Conclusion: Center‐Periphery Bargaining in the Age of Democracy

2. A Member State which decides to withdraw shall notify the European Council of its intention. […] the Union shall negotiate and conclude an agreement with that State, setting out the arrangements for its withdrawal, taking account of the framework for its future relationship with the Union. That agreement […] shall be concluded on behalf of the Union by the Council, acting by a qualified majority, after obtaining the consent of the European Parliament. 3. The Treaties shall cease to apply to the State in question from the date of entry into force of the withdrawal agreement or, failing that, two years after the notification referred to in paragraph 2, unless the European Council, in agreement with the Member State concerned, unanimously decides to extend this period. Translating these two paragraphs back into our model's language: 1) the decision to withdraw has come about direct-democratically, via referendum, with deviating majorities in two of the UK's own peripheries, Scotland and Northern-Ireland; 2) bargaining on mutual trade, immigration rights and debt sharing – to name but the most salient issues – as part of the new institutional relation has yet to start; and 3) the main actors in this process will be the UK government, on the one hand, and the European Council, i.e. the heads of state and government of 27 European democracies, as well as the European Commission and Parliament, on the other. Emphasizing this unique case of national withdrawal from a supra-national union not only has topical value but also serves to highlight the three more specific contributions of this special issue. For in this instance, as much as in the other cases covered here, democratic principles and structures permeate, institutionalize, condition and are influenced by centre-periphery bargaining. That is why we group the insights provided by the contributions to this special issue into those same three questions: legitimacy, institutions and agency. In conclusion, we will return to the UK-EU case and outline possible implications that our bargaining model has for the terms the UK will be able to obtain from the EU. The first aspect of center-periphery bargaining concerns legitimacy – namely, who is entitled to speak and bargain on whose behalf, how are decisions made and approved within each party to the bargain, and how sustainable are the solutions arrived? Legitimacy, as Horne et al. (2016) remind us, is also one the main reasons for the longevity of political structures. When political regimes are legitimate, political disputes will turn on policy details rather than polity fundamentals; redistribution instead of redistricting; and power-sharing instead of power-dividing. Legitimacy and acceptance also rest on identity – in some experiments, even when alien rulers have the ability to punish, they remain at a disadvantage with native rulers (Horne et al. 2016). As a result, they conclude, “alien rulers must offer greater concessions to peripheries with a credible exit threat than native rulers who face comparable levels of discontent” (ibid. 12). The study on Czechoslovakia provides compelling real-world evidence about this. The conclusion from Basta & Bustikova's (2016) exegesis of the velvet divorce is that both the leaders of the centre (Prague) and the periphery (Bratislava) had more to gain from separation than from sticking together. Thus, the Czech state was able to continue on its radical market liberalization track while the Slovak elite could emulate their former compatriots’ exploitation of privatization, coupled with welfare state benefits to keep the population at bay. Failing this, the Czechoslovak entity lacked legitimacy in the eyes of both sides. The same happened in the case of Berne and Jura, with the latter region seceding from the former in the 1970s, albeit while remaining in the overall Swiss federation. Having been given to Berne at the Congress of Vienna in 1815, the new rulers of Jura – speaking a different language and having a different religion – never acquired a sufficient level of legitimacy, nor did they provide enough concessions, to maintain cantonal unity. Internal secession, state restructuring and legitimacy questions are also dealt with in the contribution by Swenden (2016). Looking at the partial shift from assimilation to accommodation in India, he notes how democratization – and the accompanying need to create majorities, if rule was to be perceived as legitimate – has been one of the factors that brought about de facto federalization. Thus, new states were carved out in order to accommodate linguistic minorities and the party system has become pluralized. However, the “red lines” that remain all involve both peripherality and religious distinctiveness: while Jammu and Kashmir is predominantly Muslim and the Punjab predominantly Sikh, Nagaland, Mizoram and Maghalaya are predominantly Christian and thus different from majoritarian Hindu. Also, their economic and/or demographic clout is too small to afford them sufficient bargaining power in extracting concessions – in our model's terms, exit is not credible and central dependence on them is low. The chief lesson to be learned from these three contributions, therefore, is that legitimacy depends a great deal on shared identity. To the degree that identity is not shared, compensatory institutional solutions (e.g. concessions) have to be offered. The bargaining processes leading to these solutions are heavily influenced by democratic mechanisms, such as the need to win elections (Czechoslovakia) and create a parliamentary majority (India). This brings us to the second important aspect of centre-periphery bargaining: institutions. The precise type of bargain that is struck matters both for its own longevity and for the overall stability of the system within which it is located. Anderson and Costa (2016) most directly address this institutional question. Comparing full (e.g. Belgium and the former Yugoslavia) and partial (e.g. Canada, Nigeria or India) ethno-federations with ethnic federacies (e.g. Gagauzia/Moldova), they attribute the highest propensity for survival to the middle category – a fully federalized system where only some entities rest on ethnicity. Here, the chances for stability, defined as neither re-centralization nor break-up, are higher than in systems where either the rest of country is unitary (a federacy) or where all regions are based on ethnicity. The implication of this is that where institutional designers are faced with at least one ethnically distinct region (e.g. Kurdistan in Iraq), they should refrain from redrawing all regions along ethnic lines, but also not keep the rest of the country unitary. However, the Bosnian case analyzed by Zdeb (2016) shows the dangers associated with denying a large ethnic minority its own institutional structure. With the overall state divided between the Federation of Bosnia and Herzegovina (FBiH) on the one hand, and the Republika Srpska (RS) on the other, Croats are a minority both overall and within the FBiH. This has led them to articulate demands for further territorial autonomy, although in community terms they enjoy full parity and are a majority in at least three of the ten FBiH cantons. However, Bosnia is still there, war has not erupted, ethnic autonomy has persisted, and at least some progress towards building central state capacity has been achieved. In other words, the institutional compromise between group-based parity and only partial ethnic autonomy in territorial terms seems to have succeeded. The least territorial case is that of the US, as discussed by Kincaid (2016). If peripheral nationalism has not been politically viable there, then for two reasons. On the one hand, individualist democracy has facilitated the assimilation of newcomers as citizens belonging to a given group, but not of the group as such. On the other hand, in the case of pre-existing national minorities that are territorially concentrated, concessions in the form of asymmetrical institutional arrangements have provided accommodations for indigenous groups such as the Navajo, Puerto Ricans or Samoans. In fact, Kincaid (2016) concludes that “asymmetrical tribal governance arrangements have the significant benefit of limiting undesirable foreign in-migration”, providing recipients with the benefits of both autonomy and preserving ethnic purity. The conclusion from these three contributions in terms of our model is that the type of bargain not only depends on the preferences and bargaining power of each of the respective parties to the negotiations, but that sometimes limits are either imposed from outside (as in the case of Bosnia with territorial inviolability) or exogenous to the bargain (e.g. territorial neutrality and cultural pluralism in the US). The third and final aspect of center-periphery bargaining that the contributions to this special issue have highlighted concerns political agency, without which none of the existing structures would have come about in the first place. Most illustrative in this regard are the articles by Röth et al. (2016) and Thorlakson (2016), who cover the lack of asymmetric decentralization in Turkey and the extent of economic voting in the German Länder, respectively. In the former article, political actors take centre stage by arguing for (the type of) decentralization, i.e. political or administrative, symmetric or asymmetric, local or regional. In Germany, political actors used the already existing federal structure to either argue for political change (when in opposition) or for re-election (incumbents). However, aside from this horizontal interplay, there is also a strong vertical dimension in both Turkey and Germany, which introduces territory into the equation. Thus Thorlakson (2016) looks at the extent to which economic performance at the regional or federal level is linked to punishing or rewarding regional or federal incumbents. She finds that “voters use Land elections to punish or reward federally incumbent parties for federal economic conditions.” This is undoubtedly due to the rather uniform and homogenous nature of the German federation. When it comes to assessing regional incumbents, however, regional economics had an even stronger effect on voters than federal conditions. This has important consequences for modelling centre-periphery bargaining: if regional leaders are not held accountable for regional but rather for national economic performance, then a key assumption– that of the unitary peripheral actor – is violated. If the peripheral elite and the peripheral population base their behaviour on the economic performance of two different government levels, then they are unlikely to support each other in working towards the same goal. The multilevel character of existing democracies can therefore play a significant role in creating the conditions for centre-periphery bargaining. However, the discussion by Röth et al. (2016) reveals that a central government may obstruct such “multi-levelness” deliberately to avoid creating these conditions. The authors rely on the concept of ideological distance to explain the absence of asymmetrical decentralization in Turkey, a most likely context for such a solution. Measured in economic (market liberalization) and (multi-)cultural terms on the basis of party manifestos, the current Turkish government and Kurdish parties are very far away from one another ideologically. For Röth et al. (2016), this factor accounts for the absence of concessions, where our framework would predict them, given the dependence of the center on the Kurdish region and the latter's credible exit option(s). “Kurdish minorities have consistently mobilized for regional autonomy, but national elites either declared this to be a threat to the unitary state or outright ignored Kurdish claims.” (Röth et al. 2016) Perhaps this also points to scope conditions, since the ensuing violence in Turkey, the failed coup and the reactions to it call into question its inclusion into our set of fully democratic countries. Let us now return to the case of Brexit. Drawing on the three types of insights provided by the eight articles in this special issue allows us to pinpoint several important aspects. First of all, “Brexit” came about not least because the EU was seen in many quarters as not having the legitimacy, or not having sufficient legitimacy, to make binding decisions for the UK demos, also because it questioned the notion of parliamentary sovereignty. One could also argue that the EU, or rather the other countries regarded as dominant therein, are seen as culturally too alien by a majority of English and Welsh voters. Institutionally, however, like in the case of Czechoslovakia, one could submit that separation both serves the (current) UK government in its drive towards welfare state reduction and austerity as well as those in the EU interested in further integration and single market regulation. Whether that will mean the eventual disintegration of the British Union remains to be seen. The consequences for other EU-skeptic countries still remain unclear unless we probe deeper into the actor dimension. To begin with, the referendum was called for by then UK Prime Minister Cameron after having obtained what he tried to sell as sufficient concessions from the EU. The referendum was held in June 2016. However, the EU had at least partially given into Cameron's demands shortly before.1 So in terms of our model, in addition to the significant budgetary rebate in force since 1985, the EU had in fact offered something to the UK to remain in the fold. That the referendum was called, despite these concessions, can be attributed to UK's internal party politics. Not only did the United Kingdom Independence Party (UKIP) secure close to four million votes in the May 2015 general election, but having obtained just a single seat in Westminster put additional pressure on Cameron to cater to that party's sole goal – to leave the EU. At the same time, ruling with a tight majority of not even 51% of all MPs would prove impossible without the support of those Tories who also wanted to leave. With hindsight, playing the European blame game (“the EU/the euro/EU immigration is bad for the British economy/society”) has directly helped the anti-David Cameron wing in the Tory party (and some pro-Brexit politicians) to come to power. But the UK has always been the most Eurosceptic country; in that sense, the Brexit referendum was a weak test, not necessarily likely to produce similar outcomes elsewhere. However, it also showed that what had worked in one case, with the Scottish vote on independence in 2014 – to delegate the decision on internal party heterogeneity to the people in a referendum on centre-periphery questions – need not work in another. It is perhaps ironic that the chief exporter of federalism worldwide, the UK, with many of its former colonies turned federations (Canada, USA, South Africa, India and Australia, to name but the largest), is the most opposed to federalism “at home”, in Europe. To conclude, the model should also be able to make some general predictions about the ultimate bargain. Obviously the threat of leaving the EU has already occurred, even if it is yet to be officially communicated to Brussels. But now that the vote is over, what is the cost of a “hard” vs. “soft exit” to both the EU and the UK? This involves the second parameter in the model, namely the dependence of the EU on the UK. Chief arguments pointing to a hard exit include the desire of the EU to discourage further secessions, the inseparability of the four freedoms, and the opportunity to harness some of London's financial centrality. At the same time, pragmatism would have to prevail over ideological commitment if a soft exit with significant trade and other prerogatives were to be granted. In terms of the best deal to be extracted for the UK, it would in fact have been better if – by a small but clear margin – the “remain”-camp had won, as in Scotland in 2014. In that way, both exit credibility and central dependence on appeasing the periphery would have been highest and concessions greatest. Right now, it seems that the UK is more dependent on the EU than the other way around. Sean Mueller is Ambizione Researcher (Swiss National Science Foundation) at the Institute of Politics, University of Berne. Address for correspondence: Institute of Political Science, University of Berne, Fabrikstrasse 8, CH-3012 Berne, Tel.: +41 (0)31 631 83 33; Email: sean.mueller@ipw.unibe.ch David Siroky is Associate Professor in the School of Politics and Global Studies, Arizona State University. Michael Hechter is an elected Fellow of the American Academy of Arts & Sciences, Foundation Professor of Political Science at Arizona State University, and Professor of Sociology at the University of Copenhagen.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.