil field.227 While private efforts such as these are
encouraging, the widespread and rapid adoption of these efforts will require
221 Jeffrey D. Sachs, Technological Keys to Climate Protection Scientific American (18 March
2008), online: Scientific American
James Cowan, Green Shifts Final Reach Still Unknown National Post (4 July 2008), online:
National Post
222 Sachs, ibid.
223 International Panel on Climate Change, supra note 220.
224 Sachs, supra note 221.
225 See Scott Barrett, Environment and Statecraft: The Strategy of Environmental Treaty-Making
(New York: Oxford University Press, 2003) at 261-62, 395.
226 International Panel on Climate Change, supra note 220 at 19, 21, 23, 36-37.
227 IEA Greenhouse Gas R&D Programme, CO2 Capture and Storage, Weyburn Enhanced Oil
Recovery Project, online: CO2 Capture and Storage
S.L. HSU AND R. ELLIOT REGULATING GREENHOUSE GASES
2009]
considerable governmental involvement that might usefully take the form of
command-and-control regulation.
515
Conclusion
Both constitutional and policy considerations favour two instruments for
reducing Canadian greenhouse gases: a carbon tax and the CEA Act. As we have
argued, it is clear that both the federal and provincial governments have the authority
to impose a carbon tax, and the federal authority to consider greenhouse gases under
the CEA Act is also quite solid. In addition, both the carbon tax and the environmental
assessment process enjoy the constitutional and political advantage of leaving
provincial initiatives alone. Under taxation schemes, the federal and provincial
governments are free to establish and pursue their greenhouse gas objectives without
interference from one another.
In contrast, a comprehensive federal cap-and-trade system might survive
constitutional scrutiny but would raise issues about its relationship with provincial
trading programs. Command-and-control regulation, thought to be less economically
efficient than cap-and-trade programs or carbon taxes, may nevertheless play a role in
greenhouse gas regulation since it would stand a better chance of surviving
constitutional scrutiny than would a cap-and-trade program. The Alberta government
and the federal government have both shown increased interest in carbon capture and
storage, but this strategy gives rise to enormous potential for conflict over who will
be required to capture carbon, who will store it, and where. That two levels of
government should independently pursue separate programs requiring such a great
deal of coordination is folly. Finally, the greenhouse gas intensity-based system that
the federal government is currently pursuing poses both constitutional and policy
problems.
The policy advantages of a carbon tax and of the CEA Act are quite strong. Both
draw on existing regulatory infrastructures. In the case of the carbon tax, little
additional monitoring and enforcement capability is required, as taxation at a
transaction point is something that revenue agencies throughout Canada already do
quite effectively. And while the existing CEA Act currently does a poor job of
addressing greenhouse gas considerations, relatively simple amendments by
regulation or legislation would suffice to patch its shortcomings. By contrast, there
are some fairly serious policy issues that would need to be dealt with before either a
federal cap-and-trade or command-and-control system could be put in place, and
more still with intensity-based emissions trading.
The politics of greenhouse gas regulation are changing rapidly, more quickly than
federal politicians realize. The familiar old economic doomsayers have lost
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
516
credibility. With even oil sands interests coming out in favour of a carbon tax,228 it
appears that Canadians are more willing to absorb economic pain than federal
politicians, in their pocketbook pandering, have expected. Sometimes the simplest of
solutions are the most elusive to grasp. Yet Canadians and the world would benefit
greatly if federal politicians could summon up the modest courage and foresight
needed to implement a sensible greenhouse gas reduction strategy by taking advantage
of the two most promising policy instruments: a carbon tax and the CEA Act.
[Vol. 54
228 Matthew Burrows, Petro-Giants Will Accept a Carbon Tax (7 December 2007), online:
Straight.com
Jaremko & Jason Markusoff, Oilsands Backs Carbon Tax; All Polluters Must Help Pay for Carbon-
Capture Scheme: Industry Edmonton Journal (11 March 2008), online: Tar Sands Watch
