Introduction:
As you may have heard, the Pennsylvania court system is currently hearing a challenge to the state's ludicrously gerrymandered congressional map. If that challenge succeeds, the court may redraw several congressional districts, or even the entire map. What would a fair map look like? This is my interpretation of a one. It complies with the VRA and preserves communities of interest the best they can be.
A Note About Counties
Some people like to pay close attention to county boundaries in creation of congressional maps. And while I can understand why, I ultimately think counties have very little to do with congressional maps. They are a unit of state government, and often have little to nothing to do with communities of interest, which are more relevant for congressional districts. The only argument I can think of for using county boundaries in drawing a congressional map is that counties elect officeholders, and those officeholders may control local political culture or want to run for congress. I personally don’t think congressional districts should be made with who would be running there, outside of VRA districts. So the first thing I did once I started making my map was turned off county boundaries.
Maps:
Districts:
The intent of this section is to provide an explanation of 1) why this district has the shape it does 2) how the new district differs from the old 2) what I expect the new district’s political culture to be like, and 3) what this will mean for the current representative.
PA-01
Philadelphia really should be able to elect two black representatives. The main reasons it doesn’t is that the VRA is usually interpreted by Republican legislatures to mean a district has to be majority black by a sizable margin, and the white Philadelphia establishment has pushed to maximize their power in redistricting. I’m attempting to draw districts that will elect two black members. The first has traditionally been a more WWC district that follows the river. It’s more or less specifically drawn for Philly machine boss Bob Brady, who’s represented a version of the district since 1998. It’s also been essentially controlled by the Philly machines, with the major exception being Tom Foglietta, whose tenure began as a result of Michael Myers’s (no, not that one) involvement in the ABSCAM scandal.
This district’s white population is much less Philly based, and has none of the original’s territory above the river, or its considerable Hispanic population. Instead is more of Philly’s black population, and whiter, more Republican voters around Chester. The old 1st is 46% black, and this one is only 47%. But the difference isn’t much in the black voters. It’s in the other voters. Because most of them are outside of Philly instead of inside it, Brady will have a much harder time keeping the monolithic grip on them he needs to maintain his control of the district. That is, if he escapes indictment. While it’s not guaranteed that the new 1st will elect a black representative, I think the likelihood is high.
PA-02
The current PA-02 is 61% black. As I alluded to above, this is overkill, and done with the intention of minimizing the power of the black vote. So for this iteration of the 2nd, I pulled some of those voters out, making it “only” 53% black. The big difference is that its chunk of Montgomery County is gone, and replaced mostly with more of Philly. Overall though, it’s a fairly similar district. Since Dwight Evans is only a freshman, a primary challenge might be in the works due to the new territory, but Evans is also the giant killer after toppling Fattah, so he’ll probably stay the representative.
PA-03
I tried to respect the current naming system as much as possible, which means keeping utter nonsense like 1, 2, 4, 6, 7, and 8 being in SEPA, while 3 and 5 are all the way across the state. At any rate, PA-03 in its current form has one glaring issue. It cuts in half Erie, a small-to-medium sized city in the corner of the goddamned state. There’s no non-partisan explanation for that. So I included almost all of the Erie metro in the redraw. I also decided to stick with the OH border and take in the small manufacturing towns, instead of cutting almost all the way east to Indiana (the city), like the current one does, in what was another blatantly partisan move. I did opt to keep Butler in the district, however, since it has more in common with Beaver and New Castle than with Pittsburgh or vast expanses of nothing.
This PA-03 is a decent amount bluer than the old one, going for Obama in ‘08 50.7-48.0 instead of the current one’s 46.3-52.3 margin. Of course, since 2008, this area hasn’t moved right as much as teleported, and is probably too Republican to be competitive under this map either. I’d place it at about 40-56 Trump and R+6. That’s only winnable for a Democrat if the stars align. Mike Kelly, who lives in Butler, coincidentally, would be a heavy favorite for reelection.
PA-04
This is the district that I felt most unsure about. Basically, the question was whether Harrisburg, Lancaster, and York were more similar to each other to to the rural areas that surrounded them. I eventually decided to keep them together, because while each city has an individual culture and identity, the political desires of each city is more in line with the other two cities than with the “Pennsyltucky” that surrounds them. So while it may be a bit visually odd, I believe that it’s the best that can be done from a community of interest standpoint.
The new 4th is almost perfectly balanced politically. Obama won it 53.1-46.0 in 2008, an entire .2% off of his nationwide total, and at 73% non-Hispanic white, 11% black, 11% Hispanic, and 3% Asian, it’s not far off of the nationwide racial composition. The mixture of small-scale urban and suburban residents, along with some rural ones in between the cities, also closesly mirrors the nation as a whole. In terms of political swings, I believe it’s more or less followed the nationwide results as well, making it a perfect swing district. Scott Perry represents about 2/3 of the new district, and lives in Harrisburg, so he’d be the obvious choice for Republicans to run here. And Perry, while conservative, doesn’t play as a tea party wingnut, so he might be favored here if it weren’t 2018. As for Democrats, if Lancaster mayor Rick Gray ever wanted a promotion, the time would be now.
PA-05
This district is the largest in the state, as well as the most rural. There are no major cities, and it only touches into the exurbs of Pittsburgh and Erie. The only problem with the current 5th, really, has been the inclusion of half of the city of Erie, for blatantly partisan reasons. Since Erie is now in the 3rd, the 5th takes on even more rural territory. It currently, and going back as far as 1968, has had State College. And that’s partly because there’s just no good place to put State College. It’s a medium sized city that’s younger, educated, and diversifying. The 5th is decidedly none of those things. So I took it out, and now the 5th has nothing resembling an urban center whatsoever. The largest municipality is 10,000 person Oil City. From my description, you probably guessed that it’s very Republican. McCain won it by “only” 20 points, and Trump probably won by 40 to 50 points. Congressman Glenn Thompson, from Oil City, will likely not feel the effects of the changing boundaries much.
PA-06
What to do with a city like Reading, a plurality Hispanic city of 90,000 that's in such an economic spiral the census department recognized it for having the highest poverty rate in the country? An obvious answer might be to combine it with Allentown. But that area is just a bit too big and too far away for one reasonable district. You could make an argument to combine in with Lancaster, a somewhat similar city, but you wind up with an ugly arcing district that ruins the 4th, and sort of squishes the other SEPA districts. So that leaves putting Reading in a primarily white and rural district north of the Philly metro, west of the Lehigh Valley, or adding to the outer suburbs of Philly.
Neither option is good, really, but I decided that the exurban option was less bad. For one, the other residents of the former hypothetical district have a lot of, we'll say "economic anxiety", and may be antagonistic towards the political aims of Reading. For another, there are a fairly sizable number of Reading suburbs, who would fit in less than terribly with the outer Philly suburbs and exurbs. So I choice that option. The new district is fairly compact geographically, and certainly better than the current 6th, of which it was once said "looms like a dragon descending on Philadelphia from the west, splitting up towns and communities throughout Montgomery and Berks Counties".
The new 6th went 56.0-43.0 for Obama in 2008, but wasn't as friendly to Democrats in 2012 or 2016. In 2012, the Philly portion of the district voted more like its ancestrally Republican nature would suggest. In 2016, that part of the district rejected Trump, but Reading fell in love with him. So while the current 6th is one of the most tightly divided districts in the country, being a narrow win for Romney and Clinton, this district is maybe 3-4 points bluer. The downballot Republican sympathies of the Philly suburbs mean that it's more or less a tossup in a neutral environment. Ryan Costello really has no choice but to run here, but in a year like 2018 and a slightly bluer district, he couldn't be called the favorite.
PA-07
The current 7th congressional district is a bizarre and borderline artistic testament to Pat Meehan's disdain for the concepts of communities of interest, good governance, and geographic contiguity. The reason for this is that the natural design for the 7th, which I've drawn here, a fairly compact slab of inner Philly suburbs with comparatively high diversity and education rates, is too Democratic for him to win. After Obama managed a 59-41 win in 2008, it seems doubtful that a Republican could win here, even with all the ticket splitting going on. After Clinton won this district by something like 58-38, it looks impossible. Pat Meehan still lives here, but he may simply retire rather than face near certain defeat.
PA-08
The 8th is currently probably the least egregious district in the state. This map doesn't change it by too much. It does move west a bit to make up for the loss of its southern border to the 13th, but its essential character is still the same. It's a district of Levittowns, even if Levittown itself is technically drawn out of this version. Politically, little changes either, with the 8th becoming only slightly (0% to .5%) more Democratic. Brian Fitzpatrick, a Levittown resident, may want to move a bit north to be in the district again, but the fundamentals change little. This district remains very competitive.
PA-09
You know how I said that State College didn’t really belong in either of the districts it could be placed in? Well, that includes the 9th. But the 9th is the less bad option, since it’s mostly dominated by other micropolitan areas like Johnstown, Altoona, Indiana, and Somerset. The 9th is an amalgamation of those kind of towns, and some truly rural territory to the east. It’s safely Republican, and Bill Shuster would definitely stick around. But he may also be joined by Keith Rothfus, whose 12th gets pretty thoroughly dismantled under this map.
PA-10
This iteration of the 10th district loops around the urban centers of the Wyoming Valley, hitting some small population centers like Williamsport and New Castle, but staying out of any larger cities. While it contains more of the current 10th, there’s also a lot of the 11th. Either way, it’s very Republican, even if it’s similar to the district Chris Carney held onto for two terms. Tom Marino is likely to hold onto this one, especially since Lou Barletta will be jumping out of house for a senate run next year.
PA-11
There's not a lot to say about this one. It contains the rural and small town areas to the west of Harrisburg and York, in the southern Susquehanna Valley. Gettysburg, Carlisle and Chambersburg represent the upper bound of how urban this district can get. It’s an area that’s been traditionally Republican since before the civil war and will not change now. No current representative lives in this district, but Scott Perry may want a safe Republican seat instead of sticking around in his now competitive 4th.
PA-12
The 12th is a region in decline. It’s not the post-apocalyptic manufacturing graveyard of say, Cleveland, but the blue collar union workers who once built up towns like Penn Hills no longer have as many of those blue collar union jobs, and the 12th runs along the southern border of the better-off 14th, picking up those depressed towns that used to be where Pittsburghers went because they had enough money for a 2.5 kids and a white picket fence. The 12th winds up having a larger black population, 13%, than you might expect, since Pittsburgh had somewhat less racial strife and segregation than other rust belt cities.
Politically, it’s very competitive. Obama won it 54.1-45.0, but as might be expected for a whiter less college educated population, it’s moved right. I’m not sure if Clinton or Trump won this one, but it was close either way. The current 12th doesn’t share any territory with this one, and Keith Rothfus doesn’t live near Pittsburgh, so he’d have to choose between running here, in a competitive general, or in the 9th, with a competitive primary. This district would have been good for Tim Murphy to run in if it hadn’t been for the whole...situation. As it stands it’s likely to be an open seat tossup.
PA-13
The current 13th is somewhat defensible, I suppose. It sticks close to Philadelphia, and most of the oddness in its shape comes from the 2nd taking out chunks of its black population. But I still have some issues with it, and the way I drew my 2nd only exacerbates them. The 1st and 2nd take up the heavily black areas of the city. I intentionally kept the Hispanic parts of the city out of my 2nd, for two reasons. The first is that those voters are as a whole straight ticket Democrats, so including them in the 2nd means the black population would need to be higher to guarantee black voters get their choice. The second is that Hispanics are themselves an underserved Demographic and putting them in a black VRA district sort of drowns them out. So the 13th includes both the Hispanic population of Philadelphia, and the white Philadelphia population to the north and south of it, making about half the district.
There are two choices for who should join them in the district- to the northwest are some of the richest neighborhoods in Pennsylvania, and to the northeast, following I-95, are more modest neighborhoods, including Levittown. I chose to include the latter because it seems to me that Northwest Philadelphia has more in common with the middle class neighborhoods along I-95, which are more tightly bound to the city than their geography would suggest, and because I didn't want to put a lot of rich neighborhoods into a mostly middle and lower class district. That being said, I don't particularly like this 13th. The Hispanic portion especially feels out of place. Brendan Boyle should be expected to retain this still heavily Democratic district.
PA-14
Ever seen the movie Videodrome? I love it. It’s about a mysterious sado-masochistic TV station that has evil intentions. The station broadcasts from Pittsburgh, which is portrayed as run down post-industrial wasteland. At one point in the movie, a character, before killing a man, says “See you in Pittsburgh” instead of “See you in Hell.” And in the 1980s when that movie was made, that portrayal of Pittsburgh wasn’t all that inaccurate. The city itself was a bleeding jobs and people, and was surrounded by mostly decaying suburbs, except for the growing and comparatively affluent suburbs to the north. So it made sense then to cut off the northern suburbs from the Pittsburgh based 14th, and put them in the 12th.
But it’s not the 80s anymore. It’s not even 2010 anymore, and Pittsburgh is now a revitalized city that has left its depressed steel town past behind it. So when I asked the question of which towns around the city best represent it and thus should be combined to make a full district, I decided that the answer was finally the educated an affluent suburbs to the north, between the Allegheny and Ohio rivers, extending north into the exurbs close to Butler, and a few inner suburbs to the south with similar character, most notably Mt. Lebanon. Obama won this district 57.0-42.1 in 2008, and the district’s trended left since then, meaning Clinton almost certainly won by at least 20 points. Mike Doyle would have right of first refusal here, and it’s hard to imagine him passing it up.
PA-15
The Allentown/Bethlehem/Easton area is allllllmost big enough for its own district, which is part of why the 15th's distorted shape is so appalling. I'll be honest, I filled in the last bit of the district because there was some space left over to the north before the Wyoming Valley district started. But I think it worked out well. The metro extends to the north more than the south, which is more Philly-dominated. Politically, this district swingy. Obama did very well here in 2008, winning it 56.1-42.2, but it took a turn for the Trump in 2016. The results were very close, but I believe Clinton won here by fractions of a percent. Charlie Dent could probably hold down this district easily, but he’s retiring. As it is, it would be very competitive, voting to the left of the country in 2012 and the right of it in 2016.
PA-16
This is a pretty straightforward district, very similar to the 11th. It mostly lies in Pennsylvania Dutch country, bounded by the Philadelphia metro to the east, blue-collar Biden-land to the north, the state line to the south, and the Susquehanna to the west. It’s fairly white, fairly rural, and fairly Amish, as well. Lloyd Smucker is the only Republican with a credible claim to this district, even though he’d live a few miles out of it under these lines.
PA-17
The’s a lot to say about the 17th. For one, it had the second largest Romney → Trump swing in the country. For another, it’s one of the stupidest districts in the country. It was designed to capture the areas of strength for a conservative Democrat who then immediately lost it in the primary. And for another, so it’s not as much of a Democratic vote sink as it could be, but it’s bluer than they could have drawn it, so it’s not really that competitive for Republicans, even though it could have been under different lines. It’s just a mess of competing interests that helps no one out. This 17th keeps things more compact, taking in the Wilkes-Barre, Scranton, Hazelton, East Stroudsburgh, and Carbondale areas, without much more. This has the effect of making the district more centralized and more coherent as a district focused on industrial cities, rather than a combination of industry, mining, logging, and recreation like the old one.
What this does in terms of partisanship is a slight move to the left. It started out about a point to the left of the current one, and without Shuykill County, swings a bit less hard to the right. But Trump still wins it, though only 45-51 or so instead of his 43-53 win of the current one. That's kind of astounding, really, that by ungerrymandering a Democratic sink I could make it bluer, but I told you the current 17th is stupid. As for Matt Cartwright, you can tell he was spooked by Trump's win in his district last year. He's taken a few conservative votes and has been fundraising like a madman. This district might soothe his fears a little, but it's still a majority Trump district that will be competitive unless a wave develops.
PA-18
I like how this district mostly stays away from the Pittsburgh area and is more of small industrial cities and coal mining towns. Greene County and Mt. Lebanon should not be in the same district.
In 2000, Republicans drew a truly insane district to limit Democrats to one seat in this corner of the state. In 2010 they drew something nearly as bizarre to lock Democrats out completely. Now, a completely fair district wouldn't even be close. Pro-life, pro-gun union Democrats used to dominate in this region, especially along the Monongahela. But now Pennsylvania's own little slice of West Virginia is heavily Republican. If the upcoming special produces a Republican, they'll probably stay here. If it produces a Democrat, there's no way they wouldn't jump ship for the 12th.
Summary:
I didn't draw this map with partisanship in mind, but I think it turned out pretty balanced. There are 7 safe Republican seats (3, 5, 9, 10, 11, 16, 18), 5 safe Democratic seats (1, 2, 7, 13, 14), and 6 swing seats (4, 6, 8, 12, 15, 17). But of those swing seats, Democrats are probably at an advantage in three of them (6, 15, and 17), so on the whole it's a pretty balanced map. The only district that will necessarily flip is the 7th. In the wave environment that I expect to be developing next year, Democrats could be favored in all those swing districts and maybe even have a shot in the 3rd. Which is how it should be. Republicans would be in the reverse situation if a wave we're creating for them.
Overall, I like parts of this map better than others. I'm really pleased that my urban+upscale suburbs/downtrodden suburbs division worked out as well as it did. I like most of my map in the west, really. It hits the big divisions: weary rust belt (3rd), revitalized Pittsburgh (14th), seen better days Pittsburgh (12th), West Virginia (18th), and empty land (5th). I like how I probably added a second black representative without doing anything too messy. I'm happy with my 4th, which is very competitive and just the right size to fit all three cities.
But I'm not totally pleased. I don't really like my 13th, at least partly because there seems to be 3 district chunks. The 6th mixes affluent exurbs with the poorest city in America, which could cause issues. My 9th stretches far, being almost half as wide as the state, with no dominate theme except small cities.
So what are your thoughts? Would this be a fair congressional map that balances interests even? Do you have the way I cut up counties? Did I let some egregious typos slip through? Let me know.