Should we just leave the selection of the party leader to MPs?

From 1922-1980, Labour leaders were elected by MPs in a secret ballot. While there are superficially attractive reasons for the increased trend towards democratising party leadership elections, Richard Johnson questions whether it has really given us better value and proposes that we should reconsider the merits of leaving party leadership selection to Labour MPs. More specifically, he reviews four advantages to the PLP ballot system: constitutional propriety, a better screening process, incumbent accountability, and procedural superiority.

In 2014, as a result of the Collins Review, the Labour Party abolished its tripartite electoral college structure and effectively opened the leadership selection to any member of the public. This was a serious and dramatic change, which was a culmination of a decades-long trend within the Labour Party of widening the leadership ‘selectorate’.

While there may be ostensibly attractive reasons for this democratisation of party leadership elections, it is worth pausing to understand how we got here and whether this approach has served the Labour Party well.

The origins of the party leadership role are surprisingly misunderstood. Keir Hardie is often cited as the first Labour leader, but the title really ought to belong to Ramsay MacDonald. Hardie served as ‘Chairman of the Parliamentary Party’, elected by Labour MPs after the 1906 general election, but the position was largely managerial and held little symbolic or actual power. Between 1906 and 1922 the chairmanship rotated among six men for terms of about two years or so each.

It was not until 1922, when Labour surpassed the Liberals as the second largest party in Parliament, that there was a growing opinion within the Parliamentary Labour Party (PLP) that Labour needed a high profile leader who could be seen, as Timothy Heppell writes in Choosing the Labour Leader, as possessing ‘the characteristics of a potential Prime Minister in waiting’. The PLP dumped the incumbent chair John Robert Clynes in favour of Ramsay MacDonald and changed the title of the position from ‘Chairman of the Parliamentary Party’ to ‘Chairman and Leader of the Parliamentary Party’. This was a critical symbolic change as it reflected the intention of presenting the head of the parliamentary party as the leader of the entire Labour Party, something which had not been done before.

Between 1922 and 1980, the process of electing the Labour leader remained basically unchanged. At the annual meeting of the PLP at the beginning of the new parliamentary session, nominations would be taken for leader. This is still how Labour groups on councils choose their leadership. In nearly all instances the incumbent was re-nominated without opposition.

If a leader died or resigned, MPs would be invited to put their name forward (with a proposer and seconder) for an open ballot. Labour MPs, then, cast their vote by secret ballot. If no candidate received a majority, then the candidate with the lowest vote would be removed and a subsequent ballot would be taken, usually about a week later.

For most of its history, the system of electing leader by PLP ballot was subjected to little criticism. In 1972, Labour Party conference voted against scrapping the PLP ballot, and in 1976 the PLP ballot remained the favourite method of electing a leader according to a consultation of party conference delegates.  Robert Malcolm Punnett, a political historian at the University of Strathclyde, explained, ‘It was generally regarded as a straightforward, practical, and efficient method of election’.

However, in the early 1980s, growing divisions within the party led some to see electoral reform as salvation for the party’s growing disunity. Electoral reform is a popular panacea for problems totally unrelated to electoral procedure (see current siren voices in the Labour Party calling for PR), but it often leads to worse, unintended outcomes.

At a special conference in Wembley in 1981, delegates voted to move the election of party leader from the PLP to party conference. Since then, Labour has experimented with several configurations for electing its leader. None has yet been satisfactory.

Initially, the leader was elected by party conference with votes divided into an electoral college which gave union delegates 40% of the vote and Constituency Labour Party (CLP) delegates and MPs were 30% of the vote each. Neil Kinnock and John Smith were elected under this arrangement but since neither faced serious opposition, the method of election was largely academic. As Peter Jenkins wrote in the Independent, John Smith could have been elected leader under any electoral system ‘with the possible exception of the one to select the Dalai Lama, in which Tibetan monks wander the hills searching for a child to be their priestly king’.

In 1993, Smith delivered on a promise to move the Labour leadership selection from party conference to the wider party membership. While the electoral college was retained, a few crucial changes were made. First, individual party and trade union members (rather than conference delegates) would vote for leader by postal ballot. Second, the proportion of the electoral college afforded to union members, party members, and MPs was equalised. Additionally, the Alternative Vote would be used, in which voters would rank their preferred leadership candidates.

The process was used to elect Tony Blair in 1994 and repeated once more, in 2010, when Ed Miliband narrowly won on the fourth ballot against his brother David. As is well known, while David won a majority in the party member and PLP sections, this was insufficient to counter Ed’s overwhelming dominance in the trade union and affiliated members section, in which Ed commanded nearly 60% support.

In 2014, facing criticism in the press about the legitimacy of his own election, Miliband led the party to scrap the electoral college altogether. The Collins Review eliminated the role of MPs (except at the nomination stage) and threw open the electorate beyond party and trade union members to include a new class of ‘registered supporters’. These are individuals who are not members of the party or of an affiliated union but instead pay a fee and tick a box in affirmation of the following:

‘I support the aims and values of the Labour Party, and I am not a supporter of any organisation opposed to it.’

This is a dramatic change. While all of the other parties in British politics restrict their leadership voting to party members, Labour drastically lowered the barriers to entry, potentially allowing for all sorts of mischief. Thomas Quinn, a senior lecturer in government at the University of Essex, describes the new arrangement as an ‘important departure in leadership elections in Britain’ which ‘represents a step to the use of primaries in leadership selection’.

While some may now believe that the electoral system has (or has not) delivered their preferred leadership candidate, what is obvious is that the party stumbled into this arrangement for largely short-term and spurious reasons – namely, Miliband’s own attempts in late 2013/early 2014 to increase his floundering popularity in reaction to press criticisms.

Rather than return to the electoral college, I suggest that the party might want to consider whether it would be better off leaving leadership elections to MPs. There are several advantages to returning to the PLP ballot. I will consider four: constitutional propriety, a superior screening process, incumbent accountability, and procedural superiority.

First, in a Westminster system it is ‘constitutionally appropriate’ for the party leader to be elected by MPs.  Much like we expect the Prime Minister of the day to command the confidence of a majority of the House of Commons, rather than a direct plebiscite of voters, we should expect the leaders of the parliamentary parties to command a majority of their own MPs. Ed Miliband’s failure to secure the majority support of his own MPs, I would argue, was destabilising and undermined his credibility as leader.

Second, it is reasonable to think that MPs would have the most intimate and accurate knowledge of the leadership contenders. As Jim Callaghan reflected in his autobiography, Labour MPs live with each other ‘cheek by jowl’. He wrote that in 1976 he felt no need to give statements regarding the leadership election because the MPs already ‘were fully aware of my strengths and weaknesses’.

Historically, Labour MPs showed themselves to be pragmatists in this regard. In 1963, George Brown should have been expected to have won the leadership contest given that the PLP was dominated at that time by the Right. Yet, the PLP wisely selected the Left’s candidate Harold Wilson because, as Timothy Heppel writes, they were aware that Brown’s heavy drinking (largely obscured from the public with euphemisms about Brown being ‘tired and emotional’) rendered him an unsuitable leader.

Third, the annual re-selection of the leader offered an institutionalised way to challenge and, potentially, remove a leader who had lost the confidence of the party. Every party leader between 1922 and 1980 received a clear mandate of the majority of Labour MPs – a mandate which was refreshed each year. This ensured that the party leader needed to remain sensitive to all factions within the party in order to see off leadership challenges. When a challenge did occur (which was exceedingly rare), it could be an important way of refreshing a leader’s mandate. Hugh Gaitskell’s defeats of left-wing challengers in 1960 and 1961 strengthened his position within the parliamentary party and, arguably, secured the dominance of his position in opposition to unilateral nuclear disarmament.

Fourth, there are good procedural reasons why the PLP ballot system was superior. If no candidate received a majority of support, a subsequent ballot usually would not be taken for another week. Candidates could withdraw at subsequent stages if it was felt that a particular candidate was emerging as the preferred choice. In some cases, it also gave genuinely undecided Labour MPs the opportunity to return to their constituencies to consult with party members about their vote.

Administratively, the PLP ballot system is cheap, simple, and relatively quick. Since 1983, the average time it has taken to select a new Labour Party leader is 15.5 weeks. In contrast, under the parliamentary ballot, it took 3 weeks to elect Michael Foot in 1980, 4 weeks to elect Jim Callaghan in 1976, and 20 days to elect Harold Wilson in 1963.

One of the most loudly voiced concerns about the last leadership election was that its protracted nature distracted Labour from its task of confronting the Conservative narrative about the economy, a position from which Labour never recovered. In spite of this, last year the NEC decided in its wisdom to keep the lengthy process largely intact. It took 137 days to elect Ed Miliband; it took 127 days to elect his successor Jeremy Corbyn.

Before looking to presidential systems like America or France for advice, perhaps we should heed the warning of Leon Epstein, a politics academic at the University of Wisconsin, who wrote ruefully that American political observers ‘yearn for something closer to the old British order – in which the power to choose leaders was firmly in the hands of those most familiar with the debating performance, political knowledge, ministerial ability, and personal character of the contenders’.

Are we certain that ‘democratising’ leadership elections has really given us better value?  I think we could do worse than consider the merits of an arrangement which served the Labour Party well for three-quarters of a century.


Note: This article is a modified version of a lecture delivered to the Cambridge Labour Party in June 2015. It represents the views of the author and not those of Democratic Audit or the LSE. Please read our comments policy before posting.


Richard Johnson is a doctoral candidate in Politics at Oxford University. He researches representation and elections in the United Kingdom and the United States. With Ashley Walsh, he is the co-author of Camaraderie: One Hundred Years of the Cambridge Labour Party, 1912-2012.

Similar Posts