What happens if New York’s Independent Redistricting Commission deadlocks again?
Or, what happens when immovable constitutional ambiguity meets unstoppable partisanship?
On December 12, the New York Court of Appeals resolved one of the biggest outstanding redistricting questions for the 2024 elections: will New York’s congressional districts be redrawn? In a 4–3 decision, the court said yes, guaranteeing that new districts would be redrawn and used for the 2024 election.
In 2014, voters approved the creation of a bipartisan redistricting board known as the Independent Redistricting Commission. And, following the 2020 census, the IRC was operationalized for the very first time . . . and it crashed and burned.
Though the state constitution obligated it to attempt to produce a set of maps for legislative approval two separate times, it only did so one time—and refused to do so a second time. In response, the legislature took it upon itself to draw a new set of maps, which were struck down by New York's highest court, the Court of Appeals, both for violating the state constitution’s procedural requirements for redistricting and for being an unconstitutional gerrymander. The court passed the case back to the trial court that had originally struck down the maps to draft a new set of maps.
Immediately following this litigation, a group of voters filed a new lawsuit. This suit aimed to get the IRC back to the drawing board to produce a new set of maps. These voters argued that the court-imposed maps could only apply for the 2022 elections, and that the IRC was obligated to attempt to draw new maps that would be used for the remainder of the decade. On December 12, the Court of Appeals agreed—and ordered the IRC back into action.
But the two provisions of the New York Constitution at issue are poorly drafted. Two separate sections of the constitution—sections 4(a) and 5-b(g) of Article III—ping pong between each other in establishing an order of operations. They use inconsistent language and leave far too much up for interpretation. And most importantly, they fail to clearly answer one basic question: What happens if the IRC fails to produce maps this time?
To understand why that question is difficult to answer, we need to walk through the theoretical process by which the Independent Redistricting Commission operates.
Under section 5-b(a) of Article III, the IRC consists of 10 members. Eight are partisan members, with four each appointed by the major parties’ state legislative leaders in each chamber. Once the eight partisan members have been selected, together, they pick two politically unaffiliated members.1
The IRC’s job is to produce a “redistricting plan,” which consists of a map and “implementing legislation” for that map, which is basically the legislation that actually turns those maps into state law. (There are lots of substantive requirements for the IRC to follow in adopting its maps that are relevant to the fairness of the maps but ultimately immaterial to the issue I’m describing, so I won’t detail them here.)
For the IRC to adopt any “redistricting plan,” it must do so with seven out of 10 votes. And, because there are four Democratic appointees, four Republican appointees, and two politically unaffiliated appointees, if the approval of a redistricting plan requires seven votes, it necessarily requires a bipartisan vote.
Let’s imagine that the partisan squabbles among the parties are set aside, the sea is parted, et cetera, and such a bipartisan vote takes place. The redistricting plan is submitted to the legislature. Huzzah!
Now we bounce over to section 4(b). The redistricting plan is submitted to the legislature, and the legislature votes on the implementing legislation “without amendment.” If approved by both houses—which requires just a “majority” of each house if control of the legislature is split between two different parties, but a two-thirds majority if one party controls both chambers—and signed by the governor, it passes into law. Hooray!
What if not, though? If the first redistricting plan is rejected—that is, if either chamber of the legislature rejects it, or the governor vetoes it—then it bounces back to the IRC. Subject to the same requirements as before, the IRC produces a second redistricting plan, which it submits to the legislature. Here, too, the legislature votes on the implementing legislation “without amendment.” If approved by the houses and signed by the governor, it passes into law. Woohoo!
But what if not here, either? At this point, the IRC’s job is done. The implementing legislation proposed by the IRC is in the hands of the legislature, which has the power to amend the legislation “with any amendments each house of the legislature deems necessary.” In theory, the legislature’s power here is constrained by a 2012 law, codified at section 93 of the Legislative Law, that prevents the legislature from “affect[ing] more than two percent of the population of any district contained in such plan.”2
That’s all complicated—but, in theory, it prevents the legislature from repeatedly rejecting the IRC’s recommendations and then adopting its own redistricting plan on a party-line vote. (Unless, of course, one party has a two-thirds majority in the legislature, which the Democratic Party currently does.)
So what happens if, instead of ushering in this bipartisan utopia, the IRC fails to pass a redistricting plan with a seven-member, bipartisan vote? This is where the constitution gets a little sticky.
We’re returning to section 5-b. If the IRC fails to approve a redistricting plan, subsection (g) provides that it “submit[s] to the legislature that redistricting plan and implementing legislation that garnered the highest number of votes in support of its approval[.]”
But if “more than one plan received the same number of votes for approval . . . then the commission shall submit all plans that obtained such number of votes.” (This is basically what happened in the 2022 redistricting process.) At that point, “[t]he legislature shall consider and vote upon such implementing legislation in accordance with the voting rules set forth in subdivision (b) of section 4.”
We now return to section 4(b). But the language used here doesn’t translate very well.
The implementing legislation shall be voted upon, without amendment, by the senate or the assembly and if approved by the first house voting upon it, such legislation shall be delivered to the other house immediately to be voted upon without amendment. If approved by both houses, such legislation shall be presented to the governor for action. If either house shall fail to approve the legislation implementing the first redistricting plan, or the governor shall veto such legislation and the legislature shall fail to override such veto, each house or the governor if he or she vetoes it, shall notify the commission that such legislation has been disapproved.
How does this process work if there’s more than one plan? The references here use “the”—a definite article that contemplates one piece of “implementing legislation.” Admittedly, “legislation” is a word that could be singular or plural, but viewing it as a plural noun doesn’t really make sense here. Only one redistricting plan could possibly be approved.
But at the same time, if two (or more!) redistricting plans, containing the same number of pieces of enabling legislation, are submitted to the legislature, then the legislature could only pass one piece of enabling legislation. Section 4(b) requires that the legislature vote on the implementing legislation—but does the text require that the legislature vote on all pieces of implementing legislation if the IRC has deadlocked?
I struggle to imagine how that would jibe with what section 4(b) actually says. The text states that if either house fails to pass the legislation, or the governor vetoes it, the process automatically bounces back to the IRC. But if more than one piece of legislation has been submitted, and the legislature has to pick one to advance, then at least one other piece of legislation would always fail, no? Section 4(b) obviously contemplates the submission of just one piece of legislation—it’s almost impossible to read it any other way.
I’m not sure that the only available precedent—namely, what happened in the 2022 redistricting process—necessarily answers the question. In 2022, the legislature rejected both maps.
The problem gets more acute in the exact situation we’re in right now, wherein the IRC gets a second bite at the apple. What if the IRC fails to produce a single redistricting plan now, after the legislature rejected the first one? Here, the textual ambiguities become even more apparent.
Let’s turn again to section 5-b(g). It doesn’t actually contemplate what happens here. The entirety of subsection (g) discusses the failure to produce the first redistricting plan—not what happens afterward. I assume that the text of section 4(b) controls here, because subsection (g)’s last sentence says: “The legislature shall consider and vote upon such implementing legislation in accordance with the voting rules set forth in subdivision (b) of section four of this article” (emphasis added).
I assume that the italicized text basically means to incorporate section 4(b)’s second-redistricting-plan process to this situation. Maybe that’s a stretch, given that subsection (g) doesn’t actually say that. If that’s not what it means, however, then I don’t think that the New York Constitution acknowledges at all what would happen in this situation.
So let’s return to section 4(b), which is now starting to creak under the pressure. Suppose, again, that the IRC again submits the two most popular plans. (Again, it could submit more than two, but two is easier to game out.) The same problems described above apply here. Section 4(b)’s discussion of the second redistricting plan uses definite articles to refer to “the legislation implementing the second redistricting plan,” and the legislative process clearly contemplates just one piece of legislation. But let’s assume there’s no problem with any of that.
What happens if the legislature rejects this plan isn’t terribly clear, either. “[E]ach house shall introduce such implementing legislation with any amendments each house of the legislature deems necessary” (emphasis mine). Again, “such implementing legislation” necessarily refers to a single piece of legislation.
But now the legislature has the power to “amend[]” the legislation. Which piece of implementing legislation can it amend? Either? The 2012 law actually does a better job than the constitution in acknowledging what happens here: “If two or more plans for districts in the same legislative house or for congressional districts are submitted by the commission and voted upon by the legislature, such plans shall be considered individually and not combined.”
If the legislature keeps the 2012 law intact, I think it could clarify what happens in this situation. That is, the 2012 law specifies that the legislature has the power in this situation to take either plan and amend it, but within that 2% threshold. If the legislature amends the law or ditches it altogether—which is possible, given that the law places quite significant limits on the legislature’s power—or if the law is itself unconstitutional, then I’m less confident.
The 2012 law also makes the same assumption that I did: that section 5-b(g)’s reference to section 4(b)’s voting procedures means to incorporate all of the orders of operation spelled out in section 4(b). That the drafters of the 2012 law—who intended the law to accompany the constitutional amendment—made this assumption makes me feel more comfortable about my own assumption in that respect, and perhaps says something about the intent behind section 5-b(g).
All told, I think that the New York Constitution is clear enough—but just barely—on what happens if the IRC fails to agree on a single redistricting plan for a second time. I feel confident enough that the legislature would be able to pick and choose which of the plans it votes on, and how the backup process would play out, even though this confidence is based on a necessarily loose reading of the applicable provisions.
That said, the text is obviously written poorly. Bouncing back and forth between two different sections of the constitution to spell out a single process is less than ideal, as is reading what would otherwise be deliberate drafting choices—like the use of a definite article—out of existence.
But, hey! This is the process that two separate sessions of the New York legislature, as well as 58% of voters, chose.
I use the word “politically unaffiliated” as shorthand for the literal requirement, which is that the two final members “shall not have been enrolled in the preceding five years in either of the two [major] political parties.” N.Y. Const. art. III, sec. 5-b(a)(5).
Personally, I think that the 2012 law—which was intended to supplement the 2014 amendment—is unconstitutional under section 4(b) of Article III. Section 4(b) gives the legislature carte blanche to amend the enabling legislation “with any amendments each house of the legislature deems necessary” (emphasis added), and I don’t believe that the legislature can statutorily limit the authority granted to it by the constitution in that way.