The presidentialisation of the prime minister thesis should be expunged from political science vocabulary. To the extent that the forces identified by those who pursue the thesis exist, they do not make the British prime minister more like the US president. Quite the reverse: they enhance the different and already stronger powers of the prime minister. The prime minister's offices serve a different function from that of the White House. The roles of the prime minister and the president as leaders of their parties are entirely different. The personalisation of politics is an analytically separate process, and affects parliamentary and presidential systems alike. Media representation of prime ministers as ‘presidential’ is entirely superficial; political science needs to plunge deeper into the institutional forces of presidential and prime ministerial power. The institutions of presidential and parliamentary systems are so different that any global force acting upon them are as likely to drive them further apart as lead them to converge. Prime ministers are more powerful within their systems than presidents; strengthening their powers makes them less, not more, like presidents.
John Hart's prescient prediction has certainly come to pass: a superficial thesis is discussed in all seriousness in major scholarly journals. The claim has been most popularised by Michael Foley (1993, 2000, 2002, 2004), who suggests that prime ministers have become more presidential over the past 30–40 years, and that this trend is likely to continue, recent coalition politics notwithstanding. Foley's claim is not particularly new: non-academic commentators, journalists and politicians have made the assertion for many years. George Brown, for example, justified his resignation from Wilson's cabinet on these grounds (Brown, 1972, p. 161; see also Price, 1997 for earlier claims). Similar claims, though not always so stridently put, have been made for other parliamentary systems (see, for example, Lutjen and Walter, 2000; Poguntke and Webb, 2005a,b; Bennister, 2007). In fact, the thesis has had more critics than supporters both for the UK (Hart, 1991; Norton, 2003; Heffernan, 2005; Jones, 2006; Rhodes, 2007) and for other systems (Hart, 1992; Fabbrini, 1994; Helms, 2005).
If one has to use such terms, it is certainly less grating to talk about the ‘presidentialization’ of British politics than its ‘prime-ministerialization’, and, if the essence of this shorthand form is to get across the idea that government revolves around a single person, then it is quite likely that the message will be grasped more easily, at a popular level, by concentrating on the archetype of one-person leadership in democratic systems—the American presidency. But there is always a danger that, when relatively complex terminology is popularized and debased, the popularization will feed back into scholarly debate. (Hart, 1991, p. 212).
I would like to finally put an end to the presidentialisation argument by pointing out that (i) the forces that have been identified as presidentialisation are better seen as personalisation of politics, decreasing emphasis on parties; (ii) the personalisation of politics applies equally to presidential systems; (iii) centralisation in the executive takes us further away from presidentialism and (iv) to the extent these forces increase the power of the prime minister, they cause further divergence from presidentialism (since prime ministers have always been more powerful than presidents). I make these points through a consideration of the British prime minister in comparison with the US president (which is what commentators seem to have in mind) and in relation to both behavioural and institutional claims. Presidential systems are institutionally so different from parliamentary ones that we should be instantly suspicious of any such institutional claim, but the assertion is usually made in behavioural terms. However, executive behaviour can only be analysed through the institutional form that makes it appropriate.
Presidential and parliamentary systems are institutionally completely different arrangements. Poguntke and Webb (2005a,b) recognise this fact, but posit that three processes are leading to presidentialisation. First, premiers are enhancing their power resources; secondly, premiers are less constrained by their party than once they were; and thirdly, there is a personalisation of politics. They argue that these changes occur through the executive, through the party and through elections. However, the thesis completely misspecifies the functions of the central offices of the British prime minister and the US president. Presidential offices have been enhanced to gain greater bargaining power with the legislature. The British prime minister might have gained more power at the centre of the executive in order to control it. The process of centralising in the UK enhances the prime minister's control over policy formation, already greater than that enjoyed by presidents. Second, the relationships between the party and party leader have virtually nothing in common across the systems. To the extent that British party leaders are less constrained by their party than once they were, they lead their parties rather than are subject to their control. US presidents do not lead their parties in that manner; and British parliamentary candidates are much more affected by how their leader is perceived than are Congressional candidates. Third, whilst a personalisation of politics is occurring, it is happening in both presidential and parliamentary systems. What Foley and others identify is not a presidentialisation but, gratingly or not, a prime ministerialisation of the British prime minister.
Essentially the presidentialism of the prime minister thesis is a behavioural one. The claim is that (British) prime ministers are behaving more like presidents than prime ministers. They act through Poguntke and Webb's (2005a,b) three faces—executive, party and electoral. Prime ministers are taking on roles normally associated with the head of state; at election times they are dominant, with the focus on the leaders of the main contending parties rather than the parties themselves and their policies and the prime minister is gaining power at the centre of government, dominating their cabinet, their party and the legislature. As prime ministers accrue these extra powers through their behaviour, they start to be treated in a more presidential manner by other actors in the system.
There is no question that these factors are in play, though I think Foley exaggerates some with regard to modern prime ministers and ignores earlier precedents.1 Presidents fulfil a ceremonial role not exercised by British prime ministers to the same degree. In recent years, the latter may have enlarged this function—though to no greater extent than Churchill during the Second World War. The Queen continues to carry out the ceremonial role and cement relationships, witness her recent visits to Ireland and Australia in 2011. This is not to say that the prime minister does not take an important role in foreign affairs. The speed of modern transport allows the prime minister to meet other heads of state more often than 50 years ago, when the foreign secretary would usually be the minister representing the nation abroad, both at state functions such as funerals and in diplomatic negotiations.
The leaders of the major political parties have taken on a much higher profile to become the focus of election campaigns, though this tide has ebbed and flowed. Foley makes much of the photographs of Blair in the Labour manifestos of 1997 and 2001. But this is an aberration. Whilst photographs of Labour leaders do not feature (much) in earlier manifestos, the 1997 and 2001 Labour manifestos include eight and seven photographs of Tony Blair, respectively; but this was down to 1 in 2005, and—not surprisingly—none of Brown in 2010. Neither the Conservatives nor the Liberals have ever had more than one or two photographs of their leader. This indicates the personalisation of Blair, not of prime ministers. In 2010, for the first time in the UK a party leader debate took place, echoing the US presidential debates dating from 1960 (with some earlier precedents: see Minow and LaMay, 2008). All the major British political parties have centralised their general election campaigns, instructing local candidates on issues to highlight and the correct line to take. The media concentrate their attention upon the party leaders, while the government's response to the demands of today's 24-hour media has been to centralize its message from Number 10 Downing Street. The advent of the Downing Street Press Conference has certainly introduced a more presidential air to prime ministers, though it has not yet gained the status and centrality associated with US presidents' press conferences.
These behavioural effects, as Foley (1993, 2000) makes clear, are a result of media pressures. However, they do not affect British prime ministers alone. They have had an impact upon presidents too. The ‘personalisation of politics’ is apparent throughout the democratic world (Campbell, 1998, pp. 29–33; McAllister, 2007) and has increased in the past 30 years or so. However, personalisation and presidentialisation are analytically separate: presidential as well as prime ministerial systems can become more personalized—as commentators have argued is the case in the USA (Lowi, 1985; Campbell, 1998). Furthermore, the effects of the personalisation of politics can drive presidential and prime ministerial systems further apart.
Of course presidential election campaigns concentrate upon the personalities of the candidates, and the success of one or other has a coattail effect on the success of candidates in Senate and House elections. However, these elections too have become more personalized and candidates can, and sometimes do, distance themselves from their party's presidential candidate. This happens most spectacularly at mid-term elections when the president may be unpopular, but can also be seen during a presidential campaign. In fact in presidential systems, the local candidates are more important than local candidates in parliamentary elections—and always have been. The incumbency effect is much greater in the USA than in the UK, largely as a result of the abilities of Congressmen to affect their fortunes outside of party considerations (Krehbiel and Wright, 1983; Cain et al., 1987; Wood and Norton, 1992). In the UK, removing an unpopular prime minister or executive involves voting against their party no matter who your local member is. Whilst unpopular presidents do have a partisan effect on voting for Congressmen, the latter can distance themselves from their president in the election campaigns without threatening to remove him and split ticket voting is possible and often occurs. The systems are very different and that has a massive effect on the chief executive bases of power.
It is true that in the past 50 years or so there has been increasing emphasis in the UK upon party leaders rather than election manifestos and policy commitments, but the increasing importance of the leaders rather than local candidates points to a ‘prime ministerialisation’ of British politics, not a presidentialisation. The phenomenon is taking the UK further away from the USA, not closer to it. The reason is that the British political parties are (and have long been) far more disciplined as electoral units than US parties. Even earlier in the twentieth century, when they were not so centrally organized, British political parties were more ideologically coherent. Leaders can affect the electoral fortunes of their party by influencing the ideology and party organisation (as both Kinnock and Blair did for the Labour Party) or directly through their personal appeal to the electorate (Crewe and King, 1994). The first undoubtedly has had major effects on the electoral fortunes of parties, but is a long way from presidential politics. The argument of Allen (2001) and Poguntke and Webb (2005a,b) that leaders of parties are less constrained by their party might signal a change in British party politics. It does not lead to presidentialism, quite the reverse. In the UK candidates have to follow their leader at elections and in parliament; in the USA candidates do not and often distance themselves at elections and bargain with the President over legislation. The second does indeed make the focus of the election the party leaders, but again personalisation is a better term, because of the important institutional differences between parliamentary and presidential elections even when legislative elections accompany the latter. Unfortunately for the presidentialisation thesis, there is little evidence of a leader effect in British politics (Crewe and King, 1994; Bartle and Crewe, 2002; McAllister, 2007).
While we should not overlook early manifestations of the personalisation of politics—such prime ministerial colossi as Winston Churchill or, still longer ago, Disraeli, Gladstone and Lloyd George (Jones, 2006)—it is clear that the increasing importance of the British prime minister as a de facto head of state and dominating media actor at election time and throughout government is a significant change over the past half century or so. Again, however, this shows the personalisation and not the presidentialisation of politics since one of its effects—lowering the importance of local candidates at elections—takes us further away from the US situation, not closer to it.
The presidentialism thesis also asserts that the British prime minister has centralized power into his own hands and away from his cabinet. The argument is that he is now much more than primus inter pares, the dominant actor bestriding the entire legislative edifice and reducing cabinet government to more of a mockery than a sham. This aspect of the thesis includes claims about institutional changes in the higher structures of British government to which I now turn.
The presidentialisation thesis argues that the power of the British prime minister has grown enormously, not only through media concentration on his personality, but also through the growth of his institutional resources. Again there is some truth in this: the institutional resources of the British prime minister have grown in relation to those of his ministers; but once more I will argue that this takes us further away from the case in the US presidential system and at best promotes a ‘prime ministerialisation’ thesis.
The fundamental aspect of the changing structure is the changing nature of the British cabinet. Prime ministerial government traditionally works through a cabinet that is responsible to parliament. The prime minister is the agent of parliament and ministers are agents of the prime minister. As ‘first among equals’ the prime minister is influenced by collective cabinet responsibility, just as other ministers are. The fundamental idea is that government is cabinet run, and each minister is master of his own department—within the constraints of collective responsibility and maintaining policy coherence across departments. Within their own domain ministers still remain largely supreme, likened by Lord Norton (2000) to medieval barons with their own courtiers, fighting or forming alliances with other barons to get what they want. Ministers rely upon each other for support just as they require the support of the prime minister. This system means that ministers are powerful in their own right, and so are their departments. Of course, some departments are more important than others, and some might take on importance at specific times given electoral commitments. Furthermore, some ministers will be more powerful than others, due to force of personality or the fact they are seen as leaders of important party factions. However, all ministers are powerful within their own domain. The institutional power of ministers is that they have whole bureaucracies, departments, to help them run their briefs. Traditionally, prime ministers had only small executive bodies to help them coordinate across all the departments.
There is no question that in the post-war period ministers have tended to become less dominant within their own domains and the prime minister more dominant. The business of government has grown and become more complex. Ministers are more restricted in what they can do because of constraints laid down by Europe and greater Treasury oversight. Prime ministers have always struggled to try to maintain coherence within their government; over the years they have tried many ways to coordinate functions. They have always had a central executive; the growth of the prime minister's private office has enabled him to coordinate activities to a much greater extent than ever before. This process has had a long gestation. Clement Attlee entered government in 1947 with a permanent cabinet secretariat not greatly different from when it was created in 1916. The secretariat was extended and consolidated under the Conservative governments of the 1950s and 1960s, and under Macmillan we saw the beginning of what became the political section (Burch and Holiday, 1996, p. 21). Heath doubled the number of senior Cabinet Office staff and set up specific units, importantly the 24-person Central Policy Review Staff, as part of his attempt to remodel Whitehall and provide greater strategic direction of government (Blackstone and Plowden, 1988). The Cabinet Office grew since the 1970s from around 600 staff to around 2500 at its height in during Blair's government (though not all directly serving Blair, but now around 1000 under Cameron (Blick and Jones, 2010; Cabinet Office, 2012). Staff numbers within the prime minister's office (PMO) have also increased. From under 30 in Attlee's day to around 70 under Heath, to just over 100 under Major, numbers grew to over 200 under Blair, reducing a little to 180 under Cameron (Blick and Jones, 2010; Cabinet Office, 2012). Special advisers to the prime minister, once numbered in ones or twos, increased under Thatcher, going up to eight under John Major, to 27 under Blair (Burch and Holliday, 1996; Blick, 2004, Appendix; Blick and Jones, 2010), before falling back to 18 under Brown, 12 under Cameron initially (Guardian, 2010) rising to 20 (Bennister and Heffernan, 2011). Cameron initially cut back on the centralized offices, partly in reaction to what was seen as Blair's control freakery (Kavanagh, 2001; Riddell, 2001); and the early days of the coalition government suggest a markedly lower degree of coordination in government—at least in presentation. Over even this short time, the centre has grown again.
During the same period, the office of the president of the USA has grown too. The ‘presidential branch’—that is, the large White House staff system that dominates the Executive Office—has grown massively since Roosevelt formed it in 1939. Whilst many reasons have been offered for this growth, an important if not overriding reason has been the presidents' bargaining relationship with Congress (Dickinson and Lebo, 2007). This is in marked contrast with the office of the prime minister, the major function of which is coordinating across the executive. Again a superficial look at growth within the Cabinet and PMOs suggests a presidentialisation since the president has a large staff too. But the institutional function of these offices is, in reality, quite different. Both have grown as the complexity and scope of government have grown (together with the need to deal with an ever more prevalent media), but one has its main efforts directed at the legislature, the other at the executive. Centralisation in the UK entails prime ministerialisation and not presidentialisation.
There have always been ups and downs in these central offices, with coordination being handled differently under different prime ministers. Thatcher undid most of the structural changes introduced by Heath, preferring that strategic planning be run through the PMO. Most controversies in recent years relating to the PMO have concerned the political offices. Prime ministers have always had political advisers around them (though the Civil Service Yearbook did not recognize the political office as a distinctive element in the PMO until 1983). Political advisers, and the blurring of the lines between civil servants and political appointees, are important in considering the accountability of ministers, especially if they take on what has traditionally been the ministerial role of directing civil servants, but without accountability for their actions reflecting back on to ministers (Tiernan, 2007; Eichbaum and Shaw, 2008; 2009; Maley, 2011). Poguntke and Webb (2005b, p. 14) also suggest the hollowing-out of the state causes more bilateral meetings between ministers and the prime minister reducing collective cabinet responsibility. This can be so only if it allows other ministers to distance themselves from decisions in which they were not involved, there is no sign of that. However, what they mean, I think, is that it reduces cabinet decision-making. To the extent that bilaterals replace cabinet committees, the prime minister might be strengthened, but again this is prime ministerialisation. The key difference between the UK and the US with regard to hollowing-out is parliamentary oversight which tends to be stronger in the US.
With regard to growing ‘presidentialism’, much has been made of Blair's preferring one-to-one meetings with ministers—but so did Margaret Thatcher and Anthony Eden. What did happen under Blair was that policy initiatives in areas such as education and health were often formulated by his staff and announced by him rather than being generated from the relevant departments. Some ministers claimed the same happened under Wilson, but they seemed to be kept in the loop more than under Blair. Notwithstanding, Blair's time often appeared a dual leadership, the prime minister having conceded a great deal of welfare policy to Gordon Brown in the Treasury (see Hennessy, 2005 for how they split responsibilities). Cameron has given his ministers greater freedom over policy initiative but has proved brutal in putting them down with little warning if their policies seemed too unpopular. He has thus demonstrated the strength of the prime ministership even in a coalition government.
Another significant power of the British prime minister in comparison with the US president is his ability to restructure government. In 2006 the prime minister split up the Home Office, setting up a new Ministry of Justice and a new Office for Security and Counter-Terrorism within the Home Office. Such major reorganisations of the central government machinery are entirely within the remit of the prime minister. Contrast that with the USA, where it was Congress that set up the Homeland Security Department following the events of 9:11. Dewan and Hortala-Vallve (2011) argue that the ability of the prime minister to appoint ministers, allocate portfolios and assign responsibilities for those portfolios gives her the ability to attain her optimal policy goals. In reality, prime ministers are constrained, but far less so than any US president.
One of the most important restrictions on the power of ministers comes not from the power of the PMO but from the Treasury. A system of line-by-line oversight of departmental budgets has grown up in the past 30 years, causing the Treasury to have a much greater say in policy formation and implementation than in the past. Departments are forced to justify their expenditure on new policies and on the costs of implementing past policies at a much earlier stage, and that has allowed the Treasury (working with the prime minister, it must be said) to intervene in the policies of departments.
It should also be recognized that other ministers and departments constrain ministers. One of the biggest mistakes in considering cabinet government is to think of it as 22 ‘men’ sitting round a table. Cabinet government is about departments working out policies through whole systems of meetings; and one of the most important means of coordination is through departmental briefs. These are statements of policies sent to other departments and read carefully by permanent secretaries and other senior civil servants to ensure that one department's policy does not encroach upon or cause problems for another department. Where such problems emerge meetings must be held to try to resolve them, and the ‘centre’ might be involved in this.
These institutional changes within the heart of executive government in the UK form a growing centralisation of policy-making. This should not be mistaken for presidentialisation. The growing centralisation is in the core of the central executive, since many aspects of the implementation of policy have been devolved into government and non-government agencies in what some have called a ‘hollowing-out’ of the state (Rhodes, 1994; Weller et al., 1997). To the extent that such a hollowing-out has occurred, the power of ministers, or at least their departments, has been threatened and they have thus suffered power loss from below as well as from above. The hollowing-out could be thought to resemble a more US-like administrative structure, though in reality it bears only a superficial resemblance (and, to be fair, few have made much of this as part of a presidentialisation thesis).
The entirety of the argument regarding the institutional focus at the heart of the British executive revolves around a shift from the cabinet as the centralising agent to a small bureaucracy centred on the offices of the prime minister and the Chancellor of the Exchequer. Decisions are taken in cabinet committee, many chaired by the prime minister or his allies—some of whom gain some powers beyond what their official ministerial positions might entail (Dunleavy, 1995). Full cabinet meetings have moved from being a primary decision-making body to a formal get-together. It would be a mistake, however, to confuse the cabinet with full cabinet meetings. Cabinet ministers are important figures, heading departments, engaging in meetings with each other, civil servants and the prime minister. Modifying the way in which cabinet works, even when it enhances the institutional powers of the prime minister, does not amount to a presidentialisation of the British prime minister. Nor does it signal the end of cabinet government, even if full cabinet meetings do not play the central role they once might have done.
We should not exaggerate the growing power of prime ministers. The most dominant and, by repute, powerful prime ministers in recent years have been Margaret Thatcher and Tony Blair. Both lasted a long time, which helped to cement their reputations. Both were also right-wing outliers in their cabinet and party. Neither got everything they wanted and it is not obvious that they succeeded in getting what they wanted significantly more often than John Major who took up more of a median position in his cabinet. Thatcher and Blair needed to dominate their colleagues because they were outside the majority opinion in their cabinet and party. Major was more in the mainstream of a very split party; compromise was what he wanted (Dowding, 2008a). Brown is perceived as a weak prime minister who achieved little, though Seldon and Lodge (2011) argue that, whilst almost entirely reactive, he did achieve much of what he attempted and, given the circumstances of his premiership, perhaps no less than anyone else could have achieved. If power is getting what you want, then Major (and even Brown) might have been as powerful as Thatcher and Blair; if it is getting what you want despite resistance, then Thatcher and Blair demonstrated the power of the premiership; Major did not. Prime ministers can dominate given the institutional powers they have in government, given their personality and that of their ministers, and given their role in their parties. They always have done, though institutionally too leaders have grown in importance within their parties.
Parties have developed enormously over the past 40 years. In the UK all of the major parties have professionalized their activities. During general elections, central coordination of campaigning in every constituency is now standard, with the party line for the issue of the day first faxed, now emailed to local campaign coordinators. Parties use the web, blogosphere and twitter to communicate with members and supporters. Within parliament party discipline remains high (though not as high as in the 1960s), but MPs (largely those who have no chance of gaining government posts and those who have left them) may still prove recalcitrant, though usually over non-core issues that do not threaten a government's majority or at least its programme (Norton, 1975, 1980; Cowley, 2002, 2005; Benedetto and Hix, 2007). With Cameron's coalition government parliamentary dissent is higher than it has ever been (Cowley and Stuart, 2010), and major bills, notably Lansley's Health and Social Care Bill are being threatened to a degree that single-party government has rarely faced. Even so, British commentators need to pause when considering the strength of parliament; for in the USA it is the legislature that writes the bills as well as frequently rejecting what is put before it.
Parties have become less ideological, which also gives the prime minister greater power as the need for coalition building across warring factions becomes less necessary (though of course one would expect the coalition government to reduce this power once more). Furthermore, the development of elections for leadership, especially the convoluted process involving the entire party and associated organisations that takes place in the Labour Party, has massively strengthened the hand of the prime minister. Challenging the prime minister is costly and difficult, especially in the Labour Party.
Of course the control that the leader exercises over the party is largely shaped by the demands of the media. The media uses party dissension to attack leaders; contradictory messages from different parts of the party machine are immediately invoked as signs of problems for the leaders. In response, prime ministers (up to Cameron who seems to have relaxed this grip) have taken control of the government message away from departments over all areas of policy, in order to try and avoid the charge of chaos in government. There is no question that the press concentrates more attention upon prime ministers than in the past, but this effect would still better be viewed through the lens of the personalisation rather than presidentialisation of politics. And its effects might rather be seen as strengthening the prime minister rather than presidentialising him, since presidents are so much weaker than prime ministers.
Presidents and prime ministers
All of the above point to the growing power of the British prime minister. The office itself has been strengthened with a larger personal office, centralisation of the government's press machine, the increasing importance of cabinet committees and continued decline in the coordinating role of cabinet meetings. These changes signal a growing personalisation and centralisation of politics. They do not, analytically, entail a presidentialisation of the prime minister. It is time to properly compare the roles of the British prime minister and that of the US president and to show how their institutional situations lead them to have very different powers. We can do this by delving deeper than mere institutional facts to see how they respond to the same sorts of incentives.
Underlying the presidentialisation thesis is the idea that prime ministers are getting more powerful and thus increasingly resembling presidents. However, British prime ministers have always been more powerful than US presidents (Hart, 1991; Heffernan, 2003, 2005).2 The strengthening of prime ministerial power therefore further distances it from the presidential system. The reasons why, within their structure of government, British prime ministers are more powerful than US presidents go straight to the heart of the institutional issues—which themselves demarcate professional political science from journalism.
Let us begin by considering some simple facts about the executive and legislative systems. The British cabinet is chosen from members of the legislature, and they answer directly to that legislature via regular ministerial question times. The US cabinet is not drawn from the legislature. The vast majority of the legislation passed in the British parliament emanates directly from the executive and nothing passes without at least the tacit agreement of the executive. (That might change if the UK leaves behind the two-party dominance that has existed for so long. In other parliamentary systems where coalition governments are the norm, opposition parties can successfully introduce legislation.) The situation in the USA is very different. Here members of Congress propose legislation and very little legislation actually emanates from the president's office (it is hard to estimate how much comes from presidential initiative). Indeed, a substantial proportion of votes in the US Congress is at odds with stated presidential preferences.
The Congressional Quarterly's Presidential Support Score is a well-established measure of the frequency with which lawmakers vote in accord with the president's position when this is clear.3 The scores, which may vary from 0 to 100%, reflect the percentage of times members of Congress vote in agreement with the president's position, indicating the level of agreement between the president and congressional members. Whilst they are often reported as demonstrating ‘support for the president’, they do not measure presidential influence and might better be described as ‘presidential support for Congressional vote’ (Pritchard, 1986). The measure does not distinguish between major policy commitments and minor ones. Nor does it indicate whether the president pushed his position in Congress—often a president will just endorse something Congress wants to do and leave it to legislators to drive the measure through. Furthermore, it tends to exaggerate congruence between Congressional votes and presidential views because it counts separately several votes on the same piece of legislation, including procedural votes. On big legislative issues there can often be four or five significant votes on the same Bill. Nevertheless, whilst it is a crude indication at best of the extent of agreement between president and Congress on roll-call votes over a given year, the CQ Presidential Support Score does provide some revealing evidence relevant to comparisons between the power of British prime ministers and US presidents. Obama's Presidential Support Score for 2009 was a high 96.7%, Eisenhower was at 89% in 1953, a year when he sent very little legislation to Congress. Kennedy only had a score of 80% in 1962, as did Bush in 2001. But don't be fooled by these high scores. They do not reflect presidential bills passing through Congress but only the degree to which the president supports bills passing through Congress. Imagine a British prime minister who only supports 97% let alone 80% of bills passed in parliament! This puts into perspective the weakening of party discipline in votes in the House of Commons.
Fishel (1985) suggests that around about 50% of presidential campaign promises are actually enacted in Congress. Again, however, such a figure is crude and certainly overestimates the influence of the president. Obama's health care bill illustrates the weakness of presidents. In fact, there never was an Obama health care bill. Congress enacted a bill that emerged from several different bills on health care reform; all introduced by individual Democrats and none of them identified as the president's bill. Obama entered the stage at the end, supporting a huge and complex compromise that was two separate bills written by the Democratic leadership in the House and Senate with White House involvement. Whilst Obama got health care legislation passed, it did not contain his original core proposal known as the ‘public option’. It was a legislative success; but in the British context such a sequence of events would be considered a failure that would seriously undermine a prime minister's authority. We can compare this to the travails of the Conservative Health Minister Andrew Lansley's Health and Social Care Bill. In terms of the original bill, this could be thought of as a failure in British terms. Though the final bill as enacted (at the stage I write this) is at least as close to their original aims as Obama's health care bill was to his. A British prime minister's failure is more successful than a US president's success. Whilst public disquiet and the Lords' scrutiny and rejection would give pause to and might lead to a rethink by even the most powerful governments, a Conservative government with a Thatcher or Blair majority rather than a Conservative–Liberal Democrat coalition could have forced through the bill Lansley originally desired.
Presidents, arguably, have more support for foreign than domestic policies, almost certainly because they have greater resources and information. Congressmen are confident about their own views on domestic issues, less so on foreign affairs (Canes-Wrone et al., 2008). We might compare here the growing importance of the British prime minister in foreign affairs; but in fact British prime ministers are as likely to attract criticism of their foreign as of their domestic policy. These facts derive from the important underlying differences between the US presidential and the British parliamentary system, indeed between presidential and parliamentary systems in general. Broadly speaking, prime ministers are agenda setters who face a limited number of veto players. Presidents are only rarely agenda setters and face a greater number of veto players.
Agenda setters and veto players
An agenda setter is an agent who can present (more or less) take-it-or-leave-it proposals to other players in the political game. A veto player is an individual or collective agent who can authoritatively reject any proposal that comes before it (Tsebelis, 2002). We can distinguish institutional veto players, such as parliaments, presidents, law courts; and social veto players such as political parties or professional groups whose cooperation is required to ensure that policies are implemented.
Tsebelis argues that the greater the number of veto players within a political system, the greater will tend to be policy stability and less the radical policy change. (Of course, radical policy change can occur where all veto players agree on a specific course of action, and policy instability can occur in any system.) Tsebelis and others have provided empirical support for the thesis; generally speaking, presidential systems show greater policy stability than parliamentary ones because they have greater numbers of both institutional and social veto players. A British prime minister, almost always, controls a majority in the House of Commons and so the veto power of parliament tends to be formal only. Of course, a prime minister cannot do whatever she wishes, but given that her preferences are likely to be similar to those in her party (she was chosen as leader after all) and given the authority vested in leaders in disciplined parties, she can present more or less take-it-or-leave-it proposals to her party. Today most successful amendments to bills in parliament are indeed moved by the government or its supporters and rarely (though not never) does government have to concede amendments to which it is opposed. The most effective check on governments is through the scrutiny of the Lords, but once again, whilst in a British context the Lords has proved the most effective opposition to powerful prime ministers, what they achieve in controlling the executive's legislation is slight in comparison with how the US legislature affects legislation. In the USA the legislature writes legislation, merely influenced by the executive.
These underlying structural features of the surface ‘presidential’ versus ‘parliamentary’ institutional systems provide a major basis of the difference in power between British prime ministers and US presidents. The more detailed institutional features I have discussed above provide further institutional resources that demonstrate the greater power of the British prime minister over the US president, even when the British prime minister does not control a majority in the Commons (which, historically, they have usually done; whereas, historically, US Presidents tend not to control Congress). The structural and institutional features interact with the behavioural, of course. To take a stark example, after the 9:11 attack had a British prime minister behaved as George Bush did, it is hard to imagine that, facing an angry House of Commons, they could have survived politically. But of course, had a George Bush been a British prime minister, he would not have behaved in such a way (indeed it is unlikely that someone with his particular talents would ever have made it as far as the British premiership). People react to the institutions around them, and structure goes all the way down into their behaviour (Dowding, 2008b).
I have set out the presidentialisation of the prime minister thesis in terms of behavioural and institutional claims and found both wanting. I have directly compared presidential powers with prime ministerial ones and showed the important differences, especially when we look at the deeper structural divergences between systems with different logics. I have argued that the changes that have indeed occurred have strengthened prime ministerial control—but given that prime ministers have always been more powerful than presidents this leads us away from a presidentialisation and towards a prime ministerialisation of the British system.
The fact that British prime ministers have become more powerful in the past 40 years—a claim that has much truth but is also exaggerated by some commentators—is not evidence of the presidentialisation of the prime minister. The power of prime ministers is greater than that of presidents; adding to it makes them less, not more, presidential. It shows, on the contrary, a growing prime ministerialisation of parliamentary systems. That is not to say that that power will continue to grow unchecked. Cameron, partly through personal style but largely through the nature of coalition politics, has allowed reassertion of ministerial power, whilst coalition politics has enabled greater parliamentary revolt in the Commons and, importantly, increased the de facto powers of the Lords. The Liberal Democrats, had they chosen to demand certain areas of policy control, might well have had greater influence over larger areas of the agenda. Their choice to try to shadow everything has dissipated their resources and strengthened the control of the Conservatives. Of course, coalition politics increases the potential number of veto players, both through an organized faction within the government, but also creating extra veto players within the wings of the two parties in parliament.
There is not and has never been a growing presidentialisation of the British prime minister. There has been a growing centralisation of policy and a growing personalisation of politics. These processes have been occurring in all countries, presidential, federal, semi-presidential, as well as parliamentary. We should not mistake these institutional and social forces for presidentialisation. They are analytically separate and interact differently with institutional and structural forces within presidential and parliamentary systems. Our task is to examine the centralisation and personalisation of politics and how they interact with executive roles and powers in different systems: to bundle them up obscures rather than illuminates political systems.