On February 23, 2012, a European Union (EU) drafting committee voted on a draft law that discriminates against bitumen. This was the Draft Implementing Measure to the European Union Fuel Quality Directive (Implementing Measure). The Canadian press reported the vote ended in a stalemate. The press also noted that the law would be reconsidered in the late spring or early summer (National Post; CBC; Globe and Mail).
If the EU enacts the law it will have made a step in its fight against climate change, but the market for bitumen may be negatively impacted. If the law dies, Canada can expect a higher price on the sale of its bitumen in overseas markets. This note examines the February 23 vote within the EU law making process.
To put the vote in context, it is important to understand the nature of bitumen, life-cycle emissions, EU climate change policy and the substance of the draft law. This note also includes a sketch of whether or not a trade law challenge by Canada would be successful.
In short, the result of the vote is a procedural victory for Canada’s stated policy of protecting bitumen markets. The result of the vote makes it possible for the European Council, a small body relative to the European Parliament and therefore more easily lobbied, to block enactment of the law without providing a justification. If the law is enacted, Canada cannot be assured of defeating it under international trade law.
Bitumen is a precursor to synthetic crude oil (SCO). Bitumen is sometimes said to be a form of crude oil, but it cannot be sold as such without upgrading. Industry reports account for bitumen separately from conventional crude oil. Accordingly, it is more accurate to describe bitumen as something from which we can make crude oil. Using 1.1 to 1.2 barrels of bitumen, additional energy and hydrogen (both commonly acquired from natural gas), a person can make one barrel of SCO.
It takes more energy to produce SCO from bitumen than to produce conventional crude oil (Murphy and Hall, 2010). Producers obtain the extra energy from fossil fuels, the burning of which results in GHG emissions. In short, bitumen production and upgrading it into a barrel of SCO emits more GHG emissions than are emitted as the result of exploration and production of a conventional barrel of crude oil.
Life-cycle GHG Emissions
The life-cycle GHG emissions of a fuel are its total emissions from the time of extraction to end use (typically combustion). The life-cycle of a fuel has two parts, upstream and downstream. The upstream includes all the steps it takes to get a hydrocarbon out of the ground, into a refinable form, and to the refinery gate. The downstream includes all the steps from refining through end use.
The vast majority of GHG emissions associated with gasoline and diesel come from end use, regardless of whether the fuel originates from bitumen or conventional crude oil. On average, the upstream GHG emissions from SCO production are approximately 400 percent higher than those for conventional crude oil (see, for example, Brandt, 2011). However, if one considers both the upstream and downstream emissions (i.e. the entire life-cycle), then the GHG emissions from bitumen derived gasoline and diesel are approximately 10 to 30 percent higher than for the same fuels derived from conventional crude oil.
EU Climate Change Policy
The EU committed to reducing its emissions of GHGs to eight percent below a 1990 baseline in order to combat anthropogenic climate change. It made this commitment in the 1997 Kyoto Protocol to the 1992 United Nations Framework Convention on Climate Change (UNFCCC). In 2007, the EU unilaterally committed to cut its emissions by at least 20 percent of 1990 levels by 2020. It also offered to increase its emissions reduction to 30 percent by 2020, on condition that other major emitting countries in the developed and developing worlds commit to future emissions reductions under a global climate agreement. The EU approach to meeting this commitment is to develop a separate strategy for each sector of the economy (e.g. see here), including one for the transportation sector.
The European Fuel Quality Directive and its Implementing Measure
Part of the strategy for the transportation sector is to get more energy out of a fuel for the same GHG emissions. The European Fuel Quality Directive (FQD) is directed at this goal. It is comprised of a series of directives governing a wide range of fuel characteristics (Directives 98/70/EC, 1999/32/EC and 2009/30/EC). Directive 2009/30/EC introduced a requirement to improve energy efficiency of transport fuels.
Directives legally bind each Member State of the EU as to the result to be achieved, but leave it up to them to determine the means (see Consolidated Version of the Treaty on the Functioning of the European Union (TFEU), Article 288). The EU has the authority to make laws governing the EU’s internal market, environment, energy and transport (TFEU, Article 4(2)(a), (e), (g), and (i)).
Article 7a(2) of the FQD requires a supplier of transport fuels to the EU to reduce the life-cycle GHG emissions of the basket of fuels it supplies by six percent from a 2010 baseline by the year 2020. To ensure the required emissions reductions are made, one needs a calculation for the baseline and a method of determining the life-cycle GHG emissions of each transport fuel.
One can calculate the life-cycle GHG emissions of each batch of fuel or one can assign default values to a fuel type according to the feedstock from which the fuel is made. After considering both options (see here), the committee responsible for fleshing out the details of the FQD chose default values. Using default values is less precise than batch-by-batch reporting but more administratively efficient.
The default value approach is reflected in the draft Implementing Measure to the FQD. The Implementing Measure quantifies the 2010 baseline and lists default values for end products (e.g. gasoline and diesel) by the feedstock from which they are made (e.g. conventional crude oil, shale oil, coal to liquids, natural gas, waste plastics, and bitumen). The metric employed for both the baseline and the fuel values is emissions per unit of energy contained in a fuel, in the form of grams of carbon dioxide equivalent per mega-joule of energy (abbreviated as g CO2eq / MJ).
The 2010 baseline is 88.3 g of CO2eq/MJ. The baseline represents a weighted average of fuels supplied to the EU in 2007 (2007 is the most recent year the EU reported fuel consumption to the Intergovernmental Panel on Climate Change.). The FQD requires a supplier (or group of suppliers) to beat the average baseline of 88.3 g by six percent by 2020; this makes the 2020 target 83 g CO2eq / MJ. The Implementing Measure assigns gasoline and diesel the following emissions values: 87.5 and 89.1 g CO2eq / MJ, if refined from conventional crude oil; 107 and 108.5 g CO2eq / MJ, if derived from bitumen; 131.3 and 133.7 g CO2eq / MJ, if refined from shale oil; and 172 g CO2eq / MJ, if derived from coal. (These are not all the emissions values, but they indicate the range.)
On its face the Implementing Measure is origin-neutral. The text does not distinguish one feedstock from another by country of origin, and all supply, whether originating in the EU or outside, is subject to the same rules.
The February 23 Vote
On February 23, 2012, the drafter of the Implementing Measure, the Committee on Fuel Quality (CFQ), said it had ‘no opinion’ as to whether the measure should become law. What does this mean? In short, this means the measure goes to Council and Parliament, either one of which can block it from becoming law, subject to the Commission’s power to propose an amended version. The rest of this section broadly describes the role of each of Commission, Council, Parliament, and the CFQ in the process of making and deciding on the Implementing Measure.
Law making in the EU is geared towards advancing EU policy and harmonizing the laws of Member States, while providing Member States with a sphere of autonomy and their citizens with democratic oversight of law making. To this end, the Commission, Council and Parliament play separate but related roles.
The Commission sets EU policy by issuing proposals for the creation or amendment of laws. The Commission is comprised of 27 appointed commissioners, one from each of the EU’s 27 Member States. The commissioners are required to advance EU policy independent of other interests (Article 245 of the TFEU and Article 17 of the Consolidated Version of the Treaty Establishing the European Union (TEU).
The Council represents the interests of governments of Member States, and must approve a Commission proposal before it becomes law. The Council is made up of one ministerial level official from each Member State. A Member State’s choice of official may vary based on the law being considered. For example, if the subject matter of a proposal is financial, a Member State may send its finance minister, if environmental, its environment minister. A Council member’s primary role is to advance the interests of his or her Member State (Article 16(2) of the TEU).
Parliament represents a combination of citizen and Member State interests. Under the ordinary legislative procedure, Parliament must approve a proposal before it becomes law (Article 294 of the TFEU). Parliament is made up of no more than 750 members elected from throughout the 27 Members States (736 at present). Representation is distributed on a degressively proportional basis (this is proportional representation modified to give smaller regions greater representation) (Article 14(2) of the TEU). One of Parliament’s main purposes is to exercise democratic oversight over the Commission and Council.
The Commission, Council or Parliament often choose to delegate law making power to a committee for the sake of efficiency. Such a committee is typically a creature of the Commission. The draft law the committee writes is said to be a proposal, or proposed measure, of the Commission. The CFQ is one such committee.
The EU practice of having some of its laws developed by committee is called comitology. This term also refers to the degree of oversight the committee is subject to, and the process by which its measures are adopted or rejected.
The historical context for comitology clarifies the procedure governing the CFQ. Historically, much of EU law was centrally developed by the Commission and Council, with Parliament on the sidelines. A series of EU constitutional amendments increased the role for Parliament, creating and expanding what is known today as the ordinary legislative procedure. In simplified terms, under this procedure the Commission makes a legislative proposal to both Council and Parliament. These two bodies have an opportunity to amend the proposal, and both must approve it before it becomes law. This is sometimes referred to as the co-decision procedure, as law making requires a decision by both Council and Parliament.
A law enacted under this procedure may contain a provision that delegates future law making power to a committee. This is the case with the FQD. On a proposal from the Commission, the Council and Parliament approved Directive 1999/30/EC, which introduced the 2020 Target into the FQD. The directive contained provisions to establish the CFQ and to delegate to the CFQ the power to determine the baseline and GHG life-cycle emissions values for fuels (Articles 7a (5) and 11(4)).
A committee such as the CFQ is a creature of the Commission. As noted, the Commission’s mandate is to advance EU interests. At times these interests might not fully align with those of all Member State governments (represented by Council) or a broad cross-section of EU citizens (represented by Parliament). In such cases, law making by a committee joined at the hip to the Commission may be controversial. The EU has amended the comitology procedure from time to time to address this.
Comitology procedure was first defined in a single document in 1987 (First Comitology Decision). The First Comitology Decision provided a choice of three types of procedure to govern a committee: an advisory procedure, under which the Commission could enact its own proposal regardless of the opinion of the committee, and without consulting Council or Parliament; a management procedure, under which a committee could decide whether to submit the Commission’s proposal to Council, which Council could then block; and, a regulatory procedure, under which if the committee either forwarded the Commission’s proposal to Council or took no action, the proposal went to Council, which could then block it. Under each of these procedures, the role of Parliament was limited. Parliament was advised of a law’s substance but could not block its passage, except on limited jurisdictional grounds.
In the 1990s and early 2000s conflict over financial and environmental regulation lead Parliament to call for increased oversight of law making by comitology committees. In 1999, the First Comitology Decision was replaced by a second (Decision 1999/468/EC). The Second Comitology Decision was amended in 2006. The 2006 amendment added a fourth choice of procedure, regulatory procedure with scrutiny, which gave Parliament greater oversight over laws developed by a committee and a veto power (Decision 2006/512/EC).
Under this procedure, if a committee votes in favour of its measure, the Commission must send the measure to both Council and Parliament. If a committee does not vote in favour of the measure, then the Commission is required to make a proposal to Parliament and Council on what to do next. Regardless of whether Parliament or Council receive the measure or the Commission’s proposal on a measure, either one of them may block the measure. (Parliament’s power to block a measure is still nominally bound, but it is in effect a veto power.) If either one blocks the measure it will not become law in the rejected form. The Commission retains the power to make an alternative proposal.
The ‘regulatory procedure with scrutiny’ is defined in Article 5a of the Second Comitology Decision. The Second Comitology Decision has been repealed but Article 5a survives in that it continues to apply to committees that were formed before the repeal (Regulation (EU) No 182/2011, Article 12).
The CFQ is subject to regulatory procedure with scrutiny (Articles 7a (5) and 11(4) of Directive 1999/30/EC). The Second Comitology Decision refers to the Commission submitting a draft measure to a comitology committee for an opinion (Article 5a(2)), but in practice the committee drafts the measure, subject to the law that gave the committee its power, and then votes to determine its opinion on the measure by a deadline set by the Commission. In effect, the committee votes on its own work. However, since its work is to implement a Commission policy, the opinion resulting from the vote is said to be on a proposal of the Commission.
The vote is a qualified majority vote. The CFQ has one member from each of the EU’s 27 Member States. The vote of each committee member is weighted according to a degressively proportional system. According to the weighting there are 355 votes. To affirm the Implementing Measure, at least 14 committee members, in aggregate holding 255 votes, had to be in favour. In addition, any committee member could demand that the countries in favour represented at least 62 percent of the population of the EU (Article 5a (2) of the Second Comitology Decision, and Title II of the Protocol on Transitional Provisions).
The CFQ voted on the Implementing Measure on February 23, 2012. The result was 89 for, 128 against, with 128 abstentions (Comitology Register), based on 12 countries voting in favour, eight against, with seven abstentions. How each country voted is not a matter of public record. This may appear to be a vote against the Implementing Measure. However, of the CFQ’s available choices, unfavourable, favourable or no opinion, it chose no opinion.
The importance of this vote is that it affects the ease with which the Implementing Measure can be blocked.
If the CFQ had given a favourable opinion to the Implementing Measure then the Commission would have sent the measure directly to both Parliament and Council. While either body could then have blocked the measure (Council by a qualified majority or Parliament by a majority), it would have had to justify doing so on one or more listed grounds. These include that the substance of the measure: is outside the drafting authority of the committee; violates a principle called proportionality, i.e. that the measure is not for a legitimate purpose, or is not suitable for its purpose, or is not necessary or reasonable in light of alternative means or competing interests; or, violates the subsidiarity principle, i.e. that the action proposed is more suitably implemented by a smaller political unit (Article 5a(3) of the Second Comitology Decision).
Since the CFQ took a no opinion position, the Commission must create a proposal on what to do next and send it to both Parliament and Council. The Commission may also forward the measure, and the record indicates it has. Parliament may still block the measure, and if so, it must provide a justification on a listed ground.
So far, this is the same as if the CFQ had given a favourable opinion. The substantive difference is that under this branch of the procedure, Council may block the measure without providing a justification (Article 5a(4) of the Second Comitology Decision).
This may seem like a small difference, and it may in fact be. On the other hand, it may be significant for Canada. Canada is against the Implementing Measure (see below). Given the no opinion result of the CFQ vote it may be easier for Canada to convince a qualified majority of Council members to act against the measure, since under this mode of proceeding Council is not restricted by the requirement to justify a rejection on a listed ground.
If there is broad political support for combatting climate change among the governments of Member States or their citizens, then action against the Implementing Measure by Council or Parliament may be unlikely. The compromise position for each of these bodies is to do nothing. By doing nothing, they do not act against Canada’s expressed interests and the Implementing Measure still becomes law; if Council takes no action within two months of receiving the measure, the measure passes to Parliament, and if Parliament takes no action within four months of the date the Commission forwarded the measure to Council, then the measure is deemed adopted (Article 5a(4) of the Second Comitology Decision).
Even if Council or Parliament takes action against the Implementing Measure it may survive in another form. The Commission has the right to submit a modified proposal or an alternative legislative solution after Council or Parliament rejects one of its proposals (Article 5a(4) (c), (f) of the Second Comitology Decision).
In short, the Implementing Measure remains alive. Council and Parliament may adopt it, and the do-nothing course results in the measure becoming law as a directive. If Parliament or Council opposes it, the Commission may put forward an amended version. Should it survive in its current form, Canada may consider a challenge under international trade law.
Canada’s Trade Challenge in the event the Implementing Measure Becomes Law
Canada’s Minister of Natural Resources, the Honourable Joe Oliver, has written to the Commission stating should the Implementing Measure become law Canada may challenge its implementation as a violation of international trade law (see here).
Trade law is complex and discussion of the claims and defences that could be made in a contest over the Implementing Measure would take another post (at least). That said, here is a sketch of the relevant arguments.
Canada and the EU are both members of the World Trade Organization (WTO). Annexed to the agreement establishing the WTO are agreements intended to liberalize trade in goods and services. The General Agreement on Tariffs and Trade (GATT) governs trade in goods. The GATT obligates parties not to discriminate between goods of two foreign jurisdictions or between those of a home and foreign jurisdiction provided the compared goods are ‘like’ each other (Articles I and III -the most favoured nation and national treatment obligations).
A difficulty for Canada is that the GATT permits discrimination, in at least two cases. Discrimination is permissible if the compared goods are not alike, and discrimination is permissible if it is both for a purpose listed in the GATT and is implemented in a manner that does not have an invalid effect according to WTO law (Article XX). Recognized purposes include measures: “necessary to protect human, animal or plant life or health,” (Article XX(b)); and, measures “relating to the conservation of exhaustible natural resources” provided “such measures are made effective in conjunction with restrictions on domestic production or consumption” (Article XX(g)).
To win a trade challenge to the Implementing Measure at the WTO Canada would have to show that its product is like an EU product. It would then be up to the EU to show that the discrimination is for a valid purpose and that its implementation does not have an invalid effect.
Whether Canada can show likeness depends on what is being compared.
If the comparison is between a barrel of bitumen and an average barrel of EU crude oil supply, then a finding of likeness seems unlikely. The leading WTO case on likeness is the 1996 Appellate Body decision in Japan – Alcoholic Beverages II. In it the Appellate Body affirmed the use of four factors to consider when analyzing likeness: consumer tastes and habits; the product’s end uses in a given market; the product’s properties, nature and quality; and, its tariff classification. No one factor is determinative and the weight each one is given is determined case-by-case. While there is no consumer purchasing data regarding bitumen, the other three factors point to a finding that under WTO law bitumen and conventional crude oil are different.
The two products have different end uses. Conventional crude oil can be processed by a refinery, while bitumen cannot. Bitumen must first be upgraded into SCO. This process requires inputs other than bitumen. After upgrading, the bitumen ceases to exist. In result, bitumen has a different end use than conventional crude oil. The two products have different properties. They differ significantly in appearance, in state at the same temperature, and in molecular composition. Finally, conventional crude oil and bitumen are labelled as different products under the most widely used tariff classification system. Under the Harmonized Commodity Description and Coding System of the World Customs Organization (HS Codes) the codes assigned to crude oil and bitumen are 2709.00 and 2714.10.
If the comparison is between physically similar products, like SCO and conventional crude oil, or identical end products, such as gasoline refined from SCO and gasoline refined from conventional crude oil, the outcome of a likeness analysis is less clear. An unresolved issue in WTO law is whether two end products can be found to be ‘un-like’ where: (1) the end products are physically identical (or effectively identical), but (2) are made in different ways (referred to as their processes and production methods or PPMs. A PPM that fits with the two listed conditions is said to be an unincorporated PPM).
Regardless of whether Canada can show likeness, the EU may be able to show that its discrimination is permissible. The WTO Appellate Body has previously found a law intended to limit emissions from transport fuels has a valid purpose, under Article XX(g) (US-Reformulated Gasoline. The question then becomes whether the measure has an invalid effect. WTO adjudicators tend to approach this question holistically, inquiring generally into the merit of the measure’s goal, the good will of the implementing jurisdiction, including its consideration of the interests of its trading partners, and whether the method of implementation is even-handed.
If matters get this far, the EU would likely point to combatting climate change as a compelling and internationally recognized objective, the fact that the Implementing Measure does not discriminate based on country of origin (on its face, the measure discriminates between products but not by where they come from), and the fact the measure applies to domestic feedstock producers (the measure applies to a range of feedstocks, including oil shale, which is produced in the EU. The EU also has bitumen deposits, though they are small compared to Canada’s). The EU may also point to its previous attempts to address the same problem, climate change, through diplomacy and treaty making by pointing to its assent to the United Nations Framework on Climate Change and the Kyoto Protocol.
If a WTO adjudicative body finds a measure impermissibly breaches GATT, then the drafter has an opportunity to amend it into compliance. For example, if Canada can successfully argue the measure is discriminatory in effect, perhaps by showing its SCO is like at least some conventional crude with comparable life-cycle emissions entering the EU, and that SCO is treated less favourably than such crude, then the EU may amend the Implementing Measure to re-classify that crude, or change the valuation of carbon intensity from set values to batch-by-batch. A victory by Canada at the WTO might be pyrrhic if the EU could preserve the effect of the measure through amendment.
The Implementing Measure or a variant of it will be a live issue so long as the Commission stands behind it. An initial Canadian challenge at the WTO may result in defeat of the measure, but the effect could be short lived; the EU has scope to amend an impugned Implementing Measure into compliance with WTO law.
Preparation of this note was supported in part by a grant from Carbon Management Canada.