Gerrymandering is forever

Welcome to the permanent 220-215 House

For all the recent hue and cry, mid-decade redistricting is not a new thing. Texas did the same thing in 2003, and there was a similar walkout by Democrats then, with threats of arrest. Earlier in the current redistricting cycle, North Carolina drew its maps to be more favorable to Republicans — and New York responded with a modest tune-up of its maps to help Democrats.

What is different about this time is that both sides are shedding any pretense that this issue is about better representation for the states in question. It’s explicitly about engineering a specific outcome in the midterm elections nationally and about partisan payback for the states on the other side that did it first.

And what is specifically different about Gavin Newsom’s California ballot measure is that it’s the first effort to dismantle the whole architecture of nonpartisan redistricting that was at the core of the Democrats’ strategy for so long. After the 2010-12 redistricting cycle, the House map was biased towards Republicans, so the main thrust of Democratic efforts was to replace Republican gerrymanders with “fair” maps. And it worked. Nonpartisan or court-ordered redraws in some states (Pennsylvania) combined with Republican restraint in others (Texas, initially) and Democrats flexing their gerrymandering muscles (Illinois) worked to give the 2020s map a slight Democratic skew.

The “assault on democracy” in Texas that California is countering would actually make the House more fair by the only metric that seems to matter to anyone: the national composition of the House. But this is yet another useful reminder of one of Michael Barone’s rules for life: “All process arguments are insincere, including this one.”

Having largely succeeded in the post-2020 redistricting cycle, Democrats had no more ways to make red state processes more neutral. So the only option left was to make blue state processes more ruthless. Now that both parties have completely embraced this ethos, it will be very hard to put the genie back in the bottle. Partisan legislatures will try to get as close as they can to the same kind of unit rule voting that exists in the Electoral College in their state’s House delegations. Only in the few remaining battleground states will nonpartisan redistricting be able to withstand intense partisan pressure to maximize the majority party’s advantage. With the Democratic Party’s decision to sell gerrymandering to their base as morally justified, the days of nonpartisan commissions in many states are numbered.

Everything about American politics over the last few decades has been about more and more vicious fights for less and less real estate. There are now seven swing states where there used to be almost twenty. In the last three cycles, the House was decided by a margin of fewer than 9 seats. Presidential margins have been historically close for a long time.

This is not occurring by happenstance. Both parties are just fine with this. The current redistricting battle is a product of their being fine with razor-thin wins, since it means a higher number of safe seats.

Why parties didn’t gerrymander more before this

To me, the interesting question in the current redistricting war is why it took a mid-decade process for red states like Texas and Florida to try and draw maximally aggressive maps. Blue states were more likely to embrace nonpartisan redistricting, but the red states could have easily pushed as aggressive a map as they are now in the 2021-22 redistricting cycle.

This was also true of some Democratic states. Maryland left in place a Republican district on the Eastern Shore, and New York Democrats initially left enough of their seats vulnerable to a slew of Republican pickups in 2022. But when they went back to fix this in 2024, they backed away from a total wipeout of the state’s Republican delegation, fearful of a bad court decision.

Various legal and VRA challenges loom as reasons for parties not to go all-out, though those seem to be less of a concern right now, and the VRA might be on its last legs.

But this overlooks a bigger reason for restraint, at least in Texas: future political uncertainty. At the turn of the decade, the Texas suburbs seemed to be shifting left fast, and so exurban Republicans were drawn into Trump +25 districts to withstand what was thought to be a rapid blue shift similar to that which happened in the old TX-7 or TX-32. That led to maps that were more about protecting current incumbents than drawing new Republican seats. Whether or not you think the new Texas map is good, Donald Trump was correct in his observation that the state’s Republicans left seats on the table.

Multiple developments have changed the calculus for Texas and other states looking to redraw their maps. In Texas specifically, further suburban blue shift hasn’t materialized, leading to wasted votes in suburban and exurban seats. And nationally, House results have become much more correlated to the presidential result, with only a handful of members sitting in districts won by the opposite party at the presidential level. That makes the district’s past presidential performance an even better guide to likely future House performance. Even in a wave, a Trump +10 district is safer than it would have been in 2018, and even in 2018, few such districts fell.

Partisan polarization means that parties have a much bigger margin of safety when drawing districts. Where Trump +25 seats were assumed safe over the course of a decade, the gerrymanders now create a bunch of seats just on the other side of Trump +10 or Harris +10. Over a decade — scratch that, since any new maps only need to last four years — these seats seem reasonably likely to stay red or blue.

An additional wrinkle in this is that the more gerrymandered the seat — meaning, the more oddly it’s drawn to bring in different rural and urban geographies — the more likely that demographic trends in different parts of these districts will cancel each other out and you won’t see crazy swings in district partisanship. Like a balanced stock portfolio, gerrymandered seats help parties hedge against risk in specific demographics, like Republicans in the suburbs or Democrats with Hispanics.

Big House majorities don’t matter anymore, so it’s best to lock in as many safe seats as possible

All the redistricting wars ultimately ladder up to the national battle for House control. In the last three elections, we’ve seen three of the narrowest majorities historically, where no party could afford to lose more than four of its votes on the floor.

While this happened in 2020 by coincidence, starting in 2022 this was mostly a deliberate result of redistricting, which saw the elimination of lots of swing seats. While parties did leave seats on the table, the overall result was a much more static House map, with less potential for big swings. And with the maps potentially getting a refresh in many states, the House is likely to have even fewer truly competitive seats in the midterms.

Structurally, this means that parties have to be content with narrower House majorities, with state trifectas freezing in place a large number of safe seats for their side. The old heuristics of wave elections, of seat gains of 40 seats or more, are now an anachronism. In the battle for the House, 20 is the new 40 (and maybe only 15 is).

What the current redistricting push signals is that both parties are largely fine with capping the number of seats they could potentially win at 235 or less. Razor-thin House majorities are the new normal and both parties have learned to operate within these constraints.

Thin majorities are not without complications. Just look at 2023’s Republican Speaker drama or just this year, at Republicans needing to cut off an exodus of members into the administration or at a string of deaths in the Democratic caucus.

But Republicans largely emerged from the Speaker fight of 2023 in a stronger position. The Democratic caucus was extremely disciplined under Nancy Pelosi, and the Republicans notoriously fractious, but with Trump imposing party discipline, those concerns have ebbed.

With party loyalty the totalizing force it is today, you don’t need big House majorities. Regardless of whether you have 40 votes to spare or 2, the likely margin for any maximalist signature legislation the majority wants to pass is 1 — and the failure rate is close to nil. Larger and more ideologically diverse majorities can make a caucus harder to manage. Adding the incremental seat to your majority isn’t nothing, but it’s far less valuable in an age of lockstep partisanship.

With existing guardrails being dismantled in states like California, collectively, you’ll have a House whose partisan makeup will be locked in within very narrow bands, with swings of 5 to 10 seats deciding the majority. And there’s no realistic chance that the California measure, if it succeeds, will be temporary as proponents claim. Assume it’s 2032 and Texas has just passed another gerrymandered map. Does California then just stop? I have a Golden Gate Bridge to sell you if you believe that.

Will it backfire?

Things are only true until they aren’t, and the tricky part about making proclamations based on arbitrary district lines is that the political leanings of districts can and do change. Drawing a bunch of narrower seats may end up creating future competitive seats that it will be a headache to defend, forcing a return to more conservative redraws in the 2030s.

We certainly saw this with the Hispanic shift: all the seats in the Rio Grande Valley have gone from safely Democratic to competitive. Creating a bunch of “likely” seats out of previously safe seats could mean that in a cycle or two, a few of those become vulnerable. This seems like what could happen in California under new maps if Hispanics continue to trend Republican.

But this seems like a problem the parties can deal with later by simply redrawing the lines again. Under the new take-no-prisoners rules, parties can go back to the drawing board once more or even twice more in a cycle. Remember, New York Democrats went back for another crack after their map fell apart after just one cycle, and they might go back again.

With the Democrats having sold their base on the rightness of gerrymandering, any political impetus behind returning to the old commissions is dead and gone. If California succeeds, that will signal that other states can successfully throw out their commissions on a party line vote. And that becomes yet another useful tool in the box to further optimize party performance. The end result will be a continuation of political stasis with narrow House majorities and parties voting in lockstep as far as the eye can see.