By Harm Schepel
All is clear, then: CETA’s Investment Chapter is perfectly compatible with EU Law. According to Advocate General Bot, the agreement is wholly separate from the normative (as opposed to the factual) universe of EU law, and merely protects readily identifiable ‘foreigners’ investing in the EU in the same way as it protects readily identifiable ‘European’ investors in foreign lands. From what we know of the hearing, the Advocate General provides not much more than a useful summary of the talking points offered by the Council, the Commission and the vast majority of the 12 intervening Member States, remarkably united in a bid to save the EU’s new external trade and investment policy. Clearly, the pressure on the Court to follow suit will be enormous. And yet. It is true, CETA builds strong fences to make good neighbors. But let spring be the mischief in me: CETA cannot wall out what EU Law walls in.[i]
Apples and Oranges
The Court will finally, one would hope, address the question of how Investment Treaties relate to the prohibition of discrimination in the internal market- an issue it ignored, for reasons of judicial economy, in Achmea. If it decides to do so it would be wise to ignore the Advocate General’s proposal. On the perfectly proper basis that the principle of equal treatment requires that comparable situations not be treated differently, AG Bot spends a lot of time arguing that Canadian investors operating in the Union find themselves in a completely different situation than EU investors operating in the Union since they are, well, foreigners.[ii] This line of reasoning would be misguided in the best of circumstances, but is especially intellectually dishonest in the context of the discussion of ‘locally established enterprises’ that the AG engages in.
It is easy to imagine the Investment Chapter to be about protecting ‘investors’- heroic yet vulnerable men and women beating all the odds by ‘making it’ in distant foreign lands. Reading the Chapter might even reinforce the idea. After all, Article 8.10 prohibits ‘harassment’ and ‘targeted discrimination on grounds of gender, race or religious beliefs,’ mistreatment we would normally think actual people suffer. But it isn’t really a bunch of Vancouver hipsters trying their hand at brewing biological beer in Wallonia, or an Orkney fisherman netting lobsters off Prince Edward Island that CETA caters for: in modern economic reality, investment will predominantly take the form of companies setting up companies. The latter ‘locally established enterprises’ are to be treated fairly and equitably, are not to be expropriated, and are the ones that receive damages awarded by the Tribunal.[iii] In other words, it is mostly investments, not investors, that CETA protects. And it is in this scenario that Belgium raises a rather obvious question: aren’t these ‘locally established enterprises’ treated more favorably than other local enterprises?
At this stage, it might be useful to pause and make a blindingly obvious point- so blinding that AG Bot chooses to ignore it. CETA defines a ‘locally established enterprise’ as ‘a juridical person that is constituted or organized under the law of the respondent and that an investor of the other Party owns or controls directly or indirectly.’[iv] Ipso facto, as a matter of EU Law, this makes a ‘locally established enterprise’ a European Union company with the nationality of a Member State which finds itself, as a matter of EU Law, in the exact same position as any other European Union company. As the Court held in Felixstowe:
‘[t]he status of being a European Union company is based, under Article 54 TFEU, on the location of the corporate seat and the legal order where the company is incorporated, not on the nationality of its shareholders.’[v]
AG Bot throws everything and the kitchen sink at us to distract from the issue. The only person allowed to submit a claim to the Tribunal is the investor, not the locally established company. Sure. And any compensation awarded by the Tribunal, even if actually paid to the locally established company will ‘ultimately benefit’ the investor. Right (para. 193). When he finally gets to it, though, he just spirits away the separate legal existence of the locally established enterprise. He writes:
‘Since the investor of the one Party and the locally established enterprise in the territory of the other Party must in reality be treated as one and the same (149), the question raised by the Kingdom of Belgium is whether or not discrimination exists between foreign investors, who enjoy specific substantive and procedural protection, and local investors, who do not benefit from such protection.’(para. 194)
Footnote 149, for good measure, reads: ‘As the Commission states in its observations, locally established enterprises are an extension of the foreign investor and, therefore, it is justified to equate them with the foreign investor who owns or controls them.’
To be clear, the Advocate General’s move is wholly consistent with the logic of international investment law where separate corporate legal personality is taken very seriously when it comes to granting standing to investors,[vi] but is completely ignored when it comes to the treatment of investments- to the point of allowing minority shareholders to bring derivative claims on behalf of the locally established company. But if the compatibility of CETA’s Investment Chapter with EU Law depends on whether investments are to be considered apples or oranges, surely it is the EU Law definition of an apple that controls, and not the assertion that these apples are ‘in reality’ oranges.
The doctrine of separate legal personality, moreover, is not some insignificant formalistic technicality in the internal market: it lies at the very heart of the current protection of third-country investors. By separating shareholders from the company, it allows foreign investors to ‘externalize’ the norm prohibiting intra-EU discrimination.[vii] The irony is marvelous: on the one hand, Canadian investors may not be discriminated against because their EU investments are considered separate EU legal entities, on the other hand the rights and privileges they have under CETA are not to be considered discriminatory on the grounds that their investments are but ‘an extension’ of themselves.
If this much is accepted, then CETA’s Investment Chapter is clearly incompatible with the EU Law norm prohibiting discrimination between EU companies: it forces Member States to afford treatment and, in case of breach, pay damages, to certain European companies which they don’t afford and pay to other European companies in similar circumstances. The statement in the Joint Interpretative Instrument that ‘CETA will not result in foreign investors being treated more favorably than domestic investors’- the ‘no greater rights’ trope that the Advocate General unforgivably throws out without a hint of scrutiny-(para. 200) doesn’t change the simple fact that the Investment Chapter grants autonomous and absolute standards of protection to ‘foreign investors’- and, to repeat, more often than not these Canadian investors will be, as a matter of EU law, European companies with the nationality of a Member State- that apply to them regardless of the way that domestic investors are treated.
This is not the end of it, of course, and the Advocate General will not be fazed. Any discrimination that might be considered to result from the Investment Chapter, he writes, is in any event ‘duly justified’, since it falls squarely in the caselaw permitting discrimination ‘where the difference in treatment relates to a legally permitted objective pursued by the measure having the effect of giving rise to such a difference and is proportionate to that objective.’(para. 210) Quite obviously, according to AG Bot, the purpose of encouraging investment is ‘legally permitted’, and, equally obviously, the negotiators of the Agreement could legitimately take the view that the investor-state dispute settlement system is an effective tool to achieve that objective (para. 212). Presto fatto.
Let us not, for the moment, labor the tired and true observation that the empirical evidence on the influence of ISDS on actual investment flows is inconclusive at best, and concede the point. There is still a major problem with the Advocate General’s reasoning. In the caselaw he refers to, which can be traced back to Arcelor and from there all the way back to 1977,[viii] unequal treatment is invariably an unfortunate side-effect of a measure in pursuance of some worthy objective. Under CETA, unequal treatment is the measure that is to achieve the objective of attracting investment. To follow the Advocate General, then, the Court would be required to go well beyond existing caselaw and hold that discrimination is a legitimate instrument of public policy, and not just a tolerable consequence of it.
It’s the Autonomy, stupid
If, as is to be expected, the Court’s final opinion on compatibility will hinge on the arguments about the autonomy of the EU legal order, the choice is stark. If the Court follows the Advocate General, it will reduce the concept of autonomy to nothing more than its own interpretive monopoly over EU Law, but save CETA and the EU’s new trade and investment policy. The alternative would be to turn the concept of autonomy into a preemption doctrine and ban investment law and arbitration from the Union altogether.
The starting point must be Achmea. There, the Court held that a BIT between two Member States adversely affected the autonomy of EU Law since it embodied their agreement
‘to remove from the jurisdiction of their own courts, and hence from the system of judicial remedies which the second subparagraph of Article 19(1) TEU requires them to establish in the fields covered by EU law disputes which may concern the application or interpretation of EU law.’[ix]
The Advocate General’s first move is to deny any relevance of Achmea to the question of the compatibility with EU Law of CETA’s Investment Chapter because ‘the premises which must guide the line of reasoning are different.’(para. 106) The Courts own pronouncements in Achmea on the distinction between intra-EU BITs and EU Investment Agreements clearly empowered him to do so. There, the Court recalled its settled caselaw according to which an international agreement providing for the establishment of a court responsible for the interpretation of its provisions and whose decisions are binding on the institutions, including the Court of Justice, is not incompatible with EU law, provided that that the autonomy of the EU and its legal order is respected. It then proceeded as follows:
‘In the present case, however, apart from the fact that the disputes falling within the jurisdiction of the arbitral tribunal referred to in Article 8 of the BIT may relate to the interpretation both of that agreement and of EU law, the possibility of submitting those disputes to a body which is not part of the judicial system of the EU is provided for by an agreement which was concluded not by the EU but by Member States. Article 8 of the BIT is such as to call into question not only the principle of mutual trust between the Member States but also the preservation of the particular nature of the law established by the Treaties, ensured by the preliminary ruling procedure provided for in Article 267 TFEU, and is not therefore compatible with the principle of sincere cooperation.’[x]
In fairness, making much sense of this is not straightforward. Article 8 of the BIT at issue, it will be recalled, is the applicable law clause that instructed the arbitral Tribunal to ‘decide on the basis of the law’, and to ‘take into account’, among other things, ‘the law in force of the Contracting Party involved.’ On a narrow reading of Achmea, this presents the only problem for the Court since it allows a Tribunal located outside of the EU judicial hierarchy to apply or interpret EU Law. In the first sentence of the passage above, however, the Court seems to be suggesting (by using ‘apart from’) that its problems with Treaties between Member States go beyond potential objections it might have to EU Agreements- perhaps as far as to prohibit any dispute settlement mechanisms outside of the EU legal system between themselves and their respective nationals, regardless of applicable law clauses. In the second sentence it seems to explain why- because of the principles of mutual trust and sincere cooperation- and yet limits the extent to which Member States had offended these principles to exactly the applicable law clause.
It seems unwise to infer from this that different standards as regards the need to protect the autonomy of EU Law apply to BITs and EU Agreements. On an abstract level, it is not easy to defend reasoning whereby the content of a legal good changes in function of who offends it. On a practical level, the Court would paint itself into a corner as regards the Energy Charter Treaty which is both an EU Agreement and allows claims to be brought between Member States. It should also be remembered that the Court signaled in Opinion 2/15, regarding an EU Agreement, that its problem with investment arbitration was exactly the same as it expressed it to be in Achmea, namely that it ‘removes’ disputes from the jurisdiction of courts nestled in the EU judicial hierarchy.[xi]
In fairness, the Advocate General does not explicitly state otherwise, limiting himself to claiming that the Court’s ‘approach’ in Achmea cannot be transposed to CETA and that the ‘analytical framework’ cannot be identical (para. 106 and 113). But what the exercise does allow him to is to dismiss any notion that the concept of ‘autonomy’ might amount to more than just the need to preserve ‘the exclusive jurisdiction of the Court of Justice to provide a definitive interpretation of EU Law.’(para. 114) And with that, the bulk of his Opinion becomes nothing more than an exercise in reviewing the success of the ‘scrubbed’ Investment Chapter in ‘walling out’ EU Law.
CETA’s applicable law clause limits the Tribunal to the Agreement itself and other rules and principles of international law. It may not determine the legality of a measure under domestic law, and may ‘consider’ domestic law only ‘as a matter of fact.’ In doing so, it is to follow the ‘prevailing interpretation’ given to domestic law by the relevant courts and authorities and, in any event, ‘any meaning’ the Tribunal gives to domestic law shall not be binding on those courts and authorities (Article 8.31 CETA).The Advocate General is satisfied that CETA thus interferes as ‘little as possible’ with rules of EU Law and ‘succeeds in guaranteeing a balance’ between investment protection and the preservation of the autonomy of EU Law (paras. 117 and 118). Now, there is a lot to be said about this– and no doubt a lot will be said about this, including, it is to be feared, by the Court. Reasonable people can reasonably disagree. But either way, it is hard to see how this could be the end of the matter.
Let us, for argument’s sake, suppose that CETA’s walls between the Investment Chapter and EU Law are high enough to not deprive the Court of Justice of its monopoly of the interpretation of EU Law. Would that be enough to safeguard the ‘autonomy of the EU legal order’- in other words, does the concept of the ‘autonomy of the EU legal order’ really mean nothing more than ‘the autonomy of the Court of Justice’? The Court is entitled to a little faith, and we should assume that it does not consider its ‘interpretative monopoly’ an end in itself but part and parcel of the system of ‘effective judicial review designed to ensure compliance with provisions of EU Law,’ which it considers ‘of the essence of the rule of law.’[xii]We should also assume that it is the concern for the effective exercise of rights derived from EU Law that animates the Court’s decision to oblige Member States to set up impartial and independent courts and tribunals, safely tucked away in the EU judicial hierarchy, in fields ‘covered by EU Law.’[xiii] And in case anyone would want to argue that CETA does not occupy ‘a field covered by EU Law’, let it just be repeated: CETA would allow some, but not all, European companies to bring claims against Member States for measures that are tolerated or even required by EU Law. [xiv]
Were the Court to follow Advocate General Bot, it would reduce its holding in Achmea to the proposition that it is perfectly fine to ‘remove’ disputes in fields covered by EU Law from the jurisdiction of impartial and independent courts and tribunals collocated in the EU judicial hierarchy as long as these disputes do not concern ‘the application and interpretation of EU Law’ but the application and interpretation of another set of rules that is explicitly meant to displace and contradict EU Law. And if the Court were to follow the Advocate General, it would reduce the ‘essence of the rule of law’ to the proposition that the absence of ‘effective judicial review’ is perfectly fine as long as it is not compliance with EU Law that is pursued, but compliance with another set of rules that is explicitly meant to displace and contradict EU Law. If the Court were to follow the Advocate General, it would let CETA wall out what EU Law walls in. Let Spring be its mischief.
[i] From Robert Frost’s “Mending Wall”:
‘He only says, “Good fences make good neighbours.”
Spring is the mischief in me, and I wonder
If I could put a notion in his head:
“Why do they make good neighbours? Isn’t it
Where there are cows? But here there are no cows.
Before I built a wall I’d ask to know
What I was walling in or walling out,
And to whom I was like to give offence.’
[ii] And have hence ‘assumed the risks and the costs of an investment in a foreign economic area’ and are operating an unfamiliar legal environment. Para 207.
[iii] Articles 8.10, 8.12, and 8.39 (2)(a).
[iv] Article 8.1.
[v] Case C-80/12 Felixstowe EU:C:2015:200, para 40.
[vi] This is why CETA, rightly, demands of claimant enterprises not just the nationality of a Party but ‘substantial business activities in the territory of that Party.’ Article 8.1.
[vii] See Felixstowe, above n. vi.
[viii] Bot refers to Case C-390/15 RPO EU:C:2017:174, which refers, inter alia, to Case C-127/07 Arcelor Atlantique  ECR I-9895, which in turn refers, inter alia, to Case 114/76 Bela-Mühle Bergmann  ECR 1211.
[ix] Case C-284/15 Achmea EU:C:2018:158, para 55.
[x] Achmea, above n.xv, para 58.
[xi] Opinion 2/15 (EU- Singapore FTA) EU:C:2017:376, para 292.
[xii] See eg Case C-72/15 Rosneft EU:C:2017:236, para 73.
[xiii] See Case C-46/16 Associação Sindical dos Juízes Portugueses EU:C:2018:117.
[xiv] If nothing else, this is freely admitted in Article 3 of Regulation 912/2014, OJ L257/121.