The role and powers of the Prime Minister

Written evidence submitted by Paul Webb, Professor of Politics, University of Sussex.

This submission comes in response to an invitation from the Committee. Specifically, I wish to address the final question set out in the document Issues and Questions: Prime Minister, Prerogative and Power: [1]

Should the Prime Minister be directly elected by the British people?

I would answer 'emphatically not'.

First, it is hard to see what problem of the British political system such a reform might be intended to rectify: a prominent recent example of just such a reform is provided by Israel, where the dual intention was to enhance the power of the Prime Minister within the executive and to reduce the number of parties represented in the Knesset. Neither objective was achieved, in fact, as I shall shortly explain, but in any case it is hard to imagine that any serious analysis of the British case would currently perceive the need for a stronger British Prime Minister or for fewer parties in the House of Commons. To the contrary, it is common for critics to argue the precise opposite of these things – that the Prime Minister is too easily able to dominate the Cabinet, or that the workings of the electoral system prevent greater parliamentary representation for minor parties. So why should the Prime Minister be directly elected by the people?

Perhaps a directly elected Prime Minister would be somehow more easily held to account by the people? In principle, a directly elected head of the executive might appear to be more accountable in the sense that he or she could be removed from office by popular vote. However, although it is formally true that under the present arrangement the executive and Prime Minister are only indirectly accountable via elections to the House of Commons, few would deny that this mechanism already provides a high degree of de facto accountability. Indeed, it is an often claimed virtue of First Past The Post that it enables voters to 'throw the rascals out' if they are dissatisfied with an incumbent government and Prime Minister. There is in fact a significant body of evidence to suggest that one of the considerations that voters have uppermost in mind when visiting the polling station is the relative attraction of rival party leaders: who would make the best Prime Minister? This is in part a cause and in part a consequence of election campaigns that focus a great deal of attention on these individual politicians. Parties generally place their leaders front and centre of their campaigns, the media focus relentlessly on the leaders, and the advent of televised leaders' debates only serves to further entrench this tendency. It is small wonder then that voters often think in terms of who would make the best Prime Minister. [2] Given that this is the reality of the situation, it surely has to be conceded that incumbent premiers are obliged to consider each general election a popular referendum on their own performance in government.

To repeat, it is therefore hard to see what rationale there could be for the introduction of a directly elected prime minister in the UK. In practical terms, it is unlikely to enhance the accountability of the head of the executive, and there is no great clamour for a yet stronger Prime Minister or a less fragmented party system in the House of Commons. However, lest there should be any doubt about the potential consequences of such an innovation in this country, we should learn the lessons of the Israeli experience. A law passed by the Knesset in 1992 introduced direct elections for the Prime Minister and three such elections actually took place (in 1996, 1999 and 2001 when Benjamin Netenyahu, Ehud Barak and Ariel Sharon were the respective victors.) After the last of these contests, the law was repealed. Why? Because, against all prior expectations, the reform only served to decrease the power of the Prime Minister and increase the fragmentation of the party system in the legislature, thereby rendering the country less governable than hitherto.

The direct election of the Prime Minister in Israel, 1996-2001. [3]

Under parliamentarism, the executive emerges from and is responsible to the legislature – a fusion of powers – whereas under presidentialism, there is a separation of executive origin and survival from the legislature. With the direct election of the Israeli Prime Minister in 1996, however, the head of the executive no longer emerged from the legislature but was separately elected. Israel thus no longer belonged to the parliamentary regimes category, yet neither did it fully cross into the presidential category, because while the Prime Minister was elected separately, he continued to be responsible to the Knesset; the Prime Minister could still be removed from office by a simple majority vote in the legislature. This generated a situation in which Israel had none of the advantages of either system; the executive lacked autonomy from the legislature as under a presidential separation of powers, but without enjoying the support of a disciplined parliamentary following that generally typifies parliamentarism.

For the record, Israel was not even a ‘semi-presidential’ system like the French Fifth Republic. Semi-presidentialism, [4] is a regime in which a directly elected president coexists with a government headed by a premier who rests on parliamentary confidence. Under this dual executive system, when the directly elected President has the support of the legislature, power rests in the hands of the President. However, when the President is faced with a hostile legislative majority, the Prime Minister can take control of the reins of power.

This flexible and intermediate type of regime was not exactly the case in Israel, though. When the legislative and executive majorities coincided – the results of the 1996 and 1999 elections – Israel’s political system functioned much like the French system does when the President has a supportive legislature. But, when the legislative and executive majorities did not coincide, there was no parliamentary-supported premier in Israel to lead a government backed by the legislature. In Israel, it was the directly elected Prime Minister himself who headed the government and rested on parliamentary confidence. So in this case, the Israeli Prime Minister, and the entire executive branch, found themselves confronting a hostile legislature, akin to the situation of divided majorities in a presidential regime - with the important difference, however, that presidents cannot be removed from power by a hostile legislature. In the event, when the directly elected Prime Ministers saw their legislative majorities collapse, their response was not to continue governing in a presidential manner – by, for example, building ad hoc legislative coalitions on particular issues – but either to support early elections for both Prime Minister and Parliament (as Netanyahu did in 1998) or to resign and hold new elections for only the Prime Minister (as Barak did in 2000).

In addition, there were a number of unanticipated negative effects of the directly elected premierships in Israel:

· Israeli electors took to split-ticket voting – using their prime ministerial votes to opt for the main candidates from the largest parties, and their legislative votes to 'express' their sympathy for particular causes. This actually served to weaken Labour and Likud in the Knesset and increase the multiparty fragmentation of the legislature. In other words, the electoral reform not only failed to attack the problem for which it was designed, but actually made it worse.

· The implications for governability in light of major party decline and party system fragmentation were clear. Holding together stable coalitions became an even more precarious business than it had been hitherto. The efforts of the government to pass its own legislation, or to thwart the opposition’s popular and costly bills, largely failed. The annual budgets, for example, were revised by the coalition members in the Finance Committee – at times with the cooperation of the opposition – to an extent that was previously unknown in Israel.

· One of the results of the increasingly apparent negative consequences of the electoral reform was that public support for the new system deteriorated decisively during the years it was applied. In a 1992 survey, before it was implemented, three out of four Israelis thought the direct election of the Prime Minister would be a better system of government. By the time it was repealed in 2001, only one out of four thought it was a better system of government.

What would happen if the UK directly elected its Prime Minister?

Of course, the UK is not Israel, and while it is important to be aware of what happened in this particular case, we should be careful not to read the experience as a straightforward analog of what might happen if a directly elected premier was introduced here. I would offer the following thoughts and speculations, though.

First, one thing that Britain most probably would emulate Israel in is a degree of split-ticket voting. There is already evidence that voters across the UK are more than capable of distinguishing the way that they use their votes for different offices (eg, voters in Scotland and Wales do not all choose the same parties for devolved and Westminster elections), even when obliged to make those choices simultaneously (eg, when local and Westminster elections are held on the same day). As in Israel, many would probably be tempted to split their tickets, not least in those occasional situations where one major party is less popular than the other, but it nevertheless has the more popular leader. Evidence suggests this was true in 1970, for instance, when Harold Wilson was personally more popular than Ted Heath – but the Conservatives were the more popular party and won the election. If there had been a directly elected premier, would Wilson have been returned to Number 10, only to be faced with a hostile majority in the Commons? And if, as in Israel, voters were inclined to use their choice of MP as an 'expressive' vote, would it generate more support for minor parties, thereby perhaps increasing the multi-party nature of the Commons and generating a need for more coalition-building? Some might see this as a welcome development, but inevitably many others would regard this as a threat to 'strong' government.

Second, if the UK were to have directly elected Prime Ministers, should it also follow the Israeli example of making the premier accountable to the legislature, or should it introduce presidential-style separation of powers? One obvious problem with the Israeli model whereby parliament can remove the premier is that this undermines the direct line of accountability which flows from Prime Minister to the electorate. If the head of the executive is directly elected by the people, who but the people should have the right to remove him or her? In terms of democratic theory, it is surely logical and desirable either to have a directly elected Prime Minster who can only be brought down by a direct vote of the people, or to retain the current situation: the people elect the legislature; the executive emerges from the legislature; the executive is therefore answerable to the legislature and can legitimately be removed from office by it. A directly elected premier who can be voted down by the legislature only muddies the waters from the point of view of clear democratic accountability, however.

To this extent, then, presidential-style separation of powers seems preferable. However, while the Prime Minister would be secure in office until the next election under such a system, we should not assume that this would necessarily make him or her more politically dominant or enhance the governability of the country. A Prime Minister could find himself in a position akin to that of the US President under 'divided government' – ie, when confronted by a Congress in which his political opponents are in the majority. In general terms, the separation of powers tends to expand the independence, not the compliance, of the legislature. [5] When backbench parliamentarians do not feel fear to bring down the executive by voting against their leader's wishes, they are more likely to feel free to rebel. While some might regard such a development as healthy for democracy, it should be remembered that there is a price to be paid; governments can find it harder to pilot their legislative programmes through Parliament, a situation which may lead to allegations of ungovernability and 'gridlock'. In the long run, voters are unlikely to be impressed by such developments. Therefore, while the expectation is that a directly elected premier under 'divided government' should seek to build legislative coalitions on an issue-by-issue basis, as the US President does, it would probably be advisable to furnish the head of the executive with the right to dissolve Parliament and call fresh elections, in order to avoid the risk of political immobilism.


To summarise, there is little in the current situation at Westminster which suggests the need for a directly elected prime minster, and nothing in the empirical experience of the one country that has attempted such a reform in recent times to commend it. If introduced, it would probably weaken the position of the head of the executive, and generate a greater prospect of coalitional or even minority governments. Neither of these scenarios is necessarily disastrous – they are far from uncommon across Europe – but they would constitute a significant shift in the way that politics has been conducted at Westminster, and may not be to the taste of the electorate. If directly elected Prime Ministers were to be introduced, I would recommend combining this with separation of powers and the continuing right of the premier to dissolve Parliament. Overall, however, I cannot see a compelling case for changing from the present more orthodox form of parliamentary government that the UK operates with.

18 February 2011

[1] constitutional-reform-committee/news/pm-powers-inquiry/discussion-paper-pm-powers/.

[2] H. Clarke, D. Sanders, M. Stewart & P. Whiteley (2004) Political Choice in Britain (Oxford University Press); Foley, M (2000) The Rise of the British Presidency (Manchester: Manchester University Press) ; Kavanagh , D (1995) Election Campaigning: The New Marketing of Politics (Oxford: Basil Blackwell); Mughan , A (2000) The Presidentialization of Elections in Britain (Basingstoke: Palgrave); Graetz , B & McAllister, I (1987) ‘Party leaders and election outcomes in Britain, 1974-1983' Comparative Political Studies; Rosenbaum, M (1997) From Soapbox to Soundbite : Party Political Campaigning in Britain Since 1945 (Basingstoke: Macmillan).

[3] Much of what follows draws on R. Hazan 'The failure of presidential parliamentarism : Constitutional versus structural presidentialization in Israel's parliamentary democracy' in T. Poguntke & P. Webb ( ed's ) The Presidentialization of Politics: A Comparative Study of Modern Democracies (Oxford University Press, 2005).


[4] Duverger , Maurice (1980). ‘A New Political System Model: Semi-Presidential Government.’ European Journal of Political Research 8:2.

[5] Laver, Michael and Kenneth A. Shepsle (1994). Cabinet Ministers and Parliamentary Government (Cambridge University Press).