We filed cert. Friday in the latest stage of litigation challenging the state-created practice of awarding presidential electors in a “winner take all” (WTA) manner.
The litigation has been conducted in four states—two Red states and two Blue states. Every circuit has believed itself bound by a 1968 Supreme Court summary affirmance upholding the practice, Williams v. Virginia State Bd. of Elections, 288 F. Supp. 622 (E.D. Va. 1968), aff’d, 393 U.S. 320 (1969). In one case, Baten v. McMaster, 967 F.3d 345, 364 (4th Cir. 2020), Judge Wynn offered a compelling dissent, arguing that Equal Protection law has “developed significantly” since 1968, and that the practice of WTA plainly violates modern Equal Protection standards.
The democracy argument against WTA is clear enough: WTA divides the nation into swing states, and spectator states. Most states are spectator states. Presidential campaigns, however, care only about swing states. The vast majority of spending (literally, greater than 95%) is within swing states. If Kriner and Reeves (The Particularistic President) are correct, that fact leads presidents to bend spending and regulation to benefit swing states over non-swing states. But swing states don’t represent America: Their population is older and whiter. Their industry is not the future but the past. And they are not the small states so romanticized by people who think the current system has any relationship to the system created by our Framers: Michigan is not a small state. Pennsylvania is not a small state. Florida is not a small state. There is no good reason that America should outsource the selection of our president to these few states in particular. There’s no good reason that America should outsource the selection of our president to any set of states in general. Presidential candidates should campaign to appeal to Americans in every state in America, not just the few. WTA is the reason they don’t.
The legal argument is also fairly clear (but having been fighting this issue for 5 years now, I would think that): The case has framed the challenge based both on White v. Regester, 412 U.S. 755 (1973) and Gray v. Sanders, 372 U.S. 368 (1963). The White attack is clear enough. I find the Gray challenge conceptually more interesting.
In Gray, the Supreme Court struck down a system that was structurally identical to the Electoral College, but at the state level. As we described it in our petition, in Georgia, the
Democratic Party provided each county a set number of units corresponding to the number of representatives it had in Georgia’s lower House of Representatives. Each county then conducted its own election for statewide officeholders, such as governor, and awarded all of its units to the winner in each county. This Court held that this system violated the Constitution first, because units were not allocated in proportion to population, and so unconstitutionally favored rural voters. But second, and significant here, the Court also held that even if “unit votes were allocated strictly in proportion to population,” the unconstitutional “weighting of votes would continue” because of WTA. That is, WTA requires “the candidate winning the popular vote in the county to have the entire unit vote of that county,” ensuring “votes for a different candidate [would be] worth nothing and . . . counted only for the purpose of being discarded.” The Court thus held that Georgia’s use of WTA in a context materially identical to WTA’s use in presidential elections constituted an independent constitutional violation.
Conceptually speaking, this is a challenge to a filter that gets applied in a multi-stage election process. Obviously, in a single-stage election process, there’s no problem ignoring the votes of the minority: the minority lost. But in the class of cases of which this case is an example, there’s a multi-stage election process, and the challenge is to a rule that affects an intermediate stage. So, of course, you can effectively throw away the votes of the losing candidate at the final stage of an election — that’s what winning means. But can you throw them away along the way to determining the ultimate winner of an election? That’s what we believe Gray addressed. And we believe Gray clearly says you cannot.
Having lived with this litigation for many years now, I’ve come to know the good arguments against our position, and the arguments that are missing the point. Here are a few in the latter category; I’m eager, later, to focus on the former. But let’s start with the most important conceptual error:
How would abolishing WTA have affected previous elections?
There’s no way to know this. If there were no WTA, presidential campaigns would have been run radically differently. You can’t simply apply proportional allocation to previous results and know anything about who would have won in any past election. It’s comforting at least that it is not clear, because we should be selecting methods for selecting the president based on principle not on party.
Isn’t the Electoral College in the Constitution? How can part of the Constitution be declared unconstitutional?
Yes, the Electoral College is part of the constitution. But no, winner-take-all is not in the Constitution. We’re challenging WTA; we’re not challenging the Electoral College.
If WTA has been with us forever, how can it only now be seen to be unconstitutional?
Of course, the Supreme Court has not hesitated to discover new Equal Protection interests in the context of presidential elections — see, e.g., Bush v. Gore (2000). But we believe this question was unnecessarily stymied by the summary affirmance in a case that didn’t even raise the Gray v. Sanders argument. In the late 1960s, there was a clear recognition that there were problems with the states’ use of WTA. Delaware had brought an original action in the Supreme Court to attack it, Delaware v. New York, 385 U.S. 895 (1966); the Court ducked it. The summary affirmance in Williams v. Virginia State Bd. of Elections then effectively stifled it. But obviously, lots has happened since 1968 in Equal Protection doctrine — including White and Bush. Those standards should, we believe, now be applied to the Electoral College.
Doesn’t Rucho mean you will certainly lose?
No. In Rucho, the Court refused to police partisan gerrymandering. I’ve been criticized for this, but I believe that decision makes lots of sense. I agree fully with Justice Kagan’s reading of the constitutional principles. But I find compelling the Chief Justice’s anxiety about the institutional difficulty in adjudicating actual cases. (This is a classic conflict of fidelity v. constraint: See Lessig, Fidelity & Constraint (2019)). As I summarized the argument in the Washington Post:
Roberts made this point explicitly in an earlier gerrymandering case: If the court decides a case favoring one political party over the other, the “intelligent man on the street,” as the chief justice put it, will ask why. Lawyers may cite sophisticated social science, but the man on the street, according to Roberts, would say, “that’s a bunch of baloney.” Instead, the “intelligent man” will believe that it “must be” because the court preferred one party over the other. And in the run of such cases, “the status and integrity of the decisions of this court,” Roberts insisted, would be seriously harmed.
But that concern is not present in the challenge to WTA. Because there’s no complexity to the remedy here: allocate electors proportionally. Every “intelligent man” (and woman, and anyone else for that matter) can understand that if you get 35% of the popular vote, you should get 35% of the electoral vote. There is no institutional cost, in other words, in finding the right answer. And there is no ongoing need to police the boundaries of the right answer, once the principle is clear.
If you win, won’t all states simply adopt district allocation of electoral votes, like Nebraska and Maine (sort of) do?
No. That doesn’t follow. If WTA is unconstitutional because it violates a principle of equality, then district allocation (with gerrymandered districts) is likewise unconstitutional, because it too violates that principle of equality. Indeed, district allocation could even be worse. (Here’s a wonderful site that allows you to gerrymander the presidential election: Same election, different lines: In one, Biden gets 506 electoral votes; in the other, he gets 89.)
This case is a long shot: Won’t losing make things worse?
How? Right now, there’s no recognized limit to what the states do. Already, many states are ginning up all sorts of crazy to try to reverse the effect of their own peoples’ vote. If the Court goes Rucho on us, then we’re back where we are now — fighting this unrepresentativeness politically. If it applies an equality principle, we’ll have another weapon in the arsenal.
Wouldn’t National Popular Vote, or some other constitutional reform be better?
Yes, of course it would. I support the NPV compact right now. And I would certainly support constitutional reform after that is in place. Both may be encouraged or pressed by the press of this litigation.
Your project is called “EqualVotes.US.” If you win, won’t votes still be unequal?
Yes—but not in a way that makes a partisan difference. If electoral votes were allocated proportionally, small states would still have more power per voter than big states. But small states are pretty evenly divided between Red and Blue. So that thumb on the scale — plainly intended—would have no partisan consequence. If you want complete one person, one vote, NPV is the only solution.