Not just ‘something for non-executives to do?’: Party reaction to the opening of the ‘theatre of representation’ in English Local Government

PSA 60th Anniversary Conference: Elections, Public Opinion and Parties Specialist Group (EPOP) Paper 913 on Panel 169: Electoral Systems and Political Arrangements Not just ‘something for non-executives to do?’ Party reaction to the opening of the ‘theatre of representation’ in English Local Government Mark Ewbank ESRC Doctoral Candidate, Institute of Local Government Studies, University of Birmingham This paper presents the results of a quantitative study of overview and scrutiny councillors from the three major political parties in English local government. From the change in political arrangements as a result of the Local Government Act 2000 which brought about a separation of powers in local government, the paper examines the impact on non-executive councillors and how this has affected party group control of local decision-making. Given the wealth of evidence of the strength of party groups in local authorities (Maud 1967, Widdicombe 1986, Copus 1999, Copus & Leach, 2004) there are questions over how party group behaviour has changed since this legislation (OPDM, 2002, Ashworth 2003(a&b), ELGNCE, 2004, 2006, Ewbank 2007). With new roles developed for councillors, the ‘party’ has come under pressure to submit itself to self-scrutiny, public criticism from groups and pressure to reduce the impact of camera obscura decision-making. This quantitative research aims to investigate, using an amended selection of Harmel, Janda, Tan & Heo’s (1995) variables of party change, whether party groups have modified their behaviour since the split, within Kaare Strøm’s theory of competitive (national) political parties (1990, 1999), and whether they have either relaxed group control of decision-making (as the legislation’s intent) or attempted to respond to the new institutional environment by maintaining or expanding binding methods of party group discipline, controlling communication and the selection of candidates to continue their quest for ‘policy, office or votes.’ The findings are important to understand whether the local party groups of the three national parties operate in the ways suggested by the majority of the national party literature or whether local party groups are distinct in their response to exogenous changes in the institutional environment. Background In a consultation paper in 1991 entitled The Internal Management of Local Authorities in England, the Department of the Environment highlighted that the committee-led decisionmaking system in local government was not functioning in a healthy democratic way, as decisions not taken by officers were either taken by the whole councils or a number of committees and ‘this process was often time-consuming, cumbersome […] and in practice, the ability of ordinary members, especially those in minority parties, to influence decision-making [was] likely to be limited (1991, p.10).’ The paper identified key changes that the Government wanted to make to counter some of these problems such as; 1 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 ‘[Promoting] more effective, speedy and business-like decision-making, enhancing scrutiny of decisions, increasing the interest taken by the public in local government and providing scope for councillors to devote more time to their constituency role.’ (1991, p.10) Due to Government priorities at the time and the wholly negative reaction from some areas of local government to the suggestions contained within the consultation paper, the recommendations were shelved. It was not until the subsequent change in central government control that the proposals relating to a separation of powers for local government became more concrete. In the consultation paper Modernising Local Government: Local Democracy and Community Leadership (DETR, 1998a), there was a declaration that the incoming Government intended to change political arrangements in local authorities, but subject to individual councils choosing the best option for their area. The early ideas explained that the roles of councillors would change and be clearly delineated. ‘[Some] councils will have political and management structures involving cabinet style governance or directly elected mayors with executive responsibilities. Whatever structures councils will have chosen, the key executive, representational, and scrutiny roles of councillors will be readily distinguishable.’ (1998a, p.6) The consultation paper referenced the previous research to suggest that there were too many committee meetings taking place, a distinct lack of councillor engagement in decision-making and fundamental confusion over the councillor role. Councillors would often ‘feel unable to influence events and in the absence of a real detailed knowledge of the needs and aspirations of the people (1998a, p29)’ given that ‘in most councils it is the political groups, meeting behind closed doors, which make the big and significant decisions (1998a, p.29).’ The Government’s answer to this problem in local government was a somewhat radical solution, given that some authorities had already tried some ‘innovative ways of involving the public in their decisions, [such as] adapting their committee structures to move away from the traditional functional structure (1998a, p.29).’ For example, the argument was then pushed for a separation of powers as the Government considered that the system at the time did not foster the development of ‘community leadership’ roles for local councillors. The paper also reported that because so few people knew who led their council or who was responsible for decisionmaking within local authorities, there was a need for an individual, or an executive leadership that the constituents would have a ‘direct say (1998a, p.30)’ over. The Government articulated three motivations for the splitting of powers in local government; ‘greater clarity about who is responsible for decisions; greater clarity about who has taken and should be held to account for decisions; and sharper scrutiny of those decisions (1998a, p.30)’ Because decision-making was being taken away from the committee structure and placed in the hands of a select few members of the council, the Government believed that the changes would ‘shake-up and reinvent (1998a, p.30)’ the committee system because ‘councils would be less able to operate behind closed doors without debate and review (1998a. p.30).’ Understandably, given that the separation of powers in local government was meant to be an adhibition to diminish a councillor ‘[feeling] unable to influence events (1998a, p.29),’ it seems patently counterintuitive to suggest that by removing the decision-making capacity of the majority of councillors in the separation of powers it would improve a councillor’s belief in her 2 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 or his own efficacy. However, it becomes clear at this point that central government’s concern with the local authority committee system was more influenced by the domination of the party group over authority business rather than any stronger concerns about bureaucracy, role confusion and micro-management. To this end the consultation paper marketed the idea of scrutiny committees and backbench councillors as a model that ‘could be less time-consuming, but more high profile, more effective and therefore more rewarding (1998a, p.32)’ rather than as an difficult corollary of the legislative aim of removing party control in local government. In an interview with one of the ministers responsible for the legislation, the creation of an executive rather than a body responsible for holding to account was at the forefront of the legislation. ‘I really did feel, well, we all felt, that you needed a sort of strategic driving force, which was what the cabinet system was supposed to bring you. But then that meant you were going to have a lot of councillors who weren’t in the cabinet, and you had to give them a role that was meaningful…’ (Interview with Hilary Armstrong MP, former Minister for Local Government) Superficially the Government looked to be taking decision-making authority away from the majority of councillors but it could be argued that the Government was perhaps trying to enable councillors, from both majority and opposition groups, to correct the ‘the absence of a real detailed knowledge of the needs and aspirations of the people (1998a, p.29).’ By permitting councillors, through legislation, to look beyond the party group as a leviathan over decision-making by numbers, the separation of powers confronts the party group with a scientia ipsa potentia est dilemma. Through the introduction of evidence-based, constructive scrutiny the dynamics of intra-group relations would necessarily change; thus the legislation would confront the primacy of the party group. In the subsequent Modern Local Government: In Touch with the People (DETR, 1998b) White Paper, it was argued that the committee system in local government led ‘to inefficient and opaque decision making [and a situation where] significant decisions [are] taken behind closed doors by political groups or even a small group of key people within the majority group (1998b, p.18).’ The White Paper reported that councillors have had little opportunity to direct change and make a significant contribution. Another considerable problem highlighted in the White Paper was that people were often confused as to who was really making the decisions at a local level; ‘people identify most readily with an individual, yet there is rarely any identifiable figure leading the local community (1998b, p.19).’ Hence on the grounds of these two key arguments, the paper advocated a separation of roles in local government; executive and backbench. The White Paper gave local authorities three options for political management arrangements; a cabinet and leader model, a directly elected mayor and cabinet or a directly elected mayor with a council manager – the latter two requiring popular support in a referendum. The document further argued that each of these three versions of the separation of powers would have the following advantages; efficiency, transparency and accountability. Fundamentally, the new arrangements would ‘enable people to measure the executive's actions against the policies on which it was elected (1998b, p.19).’ The subsequent guidance which followed the actual legislation, entitled New council constitutions: guidance to English Authorities (DETR, 2000) fleshed out some of the detail of how central government expected local overview and scrutiny committees to operate. Central 3 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 to this guidance was the belief that overview and scrutiny was essential to ‘hold decision makers to account (2000, para 3.44)’ and to do so successfully required ‘a change in the way members have traditionally questioned decisions (2000, para 3.44).’ The emphasis throughout the guidance on overview and scrutiny was that it should be ‘constructive and not merely there either always to oppose the executive or to rubber-stamp the executive’s decisions (2000, para 3.30)’ and work on ‘cross-cutting [issues] rather than narrow service-based [issues] about the conduct of local authority’s business (2000, para 3.21).’ To do this effectively the guidance recommended that political parties should not whip their members in overview and scrutiny committees and suggested that the committees should ‘seek views from as many communities and interested parties as necessary to get a balanced picture of the effects of policy and executive decisions (2000, para 3.40).’ The zeitgeist was that overview and scrutiny was intended to be, inter alia, an evidence-based, rigorous new function that looked beyond the typical adversarial politics that often dominated local government under the previous political arrangements. In support of this, the guidance recommended that local authorities should strongly consider the proposition that committees should be ‘chaired by members outside the majority group or by [non-voting] church or parent governor representatives (2000, para 3.30)’ in order that ‘thorough and informed reviews take place (2000, para 3.29).’ The introduction of the new scrutiny / executive split naturally implied that the behaviour of local political parties also needed to fundamentally change and councils physically required a ‘change in political culture and approach for many members (Leach et al. 2002, p.74).’ Party discipline and expectations of party loyalty had been a significant feature in the day-to-day operation of local political parties under the previous committee system, whereas the shift initiated in the Act confronted certain traditions and practices head-on. For example, in the guidance accompanying the Act, specific instructions from the Secretary of State strongly advised against such activities as using a party whip in relation to overview and scrutiny and, expressio unius est exclusio alterius, pre-overview and scrutiny party group meetings. However in the implementation of overview and scrutiny, the function may not have evolved into the model representing the chimerical vision of the guidance document. For example, the Commons Transport, Local Government and the Regions Select Committee investigated the early impact of the introduction of the new political arrangements and found that despite being introduced to counteract camera obscura party group decision-making, the new structures had seen decisions ‘continue to be made in private and [were] merely [being] “rubber stamped” or justified in public (House of Commons, 2002a, p.13).’ Further to this, evidence provided to the select committee commented that majority party non-executive members were not being as thorough in their role of holding an executive to account as other party groups; ‘Administration group members do not truly see their role as holding the executive to account at select committee meetings, rather as one of defending cabinet members, decisions and stopping the opposition groups from embarrassing their administration.’ (Essex County Council Labour Group, LGA 35, House of Commons, 2002b) Other evidence to the committee from the LGA Liberal Democrat group, representing almost five thousand councillors, reported that scrutiny was often being used ‘as an opportunity to embarrass the administration or to raise issues that they can use for political campaigning purposes (2002b)’ and in some places there was a ‘fear of retribution [which was] preventing 4 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 some non-executive members of […] ruling groups from carrying out an effective scrutiny role (2002b).’ Research published by the Office of the Deputy Prime Minister soon after the publication of the select committee report reinforced some of the findings relating to the party group’s difficult relationship with the new executive / scrutiny split. For example, the research found that although there was ‘little obvious evidence of a whip being imposed on O&S members for any of the parties (ODPM, 2002, p.74),’ they received comments such as; ‘We don’t have a whip in scrutiny, but the chief whip attends scrutiny meetings and reports back to the group on what people have done.’ (2002, p.74) The ODPM report discovered that during the introduction and development of O&S, two distinct roles had been identified; that of consensual multi-party policy development and scrutinising the executive. Whilst the first of these roles had been judged to be relatively successful, as committees meetings were ‘freer and more flexible than [the previous] committees (2002, p.75)’ and didn’t require voting procedures (which often divided old committees along party lines), the second of these roles, of holding an executive to account, proved to be more difficult in practice. The idea of councillors of the same group criticising their own colleagues in public had created an accountability impasse, where scrutiny of decision-makers was still occurring in a surreptitious way. Whilst majority party non-executives may have been scrutinizing the actions of their colleagues ‘with real vigour and robustness (2002, p.76)’ this was still happening behind closed doors. Due to this lack of criticism in public, the Act, at this stage, was not fulfilling its intention to make the decision-making process transparent - as there was still a great ‘temptation to revert to a strict adherence to group policy and to group cohesion (2002, p.76)’ Further research by the ELG team conducting research for OPDM in 2003/2004 discovered that some local political party groups were using a party whip in relation to overview and scrutiny (ODPM, 2004, p.50). Thirty-five percent of respondents also reported that pre-scrutiny party group meetings were held before committees (2004, p.50). Party group influence was also affecting the ‘call-in’ procedure granted to overview and scrutiny committees, which was often used relating to ‘the adversarial competition between political parties (2004, p.9)’ and was used to ‘create publicity [for the party group] and make political points (2004, p.9).’ Further to this, in authorities with a large majority party, it was found that party loyalties ‘made the process of challenge to the executive problematic and difficult to sustain (2004, p.10).’ Ashworth (2003) discovered that ‘group affiliation was evident throughout meetings, often from the outset, with many members still choosing to sit and operate in [party] groups (2003, p.20).’ Leach (2005) also saw that traditional party group demarcation had made overview and scrutiny difficult to establish successfully in some authorities. Ewbank (2007, p.14) discovered that political party groups often briefed members about the strategies involved in overview and scrutiny; members were also grouped together by party in committee and the group whip was still evident to a similar extent as three years previous. However, the research in this area has not necessarily focused on some of the unusual successes of the overview and scrutiny in such areas as pre-decision scrutiny, policy development, policy reviews, external scrutiny and scrutiny of health related issues. There are 5 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 many examples of cross-party working and consensual politics that show political party groups realising the benefits of scrutiny (CfPS 2004, 2006, 2007). However, ad absurdum, party group loyalty and distrust between party groups has stunted the development of the function. As Copus (1999) had prophesised, a number of councillors saw the overview and scrutiny function directly at odds with their conception of party membership and feared that by crossing the line and disobeying orders that it was ‘at best disloyal to the party, and at worst a betrayal of the group and the individual's beliefs sustaining his or her party membership (1999, p.320).’ The acceptance of the new political arrangements in local government since the Act has been one of the most difficult tasks that authorities have had to face in the last decade. Essentially, the problem could be summed up in a profound statement by Olson (1980); ‘Parties and Committees are […] contradictory and even mutually exclusive means of internal organization [sic] [in a legislature]. The importance of each is inversely proportional to the other. The more important the committees, the less important the parties, and vice versa.’ (1980, p.269) Interpretations of the Change in Institutional Environment in Local Government Colin Copus and Steve Leach (2004) were the first to create hypotheses and formal models of how they considered political party groups had responded to the introduction of the separation of powers in local government and gave details of how and why councils would tend towards one rather than another model. Using evidence accumulated in various research projects of up to thirty four case study authorities, they posited four models of increasing intensity of party control over the overview and scrutiny function, the party as ‘a partner, [minimal party control over overview and scrutiny] arbitrator, filter and Leviathan [maximal party control over overview and scrutiny] (Copus & Leach, 2004, p.339). The differences between each of these four proposed models are explained with four factors; ‘...the degree of group coherence in public; the readiness to impose a whip and use disciplinary mechanisms, the number of group meetings and the balance between open deliberation and closed decision-making.’ (2004, p 339) Copus and Leach (2004) considered that there were certain predictors that would affect how parties would react and which hypothesised model that the parties would collect around as a result of the move from one institutional environment to another. For example, the different political parties have developed in distinctive ways and thus their constituent members would behave in dissimilar ways, such as the Labour ‘corporatist perspective [where] the group is entitled to loyalty, as it is a mechanism which implements the policies supported by the electorate (2004, p. 339),’ the Conservative’s ‘individual freedom [chastened by] individual responsibility implying obligation to accept decisions (2004, p.339)’ and the Liberal Democrat’s recognition of ‘collective, rather than individual political endeavour, which brings political success (2004, p.339).’ More importantly, as each political party necessarily commands some form of cohesion for at least some points in time, Copus and Leach looked to other factors which would affect the model to which parties would tend. For example, the political balance and composition of the 6 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 authority is seen as a major factor in determining how scrutiny will operate at a council, where party discipline is likely to be tight and restrictive in competitive authorities where the control of authority can feasibly change hands (2004, p.348) compared to the reverse situation where an opposition cannot challenge because it is ‘weak or divided (2004, p.348). Another variable raised is the level of intra-party factionalisation, where party groups would be more likely to control and moderate its members where cohesion was not naturally forthcoming and the reverse situation where cohesion came naturally. Whilst the Copus and Leach (2004) explanation of the reality of the transition from the committee system to a separation of the executive and non-executive is useful in answering some of the questions about the differing levels of political party group control over the nonexecutive function, the route by which it arrives there is inadequate; basing differences in party group approaches to the split simply on arbitrary structural differences. Whilst it may be true that ‘effective scrutiny is most likely in a situation where the majority or dominant group operates a ‘group as partner’ philosophy (2004. p.351)’ and ‘effective scrutiny is least likely in a situation where the majority (or dominant) group operates a ‘group as Leviathan’ philosophy (2004. p.351),’ the fundamental hypothesis that a political party group can even move amongst these models may be somewhat flawed. Can political party affiliations prove a more accurate descriptor of the party behaviour? Furthermore, the supposition that any rational political party group can by anything other than the described Leviathan model at all could be easily contested. One of the key failures in the mostly advisory natured, practitioner-directed literature on the changes in political management arrangements arising in the Local Government Act 2000 is that they fail to acknowledge why the party behaves as it does in the political local government arena and how and why it gets to a behavioural place in the first instance, instead relying on explaining conditions where party behaviour may be different. Thus the intrinsic worth of this literature as a key to unlock general local party behaviour is low. The Local Government Act 2000, in this case, legislated for a constitutional change in local government which affected each local authority and its constituent parts such as political party groups and yet in the ten years since Royal Assent, there has been no comparison with other movements in constitutional form which have occurred directly preceding or throughout this time (France, Israel, local government in North Rhine-Westphalia, Bergen and Oslo – see Samuels & Shugart upcoming) and how parties may have reacted to each of these different types of a move in institutional operating environments. Viewing groups as distinct unitary collective actors, local political party groups may accede and possess goals in the same way that national parties may do so and in an equivalent way they aim to fulfil these goals. For example local parties may seek to achieve policy, office or votes (Strøm, 1990) in the same manner as national parties. Whilst the local political parties differ in their functions and the way that they may formulate and achieve goals set for themselves, with or without national party interference, the way that they may go about this is no different to those of their central equivalent. If national party X seeks to achieve election to control the government, what makes the component local party X different when it seeks to control the local authority? Using a framework such as a theory of party competition (Strøm, 1990, 1999) is all the more relevant as the majority of local authorities in England elects to the council by thirds in three out of every four years. The current literature about the introduction of an executive with an overview and scrutiny function disregards the similarities between constitutional shifts at a state / sub-national level 7 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 and subsequent changes in the behaviour of parties and ignores the state level arguments that parties may be discrete actors who possess goals as part of the reason for their existence. If legislation is enacted which confronts the operating space of the local political parties and forces submission to an institutional change of multiple arenas of representation; it demands an immediate reaction from a political party. Moving from a government by assembly represented by the committee system to the relatively open and more difficult environment of a rudimentary organisational separation is one of these direct threats which directly impinges on a party’s ability to satisfice their goals. The explanation of the reactions of the local political parties to this threat of change; whereby goals are made more satisficable, is the chief subject of this research. In order to answer the foremost question about why Copus and Leach (2004) found that, in practice, party groups all did seem to tend around the Leviathan model, it is crucial to take another route beyond straightforward structural heuristics (i.e. party groups tend to dominate in competitive authorities) and one that looks squarely to other examples, including that of the theories of party behaviour at a state and sub-national level, in order to begin to research and uncover the potential initial reactions and continued adapted behaviours of local party groups in English local government. The research This particular research argues that the basis of the legislation was never likely to be delivered on the grounds that parties assess action according to the goals which they desire to fulfil as part of their very existence. In terms of state-level party behaviour (Strøm, 1990), it would seem impossible for the legislation to deliver the results that were hoped for (transparent, clear decision-making with rigorous cross-party scrutiny of an executive) when central government mandated constitutional change on local government. Whilst acknowledging the various limits of the applicability of the theory in the local government context (discussed at length in the thesis), the rational choice institutionalist approach stands up as a reasonable way to deduce what has happened in this context. Looking at the effect of the Local Government Act 2000 on political party groups in English local authorities, there had been an external shock in the form of an environmental modification which ‘altered the environment in which the parties operated (Harmel & Janda, 1994, p.277).’ The contention of this research is that this environmental modification brought about by the introduction of the Local Government Act 2000 would significantly affect how political party groups were to operate and thus considerably alter their ability to achieve their primary goals (policy, office, votes). Parties have been faced with potential options for change and as a direct result of wanting to maintain performance in achieving their primary goals despite the external shock of a change in legislation, this research aims to prove that the parties thought that modifications, specifically tightening modifications, would be an appropriate way to strike against this shock. 1. The Shock: Environmental Modification – ‘the blended separation of powers’ In the introduction of a separation of powers in the Local Government Act 2000 there was a clear affront to the workings of the well-documented local party group’s camera obscura system of decision-making in local government. Ninety-five percent of English local authorities had determined that the cabinet and leader model, from the selection offered in the legislation, was the least affrontive change to their operating environment. It is likely that political party 8 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 groups understood that the parliamentary cabinet and leader system was the nearest to what had already developed in local government and were fully aware that this model ensured that the party would retain control of authorities. More significantly, this compulsory change from assembly government to the ‘least worst’ option of the rudimentary parliamentary separation of powers created a problem for political parties because of the new institutional environment it created. As one example, the introduction of overview and scrutiny committees created a situation where non-executive majority party members were expected to objectively ‘scrutinise’ their executive party colleagues in a public arena. Given that party groups have strict standing orders and regulations about what can and cannot not be said outside of the privacy of the party group, the ‘theatre of representation’ from which the public and media are, of course, excluded (Copus & Leach, 2004, p.338),’ the introduction of a new transparent, more accountable and public system of government was very likely to considerably alter party abilities to achieve their primary goals (policy, office, votes). In two examples, a fractured opposition party with a whole host of public opinions on a particular policy matter would be unlikely to attain office if the messages delivered to an electorate were dissimilar and public criticism of an administration decision from a nonexecutive administration party member would, for the opposition parties, be a veritable quarterstaff with which to beat the administration. In general terms, under the separation of powers it would be likely that an office-holding party who sought to maintain its administration would find it more difficult to maintain office if intra-party criticism moved from inside to outside the party group. (i.e. a non-executive majority party member could potentially criticise his / her colleague within a public overview and scrutiny committee rather than in the private sphere of the group meeting.) For parties in opposition the environmental modification would also alter their ability to achieve their primary goals such as feeling the need to balance oppositional politics with policy development priorities – would the party be electorally rewarded or punished for public consensus based decision making? 2. The Reaction: Party modification – ‘cracking the whip’ Harmel and Janda (1994, p277) broadened the definition of a party change for the purposes of their research to include all changes relating to ‘party rules, structures, policies, strategies or tactics (1994, p.277).’ However they specifically defined a party modification as something which affects the party’s rules and regulations. In subsequent empirical work the same authors developed a series of coding procedures in order to track party characteristics in nineteen parties from Europe to the U.S. These were a series of; ‘…coding procedures to tap the unofficial story of parties’ goals, organisation, internal distribution of power, strategy / tactics and ideological / issue orientation.’ (1995, p.2) Of Harmel, Janda, Heo & Tan’s (1995, p.25) forty-three objects of change, the following list of organisational variables are the some of the most important dimensions relating to party rules, regulations and thus strategic and tactical changes in parties that are important in this research: 9 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Relating to organisational complexity: 1. Frequency of local meetings 2. Pervasiveness of organisation Variables relating to centralisation of power and distribution of power1 3. Formulating policy 4. Controlling communications 5. Administering discipline 6. Candidate selection 7. Allocation of public policy positions Of these variables, it is necessary to replicate their meanings in a local government arena and the functions, roles and context within which they apply. As a result, the variables are modified into the following categories with explanations of the expectations of what should happen according to the theory of party competition. Of the seven central variables, five have been assessed as part of the quantitative work of the thesis. Relating to organisational complexity: 1. Local party group meetings The first variable is primarily quantitative (duration and frequency) and reflects the distinct nature of local party group meetings as a means to discuss issues as a group, in private, prior to overview and scrutiny committee meetings. The number of which and ‘the type of’ may indicate the nature of party group control within each group. As a result of the introduction of the cabinet / scrutiny split in local authorities, it would be expected, from the theoretical exposition, that pre-scrutiny meetings might take place (in both admin and opposition parties) or regular party group meetings would devote a portion of time to the discussion of the issues raised in overview and scrutiny committee meetings – both of which were contrary to the spirit of the Secretary of State’s guidance. In comparison to the committee model of decision-making, the number of these meetings (or the portion of time devoted to scrutiny in regular meetings) would be expected to be higher as a result of the need for the parties to regulate operations for their respective goals. 2. Pervasiveness of organisation (not analysed quantitatively – no discussion here) Variables relating to centralisation of power and distribution of power 3. Centralisation of Policy Formulation and Preclusory Discipline This variable reflects the actual content of any party group meetings, communications from party group leader(s) and indications of the organisation of councillors relating to a) certain issues that have already been decided, b) certain lines to be agreed amongst Harmel and Janda (1995, p. 25) use the term ‘nationalisation’ of power which I have replaced with the more appropriate ‘centralisation’ of power. 1 10 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 the wider party group and c) any implicit or explicit whipping on issues to be investigated in public (referred to as preclusory discipline as distinct from post-hoc castigative discipline). The first half of the variable, the centralisation of policy formulation, is less amenable to quantitative research is thus dealt with in other ways. The theoretical discussion would imply that preclusory discipline would be evident in the operation of overview and scrutiny and either implicit (advising councillors) or explicit whipping (mandating councillors) would be used in certain cases in order for the party to maintain control of scrutiny operations. For example, all councillors might have experienced some form of advisory or explicit control relating to their actions in overview and scrutiny. 4. Coarctation of individual councillor communication This variable seeks to assess how active / passive the control is over individual councillor behaviour in the public and private sphere; such as whether councillors are briefed about the O&S function before joining a committee and how willing or unwilling councillors are to step away from any preclusory discipline and break rank from the party group in relation to issues that are under consideration. Thus by looking into this variable it may be possible to explain how intensely preclusory discipline has been enforced or how severely castigative discipline is feared. As a result of the constitutional split, parties would be more likely to brief party members who are selected for scrutiny committees about their role and what is expected from a group perspective. It would also be expected that councillors would be more reluctant to criticise their party group (both admin and opposition parties) in a public arena such as scrutiny for fear of breaching preclusory discipline resulting in possible castigative disciple of varying strengths. 5. Administration of Castigative Discipline The administration of castigative discipline is one of the clearest ways to demonstrate party group control in local government; its severity in all cases implies that councillors are proactively chastised for their behaviour for acting against the party group. However its impact is more subtle, as whilst unlikely to be used widely in its most severe format (removal of the party group whip), the fear of castigative discipline such as a removal of SRA positions, refusals of re-nominations to (healthily remunerated) external bodies, refusals of re-nominations to committees of enjoyment or an inability for promotion to other positions, are likely to weigh quite heavily on councillor actions in overview and scrutiny. Whilst the variable may be one of the clearest indicators of a vice-like grip on councillor action from the party group leadership, its measurement is less clear. Whilst it would be expected that castigative discipline could increase because of the move from deliberation in a private arena to the public arena, it may also be the case that councillors regulate their behaviour because of this knowledge. This variable can be thus measured by assessing how councillors may fear for their position within the party group (and the associated of benefits of adherence to preclusory discipline) as a result 11 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 of their actions. Thus it would be expected that councillors would fear castigative vengeance enough to modify their behaviour in open committees. 6. Candidate selection (not analysed quantitatively – no discussion here) 7. Allocation of Positions This variable can be utilised to explore how parties are involved in the nominations for scrutiny positions and whether the group leadership has complete control or no control over the independence of the process. By assessing who or how the nominations of overview and scrutiny committees are chaired, it is apparent how much steer individual councillors or the party groups have over the operations in scrutiny. The variable is much easier to examine in a quantitative format with a large number of councillors by asking how chairs are appointed to various overview and scrutiny committees. It would be expected that the party group, as part of a way to achieve their respective ambitions, would be more likely to play a leading role in the nomination and securing of committee chairs. For administration parties, this could mean more control over the number of committees and who controls these and for opposition parties, how they allocate their positions if they are in receipt of these. It is also necessary to examine beyond the explicit, possibly enforced cohesion that may occur in party groups in local government and explore the behaviour of individual councillors and whether the party group takes priority over evidence-based consensual decision-making under the arrangements of the Local Government Act 2000. Therefore the following final variable explores the behaviour of individual councillors under the institutional change. 8. Individual-level councillor behaviour The variable looks away from the party group’s actions (Y) on councillors (X) and explores whether the impact of the legislation on party groups was more subtle. Further to looking at explicit standing orders, restrictions and expressions from the party group side, it is crucial to look at how councillors make their own decisions under the new framework. This variable can be assessed in fieldwork under a set of hypothetical situations and how councillors would behave as a result. For example, using Strøm’s policy, office, votes trade-off and the sheer incompatibility of cross-party evidentiary-based scrutiny within the theory of party competition, it is possible to gauge how certain behaviours fit into this framework. Using a hypothetical situation, one can assess whether the councillor would act in a way consistent with Strøm’s theory of competitive political parties or whether the councillor would step outside of expected behaviour and deliver the aims of the legislation. Based on the discussions above, it would be more likely that councillors, even in the absence of any explicit party group restrictions et al., that councillors would behave in a manner suited to the fulfilment of party goals (this is explained in more depth in the thesis chapters; but the idea that councillors may act in this way 12 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 because it sustains their own conceptions of being a member of a ‘party’ and thinking beyond themselves as individuals or they fear reprisals from the party group in any form, even if these are not made explicit to them beforehand (see variables 1 through 7).) Quantitative Research As part of the mixed methods research project, of which the other half is a number of case studies in authorities with different political balances and functions (ongoing at time of writing), five of the seven central variables could be and have been assessed in a questionnaire sent to a representative cross-section of councillors across local government in England. The sampling frame for the survey included local councillors in English local authorities, operating under the dominant cabinet and leader model or alternative arrangements, who were members of at least one overview and scrutiny committee at the time the sample was constituted. There were requirements that the population from which the sample would be selected should only include elected members who had a party label of one of the three main parties (Conservative, Labour and Liberal Democrat) for the sake of substantive quantitative comparative analysis and a further requirement that they should not be party group leaders who principally make the decisions that the questionnaire served to discover the reaction towards. From three hundred and thirty-two eligible English local authorities in the sampling frame, a representative twenty-percent sample was taken using stratified random sampling in order to purposely weight the sample according to populations of English government regions, leaving sixty-seven authorities in the survey sample as outlined in the table below. This weighting according to population of government region combined with a high 20% sample of English local authorities enabled the sample to be more representative of all experiences across English councils, without which the sample would have disproportionately been balanced in favour of lower-tier authorities. The method of the survey administration was paper based (with an option of online completion) due to the relative age of respondents in comparison to the rest of the general population. Within the random stratification of sixty-seven authorities, 1684 councillors were eligible to take part in the survey; of which 871 responded to the survey, giving an extremely high 51.72% response rate in the context of relatively low returns from local government councillor postal surveys. The questionnaires were purposively given a series of unique reference numbers (URNs) to document the progress of respondents returning completed scripts, making respondent identification and chase more focused. Initial quantitative study results2 Local party group meetings Councillors in the survey were asked not only if party group meetings were held that exclusively dealt with scrutiny matters but also whether regular party group meetings devoted any portion of time to scrutiny at all. Councillors who served before the introduction of the 2 (analysis is limited to ‘surface’ data analysis according to affiliation (rather than structure) due to time & space constraints.) 13 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Local Government Act 2000 were also asked a comparative question as to whether more or fewer party group meetings were held. Q3.1 Party group meetings exclusively dealing with scrutiny Question 3.1 finds that party group meetings are broadly held in around 35% of cases despite the guidance relating to group actions relating to overview and scrutiny committees. The results are consistent with ELGNCE research, but show the texture of difference between regularity (often) and rarity (less frequently reported). Q3.2 Length of time devoted to scrutiny issues in regular group meetings Q3.2 sought to discover what additional time was devoted scrutiny issues in regular group meetings (not just those held prior to or specifically about scrutiny.) The results are interesting scrutiny.) due to the higher level of time devoted to scrutiny than previously discovered in other 14 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 research (ELGNCE, 2002, 2006, Ewbank 2007). For example, Labour and Conservative respondents reported circa. 40% of regular group meetings allocating at least some time to dealing with scrutiny issues. Even more surprisingly was the affiliation, with over half of Liberal Democrat respondents reporting regular group meetings dealt with issues in overview and scrutiny, with almost 8.5% reporting that this took up to an hour or over each meeting. reporting Q4.1 Comparison to the committee system – more or fewer party group meetings Q4.1 asked respondents, who had served before the LG2000, whether more or fewer party group meetings took place. Early indications show that party group meetings are held with a similar frequency since the change; with no major differences across political affiliation. More respondents reported more meetings than respondents who reported fewer meetings but this is unlikely to be statistically significant. Given that the legislation sought to reduce the impact of group dominance in decision-making, the large cohort of those who report the same frequency of meetings occur as under the old committee system reflect that the party remains defiant, at least, in the face of institutional change. Centralisation of Policy Formulation and Preclusory Discipline Councillors in the survey were also asked about levels of preclusory discipline in relation to overview and scrutiny. Whilst explicit whipping was forbidden by the legislation and guidance, the questions were broken up according to strength of organisational discipline. 15 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Q3.3 Respondent ‘advised’ to act a certain way in overview and scrutiny Q3.3 and Q.3.4 aimed to break ‘explicit’ whipping into ‘advisory’ and ‘mandatory’ forms of whipping. Q3.3 looks at how many respondents were advised to act a certain way or received informal pressure to do so in scrutiny committees. The results are surprising, with roughly 25% of respondents reporting advisory whipping at times. Liberal Democrat councillors appear to report the highest level of advisory whipping but on less frequent occasions – whereas Labour councillors, closely followed by Conservative councillors report the most systematic advisory whipping (i.e. LAB 7.5% advisory whip ‘all of the time’ or ‘some of the time’ compared to CON 6.6% advisory whip ‘all of the time’ or ‘some of the time.’) Q3.4 Respondent ‘mandated’ to act in a certain way in overview and scrutiny 16 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Q3.4 documents that explicit whipping is in effect according to respondents in the survey, contrary to national guidelines, regulating the operation of the function. Whilst it is in effect, the levels of explicit whipping are a good 10% lower than ‘advisory’ whipping from the party group and are approximately equal across all parties at 12%. The Labour and Conservative differences are small in relation to explicit whipping ‘all or some of the time’ at 3.7% and 3.5% ‘all respectively. Coarctation of individual councillor communication Respondents were also asked about how communications outside the party group may or may party not be moderated and how they express themselves in the public and private arenas. The first question relating to this variable looks at whether councillors are briefed by the party group about how to conduct themselves in overview and scrutiny. The second and third question looks at the contrast between public and private communication. Q3.5 (below) asks respondents if they were briefed on their conduct by their own party group. The results show that the Liberal Democrat councillors were the ‘least briefed’ about their conduct with 80% reporting that this never occurred. Labour and Conservative members reported higher levels of group briefings of 27% and 26% respectively. Looking at the figures more closely, the main occasions for councillors to be briefed on conduct by the party group was when there was a controversial issue, when the system was first introduced or when the councillors first stood for election. Q3.5 Respondent’s briefed about conduct on O&S by party group 17 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Q3.7 and Q3.8 asked respondents if they had ever publicly (overview and scrutiny, press et al) or privately (confines of party group) criticised decisions made by their party. Q3.8 (privately) provides little information on its own but the contrasting of the two could be important. (see below) Q3.7 Respondents criticising their own party decisions in PUBLIC ARENA Q3.8 Respondents criticising their own party decisions in PRIVATE ARENA From the results of Q3.7 and Q3.8, it is more clear that communication of dissent is either proactively discouraged by the party (Q.3.2, Q3.3, Q3.4, Q3.5 & Q4.10 see variables above) or a desire for the individual councillor to either avoid castigative discipline or protect the party discipline brand (Q3.10 see individual behaviour variable below) or themselves within the party (Q3.9 see ind. behaviour variable below). The interesting contrast between 3.7 & 3.8 gives an insight into 18 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 how overview and scrutiny, as operating within the public arena, is somewhere that councillors may restrain their behaviour in order to keep it within Copus’ ‘theatre’. Administration of Castigative Discipline Respondents to the survey were also asked if they had ever received any punishment as a result of their actions against the group in relation to overview and scrutiny or speaking out against the group in a public arena (including overview and scrutiny.) In relation to Q3.6 which asked respondents if they had ever been disciplined in relation to speaking out against the disciplined party group decision in scrutiny, a number of councillors highlighted that they had. Interestingly, despite the local Labour party group being more famed for castigative discipline (discussion elsewhere), the Conservative respondents received more castigative discipline in Conservative comparison. Party Group Conservative Labour Liberal Democrat % of respondents punished. 5.41% 3.65% 1.38% Q3.6 Had respondents ever been disciplined as a result of their action in O&S The follow-up question Q3.61 asked what type of discipline was used to punish the offender, giving a list of closed answers rising in severity. The most common article of castigative discipline was an informal warning from the party followed by not being reappointed to an existing role (such as a committee or policy post et al.) and being removed from an appointment (such as a committee or policy post et al.). One councillor in the study reported al.). that as a result of their action they had been deselected as a candidate at a subsequent election and another councillor reported that as a result of criticising the group in O&S they were actually suspended from the party. In the branch questions where councillors had served before the Local Government Act 2000, respondents were asked if the exercise of discipline (both preclusory and castigative) had increased or decreased in relation to the committee system. Q4.3 Discipline used compared to old LG committee system 19 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Q4.3 shows that the majority of respondents believed that the level of discipline had stayed the same despite the legislation. Interestingly, more Labour respondents believed that discipline had occurred less as a result of the legislation (this may or may not be statistically (this significant). Allocation of positions Q2.5 asked respondents who were chairs / vice-chairs of overview and scrutiny how they achieved their roles. The options for answering covered all possible routes for awarding positions and have been documented below. Q.25 How chairs / vice-chairs achieve their position Overwhelmingly, Q2.5 demonstrates that the majority of chairs and vice chairs in overview and scrutiny are nominated by the political party groups that have been allocated the positions to distribute. The question allowed for multiple entry to cater for chairs of more than one allowed committee, but the typical route for nomination according to the respondents involves nomination by the party followed by verification by full council. Individual-level councillor behaviour The survey also sought to assess how respondents themselves act (or may act) regardless of any explicit action from the party group. Q3.9 (below) and Q3.10 (further below) gave respondents a Likert scale of agreement from 1 (strongly disagree) to 7 (strongly agree) with which to judge how much a statement referring to their perspective is agreed with. Q3.9 deals with a fear of castigative reciprocation from the party group and Q3.10 deals with how councillors envisage electoral concerns in relation to their scrutiny role. 20 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 criticising Q3.9 “I would be concerned about the consequences of criticising my own party in public in relation to my position within my local party group.” The results for Q3.9 are not clear at the outset without breaking down the results beyond this first stage (affiliation) but on the surface indicate that Labour and Liberal Democrat party councillors agree with the statement more strongly than the Conservative resps (LD up to a point.) Results for Q3.10 (below) are much clearer and show that concerns about electoral performance do affect how councillors communicate in public arena (such as overview and scrutiny). Q3.10 ““I would be concerned about the consequences of criticising my own party in public criticising in relation to my party’s chances at the next election.” 21 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 The survey also wanted to test a number of hypothetical situations with local councillors to provide evidence of what motivates behaviour in relation to councillor activities such as overview and scrutiny. Using Strøm’s trade-off as shorthand but using genuinely possible statements (purposively extreme & committed positions) with a Likert scale, it was hoped that it would be possible to ascertain what action councillors would take. As a comparator, at the very least, the results to the hypothetical situation should give an important contrast to how councillors believe they have acted on a regular basis. *Note Likert scale of agreement from 1 (strongly disagree) to 7 (strongly agree) Q5.11 (Vote-seeking) “If there is a problem in my ward that has been caused by a decision made by my own party, I would contact my group leader(s) to express my serious concern and try and work out a solution. There’s no point giving the opposition ammunition by opening up such a simple thing to wider debate!” such Q5.12 (Legislation’s intent & not part of POV trade-off) “If one of my ward constituents contacted me about a problem that they had encountered and it seemed like a sensible topic for an overview and scrutiny investigation, I would try and raise investigation, the item in overview and scrutiny first rather than taking it to my party.” 22 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Q5.13 (Ind Policy) “My aim as a councillor is to make a difference to the lives of people by feeding into the policy process with my ward residents in mind. If I am unpopular in the party group it makes ‘making a difference’ much harder.” Q5.14 (Individ Office) “It is reasonable to be expected to fall in line when it comes to agreeing with what the party asks me to do. If I am loyal to our cause I am more likely to get into a position of influence. With this wider area of influence, I can make life better for my constituents by having some control over the authority’s work programme.” Whilst the statements of hypothetical situations are complex because they commit respondents to extremes, they raise wider points of whether councillors orient themselves to their party or to their constituents (as encouraged with the Act.) For example, the very strong relationship in Q5.11 demonstrates an electoral concern taking priority over the function electoral created by legislation. In Q5.12, the actual premise of the legislation is tested and councillors are more likely to disagree with that course of action. The relatively strong agreement to Q5.13 and Q5.14 show concerns for impacting policy (in relation to how the party group views the councillor) and concerns for advancing careers (in relation to how obedience and submission can lead to greater things.) These questions, whilst unusual, demonstrate a deficiency in the 23 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 independence of councillors under the new political arrangements and thus near impossibility of the legislation’s intent to provide evidence-based consensus-orientated decision-making. Summative remarks The quantitative research is only half of the investigation being undertaken into the topic and is supplemented by a series of case studies at local authorities with different political balances, functions and ‘political culture.’ Whilst no firm conclusions can be drawn from the quantitative research at this point, especially at such an early stage of analysis, the survey has drawn out some important points. For example, one of the intentions of the legislation, to remove the closed-doors of the party group in favour of the public arena, has not been successful based on the range of explicit actions taken to prevent this from happening. Councillors are routinely organised with preclusory discipline in both scrutiny-specific party group meetings (circa. 35%) and regular group meetings (cir. 40%) and with the application of a whip (both advisory (cir. 25%) and explicit (cir. 12%)). Over 1 in 20 councillors of the largest party in local government have been subject to castigative discipline due to speaking against their own group in overview and scrutiny and almost a quarter of Labour councillors strongly agree that they would be concerned about the consequences of criticising their party in public in relation to their position within the local party group. Just under one third of Labour councillors and around a quarter of Conservative councillors strongly agree that they would be concerned about the consequences of criticising their own party in public in relation to the party’s chances at the next election. From a comparative point of view, the majority of councillors who have served their authorities prior to the Local Government Act 2000 report that the number of party group meetings has remained the same under the new system and the majority believe that the level of discipline has been unchanged. Given that the legislation aimed to tackle this type of political culture, on this basis and from this research, it seems unlikely to have been the case. Whilst at this point one cannot say that the type of separation of powers mandated on local government has resulted in a Madisonian ‘tyranny of the party’ (see Samuels & Shugart, upcoming) in local government in England, the impact of the group as Leviathan-esque decision-maker at sub-national level has not hugely diminished. However, given the current trajectory of the research and the qualitative work undertaken thus far, the proposition of a tyranny of the party (in direct contrast to the aim of legislation) created by the blended separation of powers (indirectly-elected cabinet / scrutiny) instead of a (directly elected mayor) clearly delineated separation of powers is not looking unlikely. References Ashworth, R (2003a) Toothless Tigers? Councillor Perceptions of New Scrutiny Arrangements in Welsh Local Government Local Government Studies Vol 29 No 2 pg. 1-19 Ashworth, R & Hunt, S (2003b) Learning Lessons from Westminster? Paper presented at the Political Studies Association Conference 15-17th April 2003 Centre for Public Scrutiny (2004, 2006, 2007) Successful Scrutiny I, II & III [London: CfPS] Copus, C (1999) The Local Councillor and Party Group Loyalty Policy & Politics Vol 27 No 3 24 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010 Copus C, Leach S (2004) Scrutiny and the Political Party Group in UK Local Government: New Models of Behaviour Public Administration 82.2, pp 331-354 Department of the Environment (1991) The Internal Management of Local Authorities in England: A Consultation Paper [London: DoE] Department of the Environment, Transport and the Regions (1998a) Modernising Local Government: Local Democracy and Community Leadership [London: The Stationery Office] Department of the Environment, Transport and the Regions (1998b) Modern Local Government: In Touch with the People Cm 4014 [London: The Stationery Office] Department of the Environment, Transport and the Regions (2000) New council constitutions: guidance to English Authorities [London: The Stationery Office] ELGNCE (2002) Report of ELG Survey Findings for OPDM Advisory Group [ELGNCE: Manchester University] ELGNCE (2004) Operating the New Council Constitutions in English Local Authorities: A Process Evaluation [London: The Stationary Office] ELGNCE (2006) Councillors, Officers and Stakeholders in the New Council Constitutions: Findings from the 2005 Sample Survey [London: The Stationary Office] ELGNCE (2007) The New Council Constitutions The Outcomes and Impact of the Local Government Act 2000 [London: The Stationary Office] Ewbank (2007) The Life of the Party: How do party political groups impact on scrutiny? [London: CfPS] Harmel, R & Janda, K (1994) An Integrated Theory of Party Goals and Party Change Journal of Theoretical Politics Vol. 6, No. 3, 259-287 Harmel, R, Janda, K, Tan, A, Heo, U (1995) Performance, leadership, factions and party change: An empirical analysis West European Politics Volume 18, Issue 1 January 1995 , pages 1 – 33 House of Commons, Transport, Local Government and the Regions Committee (2002a) How the Local Government Act 2000 is working HC 602I [London: The Stationary Office] House of Commons, Transport, Local Government and the Regions Committee (2002b) How the Local Government Act 2000 is working HC 602II [London: The Stationary Office] Leach, S, Copus, C & Snape, S (2002) The Development of Overview and Scrutiny [London: ODPM] (Maud 1967) Ministry of Housing and Local Government (1967) Management of Local Government: Volume 1 (Report of the Maud Committee) [London: HMSO] Office of the Deputy Prime Minister (2002) The Development of Overview and Scrutiny in Local Government [London: HMSO] Office of the Deputy Prime Minister (2004) Operating the New Council Constitutions in English Local Authorities: A Process Evaluation [ELG Evaluation Team, July 2004] Olson, D (1980) The Legislative Process: A Comparative Approach [New York: Harper and Row] Samuels, D & Shugart, M (tbp. 2010) Presidents, Parties, and Prime Ministers: How the Separation of Powers Affects Party Organization and Behavior [Cambridge: Cambridge University Press] Strøm, K (1990) A Behavioural Theory of Competitive Political Parties American Journal of Political Science Vol 34, No 2 Strøm, K & Müller, W (1999) Policy, Office or Votes? How Political Parties in Western Europe Make Hard Decisions [Cambridge: Cambridge University Press] (Widdicombe 1986) Committee of Inquiry into the Conduct of Local Authority Business (1986) Research Volume I: The Political Organisation of Local Authorities Cm.9798 [London: HMSO] 25 Please do not cite as the research is ongoing and any results are in draft. For further information, do not hesitate to email mxe760@bham.ac.uk Copyright PSA 2010