Francine Busby (D) running even with Brian Bilbray (R) in special election

Survey USA ^ | May 11, 2006

[AuH2ORepublican]

If President George H.W. Bush gets 53% nationwide in 1988 and gets 51% in an Iowa CD, how is that district "Democrat-leaning"? The 51% who voted for Bush would also be able to vote for the Republican for Congress, and the fact that voters in the rest of the country voted 2% more for Bush has nothing to do with it. I do think that one needs to take into account the Bush vote in that particular *state* or *region*, which in the case of Iowa, farm areas and the Upper Midwest in 1988 would lead one to believe that a 51%-Bush CD in Iowa has quite a GOP lean to it (Dukakis carried Iowa and Wisconsin).

I can see how a district in rural Alabama that gave Bush 52% in 1988 is not necessarily "Republican-leaning," but it wouldn't be because Bush got 53% nationwide, but because Alabama and the rest of the Deep South gave Bush around 60%.

[AntiGuv]

I believe that in a reverse-1994 style election the RATs will pick up a number of districts with slight GOP leans, so yes, combined with Dem leaning districts I think there's more than enough for them to take the House.

[AntiGuv]

There is a presumption today that the presidential results accurately represent the inherent partisan lean of the congressional districts. That was not the case before 1994 when Republicans clearly performed better on the presidential level than they did on the congressional level.

[Torie]

AZ-01 (Matt Salmon)

1988: R +10.96
1994: R +6.51

AZ-06 (J.D. Hayworth)

1988: R +6.29
1994: R +3.44

I think there's little doubt that Arizona was drifting leftward between 1988 and 1994. [Maybe, but Bush 2000 carried AZ-1 by 7.33, and AZ-6 by 11.38%, so perhaps not. I think I like the 1988 numbers better]

CA-19 (George Radonovich)

1988: R +1.26
1994: R +9.04

By contrast, the inland California districts have been shifting rightward, along with inland Oregon and Washington (below). [agreed]

FL-15

1988: R +16.36
1994: R +9.83

Florida was definitely moving leftward from 1988 to the mid Nineties (and beyond). [Well this part of Florida definitely was, although not as sharply as the Gold Coast]

GA-08

1988: R +9.34
1994: R +1.01

GA-10

1988: R +10.93
1994: R +2.28

Georgia's an interesting case. It's really the only one where I almost prefer the 1988 numbers to the 1992/1996 average. Then again, Clinton clearly did relatively well in Georgia by comparison to Dukakis who super-flopped.

ID-01

1988: R +6.23
1994: R +12.62

The mountain states were definitely moving to the right throughout the Nineties. [Yep, particularly northern Idaho as the kooks moved in. :)]

IL-05

1988: D +2.40
1994: D +10.32

The shift to the left in Illinois was well-underway by 1994.
[Yep, and the 5th was browning]
IA-04

1988: D +9.27
1994: R +0.85

There's no way IA-04 was a 9 pt Dem advantage in 1994.[Dukasis ran uniquely well in Iowa, so that screws the results. But Des Moines moved a bit more GOP as well. Whether that was realized by 1994, I don't know. The true number is somewhere inbetween. Bush 2000 carried the 4th by 2% by the way.]

KS-04

1988: R +3.51
1994: R +8.29

The Great Plains in general were forging hard right.[Certainly Kansas was, outside of the KC suburbs]

ME-01

1988: R +2.46
1994: D +4.77

ME-02

1988: R +1.08
1994: D +5.83

Maine, on the other hand, was definitely Dem leaning by the time 1994 rolled around; certainly not GOP leaning as in 1988. [Maybe, but it may be more that Maine is just odd. ME-1 had a Kunnebuncport factor going. Ususally ME-2 is more GOP than ME-1]

MI-08

1988: D +3.74
1994: R +0.32

The blue collar workers in the Midwest were continuing the steady drift away from the Dems that began in the early 80s.
[MI-08 is Lansing the the exurbun right wing Livingston County. It isn't blue collar. It was the growth of Livingston that was moving the numbers]
NE-02

1988: R +3.92
1994: R +12.80

See Kansas above. [Yep]

NH-02

1988: R +8.13
1994: D +0.44

There is no way NH-02 was an 8 point GOP advantage in the mid Nineties. [The truth is probably somewhere between the two numbers Bush lost by -.75%, so if the erosion is a straight line, it was only half way there in 1994.]

NJ-02

1988: R +4.86
1994: D +0.56

NJ-08

1988: R +0.57
1994: D +3.20

The leftward shift in New Jersey was also well-underway, and accelerating.

NY-01

1988: R +6.71
1994: R +0.40

Ditto Long Island. [Agreed, but I am not sure just how far along it was in 1994. At best, the truth is inbetween. Today, nY-1 is about at parity, so if it was at parity in 1994, there has not been much movement since. I agree with you completely on NJ. Big, big Dem gains there, and just go to show how suburban a state it is.]

NC-03

1988: R +6.96
1994: R +10.01

Eastern North Carolina shifted markedly to the right throughout the Nineties. [Yep, particularly the beach.]

OK-02

1988: D +4.86
1994: D +0.23

OK-04

1988: R +4.42
1994: R +9.33

And Oklahoma was definitely on a rightward track as well. [yep]

OR-05

1988: D +3.99
1994: R +0.86

Ditto the inland Pacific Northwest. [nope, that is the second district. Bush 2000 carried it by 2.5%, with a 4.5% Nadar vote. I have no idea where it was in 1994. I don't think the district has moved much. It contains some high income Portland suburbs, and Salem, and some of the norht Oregon coast. It is kind of a mixed bag.]

PA-13

1988: R +6.40
1994: R +0.46

The Philly 'burbs were moving left. [They sure were]

PA-18

1988: D +12.38
1994: D +6.02

While western PA was moving right. [Not this innner city Pittsburg district. Bush 2000 lost the district by 16%]

TX-09

1988: D +7.63
1994: R +0.59

TX-13

1988: R +0.59
1994: R +9.93

And Texas was absolutely closer to the 1994 figures above than to the Dem leaning 1988 figures. [The areas represented by these two districts certainly were.]

VA-11

1988: R +6.54
1994: R +3.76

Northern Virginia moving left. [definitely]

WA-02

1988: D +6.93
1994: D +0.81
[Bush 2000 lost by 2%, with Nadar getting 4.5%, so no, 1988 is a better measure. This district is the one just south of Vancouver on the coast.]

WA-04

1988: R +4.54
1994: R +9.01

WA-05

1988: D +3.17
1994: R +2.66

[Ya, there are the inland districts, and they were moving right in a hurry]

And again the inland Northwest, no doubt moving right.

[So a mixed bag. No one measure is without flaws. Good job!]

[AntiGuv]

For some silly reason I was thinking of the new PA-18 (Pittsburgh suburbs/exurbs) rather than the old PA-18 (Pittsburgh innercity).

Otherwise, I don't much disagree with you on anything except perhaps Arizona, which I think took a notable move leftward in the early to mid Nineties and then drifted somewhat back to the right in the late Nineties.

[Torie]

I might as well comment on Georgia:

GA-08

1988: R +9.34
1994: R +1.01

GA-10

1988: R +10.93
1994: R +2.28

Georgia's an interesting case. It's really the only one where I almost prefer the 1988 numbers to the 1992/1996 average. Then again, Clinton clearly did relatively well in Georgia by comparison to Dukakis who super-flopped. [Bush 2000 carried GA-8 by 14.5%, and GA-10 by 10%. These are racially polarized districts, in 1988 as much as in 2000. The 1988 numbers are clearly superior IMO.]

[AntiGuv]

I agree with you on Georgia. I shouldn't have said that in every instance the 92/96 figures were better than the 88 figures. But in general the 92/96 figures are superior IMHO. There are a few exceptions though.

[AuH2ORepublican]

I think the 2000 (assuming that 40% of the Naderites would have stayed home while 70% would have voted for Gore, 25% for Bush and 5% for third parties) and 2004 numbers are superior to those of 1988, 1992 or 1996, since they were the first real post-social-polarization elections and in each case went down to the wire and involved a "mainsteam conservative" Republican against a "mainstream liberal" Democrat. The 1996 results are very similar to those from 2000 if we add the Perot numbers to the Dole numbers---try it district-by-district, and you'll see that the deviation is rather minor (of course, that's not how people would have voted in 1996 had Perot not been on the ballot, otherwise Dole would have actually won with 271 EVs by barely carrying PA). The 1992 election, if you adjust for the Perot factor (I assume around 40% of the Perotistas would have stayed home, with around 58% of those who would have voted choosing Bush, 37% Clinton and 5% other candidates---Bush would have won 274-264 by barely carrying CT, IA and ME) was in a sense the last of the Cold War elections and the first of the socially polarized elections, with Bush doing less well in Northeast and big-city suburbs than in 1988 but better than his son did in 2000, but doing less poorly in rural areas than he did in 1988 but no as well as his son did in 2000. The 1988 election was affected by things like low farm product prices and job losses in the Rust Belt, as well as suburban fears of high taxes, and thus are quite dated. Just my two cents.