Political Fantasy: What If?

[Deleted]

Duverger's Law is a funny thing

[islington]

Apr 22, 2022 10:59:47 GMT @conservativeestimate said:

Duverger's Law is a funny thing

Indeed, and it tells you a couple of things.

• Don't come third.
  
• If you must come third, don't have your vote too evenly distributed.

The Alliance broke both those rules in 1983 (and 1987), just as its (more or less) predecessors the Liberals did at GEs in the 1920s, with similar consequences each time.

[nobodyimportant]

If I calculated it correctly, 3% of the voters in 1983 switching from Con to Alliance and 3% from Con to Lab in every seat would have resulted in the following changes to the results:
CON -68
LAB +50
ALL +15
SNP +2
PC +1

This gives the Thatcher government 329 out of 650, a majority of 8. It only takes a slightly unfavourable distribution of the switchers for the Conservatives to lead to a hung parliament. The seat that would lose Thatcher her majority would be Cheltenham, which after this shift would have a Conservative majority of 323 votes over Alliance.

[polupolu]

Apr 22, 2022 10:09:51 GMT Pete Whitehead said:

Apr 22, 2022 9:56:49 GMT polupolu said:

Opinion polls around the time of Argentine invasion (https://en.wikipedia.org/wiki/Opinion_polling_for_the_1983_United_Kingdom_general_election)

28th Feb 34, 34, 30 (Tory, Labour, Alliance)

15th Mar 31.5, 33, 33

31st Mar 35, 30, 33

2nd Apr Argentine invasion

12th Apr 31.5, 29, 37

14th Apr 33, 34, 30

...

26th May 51, 25, 23

31st May 48, 28, 24

14th Jun End of war

14th Jun 45, 25, 28.5

23rd Jun 51, 24, 23

With one outlier, the Tories had never been below 27% at the highest point in their unpopularity.

After the war they were rarely below 40% (indeed only just after the Bermondsey byelection) and were generally significantly higher.

The electoral calculus is absolutely brutal within that 30/30/30 sort of range, with small shifts in opinion meaning potential huge shifts in seat numbers.

Assume the Falkands factor meant 3% of those voting Tory in 1983 would otherwise have voted Labour and the same from the Alliance (not a big stretch I suspect) and I think most models will give you a hung parliament instead of a landslide.

Add in a equivalently worse set of local elections in May 1983, and in the byelections and you get an interesting situation

I doubt it. The Alliance vote was incredibly inefficiently distributed in 1983 - if that (effectively) 3% swing from Conservative to Alliance were applied uniformly it would result in very few seats changing hands. Happy to do the maths later to confirm (when I have access to a spreadsheet I can work from) but just from memory I know that there were not all that many seats where the Alliance were within 6% of gaining a Tory seat (more seats will be vulnerable to Labour).
As you say though it is a political fantasy thread but its a bit sad that you're still engaging in a forty year old fantasy that David Steels exhortations at the 1981 conference were based on any kind of reality

What effect would you guess the Falklands war had on voting patterns in the 1983 election?

I am suggesting a scenario of -6%/+3%/+3% Tory/Lab/All i.e. 36.4/30.6/28.4

I think the Alliance would have come a very poor third in the number of seats but the Tories would have lost their majority - but I would be interested in your spreadsheet as I have had to rely on more recent models.

You are clearly correct that an inefficient spread is a big factor in the number of seats; but it begins to be less of a factor as you edge beyond 30/30/30.

It might also be worth rememebring what the satisfaction ratings of Mrs Thatcher and the Government were before the Falklands:

36% satisfied with Mrs Thatcher; 29% with the Government

Consider also a few quotes:

November 1980, Geoffrey Howe: ‘the fall in output is bottoming out’;
December 1980, John Nott: ‘I think we have reached the bottom of the recession’;
January 1981, Nigel Lawson: ‘all the signs suggest that we have now more or less reached the bottom’;
May 1981, Michael Heseltine: ‘there is tangible evidence that the worst is over’;
June 1981, Margaret Thatcher: ‘the recession has just about reached the bottom’;
October 1981, Geoffrey Howe: ‘there are some clear signs that the Government’s economic policy is beginning to work ... we have passed the end of the beginning’;
January 1982, Norman Tebbit: ‘we are beginning to see signs of our policies working’;

How much economic credibility did the government have at that point? 

[carlton43]

Apr 22, 2022 11:55:25 GMT nobodyimportant said:

If I calculated it correctly, 3% of the electorate in 1983 switching from Con to Alliance and 3% from Con to Lab in every seat would have resulted in the following changes to the results:
CON -68
LAB +50
ALL +15
SNP +2
PC +1

This gives the Thatcher government 329 out of 650, a majority of 8. It only takes a slightly unfavourable distribution of the switchers for the Conservatives to lead to a hung parliament. The seat that would lose Thatcher her majority would be Cheltenham, which after this shift would have a Conservative majority of 323 votes over Alliance.

Or to a majority of 38 which IMO would have been far more likely.

[Deleted]

Apr 22, 2022 11:55:25 GMT nobodyimportant said:

If I calculated it correctly, 3% of the electorate in 1983 switching from Con to Alliance and 3% from Con to Lab in every seat would have resulted in the following changes to the results:
CON -68
LAB +50
ALL +15
SNP +2
PC +1

Again, this counterfactual leads to the scenario of Benn holding his seat. Would we have seen a proto-Corbynite leadership from 1983 to 1987 under Benn? Would Benn have stayed on in '87 when Labour lost again?

[The Bishop]

Some serious political scientists think the Falklands War made little or no difference to the outcome of the 1983 GE.

But most of those who were around and remember it will disagree. Even if the Tories had moved back into a small polling lead on its outbreak, they also lost a high profile by-election to the SDP just a week before the Argentinian invasion - their position was far from secure. Three months later, it was (and Thatcher's own ratings transformed)

[Pete Whitehead]

Apr 22, 2022 12:07:23 GMT polupolu said:

Apr 22, 2022 10:09:51 GMT Pete Whitehead said:

I doubt it. The Alliance vote was incredibly inefficiently distributed in 1983 - if that (effectively) 3% swing from Conservative to Alliance were applied uniformly it would result in very few seats changing hands. Happy to do the maths later to confirm (when I have access to a spreadsheet I can work from) but just from memory I know that there were not all that many seats where the Alliance were within 6% of gaining a Tory seat (more seats will be vulnerable to Labour).
As you say though it is a political fantasy thread but its a bit sad that you're still engaging in a forty year old fantasy that David Steels exhortations at the 1981 conference were based on any kind of reality

What effect would you guess the Falklands war had on voting patterns in the 1983 election?

I am suggesting a scenario of -6%/+3%/+3% Tory/Lab/All i.e. 36.4/30.6/28.4

I think the Alliance would have come a very poor third in the number of seats but the Tories would have lost their majority - but I would be interested in your spreadsheet as I have had to rely on more recent models.

You are clearly correct that an inefficient spread is a big factor in the number of seats; but it begins to be less of a factor as you edge beyond 30/30/30.

It might also be worth rememebring what the satisfaction ratings of Mrs Thatcher and the Government were before the Falklands:

36% satisfied with Mrs Thatcher; 29% with the Government

Consider also a few quotes:

November 1980, Geoffrey Howe: ‘the fall in output is bottoming out’;
December 1980, John Nott: ‘I think we have reached the bottom of the recession’;
January 1981, Nigel Lawson: ‘all the signs suggest that we have now more or less reached the bottom’;
May 1981, Michael Heseltine: ‘there is tangible evidence that the worst is over’;
June 1981, Margaret Thatcher: ‘the recession has just about reached the bottom’;
October 1981, Geoffrey Howe: ‘there are some clear signs that the Government’s economic policy is beginning to work ... we have passed the end of the beginning’;
January 1982, Norman Tebbit: ‘we are beginning to see signs of our policies working’;

How much economic credibility did the government have at that point? 

The point islington was making (correctly) was that the Conservatives were recovering in the polls before the Falkands war and in particular the Alliance bubble had already been burst by then. No doubt that the Falklands war gave a big and immediate short term boost (which subsequently settled down a bit) but the trajectory was already moving in their direction, as tends to happen as one moves from the mid-term towards the end of a term (even if the subsequent elections is lost - compare 1997 with 1995 or 1970 with 1968). The configuration of the polls in early 1982 was very similar to that in early 1986 and it didn't take another Falklands war to deliver a Conservative landslide the next year.

[The Bishop]

Actually the 1983-87 parliament is an interesting one as for much of that time the Tories weren't in that great a place electorally, and as far into the schedule as September 1986 it looked like they might struggle to get another majority - and then they totally turned things around in the following nine months. I don't think that has been achieved as smoothly ever since, and it should maybe be looked into more than it has been.

[nodealbrexiteer]

Apr 22, 2022 12:51:31 GMT The Bishop said:

Actually the 1983-87 parliament is an interesting one as for much of that time the Tories weren't in that great a place electorally, and as far into the schedule as September 1986 it looked like they might struggle to get another majority - and then they totally turned things around in the following nine months. I don't think that has been achieved as smoothly ever since, and it should maybe be looked into more than it has been.

Wasnt there that relaunch and a campaign around the conference about the 'next moves forward'?

[nodealbrexiteer]

Apr 22, 2022 12:07:23 GMT polupolu said:

Apr 22, 2022 10:09:51 GMT Pete Whitehead said:

I doubt it. The Alliance vote was incredibly inefficiently distributed in 1983 - if that (effectively) 3% swing from Conservative to Alliance were applied uniformly it would result in very few seats changing hands. Happy to do the maths later to confirm (when I have access to a spreadsheet I can work from) but just from memory I know that there were not all that many seats where the Alliance were within 6% of gaining a Tory seat (more seats will be vulnerable to Labour).
As you say though it is a political fantasy thread but its a bit sad that you're still engaging in a forty year old fantasy that David Steels exhortations at the 1981 conference were based on any kind of reality

What effect would you guess the Falklands war had on voting patterns in the 1983 election?

I am suggesting a scenario of -6%/+3%/+3% Tory/Lab/All i.e. 36.4/30.6/28.4

I think the Alliance would have come a very poor third in the number of seats but the Tories would have lost their majority - but I would be interested in your spreadsheet as I have had to rely on more recent models.

You are clearly correct that an inefficient spread is a big factor in the number of seats; but it begins to be less of a factor as you edge beyond 30/30/30.

It might also be worth rememebring what the satisfaction ratings of Mrs Thatcher and the Government were before the Falklands:

36% satisfied with Mrs Thatcher; 29% with the Government

Consider also a few quotes:

November 1980, Geoffrey Howe: ‘the fall in output is bottoming out’;
December 1980, John Nott: ‘I think we have reached the bottom of the recession’;
January 1981, Nigel Lawson: ‘all the signs suggest that we have now more or less reached the bottom’;
May 1981, Michael Heseltine: ‘there is tangible evidence that the worst is over’;
June 1981, Margaret Thatcher: ‘the recession has just about reached the bottom’;
October 1981, Geoffrey Howe: ‘there are some clear signs that the Government’s economic policy is beginning to work ... we have passed the end of the beginning’;
January 1982, Norman Tebbit: ‘we are beginning to see signs of our policies working’;

How much economic credibility did the government have at that point? 

I remember a book from the 80s simulating Alliance seat totals is they went about 35 and 40% then they starting winning seats in waves

[polupolu]

Apr 22, 2022 12:19:47 GMT Pete Whitehead said:

Apr 22, 2022 12:07:23 GMT polupolu said:

What effect would you guess the Falklands war had on voting patterns in the 1983 election?

I am suggesting a scenario of -6%/+3%/+3% Tory/Lab/All i.e. 36.4/30.6/28.4

I think the Alliance would have come a very poor third in the number of seats but the Tories would have lost their majority - but I would be interested in your spreadsheet as I have had to rely on more recent models.

You are clearly correct that an inefficient spread is a big factor in the number of seats; but it begins to be less of a factor as you edge beyond 30/30/30.

It might also be worth rememebring what the satisfaction ratings of Mrs Thatcher and the Government were before the Falklands:

36% satisfied with Mrs Thatcher; 29% with the Government

Consider also a few quotes:

November 1980, Geoffrey Howe: ‘the fall in output is bottoming out’;
December 1980, John Nott: ‘I think we have reached the bottom of the recession’;
January 1981, Nigel Lawson: ‘all the signs suggest that we have now more or less reached the bottom’;
May 1981, Michael Heseltine: ‘there is tangible evidence that the worst is over’;
June 1981, Margaret Thatcher: ‘the recession has just about reached the bottom’;
October 1981, Geoffrey Howe: ‘there are some clear signs that the Government’s economic policy is beginning to work ... we have passed the end of the beginning’;
January 1982, Norman Tebbit: ‘we are beginning to see signs of our policies working’;

How much economic credibility did the government have at that point? 

The point islington was making (correctly) was that the Conservatives were recovering in the polls before the Falkands war and in particular the Alliance bubble had already been burst by then. No doubt that the Falklands war gave a big and immediate short term boost (which subsequently settled down a bit) but the trajectory was already moving in their direction, as tends to happen as one moves from the mid-term towards the end of a term (even if the subsequent elections is lost - compare 1997 with 1995 or 1970 with 1968). The configuration of the polls in early 1982 was very similar to that in early 1986 and it didn't take another Falklands war to deliver a Conservative landslide the next year.

Up to a point.

There were clearly a lot of wildly divergent polls at that time.

What you can say for sure is that after the Croydon NW the preceding pattern of almost universal Labour leads vanishes.  There are exactly two polls with Tory leads after the Croydon byelection and before the Falklands (and one of those shows an 11 point Tory increase over a poll done by the same company 5 days earlier).  In the same period the Tories were third 11 times.

[polupolu]

Apr 22, 2022 13:31:37 GMT nodealbrexiteer said:

Apr 22, 2022 12:07:23 GMT polupolu said:

What effect would you guess the Falklands war had on voting patterns in the 1983 election?

I am suggesting a scenario of -6%/+3%/+3% Tory/Lab/All i.e. 36.4/30.6/28.4

I think the Alliance would have come a very poor third in the number of seats but the Tories would have lost their majority - but I would be interested in your spreadsheet as I have had to rely on more recent models.

You are clearly correct that an inefficient spread is a big factor in the number of seats; but it begins to be less of a factor as you edge beyond 30/30/30.

It might also be worth rememebring what the satisfaction ratings of Mrs Thatcher and the Government were before the Falklands:

36% satisfied with Mrs Thatcher; 29% with the Government

Consider also a few quotes:

November 1980, Geoffrey Howe: ‘the fall in output is bottoming out’;
December 1980, John Nott: ‘I think we have reached the bottom of the recession’;
January 1981, Nigel Lawson: ‘all the signs suggest that we have now more or less reached the bottom’;
May 1981, Michael Heseltine: ‘there is tangible evidence that the worst is over’;
June 1981, Margaret Thatcher: ‘the recession has just about reached the bottom’;
October 1981, Geoffrey Howe: ‘there are some clear signs that the Government’s economic policy is beginning to work ... we have passed the end of the beginning’;
January 1982, Norman Tebbit: ‘we are beginning to see signs of our policies working’;

How much economic credibility did the government have at that point? 

I remember a book from the 80s simulating Alliance seat totals is they went about 35 and 40% then they starting winning seats in waves

Yes, that is exactly right.  The difference between having two dozen seats and two hundred seats is surprisingly small.

[johnloony]

The idea that Margaret Thatcher won a landslide victory in 1983 "because of" the Falklands war is ridiculous (the fact that it is repeated so often by ignorant ignoramuses does not somehow make it true). It happened because the opposition was split between Labour and the Alliance.

The Conservative Party had already started recovering in the opinion polls from December 1981 onwards. The increase in support for Conservative Party was given a spike by the Falklands war, and that spike was temporary. If the Falklands war had not happened, the Conservative recovery would have been more gradual, and would have reached approximately the same level by 1983.

I am old enough to remember the 1983 campaign. The main issues were the economy and the far-left shambolic policies and image of the Labour Party; people didn't keep going on about the Falklands, as if that were somehow the reason for millions of people switching from Labour to Lib/SDP.

The graph of the opinion polls cited above speaks for itself: if the Falklands war had not happened, the Conservative slope would have gone up after 1982, like a mirror image of the line going down towards December 1981.

[johnloony]

Apr 22, 2022 12:16:11 GMT The Bishop said:

Some serious political scientists think the Falklands War made little or no difference to the outcome of the 1983 GE.

But most of those who were around and remember it will disagree. Even if the Tories had moved back into a small polling lead on its outbreak, they also lost a high profile by-election to the SDP just a week before the Argentinian invasion - their position was far from secure. Three months later, it was (and Thatcher's own ratings transformed)

The mistake you are making is in thinking that the sudden big increase in support for the Conservative Party which happened after April 1982 was the reason why it was far ahead in June 1983. It isn't, and it wasn't. A sudden big increase which happened for an unexpected reason was, over the next 14 months, gradually replaced by a gradual steady increase in support which would have happened anyway.

[Pete Whitehead]

Apr 22, 2022 11:55:25 GMT nobodyimportant said:

If I calculated it correctly, 3% of the voters in 1983 switching from Con to Alliance and 3% from Con to Lab in every seat would have resulted in the following changes to the results:
CON -68
LAB +50
ALL +15
SNP +2
PC +1

This gives the Thatcher government 329 out of 650, a majority of 8. It only takes a slightly unfavourable distribution of the switchers for the Conservatives to lead to a hung parliament. The seat that would lose Thatcher her majority would be Cheltenham, which after this shift would have a Conservative majority of 323 votes over Alliance.

I came up with slightly different figures (though I didn't have the spreadsheet I thought I had, so had to do it 'manually' using the Times Guide - its quite possible I missed one or two).
I had Labour +46 Alliance +13 so that equates to a majority of 20. The closest result I believe was in Aberdeen South which the Conservatives would hold by about 20 votes

[nobodyimportant]

Apr 22, 2022 16:30:22 GMT Pete Whitehead said:

Apr 22, 2022 11:55:25 GMT nobodyimportant said:

If I calculated it correctly, 3% of the voters in 1983 switching from Con to Alliance and 3% from Con to Lab in every seat would have resulted in the following changes to the results:
CON -68
LAB +50
ALL +15
SNP +2
PC +1

This gives the Thatcher government 329 out of 650, a majority of 8. It only takes a slightly unfavourable distribution of the switchers for the Conservatives to lead to a hung parliament. The seat that would lose Thatcher her majority would be Cheltenham, which after this shift would have a Conservative majority of 323 votes over Alliance.

I came up with slightly different figures (though I didn't have the spreadsheet I thought I had, so had to do it 'manually' using the Times Guide - its quite possible I missed one or two).
I had Labour +46 Alliance +13 so that equates to a majority of 20. The closest result I believe was in Aberdeen South which the Conservatives would hold by about 20 votes

The spreadsheet I used is the larger of the two here: commonslibrary.parliament.uk/research-briefings/cbp-8647/

Aberdeen South was the most marginal for me as well.

The three seats the Conservatives lose to parties other than Labour and the Alliance are Ynys Mon, Moray and Banff & Buchan.

[Pete Whitehead]

Apr 22, 2022 17:59:39 GMT nobodyimportant said:

Apr 22, 2022 16:30:22 GMT Pete Whitehead said:

I came up with slightly different figures (though I didn't have the spreadsheet I thought I had, so had to do it 'manually' using the Times Guide - its quite possible I missed one or two).
I had Labour +46 Alliance +13 so that equates to a majority of 20. The closest result I believe was in Aberdeen South which the Conservatives would hold by about 20 votes

The spreadsheet I used is the larger of the two here: commonslibrary.parliament.uk/research-briefings/cbp-8647/

Aberdeen South was the most marginal for me as well.

The three seats the Conservatives lose to parties other than Labour and the Alliance are Ynys Mon, Moray and Banff & Buchan.

Yes I got those three. We differ only on the Labour and Alliance numbers. As I wasn't using a spreadsheet and was scanning a book its more likely than not that I missed one or two (also its not exactly unheard of for there to be errors in the Times Guide)

[carlton43]

Apr 22, 2022 14:12:37 GMT polupolu said:

Apr 22, 2022 13:31:37 GMT nodealbrexiteer said:

I remember a book from the 80s simulating Alliance seat totals is they went about 35 and 40% then they starting winning seats in waves

Yes, that is exactly right.  The difference between having two dozen seats and two hundred seats is surprisingly small.

I make it 176!

[The Bishop]

Apr 22, 2022 14:32:00 GMT johnloony said:

The idea that Margaret Thatcher won a landslide victory in 1983 "because of" the Falklands war is ridiculous (the fact that it is repeated so often by ignorant ignoramuses does not somehow make it true). It happened because the opposition was split between Labour and the Alliance.

The Conservative Party had already started recovering in the opinion polls from December 1981 onwards. The increase in support for Conservative Party was given a spike by the Falklands war, and that spike was temporary. If the Falklands war had not happened, the Conservative recovery would have been more gradual, and would have reached approximately the same level by 1983.

I am old enough to remember the 1983 campaign. The main issues were the economy and the far-left shambolic policies and image of the Labour Party; people didn't keep going on about the Falklands, as if that were somehow the reason for millions of people switching from Labour to Lib/SDP.

The graph of the opinion polls cited above speaks for itself: if the Falklands war had not happened, the Conservative slope would have gone up after 1982, like a mirror image of the line going down towards December 1981.

Well another way to interpret that graph would be to note that Tory support always remained significantly higher after the Falklands War than it had been before it - backing up the idea it was a real, and permanent, boost to their fortunes. The fact people didn't talk about it all the time in the 1983 GE campaign is a red herring, they *did* talk a great deal about "Maggie" being a "strong leader" for example. Where do you think much of that came from??

To avoid any doubt, almost nobody claims it won the Tory landslide then *on its own*. Nor was it irrelevant though, at least in the view of many of us.