1This chapter discusses empirical findings related to our third and last hypothesis (H3). It argues that, in the case of provisionally enforced mixed agreements, political ratification obstacles (PROs) tend to be reduced as a function of lock-in dynamics operating at the level of both official veto players (specifically, national parliaments) and private ones like advocacy groups. It concludes, therefore, that provisional EU enforcement emerges as the key challenge for the Commission in its striving for the survival of a mixed FTA. Lock-in dynamics observed in our empirical analysis are discussed in sequence in what follows. Institutional lock-ins are empirically explored through the case study of Italy and CETA ratification. Observations on the lock-in of advocacy groups are grounded on the empirical analysis of dynamics within the CETA Domestic Advisory Group (DAG).

8.1 Mixed FTAs and the locking-in of veto players

2This section moves to further elaborate on what policy dynamics can aid the Commission when tackling PROs to ensure the successful enforcement of mixed FTAs. We specifically focus on mixed agreements as their lengthy procedure of national-level approval has been considered a key source of ratification hurdles for Brussels (Roederer-Rynning and Kallestrup 2017; Van der Loo 2017). However, based on our empirical exploration of the CETA, this section argues that mixed FTAs do not pose greater challenges to enforcement than other agreements. We posit that once a mixed FTA (like the CETA) is provisionally enforced at the EU level, lock-in dynamics will prevent a formal veto by Member States and contestation from civil-society lobbies, both conceived as veto players in new-generation EU trade.

3 With regard to official-level policy actors, we preliminarily explore lock-in dynamics arising within Italy’s parliamentary majority ever since the provisional enforcement of the CETA, in September 2017. We regard Italy as a powerful instance of the “joint decision trap” “at play” (Scharpf 1988). Shortly after the CETA’s provisional enforcement, the Italian parliamentary majority was reversed, determining a political shift unobserved in other EU Member States (European Parliament 2019b: 161). Yet, Italy’s new political leadership, albeit having campaigned against the CETA, quickly abandoned the option of a veto, keeping the FTA off the parliamentary agenda. Although Italy is a priori least likely not to resort to a veto on the CETA, the fact that it has consistently avoided a formal rejection of the agreement points well at the policy implications of unanimous voting. Namely, in ongoing decision-making systems where EU Member-State consensus is required for policy reforms to be enacted, “individual action is no longer permitted, and the veto of one or a few governments will prevent all others from correcting or abolishing the rule in response to changed circumstances or preferences” (Scharpf 2006: 848). This is especially the case with FTAs, in which gradual domestic adjustment has been the prevailing modus operandi, while a “take it or leave it” approach more of an exception (Katzenstein 1985). This makes the prospects of a simply straightforward veto quite remote, especially if an agreement like the CETA is already provisionally in force.

4 As regards civil-society advocacy groups engaging in trade veto-playing, we suggest that the CETA grants unprecedented institutional weight to this category of actors, in charge of monitoring and advising provisional implementation. In empirically backing this claim, we explore the single case of the CETA Domestic Advisory Group (DAG). Adopting a political-economy perspective, our analysis suggests that DAG members are better off being an active part of the implementation process. Lobbying advantages related to coalition-building, information-sharing, and resource-efficiency within a DAG make the prospect of advising and shaping enforcement more desirable (and indeed productive) than pushing for a straightforward veto. A national rejection of the CETA would in fact put an end to the implementation-monitoring role these actors have carved out for themselves. We ground our empirical exploration on primary documents and semi-structured interviews. In that regard, as lock-ins constitute long-term processes observable across several years in the life of a polity, the task of this section is mainly to preliminarily map dynamics that are deemed conducive to lock-ins as part of provisional enforcement (Pierson 2000).1 We move to discussing the cases of Italy and the CETA DAG in what follows.

8.2 Case studies

Italy and CETA ratification

5Qualifying as a least-likely case for H3 to be verified, Italy is methodologically powerful in reflecting how the unanimous provisional enforcement of a mixed FTA by the EU can lock Member States into a given trade regime. This section suggests that Italy’s handling of CETA ratification mirrors well mechanisms stemming from the two key independent variables accounting for the “joint decision trap”: i) unanimous decision-making; ii) a given policy qualifying as an “ongoing decision system” (Scharpf 1988). We already discussed how these two conditions are fulfilled by the CETA. Here, we move to assessing how Italian ratification dynamics mirror these two conditions, providing empirical support to our theoretical framework.

6 When the CETA was provisionally enforced, in September 2017, Italy was led by a centre-left government headed by Prime Minister Gentiloni. The free-trade oriented, liberal agenda of the Gentiloni Government made the decision essentially low-key, also given that the Italian constitutional system does not require specific parliamentary mandates ahead of a vote in the Council. In June 2018, following general elections, the Gentiloni Government was replaced by a coalition between the Five Star Movement and the League, both having heavily campaigned against the Italian ratification of the CETA. The manifesto on which the new coalition was built openly stated that both political forces would oppose any FTA, like the CETA, if it lacked regulatory guarantees for citizens (Five Star Movement, the League 2018: 54). Even more strikingly, as an MEP, the League’s leader Matteo Salvini had sponsored an EP motion calling for the rejection of the CETA as part of the ordinary legislative procedure (European Parliament 2017b).

7Against this background, the rapidity with which the new executive softened its position on the CETA after taking office is telling of the difficulties in exiting an EU policy regime when it is already in place. Ministers avoided referring to a veto, focusing instead on more specific issues within the text. Their focal point was the regulations on geographical indications (GIs) and the risk of a liberalisation of Canadian Italian-sounding products accessing the Single Market (Bartoloni 2018). Yet, most tellingly of the inherent constraints of EU unanimous voting was the position of the Italian Ministry of Foreign Affairs (MFA), directly in charge of EU policies. The MFA adopted a position of neutrality, as Minister Moavero Milanesi consistently avoided taking an explicit stance on the CETA for the entire duration of the Government. Our empirical analysis suggests that the main overview of CETA ratification provided by an MFA official came from the Italian Permanent Representative to the EU, Ambassador Massari, during a parliamentary hearing in Rome (Chamber of Deputies of the Italian Republic 2018). Coming from a non-political official, remarks were limited to acknowledging that Italy was running a CETA impact assessment across different economic sectors, as other EU Member States yet to ratify were doing (18). As observed in the theoretical formulation of our argument, the mechanism through which unanimous voting prevents Member States from simply abandoning existing policy commitments lies in the intention not to make a unilateral decision directly affecting other favourable EU partners. This is especially the case when such a decision can sanction the suspension of an existing trade commitment for all. In this sense, the high degree of integration reached by the Single Market implies that the decision of one determines a comparatively higher impact on other EU members than in policy domains where integration is still looser. Similar concerns are mirrored by the attention of the Italian MFA not to translate the position of majority parties into a formal diplomatic stance.

8Italian diplomatic cautiousness was also a function of the rapid evolution of ratification debates domestically. These shifted from initial emphasis on a veto to the more moderate acknowledgment of the need for adjustment ahead of national ratification. This change of attitude on the part of critical veto players sheds light on how the “joint decision trap” manifests itself in the context of “ongoing decision systems” like FTAs, which can undergo reforms along the way (Scharpf 1988). In this respect, it is striking to observe how Five-Star leader and Economic Development Minister Di Maio shifted from challenging the CETA tout court to a more moderate and technical examination of its provisions in view of proposing amendments and safeguards. Shortly after entering office, in July 2018, Minister Di Maio declared that “[the governing] majority [would] reject CETA” and that civil servants still manifesting support for the agreement would be removed from office (Barone 2018). By the following November, Di Maio’s position had visibly become more moderate, and his Ministry set up a task force in charge of carrying out a cost-benefit assessment of the CETA’s impact. The group specifically focused on GIs provisions within the CETA, and their impact on Italy’s agricultural sector. An otherwise low-salience policy domain at the EU level, GIs concerned the Italian Government, and farmers’ organisations gravitating around it – above all, Coldiretti. A major issue were the so-called grandfather rights of Canadian Italian-sounding products potentially accessing the Single Market as a result of the CETA. Yet, Di Maio abandoned his emphasis on a veto, pushing instead for additional safeguards on Italian GIs. Tellingly, at the third meeting of the Ministry Task Force, the Canadian Ambassador to Italy was also present, and discussions focused on how the existing CETA mechanism for registering EU GIs under Canadian copyright law could be relied upon to strengthen protection of Italian agricultural and food sectors (Ministry of Economic Development 2018a). During a question time at the Senate, Minister Di Maio recognised how CETA modifications of Canada’s Trademark Act “[could] offer a further degree of protection to [Italian] Geographical Indications, and also the possibility to raise claims against misuses and forgeries within the Canadian market […]” (Ministry of Economic Development 2018b). Further lowering the prospect of Italian CETA contestations re-sparking is the fact that ratification has been off the agenda of the new Five-Star centre-left governing coalition, as recently confirmed by Junior Minister of Foreign Affairs Di Stefano (Italian Ministry of Foreign Affairs 2019).

9 Overall, the case of Italy and the CETA’s outstanding ratification is particularly telling of how the provisional enforcement of a mixed FTA by the Council opens up a policy path that is difficult to deviate from. This is the case, we argue, even in the presence of unfavourable political majorities within Member States. In particular, challenging unanimous voting, by deviating from unanimously agreed upon policy paths, would especially clash with the interests of more Eurosceptic governments. Putting consensus into question would mean challenging the legitimacy of the very decision-making procedure allowing national executives to retain their acquis vis-à-vis community approaches to EU policies. In this light, albeit still awaiting full de jure ratification, the Commission sees the CETA de facto implemented for the most part, and arguably secured from potential debates re-sparking.

Civil Society and the CETA DAG: From demonstrations to advisory tables

10The CETA DAG was set up under the auspices of the European Economic and Social Committee (EESC), in pursuit of Chapters 22 and 24 of the Agreement (European Economic and Social Committee 2020a). The specific task of the Group is that of monitoring and advising the implementation of the CETA’s TSD chapter. Adding particular relevance to the DAG is the fact that the CETA Joint Interpretative Instrument (JII) provides for an early review of the TSD, falling under the Group’s competences (Council of the European Union 2016c). The DAG is divided into three main clusters, respectively business groups, trade unions, and NGOs whose mandates range from environmental protection to animal welfare (European Economic and Social Committee 2020b). In this section, we privilege attention to unions and advocacy NGOs, as their vocal opposition to the CETA makes observed lock-in dynamics analytically meaningful.

11DAG unions and NGOs, albeit in principle critical of a number of CETA provisions (including TSD), enjoy a number of lobbying assets arguably incentivising their membership. This section explores how these interest groups would face greater constraints if they were to operate outside of the DAG, following a national veto putting an end to provisional enforcement. An interviewed DAG member observed: “We all know, in NGOs, when we work on trade, that […] our aim is the EU Parliament at least to say no. Because once [an FTA] is provisionally implemented, it is very difficult… It drops under the radar, like CETA, in many ways, and then it also starts to be implemented.”2 In this light, advocates will arguably seize the opportunity of DAG membership to elicit desired reviews of the CETA, rather than pushing for an unlikely national veto. Thus, we argue that lobbying assets deriving from DAG membership can lock critical veto players into the implementation of an FTA they in principle oppose. Lock-ins related to the CETA DAG are observable in terms of three types of incentives making the prospect of exiting comparatively costly: i) coalitional incentives, ii) epistemic incentives, and iii) resource incentives. We discuss these in sequence.

12Coalitional incentives to DAG membership are observable within the Labour and NGO subgroups. The two emerge as particularly conducive to advocacy coalition-building. Potential collective-action problems amid interest groups are obviated by the presence of ideological proximity between actors (Rozbicka 2013: 840-841). Based on our dataset,3 trade unions in the CETA DAG fear an erosion of EU labour standards as a result of opening up to a more deregulated system like the Canadian one. They also oppose a potential EU regulatory chill on investment-protection standards. Specifically, several statements and position papers by DAG members are still highly critical of the CETA’s ICS dispute-settlement mechanism, deemed not to provide enough guarantees to national judiciaries vis-à-vis transnational corporations (CGIL 2016b; EPSU 2016). Similar concerns on ICS are shared by member NGOs, despite their mandates focusing mainly on environmental and animal-welfare issues (ClientEarth 2016a). Another key point of contact between unions and NGOs lies in the fear of lowered food standards, as a result of agricultural trade with Canada. In this respect, while unions ground their position on the fear of Canadian dumping disadvantaging EU farmers, NGOs point at the possible regulatory chill in EU standards on chemical products in agriculture and animal welfare (ClientEarth 2016b; Eurogroup for Animals 2017). Thus, DAG unions and NGOs show different background systems of “deep core” beliefs but converge in their lobbying over “secondary aspects” more strictly pertaining to how specific provisions should be designed and implemented (Sabatier 1988: 103-104). This suggests how advocacy coalition-building can be seen as a function of genuinely institutional dynamics. As DAG members institutionally interact, they leverage their ideological proximity, coordinating ahead of policy meetings, and exercising their role cohesively vis-à-vis the Commission or competing business lobbies in the Group.

13At an epistemic level, incentives from DAG membership relate to the high technicality of policy issues being discussed, which implies that effective lobbying requires ample coordination and information-sharing within partaking coalitions. In this regard, a DAG member observed how trade policy “is very complex and technical, […] less funded than other areas” and that FTAs are so comprehensive that “you need an expert for almost each chapter […].”4 Pointing at the need for information-sharing, our dataset highlights the limited number of union and NGO representatives in the DAG. Out of thirteen examined organisations, nine of them have only one delegated representative, while four have a second member accredited as an alternate.5 This arguably poses constraints to single organisations, especially in view of an upcoming amendment of the CETA TSD chapter, whose broad commitments are particularly challenging to translate into technical policy provisions. In the face of epistemic hurdles, we were reminded, for instance, how small environmental NGOs hold preparatory meetings in-between formal sessions of the Advisory Group, collectively drafting papers for discussion.6 This constitutes the epistemic side of advocacy coalition building. On the one hand, it allows advocates to better keep track of issues on the agenda, while also enabling organisations to formulate collective arguments, concerns, and policy proposals to be advanced during discussions with Commission officials.

14Regarding resource incentives, we suggest that examined organisations tend to benefit from DAG membership, especially considering the limited financial, human, and organisational resources they have available. From a political-economy perspective, the resource disbursement required to reach equal visibility by acting outside of the DAG would likely be higher than by operating within the Group. Looking at the average funding allocated by CETA DAG unions and NGOs on EU lobbying, our dataset shows that available resources are unsurprisingly lower than what is on average allocated by business groups.7 We chose the EU car-sector lobby ACEA as a term of comparison, for the latter was able to successfully demand adjustment measures on KOREU duty drawbacks, albeit not being part of an institutionalised EU implementation group. According to the EU Transparency Register, ACEA allocates between 2.75 and 3 million euros on its lobbying activities in the EU (European Commission 2020c). In contrast, the average sum available to examined CETA DAG organisations lies between 600,000 and 650,000 euros.8 Numbers and figures per se do not tell us much about the actual lobbying power of EU interest groups. As argued before, civil-society advocacy grounds its leverage on the high politicisation it can elicit on a given issue – rather than on heavily funded technical lobbying. Yet, levels of funding still help make sense of the extent to which these organisations are structurally dependent on institutions like DAGs. Crucially, limited funding translates into limited personnel, especially on trade issues. In that regard, while ACEA involves an average of thirteen people working full time on EU lobbying, examined CETA DAG members limit themselves to an average of seven (European Commission 2020c).9 Lastly, it is relevant to observe how the majority of trade unions in the CETA DAG are national organisations lacking an ad hoc EU branch, and operating mainly within single Member States.10 Thus, Brussels-based institutional windows like the EESC and the DAG constitute a key point of contact between largely nationally oriented organisations and technical discussions conducted by the Commission over specific FTA chapters.

15Overall, by taking part in DAG monitoring and advisory activities, citizen groups can see their ability to shape EU trade policies maximised, arguably more than would be the case if they were simply operating as outsiders, calling for the suspension of an agreement tout court. In this sense, advocates’ pushing for policy change, at least in the case of provisionally enforced EU FTAs, locks these actors into the very implementation of trade regimes they in principle regard as suboptimal.