Is Pennsylvania drifting towards solid GOP?

[The_Doctor]

A big question, what's the population growth trends?

Secondly, even as West PA goes Republican, Southeast PA (particularly Philadelphia and the suburbs) seem heavily trending Democratic. Pittsburgh itself voted like 80% for Clinton (or something around there).

Virginia started going Democratic because the fastest growing areas of VA were heavily Democratic (North Virginia) and they eventually began to outvote the rest of the state. From memory, PA's Trump counties are not in the same boat.

An interesting article here. I'll look at some old county results shortly.

[The_Doctor]

So, let's do Election 1980 (because I like Ronnie Reagan and it's the beginning of the last realignment). Reagan is also comparable as the last blue collar - esq Republican.

A factoid about PA by the way - PA and the Northeast are among the slowest growing states in the Union. When the Oldest & Wisest was elected, Pennsylvania cast 4,561,501 votes. In 2016, they cast 6,166,729 votes. That's a growth of 35% in 40 years. In comparison, Virginia's seen a 114% growth. To wit, VA was +13% for St. Reagan, and +7% for him in 1980.

In 1980, in "Clinton 2016" counties, 2,500,774 votes were cast. In 2016, 3,269,066 votes were cast. Comparatively, in Trump counties, 2,060,727 votes were cast in 1980 versus 2,897,663 votes in 2016. So, if you're keeping track at home, within Pennsylvania, Clinton counties grew 30.72% and Trump counties grew 40.61% based on the election results.

A fun fact: Philadelphia county cast 709,618 votes in 2016; 718,100 in 1980. It cast 690,776 votes in 2012. So, yes, Philadelphia is weakening but I would guess that the suburbs of Philadelphia are making up for this weakness.

Trump won purely because he won higher numbers in his counties than Clinton likely did in her counties.

So, actually, yeah, Pennsylvania is not becoming the new Virginia, I think. I'll do a Bush 43 v. Trump comparison too. Trump / the GOP probably can win PA in 2020 but until they make inroads and regain the blueing suburbs of Philadelphia that isn't a realignment.

I'll do a study of West Pennsylvania. I need to find a formal definition of West Pennsylvania first.

EDITED: Fixed the Clinton growth rate; it's 30.72%, not 73%.

[The_Doctor]

OK! So West Pennsylvania, defined here. I'll take out Allegheny since that contains Pittsburgh and is a Democratic stronghold (and I can't break out the suburbs outside the Pittsburgh returns for 1980).

In 1980, they cast 993,090 votes, and in 2016, they cast 1,192,862 votes. That's a 20.12% growth in vote totals since 1980. The state, remember, cast 35% more votes in 2016 than 1980, so Western Pennsylvania's growth has been 15% below the state's average, which means that the Democratic counties in Southeast Pennsylvania are possibly performing above the state average to compensate.

I would need to do county by county estimates of Trump v. Clinton and Reagan v. Carter but honestly, no, this is not the sign of a Republican Pennsylvania. This is a sign of a swing/lean blue Pennsylvania until the GOP manages to make inroads into the Philadelphia suburbs.

Trump and the alt-right people do miserably in these counties (the Philadelphia suburbs) and have to maximise their vote totals in West and Northeast Pennsylvania. which are already stratospheric. Based on Lackawanna County, Erie County, and Monroe County, the GOP could add 100,000 to 150,000 votes but without support from the Philadelphia suburbs they are at a ceiling. (That, of course, would mean PA would go Republican by 300,000 or something by 2020 if they did that). Which doesn't make it a GOP state; in fact, Kerry and Gore carried PA by about that margin.

[The_Doctor]

You must be logged in to read this quote.

Are you sure you calculated correctly? Because I'm getting 30.72% growth in the Clinton counties and 40.61% growth in the Trump counties, by your numbers, which seems correct, given the decline in Philadelphia. It's impossible for the growth rate of both to be simultaneously higher than 35%, like you got.

Yeah, you're right, sorry.  I was working off the spreadsheet. Whoops. The Clinton Philadelphia counties grew 64.11% between 1980 and 2016. But yes, that's correct. Too many numbers.

(The Philadelphia suburbs save my analysis; they're the fastest growing in the state. Sorry about the error).

[The_Doctor]

So a revised analysis then (Thanks for the catch, ossoff). Kind of riffing and developing an analysis.

So, to edit my earlier analysis, yes, Trump counties grew faster than Clinton counties between 1980 and 2016 by about 10%. Trump counties grew at 836,936 to Clinton counties growing at 768,292). The Clinton counties had a higher base (2.5 million to the Trump counties 2.060 million). So roughly +70,000 for Trump counties. And we obviously know that Trump outperformed Clinton in his counties than she did in hers.

But, it's not West Pennsylvania that's leading a realignment (or if there's one at all). Trending Republican certainly helped but they're not the fastest growing areas of the state. They're actually lagging in growth in Pennsylvania compared to the rest of the state. Philadelphia, Pittsburgh, and the suburban growth in the collar counties around Philadelphia and Centre is what's keeping Pennsylvania competitive for the Democrats and balancing out (roughly) the GOP increased strength in West Pennsylvania. As I said, Clinton Philadelphia suburbs grew 64% between 1980 and 2016 in election returns. As the Democrats lose strength in West Pennsylvania, they're picking it up in the suburbs of Philadelphia.

The operative question, if Pennsylvania was trending Republican, how are the fastest growing counties stacking up? Notably, if you look at the link about projected growth, the counties that are fastest growing are rooted in the southeast and central Pennsylvania. For example, Centre swung from Reagan +12 to Clinton +2 (and it was +10 Bush in 2000). They're projected for a 22% growth. Clinton County, PA (heh) swung from Reagan +12 to Bush +4 to Trump +34%. I'm not sure why Centre swung to the Democrats in 2016 but Clinton swung GOP for example.  It could be that Clinton, casting 13,000 votes is smaller than Centre (which cast 77,000), and probably attracts the type of Trump voters migrating there. 

An interesting factoid in York and Lancaster Counties, which were crucial to Trump: they were also huge for W. and Reagan. Lancaster backed Reagan 67-25 and Bush 66-31% and Trump 56-37% (a dropoff from W.). York County 60-33% Reagan, 60-36% Bush, 62-33% Trump. So we're seeing in one county Democratic strength but in the other stable roughly GOP strength. York is closer to the center of the state; Lancaster is nearer the Philadelphia suburbs.

Overall, I don't think the evidence points to a GOP realignment in Pennsylvania. Western PA's growth is slower than the rest of the state; Philadelphia's suburbs are growing faster than the rest of the state, and they're trending Democratic, even if the cities stay flat. Centre, for example, is a counterfactual against any GOP realignment in PA since it has trended Democratic compared to Reagan and W.   

The state's blue collar voter are not enough to lock in PA to the GOP. They need a suburban - blue collar alliance, and so far, the suburbs lean towards going that route. So, I'd still say PA's lean is slightly lean blue and as long as the suburbs and the cities vote Democratic, the GOP hold on PA will not be like VA, where the state is rapidly becoming a Democratic stronghold because the vote in North Virginia is growing and is heavily Democratic. PA may be a tossup for a while before settling in a distinct direction.

I definitely think a non-evangelical rooted moderate Republican would win Pennsylvania very easily; the trends are tailor made for the moderate Northeastern GOP stripe of Christine Todd Whitman and Arlen Specter, but harder to win outright for the alt-right and the Southern evangelical GOP.

So, if you are talking about a possibly realignment towards a Northern moderate Republican Party, absolutely. But in the current Southern and evangelical dominated, GOP, no.

[The_Doctor]

If you think about it, Trump was a non-evangelical type moderate Republican but held a ton of Southern baggage and the Southern coalition, which probably severely held down his margins in PA. in fact, Trump only ran 2-3% ahead of downballot Republicans. (Toomey and he won roughly the same amount) That's my contention.

The PA GOP suffers, basically, from suburbanites not liking them because they're tied to the national party and the national coalition. I think that forestalls any realignment to PA's GOP.

So, let's do the map this way. I'll total up the counties with population growth and maybe break them out. Since most of PA is losing population, this is easy.

https://www.indexmundi.com/facts/united-states/quick-facts/pennsylvania/population-growth#map

[The_Doctor]

So these counties posted above 0.5% population growth between 2010 and 2014.

Allegheny
Butler
Centre
Chester
Clinton
Cumberland
Dauphin
Franklin
Juniata
Lancaster
Lebanon
Lehigh
Montgomery
Montour
Northampton
Philadelphia
Snyder
Tioga
York

They cast 2,911,392 out of 4,913,119 votes in 2000, or 59.26%. In 2016, they cast 3,771,361 out of 6,166,729 votes, or 61.16% of the votes. I didn’t do 1980 because as far as population trends go, 2000 is far back enough and was easier to calculate. They increased their vote returns by 859,969  votes. Gore won 50.60% to George W. Bush’s 46.43% whereas Trump won 48.17% to Hillary Clinton’s 47.46%.

Democrat Al Gore won 1,562,323 votes or 53.66% of the aforementioned counties. Republican George W. Bush won 1,268,879 or 43.58% of the vote. That’s a margin of 10.08%. Third parties won 80,165 or 2.75% of the vote. So, in 2000, Gore outpaced his final statewide totals in the fastest growing counties of 2010-2014 by 3.06%.

In 2016, Hillary Clinton won 2,066,679 votes or 54.80% in the same counties. Republican Donald Trump won 1,538,464 votes or 40.79%. Third parties pulled 205,549 or 5.45%. That’s a margin of 14.01%. And that’s Hillary outpacing her statewide totals by 7.34%.

So, in fact, if Pennsylvania were trending Republican, the top growing counties in the state would also be trending towards the GOP. Yet, compared to 2000 (the last time a Republican won a first term for President), the Democratic margin in these counties actually widened and even more compared to their statewide totals.  Trump won by pulling votes from the counties that lost population. (Example: he won 72% in Cameron County, which saw a population drop of 5.5%).

This is all pretty incredible given that Trump actually won the state by about .71% whereas Gore won by 4.17%.

I didn’t include Berks County which grew .50% (right at the margin) but it doesn’t substantially change the results.

And yes, even if you pull out Philadelphia, the Democratic swing is still 4.06%. Trump lost 3% from W’s 2000 numbers while Clinton gained a point. The net margin is 107,717 for the Democrats from 2000 to 2016, without Philadelphia. Her non-Philadelphia numbers are 48.43% while Trump’s was 46.70%. Had these counties determined the presidency without Philadelphia, he would still have lost.

[The_Doctor]

I might do Michigan and Minnesota tomorrow using the same metrics to determine if it's going to the GOP or not. I'll take a look at it. Downballot, however, the PA GOP isn't doing well, as aforementioned. The Democrats ran ahead of the GOP in 2016, particularly because they won Erie County.

All in all, this points to another data point that this isn't a realigning election or confirming election for the GOP. Without PA, the GOP electoral college advantage goes down to 286, and if GA flipped, exactly 270. If you throw in ME to the GOP (which it has been trending in that direction), that's 274. That's of course assuming Arizona stays GOP on the state and national level.

Anyway, this means the Democrats would probably be favored in eight years, if the GOP won 2020 and would most likely retake Pennsylvania if these trends continue. It also demonstrates that PA will be closely fought in 2020 and the GOP margin won't substantially improve if the GOP wins it. I'd say a GOP margin out of PA would be 150,000 to 200,000 at best. That's contingent on flipping/padding margins in Erie, Luzerne, and Lackawanna.

The GOP has to flip these suburbs or the fastest growing areas and realign them. They're not doing that, so PA would still continue to be a problem for the GOP without effort.

EDIT: And Pennsylvania - population 12.8 million today - is projected to be 14.1 million by 2040. So that's an increase of 1.2 million over 25 years. Relative to other states, that's very low. This, of course, puts even greater political power in the hands of the fastest growing counties.

[The_Doctor]

The reason why Obama and Hillary won Centre County is because of the influence of Penn State's University Park Campus there.

I can't find exit polls in PA to corroborate this versus W's 53-42% win in that county. I know that nationally, 18-24 split in 2000 but in 2004 they had shifted to Kerry by 9 points. But Bush won the county twice; in 2000, 53-42% and 2004, 51-47%. Admittedly the margin shrunk from 9 to 4 points in 4 years... though Santorum carried it 62-34% in 2000.

Obama swept it by 12 points in 2008, then won it 1 point in 2012, then Hillary won it by 2 points in 2016. For the record the township data does say that in 2016, the state college borough voted by 65-27% Clinton. But nationally, in both 2008 and 2012, young voters swung to the Democrats by wide margins.

Are we sure that this is the reason? This county has undergone definitely a transition from being a reliably Republican county to a Democratic county with about 75,000 votes cast.

[The_Doctor]

You must be logged in to read this quote.

I agree, I was a bit tired when I wrote this. But I think upon review, I’m talking about people who would make it out of the GOP primaries. They’re more likely to be strong anti-immigration types like Donald Trump or Ted Cruz than moderate Republicans.

If you’re thinking Marco Rubio, that’s a possibility for someone who could pick up moderate suburban support and still retain West Pennsylvania’s growing GOP partisans. But increasingly, if you look at the swing of the suburbs, they’re not merely responding to specific partisans. They’re responding to the swing of the national GOP. 20 years ago, when George W. Bush ran for President as a far more moderate Republican on many issues (compassionate conservatism), he did far better than Trump did in the suburbs.

But I’ll get to that in the rest of your post.

You must be logged in to read this quote.

Correct. But Specter’s brand of Republican is actually a winner in the Northeast. It’s the philosophy that has enabled New York and New England Republicans to flourish. The Party’s shift rightwards has eliminated the Republican Party’s federal strength in the Northeast and New England because people like Specter have been forced to switch parties or fail to straddle the primaries and general election.

He won in 2004 against Pat Toomey only because of George W. Bush’s and the national Republicans coming to his aid. So I agree. But I also would speculate that Specter’s brand of conservatism was far more viable in the Northeast than Rick Santorum (who posted smaller margins than Specter in the General) or Pat Toomey (who twice, has posted slender wins).

The Republican Party, however, in Pennsylvania and nationally has changed. The Philadelphia suburban GOP voters are now becoming extinct because they’re switching to the Democrats.

You must be logged in to read this quote.

I agree completely. The protectionist message and immigration hardline played well in speeding up the realignment of West Pennsylvania towards the Republican Party. George W. Bush was a free trader and soft on immigration which angered West Pennsylvania and didn’t pick up enough suburban voters in the Philadelphia area to win (because he was socially conservative).

However, I contend Trump did not create a long lasting formula to win Pennsylvania. He drew from the counties that were losing population the most and pushed the Philadelphia suburban counties to embrace Hillary Clinton. The fastest growing counties, on balance, swung to the Democrats, as I showed.

You must be logged in to read this quote.

Well, that’s an interesting hypothetical. But let’s assume a conventional limited government Republican with protectionist and an immigration hardline (basically Coolidge of 1924, now that I think of it). Such a Republican would probably win Pennsylvania but it wouldn’t be a landslide. For instance, let’s do it this way. Let’s add all the counties Toomey won that Trump didn’t and see what we get. I’ll just recreate the margin.

The President won Pennsylvania by 44,292 votes. I’ll add the counties where Toomey won but he didn’t.

COUNTY | CLINTON MARGIN | TOOMEY MARGIN | Cumulative Swing

Bucks … 2,699 | 18,189 | 65,180 votes
Centre … 1,184 | 1,040 | 67,404 votes  
Chester … 25,568 | 6,110 | 99,082 votes

I won't do a full workup because I think this demonstrates the point well enough.

So, the Republican margin, at their best performances, might have been a 100,000 vote margin out of Pennsylvania for Trump. What’s interesting is that Toomey didn't, for the most part, run too far ahead or behind of Trump. Over 90% of Toomey voters were also Trump voters.

Roughly put, even the best GOP margin out of Pennsylvania might be 100,000 in 2016 and maybe 200-300,000 in 2020 (assuming Lackawanna and Monroe counties flips). In a 6.5 million electorate that votes in 2020, that translates into a 51-47% win roughly. A strong win but doesn't make Pennsylvania a GOP state.

The fundamental problem is that even with someone like Rubio, I don’t see the suburbs flipping GOP strongly. It would require the GOP to adopt a very different platform.

Not sure if you agree or disagree. But that’s my feeling anyway.

[The_Doctor]

Of interest, 18-24 voted 50-45% for Clinton in Pennsylvania. Pennsylvania is 81% white (so I'd assume 78-79% white in 2020). Trump's Administration probably dictates how this age cohorts would view the GOP and vote accordingly in 2020 and 2024. If he stabilizes and demonstrates he's a viable GOP model, younger Pennsylvanian whites might go GOP more heavily. If not, I can see the rising Generation Z types revolting and putting PA in the blue column again. To date, downballot, I still haven't seen the GOP strength that's been demonstrated elsewhere in the Midwest on the row office level. 

[The_Doctor]

Of interest, 18-24 voted 50-45% for Clinton in Pennsylvania. Pennsylvania is 81% white (so I'd assume 78-79% white in 2020). Trump's Administration probably dictates how this age cohorts would view the GOP and vote accordingly in 2020 and 2024. If he stabilizes and demonstrates he's a viable GOP model, younger Pennsylvanian whites might go GOP more heavily. If not, I can see the rising Generation Z types revolting and putting PA in the blue column again. To date, downballot, I still haven't seen the GOP strength that's been demonstrated elsewhere in the Midwest on the row office level.  

The 18-24 group was quite a spectacular change in PA. Of all the age groups, this was really the only one which underwent a fundamentally large shift to the Republican. Every other group was only small improvements, if that. It should be noted that most of the 18-24 year olds in 2016 would not be previous Obama voters.

Might be useful to keep an eye on Trump's approvals among 18-29 year olds in PA. I think he could just as easily lose this group than hold them, and overall it is in the GOP's best interests to make a positive imprint on them now... except that it's hard to see that being the case. Nationally at least, Trump is absolutely bombing with 18-29 year olds. It's consistently his worst age group.

Similar shift in WI, too. Correct this might be a shift among younger whites in the Rust Belt. We'll have to see.

[The_Doctor]

The reason why Obama and Hillary won Centre County is because of the influence of Penn State's University Park Campus there.

I can't find exit polls in PA to corroborate this versus W's 53-42% win in that county. I know that nationally, 18-24 split in 2000 but in 2004 they had shifted to Kerry by 9 points. But Bush won the county twice; in 2000, 53-42% and 2004, 51-47%. Admittedly the margin shrunk from 9 to 4 points in 4 years... though Santorum carried it 62-34% in 2000.

Obama swept it by 12 points in 2008, then won it 1 point in 2012, then Hillary won it by 2 points in 2016. For the record the township data does say that in 2016, the state college borough voted by 65-27% Clinton. But nationally, in both 2008 and 2012, young voters swung to the Democrats by wide margins.

Are we sure that this is the reason? This county has undergone definitely a transition from being a reliably Republican county to a Democratic county with about 75,000 votes cast.

Remember that in places with major colleges/universities, the decisive voting bloc is not the college-aged students (who, in many cases, have their votes count toward their hometowns, not their college towns), but the professors, lecturers, and other staff who work in these universities. These people are pretty much all highly educated, and thus tend to vote for left-leaning candidates.

In the case of Centre County, people who live in and around State College tend to vote Democratic, while people in the outer rural areas tend to vote Republican. Aside from Obama and Hillary, this county also voted for Bill Clinton in 1992 and 1996. Note that Bush won college-educated voters overall in 2000 and 2004, which is probably why he managed to win this county during his elections.

OK that makes more sense. Thanks.

I guess I've exhausted PA talk. I hope the OP thinks I answered him in detail anyway. It was an interesting topic.