In Part 1 of this article, we reviewed population changes since both the 2020 Census and the Dobbs decision. Although shifts in the distribution of population can be interesting, it directly relates to political power through reapportionment. With any national population estimate released by the Census Bureau, observers often turn ahead to 2030 to try to tease out the results of the next reapportionment cycle.
It bears repeating that population estimates can deviate from reality and between one another. Extrapolating from the 2020 National and State Estimates for population, the last official estimate before the official reapportionment, did not precisely yield the actual population totals recorded in the 2020 Census, which had errors of its own. The official reapportionment showed New York, Minnesota, and Rhode Island each holding an extra seat at the expense of estimated gains for Texas, Florida, and Arizona — a difference of three seat transfers out of ten expected by the annual estimate. Keep this in mind for any review of population data even as 2030 approaches.
If reapportionment was conducted based on the figures released for July of this year, Texas would gain two seats; Florida, Arizona, and Idaho would gain one each; New York, Illinois, and Minnesota would each lose one; California would lose two. Interestingly, this would be a transfer of five seats from states that lean to the left of the country to states that lean to the right of the country. This would also be a gain of four electoral votes for states which voted for Trump in 2020, or five when counting Arizona, which the former President carried in 2016. While control of redistricting in 2030 is yet to be determined, it isn’t difficult to imagine this “baked-in” population change gives Republican legislators more seats to play with in their potential gerrymanders.
Compared to the projection that could have been made based off of last year’s data, this represents a single seat “transferring” from California to Texas. This confirms both a slow continuation of the relative population loss from California (and New York) starting from the pandemic, and a continuing trend towards a more southern population distribution.
A cursory glance at the map above shows relative population and political power increasing in the South and Interior West, two regions that once were solidly Republican but have shown some signs of weakening partisanship, especially in the past decade, except in Florida. Conversely, states losing influence include two in the Midwest and two coastal states, all of which have robust Democratic leanings, except for Minnesota, which nearly flipped in 2016.
While a hypothetical reapportionment for recent data is interesting, the true impact population redistribution has on redistricting only materializes after the Census in 2030. Should Republicans be able to retain control in the South, especially in the growing titans of Texas and Florida, these seat assignments will only serve to further the existing gerrymanders there, lending legislatures flexibility to pack and crack constituencies as desired. Meanwhile, Democratic legislatures and ostensibly-independent redistricting commissions in states losing districts could be the site of another decennial power struggle to influence the retention of certain districts as some come to be dismantled. Conditional on the retention of those political institutions, Republicans generally stand to gain from these population changes.
Again, the above is simply one possible scenario of the many that may lie ahead. There are still over six years until the data that will decide reapportionment will be collected; and control of state legislatures and redistricting commissions could be even more critical to determining the makeup of the House of Representatives. There are a few ways Democrats could benefit from growth in red states.
Firstly, as new congressional districts are assigned, even in deep red states, each district geographically “shrinks” — capturing fewer voters than the previously-bloated districts, which eventually forces gerrymandered maps to create packs, which give the disfavored party seats, instead of cracks, which precludes the out-party from victory. This can be seen in this decade’s maps in Texas, where Democratic gains in the Austin metro “forced” the creation of a new, solidly-Democratic district in Austin despite unilateral Republican control of redistricting. This effect, compounded across the South where Democrats have been gaining as of late, would likely result in the creation of more Democratic districts than extrapolating from statewide partisanship suggests.
Population loss also can be unpredictable. Even as Democratic states lose population, it has generally been the rural, Republican areas within those states that shrink the most relative to the country. This means that a Democratic state such as Minnesota could draw a map that gave Republicans three seats to the Democrats’ four without intentionally tilting the scales, meaning the GOP would lose a seat despite the seat loss occurring in a Democratic state.
And like everything in politics, a good portion of the outcome of redistricting depends on the voters themselves. If the newcomers to Texas lean left, then the apparent growth in a red state could prove to benefit Democrats should the governorship flip or if enough suburban precincts see net growth for Democrats to force Republican districts further out to the exurbs, clearing land for new Democratic suburban seats. Similarly, if Republicans are depopulating places such as California, the remaining voters will lean more Democratic, aiding in the drawing of more Democratic districts.
But with every caveat, there is a second-order disclaimer. Republicans could hold control in red states, benefit from population loss in urban cores, and perhaps newcomers to the South are more politically mixed than Democrats would hope. There is a lot of uncertainty with conjecture — partisan observers ought to pin their hopes on more conventional indicators of political support such as election outcomes, which are bound to be more elucidating than population totals.
This conjecture also does not include the possibility of Census inaccuracy. As described earlier, three seats — about a third — of the projected seat changes did not materialize. With sixteen projected above, any systematic undercounting or overcounting could result in an appreciably different outcome. Regardless of the specific numbers, the most significant indicator of how the maps will turn out in the next decade will be control of the process in 2030. This is decided by both the outcome of legislative and gubernatorial elections, but also in the selection of independent redistricting committees. That will be the key to deciding the contours of the next decade’s battlefield as it has been, and always will be so long as we elect our representatives by district.
I make election maps! If you’re reading a Split Ticket article, then odds are you’ve seen one of them. I’m an engineering student at UCLA and electoral politics are a great way for me to exercise creativity away from schoolwork. I also run and love the outdoors!
You can contact me @politicsmaps on Twitter.