How Starmer has improved Labour's poll results

This article was made possible by our Patreon supporters! Special thanks in particular go to our amazing $5 supporters: Ville Forsman, Bert Rothkugel, Alan Spence, sn, Stephen Kaar, Chloe Hopkins, Alex Wilson, Josuke Higashikata, Harry Jackson, Scott Folan

To support Stats for Lefties on Patreon, visit https://www.patreon.com/leftiestats

---

After Keir Starmer had been Leader for a month, I examined opinion poll averages to see what effect his election had had on Labour's vote share (as well as his own personal ratings). At the time, the effect was notable but not sizeable: Labour had gained slightly in vote share, Starmer had a positive net approval rating, and he was doing better in 'Best Prime Minister' polls than Corbyn.

The overall impression was thus that he had made a promising start, but nothing too unremarkable compared to past Leaders. Yet now, three months on from Starmer's election as Leader, "unremarkable" is the last word I would use to describe the polls.

As the graph below shows, Labour's average vote share in polls and their estimated seat total has risen sizeably since Starmer won. From averaging 29% in March, Labour has risen to 38% in June - a higher share of the vote than Labour won in 2019, 2015, 2010 or 2005 and just 3pts below what the party won within Great Britain in 2017 (41%).

In this article I'll examine how the polls have improved for Labour in terms of voting intention and leadership approval ratings.

Voting intention

In Keir Starmer's first month as Leader (April), there was little-to-no overall change in the polls. The Conservatives remained 20pts ahead in the polls, with Labour averaging 31% of the vote - slightly more than in March, but still less than in the 2019 election.

This has now changed. In June 2020, the Conservatives have fallen to 43% (-8pts since April) and Labour has risen to 38% (+7pts), whilst the Liberal Democrats have risen slightly to 8% (+1pt). The SNP and Greens remain unchanged on 4% each. The Conservatives' lead has been reduced from 20pts to just 5pts in three months, a smaller lead over Labour than in 2019, 2015 or 2010.

My full poll average in June, with changes since April, is as follows:

Tories 43% (-8), Labour 38% (+7), LD 8% (+1), SNP 4% (-), Green 4% (-), BXP 1% (-)

In terms of estimated seats in Parliament, the changes are even more dramatic. In April, I estimated that the Tories' 20pt lead would have resulted in a Conservative majority of 156 seats - the party's biggest Parliamentary majority since 1935. In June, however, this estimated majority has vanished.

I would now estimate that the Tories' 5pt lead would result in a Conservative majority of just 12 seats, a substantial decline from their current overall majority of 80 seats. It is virtually identical to the 10-seat majority achieved by the Conservatives in the 2015 general election.

My full seat estimate in June, with changes since April, is as follows:

Tories 331 (-72), Labour 235 (+66), SNP 51 (+3), LD 9 (+2), Plaid 4 (+1), Green 1 (-), Other 19 (-)

There's no question, therefore, that Starmer has overseen a substantial improvement for Labour in opinion polls. Labour's poll average in June (38%) is the party's highest in a monthly average of polls since August 2018 (39%), whilst Labour's estimated seat total of 235 seats is the party's highest in a monthly seat estimate since July 2019 (251 seats).

Voting intention - Remainers and Leavers

So where is this support coming from? Plenty of Labour supporters, myself included, assumed that Starmer would struggle with Leave voters and would win back far more Remain voters than Leave voters. Yet that isn't what has happened. Let's look at Remain voters first.

Labour 54% (+6), Tories 21% (-6), LD 11% (-1), SNP 7% (+1), Greens 4% (-1)

As you can see, amongst people who voted Remain in 2016 there has been a 6pt swing from the Conservatives to Labour since April. Labour has now risen to 54% amongst Remain voters, which is higher than the 49% of the vote that Labour won amongst Remainers in 2019. It is notable that 54% is just below the 55% that the party won amongst Remain voters in the 2017 general election.

The Conservatives, meanwhile, are averaging 21%, and whilst this is slightly higher than the 19% that the party won in 2019, it is lower than the 24% that they won amongst Remain voters in 2017.

The Liberal Democrats (11%) have largely collapsed amongst Remain voters, losing nearly half of their support amongst Remain voters compared to 2019 (when they won 21%) and even polling slightly below their 2017 result amongst Remainers (12%).

Let's look at Leave voters now.

Tories 67% (-10), Labour 19% (+7), LD 3% (+1), Brexit 2% (+1), SNP 2% (-), Green 2% (-1)

As the graph demonstrates, there has been a sizeable decline in support for the Conservatives amongst Leave voters since April. After averaging an impressive 77% amongst Leavers in April, the Tories have declined by 10pts and averaged 67% in June. This also represents a decline compared to their 2019 result amongst Leave voters (74%) but is still slightly higher than their 2017 result amongst Leavers (64%).

Labour, meanwhile, still has not managed to return to the level of support that it had amongst Leave voters in 2017 (24%). However, its June average of 19% amongst Leave voters is still a sizeable increase compared to April (12%) and the 2019 election (14%).

No other party averaged more than 3% amongst Leave voters in either April or June, so changes for smaller parties amongst Leave voters are not statistically significant.

As you can see, Labour has increased its support by 6pts amongst Remain voters and by 7pts amongst Leave voters - meaning that it has won back more slightly Leavers than Remainers since Starmer became Labour Leader. There has also been a bigger swing to Labour amongst Leavers (9%) than amongst Remainers (6%) since April. This very much suggests that Starmer's history of campaigning for Remain has not alienated Leave voters.

In fact, because Labour started with a smaller number of pro-Leave supporters, the 7pt increase in support amongst Leavers is arguably more significant: the party's support amongst Leavers has increased by 58%, compared to an increase of 13% amongst Remainers. All in all, Starmer is not showing any particular weakness with Leave voters in terms of voting intention.

Approval Ratings

So what about Keir Starmer's individual approval ratings?

In short: they're good.

Very good.

On average in June, 41% of voters approved of Starmer and 22% disapproved, a net approval rating of +19. This is the highest average net approval rating for any Labour Leader since 2007.

Approve 41% (+10), Disapprove 22% (+3), Don't know / neither 36% (-14)

It is difficult to emphasise just how impressive these numbers are. Starmer's 41% approval rating is itself impressive: Ed Miliband never once averaged above 40% approval during the five years that he led the party. Corbyn, meanwhile, only managed to do so in June and July 2017.

Starmer's "net" rating, meanwhile - the difference between his approval and disapproval - is even more impressive. Corbyn only ever managed a net positive rating once: in July 2017 (+4), whilst Miliband only achieved it in September and October 2010 (+12 in both months). Starmer's average net approval, meanwhile, is a stunning +19.

One stat that worried Labour supporters in April was the fact that half of voters didn't know anything about Starmer. It was unclear what these voters would think once Starmer became more widely known. Well, we now have our answer: they like Starmer. The percentage of voters without an opinion has declined from 50% to 36% (-14), but the percentage disapproving of Starmer has barely changed (+3). The percentage approving of Starmer, meanwhile, has increased substantially (+10).

But, as with voting intention, the question remains: where is Starmer's support coming from?

Approval Ratings - Remainers and Leavers

In terms of voting intention, Starmer has won back slightly more Leavers than Remainers since April. However, the same cannot be said of his approval ratings. Let's look at Remain voters first.

Approve 57% (+10), Disapprove 14% (+2), Don't know / neither 29% (-13)

Amongst Remain voters, Starmer's approval started positively (47% approve, 12% disapprove) and has only gotten better. Since April, the percentage of Remainers who approve of Starmer has risen by 10pts, with the percentage who disapprove barely changing (+2). The percentage of Remainers without an opinion has fallen from 42% to just 29% (-13), and most of these voters now have a positive opinion of Starmer.

However, amongst Leave voters, the picture is somewhat different - not bad as such, but not as good.

Approve 30% (+8), Disapprove 35% (+6), Don't know / neither 35% (-15)

Since April, the percentage of Leavers with no opinion of Starmer has fallen from 50% to 35% (-15). However, unlike with Remainers, these voters were more evenly divided in their impressions of Starmer. The percentage approving of Starmer rose by 8pts, whilst the percentage disapproving of him rose by 6pts. This did narrow the gap slightly, however - Starmer's net approval amongst Leavers rose from -7 to -5.

More notable is the fact that 30% of Leave voters have a positive opinion of Keir Starmer. Given the fact that most Leave voters support the Conservatives, and the fact that Starmer was a vocal supporter of Remain, it's impressive that a sizeable 30% of Leavers approve of the Labour Party's leader.

Best Prime Minister polling

One particular problem that Corbyn always faced in polling was that, when asked who would make the best Prime Minister, voters always said the Conservative Leader. Over the five years that he led the Labour Party, the closest that Corbyn came to outpolling the incumbent Prime Minister on this question was in October 2017, when Theresa May had an average lead of 1pt.

Starmer has not yet polled ahead of Boris Johnson on average. But since becoming Leader, he has narrowed the gap between him and the Prime Minister significantly. In April, Boris Johnson had a 24pt lead on the question of who would make the best Prime Minister; that has now fallen to 7pts. In one poll conducted at the very end of June, Starmer polled 2pts ahead of Boris Johnson.

Boris Johnson 41% (-5), Keir Starmer 33% (+11), Don't know / neither 27% (-5)

Amongst Remain voters, Starmer has a commanding lead that has only increased since April:

Boris Johnson 21% (-5), Keir Starmer 53% (+9), Don't know / neither 26% (-4)

Amongst Leavers, however, Boris Johnson has an overwhelming lead. Having said that, Starmer has gained more support amongst Leave voters (+10) since April than amongst Remain voters (+9).

Boris Johnson 61% (-9), Keir Starmer 17% (+10), Don't know / neither 20% (+2)

The net swing to Starmer since April amongst Remain voters was 7pts, whilst the net swing to Starmer amongst Leave voters was 10pts.

In short, whilst Starmer remains behind on the question of who would make the best PM, he is rapidly closing the gap and is doing so at a largely equal rate amongst Remain and Leave voters.

Conclusion

So, let's summarise the key changes since April:

• In voting intention polls, there has been a net swing to Labour of 8pts

• The Tories' average poll lead has fallen from 20pts to 5pts 

• In terms of seats, the Tories' estimated majority has fallen from 156 seats to 12 seats

• Starmer's net approval is now  +19, the highest of any Labour Leader since 2007

• Boris Johnson's lead on the Best PM question has fallen from 24pts to 8pts

It certainly seems as if the more that people get to know Starmer, the more that they like him. Starmer has presided over a dramatic increase in Labour's poll averages and in his own approval ratings.

However, we will have to see if this continues. The trend is very, very positive but Labour still has yet to poll ahead of the Tories in any poll. We will have to see what happens in the months to come, and in the 2021 local elections, to get a better sense of how Starmer has transformed Labour's fortunes. I very much hope that we continue to see improvements.

Will Labour ever support proportional representation?

This article was made possible by our Patreon supporters! Special thanks in particular go to our amazing $5 supporters: Ville Forsman, Bert Rothkugel, Alan Spence, Vincent Calabrese, sn, Stephen Kaar, Chloe Hopkins, Alex Wilson, Josuke Higashikata, Harry Jackson, Scott Folan

To support Stats for Lefties on Patreon, visit https://www.patreon.com/leftiestats

---
 
After several years of sitting on the sidelines following the failed Alternative Vote referendum, the issue of electoral reform has returned to relative prominence in the Labour Party. During the 2020 leadership campaign, Keir Starmer promised to hold a constitutional convention, and argued that "we’ve got to address the fact that millions of people vote in safe seats and they feel their voice doesn’t count". A poll from YouGov, meanwhile, showed that 76% of Labour members support proportional representation.

In this article I'll explore Labour's history with electoral reform, consider the likliehood of Labour endorsing proportional representation (PR) and look at what the results of recent elections would have been if Britain used PR to elect its MPs.

Background

The Labour Party has a complicated history with electoral reform. Even when it was a minority party with less than 50 seats in the House of Commons, the party was unconvinced by PR; in January 1914, the party's annual conference rejected PR by an overwhelming margin (66% to 34%) and rejected AV by a similar margin (1).

Labour did eventually endorse AV during the 1917 debates around electoral reform and came very close to seeing it enacted - however, AV was ultimately dropped from the Representation of the People Bill in February 1918 and it was never introduced (2). Labour proposed introducing AV after the party formed a minority government in 1929; however, the Bill failed again (3). Labour lost office in 1931 and did not return to power for another fourteen years, by which time electoral reform had completely dropped off the agenda of any Labour government.

Between 1930 and the 1990s, Labour showed little to no interest in changing the First-Past-The-Post (FPTP) voting system, even though it lost office in 1951 largely as a result of the electoral system - Labour won the popular vote in 1951 by over 200,000 votes, but the Conservatives won a majority of seats!

Despite this, it was only in the 1990s that electoral reform once again became a serious consideration for Labour. After the 1997 landslide, Labour introduced PR for European elections, used PR to elect the new devolved assemblies and introduced a new preferential voting system (the Supplementary Vote) for local and regional Mayors. However, PR was never introduced for general elections, and none of Blair's successors (Brown, Miliband and Corbyn) showed any particular enthusasism for proportional representation.

AV, however, was a different question entirely. In its 2010 manifesto, Labour made its first unambigous endorsement of electoral reform since the early 1930s when it promised that it would hold a referendum on the Alternative Vote if it won a fourth term. Although the party lost, a referendum on AV was subsequently held in 2011 by the Conservative-Lib Dem coalition. Labour officially remained neutral, but Ed Miliband campaigned for AV in the referendum. AV, however, lost in a landslide.

It has now been nearly 90 years since Labour last made a serious attempt to reform the electoral system. After losing four elections in a row, it is not surprising that electoral reform is once again being discussed by Labour. But will Labour actually support it?

Will Labour support PR?

One important thing to remember about Labour that trade unions have significant influence within the party's policy-making process. This is why the polls showing that members support PR are not as significant as one might think. Trade unions control 50% of the votes at the party's annual conference, as well as 34% of seats on the National Executive Committee (NEC), meaning that trade union support is vital if any policy is to pass. This is not a given: according to pro-PR group Make Votes Matter, GMB and Usdaw both support FPTP. These two trade unions alone hold 4 of the 38 seats on the NEC and a significant number of votes at the party conference.

In order to succeed, any electoral reform policy would need the support of:

• a majority of NEC members (to ensure that the policy appears on the agenda at the party conference)

• a majority of delegates at the party conference (50% of whom represent members, and 50% of whom represent trade unions) 

• the Labour Leader (to ensure that the policy is implemented when Labour wins)

The NEC

By my estimate, of the 38 NEC members, 7 are opposed to proportional representation: CLP rep Jon Lansman, the USDAW and GMB reps who will vote in line with their union, Richard Leonard MSP and Margaret Beckett MP.

Just 2 members of the NEC are supportive (Starmer and Jonathan Reynolds MP). The remaining 28 have no clear opinion. 20 votes are needed for a majority on the NEC, so supporters of PR would need to convince 18 other NEC members to either support PR or at least allow conference to vote on the issue.

Conference

In any vote at conference, the votes of members and trade unions are tallied seperately. The two sections then make up 50% of the overall vote. This means that even if delegates from Constituency Labour Parties (CLPs) overwhelmingly vote for a proposal, it can still fail if trade unions overwhelmingly vote against it.

In 2018, for example, one vote saw 89% of CLP delegates vote in favour of a proposal, but 91% of trade union votes were cast against the proposal. The overall result was thus 49% for, 51% against. Winning the support of trade unions is thus absolutely crucial to successfully passing policies at Labour Party conference.

Will it ever happen?

The probability of Labour endorsing proportional representation has risen recently, because the result of the 2019 election means that Labour will need to win 124 constituencies in 2024 that were won by other parties in 2019 (mostly by the Conservative Party). That's more-or-less impossible. Labour has gained over 100 seats just twice since the Second World War: 147 gains in 1997 and 239 gains in 1945.

It is, however, entirely possible for Labour to form a minority or coalition government in 2024. In that context, smaller parties like the SNP, Liberal Democrats, Greens and Plaid Cymru are likely to demand PR as part of any coalition agreement. Given that many people in the Labour Party believe that a majority government is out of reach for the party for the foreseeable future, it is not inconcievable that the party could agree to introduce PR in order to prevent the Conservatives from winning a majority in the future.

What would British elections look like under PR?

It is difficult to know exactly what the results of past elections would have been if Britain had used PR. Voters would vote differently under PR, meaning that smaller parties might have won more votes than they actually did in previous elections that used FPTP. Not only that, but there are loads of different proportional voting systems: the Additional Member System (AMS), the Single Transferable Vote (STV) and party list proportional representation (List PR) have all been used just in the UK alone for different elections at some point over the past five years.

However, I'll try in this section to give you an approximation of what the results might have been.

I'll be showing you what the results would have been if Britain had used List PR, as that's the system that we used for European elections from 1999-2019. I've used the regions and nations of the UK as the multi-member constituencies, as those are the constituencies that were used in European elections.

In 2010, the seat result would have been:

In 2015, the result would have been:

In 2017, the result would have been:

In 2019, the result would have been:

These results would have allowed for the formation of Conservative/Lib Dem coalitions in 2010, 2017 and 2019. Equally, they would have allowed for the formation of Labour/Lib Dem coalitions in 2010, 2017 and 2019 - although in 2017 and 2019 the Labour/Lib Dem coalition would have also needed the support of the Greens and the SNP.

----

Citations

(1) Pugh, Margin, Speak for Britain: a New History of the Labour Party. Vintage Books, London, 2010. pp 83-84

(2) Pugh, Margin, Speak for Britain: a New History of the Labour Party. Vintage Books, London, 2010. pp 121-122

(3) Pugh, Margin, Speak for Britain: a New History of the Labour Party. Vintage Books, London, 2010. pp 210-211