Abstract
Despite the significant attention devoted to their birth and death, the day-to-day operation of coalition government remains understudied. This article addresses this lacuna and sheds light on the dynamics of coalition governance by examining the interplay between macro-level institutions, meso-level values and micro-level practices. Focusing on the Conservative – Liberal Democrat Coalition that governed the United Kingdom between 2010 and 2015, this analysis reveals the extent to which the everyday practice of coalition governance is flexible, contingent, and proceeds through informal negotiation and accommodation. It also draws attention to the dilemmas faced by coalition actors in terms of reconciling competing loyalties and appeasing a wide range of audiences. Through this analysis, the article makes an important distinction between the ‘rules-in-form’ and ‘rules-in-use’ of coalition governance and between the different ways that coalition governance is enacted on the ‘frontstage’ and ‘backstage’. Together, these findings point to an important new avenue of research for coalition scholars.
Keywords
The formation of coalition government has been a major concern of comparative scholarship, with significant attention devoted to who gets in and who gets what in terms of parties, portfolios and policies (see for example Laver and Schofield, 1990; Martin and Stevenson, 2001). Similarly, the termination of coalition government has been subject to much analysis, as scholars have sought to explain when and why coalitions fall (see Laver, 2003; Warwick, 1994, 2001). However, it has been argued that ‘[w] happens between coalition formation and termination is still poorly understood’ and that ‘the territory remains largely uncharted’ (Müller et al., 2008: 35). This lacuna is a significant one, and one that matters for two reasons. First, across parliamentary democracies worldwide, and within western Europe in particular, coalitions are the predominant government type (see Andeweg et al., 2011; Müller et al., 2008; Müller and Strøm, 2003c). Yet despite their prevalence, we know relatively little about the everyday practice of coalition governance. Second, and relatedly, some have regarded coalition as an outward manifestation of a polity’s capacity for inclusive and consensual decision-making (notably Lijphart, 2012, but compare Matthews and Flinders, 2017). However, for such normative claims to substantiated, a clear understanding of the contextual conditions under which coalition governments operate are imperative.
This article addresses this gap, and focuses on the everyday practices of the Conservative-Liberal Democrat Coalition that held office in the United Kingdom between 2010 and 2015. Compared to the parliamentary democracies of western Europe, the United Kingdom has limited experience of national-level coalition, as the structures of Westminster are purposefully calibrated to ‘manufacture’ (Rae, 1967) a legislative majority for the plurality winning party. Yet, it is precisely because of this seeming disconnect between the principles of majoritarianism and the practice of power-sharing that this case should be regarded as ‘critical’ (Flyvbjerg, 2006: 230). It provides a unique opportunity to isolate the effects (and effectiveness) of formal and informal mechanisms of coalition governance within a polity hitherto dominated by single-party majority executives. As such, this article will make a number of broad contributions to the study of coalition governance and specific contributions to the study of British government.
First, it responds to the demand for situated analyses of coalition governance (Martin and Stephenson, 2001; Müller et al., 2008; Müller and Strøm, 2003a), and by providing critical insights from the United Kingdom, makes an important empirical contribution to extant comparative scholarship, which has largely focused on the ‘consensual’ (Lijphart, 2012) democracies of western Europe (see Andeweg et al., 2011; Müller et al., 2008; Müller and Strøm, 2003c). Second, it dovetails with a burgeoning body of literature that examines how political actors have mediated the ‘dilemmas’ arising from clashes between constitutional traditions, institutional rules and governing reality (see Bevir and Rhodes, 2003; Marsh and Hall, 2016; Matthews, 2015; Turnbull, 2016). Third, its findings provide further impetus for the emerging scholarly turn towards ‘informal governance’, and the focus on the interplay between formal ‘frontstage’ posturing and informal ‘backstage’ accommodation in governance transactions (see for example Ayres et al., 2017; Friedman, 1995; Klijn, 2014). Fourth, the article offers a counterpoint to existing studies of British government, which have (understandably) largely focused on the dynamics of intra-executive relations (Bevir and Rhodes, 2003; Marsh, 2008; Smith, 1999) and political leadership (Dowding, 2013; Foley, 1993, 2000; Heffernan, 2003) in the context of single-party majority government. Finally, recognising that ‘no act of coalition politics can be understood in isolation from others that may occur earlier or later’ (Müller and Strøm, 2003a: 5), it provides a timely contribution to the study of British politics in a period when such governments are increasingly exceptional.
To develop these strands, the article proceeds as follows. The next section brings together several hitherto separate strands of literature to demonstrate the necessity of locating the operation of coalition government within its wider institutional context, and develops a critical distinction between the ‘rules-in-form’ and ‘rules-in-use’ (Ostrom, 2005) of coalition governance. Following on from this exercise, the case of the Conservative-Liberal Democrat Coalition is analysed. Drawing on a range of primary materials, including previously embargoed interviews with actors at the heart of the government, this section examines the relevance of formal mechanisms and informal relationships for the everyday practices of the Coalition, and in doing so explores the ways in which its members sought to navigate the demands of multiparty politics within a highly adversarial majoritarian polity. The article concludes by locating these findings within a series of theoretical debates regarding the relationship between ‘frontstage’ and ‘backstage’ governance practices, and identifies a number of areas through which this research can be developed.
The rules-in-form and rules-in-use of coalition governance
Designing the rules-in-form
While significant scholarly attention has been devoted to coalition formation and termination, the day-to-day operation of coalition government has been relatively understudied. Indeed, this ‘stages approach’ has been criticised for neglecting the dynamic, cyclical and anticipatory character of coalition governance, which requires its participants to ‘anticipate and influence what will happen from the time they form their government until the time of their next election’ (Müller et al., 2008: 10–11; see also Warwick and Druckman, 2006). As suggested, coalition governments are required to remain alert to the potential for preference divergence, political opportunism and (unforeseen) external shocks in order to avoid their untimely demise. Recognising this reality, several studies have drawn attention to the variety of formal arrangements that coalitions adopt to manage the dispersal of office and policy payoffs, the resolution of disputes, and the timing of future elections.
Perhaps the most widely utilised tool is the formal coalition agreement, typically published immediately after the conclusion of coalition negotiations (see Indridason and Kristinsson, 2013; Müller and Strøm, 2008). Coalition agreements have been variously described as a ‘register of policies that coalition parties wish ministers to implement’ (Moury, 2011: 386), a ‘pre-commitment device by which the negotiating parties bind themselves to the mast’ (Müller and Strøm, 2008: 165), or quite simply the ‘bible’ (De Winter et al., 2000: 322). Coalition agreements have been regarded as playing an important role in reconciling ‘the fundamental tension between standing apart and sitting together’ (Timmermans, 2006: 264), constituting ‘a basic method for containing ministerial, or party, drift in cabinets’ (Indridason and Kristinsson, 2013: 825), and nearly all agreements lay down (some of) the intra-coalition rules of the game (although very few deal with the specifics of portfolio allocation publicly) (Müller and Strøm, 2008: 175). Coalition agreements are also regarded as good indicators of the attention an issue will receive (Moury, 2011; Müller and Strøm, 2008; Walgrave et al., 2006).
By ‘constraining the actions and policies that can be pursued after the government is formed’ (Strøm et al., 2010: 521), coalition agreements thus play an important role in reducing uncertainty. However, they cannot fully resolve the challenge of enforcement, and it is common for coalitions to rely on a combination of measures to ‘police the bargain’. While the risk of agency loss in the process of delegation from the cabinet to individual ministers is commonly experienced (see Andeweg, 2000), such challenges are more acute in multiparty cabinets as ministers may also hold partisan motivations to exploit their office. Moreover, while parties may be able to monitor ‘their’ ministers, they enjoy no such control over ministers from other parties (Thies, 2001: 580). To manage these risks, coalitions could establish rules to screen the selection of candidates, for example, granting all coalition partners veto powers over appointments. In the real-world of coalition politics, however, such controls are not the norm. As Strøm et al. (2010: 521) argue ‘given that government office often is the main goal of party leaders and their lieutenants, it is not surprising that they do not want to jeopardize such ambitions by granting veto rights over cabinet appointments to their coalition partners’.
Instead, evidence suggests that coalitions are more likely to institute monitoring and reporting arrangements to oversee policymaking. One such tool is the so-called ‘watchdog’ (Müller and Strøm, 2003b) or ‘hostile’ (Thies, 2001) junior minister, whereby parties appoint junior ministers to departments headed by ministers from different parties in order to keep tabs on their partners. The literature suggests that the prevalence of such junior ministers is contingent upon a number of institutional and political factors. Thies (2011), for example, demonstrates that the practice is most used in countries with few institutional checks upon ministerial discretion and is most concentrated upon those portfolios deemed salient to the monitoring party. These findings are echoed by Lipsmeyer and Pierce (2011), whose research also shows that watchdog junior ministers are more likely as the ideological distance between the minister and the coalition increases, and by Falcó-Gimeno (2014), who demonstrates that when parties’ policy interests overlap, but their programmatic positions are tangential, they are more likely to allocate watchdogs.
Coalitions can also establish specific bodies for the management of conflicts between the parties, which vary in terms of venue and formality. Müller and Strøm (2003b), for example, distinguish between six different arenas (inner cabinets, issue-specific cabinet committees, coalition committees, mixed committees of ministers and parliamentary leaders, committees of parliamentary leaders, and party summits), and their comparative research identifies the various ways in which such arenas are utilised by the coalitions of western Europe. Elsewhere, Andeweg and Timmermans (2008) group these arenas according to three categories (those internal to the cabinet, those external, and ‘mixed’), and distinguish between those arenas most commonly used, and those which are used for the most serious conflicts. In seeking to explain the different approaches to conflict resolution, their research suggests that the existence of a coalition agreement is a key predictor: coalitions that have adopted a coalition agreement are more likely to rely on internal arenas, whereas those without one are more likely to resort to external arenas to resolve their disputes.
In combination, such mechanisms function as an institutional check on the actions of ministers. At the same time, the formal constitutional rules of a polity, such as those regarding votes of confidence or the dissolution of the legislature, also structure the wider operating environment of coalition governance. Saalfield (2008), for example, has shown that coalition cabinets are more likely to endure if they do not have to survive an investiture vote in the legislature, whereas positive parliamentarianism increases the risk of cabinet replacement. Similarly, Diermeier and Fedderson (1998: 611) have shown that provisions for votes of confidence are positively associated with coalition cohesion, providing an ‘incentive for ruling coalitions to vote together on policy issues that might otherwise split them’. In contrast, coalitions are more vulnerable to early termination where prime ministers enjoy unilateral dissolution powers, which provide heads of government the opportunity to cut their electoral losses or exploit favourable external circumstances (Saalfield, 2008: 354).
Implementing the rules-in-use
The above studies draw attention to the variety of formal arrangements that are intended to enable coalitions to oversee their policy objectives, manage disputes, and insure against early dissolution. Yet while these studies have greatly enhanced our understanding of the ‘rules-in-form’ that structure the inter-election period, we still know relatively little about their relevance to the everyday practice of coalition governance. Indeed, one of the critical insights of neo-institutional scholarship is the extent to which the practised ‘rules-in-use’ can diverge from the prescribed ‘rules-in-form’ (see Ostrom, 2005). In some respects, this knowledge gap reflects the main theoretical and methodological traditions in the field of coalition studies, in which formal game-theoretic bargaining models and large-n quantitative analyses predominate. Undoubtedly, such approaches have made important advances in terms of hypothesis development, data generation, and the identification of empirical regularities. Nonetheless, the theoretical parsimony and empirical scale of such studies has often precluded a detailed understanding of the interplay between the formal and informal dynamics of coalition governance.
This gap in our knowledge matters. While it has been suggested that a polity’s constitutional architecture is ‘[e]xogenous to the coalition game in the sense that the actors in this game cannot have any realistic hope of changing [it], at least in the short-term’ (Müller and Strøm, 2003a: 4), this ‘game’ takes place within the long shadows that it casts. The practice of coalition governance is inexorably bound up with the traditions and norms to which this architecture gives rise, and the institutional ‘stickiness’ to which Müller and Strøm allude merely underscores this point. It is for such reasons that Blondel and Müller-Rommel caution against placing too much weight on formal institutional structures, suggesting that coalition governance is the product of the ‘interplay between “values” and the “structural instruments”’. Recognising the stickiness of such ‘exogenous’ institutions is not to say, of course, that they are static or impervious to change. As Thelen (2002: 101) has argued, the seemingly ‘remarkable resilience of some institutional arrangements’ may serve to obscure ‘ongoing subtle shifts beneath the surface’. Nor does a focus on traditions and norms assume that they are unconsciously inherited or unquestioningly maintained. Indeed, an important strand of interpretivist scholarship has argued that traditions are ‘contingent, produced by the actions of individuals’ (Bevir and Rhodes, 2003: 4, 33), and that the dilemmas which can arise when a ‘new idea stands in opposition to existing beliefs or practices’ may force a ‘reconsideration of these existing beliefs and associated traditions’ (Bevir and Rhodes, 2003: 36; see also Marsh and Hall, 2016; Matthews, 2015; Turnbull, 2016). Instead, the challenge for those who seek to explain the operation of coalition government is to unpack the relationships that exist between macro-level institutions, meso-level values and micro-level practices.
To meet this challenge, it is necessary to bring ‘country differences’ (Blondel and Müller-Rommel, 1993: 10) back into the field of study, and to produce ‘thicker’ descriptions that account for how they are experienced by those at the heart of coalition government. In recent years, qualitative case-driven research has fallen out of favour, criticised for being empirically rich but theoretically poor: ‘the basic features of coalition politics’ being ‘better known than understood’ (Müller and Strøm, 2003c: 559). Yet, despite such criticisms, case studies have several distinct advantages, particularly in terms of understanding political actors as conscious, social beings whose experiences, motives and subjective interpretations are an important part of the causal process (Bennett and Elman, 2006; Yin, 2003). As such, case-driven research can illuminate hitherto hidden practices of coalition governance, the contingency of traditions and norms, and how the ‘rules-in-form’ are interpreted on the ground. Moreover, as well as providing detailed empirical knowledge, the analysis of carefully selected cases can offer insights of wider comparative relevance and can contribute to the development of theory (see Flyvbjerg, 2006). In this respect, the case of the Conservative-Liberal Democrat Coalition, which governed the United Kingdom between 2010 and 2015, provides a critical opportunity to disentangle the contingent effects of formal institutional structures and informal governing norms on the everyday practice of coalition governance.
To do so, the following section deploys two-step research design. First, to capture the rules-in-form of the Coalition’s governance arrangements, the analysis utilises the comparative institutional framework developed by Müller and Strøm (2003c). This framework organises the formal institutional structures of coalition governance into 10 separate dimensions, and enables the United Kingdom’s experience to be located within a wider comparative context. Second, in order to capture the rules-in-use of the Coalition’s governance arrangements, the analysis draws upon the tools of case-driven research to examine the extent to which these formal rules mattered for the everyday practices of the Coalition. In doing so, it drills down to address a number of inter-connected issues such as the ways in which its members sought to reconcile the imperatives of multiparty politics with the norms of majoritarianism, the importance of less visible or more informal ‘backstage’ dimensions of coalition governance, and the deleterious impact of ‘frontstage drama’ (Friedman, 1995) in terms of governing trust.
Each step rests on the systematic analysis of a wide range of relevant empirical material. To specify the formal dimensions of coalition governance, official materials were examined, including the (revised) Ministerial Code (Cabinet Office, 2010), The Coalition: Our Programme for Government (HM Government, 2010b), the Coalition Agreement for Stability and Reform (HM Government, 2010a) and the Cabinet Manual (Cabinet Office, 2011). In turn, to identify the frontstage rhetoric of coalition governance, ministerial speeches, statements and other relevant government papers were analysed. To identify the informal backstage dimensions of coalition governance, it was necessary to look beyond such public proclamations. Face-to-face interviews were carried out with key actors during the final 12 months of the Coalition, including ministers, government and opposition backbenchers, and civil servants at the heart of the action. To encourage candour, interviewees were offered anonymity. And interviewees were indeed candid! As such, interviewees are identified in broad terms only (such as ‘Conservative minister’ and ‘government backbencher’) in order to ensure anonymity. The second stage of the research also drew upon a number of other sources, in particular the Institute for Government’s ‘Ministers Reflect’ database, which includes verbatim transcripts of interviews with ministers that served under the Coalition (58 ministers in total, 15 of which sat in cabinet). Other sources included relevant select committee reports, along with their accompanying written and oral evidence (including House of Lords Constitution Committee, 2014; Justice Select Committee, 2010); the small number of political memoirs that have since been published (such as Clarke, 2016; Clegg, 2016; Laws, 2016a) and extra-parliamentary sources such as leaders’ party conference speeches. The net result of this research is an analysis that enables the United Kingdom’s experience of coalition governance to be situated within the wider comparative context, while still being attuned to prevailing country ‘differences’. The next section puts this enabling into effect.
Coalition governance on the frontstage and backstage
Adapting the institutional environment
On 11 May 2010, just 5 days after the inconclusive general election, the Conservatives and the Liberal Democrats announced their intention to form the United Kingdom’s first executive coalition of the post-war period. Despite the seeming ideological distance between the two parties, their respective leaders, David Cameron and Nick Clegg, were optimistic about the prospects of the deal brokered. Indeed, for the new Deputy Prime Minister, this coalition provided the opportunity for ‘a new kind of government’ and ‘the new politics I have always believed in: diverse, plural, when politicians of different persuasions come together to overcome their differences’, which reflected his long-standing aspiration to emulate the coalitions of continental Europe (British Broadcasting Corporation (BBC), 2010).
In practical terms, the realpolitik of governing together within a polity organised around the principles of adversarial power-hoarding resulted in the rapid development of mechanisms to manage expectations and minimise disputes, as detailed in Table 1 below. Just 9 days after its formation, the Coalition published its coalition agreement, comprising two publicly available documents. The first, and most substantive, of the two set out the Coalition’s policy agenda. At over 13,000 words in length, spanning 31 distinct policy areas, and with over 400 separate policy commitments, The Coalition: Our Programme for Government (HM Government, 2010b) can be regarded as a large and comprehensive statement of intent (see Müller and Strøm, 2008: 176 for comparative data). Nonetheless, while comprehensive, the agreement was incomplete as there were several policy areas for which compromise was politically inimical or ideologically impossible. In relation to these issues, the Programme for Government included provisions for the Liberal Democrats to ‘abstain in any vote’, ‘maintain opposition’ or ‘to make the case for alternatives’.
The institutional rules of coalition governance.
Questions and coding derived from Müller and Strøm (2003: 20–21).
Aggregate of 238 observations across 13 cases (Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Italy, Luxembourg, Netherlands, Norway, Portugal, Sweden). Data derived from Muller and Strom (2003: 574).
The shorter second document, Coalition Agreement for Stability and Reform (HM Government, 2010a), delineated a series of ‘expectations’ regarding the composition and functioning of government; its policy and legislative programme; support for the government in Parliament; and public appointments. Several of its 18 rules stood in contrast to comparative experience and constitutional precedent. On ministerial appointments, for example, the proportional distribution of cabinet, ministerial and whip positions was entrenched; and the capacity of the Prime Minister to make appointments, redesign portfolios or undertake reshuffles was further limited by the veto of the Deputy Prime Minister. Elsewhere, provisions were included to relax collective responsibility ‘where it is explicitly set aside’, while also allowing for the consideration of ‘any other exceptions’ that may later arise. In turn, the Ministerial Code was revised in 2010 to allow collective responsibility to be relaxed in specific instances (Cabinet Office, 2010: 3). The Coalition Agreement for Stability and Reform also set out expectations regarding the application of the party whip on all issues in the coalition agreement and on ‘all matters of confidence’. It was, however, silent on all other aspects of parliamentary behaviour such as the questioning of ministers or behaviour within legislative committees. While neither document entailed provisions for the event of a breakdown, such as triggering automatic elections, the Programme for Government did pledge to ‘legislate to make provision for fixed-term Parliaments of five years’, subsequently enshrined in the Fixed-terms Parliament Act 2011. This pledge contrasted sharply with constitutional precedence, wherein the prime minister had previously enjoyed unfettered powers to dissolve Parliament at a time of their choosing. It was, however, one of the Liberal Democrat’s key manifesto commitments, and one of the party’s ‘red lines’ in the coalition negotiations, as it sought to insure itself against strategic dissolution and the subsequent risk of electoral misfortune (see Clegg, 2016: 128).
To police the coalition bargain, a number of strategies were pursued. In accordance with the provisions of the coalition agreement, ministerial positions were dispersed on a near-proportional basis, with the balance tipped slightly in favour of the Liberal Democrats (Table 2 below). Portfolios were only allocated after the policy platform was agreed, a ‘sequence’ which Clegg (2016: 30) ‘insisted on’ as he ‘didn’t want to lose leverage over those decisions by getting drawn into early horse trading about ministerial posts’. However, in terms of portfolio salience (Warwick and Druckman, 2006), the picture is more equivocal. Data from the Comparative Manifesto Project show that the issues of education and the environment received significantly more attention in the Liberal Democrats’ manifesto when compared to the Conservatives’ manifesto (see Debus, 2011). However, while the party won control of the Department for Energy and Climate Change, they were unable to secure the education brief. Moreover, rather than advocating for one of the ‘great offices of state’, the Deputy Prime Minister instead chose the Cabinet Office (albeit with responsibility for overseeing the Government’s constitutional reform agenda). In a similar vein, both parties installed ‘watchdog’ junior ministers to oversee departments headed by their partner. However, this shadowing was incomplete, and several important departments controlled by the Conservatives lacked a Liberal Democrat junior minister, including the Department for Environment Food, and Rural Affairs (2010–2012), the Department for International Development (2010–2012) and the Foreign Office (2012–2015).
Distribution of ministerial positions, 2010–2015.
LD: Liberal Democrat.
This encompasses only those departments headed by a Secretary of State and HM Treasury.
Finally, to resolve disputes, a number of bodies were established. The main formal arena was the Coalition Committee, an official cabinet committee with terms of reference to ‘manage the business and priorities of the Government and the implementation and operation of the Coalition agreement’. This Committee was co-chaired by the Prime Minister and the Deputy Prime Minister, and had equal (not proportional) representation from both parties. A smaller ‘informal working sub-group’, the Coalition Operation and Strategic Planning Group, was also established with a remit to ‘consider and resolve issues relating to the operation of the coalition agreement, the longer-term strategic planning of government business and to report as necessary to the Coalition Committee’. Together, these two bodies were intended to provide a two-stage conflict resolution process. The Coalition Operation and Strategic Planning Group was envisaged as an intermediate arena where most coalition matters would be dealt with, and the Coalition Committee was envisaged to be the final arbiter for any unresolved issues (for an overview, see Hazell, 2012). In addition to these two formal bodies, an informal ‘inner-cabinet’ quickly emerged, which became known as ‘the Quad’ and comprised four key ministers from the two parties; Cameron, George Osborne, Clegg and Danny Alexander.
The formal architecture of coalition governance detailed above was comprehensive, encompassing all of the ex-ante and ex-post tools detailed in the preceding section. Evidence suggests that this formal machinery assumed an important frontstage role in terms of ‘setting the scene’, providing a series of signals and cues to a number of different audiences about how the Coalition would operate. Of particular significance was the Programme for Government coalition agreement. Several interviewees spoke of how its publication served to bind the two parties to their shared endeavour, particularly at the outset. One Conservative cabinet minister described this ‘blueprint’ as a ‘fairly comprehensive document with no room for squabbles’ (interview with author), and another minister praised the coalition agreement for making explicit the ‘huge compromises’ between the two parties, which meant that ‘we were all going into this with eyes wide open’ (interview with author). Moreover, for the Liberal Democrats, so long out of power, the agreement was a source of optimism, as reflected in David Laws’ (2016a: 18) account of ‘jubilant’ MPs who were ‘delighted by just how much of our manifesto we had managed to negotiate into the document’. Interviewees also reflected on the role of the Programme for Government in coalescing the wider party around this shared set of policy objectives. The internal democracy of the Liberal Democrats meant that the agreement received the active endorsement of its members, and a number of specific provisions were identified as providing important assurances to those who had expressed reservations about the prospect of power-sharing with the Conservatives (see Laws, 2016a: 10–18). In contrast, several Conservative backbenchers expressed serious concerns about their lack of consultation. One stated that ‘there was absolutely no deliberation. It felt as though you were being hit by an oncoming train’ (interview with author). Another suggested that the parliamentary party was ‘effectively bounced into it’ (interview with author).
Later into the Coalition’s time in office, the agreement also assumed prominence at moments of ‘high drama’, when public recourse was made to its provisions. This point was vividly illustrated during 2012 when, in response to the collapse of Conservative support for the House of Lords Reform Bill, the Liberal Democrats withdrew their support for constituency boundary reforms. While both policies appeared in the Programme for Government, they were unrelated. Nonetheless, this apparent infidelity to the agreement was used to justify inter-party dissent, as the Deputy Prime Minister deplored the Conservatives’ ‘pick and choose’ attitude towards political reform and their failure to ‘honour’ the ‘contract’ (BBC, 2012). The agreement also featured in critical negotiations between the Coalition’s leaders, and was often invoked in private to head-off potentially damaging inter-party disputes by ensuring a united front on politically contentious policy commitments such as National Health Service (NHS) reforms and public sector cuts (see Clegg, 2016; Laws, 2016a). By setting the broad policy agenda and providing a channel of overarching accountability, the Programme for Government operated as an important, often symbolic, backstop. In terms of the day-to-day operation of the Coalition, however, it assumed less significance. As one senior civil servant explained, while the Programme for Government ‘provide[d] a bit of discipline in terms of prioritisation’, it was not ‘in any way the main focus of government activity’ (interview with author).
Establishing informal governance arrangements
Instead, it was the less visible backstage that was of critical importance in terms of making policy, managing tensions and mitigating the impact of frontstage drama. Here, formal party labels mattered less, and effective informal governance was often the product of shared political values and norms, a common sense of endeavour, and personal amity and accord. In terms of shared political values and norms, the overlap between the two parties was much greater than should be anticipated from a coalition comprising supposed ‘arch enemies’ (Laws, 2016a: 559). Having had no previous experience of working together, several interviewees were surprised at how much alignment existed between the two parties. As one Conservative cabinet minister recalled: A Liberal Democrat minister … turned to me and said ‘we had no idea that you lot were so liberal’. I replied, saying we had no idea they were so conservative … [W]e were quite surprised by how close we stood together on issues (interview with author).
Of course, this overlap was not complete, and there remained clear differences between the parties on a range of issues. Nonetheless, the existence of such differences was frequently cited as a positive force in terms of robust policymaking. As Conservative minister Stephen Crabb explained, this ‘party political tension [was] on the whole a good thing’ because ‘it meant that issues and policies had to fought over a bit harder … [and] thought through a bit more’ (Crabb, 2016).
At the same time, the two parties found common political ground in their opposition to the previous Labour government. As David Laws (2016a: 560) made clear, ‘much of our agenda could be defined against the previous Labour government’, and there was a ‘distinct sense of overarching mission – to clear up the economic mess we inherited from the previous administration’. Such a common sense of endeavour was particularly evident within departments, and several interviewees spoke of how a shared departmental mission fostered an esprit de corps. Reflecting on the painful process of welfare reform, Liberal Democrat Minister for Pensions Steven Webb (2015) said ‘all the ministers sat round the table with the list of horrible things to do … And that bonded us … in a bizarre kind of way’. This point was reiterated by his Secretary of State, Iain Duncan Smith, who described how they became ‘very good friends [with] a high degree of trust between us. We didn’t have any problems about sorting out different coalition priorities’ (Duncan Smith, 2016). The importance of trust was reiterated by several other ministers. ‘Over time trusts builds up’, one Conservative minister explained, and ‘[w]e discovered that actually governing together was easier than we thought it was going to be and quite effective’ (interview with author).
It was therefore unsurprising that ‘a lot of personal friendships sprang up’ (cabinet minister, interview with author), and several interviewees reflected on how personal amity and accord was a vital channel of informal governance. In many instances, personal connections pre-dated the Coalition. One Conservative minister, for example, recalled how an earlier school-gate acquaintance with a Liberal Democrat colleague contributed to effective cross-departmental working, and another recalled how their existing friendship with the spouse of one of their coalition colleagues provided a common bond during their time in government together (interviews with author). The significance of such amity cannot be overstated. David Laws, for example, recalled how the senior members of the coalition – Cameron, Clegg, Osbourne, Alexander, Letwin and Laws – ‘all got on very well’, thus avoiding ‘the type of personal animosity which would otherwise have undermined the government’ (Laws, 2016a: 560). In particular, the shared background, age and social outlook of the Prime Minister and his Deputy was widely regarded as vital in cementing the Coalition. As one Labour backbencher drily observed, ‘Clegg was also socially comfortable with Cameron. Both public school boys, both the same height, same good looks, same haircut. A very easy Rose Garden love-in for them’ (interview with author). Moreover, while Nick Clegg’s (2016) subsequent memoirs were often highly critical of his erstwhile colleague, David Laws (2016a: 560) credited the two with ‘establishing and maintaining [a] friendly atmosphere, which lasted right through to the end’.
Of course, not all coalition relationships were as harmonious as those described above, as revealed in the memoirs of several coalition members (Clarke, 2016; Clegg, 2016; Laws, 2016a). However, there is little evidence to suggest that such discord was solely the product of partisan differences, and as one senior Conservative minister made clear, ‘the divisions in government are by no means always on party lines’ (interview with author). Nonetheless, despite the importance of these channels of informal governance, the backstage management of the Coalition was subject to a number of frontstage pressures, in particular the challenge of reconciling competing loyalties, the management of the wider parliamentary party, the desire to differentiate, and the pressure of prospective elections. More broadly, the day-to-day practice of coalition was beset by a number of institutional limitations and incongruencies as, despite the raft of initiatives detailed above, the Whitehall machine did not readily adapt to the demands of multi-party government.
Navigating the dilemmas of coalition governance
As we have seen, coalition partners typically appoint ‘watchdog’ ministers to guard their party’s interests. In office, however, ministers are required to reconcile competing loyalties – to the party, to the government, and to their department – and several interviewees reflected on the tensions which resulted: I would be getting some mood music out of Nick Clegg’s office that this was all far too collegiate and cooperative when they wanted more grit in the oyster. I think well I can’t just be grit in the oyster because I’m also a minister in the Government having ministerial team meetings with the Home Secretary. And I can’t behave like I’m an opposition MP, I’m a government minister (Browne, 2015). But my view was if you were the Minister of State, with a Conservative Secretary of State … then you were the Lib Dem minister for that department. I know they allocate responsibilities within the department, but you were also the Lib Dem minister that should be looking at other things in that department that weren’t necessarily your primary responsibility (Wallace, 2015, emphasis added).
Other interviewees suggested that the demands of departmental business resulted in ‘surprisingly little contact with other ministers’ from their own party (Laws, 2016b), as ‘the departments did tend to cocoon you’ (McNally, 2015). Such tensions were further exacerbated by the way that Nick Clegg utilised the Liberal Democrat’s share of appointments, where a conscious trade-off was made between depth and breadth. Several ministers reflected on the sense of isolation in which this strategy resulted. In the Department for Transport, for example, Baroness Kramer explained how ‘as a Liberal Democrat, I was quite isolated and I would think “if only we were all Liberal Democrats in this department, we would really crack this problem”’ (Kramer, 2016).
At the same time, the management of the wider parliamentary party was often fraught. As described above, many Conservative MPs felt ‘bounced’ into the Coalition, which resulted in animosity between the front and back benches. Indeed, one senior backbencher suggested that ‘the tribal loyalty between the Conservative backbenchers and the government is diluted [as] you start from the premise that “this isn’t my administration, it’s a coalition”’ (interview with author). Moreover, there existed a degree of enmity between the two parties’ backbenchers. As one senior backbencher put it, ‘there was a sort of barmy post-marital jolly period … but we have reverted to normal hostilities’ (interview with author). The loosening of party loyalty was evidenced by the fact that the 2010–2015 parliament was the most rebellious of the post-war period, and despite provisions for the application of the whip on all Programme for Government pledges, Coalition MPs rebelled in 35% of all divisions (Cowley, 2015). Indeed, one cabinet minister suggested that their party’s backbenchers ‘tend to angrily blame the Liberal Democrats as an excuse for voting against the government’ (interview with author). Yet despite their prevalence, rebellions were rarely in concert, and government defeats were relatively infrequent. Reflecting on this result, several interviewees suggested that a degree of rebellion had been permitted as a means of managing intra- and inter-party tensions. Indeed, some interviewees speculated that these rebellions were stage-managed by the whips. ‘They’re taking it in turns’, suggested one Labour backbencher; ‘there’s a clear policy of “you’re allowed to rebel this week so that you can go back to your constituency and say you don’t agree with this or that”’ (interview with author).
Frustrations were further compounded by a number of institutional limitations and incongruencies, as the Whitehall machine remained unaligned to the exigencies of coalition. As Conservative minister David Willets drily observed, ‘the machine found it hard to handle coalition, especially at first’ (Willetts, 2015). Several junior Liberal Democrat ministers drew attention to the asymmetry of resources and support, such as that offered by special advisers. Despite the existence of two distinct parties within government, the rules regulating their appointment within the Ministerial Code were unchanged, which allowed cabinet ministers to appoint a maximum of two advisers, and for ministers attending cabinet to appoint one (Cabinet Office, 2010: 6). This provision had clear consequences for the Liberal Democrats’ policy capacity: I depended a lot on my spad [special adviser]. But again, I only had a small part of his time. He was a very good spad, but one that worked across several departments, whereas the Conservatives had, I think, three spads totally focused on transport and on getting a much more Conservative-biased transport agenda established (Kramer, 2016).
In a similar vein, Nick Clegg was irritated to discover that ‘[w]hile the 200 or so staff in Downing Street leapt into action to support the new Prime Minister, I was given just a single civil servant’, which left the Deputy Prime Minister ‘trying to do my job at one end of the coalition see-saw with a tiny fraction of the support provided at the other’. It was ‘embarrassingly clear’, he concluded, that the civil service ‘had no idea how to serve a coalition government … and there was no substantive thinking … as to how it might effectively become so’ (Clegg, 2016: 53–4). Indeed, Clegg’s (2016: 94) memoirs suggest that this institutional asymmetry, along with Number 10’s determination to protect ‘Cameron’s image as the government’s commander-in-chief’, was significant in the souring of his relationship with the Prime Minister, as he became increasingly frustrated with his inability to assert himself as co-author of the Coalition.
There was also evidence to suggest that cabinet and its committees lacked the capacity to facilitate cross-party discussion and collective decision-making. Whitehall observers have long lamented the superficiality of cabinet discussions, and under the Coalition there was little evidence to suggest its revival. Indeed, one cabinet minister described its meetings as a ‘very odd setting’ where ‘business is contrived to avoid any possibility of anyone debating what is actually the biggest political issue of the day’ (interview with author). A number of interviewees were also critical of how several committees were run by their respective chairs. One Conservative minister described the cabinet committees as ‘very poor. Nick Clegg ran all of ours’ and he took too much on. He couldn’t cope with the workload. Easy decisions were taking weeks’ (Djanogly, 2015). Another suggested that some committees had been willfully neglected as a means of quashing inter-party tensions: It becomes clear very early on [that] the Economic Affairs Cabinet Committee is not meeting and is not a functioning committee … George [Osborne] didn’t want to have Vince [Cable] coming in regularly and speaking across the whole range of domestic and economic policy (Willetts, 2015).
It should be noted that not all interviewees shared this pessimism. Oliver Letwin, for example, regarded cabinet committees as ‘very, very useful things because you could have the formal debates and you could also have informal discussions outside them and then they’d crystallise in the formal debates’. Moreover, by suggesting that ‘the committee structure combined with the Programme for Government did provide a framework within which those detailed and ad hoc discussions could always be set’, Letwin ascribed to the committee system an important role in the meta-governance of the Coalition (Letwin, 2016).
Nonetheless, most interviewees acknowledged that the formal machinery established to manage coalition disputes had been superseded by informal arrangements. In particular, the Quad quickly assumed prominence as ‘the real engine-room of decision-making’ (Clegg, 2016: 79). In this important backstage venue, ‘meetings were always open and frank … and were largely conducted in good faith and remained constructive long after coalition cooperation in other departments had broken down’ (Clegg, 2016: 79). This result led to accusations that the Quad had undermined collective decision-making. One senior Conservative minister suggested that ‘far too many things go to the Quad’, which had become an ‘unhealthy’ substitute for formal cabinet discussion (interview with author).
Yet precisely because so many issues were resolved informally and in private, the Liberal Democrats were concerned that ‘no-one knew about’ their ‘positive impact on policies’ (Laws, 2016a: 123). Unsurprisingly, a tension emerged between the competing demands of ensuring intra-government cohesion while maintaining inter-party individuality, the so-called ‘unity-distinctiveness dilemma’ (Boston and Bullock, 2009). This desire to differentiate was most apparent during party conference season, when the leaders of each party sought to appease the party faithful and reach out to the wider electorate with eye-catching policy announcements and claims to coalition successes. Yet while such set-pieces were played out for the benefit of these frontstage audiences, evidence suggests that behind the scenes, these events were carefully stage-managed: Each year during the coalition, we would discuss for some weeks the policy announcements that each party wanted to make at its conference … Each party would draw up its list of policy priorities in private … and we would then compare lists to see if agreements could be made … [T]he two lists [would] then go to a political Quad to seek agreement (Laws, 2016a: 202).
Nonetheless, while the Quad was intended as a clearing house for such policy announcements, it is evident that frontstage considerations often prevailed, as illustrated by David Cameron’s continued willingness to claim credit for policies that owed their existence to the Liberal Democrats. This practice extended to the latter’s flagship policy of raising the income tax threshold, which Cameron presented to the party membership as an implied Conservative success: ‘We’ve given real help to the poorest and most vulnerable. We’re taking over a million of the lowest-paid people out of tax altogether’ (Cameron, 2011).
There was also a significant temporal dimension to the widening of the gap between private backstage discussion and public frontstage performance, which was driven by the pressure of prospective elections. As anticipated by the ‘unity-distinctiveness’ dilemma, the approach of the 2015 general election witnessed each party distancing itself from its coalition partner, and at times outright disparagement prevailed, particularly on the part of the Liberal Democrats: Cows moo. Dogs bark. And Tories cut. It’s in their DNA … They’re not even pretending they want us to be ‘all in this together’ any more. The Conservative plan for the next government is an ideological lurch to the right. They have gone from being the self-proclaimed heirs to Blair to Nigel Farage in white tie (Clegg, 2015).
Nick Clegg’s (2016: 38) memoirs paint a vivid picture of his increasing frustration with the Conservatives, who he came to regard as ‘the intellectual magpies of British politics – brilliant at picking up anything that shimmers and shines, shamelessly pinching other people’s prized possessions for their own purposes’. Indeed, the way in which David Cameron announced in 2014 that ‘that a future Conservative Government will raise the tax-free personal allowance from £10,500 to £12,500’ (Cameron, 2014) can be regarded as evidence of this supposed kleptomania.
Overall, therefore, the day-to-day governance of the coalition was increasingly challenging as the parliament progressed. Indeed, the Coalition endgame was widely regarded as the most ineffective period of government. Vince Cable (2015), for example, described the last few months of government as ‘quite ragged and frustrating … all we did was scrap the whole time … [having] endless, endless fights over who gets the credit for things’. Within departments, a de facto purdah commenced several months before the election was formally announced (senior civil servant, interview with author), and there was a wide sense that the Coalition had ‘run out of steam’ (Conservative backbencher, interview with author) and ‘run out of things to agree on’ (Labour frontbencher, interview with author). A number of Liberal Democrats also expressed their exasperation at the Conservatives’ seeming shift to the right, with David Laws (2016a: 560) remarking that the supposedly ‘“compassionate Conservative” agenda did not survive for long once in government’. However, it is possible to construct an alternative reading of this growing frustration, which draws attention to the Liberal Democrats’ transition from the edges of opposition to the heart of government, and the extent to which optimism and principled purity were confronted by realism and political pragmatism. Several interviewees commented on the challenge of reconciling these tensions, with one suggesting that ‘we never resolved whether we wanted to be a governing party or whether we wanted to be an opposition party within government’ (Browne, 2016).
Conclusion
This article has sought to explain the everyday practice of coalition governance, as experienced by the Conservative-Liberal Democrat Coalition between 2010 and 2015. The inconclusive general election result of 2010 provided an important opportunity to experiment with a form of politics largely untested at the national level, where the norms of adversarial majoritarianism prevailed. As this article has shown, this experiment resulted in the creation of a series of formal mechanisms to accommodate the exigencies of power-sharing within an institutional framework more attuned to the demands of single party majorities. In some respects, these institutional innovations were typical of the broad trends in coalition governance identified in comparative scholarship (Müller and Strøm, 2003b). However, whereas many countries in western Europe have extensive experience of coalition cabinets, and have gradually adapted their institutional terrain over several decades, the substantial programme of measures implemented in the United Kingdom was marked by its rapidity. The speed by which these changes were instigated reflects the majoritarian structures of the British polity, in particular the capacity for institutional innovation afforded by a constitution with few formal veto-points. Nonetheless, despite the seeming incongruence of consensual power-sharing within a polity steeped in majoritarian tradition and norms, the Conservative-Liberal Democrat Coalition proved comparatively durable and lasted for the full 5 year term specified by the Fixed-term Parliaments Act 2011.
However, while this machinery provided a series of formal ground rules that structured the inter-election period, this article has shown how coalition governance often proceeded through informal channels, and rested on shared political values and norms, a common sense of endeavour, and a degree of personal amity and accord. At the same time, the article has also revealed how the backstage practice of the coalition was often challenged by a number of frontstage pressures, as coalition actors were required to reconcile a number of competing loyalties (party, government, department) and appease a wide range of audiences (backbenchers, activists, voters). Indeed, this article demonstrates that while the ‘backstage’ of coalition provided the space for actors to ‘relax from their roles, step out of character and work with their dramaturgical teammates to prepare for the front stage performance’ (Friedman, 1995: 93), accommodation and compromise was often ‘bounded’, occurring within the shadow of prevailing institutional norms. The article has also underlined the extent that coalition governance was subject to important temporal dynamics, as the optimism surrounding the birth of the coalition gave way to the reality of coalition life, and the prospect of coalition death encouraged its partners to focus on securing their individual legacies.
More broadly, this article points to an important new avenue of research, which moves away from large-n comparative research and towards detailed case analyses that treat coalition actors as situated, contingent, and contextually bound. By specifying the different dimensions of formal and informal coalition governance, and the different frontstage and backstage arenas in which these transactions occur, this article provides a useful and transferrable analytical framework to capture the multi-dimensionality of coalition governance. In turn, by identifying the key challenges that exist in terms of frontstage pressures and institutional limitations, this article directs attention to the way in which coalition actors navigate the demands of multiparty politics, manage the multiple roles that they must occupy, and respond to the dilemmas to which these challenges give rise. At the same time, there are a number of specific issues that warrant further investigation. The case analysed in this article focuses on a system of government ill-suited to the demands of coalition, and on two parties unused to governing together. A question mark therefore hangs over the extent to which previous experience generally (whether a party has experience of coalition) and specifically (whether the parties have experience of working together) affects the interplay between the formal and informal dimensions of coalition governance. Such research is undoubtedly warranted. Existing scholarship has furnished us with a great deal of knowledge regarding the birth and death of coalitions, and by taking this agenda forwards, the net result will be an accumulation of knowledge that enables scholars to understand and explain the everyday practice of coalition governance.
