The UK with Dems/GOP

[Sol]

Which is a long and hyper-focused way of bringing us to the increasing difficulty of doing this kind of project, as fun as it undoubtedly is. American voting patterns have become increasingly strange from a European perspective (and often look even stranger once look you at detailed results) which makes getting things to fit rather hard. And one reason for this is that the divergent political histories of the United States and Western Europe mean that often the places that have the most in common in most respects are now very different in other ways: in this case it means that even the worst patches of postindustrial despair in Britain are nothing like what can be found in the various Rust Belts of the United States, but similar comments apply to other types of area.

This is obviously true but I think it's also what makes scenarios like this fun and worth doing--they underline the ways in which these divergences create kind of shocking disparities--i.e. like Safe Labour West Virginia or Democratic Home Counties.

The parts where I think this kind of stuff really often chokes is with culturally divided patterns and with sectionalism. For example, West Virginia does seem like the obvious parallel to South Wales, but given that Wales is a rather distinct country it's complicates making the Valleys a Republican stronghold, given that a British GOP would likely have a rather strong little Englander-type attitude. On the flipside, Black and White voters in the UK are less tightly aligned to their respective party preferences than in the U.S., but importing those patterns to prognosticate in, say, Alabama seems to be a dead end.

[Sol]

In any case, at the risk of seeming dumb, I made the northeast corridor with UK-style parties. Might do the rest of the country in a bit.



This is a bit generous to the Tories--it's assuming a 2019 style right-wing overperformance. A lot of these would be more favorable to Labour normally (thinking of PA-13, NY-24, VA-08 maybe) and then a lot more would be quintessential marginals (MA-04, MA-09, NJ-12, NJ-01, all the Torie SEPA districts, maybe NH-01 and NY-10). District lines and the VRA also have a certain effect--New Jersey would probably have a few more Labour seats with compact districts, while Maryland's gerrymander of Montgomery County actually works as a Tory gerrymander, since the parts of Montgomery County near DC have a slight Labour lean while most of the outer DC area and the panhandle (save Cumberland and Garrett County) are hyper-Tory.

I should also probably say that I don't have an especially deep knowledge of U.K. politics so probably a lot of this is wrong. Also there were certain areas I was unsure about--Long Island in particular, which I ended up using Essex as a parallel for--so I'd welcome corrections. I also just realized that I completely forgot about the Lib Dems (lol) so I'd welcome suggestions about where they might win--my intuition at first blush is MA-05, NY-10, NJ-05, NJ-12, and VA-11, but that seems a little too favorable to Labour.

[Sol]

Even in 2019 Labour didn't lose the most abjectly poor parts of 'traditionally working class' Britain (although their position did weaken considerably). The places in the red wall that fell were generally the areas with high homeownership, lower poverty and an elderly age profile. I'm not an expert on West Virginia but if one of the districts is considerably less poor than the state then that one would have fallen, but in general even in 2019 the Tories still fell short in areas as deprived as West Virginia is.

WV-2 is very likely the most likely bit of West Virginia to fall--it has Putnam County and a lot of the not-traditionally industrial parts of eastern WV, plus the DC exurbs near Harpers Ferry which seem like a Tory sort of place. There's a good argument too for flipping PA-17 and PA-14 too then in that case, as well as maybe VA-09 and maybe PA-12?

On the other hand the Tories would easily be winning districts like ME-02, NH-02 (the college towns would easily get outvoted) and CT-02. These are the type of districts which would have long been Tory as they aren't poverty stricken and aren't anchored on major urban centres. They *may* have gone Labour in the Blair landslides but in general they are the type of area that Labour would have been dead and buried in since roughly the time Harold Wilson was PM.

I agree the two NOVA districts should be Labour though, the Tories do horribly with government workers even affluent ones. They probably would still hold VA-10 though.

I kind of have to beg to difffer here. ME-2 in particular is a pretty poor area--lowest median household income in New England, and more in line with the WV districts discussed above--and historically a major center of industry. I suppose there's an argument for it having flipped in 2019 but a place like that doesn't seem like somewhere where Labour would have been "dead and buried."

NH-02 is a bit more mixed, but has some similar areas (thinking of Coos County) and the ex-industrial but suburban weird place of Nashua which doesn't honestly make much sense in the American context either. I don't know much about CT-02 to be honest, but it voting Conservative would make some sense.

Wrt: Virginia, that's a hard area. A big thing which made me put those inner suburban districts in the Con column was that a lot of the suburban employment in NOVA is in industries which lean more right-wing, at least in the American context--a lot of corporate headquarters, military contractors, engineering, in addition to your typical government employees. Are there equivalent dynamics in parts of London?

[Sol]

The embarrassment of revealed ignorance continues, with a pass at the South (excluding the southern states which are a part of the Northeast corridor, like Maryland, Virginia, and Delaware).



There were also some very difficult elements here. In particular, there are several parts of Southern states, particularly in Alabama, the Carolinas, and Tennessee, which had and still have to some extent an industrial economy, one which developed in the early and mid 20th due to the South's ununionized and cheap labor. These places mostly vote Republican these days--the industrial labor isn't and never really has been unionized to the same extent as the North, except in Alabama iirc. I most went with giving these areas to the Conservatives, with exception of urban centers and areas with a union tradition.

The other big question is racial polarization, and that was a bit easier because I just decided that it would basically pattern in the same way as the Democrats and Republicans do irl. That said, I did decide that Cubans would likely be Tory-leaning in a stronger way than irl, with a more overtly socialist Labour and a less overtly racist Conservative party.

Again, this is a rough election year for Labour (or should it be Labor?). In a better year they'd have MO-05, MO-08, KY-01, KY-02 and AL-05, with a decent shot at winning classic marginals NC-08, NC-09, and mayyybe TX-14.

It's funny how in a lot of places these actually look a lot more like real results--that may actually be a Texas congressional map from the earlier part of last decade, and Georgia is of course identical to the map from 2015 to 2019. Really shows how much more the GOP has held up rich southern urban areas until the recent past.

[Sol]

Even in 2019 Labour didn't lose the most abjectly poor parts of 'traditionally working class' Britain (although their position did weaken considerably). The places in the red wall that fell were generally the areas with high homeownership, lower poverty and an elderly age profile. I'm not an expert on West Virginia but if one of the districts is considerably less poor than the state then that one would have fallen, but in general even in 2019 the Tories still fell short in areas as deprived as West Virginia is.

WV-2 is very likely the most likely bit of West Virginia to fall--it has Putnam County and a lot of the not-traditionally industrial parts of eastern WV, plus the DC exurbs near Harpers Ferry which seem like a Tory sort of place. There's a good argument too for flipping PA-17 and PA-14 too then in that case, as well as maybe VA-09 and maybe PA-12?

On the other hand the Tories would easily be winning districts like ME-02, NH-02 (the college towns would easily get outvoted) and CT-02. These are the type of districts which would have long been Tory as they aren't poverty stricken and aren't anchored on major urban centres. They *may* have gone Labour in the Blair landslides but in general they are the type of area that Labour would have been dead and buried in since roughly the time Harold Wilson was PM.

I agree the two NOVA districts should be Labour though, the Tories do horribly with government workers even affluent ones. They probably would still hold VA-10 though.

I kind of have to beg to difffer here. ME-2 in particular is a pretty poor area--lowest median household income in New England, and more in line with the WV districts discussed above--and historically a major center of industry. I suppose there's an argument for it having flipped in 2019 but a place like that doesn't seem like somewhere where Labour would have been "dead and buried."

NH-02 is a bit more mixed, but has some similar areas (thinking of Coos County) and the ex-industrial but suburban weird place of Nashua which doesn't honestly make much sense in the American context either. I don't know much about CT-02 to be honest, but it voting Conservative would make some sense.

Wrt: Virginia, that's a hard area. A big thing which made me put those inner suburban districts in the Con column was that a lot of the suburban employment in NOVA is in industries which lean more right-wing, at least in the American context--a lot of corporate headquarters, military contractors, engineering, in addition to your typical government employees. Are there equivalent dynamics in parts of London?

I would've thought PA-12 (and PA-09) would've been more likely to flip than say PA-14, though admittedly PA-14 does have some exurban areas so could've been a narrow 2019 flip (after being quite comfortably Labour in 2017).

My logic for that was that PA-14 does have quite a bit of exurban, and frankly in some parts, suburban areas, which more closely fits the description of a flipping district which vileplume gave. If you look at income maps of the Pittsburgh area lots of PA-14 is very poor but there are also big chunks of Westmoreland and Washington counties which are quite prosperous.

I don't the demographics of PA-12 and PA-09 super well, but both have very poor industrial areas. Though there is the much more Tory seeming area of Lebanon in PA-09.

[Sol]

Of course another factor in the case of NOVA is that it's quite diverse--massive Salvadoran, Korean, Indian, and African-American communities, and in some parts a bit poorer than it's generally expected to be (thinking of SE Fairfax and Eastern Prince William). That was why I had initially suggested VA-08 as a district Labour could win, since it has some of those areas and the socially liberal folks who are in the genuinely urban parts of northern Virginia.

[Sol]

The embarrassment of revealed ignorance continues, with a pass at the South (excluding the southern states which are a part of the Northeast corridor, like Maryland, Virginia, and Delaware).



There were also some very difficult elements here. In particular, there are several parts of Southern states, particularly in Alabama, the Carolinas, and Tennessee, which had and still have to some extent an industrial economy, one which developed in the early and mid 20th due to the South's ununionized and cheap labor. These places mostly vote Republican these days--the industrial labor isn't and never really has been unionized to the same extent as the North, except in Alabama iirc. I most went with giving these areas to the Conservatives, with exception of urban centers and areas with a union tradition.

The other big question is racial polarization, and that was a bit easier because I just decided that it would basically pattern in the same way as the Democrats and Republicans do irl. That said, I did decide that Cubans would likely be Tory-leaning in a stronger way than irl, with a more overtly socialist Labour and a less overtly racist Conservative party.

Again, this is a rough election year for Labour (or should it be Labor?). In a better year they'd have MO-05, MO-08, KY-01, KY-02 and AL-05, with a decent shot at winning classic marginals NC-02, NC-08, NC-09, and mayyybe TX-14.

It's funny how in a lot of places these actually look a lot more like real results--that may actually be a Texas congressional map from the earlier part of last decade, and Georgia is of course identical to the map from 2015 to 2019. Really shows how much more the GOP has held up rich southern urban areas until the recent past.

The South is the massive problem area for this exercise as if you picked up the Tory Party and dumped them into America with a near-unchanged platform, they'd do horribly with Southern Evangelicals many of whom probably go third party. I'm sure if forced to choose the greater part would pick the Tories over Labour (particularly the rich suburban every-Sunday churchgoing types) but the very poor evangelicals in the backwaters would more not vote, back some Farage-esque third party or if very poor maybe even back Labour as the Tories aren't giving them as much of the cultural incentive to override their economic interests. Labour would do much better than in this map for these reasons, probably with a hard right insurgent party winning seats too.

I mean, the whole point of this exercise is that that's not what one does--rather it's about coming up with the closest fit for the parties involved's coalitions and stretching that in a case where that's not obvious. I obviously would have given UKIP some seats if I was modeling earlier elections, but I wasn't.

Precise realism isn't the goal--the point here is just a thought experiment.

[Sol]

I think you might be underestimating the extent to which both PA-09 and PA-12 are both quite ex-industrial.

In PA-09 everything north of Berks is ex-coalmining areas, which is why Democrats used to win a lot of those bits. That certainly doesn't mean they wouldn't have flipped in 2019 of course--independent of Labour collapsing with rural voters there's also a decent counterweight in Lebanon and northern Berks, which is more ag oriented.

PA-12 doesn't really have much in the way of Harrisburg commuterland fwiw--Perry is part of the MSA but it's a quite peripheral area to that metro area and is a pretty small percentage of the district's population. I can buy it flipping for sure but Labour probably has a high floor on account of State College and the small post-industrial cities and towns (Williamsport is a big place!)

[Sol]

The embarrassment of revealed ignorance continues, with a pass at the South (excluding the southern states which are a part of the Northeast corridor, like Maryland, Virginia, and Delaware).



There were also some very difficult elements here. In particular, there are several parts of Southern states, particularly in Alabama, the Carolinas, and Tennessee, which had and still have to some extent an industrial economy, one which developed in the early and mid 20th due to the South's ununionized and cheap labor. These places mostly vote Republican these days--the industrial labor isn't and never really has been unionized to the same extent as the North, except in Alabama iirc. I most went with giving these areas to the Conservatives, with exception of urban centers and areas with a union tradition.

The other big question is racial polarization, and that was a bit easier because I just decided that it would basically pattern in the same way as the Democrats and Republicans do irl. That said, I did decide that Cubans would likely be Tory-leaning in a stronger way than irl, with a more overtly socialist Labour and a less overtly racist Conservative party.

Again, this is a rough election year for Labour (or should it be Labor?). In a better year they'd have MO-05, MO-08, KY-01, KY-02 and AL-05, with a decent shot at winning classic marginals NC-02, NC-08, NC-09, and mayyybe TX-14.

It's funny how in a lot of places these actually look a lot more like real results--that may actually be a Texas congressional map from the earlier part of last decade, and Georgia is of course identical to the map from 2015 to 2019. Really shows how much more the GOP has held up rich southern urban areas until the recent past.

NC-02 looks to be coloured in red in your map. Which I'm guessing is a mistake. I could see it being LD though.

Also I think Labour would be stronger in KY-06 than KY-02? From memory KY-06 has part of the big coal region?

Yeah the non-union industrial areas are interesting. From memory outside of Alabama the industry tends to be lighter (and therefore possibly less Labour/social democratic friendly), but I could be wrong. Certainly the nature of unionisation could be very different in the South if one party was pushing harder for unionisation.

Yeah NC-02 was a mistake. I had initially colored it yellow, but upon reflection it seemed like the sort of place which would have flipped fairly recently but stuck around, a bit like other heavily university influenced constituencies. Lib Dems and Cons would certainly have a shot there too.

Wrt: KY-06, I had forgotten that they had added those areas! Certainly makes it more winnable for Labour, but the bulk of the seat is in Lexington and surrounds, which maybe don't seem as friendly?

KY-01 and KY-02 are due to the western coalfield.

[Sol]

Part 3:



There are several constituencies on here which probably would have been Labour voting most of the time--thinking MI-01, OH-06, maybe WI-07--but which fell because of the right's overperformance last election. MI-09 is probably a classic swing district, and IL-17, IN-02, and IN-07 are winnable as well.

There were a few judgements calls as well. Both Indianapolis and Columbus are hard to compare to cities in the UK--they're both expanding, "new growth" cities with smaller histories of industry. I decided to split the difference and give Columbus to the Lib Dems and Indy to the Cons, since the latter is a lot less professional/managerial. (You can probably tell that I'm relying a lot on voting patterns from before our current alignment.)

Another call I made was to have Labour drastically underperform the Democrats in a lot of the rural agricultural Midwest. Democratic strength in places like Iowa or Western Wisconsin is fairly recent and is the outcome of the ag crisis and a higher degree of secularism. Similarly agricultural areas in the UK vote much more decisively to the right, iirc.

Chicago was an interesting area to do--you can really see the north-south class divide which is invisible in current election maps.

[Sol]

(snip)

Another good attempt which is mostly agree with. Nit-picking, i suspect Minnesota 8th would very likely have flipped Conservative in 2019, the actual Iron Range is quite small and Labour would need very large margins out of it and Duluth to overcome the agricultural areas (which they would have got pre-2019). Ohio 14th would also likely have voted Conservative as its mostly suburban/exurban Cleveland. I also don't think Michigan 1st would be a swing district, the Upper Peninsula would be winnable but the Lower Peninsula part is either agricultural or vacation homes, both awful places for Labour. Its a similar case for Wisconsin 7th (but the Green Bay based 8th district would certainly be winnable for Labour). I also suspect Indiana 8th (SW Indiana) is post-industrial enough to be won pre-2019.

Wrt: MN-08, I think that Labour probably would have just held on. The Iron Range is actually quite a lot of the seat--it's just shy of an outright majority. Quite a bit of MN-08 is also not especially agricultural--a lot of it is historic logging territory actually so I assume Labour would do a bit more similarly to Democrats irl. Plus I could see Labour actually doing even better with natives. Similar story with WI-07 and MI-1 except the proportions are different, but all have a strong historic Labour base as they did irl--the whole industrial zone by Lake Superior historically in the U.S. was far left and union heavy even in comparison to other similar areas in the rest of the country, and would consequently probably give Labour insane margins even if they were slightly cut down in 2019.

I don't know the Cleveland area especially well, but I get the impression that OH-14 is heavy on the kind of lower middle class "white ethnic" voters who also predominate in places like IL-03. Cleveland also seemed to me to resemble the kind of place, like Merseyside, where Labour might overperform due to the conservatives being uniquely weak. There's also Ashtabula which seems like it would give Labour a strong floor.

WI-08 is interesting--it doesn't seem to me like a Labour-friendly seat, given the importance of tourism and that the Fox Cities are sort of quintessential right wing voting areas--small, not too unionized, homogenous, and relatively middle class and economically stable. But I easily could be wrong!

You're definitely right wrt: IN-8; I always forget the coalmining history of that part of the country. There's a decent argument for Labour also having a shot at Mary Miller's district too, lol.

Something I didn't really talk about but which matters also I think are kind of parochial or historic influences on voter behavior which are particularly pronounced in the Midwest, like the sort of inordinate strength Republicans have in Cincinnati or Eastern Wisconsin. Not really sure how to model these--it doesn't matter a ton anyway given that the vagaries of the lines kind of take away that question, but it's sort of an interesting thought experiment.

[Sol]

It's a little bit disappointing that this exercise is based on 2019, since it seems like the size of that result is obscuring patterns that might otherwise be interesting. The Liberal Democrats seem like mostly an afterthought so far; the enormous size of American districts would certainly hurt them, but it seems like they should be a factor in the sort of places that might be too rich to vote Labour (Westchester County, Connecticut, the northern suburbs of Chicago, maybe the rich parts of Southern cities).

Yeah, that's fair enough.

To be honest I think I keep on forgetting about the Lib Dems, in part because I'm trying to correct for a bias I have towards wanting the Labour Party to do well. Most of the places I would have pegged as Lib Dem-maybe some of the NoVA seats, SC-01, IL-09, maybe some of the Broward seats--would have otherwise gone with the Tories, and that felt kind of like hurting the Conservatives out of bias. Meanwhile the only Labour-held seats that seem like it could go Lib Dem thus far are NC-02 and MD-03.

[Sol]

Here is a (very!) rough pass at the western U.S.:




A few necessary caveats:
-California means that one really needs a mental model for Asian-American voting patterns, especially for ethnic groups which are less common in the UK. Since it sounds like Chinese-British people are fairly Tory, I modeled them accordingly. For Japanese and Korean-Americans, I also had them going to conservatives, since IIRC those groups tend to be higher income. Other East Asian immigrant groups often skew a bit more working class, so I have them going to Labour. Apologies if this is grimly ignorant.
-I don't have a great sense of the distribution of a lot of Asian-American ethnic groups in west coast cities, especially in the Bay Area. I may be consequently overestimating the conservatives.
-I colored the wrong Colorado district Lib Dem--I meant to color CO-02 yellow and CO-07 blue.

Other notes:
-The West Coast seems like the sort of place where Lib Dems would be especially strong. I gave them several seats in the Bay and Pramila Jayapal's district (lol) (which would normally go Labour). There are a lot of "very bougie but extremely socially liberal/culturally Democratic" places on the west coast--Boulder, much of the richer parts of the bay, Hollywood--and it's hard to see a better party for those places.
-The Bay Area has several seats--thinking especially of CA-05 and CA-11--which are very socioeconomically heterogeneous. Since this was a Conservative year, I had Napa outvote Vallejo and the Tri-Valley outvote Richmond, but obviously most years it would be different.
-Los Angeles is notably more Left-leaning than San Francisco
-In better years, Alaska, AZ-01, AZ-02, NM-02, CA-12, CA-25, CA-36 and Montana would be winnable for Labour, in addition to the ones already mentioned.