Yesterday I created lists of the biggest gainers and losers among congressional districts over the period of 2000-10, but only hinted at the changes in racial composition that were underlying the overall population changes. A longer post about the racial composition (analogous to this one I did a year and a half ago) changes is in the works, but as part of that I conceived of this table... which really would have worked better with yesterday's piece, so I'm giving it its own home here. It shows the numeric change in each district, broken down by the numeric change among each race in each district.
What should stand out here is that among the 25 biggest gainers, in most of the districts, the combined non-white gains exceeded the (non-Hispanic) white gains. Among the few that didn't, some are districts that are either heavy on retirees (AZ-02, FL-05), some have a large Mormon population (AZ-06, UT-03), with a few a little harder to classify (GA-09 is sort of the exurban white flight receptacle from the rest of the Atlanta area, and ID-01 is a mix of a lot of Mormons and a lot of white flight from southern California). As always, as I've cautioned many times before, these districts aren't an immediate panacea for Democrats and look to stay fairly red for the short term; with most of these districts full of kids (kids who aren't likely to grow up to be Republicans, though!), gains at the ballot box are going to unfold slooooowly.
With the Census Bureau having completed its gradual rollout of data from all the states last week, I've finally gotten around to assembling data from all the various congressional districts into one place. While the actual population gain or loss in each district isn't as important a number, for SSP purposes, as the number of people each district will need to shed or gain as part of the redistricting process (which you can see in the various posts we did as each state's data came out), the overall gain and loss is an important part in the overall picture of where people are moving to and from (and where they're being born). Just the numbers of people moving in or out isn't as helpful as knowing who exactly these people are, and we'll delve a little more deeply into the changing racial compositions of the CDs in the next day or two... but for now, here are the overall population change numbers.
You're probably noticing, "Wow, that's a lot of Republican districts." That's certainly true, but these are also districts that (as we'll see when we talk about changing racial composition), for the most part, aren't becoming more Republican; people tend to bring their values with them rather than undergoing some magical David Brooksian conversion experience once they move in from the city, the inner-ring suburbs, or another country. Some of these districts are ones where much of the gains are Hispanic (like NV-03 or TX-10, or just about any California district on the list); in the case of GA-07, it's becoming more African-American. That isn't to say that these are all on the verge of becoming blue, of course; with much of these districts' non-white populations under 18, it'll be a gradual process. And redistricting is likely to de-diversify at least some of these districts, with some of the closer-in suburban portions of these districts (note that many of these districts are the ones right on the cusp of suburb and exurb) to be given to lower-population urban districts that need to expand outward, with the remaining parts of the districts staying red. (GA-07, again, is a case in point; the innermost parts of Gwinnett County, which are pretty diverse today, probably will need to get added on to underpopulated GA-05, leaving the rest of the district in very Republican-friendly condition.)
You may recall I did this same thing a year and a half ago when the 2008 estimates came out; there's been very little change to the list since then, although with some swapping of places. Despite its position at the absolute epicenter of the housing bubble, NV-03 moved up from 4th to 1st place, past the two Arizona districts and TX-10. Districts that fell out of the top 25 in 2008 include GA-06, TX-03, CO-06, FL-25, IL-14, and FL-06, replaced by VA-10, FL-12, TX-28, TX-23, TX-04, and ID-01.
Bono Mack (R)
And here are the biggest losers, looking every bit as heavily Democratic as the list of gainers is Republican. However, if you go through the list line by line, you'll notice that very few of these districts are even remotely-considered as being on the chopping block. That's partly because many of these are VRA seats, or otherwise set up by Republican legislatures as Democratic vote sinks (PA-14, for example). The most obvious exceptions up for elimination are PA-12, which almost everyone concedes is gone with the wind, OH-10, which is set to get mashed with OH-13, and possibly IL-17, ironically one of the few GOP-held seats on the list (although it might instead wind up getting turned into a significantly bluer district by the now-Dem-controlled Illinois legislature). Instead, as I mentioned earlier, many of these districts are going to wind up reaching out further into the suburbs... in many cases, expanding to follow the same constituents who just moved out of the city (for instance, all the Detroit residents who moved across 8 Mile into MI-12).
Today's the last day of Census data releases, meaning we have the complete set of all 50 states now. The Census Bureau released some data summarizing the entire nation, including what you'd think was the single most important bit of all, considering the way they hyped the announcement: the new population center of the U.S., still in south-central Missouri, but moving 30 miles to the southwest, now near Plato, MO. Perhaps more interestingly, they summarized the country's demographic change as a whole: that starts with the nation's Hispanic population crossing the 50 million mark, now up to almost 17% of the nation's population. Hispanics and Asians both grew at a 43% rate, and people checking "2 or more" races rose at a 32% rate. The non-Hispanic white share of the population fell from 69% to 64%. They also found a country that's more urban than ever before, with 84% of the country living in metropolitan areas now.
I know you're all champing at the bit to find out what happens in Maine, but there's this other state called "New... Something" that we should probably get through first. New York is one of only two states to lose two seats, from 29 down to 27. (Ohio was the other one.) New York's new target is 717,707, up from about 654K in 2000. Thanks to a few hundred votes in a couple of state Senate races that tipped that chamber's balance, the GOP managed to hold on to one leg of the redistricting trifecta, meaning that instead of a shot at a 26-1 Dem map, there's probably just going to be a shared-pain map instead with a GOP loss upstate and a Dem loss in the NYC metro area. That's despite the fact that New York City itself actually grew a bit, to 8.175 million, still by far the nation's largest city. (There are moves afoot toward an independent redistricting commission, but this doesn't seem likely to happen.)
In general, the heaviest losses were in the western part of Upstate, with the state's two biggest losers the Dem-held 27th (Buffalo) and 28th (Rochester). On the other hand, losses also popped up rather patchily in parts of the outer boroughs (especially the 11th in the black parts of Brooklyn... without much seniority, Yvette Clarke may wind up with the shortest straw among the NYC delegation) and Long Island (Peter King's 3rd... which would be a prime target for the 2nd seat to evaporate, if only the Dems controlled the trifecta here). The big gainers were both urban (Jerry Nadler's 8th, probably fueled not so much by growth in Manhattan as among Orthodox families in Borough Park in Brooklyn) and exurban (Nan Hayworth's 19th, at the outermost reaches of the NYC metro area).
While none of the districts in New York seem to be undergoing the kind of rapid demographic transformation that threatens the red/blue balance in any place like we've seen in Texas or California, a few districts are worth looking at just as an indicator of what an interesting tapestry New York City is. Take the 5th for instance (another possibility for wipeout, given its strange position straddling Nassau County and Queens, and Gary Ackerman's non-entity-ness): it's moved from 44% non-Hispanic white, 5% non-Hispanic black, 24% non-Hispanic Asian, and 24% Hispanic, to 36% white, 4% black, 33% Asian, and 26% Hispanic, close to an Asian-plurality, thanks to growth in the Asian community in Flushing. A few districts in New York City are getting whiter, thanks to hipsters and gentrifiers: the 11th moved from 21% white and 58% black to 26% white and 53% black, while the 12th moved from 23% white and 49% Hispanic to 27% white and 45% Hispanic. The Harlem-based 15th went from 16% white, 30% black, and 48% Hispanic, to 21% white, 26% black, and 46% Hispanic, while the remarkably complex, Queens-based 7th went the other direction, from 28% white, 16% black, 13% Asian, and 40% Hispanic to 21% white, 16% black, 16% Asian, and 44% Hispanic.
Now for the maine event! (Rim shot.) Maine's a lot like Rhode Island and New Hampshire in that the long-standing boundary between its two districts rarely seems to budge much, and this year won't be any different. Maine's target is 664,181, up from 637K in 2000. The disparity of a little more than 4,000 people means things won't change much; the Republicans control the redistricting process this year but there's not a lot of fertile material here for them to try to make swingy ME-02 much redder.
Rhode Island doesn't offer much for redistricting fans to sink their teeth into: it has two districts that are about equally blue, the Dems control the redistricting trifecta, and the disparity between the two districts, while not New Hampshire-close, requires only minimal boundary-shifting. Rhode Island's target is a tiny 526,284 (only up from 524K in 2000... Rhode Island had the smallest growth, percentage-wise, of any state over the decade, putting it 2nd overall behind only Michigan, which actually lost population). If this continues, there's the distinct possibility we could see Rhode Island reduced to one House seat come 2020. Also worth noting: Rhode Island had a lot of Hispanic growth over the decade, not quite on par with the Southwest but high for the Northeast; it went from 8.5% Hispanic to 12.4%, and Providence moved to a Hispanic plurality.
South Carolina is gaining one seat to move from six to seven; its new target based on 7 seats is 660,766 (it was 668K in 2000, so every district gained significantly over the decade). With the GOP holding the trifecta and much of the growth seeming to come among white retirees, look for the creation of one more Republican-friendly seat... with one possible wild card, that the Obama DOJ might weigh in and push for a second African-American VRA seat (theoretically possible if terribly ugly, as SSP's crack team of freelance mapmakers have shown here). The biggest growth has come in the coastal Low Country, rather than the fiercely evangelical uplands; I'd expect Charleston and Myrtle Beach, both part of SC-01 for now, to wind up each anchoring their own districts.
West Virginia is staying at three seats for now, although it might be headed for two seats in 2020, given its slow growth and low targets; its target is 617,665, only up from 603K in 2000. The 3rd, in coal country in the southern part of the state, is losing population (though not as fast as one might suspect); the 2nd needs to shed an amount equivalent to what the 3rd needs to gain, leaving the 1st pretty stable. Much of the state's growth is in the far east tip of the Panhandle (in the 2nd), especially Berkeley County, which serves as Washington DC's furthest-out exurbs. Dave Wasserman, who seems to get all the good redistricting-related gossip, says that while the obvious solution (moving Mason County from the 2nd to the 3rd, and calling it a wrap) still seems likely, the Dems who control the redistricting trifecta might want to cobble together a slightly Dem-friendlier 1st along the state's northern boundary that includes both Morgantown and the Panhandle exurbs (the only counties in the state that are getting bluer).
Today's Census data dump is three slow-growth northern states: Massachusetts, Michigan, and New Hampshire. Massachusetts is set to lose one seat (from 10 down to 9), meaning its new target is 727,514 (up from about 634K in 2000). Interestingly, the growth among all the districts was pretty consistent, with only about 20,000 difference between the state's largest and smallest districts. Estimates over the decade had shown Boston losing population, but in the final count it did eke out a small gain.
With no clear loser on the population front among the districts, that makes the question of who draws the shortest redistricting straw even more complicated... unless someone reverses course and decides to retire, either to challenge Scott Brown (most likely Mike Capuano or Stephen Lynch) or to call it a career (John Olver). Olver's 1st did wind up being the smallest by a small margin, so the most talked-about mashup of the 1st and 2nd may well happen; alternatively, based on seniority the axe could fall on the delegation's newest member, William Keating. At any rate, with Dems firmly in charge of the process, don't look for any of these districts to lose their bluish hues; the main question is who gets left without his musical chair.
When it was revealed in December that Michigan was the only state out of 50 that actually lost population since 2000, it was clear that the state's urban districts were in a world of hurt... but I have to admit I'm still surprised at the way that Detroit has utterly cratered. The Motor City, at one point the 4th largest city in America, is now down to 15th, with a population of 713,777 (now smaller than johnny-come-latelies like Columbus, Austin, and Charlotte). The 13th may be the 2nd least populous district in the country at this point (after WY-AL). I briefly had to wonder whether we might actually see Detroit turned into one CD, mostly contiguous with the city boundaries (since it's now about the same population as an ideal district), but I can't imagine that the Obama administration's DOJ would allow the state GOP (which controls the redistricting trifecta) to pack only one overwhelmingly African-American VRA district when the population is there to support two, albeit two that will have to reach significantly into the suburbs now.
Michigan's current target is 705,974 (based on the drop to 14 from 15 seats), up from about 663K in 2000. That means that six of its districts (the Upper Peninsula-based 1st, the Flint-and-Saginaw 5th, and the 9th and 12th in Detroit's northern suburbs, in addition to the 13th and 14th) outright lost population over the decade. With the 9th and 12th also big losers, and with the VRA looming over the 13th and 14th, this all seems to confirm what most people are expecting, that Gary Peters and Sandy Levin are going to get much better acquainted with each other in a Dem primary. If you go further out into the districts that contain Detroit's exurbs (the GOP-held 8th and 10th), those are the two districts in the state that actually need to shed some population.
These two district states are really drama-free, and New Hampshire might be the least dramatic of all. The two districts in the state stayed remarkably balanced (as they always do... the state has had two districts since the 1800s, with the boundaries rarely moving much), to the extent that the 1st needs to pick up only 254 people from the 2nd. I'll leave it to the good folks in comments to argue over which ward in Hooksett should be the one that gets moved. (New Hampshire's target was 658,235, up from 618K in 2000.)
Florida was one of the nation's biggest gainers, both in terms of overall numbers (18,801,310, up from 15,982,378 in 2000) and House seats (up two from 25 to 27, making it the only state besides Texas to gain more than one seat). Florida's new target is 696,345, up from 639K in 2000.
Most of the state's gains come in what's called the I-4 corridor, reaching from Tampa Bay through Orlando over to Daytona Beach and down the Space Coast. (Of course, that's not consistent from district to district; the only district in the state that lost outright population is FL-10 in St. Petersburg, and Tampa's FL-11 will also need to gain voters.) FL-05, centered in Pasco and Hernando Counties north of Tampa, is now one of the largest districts in the nation, in fact. Both of the new districts seem likely to be centered somewhere in the I-4 corridor, although there was enough growth in the Miami area that it will need to expand a little, too, shifting in-between districts like the 13th and 16th a step to the north. (Miami area growth was concentrated in FL-25 in Miami's westernmost suburbs; the rest of south Florida, especially the Gold Coast, seemed pretty stable). Despite the GOP-held trifecta, predicting the final map right now is a bit of a fool's errand, though, considering that the effect of Florida's Fair Districts initiatives will probably need to be filtered through the courts and the DOJ.
Florida, as you'd expect, is one of the states showing large-scale Hispanic growth. That's not as clear-cut in the Democrats' favor as it is in other states, in that it has a large Cuban community, although that's largely limited to the Miami area and Cubans are becoming a smaller percentage of the total Hispanic community even there. Hispanic growth in central Florida tends to be Puerto Rican and Central American. The state as a whole moved from 65% non-Hispanic white, 14% non-Hispanic black, and 17% Hispanic in 2000 to 58% white, 15% black, and 22% Hispanic in 2010. While the most heavily Hispanic districts, naturally, remain the three Cuban districts in the Miami area, most of the biggest increases in Hispanic percentage have come in central Florida. In particular, see FL-08 (18% Hispanic in 2000, 26% Hispanic in 2010), FL-11 (20% Hispanic in 2000, 28% Hispanic in 2010), and FL-12 (12% Hispanic in 2000, 21% in 2010). Could we see one of the new districts be a Hispanic-majority VRA district that joins Tampa, Lakeland, and Orlando? The biggest Hispanic percentage increase might surprise you, though: Debbie Wasserman Schultz's FL-20, which went from 21% to 31%, apparently based on a lot of Cuban movement to the suburbs further north).
Wasserman Schultz (D)
Georgia is gaining one seat, from 13 to 14, and with that in mind, its new target is 691,975 (up from 630K in 2000). Pretty much all decade, those in the know have been expecting Georgia's 14th seat to fall in Atlanta's northern tier of suburbs, where the state's fastest growth has been in distant exurban (and virulently red) counties like Cherokee and Forsyth. The new data basically confirms that, with the heaviest gains in suburban/exurban GA-07 (worth noting: Newt Gingrich's old stomping grounds, Gwinnett County, is now the state's 2nd largest county, having shot past Cobb and DeKalb Counties) and GA-09.
Perhaps most surprising is the deep deficit in GA-02, the VRA district in the state's rural South; there had been discussion of it reaching up to take in central Macon in order to make GA-08 safer for its new Republican occupant Austin Scott, and that seems even likelier now, given that may be the only way for it to retain an African-American majority. The two VRA districts in Atlanta will also need to expand outward, but third black-majority seat in the ATL area, the suburban 13th, has plenty of population to spare.
The changes in Kentucky are much less dramatic, which stays at six seats, has seen little change in its racial composition, and which probably won't even see much movement of its current boundaries. Its current target is 723,228, up from 673K in 2000. As in many states, the truly rural districts (in this case, the west Kentucky KY-01 and Appalachian KY-05) were stagnant, and will need to gain population from districts with exurban populations (KY-02, which includes Louisville's southernmost 'xurbs, and KY-06, centered on Lexington).
Today and yesterday's Census data dump is of three states that didn't gain or lose seats but will need some internal adjustment to reflect population movement from the cities and the rural areas to the suburbs: Minnesota, New Mexico, and Tennessee. (It also included three states with at-large seats that we won't need to discuss: Alaska, Montana, and North Dakota.)
Minnesota barely made the cut for retaining its eighth seat (13,000 fewer people statewide and it would have lost it), which you can see in its very low new target: 662,991 per district. (That's up from about 615K in 2000.) Despite the fact that Michele Bachmann lives there, people keep pouring into MN-06 in the outer-ring suburbs and exurbs to the north, west, and east of the Twin Cities. Only it and MN-02, taking in the southern suburbs/exurbs, will need to shed population, giving part to the rural 1st and 7th, and part to the urban 4th and 5th (and suburban-but-boxed-in 3rd). With split redistricting control, look for the parties, if they're able to agree, to settle on incumbent protection.
Talk of moving the college town of St. Cloud, currently in MN-06, into MN-08 (which would enable Tarryl Clark to run there) may be premature, as MN-08 gained enough population that it can remain about the same. In fact, the fact that it did so may say a lot about last year's election; the 8th's growth has been happening at its southern end, where the MSP exurbs begin and where new Rep. Chip Cravaack hails from, and the population growth in this area has outpaced losses in the dark-blue Iron Range to the north, Jim Oberstar's traditional turf.
New Mexico's target is 686,393, based on staying at three seats (up from 606K in 2000). Not much change needs to happen between the districts; the largely rural NM-02 will need to gain some population, probably from the southern suburbs of Albuquerque in NM-01. New Mexico has become appreciably more Hispanic over the last decade, though maybe not as dramatically as the other three border states (California, Arizona, and Texas), moving as a state from 45% non-Hispanic white and 42% Hispanic in 2000 to 40% non-Hispanic white and 46% Hispanic in 2010. That means that, since 2000, it has become the first state with a Hispanic plurality. The movement was fairly consistent among districts, with the 1st going from 42% to 48% Hispanic, the 2nd going from 47% to 52% Hispanic, and the 3rd going from 36% to 39% Hispanic (the 3rd, though, is the least-white of the three districts, thanks to an 18% Native American population, which stayed consistent over the decade).
Tennessee stays comfortably at nine seats, and its new target is 705,122 (up from 632K in 2000). It, like Minnesota, has seen a big population shift from cities and rural areas to suburbs and exurbs, as seen in the huge growth in the 6th (which half-circles Nashville on the east) and the 7th (a thin gerrymander that hooks up Nashville's southern suburbs with Memphis's eastern suburbs). In particular, western Tennessee, both in the city (TN-09) and the rural areas (TN-08) were hard-hit, with the 8th barely gaining and the 9th outright losing population. The GOP controls the redistricting process for the first time here, but with them up 7-2 in the current House delegation (and with Memphis unfixably blue), look for them to lock in current gains rather than getting aggressive with TN-05 (seeing as how Nashville could be cracked into multiple light-red urban/suburban districts, although that has 'dummymander' written all over it).
Arizona is gaining one seat, from eight up to nine, and that means that its new target is 710,224, up from 641K in 2000. Interestingly, despite the fact that it's gaining a new seat, there are still three currently-composed districts that are in a deficit and need to pick up people from elsewhere: the 3rd, 4th, and 5th. These are the three central districts in the Phoenix area that are essentially built out and can't expand in any direction (except up); meanwhile, the 2nd, 6th, and 7th can continue to expand every which way into the desert, which is precisely what they did over the decade, so look for one additional GOP-friendly seat to be carved out of Phoenix's endless suburbia (although whether it's centered in Phoenix's west or east suburbs remains to be seen... between the commission's role in deciding, and possible multiple incumbents opening up seats to run for the Senate, there really aren't any clues what will happen).
Like the other border states, Arizona has become signficantly more Hispanic over the decade, up to 29.6% Hispanic now compared with 25.3% in 2000. The Hispanic growth wasn't concentrated any one particular place: that 4% increase was closely mirrored in all the districts. The 2nd had the biggest Hispanic shift, at 7% (from 14% to 21%), while the 1st had the smallest shift, at 3% (from 16% to 19%). That dissipation of the Hispanic vote means that it's not terribly likely that a third VRA seat will be carved out, despite the fact that Hispanics are close to 1/3 of the state's population.
I'm not the first one to observe that Idaho redistricting is pretty much drama-free. Nevertheless, there's at least something interesting going on here in this small but fast-growing state: growth is very heavily concentrated in suburbs and exurbs west of Boise. For instance, the state's 2nd and 3rd biggest cities used to be Pocatello and Idaho Falls; now they're Meridian (a large suburb west of Boise) and Nampa (in Canyon County, the next county to the west). That means that the districts are kind of lopsided, and it looks like much of Boise proper, currently split down the middle, will wind up being given to ID-02. While Boise is certainly the most urbane part of the state, and it should tip the balance a bit in the blue direction (as for the past decade, the two districts have had almost identical PVIs), the 2nd should still be a long way away from somewhere the Dems can compete. (Idaho's target is 783,791, up from 646K in 2000. Look for it to get a 3rd seat in 2020.)
Wisconsin held steady at eight seats this year, and even its districts held pretty steady, too. Its target is 710,873, up from 670K in 2000. That means the only district that lost population is the Milwaukee-based 4th and even it only lost a few thousand since 2000. The main area of growth is the state's other blue stronghold, the Madison-area 2nd (must have something to do with THE BLOATED STATE GOVERNMENT AND THOSE GREEDY PUBLIC EMPLOYEES MULTIPLYING LIKE LOCUSTS!!!!1!!), which needs to give about 40,000 people to the 4th (although they'll have to pass through the suburban 5th, which sits smack dab between them). Also, it looks like Dairyland is gaining a little at the expense of the North Woods, as the 3rd will need to pick up 20K from GOP freshman Sean Duffy's 7th. Although the GOP controls the redistricting process here, thanks to their House gains in 2010 and the overall uniform swinginess of the rural counties, they're probably just going to be playing defense with their map.
Today is the flipside of yesterday's California release: states with stagnant populations and a bunch of old white people. We'll start with Connecticut, which is certainly characterized by stability: it easily retained five seats, not being particularly near either the cusp of gaining or losing, and even its five districts are pretty close to in balance with each other. Its target is 714,819, up from 681K in 2000.
Ohio is one of only a couple states to lose two seats, taking it from 18 down to 16. Its new target is 721,032, up from about 631K in 2000. The state as a whole didn't lose population (gaining 183,364), but seven of its districts did (the 1st, 5th, 6th, 9th, 10th, 11th, and 17th). The Columbus area was the only part of the state that seemed to experience robust growth; in fact, despite the state losing two seats, the 12th (a swing district held by GOPer Pat Tiberi) will actually need to shed population... much of the state's growth is accounted for in the growth in the 12th.
The numbers today don't really change the overall redistricting equation: one of the northeastern Ohio Democrats is clearly going to have to go, and while the Akron-area 13th (held by Betty Sutton) actually gained some population unlike its neighbors, it may be the one that gets dissected simply by virtue of being in the very middle (with the 9th pressuring it from the west, the 10th and 11th from the north, and the 17th from the east). As for which GOPer gets cut, I'd expected it to be one of Bill Johnson (in the 6th) or Bob Gibbs (in the 18th), but the 18th, despite its mostly rural, Appalachian flavor, seemed to hang in there better than expected, population-wise. Now I'm wondering if Bob Latta's 5th in the rural northwest, which is going to be pressured by the 9th to its north and the 4th to its east, may be a more natural target. Or here's another possibility (made likelier by the possibility that the local GOP might like rid themselves of a liability in the form of Jean Schmidt): the 2nd might be targeted, despite its decent numbers, as both the 1st to its west and the 6th to its east need to gain a ton of people (and extending the 1st east into red, suburban Clermont County would make GOPer Steve Chabot's life easier).
Pennsylvania's target is 705,688 based on the drop from 19 to 18 seats, up from about 646K in 2000. The 2nd, 3rd, 12th, and 14th all lost population. I'd really recommend looking at the Census Bureau's interactive map of Pennsylvania, as it shows exactly what's going on: the eastern half of the state gained a bit, while nearly every county in the state's western half outright lost population. In fact, there were enough gains in the east that four districts wind up needing to shed population: the 6th and 15th in the Philadelphia suburbs/exurbs, and the more rural, Pennsylvania Dutch-flavored 16th and 19th. These are all Republican-held districts, but these are all districts that moved sharply in the Dem direction from 2004 to 2008, while on the other hand, the shrinking western districts are Democratic areas but ones where the overall trend has been away from the Dems. (Interestingly, two cities that over recent decades came to symbolize dead northeastern industrial centers, Allentown and Reading, are actually rebounding, gaining around 10,000 people each and helping to grow the 15th and 16th respectively. Much of the growth in those two cities, though, as well as the small growth experienced in Philadelphia, is Hispanic.)
With the GOP in control of the redistricting process in Pennsylvania and the population losses heavily concentrated in the Pittsburgh area, it looks like the axe is going to fall heavily on fairly-new Dem Mark Critz in the odd-shaped 12th, which was designed to be a friendly district for John Murtha cobbling together Cambria County with the Dem-friendly parts of Pittsburgh's collar counties but is barely holding onto its Dem roots these days. Mike Doyle's 14th (in Pittsburgh proper), despite being the biggest population loser, is probably going to stay intact, as Republicans will need to concede at least one blue vote sink in the southwest (and probably get bluer, as it'll need to expand into the dead steel towns of the Mon Valley to its south, currently the bluest part of the 12th).
If Critz wants to stick around, he's likely to find himself either fighting Jason Altmire in a primary in the 4th or Tim Murphy in a general in the 18th (although Critz has enough of a Johnstown-area base that he might be able to pull out an upset in whatever district Johnstown winds up in, unless the GOP decides that the 9th, in the central part of the state, is red enough to safely absorb Johnstown).
The crown jewel of the 2010 Census is out: California. The nation's largest state is, well, even larger than before, at 37,253,956, up from 33,871,648. Divide that out among 53 districts (it was the first time in ages that California didn't gain a House seat, despite gaining more than 3 million residents... it gained at a rate close to the country as a whole), and you have a target of 702,905, which is up from about 639K in 2000.
It may not come as a surprise, but much of the state's growth is Hispanic. Since 2000, the state's Hispanic population grew 27.8%, while the state's non-Hispanic population was almost stagnant, growing only 1.5%. (The Asian population grew 31.5%, but that's a fairly small subset of the overall population.) In 2000, California was 46.7% non-Hispanic white and 32.4% Hispanic, but in 2010, it had drawn much closer: 40.1% non-Hispanic white and 37.6% Hispanic.
Looking at the table, you'll notice that a large number of districts have moved from white pluralities to Hispanic pluralities over the last ten years: the Democratic-controlled 17th, 23rd, and 27th, and the Republican-controlled 21st, 44th, and 45th. (The latter two were also the state's two fastest growing districts, both in Riverside County to the east of Los Angeles.) Two more GOP-held seats in the greater Los Angeles area are also dancing close to the edge of a Hispanic plurality: the 25th, and the Orange County-based 40th. Of course, that doesn't presage an immediate change in voting patterns; given lower Hispanic voter participation rates and the fact that much of the Hispanic population is under 18, changes will be slow to happen. Case in point: the 20th, where incumbent Jim Costa had a close call in 2010 despite it being a 70% Hispanic district! (One other bit of trivia: Pete Stark's 13th moved from a white plurality to an Asian plurality, the only Asian-plurality district outside of Hawaii.)
One other thing you'll notice: despite the fact that California didn't lose a seat, there is going to be substantial reconfiguration of districts, with boundaries moving from west to east. The Bay Area gained little population, and will need to give most of a seat to the Central Valley; likewise, Los Angeles County proper gained little, and will need to give most of a seat to the Inland Empire (San Bernardino and Riverside Counties). Although the Central Valley and Inland Empire tend to be Republican areas in general, most of the growth in those places has been Hispanic, to the extent that "new" seats are probably going to wind up being Hispanic VRA seats carved out of the general overlay of red; on the other hand, the Bay Area and LA proper are already Dem strongholds and have nothing but Dems to lose, so the overall effect is likely to be a wash. Of course, given that this is the first year that California switches to an ostensibly impartial commission, which has no compunction to preserve the incumbent protection intent of the 2000 map and may actually place a premium on compactness, we could see all manner of scrambling that goes well beyond what I'm describing.
While we aren't going into as much detail as we did with Texas, we're adding a few details to California that most states haven't received: each district's representative (as it's well nigh impossible to keep track of which district number is what when there are 53 of them), and the district's racial composition in both 2010 and 2000. The four categories expressed as overall percentages, left to right, are non-Hispanic white, non-Hispanic African-American, non-Hispanic Asian, and Hispanic.