Labour & Conservative slump, Reform rise in UK elections … my book
Labour & Conservative slump, Reform rise in UK elections … my book
1 Prelude to the UK local elections, 2025
2 The book: Local elections and national opinion – a model
“Reform’s rapid ascent offers risk and reward.
After tomorrow’s local elections, Nigel Farage’s party will no longer be the upstart – and there the problems may begin.
Reform’s appeal is that it represents a rebellion against a failed politics.”
Daniel Finkelstein. The Times, 30 April 2025: 21.
1 Prelude to the UK local elections, 2025
Tomorrow, Thursday
Background
Group A and Group B: parties in political space
Current opinion polls
UK general election 2024
Step ratios, comparing 2024 and 2025
Flow: party survival rates, from 2024 to 2025
Flow: within-group and between-group … distance
Flow equations
The Runcorn and Helsby by-election: a step ratio prediction
The Runcorn and Helsby by-election: a flow equation prediction
Local elections 2021
Local elections: step ratios 2021 to 2025
2 The book: Local elections and national opinion – a model
1 Introduction and overview
… Swing: uniform? proportional? linear? general? … flow … seats
2 An analysis of changing percentages – a conceptual map
3 National elections: mass and motion in political space, 1945-2024
4 Local results in national elections
5 Local government elections
6 By-elections
7 Modelling
1 Prelude to the UK local elections, 2025
Not two-party politics, but multi-party politics.
Last year, a Conservative slump – this year, also a Labour slump. So Reform wins.
Last year, my online book – this year the application of the ideas in my book,
to the UK local elections and by-election tomorrow, Thursday.
Tomorrow, Thursday
Conservatives will do very badly in UK local elections …
Labour will do badly in the Runcorn by-election …
Reform, Liberal Democrats and Greens will do well …
Group A and Group B return to Brexit-like parity after the Group A dip last summer …
… assuming the results follow national opinion polls, and relative to the relevant previous occasion(s).
Background
Last summer saw the end in Britain of fourteen years of rule by the Conservative party. Their vote had slumped in opinion polls and in local elections and this slump was repeated in the general election.
Now a year later local elections and a by-election are being held. Have the Conservatives recovered? What about the other parties? Last year Labour won a handsome majority of seats but with only an unremarkable percentage of the vote. How has being in government changed this percentage? The results for Nigel Farage’s party last year were opposite to that of Labour: Reform won five seats but a sizeable percentage, 14.3 of the vote. How has this changed over the past year?
In addressing these questions it is important to think about how we conceptualise changes in the percentage vote. The traditional approach is to think in terms of the swing from one party to another. However the dramatic slump in the Conservative vote led some analysts to the notion of proportional change. In general the notion of step ratio and a voter flow model are helpful.
There are national results and local results. How do these relate to one another?
There are many opinion polls and few elections. How do these two sources of information compare?
Group A and Group B: parties in political space
One dimension of political space is support for Brexit. The proportion of a party’s supporters who voted Remain in the 2016 referendum provides a measure of a party’s support for Brexit. Four parties have a high Remain score – referred to as Group B - and two parties have a low Remain score – referred to as Group A.
B SNP 0.83
B Labour 0.81
B Green 0.78
B Lib Dem 0.7
- Plaid C 0.5
- Other 0.5
A Conserv 0.39
A Reform 0.13
Reform also has a lower class and a lower female score and a higher age score and a low London score
Current opinion polls
The median of five recent opinion polls put the Conservatives on 21%, ranging between 20% and 23%. Labour and Reform have similar medians, 23% and 25% respectively. Lower are the Liberal Democrats with a median of 14%; and the Greens with a median of 9%.
Grouping the parties, Group A and Group B are equal, at 47%. Group A has Conservatives and Reform. Group B has Labour, Lib Dem and Green. (In other reports I have included SNP in Group B.)
A major topic of discussion is a possible pact or understanding between the two Group A parties, Conservative and Reform.
A Conservative 21% (20-23)
B Labour 23% (20-26)
A Reform 25% (25-28)
B Lib Dem 14% (11-16)
B Green 9% (8-13)
…
Group A 47% (45-48)
Group B 47% (45-49)
…
Source 1
UK opinion polls; end of polling date 21-25 April 2025; five polls
https://en.wikipedia.org/wiki/Opinion_polling_for_the_next_United_Kingdom_general_election
UK general election 2024
Last year’s general election gave a landslide in seats to Labour – but only 33.7% of the vote. Second were the Conservatives with a historic low of 23.7%. Reform narrowly beat Liberal Democrats into third place, with 14.3% as opposed to 12.2%. Greens had half that with 6.4%.
Group B was 14% ahead of Group A.
A Conservative 23.7%
B Labour 33.7%
A Reform 14.3%
B Lib Dem 12.2%
B Green 6.4%
…
Group A 38.0%
Group B 52.3%
…
Source 2
https://en.wikipedia.org/wiki/2024_United_Kingdom_general_election
Step ratios, comparing 2024 and 2025
We now compare last year’s general election with this year’s recent opinion polls. Dividing the latter by the former, gives the step ratio. Support for Labour has slumped – it is now two thirds of what it was last year, a step ratio of 0.68. Support for the Conservatives is also down, but not by as much, the step ratio being 0.89. Note that a large Conservative slump had happened before last year’s general election.
Other parties have enjoyed increased support. Most dramatically the support for Reform has almost doubled with a step ratio of 1.75. Lib Dem has had a modest increase with a step ratio of 1.15. The Greens have enjoyed a step ratio of 1.41, albeit starting from a low base.
There has been flow between Group A and Group B and also flow within each Group. So the step ratios are more modest, 1.24 for the increasing Group A and 0.90 for the declining group B – but note that Group B was larger than Group A in 2024.
A Conservative 0.89
B Labour 0.68
A Reform 1.75
B Lib Dem 1.15
B Green 1.41
…
Group A 1.24
Group B 0.90
Flow: party survival rates, from 2024 to 2025
The survival rate is the percentage of a party’s voters in 2024 who stayed loyal and continued to support the party in 2025. Less than two thirds of Labour voters stayed loyal, the survival rate being just 62%. Conservative voters were not much more loyal, with a survival rate of 69%. Lib Dems had a better survival rate, namely 77%. Finally, Reform voters were especially loyal, with the high survival rate of 89%.
A Conservative 69%
B Labour 62%
A Reform 89%
B Lib Dem 77%
Source 3: YouGov
https://d3nkl3psvxxpe9.cloudfront.net/documents/VotingIntention_MRP_250422_w.pdf
Flow: within-group and between-group … distance
Where do they go, those who do not stay with their party? Some go to another party but stay within the same group and some go to a different group – thus giving a flow between groups. For example whereas 69% of those who voted Conservative in 2024 stayed with their party, 20% moved to Reform, a different party but in the same Group A, and 10% moved to one of the Group B parties. So more movers stay in their own group than move to the other group. Indeed this is a general statement which applies to all the four parties listed below – within-group flow is greater than between-group flow. This can be understood in terms of distance: flow is greater when the distance is less.
A Reform: 89% stayed; 6% went to Conserv (A); 3% went to Group B
A Conservative 69% stayed; 20% to Reform (A); 10% to Group B
B Labour 62% stayed; 15% to A; 20% to B
B Lib Dem 77% stayed; 9% to A; 12% to B
Flow equations
Part of the table on page 1 of the YouGov Survey results shows how the current preference depends on the vote in the 2024 general election. This allows us to deduce flow equations.
Conserv now = 0.69 Cons GE + 0.04 Lab GE + 0.04 LD GE + 0.06 Reform GE + …
Lab now = 0.02 Cons GE + 0.62 Lab GE + 0.07 LD GE + 0.01 Reform GE + …
LibDem now = 0.07 Cons GE + 0.12 Lab GE + 0.77 LD GE + 0.02 Reform GE + …
Reform now = 0.20 Cons GE + 0.11 Lab GE + 0.05 LD GE + 0.89 Reform GE + …
If we apply these equations to the percentages in the General Election we obtain a Conservative now percentage of 19.0% (YouGov report 20%); Labour now 22.4% (23%); LibDem now 15.4% (16%); and Reform now 21.8% (25%). So the equations give quite a good prediction. On the one hand this is not surprising since the data is all within the data set; and on the other hand it is surprising since not all the sample are covered by these equations.
Note that the equations underpredict the support for Reform.
In a following section we shall apply the equations to predict the Runcorn by-election.
Source 3: YouGov
https://d3nkl3psvxxpe9.cloudfront.net/documents/VotingIntention_MRP_250422_w.pdf
Note: I used an Excel chart with the following equations.
= 0.69*d14+ 0.04*d15+ 0.04*d16+ 0.06*d17
= 0.02*d14+ 0.62*d15+ 0.07*d16+ 0.01*d17
= 0.07*d14 + 0.12*d15+ 0.77*d16+ 0.02*d17
= 0.20*d14+ 0.11*d15+ 0.05*d16+ 0.89*d17
The Runcorn and Helsby by-election: a step ratio prediction
A by-election is being held in Runcorn and Helsby. In the general election last year Labour won the seat with an overall majority with Reform in second place with only 18% of the vote, just ahead of the Conservatives on 16%.
Let us speculate about the result in Runcorn tomorrow. Suppose the local Runcorn results follow the national situation – suppose the local step ratios follow the national step ratios. This gives a prediction based on the national step ratios. Labour will fall to 36.1% and Reform will rise to 31.6%, just behind Labour. So Labour just manages to retain Runcorn. (The other three parties are far behind.)
But this prediction might be wrong. Labour in Runcorn might do worse than this and lose to Reform, losing local support because of bad behaviour by the outgoing Labour MP, and the Reform candidate gaining tactical votes from the Conservatives – giving the seat to Reform.
. GE SR SR-based prediction
Conservative 16.0 0.89 14.2
Labour 52.9 0.68 36.1
Reform 18.1 1.75 31.6
Lib Dem 5.1 1.15 5.82
Green 6.4 1.41 9
. 98.5 96.8
“Labour thinks it has lost Runcorn to Reform”
https://www.telegraph.co.uk/news/2025/04/29/labour-thinks-lost-runcorn-reform-charts/
The Runcorn and Helsby by-election: a flow equation prediction
A different prediction can be made using the flow equations. The step ratio and flow predictions are fairly similar for Conservatives and for Labour. But the flow prediction is lower for Reform and higher for Lib Dems.
. GE flow prediction SR SR-based prediction
Conservative 16.0 14.4 0.89 14.2
Labour 52.9 33.7 0.68 36.1
Reform 18.1 25.4 1.75 31.6
Lib Dem 5.1 11.8 1.15 5.82
Green 6.4 ? 1.41 9
. 98.5 85.2 96.8
Local elections 2021
This year’s local elections need to be compared with the local elections four years ago, in May 2021, when the same councils as now were up for election. At that time the Conservatives were enjoying support in the range 40% to 50%. This was reflected in the local election results which corresponded to a National Equivalent Vote of 40%. Labour were second with 30%, Liberal Democrats with 15% and Other with 15%.
Local elections: step ratios 2021 to 2025
Taking the percentages for 2021 given in the previous section and the current opinion poll percentages, we can obtain the step ratios – see below. These ratios indicate how tomorrow’s results will relate to the results in 2021. There will be a dramatic slump for the Conservatives, a less dramatic fall for Labour (their vote in 2021 was not very high), not much change for the Liberal Democrats, and a large increase for ‘Other’ which includes Reform.
A Conservative 0.53
B Labour 0.77
. Other 2.80
B Lib Dem 0.93
2 The book: Local elections and national opinion – a model (2024)
This is a draft online book which I produced after the local elections last year (2024) just before the general election.
The Front Page sets the scene with the expectation of a heavy Conservative defeat.
The Contents section lists the seven chapter titles and also the section titles within each chapter.
Chapter 2 provides a conceptual map introducing the step equation and the flow equations (3pages).
2 An analysis of changing percentages – a conceptual map.
Chapter 3 National elections: mass and motion in political space, 1945-2024.
This 25-page chapter looks at the course of UK general elections in the period 1945-2024. One sub-section looks at the rise of Nigel Farage and the various parties he has been associated with.
3 National elections: mass and motion in political space, 1945-2024.
Chapter 4 Local results in national elections.
This 11-page chapter continues the analysis of national elections but looks at how local results satisfy general equations. One sub-section looks at how deviations from the general equation depend on the Remain vote.
4 Local results in national elections.
Chapter 5 Local government elections.
This 25-page chapter looks at local government elections. Looking across different local areas is the swing uniform, proportional or linear?
Chapter 6 By-elections.
This 33-page chapter looks at by-elections: “… more than one recent poll has suggested that Conservative support is falling most in seats where the party was previously strongest …”. Again the concept of step ratio is applied.
Chapter 7 Modelling
This 11-page chapter looks back at each chapter and looks into the modelling aspects in greater detail.
The website for the book also gives the same above links to the chapters.
Local elections and national polls
THE END