Maine, ranked choice voting, and the National Popular Vote Compact

Maine recently enacted ranked choice voting (“RCV”) for most of its elections. Very briefly, it allows voters to rank the preference of candidates instead of just picking one.

The original Maine bill excluded presidential elections from RCV. But the legislature recently approved expanding that to presidential elections (only later to be hung up and carried over to a later legislative session).

The new 21-A Maine Rev. Stat. Ann. § 805, sub-§2, if approved in the future, would be amended to read: “The presidential electors at large shall cast their ballots for the presidential and vice-presidential candidates who received the largest number of votes in the State according to the ranked-choice method of counting votes described in section 723-A. The presidential electors of each congressional district shall cast their ballots for the presidential and vice-presidential candidates who received the largest number of votes in each respective congressional district according to the ranked-choice method of counting votes described in section 723-A.”

Here’s how RCV looks in 723-A:

Except as provided in subsections 3 and 4, the following procedures are used to determine the winner in an election determined by ranked-choice voting. Tabulation must proceed in rounds. In each round, the number of votes for each continuing candidate must be counted. Each continuing ballot counts as one vote for its highest-ranked continuing candidate for that round. Exhausted ballots are not counted for any continuing candidate. The round then ends with one of the following 2 potential outcomes.

A. If there are 2 or fewer continuing candidates, the candidate with the most votes is declared the winner of the election.

B. If there are more than 2 continuing candidates, the last-place candidate is defeated and a new round begins.

Whew. What that means is, candidates are ranked. The top-ranked candidates on each ballot are tallied. The candidate with the fewest top-ranked votes is eliminated. The ballots are retallied, this time as if that eliminate candidate weren’t there, and voters who’d cast their first-place votes for that eliminated candidate now have their second choice counted as the first choice. This proceeds in rounds until there are 2 candidates, and the candidate with the higher vote is the winner.

All well and good for the State of Maine, which may “appoint” presidential electors in the “manner” that the legislature may “direct,” and RCV certainly falls within that.

But how might this interact with a national tally of the popular vote for presidential elections—and, specifically, the National Popular Vote Compact (“NPV”)?

While Maine is not (yet?) a member to the NPV, many states are [###

Here’s the relevant text of typical common language from the NPV:

Prior to the time set by law for the meeting and voting by the presidential electors, the chief election official of each member state shall determine the number of votes for each presidential slate in each state of the United States and in the District of Columbia in which votes have been cast in a statewide popular election and shall add such votes together to produce a ‘national popular vote total’ for each presidential slate.

The chief election official of each member state shall treat as conclusive an official statement containing the number of popular votes in a state for each presidential slate made by the day established by federal law for making a state’s final determination conclusive as to the counting of electoral votes by Congress.

(One important caveat from reading this text: in the event a state chooses not to hold a popular election for president—say, as Colorado did in 1876, when the legislature simply chose presidential electors—those states would not be included in a national popular vote total.)

So, what happens in Maine if RCV takes place? How would Maine’s vote be totaled in the national popular vote?

If there are, say, five candidates on the ballot, the popular vote total couldn’t really be the votes for those five candidates. RCV suggests paring them down until you get to two, then you name the winner. And voters cast votes understanding that their choices will be eliminated if they fail to advance in the next round. An advantage of RCV for independent or minor party candidates is that voters can vote for them without “wasting” a vote.

So RCV would then narrow down the votes of Mainers to two candidates. Those two candidates would then appear to be the only “votes” for a “presidential slate” in the State of Maine—that is, the tally after the final round of RCV. It would empower Maine voters over national voters—after all, the people of Maine would typically have all of their votes allocated to a Republican candidate or a Democratic candidate, as opposed to voters in other states that may scatter some number of voters among independent or minor party candidates.

But it could produce an alternative wrinkle. In 1992, Ross Perot received slightly more votes than George H.W. Bush to place second in the statewide vote. It’s not clear how RCV would have played out. But suppose that in the penultimate round of RCV Mr. Bush was eliminated, and the final round of RCV voting included just Bill Clinton and Mr. Perot. In the national popular vote total, the Republican candidate would receive zero votes from Maine—a small state, sure, but a significant cost to a major party candidate.

In short, I think inserting RCV in Maine is fine for Maine if it’s what Maine wants to do. But it’s precisely a reason why we cannot think of a “national popular vote” total by adding up the aggregate votes cast in 51 jurisdiction. This “invisible federalism” typicall operates in a way we don’t notice—we might simply look at each state’s votes and assume we can add them together for a single popular vote total. But the decisions of individual states, like a potential RCV in Maine, could have unforeseen consequences that undermine how we think about a national popular vote total. It’s a reason why Electoral College reform to make the presidential election truly national must occur at the federal level and not at the state level.

Washington State Supreme Court upholds fines for 2016 faithless electors

In the latest of a string of litigation surrounding faithless electors, the Washington State Supreme Court has issued its decision in In re Guerra, here. Four electors cast votes for candidates other than Hillary Clinton and Tim Kaine; all four were fined $1000 each pursuant to state law. Three appealed the decision. In an 8-1 decision, the Court upheld the fines.

The Court opens by acknowledging that presidential electors perform a “federal function.” The electors argued that if they are performing a federal function, there is ample case law that suggests that Congress cannot interfere with that activity. But the Court noted that states may still holds power over them under Article II of the Constitution. And while states might not be able to interfere with certain federal functions, the Court understood the precedent of cases like Ray v. Blair and McPherson v. Blacker that the state’s power included “broad authority.” Language from Supreme Court precedent suggested that the role of the elector is to “transmit the vote of the State for president,” (In re Green) “suggesting that the Electoral College vote belongs to the State, not the individual elector.” (p. 17)

Unfortunately, the Court’s interpretation of precedent does not rely as heavily on the text of the Constitution, which states that the electors “shall make distinct lists . . . which lists they shall sign and certify, and transmit sealed to the seat of the government of the United States.” (Amend. XII.) It’s certainly plausible to argue that the state has the power over the electors, but it is a harder textual claim to say that the votes “belong[] to the state,” whatever Supreme Court precedent may say.

But, the Court also finds that the Twelfth Amendment ensures that electors meet at a time and place, cast votes for two qualified candidates, and that the Amendment “does not limit a state’s authority in adding requirements to presidential electors.” (p. 18) The Court goes on to find that cases like U.S. Term Limits v. Thornton and Powell v. McCormack, which concluded that qualifications could not be added to congressional candidates, do not extend to presidential electors. (There is ample historical support for this practice, as qualifications have regularly been added to electors, including district residency restrictions, which were raised at length in U.S. Term Limits.)

As a textual matter, the court in n.8 rejects the notion that the word “ballot” implies “personal, secret ballot.” It points out that historically, the fact that “faithless electors” can be identified suggests the practice of casting ballots has not always been in secret. I think that’s an accurate understanding of the word “ballot,” a project I’m working on.

The Court rejects a First Amendment claim once it finds that there is no personal right of the elector.

A brief dissent argues that the “power to appoint” is not the “power to control.,” and it cites Justice Jackson’s concerns in his dissent in Ray v. Blair.

In short, it’s a fairly unsurprising outcome, but it leaves some deep uncertainty, I think, about how the United States Supreme Court’s precedents in this area harmonize with the text of the Constitution. For instance, some precedent—and this court’s opinion—conflate “state” with “legislature,” where the “legislature” is the entity empowered to “direct” the “manner” of “appoint[ing]” electors.

If the case is appealed to the United States Supreme Court, it also presents an interesting wrinkle—the electors here are not forbidden from casting “faithless votes,” but are only fined if they do so. That’s a less onerous (but still significant) consequence than replacing faithless electors, like what occurred in Colorado.

The Electoral College prevents electing a vice presidential candidate separate from a presidential candidate

Recently, news about a startup campaign to elect the vice president separately from electing the president has popped up. As the vice.run movement describes itself, its goal is to “reinstate the vice presidency as a democratically elected position. Vice.run’s goal is to create a vice presidential ballot line in the 2020 election in all 50 states.”

It sounds like a fascinating idea. But under any scrutiny, it just won’t work. That’s because the Electoral College doesn’t function this way.

To begin, it’s entirely correct to say that states don’t have to list a particular president-vice president combination unless they seek ballot access for that combination, and it’s possible to think about alternative arrangements to how we typically look at elections. A couple of historic examples come to mind. In 1836, the Virginia Democratic Party despised Martin Van Buren’s running mate Richard Mentor Johnson. While most states’ electors ultimately voted for Van Buren and Johnson, Virginia’s electors voted for Van Buren and William Smith. That prevented Johnson from receiving a majority of electoral votes, so the election was sent to the Senate, which elected Johnson over runner-up Whig Francis Granger.

In the same election, Whigs chose not to field a single candidate, but fielded multiple candidates that they hoped each would receive sufficient support over the Democratic candidate, prevent the Democratic candidate from securing a majority of electors, and send the race to the House of Representatives to choose among the top three vote-getters, potentially yielding a Whig candidate.

All this is to say we’ve seen some interesting and creative attempts to use the Electoral College framework to achieve particular ends. But what about a separate vice presidential candidate?

The logic (although there isn’t much law) spelled out on the vice.run site goes as follows: a vice presidential candidate could be a separate line on the ballot. The people could then choose to vote for a particular presidential candidate and an entirely different vice presidential candidate. Particularly attractive independent vice presidential candidates may force presidential candidates to choose them as their running mates.

All interesting ideas. But utterly unworkable due to the Electoral College.

Each state receives presidential electors equal to the number of senators and representatives they receive—say, Vermont receives three. On Election Day, the first Tuesday after the first Monday in November, when voters cast a vote for Clinton-Kaine or Trump-Pence, they are actually casting a vote for three electors pledged to support Clinton-Kaine or Trump-Pence. (“Pledged” I use very loosely—some states have formal pledges, some are legally binding, others are mere informalities or general expectations.)

Then, on the first Monday after the second Wednesday in December, Vermont’s three presidential electors meet in the state, along with electors from the other states in their respective states. The electors cast two votes: one for president, and on a distinct ballot, one for vice president. (The headline of my post is a bit misleading—in fact, presidential electors are required to vote for a separate president and vice president. This post is about the people voting via ballot separately.)

Suppose now there is an election with a separate line item for president and vice president. How would we choose the electors? We simply couldn’t. We can’t pick three electors for president and another three electors for vice president; that would require a state to choose six electors, which it couldn’t do. So we’d need the same electors for president and vice president—which means we couldn’t have separate lines unless the electors were the same, or aligned with one another, which would seem to defeat the purpose of having separate lines.

Instead, the only feasible opportunity to have presidential and vice presidential candidates elected separately would look as follows: the state legislature chooses three electors; it designates that those electors must be bound to vote in accordance with the popular vote of a presidential candidate and a vice presidential candidate; then, those electors would have to vote for whichever presidential and vice presidential candidate were selected.

It’s not clear that this is permissible. Federal law requires that the electors be appointed on the first Tuesday after the first Monday in November. It’s not a straw poll to bind existing electors. So it’s unclear to me that a state could appoint its electors before the election. Additionally, if Election Day is about the appointment of electors, then there has to be a fixed number equal to the state’s total, which means you need a president-vice president slate of electors, not independent lines.

There are other confusing ways to think through this problem, like assuming there’s just one slate of electors representing all the names on the ballot, but I’m not sure these are feasible, either. And I’m also assuming the state can compel the electors to vote for the candidate they’re pledged to support—something I think is right, but certainly not uncontested. (Of course, faithless electors are a common problem in any Electoral College reform scenario, so I don’t dwell on that.) It also increases the likelihood that voters in a state choose a president and a vice president from their own state, if there are pluralities or tradeoffs in voting, which is impermissible under the Constitution—electors must vote for at least one candidate who is not an inhabitant of their state, and a president-vice president slate usually avoids this.

Unless I’m missing something—correct me if I am!—this scheme simply can’t work absent the legislature wholesale taking over the choosing of electors and having a straw poll of the people sometime before Election Day. The Electoral College anticipates one set of electors chosen on Election Day. That set of electors can exist however one sees fit, of course. But if voters have to make a choice of a president and a vice president on separate lines, there’s no feasible way to pick one common slate of electors like the Constitution demands.

"Nonjudicial Solutions to Partisan Gerrymandering"

I have posted this draft of an Essay forthcoming in the Howard Law Journal, “Nonjudicial Solutions to Partisan Gerrymandering.” Here is the abstract:

This Essay offers some hesitation over judicial solutions to the partisan gerrymandering, hesitation consistent Justice Frankfurter’s dissenting opinion in Baker v. Carr. It argues that partisan gerrymandering reform is best suited for the political process and not the judiciary. First, it traces the longstanding roots of the problem and the longstanding trouble the federal judiciary has had engaging in the process, which cautions against judicial intervention. Second, it highlights the weaknesses in the constitutional legal theories that purport to offer readily-available judicially manageable standards to handle partisan gerrymandering claims. Third, it identifies nonjudicial solutions at the state legislative level, solutions that offer more promise than any judicial solution and that offer the flexibility to change through subsequent legislation if these solutions prove worse than the original problem. Fourth, it notes weaknesses in judicial engagement in partisan gerrymandering, from opaque judicial decisionmaking to collusive consent decrees, that independently counsel against judicial involvement.

This Essay is a contribution to the Wiley A. Branton/Howard Law Journal Symposium, "We The People? Internal and External Challenges to the American Electoral Process." is

Statehood, the District of Columbia, and the Twenty-Third Amendment

There’s a renewed effort for statehood for the District of Columbia in the new Democratic-controlled House, and H.R. 51 is the proposal to do so. (DC’s non-voting representative, Eleanor Holmes Norton, has snagged bill #51 as a symbolic gesture in the past, too.)

I’d recently wondered about whether electors of the District of Columbia could cast votes for presidential and vice presidential candidates who resided in the District. As you may know, the Twelfth Amendment requires that electors cast two votes, one for president and one for vice president, “one of whom, at least, shall not be an inhabitant of the same state with themselves .”

DC isn’t a state, but the Twenty-Third Amendment gave DC presidential electors. So, could DC electors vote for an all-DC ticket? No, due to a clever phrase in the Amendment: “they shall be in addition to those appointed by the states, but they shall be considered, for the purposes of the election of President and Vice President, to be electors appointed by a state.” (Emphasis added.) In other words, whatever parts of the Constitution refer to “state” in the context of presidential electors? Those apply to DC’s electors, too.

But back to H.R. 51. The bill, like many bills, excises several blocks from the District when creating a new state. Those blocks are still the seat of government of the United States, and not a part of the state. So a bill like H.R. 51 would create a new state out of the old District, but it would basically split the District into two: new-state-District, and seat-of-government-district. The seat-of-government-district being quite small, essentially residuals made up of several federal buildings and the like.

Of course, the Twenty-Third Amendment comes back into play: “The District constituting the seat of government of the United States shall appoint in such manner as the Congress may direct.” So if DC becomes a state, it gets two Senators, at least one Representative, and at least three presidential electors. But the residual district—remember, just a few federal buildings carved out—is constitutionally entitled to presidential electors. No more than a few dozen people may live in this new seat of government—after all, the residences of DC have been put into the new state.

A little Googling revealed this point has been raised before. That is, before DC can become a state, it ought to be conditioned on a repeal of the Twenty-Third Amendment, lest an anomalous residual set of a few federal buildings is entitled to a slate of presidential electors. That hasn’t been a point of emphasis, of course—getting popular support for statehood for DC is a high hurdle, and the first one that must be surmounted—but it struck me as notable.

"Why not continue the political struggle in partisan-gerrymandering cases?"

I have this piece up at SCOTUSblog entitled, “Symposium: Why not continue the political struggle in partisan-gerrymandering cases?” It begins:

“In a democratic society like ours, relief must come through an aroused popular conscience that sears the conscience of the people’s representatives.” So wrote Justice Felix Frankfurter in his dissenting opinion in Baker v. Carr in 1962.

It was, of course, a dissent. A majority of the Supreme Court in short order reorganized state legislatures according to its own understanding of fair representation — that population should be roughly equal in each legislative district. And the majority’s basis for doing so, Frankfurter’s dissent chided, “ultimately rests on sustained public confidence in [the Court’s] moral sanction.”

The political process is a messy thing. It is laborious to educate the public on a matter and convince them of that matter’s significance. It is time-consuming to wait through election cycles to enact political changes. Impatient litigants demand the federal courts to intervene when the political process moves too slowly.

Two cheers for Florida's Amendment 4

I spend very little time on this blog on matters of policy, or the wisdom of particular laws. I spend almost all of my time discussing what the law, including the Constitution, authorizes or forbids. Occasionally, I dip into practical concerns of changes to laws, but these are Burkean by nature and tend to be contextual, offering some view of the law working with other laws, or how the law fits into a longstanding regime.

Florida enacted Amendment 4, which the people of Florida undoubtedly have a right to enact. I thought I’d dip into policy for a moment: the good of the Amendment and, in my view, the bad. And I thought I’d use the overused expression, two cheers for the amendment.

Amendment 4 amended Article VI, Section 4 of the Florida Constitution as follows (underlined text added):

(a) No person convicted of a felony, or adjudicated in this or any other state to be mentally incompetent, shall be qualified to vote or hold office until restoration of civil rights or removal of disability. Except as provided in subsection (b) of this section, any disqualification from voting arising from a felony conviction shall terminate and voting rights shall be restored upon completion of all terms of sentence including parole or probation.

(b) No person convicted of murder or a felony sexual offense shall be qualified to vote until restoration of civil rights.

I have a lot of complicated views of suffrage laws, who ought to be eligible to vote in our democracy. But post-conviction felon disenfranchisement laws are an easier case for me. I don’t understand them and generally oppose them.

The felon (and the criminal more generally) has broken the social compact (the best of a number of imperfect ways of putting it) and merits punishment. Imprisonment is the most common form of that punishment. I think we overuse imprisonment, particularly as we overvalue incapacitation as a reason for punishment. But the release from imprisonment should, I think, mark a return to society, and that includes attempts to reincorporate the ex-felon into the ordinary things of citizenship.

Collateral consequences, then, typically are not the stuff we view as what the felon merits as his just deserts. They are sanctions we attach to the ex-felon as prophylactic stuff of society—don’t work in this job, don’t live near this park, don’t seek a security clearance, don’t vote, because, well, you’re dangerous, or we simply don’t think you ought to really be a part of our society. In a way, the felon has broken the social compact and is permanently barred from certain aspects of society, even after serving time in prison.

It might be that we as a society think that the felon merits these penalties. But, I think, to the extent we desire to restore the felon into the community, I don’t know that long-standing penalties that extend well after imprisonment (again, to the extent we value imprisonment as the primary form of punishment) have much place in a criminal justice regime. I could probably spend much longer thinking and discussing such things, but this is my high-level approach right now.

But a few points about the specifics of this particular amendment.

First, I think those who have been convicted of an election-related felony should be permanently disenfranchised. This is assuredly a very narrow category, and perhaps too narrow for a carve-out in this constitutional amendment. But I do think that if you have tampered with an election—voter fraud, voter intimidation, voter bribery, etc.—you, perhaps, ought to be permanently excluded from participating in future elections. It is a natural consequence for your disruption of democracy that you are no longer allowed to play this part in it. (Of course, you may play many other parts—advocate, campaigner, endorser, opinion-writer, etc.) I think it’s an area where the punishment ought to haunt you the rest of your life, and the punishment feels particularly tethered to the crime committed. But, again, perhaps that’s too narrow and too picky.

Second, the law carves out two easy targets, murderers and sexual offenders. One sad element of these carve-outs is political. The hasty rejoinder from a proposed law that enfranchise ex-felons is, “So if you’re convicted of murder, that person can cancel out your vote?” Of course, murderers spend most, if not all, of their lives in prison (indeed, some in Florida are executed). And perhaps general extension of ex-felon enfranchisement is better than no extension at all.

But worse, I think, are how we treat sexual offenders. Rape and child molestation are two particularly heinous crimes. Others, like statutory offenses in “Romeo and Juliet” cases, far less heinous. But in all of them, we spend little time thinking about the punishment these felons merit, and far more time seeking to maximize punitive consequences as long and as far-reaching as possible.

Disenfranchising murderers or sexual offenders who have served their time of imprisonment does not seem terribly well-fitted to a theory of punishment, except that we strongly dislike their offenses and want to keep them at the periphery of society. Maybe something about their offenses merits it. When it comes to voting, however, I’m not so convinced.

I understand that the perfect cannot be the enemy of the good. But, in some ways, I wish we’d spend more time thinking critically about the nature and purposes of punishment of crimes more generally, and how disenfranchisement fits in that scheme. I think Florida’s new Amendment 4 mostly gets it right.

As a final note, the amendment is the latest of a number of remarkable decisions of the people of the several states to dilute their own voting power, from enfranchising the freedman to enfranchising women to enfranchising 18-year-olds. It’s a remarkable and rather selfless decision of the people, not foisted on the people through a judicial construction. Every decision to broaden the franchise dilutes the voters’ votes who’ve approved that broadening. It’s probably a topic worth exploring another day, but I close on that thought.

How changes to the Democratic presidential primaries may affect 2020 voting

One common phrase I’ve heard at conferences discussing proposed reforms to presidential primaries is something along the lines of, “Reformers are always targeting last cycle’s election.” That is, sometimes reformers are too myopic in trying to fix last cycle’s changes and fail to recognize the unintended consequences it could have on the next cycle—or that changes in events may alter how we view the next cycle.

I want to focus on two items for consideration well ahead of the 2020 Democratic presidential primaries. The first are changes to the Iowa caucuses. The second are changes to the primary calendar. Some speculation follows the observations of each change, with an assumption that a significant number of Democratic candidates (say, at least 10 and probably more than 15 viable candidates) will run, similar to the number in the 2016 Republican presidential nominating contest.

Iowa Caucuses

The Democratic caucuses in Iowa have long had distinctive features. Democratic voters show up at precinct caucus sites and publicly express support for one candidate or another. They group together in wings of the precinct site. Then there’s a period of “realignment,” where supporters of non-viable candidates (those with few supporters) throw their support behind another, more viable candidate. The caucus site names a winner; no popular vote is taken. Those caucus site winners are compiled into an overall distribution of delegates.

Multiple reforms appear to be shaping up for 2020. The caucus will resemble more of a traditional primary, if all the changes are approved. There will be absentee voting, which should dramatically increase popular participation. Raw vote totals will be released.

But it also means that this “realignment” may disappear. This was a potentially crucial opportunity for insurgent candidate Barack Obama in 2008. Reports widely circulated before the Iowa caucuses in 2008 that Bill Richardson, a second-tier candidate, had urged his supporters to throw their support behind Mr. Obama in the event of realignment (the same was true for marginal candidate Dennis Kucinich). The realignment period offered a kind of ranked choice voting. It also prevent too much fragmentation: three candidates—Mr. Obama, John Edwards, and Hillary Clinton—secured about 97% of precincts.

The Iowa caucus changes haven’t been finalized yet, but if the Iowa caucuses look more like a traditional primary, or the Republican caucuses in Iowa in 2016, we would expect to see more fracturing (no candidate broke 30% of the popular vote), more delegates awarded to more candidates, and, potentially, less opportunities for the kind of second-choice realignment that Mr. Obama benefited from in 2008.

2020 Calendar

Many states jockey for an early position in the primaries to exert their influence in the process. Iowa, New Hampshire, Nevada, and South Carolina have received privileged status in recent years. But by the first Tuesday in March, other states may follow suit with primaries.

In 2016, Alabama, Arkansas, Colorado, Georgia, Massachusetts, Minnesota, Oklahoma, Tennessee, Texas, Virginia, and Vermont held Democratic primaries or caucuses on the first Tuesday. In 2020, Arkansas has likely moved its primary back to May; California and North Carolina have moved their primaries up to the first Tuesday in March.

For candidates who survive the early stages, viability may come through home-state advantages, particularly among delegate-rich pools of states. Prospective candidates Elizabeth Warren (Massachusetts), Beto O’Rourke (Texas), and Julian Castro (Texas) all may find strength on this dates; so too would Kamala Harris (California), after California moves up its primary.

It’s worth noting that among the most viable challenger to Donald Trump for the Republican nomination after Super Tuesday appeared to be Ted Cruz—in part because he secured over 200 delegates that day, but largely because Texas’s primary (and 104 delegates of its 155 possible delegates he won) came on that day.

The fortune of the calendar, then, might provide added strength for certain candidates.

Speculation from procedural changes

Yes, much of this is speculative. But it’s worth considering that we often look back at a presidential primary and note certain things that occurred because of the procedures in place. The changes in this cycle will undoubtedly have some impact on the Democratic presidential primaries. We won’t really know what those look like until after the fact, but I hope thinking structurally and drawing some recent comparisons offers some useful perspective ahead of the 2020 presidential primaries.