Britain’s Quiet Electoral Earthquake
Jane Green and Marta Miori trace the impact of Britain’s ‘bloc equilibrium’ — two rival electoral coalitions whose shifting loyalties delivered Labour’s win last year but make holding power far harder than winning it.
On 23 October 2025, the voters of Caerphilly overturned more than one hundred years of Labour dominance by electing a Plaid Cymru member of the Senedd in a by-election widely expected to be won by Reform UK. The combined vote share for Labour and the Conservatives hit a proverbial electoral floor, with two insurgent parties battling it out for first place.
At first glance, this looks like another volatile and unprecedented outcome of British politics, suggesting that one can roll the dice and almost anything can happen. This would be a mistake. The result, and others like it, can be understood through the underlying structure of what is happening to the UK electorate and its elections. Volatile does not mean unpredictable – and even less so ‘impossible to make sense of’ – if we consider how voters are structured into two, relatively stable, ideological blocs.
Britain’s two-party system has long reflected an ideological divide between the political left and right. Since Brexit, the nature of this divide has shifted to accommodate ‘cultural’ values and attitudes towards Europe, but these broader sets of preferences still cluster into two electoral camps – or, in more academic language, ‘blocs’: sets of ideologically similar parties that voters are most likely to consider choosing between.
In this piece, we explain how the bloc-based nature of vote choice in Britain helps us understand two rapidly moving parts of the country’s electoral landscape. The first, already well established, is the relationship between electoral blocs and changing vote shares. Voters may be increasingly looking to a greater choice of parties, but changes in voting intentions follow a structured pattern of switching within ideological groups. Using British Election Study (BES) data, we show that changes in voting intention since the 2024 general election have largely followed this bloc dynamic. Labour have mostly lost voters to other parties on the left and to the undecided, and this pattern holds in both their more left-leaning and right-leaning constituencies. Within the right-wing bloc, Reform UK have principally gained support from former Conservatives and, to a lesser extent, from non-voters. These patterns highlight the constraints on where each party can realistically hope to draw new support ahead of the next local and general elections, and suggest that party vote shares may be bounded by bloc-specific ceilings.
The second part, equally important but less often discussed, is the link between party blocs and how vote shares translate into seats. The bloc nature of party support in Britain means that party fortunes can be swayed as one bloc becomes more unified and the other fragments. Our previous research has shown that the asymmetry between greater tactical voting on the left and the split on the right allowed Labour and the Liberal Democrats to overturn large Conservative majorities on modest vote shares in 2024 (Miori and Green 2025a). Crucially, changes in this ‘party bloc equilibrium’ can impact party fortunes without substantial transfers of votes from one bloc to the other.
Since 2024, this balance has shown signs of changing in understandable ways: the left-bloc has become more fragmented, and the right-bloc has continued to split further, with Reform gaining support. Labour fortunes are likely to hang in this changing balance. On the right, a more unified vote behind Reform threatens Labour majorities in the many seats the party won in 2024 as a result of the Conservative split towards Reform and the potential for tactical voting between the two right-wing parties. On the left, increased fragmentation lowers the threshold needed for Reform to overtake Labour in seats where the party is in contention. In more extreme cases, it can also lead to Labour no longer appearing as the default, viable option to defeat the main party from the right-wing bloc, even in seats where they currently hold a majority. This, as seems to have been the case in Caerphilly, can culminate in voters in the left-bloc rallying behind a different party altogether.
Electoral blocs and changing vote shares
Much like elsewhere in Western Europe, the bloc structure of party support in Britain has persisted despite rising electoral fragmentation. As voters splinter across a broader range of smaller parties, switching tends to take place within ideological blocs. Griffiths et al. (2025) showed how the relationship between age, education and vote choice sharpened in Britain after the EU referendum, and how these relationships persisted despite higher fragmentation in 2024 when we see the electorate as representing two party blocs. Within the left-bloc, pooled Labour, Liberal Democrat and Green party vote shares are strongly predicted by younger age and higher levels of education. On the socially conservative right, Reform UK and Conservative votes attract the support of older voters with lower levels of education. These two blocs still correspond to attitudes over Brexit, even as the ‘B-word’ has been absent from most political lexicons for some time.
Fragmentation within the ‘left bloc’
This bloc structure helps us understand changing shares of party support since the last election. Firstly, there has been significant splintering within the left-bloc. Despite the media and political focus on Reform UK, Labour’s current vote losses to Reform are small compared with those they are losing to other parties within the left-liberal bloc, and to the undecided. According to BES data, which tracked the same voters from 2024 and how they intended to vote just after May’s 2025 local elections, 18.9 per cent of 2024 Labour voters said they were undecided by 2025, 9.4 per cent would vote Liberal Democrat, 8.3 per cent would vote Green, and just 7.9 per cent said they would vote for Reform UK.
Importantly for Labour, the broad splintering of their support towards other parties on the left is replicated across their more right-leaning ‘heartlands’. The BES has very large samples of respondents – around 30,000 per online survey – which can be broken down into ‘Brexit area quintiles’. This is shown in Figure 2 below, where the proportions of Labour’s 2024 voters who switched to Reform across Leave share in the constituency are in turquoise, those who stayed with Labour are in red, the undecided and ‘will not vote’ categories are in grey, and ‘others’ in black. The pattern is similar across types of area, with only slightly greater switching to Reform (11.9 per cent) among 2024 Labour voters in the most Leave-leaning constituencies, compared to 7.7 and 8 per cent in the third and fourth quintiles.
Because Labour’s vote declined in its heartlands over a long period and increased in more metropolitan areas, the proportions of its 2024 vote were broadly similar across the country. While 26 per cent of Labour’s 2024 vote was in the first quintile, 16.5 per cent was in the second, 17.3 per cent in the third, 19.5 per cent in the fourth, and 20.7 per cent in the fifth, making the above pattern of splintering remarkably similar across the country.
To understand how Labour’s recent vote is more likely to fragment left, and how that pattern is similar across the country, we show in Figure 2 the percentage of Labour voters in each UK general election since 2015 who voted to Leave in the 2016 EU referendum (light-blue bars), as well as the percentage of voters who still supported Leave in the election in question (dark-blue bars).
The downward trend in the light-blue bars, visible everywhere but especially in Leave constituencies, shows Labour Leave voters switching away from the party during the ‘Brexit realignment’ (2017–2019). Many of these voters were Labour’s traditional ‘heartland’ voters, whom the party had already lost. The fact that these voters switched away over eight years ago will very likely make it harder for Labour to gain them back compared with reconnecting with voters from 2024 or 2019.
There has also been an ideological realignment towards the liberal left among those who have stayed with Labour. The percentage of Labour voters who still support ‘Leave’ (or ‘staying out of the EU’) in 2019 and 2024 is considerably lower than the percentage of these same voters who had voted to Leave in 2016, and this decline is consistent across constituency type. As a result, even in the country’s most pro-Brexit areas, on average, fewer than 20 per cent of Labour voters now support staying out of the EU. If supporting Brexit signals potential for Reform UK, given its strong relationship with other attitudes such as lower support for immigration and the potential for a vote against the status quo, Labour’s current voters across the country now show less potential for Reform support due to the earlier realignment.
An increasingly consolidated ‘right bloc’
In contrast to the splintered bloc on the left, Reform UK is emerging as the more viable party within a right-bloc that was previously highly fragmented in 2024. This has taken place as Farage’s party has continued to draw support from voters on the right.
The graph in Figure 3 follows the electoral history of BES respondents who intended in May 2025 to vote for Reform UK. We can see, from changes between 2017 and 2019, that a small proportion of May 2025’s Reform supporters were among those Labour lost to the Conservatives during the Brexit realignment. A more significant proportion has also moved in and out of the ‘non-voter’ category across recent elections.
However, by far the most common voting record of current Reform supporters is one of backing other parties within the right-bloc – especially UKIP in 2015, and the Conservatives in 2017–2019. Crucially, the more support Reform gains, the greater the threat it poses to the Conservatives, as right-bloc voters increasingly view Reform as the best-placed party to compete with the left.
With the exception of the much smaller swings from parties on the left, this suggests that Reform’s support may be capped by the voters still available to defect from the Conservatives. However, the party has so far been more successful than others at reaching non-voters. In Figure 3, for every 2025 Reform supporter who has come from Labour, two are 2024 non-voters. This partly reflects the larger pool of abstainers given the low turnout in the last general election, and Reform’s relative success in gaining support among them (see Green and Miori 2025b), which likely reflects some success in winning the support of 2024 non-voters who voted for the Conservatives in 2017 and 2019. As a result, the right-bloc is currently growing in size, even if this growth remains relatively small.
‘Bloc equilibrium’ and the relationship between votes and seats
Our previous research (Miori and Green 2025a) showed that Labour doubled its seat share with a 1.6 per cent increase in vote share in 2024 because of tactical voting on the left and fragmentation on the right. Tactical voting made the Labour vote more efficient as the left-bloc coalesced where Labour was best placed to win against the Conservatives. Reform’s split of the Conservative vote allowed Labour to overtake the Conservatives in many constituencies on lower vote shares. This means that any unwinding of this tactical vote for Labour, and any coalescing of voters on the right, will lead to very large changes in seat shares even as voters still support a party from the same ideological bloc.
Labour fortunes depend on continued fragmentation on the right
The bloc equilibrium underpinning Labour’s majority defines the party’s two main electoral challenges in defending the seats it holds. Based on those 2024 vote shares, Reform are now in contention – either second, or fewer than ten percentage points behind the second party – in half of Labour’s seats. These 207 constituencies are shown in Figure 4, where the total height of each bar represents Labour’s lead over Reform in 2024. Some of these distances are substantial, but two factors may make them difficult for Labour to defend.
The first are the high Conservative vote shares in many of these seats, which, if squeezed by Reform, could allow Reform to overtake Labour without attracting any of Labour’s voters. This is a factor largely outside Labour’s control, but it can determine the party’s fate in many of the seats it won from the Conservatives in 2024 on lower vote shares. This is shown by the dark-blue bars in Figure 4, which represent the share of the distance between Reform and Labour that Reform could overcome simply by taking Conservative voters.
The bars that are entirely dark blue – one-third of all Labour seats where Reform are in contention – are seats where the combined Conservative and Reform vote is greater than Labour’s share. These are right-leaning constituencies that Labour would not have won in 2024 without fragmentation on the right, explaining, for example, how they gained seats in the country’s most pro-Leave areas on a smaller share of mostly Remain supporters. As Reform continues to gain voters from the Conservatives, this leaves Labour vulnerable to losing these seats even if it retains the same level of support as in 2024.
The second point is that Labour needs the left-bloc vote to return to being united – and united behind Labour – in order to defend its current, largely tactical, majority. On the left, the broad splintering of voters towards other parties and the undecided will further lower the share needed for Reform to overtake them in competitive seats.
By definition, a consequence of high tactical voting between Labour and the Liberal Democrats in 2024 is that Liberal Democrat vote shares are very low in Labour-gained seats, and vice versa. This means Labour cannot always rely on gaining more voters from the left-bloc in the seats it has to defend against Reform – and that tactical unwinding, even in the form of splintering towards other left parties, can further lower the vote share needed for Reform to overtake Labour.
A tactical ‘re-grouping’ of voters behind Labour depends on whether voters identify Labour as the best-placed party on the left to challenge Reform. Labour may seek to rally the left-bloc by pitching Reform UK as its main rival, but this relies on retaining its position as the most viable left-bloc party in seats where Reform are in contention. In Caerphilly, the most viable left-bloc party became Plaid Cymru, ending Labour’s hold in Wales.
The Caerphilly result, then, offers both positive and cautionary lessons for Labour: a loss of support in Wales, a sign that left-bloc voters can coalesce tactically to stop Reform UK, but also a warning that they may not do so behind Labour. All of these lessons flow from understanding the bloc-based equilibrium in UK politics.