Soon after I was elected in July 2024, I was approached by a man in Cirencester Market Place who congratulated me on my win. He’d voted in every general election since 1974, he said, and this was the first time he had ever voted for a winning candidate. For most of that time, he told me, he had barely bothered looking at the names on the ballot paper. He already knew who would win, but had voted out of habit – “or maybe just stubbornness” – but knew that it would, in effect, be a wasted vote.
South Cotswolds is a new constituency, created at the 2024 boundary changes, and covers areas that had consistently returned Conservative MPs for a hundred years or more. 2024 was the first election many residents could remember in which the outcome was genuinely uncertain. The response was striking. Engagement was different. Conversations on doorsteps were different. People who had stopped paying attention started paying attention again, and we had the 6th highest voter turnout in the country.
That should not be remarkable. The fact that it is tells you most of what you need to know about what first-past-the-post does to democracy between elections – not just on the night the results come in, but across decades of people quietly concluding that their participation is pointless.
I spoke in Westminster Hall this week in a debate on first-past-the-post. This is not the first time I’ve raised electoral reform in Parliament, and I want to use this piece to explain why I keep coming back to it – and why I believe the stakes are now higher than ever.
The numbers are worse than we think
We know the headline: Labour won 63% of the seats on 33% of the votes in 2024, giving them 100% of the power. The Gallagher Index – the standard academic measure of proportionality – gave that election a score of 23.67, making it the least proportional general election in modern British history, and the fifth least proportional result anywhere in the world.
But the detail is more striking than the headline. According to Make Votes Matter, it took on average 23,500 votes to elect a Labour MP in 2024, and over 820,000 to elect a Reform MP (silver linings). That is a 35-fold difference in the value of a vote, depending solely on which party you supported. The Green Party, Reform UK and the SNP between them received more than 6.7 million votes – over 23% of the total – and shared just 2.7% of seats.
Make Votes Matter also points out that for roughly 90% of the time since 1935, Britain has had single-party “majority” governments, and not one of them had the support of a majority of voters. Some may claim this is strong government, but it is minority rule, dressed up in emperor’s new clothes, and we need to call it out.
The distortion goes deeper than election night
FPTP does not just distort results. It distorts government between elections.
The incentive is for parties to spend the majority of their campaign funds in a small number of marginal seats. Manifestos are written for floating voters in constituencies that might change hands – not for the country as a whole. And when governments make spending decisions, the incentive structure pulls them in exactly the same direction. The Towns Fund, which directed 40 of 45 allocations to Conservative-held seats, was not an aberration. It was the system working as designed.
The system was designed for two parties. We now have five (or six, or seven).
The risks are not just about fairness. They are about stability. In 2024, for the first time in modern British electoral history, four parties each received more than 10% of the vote – Labour, the Conservatives, Reform UK and the Liberal Democrats. A fifth, the Greens, exceeded 5%. Between them, parties other than Labour and the Conservatives took 42.6% of the total vote – a record high.
In a multi-party landscape, FPTP produces results that are not merely disproportionate but could appear almost arbitrary. Research published this year in Parliamentary Affairs describes the risk plainly: fragmentation does not translate into fair representation under FPTP – it produces super-majorities as a by-product of vote-splitting, leaving the opposition barely able to function and government effectively unchecked. University of Bath researchers put it even more bluntly: under volatile multi-party conditions, FPTP creates a risk of “perverse, extreme, and even seemingly random results.”
We cannot govern a twenty-first century multi-party democracy with a nineteenth-century two-party voting system and expect it to work.
Why this matters beyond the constitutional
I have been writing elsewhere recently about what I believe is the deepest inequality in modern Britain – not inequality of income, but inequality of agency; the sense that your actions matter, that your voice counts, that you have real influence over the decisions shaping your life.
People are not apathetic. They are alienated. There is a real difference. When people feel that participation matters, they engage. When the system signals that it does not, they stop – and the near-record low turnout of 59.9% in 2024 tells part of that story.
Proportional representation is more than a niche bureaucratic detail – It is a statement about whose voice counts. A system that tells millions of people their vote is wasted before they have cast it is undermining far more than an election result.
The man in Cirencester Market Place voted all those years out of stubbornness, not because he believed it would make any difference. He deserves a system in which his vote counts. So does everyone else who has quietly concluded that their participation is beside the point – not because they gave up on politics, but because politics gave up on them first.
* Roz Savage is the Liberal Democrat MP for South Cotswolds. She spoke in Westminster Hall on first-past-the-post on 24 June 2026, and is working on a paper for Compass with a working title of "Power Back to the People: The Future of Liberalism in the Twenty-First Century (2026)".
Follow @libdemvoice.org on Bluesky
Like us on Facebook
Subscribe to our feed
Sign-up for our daily email digest
A steam locomotive leaving Loughborough Central
Five new Liberal Democrat peers appointed
Has the Chancellor set the tone for a Burnham administration?
George Thomas
Chloe
TimL
David S
Nonconformistradical
12 Comments
Irony: first past the post was a winner for us in 2024. Suggest we keep our heads down and live with it.
Canada’s Liberals have adopted that stance, wise in my view.
Nice to see a Lib Dem not seeing electoral reform as a taboo. I doubt the Gallagher Index will shift any view or that people will be heartbroken over Reform getting so few seats. There should be just one fundamental question – should the results of elections depend on who people vote for or not? Of course when a party does well under FPTP, its support for PR collapses. Labour got the support of less than 20% of the electorate. and 63% of the MPs. Under a PR system, perhaps when people don’t vote, there should be an empty seat in parliament to reflect that, 260 empty seats !
I disagree that not inequality of income, but inequality of agency; is the problem – without redistribution of income and wealth – there is little agency. So much of the so called benefits bill, doesn’t go to people receiving benefits. So much household income goes on housing and child care costs, there is nothing for people to invest. Meanwhile the billionaires buy up ever more assets.
‘The man in Cirencester Market Place voted all those years out of stubbornness, not because he believed it would make any difference.’
Him and me both. And now we have to play a guessing game of ‘who should I vote for to keep Reform out?’ every election.
I’ve always thought the French system fairer. Vote for whoever you want in the first round, THEN vote tactically in the second, if your choice is eliminated.
The variations on PR are just too confusing. Wales switching, without asking, from MMP to d’Hondt meant no one really knew what their vote counted for. If anything.
@theakes: No, we didn’t “win” under FPTP in 2024, we broke even (with a share of seats roughly equal to our share of the vote. Just because the system worked for us once (because we had an electoral strategy designed with the vagaries of FPTP in mind) doesn’t mean it will next time. If our vote were to have another 2015-style collapse, chances are this would again lead to a near-wipeout in Parliament because our vote would fall further where it had further to fall. It’s a house of cards.
The Canadian Liberals are taking the same short-sighted position as Labour does here. It is not a good look and is not wise. The SNP is more sensible: even when it benefited from FPTP at Westminster, and would have done from a move to FPTP at Holyrood, it never argued for such a thing. Its collapse in vote share in 2024 led to the loss of most of its Westminster seats. We need to look at the long term, not just one election.
This article gives me real hope. So much of our politics is about winning. Beating the other lot because we all know that we have all the answers and they know squat.
But Roz is spot on. People feel a lack of agency. They have no influence over decisions in their own communities. They are alienated . Until we deal with that issue nothing else will come easily .
When i was 12 years old – 49 years ago the teacher asked the class what one thing would you like to change ? When it was my turn i said bring in PR. My view hasnt changed since then. My vote hasn’t counted once since then. Currently i have a useless Labour mp who doesn’t even answer my emails (im polite without fault)….the constituency link argument, seems a joke to me as most people arent represented by someone they voted for.
In practice, FPTP means the system is rigged against parties not named Labour or the Conservatives.
When we get 1.7% of the seats on 11.6% of the votes (2019), 1.8% of the seats on 7.4% of the votes (2017), or 1.2% of the seats on 7.9% of the votes (2015), this ceases to be an academic or philosophical debate and becomes an albatross around our neck, severely diminishing our ability to effect the change this country needs.
Whether your focus is on the NHS, the environment, the economy, housing, education or something entirely different, FPTP is stopping us from delivering the policies you want to see. We are effectively fighting with one hand tied behind our backs.
This is why I believe that, the next time there is a hung parliament (and, with our fragmented political scene, I expect that to happen sooner rather than later), switching to PR should be a non-negotiable condition for our party.
>”He’d voted in every general election since 1974, he said, and this was the first time he had ever voted for a winning candidate.”
Well I could say the same thing for every GE since 1983. However, I used my vote to great effect, helping to get all three major parties to develop and adopt “green” policies in the 1990s. I think this was a better outcome than voting for “the winning” candidate.
The trouble is the favoured alternatives suffer from a similar problem once voting patterns have been established in a given constituency.
I think Cassie’s observation on the French system is interesting as it would seem to encourage people to vote for the person they want rather than try and back “the winner”.
But still at the end of the process many will have voted for candidates not placed first…
@theakes “first past the post was a winner for us in 2024. Suggest we keep our heads down and live with it.”
If your opinion on an electoral system—any electoral system—is based on how well it worked for your party in one election, you are both short-sighted and not really much of a democrat.
The goal of a representative electoral system is not to ensure that this or that party gets elected. The goal should to be to ensure that every voter, or as close to that as possible, has direct representation that corresponds to their voting intentions; and that the representation fairly reflects the proportions of votes cast.
I have a ‘better’ record than the man in Cirencester: never elected anyone in a FPTP election since 1970 (despite Eric Lubbock getting 43% then).
But I have also never failed to elect someone with my STV ballot since 1973 (frequently myself). Even in European elections I could show my real choice with my first preference and then give later preferences for those who were more or less ok. Better than France.
STV also keeps constituency link with choice of representatives – which keeps them more responsive.
The French run-off system can be gamed. Say there’s a candidate (let’s call him Bordan Jardella) who is highly likely to get into the run-off but you are desperate to keep them out. There is another candidate (call him Micron) who is practically certain to make the run-off, and he is also the candidate you want to ultimately win. So in the first round you could vote for another candidate who you think is likely to get most votes after Micron and Jardella, even if you don’t really like them.
This seems to be strategic rather than tactical voting, because it involves voting insincerely for someone who isn’t your preferred candidate just to knock the one you dislike out of the race, confident that your preferred candidate will stay. It would backfire if too many people did it.
Scenarios exist where strategic voting would theoretically pay off in STV/AV. However, these are rather contrived, involving precise knowledge of voting and transfer patterns that is highly unlikely to be available. Party managers who attempt is find it inevitably backfires (e.g. Sinn Fein once tried a vote management scheme which resulted in a DUP candidate being elected).