The Reform Revolution

An end to two-party politics or its reinvention?

Originally published on the 'Comment is Freed' substack

Jane Green and Marta Miori

There aren’t many elections that change assumptions about British politics. But these ones did. There is now a great deal more uncertainty about what may happen in the coming years.

In some respects these local elections resembled previous ones. They tend to amplify protest against incumbents. In this case the incumbents were both Labour and the Tories. Minor parties did better than their polling would have suggested, as they usually do, and so fragmentation is somewhat amplified. Usually many voters who stay at home in local elections, or vote for someone else to send a protest, return to one of the large parties in a general election. One would still expect this to boost the Labour vote in a general election, although that may be challenged by an extended period of bleak economic pessimism.

The big change is on the right. Reform has broken through a ceiling and can now argue that voting for it is not a ‘wasted vote’ in Britain’s electoral system. Absent something transformational happening, it’s the best-placed party on the right against parties on the left. This was the assumption-changing moment of the 2025 English local election results, and the point Nigel Farage will make between now and the next UK general election, as long as he retains this level of support. 

In this post we set out our early thoughts about what conclusions we can and can’t yet take. Reform still face important challenges, co-ordination could be as strong on the liberal-left as it was in 2024, but the Conservatives face the greatest threat in their history. Their response will determine Labour’s future too, and whether we are headed for an even more fragmented political system or one that repolarises around different parties.

What the results showed for Reform

Sam gave a run-down of what the results showed for all the parties on Saturday. Reform won 677 Councillors, two regional Mayors and a new Member of Parliament. The party outperformed its national polling in the Projected National Share (30%) and National Equivalent Vote (32%), the BBC and Rallings/Thrasher estimates of the vote shares if the local elections had been fought everywhere.

In the following graph we compare Friday’s council vote shares to those achieved by UKIP and the Brexit Party in local elections against an average of their polling between 2012 and 2025.[1] It also shows general election vote shares for UKIP, the Brexit Party, and Reform UK (as well as the European Parliament share for the Brexit Party in 2019).

You can see the degree to which the 2025 local elections were a breakthrough for Reform. Furthermore, whereas UKIP and the Brexit Party struggled to translate their local and European Parliament successes into general election vote shares, because large proportions of their voters returned to the Conservatives in general elections, that assumption is now seriously in question.

 

Critically, contrary to their performance in the 2024 general election, Reform managed to translate their votes into council seats.

There are two ways of making a vote more ‘efficient’ (gaining a better seats-to-votes ratio) in Britain’s majoritarian election system; one is the Liberal Democrat strategy of concentrating votes geographically. The other is to surpass the winning threshold more broadly, benefiting from a ‘winner’s bonus’.

The following figure compares the votes to seats ratios for the parties in last Friday’s local elections to calculations of ‘efficiency’ in the 2024 general election.

Reform vastly improved on their votes to seats inefficiency in 2024. Their ‘efficiency’ at the locals was the same as the Liberal Democrats’ in 2025 and ahead of the Liberal Democrats’ in 2024. This is especially significant given that the Liberal Democrats won 14 times more seats than Reform on a lower vote share in the general election. And while neither was as efficient as Labour in 2024, that level of efficiency for Labour surpassed anything seen before.

 

As local elections were not held everywhere, it’s possible Reform benefited from a larger ‘winners bonus’ in a more rural and socially conservative cross-section of the electorate that is especially favourable to the party’s appeals. The lower turnout levels in local elections may also have helped them. There are clear limits to their support in more urban areas of the country where Labour remains stronger, and a limit in Scotland given higher levels of support for the SNP. However, if they achieved the above electoral efficiency nationally in a general election, on a similar share, they would see a dramatic increase in parliamentary representation.

But, despite this dramatic shift there are still substantial questions about whether they’ll achieve this kind of breakthrough which we’ll now turn to.

The challenges for Reform

There are limits to Reform’s ‘Labour’ vote

While Reform gained seats in many Labour areas of the country, there is the risk we assume that Labour ‘places’ tell us what Labour ‘voters’ were doing, something that’s known as the ‘ecological fallacy’ - a real challenge in reading elections.

In the general election last year, Reform gained votes largely at the expense of the Conservatives, with over 70% of 2024 Reform voters coming from 2019 Conservatives. This was a big help to Labour in achieving their large majority and 411 MPs, lowering the thresholds for them to take seats from the Conservatives.

In the 2025 English local elections, the Conservative seat tally and Labour’s seat tally both went down, which was all the worse for Labour given that this compounded their losses in 2021 (and before that too). However, this does not necessarily mean that Reform were taking large proportions of Labour voters.

These results could as easily be explained by Labour voters staying at home or voting for a party within the same ‘party bloc’ (a left-liberal party). This would be consistent with observations from the British Election Study (BES). Since 2016, voters have coalesced into a ‘Remain-left-liberal bloc’ and a ‘Leave-right-socially conservative bloc’.

This bloc-based alignment was especially in evidence in the high-fragmentation 2024 election. Voters are still more likely to be switching within blocs than between them, and Reform is still very likely taking most voters from the right. Additionally, we cannot know whether they are tapping into recent non-voters, although there are certainly more given the low turnout in 2024. Labour may, as it long has, have a problem mobilising its voters, but this mobilisation should be more achievable at a general election.

At the 2024 election, according to our analysis of BES data, only 5% of 2024 Labour voters saw Reform as their preferred (more ‘liked’) party among those they didn’t vote for. Furthermore, there was no significant variability in the proportions of 2024 Labour voters who like Reform across different types of ‘place’ in Britain, which we measure by a constituency’s Brexit share, 2024 Reform vote, the level of unemployment and the proportion of graduates.

To calculate this, we use respondents’ rank of each party on a scale between 0 = strongly dislike and 10 = strongly like. A majority of 2024 Labour voters – 51.8% – score Reform 7 to 10 points behind Labour, and a further 31.7% – have a 4-6 point gap between their Reform and Labour scores. Crucially, the 16.6% of Labour voters who do like Reform are not geographically concentrated in any particular type of seat.

The graphs below show the density of each group of Labour voters – those who do, those who might, and those who do not like Reform – across different types of constituencies.

 

All three categories follow similar patterns of distribution across the different constituencies. Labour voters who like Reform are, contrary to much commentary, not disproportionately concentrated in seats where fewer constituents have a university degree, have higher rates of unemployment, or where there was stronger support for Leave in 2016.

Even in areas where Reform did best in 2024, Labour voters are not particularly more likely to be ‘Reform-friendly’ than they are elsewhere. They may be slightly more numerous, but so too are Labour loyalists. 

This observation is important. It is true that Labour MPs in ‘red wall’ constituencies will hear more people saying they are tempted to switch from Labour to Reform. But they should not be in a significantly higher proportion to Labour voters who are less likely to view Reform positively.

On the other hand, the ‘available’ right wing vote for Reform is considerable because of the enormous damage the Conservatives did to their reputation in government and the losses they sustained as a result.

Labour’s problem is that it is Reform who are getting these disillusioned voters, not Labour, rather than Labour’s vote significantly splintering to Reform. Where Labour is losing voters, it is associated with voters’ financial worries, and it is in multiple directions as well as to ‘undecided’; a broad-based loss due to economics, greatest among adults in mid-life who feel the most economically insecure.

The left could again be more coordinated

The most important change from these elections, if continued, is that Reform is now the best second-placed contender in lots more constituencies. That changes the calculations of tactical voters in general elections and the way parties will target them.

There are, of course, limits to the amount of likely tactical voting or ‘correct’ understanding of the local tactical context in elections, and the changes we’re seeing in British politics are only going to make those kinds of estimations harder. There was much less tactical voting in 2024 among Conservative and Reform voters than there was among voters on the liberal-left.

However, one threat to Reform is more efficient coordination on the left.

The following figure shows the vote shares the Liberal Democrat and Green parties achieved in local elections between 2011 and 2024 (the bold lines) with their subsequent vote shares in the general elections that followed (the dashed lines). They typically outperform their general election shares in local elections.

 

A great deal is at stake for Labour here. Go after Reform voters by moving to the right on key issues like immigration and the liberal-left voters, as well as those who are currently undecided and stayed at home, may not coalesce behind Labour to the degree that they did in 2024.

Conversely, the ‘threat of Reform’ to Britain’s left-liberal voters, and the realisation that they pose a real danger to Labour constituencies and to their majority, may well stir up as great – perhaps even greater – strategic voting for parties best placed to beat Reform (or the Conservatives) on the liberal-left.

Furthermore, the longer the Conservatives stay close to Reform to try and get their voters back the greater the incentive for left-liberals to elect Liberal Democrats where they are best placed to win against the Conservatives.

Reform still help as well as potentially harm Labour in seats

In 2019 the Brexit Party under Farage contested Labour held constituencies, not those the Conservatives were defending, helping secure Boris Johnson’s majority. As Reform are now in contention in far a greater number of places, there is far less reason for Nigel Farage to coordinate with the Conservatives.

In 2024 Reform greatly contributed to splitting the right-wing vote and lowering the threshold for Labour to win constituencies from the Conservatives. They could help Labour even more next time if Reform are seen by right-wing voters as better placed to win constituencies in which the Conservatives are currently first or second.

Of the 182 constituencies Labour won from the Conservatives in 2024, Reform are not in second or close third (within 10 percent of the second party’s vote share) in 96 of them. Standing there made it much easier for Labour to take those seats from the Conservatives. This suggests that Labour’s hold on these constituencies is relatively secure, especially if the right’s vote is even more split.

However, Reform are in close contention (second or within 10 percent of the second party’s share) in 86 of these Labour gains from the Conservatives, as well as potentially in 121 other Labour seats. One of these constituencies is Runcorn, which Reform just narrowly won, with the Conservative vote share halving compared to July 2024 last year.

The figure below orders all these constituencies by Reform’s distance from Labour, with Conservative vote shares shaded in darker blue. In many of these seats, the Conservatives are close behind (or narrowly ahead of) Reform. If this balance were to be repeated next time it would help Labour. But it also means that further damage to the Conservative vote share by Reform would leave Labour very vulnerable. Labour are in the somewhat perverse situation of needing the Conservatives not to fall away completely.

We should be careful in reading too much into Runcorn (highlighted below in red), which demonstrated a lower turnout than in the general election and which took place because the previous Labour MP hit a constituent. But, whilst the Conservative vote share would have to drop quite significantly for Labour to lose similar seats at greater scale, these swings could be tactical, especially if right-voters realise Reform are in closer contention.

 

One concern for Kemi Badenoch must be that, whatever she does now, some Conservative MPs might defect to Reform given that they too will see Reform UK as the likelier winner in their constituency.

The death of two-party politics in Britain, or a period of reorganisation?

The split on the right at the 2024 general election has left many Labour seats where their fortunes are almost entirely dependent on the tug of war between Reform and the Conservatives. Whether this tug of war continues, or whether one party comes out victorious, will not only affect whether Labour holds onto many of its constituencies where both parties are contention but also whether British politics is going through a period of reorganisation on the right rather than moving to a ‘five party system’.

In 64 of the 121 constituencies currently held by the Conservatives, Reform are in contention (either second or in close third place within 10 percentage points of the second-placed party). These 64 seats are ordered by marginality in the graph below. Reform continue to squeeze votes from the right and if they did so in a general election it would be existential for the Conservatives, whereas Labour still have plenty of seats where Reform aren’t a threat.

 

Let’s say, as a back-of-envelope estimate, Reform take half the constituencies Labour won from the Conservatives in 2024 where they are second to Labour (43) and two-thirds of the seats where they’re second to the Conservatives (43) and we add the 5 MPs they have now. Reform would stand to have 91 MPs in the next Westminster parliament. A huge breakthrough, but not a ticket into No.10.

But a further breakdown of Conservative party fortunes could take Reform well beyond this threshold.

It would be even crazier today to try and predict what will happen in a future general election than it was a week ago. What could happen has become considerably more uncertain. As is made obvious from the above focus on different types of seats, four party competition is devilishly complicated, let alone five-party competition (and the picture is different again in Scotland).

But we also can’t be certain of the prognosis that two-party politics in Britain is ‘over’. It is true that, aside from 2017 and 2019, the British party system has been fragmenting more over time, and that this reached its highest points in 2024 and now in the 2025 local elections. But there is a possibility, too, that coordination on the left, and a different main party on the right, could come to look like two party politics again after a protracted period of fragmentation and reorganisation.

If Reform’s popularity grows, British voters faced will be faced with choices of different dominant parties in the same electoral system. It would just mean they are different from the two dominant parties than we have been used to. The uncertainty has only been heightened.

[1] With grateful thanks to Colin Rallings