Should FTAs open and close (or only open) international procurement markets?

I have recently had some interesting discussions on the role of Free Trade Agreements (FTAs) in liberalising procurement-related international trade. The standard analysis is that FTAs serve the purpose of reciprocally opening up procurement markets to the economic operators of the signatory parties, and that the parties negotiate them on the basis of reciprocal concessions so that the market access given by A to economic operators from B roughly equates that given by B to economic operators from A (or is offset by other concessions from B in other chapters of the FTA with an imbalance in A’s favour).

Implicitly, this analysis assumes that A’s and B’s markets are (relatively) closed to third party economic operators, and that they will remain as such. The more interesting question is, though, whether FTAs should also close procurement markets to non-signatory parties in order to (partially) protect the concessions mutually granted, as well as to put pressure for further procurement-related trade liberalisation.

Let’s imagine that A, a party with several existing FTAs with third countries covering procurement, manages to negotiate the first FTA signed by B liberalising the latter’s procurement markets. It could seem that economic operators from A would have preferential access to B’s markets over any other economic operators (other than B’s, of course).

However, it can well be that, in practice, once the protectionist boat has sailed, B decides to entertain tenders coming from economic operators from C, D … Z for, once B’s domestic industries are not protected, B’s public buyers may well want to browse through the entire catalogue offered by the world market—especially if A does not have the most advanced industry for a specific type of good, service or technology (and I have a hunch this may well be a future scenario concerning digital technologies and AI in particular).

A similar issue can well arise where B already has other FTAs covering procurement and this generates a situation where it is difficult or complex for B’s public buyers to assess whether an economic operator from X does or not have guaranteed market access under the existing FTAs, which can well result in B’s public buyers granting access to economic operators from any origin to avoid legal risks resulting from an incorrect analysis of the existing legal commitments (once open for some, de facto open for all).

I am sure there are more situations where the apparent preferential access given by B to A in the notional FTA can be quickly eroded despite assumptions on how international trade liberalisation operates under FTAs. This thus begs the question whether A should include in its FTAs a clause binding B (and itself!) to unequal treatment (ie exclusion) of economic operators not covered by FTAs (either existing or future) or multilateral agreements. In that way, the concessions given by B to A may be more meaningful and long-lasting, or at least put pressure on third countries to participate in bilateral (and multilateral — looking at the WTO GPA) procurement-related liberalisation efforts.

In the EU’s specific case, the adoption of such requirements in its FTAs covering procurement would be aligned with the policy underlying the 2019 guidelines on third country access to procurement markets, the International Procurement Instrument, and the Foreign Subsidies Regulation.

It may be counter-intuitive that instruments of trade liberalisation should seek to close (or rather keep closed) some markets, but I think this is an issue where FTAs could be used more effectively not only to bilaterally liberalise trade, but also to generate further dynamics of trade liberalisation—or at least to avoid the erosion of bilateral commitments in situations of regulatory complexity or market dynamics pushing for ‘off-the-books’ liberalisation through the practical acceptance of tenders coming from anywhere.

This is an issue I would like to explore further after my digital tech procurement book, so I would be more than interested in thoughts and comments!

Deviating from the GPA in bilateral or multilateral FTAs -- how good or effective is that regulatory strategy?

Fallen deviation road sign.

One of the issues procurement lawyers may be unlikely to think much about (at least going by my own experience) is the interaction of the several (and growing number of) pieces of international regulation of procurement-related trade liberalisation. And, in particular, the interaction between the World trade Organisation Government Procurement Agreement (GPA) and the procurement chapters in the multilateral and bilateral free trade agreements (FTAs) that have been mushrooming for a while now.

This is an issue I encountered for the first time when looking at the procurement chapter of the UK-Australia FTA (UK-AUS FTA, see here and here), and on which I will be submitting evidence to the International Trade Committee of the House of Lords tomorrow. In case of interest, here are my thoughts on the matter.

**Warning**, tackling this issue requires a bit of a deep dive into the 1969 Vienna Convention on the Law of Treaties, so this post may be niche (or even more niche than usual). TLDR: despite (intentionally?) deviating from the text of the GPA, FTAs are unlikely to generate practical effects where they vary or reduce GPA-based obligations, except in limited cases where the effects are only for the economic operators of the two jurisdictions signing the FTA, which is both counterintuitive and prone to litigation, especially where GPA- situations concern access to procurement remedies.

What’s the matter?

For an increasing number of jurisdictions, procurement-related trade liberalisation has become a multi-layered regulatory puzzle. Let’s take the example of the EU. The EU is a member of the GPA, as are Singapore and Canada (all of them members of the 2012 version of the GPA since 6 April 2014). The EU has also concluded procurement chapters in FTAs, eg with Canada in 2017 (EU-Canada FTA (CETA), Chapter 19), or Singapore in 2019 (EU-Singapore FTA, Chapter 9). In these chapters, the EU has adopted a GPA+ approach based on reiterating and tweaking the rules of the GPA—as opposed to the approach followed eg with the UK in 2021, where the EU-UK Trade and Cooperation Agreement explicitly incorporates the GPA (Art 277 EU-UK TCA), and then adds some additional rules.

The dual regulation of procurement liberalisation between GPA parties in bilateral FTAs (or in plurilateral agreements) can create legal issues where the rules in the later FTAs deviate from the GPA in a manner that varies or reduces GPA obligations (ie GPA-). This is the case, for example, of seemingly reduced national treatment obligations concerning suppliers of the relevant parties offering goods or services of other GPA parties. The issue is as follows (bear with me).

The GPA imposes national treatment and non-discrimination obligations as the foundation of its regulatory architecture. The GPA national treatment clause reads ‘With respect to any measure regarding covered procurement, each Party, including its procuring entities, shall accord immediately and unconditionally to the goods and services of any other Party and to the suppliers of any other Party offering the goods or services of any Party, treatment no less favourable than the treatment the Party, including its procuring entities, accords to: a) domestic goods, services and suppliers; and b) goods, services and suppliers of any other Party’ (Art IV(1) emphasis added). This creates a two-tier requirement of ‘most favoured treatment’, both between the goods, services and suppliers of two given GPA members (procuring and supplying) and across the goods, services and suppliers of all GPA parties other than the procuring party. The underlined clause leaves the possibility open for differential treatment of suppliers of a GPA party offering goods or services of a non-GPA party but extends the national treatment obligation to suppliers offering goods of services originating anywhere in the ‘GPA club’.

This is in line with the GPA non-discrimination clause, which reads: ‘With respect to any measure regarding covered procurement, a Party, including its procuring entities, shall not: … b) discriminate against a locally established supplier on the basis that the goods or services offered by that supplier for a particular procurement are goods or services of any other Party’ (Art IV(2) emphasis added). Again, the possibility is open for differential treatment of suppliers of a GPA party offering goods or services of a non-GPA party—on the implicit assumption that domestic suppliers offering goods or services of a non-GPA party are subjected to the same differential treatment.

Now, under both CETA and the EU-Singapore FTA, the regulation of these obligations is altered (and seemingly narrowed down).

The position under CETA is that ‘With respect to any measure regarding covered procurement, each Party, including its procuring entities, shall accord immediately and unconditionally to the goods and services of the other Party and to the suppliers of the other Party offering such goods or services, treatment no less favourable than the treatment the Party, including its procuring entities, accords to its own goods, services and suppliers.‘ (Art 19.4(1) CETA, emphasis added). Prima facie, this clashes with the GPA because it reduces the circle of protected goods and services accessing procurement opportunities via suppliers to Canadian and EU goods and services, and to the exclusion of those originating in other GPA jurisdictions. This is not contradicted by the non-discrimination clause, according to which ‘With respect to any measure regarding covered procurement, a Party, including its procuring entities, shall not: … (b) discriminate against a locally established supplier on the basis that the goods or services offered by that supplier for a particular procurement are goods or services of the other Party.’ (Art 19.4(2) CETA, emphasis added). Therefore, as a result of CETA, there is an open question as to whether EU/Canadian suppliers offering non-EU/Canadian (GPA) goods or services are protected from discrimination.

The position under the EU-Singapore FTA is perhaps slightly less straightforwardly in clash with the GPA, and it is that ‘With respect to any measure regarding covered procurement, each Party, including its procuring entities, shall immediately and unconditionally accord to the goods and services of the other Party and to the suppliers of the other Party treatment no less favourable than the treatment the Party, including its procuring entities, accords to domestic goods, services, and suppliers’ (Art 9.4(1) EU-Singapore FTA, emphasis added). This omits the GPA reference to suppliers ‘offering the goods or services of any Party’ (although it does not limit it to the goods and services of the parties, as in CETA) and could seem like a logical tailoring of the obligation to a bilateral situation. However, it generates doubts as to the position of EU and Singaporean suppliers offering non-EU/Singaporean goods or services originating in GPA jurisdictions, which would have had access under the GPA but may not have access under the FTA. A restrictive interpretation limiting access to those suppliers offering EU/Singaporean goods and services could be supported by the parallel modification of the non-discrimination clause relative to the GPA standard, which reads ‘With respect to any measure regarding covered procurement, a Party, including its procuring entities, shall not: … (b) discriminate against a locally established supplier on the basis that the goods or services offered by that supplier for a particular procurement are goods or services of the other Party (Art 9.4(2) EU-Singapore FTA, emphasis added). Therefore, the same issues as in CETA arise, although there is some more wiggle room for legal interpretation in this case.

In both cases, however, it is at first difficult to establish definitely whether suppliers offering goods or services from GPA parties other than the EU and/or Singapore/Canada have indeed lost protection as a result of the newer FTA obligations. Such a modification would have been implemented without consent from the relevant GPA party and, in practice, would be challenged. More importantly, then, both FTAs create at least a risk of breach of GPA obligations for the EU (and Singapore and/or Canada) in relation to the goods or services from other GPA parties now potentially being (indirectly) discriminated against as a result of their being offered by an EU/Singaporean and/or Canadian supplier (in effect, a ‘GPA-’ situation, despite the aim of the FTA to create a GPA+ regime).

Here is where the Law of Treaties kicks in.

The regulation in the 1969 Vienna Convention

1. General regulatory background regarding conflicting treaty norms

Art 30 of the 1969 Vienna Convention on the Law of Treaties (VCLT)[1] controls the matter of conflicting treaty norms. Before engaging in an assessment of its conflict rules, it is worth stressing that a conflict between treaty norms will only arise where an interpretation of the later treaty that avoids it is either not possible, or not accepted by the parties.[2] Equally, or in a mutually reinforcing manner, understanding how the conflict norms would operate can have an effect on the interpretation of an obligation under the later treaty that the parties are likely to (voluntarily) adopt. In reality, in cases of potential conflict between treaty norms, there will be a clear interpretive ‘pull’ to try to ensure coherence, or minimise divergence, through interpretation of the later treaty which takes account of the obligations owed to others through the earlier agreement. It is a practical approach which can lessen (but cannot eliminate) the difficulties resulting from treaty clashes. In particular because Art 30 VCLT applies to successive treaties relating to the same subject matter and establishing the subject matter of a treaty can in itself be a disputed issue.[3]

Explicit regulation. In case of conflict between treaty norms other than the Charter of the United Nations (which has precedence; ex Art 30(1) VCLT), Art 30(2) VCLT foresees the possibility of explicit regulation via a conflict clause, in which case: ‘When a treaty specifies that it is subject to, or that it is not to be considered as incompatible with, an earlier or later treaty, the provisions of that other treaty prevail’.

Unregulated conflicts between the same parties. In the absence of an explicit conflict clause, the VCLT establishes residuary rules, distinguishing two situations. First, in the case of successive treaties with identical parties, Art 30(3) VCLT establishes precedence for the later treaty,[4] and ‘the earlier treaty applies only to the extent that its provisions are compatible with those of the later treaty’.

Unregulated conflicts between different parties. The second situation concerns successive treaties with different parties, for example a successive treaty between some but not all the signatories of the previous treaty. In this case, Art 30(4) VCLT establishes two rules, based on a distinction between the effects of the new treaty between its signing parties, and its effects vis-à-vis the other signatories of the previous treaty. Between the parties to both treaties, the rule in Art 30(4)(a) is the same as in Art 30(3) VCLT. That is, the newer treaty takes precedence and ‘the earlier treaty applies only to the extent that its provisions are compatible with those of the later treaty’. Conversely, and due to the relativity of treaty rights (Art 34 VCLT), under Art 30(4)(b) VCLT, ‘as between a State party to both treaties and a State party to only one of the treaties, the treaty to which both States are parties governs their mutual rights and obligations.’ Or, in other words, Art 30(4) VCLT establishes that ‘the newer treaty may be applied by its parties but only by leaving intact the rights of those that are only parties to the earlier treaty’.[5]

Modification of treaties between certain of their parties only. Art 30(5) VCLT clarifies that the two rules in Art 30(4) are subject to Art 41 VCLT concerning agreements to modify multilateral treaties between certain parties only, under the rules of those multilateral treaties. Art 41 VCLT covers two situations: (a) the possibility that such a modification is provided for by the treaty; or (b) the modification in question is not prohibited by the treaty, it does not affect the enjoyment by the other parties of their rights under the treaty or the performance of their obligations, and it does not relate to a provision, derogation from which is incompatible with the effective execution of the object and purpose of the treaty as a whole.

Art 30 VCLT does not provide explicit residual rules concerning ‘triangular situations’, where the same action by one of the parties to both conflicting treaties affects the rights resulting from both treaties at the same time (albeit for different parties), and thus unavoidably breaches one of the treaties[6] due to the 'variable geometry' of treaty-resulting obligations. In that case, the possibility of international responsibility of a State for the breach of one of the treaties arises (Art 30(5) VCLT)–especially in the cases falling under Art 30(4)(b) VCLT.[7]

Given this possibility, as mentioned above, where a State is party to two treaties imposing different obligations for triangular situations, it will have a strong incentive to try to resolve the conflict through interpretation or, to the extent allowed by the respective treaties, it will have a strong incentive to comply with the most demanding treaty in order to avoid incurring in international responsibility. Where avoiding breach of either of the treaties in those ways is not possible, States retain discretion to choose which treaty to breach[8] (under the so-called ‘principle of political decision’[9]) and may have an incentive to breach the treaty that would trigger the least (potential) international responsibility, which may well be the newer treaty with a more limited number of parties.

2. Application to the procurement context

The issue of conflicting treaty norms and their potential effect on the interpretation of the deviating clauses of the later treaty is relevant in the context of public procurement, and in particular in the context of the EU’s example above, in particular given the regulatory overlap and divergence between the GPA and the EU’s FTAs with Canada and Singapore.

Modification of the GPA between certain of its parties only. Given that the EU, Singapore and Canada are GPA parties, one possibility would be to consider their bilateral agreements as a modification of the GPA. In that regard, the GPA includes a specific clause on modifications (Art XXII(11) GPA) that would trigger Art 41 VCLT. However, Art XXII(11) GPA requires acceptance of the intended treaty modification by two thirds of the GPA parties, which makes bilateral modification of the agreement impossible (other than concerning schedules of coverage, which are structurally negotiated on a bilateral basis). Therefore, the relevant analysis is based on Art 30 VCLT.

Explicit conflict clause in the FTAs? It is worth noting that the EU-Singapore does contain a (pseudo)conflict clause regarding the GPA, according to which ‘If the Revised GPA is amended or superseded by another agreement, the Parties shall amend this Chapter by decision in the Committee on Trade in Services, Investment and Government Procurement … as appropriate’ (Art 9.20 EU-Singapore FTA). CETA contains another type of (pseudo)conflict clause in relation to the WTO Agreement and other agreements (Art 1.5 CETA), according to which ‘The Parties affirm their rights and obligations with respect to each other under the WTO Agreement and other agreements to which they are party.’ It is highly unlikely that either of these clauses qualifies as an explicit conflict clause meeting all requirements of Art 30(2) VCLT (above).

An unregulated conflict. Therefore, the potential conflict between the procurement provisions in CETA or the EU-Singapore FTA and the GPA seems to be governed by the rule in Art 30(4)(a) VCLT: ie the FTA prevails in the procurement relationship between the EU and Singapore/Canada, with the GPA provisions applying only to the extent they are compatible with the FTA. In any case and regardless of this interpretation, however, the rule of Art 30(4)(b) stands and, in their relationships with other GPA members, the EU, Canada and Singapore are bound by the GPA, regardless of any conflict with their FTA.

This makes the existence of unregulated ‘triangular situations’ particularly challenging where the FTAs deviate from the GPA in a manner that changes or limits the parties’ obligations towards each other (‘GPA-‘). Given that the extensions of coverage under CETA and under the EU-Singapore FTA are only incremental above the general coverage included in the GPA schedules for each of the parties, most of the procurement opportunities covered by the FTAs will be subject to dual regulation—or, in other words, most procurement covered by the EU’s FTAs will generate ‘triangular situations’.

GPA- situations pose significant legal uncertainty but, for the reasons below, they will likely lack any practical relevance if the EU (Canada and Singapore) want to avoid international responsibility for breach of the GPA.

What is the proper interpretation of the FTAs, then?

As above, a literal interpretation of Art 19.4 CETA would suggest that suppliers of either of the parties are protected under the national treatment regime only when they offer goods or services of the parties. This literal interpretation is clearly much more restrictive than the GPA and can affect the rights of suppliers offering non-EU/Canadian goods or services, as well as the access that should be afforded to those goods and services.

Conversely, a literal interpretation of Art 9.4 EU-Singapore FTA would suggest that suppliers of either of the parties are protected under the national treatment regime, even if they offer goods or services from third parties, whether those are GPA or not (unless domestic suppliers offering goods or services from third parties are also subjected to specific differential treatment—eg exclusion—which would set the benchmark of the national treatment obligation). Such an interpretation would significantly expand the scope of the national treatment under the EU-Singapore FTA compared to the GPA in relation to (non-GPA) third country goods and services, which does not seem to plausibly represent the parties’ intent under the FTA. However, such literal interpretation would follow Art 30(4)(a) VCLT.

In contrast, a systematic interpretation that took account of the fact that Art 9.4(2)(b) EU-Singapore FTA only refers to locally established suppliers offering goods or services of the other party (ie either EU or Singapore), would suggest an implicit requirement in Art 9.4(1) that suppliers are only protected as long as they offer EU or Singaporean goods or services. While this interpretation seems more aligned with the putative intention of the parties under the FTA, it does create problems in case of dual regulation procurement (or triangular situations), as EU and/or Singaporean suppliers offering goods or services of a different GPA origin could seek GPA national treatment of the goods or services themselves, which attempted exclusion (under a ‘GPA-’ approach) would likely breach Art 30(4)(b) VCLT.

To avoid breaching their international obligations under the GPA (Art IV(1)) in relation to the goods or services of other GPA parties offered by EU or Canadian/Singaporean suppliers, the EU, Singapore and Australia must refrain from any discrimination of covered suppliers offering goods or services originating anywhere in the ‘GPA club’. This means that, by virtue of the operation of the rule in Art 30(4)(b) VCLT in relation to the goods or services of the GPA parties protected by Art IV(1)GPA, the interpretation limiting the scope of protection under Art 9.4(1) EU-Singapore FTA cannot have any practical relevance. And, more challenging, it also means that the EU and Canada need to set aside Art 19.4(1) CETA.

Conversely, EU or Singaporean suppliers offering non-GPA goods or services may attempt to push for the maximalist literal interpretation sketched above, based on Art 30(4)(a) VCLT. However, given the explicit constraints in Art 9.4(2)(b) EU-Singapore FTA, this is unlikely to succeed under the broader rules of interpretation in the VCLT.

Conclusion

The (lengthy and technical…) analysis above shows that, given the pervasiveness of triangular situations, despite (intentionally?) deviating from the text of the GPA, FTAs are unlikely to generate practical effects where they vary or reduce GPA-based obligations. However, there is an important exception to this, which concerns cases covered by Art 30(4)(a) VCLT, where the effects are only for the economic operators of the two jurisdictions signing the FTA, which is both counterintuitive and prone to litigation, especially where GPA- situations concern access to procurement remedies (this is further assessed here).

Even if these views are correct and there is no practical effect resulting from the deviation in the text of the FTAs from the GPA standard wording, the mere existence of the legal uncertainty resulting from such deviation is undesirable. It is also difficult to ascertain whether any practical effects intended by the respective negotiating teams/governments deciding on the textual deviations are neutralised by the rules in the VCLT, in which case there may be knock-on effects concerning the balance of reciprocal concessions across the procurement chapter, or even across chapters, of the relevant FTAs. At any rate, in my view, a direct incorporation of the GPA rules and an explicit conflict clause giving the GPA precedence over the relevant FTA would be a preferable regulatory technique from this perspective, and one that has already been used by the UK, eg in the EU-UK TCA.

_________________________________________

[1] https://legal.un.org/ilc/texts/instruments/english/conventions/1_1_1969.pdf.

[2] Indeed, in some cases, they can avoid the ‘conflict’ altogether by agreeing to a GPA-compliant interpretation; see eg A Orakhelashvili, ‘Article 30’ in O Corten and P Klein (eds), The Vienna Convention on the Law of Treaties. A Commentary, vol. 1 (OUP 2011) 764, 776 at para 31; see also J Klabbers, ‘Beyond the Vienna Convention: Conflicting Treaty Provisions’ in E Cannizzaro (ed), The Law of Treaties Beyond the Vienna Convention (OUP 2011) 192, 203.

[3] However, as this is not a difficulty concerning the specific field of public procurement, this submission will not address it in any detail. For discussion, see A Orakhelashvili, ‘Article 30 of the 1969 Vienna Convention on the Law of Treaties: Application of the Successive Treaties Relating to the Same Subject-Matter’ (2016) 31(2) ICSID Review - Foreign Investment Law Journal 344-365.

[4] Subject to other general principles, such as those resulting from a conflict between lex posterior and lex specialis, see R Kolb, The Law of Treaties. An Introduction (Edward Elgar 2016) 190.

[5] Klabbers, above n 2, 194.

[6] Orakhelashvili, above n 2, 792 at para 75.

[7] International Law Commission, Draft Articles on the Law of Treaties with commentaries (1966) 271, para (11), https://legal.un.org/ilc/texts/instruments/english/commentaries/1_1_1966.pdf. See also K von der Decken, ‘Article 30’ in O Dörr and K Schmalenbach (eds), Vienna Convention on the Law of Treaties. A Commentary (Springer 2018) 539, 551.

[8] Kolb, above n 4, 186.

[9] Klabbers, above n 2, 195 and 204-205.