The folks at University of North Carolina’s Carolina Demography issued new projections as to which states are expected to gain and lose seats after the 2020 round of congressional reapportionment. Eight states, all in the Midwest and Northeast, would lose a seat, while eight other states, all in the South and West, would likely gain a seat, with Texas likely gaining two. (The map above highlights this continued geographic trend of population shifting to the Sun Belt.) UNC also outlined a variety of other possible scenarios, all of which you can see in the table below:
(Note that If all the “likely” results came to pass, nine seats total would be gained against just eight seats lost, which is obviously impossible, but this is just a projection.)
Initially it would seem that these seat changes would further harm Democratic attempts to regain a majority in the House, once you also factor in the redistricting that would take place following reapportionment. However, as the analysis below shows, there may be less of an impact than you’d expect. As far as redistricting is concerned, we’ve also had to assume that those in charge of drawing the lines (typically legislatures and governors) will remain as they currently are. That’s a big assumption, of course, because partisan control in at least a few states is likely to shift by 2020. But many will stay the same, and it’s a good starting place to get a feel for what the decade after this one might hold in store.
Single-Party States
West Viriginia: 3 seats; current delegation: 3 R; projected: -1 seat
Rhode Island: 2 seats; current delegation: 2 D; projected: -1 seat
These two losses will almost certainly cancel each other out. All five seats are safe for the party currently holding them, so when the 2022 elections come around, one member of Congress in each state is either retiring or losing a primary.
Dominant-Party States
California: 53 seats; current delegation: 39 D, 14 R; projected: +1 seat
Oregon: 5 seats; current delegation: 4 D, 1 R, projected: +1 seat
Illinois: 18 seats; current delegation: 10 D, 8 R; projected -1 seat
Democrats are the dominant party in all three of the above states, and population growth trends point toward any new seats being favorable to Democrats. California’s independent redistricting commission will keep seats more competitive than you’d see otherwise, but shouldn’t prevent Democrats from seeing continued success there. Assuming Democrats retain control of the redistricting process in Oregon, they should be able to ensure a map that will elect five Democrats and just one Republican.
While Illinois currently has a Republican governor, the Democrats have veto-proof majorities in both chambers and would therefore currently be able to pass their own maps over any objections by the governor. The previous map was also a Democratic gerrymander, but because Republicans picked up two seats in 2014 that should make it possible to ensure that the lost seat is a Republican one.
Competitive States
Arizona: 9 seats; current delegation: 5 R, 4 D; projected +1 seat
Colorado: 7 seats; current delegation: 4 R, 3 D; projected +1 seat
Minnesota: 8 seats; current delegation: 5 D, 3 R; projected -1 seat
New York: 27 seats; 1current delegation: 18 D, 9 R; projected -1 seat
All of these states currently have split control of the redistricting process or, in Arizona’s case, an independent redistricting commission. Assuming this doesn’t change, this will result in the maps being drawn either by bipartisan compromise or by the courts. That would likely result in newly competitive seats in Arizona and Colorado, though Republicans would probably be favored in the Arizona seat and Democrats in the Colorado seat. Meanwhile, Republican and Democratic incumbents could face off in Minnesota and New York under a bipartisan or court-drawn map.
Setting Arizona aside, the other three states are all one piece away from Democratic control of the redistricting process (Democrats would need to retake the state Senates in Colorado and New York, and the state House in Minnesota). A Democratic trifecta would likely ensure the new Colorado seat was Democratic and the losses in the other states were both Republican incumbents (leaving aside the possibility that map-makers would go after other Republican-held seats).
“Blood from a Stone” States
Michigan: 14 seats; current delegation: 9 R, 5 D; projected: -1 seat
Ohio: 16 seats; current delegation: 12 R, 4 D; projected: -1 seat
Pennsylvania: 18 seats; current delegation: 13 R, 5 D; projected: -1 seat
North Carolina: 13 seats; current delegation: 10 R, 3 D; projected: +1 seat
These are possibly the most interesting states to look at for 2020 redistricting, because Republicans controlled redistricting in all four of these blue or swing states following the 2010 elections. So far, the GOP retains that control in all the states but Pennsylvania, where it’s split. The question then becomes: Are there any more Democratic seats left for Republicans to eliminate? Setting aside North Carolina, Obama won the other three states twice, yet Democrats hold less than one-third of their collective House seats. There are lots of Democratic voters in these states, and they have to go somewhere.
If Republicans are smart, they will likely heed the cold-blooded “walk the plank” strategy used by the Ohio GOP in 2010. Back then, Republicans created a new Democratic seat in Columbus, recognizing they could no longer try to disperse all the Democratic voters in the center of the state without putting multiple Republican congressmen at risk. This forced one incumbent Republican, Steve Austria, out of a seat, but locked in a map with 12 solidly Republican seats and just four safely Democratic seats.
When the party in power gets too greedy in drawing the lines, it can jeopardize a whole bunch of seats at once when the minority party has an unusually good election night, a phenomenon dubbed a “dummymander.” This is exactly what happened in Pennsylvania in 2006 and 2008, when Democrats ultimately won 12 of 19 congressional districts even though Republicans had created the map—one in which they spread their voters too thinly. Michigan Republicans might be able to nuke another Democratic seat, but it’s difficult to imagine their Ohio counterparts eliminating a Democratic district, or North Carolina Republicans adding a GOP seat without making other seats more vulnerable to a Democratic take-over.
Court Confusion
Florida: 27 seats; current delegation: 17 R, 10 D; projected +1 seat
Virginia: 11 seats; current delegation: 8 R, 3 D; projected +1 seat
It’s hard to know how 2020 redistricting will sort out in these states since court battles over the 2010 round of redistricting are still underway, halfway through the current decade. The Florida Supreme Court has been trying to enforce the state constitutional amendments passed in 2010 that require fairer redistricting, but still leave it up to the legislature to actually create and pass the maps. Assuming Republicans retain control of the state in 2020, a similar court battle will likely be fought over the next set of maps.
Meanwhile, a panel of federal judges declared Virginia’s map unconstitutional for packing too many African-American voters into the 3rd District. When that’s finally remedied, it should ultimately give Democrats a good shot at winning a fourth seat in 2016, though it’s currently unclear which one. But however these maps end up, both states are seeing influxes of Democratic-leaning voters. As a result, any seats that either Florida or Virginia earn will likely be competitive, such that Democrats should ultimately have a good shot at both of them.
Texas
Texas: 36 seats; current delegation: 25 R, 11 D; projected +2 seats
As Texans prefer, their state gets its very own category. It’s the only state currently projected to gain two seats, and Texas Republicans would obviously create two more GOP seats if allowed. However, given the state’s growing Latino population, there’s a good chance one of the two new seats would be required to have a Latino majority under the Voting Rights Act, which would likely lead to one new Republican seat and one new Democratic seat.
Of course, all of this—or at least, a lot of this—is subject to change. But it’s nevertheless worth investigating even at this early date, because projections from around this time 10 years ago turned out to be pretty accurate when the results of the 2010 Census came in. That may not happen again, but if it does, we’ll be all the better prepared.