As I was saying about Deprivation and the vote...

[Ghyl Tarvoke]

In another thread I once again discussed this chart below in relation to Labour's supposed problems with Working Class voters and how this mostly is not a thing that exists to the extent the media and various pundits think it does.



(I note that the Tory seats in the bottom four deciles are primarily a) seaside seats or b) in Lincolnshire or c) one of the handful of gains they managed back in June with a few exceptions none of which are particularly safe).

I added that if someone were to do a similar chart for the EU referendum it would look very different and would show the simple fact that the Leave vote was wealthier (wealth being a key word) than Corbyn's vote in aggregate.... well.... I went and did it, the absolute madman.



(I note that I did this on the quick so you can still see the blue and red outlines for many of the boxes, I wasn't going to do this perfectly... and it still took hours anyway. There may be some errors there, especially with the shading).

I used Political Scientist Chris Hanretty's projections of how each constituency voted - although this is known in some cases due to the release of ward results or the local Authority and constituency having conterminous boundaries - to give me my results. They are available on the Wikipedia article on the results of the referendum. As I stated above, many of these are just projections and so I gave an assumption of error of three points. That is I can't assume with certainty how one seat voted if the projected result was 53-47 either way or narrower. For that reason, I used a lighter shade of colour for those particularly close results. Reading Hanretty's analysis in detail I realize that I should have been stricter and made the error five points (so 55-45) but it wouldn't make that much difference. It's going to be fairly accurate or although perhaps with one or two errors around the edges.

And as you can see, I was right. Indeed, it's really noticeable how much 'Remain' drops off after you get past 25 or so least deprived constituencies. I note that there were more Remain voting seats in the 1st, most deprived, than in the 9th, second least deprived, decile (14 vs 12). This primarily represents urbanity and also the failure of Cameron and Stronger IN to do as well in the Home Counties as probably should, note how many of the seats in the 9th decile have the lighter shade of purple. Unlike with Labour, Parliament's Index of Total Deprivation is not a good predictor of either the Leave or Remain vote. To give another example, the 4th Decile contains the most Leave voting seats, with only 9 out of the 53 listed having voting Remain... but one of those nine is Streatham, held by Hanretty to be the most Remain of all seats.

The Number and Percentage of Remain seats by decile is below, holding all Hanretty's assumptions to be accurate

1st Decile (Most Deprived): 14 seats (26.42%)
2nd Decile: 17 seats (32.08%)
3rd Decile: 20 seats (37.74%)
4th Decile: 9 seats (16.98%)
5th Decile: 11 seats (20.76%)
6th Decile: 12 seats (22.64%)
7th Decile: 10 seats (18.87%)
8th Decile: 15 seats (28.3%)
9th Decile: 12 seats (22.64%)
10 Decile: 41 seats (73.21%)
Overall: 161 seats (30.21%)

What this all means I'll leave up to you. I'll just make some observations:

1. Even within the first and last deciles there was significant variation in comparison. In the first that should be obvious given it contains both Hull East and a lot of the worst parts of London. But even in smaller cities the results were notable so here are some of the results from within the 1st Decile with the Remain percentage rounded given in brackets: Liverpool Wavertree (64% Remain), Bethnal Green and Bow (69%), Hackney South and Shoreditch (78%), Liverpool Riverside (74%), Birmingham Hall Green (66.5%), Nottingham East (57%), Tottenham (69%), Manchester Gorton (62%), Manchester Central (63.5%), Birmingham Ladywood (64.5%)... Now there are some very common explanations to some of those results so I will merely point to overwhelmingly white working class Liverpool.

2. In the final decile while most seats were Remain very few were overwhelmingly so (>20 points margin) apart from the few in London there, Sheffield Hallam (the only Labour seat on the list, well was.... until, well, Jared O'Mara) and St Albans. However within that decile there is also a 68% Leave seat - Rayleigh and Wickford. There are also a few 60%+ Leave seats in the 9th Decile. Some Journalists to South Essex to discover these left behinds please.

3. There are very few safe Remain voting Tory seats below the 8th Decile, which perhaps shows how shallow and posh 'Tory Remain' was. The few exceptions are in London, where it's harder to segregate yourself as it is in the exurbs. The best example is the most deprived Tory-Remain seat now? The Cities of London and Westminster in the 5th decile. That fact should also tells us something about the limits of the analysis here. 

4. The deprived Remain voting seats were terrible for the Tories at the election. What was the most deprived Tory-Remain seat before the election? Brighton Kemptown, in the 3rd decile.

5. Most of the Labour voting seats in the upper deciles are Remain gains - many of them 2017 gains like Hallam and Stroud. The ones that weren't tended to see significant swings against Labour, such as Penistone and Stockbridge. 

6. The least deprived over 53% Leave seat is David Davis' Haltemprice and Howden


7. Again note how much light purple there is in the 9th decile. While I know in a referendum any vote anywhere is equal to any other vote there is a serious argument to be the made that it was not the North where the referendum was lost, or low turnouts in London and Scotland (whose turnouts would have been good or respectable for general elections) but in the Home Counties. Why did Cameron fail to bring these places, except the most wealthy, with him?

(I know most of you know this and are probably bored of analyzing the referendum results now, but hey, you don't have to read this)

[The Bishop]

Thank you for this.

For all the post-vote frenzy (not least from certain people on "his" side) that Corbyn had "thrown" the referendum - when in fact Labour voters supported remain by about the same proportion as the SNP, and not that far behind the LibDems - your stats show clearly that it was the Tory "heartland" vote wot won it for Brexit. There has, as you say, been remarkably little analysis of this subsequently - certainly nowhere near as much as the once almost ubiquitious "Labour's 'left behind' Brexit nightmare" pieces.

Yet another failure of our commentariat.

[J.G.Harston]

Nov 1, 2017 2:36:54 GMT Ghyl Tarvoke said:

I added that if someone were to do a similar chart for the EU referendum it would look very different and would show the simple fact that the Leave vote was wealthier (wealth being a key word) than Corbyn's vote in aggregate.... well.... I went and did it, the absolute madman.

With the "533 divided by ten" thing, it would be a better balanced diagram if 2, 5, 8 had one extra instead of 10 having three extra.

[Rose Tinted Lane]

Nov 1, 2017 11:31:07 GMT The Bishop said:

Thank you for this.

For all the post-vote frenzy (not least from certain people on "his" side) that Corbyn had "thrown" the referendum - when in fact Labour voters supported remain by about the same proportion as the SNP, and not that far behind the LibDems - your stats show clearly that it was the Tory "heartland" vote wot won it for Brexit. There has, as you say, been remarkably little analysis of this subsequently - certainly nowhere near as much as the once almost ubiquitious "Labour's 'left behind' Brexit nightmare" pieces.

Yet another failure of our commentariat.

At the risk of arguing for the sake of it, I'm not entirely sure that is the correct analysis of this data.

There are a lot of Tory seats in the 9th decile which are light purple, representing Tory-voting areas that voted 47-49.9% Remain. While in the 1st decile, there are a lot of Labour-voting areas with strong purple colouration, suggesting a lower than 47% Remain vote. This is despite Labour being estimated to vote about 65-70% Remain, and Tories estimated at 35-40%. On those numbers, it would suggest that the Tories largely did pull their weight, and it was Labour who let the Remain side down.

That isn't necessarily to blame Corbyn, although I know from anecdotal evidence on the day that lots of Labour voters didn't know the Party's official position and just assumed it was Leave, which is a failing of the leadership. What it does also potentially suggest though is that Labour MPs and Councillors in Leave-leaning areas were worried about being associated with the Remain campaign, so didn't really campaign hard to put their message across.

[The Bishop]

Hmmm, up to a point as they say.

If the media had bothered to report the dozens of speeches Corbyn made during the campaign, Labour voters might have had a better idea of where the party stood. As indeed they might have done if A Johnson had got off his lazy arse and put a bit of effort in (or even, horror of horrors, set up an actual Twitter account)

[Ghyl Tarvoke]

My point is that if Remain had done even slightly better in the prosperous agricultural districts and in the Exurban Home Counties, the 'left behind' vote would not have mattered.

I mean, look at the drop-off after the wealthiest 40 or so constituencies...

[Rose Tinted Lane]

Nov 1, 2017 16:15:59 GMT Ghyl Tarvoke said:

My point is that if Remain had done even slightly better in the prosperous agricultural districts and in the Exurban Home Counties, the 'left behind' vote would not have mattered.

I mean, look at the drop-off after the wealthiest 40 or so constituencies...

I don't necessarily disagree with you, but that probably works both ways! If Labour had rallied the 'left behind' voters as you describe them, then we wouldn't have needed the Tories to win over the rural gentry!

I do find the dip in the centre very interesting. This ties in with everything we know about far-right and right-populist movements - Poujadism appealed most heavily among the lower-middle classes, particularly among white voters, which is also true today of NF, UKIP, 5SM, Golden Dawn, etc.

[Ghyl Tarvoke]

The dip in the centre is mostly reflective of the fact that there are very middle income places in large urban centres tbh.

[Ghyl Tarvoke]

And to complete, I've gone completely mad with the Data Visualization. It's pretty ugly folks

[sirbenjamin]

Has anybody got a list of the actual seats in order of deprivation that corresponds to this? I've only worked out Brighton Pavillion...

Also, which set of Deprivation Indices data has been used? Do they just add all the wards or output areas together to make the seats, or is it more clever/nuanced?

[Ghyl Tarvoke]

Nov 3, 2017 19:21:01 GMT sirbenjamin said:

Has anybody got a list of the actual seats in order of deprivation that corresponds to this? I've only worked out Brighton Pavillion...

Also, which set of Deprivation Indices data has been used? Do they just add all the wards or output areas together to make the seats, or is it more clever/nuanced?

I used the data here

So... five most deprived are:
1) Liverpool Walton
2) Birmingham Hodge Hill
3) Birmingham Ladywood
4) Blackley and Broughton
5) Nottingham North

And the five least are:
529) South West Surrey
530) York Outer
531) Chesham and Amersham
532) Wokingham
533) North East Hampshire

[sirbenjamin]

Nov 3, 2017 19:39:30 GMT Ghyl Tarvoke said:

Nov 3, 2017 19:21:01 GMT sirbenjamin said:

Has anybody got a list of the actual seats in order of deprivation that corresponds to this? I've only worked out Brighton Pavillion...

Also, which set of Deprivation Indices data has been used? Do they just add all the wards or output areas together to make the seats, or is it more clever/nuanced?

I used the data here

So... five most deprived are:
1) Liverpool Walton
2) Birmingham Hodge Hill
3) Birmingham Ladywood
4) Blackley and Broughton
5) Nottingham North

And the five least are:
529) South West Surrey
530) York Outer
531) Chesham and Amersham
532) Wokingham
533) North East Hampshire

Marvellous, thank you.

[Merseymike]

This is anecdotal but it struck me that Leave left behind voters couldn't care less what any politicians said.

[Deleted]

Nov 1, 2017 13:25:13 GMT The Bishop said:

Hmmm, up to a point as they say.

If the media had bothered to report the dozens of speeches Corbyn made during the campaign, Labour voters might have had a better idea of where the party stood. As indeed they might have done if A Johnson had got off his lazy arse and put a bit of effort in (or even, horror of horrors, set up an actual Twitter account)

Well. The party for the many (most deprived 9 tenths) not the few (least deprived tenth) comprehensively failed to carry the many whereas the Tory remainers delivered the few successfully.

[ccoleman]

Although people are conflating deprivation with income - which in my opinion is a little misleading. For example, Kensington and Chelsea is technically one of the more deprived boroughs in the UK despite being the highest-earning on average.

I suspect when people say Labour are doing worse with the working classes, they're referring to the fact that Labour increased their vote share amongst ABs and won over C1s, while the Tories won over C2s and did better with DEs. And, of course, ABC1s voted largely for Remain - but many ABC1s, funnily enough, live in seats that are somewhat deprived or even very deprived. For example, Bethnal Green and Bow is over 60% ABC1 despite being a very deprived constituency. Cities of London and Westminster is middle-of-the-pack and is over 70% ABC1. Even very wealthy urban seats will often have pockets of high deprivation due to the presence of council estates.

And, of course, population distribution isn't equal - Labour for example could win a plurality of ABs in 2022 but the Tories would still win the majority of constituencies with the most AB voters simply due to how the population is distributed, and that wouldn't be reflected in 'most/least deprived constituencies' statistics.

It's a messy picture imo.

[Ghyl Tarvoke]

ABC1C2DE system is an anachronism and should no longer be used...

[ccoleman]

Well, that's very convincing, thanks. Still doesn't detract from my point that deprivation levels are not necessarily a reflection of average income, and therefore don't tell you what you think they tell you.

[Ghyl Tarvoke]

Nov 4, 2017 19:15:53 GMT ccoleman said:

Well, that's very convincing, thanks. Still doesn't detract from my point that deprivation levels are not necessarily a reflection of average income, and therefore don't tell you what you think they tell you.

For a start, in UK right now Income and Wealth are no longer as related to each other as they may have done in the past. As I've said elsewhere, it's pretty certain that the average Leave voter has more wealth than the average Remain voter, even if in all likelihood the average Remain voter has a higher income. This is simply a reflection of the age demographic of voters vis-a-vis patterns of Capital Ownership, especially that of housing. Note here that I'm only referring to averages, both sides contain complex multitudes.

As for the problems with the ABC1C2DE system, there are many problems with it so I suggest start reading here.

Also if we are allowing ourselves to use Ipsos-Mori after election polls, I suggest we should look at this:

[Ghyl Tarvoke]

Of course I will add despite the caveats above the idea that there is no correlation between Income and Deprivation is absurd. But this doesn't tell us what you seem to implying at all, indeed it might mean the opposite. As I already pointed out, the most deprived Tory 2017 - Remain 2016 constituency is the Cities of London and Westminster. But I don't think the Tories who live and vote there are, in general, a particularly deprived group. *But* the Tories won the seat with 46.6% of the vote while Remain won it with 71.4%. Even assuming a 1:1 overlap (loool) between Tory voters there and Remain voters, the Remain vote is much higher which means...

[Ghyl Tarvoke]

Ftr there was one referendum in the previous couple of years that was a genuine working class revolt against the establishment, and that was the Scottish Independence referendum and that vote wasn't enough to get to 50%+1 (and, yes, I know not all working class people voted YES, I know).