Though official census data won’t be published until the end of the year, Election Data Services has released an updated reapportionment projection, based on population estimates published by GIS software maker Esri. To see how things have changed over the years, I’m including EDS’s projections for 2007, 2008, and 2009. Note that in prior years, EDS used several different models for its forecasts, which is why you see two different possibilities for some states in certain years.
| State | 2010 | 2009 | 2008 | 2007 |
|---|---|---|---|---|
| Arizona | 1 | 1 / 2 | 2 | 2 |
| California | 0 | -1 / 0 | -1 / 0 | 0 / 1 |
| Florida | 2 | 1 | 1 / 2 | 1 / 2 |
| Georgia | 1 | 1 | 1 | 1 |
| Illinois | -1 | -1 | -1 | -1 |
| Iowa | -1 | -1 | -1 | -1 |
| Louisiana | -1 | -1 | -1 | -1 |
| Massachusetts | -1 | -1 | -1 | -1 |
| Michigan | -1 | -1 | -1 | -1 |
| Minnesota | 0 | -1 | -1 | -1 / 0 |
| Missouri | -1 | 0 | -1 | -1 |
| Nevada | 1 | 1 | 1 | 1 |
| New Jersey | -1 | -1 | -1 | -1 |
| New York | -2 | -1 | -1 | -2 |
| North Carolina | 0 | 0 | 0 / 1 | 0 / 1 |
| Ohio | -2 | -2 | -2 | -2 |
| Oregon | 0 | 0 | 0 / 1 | 1 |
| Pennsylvania | -1 | -1 | -1 | -1 |
| South Carolina | 1 | 1 | 1 | 0 / 1 |
| Texas | 4 | 3 / 4 | 4 | 4 |
| Utah | 1 | 1 | 1 | 1 |
| Washington | 1 | 1 | 0 | 0 |
The biggest loser here is New York, which, after two straight years of facing only a single-seat loss, is back two a two-seat drop – though only by the narrowest of margins: the Empire State would get the 436th seat in Congress, if it existed. The other clear loser is Missouri, which swaps places with Minnesota – MN hangs on to its final seat, while Missouri now fails to. And Arizona, which last year could have gained either one or two seats, now looks set to pick up just one. Meanwhile, California (0), Florida (2) and Texas (4) all wind up with their higher projections.
EDS also provides a list of states on the bubble, which you can see below:
| Seat # | State | Makes By/ Misses By |
|---|---|---|
| 431 | South Carolina | 42,248 |
| 432 | Florida | 84,802 |
| 433 | Minnesota | 15,643 |
| 434 | Washington | 12,923 |
| 435 | Texas | 38,005 |
| 436 | New York | 29,439 |
| 437 | California | 99,396 |
| 438 | Arizona | 30,157 |
| 439 | North Carolina | 51,588 |
| 440 | Illinois | 75,046 |
In addition to the states on the lists above, EDS also thinks that Nebraska and Rhode Island are also at risk of losing a seat when the final numbers come out in December.
UPDATE: In comments, Nico takes a look back at EDS’s projections for the prior decade and finds that, while good, they still missed a few things. So we are very likely to see some surprises.
Weren’t there a couple reapportionment scenarios floating around where New York had an all-Democratic delegation?
Massa messed that up, tho.
The Midwest was the best region as far as Census turnout goes, and the Chicago machine likely did its best to make sure the entire city was counted. It’s a long shot, but I wouldn’t be stunned if IL-19 survives by the skin of its teeth and replaces the 4th new Texas district, as TX had a very poor Census response rate. As something less of a long shot, I wouldn’t rule out a new district for North Carolina replacing TX-36 either.
If New York replaces Utah in the 436th spot on the list, then the current version of the DC House Voting Rights Act is even deader than it already appears to be. Expanding the House to 437 seats with 1 going to DC isn’t so palatable to Republicans if the extra seat is going to New York rather than Utah-though maybe if it’s Peter King’s seat being saved some would jump on board.
People were deriding me for saying Minnesota was only 50/50 when it came to losing the 8th district. I feel ever-so-slightly vindicated
Here is the EDS’ final projections for 2000 reapportionment. They’re good, but not perfect. Using 1999 estimates, they missed a handful of shake-ups, like Indiana and Michigan’s loss of a seat, while projecting that Montana would gain a seat. They also thought that Georgia and Florida would only gain one seat and that North Carolina would not gain any.
Using 2000 projections, they were a little better, swapping out Montana for that second seat to Georgia and noting that Indiana was number 435 and perilously close to losing a seat.
The point is that these projections are good, quite good even, but they’re not bulletproof. There will probably be a surprise or two once reapportionment numbers are rolled out. Don’t take anything as gospel just yet.