Tag Archives: ranked choice voting
The Speakership Saga and the Need for Nonpartisan Primaries
For those of us concerned about the capacity of Congress to function on behalf of the American people, we should be glad that the House of Representatives finally has a new Speaker three weeks (and a day) after Kevin McCarthy was removed from the speakership by an unprecedented motion to vacate. But for those of us concerned about the extent to which the composition of Congress is genuinely representative of the voters who elect its members, the removal of McCarthy and his eventual replacement by Mike Johnson is confirmation that the nation’s electoral system urgently needs fundamental reform.
Specifically, the ordeal of the last few weeks demonstrates that the existing system of partisan primaries under current conditions of asymmetrical polarization–where Republicans have moved further to the right to the point that MAGA extremism is the dominant faction within the party–causes extreme Republicans to win their party’s primary against a more moderate Republican alternative and then, in a Republican-leaning district, defeat the Democrat in the general election. But a majority of all the district’s voters in the general election–Republicans, Democrats, and independents–would have preferred the more moderate Republican candidate defeated in the primary over the MAGA extremist sent to Congress to represent the district.
Arizona’s second congressional district illustrates this problem and how it affects the House’s current deficiencies. In last year’s midterm elections, a Trump-endorsed MAGA extremist—Eli Crane—won the Republican primary with only 36% of the vote in a race with seven candidates. The runner-up with 24%, Water Blackman, was a more traditional Republican serving in the state’s legislature who, unlike Crane, repudiated Trump’s election denialism.
In the general election, Crane beat the Democrat, 54% to 46%. Had Blackman been the Republican nominee, he almost certainly would have done even better, pulling some more moderate voters away from the Democrat. In any event, it’s clear that a majority of all the district’s voters—Democrats, independents, and Republicans—would have preferred Blackman to be their representative in Congress rather than Crane.
Yet Mr. Crane went to Washington, where he became one of the eight Republicans who voted to vacate Kevin McCarthy from the speakership and thus precipitated the governance crisis in the House. If Blackman had been the one sent to Congress to represent the district, there is no reason to think that he would have participated in this obstructionist stunt.
Continue reading The Speakership Saga and the Need for Nonpartisan Primaries“Can deliberation cure our divisions about democracy?”
Boston Globe opinion essay by James Fishkin and Larry Diamond describing the results of their deliberative poll on various electoral reforms. The whole essay is a “must read”; here’s a bit of it:
“Partisan differences seem immovable, some say even “calcified.” However, in our recent national experiment “America in One Room: Democratic Reform” we find a very different picture. When Americans take the time to talk to each other in a civil, evidence-based way, they learn to listen to each other and often change their views dramatically, depolarizing across partisan divides. This Helena project shows thatAmericans’views are not so deeply entrenched but very much open to reason. …
“When our nationally representative sample of 600 (selected by NORC at the University of Chicago) deliberated for a weekend about these issues, Republicans often moved significantly toward initially Democrat positions and Democrats sometimes moved just as substantially toward initially Republican positions. The changes were all consonant with basic democratic values, such as that everyone’s vote should count and that our elections need to be administered in a nonpartisan way.
“Only 30 percent of Republicans initially supported providing access to voter registration online, but after deliberations, Republicans moved to majority support, joining Democrats (who overwhelmingly supported it at 81 percent). Republicans abandoned their view that “increasing opportunities for voter registration would open up more opportunities for voter fraud.” The percentage of Republicans believing that dropped 26 points (from 56 percent to only 30 percent). Republicans similarly abandoned their opposition to restoring federal and state voting rights to convicted felons upon their release from prison. Republican support for restoring voting rights for felons increased dramatically, from 35 percent to 58 percent.
“By contrast, only a minority of Democrats (44 percent) initially supported the mostly Republican proposal for each state to require its voting jurisdictions to conduct an audit of a random sample of ballots “to ensure that the votes are accurately counted.” However, after deliberation Democrats increased their support by 14 points to 58 percent (joining Republicans who supported it even more strongly at 75 percent). Getting an accurate count was a goal broadly shared, and audits with a random sample seemed a practical method of assuring everyone about the results. The same goal was served by another, initially Republican, proposal “that all voting machines produce a paper record of the vote that the voter verifies and then drops in a ballot box.” Democratic support rose from 44 percent to 55 percent, with Republican support staying at over 70 percent. …
“Deliberators were asked about ranked choice voting for primaries and general elections for Congress, state legislatures, and local elections. In all six cases, there was an overall majority in support of ranked choice voting after deliberation. While the weakest support came from Republicans, for most ranked choice voting options at least a third of Republicans were favorable. For the option of “final four or five voting” (using a nonpartisan primary to choose the top few candidates to advance to a general election using ranked choice voting), Republican support increased after deliberation from 32 percent to 43 percent. There was general dissatisfaction both before and after with our current “first past the post” method of conducting elections (after deliberation, only 33 percent of the overall sample wanted to keep it, but that included about 50 percent of Republicans).”
“Does Ranked Choice Voting Promote Legislative Bipartisanship? Using Maine as a Policy Laboratory”
Rachel Hutchinson of FairVote and Ben Reilly of the East-West Center have posted this new paper on SSRN. Here’s the abstract:
Political polarization in the United States has increased dramatically, hampering the functioning of American government. Some scholars attribute this dynamic to the use of plurality elections and posit that a ranked choice voting (RCV) system may promote greater bipartisanship. Maine’s 2016 adoption of RCV presents an early opportunity to test this theory on congressional races. Using comparative analysis, we show that bipartisan cosponsorship increased after the adoption of RCV in Maine’s swing House district but not in its safe district. These results, along with some anecdotal evidence from Alaska, which introduced RCV in 2021, provide an early suggestion that RCV may be more likely to prompt bipartisanship in competitive races than in safe seats. RCV would need to be implemented in more U.S. states and for a longer time to confirm its effectiveness in cultivating bipartisanship, and may work best in combination with other electoral and procedural reforms.
I very much look forward to reading this as it addresses an important question about the capacity of RCV to redress the problems of governance (which Rick Pildes has highlighted) associated with polarization (and fragmentation).
“Could a Third Party Finally Do It?”
Gerald Seib in the Wall Street Journal with an extensive analysis on both the recent history and current polarized climate that sets the stage for the prospect that a third-party campaign could roil the 2024 presidential election:
“A Pew analysis of congressional voting last year illustrated the trend. It found that, on average, Democrats and Republicans in Congress are farther apart ideologically now than at any time in the past 50 years. Today there are only about two dozen lawmakers who could be identified as “moderate,” compared with more than 160 a half a century ago. An analysis of congressional votes showed that both parties have moved away from the center, though Republicans have moved significantly farther to the right than Democrats have to the left.
“The balance of power between the two parties actually seems to be exacerbating the trend. The political system appears trapped in an endless loop of frustration, in which control bounces back and forth between Republicans and Democrats and neither can break out to implement a clear agenda. Each of the last five biennial elections has been a “change election,” in which control of at least one branch of the national government—the House, Senate or White House—has changed hands. Meanwhile, the anger stoked on social media alienates many Americans and drives them to the extremes of the ideological spectrum.
“It might seem that this balance of power would prompt the two parties to work together in the center, but it has had the opposite effect: With little to no margin for error, leaders of both parties seem desperate to hang onto their most reliable, ideologically motivated base voters, and afraid to take the chance of offending those base voters by compromising too much with the other side. …
“No Labels leaders insist that they see not just a chance to run but to win, based on polling the organization has done nationally and, in recent days, in eight presidential battleground states. Dritan Nesho, the group’s pollster, says that battleground-state polling found that between 60% and 70% of voters in swing states said they would be open to considering a moderate, independent presidential ticket if the main-party choice is between Joe Biden and Donald Trump, similar to the sentiment found nationally. ‘Our models suggest that if we were able to convert about six in 10 of these voters we would win the Electoral College outright with 286 Electoral College votes,’ he says.” …
The group’s leaders say they won’t necessarily field a ticket if they don’t see a chance for victory—or if their efforts seem to be succeeding in moving the major parties toward the middle. ‘If they force one or both parties more towards the center of the country, if they force the political system to, for a change, actually speak to the needs of the common-sense majority instead of the wants of their bases, and that closes off an opening for a No Labels ticket, that’s fine with us,’ says Ryan Clancy, the group’s chief strategist. ‘That is success.’ The group’s leaders also say they have no interest in being a Trojan Horse that helps Trump win.”
“… the winner-take-all nature of the presidential election system, under which no Electoral College votes are rewarded for just coming close to winning a state, makes the task even harder.”
I have made it clear in numerous previous ELB blog posts why a third-party presidential bid is so dangerous unless and until states adopt ranked-choice voting, or another majority-winner system, for awarding their electoral votes.
Primary Turnout is the Real Philadelphia Story
As reported in the Philadelphia Inquirer, the most encouraging estimates are that turnout in Philadelphia’s mayoral race will be in the “200,000s or low 300,000s.” Based on competitiveness and Democratic registration advantage in the city, we can assume most of those votes will be cast in the Democratic primary, and split between four to five competitive candidates.
After reading this, I wanted some hard numbers on eligible voters in the city. The only easily available number is the number of registered Democrats in Philadelphia: 775,596 according to the Pennsylvania Department of State. The total number of registered voters in Philadelphia is 1,025,223, according to the same source. Using those numbers, I estimate that Philadelphia’s next Mayor is likely to have been decided by about 5-8% of the city’s registered voters–and that is probably an optimistic projection (one that is fairly close to the estimate offered by the political consultant quoted in the Inquirer).
The Philadelphia Inquirer concludes:
“If the total number of votes the nominee gets falls below 100,000 — a real possibility — that would be the lowest winning raw vote total since at least the 1970s and would represent less than 10% of all registered voters, according to an Inquirer analysis of past election data.”
I hate to say it but Philadelphia needs rank choice voting.
“Beyond the Spoiler Effect: Can Ranked Choice Voting Solve the Problem of Political Polarization?”
Nate Atkinson, Scott Ganz and I have co-authored a new article, to be published in the Illinois Law Review, and now posted on SSRN in its current draft. The article is based on computer simulations that Nate and Scott conducted comparing (1) Alaska’s new “top four” electoral system, which uses the “instant runoff” form of ranked-choice voting, and (2) a variation on Alaska’s system that substitutes a “Condorcet-compliant” method of ranked-choice voting instead of the instant runoff version. A “Condorcet-complaint” electoral system is one that elects a candidate whom a majority of voters prefers to each other candidate when candidates are paired against each other head-to-head.
Nate and Scott created a profile of the electorate of each state, using Cooperative Election Study data. Interestingly, states differ significantly in the degree to which their electorates are polarized. Some states, for example, have a bimodal distribution of voters as a result of polarization. Other states, by contrast, have a distribution of voters that is unimodal, resembling a more conventional “Bell Curve”.
Nate and Scott then simulated 250,000 four-candidate elections for each state, using both “instant runoff” and “Condorcet-compliant” versions of ranked-choice voting. The results are striking. In all states, Condorcet-compliant RCV tends to elect a candidate whose position in the distribution of the electorate is closer to the electorate’s median voter than instant-runoff RCV will. The difference between the two forms of RCV, however, is significantly greater in those states with a polarized bimodal distribution of voters. In those states, the gap between the instant runoff winner and median voter compared to the corresponding gap between the Condorcet winner and the median voter is much larger.
The paper includes graphs showing these results for all fifty states. As an illustration, here (from left to right) are the electorate profile, instant runoff winners, and Condorcet winners for Arizona, one of the states with a polarized bimodal distribution:
The Problem with Plurality-Winner Elections
Election Law at Ohio State is pleased to present a one-hour webinar on the issue of electoral system design: The Problem with Plurality-Winner Elections – And Can Requiring Majority Winners Help Save Democracy? It’s scheduled for Friday, November 19, at noon ET.
Steve Huefner will moderate the discussion. I’ll present some research we’ve been doing here at Ohio State on this topic. Franita Tolson and Derek Muller, both familiar to readers of this blog, with comment as panelists. We are are delighted that Rachel Kleinfeld, Senior Fellow at the Carnegie Endowment of International Peace, who has been studying the relationship of electoral systems to political conflict, will also participate as a panelist.
The webinar will examine the role that the plurality-winner rule for congressional elections has in causing incumbents, like Sen. Rob Portman and Rep. Anthony Gonzalez (both of Ohio), to decide against running for reelection–regardless of being popular with their constituents–just because they have become out-of-step with their own party. The webinar will also explore whether alternative majority-winner electoral systems, like various versions of Ranked Choice Voting, might improve representation and reduce the risk of democratic decline in the U.S.
To register for the webinar (and more info), please click here.
Breaking round-robin ties
The USA women’s soccer team is advancing out of the preliminary round-robin stage of competition based on a “goal differential” tiebreaker used to differentiate teams with the same win-loss records. “The United States finished second in Group G on the strength of its plus-3 goal differential,” the Washington Post reports. I mention this here because a similar “vote differential” statistic is used to break ties among candidates in Round-Robin Voting. One reason for developing Round-Robin Voting as an electoral system is its straightforward comparison to round-robin competitions in sports. If the public can understand how a round-robin tournament works in the Olympics, the public can understand how round-robin competition would work in an election. The ability of the public to understand its own electoral system is an important feature of a democracy, including the public’s willingness to adopt the system in the first place. Obviously, there are other important criteria in making a choice among alternate electoral systems, including those directly relevant to how each system translates voter preferences into an overall winner entitled to govern in the name of the people. Still, as this year’s Olympics unfold, it’s a useful reminder that the idea of Round-Robin Voting should not be ruled out solely on the ground that it is too complicated for voters to understand.
Trump’s efforts to “primary” Cheney
Politico goes deep into the details of former president Trump’s concerted plans to oust Rep. Liz Cheney as revenge for her commitment to honest vote-counting. Particularly interesting are the efforts to change the rules to make it more difficult for Cheney to win:
Underscoring the urgency, Donald Trump, Jr. earlier this year threw his support behind legislation that would change Wyoming election law to make it harder for Cheney to win against a splintered field. The proposal would have implemented a runoff if no primary candidate received a majority of support in the first round of voting, thereby forcing Cheney into a one-on-one matchup against a Trump ally.
The legislature, however, voted down the bill in March. Since then, some state lawmakers have pursued other election law changes that would hinder Cheney’s prospects.
Two of the potential rule changes under consideration according to an article linked to by Politico are ranked-choice voting and a California-type nonpartisan top-two primary. I don’t know if either of those moves would be successful in blocking Cheney; much as with Senator Lisa Murkowski’s situation in Alaska, it would depend on what percentage of voters view Cheney as their first-choice preference. As I discuss in my paper on Round-Robin Voting, both Instant Runoff Voting (what “ranked choice voting” usually is) and California’s top-two system privilege the electorate’s first-choice preferences in comparison to all the preferences that voters have among the candidates in the race. In other words, if many voters really don’t want a candidate to win, perhaps because they view the candidate as authoritarian and dangerous to democracy, that preference (no matter how strong) will be downplayed in either Instant Runoff Voting or California’s top-two system.
Round-Robin Voting, by contrast, does not privilege (or downplay) any of the preferences that voters have among the various candidates and thus will treat a preference that a candidate lose equivalently to a preference that a candidate win. Without looking more closely at the Wyoming race, I can’t be more confident of my assessment, but I’m inclined to think that Cheney would fare much better in a system with Round-Robin Voting than under the current system or under either Instant Runoff Voting or California’s top-two system. The same point applies to Murkowski. The basic reason is that voters who don’t prefer Cheney or Murkowski as their first choice, but really don’t want a Trump-endorsed candidate to win, might rank Cheney or Murkowski second (and enough voters who prefer the Trump-endorsed candidate first might hold their noses and still prefer Cheney or Murkowski to a Democrat). If this is true, then Cheney and Murkowski can win the head-to-head matches that form the Round-Robin Voting competition, even if they would not have enough first-choice strength to prevail under either Instant Runoff Voting or California’s top-two system.
Thus, as one considers how the choice of an electoral system may favor Trump’s midterm efforts to purge Cheney, Murkowski, and others in his quest for a return to power, one should consider the same question in reverse: what electoral system would best protect against Trump’s authoritarian-style populism? I submit that a version of Round-Robin Voting has the best prospect of serving that democracy-protection purpose (although that hypothesis should be tested empirically with whatever data can be mustered for the task).
Tournament Elections with Round-Robin Primaries
I’ve posted a draft of this paper on SSRN. Here’s the abstract:
Round-robin voting uses ranked-choice ballots but calculates which candidates are most preferred by a majority of voters differently from instant-runoff voting. Like a round-robin sports competition, round-robin voting determines how each candidate fares against every other candidate one-on-one, tallying the number of wins and losses for each candidate in these one-on-one matchups. If necessary to break a tie in these win-loss records, round-robin voting looks to the total number of votes cast for and against each candidate in all of the one-on-one matchups—just as round-robin sports tournaments look to an equivalent total point differential statistic to break ties. When used in a primary election as the method to identify the top two candidates deserving to compete head-to-head as finalists in the general election, comparable to the use of round-robin competition as the preliminary stage of a sports tournament, round-robin voting is the electoral system best able to implement the democratic idea of majority rule.
I was delighted to have the opportunity to present an earlier draft at the University of Wisconsin Law School’s “Public Law in the States Conference” on June 23, and I’m looking forward to working with the Wisconsin Law Review on preparing the paper for publication. This draft will be revised before submission to the law review’s editors at the end of August, and therefore I very much welcome any comments that readers might email me before then.
J.D. Vance and the pressures of primaries
In a remarkably candid TIME interview with Molly Ball after announcing his candidacy in Ohio’s 2022 U.S. Senate election, for the seat being vacated by Rob Portman, J.D. Vance acknowledged that he’s pandering to ex-president Trump because that’s the only way to be viable in the GOP primary: “I need to just suck it up and support him.”
It’s believed that Portman abandoned his Senate seat, despite remaining popular among Ohio’s general election voters, in part because he didn’t want to pander to Trump for the sake of winning the primary. In this respect, Portman is in the same position as Roy Blunt of Missouri, Richard Burr of North Carolina, Richard Shelby of Alabama, and Pat Toomey of Pennsylvania, all non-Trump Republicans giving up their Senate seats.
Vance’s comment, as a kind of exclamation point on this troubling trend, vividly illustrates one of the main observations of recently released “The Primary Problem” report from Unite America: the existing system of partisan primaries, followed by plurality-winner general elections, not only affects which candidate ultimately holds office but also how candidates choose to campaign and then act in office in order to avoid being “primaried” when running for reelection. The distortion of representation is pervasive as a result of the particular institutional arrangement in which candidates compete. (An even newer report, from New America, reaches a similar assessment–“primaries incentivize more polarizing behavior among candidates and legislators”–although it is cautious in its conclusions on how best to address the issue.)
I wonder, therefore, what kind of candidate J.D. Vance would have attempted to be if he were running, not in the current system, but instead in the system of “Round-Robin Voting” that the Election Law at Ohio State program is developing. This system, which involves a variation on ranked-choice voting, requires candidates to compete one-on-one against all other candidates for the office regardless of party affiliation, to determine which candidate is most preferred by a majority of the entire electorate. If you watch until the end of this 15-minute video explaining Round-Robin Voting, you’ll see that it hypothesizes an “Opportunist” candidate attempting to position himself (or herself) in between a Trumpian Populist (like Josh Mandel, already running for this Ohio U.S. Senate seat) and a traditionally Conservative Republican (like Rob Portman). Does J.D. Vance exemplify this “Opportunist” candidate, and how would he fare in a Round-Robin Voting nonpartisan primary? Also, would Portman have run for reelection if he had been able to do so in a Round-Robin Voting nonpartisan primary?
Rather than answering these questions definitively right now, it’s instead worth keeping them in mind as the 2022 midterm campaigns unfold. The big-picture point: as much as the changing nature of American politics is caused in part by changes in voter preferences, it is also significantly a product of the particular system in which politicians operate. If that system artificially magnifies increasingly extremist tendencies in what voters want, it’s necessary to alter the system itself to undo that dangerous magnification of extremism.