UK By-elections thread, 2021-

[Filuwaúrdjan]

The main problem for Labour would be the reason for the vacancy: Hill has been accused of sexual harassment and faces an Employment Tribunal relating to the allegations later this year. Given that the accusations were known at the time of the last election he really should not have been allowed to stand again.

Anyway, this is a very odd town with very odd and not always entirely predictable politics. Labour have the biggest base and are the best organised party locally and will benefit from the fact that opposition voters are more likely to turn out in by-elections. Assumptions that you can just add up the Conservative and Brexit Party votes from the last GE and project onto a by-election are... silly... but I presume the Conservatives will make an effort (or at least run a noisy campaign to the effect) and hope that the oddities of the constituency break their way. Quite what to expect from the artists formerly known as UKIP/the Brexit Party I'm not entirely sure, but, again, presumably some effort just because of past performances and local government strength.* Which is the other issue: there might (although this isn't certain) be various independent runs from various egotistical local players that might be worth a few thousand votes, or not.

_{*What isn't good for them is that their issues are either dead (Brexit) or don't poll particularly well (lockdown scepticism), which doesn't seem like good news from a motivating-your-electorate perspective. But we shall see.}

[Filuwaúrdjan]

And to add the circumstances of his selection are likely to ruffle some feathers... which at best will just mean a few less hands to deliver leaflets, but at worse could theoretically lead to some pissed off members voting for a some quasi independent bid.

We shall see and all that, but as far as contentious selections go, this one doesn't seem to register that high on the scale - it tends to be impositions and exclusions that rankle, and here it looks as if Nick Brown and the CLP Exec came to an arrangement.

[Filuwaúrdjan]

Burnham amd Street, as you correctly point out, are much bigger names.  The role is as big as you choose to make it I guess.

Even there Burnham is much better known than Street, whose profile within the West Midlands is less than generally presumed by political journalists. And, of course, Burnham was already a figure with a national profile. Part of the issue is that these posts are actually very weak - they can lobby Westminster and stalk local news cameras, but they do not decide on fundamental matters.

[Filuwaúrdjan]

Reform Party officially announces that they will stand in Hartlepool.

It was either that or announce that they were shutting up shop.

Also confirmed there will be a second Westminster byelection (both almost certainly on 6 May) for Airdrie and Shotts, where the sitting SNP MP is standing aside so they can run for Holyrood. Looks fairly safe for the Nats on paper, though their majority over Labour fell to under 200 at the 2017 GE.

Running the by-election on the same day as the Holyrood elections is probably the best way to minimise any threat from Labour there, so would be amazed at any other date being picked.

[Filuwaúrdjan]

Local politics rarely influence national voting...

It can do, but usually only when the council in question has done something exceedingly unpopular, bad enough to damage the party brand locally. Tends not to work the other way - c.f. a number of particularly horrible Labour performances in '19 happened in areas with quite popular Labour councils!

[Filuwaúrdjan]

It now looks likely that Airdrie/Shotts will be on May 13th, a week after the main elections. Holding it on the same day apparently fell foul of electoral regulations in this Covid period.

An amusing miscalculation there.

[Filuwaúrdjan]

Focaldata are just... not very good and so not really worth paying attention to. There's a real and stupid greed for data at all costs in what passes for electoral analysis in this country at the moment, which often leads to more credence being given to things than is an entirely good idea.

[Filuwaúrdjan]

If Labour is to win a majority at any time in the non-distant future it would need to win a lot of seats everywhere.* Not in a position to pick and choose, the numbers do not work and can't. But that's for the future.

*Of course it happens that most of this country is actually quite similar and is increasingly so and that this actually explains much of our tediously volatile politics these days.

[Filuwaúrdjan]

Well, Williams has strong ties to the constituency and represented another Teesside constituency even if it doesn't share a boundary, so that would ordinarily count as quite local. Dawson represented a constituency in northern Lancashire, but is actually from the North East - except that he is in no sense local as he is from and based in Northumberland. Walker's former constituency might as well have been in Cornwall.

[Filuwaúrdjan]

Oh right - I assumed she was a teacher in Colne Valley. I expected too much!

She was, but very early in her career she was in Stockton for a while. Not that it matters, much like her political career.

I mean, I'm sorry but 'headteacher who took earlier retirement to become a consultant'...  vomit.

[Filuwaúrdjan]

Distinct rumblings that not all Tories are happy with their chosen candidate in Hartlepool.

A curious choice on paper, certainly.

[Filuwaúrdjan]

Tbf the Tories picked quite a few people like that in 2019 in seats that they still won from Labour - the likelihood remains strong that by-elections are a different beast tho.

There are circumstances when it genuinely doesn't matter who you run (and as you say the list of new MPs from 2019 features a lot of such people...), there are circumstances when it absolutely does. It's like bad behaviour from incumbents. Most of the time you'd never notice electorally, but isn't it funny how often incumbents so tarred have a habit of losing when things turn bad for their party in general?

[Filuwaúrdjan]

Reached the height of her career as David Cameron's Welsh Secretary 2010-12. She did not particularly distinguish herself in the role, but had a higher profile than any of the four (!) subsequent Welsh Secretaries, despite representing a constituency in England (though she was Welsh).

[Filuwaúrdjan]

Numbers like that with a very small sample size in a constituency poll in a by-election would normally just say 'competitive', particularly a month away from the election itself.

However... er... given who commissioned the poll and some of the 'issues' with this company's constituency polling at the last GE, it might be interesting to see the internals. There are a few things that smell odd as it is.

[Filuwaúrdjan]

Oh dear Lord it turns out that the effective sample size was 302.

[Filuwaúrdjan]

though I'd hope people are more accurate in their recall of a Westminster vote -- but without the tables we don't know.

So the internals are online and I've seen them. They are... er... well... problematic. I would suggest that we ignore this poll completely. Even if it ends up being 'right' it will only have done so by complete accident.

[Filuwaúrdjan]

Would be interested to know why... I’m quite thick when it comes to polling!

There is a wide range of red flags - again, the 'effective sample size of 302' is enough of one in itself - but this will do as a starter:



Top is the unweighted sample, bottom the weighted sample. A quick comparison to the actual GE results in the constituency may be instructive.

[Filuwaúrdjan]

Survation have argued the recalled 2019 BxP figure is plausible and goes with previous precedent. But the fact remains that dropping from 28% to 3% is of a different order from (say) 5% to 1%.

It is absolutely not plausible. And, of course, it is hardly the only problem with the GE recall figures there.

[Filuwaúrdjan]

Oh joy this poll is leading to more battles in the forever war... with the CWU claiming the poll is fine as national polls are only based on slightly bigger samples.

Hahahahahahahhahaha

[Filuwaúrdjan]

The UK press seems filled with stories about how the Tories are going to win Hartlepool and interviews with locals confirming that people are fed up with being taken for granted by Labour and are voting Tory but won't say so publicly etc. Is that just because it's the most exciting narrative or is this race actually trending Conservative now?

The question to ask yourself is how easy it is likely to be, given the current restrictions (even easing as they are), to even find people happy to speak at great length to journalists about their political preferences. Or even how easy it is for journalists to actually get there at the moment.

Press narratives about by-elections are generally not based on much. Journalists mostly do not know the first thing about the places reported from and spend most of the (usually very limited) time spent in the constituency finding anecdotes or other information that supports the line they've already decided to to take. Sometimes they end up being 'correct', but sometimes they are miles out - there was a press consensus, for instance, that the Oldham West & Royton by-election was going to be very tight between UKIP and Labour and that UKIP were perhaps even slight favourites.

Though by-elections are hard to get a handle on - even for the campaigns! - at the best of times, and these are not those. A lot of the usual tools used to gauge what is happening are less reliable than normal or flat-out off the table.

[Filuwaúrdjan]

Oldham East was a bit of an odd one actually - its easy to poke fun at the media "hot takes" given the actual result (and I did it myself) but the predominant feeling amongst Labour people *on the ground* in the days before the vote was that it was very likely going to be close.

Labour's big win was foreseen by genuinely few.

Not that odd because results do quite often come as a surprise to the various campaigns!

[Filuwaúrdjan]

Yes, there's basically no reason to take it particularly seriously (the questionable details are bad enough, but the effective sample size of three hundred is...), but you obviously wouldn't want to see numbers like that if you were the defending party regardless.

[Filuwaúrdjan]

Two things to note, neither of which should be seen as predictive, just as things to bear in mind. The first is that given the amount of resources that can now be concentrated on by-election campaigns (and that parties are happier to take the risk than they were) it seems likely that government by-election gains are not as generically freakish as they became after the early 1960s. The second is that this is really not a typical Northern post-industrial town politically. It has a long-term tendency towards electoral idiosyncrasy that in England at least can really only be matched by South Dorset (with a couple of other places - Mansfield for instance - moving up the list quite fast, admittedly) and a very distinctive local political culture.* An example to highlight would be that while in most constituencies that can be described as 'Northern' and 'post-industrial' Labour posted increases in vote-share and majority in 2015, both dropped back sharply in Hartlepool due to a very concerted UKIP challenge. It is a place apart, for better or worse.

_{*E.g. A big feature of North East politics for about fifty years now has been very strong grassroots female involvement in local Labour parties, which, to say the least, has not exactly been the case in Hartlepool.}

[Filuwaúrdjan]

Anyone who has ever canvassed in seats outside of major campaign areas (London, Bristol, Exeter etc) knows how worthless the promise rate is. They don’t get canvassed enough and a lot of it will be people saying they’d vote Labour in 2015 or before, they could have voted Tory twice since then. Sure the Tories could still win, but this is hardly damning for Labour (as the article admits).

Similar stories have floated around before other by-elections - including Oldham West & Royton - and generally it isn't indicative of anything much. One issue also is simply that turnout in by-elections is generally rather low.

[Filuwaúrdjan]

I think a lot of people need a holiday.