Turnout tests

First published in Progress magazine

Is the notion that boosted turnout will turn Labour’s fortunes around simply outlandish? Richard Angell examines its chances

We had a problem of mobilising people’, said leadership hopeful Jeremy Corbyn – with little chance of getting on the ballot, let alone winning – as he explained why Labour lost when Progress sat down with him on 11 June 2015. A lot has changed since that interview, not least the reluctant nominee receiving a stonking 59 per cent of the first preference vote on 12 September.

But the argument rehearsed by the now leader – that Labour loses because when it is insufficiently left voters who would otherwise be for Labour stay at home – is not new. In July 2012, Jon Trickett and others created a grouping called ‘Five Million Votes’. It is long gone now but the argument behind it is not. It tried to argue that New Labour ‘lost’ five million votes in office, 1.5 million – more than the million to the Tories and a higher proportion to the Liberal Democrats – to the couch. They simply stayed at home. YouGov president Peter Kellner pointed out in the November 2012 edition of Prospect the painful truth that the numbers were not true. In fact, 3.5 million of those who voted Labour in 1997 had died by 2010. Regardless, the argument was maintained that if a leftwing programme could be enthusiastically put to the British public, turnout would surge and Labour would be the beneficiary.

Keen to dispel the idea that the then leader’s office was entertaining a ‘35 per cent strategy’, Marcus Roberts in his Next Majority pamphlet for the Fabian Society in 2013 proposed a ‘40 per strategy’. The turnout of non-voters was practically all the difference between the two approaches. Roberts argued that three per cent would come from first-time voters, two from returning non voters. In doing so he signalled a huge note of caution: non-voters are simply not very leftwing. He wrote, ‘In the case of older non-voters who sat out the 2010 election, Labour will need to deftly handle their concerns on tough issues like immigration and welfare. These voters also worry about Labour on spending’.

The 30.4 per cent of the vote achieved by Ed Miliband in May 2015 might have put this argument to bed. In factpost-election research by the Trades Union Congress proved Roberts’ point: non-voters and voters who did not vote Labour alike refused to support Miliband’s party because of worries on spending and welfare.

Did this stop anyone arguing otherwise? Clearly not. Straight after the election, the party’s Twitter left launched its ‘76’ Twibbon – the percentage of the total electorate who had not voted Conservative. The idea of both uniting the left and turning out undecideds was back. Conor Pope of LabourList was straight on the case: ‘Any plan for 2020 that does not have at its heart a primary focus on winning over [previous] Tory voters is a blueprint for a third successive defeat. While simply trying to build an “anti-Tory coalition” may have been convincing in a landscape that suggested the Conservatives were no longer able to win majorities, we know now that that is not the case.’

For those who still think elections can simply be won for Labour by increasing turnout I add some words of caution. First, the facts simply do not suggest this. Labour last won an election – in 2005 under ‘He who shall not be named’ – on a 61.4 per cent turnout. It lost in 2010 and 2015 on higher turnouts – 65.1 and 66.1 per cent respectively – under leaders to the left of their predecessor. Second, low-turnout seats are disproportionately Labour seats. Ninety-two of the 100 lowest-turnout seats are already held by Labour, 95 in England and Wales. Increased turnout in those seats does nothing at all to contribute to a Labour majority or Corbyn victory. Finally, I would suggest the surge in turnout in Scotland in May – back up to 1997 levels – was not good for Labour. While many want to suggest the Scottish National party was elected on an ‘anti-austerity’ platform, it should be pointed out it planned to cut more than Ed Balls and has run schools worse than Michael Gove.

For Miliband to have won the popular vote in 2015 by galvanising non-voters alone, the voting public would have had to have been two million higher with turnout some 70.4 per cent. This is near 1997 levels and under the argument set out by the new leader and the team behind Momentum what we should be looking to expect.

Next year Labour will face the voters at the ballot box in elections for the Scottish parliament, the Welsh assembly, police and crime commissioners, local elections across England and Wales, and the London mayor and London assembly. What increases in turnout might be reasonable to expect?

At the equivalent set of local elections, the 31 per cent turnout in 2012 was – coincidentally – some four percentage points below the same elections in 2008. An increased turnout should not be difficult to achieve, but will this help Labour? Labour took 38 per cent of the estimated national vote share, the Tories took 31 per cent in 2012. It made 823 gains and Harriet Harman said it showed the party was ‘back in business’. These seats were last contested in the wake of the omnishambles budget; with Labour succeeding in creating the same outrage over tax credits, holding these metropolitan boroughs should be a given. Corbyn and his team will be looking to make gains and push up turnout as these places will be high in the demographic susceptible to an anti-austerity message – urban, working class, disproportionately working in the public sector, increasingly affected by cuts to public services and with high numbers of minority ethnic voters. If there is no Corbyn motivation in these electors, there are only worrying times ahead for Labour.

Labour goes into the London election next year with a voter pool some 300,000 votes larger than the Tories’ based on 2015 figures. Even though this was not enough to win the remaining two‑thirds of London’s target seats, in a mayoral race where only 185,000 second preferences were cast in 2012 this should be enough to make Sadiq Khan the outright winner in May. In theory, then, a higher turnout only helps Labour in the capital. Khan has none of the brand problems of Ken Livingstone – his current role as a parliamentarian means his earnings are public and he has a relationship exponentially better than Livingstone’s with London’s Jewish community, for example. The former mayor lost by just 62,538 votes. Had Labour increased turnout among Londoners by 1.1 per cent – with all those votes going to Livingstone – Labour would have won back the mayoralty three and a half years ago. Assuming Zac Goldsmith gets the same primary vote as his predecessor but more favourable second preferences, Labour would need to increase turnout just 1.4 per cent – and have all those for Labour – for the party to stay on top this coming May. Naturally, not all the votes derived from increased turnout would be Labour’s anyway, as Roberts warned. But considering Corbynmania’s London roots, it should not be impossible for Khan to expect the leader he nominated to give him a good surge in Labour votes.

Even though Labour lost two seats to the Tories in Wales in May 2015 and failed to win low-hanging fruit, Milband increased the party’s vote from 36 to 37 per cent. The same upturn in Labour support should have great dividends for Labour under the additional member system; any increased turnout in ‘safe seats’ could offset on the list any losses in the constituencies. If Corbyn’s focus on turnout is going to be greater than Miliband’s, he must be able to do the same, if not better, than his predecessor. Doing so will be handsomely rewarded as Labour seeks to hold its majority in the assembly and demonstrate Labour is on the march west of Offa’s Dyke.

Scotland, however, is a whole different ball game. Turnout since the referendum is historically high and has been punishing for Labour. In the 2011 Scottish parliament elections, the SNP took 45.4 per cent of the vote, Scottish Labour 31.7. In the general election, the SNP received 50 per cent of the vote and Labour just 24.3. If Labour wants to win back Scotland at the next general election then it needs to be winning people back from the nationalists now. The worst-case scenario is that Labour is still on a downward trajectory. Can Corbyn and Kezia Dugdale keep the SNP under 50 cent and regain at least half the vote lost between 2011 and 2015? Minimum expectations should be a Labour vote at 28 per cent. Anything worse than this would be confirmation that a recovery is not coming.

But in Scotland the election will come down to seats, not percentages. According to Scotland Votes, run by Weber Shandwick Public Affairs, 28 per cent would see Labour drop from 15 to just five constituency representatives in Holyrood and the overall Labour group one fewer than 2011. A similar change in our list vote would see Labour return just 33 Scottish parliamentarians altogether. The same percentage of the vote would gain just one additional seat – East Renfrewshire – at the 2020 general election. This really should be the least we can hope for.

By 2020, a four-point turnout bump in Labour’s benefit does not seem all that much to hope for, but clear progress towards this must be in evidence long before then. By failing in the run-up to May 2015 to dissuade voters in target seats from plumping for the Conservatives rather than Labour, we let down the nine million people nationally who voted Labour and were depending on it winning. Relying on an electoral strategy without signs it is working is too big a gamble and risks again condemning the most vulnerable to yet five more years of Tory rule.


Richard Angell is director of Progress

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s