<Alan, 03 Dec 25>
PV versus AMS and STV in Advance of Next GE
Question:
In the light of:
(1) These two earlier threads at https://chatgpt.com/s/t_693082c66b648191868cdd6997143696 (Proportional Confusion) and https://chatgpt.com/s/t_693083385aec8191a5d3b9a46b73b647 (AV vs STV);
(2) The limited time between now and the next General Election in 2029 (if not earlier);
(3) The catastrophic, predictable, adverse democratic consequences of holding that next GE under FPTP voting, including MPs being elected with as little 26.7% of valid votes cast and 14.4% of registered voters (as in 2024 – see the spreadsheet attached that you developed for me), heightened public distrust of politics and politicians in the UK, reduced willingness to turn out to vote ever again, etc , etc, etc; and
(4) The stark contrast between FPTP voting and the demonstrable advantages of PV voting (where PV means preferential voting, with voters optionally ranking any or all of the candidates on the ballot paper)
. . . could you please develop the best draft you can of a statement outlining the benefits of switching from FPTP to PV in advance of the next UK GE?
ChatGPT said:
Here’s a polished draft you can adapt and shorten for different audiences.
A Better Way to Vote: Why the UK Should Switch from First Past the Post to Preferential Voting Before the Next General Election
The 2024 General Election has made one thing painfully clear: our First Past the Post (FPTP) voting system is no longer fit for purpose in a modern, multi-party democracy.
Labour formed a large majority government with just around one-third of the national vote – the lowest vote share ever for a majority government in modern UK history.Wikipedia+1 Smaller parties collectively won record levels of support, yet millions of their votes translated into a handful of seats, while some MPs were elected on barely more than a quarter of the valid votes in their constituency – and scarcely more than one in seven registered voters, as your 2024 results analysis shows.
Independent researchers have described 2024 as “the most disproportionate general election in British history” and “a crisis of electoral legitimacy”.Wikipedia+2Nuffield Politics Research Centre+2 That should worry anyone who cares about democracy, whatever their party preference.
We do not have time to design and implement a completely new constitutional settlement before the next General Election. But we can take a major step to repair democratic legitimacy while keeping the familiar structure of one MP per constituency and one trip to the polling station:
Switch from FPTP to Preferential Voting (PV) – also known as Ranked Choice Voting / Alternative Vote – with optional ranking of candidates.
What is Preferential Voting (PV)?
Preferential Voting is a one-member, one-round system that lets voters rank candidates in order of preference:
- Put a “1” next to your first choice
- A “2” next to your second choice
- A “3” next to your third choice, and so on
Under the optional version we’re proposing:
- You may rank as many or as few candidates as you like.
- If you only want to back one candidate, you can still just mark a single “X” or “1” – your choice.
How the count works (in brief)
- All first preferences are counted.
- If a candidate has more than 50% of the active ballots, they win.
- If not, the candidate in last place is eliminated.
- Their ballots are transferred to each voter’s next preferred remaining candidate.
- Steps 2–4 repeat until someone has a majority of the continuing ballots.
This is exactly what a series of run-off elections would do, but carried out in a single round using ranked ballots. It is already used for national lower-house elections in countries such as Australia, as well as a growing number of cities and local elections around the world.
Why PV is a major democratic upgrade on FPTP (in our UK context)
1. Every MP would have a real majority in their constituency
Under FPTP, many MPs – including in 2024 – were elected with well under 40% of the local vote. Some scraped home with barely a quarter of valid votes, and an even smaller fraction of registered voters. That means most people in those constituencies actively voted for someone else.
Under PV, the winning MP must secure a majority after transfers:
- If your first choice can’t win, your vote moves to your next choice.
- The final winner is the candidate who can assemble the broadest coalition of local support, not just the largest minority.
That makes it much harder for a candidate to “slip through the middle” because the opposition was split.
2. Voters can vote sincerely – without “splitting the vote”
FPTP punishes honesty. Voters are constantly told:
“Don’t vote for who you really want – you’ll let the other lot in.”
This lessens the value of the vote, breeds cynicism, and entrenches tactical voting.
With PV:
- You can safely give your 1st preference to your true favourite – Green, Reform, Lib Dem, SNP, Plaid, independent, whoever.
- If they can’t win, your ballot simply transfers to your 2nd preference, and so on.
Your vote is never “wasted” for daring to support the candidate you actually like.
3. It reduces the “spoiler” effect and extremist wins on tiny pluralities
In multi-party politics, FPTP can elect highly polarising candidates on 30–35% (or less) because three or four broadly moderate candidates divide the remaining vote.
PV directly addresses this:
- If most voters prefer “anyone but Candidate A”, their later preferences will tend to converge on a compromise candidate.
- That compromise candidate will then win the final round – reflecting the true majority view of the constituency.
This doesn’t rig the system against any particular party or ideology. It simply ensures:
“If a majority of us would rather have B than A, C, or D, then B should be our MP.”
4. It keeps what people value about the current system
PV is not a revolution in constitutional design. It keeps the core features many British voters like:
- Same constituencies: one MP per defined local area
- Same rhythm: one day of polling, once every few years
- Same ballot familiarity: you still mark a box by a name – but can add 2, 3, 4 if you wish
It is an evolution, not a complete departure:
- No national party lists
- No multi-member “super constituencies”
- No second round weeks later
For many who are cautious about change, PV is the smallest change that delivers a big democratic improvement.
5. It encourages healthier campaigning
Because second and third preferences matter:
- Candidates are incentivised to appeal beyond their core base, seeking to be many voters’ acceptable second choice.
- Negative, highly divisive campaigning is less effective, because alienating rivals’ supporters can cost you crucial later-round transfers.
In a UK that is more fragmented than ever – with five or more parties polling serious national or regional supportNuffield Politics Research Centre+1 – we need incentives that reward consensus-building rather than pure factionalism.
6. It is already proven and practical
Unlike more complex theoretical systems, PV:
- Has been used for decades in Australia’s House of Representatives and other national, state, and local bodies.
- Has mature administrative procedures, counting software, and audit practices elsewhere.
- Has already been used in the UK for certain internal party elections and mayoral contests (before recent changes).
The UK would not be a global guinea pig. We would be adopting a well-understood, widely tested system that fits neatly into our existing electoral framework.
Why now – and why before the next General Election?
1. 2024 was a warning we cannot ignore
Independent bodies like the House of Commons Library, the Electoral Commission, the Electoral Reform Society, and various academic centres all agree: the 2024 election produced an exceptionally distorted translation of votes into seats.Make Votes Matter+4House of Commons Library+4Research Briefings+4
When:
- One party wins around two-thirds of the seats on about one-third of the vote, and
- Other parties win millions of votes for only a handful of MPs, and
- Some local MPs are chosen by barely one in seven registered voters,
we have a serious legitimacy problem – regardless of who governs.
2. Public trust and turnout are at stake
When people see that:
- Their preferred party can win millions of votes and almost no representation, while
- Others can win a “landslide” on historically low support,
they understandably ask: “Why bother?”
If we go into the next election with exactly the same rules, in an even more fragmented landscape, we risk:
- Lower turnout
- Greater disillusionment
- The further rise of parties whose principal message is, “the system is rigged”
PV won’t fix every problem in politics. But it will ensure that every MP has a genuine local majority mandate and that fewer votes feel pointless.
3. Reform is realistic within the available time
Redrawing the entire electoral system around proportional representation (e.g. STV or MMP) would be a major, complex constitutional project.
By contrast, moving from FPTP to PV for single-member, single-round elections is a manageable reform:
- It uses the same constituencies and broadly similar ballots.
- It mainly requires changes to the counting rules, training, and voter education.
- It can be legislated and implemented during a single Parliament if there is political will.
Given the time pressure before 2029 (or an earlier election), PV is the most credible, meaningful reform we can reasonably deliver in time.
Conclusion: A modest change with major democratic benefits
Switching from First Past the Post to Preferential Voting (optional ranking) before the next General Election would:
- Guarantee that every MP is backed by a real local majority
- Let people vote for who they actually want without fear of “splitting the vote”
- Reduce the chances of extreme or widely disliked candidates winning on slivers of support
- Reward candidates who can build consensus, not just fire up their base
- Keep the familiar one MP per constituency, one election day structure that people understand
In a country where trust in politics is fragile, and where 2024 has already been labelled our least representative election in modern times, we cannot afford to run the next contest under rules that systematically distort the public’s will.
Preferential Voting is not a radical experiment. It is a practical upgrade: one that honours the principle that in each community, the MP should be the candidate that most people can live with – not merely the one who happened to come first in a fractured race.
That is the minimum standard a mature democracy should expect of its electoral system. And it is a standard the UK can still meet – if we choose to change the rules before the next election.
If you’d like, I can now tailor shorter versions of this statement:
- a one-page briefing for MPs,
- a 500-word op-ed, and
- a simple leaflet version for the general public.
You said:
Yes, please. All three 🙂
ChatGPT said:
Here are all three versions, ready for direct use or adaptation:
✅ (1) One-page briefing for MPs (approx. 500–600 words)
Title: Why the UK Should Replace First Past the Post with Preferential Voting Before the Next General Election
Summary
The 2024 General Election delivered the least representative outcome in modern UK history. A government was formed on roughly one-third of the national vote, smaller parties secured millions of votes for minimal representation, and dozens of MPs were elected with support from barely a quarter of those who voted — or around one in seven registered voters. This is a worsening trajectory. With the next GE due by 2029 (or sooner), Parliament has a narrow window to strengthen democratic legitimacy.
Preferential Voting (PV) — also known as Ranked Choice or Alternative Vote — is the most achievable and impactful electoral upgrade available in the timeframe. It keeps the UK’s single-member constituencies and one-day voting tradition, but ensures each MP has genuine local majority support.
What PV Is
Under PV, voters rank candidates 1, 2, 3, etc. Ranking is optional; voters may still mark only a single “1” or “X” for one candidate. Counting works like a series of runoffs in a single round:
- Count first preferences.
- If no one has 50%, eliminate the last candidate.
- Transfer their ballots to the next preference marked.
- Repeat until someone has a majority of continuing ballots.
PV is used for the Australian House of Representatives and in many cities abroad. It is well understood, administratively feasible, and preserves our constituency link.
Why PV Improves Representation
1. Every MP has a majority mandate
Under PV, the winner must ultimately be supported by over 50% of continuing ballots. This is impossible to guarantee under FPTP, where MPs can win with 26–35% in multi-party races.
2. Eliminates “split vote” outcomes
Under FPTP, a candidate can win despite being opposed by a clear majority, simply because similar candidates divide the vote. PV ensures majorities consolidate around the most broadly acceptable candidate.
3. Encourages positive, broad-based campaigning
Candidates need not only mobilise their core vote but also appeal to supporters of other parties who may rank them second or third. This rewards moderation and discourages highly negative, polarising campaigns.
4. Reduces wasted votes and enables sincere voting
Voters can back their preferred candidate first — Green, Lib Dem, Reform, Independent, SNP, Plaid, or smaller parties — without fear of “letting in the wrong candidate.” If their favourite cannot win, their vote still counts toward their next preference.
5. Preserves what voters value: local MPs, simple ballots, one election day
PV is evolutionary, not revolutionary. It requires no new institutions, no large districts, and no party lists.
Why Now? The Case for Urgency
The 2024 election exposed structural weaknesses that will likely worsen as party fragmentation grows. With rising distrust in politics and falling turnout, we must reinforce democratic legitimacy before the next contest.
Introducing PV is a high-impact, low-disruption reform that can be legislated within a single Parliament. It is vastly simpler to implement than proportional systems, yet addresses the most damaging defects of FPTP.
Conclusion
Preferential Voting is the most realistic way to ensure that every MP commands majority support, every voter can vote sincerely, and every constituency elects a broadly acceptable representative. It is a practical, achievable reform that strengthens democracy without changing the foundations of UK parliamentary life. Parliament has the opportunity — and responsibility — to act before the next General Election.
✅ (2) A 500-word op-ed (newspaper-ready)
Title: The Next Election Cannot Be Fought on a Broken Voting System
The 2024 General Election should have been a moment of democratic renewal. Instead, it exposed a voting system in crisis. A government secured a large majority of seats on barely a third of the national vote. Millions who supported smaller parties found their votes converted into negligible representation. And dozens of MPs were elected with support from little more than one voter in four — or in some cases, one in seven of those eligible.
This is not healthy democracy. It is not stable democracy. And it is certainly not democratic legitimacy.
The cause of the problem is simple: the UK still uses First Past the Post, a system designed for a two-party world that no longer exists. In a country where five or six parties compete seriously in many constituencies, FPTP routinely hands victory to candidates opposed by most voters, while pushing millions into tactical voting or disengagement.
With another General Election looming by 2029 — perhaps earlier — we have little time to avoid a repeat of 2024’s distortions. Yet there is a practical solution ready to implement: Preferential Voting.
Preferential Voting (also called Ranked Choice or Alternative Vote) preserves everything British voters like about our system — one MP per constituency, one visit to the polling station, simple ballots — but adds a crucial improvement: it ensures that every MP has genuine majority support.
Under PV, voters rank candidates 1, 2, 3, and so on. If no candidate wins more than half of first preferences, the last-placed candidate is eliminated and their votes transfer to the next preference on each ballot. This process continues until someone has more than 50% of the continuing votes.
This simple change has profound benefits.
First, it tackles the “split vote” problem directly. Instead of a candidate winning because their opponents are divided, majority preferences consolidate. The winner is the candidate most people can accept, not the one who benefits from fragmentation.
Second, PV allows people to vote for who they actually want. Under FPTP, millions hold their nose and vote tactically, or stay home entirely. Under PV, you can put your favourite party — Green, Lib Dem, Reform, SNP, Plaid, independent — first, knowing that if they cannot win, your vote will still count for your next preference.
Third, PV encourages healthier campaigning. Candidates have incentives to appeal beyond their base, because second and third preferences matter. In a polarised age, that alone is a democratic upgrade.
And finally, PV is realistic. Unlike more sweeping reforms, it can be legislated and implemented in the time available. It does not require redrawing the map, introducing party lists, or redesigning Parliament.
We have just witnessed the most disproportionate General Election in modern British history. We cannot afford a second one. Preferential Voting is not a radical experiment. It is a practical fix that strengthens legitimacy, broadens representation, and respects voter choice — all while keeping the familiar foundations of our political system.
For the sake of trust in our democracy, we must not fight the next election on a broken system.
✅ (3) A simple public leaflet (short, plain-language)
Title: Make Your Vote Count — Why We Need Preferential Voting
Right now, the UK uses First Past the Post.
That means an MP can win even if most people voted for someone else.
In 2024, many MPs were elected with under 35% of the vote — and some with around 25%.
That’s not fair. And it’s not how a modern democracy should work.
A simple fix: Preferential Voting
It’s easy:
- Put “1” next to your favourite candidate
- Put “2” next to your next choice
- Put “3”, “4”, and so on if you want
You can rank as many or as few as you like.
If your first choice can’t win, your vote moves to your next choice.
The winner is the candidate supported by the most people overall.
Why it’s better
✔ Every MP gets a real majority
No one wins with just 25–35%. The winner must get over 50% after preferences.
✔ You can vote for who you actually like
Back your favourite first — Green, Lib Dem, Reform, Independent, anyone — without “helping the wrong side.”
✔ No more spoilers
If most people prefer Candidate B to Candidate A, B will win — even if there are several similar candidates.
✔ Fairer, friendlier politics
Candidates must work to earn second and third preferences, so campaigns become more positive and less divisive.
✔ Simple, familiar, and fast
Same constituencies. Same polling day. Same ballot paper.
Just smarter counting.
Why we need this now
The last election showed how broken FPTP has become.
We can’t risk another result where huge numbers of votes don’t count — or where MPs are chosen by only a small fraction of local voters.
Preferential Voting is the quickest, simplest way to fix our elections before the next General Election.
It keeps what works and improves what doesn’t.
Let’s make every vote count.
It’s time for Preferential Voting.
If you’d like, I can also prepare:
- A two-minute speech for a public meeting
- A one-page Q&A / myth-busting sheet
- A website-ready explainer
- Or a short version for social media (Twitter, Facebook, Instagram)
You said:
All great ideas to return to later. However, right now, I think the next challenge will be how to convince people living in Northern Ireland (STV) and Scotland and Wales (AMS), why the voting systems they’re now accustomed to using for regional assembly elections: (a) cannot instantly simply be rolled-out UK-wide for future UK General Elections from the next GE onwards; and (b) why both these ‘party-proportional’ voting systems are democratically inferior from a voter’s perspective to individual voter preference voting ( i.e. PV, with optional ranking), which is a voting system that uniquely supports the original concept of Representative Parliamentary Democracy developed centuries ago in England, where there’s one member per constituency and that member can be held accountable for their performance against the personal promises they made in order to get elected?
ChatGPT said:
Thought for 14s
Gotcha – we’re now in “how do we sell this to people who already use STV/AMS?” mode. Let’s build you a single, adaptable argument you can use in Northern Ireland, Scotland and Wales.
I’ll frame it as a persuasive explainer you can cut down later.
1. Start where they are: “Your systems are better than FPTP”
First thing to avoid: sounding like you’re insulting STV in NI or AMS in Scotland/Wales. I’d open with:
- In Northern Ireland, STV is much fairer than FPTP and has helped manage a very divided political landscape.
- In Scotland and Wales, the Additional Member System (AMS) is clearly more proportional than pure FPTP and gives smaller parties a real chance.
- These systems are improvements on Westminster’s outdated FPTP – and that’s exactly why people are protective of them.
Then pivot:
But the question for the next UK General Election isn’t “STV vs FPTP” or “AMS vs FPTP”.
It’s: what can realistically be implemented UK-wide in time, that maximises individual voter power while keeping single-member constituencies?
That’s where PV (preferential voting in one-member seats) comes in.
2. Why you can’t just plug in STV/AMS for Westminster by the next GE
This is your point (a). Key message: It’s not that STV/AMS are evil – it’s that they’re not plug-and-play for Westminster in the real world, on this timescale.
2.1. STV (as in Northern Ireland)
To use STV for Westminster:
- You’d need to replace almost all existing single-member constituencies with multi-member constituencies (3–6 MPs each).
- That means a complete Boundary Commission redesign, public consultations, and likely political battles in every part of the UK.
- You’d need new rules for how many MPs per area, how to group current seats into larger ones, how to ensure geographic and community coherence, etc.
- Counting STV for 650 seats, each with multi-member races, is a much bigger administrative and logistical lift than AV-style PV on existing boundaries.
In NI, STV works because:
- The Assembly is smaller and designed around multi-member constituencies.
- The political context (power-sharing, community representation) is specific and fragile.
Rolling that system out for every Westminster seat by 2029 (or earlier) would be a massive constitutional project – you’re talking about boundary redistributions + legislation + public education + systems + testing. Realistically, that’s 2–3 electoral cycles of work, not one.
2.2. AMS (as in Scotland/Wales)
To use AMS across the UK:
- You must decide how many list MPs vs constituency MPs (e.g. 50–50? 60–40?);
- Create regional list areas and decide their boundaries;
- Decide whether lists are open or closed (do voters choose people or just parties?);
- Decide whether the list formula is the current Additional Member formula, full MMP, or something else;
- Accept that there will now be two types of MPs:
- Constituency MPs (directly elected)
- List MPs (chosen from party lists, often seen as less “locally accountable”).
Again, that’s a major redesign of how Parliament is composed, not just a tweak to how we mark the ballot. It’s politically heavy and takes time.
So your summary for (a):
STV and AMS are not “drop-in replacements” for the next UK General Election.
They require re-drawing boundaries, changing the number and types of MPs, and re-educating the entire electorate. Done properly, that’s a 10-year project, not a 3–4 year one.
By contrast:
PV in single-member constituencies is something we can legislate and run properly by the next GE, because it keeps the same basic map and the same number and type of MPs. The only big change is how we count.
3. Why STV/AMS are party-proportional but not always voter-empowering
Now your point (b): explaining why, from an individual voter’s perspective, PV (in one-member seats) can be more “democratic” in the classical sense than systems that are mainly about party proportionality.
You don’t want to say “inferior full stop”; you want:
“STV/AMS do this thing (party proportionality) quite well – but PV does this other thing (individual representation & accountability) better, especially for the House of Commons.”
3.1. The democratic trade-off
STV and AMS prioritise proportionality across parties. That’s a valid goal, but it comes with trade-offs:
- STV (multi-member):
- Voters do get to rank individuals, which is good.
- But each area has several representatives. It’s not always clear who is “your” MP in the Westminster sense.
- Accountability is diffused: every party can claim, “you have one of ours in your area”.
- AMS / list-based systems:
- Some MPs are elected directly in constituencies.
- Others come from party lists to “top up” the result so parties’ seat totals match their vote share.
- List MPs are primarily accountable to their party managers (who place them on the list), not to specific voters in a specific place.
- Voters often say they feel less connection to list members: “I never voted for this person, they were just put in from the list.”
PV in single-member seats, by contrast:
- Keeps the original English idea of a named representative for a defined place.
- That person has to win a local majority of preferences from actual voters in that constituency.
- There is a unique, clear line-of-sight:
“You, personally, campaigned here. You, personally, won here. We, personally, can throw you out.”
That’s classic representative parliamentary democracy: one member, one constituency, direct accountability.
3.2. Voters vs parties
A helpful framing:
- STV/AMS are great at fairly dividing seats between parties.
- PV is better at fairly choosing a single representative for each community.
People in NI/Scotland/Wales already have one chamber (Assembly/Senedd/Scottish Parliament) that focuses on party proportionality. You’re not trying to take that away. Your pitch is:
“For Westminster – the UK-wide Parliament that holds the Government to account – the crucial thing is that each MP is personally chosen, and can be personally removed, by a majority in their own constituency. PV is the cleanest way to guarantee that while still letting you express full preferences.”
4. Why PV is particularly attractive if you already like ranking (Northern Ireland especially)
For voters in Northern Ireland, you can lean into familiarity:
- “You already know how to rank candidates. PV uses the same idea – but instead of electing multiple MLAs from each area, we elect one MP who has majority support after transfers.”
- “It’s STV’s core preference idea, adapted to the Westminster principle: one constituency, one representative.”
For Scotland and Wales:
- Many people like the idea that every vote “helps” their party nationally via the list.
- But they also dislike that “some MSPs/MSs don’t really belong to anywhere in particular” or feel “B-list” or party-controlled.
So you say:
“Preferential Voting takes the fairness and choice of ranking from systems like STV, but keeps the line-by-line accountability of one MP per seat that Westminster was built on. It’s the best of both worlds for the House of Commons, while your devolved parliaments continue to be more proportional.”
5. A draft “pitch paragraph” you can drop into speeches or leaflets
Here’s a condensed argument you can reuse:
In Northern Ireland, Scotland and Wales you’ve already moved beyond crude First Past the Post. STV and AMS give fairer party representation and more real choice than Westminster’s voting system. But those systems are built around multi-member constituencies and party lists. They are not something you can simply flick on for Westminster by the next General Election without years of boundary changes and constitutional wrangling.
Preferential Voting is different. It uses the ranking voters in Northern Ireland already know, but keeps the original idea of representative democracy developed here in England centuries ago: one member per constituency, directly chosen by – and answerable to – the people of that place. Under PV, your MP must win a real local majority after preferences; you can back your true favourite first without “splitting the vote”; and if your MP lets you down, you and your neighbours can combine to remove them at the next election. STV and AMS put party proportionality first. PV puts voters and local accountability first – and it’s the only serious reform we can realistically put in place UK-wide before the next General Election.
If you tell me which audience you want to hit first – e.g. NI civic groups, Scottish reform activists, Welsh Labour/Plaid members – I can tune this into a very targeted one-page note or a short stump speech in their language and context.
You said:
Many thanks for your help so far. I need to take a break now, and will return to this topic tomorrow. Meanwhile, bye for now, Alan.
ChatGPT said:
You’re very welcome, Alan — happy to help.
Enjoy your break, and I’ll be here whenever you’re ready to pick this up again.
Bye for now!