Interesting and impressively ACX-esque piece of original research. Those indices are an illuminating way of categorising past and potential future elections.
I think turnout will matter more than you suggest. I think it will continue its downward trajectory and there will be even more of a "plague on both your houses" effect, as voters who resented the corruption and/or austerity under the Conservatives and thought Labour would solve it (especially those too young to have known any government but Conservative) realise it's business as usual on both counts and become disillusioned.
This, of course, makes it even more difficult to predict the outcome, as it depends on the views of the shrinking remnant who still vote, rather than of the whole country. I'm guessing it will help the three second-tier parties, as their platforms all have an element of "we represent genuine change" and they haven't had a chance to disprove that. (Although there will probably also be an effect where the remaining supporters of both of the big two parties are the most loyal and dedicated ones, with the more swing-y ones defecting to other parties and the more cynical ones defecting to "none" - which means that of those declaring Con or Lab support in the polls, a higher proportion than usual will actually vote that way.)
I also think Starmer won't last until 2029, with how early and how quickly his approval rating is plummeting and the various bribery scandals surrounding him. I think your "Starmer scrapes by" scenario will actually be "Labour scrapes by with some other leader". (Although you have much more insight than I do into what does or doesn't result in changes in party leadership, which can look a bit arbitrary from the outside, so ICBW.)
(and, on the topic of Labour austerity, I'm boggled at the winter fuel cut, after they put so much energy into warning that the Conservatives might cut it and painting lurid pictures of how many grannies would die and promising it would never happen on their watch. It seems like really clueless politics, where the bad optics far outweigh the relatively small amount of money saved, so idk what they were thinking.)
I agree turnout will matter a lot. I'm not sure though how it will matter and in which direction - or how to fit it (simply) into this model! I'd be less sure that turnout will keep decreasing: moderation plays a big part in 'permitting' low turnout and I suspect views will be stronger about Starmer - and perhaps about the new Tory leader - next election than they were about Starmer and Sunak in July.
Interesting point about whether Starmer will last. Until this decade the idea that a leader who'd won a large majority wouldn't last till the next election was almost unthinkable - but I guess now anything is possible. However well or badly he governs, I still don't think Starmer has the level of personal self-destructiveness Boris did, so personally I'd bet he stays the course.
Totally agree on the optics of the winter fuel cut (setting aside the objective merits). It might have flown as part of a broader Budget, but to do it by itself, at the same time as the union pay rises was an unforced error.
This conversation prompted me to place a (small) bet on Starmer leaving. I couldn't find anywhere offering "next 6 months" or similar, only calendar years, so I put one on 2024 and one on 2025. Obviously the former didn't happen, but I'm increasingly confident in the latter.
Election turnout is better predicted by how close the polls are than by decade. There is no reason to expect a collapse of election turnout to an extent that distorts which policies parties should pursue.
I enjoyed this. I'm not a systems / processes person, so I have to ask, how do you introduce a 'wild card' / 'events, dear boy' factor into such modelling?
Any scenario that sees the SNP as wining 43 seats at the next GE is demented and incredible in the literal sense of that term. They are finished. Try again.
There is one further scenario which is the re-alignment of the Right. Though that does look to be receding as a possibility given the low quality of the Tory leadership candidates and the apparent seriousness of Reform.
Footnote 16 not really justified -> really not justified :-)
the hard choice[s] begin to pay off
"Despite this, the Conservatives fail to capitalise on this." isn't exactly an error, but it feels like you'd forgotten how the sentence started by the time you'd got to the end.
Laboutr may be a Freudian slip :-)
"some vote[r]s still do not want to see the Tories back", unless we're anthropomorphising votes
"as to the immigrants across the Channel" I have no idea what this clause is supposed to say, but I'm confident it isn't this. 'As do' would make grammatical sense, but not contextual sense.
"Even winter" Probably 'Every winter'
"none see the Conservatives recover to above than their 2017 or 2019 vote shares" has an extraneous 'above than'
What is a thermostatic effect with regards to politics? I ought to be able to Google this, but can't get past sites on global warming, or documents using it for politics, but not defining it. From context I'm guessing it's either the idea that there's a hard divide between left and right parties, and all that's left to fight for is the share of a given half, or the idea that the public turn against whatever party is in government with a sort of 'reversion to the mean'. The second seems born out by data, and the first seems false, so maybe the second?
Thank you - all fixed! Particularly congratulations for spotting the errors which kind of made sense, but were indeed errors (every winter / even winter)!
Is your point on Footnote 16 some kind of pun on justified indents? Otherwise I'd see 'not really justified' as perfectly acceptable (indeed, with a slightly different in nuance of meaning to 'really not justified'), though I accept that possibly under 1950s grammar rules one is preferred.
Re thermostatic effect, your second hypothesis is correct, though with regards to policies as well, not just specific parties. It refers to the fact that usually, when a right-wing party is in power, people start to get more pro left wing ideas and policies, and vice-versa. This seems to work for some things (e.g. tax and benefits) but not always for others (e.g. some social issues).
The claim that "We live in an age of increasing volatility and unpredictability" is one I see made often about all sorts of fields. Believing all of those claims would be like believing you'll have more readers than the global population in 12 years. Instead I think the future always feels more volatile than the past simply because we know what happened in the past, and we don't know what's going to happen in the future. I see no reason to believe that 2024-2029 will be more volatile than 2019-2024. Reversion to the mean encourages me to think it's likely to be less.
I should have been clearer that I was talking about political volatility.
I think that the major decrease in 'tribal' voters - 'I vote [Labour/Conservative] because I always have, and my father before me' - is a long-term trend and one that is here to stay, at least in the near future. That means political volatility, in terms of people being willing to switch parties between elections, is going to keep increasing as the generation in which this attitude was more common pass away.
I recognise we might get tribal voting back again at some point but I can't see it happening in the next five years.
I agree that we won't see a return to tribal voters (which I view as a positive thing). I don't think this necessarily feeds through into swingy elections or hysterical policies. Two of your five scenarios are low volitility compared to the last 5 years.
I admit Labour have started with rather more drama than I expected or wanted (it's supposed to take at least 5 years to go from the party angry about sleaze to the party doing the sleaze! Could not even manage 5 months?!)
How volatile something is is a measure of how likely it is to change rapidly, not a claim it 100% will change rapidly, so I'm going to stand by my position that an electorate with increasingly fewer tribal voters is one that is increasingly volatile.
“The Fragmentation Index” - Bob Howard goes back to his roots as an IT guy and fixes some slow hard drives.
Interesting and impressively ACX-esque piece of original research. Those indices are an illuminating way of categorising past and potential future elections.
I think turnout will matter more than you suggest. I think it will continue its downward trajectory and there will be even more of a "plague on both your houses" effect, as voters who resented the corruption and/or austerity under the Conservatives and thought Labour would solve it (especially those too young to have known any government but Conservative) realise it's business as usual on both counts and become disillusioned.
This, of course, makes it even more difficult to predict the outcome, as it depends on the views of the shrinking remnant who still vote, rather than of the whole country. I'm guessing it will help the three second-tier parties, as their platforms all have an element of "we represent genuine change" and they haven't had a chance to disprove that. (Although there will probably also be an effect where the remaining supporters of both of the big two parties are the most loyal and dedicated ones, with the more swing-y ones defecting to other parties and the more cynical ones defecting to "none" - which means that of those declaring Con or Lab support in the polls, a higher proportion than usual will actually vote that way.)
I also think Starmer won't last until 2029, with how early and how quickly his approval rating is plummeting and the various bribery scandals surrounding him. I think your "Starmer scrapes by" scenario will actually be "Labour scrapes by with some other leader". (Although you have much more insight than I do into what does or doesn't result in changes in party leadership, which can look a bit arbitrary from the outside, so ICBW.)
(and, on the topic of Labour austerity, I'm boggled at the winter fuel cut, after they put so much energy into warning that the Conservatives might cut it and painting lurid pictures of how many grannies would die and promising it would never happen on their watch. It seems like really clueless politics, where the bad optics far outweigh the relatively small amount of money saved, so idk what they were thinking.)
I agree turnout will matter a lot. I'm not sure though how it will matter and in which direction - or how to fit it (simply) into this model! I'd be less sure that turnout will keep decreasing: moderation plays a big part in 'permitting' low turnout and I suspect views will be stronger about Starmer - and perhaps about the new Tory leader - next election than they were about Starmer and Sunak in July.
Interesting point about whether Starmer will last. Until this decade the idea that a leader who'd won a large majority wouldn't last till the next election was almost unthinkable - but I guess now anything is possible. However well or badly he governs, I still don't think Starmer has the level of personal self-destructiveness Boris did, so personally I'd bet he stays the course.
Totally agree on the optics of the winter fuel cut (setting aside the objective merits). It might have flown as part of a broader Budget, but to do it by itself, at the same time as the union pay rises was an unforced error.
This conversation prompted me to place a (small) bet on Starmer leaving. I couldn't find anywhere offering "next 6 months" or similar, only calendar years, so I put one on 2024 and one on 2025. Obviously the former didn't happen, but I'm increasingly confident in the latter.
Very interesting! I'd be surprised if he did go in 2025, but of course a bet could still be valuable if it was good enough odds!
Election turnout has mostly risen over the last 25 years, and has not changed much over the last 100 years. See p.2 of https://researchbriefings.files.parliament.uk/documents/CBP-8060/CBP-8060.pdf
Election turnout is better predicted by how close the polls are than by decade. There is no reason to expect a collapse of election turnout to an extent that distorts which policies parties should pursue.
I enjoyed this. I'm not a systems / processes person, so I have to ask, how do you introduce a 'wild card' / 'events, dear boy' factor into such modelling?
Events indeed! COVID, Ukraine and Israel/Gaza all unanticipated in 2019.
I guess I would answer that's a big reason why we have the five scenarios, rather than thinking we can say which will happen.
Any scenario that sees the SNP as wining 43 seats at the next GE is demented and incredible in the literal sense of that term. They are finished. Try again.
There is one further scenario which is the re-alignment of the Right. Though that does look to be receding as a possibility given the low quality of the Tory leadership candidates and the apparent seriousness of Reform.
new future -> near future
Footnote 16 not really justified -> really not justified :-)
the hard choice[s] begin to pay off
"Despite this, the Conservatives fail to capitalise on this." isn't exactly an error, but it feels like you'd forgotten how the sentence started by the time you'd got to the end.
Laboutr may be a Freudian slip :-)
"some vote[r]s still do not want to see the Tories back", unless we're anthropomorphising votes
"as to the immigrants across the Channel" I have no idea what this clause is supposed to say, but I'm confident it isn't this. 'As do' would make grammatical sense, but not contextual sense.
"Even winter" Probably 'Every winter'
"none see the Conservatives recover to above than their 2017 or 2019 vote shares" has an extraneous 'above than'
What is a thermostatic effect with regards to politics? I ought to be able to Google this, but can't get past sites on global warming, or documents using it for politics, but not defining it. From context I'm guessing it's either the idea that there's a hard divide between left and right parties, and all that's left to fight for is the share of a given half, or the idea that the public turn against whatever party is in government with a sort of 'reversion to the mean'. The second seems born out by data, and the first seems false, so maybe the second?
Thank you - all fixed! Particularly congratulations for spotting the errors which kind of made sense, but were indeed errors (every winter / even winter)!
Is your point on Footnote 16 some kind of pun on justified indents? Otherwise I'd see 'not really justified' as perfectly acceptable (indeed, with a slightly different in nuance of meaning to 'really not justified'), though I accept that possibly under 1950s grammar rules one is preferred.
Re thermostatic effect, your second hypothesis is correct, though with regards to policies as well, not just specific parties. It refers to the fact that usually, when a right-wing party is in power, people start to get more pro left wing ideas and policies, and vice-versa. This seems to work for some things (e.g. tax and benefits) but not always for others (e.g. some social issues).
Footnote 16 was a joke (hence the smiley) - I was indicating greater doubt in the validity of this assumption than you were!
The claim that "We live in an age of increasing volatility and unpredictability" is one I see made often about all sorts of fields. Believing all of those claims would be like believing you'll have more readers than the global population in 12 years. Instead I think the future always feels more volatile than the past simply because we know what happened in the past, and we don't know what's going to happen in the future. I see no reason to believe that 2024-2029 will be more volatile than 2019-2024. Reversion to the mean encourages me to think it's likely to be less.
I should have been clearer that I was talking about political volatility.
I think that the major decrease in 'tribal' voters - 'I vote [Labour/Conservative] because I always have, and my father before me' - is a long-term trend and one that is here to stay, at least in the near future. That means political volatility, in terms of people being willing to switch parties between elections, is going to keep increasing as the generation in which this attitude was more common pass away.
I recognise we might get tribal voting back again at some point but I can't see it happening in the next five years.
I agree that we won't see a return to tribal voters (which I view as a positive thing). I don't think this necessarily feeds through into swingy elections or hysterical policies. Two of your five scenarios are low volitility compared to the last 5 years.
I admit Labour have started with rather more drama than I expected or wanted (it's supposed to take at least 5 years to go from the party angry about sleaze to the party doing the sleaze! Could not even manage 5 months?!)
How volatile something is is a measure of how likely it is to change rapidly, not a claim it 100% will change rapidly, so I'm going to stand by my position that an electorate with increasingly fewer tribal voters is one that is increasingly volatile.