Analysing the May results 2: Winning where it matters, losing where it hurts

The return of three voting patterns that will worry Conservatives, and encourage Labour

The governing party seemed to take heart from local council results this month, despite losing nearly half of the seats they were defending. The source of the Tory morale boost was a projection produced by Professor Michael Thrasher and Sky News which suggested Rishi Sunak might still defy the polls and secure a hung parliament at the next general election.¹ While this projection was not a forecast of the general election to come, it did put the spotlight on what appeared to be a rare piece of bad news for Labour and hope for the Tories - Labour was not matching the vote shares or leads achieved by past oppositions on the brink of a big election win. Perhaps the lead is soft? Perhaps Labour haven’t sealed the deal? Perhaps the election may still be close? Perhaps, perhaps, perhaps…

The attraction of this narrative is obvious. Media pundits are now bored by big stable Labour poll leads, and Tory MPs are terrified by them. Both are eager to seize upon any scrap of evidence telling a different story. And this is, indeed, one such piece of evidence. Labour’s performance on both Sky & Thrasher’s National Equivalent Vote (NEV), and John Curtice/BBC’s Projected National Share (PNS) is middling (see table), and below what we would expect from an opposition with huge poll leads. Labour’s lead over the Conservatives in the BBC’s Projected National Share has grown by just 4 points in the last 2 years (and hasn’t increased at all in the past year), even as the oppositions’ poll lead has surged by 16 points in the same period.

This is not a good performance for the opposition, historically speaking. But it is just one grain in a large dune of data. And the rest of the sand is blowing in a different direction entirely.

Parties’ local election performances on BBC Projected National Shares since

Source: BBC

There are two big reasons Labour are still on course to win big despite a historically mediocre PNS/NEV performance - fragmentation and efficiency. When flagging Labour’s low headline vote share Conservatives tend to avoid drawing attention to their own dismal showing. The Conservative share on the PNS this year equalled the lowest ever recorded. Voters emphatically rejected the Conservatives at the polls in local elections, as they did in all the other elections held this month. Three quarters of voters rejected the governing party. They just disagreed on who to install instead. There were strong performances by the Liberal Democrats, the Greens and by a gaggle of small local and independent groups - including various Gaza protest candidates. None of this is encouraging for the Conservatives. The Lib Dems are their main challengers in many seats, Green voters are very clearly more inclined towards Labour than the Conservatives, and while Gaza protest votes will cause a few Labour incumbents headaches, they won’t shift the dial in the general election battleground. A fragmented anti-Tory vote is still an anti-Tory vote.

The news from the general election battleground is far more encouraging for Labour. Three big voting patterns found in last year’s local elections - bigger Conservative to Labour swing in Leave areas; Conservatives doing worse where they start strongest; and tactical support for the locally strongest challenger in Conservative held seats - all recurred this year.

Labour have a mountain to climb at the general election. To gain the 150 seats needed for a healthy majority, they need to sweep the board in Leave leaning areas where Tory targets are concentrated. They are. They need to secure bigger swings where the Conservatives start with larger majorities. They are. And they need the smaller parties behind them in their target seats. They are. Winning where it counts, hitting the Conservatives most where it hurts most, rallying support where it counts - this is what efficient voting looks like. If Labour can repeat these tricks at the general election, they will be on course for a big win even if the polls narrow somewhat, and a massive win if their poll lead holds up.

Efficient voting 1: Bridging the Brexit divide

As I discussed last year, the 2021 local elections were a “peak Brexit” contest - the strongest expression in local elections of the “get Brexit done” coalition assembled by Boris Johnson, with big swings to the Conservatives across the broad terrain of Leave England, but much weaker performance in the urban and suburban islands of the Remain archipelago. Some of the many seats won then were up for election again this month (the seats where COVID delayed the 2020 elections by a year) so we can look directly at the damage done by three years of scandal, crisis and incompetence to the Brexit coalition Johnson built, and how far Labour have been able to reach across the Brexit divide.

In figure 1, we look at the pattern of Conservative and Labour change, and Conservative to Labour swing, in the 800 plus wards of the BBC keywards dataset by estimated local vote share for Leave in 2016. The message here is clear - the more a ward voted to Leave, the more it has swung from Conservative to Labour since 2021. The most Remain areas saw only a 2 point swing, but in places which voted 60% plus for Leave, the Conservative vote slumped by 14 points, while Labour’ vote was up 6 points - a 10 point swing.

Figure 1: Conservative 2021-24 change, Labour 2021-24 change and Conservative to Labour swing, by 2016 estimated Leave share

Source: BBC elections keywards 2024 dataset²

Another useful feature of local elections is that they are conducted in “thirds” in many places - all wards elect one councillor for a four year term three years out of four (with no elections in the fourth year). As a result, we can look at how the pattern of Tory to Labour swing behaves from different starting points, as we do in Figure 2. The biggest shift to Labour in the most Leave areas happened between 2021 and 2022 - while the average swing since May 2021 is 10 points, the swing since 2022 is more like 3-4 points. In May 2021, Boris Johnson was in his pomp. By May 2022, he was on the brink of being ousted. The decline of the Tories in Leave England coincides with the Johnson’s decline in Downing Street.

Tory decline in the most Leave areas has continued after Johnson’s departure, though at a slower pace. The swing since May 2022 is around 3-4 points, and the swing since May 2023 is another 1 point or so. This suggests the brief and disastrous reign of Liz Truss cost the Tories another couple of points of local election vote in Leave areas, and while Sunak has slowed the Tory decline in Leave England, he has not reversed it.

Figure 2: Conservative to Labour swings 2021-24, 2022-24, and 2023-24

Source: BBC elections keywards 2024 dataset

Efficient voting 2: Proportional swing

The easiest way for Labour to turn modest overall swings into large seat gains is by achieving proportional swing - that is, by making bigger gains where they need bigger gains, and smaller gains where gains aren’t needed. As Figure 3 illustrates, this is exactly what we saw in this year’s locals, just like last year. The stronger the Conservatives started in a ward, the further they tended to fall, with Labour the biggest beneficiary. In wards where the Conservatives started under 20%, they fell by an average of 5 points. In wards where the Tory incumbent started with 50% or more of the vote, they fell by an average of 17 points, with Labour rising by 7 points, a 12 point swing.

If this were replicated in a general election, Labour will see little gain in the seats it already holds, with its advance concentrated instead in the seats it is targeting. Such a pattern would dramatically improve the efficiency of the Labour vote, and reduce the swing Labour would need to achieve a majority.

Figure 3: Conservative 2021-24 change, Labour 2021-24 change and Conservative to Labour swing, by 2021 Conservative vote share

Source: BBC Keywards dataset 2024

Efficient voting 3: Tactical co-ordination

Proportional swings can be driven by targeted Labour advance, or it can reflect a widespread anti-incumbent mood, driven by a strong desire to remove Conservative incumbents driving larger swings in Tory held places. Another way this “throw the rascals” out sentiment can manifest is tactical voting, with Labour, Lib Dem and Green voters coalescing behind whichever challenger candidate is best placed to remove the local Tory incumbent. We observed this last year, and it happened again this month, as figure 4 illustrates. Labour did much better (and the Lib Dems and Greens worse) in wards where it was challenging the Conservatives than in wards it was defending from the Conservatives. Labour gained an average of nearly 9 points where it was challenging a Tory incumbent, with tiny rises in Lib Dem and Green vote. In wards with a Labour incumbent, the Labour rise was just one point, with a three point rise for the Lib Dems on average, and nearly four points for the Greens.

The Lib Dems benefitted from the same pattern. The party did much better (and Labour and the Greens worse) in Conservative vs Lib Dem wards than in Lib Dem vs Conservative wards. Lib Dem candidates challenging a Tory incumbent gained nearly six points on average, Lib Dems defending against a Tory challenge gained just two points. Labour support rose 2 points in Con vs LD wards but 4 points in LD vs Con wards. The Green vote rose 3 points in wards with a Lib Dem incumbent, but fell by an average of 1 point in seats where the Lib Dems were challenging a Tory incumbent.

In each case it seems voters were coming together in an effort to oust Tory councillors. Such tactical voting, particularly when combined with proportional swing, helps to explain why the Conservatives lost nearly half of the seats they defended - one of the worst “loss rates” ever recorded - despite a modest Labour performance and a fragmented opposition vote nationally.

While voters are fragmenting more than ever, one sentiment unites them: they want to kick Tories out, and will come together to achieve this wherever Tory incumbents are present. If such sentiments are strong enough to drive co-ordinated behaviour in low stakes, low turnout local elections, we may be in for a lot more of the same in the general election.

Figure 4: Parties’ 2021-24 performance by local tactical context

Source: BBC elections keywards 2024 dataset

Winning where it counts

Here, then, are three big reasons why the picture painted by aggregate projections of this year’s locals is a mirage. The Tory vote in Leave England is slumping. Labour is hitting the Tories hardest where it hurts them most. And voters from across the political spectrum are coming together to vote for anyone capable of defeating a Tory candidate locally. There are mixed messages from this year’s contests for the Labour opposition, which I hope to cover in future posts, but for the Conservative government the message is crystal clear. Voters want them out, everywhere, by any means necessary. That mood is as strong as ever and time is running out to change it.

1. Those excited about this forecast mostly ignored that a hung parliament would still see the Conservatives out of government, as they have no friends in the Commons to prop them up.

2. My thanks to Sir John Curtice, Steve Fisher, Patrick English, and the excellent BBC elections team headed by Julia Walker for help and support in developing and analysing this dataset. The views in this post, and any errors, are my responsibility alone.