Abstract
Brexit has occasioned a rightward shift in British politics as successive leaders have grappled with the difficulties of negotiating with the European Union and the vicissitudes of politics in the governing Conservative party. Explanations for the hardening of Eurosceptic preferences focus on the demands of ‘taking back control’ and the polarisation of post-referendum politics as key drivers. But they have not explored the ways in which negotiation strategies shaped – rather than reflected – domestic political developments. Drawing on two-level games accounts of ‘synergistic’ bargaining, this article argues both David Cameron and Theresa May sought to leverage Eurosceptic sentiment in their respective negotiations to make it more credible the United Kingdom would walk away if its demands were rejected. While both leaders failed to convey their resolve, they inadvertently strengthened Eurosceptic constituencies back home, contributing to the paucity – and the rejection – of their negotiated agreements.
Introduction
The United Kingdom’s protracted withdrawal from the European Union (EU) has brought countless surprises over the years. From Cameron’s announcement of an in/out referendum in 2013 to the shock decision of the British public to leave the Union in June 2016, and from Theresa May’s ‘hard’ interpretation of the Brexit mandate to the unprecedented rejection of her Withdrawal Agreement in early 2019, the Brexit process has seldom been smooth. Taken together, these distinct moments have also revealed a deeper and more puzzling process of hardening preferences in which concerns about the United Kingdom’s position in the EU have morphed into sustained support for a clearer break with the EU among key public and elite constituencies. Explanations for this rightward shift have focused either on the incompatibility of British demands for ‘taking back control’ with core principles of European integration, such that a shift towards a ‘harder’ outcome was always on the cards (e.g. Donnelly, 2018; Oliver, 2017; Rogers, 2019), or on the role of domestic factors such as May’s status as an insecure Remain supporter, the prevalence of internal party conflict in the Westminster system, the shifting balance of power in the governing Conservative party, and the polarising dynamics of the referendum in shifting opinion in a rightward direction (e.g. Allen, 2018; Heinkelmann-Wild et al., 2020; Sobolewska and Ford, 2020). But these cannot account for the extent of Cameron and May’s Euroscepticism during the talks, nor their subsequent abandonment of these positions, nor can they fully explain why criticism of both leaders focused so heavily on their failure to gain leverage over Brussels.
This article argues that to understand the hardening of preferences accompanying the Brexit process, we need to examine how the United Kingdom’s negotiating strategy contributed to change at the domestic level. Drawing on the literature on two-level games, it argues that successive British leaders deployed ‘synergistic’ strategies involving the instrumental use of domestic constraints in order to overcome their diminished bargaining power vis-à-vis the EU. These strategies involved the manipulation of key facets of the domestic political environment, including the shaping of the domestic discourse and the establishment of decision-moments, while also producing unintended consequences at the domestic level, ratcheting up hard Brexit sentiment among Leave supporters. These effects took on a path-dependent quality, not only because they consistently pushed towards a more Eurosceptic position, but also because subsequent events – the 2016 referendum and the rejection of May’s Withdrawal Agreement – prevented any recourse to previously ‘softer’ positions. Understanding the role of negotiating strategies in shaping domestic politics helps us to account for the shifting positions of Cameron and May in relation to their domestic detractors and to explain how the vision for Brexit gradually hardened over the course of the talks. Theoretically, the Brexit example highlights the distinct ways in which synergistic strategies can bring about unintended consequences for leaders, which can come back to haunt them at later stages.
The Brexit puzzle
It is perhaps a truism to note that the Brexit process, which has unfolded over the past few years in the United Kingdom, comprised a series of unexpected developments. Even before the United Kingdom voted to leave, then prime minister David Cameron’s commitment to an in/out referendum on the country’s EU membership was regarded by many observers as a rather drastic solution to the domestic pressures on Cameron and his desire to safeguard the United Kingdom’s position within the EU (Matthijs, 2013; Oliver, 2017). The subsequent victory of the Leave campaign in the 23 June 2016 referendum by 51.9% of the British electorate confounded the expectations of observers and pollsters alike and was widely regarded as a seismic and unexpected event, as indeed it still is (Sobolewska and Ford, 2020: 219). The selection of unity-candidate Theresa May as Conservative leader and prime minister in the aftermath of the referendum result surprised many, as did her decision to pursue a hard-line and uncompromising Brexit agenda, characterised by hard bargaining, anti-EU rhetoric, unattainable ‘red lines’, and an outright refusal to negotiate with stakeholders outside of the Conservative party (Jones, 2019: 28, 40). Fast-forward to early 2019 and the formerly hard-line May appears conciliatory compared to the Brexit supporters in the Conservative party and their collaborators in the Democratic Unionist Party (DUP), whose opposition contributed to the defeat of May’s negotiated deal by the greatest margin in living memory (Baldini et al., forthcoming). And, while the rise of ‘no deal’ sentiment under May’s tenure might have been considered an extreme position, under the premiership of Boris Johnson which succeeded her, the no deal threat came to characterise the prime minister’s ‘madman’ strategy of negotiating with the EU (Financial Times, 2021). While surprising enough in and of themselves, there is also a common thread to each of these key moments in the Brexit process: they are all characterised by the rise of increasingly virulent Eurosceptic positions among both elites and pro-Brexit constituencies. Over time, and when taken together, they amount to a gradual hardening of the United Kingdom’s designs for Brexit.
How can we explain these changes? Existing explanations may be fitted into two distinct – though at times overlapping – categories. To begin with, there are those which see this shift to the right as the product of the gradual realisation in the United Kingdom that the preferences of Eurosceptics could only be obtained as part of a more distant relationship. Since those who wished to ‘take back control’ did not wish to be subservient to EU rules and regulations, it makes sense that in a renegotiated settlement, softer models based on European Economic Area (EEA) membership, or ‘bespoke’ forms of market access would ultimately be rejected (Donnelly, 2018). Some observers contend Cameron was ‘unleashing demons’ by announcing the referendum pledge precisely because the demands of the Brexiteers in the Conservative party could not be easily mollified (Oliver, 2017). Others have focused on the role of specific pledges at key points, such as May’s articulation of the ‘red lines’ in her Lancaster House speech, in laying the ground for a harder Brexit (Schnapper, 2021: 372), while scholars focusing on the negotiations have emphasised the role of British misperceptions and the asymmetric balance of power in precluding British designs on a softer and more ‘bespoke’ Brexit outcome (Fabbrini, 2019; Figueira and Martill, 2020; Jones, 2019; Larsén and Khorana, 2020; Martill and Staiger, 2021; Schnapper, 2021).
Then there are those works which ascribe the tendency towards a harder Brexit to dynamics inherent in the British political system. It has been argued that the referendum itself, for instance, unleashed polarising dynamics by vastly increasing the salience of the EU question and subsuming pre-existing social and political conflicts under the Brexit umbrella (Sobolewska and Ford, 2020: 218). Others have conceptualised Brexit as a ‘revolutionary moment’ in which high levels of uncertainty regarding the eventual outcome created pressures for the radicalisation of opinion on both sides of the debate (Rogers, 2019). Moreover, the Westminster system reinforces this polarisation, since the principal fault-lines are located within, not between, the United Kingdom’s broad-based political parties (Heinkelmann-Wild et al., 2020). Still others have focused on the internal sources of political change within the Conservative party, as well as the continued threat of Farage on the Eurosceptic right (Bale, 2018; Kettell and Kerr, 2020; Lynch and Whitaker, 2018). And party positions may be easily linked to the positions of leaders, as is the case with explanations for May’s hard bargaining, many of which have focused on her status as an insecure Remainer and her fear of being deposed by the party right (e.g. Allen, 2018; Schnapper, 2021).
While both approaches highlight significant dynamics which have influenced the Brexit process, there remain key gaps in our understanding, which cannot be explained either by the inherent contradictions or by Brexit or changes in post-referendum domestic politics alone. We do not understand, for instance, why leaders like Cameron and May articulated positions more akin to those of their detractors on the right of the party during the negotiations and why they subsequently sought to row back from these positions. Not only were Cameron and May surprisingly convincing Eurosceptics, but they were also subject to ever-growing criticism from within their own ranks as their respective negotiations progressed. Nor do we understand why political opponents would castigate both leaders on the basis not of their reform proposals per se, but rather of their perceived failure in the negotiations. Throughout the talks, pro-Brexit constituencies maintained that ‘taking back control’ was not incompatible with access to the EU single market, chastising leaders for their failure to push the EU into making this concession. In other words, existing approaches cannot fully explain the dynamic nature of the Brexit process, nor how the demands of the negotiations and the domestic political environment have interacted over time.
Synergistic bargaining and domestic change
This article draws on insights from two-level games approaches to international negotiations to show how, in the absence of alternative forms of leverage, successive leaders tried to use the strength of domestic Euroscepticism in the United Kingdom as leverage vis-à-vis Brussels, exacerbating these forces without obtaining meaningful change. The literature on two-level games depicts leaders in international negotiations as occupying a space between two inter-linked ‘games’ at the domestic and international levels, respectively (Putnam, 1988). Because what is possible to achieve externally may not coincide with what is possible domestically, and vice versa, leaders often need to engage in ‘synergistic’ strategies that link together these games in order to bring both win-sets closer together. By citing domestic constraints to external audiences and external constraints to domestic audiences, leaders can shift outcomes in both domains in their favour (Putnam, 1988: 446–447). Moreover, since citing domestic constraints can increase bargaining leverage at the international level, leaders are known to engage in the ‘instrumental manipulation of domestic win sets to increase [their] bargaining power’ (Caporaso, 1997: 568). Such strategies may be especially effective where other forms of leverage are absent, such as under highly asymmetric bargaining conditions in which defection may be perceptibly costlier for the smaller party (e.g. Kroll and Leuffen, 2016).
Strategies aimed at establishing – and invoking – domestic constraints can take many forms. Leaders can establish audience costs by publicly committing themselves to a specific position (e.g. Debs and Weiss, 2016; Tomz, 2007: 821; Weeks, 2008). They can outsource ratification decisions to domestic constituents through parliamentary votes, referendums, or general elections where it is within their power to do so (e.g. König and Hug, 2000: 102; Zahariadis, 2017: 688). They can also manipulate the nature of the decision-moment to suit their needs by altering the timing, the presentation of alternatives (Slapin, 2006: 55), the threshold for ratification (König and Hug, 2000; Tarar, 2005: 405), and the boundaries of the (s)electorate (e.g. Hug and Schulz, 2007: 178; Schoppa, 1993: 354, 384). And they can engage in public diplomacy in order to shift domestic preferences such that they fit the requirements of the negotiations and the decision-moment (e.g. Caporaso, 1997; Putnam, 1988). In other words, by shaping the discourse in certain ways and tying their hands at key moments, leaders can make more credible any threats issued to other parties in the negotiations. But synergistic strategies are risky and are not guaranteed to work (Hug and Schulz, 2007: 178–179). If the threat is not sufficiently plausible (Janusch, 2018), if the constraints are easy to wiggle out of (Hix, 2002; Snyder and Borghard, 2011), if the government is too transparent in its game-playing (Caporaso, 1997; Zahariadis, 2017), or if domestic signals convey contrary information on the government’s intentions (Schultz, 1998), then negotiating partners may be unconvinced about the efficacy of the threatened outcome.
Moreover, synergistic strategic can introduce problems at the domestic level, which are unexpected at the time the strategy is initially articulated. Not only can they inadvertently increase support for the underlying threat among the population, but they can also inflate expectations about the likelihood of an agreement being reached (e.g. Larsén and Khorana, 2020). And they can undermine leaders’ ability to sell a negotiated agreement by producing negative visuals of ‘concessions’ back home while also establishing vague and easily manipulable standards for what counts as a ‘good deal’, which can subsequently be co-opted by political opponents. These domestic changes can easily come back to haunt leaders, either at the ratification stage or in future political events. Leaders may also find that efforts to make credible a threatened outcome increases domestic support for a threatened outcome faster than it increases the credibility of this outcome among external observers, increasing the risk of involuntary defection (Putnam, 1988: 453). Moreover, if sufficient conditions are not obtained, then the threatened outcome can itself come to be seen as a test of credibility, quite independent of how desirable it was in the first place. While significant decisions may therefore be initiated primarily as a display of resolve, once taken it can often be very difficult to return to the pre-decision status quo.
Methodology
The remainder of this article sets out the empirical basis of the theoretical argument offered above by discussing the use of synergistic strategies by both David Cameron and Theresa May in their respective negotiations with the EU. The empirical analysis shows that both leaders – finding themselves without alternative sources of leverage vis-à-vis the EU – sought to convince European leaders to compromise on fundamental principles by emphasising the strength of domestic opposition to the status quo and the ability of domestic actors to veto any sub-par agreement. More than merely
The empirical basis of the research covers Cameron’s renegotiation attempt over the course of 2015/2016 and the subsequent in/out referendum in 2016, as well as May’s effort to deliver Brexit from her ascension to the premiership in July 2016 to her resignation in July 2019. The article draws on the publicly available literature on the Brexit negotiations, including biographies of the leading figures involved, to build the narratives for the cases. These follow a process-tracing approach that adopts a partially selective focus on those aspects of the negotiations highlighted by the theoretical framework. In other words, the cases follow the articulation of the negotiating strategies, how leaders invoke domestic considerations, and the ways in which the negotiations feed back into domestic positions. Both cases show similar dynamics can be observed during the tenures of Cameron and May while allowing for the acknowledgement of relevant differences. Moreover, by covering two consecutive time periods, they allow us to observe how these processes contributed to the rise of a more uncompromising position among Brexit supporters.
The findings corroborate the principal assertions, made in the theoretical section above, that leaders attempted to leverage domestic constraints, that they actively colluded in supporting Eurosceptic arguments and stringent conditions for the EU to meet, and that their interactions with European leaders involved concerted efforts to play up the veracity of these constraints. The findings also suggest that these actions helped bring about changes in the domestic political environment, that the hardening of opinion made it more difficult for the leaders to sell their deals, and that this contributed to the eventual outcome of ‘involuntary defection’. That the observable implications of the theory can be demonstrated empirically does not preclude the validity of alternative explanations, but it does suggest that negotiating strategies did have a significant role to play alongside other factors in bringing about the gradual shift towards a harder Brexit.
David Cameron’s renegotiation and referendum
In his Bloomberg Speech in January 2013, Cameron announced that a majority Conservative government would hold an in/out referendum on the United Kingdom’s EU membership. The referendum pledge was the result of several developments in British politics, including rising Euroscepticism on the Conservative backbenches which Cameron hoped the referendum would put to rest (Bale, 2018: 263; Hayton, 2018: 224) as well as the related rise of the UK Independence Party (UKIP) which threatened the Conservative vote in general, and especially European, elections (Copsey and Haughton, 2014: 83; Vasilopoulou and Keith, 2019: 488). But the referendum decision also reflected Cameron’s own concerns about the future direction of the EU and his fears that the United Kingdom could find itself part of a very different Union in the future, one in which non-Eurozone members had a diminished voice and were subject to the policy preferences of the members of the inner-circle (e.g. Seldon and Snowdon, 2016: 261–263; Thompson, 2017: 439). Oliver Letwin, who was instrumental in convincing Cameron of the need for a referendum, argued that it seemed to me to be a very useful device to maximise the chances of getting what I wanted us to get from the EU. I thought that, it the leaders of other EU members states saw that we were heading towards an ‘in/out’ referendum in the UK, they would be more likely to give us protection against absorption into the euro. (Letwin, 2017: 17–18)
Both appeasing domestic detractors and solving the perceived problems with the EU required leverage, however, since Cameron would be asking the EU to compromise – if not wholly renege – on core principles of European integration, not least regarding the free movement of persons. And in the period following the financial crisis, leverage was in short supply. The United Kingdom had obtained significant concessions on previous occasions because of its status quo preference, which allowed it to drag its heels at Treaty negotiations, precluding further integration until it obtained the desired guarantees or opt-outs (Schimmelfennig, 2018: 1158). But the financial crisis had exacerbated fears of Britain’s status as a non-Eurozone member while simultaneously highlighting the ability of the rest of the EU to work around British opposition, as they had done with the 2011 Fiscal Compact (Thompson, 2017: 446). Moreover, there was little desire to re-open the Treaties – the main source of British leverage – and, even if there had been, the United Kingdom would have found itself requesting changes to the status quo, a much less powerful position. Of course, the United Kingdom could always threaten to leave if it did not obtain what it wanted from the EU, but European leaders would be unlikely to believe that Britain would sacrifice the benefits of membership simply to push for piecemeal reforms.
Cameron found the solution to his lack of leverage in the referendum, since it took the decision on membership out of his hands and vested it in the British people, not known for their embrace of the normative aspects of European integration (Carl et al., 2019: 286–289). By externalising the decision, Cameron made the prospect of exit more credible than it would be otherwise, since opinion polls showed the United Kingdom would likely vote to leave without reform, but that there existed a majority to remain in a reformed EU (Hobolt, 2016). In Cameron’s view, the renegotiation and referendum were ‘two sides of the same coin’, and he stated in his memoirs that ‘I could never have undertaken a proper renegotiation without a referendum. I needed that endpoint to focus the minds of the leaders from whom I was seeking change’ (Cameron, 2019: 622). Such a strategy would be familiar to theorists of negotiations as a form of hand-tying, where bargaining leverage comes from vesting the decision in a less compromising constituency. Of course, ostensibly tied hands can often be just as easily untied. But the subsequent inclusion of the referendum pledge in the 2015 manifesto (Conservative Party, 2015), for instance, made the commitment more difficult to wriggle out of, as did the decision of June 2015 to place the vote on a legislative footing (Smith, 2016: 333).
Cameron also made a concerted effort to cultivate a discourse at home which supported his assertion that the United Kingdom was prepared to walk away from the EU if it could not obtain revised terms of membership. The prime minister repeatedly emphasised that he had no ‘romantic attachment to the European Union and its institutions’ (Kroll and Leuffen, 2016: 1311) and that the United Kingdom would ‘remain in the European Union so long as it is in our interest to do so’ (Copsey and Haughton, 2014: 81), making it clear the United Kingdom would walk away if the EU’s offer was deemed insubstantial (Kroll and Leuffen, 2016: 1312). Comparative research on Cameron has found he was more Eurosceptic in the run-up to the renegotiation than otherwise, suggesting the prime minister undertook deliberate efforts to shape the discursive milieu (Brusenbauch Meislová, 2019a: 229–232; Seldon and Snowdon, 2016: 270). Fostering this discourse no doubt helped reinforce the message Cameron wished to convey to the EU, but such public statements also function as a form of ‘audience costs’, a tried-and-tested tactic through which leaders can raise the domestic costs of defecting from a position and thereby make their demands more credible (e.g. Tomz, 2007).
When Cameron began his renegotiation effort following his party’s general election victory on 7 May 2015, he focused heavily on conveying Britain’s resolve to exit the Union if new terms for membership could not be obtained. He noted that, wherever he went, his ‘objectives were the same. . . Insist that I was serious about securing a better settlement in order to keep Britain in the EU. Warn of the risks of under-delivering and seeing the UK exit the EU altogether’ (Cameron, 2019: 638). In conversations with Angela Merkel, Cameron pointed out that ‘If you force the British people to choose between a measure of control over who comes into the country, and staying in the EU, if you give them that binary choice, they will vote to leave the EU’ (Cameron, 2019: 638). And in a meeting with Tusk prior to the February 2016 European Council meeting, Cameron recalls he stated: If the February Council denies me these things, and the whole thing breaks up, we are going to have to have a look at our relationship with this organisation’, I told him . . . ‘We’re the third biggest contributor. We are a proud country, and if we’re treated like supplicants, then we’ll bugger off’. (Cameron, 2019: 642)
Cameron never succeeded in convincing other European leaders the United Kingdom would vote to leave. He reported: picking up from most leaders . . . that they simply did not think Britain would leave . . . ‘They don’t believe we’d leave the EU over this’, one advisor wrote to me. And they really didn’t. No matter how much I said it. (Cameron, 2019: 641)
Jonathan Hill, the United Kingdom’s Commissioner, reported to Cameron in January 2016 that he believed the United Kingdom was being ‘low-balled’, stating, ‘They need to understand that if they don’t get this right, the whole thing could unravel – with Britain leaving the EU’ (Oliver, 2017: 58). Several factors contributed to their scepticism. For one thing, European leaders were unimpressed with the timing of Cameron’s renegotiation effort, coming as it did just as the EU was dealing with the fallout from the Ukraine and Euro crises. The leaders were conscious, too, that the United Kingdom already had beneficial terms of membership, with several opt-outs from core policy areas (Matthijs, 2013: 13; Smith, 2016: 330). Exit was also viewed as a somewhat extreme – and thus implausible – option, even when taken out of Cameron’s hands. Merkel found it difficult to understand why British citizens would opt to leave, given the high economic costs (Cameron, 2019: 640), while Dutch Prime Minister Mark Rutte reportedly told Cameron he felt UK citizens were too conservative in disposition to leave the EU (Oliver, 2017: 58–60). And there was a sense in which Cameron’s gamble was too transparent, since many leaders seemed to regard the referendum as ‘a ruse to get more out of the renegotiation’ (Cameron, 2019: 641).
Cameron’s inability to convince European leaders the United Kingdom was hovering at the exit did not prevent him from obtaining a renegotiated basis for the United Kingdom’s membership, including an ‘emergency brake’ limiting access to in-work benefits for EU migrants, a ‘red card’ system allowing national Parliaments to demand amendments to EU legislation under certain conditions, and a formal opt-out from the commitment to ‘ever closer union’ (Kroll and Leuffen, 2016: 1316; Smith, 2016: 333). So the renegotiation was not a failure, and Cameron was ‘able to claim that he had secured just about enough concessions to satisfy all four “baskets” of demands’ (The Economist, 2016). But these changes were more contingent than Cameron wanted and did not fully address domestic concerns about migration or national sovereignty. And the domestic reaction was overwhelmingly negative. Cameron was castigated for having asked for too little, and then for having obtained less than he promised (Shipman, 2018: 10). Newspaper editorials claimed the deal was ‘a long way from what the PM and the Tory manifesto promised’ (Oliver, 2017: 77–78). Prominent Conservatives, including Cabinet ministers, sought to quickly distance themselves from the deal (Seldon and Snowdon, 2016: 546, 549). And, because the deal was deemed a failure, it featured little in the ensuing referendum campaign.
Cameron’s negotiating strategy, and his deployment of the referendum/renegotiation combination, did much to change the political environment in the United Kingdom throughout the period. To begin with, the referendum announcement itself galvanised Eurosceptic parties and organisations, whose reach was amplified by Cameron’s own Eurosceptic rhetoric and by his deliberate efforts to increase the salience of the question of Britain’s EU membership. It also mobilised new constituencies who had not voted in previous elections owing to the lack of perceived choice between the major parties (Goodwin and Heath, 2016: 325). Moreover, the referendum created clear opportunities for Conservatives seeking to challenge Cameron, given the Eurosceptic credentials of much of the party membership base, meaning that even nominally pro-Remain Conservatives had a strong incentive to oppose Cameron’s renegotiated agreement (Seldon and Snowdon, 2016: 548). The defection of prominent individuals from Cameron’s camp – like Johnson and Michael Gove – would prove to be highly damaging in the referendum campaign, since they helped mobilise citizens ‘that Nigel Farage couldn’t reach: the sensible, middle-of-the-road voters’ (Letwin, 2017: 25).
And Cameron’s rhetorical EU bashing would prove equally problematic as he tried to sell his renegotiated terms of membership to the British public. For one thing, Cameron had given Eurosceptics many of their best arguments by highlighting the flaws of the EU and the unsuitability of the status quo for the United Kingdom. He had also succeeded in reinforcing the notion that the United Kingdom required – and indeed could have – a new and exceptional basis for membership under the existing EU Treaties. Cameron’s demands had been deliberately vague, to calibrate them to what the EU might offer (Kroll and Leuffen, 2016: 1315), and this vagueness acted as a double-edged sword because it allowed others to suggest that the price of continued membership had not been met (Seldon and Snowdon, 2016: 546). Cameron also found it difficult to credibly defend the EU after so many months of spelling out its weaknesses. ‘Our enthusiasm for the Remain case could easily be questioned’, he noted: ‘After all, we had shifted very quickly from “big, bossy and interfering” negativity to “stronger, safer and better off” positivity. We were asking a lot of people to believe such a handbrake turn’ (Cameron, 2019: 665).
The clear lack of enthusiasm for Cameron’s renegotiation and the success of the Leave campaign at mobilising British citizens ultimately contributed to the Leave victory of 51.9% in the referendum on 23 June 2016. While a great many factors need to be taken into account to explain the Leave victory (Clarke et al., 2017; Sobolewska and Ford, 2020), the marginality of the result means that any factor – including Cameron’s Eurosceptic rhetoric – can be considered partially responsible for tilting the result towards exit. Interestingly, one of the reasons given by several prominent ‘converts’ to the Brexit cause, like Boris Johnson, was that voting ‘Leave’ was the only way to show the EU that Britain was serious about walking away leaving (Cameron, 2019: 654). Several Cabinet members felt Cameron had not worked hard enough to convince the EU that Britain really was prepared to walk away (Shipman, 2018: 10) and reasoned that ‘if we vote out, the EU will come running with proper concessions and a far better deal. In other words, they don’t think a vote to Leave is a vote to leave’ (Oliver, 2017: 96–97). The return to the negotiating table would never happen, since the Brexit result changed the game so significantly, precluding a simple return to the pre-referendum status quo. But it is instructive that Cameron’s failure to convince European leaders exit was a realistic possibility would, quite independently, motivate prominent individuals to promote a leave vote purely as a demonstration of resolve.
Theresa May’s Brexit negotiations
Cameron resigned following the referendum result, his position untenable. In his place, Theresa May emerged as Conservative leader and Prime Minister. A compromise candidate and pragmatist, May had also been a Remain supporter, if a quiet one, and sought to focus her premiership on delivering on the mandate established in the referendum (Allen, 2018: 108). May wanted Brexit to be delivered swiftly and in a meaningful way, but she was also keen for existing trade and security ties not to be undone in the process. Thus, her Brexit strategy focused on obtaining the greatest possible market access compatible with the return of key competences to the United Kingdom, including control over migration, and the renunciation of the jurisdiction of the European Court of Justice. This search for a ‘bespoke’ deal would put May at odds with the EU’s existing balance of rights and obligations for members and non-members, but also with the political interests of the EU27 post-Brexit, who were squarely focused on avoiding the risk of contagion, safeguarding core EU principles, and precluding the emergence of a bespoke ‘UK model’ of association (Laffan, 2019: 14). But the prime minister disregarded reports from her advisors that a bespoke deal would never fly and was reportedly ‘sick of people saying to me that we can’t go for an ambitious new Brexit model designed around the UK’s interests and needs (Seldon, 2019: 100).
May’s aims were not all that different from Cameron’s in seeking to maintain the economic underpinnings of EU membership but to de-link this from specific political commitments, including free movement. While Cameron attempted to negotiate such an arrangement from within, and along more conservative lines, May sought to do so from without and as part of a new model of external association (Figueira and Martill, 2020). And, like her predecessor, May lacked leverage over the EU. Since the United Kingdom was leaving, there was no goodwill, no solidarity (Rogers, 2019), and a vastly diminished ‘shadow of the future’, with Britain’s time in the EU ending and very little reason to concede anything to the soon-to-be former member state. Her response was to follow a similar strategy to Cameron by seeking to deploy the United Kingdom’s resolve to walk away from the negotiations – the ‘no deal’ Brexit outcome – as leverage in the talks. Indeed, the threat of ‘no deal’ was a core component of May’s ‘hard bargaining’ stance which emerged over the course of early 2017 (Martill and Staiger, 2021: 263). While Cameron’s strategy had failed spectacularly, May was among those who believed her predecessor had not gone far enough in convincing the EU of Britain’s willingness to walk away. The key lesson she took from Cameron’s experience was ‘to at least look like you were prepared to walk away from the talks to maximise leverage’ (Shipman, 2018: 10). This was not just May’s view, but was shared also by her ministers, one of whom remarked that ‘David Cameron’s mistake when he was negotiating before the referendum was that the Europeans never thought we would walk away. This time we will’ (Shipman, 2018: 97).
While a ‘no deal’ Brexit would damage the United Kingdom, it would damage the EU, too. Hence, if the EU could be reassured the outcome was inevitable, Britain – so it was believed – should gain at least some leverage (Schnapper, 2021). May’s rhetoric was carefully crafted to reinforce the message that the United Kingdom would walk away if it did not get what it wanted (Brusenbauch Meislová, 2019b). Among her many stock phrases, ‘no deal is better than a bad deal’ gained prominence and notoriety, committing the United Kingdom to a ‘no deal’ scenario if it did not obtain a favourable deal from the EU. The phrase’s author, advisor Nick Timothy, crafted it ‘in order to impress on the EU that May was prepared to cut and run if she did not get what she wanted’ (Shipman, 2018: 96). May did not engage explicitly in hand-tying, as Cameron had done by locking in the referendum pledge. But she did seek to leverage the Article 50 process, which established a two-year time limit for negotiating the terms of withdrawal, following which the United Kingdom would crash out of the Union. May did not design this process, and she had other good reasons for initiating it when she did – the EU refused to begin talks until notification of the intent to withdraw had been given – but in establishing a no deal Brexit as the reversion point, it essentially removed all domestic barriers to this outcome. May’s message to Brussels, backed up implicitly by the Article 50 process, was that unless she could obtain a deal that would satisfy Eurosceptics in her party, the United Kingdom would leave without a deal. To that end, May went to considerable effort to forestall domestic attempts to take ‘no deal’ off the table, on the explicit rationale that to do so would undermine the United Kingdom’s leverage in the talks (Martill, 2021).
May’s attempt to leverage the risk of a ‘no deal’ Brexit was unsuccessful, forcing the prime minister into a series of humiliating climb-downs during 2018 as she assented to EU proposals on the terms of withdrawal (Jones, 2019: 44–45; Schnapper, 2021). There are many reasons why May’s negotiating strategy was unsuccessful. Certainly, the prime minister struggled to make her ‘no deal’ threat credible. The asymmetry of effects was so significant (Hix, 2018: 14–15; Jones, 2019: 38) that few in Brussels took her claim seriously. Juncker’s chief of staff, Martin Selmayr, claimed the EU ‘never saw no deal under May as a credible threat’ (Seldon, 2019: 119). Moreover, talk of ‘no deal’ spooked business and other domestic constituencies who had been promised minimal disruption (Seldon, 2019: 417), and for this reason it was decided that any preparation for ‘no deal’ should be undertaken only at ‘the last possible moment’ (Shipman, 2018: 515). But May faced a deeper problem in having fundamentally misread the EU’s interests, especially when it came to protecting the principles of the single market (Figueira and Martill, 2020). With Brussels less susceptible to no deal than the United Kingdom, and happier to see the United Kingdom crash out than sacrifice the single market, even if the no deal threat were to be made credible, it would not have afforded May the kind of Brexit deal she sought. But while it would be true to say May’s ‘no deal’ threat would never be sufficient to obtain the kind of concessions the prime minister sought, this did not prevent her from articulating such a strategy, nor would it preclude the consequences of emphasising the viability of ‘no deal’.
One of the most significant consequences of May’s uncompromising Eurosceptic rhetoric was the role it played in hardening preferences among the Leave supporters within the Conservative party. While May was initially concerned more with the threat from Remain supporters (Timothy, 2020: 17–18), by mid-2018 it became clear the greatest opposition to her deal within the Conservative party was on the pro-Brexit right, who began to desert the prime minister when it became clear she would settle for less than the ‘red lines’ she initially set out (Schnapper, 2021; Shipman, 2018: 484). Support for May’s negotiated agreement drained away gradually over the course of the negotiations as the divergence between her stated aims and what she managed to negotiate became clearer. May’s Chequers Plan of July precipitated the resignations of Brexit Secretary David Davis and Foreign Secretary Johnson while further undermining backbencher trust in the government (Seldon, 2019: 420, 446). By November 2018, when the terms of the deal were agreed, Brexit Secretary Dominic Raab, Davis’ replacement, declared the agreement to be ‘Brexit In Name Only’ (BRINO), fatally undermining May’s trust among backbenchers, one of whom described the deal as a ‘Hotel California Brexit deal which ensures we can never truly leave the EU’ (Seldon, 2019: 494).
May’s embrace of ‘no deal’ as a possible – even desirable – outcome contributed to the growth in support for this outcome, such that it became the majority position among Leave supporters over the course of the negotiations (Kettell and Kerr, 2020: 598). By interpreting the Brexit mandate in an uncompromising manner, May set high expectations at home which could not be delivered from Brussels (Larsén and Khorana, 2020), making her effort seem not just a failure, but also one that fell clearly short of the ‘red lines’ established for a meaningful Brexit. The optics of an ostensibly strong United Kingdom being forced into concession after concession further increased Eurosceptic ire, with frequent assertions Britain was being ‘bullied’ by Brussels, even though London’s demands were never likely to be met (Martill and Staiger, 2021: 263). The vagueness of May’s demands, especially when it came to the much-vaunted ‘good deal’, also made it easy for political opponents to claim May’s deal was a ‘bad one’, thereby co-opting her rhetoric in order to oppose her negotiated agreement (Kettell and Kerr, 2020: 605). In fact, May’s shift to the right in her rhetoric only forced would-be detractors to move further to the right in order to oppose her deal, such that by the time of May’s eventual replacement in July 2019, the most significant issue was the openness of candidates to a ‘no deal’ Brexit. And May found it difficult – as Cameron had before her – to switch from her uncompromising rhetoric into the more conciliatory discourse required to defend her agreement and sell the virtues of institutionalised cooperation with the EU.
While Brussels insiders insisted that ‘no deal’ was never credible, and was in any case an outcome they were prepared for, many of May’s Cabinet colleagues were convinced that May had not conveyed the message forcefully enough (Seldon, 2019: 485). Gove and Davis both wrote to May in late 2017 complaining about the lack of no deal planning and urging a tougher approach (Shipman, 2018: 515). In late 2018, Raab reportedly told May to: ‘Go back, be tougher’, and even middle-ground Cabinet colleagues argued she ‘needed to be tougher with the EU to get what Britain needed, as Thatcher had been’ (Seldon, 2019: 485–487). Jeremy Hunt would later claim that ‘Brussels was
In the end, on 15 January 2019, the Withdrawal Agreement Bill was rejected by 432 to 202, a margin of defeat not encountered by a sitting prime minister since the 19th century. Following exploratory talks with the opposition, May whipped her party to support the ‘Brady Amendment’, sponsored by Sir Graham Brady, the chair of the 1922 Committee of backbench Conservative MPs, which envisaged the replacement of the Northern Ireland ‘backstop’ with alternative measures, requiring the prime minister to return to Brussels to request changes. May’s support for the amendment, which opposed the agreement she herself had negotiated, offers a good example of the anticipated benefits of allying with the domestic hardliners to try to obtain further leverage in the talks (Bercow, 2020: 377), although Brussels was unwilling to offer anything other than further ‘assurances’. A second vote took place on 12 March and saw the deal defeated again by 391 to 242, prompting the prime minister to seek a temporary extension to Article 50 on 20 March prior to the third vote, held on 29 March, which also saw the prime minister defeated, albeit by a smaller margin than previously (344 to 286). This third defeat fatally undermined May’s authority, and following abortive talks with the Labour party, the prime minister committed on 24 May to resigning, finally handing over to Johnson on 24 July following the Conservative leadership election. While efforts to use the rejection of the Withdrawal Agreement to push for better terms had repeatedly failed, Johnson’s selection as the hard-line candidate of choice offered no opportunity for re-thinking the United Kingdom’s strategy, once again ratcheting up the demands of Brexit supporters and their belief that Britain’s willingness to walk away would provide the key to getting more out of Brussels.
Conclusion
Britain’s negotiations with the EU, which have dominated British public life for the better part of five years, have witnessed a concerted and ever-more surprising shift to the right as opportunities for compromise or softer variants of Brexit have fallen by the wayside. Existing explanations have focused on the incompatibility between British designs on Europe and core principles of integration as a factor pushing the country into a more distant model of association as well as the tendency of the United Kingdom’s political system to foster polarisation and a hardening of preferences on Europe. But these perspectives cannot by themselves account for the complex pattern of interaction between elites and their more hard-line detractors, nor for the specific times at which Eurosceptic concerns have become more influential. This article argued that to understand the gradual hardening of the United Kingdom’s demands from Brexit, we need to acknowledge the impact successive negotiating strategies have had at the domestic level. Drawing on two-level games accounts of ‘synergistic’ bargaining strategies which seek to deploy domestic constraints as bargaining leverage, we can observe how both Cameron and May alike sought to co-opt the preferences of Eurosceptics into their bargaining strategy and understand the effects this had on the politics of Brexit over time.
Analysis of both premierships highlights a number of similarities. Both leaders sought to negotiate exceptions to the balance of rights and obligations that come with access to the EU single market, first as a member (Cameron) and then as a non-member (May). And, finding themselves without forms of leverage the United Kingdom had previously enjoyed, both leaders sought instead to threaten British exit, initially from the EU (Cameron) and later through a ‘no deal’ Brexit (May). Since both actions would impose asymmetric costs on the United Kingdom, both leaders relied heavily on domestic constraints which they themselves had encouraged throughout the negotiations, in order to convince the EU the decision was not theirs to take. While Cameron tied himself to the referendum result, May made every effort to keep ‘no deal’ on the table in the talks, before eventually citing her own deal’s repeated defeat as evidence of domestic intractability. Neither succeeded in convincing the EU that the threat of walking away was credible. But they did both succeed in exacerbating Eurosceptic sentiment domestically, giving credence to arguments on the right, undermining their own ability to promote cooperation with Brussels, establishing clear incentives for political opponents to outflank them, and building unrealistic expectations which were easily quashed. Not only did they increase support for outcomes they hoped to avoid, they also created a situation in which these outcomes became the only perceived means of demonstrating British credibility. And, once taken, these decisions could not be easily undone. Thus did the Brexit process ramp up alongside the perceived gap in credibility, as Cabinet colleagues urged first a Leave vote and then, in later years, rejection of the Withdrawal Agreement, to show the United Kingdom was ‘serious’.
What, then, is the value of conceptualising Brexit as a failed negotiating strategy? First, it helps us to understand the positions of leaders better, including their reasons for cultivating a more Eurosceptic discourse than was initially expected of them, as well as their subsequent shift into a more cooperative register. Second, it helps us understand the curious alliance between Conservative Eurosceptics and their more moderate party leaders when it comes to undertaking efforts to improve the credibility of the threat that the United Kingdom would ‘walk away’. Third, the argument helps explain how the negotiations contributed to the ratcheting up of the Brexit process over the years, as efforts to negotiate with the EU produced uncompromising narratives which would come to shape the domestic political milieu. Overall, the picture that emerges is one in which the Brexit outcome is more accidental than intentional, as positions designed for negotiating leverage have turned into preferences and as events undertaken to bolster the credibility of threats have become locked in. From this perspective, the current Brexit outcome is little more than a freak accident brought about by the sustained failure of the United Kingdom’s strategy for reforming the EU.
The broader theoretical implications of the study are twofold. To begin with, the argument highlights the diverse range of unintended consequences which the use of synergistic strategies can have at the domestic level, spelling out the distinct mechanisms through which leaders’ attempts to instrumentally utilise domestic constraints in the international game can backfire. While the potential for domestic backlash has, for sure, been acknowledged in existing studies of two-level games (e.g. Evans, 1993), the variety of mechanisms through which leaders can establish (negative) feedback remains underexplored, not least because domestic change is part-and-parcel of synergistic strategies. The study’s findings also highlight the importance of recognising the temporal dynamics across inter-linked negotiations, showing how the preferences fostered and lessons learned in one set of negotiations may come to inform subsequent negotiations in such a way as to produce path-dependent processes which span multiple sets of talks. In this way are the lessons of Brexit relevant to broader debates on international bargaining as well as to the calculus of leaders contemplating the tried-and-tested (but far from risk-free) manipulation of domestic win-sets for bargaining advantage.
Footnotes
Acknowledgements
The author would like to thank two anonymous reviewers as well as Columba Achilleos-Sarll, Alan Convery, Tim Oliver, Oliver Patel, Shaina Western, Nick Wright and the members of the IR Research Group at the University of Edinburgh and the IR in the North East seminar at Northumbria University for their helpful comments on earlier versions of the paper.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
