9.27 Myths about Ranked Choice Voting

Description of Ranked Choice Voting

Ranked choice voting (RCV) allows the voter to numerically rank candidates on their ballot in order of preference—first choice, second choice, and so forth.

Figure 9.15 shows a sample ballot for the 2020 presidential election in Maine.

RCV is sometimes called “instant runoff voting,” because the ballot-counting process resembles a series of runoff elections.[625]

In the first round of counting in RCV elections, each ballot counts as one vote for the candidate whom the voter ranked as their first choice. If any candidate receives an absolute majority of the votes, the counting process stops. If not, the candidate with the fewest votes is eliminated, and each ballot for the just-eliminated candidate is counted as one vote for that voter’s next choice. This process of counting and eliminating the weakest candidate is repeated until one candidate has the support of a majority of the ballots expressing a choice.[626]

Supporters of RCV argue that it is preferable to the conventional plurality-voting system used in almost all elections in the United States,[627] because the winner has the support of a majority of voters expressing a choice.

RCV supporters argue that it gives candidates a strong reason not to run harshly negative campaigns, because winning often requires earning the support of some voters whose first choice was eliminated.

Moreover, RCV eliminates the dilemma of voting for the lesser of two evils—instead of the candidate who most closely matches the voter’s views.

For example, RCV enables Libertarian voters to give their first-choice ranking to the Libertarian Party candidate, but to also give their second-choice ranking to the Republican candidate. Some Green voters might give their first-choice ranking to the Green Party candidate, but then give their second-choice ranking to the Democratic candidate.

In the traditional plurality-voting system, a voter who supports a minor-party or independent candidate often aids the major-party candidate whose views are farthest from the voter’s.

For example, 97,488 Floridians voted for Ralph Nader for President in 2000. If those voters had been able to express their second-choice on their ballots, George W. Bush almost certainly would not have carried Florida by 537 votes—and therefore would not have become President.

Similarly, in 2020, Libertarian presidential candidate Jo Jorgensen received more than three times as many popular votes in Arizona, Georgia, and Wisconsin (152,185) as Biden’s margin over Trump in those states, as shown in table 9.30. Without the 37 electoral votes from these three states, there would have been a 269–269 tie in the Electoral College. On January 6, 2021, the Republican Party had a majority of the House delegations and would have been in a position to choose Trump as President.[628]

History and constitutionality of ranked choice voting

RCV has been used for decades in numerous municipal elections in the United States.

As of 2024, RCV is used on a statewide basis in Maine and Alaska and by over 50 cities and counties in various states.[629]

Maine was the first state to adopt RCV on a statewide basis. In November 2016, its voters approved an initiative petition that adopted RCV for use in elections for Congress and state offices—but not President.[630]

In 2017, the Maine Supreme Judicial Court issued an advisory opinion saying that, based on its interpretation of the Maine Constitution, RCV could not be used in general elections for state offices. The court noted that RCV could be used in primary elections for state offices and in both primary and general elections for federal offices.[631], [632]

As a result, RCV was used in Maine in 2018 for both the primary and general elections for U.S. Senator and U.S. Representative—but only in the June primary election for state offices.[633]

The constitutionality of Maine’s RCV law was contested in federal court after RCV played a decisive role in Maine’s 2^nd congressional district election in 2018.

Both a federal district court and a federal appeals court upheld the constitutionality of Maine’s RCV law. Both found RCV to be a “one-person, one-vote” system.[634] These federal-court rulings characterized the objections raised against RCV as primarily differences of opinion as to what constitutes a desirable voting system—rather than valid legal arguments as to what is, or is not, constitutional.[635]

In his 2018 opinion Federal District Judge Walker wrote:

“Whether RCV is a better method for holding elections is not a question for which the Constitution holds the answer. … To the extent that the Plaintiffs call into question the wisdom of using RCV, they are free to do so, but … such criticism falls short of constitutional impropriety. A majority of Maine voters have rejected that criticism and Article I does not empower this Court to second guess the considered judgment of the polity on the basis of the tautological observation that RCV may suffer from problems, as all voting systems do. The proper question for the Court is whether RCV voting is incompatible with the text of Article I by giving the language its plain and ordinary meaning.”[636] [Emphasis added]

In 2019, the Maine legislature passed a law extending RCV to presidential elections.[637]

In 2020, the constitutionality of Maine’s law was again challenged in federal court. Federal District Judge Walker upheld Maine’s RCV law against the claim that voters were forced to vote for candidates they did not favor.[638]

RCV has also been upheld by the highest courts of Massachusetts and Minnesota in cases involving local elections.[639]

RCV was used for the first time in a presidential election in Maine in 2020 (as shown by the sample ballot in figure 9.15).

In November 2020, an absolute majority of Maine’s voters gave Biden their first-choice ranking on a statewide basis. Thus, the statewide RCV counting process ended in the first round. That is, the first-round count was equivalent to the final-round count.

Under Maine law, RCV is separately applied at the statewide level (for two electoral votes) and at the congressional-district level. An absolute majority of voters in the 1^st district (the southern part of the state) gave Biden their first-choice ranking. An absolute majority of voters in the 2^nd district (the northern part of the state) similarly gave Trump their first-choice ranking. Thus, the first-round count was equivalent to the final-round count in both districts.

Maine is not the only state that will use RCV in the 2024 presidential election.

In November 2020, Alaska voters approved an initiative petition that established a top-four multi-party primary for offices other than President and the use of RCV in general elections for all offices—including President.[640]

In 2021, the Alaska Supreme Court unanimously upheld the constitutionality of RCV.[641], [642], [643]

In 2022, RCV was used in Alaska in the general election for U.S. Senator and U.S. Representative as well as state offices, including Governor.[644]

In 2023, RCV opponents in Alaska launched an initiative petition to repeal RCV. The opponents included former Governor Sarah Palin, who attributed her loss in her 2022 congressional race to Alaska’s use of RCV. Voters are expected to vote on the question of repealing RCV in Alaska in November 2024.

Political context

A substantial percentage of supporters of RCV are supporters of the National Popular Vote Compact (NPV), and vice versa.

Moreover, opponents of RCV are very often opponents of NPV.

Trent England, Executive Director of Save Our States (the leading lobbying organization opposing NPV), also serves as a leading spokesman for Stop RCV (a lobbying organization opposing RCV).

The supporters of these anti-RCV and anti-NPV groups include the Honest Election Project, Heritage Action, and the Oklahoma Council on Public Affairs, a think tank that employs Trent England as its Vice President.

Because of the overlap of support for RCV and NPV and the overlap of opposition to RCV and NPV, the opponents of RCV and NPV have attempted to divide the electoral reform community by claiming that RCV and NPV are incompatible.[645] Their aim is to get supporters of NPV to oppose RCV, and to get supporters of RCV to oppose NPV.

Save Our States incorrectly claims that RCV is incompatible with the National Popular Vote Compact.

In 2023, Sean Parnell, Senior Legislative Director of Save Our States, submitted written testimony to the Minnesota Senate Elections Committee claiming:

“There is a fundamental incompatibility between the National Popular Vote interstate compact (NPV) and an election process used by some states called Ranked Choice Voting (RCV).”[646], [647], [648]

Jeanne Massey, Executive Director of FairVote Minnesota (the leading advocate for RCV in Minnesota),[649] submitted written testimony to a Minnesota House committee the day after Parnell’s testimony:

“I have read the opposing testimony related to RCV and National Popular Vote compatibility, and it is misleading and incorrect. The testimony comes from an organization opposed to both RCV and NPV and has a clear motive—to hurt both reforms. … I urge you to disregard the unproven, misleading argument that RCV and NPV are incompatible and support the NPV legislation before you.”[650] [Emphasis added]

A policy memorandum from Save Our States says:

“The National Popular Vote interstate compact (NPV) and an election method known as Ranked Choice Voting (RCV) are … fundamentally incompatible.”

“The incompatibility of RCV and NPV could prevent a conclusive determination of which candidate has won the presidency, causing a political, legal, and constitutional crisis and throwing the nation into turmoil.”[651]

The problem that allegedly will provoke a constitutional crisis was described in written testimony to the Minnesota Senate Elections Committee on January 31, 2023, by Parnell:

“NPV anticipates that every state will produce a single vote total for each candidate, but RCV produces at least two: an initial vote count, before the RCV process of transferring votes, and the final vote count at the conclusion of the RCV process. This would produce uncertainty, litigation, and opportunities for manipulation if NPV took effect.”[652], [653], [654] [Emphasis added]

In fact, there is no legitimate uncertainty as to whether to use the first-round count or the final-round count in computing the national popular vote.

Indeed, it would be preposterous to interpret an RCV-for-President law to mean that a state would hand voters a ballot enabling them to rank candidates according to their first, second, and other preferences—but then would ignore everything on the ballot except the voter’s first choice.

Using only the first-round count would negate the purpose of having an RCV-for-President law in the first place—namely to give voters the opportunity to rank candidates and have those rankings matter.

Moreover, the outcome of every election in every jurisdiction (state or local) that uses RCV in the United States is based on the final-round count—not just the first-round count or any other intermediate count. Nothing in Alaska’s or Maine’s RCV laws even hints that the state’s final result for President should be arrived at differently than for U.S. Senator, U.S. Representative, or the other offices covered by the state’s RCV law.

Finally, voters need to know how their vote for President will be counted before they decide how to vote.

• If only the first-round count is going to matter, many supporters of the Libertarian, Green, and other minor-party nominees for President might well choose to pragmatically give their first-choice ranking to one of the major-party candidates.
• If the final-round count is going to matter, such voters would vote their conscience and give their first-choice ranking to their genuine first choice.

Voters would be misled if the state were to provide them with a ballot allowing them to rank candidates, but then ignore all but their first-choice ranking.

In short, there is no good-faith legal argument in favor of using anything other than the final-round count produced by RCV.

There is no uncertainty about the statutory interpretation in the only two states that currently use RCV in presidential elections.

The interpretation of the RCV-for-President laws is a settled legal question in both of the states that currently use it (Maine and Alaska).

Maine passed its RCV-for-President law in 2019.

In November 2020, Maine used RCV in a presidential election for the first time.

In a Harvard Law & Policy Review article[655] written in 2019 and published in 2020, Rob Richie, FairVote’s founding Chief Executive Officer, and his co-authors discussed RCV in relation to the National Popular Vote Compact.[656]

At the time of the article, RCV had not yet been used in a presidential election. The article raised the rhetorical question of how the national popular vote total would be computed if the National Popular Vote Compact were in effect, but no presidential candidate were to win an absolute majority of the votes in the first round of RCV counting.

The rhetorical question raised by the Harvard Law & Policy Review article did not come up in Maine in the 2020 election, because Biden won an absolute majority of the votes in the first round (both statewide and in each congressional district). Thus, the first-round count was the final-round count.

In 2021, Secretary of State Shenna Bellows proposed an amendment to the state’s RCV-for-President law to eliminate any possible ambiguity in the state’s RCV-for-President law. Bellow’s 2021 proposal designated the final-round RCV tally as Maine’s final determination of its presidential vote count. The Maine Governor signed the Secretary of State’s recommended bill into law on June 17, 2021. That law provided:

“§803. Duties of Governor: The Governor shall send a certificate of the determination of the electors to the Archivist of the United States under the state seal. The certificate must state … the number of votes each candidate for President received … in the final round of tabulation under section 723A [Maine’s RCV law].”[657] [Emphasis added]

Maine had not enacted the National Popular Vote Compact into law at the time that it clarified its RCV-for-President law in 2021.

When Maine enacted the Compact in 2024, it retained the wording of the 2021 law quoted above and added the following new section specifically referring to Article III of the Compact (which is section 1304 of Maine law):

“When the National Popular Vote for President Act governs the appointment of presidential electors, the Governor has the following duties.

“As soon as possible after the canvass of the presidential count under section 723-A, subsection 7 is determined, the Governor shall send a certificate of determination containing the names of the electors and the statewide number of votes for each presidential slate that received votes in the final round to the Archivist of the United States under state seal. This final round vote is deemed to be the determination of the vote in the State for the purposes of section 1304.

“As used in this paragraph, ‘final round’ means the round that ends with the result described in section 723-A, subsection 7, paragraph C, subparagraph (1).”[658]

Alaska is the only other state that is poised to use RCV in the 2024 presidential election.

In 2022, the Alaska Supreme Court eliminated any uncertainty about the issue in its unanimous opinion upholding RCV (“Initiative 2”) in Kohlhaas v. State.

“With ranked-choice voting, the vote count is not final after the first round of tabulation. Maine’s law provided that if there were more than two candidates left ‘the last-place candidate [was] defeated and a new round [of tabulation began],’ repeating until two candidates remained and the candidate with the most votes was declared the winner. Similarly, Initiative 2 specifies that the tabulation ‘continues’ until two or fewer candidates remain and ‘the candidate with the greatest number of votes is elected and the tabulation is complete.’ According to both states’ ranked choice voting laws, the vote count is not complete until the final round of tabulation.”[659] [Emphasis added]

In summary, the issue is settled in both Maine and Alaska.

There is no uncertainty about the statutory interpretation of RCV-for-President ballot propositions that voters may enact in November 2024.

In November 2024, Oregon voters will decide whether to use RCV in federal and statewide elections, including the general election for President.[660] Oregon’s proposed RCV law explicitly states that the final-round RCV count will be the state’s final determination of its presidential vote count. The proposed Oregon RCV law reads:

“(B) If the National Popular Vote interstate compact set forth in section 1, chapter 356, Oregon Laws 2019, governs the appointment of presidential electors and the election of presidential electors in this state is determined by ranked choice voting:

(i) The determination of which candidates for presidential elector shall be declared elected in this state shall be made in accordance with the provisions of the National Popular Vote interstate compact; and

(ii) The “final determination” of the presidential vote count reported and certified to the member states of the compact and to the federal government shall be the votes received in the final round of statewide tabulation by each slate of candidates for the offices of President and Vice President of the United States that received votes in the final round of statewide tabulation.”[661] [Emphasis added]

As of July 2024, it appears that an initiative petition to adopt RCV may be on the ballot in November 2024 in the District of Columbia.[662]

The proposed RCV law in the District of Columbia, like Oregon’s, explicitly states that the final-round RCV count will be the final determination of its presidential vote count:

“If the appointment of presidential electors following any general election for President of the United States is governed by the National Popular Vote Interstate Agreement Act of 2010, effective December 7, 2010 (D.C. Law 18-274; D.C. Official Code §1-1051.01), then, in any general election for President and Vice-President of the United States using ranked choice voting:

“(1) The certification of the appointment of electors shall be made in accordance with the provisions of such Act;

“(2) The final determination of the presidential vote count reported and certified to the States that have enacted such Act, for purposes of that Act, shall be:

“(A) In an election using ranked choice voting pursuant to subsection (d) of this section, the votes received in the final round of tabulation by each slate of candidates for the offices of President and Vice President of the United States that received votes in the final round of tabulation.”[663] [Emphasis added]

Other RCV proposals that may be on the ballot in November 2024 do not apply to presidential elections.

In November 2024, Nevada voters will be voting on an initiative petition to adopt the “final 5” system for nominating candidates in the primary and RCV for the general election. However, this proposed legislation does not include President.[664]

As of July 2024, it appears that RCV legislation will also be on the ballot in Arizona and Colorado in November 2024; however, neither of these proposals covers presidential elections.[665]

Similarly, the proposal that may be on the ballot in Idaho in November 2024 does not apply to presidential elections.[666]

Footnotes

Trent England, Executive Director of Save Our States, has written:

“Maine has no power to tell California (for example) which set of numbers to use.”

“Officials in various states [belonging to the NPV Compact] would just decide, on their own and with no legal guidance, which numbers to use from Maine or any other states using RCV or similar election systems.

“The changes suggested by Secretary Bellows seek to solve this problem by reporting only the final RCV-adjusted numbers to other states on Maine’s Certificate of Ascertainment. … Officials in other NPV states could still decide to ignore Maine’s preference and use the raw numbers from the statewide canvas.”[667] [Emphasis added]

Contrary to what England says, every state—whether it is a member of the Compact or not—does have the power to tell states belonging to the National Popular Vote Compact how to treat its presidential vote count.

Every state has that power, because the Compact explicitly requires officials in states belonging to the Compact to treat every state’s final determination of its presidential vote count as “conclusive.” Specifically, the fifth clause of Article III of the NPV Compact states:

“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.” [Emphasis added]

Maine’s law specifically designates the final-round RCV tally as the state’s final determination of its presidential vote count.

Maine’s law is not a suggestion, hint, nudge, or plea to the election officials of states belonging to the National Popular Vote Compact. Those officials are legally required by the Compact to treat Maine’s final determination of its presidential vote count as “conclusive.”

Any ambiguity about how to interpret future RCV-for-President laws will be decided before any election based on the national popular vote.

Although the RCV-for-President laws that voters will consider in the November 2024 election explicitly address the issue of statutory interpretation raised by Save Our States, no “constitutional crisis … throwing the nation into turmoil” would arise if some future RCV-for-President law ever happened to be silent about this issue.

In the unlikely event that some future RCV-for-President law were to fail to address the issue, voters in the state involved would seek a declaratory judgment prior to Election Day so that they would know how their votes would be counted.

This question of statutory interpretation would arise and be resolved before Election Day for two reasons.

First, voters need to know how their vote for President will be counted before they decide how to cast it. If only the first-round count is going to matter for President, Libertarian and Green Party voters might pragmatically choose to give their first-choice ranking to one of the major-party candidates. If the final-round count is going to matter, such voters would give their first-choice ranking to their genuine first choice.

Second, courts generally apply the doctrine of laches to reject post-election challenges in cases in which the plaintiff was aware of an issue before the election but failed to initiate litigation until seeing the election results.

Thus, it would be virtually mandatory to raise this issue in court before the election and for the courts to settle the question before the election.

As explained in Dobbs and Robert’s Law of Remedies, Damages, Equity, Restitution:

“Laches is unreasonable delay by the plaintiff in prosecuting a claim or protecting a right of which the plaintiff knew or should have known, and under circumstances causing prejudice to the defendant.”[668]

As the U.S. Court of Appeals for the Fourth Circuit wrote in Hendon v. North Carolina State Board of Elections:

“Courts have imposed a duty on parties having grievances based on election laws to bring their complaints forward for pre-election adjudication when possible. They have reasoned that failure to require pre-election adjudication would ‘permit, if not encourage, parties who could raise a claim “to lay by and gamble upon receiving a favorable decision of the electorate” and then, upon losing, seek to undo the ballot results in a court action.’”[669], [670], [671]

UCLA Law Professor Richard L. Hasen—a leading expert on both election law and the law of remedies—explains that:

“laches … prevent[s] litigants from securing options over election administration problems.”[672]

In short, the question of interpreting a state’s RCV-for-President law would be litigated in the state involved, and such litigation would occur prior to the time when voters start to cast their ballots.

Whatever the outcome of litigation of this question of statutory interpretation, that state’s final determination of its presidential vote count would be made in accordance with the final judicial interpretation of its RCV-for-President law.

While FairVote and virtually all other supporters of RCV believe that the only reasonable interpretation of an RCV-for-President law that is consistent with RCV’s essential purpose is that the final-round count should be used to compute the national popular vote total, the alternative interpretation would present no operational difficulty for the National Popular Vote Compact. The Compact simply requires that its member states treat the state-of-origin’s “final determination” as “conclusive.”

Save Our States continues to complain even after Maine eliminated the alleged ambiguity.

In 2021, Sean Parnell, Senior Legislative Director of Save Our States, testified against the Maine Secretary of State’s recommended amendment that eliminates the arguable ambiguity.

In testifying before the Maine Committee on Veterans and Legal Affairs on May 11, 2021, Parnell mentioned the historical fact that in the 1992 race involving incumbent President George H.W. Bush, Bill Clinton, and Ross Perot, one of the major-party candidates came in third in Maine:

• Clinton finished first with 38.8%
• Perot finished second with 30.44%
• Bush finished third with 30.39%.[673]

One of the major-party candidates also came in third in Utah in 1992:

• Bush finished first with 43.3%;
• Perot finished second with 27.3%
• Clinton finished third with 24.7%.[674]

The major-party nominees have come in first or second in 610 of the 612 state-level counts in the 12 presidential elections between 1972 and 2020.

These two cases in 1992 have been the only times since George Wallace’s 1968 run for President that a third-party nominee has come in second in any state.[675]

Parnell testified:

“Under Ranked Choice Voting, if a third party or an independent candidate were to finish ahead of either the Democratic or Republican candidate, … the votes for that Democratic or Republican candidate get completely erased and will not be reported.

“In 1992, for example, Ross Perot finished ahead of George Bush in Maine. George Bush would have had subtracted, or never appeared in the national vote totals about 207,000 votes. The amendment that your Secretary of State has offered does not address this problem.”[676] [Emphasis added]

At a debate conducted by the Broad and Liberty group in Philadelphia in 2021, Sean Parnell said:

“If you’re just using the final votes, then if a candidate—a Democrat or Republican—ever finishes in third place in a state with ranked choice voting, … then what you wind up doing is literally zeroing out votes. If you ever have a Republican candidate or Democratic candidate finishing third place in a state with ranked choice voting, then you are literally going to watch hundreds of thousands, maybe even millions of votes, be completely erased.”[677] [Emphasis added]

Of course, Parnell is a vigorous defender of the current winner-take-all method of awarding electoral votes—a system that actually “erases” the popular votes cast for every second-place and every third-place candidate in every state in every election—that is, in all 612 state-level vote counts in the 12 presidential elections between 1972 and 2020.[678]

There is another important difference in the transferring that might have occurred under RCV in these two cases out of 612. If RCV and National Popular Vote had been in effect in 1992 when Bush came in third in Maine, and Clinton came in third in Utah, every voter in Maine and Utah would have had their vote counted for a candidate for whom they had actually voted. In contrast, the current winner-take-all system routinely transfers the voter’s vote to a candidate for whom the voter did not vote.

Moreover, Parnell’s objection also fails to acknowledge the important fact that every voter makes their choice and casts their vote with an awareness of the existing voting system. The people who voted for Ross Perot in 1992 in Maine and Utah were aware that doing so could either:

(1) switch the state’s popular-lead from one major-party candidate to the other, or

(2) result in their state’s electoral votes going to Perot—thereby potentially depriving one of the major-party candidates of Maine’s four or Utah’s six electoral votes.

Aware of the existing rules of the game, these voters cast their ballots for Perot in 1992. It is condescending to suggest that these voters were ignorant or confused about the implications of their votes. Lawmakers, voters, and the RCV voting system are not obligated to protect the two major-party candidates from the consequences of their own failure to earn enough support to come in first or second place.

Indeed, the precise purpose of RCV is to honor the voter’s second choice in case the voter’s first-choice candidate cannot win. This is not a bug of RCV, but a feature.

Given that Parnell vigorously defends the current system, which erases the popular votes cast for every second-place and third-place candidate in every state in every election, the concern about what might happen in two elections out of 612 elections is little more than crocodile tears.

Footnotes

Sean Parnell, Senior Legislative Director of Save Our States, stated in written testimony to the Maine Veterans and Legal Affairs Committee on January 8, 2024:

“Final RCV results typically take much longer to be determined than plurality voting. In 2022, the winner of Maine’s 2^nd congressional district wasn’t known until November 16, more than a week after the November 8 election, and Alaska’s results weren’t announced until November 23 for the state’s U.S. House race. In both races, however, most of the first-round vote totals were public on election night or shortly after.

“Assuming this timeline is repeated for presidential election results, the initial votes will be publicly available on or shortly after election night and will be incorporated into media and other counts of the national popular vote. These early tabulations will include hundreds of thousands and perhaps millions of votes that, within the next few weeks, will be removed from the national tally for at least one of the two major party candidates, which could easily change the outcome in a close national election. It is not difficult to predict the chaos, confusion, and crisis that would ensue if a candidate initially thought to have won under NPV is suddenly determined to have lost weeks later after having hundreds of thousands of votes erased from their national totals.”[679]

This testimony is mistaken in several ways.

First, there was nothing noteworthy—much less scandalous—about Maine taking eight days in 2022 to finish its RCV count in the competitive 2^nd congressional district. For example, in 2020, Pennsylvania took until the Saturday after Election Day to finalize its non-RCV count for President in 2020.[680]

Moreover, Alaska’s official count is not slow because of RCV. It has always been slow—in part because ballots must be physically transported from remote areas. For example, it took until November 17, 2008, to ascertain that Mark Begich won Alaska’s U.S. Senate race in a non-RCV election.[681]

Second, any slowness in releasing unofficial vote counts from an RCV state will have no effect on the operation of the National Popular Vote Compact, because the Compact uses only the official certified counts. Official results for President must be certified by the same federal deadline whether RCV is used or not.

Third, there is no technological or other reason why the unofficial counts from RCV elections cannot come out as promptly as the unofficial results of non-RCV elections. In many RCV jurisdictions, unofficial counts are already released on Election Night and updated on the same schedule as unofficial counts from non-RCV races.

RCV has only been used at the state level in Maine since 2018, and in Alaska since 2022. Election administrators established understandably cautious schedules in conducting their first few statewide elections under RCV. Any slowness in producing unofficial preliminary or official certified votes count is a transitory issue—not an inherent characteristic of RCV.

Fourth, no vote count is official until all ballots are received, validated, and counted. The delays in releasing unofficial counts of some recent elections on Election Night or shortly thereafter are not attributable to RCV, but instead to the understandable caution of election administrators in conducting their first few statewide elections under RCV. These delays in releasing unofficial counts stem from administrative decisions to withhold all unofficial counts until all absentee ballots arrive and get counted, and until all provisional ballots are validated and counted. In any event, official certified results for RCV and non-RCV races take the same amount of time.

Footnotes

Sean Parnell, Senior Legislative Director of Save Our States, has repeatedly tried to get attention for his incorrect arguments about the ambiguity of RCV-for-President laws (section 9.27.1) by making hyperbolic claims about “hundreds of thousands” or “millions” of votes being in jeopardy.[682]

First, it is important to recognize that the pool of votes being discussed by Parnell is not:

• the total number of voters in the state;
• the total number of votes received by the major-party candidates; or
• the total number of votes received by the minor-party candidates.

Instead, it is the considerably smaller number of minor-party ballots that would get redistributed when a minor-party candidate is eliminated in early rounds of RCV counting.

To get a picture about how few votes are involved in this discussion, note that only four minor-party presidential candidates received more than 1% of the national popular vote during the six presidential elections between 2000 and 2020:

• 1% for Jo Jorgensen in 2020,
• 1% for Jill Stein in 2016,
• 3% for Gary Johnson in 2016, and
• 3% for Ralph Nader in 2000.[683]

Recently, opponents of RCV and National Popular Vote (NPV) have frequently cited Alaska as a place where a question of statutory interpretation of RCV counting procedures might conceivably matter (even though the Alaska Supreme Court has already settled the issue).

We can get a rough idea of the magnitude of the difference between the first-round RCV tally and the final-round RCV tally by examining Alaska’s vote count in 2016—a year when minor-party candidates received unusually large numbers of votes.

Of the 318,608 votes cast in Alaska in 2016, Libertarian Party nominee Gary Johnson received 18,725, and Green Party nominee Jill Stein received 5,735.

For the sake of argument, let’s assume that all minor-party voters gave all of their second-choice rankings to a major-party candidate.

In particular, let’s assume that Hillary Clinton would have received 100% of the second choices of Green Party voters, and that Trump would have received 100% of the second choices of Libertarian Party voters.

The net difference between 18,725 and 5,735 is 12,990. That is, Trump’s 46,033-vote margin over Hillary Clinton in Alaska would have increased by 12,990 votes under these assumptions.[684]

These 12,990 votes represent 0.00008 of the total of 158,224,999 votes cast in the 2020 presidential election.

Using standard statistical methods,[685] the probability that 12,990 votes would change the outcome of a nationwide election is 1-in-605. Since presidential elections are conducted every four years, this would mean that 12,990 votes might matter once in 2,420 years.

Second, keep in mind that when one candidate wins an absolute majority of first choices (as happened in Maine in 2020), the RCV counting process generally stops immediately.[686] That is, the first-round count is equivalent to the final-round count, and therefore there is no possible claim of uncertainty about what count to use.

In fact, this outcome is what is most likely to occur in practice.

One presidential candidate won an absolute majority of the state’s popular vote in an average of 45 states in the six presidential elections between 2000 and 2020—that is, in 90% of the cases.

In particular, one candidate won an absolute majority:

• in all but five states in 2020[687]
• in all but 12 states in 2016[688]
• in 100% of the states in 2012
• in all but four states in 2008[689]
• in all but three states in 2004[690]
• in all but nine states in 2000.[691]

That is, even if all 50 states were to enact an RCV-for-President law, the first-round count of RCV votes would, based on history, likely be equivalent to the final-round count in an average of 90% of the states.

In short, even if there were any legitimate ambiguity in RCV-for-President laws (and there is not), this issue is unlikely to be outcome-determinative in any presidential election.

Footnotes

A group called “Keep Our 50 States” wrote to the Minnesota Senate Committee hearing the National Popular Vote Compact (SF538) on January 31, 2023, saying:

“From a process standpoint, NPVIC is the legislative equivalent of inserting first-generation hybrid technology into your brand new Tesla and expecting it to transition seamlessly to the flying cars of the future. The first state Compact legislation passed over 15 years ago and there’s been no effort to bring those bills into compliance with other changes in state election standards. Compact states that enact Ranked Choice Voting, for example, face a legal hornet’s nest if the Compact ever takes effect.”

Although RCV was not used at the state level by any state at the time when the National Popular Vote Compact was written in 2004 and 2005, it is not true that the Compact does not accommodate RCV.

In fact, leading supporters of RCV worked closely with the National Popular Vote organization in writing the Compact to ensure that the Compact would be compatible with RCV.

FairVote was the first organization to endorse the National Popular Vote Compact and is widely recognized as the nation’s leading advocacy group for RCV dating back to its inception in 1992. Its founding Chief Executive Officer, Rob Richie, was a co-author of the Compact and was a co-author of this book, starting with its first edition in 2006. Richie spoke at NPV’s first press conference announcing the launch of the book and the Compact in 2006.

The Compact anticipated the possibility that states would adopt innovative voting systems, such as RCV, in the future. Accordingly, the Compact was silent as to how a future RCV state’s presidential vote count would be tabulated and, more importantly, it explicitly made each state’s determination of its presidential vote count “conclusive” on the states belonging to the Compact.[692]

Footnotes

Sean Parnell is the Senior Legislative Director of Save Our States—an organization that lobbies against both the National Popular Vote Compact (NPV) and ranked choice voting (RCV).

Rob Richie is the founding Chief Executive Officer of FairVote, the leading advocate for RCV since the 1990s. Richie was lead author of an article discussing RCV and NPV in the Harvard Law & Policy Review that was written in 2019 and published in 2020.[693]

In his written testimony to the Maine Veterans and Legal Affairs Committee on May 17, 2023, Parnell quoted from page 159 of Richie’s law review article, saying:

“I will note that lobbyists for NPV claim that it’s not possible for there to be any conflict between the compact and RCV because the nation’s leading proponent of RCV (Rob Richie, president of FairVote) helped write the compact. This ignores a 2021 paper on this issue that Richie served as the lead author of, which noted:

‘As currently drafted, the [NPV compact] seems to assume a plurality system…. [U]sing RCV for Presidential elections in states might seem incompatible with [NPV]. Most fundamentally, which votes should be reported out for the purpose of [NPV]? Would it be the first choices among all the candidates? Or would it be the final “instant runoff” totals after the RCV tallies are completed? If that latter choice were made, what if one of the two strongest national candidates was eliminated during the RCV tally in a given state?’”

“That paper came out in August 2021, months after Maine changed its law.”[694] [Emphasis added]

The deceptive nature of Parnell’s written testimony to the Maine committee becomes apparent when one realizes that Richie’s law review article was written in 2019 and completed in 2020—not August 2021.

That is, Richie’s article was written and completed long before the law that the Maine Legislature debated and passed in spring 2021 and that the Governor signed in June 2021.

By inaccurately attributing an August 2021 date to Richie’s 2019–2020 article, Parnell gave the impression that Richie was saying that there was a conflict between RCV and the National Popular Vote Compact after passage of Maine’s law in June 2021.

Richie raised the above rhetorical questions in his 2020 article, and he then answered his own rhetorical questions in the remainder of the article. Richie never stated that the National Popular Vote Compact should not be implemented.

The spring 2021 hearing of the Maine Veterans and Legal Affairs Committee considered the issues raised in Richie’s 2020 article.

These rhetorical questions were then answered when the Maine Legislature passed, and the Governor signed, a law on June 17, 2021, specifying that the state’s Certificate of Ascertainment would contain the result of the RCV tabulation from the final round.[695]

Note that Parnell’s deceptive forward-dating of Richie’s 2020 article was not accidental or inadvertent.

• Parnell’s own footnote to Richie’s article in his 2023 written testimony to the Maine Legislature conspicuously omits any date for Richie’s published 2020 article.
• The date could not possibly have been inadvertently overlooked. It appears right at the top of page 159 of the 2020 article that Parnell quotes.[696] Figure 9.16 shows the “2020” date at the top of page 159 of Richie’s 2020 article. Parnell’s quotation comes from paragraph 2 on page 159.

Footnotes