The three cases – Canfor Corporation v. USA, Tembec Inc v. USA and Terminal Forest Products Ltd v. USA – were all brought by Canadian producers of softwood lumber that had been affected by the larger context of trade disputes between the US and Canadian lumber industries.

Following petitions from US lumber producers, the US Department of Commerce and the US International Trade Commission took various decisions in 2001 and 2002 imposing both anti-dumping duties and countervailing duties on imports of softwood lumber from Canada to the US. The three claimants sought to challenge the imposition of these duties under NAFTA Chapter 11, as measures affecting their investments in the United States. They also challenged an amendment to a US law that had the effect of passing the duties collected by US authorities on to the affected US lumber producers.

In parallel with the claims under NAFTA Chapter 11, Canada and the US engaged in the binational panel process under NAFTA Chapter 19, dealing specifically with disputes relating to antidumping and countervailing duties. The three Chapter 11 claimants were entitled to participate in the Chapter 19 process. Finally, Canada also pursued a claim against the US at the World Trade Organisation.

Two separate tribunals were constituted, and a third claim was commenced

Canfor filed its notice of arbitration on July 9, 2002, seeking $250 million (US) in compensation. Canfor named Frank McKenna, a former Canadian politician, to the tribunal, and the US appointed Conrad Harper. Shortly after this appointment, the US challenged Mr. McKenna, on the grounds that he had given a speech to a Canadian government council describing the US measures at issue as ‘harassment’ and questioning the measures’ validity.

The appointing authority under NAFTA, the ICSID Secretary-General, informed Mr. McKenna that it planned to uphold the challenge, and Mr. McKenna resigned in April 2003. Canfor then appointed Joseph Weiler, and the parties agreed on Emmanuel Gaillard as chair in July 2003.

Washington DC was set as legal seat of Canfor arbitration, due to location of dispute subject-matter

In its only substantive decision, dated January 23, 2004, the Canfor tribunal determined the legal seat of the arbitration.

Although unable to agree on a specific seat, the parties had agreed that the seat must be in either Canada or the US, putting aside earlier discussions of a possible seat in Switzerland, the UK or Mexico. The claimant favoured either Vancouver or Toronto, while the US favoured Washington DC.

The tribunal drew guidance from a set of non-binding factors in the UNCITRAL Notes on Organizing Arbitral Proceedings, together with an additional factor of neutrality on which both parties had presented submissions.

On neutrality, Canfor saw Washington DC as the least neutral option, being the capital city of the respondent and the place where the disputed measures were taken. Meanwhile, the US contended that any Canadian seat would hardly be neutral, given the importance of the lumber industry to Canada (and British Columbia and Ontario in particular). In support of Washington DC, the US suggested that hearings could be held at the ICSID offices to preserve neutrality.

In the tribunal’s view, given the limitation imposed by the parties’ agreement of a seat in either Canada or the US, a neutral seat was simply not possible. The tribunal disagreed that holding hearings at ICSID would increase the neutrality of the legal seat, since US law would still govern a Washington-seated arbitration and since Canada was (then) not an ICSID member state.

Thus, the neutrality factor did not assist here, the tribunal said.

Turning to the UNCITRAL factors, the tribunal also found no assistance from considering the suitability of arbitration law in each venue, since both were equally suitable. The tribunal added that it did not share the US concern that Canadian law was unsuitable given that Canadian courts had partly set aside the Metalclad NAFTA award in 2001 (see here).

The factors of enforcement, convenience and cost were also non-determinative, the tribunal said, since both countries were New York Convention states (affecting enforcement), and the arbitral hearings could be held in any location regardless of the formal legal seat (affecting convenience and cost).

The last UNCITRAL factor, however, was more useful. The tribunal noted that the subject-matter of the dispute was primarily measures taken by the US affecting the claimant’s alleged investments in the US. Given this, and the fact that all the other factors were equal, the tribunal held that Washington DC was the most appropriate seat.

Tribunal bifurcated jurisdictional objections and ordered production of NAFTA negotiating history

In the same decision, the Canfor tribunal also determined (though without particularly extensive reasoning) that the proceedings would be bifurcated into a preliminary jurisdictional phase and a possible later merits phase. The tribunal dismissed the US request to reserve the possibility of submitting further jurisdictional objections later, reasoning that the respondent should be in a position to raise all relevant jurisdictional objections at that stage.

While some of those objections might be joined to the merits, the tribunal said, their full details should be submitted in the preliminary phase. That phase was at least envisaged to hear the US objection relating to the claims’ overlap with the binational panel process in NAFTA Chapter 19.

Lastly, in a May 2004 procedural order, the tribunal ordered the US to produce any NAFTA negotiating history that it possessed on the relevant NAFTA clauses, including informal notes shared between the negotiating parties as well as the more formal travaux preparatoires.

US nominee to Canfor tribunal resigned following realisation of potential conflict

Following December 2004 hearings on jurisdiction in the Canfor case, the US nominee Mr. Harper recalled to the parties that (as he had earlier disclosed) he was a director of the Harvard Corporation, and added that he had recently realised that Harvard University was at that time being sued by the United States government in a contractual dispute. While the US played down any concerns over this potential conflict, Canfor asked Mr. Harper to resign. Mr. Harper agreed, and resigned from the tribunal in March 2005.

US requested consolidation of the three cases, while claimants resisted

Realising that the jurisdictional phase of the Canfor case would likely need to be repeated following a new US nomination to the tribunal, soon after Mr Harper’s resignation, the US petitioned the ICSID Secretary-General to establish a consolidation tribunal under NAFTA Article 1126.

In support of its petition, the US observed that it had filed (or planned to file) the same jurisdictional objections in all three cases, which relied on the same facts and brought the same claims against the US. As well, the US noted, Canfor and Terminal shared the same counsel. Now that the Canfor case had essentially returned to the beginning following Mr. Harper’s resignation, the US said, all three cases were at the same procedural stage, making it more efficient to hear them together, also thereby avoiding the risk of inconsistent decisions.

Meanwhile, the three claimants opposed consolidation of their claims, partly contending that the US petition was effectively an objection to the jurisdiction of the separate tribunals, and that (at least in the Canfor case) it was too late to bring this objection. The claimants also argued that any consolidation tribunal appointed would suffer from ethical problems, in that the arbitrators would have a strong incentive to accept jurisdiction in order to be paid higher fees. As well, the claimants observed, none of the members of a consolidation tribunal would be appointed by the claimants, infringing party consent in arbitration.

Lastly, the claimants noted that they were competitors, and foresaw problems in dealing with confidential business information of each claimant in a consolidated proceeding where all filings would be shared with all parties.

Consolidation tribunal began by dismissing concerns over consent and ethics

Subsequently, in its decision on consolidation dated September 7, 2005, the tribunal began by reviewing the negotiating history of the NAFTA consolidation clause, noting a quote from one negotiator that the text ‘does not resolve all the questions that may occur during consolidation’.

The clause had primarily been inserted to avoid ‘procedural harassment’ of states, the tribunal found, even if it had been expanded later in the negotiations to permit not only states but also investors to request consolidation.

The tribunal then dismissed the claimants’ general concerns with consolidation. On the question of party consent, the tribunal noted simply that the NAFTA states had agreed to include such consolidation procedures in their treaty, which permitted consolidation even without claimants’ direct consent. The claimants must have been aware of this possibility when they commenced arbitration under the NAFTA process, the tribunal added.

As for the alleged ethical problems, the tribunal commented that it was ‘not uncommon in arbitration’ that arbitrators would decide to uphold jurisdiction, thereby continuing a case and, unavoidably, generating higher fees for themselves. This was not an instance of having a ‘financial interest in … the outcome of the case’, as prohibited under the Non-Waivable Red List of the IBA Guidelines on Conflicts of Interest. Instead, it was merely a similar professional duty as owed by legal practitioners or surgeons when advising clients on the prospects of a legal case or a medical operation, the tribunal said.

Tribunal clarified meaning of NAFTA consolidation clause

The tribunal next considered numerous issues relating to the proper interpretation of Article 1126 and its role in ruling on consolidation.

First, the tribunal clarified that the burden of proof lay with the party seeking to establish that there was a ‘question of law or fact in common’ in the three cases. There was no scope to impose a high bar on consolidation, in the tribunal’s view; the clause’s requirements must simply be met.

Under Article 1126(8), the original separate tribunals ‘shall not have jurisdiction to decide a claim, or a part of a claim, over which a Tribunal established under this Article [i.e., a consolidation tribunal] has assumed jurisdiction’. The claimants contended that the word ‘jurisdiction’ here indicated that the consolidation tribunal could not hear any jurisdictional objections, because it had already ‘assumed jurisdiction’. However, the tribunal dismissed this, finding that ‘assuming jurisdiction’ simply meant ‘taking over the proceedings’. Thus, the US had not waived its jurisdictional objections, and the consolidation tribunal was not bound to accept jurisdiction, if it decided in favour of consolidation.

Similarly, the tribunal confirmed that requesting consolidation did not constitute a jurisdictional objection. However, the tribunal agreed that the US could not bring a jurisdictional objection to the consolidation tribunal (if it decided to consolidate) if the US had not brought that objection to the original separate tribunal within the appropriate timeframe (i.e., no later than in the statement of defence). (Given the early procedural status of the three cases, however, this finding did not appear to restrict the US filings in the case.)

Next, the tribunal addressed the claimants’ argument that Article 1126 referred to consolidation of ‘claims’, which, the claimants said, did not include ‘jurisdictional objections’. In the tribunal’s view, however, a claim could only be disposed of by also disposing of any jurisdictional objections against that claim. Thus, the consolidation tribunal had power to order consolidation of the full case (or part of the case, if thought appropriate, such as only consolidating the jurisdictional objections).

Further, the English and French versions of Article 1126 referred to consolidation where there was ‘a question’ of law or fact in common, while the equally authentic Spanish version referred to ‘questions’ of law or fact. To reconcile these, the tribunal said, it was appropriate to require at least one question in common.

It remained unclear though, whether the common question(s) needed to have already arisen in the separate cases, or were merely anticipated to arise at a later procedural stage. For the tribunal, consolidation could not occur simply because a common question was anticipated to arise. Instead, the requesting party must show with ‘a degree of certainty’ that the question(s) would arise in each proceeding.

Test for consolidation is objective, depending on factors of time, cost and conflicting decisions; tribunal downplayed confidentiality concerns

The tribunal then turned to the meaning of the ‘interests of fair and efficient resolution of the claims’ in Article 1126(2). The US argued that this was an objective standard – simply whether consolidation was ‘fair’, rather than whether it was fairer or more efficient than maintaining separate proceedings. To determine this, the US said, the tribunal should consider factors of time, cost, and the avoidance of conflicting decisions. The claimants, meanwhile, urged a comparison with separate tribunals, and called for consideration of a wider range of factors, including the parties’ wishes.

On this question, the tribunal largely sided with the US, viewing the phrase as objective rather than relative to separate tribunals. Nevertheless, it said, comparison with the alternative was a ‘guiding test’, and in this regard, the US proposed factors of time, cost and conflicts were relevant.

Thus, the more advanced each separate case was, the less likely a tribunal would be to order consolidation, since less time would be saved. Overall, consolidation may take longer than resolving separate claims, which was less beneficial for claimants, but more efficient for states. Resisting the claimants’ suggestion that conflicting decisions were only relevant where the parties in each case were related (as in the then-recent CME/Lauder situation), the tribunal held that avoiding conflicts was always desirable.

Moreover, there was no indication in the text that a smaller number of claims would be less likely to be consolidated, as the claimants had argued. Similarly, there were no concerns in this case over consolidation purely for tactical or abusive reasons.

Lastly, the factor of confidential information, pressed by the claimants, was held not determinative by the tribunal. While acknowledging that confidentiality was the major reason adopted by the Corn Products tribunal in deciding not to consolidate another set of unrelated NAFTA claims, the tribunal here held that it was equipped to handle complex confidentiality issues, including by redactions in filings. The tribunal added that ‘[t]he general trend in investor-State arbitration is transparency of process, a trend to which the Consolidation Tribunal subscribes’.

Questions of procedure, delay and estoppel were addressed

Further questions of procedure remained for consideration. First, the US suggested that the tribunal might order several consolidations – once now on jurisdiction, and perhaps again on merits later. However, the tribunal rejected this, finding that the NAFTA text contemplated a single decision on consolidation, with the tribunal’s mandate changing thereafter to ruling on the issues consolidated.

The parties had also debated the stage at which a consolidation tribunal would begin its analysis; if one separate tribunal had already ruled on jurisdiction, the consolidation tribunal might arguably be bound by that ruling in respect of that claimant. However, the tribunal saw no need to resolve that question, since it did not arise in this case. Nevertheless, it added that, if a separate tribunal had already reached a decision on merits, there may no longer be any ‘question of law or fact in common’, and so consolidation would not likely be ordered.

Last, Tembec separately contended that the US was barred from requesting consolidation either by the doctrine of laches (delay) or by estoppel. (Canfor and Terminal, meanwhile, argued that this issue was already subsumed in the analysis of a ‘fair … resolution of the claims’.)

On this issue, the tribunal noted that laches did indeed bar claims for consolidation in some domestic jurisdictions such as New York. However, the tribunal said, laches arose under common law principles of equity, which did not necessarily apply in public international law claims. In any case, the delay here was only 12-18 months, much smaller than the delays of 20-80 years in cases applying laches. As for estoppel, while agreeing that this was a principle of international law, the tribunal saw no representation from the US that it would not request consolidation.

Tribunal saw common questions of law and fact; good reasons to order consolidation, unlike in Corn Products case

Applying this extensive analysis to the case at hand, the tribunal ultimately found several common questions of law and fact arising in the cases. The US had presented (or had shown with a ‘degree of certainty’ that it would present) the same jurisdictional objections in each case, relating to the interaction with Chapter 19 and the claimants’ status as investors with investments in the US. The claimants’ claims on the merits were the same, relating to the US measures taken against the Canadian softwood lumber industry allegedly to achieve a ‘political result’. While Tembec claimed that it had some distinct products compared to the other claimants, requiring a distinct analysis, the tribunal held that this was just one question, and did not outweigh the other commonalities.

Further, the tribunal noted that none of the three cases had progressed too far to make consolidation inefficient. While Canfor would suffer some delay (since its case was furthest advanced), it would suffer delay in any event since one arbitrator had withdrawn. Costs would be lower for the US in a consolidated proceeding, and only slightly higher for each claimant, the tribunal said, noting that Canfor could re-use most of its filings already prepared. As well, there was a risk of inconsistency if the cases were not consolidated.

Recalling its view that (unlike the Corn Products tribunal’s view) it could handle any issues of confidentiality between the three competitor claimants, the tribunal thus ordered consolidation. It added that its decision was distinguished from the Corn Products tribunal’s decision not to consolidate on the grounds that the separate claims in that case were less procedurally aligned, the claimants were more different, and only jurisdictional issues were common.

As a result, the two separate tribunals in the Canfor and Tembec cases ceased to function (no tribunal had been composed in the Terminal case), and the consolidation tribunal took over the proceedings.

Tembec withdrew from consolidated case

On June 6, 2006, the consolidation tribunal then issued its Decision on Preliminary Question, ruling on the US objection that the claimants’ claims were barred by NAFTA Article 1901(3).

Before addressing this, however, the tribunal addressed the fact that one claimant, Tembec, had sought to withdraw from the consolidated proceedings in December 2005, and had filed a request for set-aside of the consolidation order in DC courts. After consulting with the parties, the tribunal issued an order withdrawing Tembec from the case, but not ruling on whether this withdrawal was done with or without prejudice to a new claim by Tembec. In the tribunal’s view, that question was to be decided by any future new tribunal to which Tembec might apply.

The costs of Tembec’s participation were not determined at this stage.

Textual analysis of exclusion for anti-dumping/countervailing duty law suggested no jurisdiction

Turning to the US objection, the tribunal then began an extensive analysis of Article 1901(3).

This clause provided that ‘no provision of any other Chapter of this Agreement shall be construed as imposing obligations on a Party with respect to the Party’s antidumping law or countervailing duty law’. In the state’s view, to the extent that claims were with respect to US anti-dumping or countervailing duty (AD/CVD) law, they could not be brought to Chapter 11 investor-state arbitration.

The claimants argued that the clause meant only that other NAFTA obligations (such as those in Chapter 11) must be ‘construed’ in a particular way, rather than that it barred the application of those other obligations. Further, the claimants argued that a Chapter 11 arbitration agreement was not an ‘obligation’ covered by Article 1901(3), and that, in any event, the Chapter 11 obligations here perhaps ‘arose out of’ AD/CVD laws, but were not obligations ‘with respect to’ those laws, and so were not captured by Article 1901(3).

However, the tribunal found that an investor-state arbitration agreement under Chapter 11 was an ‘obligation’, and that ‘with respect to’ should be interpreted broadly, to capture the US obligations here. For the tribunal, the real question was whether Article 1901(3) applied only to the adoption and content of AD/CVD laws, or whether it also covered measures taken by the respondent in application of those laws.

On this issue, the tribunal drew assistance from the general definition of ‘antidumping law and countervailing duty law’ in Chapter 19, which defined the term to ‘include … relevant statutes, legislative history, administrative practice and judicial precedents’. The tribunal agreed with the claimants that ‘administrative practice’ referred to ‘written policy guidance of general application’ and did not include the specific determinations made by the Department of Commerce and the ITC in relation to the claimants. However, alongside noting that the definition was broad and inclusive, the tribunal agreed more generally with the US that there was no difference between law and the application of that law.

As such, after this textual analysis, the tribunal found that the US application of AD/CVD law to the claimants was captured by Article 1901(3), prima facie excluding it from Chapter 11 review.

Treaty context did not assist, but treaty objectives also suggested no jurisdiction

Turning to analyse the context of Article 1901(3), the tribunal agreed that, since the NAFTA Chapter 20 inter-state dispute settlement mechanism explicitly excluded Chapter 19 matters, it might make sense to say that the Chapter 11 investor-state mechanism should similarly do so. However, the tribunal noted, Chapter 11 did not explicitly say so.

Further, the tribunal drew no guidance from Article 1112(1)’s provision that other NAFTA chapters were to prevail over Chapter 11 where inconsistent, since the tribunal saw no inconsistency pleaded here.

Similarly, the US argued that Article 1115 set out a purpose of Chapter 11 as establishing an investor-state mechanism ‘without prejudice to’ the Chapter 20 inter-state mechanism. For the US, this demonstrated that the drafters never foresaw any possibility of overlap between Chapter 11 and Chapter 19, or the overlap would have been provided for, like the Chapter 20 overlap. However, the tribunal noted that this argument could equally run in reverse; Chapter 11 might be argued to contemplate overlap with Chapter 19 precisely because it was not mentioned.

As for the objectives of NAFTA, the tribunal found that the treaty generally sought to avoid parallel proceedings in order to ‘create effective procedures … for the resolution of disputes’, as Article 102(1)(e) recognised. For the tribunal, this was seen in the waiver provisions in Chapter 11 barring parallel domestic claims, and in the separation of financial services claims into Chapter 14. Where parallel proceedings were permitted, the tribunal found – such as investor-state and inter-state cases – the treaty explicitly said so.

Moreover, the potential Chapter 11 proceedings here would indeed be parallel proceedings to the Chapter 19 binational panel proceedings, in which the claimants were entitled to participate, since the claims were based on the ‘same factual matrix’, even though one applied domestic law and the other international law, and even though different remedies would result.

Similar exclusions for tax and immigration further supported tribunal’s view; jurisdiction was declined over AD/CVD claims

The tribunal then compared Article 1901(3) with other NAFTA clauses that excluded investor-state dispute settlement on certain issues. The tax exception in Article 2103, for instance, used similar language to Article 1901(3), and had the effect of barring investor-state claims over tax matters. Article 1607’s wording similarly barred investor-state claims over immigration matters.

Turning to supplementary means of interpretation, the claimants pointed to a US explanatory note regarding NAFTA stating that Chapter 11 applied to ‘all government measures’ except for financial services measures (which were covered in Chapter 14). The tribunal was not swayed by this apparently categorical statement, noting that, for instance, immigration measures were also excluded under Article 1607.

While the claimants protested that the US interpretation would leave a hole in protection for the AD/CVD treatment of investors, the tribunal recalled the ‘the reality of treaty negotiations between States with which investors have to live’, and observed that Chapter 11 already represented a ‘great step forward’ for investors.

Thus, the tribunal concluded that ‘the entire Chapter Eleven does not apply with respect to the antidumping law and countervailing duty law of a State Party’, and declined jurisdiction over the claimants’ claims in this regard.

‘Byrd Amendment’ claims remained within jurisdiction

However, the claimants complained not only of their treatment by US authorities under AD/CVD law, but also of the effect of the so-called Byrd Amendment. This amendment to US law provided that AD/CV duties collected from overseas producers would be passed directly onto affected US producers who participated in the domestic complaint leading to the imposition of those duties.

According to the claimants, this amendment created incentives for US producers to push for frivolous AD/CVD petitions, knowing that they would financially benefit from any success. Further, the claimants said, it discouraged settlement and instead encouraged financial sanctions, and also ‘distorts the United States market place in favour of the domestic United States industry’.

While the US argued that the Byrd Amendment was also part of AD/CVD law and was therefore similarly exempt from scrutiny under Article 1901(3), the tribunal disagreed.

US failure to notify NAFTA partners of amendment disqualified it from relying on exception

For the tribunal, Chapter 19 did define AD/CVD law to include ‘statutes’, but it also imposed conditions on including any amendments to those statutes within the definition of AD/CVD law. In particular, the US was required to notify and consult with its NAFTA partners before amending any aspects of its AD/CVD law. Rejecting the US plea that its failure to do so here was a ‘harmless omission’, the tribunal held that notification was part of an important compromise in the Chapter 19 negotiations: NAFTA states would be permitted to keep their own AD/CVD laws rather than being forced to harmonise, but only on the condition that they properly notified and consulted on any changes.

The US responded that Canada (and the claimants) were well aware of the amendment, since it had been widely reported in the press, and since Canada had made submissions on it in the context of related WTO proceedings. As well, the US said, it had only a limited timeframe available to notify given that the amendment had been tacked onto another bill passing the US Congress at the time.

The tribunal dismissed these responses, finding the US domestic timeframes irrelevant to its NAFTA obligations, and holding that the claimants’ actual knowledge (if any) of the amendment did not absolve the US of its strict obligation to notify.

The US lastly protested that the tribunal’s view would allow states to ‘shield’ AD/CVD amendments from Chapter 19 scrutiny simply by failing to notify them to NAFTA partners, since the amendments would then not be part of the ‘AD/CVD law’ subject to Chapter 19. However, the tribunal held that it was obliged to assume that states would perform their obligations in good faith, and that a state would therefore not deliberately breach the notification obligation in order to avoid other NAFTA obligations.

As a result, the claimants’ claims regarding the Byrd Amendment fell within the tribunal’s jurisdiction (the tribunal added that the claims disclosed a prima facie case of a national treatment violation). It was irrelevant that the amendment was in the process of being repealed, the tribunal confirmed, since the complaints related to the period when it was in force.

Canada and US reached inter-state settlement, leading to withdrawal of NAFTA claims; US sought recovery of costs from Tembec

Around the time of the consolidation tribunal’s jurisdictional decision in June 2006, Canada and the United States were undergoing broader negotiations aimed at a settlement of the entire cross-border softwood lumber dispute. These negotiations concluded with the entry into force of the inter-state Softwood Lumber Agreement in October 2006.

As part of this inter-state agreement, Canfor and Terminal agreed to withdraw their claims against the US, and these parties all agreed not to seek any costs orders from each other and to share the relevant costs of arbitration equally.

Tembec argued that it was also in the same position, having similarly agreed not to claim any costs from the US and that the US would not claim costs from Tembec. However, the US disagreed with this view, contending that, following Tembec’s decision to ‘unilaterally abandon[] its claim in the eleventh hour’, the company should be responsible for the US costs of the separate Tembec tribunal and some portion of the consolidation tribunal.

At the outset of its final decision on costs of July 19, 2007, the tribunal noted that this claim from the US (for around US$150,000 in costs) might appear strange given that the state had just agreed to pay Tembec US$242 million under the Softwood Lumber Agreement. However, the tribunal noted the US position that this was a point of principle; the state sought to dissuade future NAFTA claimants from withdrawing their claims at a late stage without suffering any consequences.

Inter-state agreement did not cover Tembec’s costs

The tribunal firstly addressed Tembec’s contention that, like Canfor and Terminal, the matter of its costs was already agreed on in one of the side agreements to the inter-state Softwood Lumber Agreement.

The tribunal observed that an early version of one of these agreements, the Termination of Litigation Agreement (TLA), did refer to Tembec’s NAFTA claim, providing that neither party would claim any costs in relation to it. However, the tribunal noted, this version of the TLA was never signed by Tembec or the US. Similarly, a later version of the TLA was signed by Tembec, but not the US. The subsequent Settlement of Claims Agreement (SCA), which the tribunal found had replaced the TLA, was signed by both parties, but did not mention Tembec’s NAFTA claim. In the tribunal’s view, Tembec should have known that the SCA did not mention its NAFTA claim, and should not have signed the document if it objected to this. (As we detailed here, Tembec’s view that it had been misled into agreeing to a settlement that permitted the US to seek a costs order against it formed part of its unsuccessful contentions challenging the consolidation order before US courts.)

Thus, the tribunal held, there was no mutual agreement between Tembec and the US not to claim any costs from one another in relation to the NAFTA claim. Instead, allocation of costs was to be determined by the tribunal itself, under its preferred approach of ‘costs follow the event’.

Tembec fought adverse costs order, but tribunal frowned upon its unilateral withdrawal

Tembec offered several responses to the suggestion that it was the unsuccessful party and should therefore bear the US costs in the case.

First, the company noted that it had sought a set-aside of the consolidation order, suggesting that the US had not necessarily prevailed. However, the tribunal rejected the view that merely filing a set-aside could alter the outcome of the arbitration, and further noted that US courts had since denied the set-aside bid (as we discussed in 2008 here).

Second, Tembec protested that the US had been too late in requesting consolidation. However, the tribunal noted that it had already rejected this contention in ordering the consolidation.

Third, Tembec argued that it could not be compelled to finance the US interest in having the proceedings consolidated, which also raised novel legal issues for which neither party should bear the burden. The tribunal acknowledged these views, but nevertheless recalled that Tembec itself had unilaterally withdrawn from the case, an action which must come with consequences. For the tribunal, a party that unilaterally withdrew from a case was to be treated as the unsuccessful party, as it was in many domestic legal systems.

Fourth, Tembec alleged that the US was simply targeting it for opposing consolidation in the first place, while treating Canfor and Terminal differently. However, the tribunal observed that the US had a right to claim its costs, and that its claims were not abusive; furthermore, the tribunal noted, Canfor and Terminal had not unilaterally withdrawn from the case.

Thus, the tribunal concluded that Tembec was to bear all of the US costs apportioned to the Tembec proceedings.

Tembec bore all US costs for its share of the proceedings, while others split equally

Turning to the actual apportionment of the various fees, the tribunal began by swiftly deciding, based on the parties’ prior agreement, that the Canfor tribunal fees of around US$372,000 were to be shared equally between Canfor and the US. Since no Terminal tribunal was constituted, meanwhile, no arbitral fees had been incurred. Furthermore, the parties were to bear their own legal fees in these two cases. As for the separate Tembec tribunal, the consolidation tribunal ordered Tembec to bear the full US$34,000 cost.

The fees of the consolidation tribunal itself required a more complicated apportionment, in the tribunal’s view.

The arbitral costs of Phase I of the consolidation case (from its establishment until Tembec’s withdrawal) amounted to around US$478,000. All three claimants participated in this phase. Two-thirds of these costs were attributable to Canfor, Terminal and the US, the tribunal held, with that amount split 25% Canfor, 25% Terminal and 50% US. One-third was attributable to Tembec and the US, with Tembec to bear that full portion.

Phase II of the consolidation case (from Tembec’s withdrawal until the jurisdictional decision) involved only two claimants, Canfor and Terminal, and around US$328,000 in costs. The US would bear half of these, and the two claimants would bear equal shares of the other half.

In Phase III (from the jurisdictional decision until this costs decision), Tembec again participated, and the arbitral costs were around US$189,000. Given the predominant focus on Tembec’s costs claims in this phase, the tribunal apportioned these 90% to Tembec and the US, and 10% to Canfor, Terminal and the US. While the latter portion was split as for Phases I and II, Tembec would bear all of the former portion.

Thus, the final costs apportionment was around US$166,000 for Canfor, US$329,000 for Tembec, US$166,000 for Terminal and US$333,000 for the US.

As for the US legal costs in relation to Tembec, the tribunal viewed the state’s claim of US$94,720 as ‘very modest’, and awarded it entirely against Tembec.

(As we detailed here, Tembec unsuccessfully sought to set aside the tribunal’s costs order before the US courts.)