It is perhaps not immediately obvious that local elections might be of interest or relevance to the Brexit process. After all, such elections are meant to be about who can best empty the bins rather than who can cut the best deal with Brussels.
However, in practice, the broad ups and downs in party performance in local elections tend to reflect the national popularity of the parties. And given the prominence of the debate about Brexit in our national politics, we therefore cannot discount the possibility that Brexit might have had an influence on how people voted in the local elections.
After all, Brexit certainly played a role in shaping the dynamics of party support in last year’s general election. Support for the Conservative party increased between 2015 and 2017 amongst Leave voters, whereas it fell amongst those who voted Remain.
Conversely, though the relationship was not as strong, Labour advanced more amongst Remain voters than it did amongst their Leave counterparts. These patterns were also reflected in the geography of party performance.
The Conservative party gained most ground in areas that voted most heavily for Leave but fell back in constituencies where support for Remain was highest. Labour, meanwhile, saw its vote increase more in Remain voting areas than in places where Leave secured most support.
So, a key question about the local elections is whether a similar pattern can be discerned. Do the results confirm the evidence of the 2017 election that the Brexit divide has become more closely aligned with whether people vote Conservative or Labour?
Here we address that question, relying on a collection of the detailed voting figures in just over 900 wards that were collected on Thursday night and Friday by the BBC.
The seats up for grabs in this year’s local elections were in the vast majority of cases previously fought over in 2014, on the same day as the European Parliament elections. Crucially, this was long before the EU referendum as well as last year’s general election. Consequently, the baseline against which the parties’ performance was being measured was unaffected by voters’ reactions to the Brexit process.
Therefore, if Brexit influenced how people voted the same way as it did in last year’s general election this would mean that the ups and down in party support should have been different between Remain and Leave voting areas even if Brexit was playing no greater a role in shaping how people voted than it did twelve months ago.
Table 1 shows the average change in party support since 2014 broken by the level of support for Remain in the council district in which each ward was located. It is confined to those wards which were fought by the Conservatives, Labour and the Liberal Democrats in both 2014 and 2018.
It is immediately apparent that whereas on average the Conservative vote flatlined in areas that voted most heavily to Remain, it increased by almost ten points where Leave had performed strongly.
There seems to be little doubt that the increased concentration of Conservative support amongst Leave voters that was in evidence in last year’s general election has been replicated in the local elections.
One reason why this was the case is quite clear. Back in 2014 UKIP were riding high. That support disappeared in these local elections, just as the party’s vote collapsed in the 2017 general election, not least because in many wards the party failed to nominate a candidate.
The table shows that the level of support for all other parties combined, including UKIP, fell more heavily in areas that voted strongly for Leave than it did in those that backed Remain. In 2014 UKIP had performed best in places which went on in 2016 to vote most heavily for Leave. Consequently, the fall in UKIP (and thus Other) support in these local elections was greatest where UKIP had previously been strongest, and thus in Leave voting areas.
That meant there was greater scope for all of the parties to register an increase in their support in Leave voting areas. But, on average, only the Conservative party seems to have been able to take advantage of the fall in UKIP support. Neither Labour nor the Liberal Democrats advanced more strongly in the local elections in places that voted heavily for Leave.
This represents strong circumstantial evidence that, as compared with 2014, the Conservatives were more successful than their rivals in garnering the votes of those who four years ago had been part of the UKIP bandwagon, and that this was an important element in the party’s relative success in Leave voting areas.
That said, in contrast to the swing between the 2015 and 2017 general elections, Labour’s performance was not markedly worse in Leave voting areas than in those that voted heavily to Remain. So, there is little evidence here of Labour’s vote having become more orientated amongst Remain voters.
Most likely this is because although in general the Conservatives were the disproportional beneficiaries of UKIP’s decline, they were not exclusively so.
In particular, Labour also seem to have profited from UKIP’s difficulties in those wards where Labour had previously been strong. On average Labour’s vote increased by 8.4 points in wards where the party had shared first and second place with UKIP in 2014, compared with 5.3 points in wards where the battle last time had primarily been between UKIP and the Conservatives.
Meanwhile, if we look at how the parties’ share of the vote in this year’s local elections compares with what they achieved in the 2015 local elections (held on the same day as the 2015 general election), we find that Labour’s support did increase somewhat less in areas that voted heavily for Leave, even though, once again, support for Other parties fell more heavily in such wards (see Table 2). At the same time, on this measure too, the Conservatives performed much better in places that voted strongly for Leave than in those that backed Remain.
So, as in the 2017 general election, Leave voting areas – and thus most likely Leave voters – swung to the Conservatives in this year’s local elections, not least as a result of their relative success in winning over former UKIP supporters.
In Labour’s case the evidence that the party might have advanced more amongst Remain voters is less strong (just as was the case in 2017), not least because it too sometimes benefitted locally from UKIP’s collapse, though there is some evidence of such a pattern.
In any event, Labour’s loss of seats in some Leave voting areas is clearly more a reflection of the relative strength of the Conservative advance in such areas than a manifest weakness in Labour’s own performance in such circumstances – but, of course, under first past the post it is relative performance that matters.
But this still leaves us with an important question. Is the link between party performance and the outcome of the Brexit referendum any stronger now than it was in 2017? Does the pattern of voting in the local elections simply represent a replication and continuation of the behaviour that was already in evidence twelve months ago?
Or does it provide reason to believe that the link between Conservative support and voting Leave has intensified further as a result of how voters have responded to the Brexit negotiations so far? That seems to be what some Brexiteers think has happened.
To address this question, we have to adopt a rather different approach. The county council elections that took place last year were, for the most part, held in different parts of England and used different ward boundaries than this year’s local contests.
There is thus limited opportunity to compare the pattern of voting in the two ballots directly. In any event, those county elections were held five weeks before the general election and the balance of party support changed significantly in the interim.
However, in some instances it is possible to aggregate the outcome of the 2018 local elections into complete parliamentary constituencies and compare the total tally for the parties with the outcome of the general election locally in 2015 and 2017. Table 3 shows, for just over 50 constituencies where it is possible to undertake this aggregation, the mean difference between the level of party support in the local elections and the level in both the 2015 and 2017 general elections.
One key pattern is apparent. As compared with 2015 the Conservatives performed better in heavily Leave voting areas than in predominantly Remain voting places. The party’s vote held up in the former, whereas it was eight points lower in the latter. However, this pattern is not in evidence if we compare the outcome of the 2018 local elections with what happened in 2017 – indeed, if anything, the fall away in the Conservative vote since last year was rather greater in Leave areas than Remain ones.
There is therefore no reason to believe that the outcome of the local elections represents any kind of endorsement by Leave voters of how the government has been handling the Brexit negotiations during the last twelve months.
The local elections do not, then, provide any evidence that the link between voters’ attitudes towards the EU and their willingness to vote Conservative has strengthened further. But they do suggest that the reshaping of the electoral terrain that became apparent in the 2017 election was not a short-lived development.
The elections confirm that the Conservative party now finds itself supported by a predominantly pro-Leave electorate, and that the party thus has an electoral incentive to deliver a Brexit that meets the aspirations of Leave voters. During the next few weeks and months we will see how the Prime Minister deals with that challenge.
By Professor John Curtice, senior fellow at The UK in a Changing Europe. This piece originally featured on the WhatUKThinksEU.