In this post:
Federal appeals court rules in favor of Ohio getting Census data early
Winners and losers in the current apportionment
Who controls redistricting in those states
What are the likely seats gained/lost in these states because of apportionment
This morning the 6th Circuit Court Appeals in Ohio reversed a lower court decision that the Buckeye State did not have standing to sue the Census Bureau for delayed data it needs for redistricting. It’s not as important a decision as it might seem since the Bureau admitted that it could provide the data in a legacy format as early as August 16th and the state agreed that was early enough for its purposes. The admission by the Bureau eliminates one of the reasons the district court ruled there was no standing: redressability. The lower court believed it was impossible for the Bureau to deliver data sooner than September 30th and ruled that any remedy the court issued could not be enforced. The appeals court ignored the district court’s argument on another prong of the standing doctrine: injury-in-fact. The lower court ruled that since state law in Ohio directs the states to use other data for redistricting if the Census data is not available, there was no actual injury in the delay. The appeals court merely noted that there was an injury-in-fact.
Ohio is one of the six states that is losing a seat in the House of Representatives based on the apportionment data from the Census Bureau. While all states that have more than one seat need more detailed census data to draw their district maps, the ones that gain or lose seats cannot afford the delay. States whose apportionment remains unchanged do have the luxury of redistricting for the 2024 cycle (subject to state law) if need be. Considering the delays in completing the 2020 decennial census, courts would likely not entertain equal protection challenges to districts as they are currently drawn so long as new districts are implemented in the next election cycle. However, it does not appear that any state is entertaining that option as for most they will be able to use the census data when it is available in August or September.
This year the number of states either gaining and losing seats through apportionment is the lowest it’s been since 1920. Population growth in the United States between 2010 and 2020 was the slowest in about a hundred years. Nevertheless, only three states lost population during this ten-year period (West Virginia, Mississippi, and Illinois). This is the largest number of population-losing states since the 1980s. Of these three states, only Mississippi did not lose a seat in this year’s apportionment.
As discussed in the previous post in this series, states can lose seats without losing population because the number of representatives in the House is fixed and apportionment is based on relative share of the national population. In fact, one state that lost population did not lose a seat (MS) while five states that saw population gains lost seats (CA, MI, OH, PA, NY). Overall, seven states each lost one seat while five states gained one seat (FL, CO, OR, NC, MT) and one state gained two (TX).
There are seven states with changing apportionments in which redistricting is entirely controlled by the legislature. Four of those state legislatures are controlled by the GOP. These states will collectively pick up four districts and lose one. Just one of the states (Illinois) has a legislature controlled by the Democrats, and it is losing a seat. The other two states have split control.
Four of the other six states use independent commissions to redistrict. New York and Ohio have odd hybrid systems where commissions play a role, but the legislature can have the final say (see the notes in Table 1 above).
One might assume that Republican control means redistricting will favor Republicans, Democratic control means it will favor Democrats, and commission control means seats will be competitive. In general there is some truth to this idea, but there are constraints. For instance, West Virginia - which has a state government completely controlled by the GOP - represents a total Republican loss. The Mountain State is losing a seat and no matter what they do it will be a Republican loss. The same would be true for Democrats in Massachusetts if that state were losing a seat (it’s not).
States have to roughly equalize the population in each of their districts and must not run afoul of the requirements of the Voting Rights Act (such as it is after Supreme Court gutting). While Texas is gaining two seats, it already has some heavily gerrymandered districts with Republican incumbents such as Dan Crenshaw (see his district map below).
States are not just confined by equal protection rules in drawing districts, but geography. The population of a state may increase, but not it’s size. So, Texas has to squeeze two more districts into its borders in a way that will incorporate voters currently in other districts. If the Lone Star State wants to draw two safe Republican districts it may find that to do so will require making other gerrymandered districts - like Crenshaw’s - more competitive. The same dynamics exist for the other states that need to add or lose a district, although the gerrymandering aspect may be less important in, say, Oregon than it is in Texas.
Considering these factors, redistricting alone is likely to create four GOP losses, one Democratic loss, and one Republican gain, with the remainder being either competitive or unclear (Texas, for example, is unclear). However, there already ongoing attempts to draw more partisan districts in states with no apportionment change. New Hampshire, for instance, will maintain its two (Democratic) House seats but the Republican-controlled legislature is considering to pack Republican towns into one district to elect a conservative Republican. Until November, the state legislature was controlled by Democrats - providing us with a tangible example of how important down-ballot races are.
We still do not know whether New York will formally contesting the enumeration that cost it a seat. States have done so occasionally over the years, but not one has ever been successful. I expect New York’s will also be unsuccessful, but there may be reason to believe otherwise if there is information that is not currently publicly available. Despite that potential wrinkle, apportionment is set and the next step in the process is redistricting. Block group level data (that is, small geography population and demographic data) from the Census Bureau will be released at the end of the summer - perhaps by mid-August per today’s 6th Circuit decision. Once that data is available, states will be able to draw their districts. In the meantime, it will be helpful to take a look at what results we might expect from redistricting. Several analysts have mocked up potential maps for a number of states. Future posts in this series will take a look at their findings and projections.