The New York Times reported
The Supreme Court agreed on Tuesday to hear a case that will answer a long-contested question about a bedrock principle of the American political system: the meaning of ‘one person one vote.'
The court has never resolved whether voting districts should have the same number of people, or the same number of eligible voters. Counting all people amplifies the voting power of places with large numbers of residents who cannot vote legally, including immigrants who are here legally but are not citizens, illegal immigrants, children and prisoners. Those places tend to be urban and to vote Democratic.
In the process, though, several judges have acknowledged that the Supreme Court’s decisions provide support for both approaches. The federal appeals court in New Orleans said the issue “presents a close question,” partly because the Supreme Court had been “somewhat evasive in regard to which population must be equalized.”
Even if counting only adult citizens is the correct approach, there are practical obstacles. “A constitutional rule requiring equal numbers of citizens would necessitate a different kind of census than the one currently conducted,” Nathaniel Persily, a law professor at Stanford, wrote in 2011 in the Cardozo Law Review.
FiveThirtyEight looked at the Census’s American Community Survey to try to determine the effect of changing to counting voters.
Most legal scholars are skeptical that this case will change the way House seats and Electoral College votes are apportioned to states every 10 years; after all, the Constitution pretty clearly references the role of the Census’s population count in this process.
But let’s lay out a hypothetical for a minute. What would happen if counts of voters rather than people were used for both reapportionment (allotting seats among the states) and redistricting (drawing boundaries within states)? States with large Latino populations would be penalized. Based on the 2013 ACS data, California would lose six House seats, Texas would lose four and New York one, while a smattering of other states would each gain one seat.
Next, the partisan implications: Which party would gain or lose those seats? To arrive at an estimate, we calculated the percentage of eligible voters currently living in districts represented by Democrats and Republicans in each state and applied that percentage to the “new” seat total. For example, Republicans currently represent 25 of Texas’s 36 districts (69 percent), but an even higher share (73 percent) of Texas’s eligible voters. If Texas lost four seats and only adult citizens were counted, the GOP seat edge might expand to 24-8.
If we repeat this calculation across all states, a redistribution of representation based on eligible voters could result in a net gain of roughly eight seats for House Republicans, who are already sitting on a historically large majority of 245 seats.
What’s interesting, however, is that the possible shift would affect only the closest of presidential elections. If you were to replay the 2012 election under the new lines, President Obama would win 331 electoral votes, only one fewer than he actually did.