Democracy without Solidarity: Political Dysfunction in Hard Times – Introduction to Special Issue

Abstract

Why do democratic institutions struggle to maintain their vitality and legitimacy in hard times? In this special issue of Government and Opposition, we identify a loss of solidarity as the root cause of Western political dysfunction over the past decade. The argument is developed in four parts. The first part is theoretical insofar as it sketches the causal mechanism that describes what we mean by democratic dysfunction. Here we set out some of the key concepts that are central to our project. The second part is empirical insofar as it offers four negative illustrations of the fundamental problématique, which gives us the opportunity to suggest why this collection of research articles is relevant to the contemporary debate on democracy and its discontents. The third part explores the many possible sources of democratic dysfunction, which we have organized around two thematic clusters. Here we introduce the other articles in our special issue. The fourth and last part suggests implications of living in a democratic world with waning solidarity, allowing us to draw preliminary conclusions and suggest avenues for future research.

A good and sound constitution is one under which the law holds sway over the hearts of the citizens; for, short of the moment when the power of legislation shall have accomplished precisely that, the laws will continue to be evaded. (Rousseau 1985 [1772]: 4)

Over the past decade, growing institutional dysfunction and rising popular discontent have characterized the politics of the Western democratic world. Almost everywhere you look, democratically elected governments are struggling for legitimacy. They are either mired in institutional gridlock, hamstrung by international commitments, threatened by populist anti-establishment or anti-system insurgencies, unable to cope effectively with some of the adverse consequences of globalization, or bereft of healthy growth rates due to secular stagnation or crippling public and private debt overhangs. ¹ In a few cases, governments have even been elected on outright anti-democratic platforms and, once in office, have started to chip away at pluralist democratic principles and procedures that had been taken for granted since the early 1990s. ²

In the end, you can have the best political institutions in the world, but if the people who live within them do not want to use them the way they were designed to function, then those institutions will not work. Populists across the political spectrum will be quick to exploit a crisis and question the legitimacy of the existing political system (Elchardus and Spruyt 2016; Houle and Kenny 2016; Moffitt 2015). The challenge politicians face is to make people want to use common institutions properly and to agree on what constitutes proper use in the first place. This is the difficulty that Jean-Jacques Rousseau faced in his ‘considerations on the government of Poland and on its proposed reformation’ (Rousseau 1985 [1772]). ³

Politicians and policymakers continue to wrestle with this problem at all levels of government today. Moreover, better institutions or ‘structural reforms’ were not the answer for Rousseau and they are not the answer now: ‘What is impossible is to make laws that the passions of men will not corrupt – just as they had corrupted the laws previously in effect; and to foresee and evaluate all the forms this corruption will take is, perhaps, beyond the powers of even the most consummate statesman’ (Rousseau 1985 [1772]: 3).

Democratic political institutions seem particularly vulnerable to abuse. The fall of communism was meant to herald the end of history in which liberal democracy would emerge uncontested. More than 25 years later, confidence in the ascendance of democracy has diminished significantly (Fukuyama 2014a). To some extent, that confidence evaporated almost from the outset. ⁴ Partly this is due to the surprising resilience of alternative forms of political organization; ⁵ partly it is due to the spread of democratic institutions within an illiberal context (Zakaria 2007); and partly it is due to the degeneration of democratic practice in those countries where it has long been established (Fukuyama 2014b: part IV).

The purpose of this introduction is to explore that vulnerability, focusing primarily on the problem of democratic decay. The argument has four stages. The first is theoretical insofar as it sketches the causal mechanism that describes democratic dysfunction. This is where we set out some of the key concepts that are central to our project. The second is empirical in that it offers four negative illustrations of the basic problématique. This gives us the opportunity to suggest why the special issue is relevant to the contemporary debate. The third explores the many possible sources of democratic dysfunction. Here we introduce the other articles in the issue, organized around two thematic clusters. The fourth suggests implications, allowing us to draw some conclusions and suggest avenues for future research.

WHEN DEMOCRACY BECOMES DYSFUNCTIONAL

At its core, democracy is a question of values or aspirations rather than specific institutions or procedures. ⁶ This is especially true for what we understand today as liberal democracy. ⁷ Democracy is representative, which means it should be open to a plurality of voices. Democracy is participatory insofar as it relies on some kind of personal involvement. Democracy is also supposed to be equitable. This does not always amount to ‘one person, one vote’, but it does mean that relative influence is a matter of justice. It also indicates that some form of checks and balances will protect minority rights. Democracy is accountable, which means it should be effective at delivering what the people want. Democracy is predictable: the people may not know the outcomes of elections before they happen but they should know – at least in procedural terms – how elections will be decided, when, where and by whom. The exact content of the procedures matters less than their consistent application. At a minimum, the people should have a good sense of how succession will be managed both in the alternation between rising and falling political groups and in the passage of one generation to the next. Finally, democracy is adaptive. Although the people may cherish democratic institutions, those institutions will include some mechanism for the people to update them to meet the requirements of a changing society.

This group of six characteristics (or values) is what distinguishes democracy from other systems of government. Those systems include arrangements that are not representative; where relative influence is not a question of justice; where politicians do not have to answer for their performance; where the procedures used to select politicians or to manage succession are opaque, inconsistent and unpredictable; or where the power to reform those procedures is controlled by a select few. Those other systems of government may dress themselves in the trappings of democracy. They may hold elections. Multiple groups or parties may participate. They may make concessions to minorities or other marginalized interests. They may be responsive to policy failure. And they may even try to signal who is next in line to pick up the reins of power. But these institutions are not democratic unless they meet the essential criteria, and no one is really fooled by the pretence if they do not. Democracy is like an act of obscenity in that respect: hard to define, but you know it when you see it. ⁸

Democracy is recognizable but not monolithic. It manifests itself differently, both in form and in function, across countries and over periods of time. ⁹ Democracy is like any obscene act in this respect as well. Of course, parallels with acts of obscenity are obviously inappropriate if we think in normative categories with democracy as ‘good’ and obscenities as ‘bad’. If we hold those categories – good and bad – to the side for a moment, the comparison is revealing because any appreciation of democracy (like any act of obscenity) is social, inter-subjective, and changes over time.

To understand the significance of this comparison, it is useful to establish a point of contrast. Democracy is not like beauty because beauty is too idiosyncratic and individualistic. That is why we talk about beauty as being in the ‘eye of the beholder’. Beauty is democratic in the sense that everyone gets to have their own aesthetic. But democracy is not necessarily beautiful. Everyone has to appreciate the same set of institutions and for much the same reason. Put another way, democracy is almost vulgar – like Keynes’ beauty contest, where the goal is to pick out what everyone else is most likely to appreciate – which is what you would expect from a process that is so inclusive, accountable and transparent.

The difference from acts of obscenity is that democracy is only vulgar within limits; obscenities are what free speech or artistic expression become once those limits are transgressed. The limits for what is acceptable for democracy are different from those for free speech or artistic expression. Politics is, after all, only a subset of culture. What matters is that those limits are social and not individual. When Justice Potter Stewart described his threshold test for obscenity in Jacobellis v Ohio in 1964, as he was judging whether the Louis Malle motion picture The Lovers constituted ‘hard-core pornography’ or not, he put it as follows: ‘I shall not today attempt further to define the kinds of material I understand to be embraced within that shorthand description [i.e. “hard-core pornography”], and perhaps I could never succeed in intelligibly doing so. But I know it when I see it, and the motion picture involved in this case is not that (emphasis added; Jacobellis v Ohio 1964). What Justice Stewart meant is that we all recognize the departure from common decency. He did not mean that we each have our own standards. We know democracy when we see it, because it remains within the boundaries of whatever society – no matter how different – is willing to accept (Jacobellis v Ohio 1964). By implication, we also know when the performance of democratic institutions strays beyond those accepted boundaries. This very subjective measure of democratic performance is the dependent variable in our project. The goal is to explain the mechanisms by which a set of political institutions could move from what most people would regard as a well-functioning democracy to what most people would regard as democratic dysfunction.

The vagueness with which we measure movement in the dependent variable is going to be problematic for many. ¹⁰ Part of the problem is scalar. There are huge zones of indifference within which variations in institutional performance matter little for actual perceptions of democratic performance. Another part of the problem is perception itself. The six values we regard as the central characteristics of democracy – representativeness, participation, equity, accountability, predictability and adaptability – operate across multiple connections between the individual and the state and therefore create many different kinds of perception. The way some architects use the notion of hapticity might be an appropriate analogy. The haptic aspect of architecture involves more than just seeing; it involves a range of different sensations, including peripheral vision. ¹¹ When people ‘see’ democracy they do something similar by trying to bring many different assessments into play simultaneously.

The Stewart approach to recognizing what is and is not acceptable performance from democratic institutions as well as acceptable conduct of democratically elected politicians, raises important questions about how society sets standards for political behaviour. The answers depend on two different sets of factors. One set consists of the social values related to the key elements in the political process – like representation, participation, accountability, equity and justice. Such values change only very slowly over time, not least because they tend to be encoded in formal institutions. We can think of this set of variables as ‘political culture’ along the lines pioneered by Gabriel Almond and Sidney Verba (1963). Because political culture is such a slow-moving variable, it does not look to be a likely candidate for determining sudden changes in democratic performance. ¹² Democracy can break down if the political culture changes around it, but usually the problem unfolds slowly enough that societies can mobilize to change the institutions of democracy in response (Dalton 2000). This possibility is not the focus of our project, even though it will feature on the margins.

A second set of variables relates to how society perceives itself or is bound together. This is what Robert Dahl (1989) called ‘the problem of inclusion’. We can bracket this set of variables with the term ‘solidarity’. This bracketing overlaps considerably with notions of identity. The reason for using a narrower conception of solidarity as opposed to a broader notion of identity is to capture a specific subset of identity-like variables for analysis. ¹³ Solidarity can be existential, it can be procedural, it can be distributive, and it can be aspirational. As a concept, it includes the bonds that form in opposition to a common enemy; through participation in the same institutions or respect for the same set of rules; as a result of efforts to ensure that all members of society have the same opportunities; and when members of society work together to achieve a higher purpose. Moreover, while these different kinds of social bonds are complementary, they are not necessarily interconnected and may be independent from one another. You do not have to participate in the same institutions to stand together against a common enemy or to embrace equality of opportunity in pursuit of a common aspiration.

This second set of variables is more fragile and can change much more quickly than the values that constitute political culture. Common enemies disappear or become less threatening, actors develop new strategies for participating in shared institutions, rules can lose their appeal or can be reinterpreted over time, efforts to promote equality of opportunity become burdensome or costly, and collective aspirations become less appealing. Hence solidarity can evaporate even as political culture remains undiminished. This dissipation of solidarity against a backdrop of unchanging political culture matters insofar as making sweeping reforms to political institutions is not always (or even often) going to be an appropriate response. Fundamental institutional reform is too time consuming and it would also be in conflict with underlying social values. ¹⁴ The evaporation of solidarity also matters because the weakening of bonds changes perceptions of the scope and composition of the ‘demos’ as some groups drop out while others are pushed away. Everyone may hold the same values for representation, participation, accountability and all the rest of it, but they will hold different perceptions about the legitimacy of protecting these values for different groups within the same political institutions. Again, there is a close link between solidarity and identity politics.

The evaporation of solidarity turns the implications of democratic characteristics upside down. Representation becomes polarizing as political leaders fight to privilege specific groups within existing institutions (Lowi 1971). Mobilization becomes intimidating. Institutional checks and balances become leverage points for blocking minorities to extract concessions (Crozier 1973). Voters hold their officials to account for their pursuit of ever-narrower definitions of self-interest. Predictability gives way to uncertainty as politics oscillates between pluralist stagnation and populist outburst. Efforts to reform institutions quickly collapse into zero-sum distributive political games (Thurow 1980).

Moreover, this process continues until new social bonds are formed or until the institutions can be modified to fit the lower level of social cohesion through the retreat of the public sphere, decentralization or secession. This is the theoretical mechanism connecting a weakening of the bonds that bind society together to the performance of democratic institutions. Democracy without solidarity is dysfunctional because its participants no longer view their institutions as representative, equitable, accountable, predictable or even open to reform. ¹⁵

FOUR NEGATIVE ILLUSTRATIONS

As mentioned, the range of more-or-less acceptable democratic performance is wide. Once that performance crosses beyond the threshold, however, the dysfunction is easy to recognize. To illustrate the problem, it helps to focus on extreme cases where it would be hard for reasonable people to disagree that we are indeed witnessing democratic dysfunction. The examples relate to different aspects of the democratic process, including legislative procedures (United States), government formation (Belgium), direct participation (European Union) and constitutional reform (Italy). A wider set of illustrations could touch on other democratic features. All four cases undercut key characteristics of democracy – whether they are representative, participatory, equitable, accountable, predictive or adaptive – though, as we shall see, the evaporation of any sense of solidarity is doing most of the heavy lifting. The goal of this section is therefore only to be evocative. The articles in this special issue explore a wider range of topics. They also move us closer to the boundary between dysfunction that is easily recognizable and that which is not. The individual cases in the collection are more interesting in the sense that they are more challenging to observe and much less dramatic. That is not so with our four illustrations below.

The United States

The debate about legislative procedures in the US is unusually salient. Senators and House Representatives have always complained about abuse of procedure. Nevertheless, this time it is different. The 2011 debt ceiling debate and the 2013 government shutdown are the two most prominent examples. ¹⁶ The debt ceiling debate centred on the pace of fiscal consolidation: Republicans in the House of Representatives wanted deep and rapid cuts in expenditures while Democrats in the Senate wanted to protect entitlements. The problem is not the disagreement but the use of the debt ceiling as a leverage point. Ostensibly, the debt ceiling is a piece of legislation that determines whether Congress will allow the US Treasury to raise funds to pay the difference between the expenditures and revenues that members of the same legislative chambers had authorized (albeit across multiple legislatures). On a more technical level, the question is whether Congress will allow the Treasury to raise new debt in order to redeem existing debt as government bonds mature. Without those additional resources, the Treasury would have to default on existing obligations. Dissidents in the House delayed approval of any increase in the authorized debt ceiling until a technical default became imminent. At that point, they relented and allowed the Treasury to roll over existing debt. As a result of this brinkmanship, credit-rating agency Standard & Poor’s felt compelled to downgrade the creditworthiness of the US federal government.

The 2013 government shutdown centred on divisions between the House and Senate over the passage of a continuing resolution. Normally, the federal government raises revenues and authorizes appropriations through a lengthy process that starts with a budget proposal from the administration and concludes with joint legislation by the two houses of Congress. The relationship between the administration’s proposal and the resulting legislation is not very close. Congress controls the power of the purse, after all. Nevertheless, Congress usually does (or did) agree on a budget that allows for multi-year authorizations, current appropriations, and some kind of matching programme for revenues. After the Democrats lost control over the House in the 2010 midterm elections, however, the two houses of Congress were unable to agree on a budget. This did not prevent Congress from passing specific legislation to appropriate resources already authorized, but it did complicate efforts to pass new authorizations and so to keep up the pace of appropriations.

That was precisely the point for many new Republican ‘Tea Party’ members of the House of Representatives. Their strategy was to force Congress to rely on continuing resolutions in order to maintain basic government functions until they could push for agreement on significant cutbacks in future outlays. This strategy intensified in 2013 as a number of House Republicans tried to hold up the budgeting process in order to defund the Affordable Care Act, which was a signature legislative achievement for the administration of President Barack Obama. Since the fiscal year ended on 1 October 2013 without a continuing resolution to authorize essential appropriations, the government had to shut down a number of functions. The popular reaction was as angry as it was swift, which forced Congressional leadership to move quickly and come to an agreement. However, as a consequence, the credit-rating agencies moved once again to cut their assessment of the creditworthiness of the federal government.

The actions by the credit-rating agencies are only one indicator of the widespread perception of dysfunction during these two episodes. Other indicators include the very low level of public approval for Congress as an institution (by any historical standard), the increasing volatility in electoral outcomes, and the intensity of conflict within and between political parties (and Congressional caucuses). The budgeting procedures are only the most obvious illustration that something is not working. The aggressive use of the cloture rule, the silent holds on a whole series of executive appointments, including for a Supreme Court vacancy, and the increasingly frequent breaches of decorum all point in the same direction. Moreover, the problem is not that American citizens have become less enamoured with democratic values. It is that they have become less tolerant of one another and more polarized into competing communities – rich and poor, north and south, young and old, urban and rural, religious and secular, white and non-white. These cleavages are not only deep but also reinforcing. That is why some observers suggest that the problem with American democracy is even worse than it looks (Mann and Ornstein 2012). ¹⁷

Most of the explanations for this problem lie in identity politics. The institutions for American democracy are not themselves at fault. The strategies for using them that different groups deploy are what count. There is plenty of blame to go around. Much of the recent literature focuses on the Tea Party, the Conservative movement and the white demographic groups most threatened by the increasingly multicultural nature of American society. ¹⁸ American democracy is not dysfunctional because Americans stopped being democrats (with a small ‘d’); it is dysfunctional because any sense of solidarity has broken down in American society.

Belgium

The illustration from Belgium focuses on the period of 541 days that directly followed the 13 June 2010 federal elections. That was the period during which Belgium broke the world record for government formation. The 2010 elections had resulted in very different outcomes in the north and south of Belgium. The overall winner in the south of the country was the francophone socialist party (Parti Socialiste – PS). The winner in the north was the New Flemish Alliance (Nieuw-Vlaamse Alliantie – N-VA). The francophone socialists represent continuity with Belgium’s history as a consociational democracy. Although it was briefly eclipsed by the liberal party (Mouvement Réformateur – MR) in the 2007 elections, the PS returned in 2010 with a double-digit lead over its nearest competitors in the francophone electoral districts. The situation in the north of the country was altogether different. The N-VA had no roots in Belgium’s consociational past. It had risen to power in Flanders through an electoral alliance with the once-hegemonic Christian democrats (Christen-Democratisch & Vlaams – CD&V), but the N-VA severed its ties with the CD&V during the tumultuous period that followed the 2007 elections. The Flemish nationalists of N-VA emerged in their first solo electoral contest with the lion’s share of the Flemish vote (Abts et al. 2012).

The geographical juxtaposition of continuity and change in the 2010 contest underscored just how different politics had become in the two language groups. Where the PS had a long tradition of working through Belgium’s consensual political institutions and consociational arrangements, the N-VA drew support with the promise to overturn the politics of accommodation and to repatriate both more powers and more voters for the Flemish electorate. Both arguments harkened back to the period when the N-VA was in alliance with the CD&V. The difference is that where the CD&V, as a traditional power-holder in the Belgian political system, was ultimately prepared to compromise in the interests of national unity, the N-VA was willing to prevent a federal government from coming into existence until its demands were met. From September 2010 to July 2011, the N-VA leadership rejected any compromise on its core demands. It also showed both the ability and the willingness to punish any other Flemish political party that sought to break ranks and cut a separate deal with the francophone socialists. Even after the N-VA announced its intention to stay out of the government, the other Flemish parties remained reluctant to be seen as too accommodating to the francophone socialists (Hooghe 2012). ¹⁹

The impasse only ended as a result of external pressure. International investors were slow to wake up to the fact that Belgium was operating without a federal government. Although Belgium has a large stock of sovereign debt, it also has a good reputation for public debt management. As other euro area countries came under scrutiny, investors began to look more closely at Belgium as well. And when it seemed that even the traditional political parties were unable to form a government once the N-VA had withdrawn from coalition talks, international investors began to cut their exposure to Belgian sovereign debt markets. Events threatened to spiral out of control as the spike in government bond yields rendered Belgium’s outstanding obligations increasingly unsustainable. The European Commission declared its intention to sanction the Belgian government if it did not undertake sufficient austerity measures, and Standard & Poor’s added to the pressure by downgrading Belgium and announcing a negative outlook for the future. The not unreasonable justification was that the Belgian government lacked the capacity to undertake necessary reform measures (Standard & Poor’s 2011). Belgium’s political elites duly responded by forming a workable federal coalition government in early December 2011 (Hooghe 2012).

The diagnosis of the problem was swift and focused on two different lines of argument. One is that it is difficult to legitimate a federal political organization through political parties that operate only at the regional level or within specific linguistic groups. It is even more challenging when the different regions have differently structured voter preferences (Boonen and Hooghe 2013). Within this line of argument, the progressive decentralization of political representation plays the crucial role in the Belgian drama. Once Belgium was a unified country. Today it is something closer to a mini version of the European Union (EU) (Sinardet and Bursens 2010). And, as the Belgian political scientist Marc Hooghe (2012: 137) has argued, ‘it is extremely difficult to keep a federal system fully operational if there is no federal loyalty, let alone political institutions that provide incentives for developing federal loyalty’. The second line of argument focuses on the style of politics rather than the level of representation. Belgian political institutions work to maintain a consociational democracy; they cannot maintain a majoritarian pattern of government. ²⁰ As Belgian elites have developed new strategies for interacting, well-worn formulas for consensus building have gradually broken down. Belgians have not stopped being democrats; they have just stopped wanting to treat each other as they did in the past. Belgian democracy has been functioning poorly as a consequence.

The European Union

Our third illustration comes from the tension created by national referendums around different aspects of European integration. ²¹ Such referendums strain relations between national electorates and their governing elites. They also strain relations between the member state holding the referendum and the other member states of the EU. This two-fold tension has long been evident. The first Norwegian referendum on EEC membership in 1972 not only provoked turmoil at the domestic level but also between Norway and its closest European partners. So did the Danish referendum on the Maastricht Treaty in 1992 (Einhorn 2015). Most recently, we have seen those same tensions flare up between the United Kingdom and the rest of Europe after the former’s decision to vote to leave the EU in a referendum held on 23 June 2016. The vote for ‘Brexit’ not only created frictions between the UK and the rest of Europe, but also between England and Wales – where majorities voted to ‘leave’ – on the one hand, and Scotland and Northern Ireland – where majorities voted to ‘remain’ – on the other.

As European integration has progressed, these referendums have become increasingly important. They have also become more frequent. The initial response to the EU’s democratic deficit was to try to bring Europe ‘closer to the people’ by giving increasing powers to the European Parliament, and by changing the way Europe’s heads of state and government undertake major treaty revisions. None of these actions however slowed the pace of referendums or ensured more popular attachment to Europe. On the contrary, the European Constitutional Treaty – which was the first major treaty revision organized using the new, more inclusive procedures – was rejected by overwhelming majorities in popular referendums in 2005 held in France and the Netherlands, two of the EU’s founding members. When Europe’s leaders sought to re-engineer those amendments, they did so in a more traditional manner and then looked for ways to avoid holding popular referendums (and to overturn referendum outcomes, like that expressed in Ireland in 2008, that might hold up ratification) (DeVuyst 2012).

The EU strategy of trying to avoid referendums has not worked. Neither has a related strategy of trying to avoid or defer important amendments to the founding documents of the EU. The problem is that referendums on Europe have become instruments to unite political groups within the country against the threats that appear in the outside world, to consolidate hold over political power while dividing the opposition, and even to humiliate the government.

The July 2015 Greek referendum on the conditions attached to that country’s EU-IMF bailout shows the us-against-the-world dynamic. The question did not make sense given that the bailout conditions changed the moment the referendum was called, the financial turmoil caused by the referendum itself imposed an enormous cost on the Greek economy, and the vote to reject the conditions empowered the government to accept even worse terms than had been offered. Nevertheless, Greek Prime Minister Alexis Tsipras viewed the referendum as an essential instrument to unite the country. He also saw it as a strategy to extract further concessions from other European governments (Pitsoulis and Schwuchow 2015). The result was not so much a better deal on paper as an increasing wariness among Greece’s European partners about the potential for further rounds of brinkmanship.

The British in-or-out referendum on EU membership illustrates the tactical use of referendum politics by Prime Minister David Cameron to unite his Conservative Party, consolidate power and divide the opposition. It was clear from the outside that Cameron was less concerned about EU membership when he called for a referendum in January 2013 than about retaining power through the May 2015 elections. What was less apparent was whether and to what extent he had considered the potential costs (Matthijs 2013, 2014b). Once Cameron emerged from the 2015 general elections with a narrow overall majority, he had to live up to his referendum commitment. The result in the June 2016 referendum was a disaster for Cameron’s premiership, and arguably also for his country and for European integration. Nevertheless, Cameron did succeed in uniting the Conservative Party and dividing the Labour opposition. Again, this came at the price of dividing Europe.

The April 2016 Dutch referendum on the association agreement between the EU and Ukraine shows how direct democracy can be used to embarrass the government. The Dutch opposition Party of Freedom (Partij Voor de Vrijheid – PVV) called for the referendum under legislation that allows for a citizens’ initiative to demand a vote on controversial matters of EU policy. The government could not refuse to hold the vote but it could discourage voters from participating in the hope that there would be no quorum (set at 30 per cent). That strategy narrowly failed as just enough Dutch voters (32.3 per cent) turned out for the referendum to be valid. Since most of those voters were motivated more by a desire to humiliate the government than by any understanding of the intricacies surrounding EU–Ukrainian trade relations, the vast majority voted against EU policy. As a result, the Dutch government had to figure out what – if anything – could be done to satisfy this political expression of Dutch public opinion. The rest of the EU and Ukraine were left to ask the same question. ²²

These three recent referendums – in Greece, the UK and the Netherlands – in particular have left lasting scars on the European project. These scars can be seen in the attitudes of many governments in northern Europe towards the government in Greece. They also show up in the difficult negotiations to extricate Britain from the EU. And they appear in the hesitance of a Dutch government faced with the prospect that the opposition can use any piece of EU legislation against it. The British exit is the most important of these developments and yet it is worth underscoring that the problems posed by the British referendum are not unique. European political leaders cannot ignore the possibility that other governments will use referendum politics to unite their electorates against the obligations of membership (even if the goal is not to exit) or that opposition politicians will use referendums to hold up European legislation. Moreover, these conditions apply to the routine business of European integration. When thoughts turn to fundamental reforms or new initiatives to relaunch the project, the constraints implied by popular democracy become even more binding. The European project faces an existential crisis. That crisis emanates from the dysfunction of European democracy. And once again, the lack of European solidarity is to blame. ²³

Italy

The final illustration centres on Italian Prime Minister Matteo Renzi’s efforts to reform his country’s political system after he assumed power in February 2014. These attempts culminated in a referendum that was held on 4 December 2016. The outcome of the constitutional referendum resulted in an emphatic rejection of Renzi’s proposed reforms. It is, however, revealing to explain the context, sketch the reforms and explore the underlying political dynamics at play during the lead-up to the referendum. The battle over Italy’s constitution was no less dramatic than the stories about Belgium, Europe or the US. Italy may face an existential crisis of its own in the absence of any constitutional reform.

The context is a lack of reform at all levels of government and across economic institutions as well. Like most European countries, Italy has been engaged in a massive overhaul of its welfare state (Ferrera and Gualmini 2000). More than most European countries, Italy has also undergone an overhaul of its political system. The end of the Cold War made it possible for formerly communist politicians to participate in national government. The collapse of the Italian ‘First Republic’ faced with overlapping bribery, corruption and organized crime scandals in the early 1990s forever changed the landscape of political parties. The introduction of a more majoritarian electoral system to encourage greater left–right polarization altered the practice of coalition formation. And the entry of media mogul Silvio Berlusconi into mainstream Italian politics fundamentally redefined the structure of interests on the centre-right (Ginsborg 2001).

Four things that did not change in Italy were the major political personalities; the high degree of personalization in politics; the corresponding dependence of governing coalitions on the adherence of small parties, groups or ‘currents’; and the basic structure of institutions as set out in the Italian constitution. These four elements are institutionalized in the way political parties promote candidates, form coalitions and support governments. And they explain why it is necessary for Italy to reform through broad consensus or not at all.

The new political generation was looking to facilitate reforms by removing many of the veto points that made consensus a requirement. Led by Matteo Renzi, who became Italy’s prime minister in 2014, they would have liked to change things in such a way as to strengthen institutions over individuals and to reduce the number of effective veto points (Fabbrini and Lazar 2016). The two main instruments were a new electoral law and an amended constitution.

It is easiest to present the proposed constitutional amendments first even though they were rejected in early December 2016. The goal of the Renzi government was to eliminate the balanced bicameral structure of the Italian parliament. They wanted to do so by downsizing the Senate and having its members seconded by regional governments rather than being directly elected. This way the Senate could have no influence on the stability of the national government. It would play a role in the legislative process but it would no longer be required to deliver a majority to install the government and it would be unable to institute motions of (no) confidence. Hence the government would depend only on the Chamber of Deputies in the new, asymmetrical bicameral arrangement.

The new electoral law applied only to the Chamber of Deputies and ensured that the largest political party would have a stable majority to govern (Baldini and Renwick 2015). This assurance came in the form of a two-tier majority bonus. The largest political party winning more than 40 per cent of the vote in the first round of national elections would be guaranteed at least 55 per cent of the seats in the Chamber; if no political party won more than 40 per cent of the vote, then the two largest political parties would face off against one another in a second-round runoff for the majority bonus. Moreover, the parties would compete on the basis of closed lists. In other words, the party leadership could determine who was seated once the votes had been counted at the end of the process. This electoral law came into effect on 1 July 2016, but at the time of writing looks dead in the water.

Since the people did not vote their support for the constitutional reforms during the December 2016 referendum, the Italian constitution was not amended and Italy would have a balanced bicameral legislature with two very different electoral systems for the two chambers. The prospects of forming a stable government under such an arrangement are slim. The prospects of forming a government capable of proposing a new set of reforms for the Italian constitution are bordering on non-existent.

During the campaign, the coalition in support for the constitutional reforms fractured. Most polls suggested that the ‘no’ vote in the constitutional referendum was likely to win, as indeed it did. Part of the explanation was a tactical error by Renzi, who made an early decision to personalize the referendum by saying that he would step down as prime minister if the reforms were rejected. Part of the problem was that Renzi’s enemies within his own political party worried more about what he would do to them if the referendum was successful than what would happen to Italy if it failed. Meanwhile, much of the Italian electorate was deeply disenchanted with the whole political process. They did not buy into the dream of a more effective political system the way Renzi described it and they would have rather voted to replace the whole of the Italian political class. Their vote against the constitutional reforms only served to underscore the depths of Italy’s democratic dysfunction.

A MORE GENERAL PROBLEM OF DYSFUNCTION – OVERVIEW OF THE SPECIAL ISSUE

The extreme cases we highlighted above are useful to suggest a more general problématique of democratic dysfunction. Democracy is not only performing below expectations, but it is also suffering from external constraints and inadequate responses. To explore this further, this special issue brings together six diverse articles that all look at different areas of interest. These articles are organized into two thematic clusters of three articles each.

The first cluster of three articles deals with the new sources of tension between national democracy and European integration, going beyond the old debate on the EU’s ‘democratic deficit’. ²⁴ One new fault line is in Central and Eastern Europe, where the EU is clashing with democratically elected governments’ undemocratic choices in both Hungary and Poland. This is the source of tension R. Daniel Kelemen (2017) explores. Building on the existing literature on subnational authoritarianism, Kelemen shows how efforts to strengthen democracy at the European level might create perverse incentives for politicians in other parts of the EU to excuse authoritarian backsliding by ideologically affiliated governments.

The other new manifestations of democratic dysfunction in the EU are the use of national democratic vetoes and checks and balances in the north to limit solidarity with the south, and the international constraints put on southern EU member states by the EU in choosing between various modes of economic adjustment in response to crises. Manuela Moschella (2017) examines the impact of democratic oversight over bailout conditions by legislatures in wealthy northern creditor countries. What she highlights is the asymmetry of influence. For Moschella, democracy appears to matter more in some countries than it does in others. Such asymmetry not only warps perceptions of European institutions but also fosters perceptions of inequity among all parties.

The constraints implied by European commitments are a separate concern. These constraints are particularly problematic for countries seeking conditional assistance. As Matthias Matthijs (2017) explains, such countries do not perceive their plight as warranted and hence do not accept the logic of conditionality. They will bow to that logic, but it is a question of power rather than appropriateness. Moreover, the act of acceptance raises questions about both democratic legitimacy and national autonomy, as Matthijs shows how debtor countries can neither choose between different adjustment methods nor influence the EU decision-making process. Furthermore, the results of the EU’s crisis policies in the euro periphery have been more than problematic from an economic output point of view. The Greek case is only an extreme version of this dynamic. All countries subject to conditionality have experienced it to a greater or lesser extent. The return of a wide North–South gap in economic performance serves to further undermine any claim of European solidarity.

The second cluster of three articles revisits Albert O. Hirschman’s (1970) well-known trilogy of ‘exit, voice, and loyalty’. Here we explore how democratic systems respond to the pressure of dysfunction. For Jonathon Moses (2017), that response takes place through ‘exit’ or emigration as disaffected citizens look to move to other countries in search of opportunities to build a better life. This response is particularly easy in the EU, given the existence of free movement of people within its single market. But it is also riddled with problems. Moses argues that the exit of qualified citizens from lagging EU member states to more prosperous ones leaves fewer voices behind to push politicians to improve national democratic performance at home.

The question is whether politicians are really responsive to ‘voice’ in the first place. That question lies at the centre of Wade Jacoby’s (2017) article, which focuses on the increasingly frequent use of grand coalitions. Such inclusive governments are effective insofar as they create enhanced stability in the face of a divided electorate. Jacoby’s concern is that grand coalitions may also create the impression of depriving the electorate of a meaningful voice, as there is no longer any element of real choice between policies. As a result, the electorate is driven away from the centre and towards the extremes of both right and left. Stability in the present is purchased at the expense of a lack of voice, which may result in greater instability in the future.

What remains unclear is how much such responsiveness matters and how it affects citizens’ ‘loyalty’. The answer may be found where sustained government intervention is absent rather than where it is present. This is the argument Desmond King (2017) makes to explain the manifest shortcomings of the civil rights agenda in the US. Although the Congress was able to pass progressive legislation and to introduce affirmative action, such efforts mattered little when left unattended. The same political obstructionism that we describe in the first of our four illustrations – which dealt with the US – has played a major role in ensuring that the benefits of civil rights legislation have been limited. Worse, King shows how the minority groups who should have been protected by the state were left exposed to the full force of the recent economic and financial crisis. The consequences of the crisis for minorities, especially African Americans, have been many times worse than for white Americans.

DEMOCRATIC DYSFUNCTION: IMPLICATIONS AND FUTURE RESEARCH

Implications

The most obvious implication that emerges from our collective analysis is that democratic institutions balance delicately on top of a mix of identities, incentives and perceptions. As we observed at the beginning of this introduction, you can have the best political institutions in the world, but if the people who live within them do not want to use them the way they were designed to function, then those institutions will not work. Identities are slow to change but can be exploited by populists to derail democratic decision-making. Incentives put in place to make a democratic system function properly can be used to undermine the system. Perceptions are quick to change and can lead electorates to believe that the system as a whole is no longer delivering the goods, making space for other – less democratic – arrangements to gain legitimacy over time.

A second implication is that commitment to democracy is a collective action problem at least as much as it is an individual endeavour. Well-functioning democracies are quick to generate complacency; dysfunction is an incremental process that is difficult to reverse. While an individual’s commitment to democracy as a system can be sustained by multiple factors, including self-interest, the collective action problem can only be overcome if there are sufficient amounts of solidarity present within the system, be it a city, a state or a union of states. Solidarity is precarious from the point of view that it takes a long time to grow and consolidate but can evaporate very quickly. With waning solidarity, democratic institutions will slowly decay.

A third implication is that participation has to extend far beyond the act of voting. The reason for this is simple. People cannot perceive their political institutions from behind the veil of ignorance. Therefore, they have to experience them first-hand. This means not only accepting personal responsibility for making democratic institutions function but also facing external evaluations of personal performance. Perhaps most importantly, it means alternating roles between governing and governed. Political elites play a key role in this process and will only prove durable as long as they are open and transparent enough for new members to rise and join and for existing members to fall and leave.

Future Research

Part of the problem in our current lack of understanding of the weakening of established democracies and growing political dysfunction lies in the fact that much of the literature linking economic development and democratic systems focuses on transitions from authoritarian to democratic rule in developing countries (Lipset 1959; O’Donnell and Schmitter 1986; Przeworski and Limongi 1997). By contrast, much less has been written on how to sustain the relative strength of democracy in advanced industrial states, which has largely been taken for granted. Both input legitimacy (electoral choice) and output legitimacy (rising prosperity) were implicitly assumed (Scharpf 1991, 1999; Schmidt 2012), and the comparative politics field instead focused on topics such as ‘varieties of capitalism’, the politics of welfare state retrenchment and the effects of globalization on such regimes (Garrett 1995, 1998; Hall and Soskice 2001, Pierson 1994).

As a result, while we have a good idea of how poor countries become democratic and capitalist, we have a much shallower understanding of how rich countries maintain that sweet spot where democracy and markets positively reinforce one another. In the wake of the global financial crisis and the euro crisis – with examples as diverse as Syriza in Greece, the National Front in France, Podemos in Spain, Donald Trump’s stunning rise to the presidency of the United States, and the ‘Brexit’ vote to leave the EU – continuing to surprise us, we have surely learned that democratic politics and public policymaking can undermine each other as well as reinforce one another. In such a world, understanding how political legitimacy is sustained when parties and party politics are much constrained is more important than ever.

Moreover, we need a better understanding of why, since the late 1970s, all of the states across the advanced industrial world have experienced a persistent weakening in all democratic indicators (Mair 2013). Election turnout has fallen; electoral volatility has increased; public trust in democratic leaders and institutions is at an all time low; political participation has declined steadily; and the gap between democratic aspirations and satisfaction with the way democracy works has never been wider. Yet despite all this, not enough attention has been paid to what happens to the resilience of democratic institutions when these factors combine with long periods of economic stagnation, outright declines in income, falling average wages, unsustainable debt, mounting levels of income inequality and recurring booms that only seem to benefit the few.

This is what is new and dangerous in the current moment. Democratic deficits are systemically stressful, but largely survivable; when combined with permanent political dysfunction, they can become toxic.