Public procurement, Brexit and the WTO GPA

Kamala Dawar
Kamala Dawar

In a recent briefing paper, I illustrate that the sheer value of public procurement contracts makes them important both economically and for providing society with essential public goods, services and infrastructure. In 2013/14, the UK public sector accounted for 33% of UK public sector spending and 13% of GDP3 – so ensuring good public procurement policy is beneficial to markets and taxpayers.

Currently, the UK’s procurement laws fall under the application of the EU’s 2014 Procurement Directives for Goods and Services, Utilities and Concessions. The EU has also negotiated the coverage of the WTO Government Procurement Agreement (GPA) on behalf of all 28 EU Member States and various PTAs, most recently the EU-Canada CETA, which includes a comprehensive chapter of public procurement provisions. The Great Repeal Bill aims to revoke the European Communities Act 1972 and to incorporate current applicable EU law into an Act of Parliament. Additionally, following the Devolution Settlement of 1998, certain competences – including public procurement – were devolved to Northern Ireland, Wales and Scotland . So, unless the laws affecting devolved issues are unilaterally scrapped by Westminster as a consequence of Brexit, the Great Repeal Bill will result in decentralising government procurement legislation. This could potentially fragment a coherent UK-wide procurement strategy towards the WTO GPA, as well as in its PTAs.

From a negotiator’s perspective, a prerequisite to repositioning the UK’s trade terms post-Brexit is therefore going to involve establishing the UK’s Most Favoured Nation commitments under
the GATT and GATS, with all the other 164 or more Members – including the EU. Some commentators have argued otherwise – that the UK is already a WTO Member with independent rights and obligations, including those relating to its MFN coverage in goods and services. This seems an optimistic and overly simplistic interpretation
 in the case of services markets committed under the GATS schedules, where the UK’s commitments are set out both independently
and jointly with the EU. The UK will need independently to set out its GATS Schedule whether or not it is certified by other WTO Members, because the UK needs a schedule upon which to trade. So, it is not until the UK has formally determined its MFN coverage under the WTO that the UK can seek to negotiate a more favourable trade agreement with the EU or other 3rd parties. After Brexit, a pragmatic short-term solution would be
to retain current regulations for the award procedures under the Great Repeal Act, but without conferring their benefits to suppliers from third parties without reciprocal arrangements. The freedom from the imposition of EU Procurement Directives will have implications for the
 UK’s internal procurement policy.

As a consequence of
the devolution settlement of 1998, public procurement became an area of responsibility for the devolved governments of Northern Ireland, Scotland and Wales. Following Brexit, the different parts of the UK will no longer be forced to apply the same public procurement rules, and different policy objectives are likely to appear in the award of procurement contracts, promoting different local economic development and social goals.

In 2013/14, the UK public sector spent a total of £242 billion on procurement of goods and services. There is political pressure to use this sum to pursue a variety 
of public policy aims, such as promoting small and medium-sized enterprises (SMEs) or encouraging local growth. Indeed, both of these objectives were stated
aims of the UK’s coalition government of 2010-2015, which in 2013/14, set a target for central government to procure 25% of goods and services by value from small and medium-sized enterprises. The 2015 Conservative manifesto included a pledge to increase the percentage spent with small and medium-sized enterprises to a third.

Recent WTO disputes indicate that procurement policies promoting industrial or environmental policies are actionable under various multilateral agreements, even if they have been exempted from the WTO GPA commitments. The UK will need to ensure that horizontal policy objectives implemented through devolved procurement awards are in compliance not only with the WTO GPA, but also under other multilateral rules including the WTO Agreement on Subsidies and Countervailing Measures, the GATT, GATS and the TRIMs.

The Court of Justice of the EU (CJEU) also issued a revised interpretation of the concept of an undertaking for the purposes of the application of EU competition law to encompass economic agents engaging in a combination of both economic and non-economic activities. In the EasyPay case, the CJEU determined that an activity will be considered as economic – unless it has links with another activity that fulfils an exclusively social function – based on the principle of solidarity and entirely non-profit making. Moreover, such an activity must, by its nature, aims and the rules to which it is subject, be ‘inseparably’ connected to its social function.

The EasyPay Judgment was significant in departing from existing case law to confirm that for the purposes of
the application of EU competition law, an undertaking is
any entity – even a procurement agency – engaged in an economic activity, irrespective of its legal status and the
way in which it is financed. And further, that any activity consisting in offering goods and services on a given market is an economic activity. Of additional relevance here is
the 2014 High Court in England ruling on the application of the UK competition rules to tender design of an exclusive concessions contract tendered to Luton Operations to run
a bus service between the airport bus terminal and central London. In this case, the contracting authority was found to have abused its dominant position by negotiating a seven- year deal with the successful bidder when there would have been sufficient capacity for a second operator after three years. The long exclusivity generated a higher return for Luton Operations, which was held to be bad for consumers and an abuse of dominance in the buying market.

Promoting this integrated approach to implementing competition and public procurement law and policy may also be helpful in counterbalancing the centrifugal forces of devolution, undermining the benefits of competition
in public procurement. For example, this could involve centralised monitoring of horizontal policy objectives through procurement awards in the different regions of
the UK, following a similar assessment and surveillance system. This would also provide a more transparent, coherent and competitive framework for potential bidders, which would be of particular benefit to SMEs that wish
to enter these lucrative markets. Currently the EU has 
one of the most developed State aid control systems
in the world. The UK is likely to continue to apply some form of State aid control following Brexit. Providing the
 UK competition authority with the mandate to oversee 
the monitoring and enforcement of competition law, State aid control and public procurement rules would help to ensure that decentralised legislation conforms to WTO commitments towards non-discrimination and subsidy control. Such centralised supervisory powers could also act as a counterweight against legal fragmentation, which could disproportionately undermine economies of scale
as well as the benefits of competition and value for money in public procurement following the devolution of these competencies.

The briefing paper concludes by hoping that what could be a relatively straightforward discussion concerned with improving transparency, competition and value for money when awarding public procurement contracts, is not overshadowed by complex sequencing of negotiations, intra-UK jurisdictional divergences, and intractable political legacies with the EU.

Kamala Dawar is a Lecturer in Commercial Law at the University of Sussex. The briefing paper was published for the UK Trade Policy Observatory on 28 March 2017. The opinions expressed are those of the author alone and do not necessarily represent the opinions of the University of Sussex or the UK Trade Policy Observatory.

Can a notification under article 50 be unilaterally withdrawn?

Paul Eden
Paul Eden

Now that the European Union (Notification of Withdrawal) Act has received Royal Assent, the UK government is on track to meet its deadline of invoking Article 50 of the Treaty of European Union (TEU) by the end of March. Whilst it seems inevitable that the UK will indeed invoke Article 50 a key question that remains is whether we can change our minds and stop the whole process, perhaps when we are a year down the line or if there was a change in government in the UK.

In the Miller case it was common ground between the parties “that notice under Article 50(2)… cannot be given in qualified or conditional terms and that, once given, it cannot be withdrawn” (at [26]) although, as Lord Carsworth noted in his dissenting judgment, this assumption is “possibly controversial” (at [261]). This blogpost addresses the possible legal basis of any unilateral right to revoke a notification of a notice of withdrawal made in accordance with Articles 50(1) and 50(2) TEU by examining, first, the potential applicability of Article 68 of the 1969 Vienna Convention on the Law of Treaties (VCLT). Secondly, the blogpost examines Article 50 TEU to determine whether the express wording implicitly excludes the right of unilateral revocation of notification of intention to withdraw from the EU. Finally, the blogpost argues that even if the other members of the EU unanimously agreed to allow the United Kingdom to revoke a notification of withdrawal, this might amount to an amendment of Article 50 TEU and present practical problems for Member States whose national law requires specific constitutional obligations at a national level for amendments to EU Treaties to take effect.

The applicability of the Vienna Convention on the Law of Treaties

The 1969 Vienna Convention on the Law of Treaties is a multilateral treaty governing the law relating to the key aspects of treaty conclusion, application, interpretation and termination. It is only binding with regard to treaties concluded between states that are parties to the VCLT but many of its provisions are regarded as reflecting customary international law and are binding on parties and non-parties to the VCLT alike.

The starting point for many of the arguments that a notice of withdrawal made in accordance with the provisions of Article 50 TEU can be unilaterally revoked appears to be Article 68 VCLT which states that

A notification or instrument provided for in Articles 65 or 67 may be revoked at any time before it takes effect. [Emphasis added]

The italicised words are often omitted in discussions about the applicability of Article 68 VCLT to the Article 50 TEU debate. Article 65 VCLT is concerned with the procedure to be followed where a party to a treaty invokes either a defect in its consent to be bound or a ground for impeaching the validity of the treaty. Article 67 VCLT is concerned with the modalities of this particular notification process. As the proposed British withdrawal from the European Union is not based on either a defect in its original consent to be bound or on an assertion that the treaties on which the EU is based are invalid, there does not appear to be a basis for the application of Article 68 VCLT unless, as has been asserted in a House of Commons Library Briefing Paper, “it is a general principle of international law” (at page 10).

The VCLT cannot apply directly to the Treaty on European Union (TEU) because France and Romania have never been parties to the VCLT and Malta only acceded to the VCLT after the TEU was signed on 13 December 2007. In several cases, the Court of Justice of the European Union (CJEU) has acknowledged that the VCLT may still be relevant to the extent that its provisions reflect customary international law. See Case C-162/96 Racke GmbH & Co v Hauptzollamt Mainz [1998] ECR I-3688 [24] and Case C-386/08 Brita GmBH v Hauptzollamt Hamburg-Hafen [2010] ECR I-1289 [42].

Is Article 68 VCLT a customary norm of international law?

In the North Sea Continental Shelf Cases the International Court of Justice (ICJ) acknowledged that “fundamentally norm-creating” provisions of a multilateral treaty can constitute the foundation of a rule of customary international law binding on states (at [72]). It is however questionable whether the essentially procedural stipulation contained in Article 68 VCLT would fall within the category of a fundamentally norm-creating provision that can potentially form the basis of a general rule of law in any event. In the Case Concerning the Gabčíkovo-Nagymaros Project (the Danube Dam case) the ICJ noted that Articles 65-67 VCLT contained “procedural principles which are based on an obligation to act in good faith” (at [109]) and that both parties to that case were agreed that these Articles “if not codifying customary law, at least generally reflect customary international law” (ibid). It must be stressed however that it was the parties to the case and not the ICJ itself that asserted that Articles 65-67 VCLT reflected customary international law and, given the fact that there have been numerous reservations to Article 66 VCLT, this assertion of the customary status of Articles 65-67 VCLT should be approached with caution.

Some commentators are doubtful that Article 68 VCLT can be considered customary law (see, for example, A Tzanakopoulos “Article 68” in The Vienna Convention on the Law of Treaties: A Commentary Volume II (O Corten and P Klein (eds)) (OUP, 2011) 1565) but, even if Article 68 VCLT is to be regarded as a customary norm, its customary status must been seen in the context of the “complex and inter-related” (Sir Ian Sinclair, The Vienna Convention on the Law of Treaties (2nd edn, Manchester UP, 1984) 159) provisions contained in Articles 65-68 VCLT. Put simply, Articles 65-68 VCLT contain an obligation subjecting a particular class of treaty withdrawals to a dispute settlement process. The outcome of this dispute settlement process may well be the realisation that the withdrawing state’s consent to be bound was not in fact defective and/or there were in fact no valid grounds for impeaching the treaty. Consequently, it is entirely appropriate to give effect to the possible success of a dispute settlement process held in accordance with Article 66 VCLT by expressly providing that the original instrument of withdrawal can be revoked in Article 68 VCLT.

What is generally agreed is that Article 68 VCLT reflects procedural principles stemming from an obligation to act in good faith and, in the Interpretation of the Agreement of 25 March 1951 between the WHO and Egypt Advisory Opinion, the ICJ stressed the need to “have reasonable regard to the interests of the other part[ies] to the treaty” (at [47]) when exercising a right of withdrawal from a treaty.

Interpreting Article 50 TEU

Even if no right to revoke a notification of an intention to withdraw from the EU can be found in customary international law, such a right might be found, expressly or impliedly in the wording of Article 50 TEU itself. Alternatively, if one believes that Article 68 VCLT embodies a general principle of customary international law that permits the revocation of a notification of withdrawal at any time before it takes effect, the wording of Article 50 TEU may implicitly exclude the application of this alleged general principle.

A recent legal opinion entitled In the Matter of Article 50 of the Treaty on European Union, written by five eminent QC’s including a former Judge of the Court of Justice (Sir David Edward) and a former Advocate General (Sir Francis Jacobs), argues that

The language of Article 50 does not require a Member State’s decision to withdraw from the European Union to be irrevocable or unconditional prior to it being notified. The use of the word ‘intention’ in Article 50(2), and the present tense ‘which decides’, rather than ‘has decided’, allows for the possibility that a Member State may change its decision and, therefore, its intention. [49(ii)]

It is equally true to say that Article 50(2) TEU does not expressly permit a conditional or revocable notice of withdrawal and there is no evidence to suggest that the United Kingdom intends to submit a conditional notice of withdrawal in any event. Further, as the TEU is equally authentic in 23 languages (Article 55 TEU), any argument entirely premised on the use of the present tense of the verb “decides” in the English text of the TEU is rather less than convincing.

As the European Union (Notification of Withdrawal) Act satisfies the constitutional conditions laid down by the UK Supreme Court in the Miller case, the UK has satisfied the obligation contained in Article 50(1) TEU that any decision to withdraw from the EU must be in accordance with the withdrawing state’s constitutional requirements. Professor Paul Craig (‘Brexit: a drama in six acts’ (2016) 41 European Law Review 447, 464) has argued that if a Member State subsequently changed its mind

it could be argued that if this occurred there would then no longer be a valid decision to withdraw, since the original decision had been changed in accordance with national constitutional requirements.

This argument cannot be reconciled with the doctrine of inter-temporal law that acts should be judged in the light of the legal position at the time of their creation. Even if the European Union (Notification of Withdrawal) Act had contained a provision giving both Houses of Parliament the right to veto any withdrawal agreement (the so-called “meaningful vote” amendment inserted by the House of Lords but rejected by the House of Commons), Article 27 VCLT embodies the customary norm that a state “may not invoke the provisions of its internal law as justification for its failure to perform a treaty” and thus any such provision could not be invoked as a justification to revoke a notice of withdrawal unless there was a pre-existing right in either general international law or specifically provided for by the wording of Article 50 TEU.

Does Article 50 TEU embody a right to revoke a notification of withdrawal from the EU?

The key problem with the argument that a right to revoke is implicit in the wording of Article 50 TEU is that Article 50(3) TEU provides that, in the absence of a withdrawal agreement, “[t]he treaties shall cease to apply to the State in question…. two years after the notification [of intention to withdraw]…, unless the European Council, in agreement with the Member State concerned, unanimously decides to extend this period.” The requirement that an extension of the two-year period requires unanimous consent is in sharp contrast to the qualified majority required for the withdrawal agreement under Article 50(2) TEU. Further, Article 50(5) TEU makes it clear that any state that has withdrawn from the EU under Article 50 that wishes to re-join must make a new application under Article 49 TEU. The wording of Articles 50(3) and 50(5) TEU taken together support the conclusion that the unilateral revocation of a notification of withdrawal is not permitted. As noted above, the Interpretation of the Agreement of 25 March 1951 between the WHO and Egypt Advisory Opinion is authority for the proposition that the United Kingdom has an obligation to have reasonable regard for the interests of the other Member States of the EU when exercising its right of withdrawal under Article 50 TEU. Any unilateral right to revoke a notice of withdrawal would not have reasonable regard for the interests of the other Member States of the EU given the wording of Articles 50(3) and 50(5) TEU.


While there is no doubt that the other Member States of the EU could unanimously agree to allow the United Kingdom to revoke a notification of withdrawal, if the Court of Justice of the European Union (CJEU) did not accept that the Member States implicitly possessed this power under the current wording of Article 50 TEU, this magnanimous act would amount to an amendment of Article 50 TEU and this might present practical problems for Member States whose national law requires specific constitutional obligations at a national level for amendments to EU Treaties to take effect (such as referendums etc.).


Paul Eden is Senior Lecturer in Law at the University of Sussex. This  article was originally published in the UK Trade Policy Observatory’s blog. The opinions expressed are those of the author alone and do not necessarily represent the opinions of the University of Sussex or UK Trade Policy Observatory.

Trump’s immigration ban and Europe: the courage to raise a mirror and look beyond it

Nuno Ferreira

Since President Donald Trump issued an Executive Order on 27 January 2017 entitled ‘Protecting the Nation from Foreign Terrorist Entry into The United States’, traditional media outlets and social media have been in a frenzy of activity. Protests have taken place around the world and there have been reports of dramatic scenes at airports. Indignation against the Executive Order was understandable: although couched in the language of prevention of terrorism and protection of national interest, the discriminatory and arbitrary character of the measure was apparent. Indeed, the Executive Order suspended the issuance of ‘Visas and other Immigration Benefits to Nationals of Countries of Particular Concern’, namely Libya, Iran, Iraq, Somalia, Sudan, Syria and Yemen. Moreover, the Executive Order suspended the U.S. Refugee Admissions Program (USRAP). Such discrimination against Muslim-majority countries and arbitrary refusal of refugees rightly appalled politicians and civil society alike in countries around the world – especially when considering this measure in the context of all Donald Trump’s initiatives, statements and background of a discriminatory and bigoted character. This original Executive Order has been in the meantime suspended by a US federal judge, following a string of lawsuits and on the basis of a range of legal grounds. Yet, a revised, softer version has been issued on 6 March 2017, which predictably prompted immediate academic criticism and judicial actions. And whilst most people – including myself – may find these orders abject measures and a violation of international and human rights law, including children’s rights and refugee rights, I cannot help but looking around me and wondering how much better Europe is faring in dealing with its borders, its migration policy, and its international protection obligations.

Looking in the mirror and dealing with the reflection

In the midst of these events, a question kept cropping up in my mind: how different are we, in Europe, really? In Calais, a wall has been built with British funds. In Hungary, a wall prevents refugees from reaching the country and seeking asylum, and all asylum seekers are to be automatically detained, despite not having committed any criminal act. In Greece and Italy, ‘hotspots’ aim to ‘assist frontline Member States which are facing disproportionate migratory pressures at the EU’s external borders’, but the system has been accused of instead ‘abusing, misleading and expelling’. Refugees are reported to be currently dying of cold across Europe. EU resettlement programmes have been accused of cynicism and of being disgraceful. Australian-style proposals to ‘offshore’ EU asylum have been rightly labelled as ‘mass expulsions’ and ‘fiscally irresponsible, morally bankrupt, and increasingly unsustainable politically’. The EU-Turkey agreement to resettle thousands of ‘irregular migrants’ reaching Europe has been widely criticised as a source of human rights violations, and illegal and ineffective. The cooperation agreement with Libya has been criticised for endangering human rights. After only resettling 350 children from Calais to the UK out of the 3,000 expected, the UK will end the Dubs scheme. Europe’s own visa policies have long been accused of being inherently discriminatory. The list is endless.

Other commentators have not failed to notice this shocking state of affairs in our own backyard: Nikolaj Nielsen, for example, writes about European Union’s hypocrisy. This is in no way downplaying the seriousness of Donald Trump’s Executive Order – it is clear that it is a blunt instrument, with no appropriate reasoning sustaining it. And this criticism does not aim either at equalising the way European leaders and Donald Trump deal with such matters – evidently, there are significant differences in form and content. Nonetheless, that does not make European policy unproblematic. In fact, there is much to be done to render European law and policy more humane.

Looking beyond the mirror

This is not to say that European organisations and countries have ignored their human rights and international protection obligations completely. So, credit where it is due. Germany has taken more than its fair share of refugees in the European context. European Union institutions have pursued relentlessly a refugee relocation ‘burden-sharing’ deal on the basis of the principle of solidarity. The European Commission, in particular, has frequently put pressure on Member States to act upon their solidarity obligations in the field of migration. Furthermore, Advocate General Paolo Mengozzi, in an Opinion to the Court of Justice of the EU, has argued in favour of public authorities in the EU having to issue visas on humanitarian grounds to fulfil their international protection obligations. Although the Court of Justice eventually decided against the existence of this obligation under EU law, the debate that this case prompted was very positive. Nonetheless, economic pressures, populist movements across Europe and lack of political solidarity between European countries are preventing us from fulfilling our obligations in a comprehensive and timely fashion. Crucially, any progress we have made on this front is constantly under threat.

So what is preventing us from doing much better? The usual: lack of political will and leadership. As there is no magic solution to improve those, it is important to discuss other possible, more immediate solutions. Although there may be plenty of room for improvement of the Common European Asylum System (CEAS), the greatest causes for concern seem to relate to the ways the system is (mis)implemented. For example, the Dublin regulation mechanism does allow for countries to process asylum requests even when they are not the EU member state the claimant first reached (as asserted by the Court of Justice of the EU in NS v SSHD); yet, the Commission has proposed resuming transfers to Greece, ignoring the extremely poor reception conditions in Greece and the need to pursue ‘burden sharing’ policies instead. So, the Dublin regulation mechanism needs to be applied more sensibly. The current CEAS could be complemented with greater economic support for EU member states accepting to process applications of asylum seekers who first reached the EU through another member state – not that they should be doing it for economic benefits alone, as humanitarian and legal grounds should suffice, but greater economic support could help some member states become more supportive of greater application of the principle of solidarity amongst EU member states.

Moreover, if we are not to throw away what CEAS has achieved and simply improve the way it is implemented, we need to ensure that any implementation measure is compliant with the highest human rights standards. Working closely together with the Council of Europe and ensuring absolute respect for the minimum standards established by the European Court of Human Rights is essential in this context. Although accession by the EU to the European Convention on Human Rights is on hold since the Court of Justice of the EU issued its Opinion 2/13, there is still plenty of room for close collaboration with the Commissioner for Human Rights and the other Council of Europe bodies concerned with asylum related matters. This should most likely lead to the suspension of the agreements with Turkey and Libya, mentioned above. Although much more contentious, perhaps the EU should also consider a much stronger stance with regard to member states ferociously undermining asylum-seekers’ right to international protection, such as Hungary (discussed above). As a last resort, this could mean triggering Article 7 of the Treaty on the European Union, which could lead to the suspension of the voting rights of the member states in question in the European Union bodies. Admittedly, this is the least palatable measure, especially considering that this provision has never been used and that we are living in the midst of Brexit. The EU would perhaps not wish to poke the pride and sovereignty of any member state, in an attempt to avoid more exits. And yet, in the long run, tolerating club members that are in many respects not worthy of membership to the club only serves to denigrate the reputation of the club itself.

Whether we call it a ‘crisis’ or not, the truth remains that the right to international protection of thousands, if not millions, of individuals is being violated. The EU and its member states need to act swiftly. Focus and resources should be redirected from protecting ‘Fortress Europe’ to safer and more human rights compliant alternatives, such as allowing refugees to fly away from persecution and reach Europe through resettlement programmes. The causes of refugee movements should also be addressed more forcefully, by tackling head-on military conflicts and humanitarian crises pushing people out of their houses. None of this is to soften condemnation of Donald Trump’s immigration ban in all its ugliness. Yet, it is essential that Europeans across the whole continent remain alert and engaged: we are failing in our human rights and international protection obligations as well. Perhaps not to the same extent. Perhaps not so blatantly, arbitrarily or idiotically. But still, we need to do a lot more for all refugees in Europe and those trying to reach Europe, if we wish to be able to criticise other countries with a clear conscience. So, whilst remaining united and active against any measures such as Donald Trump’s immigration ban, let us keep working for a better deal for refugees knocking at our own door as well. Donald Trump’s immigration ban is a massive wake up call to all of us on how human rights anywhere are so fragile. To make it worse, Donald Trump’s extreme right-wing views are offering an extra boost to the already growing Europe’s own extreme right wing movements. Greater humaneness across all policies must be the European answer. Individuals and NGOs around the world have already showed us the power of their voices. Now political leaders across Europe need to also do the same – and in actions, not just words.

Nuno Ferreira is a Professor of Law at the University of Sussex. A previous and shorter version of this blog post has been published in the Forced Migration Forum on 6 March 2017, and can be found on Many thanks to Carmelo Danisi, Moira Dustin, Nina Held, Paul Statham and Alex Aleinikoff for their comments on previous drafts.