Introduction

One striking characteristic of the European Union (EU) as a political system is the prevalence of non-majoritarian institutions and forms of indirect or technocratic rule (Gerardin et al. 2005; Zielonka 2007). Non-majoritarianism, whether as a feature of modern governance arrangements or as a historical phenomenon, is of particular relevance to discussions about democracy and its sustainability. Put simply, in one account, the drift towards non-majoritarianism can be seen as integral to the detachment of key areas public policy from popular scrutiny. In another, non-majoritarian institutional designs are taken to be one of the most effective means of protecting democracy from ‘the tyranny of the majority’. These alternative tendencies are perhaps best understood in terms of an inherent tension at the heart of liberal government between the logics of attempting to protect the principle of representation on the one hand and the aspiration to insulate democracy from the worst excesses of mob rule on the other. This formulation presumes that democratic government is a delicate balancing act, where the task of institutional designers and those charged with working within institutional settings is to secure some sort of equilibrium between these two central operating principles. However, the positive case for non-majoritarianism—visible in recent intellectual and policy history—is often framed rather differently. In this version, non-majoritarian institutions are a central component of efficient economic governance, effectively safeguarding the market from the supposed potential of representative democratic processes to generate suboptimal policy choices. This in turn yields two institutional design impulses: the delegation of economic governance responsibilities to technocratic bodies and the movement of economic policy competences to supranational bodies beyond the reach of national politics (Slobodian 2018). This is where the EU, a project of supranational market making, becomes of special interest to scholars of non-majoritarian institutions.

The basic design of the classic EU model, in place since the early 1950s, is notable for both the prevalence of non-majoritarian institutions and the relative marginality of institutional sites that work from a direct democratic mandate. Indeed, one influential reading treats the EU as the product of an ongoing act of delegation, undertaken by member-state principals, to novel supranational agents charged with the task of making, regulating and protecting the integrity of an integrated European market order. The non-majoritarian character of the EU finds expression across its various governance methodologies. For example, ‘agentification’—the creation of multiple autonomous agencies—has been central to the regulatory mode of governance (Dehousse 2008), while the creation of de novo bodies such as the European Central Bank (ECB) sits at the heart of what some have called transgovernmental governance and others have more recently labelled the ‘new intergovernmentalism’ (Bickerton et al. 2015). Thinking along another dimension, we might also argue that non-majoritarian forms of rule, while integral to the ‘normal governance’ of integration (market making/regulation), have become the key features of EU ‘crisis governance’ (Scicluna and Auer 2019). Indeed, it is in the aftermath of the Eurozone crisis and the emergence of the EU’s post-crisis economic governance regime that the potentially coercive nature of non-majoritarian institutions has been rendered most obviously visible (Carstensen and Schmidt 2018; Kreuder-Sonnen 2016).

In this paper, we explore a puzzle that is often ignored as a puzzle. Put crudely, we need to know how non-majoritarian institutions came to such prominence in the EU, despite the fact that the institutional design of the EU was forged largely by officials representing parliamentary democracies in the supposed ‘golden age’ of democracy following World War II. There are two standard answers, both of which we explore in this paper and neither of which we find fully persuasive. The first, associated with long-standing currents within EU studies, treats non-majoritarianism as contextual, reflecting in particular the epistemic conditions within which those formative institutional design decisions were made in the 1950s. The second, emerging from the literature in critical political economy, interprets the decisive shift towards malevolent non-majoritarianism in the EU as a product of the neoliberal turn of the late 1970s and early 1980s. Both answers have merits. The first situates the non-majoritarian character of key aspects of contemporary European economic governance in the particular historical context of the EU’s foundations. The second pinpoints the key association between non-majoritarian governance and claims that seem to be central to liberal economic thought.

Our argument is compatible with both of these premises. However, and in addition, we argue that the prevalence of non-majoritarian institutions in the EU must be understood in terms of the longer trajectory of liberal theories about economics and governance—a trajectory that finds expression in both the post-war period leading up to the Treaty of Rome of 1957 and the more recent neoliberal turn that took shape some 20–30 years later. Here we point to a strong tradition of arguments and practices that seek to justify the constraint of popular governance by a range of non-majoritarian institutions, such as courts, independent central banks and upper houses. The EU has, in this context been shaped especially by two successful waves of liberal depoliticization in the latter half of the twentieth century: the ‘disciplined democracy’ currents of post-war liberalism, and the more familiar neoliberal turn since the mid to late 1970s. These two waves coincide with the two big leaps of European integration: the foundation period of the 1950s and the push of the 1980s to early 1990s that led to the internal market and the Maastricht Treaty on European Union.

Our argument here is further anchored in a normative critique of the standard case for non-majoritarianism—the emergence of ‘governmental entities that (a) possess and exercise some grant of specialised public authority … but (b) are neither directly elected by the people, nor directly managed by elected officials’ (Stone Sweet and Thatcher 2002: 2). This case tends to rest on the claim that there is no necessary conflict between the use of non-majoritarian institutions and democratic ideals. Popular majorities might not be able to translate their will into policy in the short term, but the constitutional lock-in of legality and the rule of law should protect against any concerns that input legitimacy is infringed by the growth of non-majoritarianism. Indeed, defenders of non-majoritarianism tend to place significantly more emphasis on output legitimacy as the key principle on which such governance techniques should be judged (Majone 1996, 1998). In short, if policy outputs are consistent with the purposes assigned to the institution at the point of delegation, then the output legitimacy criterion has been fulfilled. In contrast, we argue that such an approach is problematic in several regards. Aside from the empirical and counterfactual problems in demonstrating that non-majoritarian institutions do produce optimal or better policy outcomes, there is a democratic theory objection that policy delegation of this kind is likely to favour elite preferences at the expense of a broader democratic mandate. Put another way, non-majoritarianism has the potential to lock in anti-majoritarian impulses. The risk that policy institutions become not only disembedded from wider social purpose, but also actively coercive in relation to that social purpose can be construed as a form of ‘structural domination’ (Pettit 2012: 39, Ibsen 2021).

Non-majoritarian institutions in EU studies

The non-majoritarian character of key aspects of the institutional design of the early European Communities is often ascribed to the contemporaneous fashion for technocratic managerialism (Forsyth 1996; Radaelli 1999). For example, there are echoes of functionalist thought in the creation of the European Commission (the ‘High Authority’ in its earliest manifestation), as an embodiment of the idea that public welfare could be maximized through optimally situated institutions designed to deliver rational governance of human affairs (Mitrany 1966). As is well documented, the fashion for technocratic managerialism in the post-war years was motivated by two impulses that together sought to either sideline or displace societal allocation through politics. The first reflected a quasi-technical belief that modern, complex industrial societies (of whatever ideological hue) were best governed though the application of managerial reason. The second was rooted in a normative revulsion at unprecedented violence that could be unleashed between countries governed by political ideologies (particularly when fused to nationalism). These two motivations—the ‘end of ideology’ and the imperative to avoid further global conflict—intersected in many ways and had the net effect of posing questions about the scope and limits of democratic forms of societal allocation (Esmark 2020).

The story of the institutional design of the early European Communities, starting with the European Coal and Steel Community in 1951, reflected these so-called high-modernist impulses (Walters and Haahr 2005). Meanwhile, early theorists of regional integration (for whom Europe provided a compelling case from which to generalize) understood the resultant supranational institutions in ways that were consistent with their theoretical priors. For their part, neofunctionalists were less interested in understanding the origins of the European Commission and the European Court of Justice, and were rather more concerned in understanding their dual role as drivers of integration on the one hand and focal points for the interest-driven action of organized producer groups on the other (Haas 1964, 1968). In the case of the former, neofunctionalist reasoning suggested that the cumulative and functional logic of integration could be managed and stimulated through the agency of purposive supranational institutional agents. The latter revealed a key neofunctionalist premise about the nature of distributive politics in advanced industrial societies. Influenced by the pluralist view of the state that emerged within 1950s American political science, neofunctionalists worked with a group-based understanding of politics as the competition among interests for voice and influence in decision-making processes. Integration, viewed through this analytical lens was the shift of pluralist group-based interest politics from the national to the new regional level (Lindberg 1963).

Early critics (Hoffmann 1966; Hansen 1969) were able to cite theoretical and empirical deficiencies in the neofunctionalist prospectus. Put crudely, these deficiencies amounted to a radical underestimation of the continued importance of the state in general and national (member-state) governments in particular. Subsequent intergovernmentalist scholarship worked up a more elaborate account of supranational non-majoritarian institutions in the integration project. The developed ‘liberal intergovernmentalism’ associated with Moravcsik (1998, 2002) downplays both technocratic and ideological logics of integration, preferring instead to see positive sum integration achievements as the product of institutionalized bargaining among member-state whose preferences, in turn, are sourced in processes of domestic political exchange under conditions of asymmetric interdependence. The relative marginalization of supranational institutions in this perspective is not merely a matter of the theoretical primacy accorded to inter-governmental exchange. In addition, supranational institutions are themselves understood as the result of acts of delegation where member-state ‘principals’ transfer aspects of decision-making competence to supranational ‘agents’.

In the classic principal-agent account of delegation as applied to European integration, the choice to transfer aspects of decision-making authority to non-majoritarian entities reflects convergent member-state preferences around the identification of collective action problems that are logically resolved through the creation of supranational solution sets (Pollack 2003). As well as putting supranational institutions in their place, this approach also tends to downplay and/or contest concerns about a democratic deficit in the EU (classically, Moravcsik 2002). After all, if member-states are the essential sources of authority in the EU, if those member-states function as representative democracies and if the mandates of supranational institutions are tightly constrained, then it follows (from this perspective) that concerns about an absence of input legitimacy into EU policy-making become seriously misleading.

In some accounts of European integration, which, it must be remembered, involves delegation across levels of governance as well as to new governing agents, the non-majoritarian turn is a means to the efficient and effective realization of the EU’s mandate to create a functioning and properly regulated single market (Majone 1998). In Majone’s classic account, the EU should be understood as a regulatory state, an entity created specifically to create that common market across the member-states. The other classic state functions (internal/external stabilization and redistribution) are to be left to member-states. Majone’s broader analytical sweep is much more than a simple national–supranational division of labour argument. In his account, the evolution of the regulatory state is a broad cross-national phenomenon (Majone 1994). This is an important point. One of the central claims in Majone’s work is that this is not a sui generis feature of the EU, but rather evidence of a generalized tendency across advanced democracies. Indeed there are obvious connections to other influential ideas such as the ‘competition state’ (Cerny 1997; Pedersen 2010) or perhaps Streeck’s (2014) idea of the ‘consolidation state’. Thus, as Daniel Wincott (2006) has pointed out, Majone’s discussion of the regulatory state should be read not only as a compelling analytical capture of how the EU works, but also as a clear normative prescription about how the EU should work. In his 1996 working paper ‘Temporal inconsistency and policy credibility’ (Majone 1996), Majone lays this line of thinking out clearly (see also Majone 2001). Democracies are naturally short term, and therefore have to delegate to non-majoritarian institutions. This casts the EU in strictly neoliberal terms and makes a normative argument against the EU having any redistributive capacity. As such, concerns about the coercive or anti-majoritarian potentials of non-majoritarian institutions are set aside in the quest for a positive theoretical argument for non-majoritarianism.

Majone’s positive endorsement of non-majoritarian institutions takes us a step on from thinking about non-majoritarianism in European integration as either the logical institutional complement to functional integration and/or simple acts of principal-agent delegation with minimal implications for representative democratic politics. Indeed, the standard positions in EU studies discussed above are all forced to move the goalposts when it comes to assessing the democratic character of the EU. Neofunctionalists took their cue from pluralist political science to reimagine democracy as interest-driven producer group politics in which the state acts as a kind of ‘telephone exchange’, adjudicating among competing interest-based claims. Liberal intergovernmentalism, following the broad thrust of liberal institutionalist IR, is quite explicit in pointing out that integration is likely to be an attractive option for governments keen to both escape the pressure of domestic demand overload and enhance their own autonomy vis-a-vis domestic constraints (there are echoes of Milward’s ‘rescue of the nation state’ thesis here—Milward 2000). In this account, democracy—firmly rooted at the national level—serves two obvious purposes. First, as we have seen, the existence of intra-national democracy within the member-states effectively allows the EU’s legitimacy claims to be judged exclusively in output terms. Second, domestic processes of contestation (again conceived largely in producer group terms) are central to the formation of member-state bargaining preferences that are so central to the intergovernmentalist modelling of the integration process. Democracy becomes both a non-problem and a troublesome constraint on liberal states, which—in turn—augment their autonomy through acts of non-majoritarian delegation. In Majone’s version, any attempt to democratize integration (defined as the act of market making on a transnational scale) is misconceived since market making must be protected from those input pressure that might seek to subvert the simple task for which non-majoritarian institutions are trustees.

Dominance, democracy and liberal government

Our argument so far is that non-majoritarian economic governance, particularly when conducted at the supranational level, has the capacity to augment democratic deficits. Moreover, the positive case for non-majoritarian economic governance often emphasizes output legitimacy arguments that, by definition, tend to discount the importance of direct input legitimacy, which would be the most direct means to channel popular will or social purpose into economic policy decisions. The EU is an interesting site for this discussion, not least because some active proponents of non-majoritarianism in economic policy have developed their claims around and in relation to European integration. The EU’s ‘baked in’ technocratic tendencies, first visible in the institutional design decisions of the 1950s, are of further interest since they would seem to offer a foundation for later forms of coercive economic governance—such as the resort to austerity solutions in the initial phases of the Eurozone crisis and the development of post-crisis economic governance institutions and tools—that later characterize the supposed neoliberal phase of European integration. As indicated above, we do not seek to displace that cumulative account, but rather to supplement it by situating the story of non-majoritarianism within a broader understanding of the trajectory of thinking on liberal government.

In recent decades, a widespread literature has emerged on the tensions between neoliberal ideology, economic regulation and democracy. This can be seen vividly in Wolfgang Streeck’s recent work, which traces the increasing tension between capitalism and democracy (Streeck 2011, 2014, 2016) or in Wendy Brown’s Undoing the Demos (2015), which maps the corrosive effects of neoliberal ideology on the possibility of popular governance. On an international level, Quinn Slobodian’s historical account in Globalists (2018), echoing aspects of Stephen Gill’s earlier work on ‘new constitutionalism’ (Gill 1998), shows how the restriction of democracy is hard-wired into the neoliberal conception of international order, where the privileging of market freedoms and the mobility of property in the international institutional framework, can function as a constraint and limit on the working of popular democracy at the national level, effectively insulating markets from national democracy.

While much of our analysis in this article builds on the insights derived from this literature on democracy and neoliberalism, an exclusive focus on the novel aspects of neoliberalism tend to obscure some of the historical continuities associated with the problematic status of democracy and popular participation in liberal economic governance. As such, while the neoliberal wave has involved a restriction of democratic and popular participation in economic governance, both liberal anti-democratic ideas and non-majoritarian institutions clearly pre-date the full-scale emergence of neoliberalism from the 1970s onwards, and certainly cannot be reduced to secondary effects of that turn.

A main weakness in identifying the tension between liberalism and democracy in the neoliberal era lies in the mischaracterization of this era of ‘embedded liberalism’ (Ruggie 1982) as an unproblematized golden age for democracy. We can see this when Streeck tries to describe how.

Democratic capitalism was fully established only after the Second World War and then only in the ‘Western’ parts of the world, North America and Western Europe. There it functioned extraordinarily well for the next two decades—so well, in fact, that this period of uninterrupted economic growth still dominates our ideas and expectations of what modern capitalism is, or could and should be (Streeck 2011, 5)

While we acknowledge that the post-war moment involved a large democratization of society, we need at the same time to recognize the countervailing tendencies at work at the time (see also Goodman and Pepinsky 2021). As we will note in the section “Post-war liberalism and the treaty of Rome” below, there was also strong scepticism of popular rule and parliamentary power, which had a strong institutional footprint in the design of the EU institutions in the first decades of European integration. This scepticism of democracy was rooted in a very different set of preoccupations than the later neoliberal concerns. In the latter case, the main concerns were characterized by a fear of inflation and the quest for monetary stability, while in the first the main worry is the preservation of individual rights from the vicissitudes of politics. However, the remedies are surprisingly similar, namely the delegation of powers from elected parliaments to non-elected institutions of experts in order to shield these liberal preoccupations from the supposed danger and unpredictability of popular politics, as seen in the political crisis of the interwar years (Wilkinson 2021, 19).

This uneasiness with democracy as potentially dangerous and unruly, and thus in need of constraint and limitation, is not only a product of the twentieth century, but can rather be traced to the formation of liberal economic and political thought from the eighteenth and nineteenth centuries. In Models of Democracy, David Held describes classical liberal democracy as essentially a form of ‘protective democracy’ (Held 2006, 99). This protection means, on the one hand, using the state to protect life and property against the mob, and, on the other hand, using the division of powers, rule of law and (limited) representation, to protect individuals against the state. Proponents of liberal democracy envisioned protective institutions as necessary in order to protect individuals against the state, and to protect executive state power against democracy. This resulted in a set of anti-majoritarian institutions, such as powerful political courts with appointed (elite) officers and black letter constitutional limits to democracy. This governing philosophy of the prevalence of anti-majoritarian institutions can be described as ‘Madisonian’ (Bickerton 2011), in that it draws heavily on ideals of veto points, division of powers and mixed constitutions that James Madison laid out in the Federalist Papers (Madison, Hamilton & Jay, 1961). As such we try to historicize the question in two ways. First we identify the roots of the prevalence of non-majoritarian institutions as not only based in neoliberalism, but also in post-war liberal ideas of ‘self-disciplined democracy’ and ‘militant democracy’. Secondly, we embed this historical discussion of EU institutions, in a broader theoretical reading of the liberal political tradition. As well as looking at the temporal comparisons, our approach will also situate the EU in a spatial comparative context. Instead of seeing the EU as a primarily a sui generis institutional set-up, with its institutional structure determined by the unique interplay of inter-European governance structures and continued national sovereignty, we argue for the value of thinking about the EU institutions as parts of broader institutional and ideational trends, affecting nation states as well as international governance. Indeed, the key EU institutions established in the neoliberal era, notably the European Central Bank and the subsequent Stability and Growth Pact, were part of a broader wave, dating from the mid-1970s, of central bank independence and the establishment of fiscal and budgetary rule frameworks across the democratic capitalist world. In post-war liberalism, the move towards self-disciplined democracy and the fashion for powerful constitutional courts as checks on parliamentary sovereignty was rolled out in a range of European (and non-European) states in the wake of World War II, significantly before the emergence of the European Court system (on the idea of the ECJ as a form of specialized constitutional court, see Vesterdorf 2006).

On a more general level, the dominance of non-majoritarianism in the EU is also comparable most prominently with the US Constitution. Comparisons of the two systems often focus on the relative levels of centralization and the lack of a formative ‘Hamiltonian moment’ for the EU.Footnote 1 The anti-majoritarian slant of the US Constitution, though later watered down by the introduction of Senate elections and the expansion of the suffrage, was deliberately conceived. Madison argued in Federalist no. 63 that the defining characteristic of the American Constitution ‘lies in the total exclusion of the people in their collective capacity’ (Madison, Hamilton and Jay, 1961, 385, emphasis in the original). Instead, the version of liberalism typified by Madison proposed representative government and division of powers as explicitly non-democratic means of governing the polity and preventing what John Adams (2016) was, at the same time, calling ‘the tyranny of the majority’. In this understanding, representatives do not exist to channel popular sentiment into government. Rather the people surrender their transformative power to a small elite club of representatives, whose job is to protect government from both the mob and the demagogues that the mob might seek to bring to power.

In the field of International Relations theory, the anti-majoritarianism of the EU is also paralleled in the some of the central ideas of liberal institutionalism. In Democracy-Enhancing Multilateralism (2009), Keohane, Macedo and Moravcsik advance ideas of international governance that correspond with Majone’s take on the EU discussed above. When it comes to democratic legitimacy, international institutions should not be evaluated on their own democratic merits, but rather in relation to their effects on national democracies. Here they argue for a paradox, namely that by limiting the policy scope of nation states, international institutions entrench democracy through the advancement of ideas of good governance. As in the self-disciplined democracy of post-war liberalism, democracy is understood to be at its most effective when it is subject to constraint.

Post-war liberalism and the treaty of Rome

The founding treaties of what is now the EU are often read through the lenses of international diplomatic exchange, the discrepancies between and the intersections among the national interests of ‘the Six’ and the resultant bargains and side-payments. This undoubtedly helps us to understand key aspects of early European integration, particularly in terms of the inclusion and exclusion of certain policy areas (such as agriculture and social policy, respectively), but it says less about the governing philosophies informing the treaties themselves. There have been some notable discussions of the inputs of particular traditions of economic thinking into the treaties that seek to establish, inter alia, the influence of ordoliberal thought on the founding texts of European integration (Buch-Hansen and Wigger 2010; Maes 2002; Warlouzet 2019) and the treaty-specific connotations of key economic concepts (Krarup 2021). The latter are certainly important for our purposes here, but as we argue below, it is also worth considering the broader spectrum of liberal theories of government—themselves situated within the very particular contextual conditions attending the post-war consolidation of west European democratic capitalism—that sit beneath the EU’s basic treaties and inform subsequent institutional designs.

The immediate post-war era is often presented as the golden age of social democracy and, by extension, the golden age of (national) electoral democracy. In Post-Democracy Colin Crouch describes this period as a ‘democratic moment’ where the Keynesian consensus meant that ‘democratic capacity concentrated in the nation state’ (Crouch 2004, 7–8) was able to control and democratize capitalism. This idea is not without foundation, and it finds expression in Ruggie’s famous depiction of the institutional design of Bretton Woods as ‘embedded liberalism’ (Ruggie 1982), Peter Mair’s discussion of the subsequent hollowing out of ‘party government’ (Mair 2013) and indeed in the work of Streeck on democratic capitalism discussed above.

In this analysis, the non-majoritarian, non-democratic institutional framework of the early European Communities is often seen as something of an outlier or, at a stretch, complementary to the post-1945 democratic compact. The idea here would be to treat early European integration as a device to aid the opening of west European economies, while not undermining the pursuit of nationally sourced social compacts in the member-states (Rosamond 2017). This is compatible with the clear delineation in the founding Treaties between market making (which is posited as a key supranational task) and market correction (which is not) (Scharpf 2010). Of course, read differently, this division of labour appears less as an expression of the embedded liberal moment, and rather more as the historic source of coercive anti-majoritarian economic governance in the EU. As indicated earlier, the negative effects of supranational non-majoritarianism are typically assumed to kick-in later, once the neoliberal epoch is in full swing following the demise of the post-war ‘golden age’ of democratic capitalism. In other words, the malevolent qualities of non-majoritarian institutions come to be read as unintended path-dependent effects of the contextually peculiar moment of institutional design some 30–40 years earlier.

We will argue that this periodization is too simplistic. While the post-war period was, in comparative historical perspective, a time of ascendence for (some forms of) popular participation in politics and democratic regulation of the economy, this is not the entire picture. Alongside the undoubted strain of social democratic sentiment in play at the time, there was also a significant element of deep scepticism of popular sovereignty in the post-war constitutional settlement (Müller 2012, 39). Michael Wilkinson describes how in the post-war era ‘European elites attributed the collapse of interwar liberal democracy to over-politicization’ and that the relationship between state and mass democracy therefore had to be ‘reconstituted through a process of internal depoliticization’. (Wilkinson 2021, 74). This depoliticization should form an antidote to the excess of popular sovereignty and mass participation that, in this analysis, had led to the political chaos of the interwar years and the rise of fascism. This led many post-war politicians to favour strong checks on the power and policy scope of elected parliaments via the construction of non-majoritarian institutions in the judicial or administrative spheres.

The reasoning was that the potential of sovereign parliaments to hand power to would-be tyrants should be countered by the transfer of powers from elected officials to jurists in newly constructed institutional courts, or to technocrats in the administrative state. Peter Lindseth, who is generally favourable towards this development, describes the situation in France and Germany at the time as follows:

The major constitutional accomplishment in Western Europe after 1945, apart from the development of effective judicial mechanisms for the protection of individual rights, would in fact be the discovery of a workable balance between traditional parliamentarism and the broad displacement of legislative power out of the parliamentary realm and into the executive and technocratic sphere. (Lindseth 2004, 1385).

Thus when the nascent EU institutions developed first in the Treaty of Paris (1951) through to the EEC Treaty (the Treaty of Rome) (1957) were characterized by a high level of technocratic and juridical institutionalization, rather than elective mechanisms, it was not in contradiction with the general democratic spirit of the post-war era. Rather, these impulses towards supranational level market making and economic governance can be understood as close parallels to the analogous moves towards constitutionalization and depoliticization of the policy process that were characteristic of the spirit of the time.Footnote 2 The delegation of power from the majoritarian politics of national parliaments to the non-majoritarian framework of the EEC, was part of a broader attempt at delegation from the directly elected national parliaments, to a variety of non-elected institutions of bureaucratic and juridical nature. As such the ‘democratic deficit’ was baked into the European project from the very earliest conception through the dominance of non-elected institutional forms of decision making, with the central role of the ECJ in the institutional framework (Alter 2010) and the Commission—an unelected core executive at the heart of the bureaucratic apparatus of the EU. Moreover, the economic and political liberalism of the post-war era was not completely incompatible with the dominance of Keynesian macroeconomics. While the Keynesian project sought to save liberal capitalism, by combining it with the sort of social democratic politics that especially ordoliberalism tried to shield against, Keynesianism also exhibited what Geoff Mann describes as ‘an unshakeable faith in techno-bureaucratic expertise’ (Mann 2017, 55). This meant that in practice the sort of depoliticized non-majoritarianism functioned rather smoothly with a certain form of Keynesian technocracy.

We identify two main proximate intellectual sources of this development: the idea of ‘militant democracy’ developed by Karl Loewenstein and the ordoliberal tradition most associated with Wilhelm Röpke and Walter Eucken. Both strands of theory developed analyses of the interwar crisis as an excess of popular power and both argued for institutional fixes to retain a stable liberal and capitalist society in the era of mass democracy. In these traditions, non-majoritarian solutions are designed precisely to curtail or dilute popular power. In ordoliberalism, the goal is to use the constitution to protect the market economy from the corrosive effect of mass democracy and popular power, while the aim of militant democracy is to protect the very institutions of democracy against the inherent dangers associated with f mass participation in politics.

Militant democracy

The idea of the self-disciplined democracy grew out of a particular post-war interpretation of the rise of fascism. In this analysis, the problem of fascism was rooted in an excess of popular sovereignty that threatened rule of law. Consequently, democracy in the post-war era had to restrict popular sovereignty in favour of constitutional protections. The earliest formulation of these ideas came from German-Jewish émigré Karl Loewenstein, and his idea of ‘militant democracy’. Loewenstein’s account of militant democracy was formulated across two articles in American Political Science Review, both published in 1937 (Loewenstein 1937a, 1937b), in which he analysed how democratic systems can combat fascism by actively suppressing authoritarian forces, and at the same time adopting elements of the authoritarian playbook to defend themselves against threats. This involved two main elements—the first an active ban of the organization of fascist and non-democratic parties, and secondly the use of strong constitutional courts to restrict the power of legislative majorities (Loewenstein 1937a). He described how democracies needed to undergo a ‘deliberate transformation of obsolete forms and rigid concepts into the instrumentalities of “disciplined”, or even—let us not shy away from the word—“authoritarian” democracy’ (Loewenstein, 1937b, 657).

Loewenstein’s ideas were especially influential in a German setting, both with the ban on extremist parties on the left and right, and in the strong constitutional court that was a feature of the constitution of the new Federal Republic (Müller 2011). They also acquired traction within debates in democratic and constitutional theory where the broad-spectrum idea of restricting basic rights to secure and stabilize democratic exchange has been much discussed (Cappoccia 2013; Malkopoulou & Norman 2018). The part that spread to the European level was the concept of a strong constitutional framework, and a range of non-majoritarian institutions, primarily the European Court of Justice, as a way of supervising and policing this framework. Jan Werner Müller puts it like this:

European integration was part and parcel of the new ‘constitutionalist ethos’, with its inbuilt distrust of popular sovereignty, and the delegation of bureaucratic tasks to agencies which remained under the close supervision of national governments. Member countries consciously gave powers to unelected institutions domestically and also to supranational bodies in order to ‘lock in’ liberal-democratic arrangements, and to prevent any backsliding towards authoritarianism (Müller 2011: 149)

Thus, the’militant democracy’ turn in inter- and post-war liberal thought reconciles democracy-protecting constitutional depoliticization at the domestic level with the creation of European-level governance institutions that manifestly suffer from what is typically known as a democratic deficit. The idea, and its specific connection to European integration, has received renewed interest in the light of two much more recent political tendencies: the justification for limitations on freedom of association and religious freedoms in the light of Islamicist terror attacks in the West (Accetti and Zuckerman 2017) and proposals to use militant democratic solutions against the current crisis of democratic backsliding within EU member-states, most notably Hungary and Poland (Wagrandl 2018).

Ordoliberalism

In parallel with militant democracy, Ordoliberalism emerged from at the interwar years, with a distinct critique of majoritarian politics. Overall, the ordoliberal project can be seen as a way of reacting to the challenge to liberal economic governance presented by the rise of mass democracy in the early parts of the twentieth century. The rise of mass politics meant that stark new challenges emerged for the liberal state, with demands rising from a plethora of new political movements and pressure groups. In order to control the potential ungovernability of mass democratic society, a strong constitutional framework needed to be constructed in order to protect the market economy and the integrity of state institutions.

Eucken here describes how ‘[d]emocratisation grants political parties and the masses and the interest groups organised by them a massively increased influence on the government of the state and thus on economic policy as well’ (Eucken 1932, cited in Bonefeld 2017). Representative democracy, such as that which had briefly taken hold in Germany after 1918, was held by Eucken to have politicized the economic order to an unprecedently dangerous degree (Wegner 2019). This scepticism of democracy was not unique to the ordoliberals. Indeed, as we saw above, key parts of the classical liberal tradition held a very ambivalent view of democracy. Of course, liberal political thought changed significantly over the course of the nineteenth and twentieth centuries, with the ambivalent position coming under pressure from societal movement of workers, women and other excluded groups to accept universal suffrage and a general widening of the repertoire of political and social rights granted to citizens (Mulvad and Stahl 2019; Therborn 1977).

It is as a reaction to this development that the ordoliberal project must be understood (Manow 2001). Rather than accept the democratization of liberal society as a fait accompli, the ordoliberals wanted to tame and control the new mass democracy. Working from what Dyson (2019) labels a ‘conservative-liberal mindset’, ordoliberals feared that the alternative to such a domestication of mass democracy would mean the end of liberal society of free markets and private property. Röpke in The Social Crisis of Our Time cites classic liberal figures such as Madison, Tocqueville and John Stuart Mill, as inspirations for a view that.

democracy – and democracy more than any other political system – can lead to the worst forms of despotism and intolerance if bounds are not set to it by other principles and institutions, and it is this limitation in all its aspects that we must call the liberal content of a political structure. (Röpke 1950 [1942], 85)

The solution here is a rule-based order, set through an ‘economic constitution’ (Eucken 1992, 314). The constitutionalization of the principles of property and economic competition would put the principles of economic liberalism at the very core of the political order, alongside the framework of the political regulation of the state (Böhm, Eucken, and Grossmann-Doerth 1989 [1936]).

Put briefly, the ordoliberal impulse was bound up closely with the idea of protecting the liberal economic order of freely competitive markets against the distortive and destabilizing pressures of representative democratic politics. Without denying the nuance and complexity of ordoliberal thought—not to mention the changes of emphasis in the project over time—there is nevertheless a clear pathway in this tradition towards advocacy of authoritarian means to secure economic freedom (Wilkinson 2019). Non-majoritarian bodies become central players in the project of ensuring that economic order remains viable. In the contemporaneous work on militant democracy, the central objective was to prevent democracy from destroying itself. Once again, the attenuation of popular power is consistent with non-majoritarianism.

Put simply, the idea of militant democracy and the emergent tradition of ordoliberalism converge on a common solution set—non-majoritarianism—to the diverse problems that they identify. This solution, in turn, became a positive philosophy of government that was central to the epistemic context of the EU’s founding treaties. Neither approach trusts democracy to either look after itself or to look after market, in part because democracy is understood in a crude plebiscitary sense. Both share a fear (perhaps an expectation) that the will of the people will be automatically mobilized against democracy itself and/or against market society. The depiction of democracy as inherently prone to tyrannies of the majority is an important premise for both militant democrats and ordoliberals since, by definition, it underestimates the capacity of democratic systems to design institutions and procedures that facilitate both the flow of popular preferences into policy-making and the protection of the rights and interests of minorities.

Neoliberalism and the crisis of democracy

When seen though the genealogy of non-majoritarian institutions in EU economic governance, neoliberalism does not stand as point of divergence, but rather a continuation and radicalization of already existing tendencies. The anti-majoritarian tendency in liberalism re-emerged in force with the crisis of the post-war compromise. Despite the initial impact of anti-majoritarian thinking on the design of post-war institutions set-up, electoral democracy did indeed have some of the effects feared by liberal sceptics such as Eucken and Röpke. Universal suffrage meant that voters demanded, and politicians delivered, a discernible expansion of tax-financed welfare services across Western European countries, and full employment policies—as predicted by Kalecki (1943)—did push the balance of power on the labour market in favour of labour and unions. As is well established, during the era of Keynesian dominance, such scepticism had little place. But with the advent of the economic turbulence of the 1970s and the loss of legitimacy for policies associated with Keynesian demand management, an economically focused critique of democracy became politically salient. The reason was the perceived inability of the prevalent instruments of Keynesian demand management to curb the rising economic threats of unemployment and inflation.

While much of the discussion of neoliberalism has evolved round the issue of state versus markets, it can just as well be seen in terms of debates about the supposed excesses of democracy in the post-war welfare state. Indeed at the time, the stagflation crisis of the 1970s was seen as a symptom of the increasing inability of the democratic state to secure the stability necessary for the continued functioning of the capitalist market economy. The principal problem identified within this discourse was the apparent inability of governments to secure price stability and curb rising inflation. For example, in 1975 the newly formed Trilateral Commission published a highly impactful report—entitled The Crisis of Democracy—that sought to trace the relevant pathologies of what it labelled as a ‘crisis of ungovernability’ (Huntington et al. 1975). The excess of democracy and the power of special interest groups, such as unions and new social movements, had left western states ‘overloaded’ and unable to successfully manage societal inputs into the state.

According to this account, associated as it was with the rise of the ‘new right’ across the capitalist democratic world, in the economic sphere overload was especially manifest in inflation and what was seen as the irresponsible growth of public expenditure. Democratic states, unable to resist the demands placed upon them by populations and interest groups, inevitably turned to inflationary policies or took on increasing amounts of debt.

Inflation can be considered a direct result of the ungovernability of Western democracies. It is an easy answer to the tensions of growth. The less a society is capable of facing them, the readier it is to accept inflation as a less painful solution. (Crozier et al. 1975, 37)

Of course, it is important to recognize that contemporaneous analyses, from across the intellectual and political spectrum, were converging on a perception that the governing formula associated with the post-war democratic capitalist compact had, by the mid-1970s, come under severe stress (Block 1981). Disagreement did not turn on whether or not there was a fiscal crisis of the state in advanced capitalist countries, but rather on the sources of that crisis and the resultant implications for governance. For example, O’Connor’s classic Marxist account (O’Connor 1973) draws attention to the legitimacy concerns that arise from slowing economic growth and the mismatch between government revenues and increasing expenditure. But key to his analysis is an understanding of the crisis as endemic to capitalism, which in turn illustrates the state’s complex dual role in relation to the population on the one hand and to capital on the other. The new right version of the overload thesis was much less concerned with the democratic legitimacy implications of the crisis of capitalism and much more focused on how an excess of democracy was posing a threat to the maintenance of capitalism itself.

First and second generation neoliberalism

In the theoretical realm, this revival of scepticism of democracy had several strands. The ‘first generation’ version of ordoliberal or Hayekian neoliberalism, was based on the experiences of the interwar years, while the second generation, centred around the neoclassical liberalism of the Chicago School, was inspired more by the problems that were identified as endemic to the post-war order (Burgin 2012, Stahl 2019).

Hayek first laid out the dangers of the unbridled economic consequences of the democratic welfare state in The Road to Serfdom (Hayek 1944), and his later Law, Legislation and Liberty imagined a liberal constitution that would constrain direct popular influence on the rule of law, by restricting the suffrage and transplanting legislative capacity to an upper house with 15-year terms (Hayek 2013), as a way of countering the corrosive effects of mass democracy on liberty and prosperity. Wilhelm Röpke imagined the independent central bank as the foremost bulwark against economic democracy. In A Humane Economy Röpke states that: ‘Never has it been more essential to keep money out of politics than in our age of mass democracy’ and that ‘[i]ndependent central banks seem to be among the Bastilles which give our modern Jacobinism no peace until they are razed to the ground’ (Röpke 1960, 195).

While these critiques were first developed against the background of the political turmoil of the interwar years, the economic critique had less of an immediate impact than the political critique discussed above. The high growth rates of the post-war era rendered less credible the pessimism towards mass democracy voiced by the first generation of neoliberals. The instruments of Keynesian demand management and social compromise did indeed seem to make democratic capitalism economically workable, while arguably the extant non-majoritarian institutions kept democracy in place. Therefore, it was not until the 1970s that this economically focused critique of democracy became politically salient, not least because of the perceived inability of the Keynesian paradigm to fix the macroeconomic imbalances followed by the broader crisis of legitimacy that followed.

But there was also a new generation of critics of social democracy, developed with inspiration from Milton Friedman and the Chicago school version of neoclassical economics. Instead of framing the critique of democracy in the form and style of the political economy of Hayek and the ordoliberals, this new critique was rooted in the latest ‘cutting-edge’ neoclassical economics and the highly mathematicized econometric models associated with it. Much of this new generation was inspired by Milton Friedman's work, but it found its most influential formulation in the new classical macroeconomics of the likes of Robert Lucas, Thomas Sargent and Edward Prescott (Stahl 2020; Offer and Söderberg 2016).

The main thrust of this newer critique was that government action in the economy, however well-meant and benign, is doomed to fail because of the anticipations of the rational actors who populate markets. The adaptive expectation of these actors would automatically price the long-term fiscal and monetary consequences of current government policies into their current investment decisions (an updating and revival of the idea of ‘Ricardian equivalence’). As a consequence, the discretionary policy scope of governments and parliaments had to be restricted in favour of policy rules. Lucas and Sargent formulated this as ‘the need to think of policy as the choice of stable rules of the game, well understood by economic agents’ and further added that ‘policies (…), such as monetary instability and deficit financing, have the capacity only to disrupt’ (Lucas and Sargent 1979, p. 15).

The problem here was often framed in term of the inconsistency and unpredictability of democratic policies. As Kydland and Prescott (1977)—like Lucas, winners of the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel—laid it out, the problem of democratic governance of fiscal and monetary policy was the inability of elected governments to make credible commitments. Because of the ability of future governments to renege on promises made in the present and the natural short-term thinking of office seeking politicians, rational investors in the private sector would be forced to assume an expectation of rising taxation or inflation. This would in turn lead to lower levels of investment, depressing future economic growth. The solution was framed in terms of removing discretion and setting up time consistent policy rules (Clift 2020).

In other words, the interpretation of the neoclassical liberals was that the crisis of macroeconomic policy of the 1970s was basically a contradiction between the needs of investors in the market on the one hand and the aims of the voting public in the political realm on the other. In order to solve the resultant macroeconomic problems, this contradiction had to be resolved in favour of investors. Here the aim was not a rollback of the state, but instead an internal reorganization of the state, such that the balance of class forces within the state was tipped in favour of investors.

The point here is to highlight the distinctive quality of the intellectual and political moment within which the EU’s own neoliberal turn was situated. In so far as the developments in economic thought cited above had concrete political expressions, then these can be seen in the general trend towards the depoliticization of monetary policy (via the creation of operationally independent central banks), the growing emphasis on fixed, quasi-constitutional rules rather than political discretion in macroeconomic policy-making (Menéndez 2022) and a general diminution in the scope of the state to manage the economy. The translation of these impulses to the supranational level in the EU is discussed in great detail elsewhere including in this special issue (for example, Menéndez 2022, Ryner 2022), but of course the most obvious expression of this depoliticized, rules-based, non-discretionary regime is found in the institutional design of the Euro.

Conclusion

Our analysis suggests a new perspective on the discussion of economic liberalism and democracy. In contrast to much of the scholarship within International Political Economy and EU studies, this perspective stresses the continuities, rather than the contrasts between the embedded liberalism of the post-war era and the neoliberalism of recent decades. We do this by broadening the perspective from a focus on solely the ‘economic’ side of non-majoritarian institutions to an analysis of the broader political and constitutional role that they play in varieties of liberal thought. By focussing on the prevalence of non-majoritarian institutions in this broader context, the break from embedded liberalism to neoliberalism, looks more like a difference in degree, than a difference in kind, with aspects of liberal governance being intensified over time.

We believe that this perspective not only adds historical nuance to our understanding of European political and economic development, but also allows us to more fully grasp the current moment of crisis. The contradictions facing Europe at the moment not only have their roots in the economic tensions created by the neoliberal project since the 1970s and 80 s, but rather to much deeper-set liberal ideas of depoliticization that have been part of the European integration project since its inception. What is currently visible is not simply a crisis of the neoliberal mode of policy-making in the EU, but rather the potential unravelling of the foundations of the political–economic architecture set-up in the post-war moment.

But this is not only a story of continuity. While the neoliberal movement contains clear continuities with post-war liberalism, when it comes to scepticism of popular power, there are also clear differences. In post-war liberalism the focus was to secure the core of democratic capitalism, through constitutional safeguards of basic building blocks like individual rights and private property, against encroachment by popular forces. It was to this end that ideas of depoliticization and anti-majoritarian institutions were mobilized. While (especially) the early ordoliberals had broader critiques of public welfare or union power, the political realities were that the post-war social compromise allowed for the rise of the welfare state and powerful labour unions. The depoliticization was to serve as a shield against fascism or communism, the mid-twentieth century heirs to Madison’s mob-driven demagogues of the 1780s.

In the neoliberal wave, the project became more ambitious. It was not only the basic conditions of capitalism that needed to be shielded from popular participation, but also (and especially) the ability of democratic governments to use monetary and fiscal policies to redistribute property and income that needed to be curtailed. Put bluntly, depoliticization was to serve as a shield against social democracy.

It is notable how, in Majone’s work, the issue of economic efficiency and the idea of individual human rights are bound together and legitimized in a seamless way. Again this—the conflation of human rights and economic liberalism—is a generic feature of liberal thought, particularly in the euphoric ‘end of history’ moment of the early 1990s that followed the fall of the Berlin Wall. However, in a post-crisis world this intermingling—where the defence of individual rights is tied into an institutional framework that undermines economic sovereignty and restricts the political autonomy of democratic majorities—becomes potentially toxic. Throughout the Eurozone crisis, non-majoritarianism performed their normal function of acting as if the governance of the crisis was not an appropriate matter of political contention. However, the severe imbalances that emerged in Europe as the global financial crisis morphed into a supposed crisis of sovereign debt in the Eurozone raised deep distributive questions that were inevitably politicized at the domestic level, leading not only to anti-austerity movements of the left, but also to Euroscepticism and forms of nativist populism on the right. In the case of the latter, the evidence of recent years seems to suggest that the societal backlash against the effects of economic depoliticization can spill over into a backlash against the very frameworks that exist to protect individual and minority rights.Footnote 3