A couple of days ago, I had a brief altercation with my father over the recently-proposed redistricting plan in Texas. Among other things, he claimed that the Republican-backed proposal was a “naked power grab” that clearly violated “the will of the people”. I’ve been thinking about those claims, and finally decided to put them to a rough empirical test.
My neutral hypothesis was that a fair districting division would result in each party holding a number of seats proportional to the overall percentage of the vote they received. If one party winds up holding substantially more seats than that, the districting is unfairly skewed to benefit them over their rivals.
The Texas Secretary of State has 2002 election result totals available on the web, so we can test this hypothesis. Slightly collapsed and added together, the 2002 results looked like this:
| Office | Republican | Democrat | Other |
| District 1 | 66,654 | 86,384 | 0 |
| District 2 | 53,656 | 85,492 | 1,353 |
| District 3 | 113,974 | 37,503 | 2,656 |
| District 4 | 67,939 | 97,304 | 3,042 |
| District 5 | 81,439 | 56,330 | 2,139 |
| District 6 | 115,396 | 45,404 | 3,237 |
| District 7 | 96,795 | 11,674 | 58 |
| District 8 | 140,575 | 0 | 10,351 |
| District 9 | 59,635 | 86,710 | 1,613 |
| District 10 | 0 | 114,428 | 21,196 |
| District 11 | 68,236 | 74,678 | 1,943 |
| District 12 | 121,208 | 0 | 10,723 |
| District 13 | 119,401 | 31,218 | 0 |
| District 14 | 102,905 | 48,224 | 0 |
| District 15 | 0 | 66,311 | 0 |
| District 16 | 0 | 72,383 | 0 |
| District 17 | 77,622 | 84,136 | 2,046 |
| District 18 | 27,980 | 99,161 | 1,785 |
| District 19 | 117,092 | 0 | 10,684 |
| District 20 | 0 | 68,685 | 0 |
| District 21 | 161,836 | 56,206 | 4,051 |
| District 22 | 100,499 | 55,716 | 2,869 |
| District 23 | 77,573 | 71,067 | 1,912 |
| District 24 | 38,332 | 73,002 | 1,560 |
| District 25 | 50,041 | 63,590 | 2,495 |
| District 26 | 123,195 | 37,485 | 3,998 |
| District 27 | 41,004 | 68,559 | 2,646 |
| District 28 | 26,973 | 71,393 | 2,054 |
| District 29 | 0 | 55,760 | 2,833 |
| District 30 | 28,981 | 88,980 | 1,856 |
| District 31 | 111,556 | 44,183 | 5,745 |
| District 32 | 100,226 | 44,886 | 2,790 |
| Total Votes | 2,290,723 | 1,896,852 | 107,635 |
Net result — Republicans 53%, Democrats 44%, Other (mainly Libertarian and Green) 3%. Since there are currently 32 Congressional districts in Texas, the neutral hypothesis leads to the conclusion that a fair districting would lead to a seat distribution of 17 Republican, 14 Democrat and 1 Other.
If we drop the 3rd party vote totals and just consider the Republican vs. Democrat count, the results shift to 55% Republican and 45% Democrat, which means 18 seats for the Republicans and 14 for the Democrats.
The actual seat distribution that resulted from the 2002 election was 17 seats for the Democrats, and 15 for the Republicans. In other words, the current districting division is clearly skewed to favor the Democrats by 2 to 3 seats. The “will of the people” is clearly being frustrated by the current setup.
Now, this doesn’t necessarily prove that the Republican proposal is fair. Tom Delay’s comment that he wants the Republicans to have 20 seats implies, on the basis of the 2002 numbers, that the Republican plan is skewed to favor the Republicans by about the same extent that the current setup favors the Democrats. Still, in terms of “naked power grabs” and violation of the “will of the people”, the Republican plan seems no worse than the one it would have replaced. And that means the Texas Democratic representatives broke the law (yes, it is illegal in Texas for a representative to deliberately fail to appear in the statehouse for purposes of breaking a quorum) and shut down the legislature to avoid living under a proposed law that is simply the mirror image of the one they thought was just peachy when they were on the winning side.
And that, Dad, is why I think the Texas Democrats were throwing a snit and acting like crybabies because they couldn’t handle being on the losing side.
UPDATE: I did a little bit more digging. In the 1990 Texas redistricting, the Democrats controlled the legislature. The plan they came up with allowed them to capture 70% of the congressional seats in the state with only 50% of the vote — based on the current number of seats that would have been a 22-10 split favoring the Democrats when a fair division would have been 16-16.
I wonder if you found that as objectionable as the current Republican plan. My guess is no.
September 29th, 2003 at 6:01 pm - Edit
Hi,
I’m a tenth grade IB student. I was having a a bit of trouble researching my government project on Texas. I just wanted to say thank you for this information because this helped me a whole lot!
Thanks a bunch!
October 12th, 2003 at 2:50 pm - Edit
Craddick’s proposed district will hog most of the oil, (Permian Basin,) most of the water,(Lake Ivey, Lake Brownwood, Lake Proctor, Lake Buchanan, Lake LBJ, the Colorado river, the two Llano rivers, and the San Saba river,), most of the pecans, (San Saba County), most of the cattle,( San Saba auction barn, and Mason auction barns), and to top all of the above off, it will include Luckenback and Lowake!
November 6th, 2003 at 5:54 pm - Edit
Thanks for this info!!!
I knew something like this had occured but didn’t have the specific facts…now I do…thanks for this tidbit….
December 30th, 2003 at 10:38 am - Edit
The Democrats have always redrawn the map since the 60’s. I’m in my 50’s and I go back only that far. This is the first time Democrats didn’t get to draw the map.
Things always change. I can’t beat my brother at pool anymore either, but I wouldn’t run off to another state to avoid a match.
Of course this map is more serious than a pool match. But, my guess is that you cannot assume this map will create a huge gap in reps. If President Bush has success then the gap will be there in 2003. If there isn’t a Bush success then the gap will be much much less.
August 19th, 2004 at 11:40 pm - Edit
The skewing is accomplished by concentrating the opponent into a few districts where they can have a high percentage of the vote, but then be losers in the majority of the districts.
8 of the 32 seats were unopposed, 5 Republican, 3 Democrat. In the other 24 districts, Republicans got 55% of the vote, but only 10 of 24 seats (42%). 9 of the 12 highest margin of victories in challenged races were Republican majorities.
Any vote over the majority is a surplus vote, because the seat is already won. In all 32 seats, 50% of the Republican vote were surplus votes, while 40% of the Democrat votes were surplus. In the 24 challenged races, 40% of the Republican vote was surplus, compared to only 25% of the Democrat votes.
Unwittingly, the Democrats are concentrating themselves nationally, to their detriment in the Presidential election. With som many Democrat votes concentrated in the DC-Boston corridor, Detroit, Chicago, and the Pacific coast, they are helping create a situation where they can win the national popular vote, but lose the Electoral College vote. One of the reaons behind the concept of the Electoral College is to force Presidential candidates to be broad-based and not regional (you need to carry more than just New York and California to win).