What role did the party’s affiliated trade unionists play in the contest to choose Labour’s new leader? How did their votes shape Jeremy Corbyn’s landslide victory on 12 September 2015? It is an important issue because the party’s critics claimed that trade union voters might swamp the party’s individual members in the leadership ballot. After all, in 2010, Ed Miliband had apparently won on the back of such union votes.

The 2015 leadership election was based on a new arrangement in which Labour would elect its leader through a single electorate, made up of three distinct categories of voter. Alongside party members, Labour supporters would be able to register, pay a token fee, and received a vote (so-called ‘sign-ups’). So too would affiliated members from the trade unions who could enrol to get a ballot (‘opt-ins’). Unlike the old electoral college system, ballots would be weighted equally and each participant would be limited to just one.

Critics pointed out that, unlike the old system in which they were capped at a one third share, there was no limit to the number of trade union members who might opt in. In the 2010 Labour leadership contest, well over 2,500,000 ballots had gone out to members of trade unions. Less than 180,000 had gone out to the party’s members. How did participation by such trade unionists, opting in, shape the vote in 2015?

Of course, few people had expected the new format to be put into operation so quickly. But immediately following the May 2015 general election defeat, Ed Miliband stood down and the process of choosing his successor began. The party established a framework in which the new electoral register could be put together while trade unions began the process of getting members to enlist so that they too could take part. Given the simplicity of the procedure, some commentators suggested that, using phone banks, they would be able to get large numbers to opt in, ensuring that trade unionists continued to play a fundamental part in the election process.

A day or so after registration closed on 12 August, Labour reported that the total electorate for the contest comprised something like 610,000 individuals. There had been a massive last minute rush to take part with something like 150,000 registering in the final couple of days. Something like 120,000 had paid the £3 fee to get a vote as a sign-up. Of the remainder, however, less than 190,000 were affiliated members who had opted in (most of whom were trade unionists – a handful coming from socialist societies). The predicted influx had not taken place.

Just about 300,000 of the electorate were paid up party members. Individual membership of the party had risen slightly in the run up to the May general election to around 200,000: since then it had surged forward with an increase of about 50 per cent. So, contrary to some of the predictions, trade unionists did not dominate the new arrangements. Their numbers represented less than two thirds of those of party members, and less than a third, overall, of the electorate.

The figure of 610,000 voters that Labour reported, however, proved to be an exaggeration. Checking through the new list, the party found a sizable number who were ineligible – for example, those who were not on the electoral register, or those who were in arrears for their membership fee. In a relatively small number of cases some were struck off for being supporters of other political parties. At the end of August, the party indicated that the electorate had come down to around 550,000.

The biggest shift was in the number of affiliated members, in effect trade unionists who had opted in. The number of members came down by about 7,000 while the sign ups to the £3 fee moved down by about 10,000. By contrast, press reports indicated that affiliated members had been reduced by over 40,000 to just under 150,000 – largely because a high number of opt-ins were already individual members and received a ballot by that route.

In announcing the result on 12 September, Labour stated that the final electorate had been 554,272. In a turn-out of 76.30 per cent, 422,662 had voted: 245,520 members, 71,546 affiliated opt-ins and 105,598 sign ups. In voting, the ratio of trade unionists to individual members had fallen even further from that announced when registration had closed back on 12 August. Trade unionists comprised less than 30 per cent of size the party’s membership and less than 17 per cent of the total electorate.

Labour did not offer a precise breakdown of the three types of voter who comprised the total electorate. But based on the figures offered at the end of August, which suggested 147,134 trade unionists had signed up, the turnout of affiliated members was meagre: less than 50 per cent of those who received a ballot had returned it. For members and for sign-ups the turnout rate was well over 80 per cent in each case. The trade unions had struggled to register members to vote, many of whom already had ballots: they had found it even harder to get those that had opted in to return their completed voting form.

These statistics suggest that trade unionists came nowhere near playing the kind of role that they had done in previous contests to elect the Labour leader. Between 1994 and 2010 they had had one third of the votes in the electoral college (a small proportion of which went to the socialist societies). In 2015, they compromised just over 30 per cent of the original electorate. In terms of the ballot papers that went out, this proportion fell to around 27 per cent. In terms of votes returned, it came down further to under 17 per cent.

Of course, trade unionists might still have played a decisive role in shaping the outcome of a contest that was especially close. But in 2015 Jeremy Corbyn led decisively among all three groups of voters, getting 49.5 per cent of members first preference votes, and 57.6 of those from opt-ins. Most dramatically, though, he received 83.8 per cent of those votes from the registered supporters. In a sense the trade unionists simply reflected a pattern of support across the labour movement’s grassroots.

The analysis offered here does not mean that trade unions are unable to intervene in the election of the Labour leader. Trade union donations may have played some role in helping candidates to organise their campaigns and canvass for votes across the party. The party’s affiliated trade unions can still nominate individuals: many offered Jeremy Corbyn endorsements during the summer of 2015. But the capacity of such nominations to shape the outcome appears to have been dramatically reduced given the manner in which the ballot was conducted as well as the reduced share of the electorate taken by these affiliated members.

One last point concerns the number of voters. More voters took part in the 2015 leadership contest (over 420,000) compared to 2010 (around 375,000). Neither, however, matches up to the 1994 contest at which Tony Blair was elected. With a far higher affiliated membership at that time, over 950,000 ballots were returned. Given the state of party membership and organised labour, it is unlikely that we will ever again see a contest with such a level of participation. At the same time, however, trade union participants did not shape the result in the manner in which they did five years earlier.

Note: this post originally appeared on the LSE Politics and Policy blog. It represents the views of the author, and not those of Democratic Audit UK or the LSE. Please read our comments policy before posting

Mark Wickham-JonesMark Wickham-Jones is Professor of Political Science at the University of Bristol.