Disclaimer

#

This is a follow up to my original article about Democracy as well as its Addendum, it builds off a lot of the topics and terminology introduced in that article and calls back to it several times throughout, so I recommend you read it before this one if you haven’t already!

---

Written between the 16th and 19th of December 2024

---

Across the Pond

#

A long wait

#

Well bugger me sideways, those yanks really are slowpokes aren’t they?

It’s been a month since the US election, and at the time I’m writing this (Dec 16th) most people don’t actually have the full results. News agencies like the Associated Press and CNN are still listing some areas as “99% counted”, Wikipedia has the same problem, they’ve been stuck at 99% for weeks now.

That’s for a number of reasons, a lot of it has to do with state law. You see, even though the US Presidential election is for a nationwide federal office, how Presidents are elected is still decided at the state level, different states have different laws on all sorts of things, how the ballot is ordered, which candidates and parties can get on the ballot, and more importantly, how the votes are counted.

Different states have different rules on how postal votes are counted, some simply throw out any ballots they get after election day, others will have a cut off point after the election where they can still count. Postal ballots also  have to be signed for, some states will simply throw out a ballot where the signature doesn’t match other signatures they have on file for that voter, others will give them a certain amount of time to “cure” the ballot, to rectify any issues and have it counted, before it’s thrown out.

Then there are what’s called “provisional ballots”, if a voter’s identity is in doubt, or some other issue with them comes up, most states allow them to cast a “provisional ballot” that will be held separately from the rest of the votes, the voter than has a certain amount of time to clarify those issues and allow the ballot to be counted, the amount of time again depends on the state.

Lastly there’s “certification”, certification is when the final results of the election are published and confirmed, every candidate listed with every vote they got, this becomes the legal final result, different states have different deadlines for when their results have to be certified, for some states that deadline is under a week after the election, for others it’s over a month.

What this adds up to is a system where some states will resolve any issues they have with their ballots, and count or discard all of them, very quickly, and others will drag things out for months.

Now you might ask, why does that matter? We already know who won. And that’s true, the morning after election day it was confirmed that Donald Trump had won the election and would be the 46th US President, it was obvious he was going to win even before then as he progressively knocked over all of the key swing states throughout the night, but it’s a question of how he won.

I wanted to look at the results in detail, not just how many states he won but how many votes he got, how many it was compared to 2020, how many votes were between him and his rival Kamala Harris, that sort of thing, and you can’t look at these kinds of details unless you have all the results.

I started to get really impatient, I’ve been planning to write another addendum to the democracy article for a while now, talking about a number of different topics but the US election was the first of them, I decided not to start writing until I could do that.

I started feeling the itch, I took lots of notes to prepare, made a lot of new graphs for various other elections and started doing lots of tweaks to the original article, some of them were small corrections here and there but most of them were new assets and more clear explanations of some of the topics, I was happy with these improvements but it wasn’t enough, I wanted to start writing something new, which prompted me to actually look into what was causing the holdup with the US results.

And in doing so I found something interesting, I looked up how many states had actually certified their results on a website called “270 to win” and found out about half were there, but the site actually mentioned something strange, the deadline for all of them to be in was apparently December 11th.

Huh, what’s odd, why are there only half of them then? Looking up the state deadlines helped clarify things, some states just ignore this nationwide deadline and set their own deadlines later in the month, but then I looked into the date with the furthest deadline, Arkansas, it has a deadline of December 20th.

I looked it up and it turned out they actually certified their results early, on December 9th, which made me realise something, they’re not actually still counting.

The results are all in, it’s just that the editors of Wikipedia and all the major news outlets haven’t actually bothered to go and get them yet.

DIY

#

I realised that I could just go to each state’s website, get their results and tally them up, and get the full results, but I wasn’t too keen on that, there are 50 states and Washington DC gets a vote too, so that’s 51 separate results pages I would have to go and fetch, a colossal task.

But very quickly I caved, my impatience got the better of me, so from the evening of the 15th to the morning of the 16th I worked on this massive spreadsheet with the full results of every state.

Keep in mind that there’s room for some human error here, I put this list together in around 8 hours and I was very tired by the end of it, I even missed off 2 states (Arkansas and Montana) in my first go around and made a bunch of other mistakes like putting in the wrong numbers when writing down vote totals, giving the votes to the wrong candidate, missing candidates entirely, I caught all of these straight away except for the missing states and I’m relatively sure I got everything right in the end but just saying, this was put together by one guy when it’s the kind of work a whole team should probably have been doing.

It was easy enough to find each state’s results page, I either looked up the state’s Secretary of State (an official who in most states is responsible for, among other things, election administration) or if that didn’t work “[state name] elections gov” which would almost always quickly bring up the page of some sort of government elections commission that would have a results page.

Many states made it easy, they would have a straightforward webpage or PDF file with all their results: Each candidate, the votes they got and the turnout, with the total number of votes next to the number of registered voters.

Some states made things more difficult, they would put the turnout stats on a separate page forcing me to go digging to find out how many didn’t vote, some didn’t list turnout at all but would have another page with voter registration data, allowing me to subtract the number of people who voted from their number of registered voters and work things out that way.

Others still made things even more difficult, they would store their results in clunky Excel spreadsheets I would have to download and then manually copy the results from, or at their worst, they would screw up entirely.

Some states simply didn’t bother to calculate turnout yet (Maine), or didn’t provide up to date turnout/voter registration info on their websites (looking at you Minnesota, Mississippi, Missouri, Tennessee, Texas, Virginia) and some (Illinois, New Mexico, Oklahoma, Pennsylvania) had websites that were completely inaccessible, spitting out 403 or security errors when I tried to use them and forcing me to use Wikipedia and local news sources as a backup, this means I don’t have the full data on how many people were registered to vote but chose not to, but I was able to get the most important thing, the full voting results, and here they are in graph form:

_{US 2024 presidential election results - Voters only (Detailed)}

The 5 Genders of Bobby Kennedy

#

You can see that this graph is a lot more complex looking than my earlier US election graphs, that’s because for earlier elections I took the results from Wikipedia, which only has 5 options: Republican, Democrat, Libertarian, Green and “Other”.

This graph is based on my full dataset, which has 53 options, including 51 candidates, “Write-Ins” (many states allow voters to write the name of who they want to be president instead of just picking from the options on the ballot) and “None of these Candidates”, which was an option in just 1 state, Nevada.

That doesn’t actually mean there were 51 people on the ballot, most of these 51 are individual candidates but some of them are the same candidate under different names, remember how I said different states have different rules for who gets on the ballot? This is where that comes into play.

For example, Jill Stein usually ran as a candidate for the Green Party, which is the party she represented nationwide, but in some states the Green Party wasn’t allowed to run, so she had to run as an independent. In one state, Kentucky, a party called the Kentucky Party was created just so she could get on the ballot there.

Robert Kennedy Jr, who usually runs as an independent, sometimes had to run under a party label to get on the ballot, sometimes this was the “Independent Party”, sometimes it was the “American Independent Party”, sometimes the “We The People Party” and sometimes the “Natural Law Party”.

One independent candidate, Cornel West, ran under 7 different party labels as well as an independent.

The most bizarre one though was probably the Constitution Party, their guy Randall Terry sometimes had to run as an independent, in one case he ran under the “US Taxpayers Party”, but in a few states the Constitution Party nominated a completely different candidate entirely, Joel Skousen.

Guys, if you can’t even work out who your candidate is, you probably shouldn’t be running in an election.

Then you might also have noticed the little orange and pink lines between Trump and Harris, these lines are from parties based in New York. New York has a system called “fusion voting” where 2 parties can nominate the same candidate on the ballot paper, the orange line is for the New York State Conservative Party, which nominated Trump, the pink line is for the Working Families Party, which nominated Harris, the idea behind this is to allow people to support a smaller party without splitting the vote.

Anyway, if we take all of that complexity out, here’s what the results look like:

_{US 2024 presidential election results - Voters only}

This graph is actually still a bit more complex than the ones based on Wiki, but it’s much more simple than the other version, since so many votes were only for Democrats and Republicans you won’t even notice much of a difference, but here are the changes:

I narrowed down the 53 options to just 7, I picked 5 major parties (Republican, Democrat, Libertarian, Green and Independent - Kennedy) and put all the votes for their candidates together, so if you voted for Trump as “New York Conservative” here it won’t be shown separately, it’s bundled in with the rest of his votes under Republican.

Then I left Write-Ins as their own category and bundled all the other small parties and candidates together into “other”.

I thought it wouldn’t be fair to leave Kennedy out and bundle him into “other” since he got more votes than the Libertarians did, if they get to be shown separately he does too.

Dissection

#

So, what do these results add up to? Well, they tell us quite a few things.

• First, the total number of valid votes is 155,290,437, that’s a drop of just over 3 million votes compared to last time.

• Second, Trump got 77,303,425 votes, that’s about 3 million more votes than he got in 2020.

• Third, Harris got 75,018,929 votes, that’s around 6 million fewer votes than Joe Biden got.

• Fourth, the votes for all other candidates, that’s including Stein, Kennedy, Oliver (the Libertarian candidate), the write-ins and all other minor parties totals up to 2,968,093 votes, a slight rise from 2,922,155¹ last time.

That tells us while Trump has gained the result is primarily down to a Democrat loss, if the Dems had simply kept all or most of the votes they got in 2020 they still would’ve won, and the Dems haven’t lost because of vote splitting either, even if every single person who voted third party voted Democrat Harris still would’ve lost, much more narrowly, but she still would’ve lost.

That would never have happened anyway, a bunch of those minor candidate voters, like a lot of those voting for the Libertarians or Kennedy, are right wingers who if you forced them to choose would’ve probably chosen Trump over Harris, but it’s worth emphasising that minor parties aren’t to blame for Kamala’s defeat.

Trump won the popular vote for the first time, something he’s never done before, so he hasn’t relied on the USA’s disproportional election system to win this time, but he’s gotten a bigger piece of a smaller pie.

These are all things that I assumed had happened, guessing them either on the day of the vote or a few days after, but I didn’t want to say them until I had the numbers, now I do.

So, why did Harris lose all those voters that Biden won? To me the answer is simple, it’s a decisive rebuke to “Blue No Matter Who”.

I summed up Blue No Matter Who in the original democracy article, a sort of “lesser evil-ism” taken to its worst extreme:

This political hostage situation has massively escalated with the rise of Donald Trump, who has pushed the Republicans and the right to greater and greater extremes since he won the Republican nomination in 2016, anyone to the left of his right wing base has faced pressure to vote “Blue No Matter Who”: To support the Democrats no matter how much they might dislike the Democratic candidate, just for the sake of stopping Trump.

Establishment Democrats have gained an intense level of bargaining power as a result as they can resist pressure from critics by simply warning “if you don’t vote for me, you get Trump”, Hillary Clinton and Joe Biden were widely regarded as untrustworthy and unlikeable candidates but millions of people who didn’t like them or even outright hated them still voted for them anyway, with many lashing out at those who refused to do the same.

The problem is it’s a short term fix that creates bigger longer term problems, because “Blue No Matter Who” means those who would normally be the most confrontational about the problems with the Democrats are instead enabling them the Democrat candidates get more complacent and arrogant, they themselves become extreme and unelectable as a result.

And the more unpalatable the candidates become the less “Blue No Matter Who” works as a strategy, as voters decide to tap out rather than support candidates they find completely unacceptable.

The “lesser of two evils” becomes harder and harder to sell to the electorate if they keep getting more evil, in 2020 it successfully staved off Trump but it led to an emboldened Joe Biden in 2024, who is now destroying public confidence by continuing to run despite struggling to even form coherent sentences in public appearances, especially his recent debate with Trump, making Trump look good by comparison even as he lies through his teeth at every turn and plots to use every lever at his disposal to rig the system in his favour if he slips back into office.

Eventually the Democratic leadership did realise the damage they were causing by letting Biden stay on, although Biden initially resisted eventually the pressure became so strong that he realised the game was up and decided to pass the torch, but it was too late by then.

Biden dropped his candidacy just 4 months before the election, and quickly endorsed his Vice President, Harris, as his preferred candidate for the Democrats’ Presidential campaign and gave her his financial warchest.

The short time frame meant that very few people were willing to face a divisive leadership contest in the party, and so the vast majority of Democratic Party delegates simply switched their support from Biden to Harris, within a day she had gotten enough endorsements to have the nomination and launched her campaign.

At first it seemed like this could be a big boost to the Dems, Trump and the Republicans had put all their energy into attacking Biden, especially for his age and cognitive decline, suddenly they had a new opponent, someone younger and actually able to give a speech, this seemed like something that could take the wind out of their sails.

But ultimately she was bogged down for a number of reasons, having been Biden’s Vice President for 4 years she wasn’t able to escape the bad reputation that came with Biden, she didn’t seem like as fresh of a face as the Dems were hoping, their reputation on the economy was poor², the crisis at the Mexican border massively escalated (although the numbers of arrivals went down in 2024, it wasn’t enough to change minds), and foreign policy was a big sticking point.

Afghanistan was a disaster that, although much of it was caused by Trump mostly landed on Biden’s lap, Biden started his term at the end of it and he had to endure the images of the evacuation of Kabul, which looked eerily remnant of the Fall of Saigon in the 1970s.

Then there’s Ukraine, the inability to break the military deadlock between Ukraine and Russia left Americans exhausted by the seeming prospect of yet another forever war with dubious prospects of success, and then there was the explosion of the Israel/Palestine conflict.

Biden strongly supported Israel even as it escalated to extreme after extreme, despite calling for a ceasefire in Gaza he effectively did nothing when the Israeli leadership said no, he kept talking about his goal being a two state solution even when the Israeli Prime Minister rejected the idea, it became blatantly obvious that US foreign policy and the views of average Americans didn’t match what the Israeli government was doing, yet Biden kept supporting them anyway, although Harris was more critical of Israel than Biden she didn’t break from this policy, alienating many progressives.

She also tried to appeal to moderate Republicans, platforming Republican politicians under the slogan “country over party”, hoping that traditional Republican supporters shocked by Trump’s democratic backsliding and general repulsiveness would switch sides³, she changed some of her policies to tempt them, she dropped her pledge to ban fracking, she boasted about being a gun owner who would shoot home intruders, she touted her role as a prosecutor to use a tough on crime message against Trump the criminal; it didn’t pan out, the vast majority of Reps stayed Rep.

Then there was the fact that her campaign tried to turn the election into a referendum on Trump, in keeping with Blue No Matter Who logic they would often try and turn scrutiny on their policies into scrutiny towards Trump, I remember one particular clip where leading Democrat Pete Buttigieg was having a conversation with an undecided would-be voter who was concerned about climate and leaning towards the Green Party, someone exhausted by the two party system, and he responded to her by doubling down on the realpolitik argument of “its Harris or Trump”, trying to turn the debate from Harris vs Stein back to Harris vs Trump, it didn’t work, she kept repeating that she wasn’t voting for Trump and he kept going for the Trump talking points anyway.

The fundamental problem was that his argument was a vote for anyone other than Harris (or not voting at all) was effectively a vote for Trump while she didn’t see it that way: A vote for Trump is a vote for Trump, Harris for Harris, Stein for Stein.

Now personally, I can see the realpolitik argument here, with the USA’s heavily entrenched two party system it’s not necessarily wrong to say refusing to vote for the lesser evil can enable the greater evil, I would argue that this might be a necessity to get out of the politics of evil in the first place, but let’s set that aside for a moment. Even if you view things that way, it doesn’t matter, because you’re already convinced, if a voter doesn’t see things this way you have to argue on their terms, because no one is owed votes, you have to earn them, “but Trump” doesn’t do that.

It worked in 2020 because the US had just experienced 4 years of a Trump presidency, in 2024 there was a 4 year gap from that, if it was a referendum on anyone it was a referendum on Biden, and by extension Harris, and they lost that referendum. That sect of 6 million Biden voters who didn’t join Harris, some of them defected to Trump but a lot of them simply stayed home, they were so fed up with the establishment that they refused to save the lesser evil, rejecting the politics of evil entirely.

Considering the damage Trump could do that will be an extremely bitter pill to swallow for many, I don‘t know what I’d do in this situation, I probably would’ve put my anger aside and backed Harris because the prospect of the greater evil was just that bad, I made a similar choice in my own country’s elections this year, but I also completely understand those who refused to.

As I wrote in the original democracy article the “lesser of two evils” becomes harder and harder to sell to the electorate if they keep getting more evil, the war in Gaza is a textbook example, those who refused to vote for Harris because of the US government’s enabling of mass war crimes were told that those war crimes would get worse under Trump, they were effectively told the choice was “less genocide” vs “more genocide” and that doubling down on “no genocide” was selfish, arrogant, privileged or any other number of things, this level of blame and absurdism didn’t guilt them into coming back, it convinced them to tap out, the level of madness and absurdism had reached a fever pitch.

They were told that this election was an existential crisis, an inflection point where America could continue with a democratic experiment or devolve into a democracy, yet despite this being a crisis where everything was at stake it was always the voters who were expected to compromise for the party and not the other way around.

There was also just the fundamental problem that Kamala Harris was never a popular candidate in the first place, in the 2020 primaries (the elections parties hold to decide their candidates in the US) she didn’t even make it onto the ballot because she dropped out early, having failed to break into the top tier of leading candidates, in 2024 the primaries were a rubber stump for Harris because she was Biden’s choice and the party core felt it didn’t have the time to hold a debate, not because she actually won the votes.

The result was that the Democrats lost their lead in all their strong demographics, women, ethnic minorities, college educated, the young, it all evaporated, the only lead the party still has is with black voters, everything else is about 50/50.

That also makes this a massive blow to the Pied Piper strategy, as I mentioned in the first democracy article the Democrats repeatedly supported more extreme right wingers like Trump and his ilk in these last 3 elections, thinking that it would make the Republicans unpalatable to the electorate and force them towards the Democrats, we have the scoreboard now, 1-2, twice now this strategy has blown up in their faces and enabled the very people they were trying to stop.

But even then it’s important to put things into perspective and remember that Trump has not won over the people of the USA, he narrowly missed winning over 50% of voters and if we were to count non-voters too any idea he has a “mandate” to do anything would quickly vanish, the USA hasn’t become Trump country.

Trump will be a horrible president for all sorts of reasons, he’s corrupt as all hell, his foreign policy is erratic, he’s potentially a sexual predator and a rapist, he lies pretty much every time he opens his mouth, and he doesn’t even respect the most basic fundamentals of democracy.

When he lost the 2020 election rather than accept the result he told his supporters the vote had been rigged and that he was the true winner, he tried to convince the US Congress to accept fake election results that declared him the winner in key swing states that he lost, then when that didn’t work he tried to convince his Vice President, Mike Pence, to discard the results in those states entirely, effectively disenfranchising millions of people just to stay in power, when Pence refused he sent a mob of his supporters down to Congress to “motivate” him and the other representatives, telling them “if you don’t fight like hell you’re not going to have a country anymore” and that they couldn’t let the votes be counted, and so they stormed the building to try and stop the count.

He was making the same claims of fraud this time around as well, then when he won suddenly the results were legitimate, I guess the deep state set their alarm clocks for the wrong time?

If there wasn’t any moment beforehand that did it, Jan 6 should’ve totally discredited Trump as a Presidential candidate or a candidate for much of anything at all, there’s a very serious risk that now with his hands on the levers of power again he’ll do everything he can to take revenge on his opponents and disenfranchise voters so that a loss like the one he faced in 2020 can never happen again.

The fact that around 77 million people were willing to overlook that (or support it) is very worrying, but it’s worth remembering that US politics can always be split into 3 roughly equal thirds, the blue third, the red third and the grey third, Dems, Reps and Non-Voters.

This time the blue third was slightly smaller, the red and grey were slightly bigger, it hasn’t all turned red, that’s important to remember.

A textbook example of this is North Carolina, a state Trump narrowly won this election, in the state Governor’s elections the Democrats beat the Republicans by nearly a million votes, why? Nearly a million more people showed up. It reminds me of the example I gave from the UK in the first article, the Vale of Glamorgan, where the party balance of power was decided pretty much entirely by who showed up to the polls.

If the UK election this year was the perfect showcase for the impacts of disproportionality, the US elections are the perfect showcase for the impacts of turnout.

Back in the Channel

#

A Fog From Rhetoric

#

Speaking of the UK, turnout is having an after effect here too.

Lately the government has been going through a massive popularity crisis, the lead they had in the opinion polls before and during the election has vanished, they’re about neck and neck with the Conservatives, each on around 30% support, with Reform on 20%, the remaining 20% mostly made up by the Liberal Democrats and Greens in distant fourth and fifth place.

Really, no one is very popular, but that’s the balance of power when people are forced to pick a side.

Worst polling governments at this point post-election:

Thatcher 1979 - 40%
Blair 2005 - 39%
Cameron 2015 - 38%
*Starmer 2024 - 30%*

Worst polling opposition parties:

Hague 2001 - 30%
Major 1997 - 28%
Foot 1983 - 26%
*Sunak 2024 - 26%* https://t.co/DV8dAcysZw pic.twitter.com/QaMGGMPxq3

— Owen Winter (@OwenWntr) October 8, 2024

Labour’s plunge is down to a number of scandals, pretty much right after the election they faced a number of corruption scandals, it emerged that the Prime Minister had been taking a large number of “freebies”, gifts from wealthy influencers, his Chief of Staff Sue Gray resigned after it emerged that she was being paid an exorbitant fee, higher than the PM himself, probably not helped by the fact that she was the “independent” advisor who had investigated Boris Johnson over his rule breaking during COVID. 

To be clear, Boris was absolutely guilty, but it doesn’t look good if his investigator who brings down the government goes on to take a senior, very well paid, job with the government that replaced him.

Oh and here’s a good one, after the party denounced “vanity photographers” a few years ago Labour’s Deputy Prime Minister has turned out to be charging the taxpayer 68,000 quid a year for her own dedicated photographer, great.

Although all this was a drop in the ocean compared to the mountains of corruption under the Tories, it’s a very bad look for a party that ran a campaign on cleaning up government from sleaze and waste to then be caught knee deep in sleaze and waste.

Then there’s the messaging on the economy, the rhetoric around there being “no money left” has continued and worsened, the government has since claimed that there is a 22 billion pound “black hole” in the government’s finances that needs to be plugged, which will apparently be done through 40 billion pounds worth of tax rises unveiled in the new budget.

The problem with this is that the Office for Budget Responsibility, where the government has claimed the figure comes from, hasn’t actually backed those numbers up, and has instead given a much lower figure of around 9 billion, other estimates have fluctuated around the 13 billion mark. A lot of the 40 billion to plug the apparent hole is also going to come from increasing a tax called National Insurance, the issue there is that this is a tax Labour specifically promised not to raise in their election manifesto, they’ve said that this doesn’t count because the part of NI they’re raising is related to employers, not workers.

There’s a lot more going on behind the scenes, but that’s the core of it.

But even still a lot of people are shocked at how quickly the government’s reputation has tanked, to me it’s not actually a surprise at all. The government’s approval rating has been sat at around 18% for a while now, with polling site YouGov saying the ratings are 18% approval, 23% don’t know, 59% disapproval, there was a brief spike of “don’t know” up to around 40% after the election, then that dropped down, disapproval went up, approval went down, bringing us to where we are now.

The thing is, if you keep turnout in mind, this isn’t actually a big surprise at all.

When I did my election graphs for the addendum of the original democracy article I of course made one that included non-voters, and once non-voters were added into the stats it turned out that just 20% of the public had voted for Labour, so I thought with all that considered 18% actually isn’t that bad at all, they’ve only lost 2% of their voters and considering the amount of scandal and bad press that they’ve come into office with that’s not so bad, most of the massive drop is from people who didn’t vote for Labour anyway.

The reason it seems so crazy on the surface is because if you don’t put much attention on non-voters and only pay attention to those who voted, well, Labour got about 33% of the vote, so from there it seems like a massive drop, from 33 to 18 is a loss of 15, that’s just under half their support base vaporised in a month! When you count non-voters, you realise that that 33% was a myth to begin with, they may have won 33% of the vote, but they didn’t win over 33% of the public.

But then the plot thickens, you might remember how I mentioned in the original article that turnout is only calculated as a % of those who have registered to vote, and that in the UK registering is pretty much optional, meaning that the number of non-voters is actually even larger than turnout statistics would suggest. Well, recently the Electoral Reform Society published a report called “A System Out of Step: The 2024 General Election”, it mostly focuses on disproportionality and on predicted what the results could’ve looked like under different election systems, but one section focused on turnout and it actually gave an estimate of how many people probably have the right to vote, but haven’t registered to do so: 8,199,867. 

When I added this estimate of the unregistered to the number of non-voters on my graph guess what happened? Labour’s share of support dropped from 20% to just over 17%.

_{UK 2024 general election results - Voters, Non-Voters and Estimated Unregistered}

So actually they might not have lost many people at all, they might have even kept or gained a little bit of support from people who didn’t vote for them in the first place, undecided or opposition voters who want to give the new government a chance before they judge them too harshly, by these standards Labour are actually doing quite well for themselves given the circumstances, it’s just that in comparison to the rest of the population they never had a very large support base to begin with. It seems like Keir Starmer’s real name is Billy, Billy No-mates.

Just as a reminder, I voted Labour in that election, so this isn’t some joyful stab from someone who always hated them, I have a lot of reservations about them, the economic rhetoric, the slipperiness, and especially the fact that they’re wishy washy about whether or not we should stop giving large arms shipments to a country committing mass war crimes (something we really shouldn’t be in two minds about) whether I would vote for them again is a very open question, but I do actually think they’ve done some good things, especially around rail reform.

Just 2 months after getting in they resolved a rail strike with a pay package, the previous government had failed to do it for 2 years because they were trying to force through reforms that would’ve saved less money than the strikes were costing, rail operation is being handed back to the government which could be a gateway to badly needed reforms around things like ticketing.

In other areas Hereditary Peers, who make up a large chunk of our unelected house of Parliament, the Lords, are about to lose their “jobs”, a step towards getting rid of the whole thing, they’re pushing a new worker’s rights bill that cracks down on unjustified sackings of workers and offers more rights to those on zero hours contracts, a renter’s rights bill that will crack down on no fault evictions, rental bidding wars and other issues, and a bill to hold the water companies to account, they’ve been their enriching their shareholders while they polluted our rivers for years, now they’ll lose their bonuses if they can’t get the muck under control and if they keep behaving badly the water bosses could be liable for prison sentences. Seriously punishing the rich for their misconduct? What a novel idea!

So while the sticking points are very angering I don’t see myself as a natural enemy of the party, I’d just prefer it if they didn’t get so corrupt and made more of a clean break with people committing the worst crimes against humanity imaginable, if the Irish can do it why can’t we?

And on top of that if the Conservatives or really any other party grabbed power in this way, even a party I fully agreed with, I would still be talking about this, because I fundamentally don’t think we can call ourselves a democracy if our election system gives absolute power to a party only a tiny fragment of the public supported, even if I’m in that fragment. I would rather we have a government with genuine legitimacy, a support base genuinely rooted in the people, something we haven’t had for a generation, if ever.

There’s a lot more to democracy than this but this is the basic entry level, elementary, if we aren’t even getting that right what credibility do we have?

Which naturally brings me to the next question, how do we get on that entry level?

Speculating about Systems

#

Reintroducing PR

#

Proportional Representation of course, that’s an election system designed to closely match how people voted, as I described in the original article. As I mentioned in that article there are 2 forms of PR in use today, the Single Transferable Vote and the Party List system.

Under STV you split the country into a bunch of different electoral districts, the constituencies, each having several representatives.

Each constituency chooses their representatives through a “preference” system, you vote by numbering the election candidates, 1 for the candidate you most want to be elected, 2 for your second choice candidate and so on, in Ireland you only have to number the first preference, the numbering after that is optional, but most people do give out a lot of preferences.

What makes this system proportional is two things:

First, since each constituency has multiple representatives there’s no “winner takes all” problem, unless a political party is extremely popular they won’t be able to get a monopoly over a constituency.

Second, the preference system means very few votes will be wasted, your first preference candidate might not win, but if they don’t then your vote will go on to your second preference, if they don’t make it it’ll go on to your third preference, and so on, until eventually it reaches someone that can win.

STV is the system Ireland uses, one of just 3 countries that hold national elections in this way: Ireland, Malta and Australia.

The Party List is a much more common model around the world and it’s more simple, under the list system you either split the country into a few very large election districts or you don’t split it at all and just hold the election in a single constituency covering the whole country.

Parties put forward lists of their candidates and they’re elected based on how much of the vote they get, so say there are 100 seats up for grabs and a party has 100 candidates on their list, if they get 50% of the vote they’ll get 50 seats, filled in order by the first 50 people on the party’s list.

This system is proportional because as with STV there are lots of seats up for grabs in each constituency, no winner takes all, and while it doesn’t allow for preferences there are still very few wasted votes, the larger constituencies (or single constituency) mean that parties can get elected even on very small numbers of the vote, only those voting for very unpopular parties will lose out.

In both models as many votes as possible will count and be reflected in the final result.

I very much had a bias towards STV initially, Ireland’s use of it is the example I used for PR in the original article, a positive role model, I mentioned it as part of my ideal democratic system, although I did mention that it could have problems that List Systems wouldn’t and said there was no silver bullet for every problem.

I pretty much viewed the STV system as clearly superior to Lists, the preference voting allows voters to express the most choice possible, so that even those who don’t get the candidate they wanted can still impact the result, and the system doesn’t concentrate power in the political parties.

In a List system the parties decide the order of their lists, so they have nearly as much influence over who gets elected as the voters do, that’s really bad since it deprives voters of a lot of choice: What if they like a certain candidate from a party but not its leadership? What if a party can punish members who don’t play along with the leadership by punting them down lower on the list?

This is a fixable problem, around half of the list systems are “open list” where voters can vote for individual candidates in one way or another, overriding the order of the list and bumping up candidates they personally like, but this can lead to very complex looking ballots. Of course, a complex ballot is nowhere near as big a problem as a shitty voting system that doesn’t represent how people voted, but it is something to consider.

Then there’s the issue of independents, how does an independent candidate who doesn’t want to run as part of a party work in a list system?

The whole origin of a list system’s proportionality is from the lists, you put forward as many candidates as there are seats, the % of the vote you get determines how many of those seats you get, that doesn’t work if you run as an independent, if you run on your own it won’t matter if you get 1% of the vote, 5%, 10% or 50%, the result will be the same, it’s just you getting elected, instead of those votes getting to decide the makeup of 50 seats they only go towards 1, that’s massively disproportional.

There isn’t really a solution for this, other than just not letting independents run, which again causes a lot of choice to be lost, the only real way for independents to work in this system would be for them to band together as an alliance with a list of their own, becoming another political party in all but name.

STV by contrast can bring proportional results, with a choice between individual candidates that includes independents, under a simple form of voting, do you know your numbers from 1 to 10? Then you should be able to vote just fine under STV.

After I wrote the original democracy article I did try to challenge my bias, I went looking for arguments that promote the list model over STV, those I found weren’t very convincing: “There’s no reason to use STV”, “having independent candidates is overrated, they should just form a party!”, these were the kinds of points I came across.

Another thing I looked into was what’s been going on in Wales. Wales currently uses the AMS system, the German model, but Welsh politicians have been talking about moving to a PR system for some time, and they picked a closed party list system. The thing is though, this was a pretty terrible role model, only furthering my bias, I read into why the system was picked and the reasons weren’t good at all.

Back in 2017 the Welsh parliament commissioned a report into a new voting system for Wales, based on 10 key principles: Government Accountability, Member Accountability, Proportionality, Equivalent status [all representatives having the same role], Equivalent mandates [all representatives representing the same sized areas], Diversity, Voter Choice, Boundaries, Simplicity and Sustainability.

They outlined 3 systems that could meet these principles: STV, a “Flexible” List system (a variant of Open List) and the existing AMS model. STV was their main preference, with the sticking point that because of their diversity principles, they wanted it to be a legal rule that parties should have to have 50% of their candidates be men and 50% women, if this rule couldn’t be implemented they preferred the “Flexible List” instead, arguing that the Lists should be “Zipped” meaning that the candidates on a party’s list would constantly alternate between men and women, so if you had a woman in your no.1 place on the list, no.2 has to be a man, no.3 a woman and so on, AMS was considered a fallback “status quo” option⁴.

The important thing that stood out to me was that they emphasised that each of these options gave voters choice, although they were so keen on pushing for diversity they wanted to make 50/50 the law, ultimately voters could still choose who they wanted, just because the candidates were 50/50 didn’t mean voters were forced to vote that way. Lack of choice was specifically why they rejected a Closed List option, you couldn’t choose who you were electing and so you couldn’t hold them accountable⁵.

Ultimately though the parliament decided to ignore that advice and go with a closed list anyway⁶, their final 5 member election reform committee had 2 members following the original recommendations, preferring STV⁷ (or, failing that, an Open List) while the other 3 chose the Closed List, arguing it would “facilitate strong, cohesive political parties” and had “Potential, if combined with diversity measures, to remove the risk of one gender being favoured above the other by the electorate”⁸.

This was pretty much actually what I feared from a list system, taking choice away from voters and handing it to the parties, I support having diverse representation across the board, but especially when it comes to politics, it makes sense that an institution responsible for an entire country should represent as much of it as possible, but I don’t think that should come from force, it especially shouldn’t be placed above voter choice.

Women can represent constituents that are men and vice versa, it doesn’t make sense that a party should be forced to choose a man just because their last candidate was a woman, or the other way around, they should be able to choose for themselves which candidates they want to run and voters should be able to choose who they want to vote for.

But here’s the kicker, the Welsh government has now actually dropped those quota plans entirely because there’s a risk they might be illegal, but the new closed system is still going ahead, meaning there isn’t even a debate between diversity and choice, there’s just a loss of choice.

Don’t get me wrong, I would take a list system over a system that isn’t proportional, like the First Past the Post crapshoot we have here in the UK, not being able to choose an individual candidate is better than an individual candidate that barely anyone wanted being able to sneak in and “represent” everyone, but STV just seemed clearly like the more democratic option.

PR in Action

#

Which brings me to the Irish. They had an election recently too, I was relatively familiar with the Irish election season because Instagram’s wonky geolocation for ads meant that I kept getting promotions from Irish political parties, mostly from the lefty People Before Profit party but plenty of others as well, there were ads from the Greens, Social Democrats, the occasional independent, and the incumbent Fine Gael party.

The fine print of Irish politics is beyond me, after all I don’t live there and it’s not my country, but I was interested because it was an opportunity to see a Proportional Representation system in action, since they use STV.

On the last day of the election count I checked in on the elections page from the Irish state media RTÉ, I was impressed by how thorough they showcased the election process, they didn’t just include each constituency and the representatives it elected but it showed every count.

With how STV works there are a lot of counts for each constituency, first they set a “quota”, anyone that passes this quota with their first preference votes is immediately elected, after that the candidates with the least votes start getting removed from the contest and passed on to their next preferences, this continues until enough candidates pass the quota to fill all the seats or they run out of candidates.

To be elected, a candidate must generally reach the quota for the constituency. The last seat can be filled by a candidate who did not reach the quota if all the other candidates have been elected or eliminated.

The quota is calculated by dividing the total valid poll by 1 more than the number of available seats (if there is a number to carry over, it is ignored), and then adding 1.

For example, in a 4-seat constituency with a total valid poll of 25,000, the quota is:

25,000 (the total valid poll) divided by 5 (1 more than the number of seats), which is 5,000. Then add 1. The quota is 5,001.

The total valid poll is the total number of votes minus the number of spoiled papers.

Each time a candidate is elected or eliminated and their votes transferred, the votes have to be counted again, until everyone is elected, with RTÉ’s website you could see this playing out in each constituency, it was actually very interesting to watch.

But when I checked out the full results on Wikipedia a lot of things didn’t seem right, there were big discrepancies between votes and seats.

The top 3 parties, Fianna Fail, Sinn Féin and Fine Gael all lost votes but gained seats, Sinn Féin in particular lost almost 6% of the vote but went in with 37 seats and came out with 39, People Before Profit had a small increase in their vote share but lost 2 of their 5 seats, the Green Party managed to get 3% of the vote while the 100% Redress Party got 0.3% of the vote, 66,911 votes vs 6,862, yet they each came out with the same number of seats.

Parties failing upwards, succeeding downwards, balances of power that didn’t match how people voted, this looked just like the disproportional nightmares I’d called out in the original article when I was looking at First Past the Post in India. I started to think, was I wrong all along? Have I idealised a system that’s actually broken?

I already had a clue to how some of this could be explained, preferences, the share of the vote these parties were listed as having was the first preference vote but it’s not always people’s first preferences that are counted, sometimes their first preference doesn’t make it and their vote goes to their second, third, fourth, even fifth or sixth choice instead, but that didn’t quite settle my discomfort, so I looked closer.

I looked into the constituency where 100% Redress won their seat, Donegal, it was actually the only one they contested. Their candidate, Charles Ward, started off in 5th place with those original 7000 or so votes, but with each round as other lower scoring candidates were eliminated and their votes redistributed he gained votes, sometimes just a few, sometimes a few hundred, then several thousand, by the time the count ended he’d moved up to 4th place with over 11,000 votes.

The Greens by contrast all had their candidates get eliminated and the votes that went to them ended up getting passed on. All except for one, the party leader Roderic O’Gorman, who was elected in the Dublin West area. He started with about 2000 votes, after transfers he had around 6000 and was elected in 5th place.

So after transfers 100% Redress and the Greens went from 66,000 votes vs 7000 to 6000 vs 7000, now the balance of power makes a lot more sense.

On paper the Greens had 3% of the vote, but that 3% was divided across lots of different constituencies where it was outmatched by the vote share of the other parties.

That’s unfortunately an inevitable consequence of dividing the vote across relatively small constituencies, the much larger ones involved in a list system make this kind of thing much less likely.

The overall result was still very proportional, with only a roughly 2% difference between votes and seats in the results, far lower than the crackpot kind of results disproportional systems like FPTP can spit out, but this was an insight into the drawbacks of STV.

_{2024 Irish General Election Results - Votes and Seats}

Contrasting with this negative example of STV I came across a positive example of a list system. Me and Massi found ourselves talking at one point about a political party he found interesting: Volt Germany, a branch of Volt Europa, a centrist party that calls for the EU to unite as its own country, a United States of Europe if you like. While we were talking about Volt I noticed something interesting, they actually do have some representation in parliament, not in the German Parliament, the EU Parliament.

We found out that unlike the German national elections which have their 5% election threshold, where any party getting under 5% of the vote doesn’t get into parliament, in the German elections to the EU Parliament there’s no threshold at all, the German Supreme Court actually abolished it because it violated “the principles of equal voting rights and equal opportunities for the parties”.

Why didn’t they do that for the national parliament as well? Apparently it’s because they see a threshold as leading to greater stability, and argue stability is more important in the national parliament than the EU one, they’ve since softened their stance on the idea of an EU threshold after the EU itself pushed for one to be adopted, but so far it still hasn’t happened.

Because the EU elections don’t have a threshold Volt were able to get their first seat in the 2019 elections, where they got just 0.67% of the vote, in the latest EU elections in 2024 their vote share jumped up to 2.57%, giving them 3 seats.

I was really interested seeing this because it proved exactly what I said in the original article:

Is [the threshold] so bad? Well, systematically not too much, only parties with a rather small vote share are being screwed over by this system, but think about it this way, small parties don’t usually have the benefits of easy media attention and an established voter base that the big parties have, they need to grow those, if an electoral threshold doesn’t exist they can get into parliament with a modest but still noteworthy amount of seats, that means they’re making speeches in parliament, they have voting records that tell you what they’re really about, if they deliver for their voters attention can spread around, public confidence can grow and they can build that community, potentially being able to take on the larger parties and maybe even replace them!

If you do have a threshold, it’s much harder for that process to start, those small parties can’t get their foot in the door, which means they can’t get the exposure, which means they don’t grow, dooming small parties to irrelevance.

Now maybe some of those parties might be extremists, the kind of organisations you don’t want to get off the ground, but others might be totally reasonable or positive parties that never get to have their ideas tested, the threshold doesn’t discriminate between extremist and reasonable, just small and large, and the large parties tend to be the establishment ones.

When you don’t have this arbitrary barrier to entry smaller parties get a shot at proving themselves that they otherwise wouldn’t have gotten, Volt more than doubled their vote and seat share since the last elections and I’m sure a large part of that is down to the fact that they managed to get their foot in the door, having a representative in the EU Parliament gave them a chance to show people what they were about and that a vote for them is worth it, if there was a threshold those voters most likely would’ve deserted Volt after seeing that their vote had gone to waste.

Germany’s EU election system stands out for other reasons too, unlike the national election system the EU elections run on a straight party list system, they’re also ran in a single national constituency.

The result is a series of election results that are highly proportional, if you look at the graphs for the votes and the seats they’re practically identical, the only difference is that the very bottom scoring parties, the ones with less than .5% of the vote, get eliminated because there just aren’t enough seats to accommodate them and as a result the parties left get their share inflated by about .3% going from votes to seats, which is of course an inevitable result:

there’s only so many seats a legislature can have and not everyone can win, that does mean some votes will be “lost”, eventually if you keep dividing a number of seats by smaller and smaller percentages you’re going to get numbers lower than 1 and you can’t give a party a third of a seat or half a seat, that’s a natural limitation built into the formula, when the Monster Raving Loony Party runs and gets less than 0.1% of the vote across the entire country they’re not going to get a seat and that’s fair enough

_{2024 EU elections in Germany - Votes vs Seats}

This is something I really like, an election with very few losers where if you go out and vote you’re very likely to get someone in, so many voices that can actually be heard, it seems a lot more like democracy to me.

It’s probably one of the most proportional election systems in the world, a single nationwide constituency for elections is something very rare, it’s used for the EU Parliament in 23 of the 27 EU member states (the 4 that don’t are Belgium, Ireland, Italy and Poland if you’re curious) and on the national stage by a select few countries, including the Netherlands and Israel⁹.

I plugged the Netherlands’ latest elections into the graphs and they produced similarly proportional results, 26 parties contested the election and 15 of them got seats, the remaining 11 each got .51% of the vote or less.

_{2023 Netherlands General Election - Votes vs Seats}

Israel’s results were slightly more distorted, but that’s because their system uses a threshold of 3.25%, equivalent to 4 seats in their parliament, the Knesset, it was originally 1% but was between the 1980s and the 2010s.

_{2022 Israeli General Election - Votes vs Seats}

Despite that it might actually be more democratic than the system in the Netherlands, since the Netherlands has a second branch of its parliament, the Senate, that the public don’t get to vote for. Then again Israel is also governing several million people that don’t get a vote at all, so maybe not.

The Trade of Values

#

So, looking at this, here’s the tradeoff I’ve learnt.

If we’re assuming the constituency boundaries are fair, there are enough seats up for grabs, and there are no thresholds or any other funny business like that, a Party List model has a much lower natural threshold for candidates to be represented but votes for losers are totally binned, STV has a much higher threshold but voters for the losing candidates get a second chance through transfers.

Under both systems as many votes as possible are represented, but in different ways.

It’s basically a question of what you value more: As proportional a balance of power between the parties as possible, or as many votes counting as possible? A question of what you consider when you vote, are you voting for a party or a person?

What do you want your representative to actually represent, your local area or the country at large? Does there need to be a “constituency link” in parliament or is that something that could be handled by better local government instead?

How much do you value the ability to step out of party politics?

Those questions are a lot more nuanced than I really thought they would be, I know I definitely value the idea of parties not having a monopoly over politics, and there’s a perfect example of that that came up recently.

Normally MPs in our parliament are under something called the “whip”, which is basically the party line on how to vote, every party has its own “chief whip”, an MP whose job it is to basically tell the other party MPs how to vote, and discipline them if they don’t do as they’re told.

That disciplining can go as far as kicking an MP out of their party if their rebellion is seen as too serious. The risk of that means that many MPs will simply do what the party tells them, with the balance of power of many votes on key issues being strictly on party lines.

Sometimes however, there will be “free votes”, where parties don’t “whip” their MPs and let them decide for themselves how to vote, this recently came up with a vote on a proposed assisted dying bill (which aims to give those with a terminal illness with less than 6 months to live the right to end their own lives, subject to a process of legal safeguards) which would decide if it would move to the next stage of lawmaking or be dropped.

Because the vote was free MPs were doing a lot of soul searching, talking to activists on both sides, asking their constituents for input and reflecting on their own personal values, sometimes secular and sometimes religious, there were debates over whether this idea of a “right to die” should exist, whether the safeguards were enough to stop people making choices they might have regretted had they lived to think about it longer, or to stop the risks of coercion and exploitation, whether it was moral to pass something like this into law when the care system is in a poor state.

This debate was being given prime time news coverage, all these different groups airing out their views, my family was debating their own thoughts, this seemed more like, well, democracy.

The whole time the vote was so up in the air that nobody seemed to know how it would go, in the end 605 of the 650 members of the House of Commons, the elected half of our parliament, turned up to vote, and the vote passed by just 55 votes, a narrow majority allowing the bill to move to the next stage.

Except for the Green Party and several of the parties representing Northern Ireland every party, large and small, was split on the vote, Labour MPs were mostly in favour, Conservatives mostly against, but they each had members on both sides, the Reform Party with just 5 MPs had 3 of them vote yes and 2 of them vote no.

This was actually the first free vote since 2019, there have only been 18 of them in the last 10 years, out of 2186 overall votes in the same period, usually MPs are “whipped” and they do as they’re told, even the most rebellious MPs have only dared to defy it around 10-20% of the time in the past¹⁰.

Of course free votes and rebellions bring in their own questions, do voters want their MPs to rebel? If you voted for them as an individual, maybe you would, if you voted for them because of the party they stood for you might see it as betrayal, what is your MP’s job at the end of the day? To represent themselves? Their party? The broad, sometimes contradictory range of opinions their constituents might have? It’s complex.

And of course sometimes you do want MPs to do as they’re told, if you had rebellions on everything it would be much harder to get the daily business of politics done, but some level of independence I think is important and I think it would be harder to get that in a party list model where parties have so much more power over the system, when they already have a lot.

But on the other hand comparing the graphs of the votes and the results list systems can produce, where the seats and the votes are nearly 1:1, that’s something I see great value in, the confidence that voters can have that what they choose is what they get, that they really do hold the balance of power, that would be a very precious thing.

It’s difficult, isn’t it?

The Best of Both Worlds?

#

Is there some sort of way to bridge the gap and have the best of both worlds? Maybe.

Here in the UK we currently have 2 houses of Parliament, the elected Commons and the unelected Lords, and it has long been suggested that the Lords should be replaced by an elected body instead.

Maybe one of the houses could be voted in by STV and the other by a list system? I did think about mentioning this as a kind of blueprint when writing the original democracy article, but I didn’t in the end because I was more interested in starting a conversation about democracy than coming up with constitutional fantasies.

But hey, why not experiment right? If you’ve come this far you probably have some interest or patience to listen to my political preaching anyways.

This kind of 2 way street could be a way to ensure maximum representation, but it does come with its own drawbacks, if you have 2 different elected branches of parliament what happens if their memberships differ too much? What if they have very opposing views on certain issues and start political tugs of war over lawmaking? Right now we have a clear divide, the Commons is superior to the Lords, both in the balance of power and its legitimacy, the Commons are elected, even if by a very shitty election system, the Lords are not, and that means their say is worth less.

If there were 2 houses each equally elected, both chosen proportionally by the people, would they both have a mandate? And what do you do when they disagree?

Thorny question. Okay, so what if we somehow merged the two voting systems in a single house of parliament instead?

That’s kind of the idea behind AMS, you have the constituency representatives and the regional/nationwide representatives, AMS mixes together First Past the Post for the constituencies and the Party List system for the regional/nationwide representatives, with the number of regional reps ensuring the overall balance of power is proportional to how people voted.

Like I said when talking about Germany as an example in the original article AMS isn’t a proportional system as it is, because First Past the Post isn’t proportional, even though the end balance of power will be proportional (as long as you don’t use something like a threshold) the constituency MPs are chosen by a system that just doesn’t match how people voted, it’s only a partly proportional model.

But what if the constituency MPs were chosen by a different system instead? We’ve actually floated with this idea before, in the late 1990s a committee called the Jenkins Commission suggested a model called “Alternative Vote +”:

Voters vote once with the Alternative Vote (in which voters rank candidates in order of preference) to elect a candidate in each constituency, and vote again on a small top-up list to make the overall result more proportional.

With the top-up list, voters can either select their favourite party or choose their favourite candidate from the top-up list and the votes are then allocated to represent each party’s share of the votes proportionally

Now even this still wouldn’t be proportional because as I mentioned when talking about Australia before AV only has one seat up for grabs per constituency, making for a winner takes all system that has a lot of the same problems as First Past the Post.

But the only difference between AV and STV is that there are several seats per constituency rather than just one, so what if we had an “STV +” system instead?

You vote twice, first for your constituency representatives, numbering them with as many preferences as you like, then second for the party list, you just put a tick or a cross or something like that next to the party that you support.

Then we count the votes, elect the constituency MPs first and see how their share of the seats matches their share of the vote, for parties that don’t have enough we start handing them seats from the lists to compensate.

Something like this could have a lot of benefits, your vote would be extremely likely to count and be accurately represented in how the parliament is made up, independents could have a say if people want them in, you’ve got local representatives and a fair balance of power on the national stage, it could actually tick all the boxes (proportionality, accountability, choice) and (what I see as) the majority flaws of each the two systems would be cancelled out by the other.

STV’s problem of the (first preference) votes not being matched by the seats would be solved by the List Seats balancing things out, while the List System’s problem of voters for the lowest scoring parties being unrepresented would be solved by the STV seats: The preferences mean that your constituency vote will almost definitely count, and while some candidates inevitably have to lose because there won’t be enough seats for all of them small parties that can’t get enough of the vote to be elected on the national level with the lists could maybe be elected by running a local campaign in a constituency instead.

It could also solve the question of what you’re voting for when you vote for an MP, a person or a party? With the constituency vote you’re voting for a person, either because you trust their judgement or you think they’ll represent your own views well, with the list vote you’re voting for a party, hopefully because you think their policies are the best for the country.

But of course a model like this would have its own drawbacks too:

• Having 2 different voting systems used for the same parliament in the same election could be confusing or complex. Again, far from the most important issue, most voters can figure out how to number some boxes and draw a cross and any difficulties that could come with complexity are less important than a system that makes for a fair result, but it’s still something to consider.

• MPs would be elected by different systems with different responsibilities (Constituency MPs would be responsible for their local constituents, List MPs wouldn’t, that’s how it works in Germany for example), would that cause friction between them? Be divisive? Raise questions over which of them is more legitimate? Or would it not matter?

• Varying size. The problem with a system that relies on list seats to “compensate” for the disproportionality of a constituency system is that you don’t know how many seats you’ll need in compensation, for example Germany’s parliament changes in size every election and it usually gets bigger:

The minimum legal number of members of the Bundestag (German: Mitglieder des Bundestages) is 598; however, due to the system of overhang [when the constituency vote gives a party more seats than the list vote would] and leveling [when a party is given list seats because the constituency seats didn’t give them enough to match their share of the vote] seats the current 20th Bundestag has a total of 735 members, making it the largest Bundestag to date and the largest freely elected national parliamentary chamber in the world.

Building Blocks

#

Every system has its benefits and its problems, it’s just that some have more problems than others, any kind of PR is better than not having PR, because not having PR means the system will be erratic, unreliable and often unrepresentative. I could point to all the old examples from my own country, but another one I looked at recently was the latest election from our archenemies, the French.

They have a quite rare voting system called the Two Round System. They run the elections just like we do, First Past the Post, but the twist is that if no candidate gets a high enough % of the vote, it has to be held again, that’s called the second round.

To be elected in the first round, a candidate is required to secure an absolute majority of votes cast, and also to secure votes equal to at least 25% of eligible voters in their constituency. Should none of the candidates satisfy these conditions, a second round of voting ensues. […] Only first-round candidates with the support of at least 12.5% of eligible voters are allowed to participate, but if only one candidate meets that standard the two candidates with the highest number of votes in the first round may continue to the second round.

This system can be just as disproportionate as regular FPTP, if not more, and usually you’ll find that the leading parties have their level of representation inflated by 10 or even 20% compared to the votes, something that allows them to form a majority government even if they actually win didn’t the majority of the votes. This time however, it only gave them a boost of 5.5%, and the disproportionality wasn’t enough to benefit anyone, the French parliament is in a nearly even 3 way split between the country’s big political alliances, the far-left New Popular Front, the centrist Ensemble and the far-right National Rally, leaving the country completely paralysed, the last government they tried to form collapsed a few weeks ago.

_{2024 French General Election - First Round Votes/ Seats, Second Round Votes/ Seats, Total Seats}

So if anyone tells you that not having PR is a way of ensuring “stability”, they’re talking bollocks, if anything PR guarantees the only form of stability a voting system truly can, that what you choose is what you’ll get. Sometimes non-PR systems get close to that, sometimes they’re so far away they might as well be in the fucking Bermuda Triangle, and that’s not stability at all.

So building PR is, to me, a must, how to go about it and what values to weigh over others is a complex question, but thankfully some people have a solid idea on how to do it.

When I’ve looked to PR advocacy groups to cite or inspire this series so far I’ve usually looked to one of the big UK groups, the Electoral Reform Society, they do a lot of good work and their explainers on different election systems have been very helpful, but they have their drawbacks. They’re biased when it comes to how to do PR, they favour STV and make no secret of it, they also campaign on other issues and one of them, calling for the voting age to be lowered to 16, is one I’m really not sold on.

But there’s another group, Make Votes Matter, that’s actually much more in tune with my values, they’re a single issue advocacy group, they call for PR and nothing else, and they don’t actually label one specific system over another, what do they call for instead? Letting the people decide, they’re calling for a “Citizens Assembly” to be given the task of picking through all the competing values and systemic complexities to figure out what voting should really look like:

A citizens’ assembly is a large group of ordinary people selected in a similar way to a jury, but with care taken to ensure it is representative of the population at large. They are given the opportunity to hear from and cross-examine experts, to deliberate and reach recommendations.

And this idea has a bunch of support from UK political parties, not the biggest two of course, the ones who’ve benefited the most from the current system, but in this increasingly volatile political atmosphere for all we know they might not be the biggest for much longer. The status quo has lasted a long time, but as recent events have shown there are years where nothing changes and days where everything changes.

---

Assets

#

US Election Data Table

#

• 2024 US Presidential Election Results by State (Spreadsheet)

Election Graphs

#

UK Parliament Election 2024

#

• Voters + Non-Voters + Estimated Unregistered

German EU Election 2024

#

• Votes

• Seats won (Result)

Netherlands Parliament Election 2023

#

• Votes

• Seats won (Result)

Israeli Parliament Election 2022

#

• Votes

• Seats won (Result)

French Parliament Election 2024

#

First Round

#

• Votes
• Seats won (Result)

Second Round

#

• Votes
• Seats won (Result)

Total

#

• Total Seats

Changelog

#

• Edit 1 - 20/12/24 - Added footnote on elections with single national constituencies
• Edit 2 - 24/12/24 - Minor formatting changes, typo fixes, correction on minor party vote share in the US (it’s a slight increase from 2020 to 2024, not a drop), added embed of Pete Buttigieg talk
• Edit 3 - 25/12/24 - Minor typo fix and formatting change

Footnotes

#