RelationDigest

Sunday, 7 July 2024

#GE24 – the end of “swing”…

"Swing" is a term now used by some simply to mean the change in vote share for a particular party, but actually that is not really what it was designed to measure. "Swing", in fact, is specifically the change in vote share of one party versus another, a…
Read on blog or Reader
Site logo image Ian James Parsley Read on blog or Reader

#GE24 – the end of "swing"…

By ianjamesparsley on 7 July, 2024

"Swing" is a term now used by some simply to mean the change in vote share for a particular party, but actually that is not really what it was designed to measure. "Swing", in fact, is specifically the change in vote share of one party versus another, and for many years it was measured on the famous BBC "swingometer". In 2024, however, it was rendered tragically redundant.

For example, in 1945 the Conservatives "and Allies" (assorted friendly Liberals and Unionists) received 36.2% of the vote across the UK, while Labour received 49.7%; the outcome in the Commons was a Labour lead over the Conservatives of 196 seats, and over all other parties combined (the "overall majority") of 146. At the next election in 1950, Conservatives "and Allies" received 43.4% and Labour 46.1%, an altogether closer run thing with the Conservatives up 7.2 points and Labour down 3.6; this reduced Labour's lead in seats over the Conservatives to 17 and over all other parties combined to just 5. This rise of 7.2 points for one major party and decline of 3.6 for the other is collectively 10.8, which is then divided in two (to account for one going one way and one the other) to provide a "swing" of 5.4%. Fundamentally, this means that on average 5.4 people in every 100 voted Labour in 1945 but then voted Conservative in 1950; of course, it is not as simple as this - some may not have voted and one occasion but done so on another, some may have switched to a different party, and so on - but the net effect is as described.

In 1951, for example, Labour's vote actually rose 2.7 points to 48.8%, but the Conservatives' rose 4.6 points to 48.0%; this left Labour still ahead in terms of total vote, yet the net swing of almost 1% (i.e. the net effect of nearly one in every 100 voters switching from Labour to Conservative) gave the Conservatives in the Commons a lead over Labour of 26 seats and an overall majority of 17. In 1955 a swing of just over 2% to the Conservatives put them ahead in terms of votes and extended their lead in seats to 67 over Labour and and overall majority of 58. Yet the concept, at this stage, had not yet been explained.

The 1959 election was covered by the BBC much as it still is, and it was on this programme that the concept of swing, and (Sir) David Butler's famous phrase "If the whole country behaves like [Constituency X], we can predict that..." first became popularised. In the era before exit polls, the election programme consisted of waiting for the first results and then extrapolating what they meant for the whole of Great Britain. In 1959, Labour had hopes of overturning the Conservatives' comfortable majority and winning; there were certainly few who predicted anything other than a close result. It was then that David Butler explained his concept; should around 2 in 100 people choose to switch from Conservative to Labour, Labour may expect to take over in government. Yet the first results coming in showed almost precisely the opposite; and "Butler" (they preferred surnames in those days!) explained that should the whole country behave as the early constituencies were behaving, the swing the other way would mean a Conservative overall majority of over 100. As it happened, constituencies in the North West and Scotland did show a swing to Labour, while the rest of the country swung to Conservative, leaving an overall majority of exactly 100. Until 1970 and, realistically, 1992, "swing" became the way of predicting election outcomes early on election night, as it was explained that if the whole country behaved like Guildford, Cheltenham, Salford or Torbay (invariably the early declarers) and matched their "swing" one way or the other, then this would be the outcome across Great Britain in terms of seats in the Commons. This could lead to drama (as in 1964 when the outcome was right on the boundary between an overall majority for Labour or Labour being just short of an overall majority all night); debate (as in 1970 when Bob McKenzie used his "swingometer", a type of hanging arrow, to declare "on this evidence I would not hesitate to say the government is defeated" as Harold Wilson's Labour unexpectedly lost power); uncertainty (as in February 1974 when the result was a hung parliament but also when the "swingometer" struggled with a dramatic rise in the Liberal and Scottish Nationalist vote) and controversy (as in October 1974, when the BBC survey of voters suggested a large Labour majority but early "swing" suggested it would be a very small one if it even existed at all). Nevertheless, at least for England and Wales, it remained the case that you could extrapolate from the "swing" in votes to or from the Conservatives to or from Labour what the overall outcome was likely to be.

Until 2024, when the electoral outcome decisively lost its "swing"...

At this year's General Election, it was projected that Labour would need a uniform swing of 12.5% to secure an overall majority in the Commons, after the Conservative majority from 2019 was raised notionally on new constituency boundaries to 92. In other words, to turn that around into a Labour overall majority of just one seat, Labour would need 12 and a half people to switch from voting Conservative in 2019 to Labour in 2024 - or, at least, that net effect. Early in the night, however, it became apparent that the swing to Labour was settling at an average of around 11%, enough to be lead party in the Commons but definitely not enough to secure an overall majority if applied universally. If the whole country behaved like Houghton and Sunderland South or like Newcastle Central and West, there would be a hung parliament.

The thing is, the whole country did on average behave like those places. The swing from Conservative to Labour was 10.8%, enough for a comfortable Labour lead over the Conservatives, but not for an overall majority in the Commons. So how did Labour end up with an overall majority of 174 seats? Because the whole concept of "swing" collapsed...

In Labour-held seats voters, largely not particularly enamoured with the party, did not bother to vote for it in large numbers (this even applied to the new prime minister's own constituency); typically, its vote share actually fell in those constituencies and the swing from Conservative was scant (not least because there often was not much of a Conservative vote to get a swing from anyway). Added to this, even in Conservative-held seats where Labour was not the main challenger, voters turned away from Labour and instead voted for the most likely winner - most often the Liberal Democrats, but in some instances the Greens, an independent, Plaid Cymru or perhaps (in more Leave-orientated areas) Reform UK. In some instances, voters could not even decide who the main challenger was but, for as long as it was not clearly Labour, the Labour vote share generally fell (bizarrely, amid the mayhem from their point of view, the Conservatives even gained the Leicester East seat as voters also turned away from the Labour candidate but the competing vote split between a well known independent and a raft of other parties and assorted others leaving the Conservative ahead). However, in Conservative-held seats where Labour was the main challenger, the swing generally exceeded the necessary 12.5% - and, because it was only actually in these constituencies that this actually mattered to the outcome in terms of seats, that gave Labour a huge overall majority.

Therefore, the "whole country" simply does not behave like anywhere in particular any more. As a result, the whole concept of "swing" has, sadly, had its day...

Comment
Like
You can also reply to this email to leave a comment.

Ian James Parsley © 2024.
Manage your email settings or unsubscribe.

WordPress.com and Jetpack Logos

Get the Jetpack app

Subscribe, bookmark, and get real‑time notifications - all from one app!

Download Jetpack on Google Play Download Jetpack from the App Store
WordPress.com Logo and Wordmark title=

Automattic, Inc.
60 29th St. #343, San Francisco, CA 94110

at July 07, 2024
Email ThisBlogThis!Share to XShare to FacebookShare to Pinterest

No comments:

Post a Comment

Newer Post Older Post Home
Subscribe to: Post Comments (Atom)

Three Things #4

1. ͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏     ­͏   ...

  • [New post] Wiggle Kingdom: April Earnings on Spring Savings!
    Betsi...
  • [New post] Balancing the ‘E’ and ‘S’ in Environment, Social and Governance (ESG) crucial to sustaining liquidity and resilience in the African loan market (By Miranda Abraham)
    APO p...
  • Something plus something else
    Read on bl...

Search This Blog

  • Home

About Me

RelationDigest
View my complete profile

Report Abuse

Blog Archive

  • August 2025 (39)
  • July 2025 (59)
  • June 2025 (53)
  • May 2025 (47)
  • April 2025 (42)
  • March 2025 (30)
  • February 2025 (27)
  • January 2025 (30)
  • December 2024 (37)
  • November 2024 (31)
  • October 2024 (28)
  • September 2024 (28)
  • August 2024 (2729)
  • July 2024 (3249)
  • June 2024 (3152)
  • May 2024 (3259)
  • April 2024 (3151)
  • March 2024 (3258)
  • February 2024 (3046)
  • January 2024 (3258)
  • December 2023 (3270)
  • November 2023 (3183)
  • October 2023 (3243)
  • September 2023 (3151)
  • August 2023 (3241)
  • July 2023 (3237)
  • June 2023 (3135)
  • May 2023 (3212)
  • April 2023 (3093)
  • March 2023 (3187)
  • February 2023 (2865)
  • January 2023 (3209)
  • December 2022 (3229)
  • November 2022 (3079)
  • October 2022 (3086)
  • September 2022 (2791)
  • August 2022 (2964)
  • July 2022 (3157)
  • June 2022 (2925)
  • May 2022 (2893)
  • April 2022 (3049)
  • March 2022 (2919)
  • February 2022 (2104)
  • January 2022 (2284)
  • December 2021 (2481)
  • November 2021 (3146)
  • October 2021 (1048)
Powered by Blogger.